E Motif ET V6 - M.MOAM.INFO (2025)

Oct 13, 2015 - Dr. Judah Folkman. The same .... tutions, Dana-Farber Cancer Institute, Boston;. University ... 15 patients enrolled in Dana-Farber Cancer Insti-.

E Motif ▶ E-Box

E T V6 Stefan K. Bohlander Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand

Synonyms ETS variant gene 6

rearrangements include balanced translocations with a great number of different chromosome partner bands as well as unbalanced translocations and deletions. The latter two rearrangements lead to the loss of genetic material from 12p. Molecular cytogenetic studies showed that more than half of the observed balanced translocations of 12p have breakpoints that involve the ETV6 gene. There are currently more than 40 different 12p translocations described that involve the ETV6 gene. ETV6 was originally identiﬁed as the fusion partner of the ▶ platelet-derived growth factor receptor beta gene (PDGFRB) in a balanced t(5;12)(q31:p13) translocation from a case with chronic myelomonocytic leukemia (CMMoL). Initially, ETV6 was called TEL (translocation ets leukemia gene) but was later renamed as ETV6 (ets variant gene 6) to avoid confusion with the abbreviation for telomere.

Definition A gene encoding an ets domain transcription factor located on human chromosome 12 band p13. It is a frequent target of ▶ chromosomal translocations.

Characteristics Discovery The short arm of chromosome 12 is a hot spot for chromosomal rearrangements in diverse types of hematological malignancies. These # Springer-Verlag Berlin Heidelberg 2017 M. Schwab (ed.), Encyclopedia of Cancer, DOI 10.1007/978-3-662-46875-3

Protein Domains ETV6 is a member of the ets (E-26 transforming speciﬁc) family of transcription factors. All ets family proteins share a very conserved protein domain of about 88 amino acids in length, the so-called ets domain (Fig. 1). The ets domain is a sequence-speciﬁc DNA-binding domain but also mediates protein-protein interaction. It is evolutionarily highly conserved and found in invertebrates such as Drosophila and C. elegans. The ets domain of ETV6 is more closely related to the ets domain of the Drosophila protein yan than to ets domain of the human ETS1 or SPI1 (PU.1) genes.

1446

E T V6

“Unproductive fusions”

Intron 1: BTL ACS2 TTL PER1-AS

Intron 2: MN1 MDS1-EVI1 BTL CDX2 PAX5 HLXB9 MDS2 STL

Intron 3: MN1

Intron 4: ARNT

Intron 5: RUNX1

Central domain

Pointed

ETV6

Transcription factors

ets 452aa

1 Intron 4: PDGERB ABL1 NTRK3 JAK2 Protein tyrosine kinases

Intron 5: JAK2 ABL1 ABL2 JAK2 NTRK3 FGFR3 SYK

E T V6, Fig. 1 Diagrammatic representation of the ETV6 protein with the position of the breakpoints of the various fusion partner genes

The other evolutionarily conserved domain in ETV6 is the N-terminally located pointed or SAM (sterile alpha motif) domain. This domain is even more highly conserved in evolution than the ets domain and is found in many ets family members as well as in many other transcription factors and signal transduction proteins. The pointed domain serves as a homo- and heterodimerization module. ETV6 Fusion Partners in Cancer After the initial cloning of ETV6, several other fusion partners of ETV6 were identiﬁed in quick succession. There are now well over 20 fusion partners of ETV6 described, and the list is still growing. The various ETV6 fusions can be assigned to three groups: (i) protein tyrosine kinases (PTK), (ii) transcription factors and others, and (iii) “unproductive” fusions, i.e., fusions that do not result in an obvious fusion protein. Protein Tyrosine Kinase Fusion Partners of ETV6 The ﬁrst identiﬁed fusion partner of ETV6 was the protein tyrosine kinase (PTK) platelet-derived

growth factor receptor beta gene. The ETV6/ PDGFRB fusion protein is a constitutively active PTK and is the critical product of this translocation. In the ETV6/PDGFRB fusion protein, the N-terminal portion of ETV6, which includes the pointed domain, is fused to the C-terminal two thirds of the PDGFRB protein, which includes the tyrosine kinase domain of PDGFRB. This general structure, i.e., the pointed domain of ETV6 in the N-terminal half and the tyrosine kinase domain of the fusion partner in the C-terminal half of the fusion protein, is characteristic of all ETV6/PTK fusions. Many studies have shown that the pointed domain of ETV6 serves as a dimerization module for the fusion protein. Dimerization of the fusion protein leads to the constitutive activation of the PTK domain, which results in autophosphorylation of the fusion protein as well as phosphorylation of cellular proteins like rasGAP, Shc, SH-PTP2, SH-PTP1, CRK-L, CBL, paxillin, and STATs. Expression of ETV6/ PTK fusions in the interleukin 3-dependent hematopoietic cell line Ba/F3 leads to factorindependent growth. Several ETV6/PTK fusion proteins have also been assayed in murine bone

E T V6

1447

E T V6, Table 1 Translocation partners of ETV6 ETV6 fusion Translocation partner Tyrosine kinases PDGFRB t(5;12)(q31;p13) ABL1 t(9;12)(q34;p13) ABL2 t(1;12)(q25;p13) JAK2

t(9;12)(p24;p13)

NTRK3

t(12;15)(p13;q25)

FGFR3

t(4;12)(p16;p13)

▶ SYK t(9;12)(q22;p12) Transcription factors and cofactors RUNX1 t(12;21)(p13;q22) MN1 t(12;22) (p13;q11) ARNT t(1;12)(q21;p13) “Unproductive fusions” MDS1/EVI1 t(3;12)(q26;p13) BTL t(4;12) (q11-q12;p13) CDX2 t(12;13)(p13;q12) PAX5 t(9;12)(q11;p13) HLXB9 t(7;12)(q36;p13) MDS2 t(1;12)(p36.1;p13) STL t(6;12)(p13;q23) ACS2 t(5;12)(q31;p13) TTL PER1 antisense

t(12;13)(p13;q14) t(12;17)(p13;p13)

Disease

CMMoL AML, ALL AML-M3,-M4, T-ALL Pre-B-cell ALL, T-ALL Congenital ﬁbrosarcoma, mesoblastic nephroma, secretory breast carcinoma, AML Peripheral T-cell lymphoma MDS ALL AML and MDS AML-M2 MDS, CML AML-M0 AML ALL AML MDS B-ALL MDS-RAEB, AML, AEL ALL AML

marrow transplantation or transgenic mouse models where they lead to various hematological diseases like myelo- or lymphoproliferative syndromes. The different protein tyrosine kinase genes that have been found to be fused to ETV6 are listed in Table 1. It should be noted that most of these fusions are rather rare. For several partners just one or a handful of cases have been described in the literature. ETV/PTK fusions are not only found in various types of leukemia, but the ETV6/NTRK3

fusion has also been found in solid tumors such as congenital ﬁbrosarcoma, mesoblastic nephroma, and secretory ▶ breast cancer. Transcription Factors and Other Fusion Partners of ETV6 There are only two fusions of ETV6 with non-PTKs for which the transforming potential of the fusion protein could be established. This is the very common ETV6/▶ RUNX1 and the much rarer MN1/ETV6 fusion. The ETV6/RUNX1 Fusion

The ETV6/RUNX1 (TEL/AML1) fusion was the second fusion of ETV6 that was identiﬁed. It was soon recognized that the ETV6/RUNX1 fusion is the most common fusion gene found in childhood ▶ acute lymphoblastic leukemia, which is present in up to 25% of all childhood B-ALL cases. The ETV6/RUNX1 fusion protein, which is the critical fusion in this translocation, comprises the pointed domain of ETV6 as well as the DNA-binding and transactivation domain of RUNX1. Many ETV6/RUNX1-positive ALLs have interstitial deletions (detectable by ﬂuorescence in situ hybridization, FISH) of the short arm of the non-rearranged chromosome 12, which encompass the ETV6 locus. Even in cases in which no interstitial deletion of the non-rearranged ETV6 allele can be detected by FISH, there is no expression of the wild-type ETV6 allele suggesting a very important role for ETV6 loss of function in the pathogenesis of ALL. Little is known about mechanisms by which the ETV6/RUNX1 fusion protein causes leukemia. Several attempts to establish ETV6/RUNX1 transgenic or bone marrow transplant leukemia models have failed or yielded leukemia only after a very long latency. There is some evidence that leukemogenesis by ETV6/RUNX1 is accelerated if cell cycle regulation is compromised by other mutations. Some hints as to the function of ETV6/RUNX1 have been gleaned from reporter gene assays. In these experiments, the ETV6/RUNX1 fusion protein behaves as a strong transcriptional repressor

1448

on AML1 target genes. The pointed domain and the central domain of ETV6 have been shown to recruit transcriptional corepressors like SMRT, N-CoR, and mSin3A. The MN1/ETV6 Fusion

The MN1/ETV6 fusion is found in some patients with ▶ acute myeloid leukemia (AML) or ▶ myelodysplastic syndrome with a t(12;22) (p13;q11) translocation. MN1 was originally found as a gene disrupted by a translocation in meningioma. The critical MN1/ETV6 fusion protein, which is able to transform murine NIH3T3 ﬁbroblasts in vitro (NIH3T3 transformation assay), contains transcriptional activation domains derived from MN1 and the ets DNA-binding domain of ETV6. This is the only known example in which the ets domain of ETV6 seems to be critical for transformation. The ETV6/ARNT Fusion

Another rare ETV6/transcription factor fusion is the ETV6/ARNT fusion gene reported in one patient with AML and a t(1;12)(q21;p13), which links the N-terminal portion of ETV6 including the pointed domain to almost the complete ▶ aryl hydrocarbon receptor nuclear translocator (ARNT). “Unproductive” ETV6 Fusions A large number of chromosomal translocations involving ETV6 have been cloned, in which the breakpoints lie in intron 1 or 2 of ETV6. These translocations result in ETV6 fusions that contain only the 54 amino-terminal amino acids of ETV6 and lack any of the important protein domains of ETV6 like the pointed or ets domain (Table 1). Almost all of these translocations have been identiﬁed in only one or at the most a handful of leukemia cases. It is very likely that most, if not all, of these translocations do not result in the formation of a transforming ETV6/other fusion protein but rather that they lead to the transcriptional upregulation of genes adjacent to the translocation breakpoints. This has been elegantly shown for the t(12;13) (p13;q12) which results not only in the formation of a fusion between ETV6 and the caudal-related

E T V6

homeobox gene CDX2 but also in the upregulation of a transcript that codes for the full-length CDX2 protein. In a bone marrow transplant model, the upregulation of the wild-type CDX2 protein and not the expression of the ETV6/CDX2 fusion protein is critical for the development of leukemia in a murine bone marrow transplantation model. Putative Tumor Suppressor Gene and Physiological Function There are several lines of evidence that suggest that ETV6 might function as a ▶ tumor suppressor gene. In up to 70% of childhood ALL cases with ETV6/RUNX1 fusions, there is concomitant deletion of the non-rearranged ETV6 allele. Deletions of the short arm of chromosome 12 are frequently found in a broad spectrum of hematological malignancies, and the common region of deletion was mapped to a small genomic region including ETV6 and CDKN1B. In addition, several studies showed that even if no deletion of the ETV6 locus can be detected, there was no expression of ETV6 at the mRNA level or absence of the ETV6 protein. In vivo and in vitro studies provide additional evidence that ETV6 might function as a tumor suppressor gene. ETV6 expression inhibits growth in soft agar of RAS transformed NIH3T3 ﬁbroblasts cells and leads to the differentiation of erythroleukemia cells into erythrocytes. Additionally, the expression of ETV6 in serum-starved NIH3T3 cells induces apoptosis. Reporter gene assays have demonstrated that ETV6 is a strong transcriptional repressor, which requires corepressors like N-Cor, mSin3, and SMRT. Targeted deletion of the murine Etv6 gene demonstrated that Etv6 is essential for yolk sac angiogenesis and the establishment of deﬁnitive hematopoiesis in the bone marrow, while hematopoiesis in the yolk sac and fetal liver was not affected. Etv6/ mice die between day 10.5 and 11.5 of embryonic development due to a defect in yolk sac ▶ angiogenesis and widespread apoptosis of mesenchymal and neural cells. Furthermore, Etv6 could be shown to be an essential regulator for the maintenance of hematopoietic stem cells in adult murine bone marrow.

E2A-PBX1

References Bohlander SK (2005) ETV6: a versatile player in leukemogenesis. Semin Cancer Biol 15:162–174 Golub TR, Barker GF, Lovett M et al (1994) Fusion of PDGF receptor b to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell 77:307–316 Hock H, Meade E, Medeiros S (2004) Tel/Etv6 is an essential and selective regulator of adult hematopoietic stem cell survival. Genes Dev 18:2336–2341 Rawat VPS, Cusan M, Deshpande A et al (2004) Ectopic expression of the homeobox gene Cdx2 is the transforming event in a mouse model of t(12;13)(p13; q12) acute myeloid leukemia. Proc Natl Acad Sci U S A 101:817–822

E100 ▶ Curcumin

E11 Antigen ▶ Podoplanin

E2A-PBX1 Brandy D. Hyndman1 and David P. LeBrun1,2,3 1 Department of Pathology and Molecular Medicine, Queen’s University Cancer Research Institute, Queen’s University, Kingston, ON, Canada 2 Protein Function Discovery Group, Queen’s University, Kingston, ON, Canada 3 Division of Cancer Biology and Genetics, Cancer Research Institute, Queen’s University, Kingston, ON, USA

Definition E2A-PBX1 is an oncogenic transcription factor expressed in neoplastic cells consequent to a somatic chromosomal translocation (t(1;19)) in

1449

some cases of ▶ acute lymphoblastic leukemia (ALL).

Characteristics Leukemia cells commonly contain somatic chromosomal translocations that contribute to the induction and, presumably, perpetuation of the disease. They do this by altering proto-oncogenes that reside close to the involved breakpoints on the participating chromosomes. Translocation (1;19) (q23;p13.3) is the second most frequently observed recurrent translocation in ALL, detectable in approximately 5% of cases using conventional cytogenetics. In the vast majority of instances, t(1;19) fuses coding regions from two genes, E2A and PBX1, that reside respectively on chromosomes 19 and 1. Subsequent transcription and pre-mRNA splicing lead to expression of E2A-PBX1, an abnormal, chimeric transcription factor with potent oncogenic activity. Abundant evidence supports the notion that E2A-PBX1 contributes to the abnormal accumulation of primitive lymphoid progenitors that characterizes ALL by deregulating the transcription of key target genes. Clinical Aspects ALL is primarily a pediatric disease, so most patients with t(1;19)-positive and E2A-PBX1expressing leukemia are children. Although the neoplastic cells in ALL most often have immunophenotypic (▶ ﬂow cytometry) and genotypic (▶ molecular pathology) features characteristic of the very early stages of B-lymphoid development, prior to the initiation of immunoglobulin gene rearrangement or expression, the vast majority of cells associated with t(1;19) generally manifest characteristics typical of more mature “pre-B cells,” including rearrangement of the immunoglobulin heavy-chain gene locus and expression of cytoplasmic, but not surface, immunoglobulin heavy-chain protein. Other clinical features associated with t(1;19)-positive ALL include especially high leukocyte counts at presentation, non-Caucasian race, and central nervous system involvement. Although the translocation was originally associated with

1450

E2A-PBX1

AD1

AD2

bHLH

E12 HD PBX1 AD1

AD2

E2A-PBX1

E12/E47

E2A-PBX1

PBX/HOX/MEINOX Target gene

HOX

E2A-PBX1, Fig. 1 Structure and function of E2A-PBX1. (a) Translocation 1;19 results effectively in fusion of the amino-terminal two-thirds of the E2A proteins (this portion is identical in the E12 and E47 isoforms) with most of PBX1. The vertical line indicates the point of fusion. AD1 and AD2, transcriptional activation domains 1 and 2; bHLH, basic helix-loop-helix domain; HD, homeodomain.

(b) A model illustrating hypothetical mechanisms of neoplastic transformation by E2A-PBX1. The oncoprotein may deregulate the expression of critical genes normally controlled by PBX/HOX/MEINOX complexes, alter the function of transcriptional co-activators, or impair the function of wild-type E12/E47

unfavorable clinical outcomes, the more intensive treatment regimens used currently have largely or completely abrogated this association.

The PBX1 gene, at chromosome band 1q23, was identiﬁed through its involvement in the 1;19 translocation. Its protein products, PBX1a and PBX1b, contain a homeodomain, an evolutionarily ancient domain involved in DNA binding and protein-protein interactions (▶ homeobox genes). PBX1 binds to DNA and regulates the transcription of target genes in cooperation with other homeodomain-containing proteins of the HOX and MEINOX classes. These physical and functional interactions have important roles in embryonic development and tissue homeostasis. In particular, important roles in hematopoiesis are well documented.

Wild-Type E2A and PBX1 Gene Products The E2A gene resides at chromosome band 19p13.3 and encodes two proteins, called E12 and E47, which are generated by alternative splicing of exons (Fig. 1). E12 and E47 (or “E2A proteins”) possess a C-terminal basic helix-loophelix (bHLH) domain that mediates homo- or hetero-dimerization as well as binding to DNA at sites that contain the consensus sequence CANNTG (the ▶ E-box). E12 and E47 function as transcriptional activators, inducing the transcription of genes that lie in the general vicinity of the E-boxes to which they bind. The E2A proteins have important roles in regulating lineage-speciﬁc cellular differentiation; their contributions to various aspects of lymphocyte development have been especially well delineated.

Structure and Function of E2A-PBX1 The recombination of exons brought about by t(1;19) essentially fuses the amino-terminal two-thirds of the E2A proteins (a portion that is invariant between E12 and E47) to most of PBX1. The E2A-encoded portion includes two

EAP1

transcriptional activation domains capable of inducing target gene transcription by recruiting transcriptional co-activators (▶ chromatin remodeling), whereas the PBX1-derived portion includes the DNA-binding homeodomain. E2A-PBX1 can function as a potently transforming oncoprotein in several cellular lineages. For example, enforced expression induces lethal lymphoproliferative diseases in transgenic mice and aggressive myeloproliferative diseases in a murine bone marrow transplantation model. The impaired cellular differentiation and accelerated proliferation that are associated with E2A-PBX1 expression result from the cumulative or cooperative effects of physical or functional interactions between E2A-PBX1 and other macromolecules. The available experimental evidence supports a general model, the basis of which was originally suggested by the nature of the functional domains that are brought together in the oncoprotein (Fig. 1). Largely by means of its PBX1-encoded portion, E2A-PBX1 retains the ability to bind cooperatively with HOX proteins and cognate PBX/HOX binding sites on the DNA. This probably results in the abnormal expression of target genes whose transcription is normally regulated by PBX/HOX/MEINOX complexes. The retention of potent transcriptioninducing potential by the E2A portion of the oncoprotein and the documented involvement of HOX proteins (and, by implication, the target genes that they regulate) in normal and leukemia-associated hematopoiesis are consistent with this mechanism. In a perhaps complementary mechanism, the interaction with E2A-PBX1 may alter the function of transcriptional co-activator proteins, including the histone acetyltransferases p300 and CREB-binding protein (CBP) (chromatin remodeling in cancers), so as to promote neoplastic transformation in a manner at least superﬁcially analogous to oncoproteins encoded by DNA tumor viruses such as Simian virus 40. Finally, E2A-PBX1 may exert dominant inhibitory effects on the wild-type E2A proteins, as these possess tumor suppressor activities. The E2A locus is involved in another translocation, t(17;19)(q22;p13), seen relatively rarely in

1451

cases of ALL. Here, a similar portion of the E2A proteins is fused to a portion of the transcription factor hepatic leukemia factor (HLF), including the DNA binding domain. These observations indicate the promiscuity of the E2A locus with respect to translocation partners and suggest the involvement of common oncogenic mechanisms in cases of ALL associated with t(1;19) or t(17;19).

E References Abramovich C, Humphries RK (2005) Hox regulation of normal and leukemic hematopoietic stem cells. Curr Opin Hematol 12:210–216 LeBrun DP (2003) E2A basic helix-loop-helix transcription factors in human leukemia. Front Biosci 8:206–222 Murre C (2005) Helix-loop-helix proteins and lymphocyte development. Nat Immunol 6(11):1079–1086 Pui CH, Relling MV, Downing JR (2004) Acute lymphoblastic leukemia. N Engl J Med 350:1535–1548

E3 Ubiquitin Protein Ligase ▶ CHFR

Eag ▶ Ether à-go-go Potassium Channels

Eag1 ▶ Ether à-go-go Potassium Channels

EAP1 ▶ Securin

1452

Early B-Cell Factors Daiqing Liao Department of Anatomy and Cell Biology, UF Health Cancer Center, University of Florida College of Medicine, Gainesville, FL, USA

Synonyms COE; Collier-Olf-EBF; Early B-cell factors; EBF; O/E; Olf; Olfactory neuronal transcription factor; Olfactory/early B-cell factors

Definition Early B-cell factors are a group of DNA-binding transcription factors containing the helix-loophelix (HLH) domain. Within their highly conserved DNA-binding domain (DBD), a sequence motif consisting of an atypical zinc ﬁnger (H-X3C-X2-C-X5-C) is unique to this family of proteins.

Characteristics Early B-cell factor 1 (EBF1) was initially isolated in 1993 from the nuclear extracts of a murine pre-B-cell line through oligonucleotide afﬁnity chromatography. The DNA sequence of the oligonucleotide was derived from the promoter of the mb-1 gene, encoding an immunoglobulinassociated protein that is only expressed in the early stages of B-lymphocyte differentiation. EBF1 is essential for B-cell development, as mice lacking EBF1 gene do not produce functional B cells and immunoglobulins. The cDNA encoding a related factor called Olf-1 was identiﬁed in the same year by screening rat cDNA clones that could activate reporter gene in yeast under the control of a synthetic promoter containing DNA elements derived from olfactory speciﬁc genes. The EBF family is found only in the animal kingdom from Caenorhabditis elegans to

Early B-Cell Factors

humans. In the mouse and human genomes, there are four paralogous genes of the EBF family (EBF1–4). In humans, they localize at chromosomes 5q34, 8p21.2, 10q26.3, and 20p13 for EBF1–4, respectively. Like a typical DNA-binding transcription factor, EBF proteins contain well-deﬁned modular structural and functional domains (Fig. 1). The DBD near the N-terminus consists of approximately 200 residues whose sequence is exceedingly well conserved throughout evolution with >75% sequence identity between distant species. This family of proteins binds directly to DNA sequences with a consensus of 50 -CCCNNGGG-30 as homo- or heterodimers. The region following the DBD resembles the conserved domains called TIG-IPT (immunoglobulin-like fold, plexins, transcription factors, or transcription factor immunoglobulin). The function of this region in EBF has not been determined, although the TIG-IPT domain may be involved in homo- or heterodimerization in other transcription factors containing this sequence. Located next to TIG-IPT is the HLH domain that includes two helices with similar sequence. This feature distinguishes the EBF family from other HLH-containing transcription factors such as Myc, Max, and MyoD that usually contain two dissimilar amphipathic helices. The EBF HLH domain is probably involved in dimerization, as proteins with the deletion of this domain could not form stable dimer in solution. The C-terminal domain is highly rich in serine, threonine, and proline residues. It is less conserved but is important for transcriptional activation; nonetheless, mutant without this domain could still activate transcription. Widespread expression of EBF is detected in diverse cell types such as adipocytes and neuronal cells and during different developmental stages such as limb buds and the developmental forebrain. The EBF orthologs in both Drosophila melanogaster and Caenorhabditis elegans are implicated in neurogenesis. During mouse embryogenesis, EBF members are expressed in early postmitotic neurons from midbrain to spinal cord and at speciﬁc sites in the embryonic

Early B-Cell Factors Early B-Cell Factors, Fig. 1 The structural features of EBF family of transcription factors. Shown are the four paralogs (EBF1–4) and their corresponding accession numbers (GenBank or SwissProt database). Speciﬁc domains are shown in different color. Numbers refer to the position of amino acid residue in each protein. The signature zinc ﬁnger motif of the EBF family of proteins in complex with a zinc ion is also depicted. The different domains are DBD DNA-binding domain, TAD transactivation domain, COE the signature sequence of EBF family, ZBM zincbinding motif, IPT/TIG the immunoglobulin-like fold domain, and HLH helixloop-helix motif

1453 EBF1 (NP_076870) DBD 50

376 412

157 170

251

IPT/TIG

TAD

591 COE/ZBM

HLH 369 428

252 EBF2 (NP_034225) DBD 49

367 403

156 169

251

IPT/TIG

TAD

575 COE/ZBM

HLH 360 419

251 EBF3 (Q9H4W6) DBD 50

377 413

157 170

251

IPT/TIG

TAD

596 HLH 370 429

COE/ZBM R

S 161 C M I E

Zn2+

164 CC D

H 157

252

K K S C 170

EBF4 (Q9BQW3-2) DBD 51

158 171

370 406 252

IPT/TIG

TAD

544 COE/ZBM 253

forebrain, indicating that EBF proteins may be involved in regulating neuronal maturation in the central nervous system (CNS). Interestingly, EBF1 is abundantly expressed in the striatonigral medium spiny neurons (MSNs). EBF1 deﬁciency in mice results in markedly reduced number of striatonigral MSNs at postnatal day 14 (P14), although such neurons are properly speciﬁed in EBF1 / mice by P0. Thus, EBF1 appears to be a lineage-speciﬁc transcription factor essential to the differentiation of striatonigral MSNs. Targeted deletion of mouse EBF2 results in defects in peripheral nerve morphogenesis, migration of hormone-producing neurons, projection of olfactory neurons, and cerebellar development. In addition to B-cell development and neuronal differentiation, EBF transcription factors are

HLH 364 422

involved in other developmental processes. For example, EBF2 is a regulator of osteoblastdependent differentiation of osteoclasts, and EBF2-deﬁcient mice have reduced bone mass. Furthermore, EBF1 induces adipogenesis in NIH-3 T3 ﬁbroblasts. The expression of EBF proteins in multiple tissues and their involvement in diverse developmental pathways suggest that they have fundamental cellular functions and their roles in lineage determination may be achieved through cooperation with other tissue-restricted factors. The four paralogs of EBF members are highly similar to each other at the amino acid sequence level. It is therefore surprising that they have quite distinct functions, as EBF3-deﬁcient mice exhibit neonatal lethality before postnatal day 2, and

1454

EBF2-null mice are much smaller than their wildtype littermates, with body weight of the former being less than one half of that of the latter 30 days after birth. The N-terminal 50 amino acid residues and the entire C-terminal transactivation domain are the most divergent regions among these paralogs, although these regions are essentially identical among corresponding orthologs in different mammalian species. Therefore, functional speciﬁcity of individual EBF member may be determined by these divergent domains or through speciﬁc regulation of their expression. Indeed, the promoter sequences of the four paralogs appear quite distinct. Potential Roles in Cancer The EBF3 locus at chromosome 10q26.3 is biallelically altered by genomic deletion and/or promoter hypermethylation in most cases of high-grade brain tumors. In a small number of examined clinical samples, the EBF3 locus is inactivated in 50% of grade II, 83% of grade III, and 90% of grade IV brain tumors. EBF3 is expressed in normal brain cells but is silenced in brain tumor cells. Thus, it is likely that EBF3 may be a tumor suppressor in the brain. EBF3 might also restrict abnormal proliferation in cancer of other tissue origins, as epigenetic silencing of EBF3 occurs in cancer cell lines derived from the breast, colon, bone, and liver. Consistent with epigenetic silencing of the EBF3 locus, EBF3 expression can be reactivated by treating cells with 5-aza-20 -deoxycytidine, a demethylating agent, and trichostatin A, an inhibitor of histone deacetylases. Ectopic expression of EBF3 causes cell-cycle arrest and apoptosis in cancer cells through regulating the expression of genes involved in cell-cycle control. Speciﬁcally, EBF3 directly activates genes encoding the Cip/ Kip family of inhibitors of cyclin-dependent kinases such as p21cip1/kip1 and p27kip1. Conversely, EBF3 can repress the expression of genes responsible for cell proliferation and survival such as ▶ cyclins A and B, CDK2, Daxx, and Mcl-1. Therefore, EBF3 may act as a tumor suppressor by regulating expression of speciﬁc set of genes.

Early Detection

In mouse B-cell lymphomas, retroviral insertions occur frequently in two genetic loci encoding Evi3 (ecotropic viral integration site 3) and EBFAZ (EBF-associated zinc ﬁnger protein, also known as OAZ). Such viral integration results in heightened expression of EBFAZ or Evi3. These two proteins are highly similar to each other, with each containing 30 Krüppeltype zinc ﬁngers. Via several zinc ﬁngers near its C-terminus, EBFAZ or Evi3 binds to EBF, and it is suggested that overexpression of EBFAZ or Evi3 in B-cell leukemias and lymphomas causes aberrant expression of EBF1 target genes that might contribute to tumorigenesis in B cells. In support of this notion, Evi3 is signiﬁcantly expressed in most human acute myelogenous leukemias. EVi3 expression is abundant in human hematopoietic progenitors and declines rapidly during cytokine-driven differentiation. Interestingly, Evi3 or EBFAZ could either repress or activate EBF-mediated transcription, depending on cell type and gene promoter. Therefore, the precise implications of the EBFAZ–EBF pathway in cancer development remain to be determined. Finally, focal deletions of theEBF1 locus have been detected in signiﬁcant cases of B-progenitor ALL (acute lymphoblastic leukemia).

References Liberg D, Sigvardsson M, Akerblad P (2002) The EBF/Olf/Collier Family of transcription factors: regulators of differentiation in cells originating from all three embryonal germ layers. Mol Cell Biol 22:8389–8397 Zardo G, Tiirikainen MI, Hong C et al (2002) Integrated genomic and epigenomic analyses pinpoint biallelic gene inactivation in tumors. Nat Genet 32:453–458 Zhao LY, Niu Y, Santiago A et al (2006) An EBF3mediated transcriptional program that induces cell cycle arrest and apoptosis. Cancer Res 66:9445–9452

Early Detection ▶ Prostate Cancer Diagnosis

Early Genes of Human Papillomaviruses

Early Genes of Human Papillomaviruses Massimo Tommasino1, Rosita Accardi1 and Uzma Hasan2 1 Infections and Cancer Biology Group, International Agency for Research on Cancer, Lyon, France 2 CIRI, Oncoviruses and Innate Immunity, INSERM U1111, Ecole Normale Supérieure, Université de Lyon, CNRS-UMR5308, Hospices Civils de Lyon, Lyon, France

Definition Human Papillomavirus (HPV) 16 has six early genes that are transcribed from the same DNA strand. As in other viruses, the function of the early proteins is to alter several cellular events to guarantee the completion of the virus life cycle. In addition, three early proteins, E5, E6, and E7, are also involved in the induction of malignant transformation of the infected cells. The cutaneous HPV types have a similar organization of the early region, with exception that the majority of these types lack the E5.

Characteristics HPV constitutes of a heterogeneous group of viruses from the Papillomaviridae family. The HPV phylogenetic tree has been designed based on the homologous nucleotide sequence of the L1 capsid protein. So far, 92 HPV types have been fully sequenced, of which 60 belong to the alpha genera and the other 32 belong to the beta and gamma genera. Based on their tissue tropism, HPVs can be divided in cutaneous and mucosal HPV types. The mucosal HPV types are included in the genus alpha together with certain benign cutaneous HPV types, whereas the beta and the gamma genera exclusively consist of cutaneous HPV types. Biological and epidemiological studies have clearly demonstrated that certain mucosal

1455

HPV types are associated with ▶ Cervical cancer. On the contrary, the role of cutaneous types in carcinogenesis is still under debate, although several lines of evidence support their association with nonmelanoma skin cancer (NMSC). The mucosal HPV type 16 (HPV16) is the most frequently found HPV genotype in cervical cancers worldwide, and thus its early gene products are the best studied and characterized (Fig. 1). HPV16 E6 and E7 are the major transforming proteins. Three different lines of evidence have demonstrated the involvement of E6 and E7 in cervical carcinogenesis: 1. The ﬁrst indication came from the analysis of HPV-infected cells, which showed that viral DNA is randomly integrated in the genome of the majority of cervical carcinomas. Integration leads to the disruption of several viral genes with preservation of only the E6 and E7, which are actively transcribed. 2. The discovery that E6 and E7 proteins are able to induce cellular transformation in vitro conﬁrmed their oncogenic role. Immortalized rodent ﬁbroblasts can be fully transformed by expression of HPV16 E6 or E7 protein. These rodent cells acquire the ability to grow in an anchorage-independent manner and to be tumorigenic when injected into nude mice. In addition, HPV16 E6 and E7 together are able to immortalize primary human keratinocytes, the natural cellular host of the virus. In agreement with the in vitro assays, transgenic mice coexpressing both viral genes exhibit epidermal hyperplasia and various tumors. Similar to the mucosal HPV types, E6 and E7 from certain cutaneous HPV types of the genus beta (e.g., HPV8 and 38) display transforming properties in in vitro and in vivo models. Independent studies have demonstrated that HPV16 E6 and E7 proteins do not induce cellular transformation via a “hit and run” mechanism, but continuous expression of both proteins is required for the maintenance of the malignant phenotype. 3. Finally, biochemical studies have clariﬁed the mechanism of action of E6 and E7. The viral

1456

Early Genes of Human Papillomaviruses

Early Genes of Human Papillomaviruses, Fig. 1 Organization of the early region of HPV16. The six early genes are represented by using different colors. The nucleotide positions of each early gene and the predicted size in amino acids are also shown. Several polycistronic transcripts have been identiﬁed, which comprise of 2–3 early genes in different combinations and are

most likely transcribed from the P97 promoter. The function of each transcript is not known, but some of them are only transcribed at different stages of differentiation. Several lines of evidence indicate that the alternative splicing in E6 and E7 transcripts play an important role in the translational regulation of the viral proteins

oncoproteins are able to form stable complexes with cellular proteins and alter, or completely neutralize, their normal functions. These events lead to the loss of control of cell cycle checkpoints, apoptosis, and cellular differentiation.

protein is a transcription factor that can trigger cell cycle arrest or apoptosis in response to stress or DNA damage. E6 binds to a 100 kDa cellular protein, E6AP (E6-protein), which functions as an ubiquitin protein ligase (E3). The E6/E6AP complex then binds p53, which becomes very rapidly ubiquitinated and as a consequence is targeted to proteasomes for degradation. Since the major role of p53 is to safeguard the integrity of the genome by inducing cell cycle arrest or apoptosis, cells expressing HPV16 E6 show chromosomal instability, which greatly increases the probability that HPV-infected cells will evolve toward malignancy. Additional ﬁndings have demonstrated that HPV16 E6 also associates with the transcriptional regulators, CBP and

E6 protein. HPV16 E6 is a small basic protein of 151 amino acids. The major structural characteristic of E6 is the presence of two atypical zinc ﬁngers. At the base of these zinc ﬁngers are two motifs containing two cysteines (Cys-X-X-Cys), which are conserved in all E6 HPV types. The best characterized HPV16 E6 activity is its ability to induce degradation of the tumor suppressor protein p53 via the ubiquitin pathway. This cellular

Early Genes of Human Papillomaviruses

1457

Neutralisation of p300 function

E6AP

p53 degradation

DNA damage

p300 p53

Transcriptional activation

p21

G1 arrest

p53 stabilization

Bax Apoptosis

Early Genes of Human Papillomaviruses, Fig. 2 p53 pathways targeted by HPV16 E6. When cells are exposed to DNA-damaging agents, e.g., X-rays, the half-life of p53 can be greatly increased by posttranslational modiﬁcations (phosphorylation). In turn, p53 can either activate the transcription of the cyclin-dependent kinase (CDK)

inhibitor, p21WAF1/CIP1, leading to a G1 arrest and DNA repair before replication, or activate the transcription of the proapoptotic gene Bax with consequent induction of apoptosis. Cells expressing HPV16 E6 protein are resistant to the cell cycle arrest or apoptosis induced by DNA-damaging agents

p300, with resulting inhibition of p53-driven transcription. Thus, HPV16 E6 neutralizes p53 by two distinct mechanisms; the ﬁrst is mediated by the p300/CBP association, while the second occurs via binding to E6AP to promote p53 degradation (Fig. 2). Interestingly, the E6 protein from cutaneous HPV types is not able to induce degradation of p53. In fact, it has been shown that the beta HPV38 can inactivate p53 function by inducing accumulation of its antagonist, DNp73. HPV16 E6, as well as E6 from certain cutaneous HPV types, can also interfere with the apoptotic pathways via its association with Bak, a member of the Bcl-2 family. Analogously to its effect on p53, E6 induces Bak degradation via the ubiquitinmediated pathway. Several p53-independent cellular pathways, which are altered by the E6 molecule, have been identiﬁed. HPV16 E6 is able, through its association with E6AP, to promote the degradation of the transcriptional repressor NFX1-91 and consequently to activate the transcription of the hTERT (human telomerase reverse transcriptase) gene encoding the catalytic subunit of the telomerase complex. This effect directly results in

telomerase activity upregulation, a key event in the immortalization of primary keratinocytes. In addition, HPV16 E6 is able to interfere with cell mobility, through interaction with the human homologue of the Drosophila discs large protein (DLG). Also in this case, E6 binding leads to degradation of the cellular protein. However, different E6 domains are required to induce degradation of p53 and DLG. Deletion of the carboxy terminus of HPV16 E6 abolishes its binding to DLG without inﬂuencing its ability to promote p53 destabilization. Another cellular target of HPV16 E6 is paxillin, a protein involved in transducing signals from the plasma membrane to focal adhesions and the actin cytoskeleton. The fact that E6 from the oncogenic HPV16, but not E6 from the low-risk HPV types 6 and 11, is able to bind paxillin suggests that this interaction has a role in the carcinogenesis of HPV infection. Several factors have been described to minimize or prevent exposure of HPV to the immune system. It has also been shown that HPV16 E6 interacts with interferon regulatory factor-3 (IRF-3), a positive transcriptional regulator of the INF-b promoter, which is activated in response to virus infection.

1458

E6 binding inhibits IRF-3 transactivation function. Thus, this E6-induced event may enable the virus to circumvent the antiviral response of the infected cell. The adhesion between keratinocytes and antigen presenting cells, i.e., Langerhans cells, in the epidermis is mediated by E-cadherin. It has been shown that E6 can reduce the levels of cell surface E-cadherin on keratinocytes, thereby limiting the presentation of viral antigens to the Langerhans cells and promoting viral survival. Furthermore, HPV16 E6 has been shown to suppress innate immune responses mediated by a family of toll-like receptors (TLRs) that are key sensors of evading pathogens. E6 is able to block the promoter of TLR9 which has been identiﬁed to recognize dsDNA sequences. Finally, E6 associates with ERC 55, a putative calcium-binding protein located in the endoplasmic reticulum. However, the biological signiﬁcance of this interaction is still unclear. E7 protein. HPV16 E7 is an acidic phosphoprotein of 98 amino acids, which is structurally and functionally related to a gene product of another DNA tumor virus, the adenovirus E1A protein. On the basis of the similarity in primary structure between the two viral proteins, they can be divided into three domains: conserved region 1–3 (CR1–3). Mutational analysis of HPV16 E7 has demonstrated that all three regions are important for the in vitro transforming activity of the molecule. CR3 contains two CXXC motifs involved in zinc binding and essential for the stability of the protein. Independent studies have demonstrated that this viral protein is located in the nucleolus, nucleus, and cytoplasm. Indeed, it has been shown that the E7 molecule associates with cytoplasmic and nuclear proteins. The best understood interaction of E7 with a cellular protein is that involving the “pocket” proteins, pRb, p107, and p130. The pocket proteins are central regulators of cell cycle division. They negatively regulate, via direct association, the activity of several transcription factors, including members of the E2F family (E2F1–5), which are associated with their partners, DPs. Under normal cell cycle regulation, phosphorylation of pRb, which is mediated by cyclin-dependent kinase (CDK) activity, leads to the disruption of pRb/E2F

Early Genes of Human Papillomaviruses

complexes, with consequent activation of E2Fs. HPV16 E7 binds the pocket proteins and, analogously to the phosphorylation, results in the release of active E2Fs which in turn activate the transcription of a group of genes encoding proteins essential for cell cycle progression, such as cyclin E and cyclin A. As described for the interaction between E6 and p53, HPV16 E7 protein is able to promote the destabilization of pRb through the ubiquitin–proteasome pathway (Fig. 3). Similarly, the beta HPV38 E7 binds to pRB and promotes its degradation. This property is not shared by all E7s from the different HPV genotypes. Indeed, E7 from the benign HPV1 can efﬁciently associate with pRb without inducing its degradation. It is likely that the E7-induced pRb degradation represents a more effective way to neutralize the function of the cellular protein. The other two members of the pocket protein family, p107 and p130, are involved in controlling additional cell cycle checkpoints; p130 exerts its transcriptional regulatory function during the G0/G1 transition, while p107 is active in the G1/S transition and in the G2 phase. Analogously to pRb, HPV16 E7 protein associates with p107 and p130, inactivating key cell cycle checkpoints. Besides targeting the pocket proteins, E7 can alter cell cycle control by additional mechanisms. The HPV16 E7 protein is able to associate with the CDK inhibitors p21WAF1/CIP1 and p27KIP1 causing neutralization of their inhibitory effects on the cell cycle. Cells coexpressing HPV16 E7 and p21WAF1/CIP1 or p27KIP1 are still able to enter S phase, while in the absence of E7 cells are arrested in G1 phase. HPV16 E7 can also directly and/or indirectly interact with cyclin A/CDK2 complex. The biological function of this interaction remains to be elucidated, but it is possible that E7 may act by redirecting the kinase complexes to a different set of substrates. Other cellular proteins involved in transcriptional regulation have been identiﬁed as HPV16 E7 targets. HPV16 E7 binds the TATA boxbinding protein (TBP) and the TBP-associated factor TAF110, indicating that the viral protein is able to interfere with the basic transcriptional machinery of the host cell. Furthermore, E7

Early Genes of Human Papillomaviruses

1459

Early Genes of Human Papillomaviruses, Fig. 3 Deregulation of the restriction point (R) by HPV16 E7. E2F transcription factors form heterodimer complexes with members of the DP family and regulate the transcription of several genes during the cell cycle. In quiescent cells, pRb is present in a hypophosphorylated form and associates with E2F molecules, thereby inhibiting their transcriptional activity. When quiescent cells are exposed to mitogenic signals, genes encoding

the G1-speciﬁc D-type cyclins (D1, D2, and D3) are activated. Subsequently, cyclins associate with a catalytic subunit, CDK4 or 6, and after transport into the nucleus, the kinase complexes phosphorylate pRb in mid-G1 phase causing release of active E2F/DP1 heterodimer complexes and progression through the restriction point (R). E7 binding to pRb mimics its phosphorylation. Thus, E7-expressing cells can enter S phase in the absence of a mitogenic signal

associates with AP1 complex, activating its transcriptional activity. Similar to other stimulators of proliferation (e.g., c-myc or adenovirus E1A), the HPV16 E7 protein, besides the ability to deregulate the cell cycle, also promotes apoptosis. Expression of HPV16 E7 in normal human ﬁbroblasts (NHF) or in human keratinocytes results in a cytocidal response, which displays the typical features of apoptosis and is much more evident in the absence of mitogenic signals. This E7-induced apoptosis requires pRb inactivation and is mediated by p53-dependent and -independent pathways. It is likely that the E7-induced apoptosis represents a cellular response elicited by the loss of cell cycle control. Interestingly, E6 protein is able to completely abrogate the E7 activity in promoting apoptosis. Thus, both viral proteins are required to

induce transformation of the host cells. Although E7 is constitutively expressed in HPV16associated lesions and therefore appears as candidate antigen for a speciﬁc immune response, the immune system fails to produce an efﬁcient defense against tumor outgrowth in affected patients. As in the case of E6, E7 has also evolved to escape immune surveillance. E7 has the ability to bind and prevent the transcription factor IRF-1 from activating the INF-a and INF-b promoters. E7 is also capable of downregulating innate responses via suppressing TLR9 expression. Furthermore, in HPV16 E6 and E7 transgenic mice, the transgene product E7 does not induce an immune response. However, upon vaccination with E7, anti-E7 antibodies were produced without causing signs of autoimmune disease. In contrast, E7-speciﬁc cytotoxic T lymphocytes (CTL) were

1460

not detected after immunization. Therefore, the E7 transgene expression induces speciﬁc immunological tolerance at the CTL level. E5 protein plays an early role in the HPV-induced transformation. Studies on ▶ bovine papillomavirus (BPV) have provided evidence that E5 is a potent oncoprotein. It has also been shown that HPV16 E5 is able to induce cellular transformation, although with less efﬁciency than BPV E5. HPV16 E5 is a small hydrophobic protein, which is located in the endoplasmic reticulum (ER), nuclear membrane, and cytoplasmic vesicles. E5 is able to enhance growth factor-mediated signal transduction to the nucleus, resulting in stimulation of cellular proliferation. BVP1 and HPV16 E5 associate with the 16 kDa subunit c of the vacuolar H + -ATPase, which is responsible for acidiﬁcation of membrane-bound organelles, such as Golgi, endosomes, and lysosomes. It has been shown that BPV1 E5 induces alkalinization of Golgi, and that this activity is linked to its in vitro transforming activity. Mutations in E5 that abolish the interaction with 16 kDa subunit c abrogate Golgi alkalinization and cellular transformation. HPV16 E5-mediated immune evasion also involves suppressing the expression of the major histocompatibility complex class I (MHC I) and antigen processing via the TAP pathway, reﬂecting the lack of antigen presentation to CTL. Since the integration of viral DNA, which occurs in tumor cells, results in a loss of E5 gene expression, it is clear that E5 is involved in early events during the multistep process of cervical carcinogenesis and that its function is no longer required after the establishment of the transformed phenotype. E1 and E2 are involved in the regulation of viral DNA replication. The function of E1 is to control the replication of viral DNA. E1 contains the cyclin-binding RXL motif and is able to associate with cyclin E and A. Consistent with these ﬁndings, E1 is phosphorylated by cyclin E or A-associated kinase. Mutation of E1 phosphorylation sites results in a reduction of HPV DNA replication, supporting the idea that the E1/cyclin association plays an important role in viral DNA replication. Moreover, it has been shown that E1 has an ATPase and helicase activity. E1 forms a

Early Genes of Human Papillomaviruses

stable complex with E2 and binds to the replication origin of HPV in order to recruit cellular factors essential for DNA replication. Furthermore, E1 has been found associated with components of the cellular DNA replication machinery, e.g., DNA polymerase a. E2 regulates the transcription of the early genes. In addition to controlling viral DNA replication together with E1, E2 is able to negatively or positively regulate the transcription of the early genes. Like all transcription regulatory factors, E2 has an amino-terminal transacting domain and a carboxy-terminal DNA-binding domain, which recognizes four cis elements (ACCN6GGT) in the long control region (LCR) of the HPV genome. These two domains are separated by a central region (hinge), which is important, together with the amino-terminal domain, for the nuclear localization of the molecule. Whether E2 binding results in repression or activation of the promoter of the early genes is dependent on the position of the E2-binding site in the LCR. E2 binding to the promoter-distal or -proximal elements leads to a positive or negative regulation of the promoter, respectively. E4 is probably involved in virus maturation and/or replication. E4 is a late protein expressed from the early region of the genome. Most of the studies have been performed on E4 from the cutaneous HPV type 1. In these lesions E4 is present at very high levels and several E4-derived proteins have been detected. The primary product is a 17 kDa protein, which is expressed from E1^E4 transcript. E4 associates with and disrupts the cytoplasmic keratin network. The biological signiﬁcance of this E4-induced event is not fully understood. It has been proposed that E4 plays a role in the productive phase of the infection establishing a favorable condition for viral maturation. Clinical Relevance HPV16 infection results in the induction of a benign proliferation, which, after a long latent period, can progress to invasive cancer. Persistent HPV infection is necessary for the development of the malignant lesion. This requirement is explained by the fact that viral proteins, in order to induce full malignant transformation of the host

Early-Stage Ovarian Cancer

cells, have to cooperate with an activated cellular oncogene. Accumulation of mutations in cellular genes, which possibly lead to activation of an oncogene, requires continuous proliferation. This is achieved by the abilities of E6 and E7 to respectively neutralize apoptotic pathways and to induce unscheduled proliferation. Therefore, a possible approach to induce regression of an HPV-positive lesion is to target the biological functions of E6 and E7. This possibility is supported by ﬁndings, which clearly demonstrate that continuous expression of the two viral genes is necessary for the maintenance of the host cell transformed phenotype. Thus, we can predict that a blocking of the activity of E7 can lead to a rapid exit from the cell cycle. Neutralization of E6 function should result in an even more efﬁcient way to induce regression of the HPV lesion. As described above, E7 has a dual activity, being able to induce proliferation and apoptosis. E6, acting upon p53-dependent and -independent pathways, completely abolishes the E7-induced apoptosis. Thus, we could imagine that in cells expressing E6 and E7 genes, the block of only E6 functions may push the balance between proliferation and apoptosis in favor of the latter causing regression of the HPV lesion. Alternative targets are E1 and E2, which are involved in viral transcription and replication. Several approaches to neutralize the early viral proteins are under investigation. These include strategies to block the transcription or translation of viral genes or to identify small molecules able to speciﬁcally associate with and inactivate the viral proteins. In addition, HPV E6 and E7 are able to downregulate the innate and adaptive immunity. Identiﬁcation of strategies to reactivate the immune response in HPV-infected cells may favor the clearance of the infection preventing the development of cervical diseases.

References Campo MS (2006) Papillomavirus research: from natural history to vaccine and beyond. Caister Academic Press, Norfolk

1461 Munger K, Baldwin A, Edwards KM et al (2004) Mechanisms of human papillomavirus-induced oncogenesis. J Virol 78:11451–11460 O’Brien PM, Campo MS (2003) Papillomaviruses: a correlation between immune evasion and oncogenicity? Trends Microbiol 11:300–305 Tommasino M (ed) (1997) Human papillomaviruses in human cancer: the role of E6 and E7 oncoproteins. Molecular Biology Intelligence Unit, Landes Company, Austin zur Hausen H (2002) Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer 2:342

Early-Stage Ovarian Cancer Behrouz Zand and Ralph S. Freedman UT MD Anderson Cancer Center, Houston, TX, USA

Synonyms Stage IA–C ovarian cancer

Definition Early-stage ▶ ovarian cancer generally refers to malignancy of the ovary that is conﬁned to one or both ovaries (International Federation of Gynecology and Obstetrics [FIGO] stage I A and B). • Stage IA is growth limited to one ovary. • Stage IB is growth limited to both ovaries. Once the tumor involves the surface of an ovary, or there is capsule rupture, or presence of ascites or pelvic washings containing malignant cells (IC), the cancer is more advanced and prognosis for the patient is less favorable.

Characteristics ▶ Ovarian cancer is a relatively common gynecologic malignancy. Approximately 90% of ovarian cancers are of the epithelial cell type developing

1462

from the ovarian surface epithelium. Epithelial ovarian cancer peaks in incidence between 55 and 65 years of age and over 60% has serous histology. Less common epithelial cell cancers include primary mucinous, endometrioid, clear cell, and undifferentiated carcinomas, which comprise about 20–30% of the epithelial types. Non-epithelial types of ovarian cancer including germ cell tumors and sex cord stromal tumors comprise about 10–20% of all ovarian cancer cases. Non-epithelial ovarian cancers have overall a better prognosis than epithelial types and also occur at higher proportions in younger patients. Non-epithelial ovarian cancers mainly include germ cell types (dysgerminoma, immature teratoma, endodermal sinus tumor) and sex cord stromal cell types (▶ granulosa cell tumors, SertoliLeydig cell tumor). Overall they have a better prognosis than the epithelial types in part because they tend to be diagnosed at an early stage. They are also more commonly seen in women in their reproductive years. The non-epithelial malignancies particularly the germ cell tumors are more chemosensitive. A variety of other rare types of non-epithelial ovarian cancers including sarcomas, lipoid cell, and small cell or ▶ neuroendocrine tumors have poor prognosis but only contribute to 0.1% of all ovarian cancers. Unfortunately, most epithelial ovarian cancers are diagnosed at a more advanced stage (stage II to IV). This makes ovarian cancer the most lethal gynecological cancer, except in developing countries where ▶ cervical cancer remains a major contributor to cancer deaths among women. Approximately 20% of patients with epithelial ovarian cancer can be diagnosed at an earlier stage of disease. Surgery is usually the ﬁrst component in the treatment of most ovarian cancers of any histological type. Patients who are suspected of having an ovarian malignancy should undergo surgical exploration by a gynecological oncologist to conﬁrm the diagnosis and to determine the stage and histological grade, followed by a rigorous effort to remove as much of the malignancy as possible. The type and duration of any required subsequent chemotherapy are determined in most cases by the accuracy and completeness of the surgical procedure and the histological type and

Early-Stage Ovarian Cancer

grade. In both early-stage epithelial and non-epithelial cancers, there are certain situations where postoperative chemotherapy may not be indicated, hence the importance of obtaining adequate staging and pathological assessment in early-stage disease. Preoperative Workup of an Adnexal Mass The diagnosis of ovarian cancer remains a clinical and technological challenge despite the development of some new approved tests. Patients found to have a pelvic mass on physical examination should be evaluated with transvaginal pelvic sonography and a measurement performed of their ▶ serum biomarkers cancer antigen 125 (CA125) levels. CA125 is a mucin-like protein of high molecular mass estimated at 200–20,000 kDA. CA125 cell surface expression is upregulated when cells undergo metaplastic differentiation into a Müllerian epithelium. CA125 is the most extensively studied biomarker for possible use in ▶ ovarian cancer ▶ early detection. Its expression is elevated in some cases of endometriosis. Characteristics of masses that are more suggestive of malignancy include: size greater than 8 cm, a solid as opposed to a uniform cystic, consistency, immobility with an irregular shape, bilaterality, and the presence of ▶ ascites. A malignancy may be found with any one of these ﬁndings. A serum CA125 level of >35 U/mL combined with these features would suggest malignancy, although an elevated CA125 on its own is not deﬁnitive, as an elevated CA125 value can also occur commonly in patients with a variety of benign conditions such as ovarian hemorrhagic cysts, endometriosis, uterine ﬁbroids, adenomyosis, ectopic pregnancy, pelvic inﬂammatory disease, appendicitis, and colitis. Speciﬁcity and positive predictive values for CA125 level measurements are consistently higher in postmenopausal women compared with premenopausal women in whom there is a higher frequency of functional and nonfunctional ovarian cystic or tubal inﬂammatory conditions that could confound the diagnosis. In a postmenopausal woman, any CA125 elevation accompanied by a palpable or transvaginal ultrasound documented pelvic mass should increase the suspicion for

Early-Stage Ovarian Cancer

pelvic malignancy although there are medical conditions such as hypothyroidism, congestive heart failure, and cirrhosis that can sometimes contribute to false-positive results. In some instances it may be necessary to repeat the CA125 test for conﬁrmation. The same assay and laboratory should be used for all repeat values. Another caveat related to CA125 is the fact that only 50% of women with stage I disease have an elevated CA125; therefore, a patient with a normal CA125 and adnexal mass does not exclude an early ovarian malignancy. This serves to tell us that there is no substitute for a good history, physical examination, and clinical judgment. Most early ovarian cancers are asymptomatic. Some non-epithelial tumors can present with pain due to tumor complications such as hemorrhagic necrosis or torsion. Additionally, b-hCG, L-lactate dehydrogenase (LDH), and ▶ alpha-fetoprotein (AFP) levels may be elevated in the presence of certain malignant germ cell tumors, while inhibin A and B levels can be elevated in some granulosa cell tumors of the ovary. These latter markers should also be measured when surgical removal of an ovarian mass is being considered in any women during the reproductive period. Surgical Staging If the suspicion of malignancy is high, a computed tomography (CT) scan of the abdomen and pelvis with intravenous and rectal contrast can be helpful preoperatively for identifying the extent of disease in the abdomen and pelvis. It can also provide useful information for discussion with the patient regarding the nature and extent of surgical procedures that may be needed as well as any potential fertility-sparing procedures if appropriate. It is recommended that a gynecologic oncologist be involved in these discussions and performs the surgical procedures whenever possible. Some oncologists ﬁnd it helpful to place ureteric catheters on the day of surgery to aid in the location of the ureters, especially where the mass is very large and appears to be immobile or there has been a prior history of endometriosis or pelvic inﬂammatory disease. Surgical pathological staging should be undertaken regardless of the patient’s age or

1463

desire for fertility. The diagnosis can usually be conﬁrmed at frozen section on the affected ovary removed. It may happen that a younger patient desirous of preserving her fertility status has to undergo emergent surgery in a situation where either surgical expertise or frozen section and gynecologic pathology expertise are unavailable. In such a situation where there is also a suspicion of malignancy, the affected ovary should be removed completely for diagnosis and the patient referred to a center of expertise. There is not universal agreement as to the extent of the surgical staging that is required in early ovarian cancer. At a minimum the procedure should include peritoneal washings; careful inspection of the abdominal cavity to search for indications of spread outside of the ovaries, metastatic deposits which should be biopsied; and random peritoneal biopsies from the pelvis, paracolic gutters, the diaphragm, and the omentum. Either pelvic and para-aortic lymph node sampling or full ▶ lymphadenectomy can provide additional information that can result in upstaging if lymph node metastases are found. A woman who has completed her family should undergo a total hysterectomy and bilateral salpingo-oophorectomy in addition to the surgical staging. Premenopausal patients who are interested in fertility conservation should be offered the option of unilateral salpingo-oophorectomy along with routine staging, with preservation of the contralateral ovary and the uterus – provided that these appear normal at the macroscopic level. In early-stage disease, it is very important to establish that the disease has not spread beyond the ovaries or the pelvis since accurate staging will help determine the prognosis for the individual patient, the type and intensity and duration of postoperative chemotherapy, and in certain cases whether adjuvant chemotherapy is even indicated. Today with appropriate staging and appropriate adjuvant chemotherapy, 90% of patients with stage I ovarian cancer generally survive 5 years and more. ▶ Meta-analysis of the largest and best conducted randomized trials has conﬁrmed the beneﬁcial effects of adjuvant chemotherapy in “early-stage” ovarian cancer, but any beneﬁt of chemotherapy after “optimal” surgical staging on

1464

progression-free or overall survival could not be shown. Optimal staging, even when deﬁned, was performed in only a minority of these patients; thus, these studies were not designed or powered to address the question. A reasonable default position would be to offer adjuvant chemotherapy treatment to patients who were adequately staged and who have high risk factors (stage IA or B, with grade 3 or clear cell cancers, or stage IC) and to offer restaging to those patients whose staging were suboptimal or accept a higher bar for not using chemotherapy in such patients or to initiate treatment at ﬁrst recurrence. Although staging (▶ Staging of Tumors) for early-stage epithelial ovarian cancer has traditionally been performed by open laparotomy, there is an increasing body of evidence supporting the alternative use of a laparoscopic procedure. These studies have shown that laparoscopic staging and laparotomy staging result in similar surgical outcomes and staging adequacy in terms of nodal yield, omental size, and accuracy of identifying metastatic disease and safety. At present there are no prospective, collaborative trials comparing the use of laparoscopy staging with laparotomy staging for presumed early-stage epithelial ovarian cancer; therefore, it is difﬁcult to draw any deﬁnite conclusions about laparoscopic staging in early ovarian cancer. In the hands of suitably trained and experienced laparoscopic gynecologic oncologists, there is good reason to believe that adequate and safe staging and surgical treatment by interventional laparoscopy may become possible in the future. First-line adjuvant chemotherapy for epithelial cell ovarian cancer generally includes a combination of carboplatin and a taxane, either ▶ paclitaxel or ▶ docetaxel, usually for six cycles. In the USA both drugs are frequently utilized in earlystage disease. In Europe and the UK, single-agent carboplatin may be preferred because of concerns regarding toxicity particularly chronic toxicity such as peripheral neuropathy. Some have suggested that the number of cycles of chemotherapy should be reduced from six to three; there are no adequately powered studies to determine

Early-Stage Ovarian Cancer

which is better for early-stage disease. Since some patients will fail either due to understaging or to resistance to chemotherapy and ovarian cancer recurrences are rarely curable, a reasonable rationale is provided for treating patients with the standard six cycles that is used for higherstage tumors. Monitoring patients closely for development of toxicities that could become irreversible from the chemotherapy is important. Non-epithelial Ovarian Cancer: Surgical Staging and Treatment Germ Cell Tumors

Dysgerminoma is the most common germ cell malignancy and accounts for 30–40% with 65% in stage I. Dysgerminomas have a higher rate of developing bilaterally (10–15%) than other germ cell tumors; therefore, careful inspection of the contralateral ovary for tumor is mandatory. Careful surgical staging should be done to determine the extent of occult metastatic disease. The minimum treatment is a unilateral salpingo-oophorectomy. If future fertility is desired, then the contralateral ovary, tube, and uterus can be left in situ even when metastasis is present, as these tumors are highly chemosensitive. Five percent of dysgerminomas occur in patients with abnormal XY karyotype (i.e., gonadal dysgenesis). In the latter patients, both gonads should be removed although the uterus may be left in situ for possible embryo transfer. Any suspicious lesion on the contralateral ovary should be biopsied and resected if found to be malignant (preserving some normal ovary) due to higher rate of bilaterality, but routine biopsy or wedge resection of a normal contralateral ovary is not indicated. A patient with stage IA dysgerminoma has a 5-year survival of 95% after a unilateral salpingooophorectomy alone. Immature Teratoma

These also require full staging and can be treated with fertility-sparing surgery or total hysterectomy and bilateral salpingo-oophorectomy. Very

Early-Stage Ovarian Cancer

few are bilateral and therefore contralateral biopsy is not indicated. Typically these tumors occur in combination with other germ cell tumors as mixed germ cell tumors, and less than 1% account for pure immature teratoma in all ovarian cancers. Overall the 5-year survival rate for patients with all stages of pure immature teratomas is 70–80% and 90–95% for patients with surgical stage I lesions. Endodermal Sinus Tumor

EST; Also called “yolk sac” tumors, ESTs have a median age at diagnosis of 18 years. The primary treatment consists of unilateral salpingooophorectomy with frozen sections and removal of any gross disease. As with other germ cell tumors, fertility-sparing surgery is appropriate and all patients will need chemotherapy. The malignant germ cell tumors as a group, dysgerminoma, immature teratoma, and endodermal sinus tumor, are very sensitive to a variety of chemotherapy combinations such as BEP (▶ bleomycin, ▶ etoposide, ▶ cisplatin), VBP (vinblastine, bleomycin, cisplatin), and VAC (vincristine, actinomycin D, ▶ cyclophosphamide). Sex Cord Tumors

Granulosa cell tumors (GCT) are bilateral in only 2% of patients, so a unilateral salpingooophorectomy is indicated for stage IA tumors in children or in women of reproductive age. Surgical staging should be performed, and any suspicious ovarian lesion or enlargement should be biopsied. In postmenopausal women or those who do not desire future fertility, total abdominal hysterectomy-bilateral salpingo-oophorectomy (TAH-BSO) should be performed. If uterus is left in situ, preoperative pathologic assessment of the endometrium should be performed to exclude endometrial carcinoma or atypical hyperplasia as GCT typically secrete unopposed estrogen that can affect the endometrium. The prognosis of GCT is dependent on the stage. Stage I patients may be cured. However, late recurrences are sufﬁciently frequent to warrant close follow-up. Effective adjuvant

1465

chemotherapy for early-stage disease has not been determined. Sertoli-Leydig Cell Tumors

These occur in the 3rd and 4th decades of life and account for 0.2% of all ovarian cancers. They are typically low-grade malignancies, although some may be poorly differentiated. These tumors secrete androgens and can lead to virilization in 70–85% of patients. These tumors have a less than 1% chance of being bilateral, and so the treatment of choice is a unilateral salpingo-oophorectomy and evaluation of contralateral ovary in women in the reproductive period. Postmenopausal women may undergo TAH-BSO. The 5-year survival rate is 70–90% and recurrences are uncommon unless it is not poorly differentiated. Fertility Preservation In young patients with non-epithelial ovarian carcinoma or low malignant potential epithelial ovarian tumors (LMPT) conﬁned to a single ovary, the surgical treatment should in most cases be conservative with preservation of the uterus and contralateral normal-appearing ovary even with invasive implants that are localized to the affected ovary. Routine biopsies of the macroscopically normal-appearing contralateral ovary identify a low rate of occult disease (99% of tumor growth. Soluble forms of both mouse and human endostatins are currently produced in yeast, insect, and mammalian cells.

Structure and Function The precursor for endostatin is collagen type XVIII, a special kind of collagen found at the basement membrane of endothelium and epithelium. Collagen XVIII along with collagen XV

1530

Endostatin Endothelial cells

Pericytes/smooth muscle cells

Blood vessel lumen

Basement membrane Collagens Type I, IV, XVIII etc Laminins Perlecan Nidogen (entactin) Fibulin

NC11

Collagen type XVIII

NC1

NH2

COOH

Non-collagenous domains

Hinge

Trimerization domain

Proteolysis by cathepsin L MMPs

Endostatin

Endostatin, Fig. 1 Generation of endostatin from the vascular basement membrane

belongs to a subfamily called multiplexin. Collagens normally have a contiguous, long stretch of triple-helical region (collagenous domain) which ends at the carboxyterminus with a globular, non-collagenous domain (NC1). Collagenous domains are characterized by repeated sequence of glycine–hydroxylysine–hydroxyproline amino acid residues. The multiplexin subfamily of collagens has multiple interruptions in the collagenous domain. Collagen type XVIII, for example, has 11 non-collagenous (NC1–NC11) domains ﬂanking ten collagenous domains (Fig. 1). Endostatin is generated from the NC1 domain, which contains a trimerization domain at the amino terminus followed by a protease-sensitive hinge region which culminates at the carboxyterminus as endostatin. Cathepsin L and, to some extent, matrix metalloproteinases cleave at the hinge region of NC1 domain to release endostatin.

X-ray crystallographic studies have shown that endostatin is a compact globular molecule. Schematic diagram shows the complex architecture of vascular basement membrane/extracellular matrix. In addition to providing the scaffold for vascular assembly, the matrix contains a number of sequestered growth factors which can be locally released to modulate angiogenic response. The domain organization of collagen type XVIII is shown. Endostatin is generated by the proteolysis of the carboxyterminal NC1 domain. A number of positively charged arginine residues are located at the major alpha-helix aligning toward the surface of endostatin. Substitution of arg residues affects heparin binding but does not alter the biological activity. A single point mutation at proline 125 to alanine has been found to increase the biological activity of human

Endostatin

endostatin. Comparison of amino acid sequences between various organisms reveals substitutions of proline to a different amino acid residue at this site. Endostatin sequence from chicken and Xenopus shows proline to alanine substitution. Japanese puffer ﬁsh and drosophila (fruit ﬂy) have glycine at this position. C. elegans has an aspartic acid at this location. However, no polymorphism has been found in human collagen XVIII/endostatin at this position. Some studies suggest even shorter fragments of endostatin are biologically active and inhibit angiogenesis. Mechanism of Action Endostatin binds to a5b1 integrin and induces clustering within the lipid rafts. In addition, glypican 1, a heparan sulfate glycosaminoglycan, serves as a low-afﬁnity binding site for endostatin. Two phenylalanine residues (F31 and F34) are found to be critical for glypican binding. Mutation of these two residues affects endostatin binding to glypican 1. The integrin-binding site of endostatin has not yet been identiﬁed. Clustering of integrinbound endostatin activates ▶ Src kinase and inhibits RhoA activity leading to the disruption of focal adhesion. Associated changes in the actin stress ﬁbers ultimately affect endothelial cell migration. Endostatin is also known to affect a second signaling pathway, the ▶ Wnt signaling pathway, leading to the proteosomal degradation of b-catenin. Reduced levels of cytoplasmic b-catenin block T-cell factor (TCF)-mediated transcription of ▶ cyclin D1 and c-myc. Repression of cyclin D1, one of the critical mediators of cell cycle progression, arrests endothelial cells at the G1 phase of the cell cycle. Other reports have identiﬁed cell surface-bound tropomyosin, ▶ vascular endothelial growth factor (VEGF) receptor 2, and matrix metalloproteinase-2 (MMP-2) as potential targets for endostatin. Genetic Variations Mutations in collagen XVIII/endostatin gene have been identiﬁed in humans. Absence of collagen XVIII/endostatin is associated with Knobloch syndrome (an autosomal recessive disease that is characterized by ocular defects leading to

1531

retinal detachment, macular degeneration, and blindness). In mice, collagen XVIII/endostatin gene knockout did not affect developmental angiogenesis but however showed progressive loss of vision. Other mutation, such as D104N substitution in endostatin, has been observed in Knobloch syndrome patients (heterozygous, with one allele truncated and the other with a mutation leading to D104N substitution) as well as in many cancer patients. D104N endostatin does not affect the biological activity of endostatin but showed reduced afﬁnity to laminin. Cancer patients having D104N mutation in endostatin did not show any correlation with the disease progression and survival. These studies suggest that loss of collagen XVIII/endostatin does not paralyze the entire vascular system but leads to some selective changes associated with the eye. In contrast, increased expression of collagen XVIII/endostatin has been postulated to reduce the risk of cancer development. Normal serum levels of endostatin in human are between 25 and 30 ng/ml. Higher levels (about 70%) of endostatin have been observed in Down syndrome patients who have an extra copy of chromosome 21 (trisomy). Collagen XVIII gene is located in chromosome 21. In comparison to age matched, control healthy population, a decreased risk for cancer development has been noticed in the Down syndrome patients. Folkman and Kalluri have hypothesized a physiological tumor suppressive role for endostatin based on these observations. Preclinical Studies Endostatin inhibits tumor growth in several model systems. More than ﬁfteen different types of human tumor cell lines (e.g., ▶ ovarian cancer, ▶ breast cancer, glioblastoma, and renal cancer) transplanted into immunodeﬁcient mice were inhibited, to varying degrees by endostatin treatment. Endostatin treatment was successful when initiated at an early stage of tumor growth. In a few studies, endostatin treatment induced the regression of established tumors. Furthermore, twice daily injections of endostatin were found to be more effective than daily bolus injections. Serum half-life (pharmacokinetics) and bioavailability

1532

are responsible for the schedule-dependent differences seen in the efﬁcacy of endostatin treatment. This notion was further supported by the observation that tumor growth was better inhibited when endostatin was delivered by mini-osmotic pumps and in slow-release formulations. Endostatin is well tolerated in mice and does not show any toxicity.

Clinical Trials In 1999, human trials were initiated at three institutions, Dana-Farber Cancer Institute, Boston; University of Wisconsin, Madison; and MD Anderson Cancer Center at Houston. Recombinant endostatin expressed in yeast was used in these phase I trials. A total of 61 patients with advanced disease of different tumor types were enrolled into these studies. Recombinant endostatin was administered either as a 20 min or 1 h intravenous infusion. Dosages ranged from 15 to 600 mg/m2/day. There was no dose-limiting toxicity observed in all the three trials. Serum half-life of endostatin was found to be between 10 and 11 h. In two of the 25 patients treated at the MD Anderson Cancer Center, there was evidence of minor antitumor effect. Out of the total of 15 patients enrolled in Dana-Farber Cancer Institute, there was a minor response in a patient with a pancreatic neuroendocrine tumor, while two of the other patients showed stabilization of the disease. In the University of Wisconsin trial, there was no objective clinical response in any of the 25 patients treated with endostatin. Dynamic CT scans of the patients treated with endostatin showed some evidence of changes in the microvessel density. Other imaging methods to study functional changes in tumor vasculature indicated changes in blood ﬂow and metabolism following endostatin treatment. A deﬁnitive antiangiogenic effect in treated patients could not be documented due to the lack of suitable biomarkers to validate changes in tumor angiogenesis. Another phase I study was undertaken at the Vrije Universiteit Medical Center in the Netherlands. Thirty-two patients received recombinant endostatin as a continuous infusion for 4 weeks.

Endostatin

Treatment was continued after 1 week of rest by twice daily subcutaneous injections. This trial also noted no adverse side effects in the treated patients, and again there was no objective clinical response. However, two patients had a longlasting stable disease. Since one of the patients with pancreatic neuroendocrine tumor showed partial response in phase I trial, a phase II trial was initiated at multiple centers. Forty patients with advanced neuroendocrine tumors were treated with recombinant endostatin by daily subcutaneous injections. Even though a steady-state level of potentially effective serum concentration of endostatin was achieved in these patients, no partial or clinical response was observed. While it is disappointing to note that the antitumor effects seen in experimental animals could not be replicated in these early clinical trials, a lot has been learned on the pharmacological properties of endostatin. Future studies will focus on combining endostatin treatment with other modalities to improve the antitumor effects. Indeed, the inhibition of tumor growth by radiotherapy and ▶ chemotherapy is potentiated by endostatin treatment. Furthermore, ▶ gene therapy approaches using viral vectors have shown promising effects in experimental animals. In fact, adeno-associated virus (AAV)-mediated expression of endostatin is in the early phases of clinical development. The surgical removal of tumors followed by chemotherapy in combination with endostatin treatment is a promising strategy for cancer treatment.

References Benezra R, Raﬁi S (2004) Endostatin’s endpoints – deciphering the endostatin antiangiogenic pathway. Cancer Cell 13:205–206 Folkman J (2006) Antiangiogenesis in cancer therapy – endostatin and its mechanism of action. Exp Cell Res 312:594–607 Kalluri R (2003) Basement membranes: structure, assembly and role in tumour angiogenesis. Nat Rev Cancer 3:422–433 Marneros AG, Olsen BR (2005) Physiological role of collagen XVIII/endostatin. FASEB J 19:716–728

Endothelial-Derived Gene-1

Endothelial Cell-Specific Molecule-1

1533

immunodeﬁcient mice, EG-1 overexpressed cells develop into larger xenograft tumors (Sato et al. 2004).

▶ Endocan

Endothelial Transglutaminase ▶ Transglutaminase-2

Endothelial-Derived Gene-1 Mai N. Brooks Surgical Oncology, School of Medicine, University of California, Los Angeles, CA, USA

Synonyms Magicin; Med28

Definition Endothelial-derived gene EG-1 was discovered in 2002. Although ﬁrst cloned from a human endothelial cell cDNA library (▶ Angiogenesis), EG-1’s transcript has been shown to be present in other cell types as well, particularly in epithelial cells. The calculated mass of EG-1 is 19,520.2 Da based on amino acid sequence alone, whereas the native complete protein is ~22 kDa. The homology between the human EG-1 peptides to its mouse counterpart is 94.9%, 95.5% to its rat counterpart, and 31% to the Drosophila one.

Characteristics EG-1 is strongly associated with cellular proliferation. Overexpression of EG-1 achieved by transfection results in increased proliferation of multiple human cell lines in culture. When these cells are injected subcutaneously into

Mechanisms Overexpression of EG-1 results in activation of the mitogen-activated protein kinase (MAPK) pathway (▶ MAP kinase), which has been shown to be crucial in promoting cellular proliferation. This manifests as increased levels of phosphorylated p44/42 MAP kinase, phosphorylated JNK (Jun-terminal kinase) (▶ JNK Subfamily), and phosphorylated p38 kinase. EG-1 overexpression also results in c-Src activation. c-Src is a member of the Src family of cytoplasmic tyrosine kinases that regulate cell growth, differentiation, cell shape, migration, and survival (▶ Src). c-Src has been reported to be overexpressed and to play a role in human carcinomas of the breast, colon, and others. Src family tyrosine kinases are often activated by ▶ receptor tyrosine kinases, such as EGF-R (▶ epidermal growth factor receptor) (EGR ligands) or PDGF-R (▶ platelet-derived growth factor receptor) (PDGF). Via its proline-rich region, EG-1 binds to the Src family of protein tyrosine kinases c-Src and Yes and possibly FYN and Hck (hemopoietic cell kinase). As a result of EG-1 binding, c-Src then becomes catalytically active. However, EG-1 is not a direct substrate of c-Src, nor does it increase c-Src expression (Lu et al. 2005). Two proteins with identical sequences to EG-1 have been identiﬁed: Magicin for merlin and Grb2-interacting cytoskeletal protein in 2004 (Wiederhold et al. 2004). Magicin is described to associate with the actin cytoskeleton and is proposed to have a role in receptor-mediator signaling at the cell surface. Magicin binds directly to Grb2 (growth factor receptor bound 2 protein). Med28 is a member of the Mediator, a multiprotein transcriptional coactivator that is expressed ubiquitously in eukaryotes for induction of RNA polymerase II transcription by DNA-binding transcription factors. As Med28, this protein is one subunit of the “adaptor” that bridges RNA polymerase II with its DNA-binding

1534

regulatory proteins and transduces both positive and negative signals. In summary, EG-1 is an important protein that has multiple interactions with crucial cellular pathways involved in cellular proliferation, cytoskeletal function, and transcriptional regulation. EG-1 in Human Cancer Immunohistochemistry of human samples demonstrates that EG-1 is present in the nucleus as well as in the cytoplasm and possibly in the cell membrane. There are signiﬁcantly higher levels of EG-1 peptides in cancer specimens of the breast, colon (Colon cancer), and prostate (Prostate carcinoma), in comparison with their benign counterparts (Zhang et al. 2004) (▶ Cancer). EG-1 has been detected at elevated levels in sera from breast cancer patients (▶ Serum biomarkers). In human urine, EG-1 appears primarily at a larger molecular weight, suggesting that these peptides may co-aggregate or associate with other moieties in urine (Biomarkers). Thus, EG-1 may be secreted or it may be shed with cell death/turnover.

Endothelial-Derived Gene-1

Cross-References ▶ Angiogenesis ▶ Breast Cancer ▶ Cancer ▶ Chemotherapy ▶ Epidermal Growth Factor Receptor ▶ Epithelial Tumorigenesis ▶ Estrogen Receptor ▶ JNK Subfamily ▶ MAP Kinase ▶ Molecular Therapy ▶ Mouse Models ▶ Neoadjuvant Therapy ▶ Platelet-Derived Growth Factor ▶ Prostate Cancer Clinical Oncology ▶ Receptor Tyrosine Kinases ▶ Serum Biomarkers ▶ SiRNA ▶ Small Molecule Drugs ▶ Src

References Translational Aspects Studies have shown that endogenous EG-1 can be targeted to inhibit breast tumor growth (Lu et al. 2007). This inhibition, whether delivered via siRNA lentivirus (▶ siRNA) or polyclonal antibody, results in decreased cellular proliferation in culture and smaller xenograft tumors in mice (▶ Mouse models). The effects are shown in both ER (estrogen receptor)-positive human breast cancer MCF-7 cells (▶ Estrogen receptor) and in ER-negative MDA-MB-231 cells. As breast cancer is the most common malignancy diagnosed in women and as one-third of these patients will die of their disease, a novel target for breast cancer therapeutic development such as EG-1 would be very useful (▶ Breast cancer). Because EG-1 is unique, its use may not be redundant to other gene products/potential targets involved in other molecular pathways (▶ Molecular therapy; ▶ small molecule drugs). Further preclinical studies are warranted to explore the usefulness of targeting EG-1 for future cancer therapy.

Lu M, Zhang L, Maul RS et al (2005) The novel gene EG-1 stimulates cellular proliferation. Cancer Res 65:6159–6166 Lu M, Zhang L, Sartippour MR et al (2006) EG-1 interacts with c-Src and activates its signaling pathway. Int J Oncol 29:1013–1018 Lu M, Sartippour MR, Zhang L et al (2007) Targeted inhibition of EG-1 blocks tumor growth. Cancer Biol Ther 6:936–941 Sato S, Tomomori-Sato C, Parmely TJ et al (2004) A set of consensus mammalian mediator subunits identiﬁed by multidimensional protein identiﬁcation technology. Mol Cell 14:685–691 Wiederhold T, Lee MF, James M et al (2004) Magicin, a novel cytoskeletal protein associates with the NF2 tumor suppressor merlin and Grb2. Oncogene 23:8815–8825 Zhang L, Maul RS, Rao J et al (2004) Expression pattern of the novel gene EG-1 in cancer. Clin Cancer Res 10:3504–3508

See Also (2012) Biomarkers. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, pp 408–409. doi:10.1007/978-3-642-164835_6601

Endothelins

Endothelial-Leukocyte Adhesion Molecule 1 (ELAM1) ▶ E-Selectin-Mediated Adhesion and Extravasation in Cancer

Endothelin-2 ▶ Endothelins

Endothelins Matthew J. Grimshaw Breast Cancer Biology Group, King’s College London School of Medicine, Guy’s Hospital, London, UK

Synonyms EDN; Endothelin-2; ET; ET-2; vasoactive intestinal contractor; VIC

Definition Endothelins (ETs) are a family of three similar, small peptides that are among the strongest vasoconstrictors known and play a key part in vascular homeostasis. Endothelins have numerous roles in tumors including modulating ▶ angiogenesis and blood ﬂow, inducing mitogenesis and ▶ invasion of tumor cells, immune activation, and protecting cells from ▶ apoptosis.

Characteristics Endothelins (ETs) are a family of small, structurally related, vasoactive peptides that have a variety of physiological roles in many tissues, notably vascular homeostasis. The “ET axis” consists of

1535

three peptides, two receptors, and two activating enzymes (Table 1). Some examples of the roles of ETs in both normal physiology and pathological conditions are shown below: 1. Blood vessels: Maintain basal level of vasoconstriction (“contraction of blood vessels” which controls blood pressure). Involved in the development of hypertension and atherosclerosis. 2. Heart: Affect the force and rate of contraction of the heart. Mediate hypertrophy and remodeling in congestive heart failure. 3. Lungs: Regulate the tone of airways and blood vessels. Involved in pulmonary hypertension. 4. Kidney: Controls water and sodium excretion and acid–base balance. Participate in renal failure. 5. Brain: Modulates cardiorespiratory centers and hormone release. 6. Cancer: Numerous tumors – including carcinomas of the breast, lung, prostate, and ovary – produce one or more of the ETs and their receptors, and there are many potential roles of the “ET axis” in cancer. Mitogenesis: Endothelins have a mitogenic (“growth promoting”) effect on both tumor and stromal (i.e., the noncancer cell component of a tumor including blood vessels, immune cells, and ﬁbroblasts) (stroma) cells and enhance tumor growth. Tumor angiogenesis: Angiogenesis is the growth of new blood vessels and is critical for the growth of a solid tumor. ETs stimulate angiogenesis within solid tumors by acting directly on endothelial cells to modulate proliferation, migration, invasion, and morphogenesis. ETs also modulate angiogenesis indirectly through induction of the angiogenic cytokine ▶ vascular endothelial growth factor (VEGF) expression. ETs also affect blood ﬂow through the established tumor vasculature due to their vasoactive nature. However, the effect of ETs on blood ﬂow appears to be tissue and tumor speciﬁc. Tumor invasion and metastasis: Invasion is the process by which cancer cells spread beyond the border of the tumor enabling the tumor to grow and spread. ETs stimulate invasion of

1536

Endothelins

Endothelins, Table 1 Genes and peptide sequences of ET axis members. Amino acids that differ in ET-2 and/or ET-3 from ET-1 are marked in bold/red Gene EDN1 EDN2 EDN3 EDNRA EDNRB ECE1 ECE2

Mapping position 6p24 1p34 20q13 4q31 13q22 1p36 3q28-29

Peptide or protein ET-1 ET-2 ET-3 ET-RA ET-RA ECE-1 ECE-2

several types of tumor cells including ▶ P-Glycoprotein family cells, Ewing’s sarcoma and neuroblastoma cells, and ▶ breast cancer cells. Stimulation of, for instance, breast tumor cell lines with ETs leads to an invasive phenotype via several autocrine and paracrine mechanisms including induction of ▶ matrix metalloproteinases (MMPs) and ▶ chemokine receptors and the activation of macrophages. Protection from apoptosis: ETs can protect several cell types – including tumor cells, macrophages, and endothelial cells – from apoptosis (“programmed cell death”) induced by cellular stresses including ▶ hypoxia, serum starvation, and chemotherapeutic agents. Immune modulation: Trafﬁcking, differentiation, and activation of tumor-inﬁltrating immune cells are all modulated by ETs. The Endothelin Axis. The ET axis consists of three 21 amino acid (aa) peptides (ET-1, ET-2, and ET-3) (Fig. 1), two G-protein-coupled receptors (ET-RA and ET-RB), and two membrane-bound endothelin-converting enzymes (ECE-1 and ECE-2). ET-1 was initially found in the conditioned medium of cultured endothelial cells, and its activity as a potent vasoconstrictive peptide was described. ET-2 and ET-3 were rapidly described following ET-1’s discovery, and further roles in a variety of tissues have been described. The three ET isoforms – which are highly conserved in human, rat, and mouse – derive from three separately regulated genes yet have a similar structure. Human ET-1 derives from a 212 aa precursor, preproendothelin-1. The removal of the signal sequence generates the 195 aa proendothelin-1, which is further processed to release the intermediate 38 aa “big

ET isoform peptide sequence CSCSSLMDKECVYFCHLDIIW CSCSSWLDKECVVFCHLDIIW CTCFTYKDKECVYYCHLDIIW – – – –

* * S * * * S S L C C NH2 ** M D K E

C H V Y

I I

W COOH

Endothelins, Fig. 1 Structure of ET-1. ET-1 is a 21 amino acid peptide with a hydrophobic C-terminus and two disulﬁde bonds at the N-terminus. ET-2 and ET-3 are structurally similar to ET-1, differing by two and six amino acids, respectively. The amino acids which differ in the ET-2 sequence are indicated by “*,” while those which differ in ET-3 marked by “*”

ET-1.” ECEs hydrolyze big ET-1 to yield the active 21 aa ET-1. The gene for each ET has a distinct pattern of tissue expression: ET-1 is expressed by endothelial cells of many organs, ET-2 is in the ovary and intestine, and ET-3 is found in the brain. There is a relatively low basal level of synthesis of ETs, but these genes are readily inducible by inﬂammatory stimuli. Endothelin Receptors. Two receptors for ETs have been characterized: ET-RA (also known as EDNRA or ETAR) and ET-RB (EDNRB, ETBR). Both receptors are expressed in a wide variety of tissue types. ETs bind these receptors with varying afﬁnity: ET-RA binds ET-1 ET-2 > ET-3, but ET-RB shows no selective afﬁnity for any ET subtype. Binding of the ligands to these G-protein-coupled receptors (GPCRs) modulates several overlapping signaling pathways resulting in the activation of phospholipase C and MAPK

Endothelins

pathways, an increase in intracellular calcium and the induction of immediate early genes. Induction of Endothelin Expression by the Tumor Microenvironment. One region of tumor, compared to another, may differ in the levels of hypoxia, cytokine concentration, immune inﬁltrate, vascularization, necrosis, etc. The “▶ tumor microenvironment” – particularly hypoxia and soluble factors such as cytokines – modulates expression of numerous “pro-tumor” genes, including those of the ET axis. Transcriptional regulation of numerous hypoxia-responsive genes is via the hypoxia-induced transcription factor, HIF-1, which initiates transcription of genes whose promoter contains a hypoxia response element (HRE). Hypoxia induces ET axis transcription in several cell types including endothelial and tumor cells. There is a functioning HRE in the antisense strand of the promoter of ET-1, and induction of ET expression by hypoxia is via HIF-1. Endothelin Receptor Antagonists. The role of ETs in vasoconstriction has led to the development of several antagonists of the ET receptors that are currently under investigation for the treatment of hypertension, heart failure, and renal disease. Several are now in phase II and III trials for the treatment of various neoplasms, particularly prostate cancer. ET receptor antagonists hold the attractive possibility that they will “hit” several different cell types and mechanisms of cancer progression. Of the antagonists available, it is the modiﬁed peptide-based antagonists BQ123 (ET-RA antagonist) and BQ788 (ET-RB antagonist) that have been used extensively both in vitro and in vivo. Small molecule antagonists such as atrasentan, a highly selective ET-RA antagonist, have been used clinically. These antagonists can be administered orally, are well tolerated, and have few toxic side effects. ET receptor antagonists inhibit proliferation of Kaposi sarcoma cells, Ewing’s sarcoma and neuroblastoma cells, melanoma cells, and ovarian carcinoma cells. As well as the commercially produced antagonists, it is of interest that ET activity may be modiﬁed by dietary factors. An extract of red wine polyphenols causes inhibition of ET-1

1537

synthesis in endothelial cells; this is associated with modiﬁcations in phosphotyrosine staining, indicating that the active components of red wine cause speciﬁc modiﬁcations of tyrosine kinase signaling. Green tea polyphenol epigallocatechin3-gallate inhibits the ET axis and downstream signaling pathways in ovarian carcinoma. Endothelin Expression in Cancer Numerous types of tumors produce one or more of the ETs and their receptors. However, the expression and actions of ETs in cancer are incompletely described and are tumor-type speciﬁc. ET axis expression is increased in many types of tumor, yet in several types of tumor, expression of the ET axis – particularly the receptors – is decreased in neoplastic tissue. For instance, in carcinomas of the breast, both ET-RA and -RB are increased, yet in prostate cancer, ET-RB is decreased, while in lung cancer, ET-RA is downregulated. The function of the ET-RB receptor in tumors is particularly enigmatic; in some cases, such as breast cancer, ligand binding to ET-RB initiates several pro-tumor actions, such as promoting invasion, yet in prostate cancer, the loss of ET-RB expression is postulated to increase ET-1 peptide in the tumor due to the loss of ET-RB’s ET clearance function. Ovarian Carcinoma. ETs have several roles in ovarian tumors including promoting growth and invasion, and ETs stimulate both tumor and stromal cells. Ovarian carcinoma cells secrete ET-1, which acts as an autocrine growth factor via ET-RA and also has a paracrine growth effect on the ﬁbroblastic cells via both receptors. ETs acting through ET-RA promote invasion of the tumor cells by upregulating secretion and activation of MMPs. ET-RA is found in both tumor cells and intratumoral vessels, whereas ET-RB is expressed mainly in endothelial cells, and ETs induce angiogenesis via hypoxia-inducible factor (HIF)-1a and VEGF. Atrasentan decreases growth of ovarian xenografts in mice, and this is associated with decreased angiogenesis and MMP expression and increased apoptotic tumor cells. Prostate Cancer. The role of ETs has been studied in both the normal and transformed prostate. ETs are produced in the normal prostate gland by epithelial cells and are found in high

1538

Endothelins

ET-1 ET-2

Survival chemotaxis invasion

ET-1 ET-2 Tumour cell

Macrophage

Chemotaxis activation MMP induction

Chemotaxis

xis ota em Ch

= ET-RA MMPs

ET-1

m Tu Smooth muscle cell

Proliferation angiogenesis

ion

as nv

ll i

Endothelial cell

Proliferation chemotaxis invasion angiogenesis

= ET-RB

Endothelins, Fig. 2 Putative roles of ETs in breast tumors. TAMs express inﬂammatory cytokines that may induce ET expression by both tumor cells and macrophages, which release ETs and express both receptors. Microenvironmental factors, including hypoxia and the ETs themselves, further stimulate the ET axis. Stimulation of tumor cells and macrophages with ETs leads to

chemotaxis of these cells, induction of MMP activity and cytokine expression, and invasion of the tumor cells. ET-2 also protects tumor cells from apoptosis and activates macrophages. Endothelial cells and VSMCs express ET-RB and ET-RA, respectively. Stimulation of endothelial cells and VSMCs with ETs may stimulate angiogenesis

concentrations in seminal ﬂuid (up to 5 mg/l). ET-RA and ET-RB are found in normal prostate tissue, but in the malignant prostate, there is a loss of ET-RB and increased levels of ET-1. Roles for ETs in prostate cancer include growth promotion, apoptosis inhibition, and bone formation. Atrasentan delays time to progression in prostate cancer in phase III clinical trials. ETs modulate nociception (“sense of pain”), and ET-RA antagonism attenuates prostate carcinomainduced pain. Breast Cancer. There are numerous potential consequences of ET expression in breast tumors that may lead to a more aggressive tumor cell phenotype (Fig. 2). These include the induction of invasion and angiogenesis via stimulation of tumor cells and ▶ tumor-associated macrophages (TAMs). There is increased expression of the ET

axis in invasive ductal carcinoma (IDC) of the breast compared to the normal breast; lymph node metastases (▶ metastasis) have a higher degree of ET staining still. Cells expressing ETs and their receptors in IDC include the tumor cells, the CD68+ macrophage inﬁltrate, and the endothelial cells. ETs have a role in recruiting TAMs – macrophages express both ET receptors and chemotax toward ETs via ET-RB and a MAPK-mediated signaling pathway. Exposure of macrophages to ETs leads to an “activated” phenotype and cytokine secretion. Macrophages not only react to ETs but also produce ETs themselves, and the TAMs contribute to the ETs in the breast tumor microenvironment. ETs induce expression of chemokine receptors including CCR7 and potentiate the

Endothelins

response of breast tumor cells to chemokines including CXCL12 and CCL21, which modulate the organ speciﬁcity of breast cancer metastasis. A further potential function of ETs in breast cancer is the modulation of angiogenesis. Boyden chamber (a mixed ET-RA/B antagonist) inhibits tumor vascularization and bone metastasis in a murine model of breast carcinoma cell metastasis. Expression of ET-RA predicts unfavorable response to neoadjuvant chemotherapy in locally advanced breast cancer. ▶ Melanoma. The ET axis may be a promising therapeutic target for the treatment of melanomas. Activation of ET-RB promotes melanocyte precursor cell proliferation while inhibiting differentiation, two hallmarks of malignant transformation. In melanoma cell lines, ETs prevent apoptosis, and ET-RB antagonists cause an increase in cell death. ETs are also involved in angiogenesis in mouse models of melanoma. In vivo, BQ788 slows growth of human melanoma tumors in nude mice. A phase II study of bosentan as monotherapy in patients with stage IV metastatic melanoma showed disease stabilization in 6 of 32 patients. ▶ Lung Cancer. ET-1 has been proposed as a prognostic marker in non-small cell lung carcinoma (NSCLC). There is higher expression of ET-1, ET-RA, and ECE-1 in lung tumors compared to the normal tissue, while ET-RB is decreased. Interestingly, ET-1 is increased in the breath condensate of NSCLC patients, and this could potentially be used as a noninvasive test for early detection of NSCLC. ▶ Bladder Cancer. The ET axis, particularly ET-RB, is overexpressed in bladder cancer. Patients with ET-RB expression tend to have organ-conﬁned tumors and no vascular invasion, and as such, ET-RB is associated with favorable disease-free survival. When metastatic bladder carcinoma cells were injected into mice treated with atrasentan, there was a dramatic reduction of metastases to the lungs. ▶ Nasopharyngeal Carcinoma. Elevated plasma big ET-1 is associated with distant failure in patients with advanced-stage nasopharyngeal carcinoma.

1539

▶ Cervical Cancer. In human papillomaviruspositive cervical cancer cells, ET-RA mediates an ET-induced mitogenic effect. Atrasentan inhibits growth and angiogenesis in cervical cancer xenografts.

Cross-References ▶ Angiogenesis ▶ Apoptosis ▶ Bladder Cancer ▶ Breast Cancer ▶ Cervical Cancers ▶ Chemokines ▶ Hypoxia ▶ Invasion ▶ Lung Cancer ▶ Matrix Metalloproteinases ▶ Metastasis ▶ Nasopharyngeal Carcinoma ▶ P-Glycoprotein ▶ Tumor-Associated Macrophages ▶ Tumor Microenvironment ▶ Vascular Endothelial Growth Factor

References Grimshaw MJ (2005) Endothelins in breast tumour cell invasion. Cancer Lett 222(2):129–138 Kedzierski RM, Yanagisawa M (2001) Endothelin system: the double-edged sword in health and disease. Annu Rev Pharmacol Toxicol 41:851–876 Nelson J, Bagnato A, Battistini B et al (2003) The endothelin axis: emerging role in cancer. Nat Rev Cancer 3(2):110–116

See Also (2012) Atrasentan. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 303. doi:10.1007/978-3-642-16483-5_444 (2012) Boyden Chambers. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 465. doi:10.1007/978-3-642-16483-5_696 (2012) Endothelin Converting Enzyme. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1254. doi:10.1007/978-3-642-164835_1903 (2012) ET-RA. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1339. doi:10.1007/978-3-642-16483-5_2024

1540 (2012) ET-RB. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1339. doi:10.1007/978-3-642-16483-5_2025 (2012) G-protein Couple Receptor. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1587. doi:10.1007/978-3-642-16483-5_2294 (2012) Invasive Ductal Carcinoma. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1906. doi:10.1007/978-3-642-16483-5_3134 (2012) Lymph Node Metastases. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 2116. doi:10.1007/978-3-642-16483-5_3444 (2012) Prostate Cancer. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 3009–3010. doi:10.1007/978-3-642-16483-5_6576 (2012) Stroma. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 3541. doi:10.1007/978-3-642-16483-5_5532 (2012) Stromal Cells. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 3544. doi:10.1007/978-3-642-16483-5_5535

Endotoxin-Induced Factor in Serum

Enzymes ▶ Antioxidant Enzymes

Enzymic Mouth to Mouth Feeding ▶ Substrate Channeling

Ep ▶ Erythropoietin

EpCAM Endotoxin-Induced Factor in Serum ▶ Tumor Necrosis Factor

M. Asif Chaudry University Department of Surgery, Royal Free and University College London Medical School, London, UK

Engineered Antibody Synonyms ▶ Diabody 17-1A; EGP-2; EGP34; EpCAM; Epithelial cell adhesion molecule; ESA; GA733-2; HEA125; KSA; MK-1; TROP-1

Enlarged Breast Male ▶ Gynecomastia

ENPP2 ▶ Autotaxin

Enteropeptidase (TMPRSS15) ▶ Serine Proteases (Type II) Spanning the Plasma Membrane

Definition A membrane protein found on all simple epithelia to varying degrees is a type I membrane protein. It is a pan-epithelial differentiation antigen expressed on the basolateral surface of all carcinomas, varying in density and glycosylation. As a homotypic cell adhesion molecule, it is intimately integrated within the cadherin-catenin and ▶ Wnt signaling pathways. It modulates the expression of proto-oncogenes such as ▶ Myc oncogene. Its status as a pan-carcinoma antigen has rendered it an attractive target for cancer ▶ immunotherapy.

EpCAM

Characteristics Structure and Function Structure

This 37 kDa protein is formed from 314 amino acids (aa) of which only 26 aa face the cytoplasm. The extracellular component contains three domains: the ﬁrst is novel and is the site to which most of the antibodies developed are targeted (323 ~ A3, 17-1A and others). The second is similar to EGF-binding proteins 1 and 6 and thyroglobulin. The third has a novel structure that also has similarities with EGF. The intracellular portion of the antigen has a tyrosine phosphorylation site, the signiﬁcance of which is uncertain. Tissue Morphogenesis

EpCAM is essential for stable adhesion formation and tissue morphogenesis similar to adhesion molecules: ▶ carcino-embryonic antigen (CEA) and ICAM-1. The mechanism by which cytoskeletal and intracellular elements mediate this function is being characterized. EpCAM inhibits intercellular adhesion mediated by E-cadherins, in turn interacts with a-, b-, and g-catenins forming the cadherin-catenin complex. Catenins link cadherins with the actin cytoskeleton and form complexes with other proteins. Cadherins are crucial for the establishment and maintenance of epithelial cell polarity, morphogenesis of epithelial tissues, and regulation of cell proliferation and apoptosis. Their association with b-catenin is particularly interesting as this is a component in the ▶ Wnt signaling pathway that regulates the expression of proto-oncogenes such as c-Myc: fundamentally associated with tumor development. Wnt glycoproteins are signaling molecules that regulate cell-to-cell interaction during embryogenesis. Wnt proteins bind to receptors of the Frizzled family. Through several cytoplasmic relay components, the signal is transduced to b-catenin, which is stabilized, accumulates in the cytoplasm, and enters the nucleus, where it binds a lymphoid enhancer factor/T-cell factor transcription factor. Together, b-catenin and lymphoid enhancer factor/T-cell factor activate expression of many target genes, such as Myc

1541

oncogene, VEGF, and cyclooxygenase-2, all associated with neoplasia. EpCAM directly impacts the cell cycle by upregulating c-Myc and cyclin A/E. Human epithelial cells expressing EpCAM reduce growth factors’ dependency and increase metabolism and colony formation. Inhibition of EpCAM expression with antisense nucleic acid reduces proliferation and metabolism in human carcinoma cells. The intracellular domain is essential for these effects. EpCAM adhesive properties promote calciumindependent homotypic cell sorting. Cells transfected to express EpCAM are sorted from cells of the same line that do not normally express EpCAM. It also inhibits invasive growth in cell colonies. Both activities are inhibited by antiEpCAM antibodies. The function of EpCAM thyroglobulin domain is being actively investigated. These domains commonly inhibit cathepsins: cysteine proteases frequently produced by tumor cells and known to be involved in metastasis. Pattern of Tissue Expression Normal Tissue

EpCAM is present on all normal epithelia excluding stratiﬁed squamous epithelia. Within the gastrointestinal (GI) tract, colonic expression is greatest and gastric lowest. Glandular GI epithelium displays a marked expression gradient from crypts to the apex of villae. Abnormal Tissue

Carcinomas and actively proliferating tissues show increased and differential expression of EpCAM. Expression correlates with differentiation in gastric lesions. Immunochemical and mRNA studies show that well-differentiated tumors are more expressive than those less differentiated. Normal background mucosa shows weak expression, but interestingly areas of ▶ Barrett esophagus or metaplasia are highly expressive. Ninety percent of colorectal carcinoma cells express EpCAM but in a differential form. Modiﬁcations include variable glycosylation

1542

analogous to tumor-speciﬁc antigens such as colonic tumor antigen MUC1. EpCAM exists in the cell membrane of colon carcinoma cells as a high afﬁnity noncovalent cis-dimer. Dimers on opposing membranes can associate via a headto-head interaction to form tetramers with moderate afﬁnity consistent with reversible intercellular associations. It is not known how exactly antibody binding correlates with variable glycosylation or oligomerization in a functional or structural sense; further investigation is required. Tissue microarray assessment of EpCAM expression in 3,900 tissues of tumor of stratiﬁed stages and grades of 134 different histological subtypes sourced from head and neck, lung, gastrointestinal, breast, urogenital, and mesenchymal tumors showed 75% tumor categories expressed EpCAM. At least weak EpCAM expression in >10% of tumors was observed in 87 of 131 different tumor categories. Colon cancer (81%), ▶ gastric cancer, pancreas cancer (78%), and ▶ lung cancer revealed a high proportion of strongly positive tumors suggesting EpCAM is an attractive target for pan-carcinoma ▶ immunotherapy. Paradox of Expression with Advance in Carcinomas A simple linear pattern of an increase in EpCAM expression with the progression of all tumors is not seen. The exact nature of EpCAM temporal expression vis-a-vis the grade of different tumor types remains to be stratiﬁed. The functionally paradoxical upregulation of EpCAM with disease progression in colorectal, breast, prostate, and upper GI carcinomas remains unexplained. This is intriguing in metastatic carcinoma in which degradation of intercellular adhesions is a primary feature. Perhaps, EpCAM is upregulated in response to other intra- and extracellular processes that promote destruction of tissue adhesion and morphology, maintaining its constitutional stabilizing function. Conversely in some tumors, e.g., colorectal cancer, a loss of EpCAM expression is associated with increased local recurrence risk and a diffusely inﬁltrative morphology but not distant recurrence. In prostate cancer, reduced EpCAM expression correlates

EpCAM

with a higher Gleason score, but expression is higher on hormone-refractory tumor tissue than at earlier stages. In cholangiocarcinomas, squamous cell carcinoma of the head and neck and esophagus increased expression correlates with reduced survival. The relative loss of EpCAM expression in patients with gastric cancer is associated with a signiﬁcant reduction in survival indicating that loss of EpCAM expression identiﬁes aggressive tumors especially in patients with stages I and II disease. Data from a Dutch study compared p53, ▶ CD44, E-cadherin, EpCAM, and c-erB2/neu in tumors of 300 patients, investigating the extent of lymph node clearance. Patients without loss of EpCAM expression of tumor cells (19%) had a signiﬁcantly better 10-year survival compared to patients with any loss: 42% versus 22%. The prognostic value was stronger in stages I and II and independent of the TNM stage. Similarly, in breast cancer, a relative reduction in EpCAM expressing disseminated tumor cells in the bone marrow of patients is associated with a relatively poorer prognosis, whereas an increase in EpCAM-positive tumor cells in lymph nodes and peripheral blood is associated with reduced survival. In prostate cancer patients staged as M0 (no metastasis), BR (biochemical PSA relapse), and M1 (established metastasis), the presence of bone marrow EpCAM expressing tumor cells signiﬁcantly increased with progression from M0 to BR and M1 stages from 9 to 16–33%. These cells had double the chromosomal aberrations compared to cytokeratin-positive tumor cells. There was only a small overlap between EpCAM+ and CK+ DTC populations of 9.5%. EpCAM marked a cohort of DTC in ca prostate patients that unlike CK+ DTC expanded during biochemical relapse and had a phenotype different from that of CK+ tumor cells. This differential expression of EpCAM as compared to cytokeratin indicates a speciﬁc functional role for EpCAM in the development of metastatic precursors over and above the simple role of an adhesion molecule once thought. The possible interaction of EpCAM with immune escape remains to be elucidated. Once neoplastic transformation has taken place, a reduced EpCAM expression is an

EpCAM

indicator of a more aggressive tumor phenotype with increased ▶ invasion, ▶ metastasis, and mortality. This seemingly contradicts a study that suggested EpCAM silencing leads to reduced invasive potential of tumor cells. Breast cancer cell lines were grown in Matrigel invasion/migration chambers. Cells in which EpCAM expression was silenced with ▶ SiRNAs showed a reduction of 35–80% in proliferation, 92% in cell migration, and 96% in cell invasion without increase in cell death or apoptosis. There was however an increase in E-cadherin, a-catenin, and b-catenin. This may be due to silencing of the inhibition that EpCAM exerts on E-cadherin. Alternatively, EpCAM gene silencing may lead to decreased cytoplasmic b-catenin through an increase in its association with the E-cadherin adhesion complex. Hence, reducing EpCAM may decrease b-catenin availability for the Wnt pathway and activation of its target genes downstream. A process whereby EpCAM expression increases up to a point after which destabilizing factors predominate and its expression is no longer stimulated is plausible. Once this point is reached, which may be variable according to the particular tumor type, tumors are less stable and display greater invasive and metastatic potential. A comprehensive study of the temporal expression of EpCAM during tumor progression is currently lacking. This would be useful as a predictor of the efﬁcacy of immunotherapy in individual patients according to their tumor stage. A tenfold reduction in expression of EpCAM is seen in ▶ circulating tumor cells compared to primary tumors from whence they emerged and established metastases. Absence of homotypic adhesions stimulating EpCAM expression in the vascular microenvironment may be the causal link. EpCAM-targeted immunotherapy may be more effective in established tumors or metastases as opposed to ﬂuid borne disease. This does not preclude ascites due to peritoneal metastases for which treatment with trifunctional antibodies is effective. However, the efﬁcacy of destroying blood-borne circulating tumor cells in a hope of eradicating minimal residual disease may be limited.

1543

EpCAM-Targeted Immunotherapy Immunotherapy manipulates competent host immune system inducing tumor growth inhibition, regression, or cytolysis. Approaches include the use of monoclonal antibodies and their derivatives, hybrid bispeciﬁc (trifunctional) antibodies, tumor cell vaccines, anti-idiotypic antibodies, and dendritic cell vaccines. Pure immunomodulatory cytokines have been used to enhance the effect of MAbs. Mechanism of Tumor Inhibition The mechanisms by which anti–EpCAM antibodies exert tumor inhibition in vivo remain controversial. Cytotoxic mechanisms include antibody-dependent cellular cytotoxicity (ADCC) mediated by natural killer cells and T lymphocytes, complement-mediated cytolysis (CMC), and opsonization promoting phagocytosis mediated by PMNs. The question of whether anti- EpCAM antibodies directly inhibit tumor cell proliferation remains unanswered. It could be postulated that EpCAM antibodies directly interfere with the activation of the Wnt pathway causing downregulation of c-Myc: this remains untested. The majority of anti-EpCAM antibodies produced are speciﬁc for epitopes within the ﬁrst of two EGF-like domains in the extracellular segment of EpCAM (Fig. 1); none have been shown to mimic the dimerization/tetramerization that EpCAM undergoes on ligation or to interfere with downstream gene activation or cell proliferation in vivo. EpCAM antibodies do obliterate EpCAM-mediated homotypic cell-sorting activity in vitro; this effect may be a competitive event preventing dimerization alone. Although it is unlikely that a similar competitive effect takes place in established tumors, an investigation to see any effect on the establishment of metastases would be interesting. A comparison between any differences in the cytotoxicity of antibodies according to the functional EpCAM domain targeted is awaited. It is possible that the majority of these antibodies work to opsonize cells alone – inducing the cytolytic mechanisms mentioned above – particularly as no physiological ligands for the extracellular domain

1544

EpCAM

Extracellular domain 0 21 27 59 66

Cytoplasmic domain 135

267 288

314

Cyst SP

AUA1 Ber-EP4 CO17-1A C215

EGF I

GA733-2 MOC31 VU-1D9 K931 323/A3 KS1/4 Mh99

EGF II

2GB

Cystein poor region

311/K1

MM104

EpCAM, Fig. 1 Protein domain structure

of EpCAM other than EpCAM itself have been identiﬁed. EpCAM forms a complex with the tight junction protein Claudin-7 within its intramembranous segment; the physiological signiﬁcance of this is not yet known although an effect on apoptosis resistance in tumors is intriguing. A ﬂurry of interest followed the assertion that LAIR-1, a member of the inhibitory group of the immunoglobulin-like receptors, was a novel receptor for EpCAM. Speculation that neoplastic cells escape immunological surveillance and clearance by interacting with LAIR-1 via EpCAM gaining selective advantage for their growth, spread, and dissemination was nulliﬁed when the original paper by Meyaard et al. was retracted because the observed binding of the LAIR-1 to EpCAM transfected cells was an artifact, attributed to the contamination of the LAIR-1 fusion protein preparation with an antihuman EpCAM monoclonal antibody.

Clinical Trials (Table 1)

Monospecific Murine Antibody

The murine Ig2a antihuman 17-1A monoclonal antibody edrecolomab was the ﬁrst immunotherapeutic agent licensed for use in large-scale human antitumor immunotherapy trials. Initial trials in patients with advanced colorectal cancer showed little improvement in morbidity or mortality. Augmentation with interferm and GM-CSF increased ADCC with associated tumor lymphocyte inﬁltration and complement deposition. Patients with greater ADCC survived longer.

In 1994, 189 patients with Dukes C CRC were randomly assigned to adjuvant therapy with edrecolomab or resection alone. Survival at 3 years was 72% for the edrecolomab cohort and 62% for surgery alone. Further follow-up at 7 years showed signiﬁcantly reduced mortality (32%), disease recurrence (23%), and metastases leading to further phases II and III trials. In 2002, Punt published results of a trial of 2,761 patients randomized to MAb 17-1A monotherapy, 5-FU and folinic acid or 5-FU + edrecolomab. No additional beneﬁt was seen by adding immunotherapy to the standard chemotherapy regimen at 26 months. Immunotherapy alone was associated with signiﬁcantly shorter disease-free survival. Edrecolomab was removed from circulation. The discrepancy between preclinical and clinical ﬁndings has led to much debate. What are the reasons for this discrepancy? EpCAM expression density varies at different stages of tumor growth suggesting patient antigen positivity should be assessed prior to clinical use. EpCAM density is a proven predictor of survival in breast cancer patients. As a murine antibody, edrecolomab induces a neutralizing humoral response in humans resulting in a short serum half-life. Foreign MAbs are rapidly cleared as immune complexes depositing in the liver, greatly reducing bioavailability. Reduced compatibility with human effector cells may also be signiﬁcant. EpCAM-targeted immunotherapy to date has targeted advanced disease: its value weighed against classic adjuvant treatments. The effect of such immunotherapy on earlier, less established disease or cancer models, is unknown.

EpCAM

1545

EpCAM, Table 1 Trials to assess efficacy of EpCAM-targeted immunotherapy for intra-abdominal carcinomas Author Weiner et al. (1986)

Herlyn et al. (1994)

Patients 27 metastatic adenocarcinoma of the colon or pancreas Nine CRC

Herlyn et al. (1994)

54 CRC

Fagerberg et al. (1995)

Six CRC

Ragnhammar et al. (1995)

86 Adv CRC

Riethmuller et al. (1998)

189 Dukes C

Shetye et al. (1998)

20 Adv CRC

Hjelm et al. (1999)

20 Adv CRC

Punt et al. (2002)

2761st III CRC

TRION Pharma, Fresenius (2003)

23 symptomatic ascites Ca ovary

Heiss (2005)

Eight peritoneal carcinomatosis

Treatment Passive MAb171A preceded by 4day gIFN Active antiidiotypic CO171A aluminum hydroxide precipitated Active polyclonal goat and monoclonal rat anti-idiotypic CO17-1A Active antiidiotypic CO171A Passive murine MAb17-1A (76) or chimeric MAb171A (10) Passive observation or MAb17-1A adjuvant Passive single infusion MAb171A + GM-CSF Passive MAb171A + IL-2 + GMCSF Passive multicenter; (1) 17-1A MAb/5FU/LV or (2) 5 FU/LV or (3) 17-1A MAb Passive trifunctional multicenter open label intraperitoneal Removab Passive trifunctional, 4–6 applications intraperitoneal

Results No objective clinical markers. Serum tumor markers reduced in 36%. 11 developed Ab3 response Three patients developed Ab3 response to Ab2 determinants

Conclusions MAb 17-1A safe for clinical use. Evidence of anti-idiotypic response Marginal success

Majority developed Ab3 response; 30% developed delayed type hypersensitivity

Anti-idiotypic CO17-1A effective in stimulating long-term immunity in cohort

Six patients developed T-cell immunity; 5 mounted Ab3 response All patients developed antiidiotypic Abs increased by GM-CSF; c-MAb less response and more allergic side effects than MAb 7-year evaluation, mortality decreased by 32% and recurrence by 23%

Small study evidence of anti-idiotypic response

Increased tumoral PMN, monocytes, and T lymphocytes One patient partial remission, 2 patients stable disease for 7 and 4 months 3-year surv DFS (1) 74.7% 63.8% (2) 76.1% 65.5% (3) 70.1% 53.0%

Well-tolerated 22 of 23 patients ascites free at day 37

Patients with Ab2 response – median survival 9/12

Therapeutic effect maintained after 7 years, mortality/recurrence reduced Increased TILs representing ADCC and CTLs No augmentation of effect of MAb 17-1A Addition of edrecolomab to standard therapy does not improve the disease outcome. Panorex withdrawn Effective treatment of malignant ascites phase III for all-cause malignant ascites underway

Seven of eight patients no further paracentesis needed. Eradication of tumor cells in ascites

ADCC antibody-dependent cell cytotoxicity; CRC colorectal cancer; CTL cytotoxic T cells; DFS disease-free survival; GM-CSF granulocyte-macrophage colony-stimulating factor; IFN interferon; MAB monoclonal antibody; PMN polymorphonuclear cells; TIL tumor inﬁltrating lymphocytes

1546

What Are the Solutions? Humanized Antibody

A human IgG antibody, MT201 (adecatumumab) combines binding afﬁnity similar to edrecolomab with considerably enhanced ADCC potency with human gastric carcinoma cell lines. Addition of human serum containing IgG or human peripheral blood monocytes halves MT201 ADCC but abolishes that of edrecolomab: indicating the importance of human anti-mouse antibodies (HAMA) and compatibility of syngeneic effector cells. MT201 reduces tumor growth in xenotransplanted HT-29 CRC cells in nude mice but only to a level similar to edrecolomab. It is hoped that human effector cells with greater type speciﬁcity of Fcg receptors will facilitate ampliﬁed tumor inhibition clinically. Three clinical trials are currently underway: two phase II studies with metastatic breast cancer and early-stage prostate cancer patients, respectively, and a phase I study testing the safety of a combination with Taxotere. Bispecific Antibodies

Structure and Rationale for Development (Fig. 2) Normal IgG molecules compose of Fc and FAb segments. The monospeciﬁc FAb segment binds to speciﬁc epitopes on antigens, whereas the Fc portion recruits cells expressing Fc receptors (e.g., FcgR) such as macrophages. These are described as being bifunctional and monospeciﬁc. In trifunctional antibodies, the two halves of the FAb segment have different speciﬁcity: they are bispeciﬁc and trifunctional. Both edrecolomab and MT201 are bifunctional antibodies: IgG1 and IgG2a, respectively, with active components being the anti-light chains and Fc portions. Zeidler (1999) successfully constructed a bispeciﬁc/trifunctional targeting both and CD3 (BiUII or Removab). The rationale being that ADCC is complemented by the presence of CD3 + T lymphocytes in addition to macrophage/monocytes, NK, and dendritic cells, known to express FcgR binding to the Fc portion

EpCAM

of the antibody. This antibody consists of a murine IgG2a associated with an anti-light chain and rat IgG2b associated with anti-CD3. In Vitro Cytotoxicity

In vitro experiments of ADCC with cell lines, effector cells, and BiUII demonstrated increased production of interleukins IL-1b, IL-2, IL-6, IL-12, and DC-CK1. Simultaneous stimulation of accessory cells and T lymphocytes leads to antigen presentation to T lymphocytes inducing immunomodulation and cytotoxicity. BiUII induces production of IL-2 in the presence of + cells activating accessory and T-cells without the requirement of exogenous IL-2. An immunologically self-supporting tri-cell complex is formed which is efﬁcient for immune cell activation. Cytolysis occurs within 1–3 days. The mode of cell death is characteristically necrotic and not apoptotic. Lymphocytes with pore-forming perforin proteins surround the tumor cells causing cytolysis. Prolonged Antitumor Immunity In Vivo

Another bispeciﬁc antibody: BiLu induces longlasting antitumor immunity consisting of both humoral and cell-mediated responses when administered intraperitoneally in a murine syngeneic model. It targets murine CD3 and human EpCAM. The Fc portion is identical to. The human CD3 counterpart of BiLu: Catumaxomab/Removab has shown promising results. Clinical Trials (Table 2)

A Phase I/II study for the treatment of ovarian cancer patients with symptomatic ascites has now been completed (23 patients) showing that Removab was safe and effectively reduced ascitic ﬂow and tumor cell content. A substantial Phase II/III trial assessing efﬁcacy in patients with all causes of malignant ascites including primary gastrointestinal tumors commenced in September 2004 (250 patients), and a Phase IIa study of platinum refractory ovarian cancer patients is also underway. Finally, a Phase I/II study of patients with peritoneal carcinomatosis due to GI

EpCAM

1547

Wnt

Frizzled G protein

Kremen

Dkk

LRP5 / 6

PAR1 Frodo Dsh

Dpr β-Arr1

Diversin

GBP

lα

GSK3

APC

Cytoskeleton

l CK

β-Catenin pathway

C PP2A-C Axin

TAK1

PP2A-B NLK β-TrCP P N

β-Catenin

NF I/S C SW TCF BCL19 Pyg Smad4

C Ub Ub Ub Ub

EpCam

Target genes

E Cadherin

ICAT

Conductin

αCD3

Rat IgG2b

αEp-CAM

Mouse IgG2a

Tumor cell

T-cell 1 2 CD40L CD28 CD2 CD40 B7, 1-2 LFA-3

ADCC FcγRI + cell

IL-12 IL-2 IL-1

IL-6 TNF-α DC-CK1

Stimulation

EpCAM, Fig. 2 Structure rationale for the development of bispeciﬁc antibodies

Myc Nkd

Twin

Cyclin Dl En-2

PPARd

Xbra

ID2

Siamois

Xnr3

MMP7

TCF-1

Ubx

1548

EpCAM

EpCAM, Table 2 EpCAM targeting trials underway Therapeutic (alternative name) Catumaxomab (Removab ®)

Class Trispeciﬁc antibody; mouse IgG2a/ rat IgG2b hybrid

Ongoing or completed trials Phase II/III in ovarian cancer Phase II in gastric cancer

Proxinium ® Vivendium ® (VB4-845)

Immunotoxin; single-chain antibody pseudomonas exotoxin fusion

IGN-101 (edrecolomab)

Vaccine for induction of antiidiotypic antibody response

Adecatumumab (MT201)

Fully human IgG1 MAb

EMD 273066 (huKS-IL2)

Fusion of humanized MAb KS1/4 with human IL-2

Phase II/III in head and neck cancer Phase I/II in bladder cancer Phase II in various adenocarcinomas Phase II/III in non-small cell lung cancer Phase II in metastatic breast and early-stage prostate cancer Phase I in metastatic breast, plus Taxotere Phase I in hormone-refractory prostate cancer

Company Trion Pharma/ Fresenius Biotech Viventia

Aphton

Micromet, Inc./ Serono

Lexigen, Inc./ Merck KGaA

EpCAM epithelial cell adhesion activating molecule; IP intraperitoneal; IV intravenous; MTD maximum tolerated dose; SC subcutaneous

tumors but without symptomatic ascites is underway. Preliminary results are promising indicating the utility of Removab in the treatment of minimal ﬂuid borne disease and micrometastases.

Cross-References ▶ Barrett Esophagus ▶ Bispeciﬁc Antibodies ▶ Carcinoembryonic Antigen ▶ CD44 ▶ Circulating Tumor Cells ▶ E-Cadherin ▶ Gastric Cancer ▶ Immunotherapy ▶ Invasion ▶ Lung Cancer ▶ Metastasis ▶ MYC Oncogene ▶ Pancreatic Cancer ▶ Prostate Cancer ▶ Prostate-Speciﬁc Antigen ▶ SiRNA ▶ Wnt Signaling

References Baeuerle PA, Gires O (2007) EpCAM (CD326) ﬁnding its role in cancer. Br J Cancer 96(3):417–423 Chaudry MA, Sales K, Ruf P, Lindhofer H, Winslet MC (2007) EpCAM an immunotherapeutic target for gastrointestinal malignancy: current experience and future challenges. Br J Cancer 96:1013–1019 Trzpis M et al (2007) Epithelial cell adhesion molecule: more than a carcinoma marker and adhesion molecule. Am J Pathol 171:386–395

See Also (2012) Antibody. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 208. doi:10.1007/978-3-642-16483-5_312 (2012) Antisense nucleic acid. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, pp 220–221. doi:10.1007/978-3-64216483-5_336 (2012) Beta-catenin. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 385. doi:10.1007/978-3-642-16483-5_889 (2012) Cadherins. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, pp 581582. doi:10.1007/978-3-642-16483-5_770 (2012) Chromosomal aberrations. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 838. doi:10.1007/978-3-642-164835_1138

Eph Receptors (2012) Cyclooxygenase-2. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1035. doi:10.1007/978-3-642-16483-5_1435 (2012) EGF. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1211. doi:10.1007/978-3-642-16483-5_1824 (2012) Humanized antibodies. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1760. doi:10.1007/978-3-642-16483-5_2863 (2012) ICAMs. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1803. doi:10.1007/978-3-642-16483-5_2938 (2012) Interferon. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1888. doi:10.1007/978-3-642-16483-5_3090 (2012) Interleukin. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1892. doi:10.1007/978-3-642-16483-5_3094 (2012) Matrigel. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 2182. doi:10.1007/978-3-642-16483-5_3551 (2012) MUC1. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, pp 2384–2386. doi:10.1007/978-3-642-16483-5_3870 (2012) P53. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 2747. doi:10.1007/978-3-642-16483-5_4331 (2012) Phagocytosis. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 2840. doi:10.1007/978-3-642-16483-5_4493 (2012) PSA. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, pp 3111–3112. doi:10.1007/978-3-642-16483-5_6738 (2012) Thyroglobulin. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 3687. doi:10.1007/978-3-642-16483-5_5805 (2012) Trifunctional antibody. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 3783. doi:10.1007/978-3-642-16483-5_5976

Eph Receptors Diego Arango CIBBIM - Nanomedicina Oncologia Molecular, Vall d’Hebron Hospital Research Institute, Barcelona, Spain

Definition Eph receptors are tyrosine kinases (RTK) bound to the extracellular membrane that function as

1549

“switches” that upon activation by ephrin (EFN) ligands initiate signaling cascades that regulate numerous developmental processes, particularly in the vasculature and nervous system.

Characteristics The name of the Eph receptors is derived from the name of the cell line used to characterize them initially (erythropoietin-producing hepatocellular carcinoma cell line). These receptors and their ephrin (EFN) ligands constitute the largest family of ▶ receptor tyrosine kinases (RTK) known to date. Eph receptors are integral membrane proteins with a conserved N-terminal domain responsible for ligand binding, followed by a cysteinerich region and two ﬁbronectin type III repeats which are essential for dimerization and interactions with other proteins. The intracellular region of these receptors contains a juxtamembrane domain, a conserved kinase domain, a sterile alpha motif (SAM), and a PDZ-binding motif. There are at least 15 Eph receptors in the human genome. Based on sequence homology and the preferred type of ephrin ligands that they bind, the Eph receptors can be divided into two subclasses, EphA and EphB. The ligands of the Eph receptors (ephrins) are also bound to the extracellular membrane. Ephrins of the A subclass are bound to the membrane through a GPI anchor (glycosylphosphatidylinositol), whereas members of the B subclass have a transmembrane domain. At least ﬁve ephrins of the A subclass and three of the B subclass have been described. The signaling cascade is initiated by the binding of a membrane-bound ephrin and an Eph receptor in neighboring cells. This leads to the autophosphorylation of the Eph receptor on several tyrosine residues and the activation of the tyrosine kinase activity. In addition to the “forward” signaling elicited by the Eph receptors, the ephrin ligands are also able to transduce a signal upon interaction with their cognate Eph receptor. This is often referred to as “reverse” signaling. The signaling cascade initiated upon ligand

1550

binding regulates many developmental processes and has an important role in tumor initiation and progression in different tissues. Normal Eph Signaling The role of Eph signaling in the nervous and vascular systems during normal development has been characterized in detail. During normal embryogenesis, Eph signaling has an important role in the development of the nervous system. Eph signaling regulates the migration of neural crest cells, the formation of the corticospinal tract, the boundary formation between hindbrain segments (rhombomeres), the establishment of neural topographic maps, and the formation and functional properties of neuronal synapses. Eph signaling also plays an important role in the angiogenic process by restricting arterial and venous endothelial mixing. In addition, interactions between Eph receptors and ephrin ligands mediate cytoskeleton organization, cell migration, and substrate attachment. The Eph family also has an important physiologic role in the normal intestinal epithelium. EphB2 and EphB3, together with their ligand ephrin-B1, regulate proliferation and cell positioning within the intestinal crypts. Eph Signaling in Cancer Deregulation of the levels of expression and normal Eph signaling are commonly observed in tumors of various origins. This is not surprising since aberrant Eph signaling can interfere with processes that are crucial during malignant transformation such as cell attachment, ▶ migration, proliferation, cytoskeleton organization, and ▶ angiogenesis. Eph/ephrin activation has been implicated in several ▶ signal transduction pathways contributing to the tumorigenic process. For instance, Akt/PI3K (phosphatidylinositol 3kinase) has been shown to be implicated in the increase in proliferation and migration of endothelial cells after activation of EphB4 with its ligand ephrin-B2. In addition, Eph signaling can regulate cell ▶ motility by modulating the activity of other signal transduction proteins such as ▶ focal adhesion kinase (FAK) and ▶ Rho. The

Eph Receptors

important role of Eph signaling in the modulation of the cellular attachment to the extracellular matrix seems to be regulated through the modulation of integrin-mediated adhesion. Most of the studies in the literature report increased levels of expression of Eph receptors and/or ephrin ligands in most of the tumor types studied, compared with the respective normal tissue. For instance, EphA2 is overexpressed in melanomas; EphA1 shows elevated levels in breast, liver, and lung tumors; and EphB4 has been reported to be overexpressed and signiﬁcantly contribute to tumor progression in prostate, bladder, breast, and head and neck tumors. Although the general view emerging is that Eph receptors and ephrins may function as ▶ oncogenes in the sense that elevated levels or kinase activity promotes tumor formation and/or progression, Eph receptors may be important ▶ tumor suppressor genes in some tissues. This is the case of EphB2 and EphB4 in the intestine. The expression of these two EphB receptors is reduced in colorectal tumors compared to the normal intestinal cells and the premalignant lesions. The levels of expression of EphB2 and EphB4 negatively correlate with tumor progression, and mechanisms of inactivation of these receptors include somatic mutations and hypermethylation of ▶ CpG islands situated within the promoter regions regulating their expression. Moreover, animal studies clearly support the tumor suppressor role of EphB2 in colorectal tumors, and germ line mutations in this gene seem to predispose to prostate and possibly to colorectal cancer.

References Alazzouzi H, Davalos V, Kokko A et al (2005) Mechanisms of inactivation of the receptor tyrosine kinase EPHB2 in colorectal tumors. Cancer Res 65:10170–10173 Committee EN (1997) Uniﬁed nomenclature for Eph family receptors and their ligands, the ephrins, Eph Nomenclature Committee. Cell 90:403–404 Davalos V, Dopeso H, Castano J et al (2006) EPHB4 and survival of colorectal cancer patients. Cancer Res 66:8943–8948

Epidemiology of Cancer Holder N, Klein R (1996) Eph receptors and ephrins: effectors of morphogenesis. Development 126: 2033–2044 Surawska H, Ma PC, Salgia R (2004) The role of ephrins and Eph receptors in cancer. Cytokine Growth Factor Rev 15:419–433

Ephelide ▶ Carney Complex

Epidemiology of Cancer Elizabeth B. Claus Department of Epidemiology and Public Health, Yale University School of Medicine, New Haven, CT, USA

Synonyms Cancer epidemiology

Definition Is the study of the incidence, distribution, and, ultimately, the prevention and control of cancer within the general population.

Characteristics The discipline of cancer epidemiology is a relatively young one, with much of the methodology developed over the past 50 years. Prior to the advent of formal methods of collection and analysis of cancer incidence and risk factor data, associations were generally the result of reports or observations of astute clinicians or scientists. The literature is full of fascinating stories of early attempts at epidemiologic cancer studies such as that of the nineteenth-century physician,

1551

Alfred Haviland, who created elaborate maps of cancer deaths in England and Wales using national mortality statistics. One of the ﬁrst and probably most well-known reports of a relationship between a risk factor and the occurrence of cancer occurred in 1950, with the publication of several case/control studies detailing the association between cigarette smoking and the development of lung cancer. Study Design There are a number of basic study designs in cancer epidemiology. Descriptive cancer epidemiology examines how cancer incidence and mortality rates vary according to demographic characteristics of the study population such as geographic location, race, and sex. One may further analyze such information by categories of age as well as by birth cohort and time period. Ecological studies generally examine aggregate measures of risk and cancer outcome such as median income and cancer incidence across counties of a given state in an effort to identify an association between the two. Alternatively, many epidemiologic studies are based on the use of individuals as the study unit rather than larger groups, or populations of study subjects. Within this category of analysis, there are essentially three study designs (with several variations) including the cross-sectional, case/control, and cohort study design. Cross-sectional study designs allow for the consideration of a reference population at a given point in time. In a case/control study, the frequency of a particular risk factor among individuals with a cancer of interest (cases) is compared with that among individuals without cancer (controls). Case/control studies have been instrumental in the identiﬁcation of numerous important cancer risk factors including the association between family history and breast cancer as well as between tobacco and lung cancer. In a prospective or cohort study, researchers assemble a cohort of healthy individuals who provide information on risk factors of interest at a baseline point in time. The study subjects are then followed prospectively until they develop cancer or the study is

1552

completed. An advantage of this type of study design is that risk factor data is collected before any cancer is diagnosed, thus reducing the amount of recall bias associated with disease status in the reporting of risk factor data such as family history information. There are many well-known examples of cancer cohort studies including the Atomic Bomb Casualty Commission established in 1947 to study the effects of exposure to radiation with outcomes such as leukemia, the British physicians cohort from the 1950s that examined the association between smoking and lung cancer, and the Nurse’s Health Studies I and II, started in 1976 and 1991, respectively, which have examined a wide range or hormonal and dietary risk factors (among others) in the development of breast and other cancers. Risk Factors Categories of cancer risk factors are many and include infectious agents, diet and lifestyle factors, endogenous and exogenous hormonal components, and genetic factors, to name a few. It is currently popular to divide cancer risk factors into two broad categories deﬁned as genetic and environmental. This has come about because of the many laboratory-based advances in the identiﬁcation of genetically transmitted or regulated diseases such as cancer, leading to the emergence of a new ﬁeld of investigation, that is, the genetic epidemiology of cancer. For cancers, a small subset of cases exist that are attributable to rare inherited cancer susceptibility genes. The majority of cases appear to be due to sporadic mutations that may be a result of genetic or environmental events or the result of an interaction between genetic and environmental factors. Much of traditional epidemiologic methodology has been adopted for use in genetic epidemiology. In addition, new methods speciﬁc to genetic epidemiology have been developed including the twin, adoption, and pedigree study designs as well as segregation and linkage analyses. The many advances in genetic testing, as well as in some instances, prevention, or treatment options associated with particular cancer diagnoses, has led to a surge of interest in the availability of personalized risk estimates in the clinical

Epidemiology of Cancer

setting and hence the development of cancer risk assessment models used to generate these estimates. Cancer risks may be presented in both relative and absolute terms and may deﬁne risk for a discrete period of time or over a lifetime. A wide variety of statistical methods exist to estimate cancer risk, with the most fully developed existing in the area of breast and ovarian cancer. Although the concept of risk assessment is not new to the ﬁelds of medicine or genetics, the use of detailed genetic information on a large population-based scale is, with all the associated difﬁculties of presentation and interpretability. The ﬁeld of cancer epidemiology is an exciting scientiﬁc discipline, which is able to adapt well to new information and technology. New developments in the ﬁeld include the integration of biomarkers into exposure data, the inclusion of both molecular genetics and environmental risk factor data into study designs in an effort to explore the complex interaction between genotype and the environment, the creation of international databases via the Internet, and the merging of large databases that combine risk factor information with cancer incidence and mortality data. All of these advances should continue to assist scientists and health-care professionals in the identiﬁcation of individuals at increased risk of developing cancer so that screening and prevention regimes as well as treatment plans may be developed.

Cross-References ▶ Alcoholic Beverages Cancer Epidemiology ▶ Obesity and Cancer Risk

References Claus EB (2000) Risk models in genetic epidemiology. Stat Methods Med Res 9(6):589–601 Samet JM, Munez A (1998) Epidemiologic reviews: cohort studies. Am J Epidemiol 20(1):1–136 Szklo M, Nieto FJ (2000) Epidemiology. Beyond the basics. Aspen, Gaithersburg

Epidermal Growth Factor Inhibitors

Epidermal Growth Factor Inhibitors Deepa S. Subramaniam1, Marion Hartley2, Michael Pishvaian3, Ruth He3 and John L. Marshall3 1 Georgetown University Hospital, Washington, DC, USA 2 Ruesch Center for the Cure of Gastrointestinal Cancers, Lombardi Comprehensive Cancer Center, Georgetown University, Washington, DC, USA 3 Lombardi Comprehensive Cancer Center, Georgetown University, Washington, DC, USA

Definition The EGFR receptor (EGFR) is a member of the HER family of receptor tyrosine kinases that includes EGFR itself (ErbB1/HER1), ErbB2 (HER-2/Neu), ErbB3 (HER3), and ErbB4 (HER4). These proteins are classic membranebound tyrosine kinase receptors whose activation is typically ligand dependent. The principal ligands for EGFR are EGF and TGF-a. Other ligands include amphiregulin, heparin-binding EGF, the poxvirus mitogens, epiregulin, and b-cellulin.

Characteristics EGFR Activation and Downstream Signaling Receptor activation results in homo- or heterodimerization and autophosphorylation of c-terminal tyrosine residues. Receptor activation enables the docking of cytoplasmic proteins that bind to speciﬁc phosphotyrosine residues and initiate several cell signaling pathways. These pathways include the Ras-Raf-MAPK pathway, the PI3K-AKT pathway, the protein kinase C pathway, the

Modiﬁed version of Pishvaian M, Marshall JL, He R, Wang D (2012) Epidermal growth factor inhibitors. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin Heidelberg, pp 1271–1275. doi:10.1007/978-3642-16483-5_1931

1553

STAT pathway, and the src kinase pathway, all of which play important roles in tumor cell proliferation, invasion, migration, and inhibition of apoptosis. Reports have also demonstrated that EGFR can be found in the nucleus where it can act as a transcription factor. EGFR activation does not initiate linear downstream pathway signaling, but rather can activate multiple pathways that cross-connect intracellularly. The pattern of activation is often cell/tissue speciﬁc and likely contributes to the rich variety of biological responses to EGFR activation. Ligand activation of EGFR also results in receptor downregulation, mediated through endocytosis and ultimately Cbl-mediated EGFR ubiquitination and degradation (Marshall 2006). Markedly, EGFR can also be activated in a ligand-independent manner. This aberrant EGFR activation can result from receptor overexpression, gene ampliﬁcation, activating mutations, or loss of regulatory mechanisms. Molecular Mechanisms of Targeted Therapies Anti-EGFR therapies include monoclonal antibodies (mAbs) that recognize EGFR and small molecule inhibitors of EGFR tyrosine kinase activity (TKIs). Cetuximab is the ﬁrst anti-EGFR mAb to be developed and eventually US Food and Drug Administration (FDA) approved. This agent prevents receptor dimerization through steric inhibition of the extracellular domain of EGFR. Cetuximab also promotes receptor internalization and degradation without receptor activation, resulting in receptor downregulation and reduced cell surface expression levels of EGFR. Cetuximab also blocks the transport of EGFR into the cell nucleus, thus inhibiting any direct effects on DNA transcription and/or repair. Finally, cetuximab has the potential to kill its target cells by mediating antibody-dependent cell-mediated cytotoxicity (ADCC) and complement ﬁxation. The TKIs are competitive inhibitors of adenosine triphosphate (ATP). They block the enzymatic activity of the intracellular domain of EGFR. Because of their mechanism of action, these TKIs can block EGFR mutants that lack an extracellular domain and block ligandindependent receptor activation.

1554

EGFR Mutations A number of genetic mutations of EGFR have been found in cancer cells. These mutations generally result in constitutive activation of receptor signaling and, in this way, become independent oncogenic drivers. Many of these mutations predict responsiveness to targeted therapies, especially in non-small cell lung cancer. Many somatic gene mutations – called activating or sensitizing mutations – are associated with an increased response to TKIs and are clustered within exons 18–21 of the receptor kinase domain, near the ATP-binding region. Two of the most prevalent sensitizing mutations include an Exon 19 deletion and a substitution mutation in Exon 21 (L858R). These, and other, sensitizing mutations result in structural changes that confer exquisite sensitivity to TKIs, wherein the presence of EGFR mutations in a patient’s tumor results in increased responsiveness and signiﬁcantly improved progression-free survival compared with a patient whose tumor does not harbor such EGFR mutations. Conversely, certain mutations, such as an Exon 20 insertion, can confer primary resistance to targeting by ﬁrstgeneration TKIs, and other mutations, such as the T790M point mutation in Exon 20, can confer secondary (acquired) resistance after initial TKI therapy. Furthermore, TKI resistance may also result from activation of previously redundant growth signaling pathways, such as ras, raf, and PI3K; loss of the tumor suppressor PTEN pathway; or activation of type I growth factor receptor pathways, such as the IGF-R (insulin growth factor receptor) pathway. EGFRvIII is a tumor-speciﬁc oncogene expressed in a third of all primary or de novo glioblastomas (GBM), but not in normal tissues. The mutation arises from an in-frame deletion, resulting in loss of the majority of the extracellular ligand-binding domain, leading to a constitutively active ligand-independent tyrosine kinase. EGFRvIII-positive tumor cells may induce growth in EGFRvIII-negative cells via paracrine signaling. Expression of EGFRvIII in primary glioblastoma is linked to a poorer long-term patient survival (Schuster et al. 2015).

Epidermal Growth Factor Inhibitors

Only 10–20% of all patients with GBM respond to EGFR inhibitors, despite the fact that most GBMs overexpress EGFR. In a study involving 59 patients, 30 (approximately 50%) expressed EGFRvIII, but only 13 of those 30 patients were seen to respond to TKIs (43% of EGFRvIII-expressing GBM). If patients had tumors that expressed EGFRvIII together with PTEN (14 out of 59 patients), 11 of these 14 (78%) responded to TKIs. PTEN is a tumor suppressor whose expression is frequently lost in GBM and acts as an inhibitor of the PI3K/AKT pathway. Loss of PTEN confers resistance to EGFR inhibitors, presumably because of the persistent activation of the PI3K pathway downstream of EGFR. Nevertheless, only about 25% of the patients (14/59) actually exhibited this coexpression of EGFRvIII and PTEN. In fact, mutations are often cancer-type speciﬁc. For example, while approximately 30–50% of GBMs and head and neck cancers exhibit expression of EGFRvIII, only about 1.5% of non-small cell lung cancers (NSCLC) and 0% of metastatic colorectal cancers express EGFRvIII. By contrast, 10–25% of patients with NSCLC, very few head and neck cancers, and no GBMs contain EGFR tyrosine kinase (EGFR-TK) gene mutations. Rindopepimut consists of an EGFRvIIIpeptide vaccine conjugated to keyhole limpet hemocyanin (KLH) and generates a speciﬁc immune response against EGFRvIII-expressing glioblastoma. A randomized phase II trial of rindopepimut or control (KLH alone) with bevacizumab in recurrent EGFRvIII-expressing glioblastoma GBM demonstrated an overall survival advantage in the vaccine-treated population, with a hazard ratio of 0.47 (p = 0.0208) and 30% of patients in the treatment arm being alive at 18 months, compared with 15% in the control arm (Reardon et al. 2015). Additionally, a large randomized trial of the current standard of care (temozolomide) with or without rindopepimut is ongoing in patients with EGFRvIII-expressing GBM [NCT01480479]. Surprisingly, there is a lack of reported assessments of EGFR-TK mutations in colorectal cancers.

Epidermal Growth Factor Inhibitors

EGFR Inhibitors in GI Tumors EGFR is implicated in the pathogenesis of several cancers including CRC, pancreatic cancers, NSCLC, head and neck cancers, breast cancers, and brain cancers, and overexpression of EGFR has been associated with metastasis, chemotherapy resistance, and poor outcome. The receptor has emerged as a rational target for anticancer treatment in these listed tumors. MAbs to EGFR and tyrosine kinase inhibitors to EGFR kinase have been developed with the aim of inhibiting EGFR signaling. Cetuximab and panitumumab are two mAbs to EGFR that are approved by the FDA in the treatment of CRC. Since the original publication of this book chapter, a great deal has been learned about the incorporation of EGFR receptortargeting mAbs in the treatment of mCRC. Both cetuximab and panitumumab are now approved as ﬁrst-line agents, as well as in the refractory setting, and are guideline approved for use in the second-line setting. The most important breakthrough has been the recognition of Ras mutations conferring resistance to these anti-EGFR antibodies. Originally, the only presumptive biomarker for selection of patients for treatment with EGFR antibodies was the presence of EGFR receptors on the patient’s colon cancer cells. This proved not to be useful in discerning beneﬁt. Important clinical trials using the EGFR antibodies alone, as well as in combination with chemotherapy, have consistently demonstrated that patients whose tumors have one of several RAS mutations fail to respond to this treatment. In fact several studies have demonstrated potential harm when the antibodies are given to patients whose tumors harbor these mutations. This research ﬁeld moved more rapidly than clinical uptake. It is important to ensure that adequate tumor has been tested for RAS mutations prior to considering one of these agents in any line of therapy. As a consequence of this improved enrichment, the beneﬁt seen when using anti-EGFR mAbs in the properly selected RAS wild-type patients has signiﬁcantly improved. This has resulted in improved outcomes in both

1555

progression-free survival and overall survival and response rate in patients appropriately selected. However, much remains to be learned about the appropriate utilization of these compounds in the sequential therapy of mCRC. Increasingly, B-raf is becoming recognized as a potential pathway of resistance, but is less ﬁrmly established as a marker. Cetuximab and panitumumab have been compared head to head in a randomized clinical trial involving patients with chemotherapy-refractory mCRC, and both mAbs demonstrated very similar activity, clinical beneﬁt, and toxicity (Price et al. 2014). Cetuximab has been tested in a randomized phase III clinical trial in the stage III adjuvant setting and failed to demonstrate beneﬁt (Taieb et al. 2014). In addition, cetuximab was tested in the perioperative setting for patients with liver metastasis and was likewise found to be of no additional beneﬁt (Primrose et al. 2014). It is clear that these compounds have a signiﬁcant role in the traditional metastatic setting, but their role in the neoadjuvant or adjuvant setting has not been fully established and is currently negative. A survival advantage of adding cetuximab to radiation therapy was demonstrated in a randomized phase III study comparing concurrent cetuximab/radiation therapy with radiation alone in locoregionally advanced head and neck squamous cell carcinoma. Since then, the role of cetuximab in concurrent chemoradiation has been tested in esophageal/gastric cancer, rectal cancer, and pancreas cancer in phase II clinical trials. Patients with localized esophageal/gastric cancer received radiation and concurrent chemotherapy of cetuximab, carboplatin, and paclitaxel. Approximately 67% of the patients had a complete clinical response, and 43% were found to have a complete pathologic response at surgery. A conﬁrmatory phase III study is being planned. Neoadjuvant chemotherapy and chemoradiation in rectal cancer treatment decreases local relapse rate of rectal cancer with higher sphincter preservation rate compared with postoperative chemoradiation. Cetuximab has been evaluated in combination with a capecitabine/oxaliplatin (CapOx)-based regimen in concurrent

1556

neoadjuvant chemoradiotherapy of patients with rectal cancer in phase II studies. The combination therapy was found to be feasible and resulted in pathologic complete response (CR) in some patients. In the EXPERT-C trial, which compared “neoadjuvant oxaliplatin, capecitabine, and preoperative radiotherapy with or without cetuximab, followed by total mesorectal excision in patients with high-risk rectal cancer,” the presence of cetuximab was found to signiﬁcantly increase patient response rate (RR) and overall survival (OS), but not CR, as long as patients had KRAS/ BRAF wild-type rectal cancer (Dewdney et al. 2012). Management of borderline resectable pancreas cancer is challenging, and neoadjuvant chemotherapy or chemoradiation is controversial in those patients. The role of cetuximab in combination with gemcitabine in concurrent neoadjuvant chemoradiation was examined in a phase II clinical trial. Two of ten patients exhibited partial response. Six patients went on to margin () resection, including one patient each with borderline resectable and unresectable disease prior to therapy. The role of cetuximab with neoadjuvant chemoradiation in downstaging pancreas cancer for resection needs to be conﬁrmed by a phase III randomized study. More than 80% of patients treated with cetuximab experience an acneiform rash. EGFR is expressed in the epidermis, sebaceous glands, and hair follicle epithelium. EGFR plays a role in the normal differentiation and development of skin follicles and keratinocytes. Results following the study of mice with EGFR gene knockout, or dominant negative mutational status, predict that EGFR inhibition in humans is feasible but may be associated with cutaneous toxicity. Indeed skin rash is the most common toxicity following EGFR inhibitor therapy (including anti-EGFR mAbs and small molecule inhibitors of EGFR tyrosine kinase activity (TKIs)). The ability of the presence and intensity of a skin rash to predict the response to EGFR inhibitors was investigated in a subgroup analysis from the BOND study. An increase in RR, median time to progression (mTTP), and median OS was all found to correlate with a higher grade of skin toxicity to cetuximab.

Epidermal Growth Factor Inhibitors

The same phenomena were observed with panitumumab treatment. To take a step further, the question was posed as to whether to dose escalate anti-EGFR antibody therapy until a skin rash is detectable. In a phase II study, patients who had no, or mild, skin reactions in response to a standard cetuximab dose were subjected to dose escalation (up to 500 mg/m2), after which an improved tumor response rate was generally observed. TKIs did not mirror the clinical activities and toxicity proﬁles of anti-EGFR antibodies (cetuximab and panitumumab) in CRC; when administered as single agents, TKIs showed minimal activity in mCRC. When patients with mCRC were treated with a TKI plus ﬂuoropyrimidine-, oxaliplatin-, and irinotecanbased chemotherapy regimens, clinical response rates were observed to range from 24% to 74% in phase II studies. However, TKIs were found to increase grade 3 and 4 toxicities, and some of the trials had to close prematurely due to intolerable adverse effects. To conﬁrm the clinical beneﬁt of TKI therapy in mCRC, a phase III study was carried out that administered bevacizumab-based induction chemotherapy to patients and randomized those who were subsequently free of disease progression or the need for surgery to maintenance therapy with bevacizumab, with or without erlotinib. The target accrual was 640 patients. On ﬁnal analysis, median overall survival from the beginning of maintenance therapy was signiﬁcantly greater following bevacizumab plus erlotinib compared with bevacizumab therapy alone. Median progression-free survival (PFS) was also improved (Tournigand et al. 2015). Erlotinib in combination with gemcitabine demonstrated beneﬁt in patients with pancreatic cancer in a phase III study. To the best of our knowledge, results from this trial were the ﬁrst to demonstrate a clinical beneﬁt following the use of a TKI in combination with chemotherapy. Cetuximab was found to have encouraging activity when combined with gemcitabine and concurrent radiation in localized pancreatic cancer. In addition, cetuximab showed clinical efﬁcacy in combination with gemcitabine/oxaliplatin (GEMOXCET) in the treatment of previously

Epidermal Growth Factor Inhibitors

untreated patients with metastatic pancreatic cancer. Further evaluation in a phase III trial is warranted. Treatment choice for hepatocellular carcinoma (HCC) is limited due to low response rate and transient response following most cytotoxic chemotherapy. Although cetuximab showed very modest activity in HCC, erlotinib demonstrated clinical efﬁcacy. Thirty-eight patients with HCC were treated with erlotinib at a dose of 150 mg daily, and 32% of these patients were found to have PFS at 6 months. Disease control was seen in 59% of the patients. Median overall survival time was 13 months. The role of TKIs in HCC treatment needs to be conﬁrmed in a phase III study. EGFR Inhibitors in Lung Cancer The development of EGFR inhibitors as a therapeutic option in NSCLC has been moving at a rapid pace. Geﬁtinib (Iressa, AstraZeneca), a ﬁrstgeneration reversible small molecule inhibitor of EGFR, initially received accelerated approval from the FDA after phase II clinical studies of this agent showed a promising clinical response in NSCLC patients (Lynch et al. 2004). However, subsequent large phase III trials in the United States failed to conﬁrm these ﬁndings, leading the FDA to restrict geﬁtinib use to only those patients who had already been on this medication with good response. However, in Europe and China, multiple randomized clinical trials have demonstrated the superiority of geﬁtinib over standard platinum-doublet chemotherapy in EGFR-mutant NSCLC (I-PASS trial), allowing the reapproval of geﬁtinib in the United States in 2015. Erlotinib (Tarceva, Genentech) is also a ﬁrst-generation reversible TKI, which, likewise, competes with ATP for the ATP-binding site of the TK domain of EGFR. Erlotinib’s initial approval came after the BR.21 phase III trial demonstrated an overall survival beneﬁt of 2 months compared with placebo in second- and third-line NSCLC (Shepherd Frances et al. 2005). Subsequently, erlotinib has demonstrated a statistically signiﬁcant improvement in progression-free survival compared with platinum-doublet chemotherapy in the frontline treatment of advanced EGFRmutant NSCLC (EURTAC and OPTIMA). Both

1557

geﬁtinib and erlotinib are oral medications and are usually well tolerated. The most common adverse effects of these TKIs are acneiform rash and diarrhea; less commonly reported are transaminitis and stomatitis, with rare cases of interstitial lung disease. Initially, it appeared that certain demographic features, such as female sex, Asian ethnicity, never-smoker status, and adenocarcinoma histology correlated with higher response rates to TKIs. However, it is now known that these demographic factors only serve as surrogates for a higher prevalence of sensitizing EGFR mutations, which are identiﬁed even in the absence of such demographics. As these mutations are exceptionally rare in squamous NSCLC, the current recommendations of the National Comprehensive Cancer Network (NCCN) are to perform EGFR mutation testing routinely in adenocarcinoma tumors, tumors with mixed adenosquamous histology, or very poorly differentiated NSCLC tumors without clearly deﬁned histology. Second-generation TKIs have been developed that are irreversible inhibitors that bind to the TK domain of EGFR and ErbB family members. Such inhibitors include afatinib and dacomitinib. Afatinib (Gilotrif, Boehringer Ingelheim) is FDA approved for patients with tumors harboring Exon 19 deletions or Exon 21 L858R substitution mutations. There are no head-to-head comparisons of these second-generation TKIs with ﬁrstgeneration TKIs in EGFR-mutant NSCLC. LUX-Lung 3 and LUX-Lung 6 are two phase III randomized trials that compare afatinib with cisplatin-doublet chemotherapy (a pemetrexed combination in LUX-Lung 3 and a gemcitabine combination in LUX-Lung 6) in frontline treatment of EGFR-mutant NSCLC. These trials demonstrated a clear PFS advantage following afatinib therapy, leading to this agent’s FDA approval. In early phase clinical trials, dacomitinib also showed promise as an active agent in EGFR-mutant NSCLC. This agent is being explored in Her2mutated NSCLC; one study involving 26 subjects demonstrated a 12% response rate (3 of 26 subjects). However, dacomitinib does not appear to have much activity in Her2-ampliﬁed NSCLC. A risk of diarrhea, skin rash, and stomatitis limit the use of dacomitinib and afatinib.

1558

Third-generation EGFR TKIs include the irreversible and potent, pyrimidine-based EGFR inhibitors such as rocelitinib (Clovis Oncology), AZD9291 (Astra Zeneca), and WZ4002 and HM61763 (Hanmi Pharmaceuticals/co-licensed with Boehringer Ingelheim). These drugs appear to target both activating and resistance mutations, such as T790M, which accounts for over 50–60% of cases of acquired resistance to the ﬁrstgeneration TKIs (e.g., erlotinib and geﬁtinib). The third-generation drugs do not target wildtype EGFR and thus avoid the toxicitiesassociated wild-type receptor targeting, including skin rash and diarrhea. Rocelitinib demonstrates an objective response rate of 60% and a disease control rate of 90% in patients with an acquired T790M resistance mutation. In patients with T790M-negative tumors, a response rate of 37% is observed (Sequist et al. 2015). Key side effects include hyperglycemia (due to inhibition of IGF-1R/IR), diarrhea, nausea, fatigue, QT prolongation, and cataract formation. A phase I dose expansion cohort study of AZD9291 in 60 subjects demonstrated an objective response rate of 83% at a dose of 160 mg and 63% at a dose of 80 mg (Ramalingam et al. 2015). Approximately 79% of patients remained disease progressionfree at one year; the disease control rate was 93% at the 80-mg dose, which is now being explored in the frontline phase III FLAURA trial, comparing AZD9291 with erlotinib in EGFR-mutant NSCLC. It appears that a newly identiﬁed EGFR mutation, namely, C797S in Exon 20 of EGFR, may mediate resistance to AZD9291, at least in select cases. A major drawback of all three generations of EGFR inhibitors is the lack of central nervous system (CNS) penetration: low cerebrospinal ﬂuid (CSF) and plasma concentrations are achieved at standard systemic doses. Therefore, patients with good control of systemic disease may fail therapy if they have brain metastases or leptomeningeal carcinomatosis. A couple of newer EGFR inhibitors with promising CNS penetration in preclinical models are entering clinical trials, including tesavatinib (Kadmon Pharma) and AZD3759 (Astra Zeneca).

Epidermal Growth Factor Inhibitors

In addition to TKIs, cetuximab has been tested in NSCLC with cytotoxic chemotherapy. The role of anti-EGFR mAbs in lung cancer is not very clear as these studies are still ongoing. NSCLC patients respond differently to EGFR inhibitors, especially to TKIs, as suggested by several clinical trials. K-ras is a downstream molecule in the EGFR-signaling pathway. K-ras mutations are consistently shown to be related to TKI resistance (Eberhard et al. 2005). Other processes such as Her2 gene ampliﬁcation and Akt phosphorylation have been reported to correlate with clinical response to TKIs (Stasi and Cappuzzo 2014). As with other targeted therapy in oncology, the molecular target of EGFR inhibition needs to be further deﬁned in NSCLC. If this is accomplished, we can individualize anti-EGFR therapy to achieve the best possible outcomes.

References Dewdney A, Cunningham D, Tabernero J, Capdevila J, Glimelius B, Cervantes A, Tait D, Brown G, Wotherspoon A, Gonzalez de Castro D, Chua YJ, Wong R, Barbachano Y, Oates J, Chau I (2012) Multicenter randomized phase II clinical trial comparing neoadjuvant oxaliplatin, capecitabine, and preoperative radiotherapy with or without cetuximab followed by total mesorectal excision in patients with high-risk rectal cancer (EXPERT-C). J Clin Oncol 30(14): 1620–1627. doi:10.1200/JCO.2011.39.6036. Epub 2 Apr 2012 Eberhard DA, Johnson BE, Amler LC et al (2005) Mutations in the epidermal growth factor receptor and in KRAS are predictive and prognostic indicators in patients with non-small-cell lung cancer treated with chemotherapy alone and in combination with erlotinib. J Clin Oncol 23:5900 Lynch TJ, Bell DW, Sordella R et al (2004) Activating mutations in the epithelial growth factor receptor underlying responsiveness of non-small-cell lung cancer to geﬁtinib. N Engl J Med 350:2129 Marshall J (2006) Clinical implications of the mechanism of epidermal growth factor receptor inhibitors. Cancer 107(6):1207–1218 Price TJ et al (2014) Panitumumab versus cetuximab in patients with chemotherapy-refractory wild-type KRAS exon 2 metastatic colorectal cancer (ASPECCT): a randomised, multicentre, open-label, non-inferiority phase 3 study. Lancet Oncol 15(6):569–579 Primrose J et al (2014) Systemic chemotherapy with or without cetuximab in patients with resectable colorectal

Epidermal Growth Factor Receptor liver metastasis: the New EPOC randomised controlled trial. Lancet Oncol 15(6):601–611 Ramalingam SS et al (2015) AZD9291, a mutant-selective EGFR inhibitor, as ﬁrst-line treatment for EGFR mutation-positive advanced non-small cell lung cancer (NSCLC): results from a phase 1 expansion cohort. J Clin Oncol 33 (suppl): abstr 8000 Reardon DA, Schuster J, Tran DD, Fink KL, Nabors LB, Li G, Bota DA, Lukas RV, Desjardins A, Ashby LS, Duic JP, Mrugala MM, Werner A, Hawthorne T, He Y, Green JA, Yellin MJ, Turner CD, Davis TA, Sampson JH, The ReACT Study Group (2015) ReACT: Overall survival from a randomized phase II study of rindopepimut (CDX-110) plus bevacizumab in relapsed glioblastoma. J Clin Oncol 33(15):2009 Schuster J, Lai RK, Recht LD, Reardon DA, Paleologos NA, Groves MD, Mrugala MM, Jensen R, Baehring JM, Sloan A, Archer GE, Bigner DD, Cruickshank S, Green JA, Keler T, Davis TA, Heimberger AB, Sampson JH (2015) A phase II, multicenter trial of rindopepimut (CDX-110) in newly diagnosed glioblastoma: the ACT III study. Neuro Oncol 17(6):854–861 Sequist LV et al (2015) Rociletinib in EGFR-mutated non–small-cell lung cancer. N Engl J Med 372:1700–1709. doi:10.1056/NEJMoa1413654 Shepherd Frances A et al (2005) Erlotinib in previously treated non–small-cell lung cancer. N Engl J Med 353:123–132 Stasi I, Cappuzzo F (2014) Second generation tyrosine kinase inhibitors for the treatment of metastatic non-small-cell lung cancer. Transl Resp Med 2:2. doi:10.1186/2213-0802-2-2 Taieb J et al (2014) Oxaliplatin, ﬂuorouracil, and leucovorin with or without cetuximab in patients with resected stage III colon cancer (PETACC-8): an openlabel, randomised phase 3 trial. Lancet Oncol 15(8):862–873 Tournigand C et al (2015) Bevacizumab with or without erlotinib as maintenance therapy in patients with metastatic colorectal cancer (GERCOR DREAM; OPTIMOX3): a randomised, open-label, phase 3 trial. Lancet Oncol. doi:10.1016/S1470-2045(15)00216-8. Published online 13 Oct 2015

Epidermal Growth Factor Receptor Christina L. Addison Cancer Therapeutics Program, Ottawa Hospital Research Institute, Ottawa, ON, Canada

Synonyms EGFR; ERBB; ErbB-1; HER1; PLG61

1559

Definition The epidermal growth factor receptor (EGFR) is the prototype cell surface receptor for the epidermal growth factor (EGF) family of soluble protein ligands. The receptor itself is part of a family of homologous transmembrane tyrosine kinase receptors that includes EGFR (ErbB-1), HER2/cneu (ErbB-2), Her 3 (ErbB-3), and Her 4 (ErbB4). EGFR is a 1186aa, 134 kD protein and was the ﬁrst receptor described to have intrinsic tyrosine kinase activity directed toward not only to itself (autophosphorylation) but also to various downstream target proteins. Following binding of its ligands, EGFR can elicit intracellular signals that modulate a wide variety of cellular functions including cell proliferation, survival, migration, and differentiation. The EGFR gene maps to chromosome 7p11-13, and it is often ampliﬁed or rearranged in glioblastoma multiforme, non-small cell lung cancer, and prostate cancer.

Characteristics EGFR Structure EGFR is transcribed from a single 26 exon gene on chromosome 7p11-13, and upon translation, it generates a 1186 amino acid mature transmembrane glycoprotein (Fig. 1). The amino terminal 622 amino acids form the extracellular domain which contains two cysteine-rich domains (CR1 and CR2), which comprise the ligand binding domain of the protein. Following this, the transmembrane domain separates the extracellular domain from the intracellular 542 amino acid carboxy-terminus of the protein. The intracellular domain contains three separate motifs, the juxtamembrane domain, the kinase domain, and a carboxy-terminal tail. The juxtamembrane domain primarily plays a role in feedback attenuation of receptor signaling. The kinase domain becomes activated following ligand binding and receptor dimerization, and it then phosphorylates numerous substrates on tyrosine residues. The kinase domain of the EGFR family of receptors is highly conserved with the exception of ErbB3

1560 Epidermal Growth Factor Receptor, Fig. 1 EGFR structure. EGFR is a transmembrane receptor tyrosine kinase whose domain structure is illustrated herein. In the extracellular region, the CR1 and CR2 domains form the ligand binding domain. Intracellularly, the tyrosine kinase domain and a number of tyrosine residues which can become phosphorylated following receptor activation play important roles in EGFR signal transduction. EGFR ligands are grouped into three classes depending on their abilities to bind various EGFR family members in addition to binding EGFR

Epidermal Growth Factor Receptor

EGFR Ligands Group 1 Group 2 EGF BTC TGF-α HB-EGF AR EPR EPG

Group 3 NRG-1 NRG-2

CR1

Ligand binding

CR2

Kinase Domain

Y845

Stat5 and full receptor activation

Receptor trafficking PKC activation Ubiquitin-mediated receptor degradation MAPK activation AKT activation MAPK activation EGFR inactivation via dephosphorylation

which has amino acid substitutions at critical sites which results in it lacking tyrosine kinase activity. Within the carboxy-terminal tail of EGFR, there are a number of tyrosine residues that are autophosphorylated, which following their phosphorylation can serve as binding sites to recruit additional signaling proteins leading to activation of a number of downstream signal transduction pathways (Fig. 1). The carboxy tail also has additional tyrosine residues that are known to be phosphorylated by other signaling molecules, for example, Src kinase, which also plays a role in modulation of receptor signaling. EGFR Ligands EGFR family ligands can be primarily divided into three groups based on their afﬁnity for binding the various EGFR family of receptors. As

AP2 PLCγ

Y974 Y992

Cbl Grb2 PI3K Grb2 SHP1

Y1045 Y1068 Y1086 Y1101 Y1148 Y1173

Src-induced phosphorylation Autophosphorylation

EGFR receptors can homodimerize or heterodimerize, binding of ligands and subsequent signaling that occurs following this binding will be inﬂuenced by the expression of the combination of different EGFR receptors in cells. Group 1 EGFR ligands represent those that bind speciﬁcally to the prototypical family member EGFR. These include EGF, tumor growth factora (TGF-a), epigen (EPG), and amphiregulin (AR). The second group of ligands includes those that bind EGFR but additionally bind ErbB4. Group 2 ligands include betacellulin (BTC), heparin-binding epidermal growth factor (HB-EGF), and epiregulin (EPR). Group 3 comprises the neuregulin (NRG) ligands, which is further divided into two subgroups. Subgroup 1 is comprised of NRG-1 and NRG-2 which bind to both EGFR and ErbB4, while subgroup

Epidermal Growth Factor Receptor

2 contains only those neuregulins that bind ErbB4, namely, NRG-3 and NRG-4. It should also be noted that no known ligands bind to ErbB2. EGFR Signal Transduction Upon ligand binding, EGFR homodimerizes or heterodimerizes and undergoes subsequent internalization and activation of the kinase domain of the receptor. This results in phosphorylation of the receptor at a number of tyrosine residues in its cytoplasmic tail, most notably at Y974, Y992, Y1045, Y1068, Y1086, Y1148, and Y1173. These phosphorylation events create docking sites for a number of other proteins that contain Src homology 2 (SH2) domains or phosphotyrosine binding (PTB) domains (Fig. 1). One of the most important downstream signaling events following EGFR activation is the subsequent activation of the Ras/Raf/MEK/ MAPK pathway via association of phosphorylated Y1068, Y1086, or Y1148 residues with the adaptor protein growth factor receptor-bound protein-2 (Grb2). The SH3 domains of Grb2 are constitutively associated with SOS (son of sevenless), an exchange factor of Ras GTPase. Besides interaction with SOS, Grb2 SH3 domains are capable of association with several additional proteins, including dynamin and Cbl, which both play a role in the regulation of EGFR endocytosis. Binding of the Grb2 and SOS complex to the EGFR places SOS in proximity to Ras, thus leading to GTP-loading of Ras and subsequent activation of Ras effectors, including Raf kinases and PI3K (phosphatidylinositol 3-kinase). Raf initiates a cascade of phosphorylation events including the phosphorylation and activation of the MEKs (MAPK/ERK kinases) and ERKs (extracellular signal-regulated kinases). Other important interactions include phosphorylated Y992 of EGFR with phospholipase Cg (PLCg), which upon binding becomes activated and in turn activates protein kinase C (PKC), a serine-threonine kinase. This enzyme, which has two SH2 domains, catalyzes the hydrolysis of PIP2, generating the second messengers DAG (1,2-diacylglycerol) and IP3 (inositol trisphosphate). IP3 diffuses through the cytosol

1561

and releases stored Ca2+ (Calcium) ions from the ER (endoplasmic reticulum). DAG is the physiological activator of PKC (protein kinase C), which in turn leads to phosphorylation of various substrate proteins that are involved in an array of cellular events including enhanced cellular migration. Phosphorylated EGFR has also been shown to recruit and activate the p85 subunit of PI3K. Recruitment of PI3K may be mediated by the docking protein GAB1 (Grb2-associated binder1). Once PI3K is recruited and becomes activated, it phosphorylates membrane bound PIP2 (phosphatidylinositol (4,5)-bisphosphate) to generate PIP3 (phosphatidylinositol-3,4,5trisphosphate). The binding of the released PIP3 to the PH domain of Akt anchors Akt to the plasma membrane and allows its phosphorylation and activation by PDK1 (phosphoinositidedependent kinase-1). Akt then phosphorylates several substrates thereby regulating cell survival. A number of other autophosphorylation sites play a role in regulating the duration of activation of EGFR. For example, the autophosphorylation at Y1045 allows association with the Cbl protein, an E3 ubiquitin ligase which subsequently modiﬁes EGFR and targets it for ubiquitin-mediated proteasomal degradation. Another mechanism by which EGFR signaling may be attenuated is via interaction of phosphorylated Y1173 with the protein tyrosine phosphatase SHP1, which dephosphorylates the autophosphorylated tyrosine residues on EGFR and hence halts its continual interaction with other signaling molecules. EGFR also activates several other proteins including FAK (focal adhesion kinase), paxillin, caveolin, E-cadherin, and CTNN-beta (cateninbeta). FAK activation links EGFR with regulation of cell motility, and caveolin, cadherin, and CTNN-beta are involved in cytoskeletal regulation. EGFR Transactivation In addition to activation of EGFR via ligand binding and subsequent autophosphorylation, EGFR may be activated by other signaling pathways in a process known as transactivation. For example, Janus tyrosine kinase 2 (JAK2) can phosphorylate

1562

speciﬁc residues on the cytoplasmic tail of EGFR following its own activation by growth hormone or prolactin interactions with their receptors. The intracellular kinase Src has also been shown to directly phosphorylate EGFR on speciﬁc residues such as Y845 and Y1101 following Src activation by a number of pathways including other growth factor receptors and integrins. A number of G protein-coupled receptors have also been shown to transactivate EGFR following binding their own agonists, such as endothelin-1, bombesin, thrombin, or lysophosphatidic acid. EGFR in Cancer Given that EGFR signaling results in modulation of cellular proliferation and migration, it is not surprising that EGFR has been shown to play a signiﬁcant role in cancer. There are a number of mechanisms that result in the dysregulation of EGFR signaling in cancer including: (1) increased production of ligands, (2) increased expression of EGFR protein, (3) acquisition of EGFR mutations which give rise to constitutively active EGFR variants, (4) impairment of pathways which control downregulation of EGFR signals, and (5) increased cross talk with other growth factor receptor systems. Increased EGFR Ligand Production

In many tumor types, increased expression of EGF and TGF-a has been noted, and these ligands act in a paracrine and autocrine manner to further drive tumor cell proliferation in an EGFR-dependent manner. In fact, stimulation of ﬁbroblasts overexpressing EGFR with EGF or TGF-a resulted in their enhanced proliferation and transformation. Transgenic mice engineered to overexpress TGF-a under the mammary-speciﬁc MMTV promoter developed mammary hyperplasia and adenocarcinoma of mammary tissues. It has also been shown that co-expression of TGF-a with EGFR in many human tumors correlates with higher proliferative indices and an overall worse patient survival. The EGFR ligand amphiregulin has also been shown to be upregulated in tumor tissues in patients, and its overexpression in tumor cells modulates proliferation, anchorage independent growth, and response to chemotherapy drugs.

Epidermal Growth Factor Receptor

Increased EGFR Protein Levels

Increased EGFR expression in ﬁbroblasts led to increased cellular transformation and migration independent of ligand binding, likely as a result of increased spontaneous receptor dimerization and subsequent activation. Subsequently, in vivo transgenic mouse models overexpressing EGFR under control of mammary-speciﬁc promoters resulted in the development of mammary hyperplasia that led to dysplasia and tubular adenocarcinomas in lactating animals. It should be noted, however, that forced overexpression of EGFR in other tissues including urothelium, glial cells, or esophageal keratinocytes resulted in increased cellular proliferation but not carcinoma. The family member ErbB-2, however, has been shown to have more potent transforming ability compared to EGFR, as its forced overexpression in a tissuespeciﬁc manner resulted in carcinoma of the breast and the skin. In human cancer, both EGFR and ErbB-2 have been implicated in the development and progression in a variety of different tumor types. The overexpression of this family of receptors has been shown in the majority of solid tumors and is reviewed elsewhere (Normanno et al. 2006; Arteaga 2002; Mendelsohn and Baselga 2000; Spaulding and Spaulding 2002). On average, 50–70% of lung, colon, and breast carcinomas have been shown to overexpress EGFR. In most cases, receptor overexpression is a result of gene ampliﬁcation; however, this varies according to tumor type. For example, EGFR ampliﬁcation or overexpression occurs in ~40–60% of glioblastoma multiforme tumors and ~20–35% of non-small cell lung carcinoma (NSCLC); however, in other tumor types, this incidence of ampliﬁcation may not be as high. EGFR Mutations

Mutations in EGFR in tumor cells are quite frequent. EGFR mutations can be divided into three main groups, namely, (1) extracellular mutations, (2) intracellular mutations, and (3) mutations within the kinase domain. Extracellular mutations predominantly involve deletions of speciﬁc exons encoding all or part of the extracellular domain of EGFR; however, some characterized mutations

Epidermal Growth Factor Receptor

have also included duplications of some regions. Most of the deletion mutations give rise to truncated versions of the receptor which are constitutively active and usually escape normal receptor downregulatory mechanisms. Similar to the extracellular mutations, the characterized intracellular mutations also predominantly arise from deletions or duplications. These usually result in truncation of the c-terminus of the receptor. Finally, a number of mutations that alter the tyrosine kinase domain have been identiﬁed. Although exons 18–24 are known to encode the tyrosine kinase domain, the identiﬁed kinase domain mutations are all restricted to exons 18–21. The most frequent mutations here include: (1) small deletions of 6–7 codons in exon 19 that affect amino acids 746–753 of the EGFR protein, (2) point mutations such as the missense mutation L858R in exon 21 or G719A/C in exon 18, and (3) small duplications or insertions in exon 20. Generally, these mutations are centered around the ATP binding site of the kinase domain and are associated with increased tyrosine kinase activity of the mutant receptors. Impairment of EGFR Downregulation

In normal cells, EGFR downregulation and attenuation of its signaling occur as a result of receptor internalization and subsequent degradation of ligand-activated receptors. It is known that this downregulation is dependent on the cytoplasmic tail of EGFR, and mutant EGFR lacking portions of the cytoplasmic tail is able to transform ﬁbroblasts in a ligand-dependent manner. The E3 ubiquitin ligase c-Cbl plays a signiﬁcant role in this process, and both EGFR variants that lack ability to bind c-Cbl or mutant c-Cbl variants that cannot ubiquitinate EGFR both result in increased cellular proliferation. Co-expression of ErbB-2 also results in decreased downregulation of EGFR, as it promotes formation of EGFR/ ErbB-2 heterodimers which are not readily downregulated by the same mechanism that controls degradation of EGFR homodimers. Differential ligand stimulation can also modulate EGFR degradation, as it has been shown that EGF-induced EGFR activation results in efﬁcient EGFR degradation, while TGF-a-induced EGFR

1563

activation predominantly results in recycling of EGFR back to the cell surface where it can then be restimulated. EGFR Cross Talk with Other Receptors

In addition to the modulation of EGFR activity following interaction with other EGFR family receptors, EGFR signaling is also affected by the activity of other receptor tyrosine kinases, cell adhesion molecules, cytokine receptors, ion transport channel proteins, and G protein-coupled receptors (GPCR). The most well-characterized EGFR interaction with other receptors is that with its family member ErbB2. This heterodimer is the most potent inducer of cellular transformation and mitogenic signaling compared to other EGFR family heterodimers. Activation of another important growth factor receptor in cancer, c-met, has also been shown to modulate the activity of EGFR and production of EGFR ligands in tumor cells. EGFR activity has also been shown to be modulated by cell adhesion molecules such as the integrins following their binding to extracellular matrix (ECM) proteins. The exact mechanism of this is unclear, however, likely involves their association with EGFR in cell surface molecule clusters along with other adaptor and signaling proteins. Common downstream signaling targets between EGFR and integrins may also facilitate this interaction, such as PI3K or Src. EGFR transactivation has also been shown to be modulated by GPCR agonists in an EGFR ligandindependent manner. This EGFR transactivation results in enhanced cell proliferation and migration. GPCRs may also contribute to EGFR liganddependent EGFR activation, as GPCR stimulation by agonists may lead to enhanced membrane bound matrix metalloproteinase (MMP) activity which then cleaves membrane bound EGFR ligand precursors that subsequently bind and activate EGFR. Therapies Targeting EGFR Given the established role of EGFR in a number of human cancers, it has become a main focus of targeted cancer therapy in the past 10–15 years. Anti-EGFR agents include mAbs (monoclonal antibodies) targeting the EGFR extracellular

1564

receptor domain and small molecule TKIs (tyrosine kinase inhibitors) targeting the EGFR intracellular kinase domain. Both mAbs and TKIs have demonstrated encouraging results as monotherapies and in combination with chemotherapy and radiotherapy. More detailed discussion on anti-EGFR therapies and their results can be found elsewhere (Ganti and Potti 2005; Baselga 2001; Khalil et al. 2003; El-Rayes and LoRusso 2004; Johnston et al. 2006). A number of mAbs to EGFR have been developed and tested in clinical trials, but the most extensively tested one to date is cetuximab (aka Erbitux). Cetuximab has a higher afﬁnity for binding EGFR than EGFR ligands, hence, acts as a competitive inhibitor to block EGFR activation, in addition to facilitating EGFR internalization and degradation following mAb binding. Preclinical studies with cetuximab suggested that it could inhibit tumor cell proliferation and induce apoptosis following a G1 cell cycle arrest. Cetuximab further inhibited tumor growth in xenograft models where it additionally showed an ability to inhibit tumor vascularization. These preclinical studies supported its use in clinical trials, where early Phase I studies demonstrated excellent patient tolerability with relatively minor side effects. Following Phase III clinical analyses, cetuximab demonstrated antitumor efﬁcacy as a single agent or in combination with other anticancer regimens, which has led to its approval for use clinically in certain advanced cancers. Other mAbs to EGFR are presently being tested and are in various stages of clinical studies. Similar to mAbs, a number of small molecule TKIs have been developed to target EGFR. The primary mechanism of action is via the ability of the small molecule to act as an ATP mimetic for EGFR, thereby resulting in selective inhibition of its autophosphorylation and subsequent signaling activity. The two EGFR small molecule inhibitors most extensively studied are geﬁtinib (aka Iressa or ZD1839) and erlotinib (aka Tarceva or OSI774). Both have been shown to effectively inhibit tumor cell proliferation in vitro and tumor growth in vivo in preclinical studies. Following testing in Phase I studies where clinical activity

Epidermal Growth Factor Receptor

was most favorable, the majority of advanced clinical testing was performed in patients with NSCLC. Both geﬁtinib and erlotinib were shown to increase the survival of advanced NSCLC patients. However, it appears that a small subset of patients are sensitive to these agents while the vast majority do not derive signiﬁcant antitumor beneﬁt from these drugs. The majority of sensitive patients appear to have either EGFR ampliﬁcation or activating point mutations that confer sensitivity. However, acquisition of other point mutations in EGFR have also been associated with gained resistance to these agents. To circumvent these issues, a new class of “second-generation” EGFR inhibitors has been generated and is currently under clinical evaluation. These “irreversible” EGFR receptor inhibitors have the added beneﬁt of binding to the ATP binding site of the kinase domain in a covalent and irreversible manner following reaction of a nucleophilic cysteine residue. Additionally, as the ﬁrst-generation inhibitors such as geﬁtinib and erlotinib were selective for inhibition of EGFR, most of the second-generation inhibitors target more than one EGFR family member. CI-1033 (aka canertinib or PD183805) is one such irreversible inhibitor which actually targets all three of the kinase domain-containing members of the EGFR family. CI-1033 is more effective in inhibiting tumor cell proliferation in vitro than the ﬁrstgeneration EGFR inhibitors, and it has demonstrated antitumor efﬁcacy in EGFR- and ErbB2dependent tumor xenograft models in preclinical studies. While still in clinical testing, early studies have shown that the drug is reasonably tolerated by patients; however, a maximum tolerated dose was established as a result of some grade 3 toxicities. A number of Phase II and III studies are currently underway with this agent in a variety of tumor types; however, we will have awaited the outcome of the mature clinical data to determine whether acquired tumor resistance to these inhibitors also develops. Summary EGFR is a critically important mediator of tumor growth and remains an important therapeutic

Epidermal Growth Factor-like Ligands

target. However, regulation of its activity and signal transduction by other cellular receptors or adhesion molecules require further analysis in order to determine what role this receptor cross talk may play in mediating the response of tumor to anti-EGFR therapeutic strategies. It will also be important to understand how different EGFR ligands affect not only EGFR signaling and receptor trafﬁcking but also how they may modulate tumor response to EGFR targeting agents. Finally, understanding the tumor characteristics that are important for response to EGFR-targeting agents will facilitate identiﬁcation of cancer patients most likely to beneﬁt from these agents in future treatment strategies.

1565

Epidermal Growth Factor Receptor (EGFR) Inhibitors ▶ Receptor Tyrosine Kinase Inhibitors

Epidermal Growth Factor-like Ligands Aleksandra Glogowska and Thomas Klonisch Department of Human Anatomy and Cell Science, College of Medicine, Faculty of Health Sciences, University of Manitoba, Winnipeg, MB, Canada

Cross-References Synonyms ▶ HER-2/neu ▶ Herceptin

References Arteaga CL (2002) Overview of epidermal growth factor receptor biology and its role as a therapeutic target in human neoplasia. Semin Oncol 29:3–9 Baselga J (2001) The EGFR as a target for anticancer therapy–focus on cetuximab. Eur J Cancer 37(Suppl 4):S16–S22 El-Rayes BF, LoRusso PM (2004) Targeting the epidermal growth factor receptor. Br J Cancer 91:418–424 Ganti AK, Potti A (2005) Epidermal growth factor inhibition in solid tumours. Expert Opin Biol Ther 5:1165–1174 Johnston JB, Navaratnam S, Pitz MW, Maniate JM, Wiechec E, Baust H, Gingerich J, Skliris GP, Murphy LC, Los M (2006) Targeting the EGFR pathway for cancer therapy. Curr Med Chem 13:3483–3492 Khalil MY, Grandis JR, Shin DM (2003) Targeting epidermal growth factor receptor: novel therapeutics in the management of cancer. Expert Rev Anticancer Ther 3:367–380 Mendelsohn J, Baselga J (2000) The EGF receptor family as targets for cancer therapy. Oncogene 19:6550–6565 Normanno N, De Luca A, Bianco C, Strizzi L, Mancino M, Maiello MR, Carotenuto A, De Feo G, Caponigro F, Salomon DS (2006) Epidermal growth factor receptor (EGFR) signaling in cancer. Gene 366:2–16 Spaulding DC, Spaulding BO (2002) Epidermal growth factor receptor expression and measurement in solid tumors. Semin Oncol 29:45–54

EGF family; EGF-like ligands; Growth Factors

Definition Epidermal growth factor (EGF-)-like family members bind to and activate EGF receptor tyrosine kinases ErbB-1, 2, 3, and 4, also named HER1-4. This triggers the activation of intracellular signal transduction pathways, resulting in cellular proliferation and differentiation. All members of the EGF-like family are produced as membrane-anchored precursors and are processed and released through the action of speciﬁc membrane-bound proteolytic enzymes of the sheddase family. Epidermal Growth Factor- (EGF-) like family consists of at least twelve members: Epidermal growth factor (EGF), transforming growth factor alpha (TGFa), heparin binding-EGF like growth factor (HB-EGF), amphiregulin(AR), betacellulin (BTC), epiregulin (EPR), epigen, cripto, and neuregulins 1–4 (NRG1-4). These EGF-like members are deﬁned by three characteristics: (a) they display high afﬁnity binding to membrane-bound epidermal growth factor tyrosine kinases receptors ErbB1-4, (b) they are eliciting a

1566

mitogenic response in EGF-sensitive cells, and (c) they possess at least one primary EGF structural motif of 50–60 amino acid (aa) residues with the general sequence XnCX7CX2-3GXCX10-13CXCX3YXGXRCX4LXn embedded in the transmembrane precursor molecule. Usually present as multiple structural units in the extracellular domain of EGF-like ligands, this conserved cysteine-rich EGF-like motif is crucial for the binding to ErbB receptors. The EGF-like ligands are proteolytically cleaved and liberated from the extracellular region of the transmembrane precursor. Ligands, either membrane-bound or soluble, can bind to homoand heterodimer combinations of the four currently known membrane-anchored tyrosin kinase receptors ErbB1 (EGFR, HER-1), ErbB2 (Neu antigen, HER-2), ErbB 3 (HER-3), and ErbB4 (HER-4). Of all possible combinations, ErbB2 homodimers appear not to be stable. EGF, AR, and TGFa bind preferentially to EGFR, while BTC, HB-EGF, and EPR bind to both EGFR and ErbB4. NRG1 and NRG2 preferentially bind to ERbB3 and ERbB4, whereas NRG3 and NRG4 bind to ErbB4. Growth factor binding induces homo- or heterodimerization of the receptors and stimulates intracellular proteintyrosine kinase activity resulting in auto- and crossphosphorylation of key tyrosine residues in the C-terminal domain in the ErbB homo- or heterodimer. This autophosphorylation sites are docking stations for several proteins that associate with the phosphorylated tyrosine residues through SH2 domains (Src homology 2). This initiates downstream ▶ signal transduction cascades, among them being the ▶ MAPK, AKT, and ▶ JNK pathway, which leads to DNA synthesis, cell proliferation, alterations in apoptotic pathways, and cell migration and adhesion (Jorissen et al. 2003; Yarden 2001). Epidermal Growth Factor, EGF, was originally isolated from the mouse submaxillary gland as a stimulatory of eyelid opening and tooth eruption in newborn rodents. Other sources of (pro) EGF are the kidney*, epidermis* (*mainly membrane-bound proEGF), pancreas, small intestine, and brain. Human EGF precursor (preproEGF) is a large protein consisting of 1207 aa containing nine EGF motifs in the extracellular proEGF region. The EGF motif closest to

Epidermal Growth Factor-like Ligands

the transmembrane domain corresponds to soluble EGF (Cohen 1962). The enzyme responsible for cleavage of the extracellular proEGF domain is ADAM 10 (A Disintegrin And Metalloprotease domain 10) (Dong and Wiley 2000). EGF null mice do not show a signiﬁcant phenotype. Transgenic overexpression of mouse EGF targeted to villous enterocytes of the jejunum and ileum causes increased villous height and crypt depth in these intestinal parts. Transgenic mice overexpressing human proEGF-driven by the b -actin promoter have revealed that (pro)EGF is the active ligand for the EGFR in germ cells and proper EGF expression is important for the completion of spermatogenesis (Harris et al. 2003; Wong et al. 2001). Transforming Growth Factor alpha, TGFa, derives from a 160 aa precursor (preproTGFa). This protein was originally characterized by its capacity to induce oncogenic transformation in rat kidney ﬁbroblasts. TGFa is targeted preferentially to the basolateral compartment of polarized epithelial cells where it is cleaved by TNF alpha converting enzyme (TACE)/ADAM17. TGFa mRNA and protein have been detected in various adult tissues, including pituitary gland, brain, skin keratinocytes, and macrophages (Burdick et al. 2000). Transgenic animals with overexpression of TGFa display hypertrophy of skin and hyperkerotosis with alopecia. These animals also have stunted hair growth and psoriasislike lesions. TGF a null mice show a very mild phenotype which includes wavy fur and whiskers (Harris et al. 2003; Mann et al. 1993). Heparin binding-EGF like growth factor, HB-EGF, derives from a 204 aa molecule. Human HB-EGF was identiﬁed in conditioned media of human monocytes/macrophages and was puriﬁed from phorbol ester-treated human U-937 macrophages as a 22-kDa factor with strong afﬁnity for heparin (Higashiyama et al. 1991, 1992). HB-EGF is expressed in a wide variety of hematopoietic cells, endothelial cells, vascular smooth muscle, and epithelial cells. Cleavage of proHB-EGF may involve ADAM9, 10, 12, and 17 (Suzuki et al. 1997). Overexpression of human HB-EGF decreases the weight and growth rate in these transgenic

Epidermal Growth Factor-like Ligands

mice, and this may involve alternation of insulinlike growth factor binding protein (IGFBP) pathways. Moreover, histological analysis showed overexpression of HB-EGF exclusively in kidney, liver, lung, and stomach. Employing several mutant mice lacking HB-EGF expression revealed a critical role for HB-EGF in cardiac valve formation and normal heart function. HB-EGF null mice develop heart failure as a result of grossly enlarged ventricular chambers (Asakura et al. 2002; Harris et al. 2003). Amphiregulin, AR, is synthesized as a 252 aa transmembrane precursor. AR was originally isolated from the phorbol ester-treated human breast adenocarcinoma cell line MCF7. ProAR is expressed in many human tissue including ovary, placenta, pancreas, testes, lung, cardiac muscle, breast, kidney, spleen, and colon. In polarized epithelial cells, cleavage of basolateral proAG is facilitated by TACE/ADAM17 to release a 78–84 aa growth factor. Posttranslational modiﬁcation results in several different cell surface and soluble isoforms (Brown et al. 1998). Keratin-14 promoter-driven overexpression of AR in the basal epidermal cell layer in transgenic mice induces severe, early-onset skin pathology, resembling psoriasis. Histological examination of the skin in these AR overexpressing mice revealed hyperkeratosis, focal parakeratosis and acanthosis, mixed leukocyte inﬁltration, including CD3-positive T cells and neutrophils in the epidermis and dermis. These mice also display tortuous dermal vasculature. AR null mice similar to EGF null mice do not display any gross or histological abnormalities (Inui et al. 1997). Betacellulin (BTC) has been isolated from conditioned media from a pancreatic b cell tumor cell line. BTC is also expressed in a variety of mesenchymal and epithelial cell lines and in many tissue including pancreas, liver, kidney, and small intestine. Membrane-bound proBTC is processed by ADAM10 to release soluble BTC (Dunbar and Goddard 2000; Sasada et al. 1993). Transgenic chicken actin promoter-driven BTC overexpression in mice causes bony deformations of the skull, pulmonary hemorrhage syndrome, and complex eye pathology. Transgenic animals showed decrease in the weight of pancreas and

1567

increase weight of the eye, lung, and spleen (Schneider et al. 2008, 2009). Epiregulin, EPR, was initially puriﬁed from conditioned medium of the mouse ﬁbroblastderived tumor cell line NIH-3 T3 clone T7. EPR is expressed in adult pancreas, liver, kidney, and small intestine. Human epiregulin encodes a 46 aa growth factor which exhibits 24–50% identity with the sequences of other EGF-like ligands. Epiregulin exhibits a bifunctional regulatory property in that it inhibits the growth of several epithelial cell lines, like NIH-3 T3, and stimulates the growth of ﬁbroblasts, primary rat hepatocytes, human smooth muscle cells, and various other cell types, including COS-7 (monkey) and Swiss 3 T3 (mouse) (Shirakata et al. 2000). Epiregulin can bind to and activate EGFR and ErbB4. EPR can also transactivate ErBB2 and ErbB3. Although EPR displays lower afﬁnity towards EGFR, EPR-mediated receptor activation is more sustained when compared with EGF or BTC. The lower afﬁnity of EPR to EGFR results in enhanced dissociation of EPR from the ligand-receptor complex during receptor endocytosis and renewed EGFR activation by recycled EPR. EPR-deﬁcient mice develop chronic dermatitis (Komurasaki et al. 1997). Neu differentiation factors (NDF, heregulins, Neuregulins, Glia Growth Factors-GGFs, Acetylocholine Receptor Inducing Activity-ARIA) were isolated from mesenchymal cells based on their ability to elevate phosphorylation of ErbB proteins. At least four different genes of neuregulins are known. NRGs are widely expressed in neurons of the central and peripheral nervous system. They are known to play important functions during neuronal development and migration (Britsch 2007). Furthermore, NRG-1 is also involved in heart development. Mice defective in genes encoding either NRG-1 or the receptors ErbB-2 or ErbB-4 display identical failure of trabecular formation in the embryonic heart, consistent with the notion that trabeculation requires NRG-1-mediated activation of ErbB-2/ErbB-4 heterodimers (Kuramochi et al. 2004). Epigen encodes a protein of 152 aa that contains EGF-like features. Epigen has 24–37%

1568

identity with EGF, TGF /, and Epiregulin. In epithelial cells, Epigen stimulates the phosphorylation of ErbB1 and causes activation of MAP kinase signaling pathways. Epigen also activates genes under the control of the SRE (Serum Response Element). Epigen is a mitogen for HaCaT cells and this activity can be signiﬁcantly reduced by a blocking antibody to the receptor ErbB-1 (Strachan et al. 2001). Cripto-1 (Cr-1) is a glycoprotein member of 188 aa. Cr-l has been implicated in development. Expression of Cr-l was reported in trophoblast and embryoblast of 4-day-old mouse blastocysts, and the myocardium of developing heart tubes in 8.5day-old mice embryos. Transgenic murine embryos lacking Cr-l are devoid of cardiac speciﬁc gene expression (a- and b- myosin heavy chain, myosin light chains 2A and 2B) (Ozcelik et al. 2006). Antisense RNA silencing studies showed that inhibition of Cr-1 expression inhibits the growth of human colon carcinoma cell lines expressing this factor (Normanno et al. 2004). EGF-like Ligands and Cancer Members of the EGF-like ligand receptor system are often ampliﬁed in various cancers. EGFR overexpression in brain tumors (oligodendroglioma, glioblastoma) and in carcinoma of the stomach, thyroid, lung, and breast (for the latter two also Neu/ErbB2 ampliﬁcation) makes these tumors targets for the actions of EGF-like ligands and anti-cancer drugs directed at inhibiting ligand-mediated ErbB1/2 activation. EGF-like ligands affect tumor cell growth, differentiation, and metastasis. Ampliﬁcation and ligand-induced activation of EGFR correlates with increased tumor cell migration, matrix degradation in vitro, and enhanced tissue invasiveness in vivo. EGFR activation as a result of the prior activation of G protein coupled receptors (GPCRs) involves the proteolytic cleavage of membrane-anchored EGF-like ligands by GPCR-mediated activation of members of the metalloproteinase super family. The GPCRmediated EGFR activation in COS-7, HEK-293, and breast cancer cells involves cleavage of

Epidermal Growth Factor-like Ligands

proHB-EGF, while in colon epithelial cells and head and neck squamous cell carcinomas proTGFa has been implicated (Olayioye et al. 2000; Yarden 2001). Brain Tumor Malignant human gliomas are the most common form of primary tumors of the central nervous system and display an invasive growth pattern. Among the genetic alternations found in these tumors, p53 inactivation and PDGF/PDGFR activation represent early events, whereas the loss of chromosome 10, gene ampliﬁcation, and rearrangement of genomic EGFR are late events in glioma carcinogenesis. Coampliﬁcation of TGFa and EGFR in human glioma cell lines and primary glioma tissues coincides with sustained glioma proliferation and suggests an autocrine growth loop promoted by this EGF-like ligandreceptor pair, especially in high grade glioma. Antisense-mediated downregulation of TGFa expression was shown to inhibit glioma growth (Harris et al. 2003; Nair 2005; Perry et al. 2002; Tang et al. 1997). Carcinoma of the Lung EGF-like growth factors play an important role in the pathogenesis and progression of nonsmall cell lung cancers (NSCLC) comprising large cell cancer, squamous cancer and adenocarcinoma, and small cell lung cancer (SCLC). Both the presence of cytoplasmic EGFR, but not its membraneanchored form, and/or higher EGFR expression, either in its cytoplasmic or membrane-bound form, together with the ampliﬁcation of TGFa are all signiﬁcantly associated with poor patient survival rates (Nair 2005; Olayioye et al. 2000; Rusch et al. 1997; Wu et al. 2007). Head and Neck Squamous Cell Carcinoma (HNSCC) GPCR-EGFR transactivation via GPCRs for gastrin releasing peptide and lysophosphatidic acid (LPA) or carbachol results in a matrixmetalloproteinase-dependent enhanced invasiveness and growth of HNSCC cells in vitro. This

Epidermal Growth Factor-like Ligands

coincides with the release of TGFa and AR, but not EGF or HB-EGF, into the supernatant of HNSCC. Enzymatic processing of proamphiregulin and proTGFa upon treatment with lysophosphatidic acid (LPA) or carbachol involves the activation of TNF-converting enzyme (TACE/ADAM-17) and downstream activation of EGFR-induced MAPK signaling pathways in HNSSC (Grandis and Tweardy 1993; Gschwind et al. 2003). Breast Cancer TGFa stimulates growth and differentiation of mammary epithelial cells and is implicated in the pathogenesis of human breast cancer (Zheng 2009). High expression of AR has been detected in several human breast cancer cell lines and primary human breast carcinomas. AR acts as an autocrine/juxtacrine growth factor in human mammary epithelial cells transformed with an activated c-Ha-ras proto-oncogene or overexpressing Neu/ErbB2 (Normanno et al. 1994). EGF and AR may modulate invasion of metastatic breast cancer cells by increasing the expression of matrix-metalloproteinases. EGFR activation by GPCRs in breast cancer cells appears to be mainly facilitated by HB-EGF. Furthermore, estradiol treatment of breast cancer cells causes EGFR activation by the MMP-2- and MMP-9-mediated release of HB-EGF (Olayioye et al. 2000). Colorectal Cancer Most human colon cancer cell lines express TGFa, AR, and cripto (Wu et al. 2009). Cripto and AR appear to be suitable markers for human colorectal cancer tissues since transcripts for both EGF-like ligands are detected in 60–70% of primary or metastatic human colorectal cancers but are present in only 2–7% of normal human colonic mucosa. Also, immunoreactive AR was reported in primary and metastatic colorectal tumors but not in normal colon (Normanno et al. 2004; Ohchi et al. 2012). Agonists to the GPCR prostaglandin E2 receptor and the M3 muscarinic receptor lead to a metalloproteinase-/

1569

TACE-dependent processing of TGFa resulting in EGF-R transactivation and proliferation of colon cancer (Caco-2, LoVo, and HT-29) cell lines (Fiske et al. 2009).

Cross-References ▶ Adhesion ▶ Amphiregulin ▶ Cripto-1 ▶ JNK Subfamily ▶ Pleiotrophin ▶ SH2/SH3 Domains ▶ Signal Transduction ▶ Transforming Growth Factor-Beta

References Asakura M, Kitakaze M, Takashima S, Liao Y, Ishikura F, Yoshinaka T, Ohmoto H, Node K, Yoshino K, Ishiguro H et al (2002) Cardiac hypertrophy is inhibited by antagonism of ADAM12 processing of HB-EGF: metalloproteinase inhibitors as a new therapy. Nat Med 8:35–40 Britsch S (2007) The neuregulin-I/ErbB signaling system in development and disease. Adv Anat Embryol Cell Biol 190:1–65 Brown CL, Meise KS, Plowman GD, Coffey RJ, Dempsey PJ (1998) Cell surface ectodomain cleavage of human amphiregulin precursor is sensitive to a metalloprotease inhibitor. Release of a predominant N-glycosylated 43-kDa soluble form. J Biol Chem 273:17258–17268 Burdick JS, Chung E, Tanner G, Sun M, Paciga JE, Cheng JQ, Washington K, Goldenring JR, Coffey RJ (2000) Treatment of Menetrier’s disease with a monoclonal antibody against the epidermal growth factor receptor. N Engl J Med 343:1697–1701 Cohen S (1962) Isolation of a mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the new-born animal. J Biol Chem 237:1555–1562 Dong J, Wiley HS (2000) Trafﬁcking and proteolytic release of epidermal growth factor receptor ligands are modulated by their membrane-anchoring domains. J Biol Chem 275:557–564 Dunbar AJ, Goddard C (2000) Structure-function and biological role of betacellulin. Int J Biochem Cell Biol 32:805–815 Fiske WH, Threadgill D, Coffey RJ (2009) ERBBs in the gastrointestinal tract: recent progress and new perspectives. Exp Cell Res 315:583–601

1570 Grandis JR, Tweardy DJ (1993) TGF-alpha and EGFR in head and neck cancer. J Cell Biochem Suppl 17F:188–191 Gschwind A, Hart S, Fischer OM, Ullrich A (2003) TACE cleavage of proamphiregulin regulates GPCR-induced proliferation and motility of cancer cells. EMBO J 22:2411–2421 Harris RC, Chung E, Coffey RJ (2003) EGF receptor ligands. Exp Cell Res 284:2–13 Higashiyama S, Abraham JA, Miller J, Fiddes JC, Klagsbrun M (1991) A heparin-binding growth factor secreted by macrophage-like cells that is related to EGF. Science 251:936–939 Higashiyama S, Lau K, Besner GE, Abraham JA, Klagsbrun M (1992) Structure of heparin-binding EGF-like growth factor. Multiple forms, primary structure, and glycosylation of the mature protein. J Biol Chem 267:6205–6212 Inui S, Higashiyama S, Hashimoto K, Higashiyama M, Yoshikawa K, Taniguchi N (1997) Possible role of coexpression of CD9 with membrane-anchored heparin-binding EGF-like growth factor and amphiregulin in cultured human keratinocyte growth. J Cell Physiol 171:291–298 Jorissen RN, Walker F, Pouliot N, Garrett TP, Ward CW, Burgess AW (2003) Epidermal growth factor receptor: mechanisms of activation and signalling. Exp Cell Res 284:31–53 Komurasaki T, Toyoda H, Uchida D, Morimoto S (1997) Epiregulin binds to epidermal growth factor receptor and ErbB-4 and induces tyrosine phosphorylation of epidermal growth factor receptor, ErbB-2, ErbB-3 and ErbB-4. Oncogene 15:2841–2848 Kuramochi Y, Cote GM, Guo X, Lebrasseur NK, Cui L, Liao R, Sawyer DB (2004) Cardiac endothelial cells regulate reactive oxygen species-induced cardiomyocyte apoptosis through neuregulin-1beta/ erbB4 signaling. J Biol Chem 279:51141–51147 Mann GB, Fowler KJ, Gabriel A, Nice EC, Williams RL, Dunn AR (1993) Mice with a null mutation of the TGF alpha gene have abnormal skin architecture, wavy hair, and curly whiskers and often develop corneal inﬂammation. Cell 73:249–261 Nair P (2005) Epidermal growth factor receptor family and its role in cancer progression. Curr Sci 88:890–898 Normanno N, Ciardiello F, Brandt R, Salomon DS (1994) Epidermal growth factor-related peptides in the pathogenesis of human breast cancer. Breast Cancer Res Treat 29:11–27 Normanno N, De Luca A, Maiello MR, Bianco C, Mancino M, Strizzi L, Arra C, Ciardiello F, Agrawal S, Salomon DS (2004) CRIPTO-1: a novel target for therapeutic intervention in human carcinoma. Int J Oncol 25:1013–1020 Ohchi T, Akagi Y, Kinugasa T, Kakuma T, Kawahara A, Sasatomi T, Gotanda Y, Yamaguchi K, Tanaka N, Ishibashi Y et al (2012) Amphiregulin is a prognostic factor in colorectal cancer. Anticancer Res 32:2315–2321

Epidermal Growth Factor-like Ligands Olayioye MA, Neve RM, Lane HA, Hynes NE (2000) The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J 19:3159–3167 Ozcelik C, Bit-Avragim N, Panek A, Gaio U, Geier C, Lange PE, Dietz R, Posch MG, Perrot A, Stiller B (2006) Mutations in the EGF-CFC gene cryptic are an infrequent cause of congenital heart disease. Pediatr Cardiol 27:695–698 Perry SW, Dewhurst S, Bellizzi MJ, Gelbard HA (2002) Tumor necrosis factor-alpha in normal and diseased brain: conﬂicting effects via intraneuronal receptor crosstalk? J Neurovirol 8:611–624 Rusch V, Klimstra D, Venkatraman E, Pisters PW, Langenfeld J, Dmitrovsky E (1997) Overexpression of the epidermal growth factor receptor and its ligand transforming growth factor alpha is frequent in resectable non-small cell lung cancer but does not predict tumor progression. Clin Cancer Res 3:515–522 Sasada R, Ono Y, Taniyama Y, Shing Y, Folkman J, Igarashi K (1993) Cloning and expression of cDNA encoding human betacellulin, a new member of the EGF family. Biochem Biophys Res Commun 190:1173–1179 Schneider MR, Antsiferova M, Feldmeyer L, Dahlhoff M, Bugnon P, Hasse S, Paus R, Wolf E, Werner S (2008) Betacellulin regulates hair follicle development and hair cycle induction and enhances angiogenesis in wounded skin. J Invest Dermatol 128:1256–1265 Schneider MR, Mayer-Roenne B, Dahlhoff M, Proell V, Weber K, Wolf E, Erben RG (2009) High cortical bone mass phenotype in betacellulin transgenic mice is EGFR dependent. J Bone Miner Res 24:455–467 Shirakata Y, Komurasaki T, Toyoda H, Hanakawa Y, Yamasaki K, Tokumaru S, Sayama K, Hashimoto K (2000) Epiregulin, a novel member of the epidermal growth factor family, is an autocrine growth factor in normal human keratinocytes. J Biol Chem 275:5748–5753 Strachan L, Murison JG, Prestidge RL, Sleeman MA, Watson JD, Kumble KD (2001) Cloning and biological activity of epigen, a novel member of the epidermal growth factor superfamily. J Biol Chem 276:18265–18271 Suzuki M, Raab G, Moses MA, Fernandez CA, Klagsbrun M (1997) Matrix metalloproteinase-3 releases active heparin-binding EGF-like growth factor by cleavage at a speciﬁc juxtamembrane site. J Biol Chem 272:31730–31737 Tang P, Steck PA, Yung WK (1997) The autocrine loop of TGF-alpha/EGFR and brain tumors. J Neurooncol 35:303–314 Wong WC, Dong M, Mak KL, Chan SY (2001) Prospects of EGF transgenic mice researches. Sheng Wu Hua Xue Yu Sheng Wu Wu Li Xue Bao (Shanghai) 33:473–476 Wu W, O’Reilly MS, Langley RR, Tsan RZ, Baker CH, Bekele N, Tang XM, Onn A, Fidler IJ, Herbst RS (2007) Expression of epidermal growth factor (EGF)/ transforming growth factor-alpha by human lung cancer cells determines their response to EGF receptor

Epigallocatechin tyrosine kinase inhibition in the lungs of mice. Mol Cancer Ther 6:2652–2663 Wu WK, Tse TT, Sung JJ, Li ZJ, Yu L, Cho CH (2009) Expression of ErbB receptors and their cognate ligands in gastric and colon cancer cell lines. Anticancer Res 29:229–234 Yarden Y (2001) The EGFR family and its ligands in human cancer. signalling mechanisms and therapeutic opportunities. Eur J Cancer 37(Suppl 4):S3–S8 Zheng W (2009) Genetic polymorphisms in the transforming growth factor-beta signaling pathways and breast cancer risk and survival. Methods Mol Biol 472:265–277

1571

Epigallocatechin Ann M. Bode and Zigang Dong The Hormel Institute, University of Minnesota, Austin, MN, USA

Synonyms Catechin; Green tea polyphenol

See Also (2012) AKT. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 115. doi:10.1007/978-3-642-16483-5_163 (2012) Apoptosis Pathways. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 244. doi:10.1007/978-3-642-16483-5_365 (2012) Betacellulin. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 385. doi:10.1007/978-3-642-16483-5_591 (2012) Cell Migration. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 738. doi:10.1007/978-3-642-16483-5_1006 (2012) DNA. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1129. doi:10.1007/978-3-642-16483-5_1663 (2012) Domain. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1150. doi:10.1007/978-3-642-16483-5_1702 (2012) Epigen. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1283. doi:10.1007/978-3-642-16483-5_1939 (2012) Epiregulin. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1291. doi:10.1007/978-3-642-16483-5_1954 (2012) G-protein Couple Receptor. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1587. doi:10.1007/978-3-642-16483-5_2294 (2012) MAPK. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 2167. doi:10.1007/978-3-642-16483-5_3532 (2012) Phosphorylation. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 2870. doi:10.1007/978-3-642-16483-5_4544 (2012) Tyrosine. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 3822. doi:10.1007/978-3-642-16483-5_6078

Epidermoid Carcinoma ▶ Squamous Cell Carcinoma

Definition Epigallocatechin is one of the several biologically active ingredients that make up the bulk of the potent antioxidant polyphenols known as catechins, which are found in green tea ▶ Green Tea Cancer Prevention.

Characteristics Next to water, tea is the second most widely consumed beverage in the world. Green tea, like oolong and black teas, is derived from the Camellia sinensis plant, but is processed immediately from fresh leaves and is protected from oxidation. The biologically active ingredients in green tea are a family of polyphenols (catechins) and ﬂavonols, which are very strong antioxidants. The catechins comprise about 90% of the bulk of green tea and include epicatechin (EC), epigallocatechin (EGC), epicatechin gallate (ECG), and epigallocatechin gallate (EGCG) (Fig. 1). The catechins are characterized by the presence of a di- or trihydroxyl group substitution on the “B”-ring and the meta-5.7-dihydroxyl substitution at the “A”-ring (Fig. 1). EGCG appears to account for 50–80% of the total catechins found in green tea and is considered to be the most biologically active. A cup of green tea (2.5 g of dried green tea leaves brewed in 200 mL of water) usually contains about 90 mg of EGCG. In addition, green tea contains a similar or slightly smaller amount (65 mg) of EGC, about 20 mg each of ECG and

1572 Epigallocatechin, Fig. 1 Chemical structures of the major green tea catechins

Epigallocatechin

OH HO A

O C

OH OC

B HO A D

O C

OH (−)-Epigallocatechin-3-gallate (EGCG)

(−)-Epicatechin (EC)

OH OH HO A

O C

OH OH

OC OH

HO A

OH OH

O OH

(−)-Epicatechin-3-gallate (ECG)

EC, and about 50 mg of caffeine. However, physiologically achievable tissue levels of EGCG appear to be in the low micromolar range (i.e., 1–7 mM). Epidemiologic and laboratory studies suggest that consumption of green tea might be associated with a decreased risk of developing skin, lung, bladder, esophageal, stomach, liver, duodenum and small intestine, pancreatic, colorectal, prostate, or breast ▶ cancer. Many chronic diseases, including cancer, are associated with ▶ oxidative DNA damage produced by free radicals. Much of the effectiveness of green tea has been attributed to its potent antioxidant activity, which is suggested to be greater than that of vitamin C or E or equivalent servings of most vegetables and fruits. Research data also suggest that some of the effects of green tea might be due to its ability to generate ▶ reactive oxygen species as a prooxidant molecule. However, direct unequivocal evidence for either mechanism of action in vivo is lacking. On the other hand, an accumulating number of research studies suggest that EGCG may speciﬁcally target and modulate distinct cancer genes or proteins, with little or no

O C

OH (−)-Epigallocatechin (EGC)

effect on normal molecules, in order to exert its anticancer effects. Cellular and Molecular Targets of EGCG Cells respond to their environment through a process known as signal transduction in which information from a stimulus outside a cell is transmitted from the cell membrane into the cell and along an intracellular chain of signaling molecules to perpetuate a response. The development of cancer is a complex, several-step process (i.e., initiation, promotion, progression) that affects innumerable genes and proteins, including signaling molecules that are critical in the regulation of many cellular functions but especially proliferation or growth. The initiation step is an irreversible period involving a process in which a normal cell becomes changed so that it has the capacity to form a tumor (i.e., preneoplastic). The change results from DNA damage that can be induced by a number of “initiators,” including radiation, ultraviolet irradiation, chemical carcinogens, and retroviruses. Under normal conditions, the DNA damage is either repaired or, if the damage is too extreme, the cell is eliminated by a tightly

Epigallocatechin

1573

controlled process of cell death called “▶ apoptosis.” If the damaged DNA is allowed to duplicate, the cell may be predisposed to cancer. The promotion step is a potentially reversible process by which actively proliferating preneoplastic cells accumulate and begin to develop characteristics, including growth factor independence, lack of contact inhibition, and resistance to apoptosis, all of which enhance the cells’ ability to proliferate by escaping normal control mechanisms. This step generally occurs over a period of many years, and environmental tumor promoters or host factors may play roles in cancer promotion and latency. The ﬁnal step of cell transformation is progression in which preneoplastic cells acquire increased metastatic potential and the ability to spread to other tissue sites of the afﬂicted organism. Research ﬁndings have shown that the dysfunction or deregulation of various cellular signaling molecules is a major factor in cancer development and prevention. The prevailing idea today is that cancer may be prevented or treated by targeting and modulating the activity of speciﬁc cancer genes or signaling proteins. Each step of cancer development could be a potential target for anticancer agents, but especially the promotion step because of its length and reversible nature. In addition, an increased interest in discovering and developing natural, nontoxic compounds as chemopreventive agents now exists. The molecular mechanisms explaining how normal cells undergo transformation induced by tumor promoters are rapidly being clariﬁed, and the mechanisms by which natural compounds such as the green tea polyphenol EGCG can act as chemopreventive agents are also being elucidated.

promoters such as 12-O-tetradecanoylphorbol13-acetate (TPA), epidermal growth factor (EGF), and platelet-derived growth factor (PDGF) and are also strongly stimulated by stresses such as ultraviolet (UV) irradiation and arsenic. The activation of these signaling cascades can result in a multitude of cellular responses including apoptosis, proliferation, inﬂammation, differentiation, and development. MAP kinases activate a variety of target proteins that are important in tumor development, including activator protein-1 (▶ AP-1) and nuclear factor-kappa B (NF-kB), which in turn may promote transcription of a variety of cancer-related genes such as cyclooxygenase-2 (cox-2). A substantial body of evidence suggests that EGCG or other green tea polyphenols inhibit the phosphorylation and activation of the MAP kinases and various components of another critical cancer-associated pathway, the phosphatidylinositol-3 kinase (▶ PI3K)/▶ Akt pathway. Green tea polyphenols have also been reported to suppress tumor promoter- or growth factor-induced cell transformation and AP-1, NF-kB, or COX-2 activation. EGCG has also been shown to inhibit the phosphorylation of the upstream epidermal growth factor family of proteins. Furthermore, consumption of green tea polyphenols has been associated with suppression of numerous markers of angiogenesis and ▶ metastasis, including expression of ▶ matrix metalloproteinases 2 and 9 and ▶ vascular endothelial growth factor. Other studies indicate that EGCG also inhibits ▶ telomerase activity to induce cell senescence and suppress DNA methyltransferase resulting in the reactivation of the ▶ tumor suppressor gene p16 INK4a.

General Anticancer Effects of Green Tea Polyphenols Green tea polyphenols have been reported to suppress cancer cell proliferation, enhance apoptosis, decrease ▶ angiogenesis, and suppress oncoprotein activation. In particular, the mitogen-activated protein or ▶ MAP kinase signaling pathways are activated differentially by various tumor promoters. The MAP kinases generally transmit signals initiated by tumor

Direct Molecular Targets of EGCG Identifying EGCG receptors or high-afﬁnity proteins that bind to EGCG is a key step in understanding the molecular and biochemical mechanism of this polyphenol’s anticancer effects. The structure of EGCG (Fig. 1) facilitates its ability to bind with varying afﬁnity to a number of proteins. Proteins that have thus far been reported to directly bind with EGCG include ▶ ﬁbronectin, ▶ laminin and the 67-kDa laminin

1574

receptor, ▶ insulin growth factor-1 receptor (IGF-1R), Pin1, ZAP-70 kinase, Fyn, ▶ BCL-2 and Bcl-xL, vimentin, the glucose-regulated protein 78 (GRP78) chaperone protein, apoptosisassociated Fas, and fatty acid synthase. However, not all of these proteins (i.e., Fas, 400 mM; fatty acid synthase, 52 mM) show a high binding afﬁnity, and the consequences or details of the binding with EGCG are not fully understood. EGCG was ﬁrst reported to inhibit cancer cell ▶ adhesion to ﬁbronectin by directly binding with this protein. However, the inhibition of cancer cell adhesion was not caused by EGCG binding to the cell-binding domain because EGCG interacted with the adjoining domain. EGCG was also reported to bind to another extracellular matrix protein, laminin, and to inhibit ▶ melanoma cell adhesion. EGCG was later found to inhibit cell growth by binding with the invasion- and metastasis-associated 67-kDa laminin receptor (67LR), which is expressed on various cancer cell types. Importantly, the Kd value for binding was 39.9 nM suggesting possible in vivo relevance. The intermediate ﬁlament protein, vimentin, which has an important functional involvement in cell division and proliferation, was also identiﬁed as an EGCG-binding protein. Vimentin displayed an even higher afﬁnity (Ki = 3.3 nM) for binding with EGCG, and the association also appeared to have a regulatory role in controlling cell proliferation. Another work has suggested that EGCG (20 mg/mL) blocks anchorage-independent growth and human breast and cervical cancer cell phenotype expression through inhibition of IGF-1R downstream signaling. Thus, EGCG appears to interact with surface proteins to suppress cancer cell proliferation. EGCG may also interact with key proteins to modulate apoptosis. The pro-survival Bcl-2 proteins are overexpressed in many cancer types and thus contribute to resistance of cancer cells to apoptosis. EGCG was reported to directly bind to the BH3 pocket of Bcl-xL (Ki = 490 nM) or Bcl-2 (Ki = 335 nM), resulting in the suppression of the antiapoptotic activity of these proteins. EGCG was also shown to directly interact with GRP78, which is associated with the multidrug

Epigallocatechin

resistance phenotype of many types of cancer cells. EGCG suppressed GRP78’s function and caused an increased etoposide-induced apoptosis in cancer cells. These results strongly suggest that EGCG can enhance apoptosis of cancer cells. Data suggest that EGCG has multiple targets and identiﬁcation of novel EGCG-binding proteins could facilitate the design of new strategies to prevent cancer and hopefully help translate the effectiveness of EGCG observed in cell and animal models to humans. Clinical Relevance Tea polyphenols have attracted a great deal of interest because of their perceived ability to act as highly effective chemopreventive agents. EGCG has been reported to cause growth inhibition, G1-phase arrest, and apoptosis in a variety of human cancer cells. Notably, EGCG’s effects appear to target only cancer cells with little or no effect on normal cells. This apparent speciﬁcity suggests that EGCG can be used in combination with traditional chemotherapeutic agents to enhance cancer cell death without harming normal cells. For example, treatment of lung cancer cells with EGCG plus ▶ celecoxib, a COX-2 inhibitor, has been shown to synergistically induce apoptosis. EGCG has also been shown to increase the toxicity of the chemotherapeutic drug, ▶ cisplatin, by severalfold in ovarian cancer cells and showed IC50 values in the mM range even for ovarian cancers that are known to be resistant to cisplatin. Furthermore, the combination of EGCG with radiotherapy has been suggested to improve the efﬁcacy of ionizing radiation in treating glioblastoma cells. Although animal and cell culture data suggest a potent anticancer effect for EGCG, reports of anticancer activity of tea polyphenols in humans are less dramatic. Phase I and II clinical trials have been performed to test the anticancer effects of oral administration of green tea but results are still inconclusive. Laboratory data clearly indicate that EGCG and other green tea polyphenols are very unlikely to have only a single target or receptor to account for all its observed activities and effects. Furthermore, based on limited bioavailability, experimental concentrations of EGCG greater

Epigenetic

than 20 mM may not be relevant to the in vivo situation. Understanding the molecular mechanisms of tea in antitumor promotion may reveal additional high-afﬁnity molecular targets for the development of more effective agents with fewer side effects for the chemoprevention of cancer. A continuing emphasis on obtaining rigorous research data and critical analysis of those data regarding tea polyphenols and other food factors is vital to determine the molecular basis and longterm effectiveness and safety of these compounds as chemopreventive agents. Large-scale and comprehensive studies using combined approaches of biochemical, molecular, animal, and clinical studies are needed to address the bioavailability, toxicity, molecular target, signal transduction pathways, and side effects of tea polyphenols for translation to humans.

1575 Bode AM, Dong Z (2006) Molecular and cellular targets. Mol Carcinog 45:422–430 Na HK, Surh YJ (2006) Intracellular signaling network as a prime chemopreventive target of ()-epigallocatechin gallate. Mol Nutr Food Res 50:152–159 Yang CS, Lambert JD, Hou Z et al (2006a) Molecular targets for the cancer preventive activity of tea polyphenols. Mol Carcinog 45:431–435 Yang CS, Sang S, Lambert JD et al (2006b) Possible mechanisms of the cancer-preventive activities of green tea. Mol Nutr Food Res 50:170–175

E See Also (2012) AKT. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 115. doi:10.1007/978-3-642-16483-5_163 (2012) Laminin. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 1971– 1972. doi:10.1007/978-3-642-16483-5_3268 (2012) Senescence. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3370. doi:10.1007/978-3-642-16483-5_5236

Cross-References ▶ Adhesion ▶ Angiogenesis ▶ Apoptosis ▶ AP-1 ▶ Bcl2 ▶ Cancer ▶ Celecoxib ▶ Cisplatin ▶ DNA Oxidation Damage ▶ Fibronectin ▶ Green Tea Cancer Prevention ▶ Insulin-Like Growth Factors ▶ MAP Kinase ▶ Matrix Metalloproteinases ▶ Metastasis ▶ PI3K Signaling ▶ Reactive Oxygen Species ▶ Telomerase ▶ Tumor Suppressor Genes ▶ Vascular Endothelial Growth Factor

Epigenetic Definition Heritable changes in genome function that occur without a change in DNA sequence. DNA ▶ methylation at ▶ CpG-islands and posttranslational histone modiﬁcations are the molecular basis for epigenetic information. Epigenetic changes of DNA inﬂuence the phenotype without altering the genotype. They consist of changes in the properties of a cell that can be transmitted through cell cycle divisions but do not represent a change in genetic information. The main epigenetic changes that target gene regulatory regions and modulate gene expression are methylation and acetylation.

Cross-References References Bode AM, Dong Z (2003) Signal transduction pathways: targets for green and black tea polyphenols. J Biochem Mol Biol 36:66–77

▶ CpG Islands ▶ Epigenetic Gene Silencing ▶ Epigenetic Therapy ▶ Methylation

1576

Epigenetic Biomarker Tina Bianco-Miotto Robinson Research Institute and School of Agriculture, Food and Wine, The University of Adelaide, Adelaide, SA, Australia

Definition Epigenetic biomarker refers to the measurement of epigenetic modiﬁcations in tissues or peripheral ﬂuids like urine, blood, plasma, serum, and stool samples, as markers of disease detection, progression, and therapy response. In particular, most research into epigenetic biomarkers has focused on assessing epigenetic modiﬁcations as markers of diagnosis, prognosis, and therapy response in cancers. However, epigenetic biomarkers have also been used as markers of disease (e.g., Alzheimer, Parkinson, diabetes, obesity), and other uses have included markers of response to environmental exposures and toxicology.

Characteristics Epigenetic Modifications ▶ Epigenetic changes include DNA methylation, ▶ histone modiﬁcations, and noncoding RNA which includes ▶ microRNAs. Epigenetic changes result in changes in gene expression without alterations to the DNA sequence. DNA methylation is the addition of a methyl group to the 50 carbon of a cytosine, which occurs predominantly at cytosine and guanine (CpG) dinucleotides and is catalyzed by DNA methyltransferases. Histone modiﬁcations are posttranslational modiﬁcations to the amino acids (lysine, arginine, serine, threonine, and proline) within histone tails and include acetylation, phosphorylation, ▶ ubiquitination, and methylation. Histone acetylation is associated with active gene transcription and is a reversible modiﬁcation regulated by the opposing actions of histone acetyltransferases and histone deacetylases. Histone methylation is associated with both active and inactive gene expression

Epigenetic Biomarker

and is regulated by histone methyltransferases and histone demethylases. Small noncoding RNAs, such as microRNAs, consist of approximately 21 nucleotides, exist naturally in the genome, are involved in the regulation of cellular functions, and can bind to partial or complete complementary mRNA targets to induce gene silencing. Epigenetic Biomarkers in Cancer In cancer, biomarkers are used for early detection (diagnosis), prognosis, and therapy response. Since epigenetic modiﬁcations such as DNA methylation and microRNAs are easy to measure in tumor tissue and peripheral ﬂuids, they are extensively investigated as biomarkers. Since the 1990s, assessing DNA methylation as a biomarker for cancer diagnosis, prognosis, and therapy response has been extensively pursued. This area has continued to grow to include the assessment of histone modiﬁcations as markers of patient outcome and microRNAs and other noncoding RNAs in tumor tissues and other peripheral samples. This has been greatly aided with the newest sequencing technologies and developments. Prostate Cancer Prostate cancer is one of the most commonly diagnosed cancers in men of developed Western countries, and globally, it is the 2nd most commonly diagnosed and 6th leading cause of cancer death in men. However, the majority of men diagnosed with prostate cancer have insigniﬁcant disease that will not cause their mortality. Current tests do not differentiate between these men and those who have clinically signiﬁcant disease and life-threatening prostate cancer, for whom early detection and treatment are necessary and whose disease is potentially curable. Several studies have shown that epigenetic modiﬁcations play a major role in prostate carcinogenesis, are capable of differentiating between benign and malignant disease, and are useful markers of response to epigenetic therapy. The only biomarker currently used for prostate cancer is serum ▶ prostate-speciﬁc antigen (PSA) levels. Studies have shown that a proportion of men without prostate cancer have high levels of serum PSA, and 22% of men with

Epigenetic Biomarker

prostate cancer have low serum PSA. Serum PSA levels cannot distinguish between patients with insigniﬁcant disease and those with clinically signiﬁcant cancer at diagnosis. Therefore, there is still a great need to identify biomarkers that differentiate insigniﬁcant from clinically signiﬁcant prostate cancers. GSTP1 DNA Methylation in Prostate Cancer DNA methylation is the most frequently studied epigenetic modiﬁcation in cancer. DNA methylation-based biomarkers for cancer are very appealing due to the high stability of DNA, the ease of analysis with the current techniques available, and the ability to assess the biomarker in body ﬂuids such as blood, urine, and saliva. The most studied epigenetic biomarker for prostate cancer is DNA methylation of the glutathioneS-transferase P1 (GSTP1) gene, which encodes an enzyme required for detoxiﬁcation and protection of DNA from oxidants and electrophilic metabolites. DNA methylation of GSTP1 has several characteristics which make it an ideal biomarker for prostate cancer: it can be easily measured in body ﬂuids, it has a higher speciﬁcity for prostate cancer compared to serum PSA, it can differentiate prostate cancer from other prostatic diseases, and different levels are associated with different stages of prostate cancer and recurrence of the disease following treatment. Although GSTP1 has a much higher speciﬁcity than serum PSA, it is still not 100% prostate cancer speciﬁc as DNA methylation of GSTP1 has been detected in other cancers. However, several studies have improved the overall speciﬁcity and sensitivity by combining GSTP1 with a panel of genes. This may be a better option as a biomarker for cancer and other diseases. Conclusion Although there has been extensive research and discovery in the ﬁeld of epigenetic modiﬁcations and their use as biomarkers, further studies are required. An exciting advancement in the development of epigenetic biomarkers is the improvements in technology, which now allow proﬁling of epigenetic alterations at a much higher sensitivity and genomic scale previously not possible. With the

1577

advancement of these new technologies, such as next-generation sequencing, and with the development of platforms for global epigenome analyses, the critical epigenetic alterations involved in tumorigenesis and other diseases and developmental processes will be identiﬁed. As we advance our ability to deﬁne and elucidate how epigenetic modiﬁcations are involved in the etiology of diseases, this will enhance our ability to use the assessment of epigenetic changes as biomarkers of diagnosis, prognosis, and therapy response and to bring epigenetic biomarkers into clinical use.

References Baylin SB, Jones PA (2011) A decade of exploring the cancer epigenome – biological and translational implications. Nat Rev Cancer 11:726–734 Enokida H, Shiina H, Urakami S, Igawa M, Ogishima T, Li LC, Kawahara M, Nakagawa M, Kane CJ, Carroll PR, Dahiya R (2005) Multigene methylation analysis for detection and staging of prostate cancer. Clin Cancer Res 11:6582–6588 Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D (2011) Global cancer statistics. CA Cancer J Clin 61:69–90 Jerónimo C, Henrique R, Hoque MO, Mambo E, Ribeiro FR, Varzim G, Oliveira J, Teixeira MR, Lopes C, Sidransky D (2004) A quantitative promoter methylation proﬁle of prostate cancer. Clin Cancer Res 10:8472–8478 Lee WH, Morton RA, Epstein JI, Brooks JD, Campbell PA, Bova GS, Hsieh WS, Isaacs WB, Nelson WG (1994) Cytidine methylation of regulatory sequences near the pi-class glutathione S-transferase gene accompanies human prostatic carcinogenesis. Proc Natl Acad Sci U S A 91:11733–11737 Nakayama M, Bennett CJ, Hicks JL, Epstein JI, Platz EA, Nelson WG, De Marzo AM (2003) Hypermethylation of the human glutathione S-transferase-pi gene (GSTP1) CpG island is present in a subset of proliferative inﬂammatory atrophy lesions but not in normal or hyperplastic epithelium of the prostate: a detailed study using laser-capture microdissection. Am J Pathol 163:923–933 Neal DE, Donovan JL (2000) Prostate cancer: to screen or not to screen? Lancet Oncol 1:17–24 Seligson DB, Horvath S, Shi T, Yu H, Tze S, Grunstein M, Kurdistani SK (2005) Global histone modiﬁcation patterns predict risk of prostate cancer recurrence. Nature 435:1262–1266 Wu T, Giovannucci E, Welge J, Mallick P, Tang WY, Ho SM (2011) Measurement of GSTP1 promoter methylation in body ﬂuids may complement PSA screening: a meta-analysis. Br J Cancer 105:65–73

1578

Epigenetic Gene Silencing

the functional importance of covalent epigenetic modiﬁcations, such as DNA methylation and histone posttranslational modiﬁcations, is becoming increasingly recognized as an important early event during carcinogenesis and tumor development.

Epigenetic Gene Silencing Kenneth P. Nephew School of Medicine, Indiana University, Bloomington, IN, USA

Definition Epigenetic: A heritable change in gene expression that is not accompanied by changes in DNA sequence. The two most studied ▶ epigenetic phenomena are DNA methylation and modiﬁcations of histone tails.

Characteristics The term epigenetics was originally coined to describe the development of phenotype from genotype. The ﬁeld of epigenetics now encompasses DNA methylation, covalent modiﬁcations of histones, nucleosome–DNA interactions, and small inhibitory RNA molecules. While it is well documented that genetic changes, such as mutations and deletions, play a functional role in silencing tumor suppressor genes in cancer cells,

Epigenetic Gene Silencing, Fig. 1 Schematic representation of cancer progression. Characteristic epigenetic features of cancer are listed. During tumor progression, accumulation of both genetic and epigenetic abnormalities contributes to carcinogenesis

Epigenetics and Cancer Epigenetic alterations are widely observed in cancer, and it has been shown that epigenetic mechanisms can be important to all phases of the cancer process, including tumor initiation, tumor progression, and maintenance of the malignant state of cancer cells (Fig. 1). Tumors exhibit two characteristic changes in DNA methylation patterns. One is genome-wide hypomethylation, primarily in repeat elements and pericentromeric regions, which is responsible for genomic instability. The other is promoter hypermethylation of normally protected ▶ CpG islands near the transcription start site of genes, which is responsible for transcriptional inactivation. Another characteristic of tumors includes hypoacetylation of histones, and HATs have been reported as upor downregulated in a number of tumors. In this entry, both DNA methylation and histone modiﬁcations are discussed in the context of cancer.

Epigenetic events in tumor progression Global Hypomethylation + (oncogene activation, genetic instability)

Normal

Region-Specific Hypermethylation (silencing of tumor suppressor genes)

Cancer

genetic and Accumulation of alities rm no epigenetic ab

Epigenetic Gene Silencing

DNA Methylation The transfer of a methyl group to the carbon 5 position of a cytosine residue within the context of a CpG dinucleotide is the only known epigenetic modiﬁcation of DNA itself. The 5-methylcytosine is the best-studied epigenetic modiﬁcation and often referred to as the “ﬁfth base” present in DNA. Although 5-methylcytosine comprises approximately 1% of the human genome, it is underrepresented in the bulk of the genome due to spontaneous deamination to thymine, which is not recognized by DNA mismatch repair systems. The CpG dinucleotides that are present are almost always methylated in normal cells, as 5-methylcytosine is widely believed to act to “silence” expression and/or retrotransposition of parasitic repeat sequences, such as Alu repeats and long-interspersed elements (LINEs) found throughout the genome. Aberrations in DNA methylation patterns are ﬁrmly associated with cancer, as evidenced by vast alterations in methylation patterns that occur during tumorigenesis, including global loss of methylation at CpG dinucleotides and regionspeciﬁc hypermethylation of distinct regions of 5-methylcytosine, located within speciﬁc CG-rich sequences known as CpG islands, often found within the promoter and associated with active genes. About 60% of human genes are associated with unique CpG islands, and it has been estimated that the human genome contains about 29,000 CpG islands. These normally unmethylated CpG islands may become methylated in cancer cells, and the event is associated with loss of expression of ﬂanking genes. It has also been estimated that aberrant methylation can accumulate in as many as 10% of the CpG islands during tumor development. Thus, abnormal de novo methylation of CpG islands in human cancer cells represents one of the most prevalent molecular markers yet identiﬁed, and the list of methylated genes identiﬁed in various tumor types continues to grow. DNA is methylated by DNMTs, a family of proteins. DNMT1 is responsible for the maintenance of DNA methylation after each round of replication. De novo methylation of DNA is the responsibility of DNMT3a and DNMT3b.

1579

Histone Modifications and Chromatin Remodeling While DNA methylation is currently the beststudied epigenetic modiﬁcation, other well-known chromatin alterations play critical roles in normal and aberrant physiological processes, including cancer. Histones are the major components of chromatin, the complex of DNA and protein found within the nucleus of eukaryotic cells. Early structural studies of chromatin revealed that 146 base pairs of DNA are “wrapped” around a core nucleosome, the fundamental unit of chromatin. A nucleosome consists of an octamer of histones, with 50 base pairs of “linker” DNA between repeating octamers. The N-terminal “tail” regions of histones extend from the nucleosome core octamer and are subject to various posttranslational modiﬁcations. Histone tails can be acetylated and deacetylated by HATs and HDACs, respectively. These enzymes are gene activators and repressors, and histone acetylation/deacetylation is important in transcriptional regulation. Additional histone modiﬁcations include methylation, by HMTs, phosphorylation, sumoylation, and ubiquitination. Overall, these and other modiﬁcations of histone tails play a critical role in chromatin compaction (e.g., tightly vs. loosely compacted chromatin), which can determine the access of various factors involved in transcription and gene expression and whether a gene is switched on (activated) or off (repressed). Furthermore, establishing transcriptional silencing of a gene involves a close interplay between histone modiﬁcations and DNA methylation, and loss and gain of DNA methylation, histone acetylation, and methylation are observed in cancer cells. The sum total of these covalent alterations in the epigenome has been referred to as the “histone code,” which can be “written” by the various modifying enzymes and “read” by various binding proteins that act to further modify chromatin and/or alter gene expression. According to the histone code hypothesis, the combination of DNA and histone modiﬁcations allows genes to go from the activate or inactive state interchangeably. The relatively new discipline of ▶ epigenomics promises to reveal novel insights into the histone code and a better understanding of normal development and human disease, including cancer.

1580

Clinical Relevance Unlike genetic changes, the epigenetic changes in cancer are potentially reversible. The ability to reactivate epigenetically silenced tumor suppressor genes and key control pathways and reverse the cancer cell phenotype is a promising strategy. Epigenetic drugs include both DNMT inhibitors (“demethylating” agents) and HDAC inhibitors (agents that cause “hyperacetylation” of histones). These agents can be used singly or in combination with currently available cancer chemotherapies. Epigenetic therapies themselves are now approved for various hematological malignancies and currently being studied in clinical trials for solid tumors. Furthermore, the combination of DNMT inhibitors with HDAC inhibitors has shown synergistic re-expression of epigenetically silenced genes and inhibition of tumor growth. Moreover, many investigators have shown that DNMT inhibitors with HDAC inhibitors can act to resensitize drug-resistant cancer cells to standard chemotherapeutic and hormonal agents. The ability to measure biochemical responses to epigenetic drugs, such as demethylation of previously hypermethylated genes, and correlate this with clinical responses has also been shown, and epigenetic therapies for chemoprevention in individuals with aberrant epigenetic alterations but have not yet developed cancer are an exciting possibility. Distinct CpG island methylation proﬁles for various cancers continue to emerge; consequently, aberrant DNA methylation and histone modiﬁcation patterns are now being investigated as potential biomarkers and pathway-speciﬁc therapeutic targets. Thus, epigenetic proﬁling of tumors could provide new therapeutic targets and epigenetic biomarkers for prognosis, such as predicting therapy outcome in patients with cancer or, ideally, early cancer detection. Clearly, attractive and promising clinical possibilities exist for epigenetic-based therapies. In summary, perhaps all human cancers are at least partially associated with epigenetic dysregulation of gene expression, forming a rational basis for future treatment strategies designed to alter this fundamental processes in cancer. Comprehensive elucidation of epigenetic

Epigenetic Gene Silencing

modiﬁcations, in both normal and diseased tissues, will allow for an extensive understanding of gene regulatory networks that control both normal and cancer phenotypes. The realization of the Human Epigenome Project (HEP), proposed as an exhaustive annotation of all histone and deoxycytosine modiﬁcations throughout the human genome, should have a major impact on cancer.

Cross-References ▶ CpG Islands ▶ Epigenetic ▶ Epigenomics

References American Association for Cancer Research Human Epigenome Task Force and European Union, Network of Excellence, Scientiﬁc Advisory Board (2008) Moving AHEAD with an international human epigenome project. Nature 454(7205):711–715. doi:10.1038/454711a Egger G, Liang G, Aparicio A, Jones PA (2004) Epigenetics in human disease and prospects for epigenetic therapy. Nature 429:457–463 Feinberg AP, Tycko B (2004) The history of cancer epigenetics. Nat Rev Cancer 4:143–153 Herman JG, Baylin SB (2003) Gene silencing in cancer in association with promoter hypermethylation. N Engl J Med 349:2042–2054 Nephew KP, Huang TH (2003) Epigenetic gene silencing in cancer initiation and progression. Cancer Lett 190:125–133 Toyota M, Issa JP (2005) Epigenetic changes in solid and hematopoietic tumors. Semin Oncol 32:521–530

See Also (2012) DNA methylation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1140. doi:10.1007/978-3-642-16483-5_1682 (2012) DNMTs. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1147. doi:10.1007/978-3-642-16483-5_1698 (2012) Epigenome. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1290. doi:10.1007/978-3-642-16483-5_1949 (2012) HAT. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 1632–1633. doi:10.1007/978-3-642-16483-5_2573 (2012) HDACs. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1635. doi:10.1007/978-3-642-16483-5_2592

Epigenetic Therapy

Epigenetic Modifications Definition Reversible, heritable changes in gene regulation that occur without a change in DNA sequence.

Cross-References ▶ Epigenetic

Epigenetic Therapy Debby Hellebrekers and Manon van Engeland Department of Pathology, GROW-School for Oncology and Developmental Biology, Maastricht University Hospital, Maastricht, The Netherlands

Definition Epigenetic therapy refers to therapy using inhibitors of DNA ▶ methylation and histone deacetylation, which reverse these epigenetic modiﬁcations. These compounds inhibit tumor growth in vitro and in vivo, which is thought to be due to reactivation of epigenetically silenced ▶ tumor suppressor genes by inhibition of promoter DNA methylation and histone deacetylation of these genes.

Characteristics DNA Methylation and Histone Deacetylation ▶ Epigenetic modiﬁcations regulate heritable changes in gene expression without changing the primary DNA sequence. The best-studied epigenetic mechanisms are DNA methylation and posttranslational histone modiﬁcations. DNA methylation is the covalent addition of a methyl group to the DNA, predominantly to the base

1581

cytosine 50 to guanine, also called a CpG dinucleotide. CpG dinucleotides are clustered in small stretches of DNA called ▶ CpG islands, often located in or near the promoter region of approximately half of all human genes. Methylation of CpG dinucleotides, which occurs nonrandomly, is an important ▶ epigenetic gene silencing mechanism. Most methylation in the human genome occurs in the noncoding DNA, preventing the transcription of repeat elements, inserted viral sequences, and transposons. In contrast, CpG islands are largely unmethylated in both expressing and nonexpressing tissues under normal conditions. Exceptions to this unmethylated state of CpG islands involve the silenced gene alleles for imprinted genes and genes located on the inactive X chromosome of females. DNA methylation is catalyzed by DNA methyltransferases (DNMTs). DNA methylation can induce gene silencing through several mechanisms. By sterically hindering the binding of activating transcription factors to gene promoters, DNA methylation can directly repress gene transcription. Another mechanism is through recruitment of several methyl-binding domain proteins (MBDs) that recognize methylated DNA, including MeCP2, MBD1–4, and Kaiso. These proteins themselves can repress gene transcription or bind proteins which cause gene silencing. The DNA helix is wrapped around a core of histone proteins. The basic amino-terminal tails of histones are subject to various posttranslational modiﬁcations, including acetylation, methylation, phosphorylation, ubiquitination, sumoylation, ADP-ribosylation, glycosylation, biotinylation, and carbonylation. The best-characterized histone modiﬁcation is histone acetylation, which is controlled by histone acetyltransferases (HATs) and histone deacetylases (HDACs). Histone acetylation generally correlates to active gene transcription, whereas histone deacetylation is associated with transcriptional repression, by blocking accessibility of transcription factors to their binding sites. DNA methylation and histone deacetylation are interconnected in gene silencing. Methylbinding domain proteins are components of HDAC complexes or recruit these complexes to

1582

methylated DNA, resulting in transcriptional silencing. Furthermore, a much more direct connection between DNA methylation and histone deacetylation exists by direct interactions between DNMTs and HDACs. DNA methylation and histone deacetylation are pivotal in X chromosome inactivation, ▶ imprinting, and establishment of tissue-speciﬁc gene expression. However, aberrant epigenetic regulation of gene expression also plays a major role in the development of human cancer. Epigenetic Abnormalities in Cancer Aberrant epigenetic silencing of tumor suppressor genes by promoter DNA hypermethylation and histone deacetylation plays an important role in the pathogenesis of cancer. According to Knudson’s two-hit model, complete loss of function of a tumor suppressor gene requires loss of function of both gene copies. Epigenetic silencing of the wild-type allele of a tumor suppressor gene by aberrant promoter hypermethylation and histone deacetylation can be considered as the second hit in this model, resulting in complete loss of function of the gene. It has become apparent that many genes, located across all chromosome locations, are epigenetically silenced in cancer cells. Examples are genes involved in cell cycle regulation and apoptosis (p14ARF, p15INK4b, p16INK4a, APC, RASSF1A, and HIC1), DNA repair genes (hMLH1, GSTP1, MGMT, and BRCA1), and genes related to ▶ metastasis and invasion (CDH1, TIMP-3, DAPK, p73, maspin, TSP1, and VHL). DNA Methyltransferase and Histone Deacetylase Inhibitors In contrast to genetic alterations, which are almost impossible to reverse, ▶ chromatin remodeling in cancers is potentially reversible. This resulted in the development of pharmacologic inhibitors of DNA methylation and histone deacetylation. By inducing DNA demethylation and histone acetylation, DNMT and HDAC inhibitors can reverse epigenetic silencing of tumor suppressor genes, resulting in reactivation of these genes in tumor cells and restoring of crucial cellular pathways. The most extensively studied DNMT inhibitors

Epigenetic Therapy

are 5-azacytidine and 5-aza- 20 -deoxycytidine [5-aza-20 deoxycytidine and Cancer], which were initially developed as chemotherapeutic agents. These nucleoside analogs are incorporated into DNA in place of the natural base cytosine during DNA replication and are therefore only active during S phase. Once incorporated into the DNA, a complex is formed with active sites of DNMTs, thereby covalently trapping these enzymes. This results in the depletion of active enzymes and the demethylation of DNA after several cell divisions. 5-Aza- 20 -deoxycytidine is the most commonly used DNMT inhibitor in assays with cultured cells. This compound reactivates dormant tumor suppressor genes by demethylation of their hypermethylated promoter, thereby restoring their normal function. This seems to be a widespread effect of 5-aza20 -deoxycytidine, because all cancer cell lines studied so far are sensitive to the DNA demethylating effects of this agent. Reactivation of silenced tumor suppressor genes might be the mechanism by which this compound suppresses growth and induces differentiation of human tumor cell lines. Examples of other DNMT inhibitors are the cytidine analogs 5,6-dihydro-5azacytidine, 5-ﬂuoro-20 -deoxycytidine, and zebularine, the small molecule RG108, which blocks the DNMT active site, and MG98, an antisense oligonucleotide that speciﬁcally inhibits DNMT1 mRNA. By inhibiting histone deacetylation, HDAC inhibitors cause accumulation of acetylated histones, leading to increased transcription of previously silenced tumor suppressor genes in malignant cells. Both naturally existing and synthetic HDAC inhibitors have been characterized. The effects of HDAC inhibitors on gene expression in transformed cells are selective; only about 2–10% of all known genes are affected by these agents. One gene most consistently induced by HDAC inhibition is CDKN1A, which encodes the cell cycle inhibitor ▶ p21. HDAC inhibitors can also relieve inappropriate transcriptional repression mediated by chimeric oncoproteins, such as PML-RARa, thereby inducing differentiation in cells harboring these translocations. HDAC inhibitors have many antitumor effects

Epigenetic Therapy

including induction of cell cycle arrest, differentiation, and/or apoptosis in virtually all cultured transformed cell types and in cells from different tumors. Epigenetic Therapy in Cancer It is clear from in vitro and preclinical studies that the clinical application of reversing epigenetic aberrations in tumor cells, called epigenetic therapy, is an exciting strategy for cancer treatment. Many agents have been discovered that inhibit DNA ▶ methylation or histone deacetylation, and the value of these compounds will be established by ongoing ▶ clinical trials. 5-Azacytidine (Vidaza) and 5-aza20 -deoxycytidine (Decitabine) represent the two most prominent inhibitors of DNA methyltransferases that are being used in clinical practice, and these drugs have been approved by the FDA for the treatment of myelodysplastic syndrome (MDS). The use of DNMT inhibitors in the treatment of MDS results from the knowledge that epigenetic gene silencing of, in particular, p15INK4b is present in poor-risk MDS subtypes and often predicts transformation to acute myeloid leukemia (AML). Multiple HDAC inhibitors are currently being tested in patients through intravenous or oral administration. Suberoylanilide hydroxamic acid (SAHA) is one of the HDAC inhibitors most advanced in development. Encouraging results were obtained in Phase I, II, and III clinical trials for patients with both hematologic and solid tumors. Other examples of HDAC inhibitors undergoing clinical testing in a range of solid and hematological malignancies are ▶ valproic acid, PXD101, NVP-LAQ824, LBH589, depsipeptide, MS-275, and CI-994. Complete targeting of epigenetic gene regulation might require a combination of chromatin modifying agents. The synergy between demethylating drugs and HDAC inhibitors in reactivation of epigenetically silenced tumor suppressor genes in vitro makes combined treatment with DNMT and HDAC inhibitors a promising epigenetic therapy. Reduction of individual doses should minimize toxic effects and optimize the therapeutic response of such combination. Encouraging anticancer activity of epigenetic therapy has been

1583

shown particularly in the treatment of hematologic disorders, but their effectiveness in solid tumors largely remains to be determined. In addition to the inhibitory effects of DNMT and HDAC inhibitors on tumor cells, by reactivation of epigenetically silenced tumor suppressor genes, these compounds have also been described to inhibit tumor endothelial cell growth and tumor angiogenesis. ▶ Angiogenesis is required for tumor growth to a size of approximately 2 mm3 but is also instrumental for tumor cells to metastasize to other locations in the body. Angiogenesis is considered to be a promising target of anticancer treatment. By inﬂuencing the gene expression proﬁle of tumor cells, DNMT and HDAC inhibitors target genes, which are regulating angiogenesis. Among the epigenetically silenced tumor suppressor genes in tumor cells are genes with angiogenesis inhibiting properties. By reexpression of these genes in tumor cells, DNMT and HDAC inhibitors might indirectly – via the tumor cells – exhibit angiostatic effects in vivo. An example of these genes is p16INK4a, which can regulate angiogenesis by modulating ▶ vascular endothelial growth factor (VEGF) expression by the tumor. Another epigenetically silenced tumor suppressor gene that inhibits angiogenesis by downregulation of VEGF is p73. The tumor suppressor ▶ maspin, which is often silenced in tumors by epigenetic promoter modiﬁcations, is also an effective inhibitor of angiogenesis. Methylation-associated inactivation of the angiogenesis inhibiting factors tissue inhibitor of metalloproteinase-2- and -3 (TIMP-2/3) is frequent in many human tumors. Thrombospondin-1 (TSP-1) has been described to be repressed by epigenetic promoter modiﬁcations in several adult cancers and can be reactivated by 5-aza- 20 -deoxycytidine. The secreted protease ADAMTS-8 (METH-2) has anti-angiogenic properties, which can speciﬁcally suppress endothelial cell proliferation, and significant downregulation of ADAMTS-8 by promoter hypermethylation, which has been described in different tumor types. Besides the indirect effects of DNMT and HDAC inhibitors on tumor angiogenesis, these compounds also directly inhibit endothelial cell growth and angiogenesis in vitro

1584

Epigenetic Therapy

DNMT-and HDAC inhibitors Inhibition of tumor cell growth by reactivation of epigenetically silenced tumor suppressor genes in tumor cells p14ARF p15INK4b p16INK4b 1

Indirect angiostatic effects by reactivation of epigenetically silenced angiogenesis-inhibiting tumor suppressor genes in tumor cells p16INK4a p73 maspin TIMP-2/-3 TSP-1 ADAMTS-8

Direct angiostatic effects by inhibition of growth (and / or migration) and sprouting of tumor endothelial cells

Epigenetic Therapy, Fig. 1 Antitumor effects of DNA methyltransferase and histone deacetylase inhibitors in vivo. (1) DNMT and HDAC inhibitors decrease tumor cell growth by reactivation of epigenetically silenced tumor suppressor genes in tumor cells. (2) Release of transcriptional repression of angiogenesis inhibiting

tumor suppressor genes in tumor cells might result in indirect angiostatic effects of DNMT and HDAC inhibitors. (3) DNMT and HDAC inhibitors directly decrease endothelial cell growth and angiogenesis, thereby exhibiting direct angiostatic effects

and in vivo. Potent anti-angiogenic activity has been described for several HDAC inhibitors, such as trichostatin A (TSA), SAHA, depsipeptide, valproic acid, butyrate, apicidin, LBH589, and NVP-LAQ824, as well as for the DNMT inhibitors 5-aza- 20 -deoxycytidine and zebularine. These drugs suppress spontaneous or VEGFinduced angiogenesis in different in vitro, ex vivo, and in vivo angiogenesis assays. Clearly, the dual targeting of epigenetic therapy in cancer treatment, inhibiting both tumor cells as well as tumor angiogenesis, makes them suitable combinatorial anticancer therapeutics (Fig. 1). By targeting multiple genes and pathways in tumor cells, as well as endothelial cell biology and angiogenesis, DNMT and HDAC inhibitors decrease the development of resistance that is associated with many of the current chemotherapeutic drugs and anti-angiogenic drugs.

Despite the promising data from clinical trials, there are several pitfalls regarding the clinical application of epigenetic therapy. An important side effect that should be taken into account in the use of these drugs is induction of global hypomethylation, which might induce tumorigenesis by aberrant activation of repetitive DNA sequences, transposons and ▶ oncogenes, induction of chromosomal instability, and mutagenesis. Furthermore, the existence of many different DNMTs and HDACs makes the development of selective inhibitors that target individual enzymes imperative.

Cross-References ▶ Angiogenesis ▶ Chromatin Remodeling

Epigenomics

▶ Clinical Trial ▶ CpG Islands ▶ Epigenetic ▶ Epigenetic Gene Silencing ▶ Imprinting ▶ Knudson Hypothesis ▶ Maspin ▶ Metastasis ▶ Methylation ▶ Oncogene ▶ p21 ▶ Tumor Suppressor Genes ▶ Valproic Acid ▶ Vascular Endothelial Growth Factor

References Egger G, Liang G, Aparicio APA et al (2004) Epigenetics in human disease and prospects for epigenetic therapy. Nature 429:457–463 Folkman J (2006) Antiangiogenesis in cancer therapy – endostatin and its mechanisms of action. Exp Cell Res 312:594–607 Hellebrekers DM, Grifﬁoen AW, van Engeland M (2007) Dual targeting of epigenetic therapy in cancer. Biochim Biophys Acta 1775:76–91 Herman JG, Baylin SB (2003) Gene silencing in cancer in association with promoter hypermethylation. N Engl J Med 349:2042–2054 Minucci S, Pelicci PG (2006) Histone deacetylase inhibitors and the promise of epigenetic (and more) treatments for cancer. Nat Rev Cancer 6:38–51

See Also (2012) Anti-Angiogenic Drugs. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, pp 207–208. doi:10.1007/978-3-642-164835_302 (2012) Chimeric Oncoproteins. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 811. doi:10.1007/978-3-642-164835_1095 (2012) Chromatin. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 825. doi:10.1007/978-3-642-16483-5_1125 (2012) DNA Methyltransferases. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1140. doi:10.1007/978-3-642-16483-5_6997 (2012) FDA. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1386. doi:10.1007/978-3-642-16483-5_2136 (2012) Histone Deacetylation. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1702. doi:10.1007/978-3-642-16483-5_2753

1585 (2012) Histone Proteins. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1706. doi:10.1007/978-3-642-16483-5_2761 (2012) Nucleoside Analogs. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 2577. doi:10.1007/978-3-642-16483-5_4164 (2012) PML-RARa. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 2931. doi:10.1007/978-3-642-16483-5_4645 (2012) Tumor Endothelial Cell. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 3795. doi:10.1007/978-3-642-16483-5_6026

Epigenomics William Chi-Shing Cho Department of Clinical Oncology, Queen Elizabeth Hospital, Kowloon, Hong Kong

Definition Epigenomics is the study based on the comprehensive analyses of the entire epigenome using high-throughput technologies, including physical modiﬁcations, associations and conformations of genomic DNA sequences. The complete set of the epigenetic landscape in a cell is known as epigenome, which is tissue speciﬁc, developmentally regulated, and highly dynamic. This variability offers a potential explanation for individual differences in phenotype. Aberrant epigenetic marks are associated with a range of complex pathologies, including ▶ cancer. The ﬁeld of epigenomics involves chromatin, the threedimensional complex of DNA, protein, and ▶ noncoding RNAs that determines the accessibility of DNA by the transcriptional machinery.

Characteristics The Epigenome and Cancer The term ▶ epigenetics was ﬁrst used by Prof. Conrad H. Waddington in 1942 as part of his model of how cell fates are established during development. It usually refers to reversible biochemical alterations of DNA and related proteins.

1586

Although it represents memories of molecular decisions that can be perpetuated through cell divisions, it does not change the DNA sequence. As epigenetic factors (such as DNA ▶ methylation, ▶ histone modiﬁcation, histone variant, ▶ chromatin remodeling, and noncoding RNA) can inﬂuence chromatin structure and dynamics, they play a signiﬁcant role in determining cell fate and are important epigenetic mechanisms that regulate gene expressions. Posttranslational modiﬁcation of chromatin facilitates conformational transition between transcriptionally permissive and suppressive chromatin structures. Dysregulation of chromatin modiﬁcation leads to pathologic alteration in gene transcription, hence to human disease. DNA methylation is binary at the cellular level, i.e., a CpG site in a given allele in a single cell is either methylated or unmethylated. Methylation of cytosine in the context of CpG dinucleotides is an epigenetic phenomenon in eukaryotes that plays a vital role in genome function and transcription regulation, including host defense of endogenous parasitic sequences, embryonic development, transcription, X chromosome inactivation, and ▶ genomic imprinting. Aberrant change in DNA methylation is a main feature of several human diseases, such as cancer and neurological disorders. Epigenetic modiﬁcations in the human genome can regulate and impact the expression of coding genes and noncoding genes, such as ▶ microRNA (miRNA), at the transcriptional level; they thus play a critical role in cancer initiation and development. Some of the epigenetic changes have been well-characterized, in which a range of modiﬁcations (including methylation, acetylation, phosphorylation, ▶ ubiquitination, deamination, citrullination, sumoylation, ADP-ribosylation and isomerization) of histones’ N-terminal tails are involved. The enzymatic activities of chromatin modiﬁers contribute to the epigenetic regulation of gene expression through the remodeling of chromatin. In addition, there are at least 16 distinct classes of histone modiﬁcations and more than a hundred different modiﬁcation sites in the major core histones.

Epigenomics

Since anomalous DNA methylation and histone modiﬁcation are often seen in cancer, epigenetic change is considered to be a characteristic of tumorigenesis. It is believed that changes in DNA methylation are molecular signatures of the tumor itself. As DNA methylation heterogeneity may differentiate the cancer group from the unaffected individuals even before other histopathological changes appear, DNA methylation ▶ biomarkers circulating in blood can be used for early screening and the follow-up of high-risk subjects. miRNAs can transcriptionally cleave or translationally repress the expressions of major enzymes participating in epigenetic processes, such as DNA methylation and histone modiﬁcations. With the regulatory effect on gene expression, miRNAs involve in many vital biological and cellular processes, including proliferation, differentiation, ▶ apoptosis, metabolism and viral infection. They are thus promising therapeutic targets for cancer interventions which may inﬂuence the epigenetic states associated with cancer. Technologies Bisulﬁde treatment converting unmethylated cytosine to uracil followed by cloning and Sanger sequencing is the gold standard for DNA methylation analysis of individual genes, which is a quantitative, allelic, contiguous, and base-pair resolution of CpG methylation approach. Development of high-throughput technology and the systematic assessment of accumulated data allow the identiﬁcation of previously unknown biological processes and disease states in terms of wholegenome proﬁles of epigenetic signatures at a high resolution. Focusing on the discovery of speciﬁc genetic risk factors, genome-wide association studies have identiﬁed many disease-associated loci. It helps us to understand the genetic basis of complex traits and motivates us to explore the epigenetic contribution to interindividual variation in complex phenotypes. The application of genomewide methylation arrays has proved very informative to investigate both clinical and biological questions in human epigenomics. There are a number of informative functional genomics and

Epigenomics

epigenomics assays available in cancer research, such as DNA methylome for proﬁling genomewide DNA methylation status, HELP tagging using massively parallel sequencing technology for genome-wide methylation proﬁling, DNaseseq for identifying genome-wide accessible chromatin regions that are hypersensitive to DNase I cleavage, ChIP-seq for investigating genome-wide transcription factor binding or histone modiﬁcations, as well as GRO-seq for studying genome-wide nascent RNA and transcriptional rate. Utilization of high-throughput sequencing also facilitates many applications to understand the transcriptional and epigenetic gene regulations. Rapid advances in the proﬁling of sequencing-based DNA methylation have enabled comprehensive comparison of the complete DNA methylome between cancer cells and normal tissues. The sequencing-based methods permit the measurement of DNA methylation in interspersed repeat sequences, which is unattainable using microarray platform. The advancement in next-generation sequencing is proceeding in a breathtaking pace. Notable increases in sequence reads per lane and read length, along with the development of pairedend sequencing, have created optimism for the future of sequencing-based methylome mapping. Epigenomic Researches and Therapies Aiming to catalog the epigenomic patterns across different cell and tissue types using nextgeneration sequencing-based methodologies, the National Institute of Health Roadmap Epigenomics was launched in 2008. A new epigenomics database has been established at the National Center for Biotechnology Information to serve as a comprehensive public resource for epigenetic and epigenomic data sets (www.ncbi.nlm. nih.gov/epigenomics). This database collects data from several large-scale projects, including the National Institute of Health Roadmap Epigenomics project, the ENCODE and modENCODE projects, as well as from smaller single laboratory studies. It enables the epigenomic data to be readily available and easily accessed by the scientiﬁc community.

1587

Growing data demonstrate that environmental and lifestyle factors can affect epigenetic patterns, such as pollution, alcohol consumption, depression, smoking, diet, obesity, and physical activity. In addition, aging is likely to drift with a loss of DNA methylation and an increased hypermethylation of CpG islands. For example, small but statistically signiﬁcant age-related demethylation of CpG motifs in the tumor necrosis factor promoter is detected in human peripheral blood cells and primary monocyte-derived macrophages from healthy subjects using pyrosequencing. Unlike genetic modiﬁcations, epigenetic alterations are potentially reversible. Exciting new developments in medicinal chemistry deliver an extensive collection of potential therapeutic compounds against chromatin regulators, which may target the catalytic activity of diverse chromatin modiﬁers/remodelers and chromatin-reading/chromatin-binding modules. There are a number of DNA demethylation agents emerging as therapeutics in the clinic. The epigenetic treatments targeting against the catalytic activity of chromatin-modifying enzymes have proved to be successful, and they have become a new anticancer therapy. As DNA demethylation therapeutics is emerging as potential clinical agents, there is a pressing need to identify DNA demethylation resistant genes and clarify their role in cancer treatment. Rapid advancements in high-throughput epigenomic technologies have led us to a new era in epigenetic biology which enhances our understanding of cancer cells. Along with the promising anticancer effects shown in some inhibitors of enzymes controlling epigenetic changes, the development of epigenetic-based therapies possesses great potential in the clinical applications against cancer.

Cross-References ▶ Apoptosis ▶ Cancer Epigenetics ▶ Carcinogenesis ▶ Chromatin Remodeling ▶ Epigenetic ▶ Epigenetic Gene Silencing ▶ Genomic Imprinting

1588

▶ Histone Modiﬁcation ▶ Methylation ▶ MicroRNA ▶ Noncoding RNA ▶ Ubiquitination

Epithelial Cadherin

Epithelial Carcinogenesis ▶ Epithelial Tumorigenesis

References

Epithelial Cell Adhesion Molecule

Baylin SB, Jones PA (2011) A decade of exploring the cancer epigenome – biological and translational implications. Nat Rev Cancer 11:726–734 Cancer Genome Atlas Research Network (2013) Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl J Med 368:2059-2074 Cho WC (2010) MicroRNAs in cancer – from research to therapy. Biochim Biophys Acta 1805:209–217 Jiang W, Liu N, Chen XZ, et al (2015) Genome-wide identiﬁcation of a methylation gene panel as a prognostic biomarker in nasopharyngeal carcinoma. Mol Cancer Ther 14:2864–2873 Mullard A (2015) The Roadmap Epigenomics Project opens new drug development avenues. Nat Rev Drug Discov 14:223–225 Roadmap Epigenomics Consortium, Kundaje A, Meuleman W, et al (2015) Integrative analysis of 111 reference human epigenomes. Nature 518:317–330 Taudt A, Colomé-Tatché M, Johannes F (2016) Genetic sources of population epigenomic variation. Nat Rev Genet 17:319–332

▶ EpCAM

See Also (2012) Acetylation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 17. doi:10.1007/978-3-642-16483-5_24 (2012) Biomarkers. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 408– 409. doi:10.1007/978-3-642-16483-5_6601 (2012) Hypermethylation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1784. doi:10.1007/978-3-642-16483-5_2910 (2012) Phosphorylation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2870. doi:10.1007/978-3-642-16483-5_4544 (2012) Sumoylation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3562. doi:10.1007/978-3-642-16483-5_5572 (2012) Transcription factor. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3752. doi:10.1007/978-3-642-16483-5_5901

Epithelial Tumorigenesis Vassilis Gorgoulis and Eleni A. Georgakopoulou Department of Histology and Embryology, Faculty of Medicine, National and Kapodistrian University of Athens, Athens, Greece

List of Abbreviations AKT

APC EGFR GSTP HPV HRAS KRAS LOH MDR1 MGMT MLF1 MYC PTEN RARRES RASSF2

Epithelial Cadherin ▶ E-Cadherin

TGF-b WNT

Nonspeciﬁc name abbreviation for a group of serine/threonine kinases involved in many cellular processes Adenomatous polyposis coli Epidermal growth factor receptor Glutathione S-transferase P enzymes Human papillomavirus Harvey sarcoma viral oncogene Kirsten rat sarcoma viral oncogene homologue Loss of heterozygosity Multidrug resistance 1 gene O6-Methylguanine DNA methyltransferase Myeloid leukemia factor 1 Cellular myelocytomatosis viral oncogene Phosphatase and tensin homologues Retinoic acid receptor responder protein Ras association domain family member 2 Transforming growth factor-beta Wingless-type MMTV (mouse mammary tumor virus) integration site family

Epithelial Tumorigenesis

1589

Synonyms Carcinoma carcinogenesis

Epithelial cancers in a simple classiﬁcation may be of three types: pathogenesis;

Epithelial

Definition Epithelial tumorigenesis relates to the process of developing and/or progressing of a tumor, originating from epithelial cells. Malignant tumors of this origin are known as carcinoma, whereas malignant tumors derived from cells of the connective tissue are known as sarcoma (Fig. 1).

Characteristics Introduction Epithelial cells are found in most organs of the human body, and 90% of human cancers originate from epithelial cells. Epithelial cancers (carcinomas) are the most common cancers among elderly, in contrast to mesenchymal cancers and ▶ hematological malignancies that are more common in younger patients and children. Characteristic epithelial cancers are ▶ skin cancer and cancers that originate from cells of the epithelial lining of the aerodigestive and genitourinary system.

• Squamous cell carcinomas (e.g., skin carcinoma, mouth carcinoma) • Adenocarcinomas (e.g., colon adenocarcinoma, lung adenocarcinoma) • Transitional cell carcinomas (e.g., bladder cancer, renal carcinoma) Predisposing Factors Age is the common risk factor for all epithelial cancers as their incidence rises with age. Genetic predisposition in epithelial cancer may be associated with syndromes where inherited mutations lead to cancer development (Table 1). Usually, in the cases of inherited cancer syndromes, the genes affected are those controlling DNA damage repair mechanisms. An example is hereditary nonpolyposis colorectal cancer (HNPCC) in which inherited germline mutations affect the DNA mismatch repair system (MMR). Various agents and conditions have been associated with predisposition to epithelial cancers. Most studied are tobacco and ▶ alcohol consumption for aerodigestive tract cancers, HPV infection for cervical cancer, ultraviolet radiation for skin cancer, ▶ H. pylori infection for ▶ gastric cancer, and ▶ obesity and inﬂammatory bowel disease for colorectal cancer.

Epithelial Tumorigenesis, Fig. 1 Illustration of the main identiﬁed events that characterize epithelial tumorigenesis

1590

Epithelial Tumorigenesis

Epithelial Tumorigenesis, Table 1 Some hereditary cancer syndromes, associated genetic mutations, and carcinomas Cancer syndrome Hereditary breast and ovarian cancer syndrome Cowden syndrome Hereditary nonpolyposis colon cancer Familial adenomatous polyposis Xeroderma pigmentosum

Genes BRCA1, BRCA2

Carcinomas Breast, ovarian, prostate

PTEN

Breast, thyroid, uterus, kidney Colorectal ovarian, hepatobiliary

MLH1, MSH2, MSH6 ▶ APC

XP A-D/ ERCC 4-6

Epithelial Tumorigenesis, Table 2 Common genetic changes in some epithelial cancers Cancer Lung Colon Oral Gastric Bladder Breast Cervical Prostate

Genetic alterations in ▶ p53, MYC, EGFR ▶ APC, ▶ KRAS, ▶ TP53 MYC, ▶ TP53 MLF1, MGMT, p16, RASSF2 ▶ HRAS, ▶ KRAS, RB BRCA1, BRCA2 EGFR, AKT, PTEN GSTP, RARRES, MDR1

Colorectal

Skin basal and squamous cell carcinoma (and melanoma)

The Multistage Pathogenesis Model of Epithelial Tumorigenesis In almost all cancers of epithelial origin, a “pattern” of multistage pathogenesis has been identiﬁed. This model comprises an initiation phase, a promotion phase, a premalignant phase, and ﬁnally malignant progression. It requires the interaction of normal cells with carcinogenic agents which lead to genetic alterations. Alterations in the affected cells are the cause of their clonal expansion and accumulation of further alterations. (Common genetic alterations in epithelial cancers are summarized in Table 2.) A critical point of genetic alterations may lead to autonomy in growth and invasive phenotype. Such cancer models have been described in several carcinomas including oral cancer, colon cancer, and skin cancer. They have been reproduced under laboratory conditions by the repeated application of chemical carcinogens in mouse models. Example of Multistage Carcinogenesis in Colorectal Cancer Initiation ! promotion ! premalignancy ! cancer Normal cell ! hyperproliferation ! early and late adenoma ! carcinoma Changes in APC/KRAS ! LOH in 18q ! LOH 17p

Multistage carcinogenesis requires the accumulation of mutations that either enhance the expression of oncogenes or block the expression of tumor suppressor genes and is in accordance with the “mutation theory” for epithelial tumorigenesis. Commonly altered genes in epithelial cancers are the tumor suppressor genes encoding the proteins ▶ p53 and RB (chromosome location 17p and 13q, respectively). P53, often referred to as the “guardian of the genome,” is one of the most studied tumor suppressor genes. P53 is inactivated in many cancers, and its role is crucial in cell cycle control. The RB gene is a tumor suppressor gene whose protein is a crucial cell cycle regulator. The Rb protein interacts with ▶ cyclin D and ▶ cyclin-dependent kinases 6 and 4 to allow the cell cycle to progress. Rb is also an E2F transcription factor regulator. The Rb protein is a target of the HPV viral protein E7, which binds to Rb and inhibits E2F control, thus allowing for malignant progression. Multiple “hits” in different genes are required for the development of malignant cell clones. According to the “mutation theory,” mutations are the events responsible for cancer development. However, the role of the epithelial microenvironment in the pathogenesis of epithelial cancer has been highlighted. Nevertheless, mutations especially in the genes that control DNA repair and cell cycle are key events for genomic instability, one of the hallmarks of malignant and premalignant lesions. Apart from mutations, another characteristic of the premalignant phase and cancer in comparison to normal epithelium is the presence of ▶ DNA

Epithelial Tumorigenesis

damage as well as ▶ DNA damage response activation. In premalignant lesions it is possible that oncogenes may cause DNA replication stress (unprogrammed and defective DNA replication) resulting in DNA damage (e.g., in the form of DNA double-strand breaks). This concept is supported by the fact that DNA double-strand breaks and DNA repair mechanisms are present in cancers and precancerous lesions. P53 activation in premalignant lesions could be a result of DNA damage checkpoint or ▶ arf action, and this activation suppresses tumorigenesis. P53 mutations relate to progression of premalignant lesions to cancer. The Role of Stem Cells in Epithelial Tumorigenesis The initial cellular origin of epithelial malignant clones is a matter of debate. Adult stem cells have been identiﬁed in carcinomas and are named ▶ cancer stem cells. Their role is uncertain. It has been proposed that cancer stem cells are the only cells that can develop clonal populations when they accumulate a critical amount of genetic alterations, in contrast to the concept that any cell may acquire clonal properties. Furthermore, cancer stem cells have been blamed for the resistance of many epithelial cancers to chemotherapy and radiotherapy. Epigenetic Changes and Epithelial Tumorigenesis Epigenetic changes are the changes that affect the expression of genes without changing the DNA sequence. Such changes are hypermethylation and hypomethylation of promoter regions and histone deacetylation. Hypermethylation of tumor suppressor genes is a characteristic of many epithelial cancers. DNA hypermethylation may be the result of exogenous carcinogens and may be conserved during mitosis. Also, histone deacetylation results in different histone types that have various functions, e.g., as DNA damage marker (▶ H2AX variant) or gene promoter regions (H2AZ variant). Epigenetic changes may also be identiﬁed in DNA areas containing genes that control cell death, cell cycle regulation, or DNA repair. Accumulation of epigenetic changes

1591

contributes to genomic instability and sequential carcinogenesis. The Interaction of Cellular Microenvironment in Epithelial Tumorigenesis The epithelial microenvironment is composed of non-epithelial cells that form the supportive stroma of the epithelium (ﬁbroblasts, lymphocytes, endothelial cells in vessels), the extracellular matrix, and produce a variety of growth factors and cytokines. The interaction between stroma and epithelial cells is important during embryogenesis for the organization and differentiation of the various epithelia in different systems of the human body. These interactions have also been proven important for the development and progression of epithelial cancer. Stroma ﬁbroblasts are involved in epithelial tumor development by remodeling the extracellular matrix and by providing the epithelial tumor cells with the required growth factors signals to enhance their autonomy in proliferating signals. Epithelial Mesenchymal Transition (EMT) An important role in the epithelial integrity is played by the junctional structures (gap junctions, desmosomes, hemidesmosomes) that adhere epithelial cells to each other and to the stromal structures, as well as the apicobasal polarity of epithelial cells and the stable direction of their division in parallel with the epithelial sheet. These features are characteristic of the epithelial phenotype. Changes in these cellular characteristics and development of a mesenchymal phenotype are a characteristic of advanced epithelial cancers. This phenomenon described as ▶ epithelial mesenchymal transition enables epithelial malignant cells to invade the connective tissue and also to metastasize in distant organs through the lymphatic or hematogenous route. The most studied epithelial mesenchymal transition molecules are ▶ TGF-b, Twist, ▶ Snail/Slug, ZEB, and ▶ Wnt. In the same sense, epithelial cells also accumulate motility and plasticity as they transform to a mesenchymal phenotype which enables malignant epithelial cells to migrate to surrounding stroma. A characteristic of EMT is the downregulation of the cellular adhesion

1592

molecule E-cadherin. E-cadherin is normally located in the cellular membrane of epithelial cells and functions as a blockage for the development and inﬁltration of malignant cells. Chronic Inflammation and Epithelial Cancers Several epithelial cancers develop on the basis of previous chronic inﬂammation. Such carcinomas are gastric cancers following chronic gastritis, esophageal cancer following chronic esophageal inﬂammation, and colorectal cancer on the basis of inﬂammatory bowel disease. The chronic inﬂammation causes oxidative stress and damage to DNA that result in malignant transformation. The inﬂammatory mediators produced by the inﬂammatory cellular populations support the cancer cells with growth signals and enable tumor progression. Summary Epithelial tumorigenesis is a complex multifactorial process. Cellular alterations that result in malignant changes are a combination of mutations, DNA damage epigenetic changes, and interactions of malignant cells with their microenvironment.

Cross-References ▶ Alcohol Consumption ▶ Cyclin D ▶ Cyclin-Dependent Kinases ▶ DNA Damage ▶ DNA Damage Response ▶ Epidermal Growth Factor Receptor ▶ Epithelial-to-Mesenchymal Transition ▶ Gastric Cancer ▶ Helicobacter Pylori in the Pathogenesis of Gastric Cancer ▶ HRAS ▶ KRAS ▶ Lynch Syndrome ▶ Mismatch Repair in Genetic Instability ▶ MYC Oncogene ▶ Skin Cancer ▶ Tobacco Carcinogenesis ▶ Tobacco-Related Cancers

Epithelial Tumorigenesis

▶ TP53 ▶ Transforming Growth Factor-Beta ▶ Xeroderma Pigmentosum

References Fearon ER (2011) Molecular genetics of colorectal cancer. Annu Rev Pathol 6:479–507 Frank SA (2004) Genetic predisposition to cancer – insights from population genetics. Nat Rev Genet 5:764–772 Halazonetis TD, Gorgoulis VG, Bartek J (2008) An oncogene-induced DNA damage model for cancer development. Science 7:1352–1355 McCaffrey LM, Macara IG (2011) Epithelial organization, cell polarity and tumorigenesis. Trends Cell Biol 21:727–735 Shan W, Yang G, Liu J (2009) The inﬂammatory network: bridging senescent stroma and epithelial tumorigenesis. Front Biosci 14:4044–4057

See Also (2012) AKT. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 115. doi:10.1007/978-3-642-16483-5_163 (2012) APC. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 234. doi:10.1007/978-3-642-16483-5_347 (2012) ARF. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 265. doi:10.1007/978-3-642-16483-5_383 (2012) Cancer stem cells. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 626. doi:10.1007/978-3-642-16483-5_815 (2012) Carcinoma. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 657. doi:10.1007/978-3-642-16483-5_848 (2012) Colorectal cancer. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 916. doi:10.1007/978-3-642-16483-5_1265 (2012) EGFR. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1211. doi:10.1007/978-3-642-16483-5_1828 (2012) Epigenetic changes. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1283. doi:10.1007/978-3-642-16483-5_1942 (2012) Epithelial cell. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 1291– 1292. doi:10.1007/978-3-642-16483-5_1958 (2012) Genomic instability. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1539. doi:10.1007/978-3-642-16483-5_2391 (2012) Helicobacter pylori. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1639. doi:10.1007/978-3-642-16483-5_2606

Epithelial-to-Mesenchymal Transition (2012) Loss of heterozygosity. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 2075–2076. doi:10.1007/978-3642-16483-5_3415 (2012) MSH2-6. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2383. doi:10.1007/978-3-642-16483-5_3860 (2012) MYC family. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2426. doi:10.1007/978-3-642-16483-5_3922 (2012) Obesity. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2595. doi:10.1007/978-3-642-16483-5_4185 (2012) P53. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2747. doi:10.1007/978-3-642-16483-5_4331 (2012) Sarcoma. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3335. doi:10.1007/978-3-642-16483-5_5161 (2012) Slug. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3439. doi:10.1007/978-3-642-16483-5_5354 (2012) Tobacco. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 3716– 3717. doi:10.1007/978-3-642-16483-5_5844 (2012) Wnt. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3953. doi:10.1007/978-3-642-16483-5_6255 Knowlton CA, Mackay MK, Huth BJ, Roedel C (2013) Hereditary nonpolyposis colorectal cancer (HNPCC). In: Brady LW, Yaeger TE (eds) Encyclopedia of radiation oncology. Springer, Berlin/Heidelberg, p 312. doi:10.1007/978-3-540-85516-3_481

Epithelial-to-Mesenchymal Transition Ayyappan K. Rajasekaran and Sigrid A. Langhans Nemours Center for Childhood Cancer Research, Alfred I duPont Hospital for Children, Wilmington, DE, USA

List of Abbreviations ECM EMT MET NF-kB SMA TGF-b1

Extracellular matrix Epithelial-mesenchymal transition Mesenchymal-epithelial transition Nuclear factor kB Smooth muscle actin Transforming growth factor-b1

1593

Definition Phenotypic alterations in which epithelial cells adopt characteristics of mesenchymal cells. Epithelial-mesenchymal transition (EMT) is a physiological process during normal development and a pathological process during cancer progression and ﬁbrosis.

Characteristics Epithelial cells line external surfaces and internal cavities of the body. A distinguishing characteristic of epithelial cells is the presence of junctional complexes, such as ▶ tight junctions, ▶ adherens junctions, and desmosomes, and segregation of plasma membrane into apical and basolateral domains. These features promote adhesion, restrict motility, facilitate intercellular communication, and permit individual cells to function as a cohesive unit. The phenotype of epithelial cells cultured in vitro or in tissues is often described as well differentiated (Fig. 1, left panel). Mesenchymal cells are spindle shaped with ﬁbroblast-like morphology, lack adhesiveness, and are highly motile. They do not have junctional complexes and specialization of the plasma membrane into apical and basolateral domains. When cells of epithelial origin show a mesenchymal phenotype under in vitro culture conditions or in tissues, they are often described as poorly differentiated (Fig. 1, right panel). The plasticity of epithelial cells enables them to convert between the epithelial and mesenchymal phenotypes. These phenotypic transformations are highly regulated by speciﬁc signaling events and molecules. Conversion of the epithelial cell to a mesenchymal phenotype is known as epithelial-mesenchymal transition (EMT) and vice versa as mesenchymal-epithelial transition (MET). EMT and MET occur during normal development as well as in cancer progression. EMT provides a mechanism for epithelial cells to overcome the physical constraints imposed upon them by intercellular junctions and to adopt a motile phenotype. The process was originally identiﬁed during speciﬁc stages of

1594

Epithelial-to-Mesenchymal Transition

Well-ifferentiated epithelial cells

Poorly-differentiated mesenchymal cells

TJ AJ

EMT

D MET

Tight junction (TJ), adherens junction (AJ), desmosome (D)

−

No junctions

Apical-basal polarity

−

Front-back polarity

Cortical actin

−

Stressfibers

Stationary

−

Motile and invasive

Cytokeratins, E-cadherin

−

Vimentin, smooth muscle actin, extracellular matrix deposition

Epithelial-to-Mesenchymal Transition, Fig. 1 Transitions between well-differentiated epithelial cells and poorly differentiated mesenchymal cells during EMT. Left panel, phase contrast microscopy and schematic diagram of well-differentiated polarized epithelial cells with characteristic apical-basolateral polarity and junctional

complexes. Right panel, cells that have undergone EMT display mesenchymal morphology with no junctions and are highly motile and invasive. The process of reversion of mesenchymal cells to an epithelial phenotype is called mesenchymal-epithelial transition (MET)

embryogenesis in which epithelial cells migrate and colonize various embryonic territories to form different organs. EMT is critical for the formation of ectoderm, mesoderm, and endoderm during gastrulation as well as for the differentiation of neural crest cells into neurons and glia of the peripheral nervous system. During embryogenesis EMT is spatially and temporally regulated in a subtle manner that is essential for normal organ development. Activation of the EMT program depends on the convergence of multiple cues that are both intrinsic to the cell and received from the microenvironment. EMT during cancer progression occurs in an aggressive and uncontrolled fashion and might facilitate the invasive and metastatic potentials of cancer cells. The phenotypic conversion of epithelial cells to mesenchymal cells involves a series

of events that includes dissolution of ▶ tight junctions, ▶ adherens junctions, and desmosomes, the suppression of molecules involved in restricting invasiveness and motility, and induction of factors that promote invasiveness and motility and gain of stem cell attributes. EMT can also occur as a partial transition when the phenotypic conversion is not complete. A characteristic feature of cells undergoing EMT in culture is the change in the organization of the actin cytoskeleton. In most cases stress ﬁbers are induced with a concomitant loss of the cortical actin ring. Although it is established that such morphological changes accompany EMT, the chronology of these events is still not deciphered. It is also not known whether all these changes are essential for induction of EMT and the metastatic potential of cancer cells.

Epithelial-to-Mesenchymal Transition

Mechanisms EMT is in part achieved by downregulation of epithelial-speciﬁc molecules and induction of proteins expressed in mesenchymal cells. One of the epithelial cell molecules extensively studied that change during EMT is the cell-cell adhesion molecule ▶ E-cadherin. During epithelial morphogenesis, ▶ E-cadherin regulates the establishment of ▶ adherens junctions, which form a continuous adhesive belt below the apical surface. The extracellular domain of ▶ E-cadherin mediates calcium-dependent homotypic interactions with ▶ E-cadherin molecules on adjacent cells, and the intracellular domain binds cytosolic catenins and links the ▶ E-cadherin complex to the actin cytoskeleton. A stable ▶ E-cadherin complex at the plasma membrane is essential for the cell-cell adhesion function of this protein. Several studies have shown that expression of ▶ E-cadherin is reduced during EMT, associated with the loss of junctional complexes and the induction of a mesenchymal phenotype of carcinoma cells. It is believed that the decrease in adhesive force following reduced expression of ▶ E-cadherin facilitates invasion and dispersion of carcinoma cells from the primary tumor mass. Methods to abolish ▶ E-cadherin function promote epithelial cell invasion into a variety of substrates, as determined by a number of in vitro and in vivo experimental systems. Loss or reduced expression of ▶ E-cadherin is also accompanied by expression of mesenchymal markers such as vimentin, smooth muscle actin (SMA), g-actin, b-ﬁlamin, and talin and extracellular matrix (ECM) components such as ﬁbronectin and collagen precursors. Upregulation of these proteins facilitates cytoskeletal remodeling and promotes cell motility (Table 1). The diverse molecular mechanisms mediated by growth factors and extracellular matrix proteins contribute to EMT. Growth factors such as epidermal growth factor (EGF), hepatocyte growth factor (HGF), or insulin growth factor II (IGF-II) promote signaling cascades through their cognate receptor tyrosine kinases, which in turn signal through various downstream effector molecules such as Ras, Src, phosphatidylinositol-3-

1595 Epithelial-to-Mesenchymal Transition, Table 1 Markers of EMT Epithelialmesenchymal markers

Signal transduction molecules/ pathways

Transcriptional regulators

Epithelialmesenchymal markers

Increased abundance/activity N-Cadherin Vimentin Fibronectin Smooth muscle actin g-Actin b-Filamin Talin Collagen precursors MMPs Stress ﬁbers Epidermal growth factor (EGF) Fibroblast growth factor (FGF) Hepatocyte growth factor (HGF) Insulin growth factor (IGF) Platelet-derived growth factor (PDGF) Transforming growth factor (TGF-b) Ras Src PI3K Wnt Notch Hedgehog GSK-3b MAPK TNFa NFkB Smurf-1 miR-138 miR-200 Snail Slug ZEB-1 ZEB-2 TCF/LEF Smads Twist1/2 E12/E47 DNA methylation Histone acetylation Decreased abundance/activity E-cadherin Cytokeratin Claudin Occludin Desmoplakin Desmoglein

kinase (PI3K), and MAPK leading to EMT. In addition, signaling pathways essential for stem cell function during development such as the Wnt, ▶ Notch, and ▶ Hedgehog signaling

1596

pathways are activated during EMT. The role of ▶ Wnt signaling has been well established during normal development as well as in EMT. Binding of the soluble ligand Wnt to its receptor frizzled inhibits b-catenin degradation and facilitates its nuclear translocation together with the TCF/LEF transcription factors to activate transcription of target genes such as cyclin D1 and Myc. The transcriptional activity of b-catenin is increased in a wide variety of cancers as well as in growth factor-induced EMT of cultured cells. A key molecule involved in the induction of EMT and extensively studied is the ▶ transforming growth factor-b1 (TGF-b1). The TGF-b growth factor superfamily comprises TGF-bs, bone morphogenetic proteins (BMPs), activins, and other related proteins. TGF-b induces EMT in epithelia either through transcriptional- or transcription-independent mechanisms. Cooperation between TGF-b and Ras/Raf/MEK/MAPK signaling is involved in the induction and maintenance of EMT. TGF-b has been shown to stimulate ERK1/2 activity in cell culture models of EMT that is required for the disassembly of junctional complexes and the induction of motility. TGF-b also activates phosphatidylinositol-3-kinase (PI3K) in a RhoAdependent manner, which has been implicated in the disassembly of ▶ tight junctions. In keratinocytes and several epithelial cell types, TGF-b treatment activates the ▶ Notch pathway by inducing the ▶ Notch ligand jagged1 and the ▶ Notch target genes TLE3, HEY1, HEY2, and HES1 at the onset of EMT. TGF-b, in cooperation with oncogenic Ras, induces EMT by the activation of the transcription factor ▶ nuclear factor k B (NF-kB). Constitutive activation of NF-kB induces EMT and metastasis, whereas inhibition of NF-kB by inhibitory IkBa suppresses EMT and metastasis in a breast tumor model. Although in different cell types TGF-b induces various signaling pathways, these signals subsequently target ▶ E-cadherin expression and the disassembly of epithelial junctional complexes to induce EMT. For example, the TGF-b type I receptor is localized to ▶ tight junctions through the ▶ tight

Epithelial-to-Mesenchymal Transition

junction protein occludin allowing for efﬁcient TGF-b-dependent dissolution of ▶ tight junctions during EMT. The epithelial polarity protein Par-6 interacts with the TGF-b type I receptor and TGF-b binding initiates Par-6 phosphorylation and activation of the E3-ubiquitin ligase, Smurf1. Activated Smurf-1 promotes degradation of local RhoA resulting in ▶ tight junction dissociation, inhibition of cell adhesion, and transition to a mesenchymal phenotype. The signaling cascades described above induce two major types of transcriptional regulators that mediate EMT, zinc ﬁnger (Snail, Slug, ZEB-1, ZEB-2) and basic helix-loop-helix (Twist, E12/E47) proteins. The transcription suppressors SNAI1 (Snail) and SNAI2 (Slug) play a central role in the induction of EMT. These zinc-ﬁnger proteins recognize E-box elements in the cognate target promoters, and SNAI1 represses the transcription of the ▶ E-cadherin gene during EMT as well as embryonic development. Factors that regulate SNAI1 by phosphorylation, subcellular localization, and transcription have been well described in development and EMT. While phosphorylation of SNAI1 in the two GSK3b phosphorylation consensus motifs targets it for export from the nucleus (motif 2) and ubiquitinylation and degradation (motif 1), phosphorylation of SNAI1 at Ser246 by p21-activated kinase (PAK1) results in its accumulation in the nucleus and induction of EMT. LIV-1, an estrogen-regulated member of the LZT subfamily of zinc transporters, is activated by STAT3, which is essential for nuclear localization of SNAI1 and suppression of ▶ E-cadherin expression during gastrulation in zebraﬁsh embryos. Further, SNAI1 expression is transcriptionally suppressed by metastasisassociated gene 3 (MTA3), a subunit of the Mi-2/NuRD transcriptional corepressor, thereby establishing a mechanistic link between estrogen receptor status and invasive growth of breast cancers. While there is great deal of knowledge about SNAI1 regulation, much less is known about SNAI2. It has been shown that SNAI2 suppresses ▶ E-cadherin expression when ectopically expressed in well-differentiated epithelial cells.

Epithelial-to-Mesenchymal Transition

HGF and FGF induce SNAI2 to suppress desmoplakin and desmoglein, thereby destabilizing desmosomes. The SMAD-interacting repressors SIP-1/ZEB2 and dEF1/ZEB1 that can be induced by TGF-b bind to the ▶ E-cadherin promoter to suppress its transcription. The basic helix-loophelix transcription factors involved in the induction of EMT are E12/E47 (E2A gene product) and Twist, both of which have been shown to repress ▶ E-cadherin expression and induce EMT. The mechanisms by which these factors suppress ▶ Ecadherin expression are not well established but recent work emphasized the importance of microRNAs and epigenetic factors. Clinical Relevance Although EMT represents a fundamentally important process for tumor dissemination and is widely believed to be an essential event involved in cancer metastasis, there are several lines of evidence to suggest that many invasive and metastatic carcinomas have not undergone a complete transition to a mesenchymal phenotype. Many advanced carcinomas of prostate, breast, squamous cell carcinomas derived from a variety of origins, including the esophagus, oral epithelium, lung, cervix, and salivary neoplasms, possess molecular and morphological characteristics of welldifferentiated epithelial cells, with the presence of epithelial junctions and apical-basolateral plasma membrane asymmetry. High ▶ E-cadherin expression was also observed in a wide variety of carcinomas and ▶ E-cadherin levels did not correlate with invasiveness and metastasis. These results are consistent with the idea that complete EMT might not be necessary for cancer cell metastasis or that cancer cells redifferentiate to an epithelial phenotype following metastasis. While EMT is well established in cultured cells, there is little evidence for EMT in vivo and if EMT occurs it is not known at what stage of tumor progression. There are several possibilities by which cancer cells could spread without undergoing complete EMT: (1) incomplete EMT by which epithelial cells partially convert to a mesenchymal phenotype acquiring invasive and metastatic

1597

potential, (2) cohort migration in which welldifferentiated epithelial cells migrate as a cluster and cause metastasis, and (3) reversion of poorly differentiated cells to a well-differentiated phenotype by mesenchymal-epithelial transition (MET) at the site of metastasis. These diverse mechanisms might be regulated by the tumor microenvironment and/or signaling pathways distinct from the molecular machinery of EMT. Thus, there are several mechanisms by which cancer cells could metastasize and EMT may represent one of the global changes associated with malignant transformation of epithelial cells. Recognizing EMT as a fundamentally important process for tumor dissemination together with the increasing knowledge about the molecular pathways leading to EMT may offer new targets for therapeutic intervention. Indeed, inhibitors of the TGF-b, ERK1/2, and PI3K/Akt pathways have shown encouraging results in the suppression of tumor progression. Further understanding of the molecular requirements of EMT will allow for more effective approaches for future therapeutic intervention.

Glossary Actin cytoskeleton The actin cytoskeleton is a dynamic structure of actin bundles and networks in the cytoplasm that provides a framework to maintain cell shape, protects the cell, and enables cell locomotion. It also plays an important role in intracellular transport. Epigenetics Mechanisms that impose a cellular phenotype without a change in its nucleotide sequence and largely achieved by covalent modiﬁcation of DNA and histone proteins through methylation and acetylation. Epithelial cell Epithelial cells line and protect both the outside and the inside cavities and lumen of the body. They regulate selective permeability and transcellular transport between the compartments they separate and are involved in secretion absorption and sensation detection. Microenvironment Biophysical and biochemical factors in the immediate vicinity of a cell that directly or indirectly affect the behavior of

1598

a cell. The microenvironment is composed of extracellular matrix homotypic and heterotypic cells, soluble factors including cytokines, hormones and other bioactive agents, and mechanical forces. Stem cell Undifferentiated cell capable of dividing and renewing itself for long time periods and with the potential to develop into different cell types.

Cross-References ▶ Adherens Junctions ▶ E-Cadherin ▶ Notch/Jagged Signaling ▶ Nuclear Factor-kB ▶ Tight Junction ▶ Wnt Signaling

References Chapman HA (2011) Epithelial-mesenchymal interactions in pulmonary ﬁbrosis. Annu Rev Physiol 73:413–435 De Craene B, Berx G (2013) Regulatory networks deﬁning EMT during cancer initiation and progression. Nat Rev Cancer 13:97–110 Kalluri R, Weinberg RA (2009) The basics of epithelialmesenchymal transition. J Clin Invest 119:1420–1428 Lamouille S, Xu J, Derynck R (2014) Molecular mechanisms of epithelial-mesenchymal transition. Nat Rev Mol Cell Biol 15:178–196 Lindsey S, Langhans SA (2014) Crosstalk of oncogenic signaling pathways during epithelial-mesenchymal transition. Front Oncol 4:358 Moreno-Bueno G, Portillo F, Cano A (2008) Transcriptional regulation of cell polarity in EMT and cancer. Oncogene 27:6958–6969 Tam WL, Weinberg RA (2013) The epigenetics of epithelial-mesenchymal plasticity in cancer. Nat Med 19:1438–1449 Thiery JP, Acloque H, Huang RY, Nieto MA (2009) Epithelial-mesenchymal transitions in development and disease. Cell 139:871–890

Epitheliasin (TMPRSS 2) ▶ Serine Proteases (Type II) Spanning the Plasma Membrane

Epitheliasin (TMPRSS 2)

Epithelium Isabelle Gross INSERM U1113, Université de Strasbourg, Strasbourg, France

Definition The epithelium is one of the basic tissue types. It lines the cavities or covers surfaces of structures throughout the body and also forms many glands. Functions of epithelial cells include protection, secretion, selective absorption, transcellular transport, and detection of sensation. Epithelia are classiﬁed by the morphology of their cells (columnar, squamous, cuboidal, or transitional) and the number of layers they are composed of (simple, stratiﬁed). They have the following structural and functional characteristics: • Epithelia form continuous sheets of cells densely packed together through the presence of numerous specialized intercellular junctions. Tight junctions are speciﬁc to epithelia and contribute to their barrier function. Desmosomal and adherens junctions ensure mechanic stability through connections with cytokeratin intermediate ﬁlaments speciﬁc to epithelia and actin microﬁlaments of the cytoskeleton. • Epithelia rest on a basement membrane called the basal lamina that acts as a scaffold and connects them to the underlying connective tissue that contains the blood vessels required for nutrient delivery and waste product disposal. • Epithelial cells exhibit polarity, meaning that they have an asymmetric organization of the cell surface, intracellular organelles, and the cytoskeleton. The apical region is deﬁned as the area lying above the tight junctions and contains the apical membrane that faces the lumen or the outer surface. The basolateral region is the side that is below the tight junctions and contains the basolateral membrane that is in contact with the basal lamina.

Epothilone B Analogue

1599

Cross-References

epothilone B analogue ixabepilone is a semisynthetic analogue of the natural product epothilone B, made by replacing the lactone oxygen of epothilone B with a lactam (azoepothilone B) designed to overcome the metabolic instability of the natural product. Chemical name: [1S-[1R*,3R*(E),7R*, 10S*,11R*,16S*]]-7,11-Dihydroxy-8,8,10,12,16pentamethyl-3-[1-methyl-2-(2-methyl-4-thiazolyl) ethenyl]-17-oxa-4-azabicyclo[14.1.0] heptadecane5,9-dione Molecular formula: C27H42N2O5S M.W. (506.7 g/mole)

▶ Adherens Junctions

See Also (2012) Epithelial cell. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 1291–1292. doi:10.1007/978-3-642-16483-5_1958

Epo ▶ Erythropoietin

Characteristics

Epoetin ▶ Erythropoietin

Epothilone B Analogue Robert P. Whitehead Nevada Cancer Institute, Las Vegas, NV, USA

Synonyms Azaepothilone B; BMS-247550; Ixabepilone; NSC-710428

Definition The epothilone B analogue, ixabepilone, is the term used to denote one speciﬁc agent of a new class of anticancer drugs, the epothilones. The epothilones A and B are derived from fermentation of the myxobacteria Sorangium cellulosum. They have been found to have potent cytotoxic activity, which, like that of the taxanes, has been linked to the stabilization of cellular microtubules resulting in the blocking of cell division at the ▶ G2/M Transition portion of the cell cycle. The

Introduction The epothilones A and B are a new class of anticancer agents isolated in the mid-1990s. They are in the macrolide class of drugs but have a mechanism of action similar to the taxanes. ▶ Paclitaxel, the ﬁrst taxane to be widely used, was found in 1971 in a screening assay for antitumor agents. It is a complex diterpene compound. Its cytotoxic activity derives from its binding to cellular microtubules stabilizing them and preventing the dynamic growth and shrinkage that occurs during normal cellular processes. In dividing cells, this leads to mitotic arrest and cell death by causing a block at the transition between G2 and M phases of the cell cycle. Paclitaxel has been found to have important clinical activity against breast, ovarian, lung, and head and neck cancers. However, some tumors such as colorectal cancer have innate resistance to this agent. Other tumors develop resistance to paclitaxel through the multidrug resistance mechanism in which ▶ P-glycoprotein removes cytotoxic agents from the tumor cell or by genetic mutations leading to altered tubulin protein. Side effects such as ▶ neutropenia and ▶ peripheral neuropathy can limit its activity and usefulness in the clinic. Its low solubility requires that paclitaxel be administered in a Cremophor vehicle which itself can induce hypersensitivity reactions. Modiﬁcations to paclitaxel to improve its solubility or reduce side effects are difﬁcult because of its complex ring structure. Therefore,

1600

when the epothilones A and B were found to have a mechanism of action similar to the taxanes, it was hoped that they might lead to more effective anticancer agents. It was found that the epothilones did show important cytotoxic activity against tumor cells when tested in in vitro cell assays, but when tested in in vivo models of cancer, only modest antitumor activity was present. This was found to be due to poor metabolic stability and other unfavorable characteristics. To overcome these problems, multiple semisynthetic analogues of the epothilones were made and tested. One of these analogues, BMS-247550, epothilone B analogue, or ixabepilone, was found to be the most effective epothilone in a variety of laboratory assays and was also active in paclitaxel-resistant tumor models. It has now been tested in the clinic and has been found to have activity in a variety of clinical cancers. Preclinical Testing Ixabepilone has shown potent cytotoxic activity when tested against a broad panel of tumor cell lines in vitro, including human P-glycoprotein family; colon carcinoma; breast, prostate, and lung cancers; ▶ squamous cell carcinoma; human leukemia; and mouse lung carcinoma. The concentration of drug at which 50% of the tested cells were killed (IC50) ranged from 1.4 to 34.5 nM. Included in this panel were cell lines resistant to paclitaxel by either multidrug resistance (MDR) due to P-glycoprotein overexpression or due to mutations of b-tubulin, the two most common mechanisms of resistance to paclitaxel. Ixabepilone is more potent than paclitaxel in causing tubulin polymerization and has similar activity to that of the parental compounds, epothilones A and B. Ixabepilone maintained its activity in whole animal systems. Using paclitaxel-sensitive human ovarian or colon tumor cell lines in the nude mouse models, ixabepilone produces comparable log cell kill and tumor growth delay as paclitaxel, while in nude mouse or rat tumor models using paclitaxel-resistant human ovarian, breast, colon, or ▶ pancreatic cancer cell lines, ixabepilone produced much greater tumor log cell kill and tumor growth delay than did paclitaxel. In contrast to

Epothilone B Analogue

paclitaxel which is usually ineffective when given orally, ixabepilone was also active in paclitaxelresistant human ovarian cancer and paclitaxelsensitive human colon cancer nude mouse models when given by the oral route. Phase I Studies in Cancer Patients In animal studies, the main toxicities of ixabepilone were related to the gastrointestinal tract, peripheral neuropathy, and bone marrow toxicity. Similar to the taxanes, when tested in cancer patients, ixabepilone required the same solvent for i.v. use, a Cremophor-based formulation (ethanol plus polyoxyethylated castor oil) which can lead to hypersensitivity reactions. After this type of reaction occurred in a patient, subsequent patients were prophylactically treated with histamine-1 (H1) and histamine-2 (H2) blockers. Schedules tested included an every 21-day cycle, weekly administration, a daily times 5 every 21-day cycle, and a daily times 3 every 21-day cycle. In each of these schedules, the drug was administered by the intravenous route. All of the phase I trials showed antitumor responses. These occurred in patients with breast, non-small cell lung, and ovarian cancers and melanoma. Some of these patients had previous treatment with paclitaxel or ▶ docetaxel. The 21-day cycle trials consisted of a 60-min i.v. infusion on day 1 repeated every 21 days. The dose-limiting toxicities for this schedule were neutropenia and sensory neuropathy. One study recommended a phase II dose of 50 mg/m2 and another trial of the same schedule suggested 40 mg/m2 as the phase 2 dose. Most subsequent trials have used the lower dose. Other toxicities commonly seen with this schedule included fatigue, arthralgias, myalgias, and vomiting. With treatments using weekly infusions or daily times ﬁve or daily times three infusions, maximum tolerated doses were lower. Neutropenia, sensory neuropathy, fatigue, and hypersensitivity reactions were seen with the weekly schedules. With the daily times ﬁve or three schedules, dose-limiting toxicity was neutropenia with fewer hypersensitivity reactions and less severe neurotoxicity.

Epothilone B Analogue

In patients given a 1-h infusion of ixabepilone, there was found to be bundling of microtubules in the peripheral blood mononuclear cells and this correlated with the plasma area under the curve, the concentration of drug measured in the blood multiplied by the time it is present. Similar microtubule bundle formation was seen in breast tumor cells obtained from a chest wall mass in a patient who showed a partial response after receiving drug on the 1-h infusion schedule. This patient was taxane refractory and the tumor expressed multidrug resistance protein. Cell death occurred in these tumor cells 23 h after the peak formation of microtubule bundles. Phase II Studies in Cancer Patients There is data from some phase II trials of ixabepilone in cancer patients. With ixabepilone given by a 1- or 3-h infusion on an every 21-day schedule, responses have been seen in a modest number of gastric or ▶ breast cancer patients previously treated with a taxane or ▶ non-small cell lung cancer patients who had previously received a platinum-based regimen. In breast cancer patients who were previously treated but had not received a taxane, a higher response rate of 34% was seen. A trial in colorectal cancer patients who had previously received an ▶ irinotecan-based regimen did not show any responses. Multicenter phase II trials in the Southwest Oncology Group of ixabepilone in previously untreated patients with advanced pancreatic cancer or chemotherapy-naive patients with hormone-refractory prostate cancer have shown encouraging results that suggest that further testing is warranted. Summary and Future Outlook The epothilone B analogue ixabepilone is a cancer therapeutic agent with a mechanism of action similar to the taxanes, stabilization of cellular microtubules leading to mitotic arrest and cell death. However, it demonstrates antitumor effects against both taxane-sensitive and taxane-resistant tumors and is clinically active against a broad spectrum of tumor types. Further testing as a single agent or in combinations in previously untreated patients is needed. Phase III trials in

1601

which it is compared to standard regimens will further deﬁne its role in cancer treatment. Ixabepilone is felt to be an important new anticancer agent that may surpass the taxanes in usefulness.

Cross-References ▶ Breast Cancer ▶ Docetaxel ▶ Gastric Cancer ▶ G2/M Transition ▶ Irinotecan ▶ Log-Kill Hypothesis ▶ Lung Cancer ▶ Neutropenia ▶ Non-Small-Cell Lung Cancer ▶ Paclitaxel ▶ Pancreatic Cancer ▶ Pancreatic Cancer Basic and Parameters ▶ Peripheral Neuropathy ▶ P-Glycoprotein ▶ Prostate Cancer ▶ Prostate Cancer Clinical Oncology ▶ Squamous Cell Carcinoma

Clinical

References Bollag DM, McQueney PA, Zhu J et al (1995) Epothilones, a new class of microtubule-stabilizing agents with a taxol-like mechanism of action. Cancer Res 55:325–2333 Goodin S, Kane MP, Rubin EH (2004) Epothilones: mechanism of action and biologic activity. J Clin Oncol 22:2015–2025 Lee FYF, Borzilleri R, Fairchild CR et al (2001) BMS-247550: a novel epothilone analog with a mode of action similar to paclitaxel but possessing superior antitumor efﬁcacy. Clin Cancer Res 7:1429–1437 McDaid HM, Mani S, Shen HJ et al (2002) Validation of the pharmacodynamics of BMS-247550, an analogue of epothilone B during a phase I clinical study. Clin Cancer Res 8:2035–2043 Whitehead RP, McCoy S, Rivkin SE et al (2006) A phase II trial of epothilone B analogue BMS-247550 (NSC #710428) ixabepilone, in patients with advanced pancreas cancer: a Southwest Oncology Group study. Invest New Drugs 24:515–520

1602

Epothilones Definition A group of microtubule-targeting agents, like taxanes, and vinca alkaloids. Epothilones were originally identiﬁed as metabolites produced by the common soil myxobacterium Sorangium cellulosum. They were found initially to have a narrow antifungal spectrum, but they also were found too toxic for use as an antifungal. Subsequently, their anticancer properties were detected. They are important and powerful options in the management of breast cancer and prostate cancer. The epothilones are a new class of cytotoxic molecules, including epothilone A, epothilone B, and epothilone D, identiﬁed as potential chemotherapy drugs. Early studies in cancer cell lines and in human cancer patients indicate superior efﬁcacy to the taxanes. Their mechanism of action is similar to that of the taxanes, but their chemical structure is simpler and they are more soluble in water. Although taxane-based therapy has been used successfully, its effectiveness is often compromised by the emergence of ▶ drug resistance. Efforts to overcome drug resistance have led to the discovery of several novel antimicrotubule agents, including the epothilones. Epothilones exhibit broad antitumor activity similar to that of the taxanes, but they are less sensitive to known resistance mechanisms. The ongoing development of microtubule-targeting agents provides new strategies for overcoming taxane resistance and may improve clinical efﬁcacy and patient outcomes.

Cross-References ▶ Drug Resistance

See Also (2012) Microtubule. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2308. doi:10.1007/978-3-642-16483-5_3734

Epothilones (2012) Taxanes. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 3614–3615. doi:10.1007/978-3-642-16483-5_6648 (2012) Vinca alkaloids. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3908. doi:10.1007/978-3-642-16483-5_6187

Epstein-Barr Virus Evelyne Manet CIRI-International Center for Infectiology Research, INSERM U1111, Université Lyon 1, ENS de Lyon, Lyon, France

Synonyms EBV; HHV4; Human herpesvirus 4

Definition Epstein-Barr virus (EBV) was the ﬁrst virus isolated from a human tumor, Burkitt lymphoma (BL). EBV is a lymphotropic g-herpesvirus widely spread in the human population: 90–95% of adults have antibodies against the virus. In the majority of cases, the primary infection occurs within the ﬁrst 3 years of life and is asymptomatic. When EBV infection occurs later in life, usually during adolescence, it results in the symptomatic illness known as infectious mononucleosis (IM). Infected individuals carry the virus all their life, in a very low number of lymphoid B cells (probably resting B cells) in their peripheral blood and lymphatic organs. Intermittent viral shedding occurs into the saliva, due to viral replication in the oropharyngeal lymphoid or epithelial tissues: saliva is the main transmission route of the virus. Since its ﬁrst discovery in 1964 in Burkitt lymphoma tumor, EBV has been found to be associated with several other human malignancies including the undifferentiated ▶ nasopharyngeal carcinoma (NPC), ▶ Hodgkin disease, rare T-cell and natural killer (NK)-cell lymphomas, gastric carcinomas, and B- and T-cell lymphomas in immunocompromised individuals.

Epstein-Barr Virus

1603

A characteristic unique to EBV is its capacity to induce the indeﬁnite proliferation or immortalization of quiescent B lymphocytes, upon their infection in vitro.

Characteristics In vitro Immortalization of Primary B Cells by EBV: The Growth Transcription Program In vitro, infection of B cells by EBV is not productive but results in the outgrowth of latently infected lymphoblastoid cell lines (LCL) (Fig. 1). Such cell lines can also be obtained by culture of peripheral blood lymphocytes (PBL) from naturally infected individuals. The phenotype of LCLs (i.e., morphology and cell surface markers) is very similar to that of antigen-

activated B cells, and a limited set of viral gene products is expressed: six nuclear proteins, the Epstein-Barr nuclear antigens (EBNA-1, EBNA-2, EBNA-3A (or EBNA-3), EBNA-3B (or EBNA-4), EBNA-3C (or EBNA-6), and EBNA-LP (or EBNA-5)), three integral membrane proteins (LMP-1, LMP-2A (or TP-1), and LMP-2B (or TP-2), two small non-polyadenylated nuclear RNAs (EBER-1 and EBER-2). This expression proﬁle is referred to as the growth transcription program or latency III (Table 1) and over 25 miRNAs. In such immortalized cell lines, the viral DNA (a 172 kpb double-stranded DNA molecule) is maintained in the nucleus as multiple extrachromosomal copies of the viral episome. Among the nine proteins expressed in latency III, EBNA-1, EBNA-2, EBNA-3A, EBNA-3C, and EBNA-LP and EBV-infected

Primary B cells

CD21 LMPs EBNAs

G0 EBV Quiescent B-cell

Activated B-cell

Immortalization

PBL of EBV-infected individuals (or IM patients) Epstein-Barr Virus, Fig. 1 Immortalization of B cells by EBV. EBV infection of resting primary human B cells in vitro causes cell cycle entry of the infected cells, establishment of latent viral infection, and conversion of the cell culture into permanently growing lymphoblastoid cell lines (LCLs), which progress to a fully immortalized state. Such LCLs can also be obtained by culture of PBL from individuals infected by EBV (mostly IM patients). The ﬁrst step of infection is the interaction of the EBV gp350 glycoprotein with its receptor, the CD21. This induces a decondensation of the chromatin, a prerequisite for expression of a subset of the viral genes that cooperate

to induce B-cell proliferation and the expression of several cellular activation markers and adhesion molecules. These latency genes code for three nuclear proteins, the EpsteinBarr virus nuclear antigens (EBNA1, 2, 3A, 3B, 3C, LP), three integral membrane proteins (LMP1, LMP2A, LMP2B), plus two non-polyadenylated small nuclear RNAs (EBERs). The EBV genome is maintained in these proliferating cells as multiple extrachromosomal episomes and is replicated by cellular proteins, concomitantly with the cellular genome, via a DNA replication origin called oriP

1604

Epstein-Barr Virus

Epstein-Barr Virus, Table 1 EBV gene latency transcription programs Transcription program Growth

Type of latency III

Default EBNA 1 only Latency

II I 0

EBV genes expressed EBNA1, 2, 3a, 3b, 3c, LP, LMP1, LMP2a and LMP2b EBNA1, LMP1 and LMP2a EBNA1 None

Occurrence PTLD, immunoblastic lymphoma (HIV) HD, NPC, gastric carcinoma BL, PEL Memory B cells in peripheral blood

PTLD posttransplant lymphoma disease, HD Hodgkin disease, NPC nasopharyngeal carcinoma, BL Burkitt lymphoma, PEL primary effusion lymphoma

LMP-1 are essential for efﬁcient transformation/ immortalization of B lymphocytes in vitro. These proteins are thought to cooperate for the initiation and maintenance of B-cell proliferation in vitro. EBNA-1 is a sequence-speciﬁc DNA-binding protein which binds to the EBV origin of replication (OriP). This interaction is required for viral DNA replication and for equilibrated distribution of the EBV episomes to the daughter cells during cell division. EBNA-1 may have other roles in EBV-induced oncogenesis as EBNA-1 transgenic mice display an increased incidence of B-cell lymphoma. EBNA-2 is a transcriptional activator which regulates the expression of all EBV genes expressed in latency III as well as certain cellular genes including CD21, CD23, and cfgr. EBNA-2 does not bind DNA directly but is recruited to EBNA2-responsive elements by the cellular sequence-speciﬁc DNA-binding factor RBP-Jk (also called CBF-1). As RBP-Jk is part of the Notch signaling pathway, EBNA-2 may mimic part of Notch signal transduction (▶ NOTCH/ JAGGED signaling in neoplasia). Although the exact function of EBNA-LP is still unknown, this nuclear factor has been found to cooperate with EBNA-2 for the transcriptional activation of the LMP-1 gene. Furthermore, co-expression of EBNA-2 and EBNA-LP in primary resting B cells previously activated through binding of the EBV glycoprotein gp350 to the EBV receptor CD21 induces cyclin D2 expression and drives resting B lymphocytes into the G1 phase of the cell cycle. EBNA-3A, EBNA-3B, and EBNA-3C are related proteins but only EBNA-3A and EBNA-

3C are essential for B-cell immortalization by EBV in vitro. However, these proteins have at least one common function: repression of EBNA2-activated transcription by directly contacting RBP-Jk and inhibiting its binding to DNA. Furthermore, EBNA-3C is able to cooperate with activated (Ha)-Ras (RAS), to induce the proliferation of primary rat ﬁbroblasts. LMP-1 (▶ Epstein–Barr virus latent membrane protein 1) is an integral membrane protein with a very short 24 aa N-terminal cytoplasmic domain, six membrane spanning hydrophobic segments, and a 200 aa cytoplasmic C-terminal domain. LMP-1 transforms rodent ﬁbroblast cell lines and Rat-1 cells expressing LMP-1 are tumorigenic in nude mice. LMP-1 acts as a constitutively activated member of the tumor necrosis factor receptor (TNRF) superfamily. It activates NF-kB transcription factor activity through a pathway that involves the recruitment of TNF-RI receptor-associated factors (TRAFs). It also induces ▶ AP-1 transcription factor activity via triggering of the c-jun N-terminal kinase (JNK). The STAT-1 transcription factor has also been shown to be a target of LMP1 which induces STAT-1 phosphorylation and thus its subsequent transfer to the nucleus. LMP-2A is an integral membrane protein with a 119 aa hydrophilic N-terminal cytoplasmic tail which contains immunoreceptor tyrosine-based activation motifs (ITAMs) followed by 12 transmembrane domains and a 27 aa hydrophilic C-terminus. LMP-2B differs from LMP-2A by the lack of the N-terminal cytoplasmic domain. Although dispensable for in vitro immortalization of B lymphocytes, LMP-2A has an important role in the biology of EBV in vivo (see below) in

Epstein-Barr Virus

mimicking the presence of a B-cell receptor (BCR) and providing important survival signals for B cells. EBV Biological Cycle In Vivo Although understanding of EBV infection biology in vivo is still rudimentary, it is believed to mimic the normal B-cell response to environmental antigen. EBV transits the epithelium and infects naive B cells in the underlying tissue. Expression of the latency genes (transcription growth program: Table 1) causes the cell to become activated, proliferate, and migrate to the follicle. After this initial clonal expansion, some EBV-infected cells undergo germinal center reactions. EBV transcription is then limited to EBNA1, LMP-1, and LMP-2 expression (the default transcription program: Table 1). These germinal center B cells will then differentiate into memory B cells in which EBV expression is turned off (latency program: Table 1). These cells constitute the long-term reservoir of EBV. The lytic viral cycle can be reactivated in these cells by signals that cause B cells to differentiate into antibodysecreting plasma cells (through antigen stimulation) and migrate to the mucosal epithelium allowing the release of viral particles in the saliva, the main route for transmission of the virus between individuals. Although EBV is considered to be a B-lymphotropic virus, it can also infect epithelial cells in vitro and is found in several EBV-associated carcinomas in vivo. A role of the epithelial cells of the oropharynx in the ampliﬁcation of virus production in vivo has long been suspected but not yet demonstrated. Viral Productive Cycle In vitro, the lytic productive cycle can be induced in EBV-infected cell lines (either LCL or cell lines established from Burkitt lymphoma) by treatments of the cells with various agents such as the phorbol ester TPA, butyric acid (BA) or cross-linking of surface immunoglobulin, etc. A key mediator of the entry into the productive cycle is the viralencoded transcription factor BZLF1 (also called EB1, Zta, and Zebra) which activates both transcription of all the EBV early genes and DNA

1605

Proliferating EBV-infected B-cells TPA BA/TSA TGF-β α-lgG BZLF1 and BRLF1

E Early genes

Viral DNA replication

Orilyt

Late genes

Production of viral particles

Epstein-Barr Virus, Fig. 2 The EBV replicative cycle. The EBV lytic replication program can be induced in vitro by treatment of proliferating EBV-infected B cells with various chemicals, by cross-linking of the surface immunoglobulins, or by expression of the EBV transcriptional activator EB1 (also called Zta or ZEBRA), product of the BZLF1 gene. In cooperation with another viral transcriptional factor (also called R or Rta), product of the BRLF1 gene, EB1 activates the expression of all the EBV early genes. EB1 also directly stimulates viral DNA replication which is dependent on viral proteins (a DNA-polymerase, a polymerase processivity factor, a single-stranded DNA-binding protein, a primase, and a helicase/primase associated protein). This viral DNA replication is initiated at replication origins, called ori lyt, which are different from the one used during latency. This lytic replication program leads to ampliﬁcation of the viral genome, synthesis of structural viral proteins, and the assembly of infectious virus particles

replication from the replication origins (Orilyt) active during the lytic cycle (Fig. 2). Clinical Relevance EBV is associated with several human malignancies in both immunocompromised and immunocompetent individuals.

1606

In immunodepressed individuals – posttransplant patients or AIDS (acquired immunodeﬁciency syndrome) patients – EBV is probably directly involved in the appearance of immunoblastic B lymphomas (in AIDS patients) or posttransplantation lymphoproliferative diseases (PTLDs) (in patients undergoing organ transplantation) due to the loss of normal cytotoxic T-cell surveillance. These lymphomas are monoclonals or polyclonals, and the cells usually express the full set of EBV genes found in LCLs proliferating in culture (latency III, Table 1). In immunocompetent individuals EBV is associated with several cancers. Endemic Burkitt lymphoma (BL) is found in certain parts of Africa and South America where malaria – which appears to act as a cofactor – is also endemic and affects mainly children from 7 to 9 years old. In these regions, BL is associated with EBV in more than 90% of cases. In lower-incidence regions, the association is only found in 20–30% of cases. BL is a monoclonal tumor characterized by a translocation of the c-myc gene (▶ Myc oncogene) to one of the immunoglobulin loci which results in an altered regulation of c-myc. The expression of EBV in the tumor cells is limited to EBNA-1, the EBER RNAs, plus several miRNAs. This proﬁle of expression is deﬁned as latency I (Table 1). Burkitt lymphoma cell lines can be readily established from tumor biopsies. On the contrary to LCLs, these cells are tumorigenic in nude mice. However, after several passages of these cells in culture, the expression proﬁle of the EBV genes has been shown to derive towards a latency III proﬁle. In ▶ Hodgkin disease, EBV is present in the Reed-Sternberg cells (▶ Hodgkin and Reed/ Sternberg Cell), in about 40% of cases, mostly of the mixed cellularity type. Expression of EBV in Hodgkin disease is characteristic of latency II (Table 1). EBV is also found associated with rare but speciﬁc types of nasal T-cell lymphomas more common in Southeast Asian populations and also natural killer (NK)-cell (▶ Natural Killer Cell Activation) lymphomas. These types of lymphomas seem to arise either after acute primary

Epstein-Barr Virus

infection or in some cases of chronic active EBV infection. The EBV expression proﬁles in these tumors are characteristic of latency I/II (Table 1). EBV is also associated with a variety of carcinoma particularly the undifferentiated nasopharyngeal carcinoma (▶ nasopharyngeal carcinoma) (NPC). NPC is associated with EBV in almost 100% of cases worldwide and is particularly common in areas of China and Southeast Asia. Genetic disposition as well as environmental cofactors such as dietary components is thought to be important in the etiology of NPC. EBV gene expression in NPC epithelial cells consists of EBNA-1, the EBER RNAs, and LMP-1, LMP-2A/-2B (in 65% of cases), plus the miRNAs. This proﬁle of expression is similar to latency II (Table 1). Several factors suggest that a reactivation of EBV (i.e., entry into the lytic cycle) precedes or accompanies the development of NPC. EBV is also found in about 10% of gastric adenocarcinomas (▶ gastric cancer) with a pattern of expression of EBV genes similar to that observed in NPC. The exact role of EBV in the development of these different tumors is not yet understood and both environmental and genetic cofactors also contribute. However, the fact that the EBV genome is present in the great majority of the cells in EBV-associated malignancies and the demonstration that the virus is present in the tumor cells at a very early stage argue for a causative role for EBV in these cancers.

Cross-References ▶ AP-1 ▶ Burkitt Lymphoma ▶ Epstein–Barr Virus Latent Membrane Protein 1 ▶ Gastric Cancer ▶ Hodgkin and Reed/Sternberg Cell ▶ Hodgkin Disease ▶ MYC Oncogene ▶ Nasopharyngeal Carcinoma ▶ Natural Killer Cell Activation ▶ Notch/Jagged Signaling ▶ RAS Genes

Epstein–Barr Virus Latent Membrane Protein 1

1607

References

Characteristics

Münz C (ed) (2015a) Epstein Barr virus volume 1. One herpes virus: many diseases. In: Current topics in microbiology and immunology, vol 390. Springer International Publishing Münz C (ed) (2015b) Epstein Barr virus volume 2. One herpes virus: many diseases. In: Current topics in microbiology and immunology, vol 391. Springer International Publishing Rickinson AB (2014) Co-infections, inﬂammation and oncogenesis: future directions for EBV research. Semin Cancer Biol 26:99–115 Thorley-Lawson DA, Hawkins JB, Tracy SI, Shapiro ME (2013) The pathogenesis of Epstein-Barr virus persistent infection. Curr Opin Virol 3(3):227–232

EBV belongs to the family of gamma herpes viruses; it infects humans and establishes a latent infection for the lifetime of the individual. Infection in childhood is usually asymptomatic, but in young adults can lead to infectious mononucleosis. Due to its association with ▶ B cell tumors as Burkitt lymphoma (BL), ▶ Hodgkin disease/ lymphoma (HL), posttransplant lymphoproliferative disorder, and epithelial tumors as nasopharyngeal carcinoma (NPC) and gastric carcinoma, EBV was the ﬁrst human tumor virus to be discovered and is now classiﬁed as a group 1 carcinogen by the WHO. In vivo, EBV does usually not replicate in B lymphocytes, but instead establishes a latent infection with deﬁned expression of the virus latent genes. In vitro, EBV infection leads to continuously proliferating lymphoblastoid cell lines (LCLs) and a different expression pattern of virus latent genes. The virus genes inﬂuence both viral and cellular transcription in the host cell. Among them LMP1 is essential for B cell transformation in vitro and behaves as a classical oncogene in rodent ﬁbroblast transformation assays. When expressed in the B cell compartment of transgenic mice, LMP1 induces the development of B cell lymphoma, whereas its expression in the murine epidermis results in hyperplasia. It is interesting to note that expression of LMP1 is variable among EBV-associated tumors, being always present in EBV-associated Hodgkin lymphoma, almost always absent in virus-positive Burkitt lymphoma, and variably present in NPC. Structurally, LMP1 is a 66 kDa integral membrane protein consisting of 386 amino acids. It can be divided into three domains: a short cytoplasmic amino-terminus of 24 amino acids, a transmembrane domain consisting of six transmembrane helices which oligomerize to form membrane patches, and a carboxy-terminal cytoplasmic domain of 200 amino acids. The cytoplasmic domain contains two carboxy-terminal activating regions (CTARs), also known as transformation effector sites (TES). The CTARs are critical for EBV’s transforming activity; while CTAR1/TES1

See Also (2012) Burkitt Lymphoma Cell Lines. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 575. doi:10.1007/978-3-642-164835_754 (2012) Lymphoblastoid Cell Lines. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, pp 2122-2123. doi:10.1007/978-3-642-164835_3454

Epstein–Barr Virus Latent Membrane Protein 1 Martina Vockerodt Department of Pediatrics I, Children’s Hospital, Georg-August University of Gottingen, Gottingen, Germany

Definition The latent membrane protein 1 (LMP1) of the ▶ Epstein-Barr virus (EBV) is an ▶ oncogene that is expressed during latent EBV infection. LMP1 is sufﬁcient for the transformation of rodent ﬁbroblast cells and is essential for efﬁcient B cell transformation. LMP1 mimics a constitutively active tumor necrosis factor (TNF) receptor and interacts with and deregulates the ▶ signal transduction network of the host cell leading to altered cell survival, differentiation, and phenotypic changes.

1608

Epstein–Barr Virus Latent Membrane Protein 1

LMP1

PI3K

Survival proliferation invasion

Akt

NH2 CTAR1

TRAF 1, 2, 3, 5

p38

231

ATF2

Target genes NFκB

AP1

351 CTAR2

JNK 386 TRADD

TRAF 3, 6

Epstein–Barr Virus Latent Membrane Protein 1, Fig. 1 Structure of LMP1 and its functional domains. The signaling domains recruit TRAFs and TRADD

which results in the activation of host cell signal transduction pathways and the regulation of host cell target genes

is essential for initial B cell transformation, CTAR2/TES2 mediates growth factor-like signals which are required for long-term outgrowth of EBV-infected cells. The region between CTAR1 and CTAR2, sometimes referred to as CTAR3, is not required for B cell ▶ immortalization.

inducing kinase (NIK) and the I-kB kinases (IKKs). This results in the activation of the canonical I-kB-dependent NF-kB pathway (involving p50-p65 heterodimers) and the noncanonical pathway, leading to the processing of p100 to generate p52-p65 heterodimers. While activation through the CTAR1 domain of LMP1 is mediated by the noncanonical NF-kB pathway, CTAR2 appears to activate the canonical pathway by utilizing TRAF6 and TAK1. CTAR2 is also important in activating the c-Jun N-terminal kinase (▶ JNK subfamily) leading to activation of the transcription factor ▶ AP-1. The phosphatidylinositol 3-kinase (▶ PI3K signaling) pathway is triggered through CTAR1 leading in epithelial cells to actin polymerization and cell ▶ motility. Other signaling pathways activated by LMP1 are the p38 ▶ MAP kinase pathway and the ▶ signal transducers and activators of transcription (STAT) pathway. The NF-kB pathway plays a key role in the activation of many genes and is essential for the transformation of B cells by EBV. NF-kB

Mechanisms Cell transformation requires activation of the host cell signaling machinery. LMP1 mediates this activation by recruiting molecules of the TNF receptor and ▶ toll-like receptor family (Fig. 1). CTAR1 interacts with the TNF receptorassociated factors (TRAF) 1, 2, 3, and 5 through the PxQxT-binding motif of LMP1. Recruitment of TRAF molecules to CTAR2 may be mediated indirectly through binding of the TNF receptorassociated death domain protein (TRADD), although much of CTAR2-mediated ▶ nuclear factor kappa B (NF-kB) activity has been shown to be TRADD independent (Fig. 1). These adaptor proteins subsequently recruit multiprotein complexes containing the NF-kB-

Erlotinib

mediates induction of antiapoptotic genes (▶ bcl2, bﬂ-1, A20, c-IAPs, c-FLIP) and downregulates proapoptotic genes as Bax (Bcl-2-associated protein x). LMP1 expression in B cells also results in the upregulation of activation markers and ▶ adhesion molecules. LMP1 also modulates the communication between EBV-infected cells and its cellular environment by upregulating the expression of a large number of cytokines (▶ interleukin-6, 10, TNF-a) and ▶ chemokines (RANTES, IP-10, interleukin-8). It is also involved in migration and invasion processes as it activates proangiogenic factors as ▶ vascular endothelial growth factor, ▶ matrix metalloproteinases, and modulators of the cytoskeleton. Although LMP1 shows no homology with any cellular protein, it functionally mimics an activated CD40 receptor, which is a costimulatory receptor required for B cell proliferation. Although LMP1 and CD40 both recruit TRAF molecules and regulate an overlapping pattern of signaling pathways and target genes, they have divergent roles in B cell development. For example, in CD40-deﬁcient LMP1 transgenic mice, LMP1, like CD40, can induce extrafollicular B cell differentiation, but in contrast to CD40, LMP1 leads to a defective germinal center reaction, characterized by splenomegaly and lymphadenopathy.

Cross-References ▶ Hodgkin Lymphoma, Clinical Oncology

1609

ER Stress ▶ Endoplasmic Reticulum Stress

ERBB ▶ Epidermal Growth Factor Receptor

ErbB-1 ▶ Epidermal Growth Factor Receptor

ERBB2 ▶ HER-2/neu

Erlotinib Bassel El-Rayes1,2, Shirish Gadgeel3, Shadan Ali3, Philip A. Philip3 and Fazlul H. Sarkar3 1 Department of Hematology and Medical Oncology, Emory University School of Medicine, Atlanta, GA, USA 2 Winship Cancer Institute of Emory University, Atlanta, GA, USA 3 Karmanos Cancer Institute, Wayne State University, Detroit, MI, USA

Synonyms

References

Tarceva

Young LS, Rickinson AB (2004) Epstein-Barr virus: 40 years on. Nat Rev Cancer 4:757–768

Definition

ER ▶ Estrogen Receptor

Erlotinib is a potent and selective ▶ epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor. Erlotinib is commercially available in tablets of 25, 100, and 150 mg formulations. Erlotinib is currently approved for use in previously treated non-small cell ▶ lung cancer and in

1610

frontline management of ▶ pancreatic cancer in combination with gemcitabine chemotherapy.

Characteristics Rationale for Targeting the EGFR. The EGFR is frequently dysregulated in epithelial cancers. Overexpression of EGFR can result in malignant transformation of cells. Patients whose tumors have overexpressed or dysregulated EGFR and/or ligand expression may have a worse prognosis. Activation of the EGFR initiates dimerization of the receptors leading to activation of the tyrosine kinase domain. The kinase in turn phosphorylates (Phosphorylation) and activates proteins in the signal transduction cascade promoting cell proliferation, angiogenesis, invasion, and survival. In preclinical models, erlotinib selectively inhibited EGFR tyrosine kinase activity in human cancer cell lines and resulted in inhibition of various tumor growths and induction of apoptosis. Erlotinib potentiated the activity of cytotoxic agents and radiation in cancer cell line as well as in animal models. Phase I Trials. The maximal tolerated dose of erlotinib was 150 mg once daily in phase I trial. The most commonly observed toxicities were ▶ acneiform rash and diarrhea. Experience in Non-Small Cell Lung Cancer (NSCLC). Three randomized trials evaluated the efﬁcacy of erlotinib in advanced NSCLC. In the ﬁrst trial, patients with previously treated advanced NSCLC were randomized to erlotinib or best supportive care. The results revealed a signiﬁcant improvement in median survival (6.7 vs. 4.7 months, p = 0.001) and survival at 1 year (31.2% vs. 21.5%) in favor of erlotinib. The median time to progression and response rates in the erlotinib and best supportive care arms were 9.7 vs. 8.0 weeks (p < 0.001) and 8.9 vs. 1.0%, respectively. The incidence of grade 3 and 4 rash and diarrhea in the erlotinib arm was 9% and 6%, respectively. A very important ﬁnding, in a multivariate analysis, was that nonsmokers and those with adenocarcinoma histology beneﬁted most from erlotinib. Based on this trial, erlotinib was approved in patients with previously treated NSCLC.

Erlotinib

Two trials compared conventional chemotherapy to the same chemotherapy and erlotinib in patients with newly diagnosed advanced NSCLC. The chemotherapy regimens evaluated were ▶ gemcitabine/carboplatin (TALENT) and ▶ paclitaxel/▶ cisplatin (TRIBUTE). No signiﬁcant difference was observed between patients receiving chemotherapy or erlotinib and chemotherapy with respect to objective response rate, survival, or time to progression. Therefore, erlotinib is not used in combination with chemotherapy to treat patients with NSCLC. Experience in Pancreatic Cancer. A randomized trial compared gemcitabine to gemcitabine and erlotinib in patients with advanced pancreatic cancer. The trial resulted in a signiﬁcant improvement in median survival (6.37 vs. 5.91 months, p = 0.01) and 1-year survival (24 vs. 17%) in favor of erlotinib. Signiﬁcant predictors for favorable outcome were performance status 0–1, locally advanced disease, and normal albumin. Treatment with erlotinib resulted in an improvement of progression-free survival (3.75 vs. 3.55 months, p = 0.009). The incidence of grade 3 and 4 toxicities was higher in the erlotinib arm with respect to rash (6 vs. 1%) and diarrhea (7 vs. 2%). The Food and Drug Administration (FDA) approved erlotinib in combination with gemcitabine for previously untreated advanced pancreatic cancer. A similar approval was granted in Europe for this particular indication. Trials in Other Tumor Types. Erlotinib has been evaluated in bile duct, gastric, ▶ esophageal, ▶ hepatocellular, ▶ colorectal (CRC), and head and neck cancers. In bile duct cancer, a multiinstitutional phase II trial evaluated single-agent erlotinib in 42 patients with advanced disease. The trial met its primary end point with a progression-free survival at 6 months of 17%. Three patients had a partial response. The Southwest Oncology Group (SWOG) performed a phase II trial of erlotinib in advanced gastric and gastroesophageal (GEJ) tumors. The response rate was 9%. All responses were observed in the GEJ patients. Interestingly, no responses were seen in the patients with gastric cancer. Philip et al. reported the results of 38 patients with advanced hepatocellular cancer treated with

Erlotinib

erlotinib. Forty-seven percent of the patients had received prior chemotherapy. The study met its primary end point with a 6-month progressionfree survival of 32%. Three patients had partial response. Erlotinib has been evaluated in CRC as a single agent and in combination with FOLFIRI, capecitabine, capecitabine/oxaliplatin, and FOLFOX. As a single agent, erlotinib was evaluated in patients with previously treated CRC. Thirty-nine percent of patients had stable disease and no responses were observed. The combination of FOLFIRI and erlotinib resulted in excessive toxicity, and the trial was discontinued. Erlotinib and capecitabine were well tolerated, and in the phase I trial, two of the nine patients with CRC had a partial response. The phase II trial is still ongoing. Two trials evaluated erlotinib and capecitabine/oxaliplatin in patients with previously treated CRC. The partial response rates and stable disease were 20–22% and 61–64%, respectively. There is still uncertainty whether erlotinib plays any role in CRC. Currently, research in CRC is focused on the use of monoclonal antibodies that target the EGFR. Erlotinib in combination with FOLFOX is currently in a phase II trial. Modest activity was observed with erlotinib in head and neck cancer patients with a response rate of 4% and disease stabilization rate of 38%. In conclusion, erlotinib has demonstrated promising activity in cholangiocarcinoma, hepatocellular, and GEJ tumors. The role of erlotinib in these diseases requires further randomized trials. Future Directions. Erlotinib has demonstrated a signiﬁcant but modest activity in a number of carcinomas. The future challenge is how to improve the activity of erlotinib. Two approaches are being evaluated in clinical trials. The ﬁrst is to select patients with a higher likelihood of beneﬁt from erlotinib. For example, patients with NSCLC and either activating mutations of the EGFR or no smoking history have demonstrated an increased beneﬁt from erlotinib therapy. Additional trials are evaluating the efﬁcacy of erlotinib in previously untreated patients whose tumors have EGFR mutations. On the other hand, a trial is evaluating the combination of erlotinib with chemotherapy in previously untreated patients with NSCLC and no smoking history. The results

1611 Erlotinib, Table 1 Ongoing trials of erlotinib in combination with targeted agents Agent Isoﬂavone Sorafenib Dasatinib Cetuximab Celecoxib RAD001

Target Akt/NF-kB VEGFR, PDGFR, Raf Src EGFR Cyclooxygenase 2 mTOR

Phase (disease) II (Pancreatic) I I I II (NSCLC) Phase I

E of these trials will determine whether erlotinib should be used in the frontline management of NSCLC in selected patient populations. The second approach is based on combining erlotinib with other targeted agents because of the redundancy of the signaling pathways and existence of independently activated survival pathways. Since the inhibition of the EGFR results in inhibition of angiogenesis, the combination of erlotinib and ▶ bevacizumab is being evaluated in a number of malignancies including hepatocellular, cholangiocarcinoma, NSCLC, and pancreatic cancer. The combinations of erlotinib with other agents targeting the signaling cascade are currently at different stages of development (Table 1).

Cross-References ▶ Achneiform Rash ▶ Bevacizumab ▶ Cisplatin ▶ Colorectal Cancer ▶ Epidermal Growth Factor Receptor ▶ Esophageal Cancer ▶ Gemcitabine ▶ Hepatocellular Carcinoma ▶ Lung Cancer ▶ Paclitaxel ▶ Pancreatic Cancer

References El-Rayes BF, LoRusso PM (2004) Targeting the epidermal growth factor receptor. Br J Cancer 91:418–424 Giaccone G (2005) Targeting HER1/EGFR in cancer therapy: experience with erlotinib. Future Oncol 1:449–460

1612

Erlotinib (Tarceva)

Mendelsohn J, Baselga J (2006) Epidermal growth factor receptor targeting in cancer. Semin Oncol 33:369–385 Philip PA, Mahoney MR, Allmer C et al (2005) Phase II study of Erlotinib (OSI-774) in patients with advanced hepatocellular cancer. J Clin Oncol 23:6657–6663 Philip PA, Mahoney MR, Allmer C et al (2006) Phase II study of erlotinib in patients with advanced biliary cancer. J Clin Oncol 24:3069–3074

Definition

See Also

Characteristics

(2012) Bile Duct. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 399. doi:10.1007/978-3-642-16483-5_616 (2012) Capecitabine. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 640. doi:10.1007/978-3-642-16483-5_828 (2012) FOLFIRI. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1440. doi:10.1007/978-3-642-16483-5_2232 (2012) FOLFOX. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1441. doi:10.1007/978-3-642-16483-5_2233 (2012) Phosphorylation. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 2870. doi:10.1007/978-3-642-16483-5_4544

Erlotinib (Tarceva) ▶ Receptor Tyrosine Kinase Inhibitors

ERM Proteins Ling Ren1 and Chand Khanna2 1 Pediatric Oncology Branch, National Cancer Institute, Center for Cancer Research, Bethesda, MD, USA 2 Comparative Oncology Program, Center for Cancer Research, National Cancer Institute, Bethesda, MD, USA

Synonyms Cytovillin; Villin 2

The entry “ERM Proteins” appears under the copyright Springer-Verlag Berlin Heidelberg (outside the USA) both in the print and the online version of this Encyclopedia.

The ERM family consists of three closely related proteins, ezrin, radixin, and moesin. ERM proteins are cell membrane and cytoskeleton linker proteins.

History, Structure, and Sequence Ezrin, the prototype ERM protein, is a 585-amino acid polypeptide, ﬁrst identiﬁed as a constituent of microvilli and shown to be present in actincontaining surface structure on a wide variety of cells. ERM proteins share homology in sequence structure and function. They are composed of three domains: an N-terminal globular domain, an extended a-helical domain, and a charged C-terminal domain. The N-terminal domain of ERM proteins is highly conserved and is also found in ▶ merlin, band 4.1 proteins, and members of the band 4.1 superfamily. This domain is called FERM (fourpointone protein, ezrin, radixin, moesin) domain. The crystal structure of moesin revealed that the FERM domain is composed of three structural modules that, together, form a compact clover-shaped structure. The C-terminal domain can extend across the FERM domain surface, potentially masking recognition sites of other proteins. Ezrin and radixin also contain a polyproline region between the helical and C-terminal domains. The cDNA sequence of radixin encodes a protein of 583 amino acids with 77% identity to ezrin. Moesin, isolated as a heparin-binding protein, consists of 577 amino acids with 74% identity to ezrin. Regulation ERM proteins are conformationally regulated. ERM proteins exist in proposed dormant forms in which the C-terminal tail binds to and mask the N-terminal FERM domain (Fig. 1). The activation of ERM protein is mediated by both C-terminal threonine phosphorylation (T567 in ezrin, T564 in radixin, T558 in moesin) and exposure to PIP2. It is likely that phosphorylation at other residues in ERM proteins is needed to maintain an open

ERM Proteins

1613

Indirect association

Direct association

CFTR, NHE-3, b2-adrenergic receptor

CD-44, ICAM-1, ICAM-2

Plasma membrane

EPB50

Kinase? PKA

PKC

GRK2 ROCK

Activated ERM monomer

pThr

Inactive ERM monomer F-actin

ERM oligomer or dimer-linked by N/C ERMAD associations

Cytoplasm

ERM Proteins, Fig. 1 Activation of ERM proteins. The dormant ERM proteins exist as monomers, dimers and oligomers with a closed conformation. The activation of ERM proteins is mediated by both exposure to PIP2 and phosphorylation of the C-terminal threonine. The C-

terminal of activated ERM proteins bind to F-actin ﬁlaments. The N-terminal domains of activated ERM proteins are associated directly with the adhesion molecules such as CD44 and ICAM-1, -2 and -3 or indirectly with other transmembrane proteins such as NHE3 through EPB50.

activated conformation for ezrin and to direct ezrin-speciﬁc effects in cells. It is also unclear what functions are ascribed to the so-called inactive closed conformation of ERM proteins. Several protein kinases have been found to phosphorylate the C-terminal threonine residue of the ERM proteins. Examples include PKCa (▶ protein kinase C family), PKCy (protein kinase C family), rho kinases/ROCK, G proteincoupled receptor kinase 2 (GRK2), and myotonic dystrophy kinase-related Cdc42-binding kinase (MRCK). Dephosphorylation of C-terminal Thr of moesin has been suggested to be a crucial step for lymphocyte adhesion and transendothelial migration. The disassembly of microvilli on lymphocyte cell surfaces caused by dephosphorylation of moesin

facilitates the cell–cell (lymphocyte–endothelium) contact. Protein phosphatase 2C is involved in the dephosphorylation of moesin through the activation of Rac1 small ▶ GTPase. Function, Distribution, Localization ERM proteins either directly associate with the cytoplasmic domains of adhesive type I membrane proteins, such as ▶ CD44, CD43, ICAM-1, ICAM-2, and ICAM-3, or indirectly associate with membrane proteins via PDZ-containing adaptors EBP50 and E3KAP. Regulated attachment of membrane proteins to F-actin is essential for many fundamental cellular processes, including the determination of cell shape, polarity, and surface structure, cell ▶ adhesion, ▶ motility, cytokinesis, phagocytosis, and

1614

integration of membrane transport with signaling pathways. There is functional redundancy between ERM proteins. This is best exempliﬁed by the phenotype of the ezrin knockout mouse. This mouse is viable at birth, suggesting the ability of radixin and moesin to ﬁll the role of ezrin during development. Interestingly, the fatal phenotype of this mouse is characterized by intestinal villous malformations seen at day 13 postpartum. The normal intestinal epithelial cells nearly exclusively express ezrin. Although ezrin, radixin, and moesin are coexpressed in most cultured cells, they exhibit a tissue-speciﬁc expression patterns. Ezrin is highly concentrated in the intestine, stomach, lung, and kidney although moesin is prominent in the lung and spleen, and radixin in the liver and intestine. Ezrin is expressed in epithelial and mesothelial cells, while moesin is expressed in endothelial cells. As indicted, the brush border of intestinal epithelial cells expresses only ezrin, and hepatocytes express only radixin. The Expression and Functions of ERM Proteins in Cancer Ezrin has been shown to be expressed in most human cancers and linked to progression in several cancers, including carcinomas of endometrium, breast, colon, ovary, in uveal and cutaneous ▶ melanoma, ▶ brain tumors, and soft tissue sarcomas. In a cDNA array (▶ microarray (cDNA) technology) analysis of highly and poorly metastatic ▶ rhabdomyosarcoma and ▶ osteosarcoma, ezrin was indicated as a key metastatic regulator. In several murine and human cancer models, suppression of ezrin protein and disruption of ezrin function signiﬁcantly reduced the metastatic phenotype despite the expression of other ERM proteins. This suggests that the redundancy provided by the other ERM proteins for ezrin does not extend to ▶ metastasis and that ezrin contributes a unique and necessary function to cells undergoing metastasis. Comparing lung adenocarcinoma with normal lung tissue, the expressions of ezrin, radixin, and moesin were decreased on mRNA level as well as the protein level. Interestingly, the high expression of ezrin was observed in the invading tumor cells in lung adenocarcinoma.

ERp60

Cross-References ▶ Adhesion ▶ Brain Tumors ▶ CD44 ▶ GTPase ▶ Merlin ▶ Metastasis ▶ Microarray (cDNA) Technology ▶ Motility ▶ Osteosarcoma ▶ Protein Kinase C Family ▶ Rhabdomyosarcoma

References Bretscher A, Chambers D, Nguyen R et al (2000) ERM-Merlin and EBP50 protein families in plasma membrane organization and function. Annu Rev Cell Dev Biol 16:113–143 Bretscher A, Edwards K, Fehon RG (2002) ERM proteins and merlin: integrators at the cell cortex. Nat Rev Mol Cell Biol 3:586–599 McClatchey AI (2003) Merlin and ERM proteins: unappreciated roles in cancer development? Nat Rev Cancer 3:877–883 Tsukita S, Yonemura S (1999) Cortical actin organization: lessons from ERM (ezrin/radixin/moesin) proteins. J Biol Chem 274:34507–34510 Vaheri A, Carpen O, Heiska L et al (1997) The ezrin protein family: membrane-cytoskeleton interactions and disease associations. Curr Opin Cell Biol 9:659–666

ERp60 ▶ Calreticulin

Erythrocytosis ▶ Polycythemia

Erythroid Colony-Stimulating Activity ▶ Erythropoietin

Erythropoietin

Erythroid Differentiation Factor ▶ Activin

Erythroleukemia Definition A form of acute myeloid leukemia where the myeloproliferation is of abnormal, immature red blood cells.

Cross-References ▶ Erythropoietin

Erythropoiesis-Stimulating Factor ▶ Erythropoietin

Erythropoietin Jiuwei Cui1 and Yaacov Ben-David2 1 Jilin University, Changchun, Jilin, China 2 Division of Molecular and Cellular Biology, Sunnybrook Health Sciences Centre, Toronto, ON, Canada

Synonyms ECSA; Ep; Epo; Epoetin; Erythroid colonystimulating activity; Erythropoiesis-stimulating factor; ESF

Definition Erythropoietin (Epo) (from Greek erythro for red and poietin to make) is a small glycoprotein

1615

hormone that is essential for the production of red blood cells. Epo promotes the survival, proliferation, and differentiation of erythroid progenitor cells (BFU-E, CFU-E) to mature erythrocytes and initiates hemoglobin synthesis.

Characteristics The Epo gene contains at least ﬁve exons and resides on chromosome 7q21-q22 in humans and chromosome 5 in mice. DNA sequences from monkey and mouse display 90% and 80% homology to human Epo, respectively. Epo is produced primarily in the kidney and to a lesser extent in the liver. It is an acidic glycoprotein hormone with a molecular weight of 34–37 kD and circulates in the blood plasma at a very low concentration (about 5 pmol/l). It is composed of a single chain polypeptide and is resistant to denaturation by heat, alkali, or reducing agents. Epo is synthesized as a 193-amino acid precursor that is cleaved to yield an active protein of 165 amino acids. It is N-glycosylated at asparagine residues 24, 36, and 83 and O-glycosylated at serine 126. Epo is also sialylated and contains two disulﬁde bonds at positions 7/161 and 29/33. The alpha form of the hormone consists of 31% carbohydrates, while the beta form consists of 24%. These two forms of Epo have similar biological and antigenic properties. The carbohydrate moiety of Epo plays an important role in the mediation of its full biological effect and the pharmacokinetic behavior of the protein in vivo; non-glycosylated Epo has a very short biological half-life. Epo is fully synthesized in its active form prior to secretion into circulation. Epo, already known as the stimulating hormone for erythropoiesis, has displayed different and interesting pleiotropic actions. It not only affects erythroid cells but also myeloid cells, lymphocytes, and megakaryocytes. This hormone can enhance phagocytic function of polymorphonuclear cells and reduce the activation of macrophages, thus modulating the inﬂammatory process. Epo also exerts diverse biological effects in many nonhematopoietic tissues and is involved in the wound-healing cascade, functions as a proangiogenic cytokine during physiological

1616

Erythropoietin

Epo

STAT5

P Y JAK2

W282 BOX2

JAK2

GAP

Y P Ras

Ras Raf

Y343

LYN Cis P Y SHP-1 P Y PI3K

Y401 Y429 Y431 Y443 Y450 Y464 Y479

Y P

GDP SHP-2 Y P

GTP

GRb2 Cbl Y P Shc Y P

Gene transcription P Y

SOS

P Y

BOX1

STAT5

P MAPK

PLCγ

Nucleus

SHIP

Y P

Cytoplasm

Erythropoietin, Fig. 1 Schematic diagram of the EpoR depicting the positions of tyrosine (Y) residues (black bars; Y) in the cytoplasmic domain and attachment sites of signal transduction proteins such as STAT5, SHP-1, and

SHP-2. Binding of Epo to its receptor results in the autophosphorylation and activation of JAK2, which in turn phosphorylates eight tyrosine residues in the cytoplasmic domain of the EpoR

angiogenesis in the embryo and uterus, and exerts tissue-protective effects as part of the innate response to stressors.

to Epo. The Epo receptor (EpoR) belongs to the class I cytokine receptor superfamily. The mouse EpoR consists of 507 amino acids with an extracellular domain, a single hydrophobic transmembrane domain, and a cytoplasmic domain. The human EpoR is a 66 kD protein comprised of 508 amino acids. It consists of eight exons spanning some 6 kb on human chromosome 19p13.3. The interaction of Epo with its receptor results in the formation of a homodimer and its subsequent internalization (Fig. 1). Dimerization of the receptor results in the autophosphorylation of Janus kinase 2 (JAK2), a protein kinase that is tightly associated with the EpoR. Once activated, JAK2 phosphorylates eight tyrosine residues located in the cytoplasmic domain of the EpoR. Phosphorylation of the EpoR leads to the recruitment and phosphorylation of a number of signal transduction proteins. One such protein is

Cellular and Molecular Regulation The synthesis of Epo in the kidney is under the control of an oxygen-sensing mechanism. Transcriptional response of the Epo gene to hypoxia is mediated partly by promoter sequences but mainly by a 24 bp hypoxia-response element located at the 30 ﬂanking region of the Epo gene bound to the hypoxia-inducible factor-1(HIF-1). Epo production is also modulated by several other factors such as hypoglycemia, increased intracellular calcium, insulin release, estrogen, androgenic steroids, and various cytokines. The biological activity of Epo is mediated by its speciﬁc receptors present at 300–3,000 copies per cell that undergo phosphorylation in response

Erythropoietin

STAT5, a transcription factor that plays an important role in the regulation of in vivo erythropoiesis. Once phosphorylated by binding to tyrosine 343 and 401 of the EpoR, STAT5 translocates to the nucleus to activate the expression of several downstream target genes. Other signaling cascades triggered by Epo binding to its receptor include phosphatidylinositol 3-kinase (PI3K) that binds to tyrosine 479 and is involved in erythroblast survival and Grb2 that binds to tyrosine 464 and is involved in the activation of the Ras pathway. Ras pathway may be required for the synergistic expansion of erythroid progenitors and precursor cells in response to Epo and stem cell factor (SCF). EpoR-mediated activation of phospholipase A2 and C also leads to the release of membrane phospholipids, the synthesis of diacylglycerol, and the increase in intracellular calcium levels and pH. Since phosphorylation of the EpoR by Epo is diminished after 30 min of stimulation, a number of tyrosine phosphatases have been identiﬁed that are involved in attenuating the signal. The tyrosine phosphatase SHP-2 binds to tyrosine 401 of the Epo receptor and stimulates erythroid proliferation, while SHP-1 binds to tyrosine 429 and inhibits proliferation. Expression of the EpoR is not restricted to hematopoietic cells and exhibits a multi-tissue distribution that includes vascular endothelial cells, muscle cells, and neurons; therefore, Epo is believed to play a physiological role in angiogenesis and cardiac and brain development. Abnormal regulation of Epo-EpoR signaling in hematopoietic cells has been associated with proliferative disorders of the bone marrow, such as polycythemia vera, a disorder characterized by erythrocytosis, as a consequence of an active mutation in the EpoR. Additionally, prolonged activation of STAT5 has been observed in cells transfected with mutant (tyrosine 429) EpoR, suggesting that STAT5 DNA binding activity may play a role in the pathogenesis of erythrocytosis. A point mutation at position 129 of the mouse EpoR gene results in constitutive activation of the receptor without stimulation with Epo. Mice infected with a retrovirus expressing this aberrant receptor develop ▶ erythroleukemia and splenomegaly. Taken

1617

together, these provide evidences that the precise control of Epo-EpoR signaling is critical for the normal proliferation and differentiation of erythroid progenitor cells. Clinical Relevance The synthesis of Epo is subject to a complex circuit that links the bone marrow and kidney in a feedback loop. Its reference interval in the blood plasma ranges between 3.3 and 16.6 mIU/ml. Patients suffering from most anemias display higher than normal concentrations of serum Epo, whereas those suffering from anemia associated with chronic renal disease have values either low or within the normal range. Epo levels are disproportionately low in anemic patients with chronic disorders as well, such as rheumatoid arthritis, AIDS, and cancer, in which inhibition of Epo production and erythroid progenitor proliferation by inﬂammatory cytokines, such as IL-1 and TNF, are thought to play major causative roles. Abnormally high concentrations may also be induced by renal neoplasms, benign tumors, polycystic kidney disease, renal cysts, and hydronephrosis. The pathophysiological excess of Epo leads to erythrocytosis that is accompanied by an increase in blood viscosity and may cause heart failure and pulmonary hypertension. Chronic kidney disease causes the destruction of Epo-producing cells resulting in hyporegenerative normochromic normocytic anemias. Epo is therefore clinically used for the treatment of patients with severe kidney insufﬁciency. In uremic patients, treatment with recombinant human Epo (rhEpo) effectively reactivates the bone marrow to produce erythrocytes and also improves platelet adhesion and aggregation. Hypertension is an important complication in the treatment of renal anemia with rhEpo. rhEpo is also used to treat nonrenal forms of anemia caused by chronic infections, inﬂammation, radiation therapy, and chemotherapy. Beyond ameliorating anemia, rhEpo has been shown to restore radiosensitivity and increase cytotoxicity of chemotherapy in the treatment of cancer-related anemia. However, clinical trials have shown increase in the relative risk of thromboembolic complications and lower survival, which raises

1618

concerns about the potential adverse effects of rhEpo in cancer patients. Additional studies show that Epo and EpoR expression also occurs in tumor cells, suggesting the potential for the generation of an autocrine or paracrine growthstimulator Epo-EpoR loop in cancer cells. Further studies will be required to investigate the effects, if any, of rhEpo therapy on disease progression and survival. For its role in stimulating the production of erythrocytes, an important application of Epo is the presurgical activation of erythropoiesis allowing for the collection of autologous donor blood. rhEpo has emerged as a novel antiinﬂammatory and cytoprotective agent, as evidenced by its physiological response to various forms of tissue injury. Accordingly, the therapeutic potential of Epo has been shown in acute renal failure, diabetic neuropathy, myocardial infarction, and cerebral ischemia. The characterization of Epo variants, such as asialo-Epo and carbamylated Epo, that retain nonhematopoietic, tissue-protective properties of Epo without stimulating erythropoiesis has uncovered new areas of research into the mechanisms of Epo-mediated signaling in nonhematopoietic tissues as well as novel clinical applications for rhEpo and its derivatives in disorders other than anemia.

References Hardee ME, Acrasoy MO, Blackwell KL et al (2006) Erythropoietin biology in cancer. Clin Cancer Res 12:332–339 Heuser M, Ganser A (2006) Recombinant human erythropoietin in the treatment of nonrenal anemia. Ann Hematol 85:69–78 Jelkman W (1992) Erythropoietin, structure, control of production and function. Physiol Rev 72:449–489 Lipton S (2004) Erythropoietin for neurological protection and diabetic neuropathy. N Eng J Med 350:2516–2517 Wojchowski DM, Gregory RC, Miller CP et al (1999) Signal transduction in the erythropoietin receptor system. Exp Cell Res 253:143–156

ESA ▶ EpCAM

ESA

ESE (EH Domain and SH3 Domain Regulator of Endocytosis) ▶ Intersectin

E-Selectin-Mediated Adhesion and Extravasation in Cancer Liang Zhong, Bryan Simoneau, Pierre-Luc Tremblay, Stéphanie Gout, Martin J. Simard and Jacques Huot Le Centre de recherche du CHU de QuébecUniversité Laval: axe Oncologie, Le Centre de recherche sur le cancer de l’Université Laval, Québec, QC, Canada

Keywords Angiogenesis; death receptor-3; extravasation; endothelium; MAP kinases; metastasis; sialyl Lewis determinants

Synonyms CD62 antigen-like family member E (CD62E); Endothelial-leukocyte adhesion molecule 1 (ELAM1); Leukocyte-endothelial cell adhesion molecule 2 (LECAM2)

Definition ▶ Adhesion of circulating cancer cells to the endothelium is a prerequisite for ▶ metastasis. It requires speciﬁc interactions between adhesion molecules such as E- and P- selectins that are present on endothelial cells and their counter-receptors on cancer cells. The speciﬁcity of this interaction constitutes the basis of the organ selectivity of metastatic colonization. Subsequently to adhesion, cancer cells form metastasis either by growing locally in the capillaries or by invading the surrounding tissues following extravasation.

E-Selectin-Mediated Adhesion and Extravasation in Cancer

Characteristics Structure E-selectin (64 kDa) is a transmembrane receptor of the selectin family that also contains L- and Pselectins. Two glycosylated forms of E-selectin are detected at 100 and 115 kDa. The extracellular part of selectins is constituted of three domains: an N-terminal C-type lectin domain, which is calcium-dependent and mediates ligand interaction; an epidermal growth factor (EGF) domain, which also regulates ligand interaction; and consensus complement regulatory protein (CRP) repeats of ~60 amino acids each, which serve as spacers to hold the other two domains away from cell surface and mediate the rolling of adhering cells (see below). The number of CRP repeats distinguishes the extracellular domain of different selectins. Eselectin has six CRP repeats. Selectins are anchored in the membrane through a single helicoidal transmembrane domain followed by a short cytoplasmic tail (Fig. 1). The cytoplasmic tail of E-selectin can trigger signaling in the endothelial cell and is connected to the actin cytoskeleton via actin-binding proteins, which are important mediators of extravasation. Expression E-selectin is expressed exclusively by endothelial cells. Its constitutive expression has been detected in the skin and parts of bone marrow

1619

microvasculature. However, in most vessels, the de novo synthesis of E-selectin is induced by proinﬂammatory molecules such as tumor necrosis factor a (TNFa), interleukin 1b (IL-1b), endothelial monocyte activating polypeptide II (EMAPII), and bacterial lipopolysaccharide (LPS). Following stimulation by TNFa, E-selectin relies on PI3K-Akt-NFκB and JNK-c-Jun pathways for its transcription. In physiological conditions such as inﬂammation, the expression of Eselectin is transient and often reaches its peak 2–6 h after stimuli. E-selectin is gradually internalized by endocytosis by clathrin-coated pits and degraded in the lysosomes. In the endothelium areas of chronic inﬂammation, E-selectin may remain upregulated. Several cancer cells have the ability to induce E-selectin. For instance, Lewis lung carcinoma cells induce E-selectin in liver sinusoidal endothelium. Moreover, highly metastatic human colorectal and mouse lung carcinoma cells, upon their entry into the hepatic microcirculation, induce TNFa production by resident Kupffer cells, triggering E-selectin expression. Clinically, patients with various cancers including breast, colorectal, lung, bladder, head and neck, and melanoma have elevated galectin-3 in their serum. In turn, galectin-3 induces secretion of proinﬂammatory cytokines by blood vascular endothelium, which triggers the expression of E-selectin. A ΤΝFa-inducible microRNA, miR-31, directly targets the mRNA of E-selectin

E-selectin-mediated adhesion and extravasation in cancer C-type lectin domain EGF domain

CRP repeats

Transmembrane domain Cytoplasmic tail

E-Selectin-Mediated Adhesion and Extravasation in Cancer, Fig. 1 Structure of E-selectin. The extracellular part of E-selectin is divided into three domains: an Nterminal C-type lectin domain, an epidermal growth factor

Plasma membrane of endothelial cell

(EGF) domain, and consensus complement regulatory protein (CRP) repeats. E-selectin is anchored in the membrane through a single helicoidal transmembrane domain, followed by a short cytoplasmic tail

1620

E-Selectin-Mediated Adhesion and Extravasation in Cancer

Step 1

E-selectin-mediated adhesion and extravasation in cancer

Step 2

Blood flow

Cancer cell Step 3 Chemokines

DR3 Integrins E-selectin CAMs

Endothelial cells

E-Selectin-Mediated Adhesion and Extravasation in Cancer, Fig. 2 E-selectin-mediated adhesion and transendothelial migration of cancer cells. Extravasation of cancer cells is a multistep process. The ﬁrst step involves the transient adhesion of cancer cells to the endothelium. It requires endothelial E- and P-selectins, and their counterreceptors (such as DR3 and CD44 for E-selectin) on cancer

cells. This step is associated with the rolling of cancer cells on the endothelium. The second step involves a ﬁrmer adhesion of cancer cells to the endothelium, which is mediated by cell adhesion molecules (CAMs) on the endothelium and integrins on cancer cells. The third step is the extravasation of cancer cells through endothelial cell-cell junctions

and downregulates its expression, suggesting its involvement in carcinoma dissemination.

distribution pattern of E-selectin enhances its ability to mediate rolling in ﬂow conditions. The sequence of events is as follows: the Ctype lectin domain of E-selectin binds its ligand on cancer cells. This primary adhesion is unstable under shear stress, which allows the rolling of cancer cells along the endothelium. In response to the E-selectin-mediated attachment, chemokines are produced and released by endothelial cells, which activate integrins on cancer cells. Integrins are capable of ﬁrmly binding to cell adhesion molecules (ICAM-1/2 and VCAM-1) on endothelial cells, allowing the extravasation of cancer cells into tissues (Fig. 2). Breast, bladder, gastric, and pancreatic carcinoma, leukemia, and lymphoma form metastasis in an E-selectindependent manner in organs as various as liver, bone marrow, skin, and lung. The interaction between E-selectin and its ligand triggers signals in both endothelial cells bearing E-selectin and cancer cells bearing the ligand. When E-selectin binds to DR3 on colon carcinoma cells, on one hand, this interaction activates not only the prosurvival ERK MAP kinase and PI3K pathways but also the

Function E-selectin recognizes the sialyl Lewis-a/x tetrasaccharide borne by glycoproteins and glycolipids on the surface of leukocytes and tumor cells. Its glycoprotein ligands include: E-selectin ligand-1 (ESL-1), P-selectin glycoprotein ligand1 (PSGL-1), b2 integrin, L-selectin, CD43/44, lysosomal-associated membrane protein-1/2 (LAMP-1/2), mucin-16 (MUC16), Mac-2, podocalyxin (PODXL), and death receptor-3 (DR3). Malignant transformation is often associated with abnormal glycosylation such as increased sialyl Lewis-a/x structures. On carcinoma cells, sialyl Lewis-a/x are mostly carried by mucins, making them major E-selectin ligands on carcinoma cells. The physiological role of Eselectin is to mediate the adhesion of leukocytes to the endothelium. In pathological conditions, cancer cells “hijack” the inﬂammatory system to interact with E-selectin. On the surface of endothelial cell, E-selectin molecules cluster in clathrin-coated pits and lipid rafts. This

E-Selectin-Mediated Adhesion and Extravasation in Cancer

1621 E-selectin-mediated adhesion and extravasation in cancer

Colon carcinoma cell

Migration

Apoptosis

Survival

p38

ERK

PI3K

DR3

Endothelial cells

E-selectin

E p38 F-actin

ERK MLC

Src

VE-cadherin/β-catenin complex

Cell retraction

E-Selectin-Mediated Adhesion and Extravasation in Cancer, Fig. 3 E-Selectin-Mediated Bidirectional Signaling in colon carcinoma cells and endothelial cells. The adhesion of colon carcinoma cells to endothelial cells involves the binding of E-selectin on endothelial cells to counter-receptors such as death receptor-3 (DR3) on colon carcinoma cells. The interaction between E-selectin and DR3 induces activation of PI3K, p38 and ERK MAP

kinases in cancer cells, which increase their motile and survival potentials. Reciprocally, the interaction triggers the activation of p38 and ERK MAP kinases in endothelial cells, which results in myosin-light chain (MLC)-mediated cell retraction, and dissociation of the VE-cadherin-bcatenin complex, and thereby destruction of adherens junctions leading to increased endothelial permeability and extravasation of cancer cells

promigratory p38 MAP kinase pathway in colon carcinoma cells; on the other hand, in endothelial cells, the interaction activates p38 and ERK MAP kinase pathways to increase the permeability of the endothelium (Fig. 3). Similar mechanism has also been observed with ESL-1, where E-selectin binds to ESL-1 on the circulating prostate cancer cell, and activates the pro-metastatic RAS-ERKcFos signal cascade in the cancer cell. Moreover, CD44 on melanoma cells can bind to E-selectin on endothelial cells and activate PKCa-p38-SP-1 pathway to up-regulate ICAM-1 on endothelial cells. This bidirectional signal transduction also characterized the tethering of leukocytes: Eselectin binding to PSGL-1 on neutrophils activates b2 integrin through the Syk-Src pathway. At the same time, E-selectin transduces signals into endothelial cells through p38 and p42/p44 MAP kinase pathways. Overall, E-selectin-mediated adhesion of cancer cells increases their metastatic potential by inducing bidirectional signaling that enhances their intrinsic motile and survival

abilities, as well as the permeability of the endothelium. E-selectin is a double-edged sword in cancer therapy, as it also allows lymphocyte inﬁltration into tumor. Some cancer cells are able to reduce Eselectin to evade immune detection: squamous cell carcinomas can recruit nitric oxide (NO)–producing myeloid-derived suppressor cells, and this local production of NO inhibits vascular E-selectin expression, preventing T cells from entering in squamous cell carcinomas. In this case, lower E-selectin level is correlated with lower survival. On the other hand, many types of cancer cells beneﬁt from E-selectin in a variety of ways. The recruitment of leukemia cells by E-selectin sequesters leukemia cells in a quiescent state, rendering them immune to chemotherapy. Given that leukemia cells can stimulate endothelial cells by themselves, they promote their own survival through E-selectin. In addition, proliferating hemangioma endothelial cells from infantile hemangioma constitutively express E-

1622

selectin, which enhances hemangioma stem cell adhesion and vasculogenesis. A new role in stem cell proliferation has been identiﬁed for E-selectin on bone marrow vascular endothelial cells: it recruits hematopoietic stem cells that express appropriate ligands, and this attachment wakes hematopoietic stem cells up by inducing their proliferation, self-renewal, chemosensitivity, and radiosensitivity.

Clinical Relevance of E-Selectin in Cancer The ﬁnding that cancer cells are recruited by E-selectin-expressing endothelial cells is of signiﬁcant clinical importance and opens several therapeutic avenues. Targeting E-selectin and Its Ligands Various strategies targeting E-selectin and its ligands are promising to suppress E-selectinmediated cancer cell adhesion. For example, antibodies against E-selectin can impair lung metastasis of colon carcinoma in mice. ESTA, an aptamer targeting E-selectin, is able to reduce metastasis of breast cancer in mice. ESTA is safe as an antagonist as it can be applied at high doses without causing overt side effects. It is of particular interest for the prevention of metastasis of ER ()/CD44(+) breast cancers. Similarly, SDA, a DNA aptamer antagonizing E- and P-selectins, also exhibits antiadhesive effect for colorectal cancer and leukemia in vitro. In mice, encouraging results have been obtained with colon cancer metastasis by using cimetidine to inhibit Eselectin expression. In clinical trials, cimetidine treatment dramatically improves the 10-year cumulative survival of colorectal cancer patients. Moreover, atrial natriuretic peptide (ANP) is capable of reducing E-selectin expression and preventing recurrence in patients with non-small cell lung cancer. E-selectin is also a target for inhibition of angiogenesis. Knocking down vascular E-selectin in mice inhibited the recruitment of endothelial progenitor cells to the tumor, thus reducing angiogenesis and tumor growth in human melanoma xenograft murine model. Another approach for reducing E-selectin-

E-Selectin-Mediated Adhesion and Extravasation in Cancer

mediated adhesion is to target ligands of Eselectin. Antibodies against sialyl Lewis-a/x determinants inhibited the formation of metastasis by human pancreatic and gastric cancers in nude mice. The antisense-cDNA for fucosyltransferase genes (FUT III/VI), enzymes producing sialyl Lewis saccharides, suppressed metastatic colonization by colon cancer cells in mice. Along the same lines, celecoxib, an inhibitor of cyclooxygenase-2, impaired the expression of sialyl Lewis-a on colon cancer cells and reduced metastasis. Still in mice, re-introduction of the glycosyltransferase B4GALNT2, which synthesizes “normal” saccharides instead of sialyl Lewis-a/x, prevented dissemination of gastric carcinoma. Soluble E-selectin as a Diagnostic Marker A soluble form of E-selectin (sE-selectin) is generated by enzymatic cleavage or when activated endothelial cells shed their damaged parts. The concentration of sE-selectin is directly correlated with its cell surface expression. sE-selectin limits E-selectin-mediated rolling by competing for binding sites on the leukocyte, thus downregulates the inﬂammatory response. sEselectin can be used as a marker of activation of endothelium and is therefore useful for diagnosis of acute inﬂammation and metastasis. Speciﬁcally, for breast cancer patients, high sE-selectin level is associated with liver metastasis. In patients with non-small cell lung cancer, high sE-selectin level correlates with poor prognosis when cancer cells express sialyl Lewis-a/x. Increased sE-selectin also characterizes patients with chronic lymphatic leukemia. For patients suffering from oral cavity cancer, higher level of E-selectin correlates with higher risk of cancer transformation and relapse. Hence, the determination of blood sE-selectin on tumor biopsies is of prognostic value. E-selectin-Mediated Capture Highly metastatic circulating cancer cells express mucins with increased sialyl Lewis-a/x (such as MUC1), and these mucins consistently expose their core epitope. Nanostructured surface coated with E-selectin offers a method to selectively

E-Selectin-Mediated Adhesion and Extravasation in Cancer

capture viable cancer cells from blood samples. This E-selectin-mediated capture allows fast analysis and elimination of circulating cancer cells. For instance, microtube surface with E-selectinfunctionalized liposomal doxorubicin speciﬁcally captured breast adenocarcinoma MCF7 cells from the perfusion, and induced signiﬁcant cancer cell death. E-selectin as a Receptor for Targeted Delivery E-selectin can serve as a receptor for the delivery of anti-inﬂammatory drugs, anticancer drugs, and imaging markers in endothelial cells. For this purpose, antibodies against E-selectin, or artiﬁcial ligands of E-selectin are conjugated to the surface of polymeric particles. These immunoparticles are used to encapsulate the agent, so they can selectively bind to E-selectin-expressing endothelial cells and get internalized, together with the agent inside them. This technique allows the speciﬁc delivery of drugs to the proinﬂammatory microenvironment harboring tumor cells. Naturally, immunoparticles targeting E-selectin can also directly compete with cancer cells to bind to E-selectin. Based on these principles, intravenous injections of two E-selectin-targeting drug-carrying immunoparticles, P-(Esbp)-DOX and P-(Esbp)-KLAK, inhibited primary tumor growth and metastasis of lung carcinoma in mice. Moreover, the “drug free” immunoparticle P-(Esbp) also exhibited antimetastatic effects by competing with circulating lung carcinoma cells. By targeting E-selectin, we can also carry out targeted gene therapy, if viral vectors are encapsulated. E-selectin thioaptamer-conjugated multistage vector (ESTA-MSV) can carry therapeutic antiSTAT3 siRNA to bone marrow vascular endothelium of mice, and infect breast cancer cells there. In vitro, anti-E-selectin lipoplexes can deliver anti-VE-cadherin siRNAs to inﬂamed primary vascular endothelial cells originating from different vascular beds, which are generally difﬁcult to transfect. Overall, E-selectin-mediated endothelial adhesion plays a key role in metastasis, which opens new avenues for therapeutic interventions aiming at inhibiting this fatal complication of cancer.

1623

Glossary Angiogenesis Angiogenesis is a multistep process that refers to the formation of new blood vessels from pre-existing ones. In cancer, angiogenesis is required to feed the tumors. It is initiated by an hypoxic signal generated by cancer cells and that leads to the expression of angiogenic agents such as vascular endothelial growth factor (VEGF) that will switch on angiogenesis and activate endothelial cells. Extravasation Extravasation refers to the passage of cancer cells from blood or lymph vessels to the surrounding tissues. MAP kinases The MAPK cascades are highly conserved signaling networks that transduce the signals elicited by stress and physiological stimuli. The ERK pathway is the best known of these cascades. It is activated, for example, through the binding of agonists to tyrosine kinase (TK) receptors, which results into auto-phosphorylation on tyrosyl residues, hence creating docking sites for adapter proteins and enzymes,followed by the activation in cascade of the GTPase Ras, the MAP kinase kinase kinase (MAPKKK) Raf, the MAP kinase kinases (MAPKK) MEK-1/2, and ﬁnally the MAP kinase (MAPK) ERK. In turn, ERK phosphorylates a number of cytoplasmic and nuclear proteins to regulate cell functioning. The functions of the signaling molecules along the MAPK pathways are regulated by scaffolding proteins such as CNK in the ERK pathway. The scaffolding proteins facilitate or restrict the enzyme/substrate interactions by modulating the mutual availability of the signaling components. Metastasis Metastasis consists in the formation of secondary tumor sites distant from the primary site. The capacity to metastasize is a characteristic of all malignant tumors.

Cross-References ▶ Adherens Junctions ▶ Adhesion ▶ Cell Adhesion Molecules

1624

▶ MAP Kinase ▶ Metastasis ▶ microRNA

References Bird NC, Mangnall D, Majeed AW (2006) Biology of colorectal liver metastases. A review. J Surg Oncol 94:68–80 Gout S, Morin C, Houle F et al (2006) Death receptor-3, a new E-selectin counter-receptor that confers migration and survival advantages to colon cancer cells by triggering p38 and ERK MAPK activation. Cancer Res 66:9117–9124 Gout S, Tremblay PL, Huot J (2008) Selectins and selectin ligands in extravasation of cancer cells and organ selectivity of metastasis. Clin Exp Metastasis 25:335–344 Jubeli E, Moine L, Vergnaud-Gauduchon J, Barratt G (2012) E-selectin as a target for drug delivery and molecular imaging. J Control Release 158: 194–206 Kannagi R, Izawa M, Koike T et al (2004) Carbohydratemediated cell adhesion in cancer metastasis and angiogenesis. Cancer Sci 95:377–384 Khatib AM, Auguste P, Fallavollita L et al (2005) Characterization of the host proinﬂammatory response to tumor cells during the initial stages of liver metastasis. Am J Pathol 167:749–759 Läubli H, Borsig L (2010) Selectins promote tumor metastasis. Semin Cancer Biol 20(3):169–177 Suárez Y, Wang C, Manes TD, Pober JS (2009) Cutting edge: TNF-induced microRNAs regulate TNF-induced expression of E-selectin and intercellular adhesion molecule-1 on human endothelial cells: feedback control of inﬂammation. J Immunol 184(1):21–25 Tremblay PL, Auger FA, Huot J (2006) Regulation of transendothelial migration of colon cancer cells by E-selectin-mediated activation of p38 and ERK MAP kinases. Oncogene 25:6563–6573

ESF (2012) Integrin. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1884. doi:10.1007/978-3-642-16483-5_3084 (2012) Sialyl Lewis-a/x determinants. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 3402–3403. doi:10.1007/978-3-64216483-5_5293

ESF ▶ Erythropoietin

ESM-1 ▶ Endocan

ESO1 ▶ NY-ESO-1

Esophageal Adenocarcinoma Landon Inge Norton Thoracic Institute, St. Joseph’s Hospital and Medical Center, Phoenix, AZ, USA

See Also (2012) Cyclooxygenase-2. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1035. doi:10.1007/978-3-642-16483-5_1435 (2012) Death receptor-3. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1065. doi:10.1007/978-3-642-16483-5_1537 (2012) Endothelium. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1258. doi:10.1007/978-3-642-16483-5_1905 (2012) Extravasation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1370. doi:10.1007/978-3-642-16483-5_2080 (2012) Galectin-3. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1490. doi:10.1007/978-3-642-16483-5_2306

Synonyms Barrett adenocarcinoma; adenocarcinoma

Oesophageal

Definition Esophageal adenocarcinoma (EAC) is an epithelial malignancy of the lining of the esophagus. EAC is distinct from squamous cell carcinoma

Esophageal Adenocarcinoma

of the esophagus, arising from the glandular, metaplastic columnar epithelium characteristic of Barrett esophagus (BE).

Characteristics Esophageal cancer is the sixth most frequent malignancy worldwide. Squamous cell carcinoma accounts for the majority of esophageal cancers; however, epidemiological data reveal increased incidence of EAC within Western countries. In particular, EAC now accounts for the majority of esophageal cancers in the USA and is one of the few malignancies whose incidence is continuing to increase in the USA (Simard et al. 2012). Fiveyear relative survival rates (2000–2007) are 47.8% for localized cancer, 8.9% for cancer with regional metastasis, and 2.9% for cancer with distant metastasis. Overall survival for EAC is poor, a consequence of the frequent presence of metastasis at diagnosis (71% of diagnoses). Epidemiology, Risk Factors, and Clinical Treatment Within the USA, analyses of data from the National Cancer Institute Surveillance, Epidemiology, and End Results program show that Caucasian males and females carry the highest risk of EAC relative to other ethnic groups. These ﬁndings are paralleled by high incidence of EAC in Western Europe and Australia, and increases in EAC incidence have been reported in Japan and Singapore. Epidemiological analyses have deﬁned several risk factors for EAC. Obesity and symptomatic gastroesophageal reﬂux disease (GERD) are currently the strongest risk factors for EAC. Casecontrol and cohort studies show that obesity, deﬁned as a body mass index >30 kg/m2, increases the relative risk of EAC by 2.4–2.8. Similarly, symptomatic GERD increases the relative risk of EAC fourfold. Additional risk factors (poor diet and smoking) have also been identiﬁed. Barrett esophagus (BE), a metaplastic response to GERD, carries a low (~0.5% per year) but

1625

signiﬁcant risk of EAC and is thought to affect 1.5–1.6% of the population in the Western world. Despite the low risk of EAC within overall BE patient population, subsets of BE patients undergo pathological progression, developing neoplastic morphology (dysplasia or intraepithelial neoplasia) within their BE lesion. In addition, histological diagnosis of dysplasia frequently occurs concomitantly with a diagnosis of invasive EAC. As such, current treatment practices are designed to identify and remove dysplasia in patients with BE, despite the understanding that the bulk of these patients will likely never progress to dysplasia and EAC. Current treatment guidelines for BE patients in the USA advocate endoscopic biopsy, combined with pathological review at 1–2-year intervals in order to diagnose dysplasia or noninvasive EAC early, allowing for endoscopic removal of the dysplastic BE/EAC lesion via radio-frequency ablation or resection. For patients diagnosed with invasive/metastatic EAC, treatment is limited to surgery and chemotherapy combine with radiation therapy. Staging determines the treatment approach with patients with early stage (I–III) treated with surgery or preoperative chemoradiation followed by surgery and chemoradiation without surgery for advanced (stage IV, metastatic) disease. Preferred chemotherapies for EAC are platinum-containing DNA cross-linkers (cisplatin, oxaliplatin, carboplatin) combined either with an antimetabolite (5-ﬂuorouracil, capecitabine), a topoisomerase inhibitor (irinotecan, epirubicin), or a mitotic inhibitor (paclitaxel). For EAC patients whose tumors display HER2-neu (ERBB2) overexpression, the addition of the HER2-neu inhibitor, trastuzumab, to standard chemotherapy has been shown to be beneﬁcial. Mechanisms of Esophageal Adenocarcinoma Pathogenesis Chronic inﬂammation has a central role in EAC carcinogenesis. Retrospective analyses of human samples harboring the pathological progression from BE to EAC, as well as murine models of EAC pathogenesis, reveal increases in several

1626

inﬂammatory molecules (IL-1b, IL-8, IL-6, COX-2) and activation of the pro-inﬂammatory signaling (NF-kB, STAT3). Supplementation with nonsteroidal anti-inﬂammatory agents (NSAIDs) reduces progression to EAC in surgical murine models of BE to EAC pathogenesis, while retrospective epidemiological analysis found that BE patients taking NSAIDs had a lower risk of EAC. Finally, establishment of a chronic inﬂammatory microenvironment within the murine esophagus via ectopic, constitutive expression of the pro-inﬂammatory cytokine, IL-1b, induces EAC disease pathogenesis comparable to the human disease process (esophagitis-BE-EAC) (Quante et al. 2012). Development of chronic inﬂammation is a consequence of GERD. The repeated exposure to the gastric reﬂuxate, comprised of gastric acid, digestive enzymes, bile salts, and ingested food and their metabolites, results in acute and chronic inﬂammation and consequently oxidative stress and DNA/tissue damage. Introduction of bile salts or reduced pH in cell culture replicates the inﬂammatory state within BE and EAC cell lines. Obesity is thought to also contribute to the chronic inﬂammatory state by either increasing the episodes of GERD or via increased systematic chronic inﬂammation associated with an obese state. Other contributors, such as diet and smoking have also been demonstrated; however, exactly how they contribute to EAC pathogenesis is still unclear. Some evidence suggests that increased intake of nitrosamines, found at high levels in processed meats as well as in tobacco, is a possible mechanism. Concurrently, there is limited evidence of genetic susceptibility for EAC, based upon studies of families with hereditary history of BE and EAC, as well as larger population-based studies of BE and EAC patients for genetic variants. Mutational Landscape of Esophageal Adenocarcinoma Several high-resolution genetic analyses of human EAC tumors have been completed (Nones et al. 2014; Weaver et al. 2014; Dulak et al. 2013). Of particular note, these studies show a high mutation frequency (8.0–9.9 mutations/Mb of DNA) and prevalence of T:A>G:C

Esophageal Adenocarcinoma

transversions. A single study found evidence of chromothripsis (Nones et al. 2014), a massive genomic rearrangement during a single catastrophic event, present in 32% of EAC tumors. Combined, these characteristics are suggested to be reﬂective of the deleterious microenvironment (GERD, chronic inﬂammation) EAC arises in. Frequent mutational inactivation of several tumor suppressor genes (CDKN2A, ARID1A, and SMAD4) are also reported. TP53 is the most frequently mutated gene across all three datasets. TP53 mutations were also found to designate dysplastic BE, and identiﬁcation of TP53 had value as a diagnostic marker of high-grade dysplasia (Weaver et al. 2014). In addition to mutational inactivation, methylation of CDKN2A is frequently present in EAC. Gain-of-function mutations in known oncogenes have not been found at a high incidence; however, other studies have reported ampliﬁcation in the ERBB2, KRAS, GATA4, ERBB1, and CCNE1 genes in EAC.

Cross-References ▶ CDKN2A ▶ HER-2/Neu ▶ KRAS ▶ Obesity and Cancer Risk ▶ TP53

References Dulak AM et al (2013) Exome and whole-genome sequencing of esophageal adenocarcinoma identiﬁes recurrent driver events and mutational complexity. Nat Genet 45:478 Nones K et al (2014) Genomic catastrophes frequently arise in esophageal adenocarcinoma and drive tumorigenesis. Nat Commun 5:5224 Quante M et al (2012) Bile acid and inﬂammation activate gastric cardia stem cells in a mouse model of Barrettlike metaplasia. Cancer Cell 21:36 Simard EP, Ward EM, Siegel R, Jemal A (2012) Cancers with increasing incidence trends in the United States: 1999 through 2008. CA: Cancer J Clin Weaver JM et al (2014) Ordering of mutations in preinvasive disease stages of esophageal carcinogenesis. Nat Genet 46:837

Esophageal Cancer

See Also Barrett esophagus (2012) In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 342. doi:10.1007/978-3-642-16483-5_527

Esophageal Cancer Ruggero Montesano International Agency for Research on Cancer, Lyon, France

Definition Esophageal cancer comprises two main types of malignant epithelial neoplasms: squamous cell carcinoma, originating from the lining squamous epithelium of the esophagus, and adenocarcinoma (also called Barrett adenocarcinoma), originating from metaplastic columnar epithelium in the lower part of the esophagus.

Characteristics Esophageal cancer is the sixth most frequent cancer worldwide. In 2001, the estimated number of deaths due to esophageal cancer amounted to about 338,000 out of a total of 6.2 million cancer deaths. Of those, more than 80% occurred in developing countries, the majority being squamous cell carcinomas. The occurrence of this cancer varies greatly in different parts of the world, with areas of high mortality rate per year in regions of South Africa, Northeast of Iran and China (30 or more per 100,000 in males and 10 per 100,000 in females). In Europe or USA, the age standardized annual mortality of squamous cell carcinoma is no more than 5 in males and 1 in females per 100,000. There are, however, areas in Europe, namely in Normandy and Brittany in France and North east of Italy, where the

This entry was ﬁrst published in the 2nd edition of the Encyclopedia of Cancer in 2009.

1627

mortality rates, at least in males, are as high as those observed in China. Adenocarcinoma of the esophagus is less frequent and occurs mainly in industrial countries. However, epidemiological data show increasing numbers of adenocarcinoma cases and this type of cancer accounts for more than 50% of all esophageal cancer in the USA. The 5-year survival rate for patients with squamous cell carcinoma and adenocarcinoma of the esophagus is similarly poor (10%), with no difference between industrial and developing countries. This is mainly due to their late detection and the poor therapy efﬁcacy. No reliable prognostic markers are available. Epidemiological studies have clearly shown that tobacco smoke (▶ Tobacco Carcinogenesis) and alcohol, together with a low intake of fresh fruits, vegetables, and meat, is causally associated with squamous cell carcinoma. It is estimated that in industrial countries, approximately 90% of this cancer is attributable to tobacco and alcohol consumption. Other risk factors are chewing of betel, consumption of pickled vegetables, and hot mate drink in Southeast Asia, China, and South America, respectively. Adenocarcinoma of the esophagus arises from Barrett esophagus, a condition in which the normal squamous epithelium is replaced by metaplastic columnar epithelium. This condition is frequently present in patients with chronic gastroesophageal reﬂux and these patients have a more than 100-fold higher risk than the general population to develop adenocarcinoma. Squamous cell carcinoma and adenocarcinoma of the esophagus show multiple genetic alterations (point mutations, allelic loss, and gene ampliﬁcation) of several oncogenes and tumor suppressor genes. The most interesting observation in both cancer types is the high prevalence of mutations (up to 80%) of the tumor suppressor gene p53. In addition a distinct pattern of p53 mutations, namely a high prevalence of G>A transitions at CpG sites in adenocarcinoma and a higher prevalence of G>T transversions and mutations at A:T base pairs in squamous cell carcinoma. There is good evidence that the

1628 Esophageal Cancer, Fig. 1 Incidence of esophageal cancer in males from selected world regions (From Parkin et al. (1999))

ESR1

Turkmenistan Southern Africa Iran China Japan South America Calvados

Western Europe Veneto

U.S.A. 0

mutations in squamous cancer types are attributable to carcinogens present in tobacco smoke. In both types of cancers, p53 mutations occur very early and are followed by the accumulation of other genetic alterations during the process of esophageal carcinogenesis. It is evident that these genetic alterations are relevant not only in the understanding of the multifocal monoclonal origin of this cancer but also to the elucidation of its multifactorial etiology (Fig. 1).

10 20 30 40 50 Incidence ASR (world), per 100,000 (all ages)

ESR1 ▶ Estrogen Receptor

ESR2 ▶ Estrogen Receptor

Cross-References

Estradiol

▶ Tobacco Carcinogenesis

Jose Russo and Irma H. Russo Breast Cancer Research Laboratory, Fox Chase Cancer Center, Philadelphia, PA, USA

References Devesa SS, Blot WJ, Fraumeni JF Jr (1998) Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 83:2049–2053 Montesano R, Hainaut P (1998) Molecular precursor lesions in oesophageal cancer. Cancer Surv 32:53–68 Montesano R, Hollstein M, Hainaut P (1996) Genetic alterations in esophageal cancer and their relevance to etiology and pathogenesis: a review. Int J Cancer 69:225–235 Munoz N, Day N (1997) Esophageal cancer. In: Schottenfeld D, Fraumeni JF (eds) Cancer epidemiology and precvention, 2nd edn. Oxford University Press, Oxford, pp 681–706 Parkin DM, Bray FI, Devesa SS (2001) Cancer burden in the year 2000. The global picture. Eur J Cancer 37(Suppl 8):S4–S66 Pisani P, Parkin DM, Bray F et al (1999) Estimates of the worldwide mortality from 25 cancers in 1990. Int J Cancer 83:18–29

Definition E2 (estradiol) or estradiol-17b is biologically the most active estrogen, and circulating estrogens are mainly originated from ovarian steroidogenesis in premenopausal women and peripheral aromatization of ovarian and adrenal androgens in postmenopausal women.

Characteristics Mechanism of Action It is generally accepted that the biological activities of estrogens are mediated by nuclear ▶ estrogen

Estradiol

receptors (ER) which, upon activation by cognate ligands, form homodimers with another ER–ligand complex and activate the transcription of speciﬁc genes containing the estrogen response elements (ERE). According to this classical model, the biological responses to estrogens are mediated by either of the two estrogen receptors, ERa and ERb. The presence of ERa in target tissues or cells is essential to their responsiveness to estrogen action. The expression levels of ERa in a particular tissue have been used as an index of the degree of estrogen responsiveness. For example, human ▶ breast carcinomas are initially positive for ERa, and their growth can be stimulated by estrogens and inhibited by antiestrogens. ERb can be activated by estrogen stimulation and blocked with antiestrogens. Upon activation, ERb can form homodimers as well as heterodimers with ERa. The existence of two ER subtypes and their ability to form DNA-binding heterodimers suggest three potential pathways of estrogen signaling: via the ERa or ERb subtype in tissues exclusively expressing each subtype or via the formation of heterodimers in tissues expressing both ERa and ERb. In addition, estrogens and antiestrogens can induce differential activation of ERa and ERb to control transcription of genes that are under the control of an ▶ AP-1 element. Sources of Estrogens Circulating estrogens are mainly originated from ovarian steroidogenesis in premenopausal women and peripheral aromatization of ovarian and adrenal androgens in postmenopausal women. Three main enzyme complexes are involved in the synthesis of biologically active estrogen (i.e., estradiol-17b) (Fig. 1): (i) ▶ aromatase that converts androstenedione to estrone, (ii) estrone sulfatase that hydrolyzes the estrogen sulfate to estrone, and (iii) estradiol-17b hydroxysteroid dehydrogenase that preferentially reduces estrone to estradiol-17b in tumor tissues. Role of Estrogens in Human Breast Carcinogenesis There are three mechanisms that have been considered to be responsible for the carcinogenicity of estrogens:

1629

• Receptor-mediated hormonal activity, which has generally been related to stimulation of cellular proliferation, resulting in more opportunities for accumulation of genetic damages leading to ▶ carcinogenesis • ▶ Cytochrome P450 (CYP)-mediated metabolic activation, which elicits direct genotoxic effects both by increasing mutation rates and the induction of ▶ aneuploidy • Reduction of the ▶ DNA repair capability, which allows accumulation of lesions in the genome

Receptor-Mediated Pathway The receptor-mediated activity of estrogen is generally related to induction of expression of the genes involved in the control of ▶ cell cycle progression and growth of human breast epithelium. The biological response to estrogen depends upon the local concentrations of the active hormone and its receptors. The proliferative activity and the percentage of ERa-positive cells are highest in Lob 1 in comparison with the various lobular structures composing the normal breast. Even though it is now generally believed that alterations in the ER-mediated signal transduction pathways contribute to breast cancer progression toward hormonal independence and more aggressive phenotypes, there is also mounting evidence that a membrane receptor coupled to alternative second messenger signaling mechanisms is operational and may stimulate the cascade of events leading to cell proliferation. This knowledge suggests that ERa-negative cells found in the human breast may respond to estrogens through this or other pathways. The biological responses elicited by estrogens are mediated, at least in part, by the production of autocrine and paracrine growth factors from the epithelium and the stroma in the breast. In addition, evidence has accumulated over the last decade supporting the existence of ER variants, mainly a truncated ER and an exon-deleted ER. It has been suggested that expression of ER variants may contribute to breast cancer progression toward hormone independence.

1630

Estradiol H3C CH3

CH3 CH3

Cholesterol

Progesterone

Prognenolone O H3C

H3C O

O H3C

CH3 OH

H3C

O H3C

CH3

H3C

CH3

H3C

O H3C

CH3 OH

H3C

O O Dehydroepiandrosterone sulfate H3C

H3C

O Androstenedione

Aromatase H3C

H3C

Testosterone O

Aromatase H3C

Estradiol-17b hydroxysteroid dehydrogenase

Estrone sulfatase

H3C

O Dehydroepiandrosterone

H3C

17-hydroxyprogesterone

17-hydroxyprognenolone

O HO

O Estrone sulfate

HO Estrone

17β-estradiol

Estradiol, Fig. 1 Steroidogenic pathways leading to the biosynthesis of estrogens (Reprinted from Russo J, Russo IH (2004) Biological and molecular basis of breast cancer. Springer-Verlag, Heidelberg)

Oxidative Metabolism of Estrogen There is evidence that the oxidative catabolism of estrogens, which is mediated by various CYP complexes, constitutes a pathway of their metabolic activation to reactive free radicals and intermediate metabolites that can cause oxidative stress. Estradiol-17b and estrone, which are continuously interconverted by estradiol-17b hydroxysteroid dehydrogenase (or 17b-oxidoreductase), are the two major endogenous estrogens (Fig. 2). They are generally metabolized via two major pathways: hydroxylation at C-16a position and at the C-2 or C-4 positions (Fig. 2). The carbon position of the estrogen molecules to be hydroxylated differs among

various tissues, and each reaction is probably catalyzed by various CYP isoforms. For example, in MCF-7 human breast cancer cells, which produce catechol estrogens in culture, CYP 1A1 catalyzes hydroxylation of estradiol-17b at C-2, C-15a, and C-16a, CYP 1A2 predominantly at C-2, and a member of the CYP 1B subfamily is responsible for the C-4 hydroxylation of estradiol-17b. CYP3A4 and CYP3A5 have also been shown to play a role in the 16a-hydroxylation of estrogens in human. The hydroxylated estrogens are catechol estrogens that will easily be autoxidated to semiquinones and subsequently quinones, both of which are electrophiles capable of covalently

Estradiol

1631

H3C

Estradiol- 17b hydroxysteroid dehydrogenase

17β-estradiol

Estrone Cytochrome P450 enzymes H3C H3C

H3C

O OH

R2 4-hydroxyestrone: R1=H; R2=OH 2-hydroxyestrone: R1=OH; R2=H

HO 16α-hydroxyestrone

R2 4-hydroxyestradiol: R1=H; R2=OH 2-hydroxyestradiol: R1=OH; R2=H

Catechol-O-methyltransferase

H3C

HO Estriol

H3C

OH OH

H3C

R2 4-methoxyestrone: R1=H; R2=OCH3 2-methoxyestrone: R1=OCH3; R2=H

R2 4-methoxyestradiol: R1=H; R2=OCH3 2-methoxyestradiol: R1=OCH3; R2=H

Estradiol, Fig. 2 Biosynthesis and steady-state control of catechol estrogens in human breast tissues (Reprinted from Russo J, Russo IH (2004) Biological and molecular basis of breast cancer. Springer-Verlag, Heidelberg)

binding to nucleophilic groups on DNA via a Michael addition and, thus, serve as the ultimate carcinogenic reactive intermediates in the peroxidatic activation of catechol estrogens. In addition, a redox cycle consisting of the reversible formation of the semiquinones and quinones of catechol estrogens catalyzed by microsomal P450 and CYP-reductase can locally generate superoxide and hydroxyl radicals to produce additional DNA damage. Furthermore, catechol estrogens have been shown to interact synergistically with nitric oxide present in human breast generating a potent oxidant that induces DNA strand breakage. Steady-state concentrations of catechol estrogens are determined by the CYP-mediated

hydroxylations of estrogens and monomethylation of catechols catalyzed by blood-borne catechol O-methyltransferase (Fig. 2). Increased formation of catechol estrogens as a result of elevated hydroxylations of estradiol17b at C-4 and C-16a positions occurs in human breast cancer patients and in women at a higher risk of developing this disease. There is also evidence that lactoperoxidase, present in milk, saliva, tears, and mammary glands, catalyzes the metabolism of estradiol-17b to its phenoxyl radical intermediates, with subsequent formation of superoxide and hydrogen peroxide that might be involved in estrogen-mediated oxidative stress. A substantial increase in base lesions observed in the DNA of invasive ductal carcinoma of the

1632

breast has been postulated to result from the oxidative stress associated with metabolism of estradiol-17b. The breast is an endocrine organ and can synthesize E2 in situ from precursor androgens via the enzyme ▶ aromatase. Breast tissue contains aromatase and produces amounts of E2 that exert biologic effects on proliferation. The effects of local production exceed those exerted in a classical endocrine fashion by uptake of E2 from plasma. Estrogens as Inducers of Aneuploidy Breast cancer is considered the result of sequential changes that accumulate over time. DNA content changes, i.e., loss of heterozygosity (LOH) and ▶ aneuploidy, can be detected at early stages of morphological atypia, supporting the hypothesis that aneuploidy is a critical event driving neoplastic development and progression. Aneuploidy is deﬁned as the gain or loss of chromosomes; it is a dynamic, progressive, and accumulative event that is almost universal in solid tumors. The extensive array of altered gene expression observed in tumors and the numerous altered chromosomes detected by comparative genomic hybridization provide evidence that aneuploidy can disrupt cell homeostatic control. The main question is whether aneuploidy is a consequence of neoplastic development or a cause of neoplastic development. One of the several mechanisms proposed for the development of aneuploidy is the failure to appropriately segregate chromosomes, for example, interference with mitotic spindle dynamics, abnormal centrosome duplication, altered chromosome condensation and cohesion, defective centromeres, and loss of mitotic checkpoints. Functional consequences of centrosome defects may play a role during neoplastic transformation and tumor progression, increasing the incidence of multipolar mitoses that lead to chromosomal segregation abnormalities and aneuploidy. In considering estrogen as a carcinogenic agent, there is evidence that it affects microtubules. The importance of these ﬁndings is magniﬁed with publications that demonstrate women on ▶ hormone replacement treatment that include progesterone have increased ▶ mammographic breast density

Estren-Dameshek Syndrome (= Variant Form)

and increased breast cancer risk than women taking only estrogen. In the center stage of the research endeavor on aneuploidy are the ▶ centrosomes that are organelles that nucleate microtubule growth and organize the mitotic spindle for segregating chromosomes into daughter cells, establishing cell shape and cell polarity. Centrosomes also coordinate numerous intracellular activities, in part by providing a site enriched for regulatory molecules, including those that control cell cycle progression, centrosome and spindle function, and ▶ cell cycle checkpoints. Although the underlying mechanisms for the formation of abnormal centrosomes are not clear, several possibilities have been proposed and implicated in the development of cancer such as alterations of checkpoint controls initiating multiple rounds of centrosome replication within a single cell cycle and failure of cytokinesis, cell fusion, and cell cycle arrest in S-phase uncoupling DNA replication from centrosome duplication.

References Chakravarti D, Mailander P, Franzen J et al (1998) Detection of dibenzo[a, l]pyrene-induced H-ras codon 61 mutant genes in preneoplastic SENCAR mouse skin using a new PCR-RFLP method. Oncogene 16:3203–3210 Hu Y-F, Russo IH, Russo J (2001) Estrogen and human breast cancer. In: Matzler M (ed) Endocrine disruptors. Springer, Heidelberg, pp 1–26 Jefcoate CR, Liehr JG, Santen RJ et al (2000) Tissuespeciﬁc synthesis and oxidative metabolism of estrogens. In: Cavalieri E, Rogan E (eds) JNCI monograph 27: estrogens as endogenous carcinogens in the breast and prostate. Oxford Press, Oxford, pp 95–112 Liehr JG (2000) Is estradiol a genotoxic mutagenic carcinogen? Endocr Rev 21:40–54 Russo J, Russo IH (2000) Biological and molecular basis of breast cancer. Springer, Heidelberg

Estren-Dameshek Syndrome (= Variant Form) ▶ Fanconi Anemia

Estrogen Receptor

1633

development at a high rate of molecular biology techniques, to have the ﬁrst cDNA encoding estrogen receptor (a) cloned by the group of Chambon in Strasbourg. In 1996, when most of nuclear receptors had already been cloned, a great surprise arose with the fortuitous isolation of a second estrogen receptor (b) cloned from a ▶ Prostate Cancer Clinical Oncology library.

Estrogen Receptor Gwendal Lazennec INSERM, Montpellier, France

Synonyms ER; ESR1; ESR2; Estrogen receptor alpha; Estrogen receptor beta

Definition Estrogen receptors are represented by two main members (estrogen receptor alpha, ERa, and estrogen receptor beta, ERb), which bind estrogens. These receptors are mainly nuclear, even though membrane receptors are also suspected to exist. Nuclear estrogen receptors belong to a large family of nuclear receptors, which are ligandactivated transcription factors able to modulate the expression of different genes.

Characteristics History The mediators of estrogens, namely, estrogen receptors, remained elusive until the synthesis of radiolabeled ▶ estradiol by Jensen and Jacobsen in 1960, which allowed the identiﬁcation of such receptors. But it took more than 20 years, with the

Estrogen Receptor, Fig. 1 Schematic representation of ERa and ERb proteins

Molecular Mechanisms of Action The genes coding for estrogen receptors ERa and ERb are located on two distinct chromosomes (ERa on chromosome 6q25.1 and ERb on chromosomes 14q22–14q24). ERa and ERb proteins have a respective size of 595 and 530 amino acids (Fig. 1). As indicated by their membership to nuclear receptors, estrogen receptors are mainly present in the nucleus of cells, even in the absence of estrogens. In addition, a small pool of ERs localize to the plasma membrane and signal mainly through coupling to G-proteins, but we will not discuss this issue here. Upon binding to estrogens, nuclear estrogen receptors dimerize, bind coactivators, and interact with DNA or DNA-bound proteins to modulate the transcription of estrogen target genes (Fig. 2). Estrogen receptors share the common structure of nuclear receptors composed of 6 domains (A, B, C, D, E, and F) (Fig. 1). The C domain (DBD) is responsible for DNA binding. Two distinct synergistic transcriptional activation functions (AFs) have been identiﬁed: the ligand-independent AF-1 located in the N-terminal A/B region and the ligand-dependent AF-2 encompassing region

1 hERα

A/B

553 E

148 214 304

1 hERβ

263 302

180

A/B

500

Ligand Binding AF-1

530 F

DNA binding

Transactivation

595

AF-2

1634

Estrogen Receptor

Estrogen Receptor, Fig. 2 Mechanism of action of estrogen receptors. CoA coactivator, E2 ▶ estradiol, ERE estrogen responsive element

ERE

GENE

CoA

AAA

E2 + ER

Effect

E (the LBD). Both ER AF-1 and AF-2 were found to act in promoter context and cell-speciﬁc fashions. Coactivators are mainly interacting with the E domain, but an increasing number of coactivators are also able to interact with the AB domain and the DBD. Estrogen receptors modulate the transcription either by binding to classical estrogen receptor element (ERE) or by interacting with ▶ AP-1, SP-1, or NF-kB-bound transcription factors. Many EREs consist of two inverted, palindromic half sites and are frequently present in the upstream regulation region of estrogen target genes. On the majority of ERE promoters, ERb is less potent activator than ERa. Interestingly, ERa and ERb form preferentially heterodimers rather than homodimers when they are both expressed in the same cell, which enables ERb to decrease ERa transcriptional potential. Both estrogen-bound receptors mediate gene transcription similarly through the ERE pathway but have opposite effects when signaling through the AP-1 pathway. Indeed, ERa activates AP-1 gene transcription, whereas ERb inhibits this pathway. Moreover, tamoxifen-liganded ERa is inactive on AP-1 elements, but ERb is able to activate the transcription response. What Have We Learned from ER Knockout Mice? Disruption of the ERa gene (aERKO animals) is not lethal, but rather the animals develop normally and exhibit a life span comparable to their wild-

Protein

type littermates. aERKO mice exhibit several abnormalities and deﬁciencies, most notable of which are the phenotypic syndromes that result in infertility in both sexes. The mammary phenotype of aERKO female mice demonstrates that embryonic mammary gland development is independent of ERa, but ERa is required for ductal elongation during puberty and complete mammary gland development in the mature mouse. The female reproductive tract of aERKO undergoes normal pre- and neonatal development but is insensitive to estrogens during adulthood. Ovaries undergo normal pre- and neonatal development, but are anovulatory during adulthood, exhibit multiple hemorrhagic cysts, and have no corpora lutea. There is also a 30–40% incidence of ▶ Ovarian Cancer tumors by 18 months of age. Male reproductive tract of aERKO displays dilation of rete testis, atrophy of the seminiferous epithelium, and decreased sperm counts, leading to infertility. Mice lacking ERb (bERKO) display a much less pronounced phenotype than aERKO. They develop normally and are indistinguishable grossly and histologically from their littermates. Sexually mature bERKO females are fertile and exhibit normal sexual behavior, but they produced substantially fewer litters as well as signiﬁcantly less number of pups per litter when compared with their wild-type littermates. This reduction in fertility is the result of reduced ovarian efﬁciency. The mutant females have normal breast

Estrogen Receptor

development and lactate normally. However, lactating glands display alveoli which are larger, and there is less secretory epithelium in ERb than in wild-type mice. Ovaries undergo normal pre- and neonatal development, but do not exhibit normal frequency of spontaneous ovulations during adulthood, and exhibit a severely attenuated response to superovulation treatment with reduced number of oocytes and multiple trapped preovulatory follicles. In addition, male bERKO mice develop prostate hyperplasia. ER disruption has also distinct effects on sexual behavior of the animals. aERKO male mice, although they rarely ejaculate and are infertile, show almost normal levels of mounts and just reduced levels of intromissions. In contrast, all three components of sexual behaviors are present and robust in bERKO males. Aggressive behavior is greatly reduced in aERKO male mice. In particular, male-typical offensive attacks are almost completely abolished, whereas lunge and bite attacks are still present. On the other hand, the aggressive behavior of bERKO is not reduced but rather elevated depending on age and social experiences. Involvement of Estrogen Receptors in Physiology and Cancer Estrogens play a critical role in many physiologic processes, including reproduction, cardiovascular health, bone integrity, immune system, cognition, and behavior. But estrogens are also involved in many pathologies including osteoporosis, endometriosis, or neurodegenerative diseases, but also cancer (breast, ▶ Ovarian Cancer, ▶ Endometrial Cancer, ▶ Prostate Cancer Clinical Oncology, ▶ Colorectal Cancer Premalignant Lesions). ERa and ERb display disparity in their tissue distribution. Studies on rat have shown that ERa mRNA is predominant in the uterus, mammary gland, testis, pituitary, liver, kidney, heart, and skeletal muscle, whereas ERb transcripts are signiﬁcantly expressed in the ovary and prostate. In humans, ERa and ERb are both present in the brain, breast, cardiovascular system, and bone. In addition, ERa is much more abundant than ERb in the uterus and liver, whereas the opposite is true for ERb in the ovary, prostate, and gastrointestinal tract. This differential distribution

1635

suggests that the two receptors could play distinct roles. The most characterized cancer location involving estrogen receptors is deﬁnitely the breast. The cumulative exposure of the breast epithelium to estrogens is one of the ﬁrst risk factors for ▶ breast cancer incidence. Estrogens are not only believed to be involved in the development and growth of breast tumors but also the late course of the disease, in their ▶ metastasis. The potential carcinogenic properties of estrogens are also suspected for endometrial and ▶ ovarian cancers. Two current hypothesis could account for estrogen carcinogenic action. First, estrogen metabolism could lead to the production of genotoxic products which directly alter DNA integrity. Second, by stimulating cell proliferation, estrogens could increase the number of cell divisions and thus the risk of replication errors, leading possibly to selective advantages for the mutated cells, which would exhibit defects in terms of apoptosis, proliferation, and DNA repair. It is interesting to note that this second hypothesis can be further divided into two subhypothesis. Indeed, ERa is generally expressed at low levels in normal breast epithelium cells (approximately 10–20% depending on the phase of menstrual cycle). Moreover, in the normal breast, it seems that there is a discordance between the cells which express ERa and the one that proliferate. For this reason, some people believe that estrogens act directly on epithelial cells to stimulate the proliferation, and others suggest that estrogens act on stromal cells, which in turn secrete soluble growth factors stimulating epithelial cell growth. In contrast to the situation observed in the normal breast, breast cancer cells express frequently ERa, and the same cells are actively proliferating. Breast cancers are ﬁrst divided into two subtypes based on whether or not the tumor cells express ERa. ERa-positive cancers represent about two-thirds of all breast cancers. The reason for analyzing ERa status is based on the facts that estrogens constitute one of the major mitotic signals for ERa-positive breast cancers. Moreover, antiestrogen drugs such as tamoxifen are commonly used as ﬁrst-line therapy against ERa-positive breast cancers. Unfortunately though,

1636

Estrogen Receptor Alpha

the use of tamoxifen is rarely associated with long-term remissions in metastatic diseases. In contrast to the situation described for ERa, it has been reported by several groups that ERb expression was lower in cancerous tissues compared with normal tissues. This is true for breast, ▶ Ovarian Cancer, ▶ Prostate Cancer Clinical Oncology, lung, and colon cancers, suggesting that this could be a general trend. Moreover, the exogenous delivery of ERb to breast and prostate cancer leads to an inhibition of proliferation of tumor cells both in vitro and in vivo. In addition, ERb is able to reduce the ▶ invasion potential of tumor cells. Moreover, ERb knockout animals display prostate hyperplasia suggesting the possible proliferation gatekeeper role of ERb. Unlike ERb, ERa levels are frequently correlated to tumor grade. Patients whose tumors express ERa have a longer interval to recurrence and an improved survival. Concerning ERb, the prognostic value of this receptor is more controversial, even though several studies have shown that ERb was associated with a longer disease-free survival. At the present time, it is not completely understood why tumor cells in some instances lose the expression of ERa or ERb. Several studies suggest that promoter hypermethylation of both genes could silence the activity of ERa and ERb promoters. On the other hand, the hyperexpression of ERa in the vast majority of breast cancers is not at all clariﬁed. In addition, the role of the numerous splice variants identiﬁed for ERa and ERb transcripts is not well established. The alternative splicing of ERa mRNA occurs frequently in breast tumors, but the expression of these transcripts is also usually conserved not only in primary tumors but also in distant metastases. In the same line, several single nucleotide polymorphisms (SNPs) in ERa gene have been associated either with an increased or a decreased risk of breast cancer, whereas the situation is much less clear for SNPs found in ERb gene.

▶ Colorectal Cancer Premalignant Lesions ▶ Endometrial Cancer ▶ Estradiol ▶ Invasion ▶ Metastasis ▶ Ovarian Cancer ▶ Prostate Cancer Clinical Oncology

Cross-References

Estrogen Receptor, Progesterone Receptor, and HER-2 Negative

▶ AP-1 ▶ Breast Cancer

References Deroo BJ, Korach KS (2006) Estrogen receptors and human disease. J Clin Invest 116:561–570 Jensen EV (2005) The contribution of “alternative approaches” to understanding steroid hormone action. Mol Endocrinol 19:1439–1442 Levin ER (2005) Integration of the extranuclear and nuclear actions of estrogen. Mol Endocrinol 19:1951–1959

See Also (2012) Coactivators. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 886. doi:10.1007/978-3-642-16483-5_1238 (2012) Estrogens. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1333. doi:10.1007/978-3-642-16483-5_2019 (2012) Nuclear Receptor. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 2571. doi:10.1007/978-3-642-16483-5_4157 (2012) Transcription Factor. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 3752. doi:10.1007/978-3-642-16483-5_5901

Estrogen Receptor Alpha ▶ Estrogen Receptor

Estrogen Receptor Beta ▶ Estrogen Receptor

▶ Triple-Negative Breast Cancer

Estrogen Signaling

Estrogen Signaling Rosamaria Lappano and Marcello Maggiolini Department of Pharmacy and Health and Nutritional Sciences, University of Calabria, Rende, Italy

Definition Estrogen signaling refers to the transduction pathways which mediate the multifaceted biological actions of estrogens. Estrogen signaling triggers both rapid and genomic responses mainly through: (1) the nuclear estrogen receptor (ER)a and ERb which act as transcription factors, (2) the

Estrogen Signaling, Fig. 1 Schematic representation of the estrogen signaling. The binding of 17b-estradiol (E2) to the classical ER modulates gene transcription by binding to speciﬁc DNA sequences (estrogen response elements, ERE) and other transcription factors complexes (TR) like AP-1 and Sp-1. ER can also be activated in a

1637

membrane-localized ERs, and (3) the G proteincoupled estrogen receptor (GPER), formerly named GPR30 (Fig. 1).

Characteristics Estrogens, the main female sex steroids, regulate many cellular processes involved in the growth, differentiation, and function of the reproductive systems and exert various biological effects in the cardiovascular, musculoskeletal, immune, and nervous systems in both men and women. In addition to their role in physiology, estrogens may inﬂuence the development of hormonedependent tumors, like breast, endometrial, ovarian, thyroid, and prostate cancer. Endogenous

ligand-independent manner by growth factors. In addition, E2 activates the GPR30/GPER-mediated signaling, which involves the epidermal growth factor receptor (EGFR) and diverse transduction pathways leading to gene transcription

1638

estrogens consist of estrone (E1), estriol (E3), and 17b-estradiol (E2) which is commonly recognized as the most potent female sex hormone. The biological responses to estrogens are mainly mediated by the ERs that initiate a complex array of cellular events upon ligand binding. Generally, these responses are divided into two categories: (i) activation of gene transcription referred to as genomic responses, which requires at least some hours, and (ii) rapid events that occur within seconds or minutes once cells are exposed to estrogens. Although these quick effects are referred to as nongenomic responses, they can lead to gene transcription confounding the aforesaid distinction. ER-Mediated Signaling Both the genomic and rapid signaling events initiated by estrogens were ascribed solely to the classical ERs. Estrogens bind to and activate the two thoroughly characterized members of the nuclear receptor superfamily, ERa and ERb, which are products of two separate genes. Like all steroid receptors, the ERs consist of structurally and functionally distinct domains. The N-terminal domain (A/B region) contains the activating function 1 (AF-1) and several phosphorylation and sumoylation sites. The DNA binding domain (DBD) (C region), which is highly conserved among all nuclear receptors, is responsible for the binding to speciﬁc DNA sequences and is involved in protein dimerization. The D region is a hinge domain which contains the nuclear localization signal and different sites involved in posttranslational modiﬁcations. The C-terminal domain (E/F region) corresponds to the activating function 2 (AF-2), which includes the dimerization and the ligand binding domain (LBD). The E/F region is also responsible for the binding to chaperones (heat shock proteins 70 and 90). ERa and ERb, which share a high degree of sequence homology except in their N-terminal domains, have similar afﬁnity for E2 and bind to the same DNA response elements. However, triggering distinct gene transcriptions and signaling pathways, ERa and ERb mediate speciﬁc responses and exert opposite effects on important cellular processes, including proliferation, migration, and

Estrogen Signaling

apoptosis. Upon ligand activation, ERa and ERb regulate the transcription through the binding to estrogen-responsive elements (ERE) located within the regulatory regions of target genes. The consensus ERE consists of a 5-bp palindrome with a 3-bp spacer (GGTCAnnnTGACC), although many natural EREs deviate substantially from the consensus sequence. For instance, the well-studied estrogen-responsive genes pS2, cathepsin-D, and progesterone receptor do not contain the perfect consensus ERE sequence. The cell-speciﬁc transcriptional response to estrogens depends on multiple factors, like the characteristics of gene promoters and coregulatory proteins. Coactivators which generally do not bind to the DNA can enhance the transcriptional activity once recruited to the promoter of target genes, whereas corepressors negatively regulate gene expression. In addition, posttranslational modiﬁcations, namely, acetylation, glycosylation, myristoylation, nitrosylation, palmitoylation, phosphorylation, sumoylation, and ubiquitination, have been reported to regulate ER functions. Estrogens interacting with other transcription factors modulate the expression of genes whose promoters do not harbor EREs. For instance, the stimulating protein 1 (Sp-1) binds to the estrogenresponsive DNA regulatory regions, then ER enhances the binding of Sp1 to the DNA contributing to coactivator recruitment. Moreover, ERa and ERb interact with the fos/jun transcription factor complex on the activator protein 1 (AP-1) sites to regulate gene expression. Ligand-independent ER activation occurs by epidermal growth factor (EGF), insulin, insulinlike growth factor 1 (IGF-1), and transforming growth factor (TGF)-b. In addition, heregulin, interleukin 2, and dopamine can also modulate ER activity. Noteworthy, the growth factoractivated pathways contribute to the hormoneindependent tumor progression toward a more aggressive phenotype. As mentioned above, estrogens induce biological effects in a rapid manner that cannot be accounted for genomic actions. Intriguingly, these effects which are insensitive to inhibitors of transcription and translation occur via

Estrogen Signaling

membrane-localized ERs. Estrogens activate several membrane-initiated transduction cascades, like cAMP/PKA and PKC pathways, endothelial nitric oxide synthase (eNOS) promoting NO release, phospholipase (PL)C-dependent inositol trisphosphate (IP3) production, calcium inﬂux, MAPK and p38 signaling, and phosphatidylinositol-3-kinase (PI3K)/AKT pathway. These estrogen-activated signals which occur in seconds or minutes are typically associated with plasma membrane receptors. In this vein, the palmitoylation of steroid receptors plays a key role as this biological process is required for their membrane localization and the rapid transduction of signals into cell biology. It should be also mentioned the ability of ligandactivated ERa in interacting directly with the IGF-1 receptor, which in turn triggers the MAPK-dependent pathway. As it concerns the interaction of membrane ERa with ErbB2 (HER-2/neu), it was involved in the resistance to tamoxifen-induced apoptosis in breast cancer cells. In addition, E2-activated ERa has been reported to transactivate the EGF receptor by a mechanism which involves the activation of G proteins, Src kinase, and matrix metalloproteinases that in turn activate the MAPK and AKT transduction pathways. GPER-Mediated Signaling In the last years, a member of the seventransmembrane G protein-coupled receptor family, named GPER/GPR30, has been implicated in mediating both rapid and transcriptional responses to estrogens. GPER exhibits many of the expected characteristics of a membrane estrogen receptor, as it binds to estrogens generating biochemical signals mediated by plasma membrane-associated enzymes. For instance, GPER stimulates adenylyl cyclase through G proteins leading to increased cAMP levels. In addition, GPER signaling occurs via the EGFR transactivation which involves a Gbg-dependent pathway and Src family tyrosine kinases. In particular, ligand-stimulated GPER activates metalloproteinases and induces the extracellular release of heparin-bound epidermal growth factor (HB-EGF), which binding to EGFR triggers

1639

downstream transduction pathways, like ERK and PI3K signaling cascades as well as intracellular calcium mobilization. Likewise, GPER regulates the transcription of several genes which play relevant roles in diverse biological processes such as cell growth, migration, and differentiation. As GPER expression and function are regulated by EGF and insulin/IGF system, it may be considered as a further player involved in the cross-talk among estrogen signaling and these growth factors, toward important biological responses in both normal and cancer cells. Estrogen Signaling in Cancer Estrogens inﬂuence many physiological processes in mammals, including reproduction, cardiovascular health, bone integrity, cognition, and behavior. Given this widespread role for estrogens in human physiology, it is not surprising that these steroids are also implicated in the development and progression of numerous diseases, including osteoporosis, neurodegenerative and cardiovascular diseases, insulin resistance, endometriosis, and a variety of tumors. In particular, estrogens are critical mediators of breast cancer initiation, progression, and metastasis. At least two current hypotheses may explain this relationship. In the ﬁrst, estrogens binding to ERa stimulate the proliferation of mammary cells, increasing the target cell number within the tissue. The increase in cell division and DNA synthesis elevates the risk for replication errors, which may result in the acquisition of detrimental mutations that disrupt normal cellular processes such as apoptosis, DNA repair, or cellular proliferation. In the second hypothesis, the metabolism of estrogens leads to the production of genotoxic products that could directly damage DNA resulting in mutations and other adverse effects. In addition to breast cancer, other malignancies including ovarian, endometrial, colon, pancreatic, prostate, and thyroid tumors have been associated with estrogens and estrogen receptor-mediated action. On the basis of the main role elicited by estrogens in cancer and the existence of multiple estrogen-activated transduction pathways, the inhibition of estrogen signaling still represents one main strategy for targeting hormone-

1640

Estrogen Synthase

dependent cancer, like breast, ovarian, and endometrial tumors. Indeed, endocrine therapy by using antiestrogens such as the selective estrogen receptor modulator (SERM) tamoxifen and aromatase inhibitors, which ablate peripheral estrogen synthesis, has been shown to signiﬁcantly improve disease-free survival. Despite the anticancer effects, initial or acquired resistance to endocrine therapies frequently occurs still representing a main concern. Cumulatively, a better understanding on the mechanisms by which estrogens may drive cancer progression is needed in order to identify novel biological targets and therapeutic strategies in hormone-sensitive tumors. In this regard, the effects of estrogens mediated by GPER should be considered in the development of novel SERMs, and the use of GPER antagonists could be included among the current pharmacological approaches in these types of cancer.

modulators: implications for new drug discovery in breast cancer. Curr Opin Pharmacol 10:620–628 Prossnitz ER, Barton M (2011) The G-protein-coupled estrogen receptor GPER in health and disease. Nat Rev Endocrinol 7:715–726 Thomas C, Gustafsson JÅ (2011) The different roles of ER subtypes in cancer biology and therapy. Nat Rev Cancer 11:597–608

References

Definition

Ascenzi P, Bocedi A, Marino M (2006) Structure-function relationship of estrogen receptor alpha and beta: impact on human health. Mol Aspects Med 27:299–402 De Marco P, Cirillo F, Vivacqua A, Malaguarnera R, Belﬁore A, Maggiolini M (2015) Novel Aspects Concerning the Functional Cross-Talk between the Insulin/IGF-I System and Estrogen Signaling in Cancer Cells. Front Endocrinol (Lausanne) 6:30 Deroo BJ, Korach KS (2006) Estrogen receptors and human disease. J Clin Invest 116:561–570 Hammes SR, Levin ER (2007) Extranuclear steroid receptors: nature and actions. Endocr Rev 28:726–741 Katzenellenbogen BS, Choi I, Delage-Mourroux R, Ediger TR, Martini PG, Montano M, Sun J, Weis K, Katzenellenbogen JA (2000) Molecular mechanisms of estrogen action: selective ligands and receptor pharmacology. J Steroid Biochem Mol Biol 74:279–285 Lappano R, De Marco P, De Francesco EM, Chimento A, Pezzi V, Maggiolini M (2013) Cross-talk between GPER and growth factor signaling. J Steroid Biochem Mol Biol 137:50–6 Lewis JS, Jordan VC (2005) Selective estrogen receptor modulators (SERMs): mechanisms of anticarcinogenesis and drug resistance. Mutat Res 591:247–263 Maggiolini M, Picard D (2010) The unfolding stories of GPR30, a new membrane-bound estrogen receptor. J Endocrinol 204:105–114 McDonnell DP, Wardell SE (2010) The molecular mechanisms underlying the pharmacological actions of ER

Estrogens are steroid sex hormones produced chieﬂy in the ovary and responsible for development and function of the female reproductive tissues such as the uterus and mammary gland. Smaller amounts are also produced by the testis in the male and the adrenal gland in both sexes and contribute to maintenance of bone density and function of cardiovascular and neurological tissues. Estrogenic hormones play a role in both the genesis and treatment of several types of cancer. In general, estrogen-related tumors are those involving tissues of the female reproductive tract, although estrogens produce liver cancers in the hamster and are probably a factor in their occurrence in humans.

Estrogen Synthase ▶ Aromatase and Its Inhibitors

Estrogenic Hormones Elwood V. Jensen National Institute of Health, Bethesda, MD, USA

Characteristics Etiology Carcinogenesis is known to be a multistep process (▶ multistep development), involving both initiation (alteration of DNA) and promotion

Estrogenic Hormones

(proliferation of the altered cells). It is generally agreed that the principal effect of estrogenic hormones is on the promotion stage, especially in tissues where growth and function are normally regulated by estrogen. It has been controversial whether estrogens, especially in physiological amounts, also cause genetic changes in a manner similar to the action of chemical carcinogens such as dimethylbenzanthracene or nitrosourea. Breast Cancer

The human malignancy most studied in relation to estrogen is ▶ breast cancer, both because of its high incidence and because its involvement with estrogens is especially striking. Much evidence indicates that estrogenic hormones play an important role in the appearance of mammary cancer, both in experimental animals and in the human. It has long been known that early menarche and/or late menopause increases the risk of ▶ breast cancer and that artiﬁcial menopause induced by ovariectomy or radiation reduces the risk, suggesting a cumulative effect of the number of ovulatory cycles on the incidence of the disease. It is also known that full-term pregnancy before the age of 20 years confers a signiﬁcant protective effect, whereas nulliparous women have an increased susceptibility to breast cancer, but the basis of this phenomenon is not clear. The effect of hormone replacement therapy has been the subject of much investigation and some controversy, but from studies, it appears that for breast cancer, the risk from unopposed estrogen is small, and the addition of progestin to the regimen makes little difference. The putative involvement of estrogens in the genesis of breast cancer has afforded an approach to its prevention that is currently under investigation. The antiestrogen tamoxifen was shown to prevent both the induction of mammary tumors by dimethylbenzanthracene in the rat and the appearance of cancer in the contralateral breast after mastectomy in the human. Following this, a large clinical trial has demonstrated that this agent, and related antiestrogens, can lower the incidence of breast cancer in women who are at high risk for developing this disease.

1641

Uterine, Cervical, and Ovarian Cancer

Exposure to estrogen, unopposed by progestin, is a major factor in the occurrence of cancer of the uterine endometrium. Unopposed estrogen replacement therapy for more than 5 years results in an elevated risk of endometrial cancer, which persists for several years after the medication has been discontinued. The addition of progestin to the estrogen replacement regimen substantially reduces this risk. In comparison with cancers of the breast and uterus, involvement of estrogens in the etiology of cervical neoplasia is less clear. Most studies have been of the effect of oral contraceptives and generally have shown some correlation between the prolonged use of these agents and the incidence of cervical cancer. As in the case of breast cancer, late menopause, which gives rise to a longer period of ovulatory activity, results in an increased risk of ovarian cancer. Similarly, pregnancy and the use of oral contraceptives, which decrease the number of ovulatory cycles, are protective. Vaginal Adenocarcinoma

During the period 1945–1955, large doses of estrogenic hormones were often administered to pregnant women with a history of miscarriage in the belief that this would protect against spontaneous abortion. Because orally active steroidal hormones were not available at that time, a synthetic estrogen called ▶ diethylstilbestrol (DES) was used. In the early 1970s, a previously rare cancer, clear cell ▶ adenocarcinoma of the vagina, began to appear in the daughters born to the DES-treated mothers, leading to the impression that estrogens in general, and DES in particular, are carcinogens and should not be used for human medication. However, the amounts administered (0.5–1.0 g) were 5,000–10,000 times the hormonally active dose, and the cancers produced were not in those persons receiving the estrogen but in their offspring, indicating that this is an in utero phenomenon and the action of the hormone is better described as teratogenic than carcinogenic. Longer follow-up has demonstrated a slightly increased incidence of breast cancer among the DES-treated mothers, but not what

1642

might be expected from such high doses of a true carcinogen. Genital abnormalities were produced in the male offspring, in keeping with a teratogenic phenomenon. Hepatoma

In certain animal species such as the hamster, liver cancer can be induced by the simple administration of estrogens. In the Western world, primary liver cancer in humans is a relatively rare phenomenon, except for individuals with cirrhosis. However, it is a major cause of death in Asia and South Africa. The introduction of oral contraceptives has led to an increased incidence of liver tumors after long-term use of preparations containing substantial amounts of estrogen. It has been reported that hepatomas can arise from the use of prolonged ▶ diethylstilbestrol (DES) therapy for prostatic cancer. Therapy When cancer occurs in tissues where growth and function depends on estrogenic hormones, in some instances, the malignant cells retain their hormone dependency, while others lose the need for continued stimulation. It is not clearly established what is the exact basis for hormone dependency and whether escape from this regulation takes place on neoplastic transformation or during subsequent tumor progression. For cancers that retain hormone dependency, depriving them of estrogen provides an effective palliative treatment, less traumatic than cytotoxic chemotherapy, which is the only recourse for the majority of non-hormone-dependent metastatic cancers. Breast Cancer

Hormone-dependent mammary tumors can be deprived of supporting estrogen either by removing the organs in which the hormone is produced, administration of substances that inhibit estrogen biosynthesis, or by giving a so-called antiestrogen that prevents the hormone from exerting its growth-stimulating effect in the cancer cells. More than a century ago, before it was known what estrogens are or that they are produced in the ovary, it was found that removal of the ovaries

Estrogenic Hormones

from young women with advanced breast cancer caused remission of the disease in some patients. But the majority of breast cancers occur in postmenopausal women, where the ovaries are no longer functional, and it was long suspected that in the older patient the adrenal glands are the source of supporting estrogen. When cortisone became available, ﬁrst for the treatment of inﬂammatory diseases, it became possible to remove the adrenal glands or the pituitary gland which controls them and maintain the patient on glucocorticoid replacement therapy. Subsequent clinical experience showed that about one-third of all the patients have mammary tumors that undergo remission when deprived of supporting hormone by any of these procedures, and endocrine ablation (the surgical removal of organs that produce hormones) became ﬁrst-line therapy for advanced breast cancer, especially after methods were developed to predict which patients will or will not respond to endocrine manipulation. When it was demonstrated that estrogens, like steroid hormones in general, exert their physiological actions in combination with speciﬁc receptor proteins, it was established that patients whose tumors contain low or negligible amounts of ▶ estrogen receptor (ER) rarely respond to any kind of endocrine therapy, whereas most, but not all, patients with ER-rich cancers beneﬁt from such treatment. Determination of estrogen receptor on excised breast cancer specimens, either by immunological or hormone-binding procedures, is now a standard clinical practice. As an alternative to endocrine ablation, hormone deprivation can be effected by inhibiting the enzymes involved in estrogen biosynthesis. This approach has the advantage that it eliminates not only estrogen arising from the ovary or adrenal gland but also that which, in some cases, appears to be produced by the tumor itself. The ﬁrst successful agent for this purpose was aminoglutethimide, which inhibits the key enzyme, aromatase, but its clinical utility has been limited by undesirable side effects. Several improved compounds have been developed including fadrozole, letrozole, vorozole, and arimidex, which show promise of increased activity with reduced toxicity.

Estrogen-Replacement Therapy

With the advent of ▶ tamoxifen, the ﬁrst antiestrogen to be tolerated on prolonged administration, this reversible treatment has largely replaced the irreversible endocrine ablation as ﬁrst-line therapy for endoplasmic reticulum (ER)-rich breast cancers. Although there are side effects from prolonged treatment as well as a slightly increased risk of endometrial cancer, the beneﬁts greatly outweigh the drawbacks. Tamoxifen and related nonsteroidal compounds such as toremifene, raloxifene, and droloxifene show curious pharmacology in that depending on species, tissue, and dose, they can act either as stimulators or inhibitors. A limitation of tamoxifen therapy is the development in many patients of an “acquired tamoxifen resistance” in which the medication no longer inhibits but actually stimulates the growth of the cancer. Steroidal antiestrogens such as Faslodex (ICI 182,780) and RU 58668 have been developed, which show only inhibitory (antagonist) but not stimulatory (agonist) action. Uterine and Cervical Cancers

Because growth and development of the uterus are stimulated by estrogen, attempts have been made to treat ▶ endometrial cancer with ▶ tamoxifen in a manner analogous to mammary cancer, but the response rate is low and variable. The most widely used hormonal therapy for this malignancy is treatment with ▶ progestin. Cervical cancer is especially sensitive to radiation and does not metastasize aggressively, and so surgery and radiotherapy are the usual therapeutic procedures, and endocrine therapy has found little application.

Cross-References ▶ Adenocarcinoma ▶ Breast Cancer ▶ Diethylstilbestrol ▶ Endometrial Cancer ▶ Estrogen Receptor ▶ Metastatic Colonization ▶ Multistep Development ▶ Progestin ▶ Tamoxifen

1643

References Holland JF, Frei E III, Bast RC Jr et al (2000) Cancer medicine, 5th edn. BC Decker, Hamilton Jensen EV (1999) Oncology. In: Oettel M, Schillinger E (eds) Handbook of experimental pharmacology vol 135/II, estrogens and antiestrogens II pharmacology and clinical application. Springer, Berlin/Heidelberg/New York, pp 195–203 Li JJ, Li SA, Gustafsson J-Å et al (1996) Hormonal carcinogenesis II. Springer, Berlin/Heidelberg/New York Lindsay R, Dempster DW, Jordan VC (1997) Estrogens and antiestrogens: basic and clinical aspects. Lippincott-Raven, Philadelphia/New York Parl FF (2000) Estrogens, estrogen receptor and breast cancer. Ios Press, Amsterdam

See Also (2012) Antiestrogens. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 209. doi:10.1007/978-3-642-16483-5_318 (2012) Aromatase. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 276. doi:10.1007/978-3-642-16483-5_394 (2012) Biosynthesis. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 415. doi:10.1007/978-3-642-16483-5_647 (2012) Cytotoxic chemotherapy. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1058. doi:10.1007/978-3-642-16483-5_1499 (2012) Endocrine ablation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1223. doi:10.1007/978-3-642-16483-5_1871 (2012) Endometrium. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1239. doi:10.1007/978-3-642-16483-5_1886 (2012) Endoplasmic reticulum. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1240. doi:10.1007/978-3-642-16483-5_1887 (2012) Glucocorticoids. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 1558. doi:10.1007/978-3-642-16483-5_2429 (2012) Hormone replacement therapy. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, pp 1733-1734. doi:10.1007/978-3-642-164835_2815 (2012) Mastectomy. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 2177. doi:10.1007/978-3-642-16483-5_3548 (2012) Teratogenic. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 3651. doi:10.1007/978-3-642-16483-5_5731

Estrogen-Replacement Therapy ▶ Hormone Replacement Therapy

1644

Characteristics

ET ▶ Endothelins

ET-2 ▶ Endothelins

ET-743 ▶ Trabectedin

Eta-1 ▶ Osteopontin

Ether à-go-go Potassium Channels Javier Camacho Department of Pharmacology, Centro de Investigación y de Estudios Avanzados del I.P.N., Mexico City, D.F., Mexico

Synonyms Eag; Eag1; KCNH1; Kv10.1

Definition Ether à-go-go potassium channels are transmembrane proteins opening in response to changes in membrane potential and allowing the movement of potassium ions.

Ion transport is crucial for maintaining proper cellular function; this important task is performed by several ▶ membrane transporters including ▶ ion channels. Among other functions, ion channels play key roles in neurotransmission, muscle contraction, metabolism, sensory transduction, ▶ apoptosis, and cell cycle progression. Because of their pivotal role in cellular function, altered expression and/or activity of ion channels leads to several diseases including cardiac arrhythmias, diabetes, and epilepsy, turning these proteins into major drug targets. Participation of ion channels in cell proliferation and cell death makes them not only fundamental elements for the understanding of ▶ cancer but also potential clinical tools both for diagnosis and therapy of cancer. Ether à-go-go (Eag) comprises a family of voltage-gated potassium channels opening in response to changes in membrane potential. Some members of the Eag family, namely, human Eag1 (h-Eag1) and human-Eag-related gene (h-Erg), are linked to major diseases. h-Erg channels have an essential role in the cardiac action potential; h-Erg mutations produce the long Q-T syndrome type 2 leading to cardiac arrhythmias and eventually – in many cases – death of the patient. A huge number of very different drugs inhibit h-Erg channels producing cardiac arrhythmias as a nondesirable side effect; thus, h-Erg has become an intensively studied ion channel when designing new drugs. In addition, both Eag1 and Erg have been found to be overexpressed in a variety of tumors. h-Erg channels have been found in leukemic cells and biopsies from ▶ colorectal cancer and ▶ endometrial cancer; hence, h-Erg expression has been suggested as a molecular marker for human neoplastic hematopoietic cells and a potential prognostic factor for colorectal cancer. h-Eag1 is the other member of the EAG family involved in cancer. In addition to being overexpressed in many human tumors, h-Eag1 possesses oncogenic properties and has a more restricted distribution in normal healthy tissues in comparison with h-Erg channels, and speciﬁc inhibition of h-Eag1 gene expression reduces tumor cell proliferation. In the

Ether à-go-go Potassium Channels

following, only oncogenic Eag1 channels will be discussed. Oncogenic Potential of h-Eag1 Channels Eag1 channels display oncogenic properties. Some characteristics of tumor cells are that they are able to grow in very low serum concentration, lose contact inhibition, and induce tumor formation when injected into immune-deﬁcient mice. Cell lines that normally do not display these characteristics acquire properties of tumor cells when forced to express h-Eag1 channels. The oncogenic potential given to the cells by Eag1 channels is speciﬁc because the expression of another type of voltage-gated potassium channel in the same cell type does not induce tumor properties like those induced by Eag1. In addition, cell lines forced to express Eag1 have a higher metabolic activity and DNA synthesis than cells lacking Eag1 or expressing a different potassium channel. Findings of these oncogenic properties of Eag1 raised immediately research on the distribution of Eag1 in normal human tissues and tumor samples and its potential use as a cancer biomarker. Eag1 as a Diagnostic Marker Eag1 mRNA is distributed in a very restricted manner in healthy tissues. It is mainly expressed in the brain, although few amounts are found in the testis, adrenal gland, and placenta, and is also transiently expressed in myoblasts. Eag1 is expressed before myoblast fusion, and channel activity has been proposed to modulate the shape of the action potential in some cerebellar cells. Mutations in the Eag1 gene have been associated to epilepsy. In sharp contrast, Eag1 mRNA, protein expression (detected by speciﬁc antibodies), and protein activity (studied with the patch-clamp technique) have been found in several tumor cell lines and in a wide variety of biopsies from human tumors including lung, mammary gland, prostate, colon, and uterine-cervix. Eag1 mRNA expression has been also suggested as a potential early indicator of tumor formation. Cervical cancer studies revealed Eag1 mRNA expression in all of the samples from carcinomas but also in some control samples from patients with normal cervix (diagnosed by pap smears studies);

1645

these patients presented either human papillomavirus (HPV) infection or alteration in nearby regions (e.g., an ovarian cystadenoma or endometrial hyperplasia) suggesting Eag1 as a potential sign of early cellular alterations. In accordance to this idea, several cancer etiological factors induce Eag gene expression (see Fig. 1). Human papillomavirus oncogenes and hormones including estrogen and progesterone upregulate Eag mRNA and protein expression suggesting a novel mechanism by which HPV and hormones promote proliferation. In addition, chemical carcinogens induce Eag expression in a mouse model of colon cancer. Besides, Eag1 mRNA is present not only in mammary gland tumors but also in some free-tumor tissues from the vicinity of the tumors, while Eag1 is not found in commercially available RNA from normal mammary epithelium. Eag is also expressed in human diverticulitis which has the potential to turn into colonic cancer. Diagnostic methods based on Eag1 expression are promising because in some cases (e.g., cervical cancer), cells can be obtained from the patients and tested for Eag1 protein expression with speciﬁc antibodies. On the other hand, it has been shown that ﬂuorescent-labeled antibodies against Eag1 show the presence of the protein in lymph nodes that had not been clinically evident; this approach represents a noninvasive optical technique to detect Eag1. The restricted distribution of Eag1 in normal tissues, the abundant and ubiquitous expression in tumors, and the potential regulation by cancer-associated factors convert Eag1 expression in an attractive cancer biomarker. Eag1 as a Therapeutic Target and Prognostic Marker Several approaches have demonstrated that Eag1 expression in tumor cells has an important role in cell proliferation. Eag1 channels are inhibited by several nonspeciﬁc potassium channel blockers; however, at the moment there are no drugs inhibiting speciﬁcally Eag1 channel activity. Imipramine (a common antidepressant drug) and astemizole (a second-generation antihistamine) both inhibit several ion channels including Eag1 and decrease tumor cell proliferation of different cell lines expressing Eag1. These data strongly

1646

Ether à-go-go Potassium Channels

Cancer-related factors (viruses, chemical carcinogens radiations, hormones, growth factors)

Eag1 channels

Channel expression Eag1 mRNA expression

Inhibition of channel activity (durgs, antibodies)

Inhibition of Eag1 mRNA expression Abnormal cell proliferation and tumor formation

Decreased tumor cell proliferation

Ether à-go-go Potassium Channels, Fig. 1 Clinical relevance of Eag1 channels. (a) Eag1 channels are overexpressed in tumor cells favoring cell proliferation; some cancer-related factors are potential inducers of Eag1 channels which are expressed at the plasma membrane but

also probably in the nucleus. (b) Inhibition of either channel activity or RNA expression decreases tumor cell proliferation. Therefore, Eag1 channels represent powerful tools both for cancer diagnosis and therapy

suggest that Eag1 channel activity is necessary in proliferation of human tumor cells. Actually, non-conducting Eag1 channels still produce tumor formation. Molecular biology strategies like RNA silencing have been proved to be very useful for inhibiting speciﬁc gene expression. This approach has been used to study the role of Eag1 in proliferation of tumor cells. Speciﬁc inhibition of Eag1 gene expression by small interfering RNA (siRNA) caused a marked decrease in proliferation of several tumor cell lines. In some cases, speciﬁc Eag1 RNA silencing inhibited tumor cell proliferation in more than 80%. Very speciﬁc monoclonal anti-Eag antibodies have been developed. Some of them do not only detect the protein but also inhibit channel activity and display antitumor properties both in vitro and in vivo. Targeting Eag1 for cancer therapy (see Fig. 1) offers at least two important advantages:

cancer cells and the side effects in normal cells should be almost insigniﬁcant. In healthy tissues Eag1 is mainly expressed in the brain which is already protected by the blood-brain barrier. 2. The well-known drug resistance showed by some cancer cells via membrane transporters would be bypassed when using speciﬁc blockers of channel activity targeting the plasma membrane Eag1 channels. Eag is also a promising tool in cancer prognosis. Eag gene ampliﬁcation was found in some samples from human colorectal adenocarcinoma and was signiﬁcantly associated with adverse outcome. Eag ampliﬁcation emerged as an independent prognostic marker for colon cancer. Eag expression has also been found in some types of leukemia. In the case of acute myeloid leukemia, Eag expression has been proposed as an independent predictive factor for reduced disease-free and overall survival expression strongly correlated with increasing age, higher relapse rates, and shorter survival. Besides,

1. Because of Eag1 restricted distribution in normal tissues, targeted cells would be mainly

Ether à-go-go Potassium Channels

high Eag1 expression is associated to poor survival in ovarian cancer patients. In summary, in addition to serving as a diagnostic marker and therapeutic target, Eag might serve also as a cancer prognostic marker. Mechanisms of Oncogenicity Molecular mechanisms explaining the oncogenic potential of Eag1 channels remain unknown; however, there are several current hypotheses which can be divided into two groups, general and particular. General hypothesis include the following: 1. Eag1 channels establish a negative membrane potential required for cell cycle progression from the G1 to the S phase. Potassium channel inhibition arrests the cells in the G1 phase of the cell cycle. 2. The negative membrane potential increases the electromotive force for calcium entry which in turn might trigger several transduction pathways. 3. Potassium movement regulates cell volume and volume changes are associated to cell proliferation, for instance, by altering nutrient concentration. Particular current hypotheses are based on more structural features. Like other voltagegated potassium channels, Eag1 protein is composed of four subunits (see Fig. 1) having six transmembrane-spanning segments numbered S1–S6; the S4 segment forms the voltage sensor due to its positively charged amino acid residues, and the loop between the segments S5–S6 forms the pore of the channel. Nevertheless, Eag1 has special sequences not shared as a whole with other potassium channels that provide potential clues on its oncogenic mechanisms and include the following: Eag1 has a nuclear targeting signal presumably to direct the channel to the nucleus. It has been demonstrated for a calcium channel that a segment of the channel encodes a transcription factor regulating the expression of several genes. Something similar should be expected to happen with Eag1 if the channel or at least part of the protein

1647

was expressed in the nucleus (see Fig. 1). Epsin is a protein participating in the endocytosis of growth factor receptors. Epsin binds to Eag1; probably this binding changes the free-epsin levels deregulating endocytosis of growth factor receptors and allowing growing signals to proceed. Calmodulin binding sites (integrating calcium signals), a nucleotide binding domain and a Pern-Arnt-Sim domain (PAS domain involved in responses to hypoxia), as well as putative phosphorylation sites by protein kinase C and mitogenactivated protein kinase (MAP kinase) described for Eag1 might also act in concert to regulate channel activity or interaction with other proteins and favor cell proliferation. Non-conducting channels produced by mutations that abolish ion permeation have been produced. These non-conducting channels fail to completely eliminate xenograft tumor formation by transfected cells. Thus, Eag oncogenic properties seem to not rely exclusively on the role of Eag as an ion channel. In addition, it is observed that Eag-expressing cells increase hypoxia-inducible factor 1 activity, vascular endothelial growth factor secretion, and tumor vascularization. Because of different regions of the protein, Eag channels might have several oncogenic mechanisms. Eag1 knockout mice have been produced. It will be very important to know how tumor development occurs in these mice. Outline Expression of Eag1 potassium channels confers oncogenic properties to mammalian cells. Because of their very restricted distribution in normal human tissues but more general distribution in tumor samples, Eag1 mRNA and/or protein expression offers potential tools for the diagnosis of a wide variety of neoplasms. Moreover, the regulation of Eag1 by cancer etiological factors like human papillomavirus, hormones, and carcinogens suggests Eag1 as a potential indicator of early cellular transformation. Speciﬁc inhibition of Eag1 produces a drastic decrease in tumor cell proliferation, making Eag1 a promising target for cancer therapy. Important advances have been achieved on the potential molecular mechanisms of Eag1 oncogenicity. In conclusion, Eag1

1648

represents a hopeful tool for cancer diagnosis, prognosis, and therapy.

Cross-References ▶ Apoptosis ▶ Cancer ▶ Colorectal Cancer ▶ Endometrial Cancer ▶ Ion Channels ▶ Membrane Transporters

References Agarwal JR, Griesinger F, Stuehmer W et al (2010) The potassium channel Ether à go-go is a novel prognostic factor with functional relevance in acute myeloid leukemia. Mol Cancer 9:18 Asher V, Khan R, Warren A, Shaw R, Schalkwyk GV, Bali A, Sowter HM (2010) The Eag potassium channel as a new prognostic marker in ovarian cancer. Diagn Pathol 5:78 Diaz L, Ceja-Ochoa I, Restrepo-Angulo I et al (2009) Estrogens and human papillomavirus oncogenes regulate human Ether à go-go-1 potassium channel expression. Cancer Res 69:3300–3306 Downie BR, Sánchez A, Knötgen H et al (2009) Eag1 expression interferes with hypoxia homeostasis and induces angiogenesis in tumors. J Biol Chem 238:6234–6240 Farias LMB, Ocaña DB, Díaz L et al (2004) Ether à go-go potassium channels as human cervical cancer markers. Cancer Res 64:6996–7001 García-Quiroz J, García-Becerra R, Santos-Martínez N, Barrera D, Ordaz-Rosado D, Avila E, Halhali A, Villanueva O, Ibarra-Sánchez MJ, Esparza-López J, Gamboa-Domínguez A, Camacho J, Larrea F, Díaz L (2014) In vivo dual targeting of the oncogenic Ether-à-go-go-1 potassium channel by calcitriol and astemizole results in enhanced antineoplastic effects in breast tumors. BMC Cancer 14:745 Gómez-Varela D, Zwick-Wallasch E, Knötgen H et al (2007) Monoclonal antibody blockade of the human Eag1 potassium channel function exerts antitumor activity. Cancer Res 67:7343–7349 Hemmerlein B, Weseloh RM, Mello de Queiroz F et al (2006) Overexpression of Eag1 potassium channels in clinical tumours. Mol Cancer 5:41 Mortensen LS, Schmidt H, Farsi Z, Barrantes-Freer A, Rubio ME, Ufartes R, Eilers J, Sakaba T, Stühmer W, Pardo LA (2015) KV 10.1 opposes activity-dependent increase in Ca2+ inﬂux into the presynaptic terminal of the parallel ﬁbre-Purkinje cell synapse. J Physiol 593 (1):181–196

Etidronate Ousingsawat J, Spitzner M, Puntheeranurak S et al (2007) Expression of voltage-gated potassium channels in human and mouse colonic carcinoma. Clin Cancer Res 13:824–831 Pardo LA, Del Camino D, Sánchez A et al (1999) Oncogenic potential of EAG K channels. EMBO J 18:5540–5547 Ramírez A, Hinojosa LM, Gonzales J d J, MontanteMontes D, Martínez-Benítez B, AguilarGuadarrama R, Gamboa-Domínguez A, Morales F, Carrillo-García A, Lizano M, García-Becerra R, Díaz L, Vázquez-Sánchez AY, Camacho J (2013) KCNH1 potassium channels are expressed in cervical cytologies from pregnant patients and are regulated by progesterone. Reproduction 146: 615–623 Rodríguez-Rasgado JA, Acuña-Macías I, Camacho J (2012) Eag1 channels as potential cancer biomarkers. Sensors 12:5986–5995 Schönherr R (2005) Clinical relevance of ion channels for diagnosis and therapy of cancer. J Membr Biol 205:175–184 Simons C, Rash LD, Crawford J, Ma L, CristoforiArmstrong B, Miller D, Ru K, Baillie GJ, Alanay Y, Jacquinet A, Debray FG, Verloes A, Shen J, Yesil G, Guler S, Yuksel A, Cleary JG, Grimmond SM, McGaughran J, King GF, Gabbett MT, Taft RJ (2015) Mutations in the voltage-gated potassium channel gene KCNH1 cause TempleBaraitser syndrome and epilepsy. Nat Genet 47(1):73–77 Ufartes R, Schneider T, Mortensen LS, de Juan Romero C, Hentrich K, Knoetgen H, Beilinson V, Moebius W, Tarabykin V, Alves F, Pardo LA, Rawlins JN, Stuehmer W (2013) Behavioural and functional characterization of Kv10.1 (Eag1) knockout mice. Hum Mol Genet 22(11):2247–2262

Etidronate Definition Etidronic acid (INN) or 1-hydroxyethane 1,1-diphosphonic acid (HEDP) is a ▶ bisphosphonate used in detergents, water treatment, cosmetics, and pharmaceutical treatment. Etidronic acid is a chelating agent and may be added to bind or, to some extent, counter the effects of substances, such as calcium, iron, or other metal ions, which may be discharged as a component of gray wastewater and could conceivably contaminate groundwater supplies. As a phosphonate it

Ets Transcription Factors

has corrosion inhibiting properties on unalloyed steel. Clinically, etidronate disodium acts primarily on bone. It can inhibit the formation, growth, and dissolution of hydroxyapatite crystals and their amorphous precursors by chemisorption to calcium phosphate surfaces. Inhibition of crystal resorption occurs at lower doses than are required to inhibit crystal growth. Both effects increase as the dose increases. Etidronate is used in radionuclide therapy, for instance, of ▶ prostate cancer (▶ Prostate Cancer Radionuclide Therapy). For patients with advanced prostate cancer efﬁcient therapy of painful bony lesions is the primary goal of interdisciplinary treatment strategies. Preservation of quality of life appears to be the main aim rather than prolongation of life. Apart from oral pain relief and local irradiation systemic treatment with radionuclides offers low-risk radiotherapeutic strategies for the palliation of painful, multifocal osteoplastic bone metastases. Depending on the radiopharmaceutical substance chosen response and reduction of pain are described in 65–80%. The duration of pain relief lasts 6–12 weeks. During this time the morphinebased medication can be reduced and in some cases withdrawn, which positively affects quality of life. After improvement of myelosuppression treatment with radionuclides can be repeated. Patients have to be hospitalized for 2 days because of protection from radiation procedures.

1649

Etoposide Synonyms VP-16

Definition With common trade names Etopophos (etoposide phosphate) and VePesid, etoposide is a cytotoxic chemotherapeutic. Most commonly it is used to treat ▶ Ewing sarcoma, testicular cancer, ▶ lung cancer, lymphoma, nonlymphocytic leukemia, and glioblastoma multiforme. Etoposide inhibits DNA synthesis, through inhibition of DNA topoisomerase II.

Cross-References ▶ Ewing Sarcoma ▶ Lung Cancer

See Also (2012) DNA Topoisomerases II. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 1141. doi:10.1007/978-3-642-16483-5_1691.

Ets Transcription Factors Cross-References ▶ Bisphosphonates ▶ Prostate Cancer ▶ Prostate Cancer Radionuclide Therapy

Etiology of Prostate Adenocarcinoma ▶ Prostate Cancer Genetic Toxicology

Jürgen Dittmer Klinik für Gynäkologie, Universität Halle-Wittenberg, Halle (Saale), Germany

Definition Ets transcription factors are deﬁned by a unique DNA-binding domain, the ETS domain, which speciﬁcally interacts with an 10 bp long DNA sequence containing a 50 -GGAA/T-30 core motif (Fig. 1) Ets stands for E26 transformation speciﬁc or E26, as the Ets sequence (v-ets) was ﬁrst

1650

Ets Transcription Factors

“Wing”

“Turn” α1

β1

ETS1 ERG2 FLI-1 PU-1

β2

α2

A C/G C/G A C A C

α3

C/A C/A C/A G

G G G G

GA GA GA GA

β3

A/T A/T A A

G/A G/A G/A G/C

β4

T/C T/C T/C A/G T A/G

Ets Transcription Factors, Fig. 1 The ETS domain. This winged helix-turn-helix domain binds DNA by a loop-helix-loop scaffold, composed of the helix(a2)-turnhelix(a3) motif and the loops between a2 and a3 (turn) and between the b strands b3 and b4 (wing). All direct contacts with speciﬁc bases of the DNA are made by residues in the a3 recognition helix while residues of the two loops contact the phosphate backbone. The resulting

neutralization of the phosphate charges is likely to induce DNA bending, as observed in Ets protein-DNA complexes. In contrast to the helices, the loops are not strictly conserved among members of the Ets family. They may, therefore, be responsible for the preference of an individual Ets protein for the sequences ﬂanking the conserved GGAA/T binding motif.

identiﬁed in the genome of the avian retrovirus E26. c-Ets1, closely related to v-ets, was the ﬁrst cellular Ets protein that was discovered. More than 30 different Ets proteins have been identiﬁed, found throughout the metazoan world including mammals, sea urchins, worms, and insects. Currently, 27 human Ets proteins are known. The Ets family is subdivided into subfamilies based on the similarity in the ETS domain (Fig. 2).

shows similarities to the sterile alpha motif (SAM) domain and is an interface for homotypic and heterotypic protein-protein interactions. In Ets1 and Ets2 proteins, the Pointed domain is the docking site for ERK1/2 allowing these kinases to phosphorylate Ets1 and Ets2 at an N-terminal threonine. In contrast, the Pointed domain of the TEL protein mediates homo-oligomerization. Most Ets proteins are transcriptional activators; others (ERF, NET, Tel, Drosophila YAN, Caenorhabditis lin-1) act as repressors. Some, such as Elk-1 and NET, can undergo activatorrepressor switching (Fig. 2). Ets proteins play an important role in transcriptional regulation. Many eukaryotic genes contain Ets DNA-binding sites and are responsive to Ets proteins. Ets-responsive genes are found among critical genes that regulate fundamental cellular processes such as proliferation, differentiation, ▶ invasion, and ▶ adhesion.

Characteristics In contrast to many other transcription factors, Ets proteins bind to DNA as monomers. Most eukaryotic cells express a variety of Ets proteins at the same time. To achieve functional speciﬁcity, Ets proteins display differences in preference for certain nucleotides ﬂanking the core motif in the Ets-responsive DNA element and, more important, for certain cooperating partners. A strong interaction with a cooperating partner may even force Ets proteins to bind to an unfavorable DNA-binding site, such as GGAG (Pax5/Ets1 partnership). In many cases, interactions with other proteins depend upon particular protein domains, e.g., for the cooperation with SRF, the so-called B domain is required, which is found in the proteins of the TCF subfamily and the Fli-1 protein. The Pointed domain, named after the Drosophila Ets protein Pointed and shared by many Ets proteins of different subfamilies,

Ets Factors and Development Some Ets factors, including Ets2, Esx, Ese-2, Fli-1, Pu.1, GABPa, and Tel, are essential for embryonic development. Disruption of the Ets2, Ese-2, Fli-1, Pu.1, GABPa, or Tel gene in mice results in early death of the embryo. Lack of Ets2 or Ese-2 leads to defects in trophoblast development and to the absence of extraembryonic ectoderm markers. Ese-2 is also involved in mammary alveolar morphogenesis. Tel null mutant embryos fail to develop a vascular network in the yolk

Ets Transcription Factors Subfamily ETS TEL ESX ETS4 ELG ERG TGF

ELF PEA3

ERF SPI

ER71

Member

1651 PNT

ETS

Ets1 Ets2 TEL TEL2 ESX (EIf3) Ehf ESE-2b (EIf5) PDEF GABPα Erg2 FIi.1 FEV EIk-1 Sap1a NET EIf1 NERF1b (EIf2) MEF (EIf4) E1AF ERM ETV-1 ERF PE-1 PU.1 Spi-B Spi-C ETV2

Ets Transcription Factors, Fig. 2 Members of the Ets transcription factor family in humans. The DNA-binding domain (ETS), the Pointed (PNT) domain, and the SRF-interacting B domain are marked. Note that most Ets proteins have several different names. The Esx and Elf proteins are grouped into two separate subfamilies. Splicing variants of the different Ets proteins are not listed. Tel translocation, Ets leukemia, Esx epithelial-restricted with serine box, Ehf Ets homologous factor, ESE epitheliumspeciﬁc Ets, PDEF prostate-derived Ets factor, ELG ets-like gene, GABP GA-binding protein, Erg ets-related

gene, Fli Friend leukemia integration, FEV ﬁfth Ewing variant, TCF ternary complex factor, Elk ets-like gene, Sap SRF accessory protein, NET new ets transcription factor, Elf E74-like factor, NERF new ets-related factor, MEF myeloid Elf-1 like factor, PEA3 polyoma enhancer A3, E1AF adenovirus E1A factor, ERM Ets-related molecule, ETV Ets translocation variant, ER81 Ets-related clone 81, ERF Ets2 repressor factor, PE PU-Ets-related, Spi SFFV provirus integration site, PU recognizes purinerich sequences

sac. Pu.1 is necessary for B- and T-cell development, erythropoiesis, terminal myeloid cell differentiation, and maintenance of hematopoietic stem cells. Fli-1 null embryos die of aberrant hematopoiesis and hemorrhaging. Deﬁciency of GABPa which is expressed in embryonal stem cells leads to embryonic death prior to implantation. GABPa is also required for the function of neuromuscular junctions. Mice lacking Esx die early after birth. Their intestinal epithelial cells fail to differentiate and polarize as result of reduced levels of the TGFbII receptor (▶ transforming growth factor b). Ets1 deﬁciency leads to defects in B- and T-cell development. In ER81 null mice, two types of mechanoreceptors, muscle spindles and

the Pacinian corpuscles, are either absent or degenerated. MEF is involved in osteogenic differentiation. Regulation of Ets Protein Activities The activities of Ets proteins are controlled transcriptionally and posttranslationally. The expression of many Ets genes are restricted to certain cell types and/or can be induced by speciﬁc extracellular stimuli, e.g., the transcription from the Ets1 gene can be activated by a variety of factors including phorbol ester, ▶ AP-1, ▶ TP53, ▶ retinoic acid, ERK1/2 (▶ MAP kinase), and HIF-1 (▶ hypoxia-inducible factor1). Many Ets proteins undergo posttranslational

1652

modiﬁcations, which have an impact on their activities. The most common posttranslational modiﬁcation of Ets proteins is phosphorylation by MAP kinases, such as ERK1/2. Phosphorylation by MAPK leads to activation of activating Ets proteins, such as Ets1, Ets2, ER81, ERM, Sap1, Elk1, PEA3, or GABPa, and loss of activity of repressing Ets proteins, such as Tel or ERF. When phosphorylated by ERK1/2, NET even switches from a repressor to an activator phenotype. It seems that MAPK-dependent phosphorylation (phosphorylation of proteins) shifts the balance between Ets-dependent activation and repression toward activation. In the case of Ets1, MAPKdependent phosphorylation enhances the transcriptional activity by recruitment of the coactivator CBP/p300 (▶ P300/CBP coactivators). Some Ets proteins are also targets of PKA, PKC (▶ protein kinase C family), CaMKII (▶ CalciumBinding Proteins), CKII, and cyclin A-dependent cdk2 (▶ cyclin-dependent kinase). CKII increases the activity of Pu.1 and Spi-B, and PKCa activates Ets1. In contrast, PKA inhibits the DNA-binding activities of ER81 and ERM, whereas CaMKII and cdk2 act as inhibitory on Ets1 and GABPa, respectively. CaMKII phosphorylates Ets1 on serines of a serine-rich region ﬂanking an autoinhibitory module that regulates Ets1 DNA-binding activity. The inhibitory effect of CaMKII on the Ets1 protein increases with each serine that is phosphorylated within the serine-rich region allowing ﬁne-tuning of Ets1-dependent transcription. A few Ets proteins, Ets1, Elk-1, and Tel, have been shown to undergo sumoylation. This posttranslational modiﬁcation inhibits transcriptional activity of Ets1 and Elk-1 and abrogates the repressing activity of Tel. When sumoylated, the activating Elk-1 protein even transforms to a repressor. Acetylation is another means nature uses to modify the activities of Ets proteins, such as Ets1 and ER81. ER81 becomes acetylated and phosphorylated in response to Her2/neustimulated signaling. Acetylation takes place on two lysines within the transactivation domain of ER81 increasing its DNA-binding afﬁnity and protein stability. Elf-1 is an example of an Ets protein

Ets Transcription Factors

that becomes glycosylated. ▶ Glycosylation affects the subcellular localization and DNA-binding activity of Elf-1. Ets Proteins and Cancer The Ets proteins Ets1, Ets2, Fli-1, and Erg are able to transform murine cells. These and other Ets proteins are also involved in human carcinogenesis and/or tumor progression. This is in line with the fact that many of these Ets proteins are targets of the ▶ Ras/Raf/MEK/ERK signaling pathway which is often deregulated in human tumors. The Ras-responsive Ets1 protein is found in different types of solid tumors, including carcinomas and sarcomas. Its overexpression often correlates with increased invasion, higher tumor microvessel density, higher grading, and unfavorable prognosis. Ets1 has been linked to the regulation of key proteases, such as ▶ matrix metalloproteases, involved in the degradation of the ▶ extracellular matrix. In tumors, Ets1 is expressed by tumor cells as well as by stromal cells. By its ability to convert endothelial cells to an angiogenic phenotype, Ets1 is involved in tumor-dependent ▶ angiogenesis. A number of other Ets proteins, such as Erg, PEA3, and E1AF, are capable of upregulating proteases and supposed to be involved in tumor progression. PEA3 has particularly been linked to mammary gland development and oncogenesis. Fli-1 and Ets1 has been shown to regulate ▶ tenascin C, an extracellular matrix protein, associated with tumor progression. In some tumors, Ets genes are subject to mutations and recombinations. The inhibitory Ets protein Tel2 has been shown to induce myeloproliferative diseases in mice by cooperating with the ▶ Myc oncogene and stimulating proliferation. Elf-1 has been implicated in tumor-associated angiogenesis. A target of Elf-1 is Tie2 (▶ receptor tyrosine kinases), a receptor tyrosine kinase involved in the activation of endothelial cells. Chromosomal translocations leading to fusion proteins containing Ets proteins are observed in Ewing tumors and certain types of leukemias. EWS-Ets fusion proteins (▶ EWS-FLI (ets) fusion

Everolimus

transcripts), most often containing Fli-1 or Erg, rarely ETV-1, E1AF, or FEV, are critically involved in the development of Ewing tumors. The fusion protein presumably acts as a transcription factor that binds through the Ets domain to Ets-responsive genes. In addition, EWS-Ets proteins have been suggested to interfere with RNA splicing. Ets fusion proteins, as found in leukemias, harbor either Tel or Erg2. Erg2 is fused to TLS, a protein structurally related to EWS. Hence, TLS-Erg2 chimeric proteins are supposed to have similar functions as EWS-Ets proteins. Tel is frequently fused to tyrosine kinases, such as PDGFRb (▶ platelet-derived growth factor), Abl (▶ BCR-ABL1), or Jak2 (▶ signal transducer and activators of transcription in oncogenesis). The Pointed domain of Tel mediates homodimerization resulting in constitutively active kinases. In another fusion protein, Tel is linked to the DNA-binding factor AML-1 (▶ runx1) which together with CBFb forms the transcription factor CBF. CBF activity is often inhibited in leukemic cells. As a result of Tel-dependent dimerization, CBF function is also blocked when AML-1 is fused to Tel.

1653

Everolimus Daniel C. Cho Beth Israel Deaconess Medical Center, Boston, MA, USA

Synonyms Aﬁnitor ® (marketed by NOVARTIS); RAD001; 42-O-(2-hydroxyethyl) rapamycin

Definition Everolimus is an orally administered inhibitor of ▶ mammalian target of rapamycin (mTOR). This agent has signiﬁcant activity in renal cell carcinoma (RCC) and is approved by the Food and Drug Administration (FDA) for treatment of patients with RCC who have failed treatment with ▶ sorafenib or sunitinib.

Characteristics References Dittmer J (2003) The biology of the Ets1 proto-oncogene. Mol Cancer 2:29 Foos G, Hauser CA (2004) The role of Ets transcription factors in mediating cellular transformation. In: Handbook of experimental pharmacology, vol 166, Transcription factors. Springer, Berlin/Heidelberg/ New York, pp 259–275 Seth A, Watson DK (2005) ETS transcription factors and their emerging roles in human cancer. Eur J Cancer 41:2462–2478 Tootle TL, Rebay I (2005) Post-translational modiﬁcations inﬂuence transcription factor activity: a view from the ETS superfamily. Bioessays 27:285–298

ETS Variant Gene 6 ▶ E T V6

Everolimus (RAD001, Aﬁnitor ®) is an ester of the immunosuppressant ▶ rapamycin which binds with high afﬁnity to FK506-binding protein 12 (FKBP12), forming a complex that inhibits the kinase activity of mTOR. mTOR is a highly conserved serine/threonine kinase which is activated downstream of Akt (protein kinase B; ▶ Akt signal transduction pathway) and regulates cell growth and metabolism in response to environmental factors (Fig. 1). Once activated, mTOR executes its biologic functions as a critical component of two distinct complexes, TORC1 and TORC2, which have differential sensitivities to the rapalogues (i.e., everolimus, ▶ temsirolimus). TORC1, which includes mTOR, LST8 (GbL), and raptor (regulatory-associated protein of mTOR), is inhibited by the rapalogues, whereas TORC2, which includes mTOR, LST8, Sin1, and rictor, is insensitive to the rapalogues. Thus,

1654

Everolimus Growth Factors RTK

PI3K

Akt mTOR

mTOR

GβL

GβL raptor

rictor TORC2

Everolimus

TORC1

S6K

4E−BP

eIF4E

Ribosome Biosynthesis

Cap-Dependent Translation (VEGF, c-myc, Cyclin D, suvivin)

Everolimus, Fig. 1 mTOR signaling pathway and inhibition of TORC1 activity by everolimus

everolimus primarily inhibits the activity of TORC1, which typically acts through its downstream effectors, the eukaryotic translation initiation factor 4E binding protein (4E-BP) and the 40S ribosomal protein p70 S6 kinase (S6K), to stimulate protein synthesis and entrance into G1 phase of the cell cycle. Mechanism of Action in Renal Cell Carcinoma While the exact mechanism of action of everolimus in RCC remains unknown, inhibitors of TORC1 likely function by attenuating the translation of critical mRNA. Everolimus inhibits the phosphorylation of 4E-BP by TORC1, allowing 4E-BP1 to remain associated with eukaryotic translation initiation factor 4E (eIF4E). Thus sequestered, eIF4E is hindered from interacting with the 50 untranslated region (UTR) of capped mRNA and initiating translation. It is now known that certain mRNA,

characterized by lengthy 50 UTR containing stem loop structures, are more dependent upon the availability of eIF4E and are therefore preferentially sensitive to the suppressive effects of TORC1 inhibition. This group of mRNA includes several gene products critical for malignant transformation and tumor progression such as ▶ vascular endothelial growth factor (VEGF), ▶ Myc oncogene, ▶ cyclin D, ▶ survivin, and ornithine decarboxylase (ODC). It is likely that suppression of translation of one or more of these critical mRNA is central to the clinical efﬁcacy of TORC1 inhibitors. While it has been advocated that the efﬁcacy of TORC1 inhibitors in RCC is primarily achieved through the inhibition of translation of hypoxia-inducible factors (HIF)-1a and HIF-2a (▶ hypoxia; ▶ hypoxia-inducible factor1), evidence that the translation of HIF-2a, believed by many to be the more relevant HIF in RCC, is completely dependent upon the activity of TORC2 has cast some doubt upon this hypothesis. Clinical Activity in Renal Cell Carcinoma After demonstrating promising activity in a phase II trial in patients with predominantly clear cell RCC (▶ Renal Cancer Clinical Oncology), everolimus was assessed in a double-blind, randomized, placebo-controlled phase III trial in patients with advanced RCC who had failed treatment with either ▶ sorafenib or sunitinib. Patients were randomized in a 2:1 fashion to receive either everolimus 10 mg once daily (n = 272) or placebo (n = 138) in conjunction with best ▶ supportive care. Patients randomized to everolimus experienced a signiﬁcantly longer progressionfree survival (4.0 months, 95% CI 3.7–5.5) compared with those randomized to placebo (1.9 months, 95% CI 1.8–1.9) with a hazard ratio of 0.30 (95% CI 0.22–0.40; p < 0.0001). Toxicities, which were more common in the everolimus group, included stomatitis, rash, diarrhea, hyperglycemia, hypercholesterolemia, hyperlipidemia, and noninfectious pneumonitis. Based on these ﬁndings, everolimus was approved by the FDA on March 30, 2009, for the treatment of patients with advanced RCC who have failed prior therapy with sorafenib or sunitinib.

Ewing Sarcoma

Conclusion Having demonstrated activity in a randomized, placebo-controlled, phase III trial, everolimus is considered a standard therapeutic option for patients who have failed either sorafenib or sunitinib and is approved by the FDA for this indication. Based on its tolerability, everolimus is now being explored in combination with other active agents in RCC as well as in the adjuvant (▶ adjuvant therapy) setting.

Cross-References ▶ Adjuvant Therapy ▶ Akt Signal Transduction Pathway ▶ Cyclin D ▶ Hypoxia ▶ Hypoxia-Inducible Factor 1 ▶ Mammalian Target of Rapamycin ▶ MYC Oncogene ▶ Rapamycin ▶ Renal Cancer Clinical Oncology ▶ Sorafenib ▶ Supportive Care ▶ Survivin ▶ Temsirolimus ▶ Vascular Endothelial Growth Factor

References Amato RJ, Jac J, Giessinger S, Saxena S, Willis JP (2009) A phase 2 study with a daily regimen of the oral mTOR inhibitor RAD001 (Everolimus) in patients with metastatic clear cell RCC. Cancer 115:2438–2445 Graff JR, Konicek BW, Carter JH, Marcusson EG (2008) Targeting the eukaryotic translation initiation factor 4E for cancer therapy. Cancer Res 68:631–634 Meric-Bernstam F, Gonzalez-Angulo AM (2009) Targeting the mTOR signaling network for cancer therapy. J Clin Oncol 27:2278–2287 Motzer RJ, Escudier B, Oudard S, Hutson TE, Porta C, Bracarda S, Grunwald V, Thompson JA, Figlin RA, Hollaender N, Urbanowitz G, Berg WJ, Kay A, Lebwohl D, Ravaud A (2008) Efﬁcacy of everolimus in advanced renal cell carcinoma: a double-blind, randomized, placebo-controlled phase III trial. Lancet 372:449–456 Toschi A, Lee E, Gadir N, Ohh M, Foster DA (2008) Differential dependence of hypoxia-inducible factors 1 alpha and 2 alpha on TORC1 and TORC2. J Biol Chem 283:34495–34499

1655

See Also (2012) Adjuvant. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 75. doi:10.1007/978-3-642-16483-5_107 (2012) Cell cycle. In: Schwab M (ed) Encyclopedia of Cancer, 3rd edn. Springer Berlin Heidelberg, p 737. doi:10.1007/978-3-642-16483-5_994 (2012) Ornithine decarboxylase. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 2656. doi:10.1007/978-3-642-16483-5_4259 (2012) Renal cancer. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, pp 3225–3226. doi:10.1007/978-3-642-16483-5_6575 (2012) Sunitinib. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 3562. doi:10.1007/978-3-642-16483-5_5575 (2012) Translation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer Berlin Heidelberg, p 3770. doi:10.1007/978-3-642-16483-5_5936

Ewing Sarcoma Heinrich Kovar Children’s Cancer Research Institute, Vienna, Austria

Synonyms Ewing sarcoma family tumors; Ewing tumor; Neuroepithelioma; Peripheral primitive neuroectodermal tumor

Definition Aggressive small round cell tumor affecting bone and soft tissue in children and young adults. Ewing’s sarcoma (ES) and peripheral primitive neuroectodermal tumor (pPNET), also called neuroepithelioma, are currently deﬁned as biologically closely related tumors along a gradient of limited neuroglial differentiation. ▶ Askin tumor is the historical designation of Ewing’s sarcoma of the chest wall. Today, all these neoplasms are summarized as Ewing tumors (ET) or Ewing sarcoma family tumors (ESFT). Although ﬁrst described in 1866 and 1890 by Lücke and Hildebrand, respectively, the disease carries the

1656

name of the American pathologist James Ewing who, in 1921, was the ﬁrst to recognize the tumor as a separate entity, which he deﬁned as diffuse endothelioma of the bone. On the genetic level, Ewing sarcoma family tumors are deﬁned by the consistent presence of a reciprocal ▶ chromosomal translocation between the long arms of chromosome 22 and either chromosome 11 (85%) or 21 (10%). In rare cases, alternative rearrangements of chromosome 22 with either chromosome 7, 17, or 2 have been reported. These aberrations result in a gene fusion that serves as a diagnostic criterion allowing to discriminate Ewing sarcoma family tumors from osteomyelitis and childhood malignancies with a similar small round cell phenotype, including ▶ neuroblastoma, ▶ rhabdomyosarcoma, non-Hodgkin lymphoma, and small cell osteosarcoma. Immunohistochemically, Ewing sarcoma family tumor cells are deﬁned by the abundant presence of the cell surface marker CD99. The exact histogenesis of the disease is not known. An origin from a very primitive, ectodermally derived, migrating cell from the neuroepithelium has been suggested based on ultrastructural and histological signs of limited neural differentiation. Experimental evidence indicates that at least part of the neural resemblance is a functional consequence of the characteristic gene fusion in Ewing sarcoma family tumors. Based on these molecular biological ﬁndings, a mesenchymal origin is currently discussed for the disease.

Characteristics Ewing sarcoma family tumors comprise about 10–15% of malignant bone tumors with a yearly incidence of 0.6 per million in the Caucasian population. The disease is rarely observed among black Africans, African Americans, and Chinese people. It typically occurs in adolescence, in the second decade of life (average age at diagnosis is 13.5 years) with a slight male prevalence. Nevertheless, infants less than 5 years of age as well as adults up to 60 years of age are occasionally diagnosed with Ewing

Ewing Sarcoma

sarcoma family tumors. The tumor usually presents as a painful swelling, rapidly increasing in size. The duration of symptoms prior to the deﬁnitive diagnosis can be weeks to months, or rarely even years, with a median of 3–9 months. In patients with metastatic disease, nonspeciﬁc symptoms such as malaise and fever may resemble symptoms of septicemia. The most frequent tumor localizations are the pelvis, the long bones of the extremities, the ribs, the scapula, and the vertebrae. Less frequently, Ewing sarcoma family tumors arise from extraosseous locations. Unlike ▶ osteosarcomas, Ewing sarcomas tend to arise from the diaphyseal rather than the metaphyseal portion of the bones and are frequently accompanied by tumor-related osteolysis, detachment of the periosteum from the bone, and spiculae of calciﬁcation in soft tissue tumor masses. About 20–25% of Ewing sarcoma patients present at diagnosis with gross, clinically detectable metastases in the lung and/or in bone and/or bone marrow. Metastases to lymph nodes or other sites like liver or central nervous system (CNS) are rare. In contrast to patients with localized disease, cure of this group of patients is very difﬁcult to achieve. Classic Ewing sarcoma is composed of a monotonous population of small round cells with high nuclear to cytoplasmic ratios arrayed in sheets. The cells have scant, faintly eosinophilic to amphophilic cytoplasm, indistinct cytoplasmic borders, and round nuclei with evenly distributed, ﬁnely granular chromatin and inconspicuous nucleoli. Mitotic activity is usually low. By means of immunohistochemistry, the tumor cells occasionally stain positive for neuroglial markers, such as neuron speciﬁc enolase, S100 protein, chromogranin A and B, or the gene product PGP9.5. This is in addition to CD99, which is highly expressed in Ewing sarcoma family tumors with consistency. The inclusion of glycogen can frequently be observed in the tumor cells. Infrequently, Ewing sarcoma is focally immunoreactive for cytokeratins. Cytogenetics and Gene Alterations The most consistent marker of Ewing sarcoma family tumors is the rearrangement of the Ewing

Ewing Sarcoma

sarcoma gene EWS on chromosome 22 (band q12) with a gene encoding for an ETS transcription factor. These proteins are characterized by a unique structure of their DNA-binding domain determining target gene speciﬁcity. In the majority (85%) of Ewing sarcoma family tumors EWS is rearranged with Fli-1, which is located on chromosome 11 (band q24). The second most frequent translocation partner of EWS in this disease is the Ets family member ERG on chromosome 21 (band q22) (10%). In rare cases of Ewing sarcoma family tumors, complex or interstitial chromosomal rearrangements fuse EWS to related Ets transcription factor genes located on chromosomes 7 (band p22), 17 (band q12), and 2 (band q36). These gene rearrangements are currently monitored for diagnostic purposes, either on the chromosomal level using ▶ ﬂuorescence in situ hybridization (FISH) or on the RNA level using reverse transcriptase polymerase chain reaction (PCR) (RT-PCR). The latter method also allows for high sensitivity detection of minimally disseminated disease in blood, bone marrow, and peripheral blood progenitor cell (PBPC) collections. The prognostic impact of RT-PCR detectable tumor cells in these samples is currently under prospective evaluation in several clinical studies. As a result of the gene fusion, a potent novel transcription factor with altered structural and functional features is expressed in the tumor cells. EWS-Fli-1 and EWS-ERG fusion proteins have been shown to render mouse ﬁbroblast cell lines and bone marrow-derived mesenchymal progenitor cells tumorigenic in animal models. Using antagonistic agents, generated by means of gene technology, involvement of these aberrant gene products in Ewing sarcoma tumor cell proliferation has experimentally been demonstrated. It is commonly assumed that the EWS-Ets chimeric transcription factors mediate their transforming properties by inappropriately activating or repressing other genes. In vitro gene transfer experiments into a number of different cell types indicated that the spectrum of genes responsive to EWS-Ets fusion proteins is context dependent. EWS-Fli-1 is toxic to most primary human cell types with only very few exceptions.

1657

Among tissues tolerant of the expression of the Ewing sarcoma oncogene are bone marrowderived mesenchymal progenitor cells. Here, EWS-Fli-1 interferes with the differentiation potential of these cells. With the advent of new technologies to modulate EWS-Ets expression in Ewing sarcoma cells, such as ▶ RNA interference (RNAi), we are just starting to understand the mechanisms underlying malignant transformation by this oncogene. Large scale gene expression proﬁling studies have identiﬁed a speciﬁc signature of EWS-Fli1 within the Ewing sarcoma transcriptome. Chromatin and epigenetic studies indicate that EWS-Fli-1 deregulates gene expression by large scale re-programing of distal gene regulatory elements and aberrant regulation of proximal gene promoters. Current functional studies attempt to separate malfunctions essential for tumorigenesis from collateral damage. No reliable genetic indicators of prognosis have been identiﬁed for Ewing sarcoma family tumors, so far. Cytogenetically, trisomy 8 and 12 accompany the characteristic rearrangement of chromosome 22 in about 44% and 12% of tumors, respectively. Additional structural changes affect chromosomes 1 and 16 in about 20% of tumors, most frequently leading to a gain of 1q and a loss of 16q and the formation of a derivative chromosome. A possible prognostic impact of these cytogenetic alterations has been discussed controversially. Genetic aberrations associated with unfavorable disease in many human malignancies, such as mutations of the tumor suppressor gene p53 and of the Ras oncogene, are infrequent in Ewing sarcoma family tumors. Besides the EWS-Ets gene rearrangement, the only molecularly deﬁned recurrent genomic alterations each occurring in 20–30% of primary Ewing sarcoma family tumors are the homozygous and the loss of STAG2 on Xq25 ▶ INK4A gene located on chromosome 9 (band p21). Preliminary retrospective data suggest an adverse prognostic impact of these aberration and of rare p53 mutations. However, in the absence of any prospectively conﬁrmed molecular prognostic marker, the extent of disease monitored by clinical imaging techniques at the time of diagnosis (computed tomography of the chest

1658

to document or exclude intrathoracic metastases and 99m-Technetium whole body radionuclide bone scans to search for skeletal metastases), microscopically detectable bone marrow micrometastases, and the histopathologically determined tumor response to initial chemotherapy still serve as the only accepted criteria for treatment stratiﬁcation. Etiology The etiology of Ewing sarcoma family tumors is not known. Neither is there evidence for genetic predisposition nor for a role of environmental exposure. A genome-wide association study identiﬁed small nucleotide polymorphisms in three regions on chromosomes 1p36, 10q21 and 15q15 displaying linkage disequilibrium in Ewing sarcoma patients potentially responsible for differences in disease incidence between Caucasians, Asians and Africans. Due to its tight association with the disease, the EWS-Ets gene rearrangement is considered the primary event during Ewing sarcoma pathogenesis. No speciﬁc recombinogenic activity has been identiﬁed as responsible for this aberration and although involvement of a viral agent in generating the chromosomal translocation has been suggested, it has not been conﬁrmed. Therapy In the pre-chemotherapy era, less than 10% of Ewing sarcoma patients survived the disease despite the well-known radiosensitivity of the tumor and despite its radical resection, calling for systemic treatment to eradicate disseminated tumor cells. Today, patients with Ewing sarcoma family tumors are treated by multimodal therapeutic regimens including radiotherapy and chemotherapy (combinations of vincristine, actinomycin D, cyclophosphamide, doxorubicin, ifosfamide, and etoposide) as well as surgical resection whenever possible. By using this treatment strategy together with optimized schedules and dose intensities, the result for patients with localized disease was improved to an overall survival rate of 60–70%. The treatment of Ewing sarcoma patients worldwide is organized in

Ewing Sarcoma Family Tumors

cooperative trials, aiming to further improve treatment outcome. In contrast, the management of primary metastatic disease and early relapse remains a clinical challenge that is currently assessed by myeloablative approaches, combining high-dose chemotherapy and total-body irradiation with stem cell reinfusion. The efﬁcacy of this therapeutic approach for high-risk Ewing sarcoma patients remains to be established. To avoid the toxic side effects of chemotherapy, future biologically tailored therapy may target the EWS-Ets fusion protein or genes downstream of this tumor speciﬁc aberration.

References Bernstein M, Kovar H, Paulussen M et al (2006) Ewing’s sarcoma family of tumors: current management. Oncologist 11:503–519 Kovar H (2005) Context matters: the hen or egg problem in Ewing’s sarcoma. Semin Cancer Biol 15:189–9656 Kovar H (2014) Blocking the road, stopping the engine or killing the driver? Advances in targeting EWS/FLI-1 fusion in Ewing sarcoma as novel therapy. Expert Opin Ther Targets. 18:1315–28. Lawlor ER and Sorensen PH (2015) Twenty Years on: What Do We Really Know about Ewing Sarcoma and What Is the Path Forward? Crit Rev Oncog 20:155–171. Gaspar N, Hawkins DS, Dirksen U et al. (2015) Ewing Sarcoma: Current Management and Future Approaches Through Collaboration. J Clin Oncol 33:3036–46. Sand LG, Szuhai K, Hogendoorn PC (2015) Sequencing Overview of Ewing Sarcoma: A Journey across Genomic, Epigenomic and Transcriptomic Landscapes. Int J Mol Sci 16:16176–215.

Ewing Sarcoma Family Tumors ▶ Ewing Sarcoma

Ewing Tumor ▶ Ewing Sarcoma

EWS-FLI (ets) Fusion Transcripts

EWS-FLI (ets) Fusion Transcripts Enrique de Alava1 and Santiago Ramón y Cajal2 1 Institute of Biomedicine of Sevilla (IBiS), Virgen del Rocio University Hospital /CSIC/University de Sevilla, Seville, Spain 2 Department of Pathology, Vall d’Hebron University Hospital, Barcelona, Spain

Definition The EWS-FLI1 fusion transcript is the result of a balanced reciprocal chromosomal translocation between chromosomes 11 and 22, which fuses the EWS gene in chromosome 22 to the FLI1 gene in chromosome 11. This fusion transcript is detected in approximately 85% of cases of the ▶ Ewing sarcoma family of tumors and is considered a tumor-speciﬁc molecular rearrangement, therefore useful for diagnosis, prognosis, and presumably speciﬁc therapeutics. Approximately 10% of Ewing tumors have fusions involving EWS and ERG genes. An additional 5% bear fusions between EWS and other less frequent genes.

Characteristics Structure Chromosomal translocations result in the genesis of chimeric genes, encoding hybrid transcripts and novel fusion proteins. Many fusion proteins contain juxtaposed functional domains usually found in separate proteins. The EWS-FLI1 fusion protein contains the aminoterminal domain of EWS and the carboxyterminal region of FLI1. EWS gene is an RNA-binding protein, which is believed to mediate mRNA transcription probably through its interaction with RNA polymerase II complex. Several forms (at least 12 types) of EWS-FLI1 exist because of variations in the location of the EWS and FLI1 genomic breakpoints. They contain different combinations of exons

1659

from EWS and FLI1, the most frequent being the fusion of EWS exons 1–7 to FLI1 exon 6–9 (type 1) and fusion of EWS exons 1–7 to FLI1 exon 5–9 (type 2). Properties of the EWS-FLI1 Fusion Transcript and Protein EWS is widely expressed in most tissues, and because of the genomic structure of the fusion, the EWS promoter drives the expression of EWS-FLI1. The aminoterminal domain of EWS, included in the fusion, has strong transactivating properties. The FLI1 gene encodes a member of the ETS family of transcription factors and its expression is highly restricted to hematopoietic, endothelial, and mesodermal cells as well as to neural crest cells. The ETS DNA-binding domain of FLI1 is included in the fusion. The resulting EWS-FLI1 protein is therefore an aberrant transcription factor. The chimeric product EWS-FLI1 can transform some cell lines in culture and can inhibit or activate diverse cellular pathways. For example, EWS-FLI1 protein can suppress transcription of transforming growth factor beta type 2 receptor gene leading to TGF-beta resistance and can activate MFNG, a member of fringe family, related to somatic development, and other genes. The action of EWS-FLI1 as a transcription factor is probably related to the cell context in which this fusion is detected. This cellular context is probably inﬂuenced by the cell type, stage of differentiation, and microenvironment and could include expression of several growth factors, for example, IGF1 and its receptor. On the other hand, it is likely that FLI1 gene is developmentally regulated, being expressed at certain times and places, and this can confer some site and tissue speciﬁcity to the transcriptional activity generated by the EWS-FLI1 fusion protein (Fig. 1). EWS and Other Specific Human Translocations EWS can fuse after chromosomal translocation to other genes, including several members of the ETS transcription factor family. In an analogous way to FLI1, EWS can be detected fused to ERG gene through a t(21;22), to ETV gene through a t

1660

EWS-FLI (ets) Fusion Transcripts

Aminoterminal domain of EWS 1

3 4

ETS domain DNA binding 7

NH2

3 4

EWS-ERG

COOH

EWS-FLI-1

ETS domain DNA binding

Aminoterminal domain of EWS 1

COOH

NH2

EWS-FLI-1

ETS domain DNA binding

Aminoterminal domain of EWS 1

COOH

7-9 NH2 Aminoterminal domain of EWS 1

3 4

ETS domain DNA binding 7

9-4 NH2

4 5 6 7 8

COOH EWS-FLI-1

EWS-FLI (ets) Fusion Transcripts, Fig. 1 The ﬁgure illustrates two sources of variability of the fusion transcripts in Ewing sarcoma, which include (a) the EWS

gene partner, and (b) the speciﬁc number of exons from each gene involved in the fusion

(7;22), to E1A-F gene through a t(17;22), or to FEV gene through a t(2;22) ▶ Chromosomal Translocation. All of these EWS-ETS gene fusion transcripts can confer a common tumorigenic phenotype of small round cells and can be found in the Ewing family of tumors. But, interestingly, EWS can fuse to other genes and be detected in other tumor types. For example, in desmoplastic small round cell tumor, EWS is fused to the tumor suppressor gene WT1 through a t(11;22); in clear cell sarcoma of soft tissue, is fused to the ATF1 through a t(12;22); in myxoid and round cell liposarcoma, can also be found fused to the CHOP gene; and to CHN gene, located in chromosome 9, in extraskeletal myxoid chondrosarcoma.

evidence that several chromosomal translocations may not be random events, but may be speciﬁcally promoted by the presence of certain DNA sequence motifs at or around certain target genes. A number of recombinogenic sequences have been described, like topoisomerase-binding sequences in leukemias or lymphomas and translin sequences in alveolar rhabdomyosarcoma and in myxoid liposarcoma. Furthermore it has been suggested that mobile elements or endogenous retroviruses may take part in gene rearrangements. In Ewing tumors, in which illegitimate recombination has been reported to occur, recombinogenic sequences have not been described in a large study of genomic breakpoints.

Genesis of the Translocation The mechanism by which the translocation is generated is another ﬁeld worth to be studied. In other words why do the breakpoints always occur in the same introns? Is that a random event or are there certain areas of these particular genes particularly prone to recombine? There is some

The EWS-FLI1 Fusion and the Pathogenesis of Ewing Sarcoma A problem to study sarcoma pathogenesis, in general, and Ewing tumor genesis, in particular, is the absence of preneoplastic lesions, as well as the lack of animal transgenic models. Therefore, to study the mechanism by which EWS-FLI1

EWS-FLI (ets) Fusion Transcripts

induces ▶ Ewing sarcoma, there are two complementary approaches: 1. Induced expression of the fusion in different cellular models. For example, induced expression in NIH3T3 ﬁbroblastic cells accelerates tumor growth in immunodeﬁcient mouse models, while expression in alveolar rhabdomyosarcoma or neuroblastoma cell lines induces a shift in differentiation, which becomes closer to that of Ewing tumor. The same experiment performed in other cell lines induces, however, cell death, showing that the cellular context matters in the pathogenesis of Ewing tumor. 2. By small interference RNA studies blocking EWS-FLI1 mRNA, cell cycle arrest and a decrease in tumor growth in animal models are observed. Both types of experiments conﬁrm that EWS-FLI1 functions as an aberrant transcription factor, although the target genes are still relatively unknown. The availability of expression microarrays in the last years, coupled to the experiments described earlier, has been useful to suggest possible target genes. EWS-FLI1 participates in Ewing tumor pathogenesis controlling cell proliferation and survival by promoting expression of IGF1, MYC, CCND-1, PDGFC, DAX1-NR0B1, and NKX2-2 and by repressing genes such as p21 WAF1 , p57 kip, TGFbRII, and IGFBP3. The genes related to the induction of the undifferentiated phenotype of Ewing tumor are largely unknown. In addition, EWS-FLI1 exerts its function with the help of other proteins such as RNA helicase A, which binds to the promoters targeted also by EWS-FLI1, enhancing its function.

Clinical Relevance EWS-FLI1 (ets) in Diagnosis The detection of the chimeric product EWS-FLI1, in the adequate morphologic and immunophenotypic context, is diagnostic of a Ewing family tumor. The fused product can be studied by ﬂuorescence in situ hybridization (FISH) or by

1661

RT-PCR in very small samples of tissue or cytological specimens. Frozen tissue is the preferred source of RNA, which can also be extracted from formalin-ﬁxed parafﬁn-embedded tissue with variable success. EWS-FLI1 detection is particularly useful in cases in which a Ewing family tumor arises in unusual locations (kidney, skin, lung, ovary, pancreas), or in patients over 40 years, or shows atypical features (epithelial differentiation, no CD99/MIC2 expression). EWS-FLI1 (ets) Fusion Type and Prognosis Ewing tumors display a moderate level of molecular heterogeneity. The variability in the chimeric transcript structure may help to deﬁne clinically distinct risk groups of Ewing tumors. In fact, two independent groups have found that the type 1 EWS-FLI1 fusion transcript is associated with less aggressive clinical behavior than the patients carrying tumors with other EWS-FLI1 fusion types (Fig. 2), regardless of stage at diagnosis, tumor location, or tumor volume. A study has shown that this particular gene fusion (type 1) encodes for a chimeric protein that functions as a weaker transcription factor than chimeric proteins encoded by other fusion types. In fact tumors with EWS-FLI1 type 1 fusions have a lower proliferative rate than their counterparts with other fusion types. So far, no clinical differences have been seen between patients with tumors bearing EWSFLI1 fusions and those having EWS-ERG transcripts. These results are still pending of conﬁrmation; a European prospective study is being conducted to assess for clinical differences among tumors having the most prevalent fusion types. EWS-FLI1 (ets) and Detection of Minimal Residual Disease EWS-ETS fusion transcripts can be detected by RT-PCR in peripheral blood and bone marrow. That demonstration may contribute to patient management and staging, although clinical relevance is still unclear. A French study has shown in 2003 that presence of circulating Ewing tumor cells and/or Ewing tumor cells in the bone marrow at the time of diagnosis is a signiﬁcant predictor of tumor relapse.

1662

EWS-FLI1 fusion type

1.0

Type 1 (n=64) No type 1 (n=35) Log rank p=0.023

0.8 Overall survival

EWS-FLI (ets) Fusion Transcripts, Fig. 2 Early studies showed that patients with Type 1 transcripts were associated to a better overall survival than all others

EWS-FLI (ets) Fusion Transcripts

0.6

0.4

0.2

0.0 0

100

120

Follow up (months)

EWS-FLI1 (ets) and Therapeutics Current efforts are focused on directly inhibiting chimeric proteins (or their downstream targets) and on immunotherapy directed at tumor cell-speciﬁc epitopes derived from chimeric products. A potentially useful approach is the delivery of siRNA in the appropriate vectors. The results of an animal model of metastatic Ewing tumor in which EWS-FLI1 siRNA has been injected intravenously are encouraging and could potentially be used in the future treatment of patients with this neoplasm.

Cross-References ▶ Chromosomal Translocations ▶ Ewing Sarcoma

References Alava E, de Gerald WL (2000) Molecular biology of the Ewing’s sarcoma/primitive neuroectodermal tumor family. J Clin Oncol 18:204–213 Hu-Lieskovan S, Heidel JD, Bartlett DW et al (2005) Sequence-speciﬁc knockdown of EWS-FLI1 by targeted, nonviral delivery of small interfering RNA inhibits tumor growth in a murine model of metastatic Ewing’s sarcoma. Cancer Res 65:8984–8992 Kovar H (2005) Context matters: the hen or egg problem in Ewing’s sarcoma. Semin Cancer Biol 15:189–196

Riggi N, Suva ML, Stamenkovic I (2006) Ewing’s Sarcoma-like tumors originate from EWS-FLI-1expressing mesenchymal progenitor cells. Cancer Res 66:9786 Schleiermacher G, Peter M, Oberlin O et al (2003) Increased risk of systemic relapses associated with bone marrow micrometastasis and circulating tumor cells in localized Ewing tumor. J Clin Oncol 21:85–91

See Also (2012) CD99. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 704. doi:10.1007/978-3-642-16483-5_941 (2012) Chromosomal translocation t(9;22). In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/ Heidelberg, pp 845–846. doi:10.1007/978-3-64216483-5_1142 (2012) Formalin-ﬁxed Parafﬁn-embedded Tissue. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1446. doi:10.1007/9783-642-16483-5_2249 (2012) Illegitimate recombination. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 1808. doi:10.1007/978-3-642-164835_2960 (2012) RT-PCR. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3322. doi:10.1007/978-3-642-16483-5_5129 (2012) Transcript. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3752. doi:10.1007/978-3-642-16483-5_5898 (2012) Transcription factor. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3752. doi:10.1007/978-3-642-16483-5_5901 (2012) Translocation. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 3773. doi:10.1007/978-3-642-16483-5_5942

Exfoliation of Cells

Exfoliation of Cells Alexandre Loktionov DiagNodus Ltd, Babraham Research Campus, Cambridge, UK

Synonyms Shedding of cells; Sloughing of cells

Definition Cell exfoliation (Latin – exfoliare – to strip of leaves) is a process of spontaneous or induced complete detachment of single epithelial cells or groups of cells from an epithelial layer.

Characteristics Exfoliation of cells is one of the main mechanisms of cell loss participating in the homeostatic control of cell population size. Cell exfoliation is a characteristic feature of epithelial tissues forming epithelial layers covering both external body surface (skin epidermis and skin appendages) and surfaces of internal cavities and passages (gastrointestinal tract, respiratory system, and genitourinary system) as well as major exocrine glands and glandular ducts (mammary gland, exocrine pancreas, biliary system, etc.). Cell exfoliation process in normal physiological conditions is closely associated with terminal differentiation and orderly loss of dying cells compensated by permanent cell population renewal. This relationship is exempliﬁed by the structure of stratiﬁed squamous (epidermal-type) epithelia where layers of maturing cells are being moved toward the surface due to continuous arrival of younger counterparts produced by proliferation in the basal layers of the epithelium. Eventual death, keratinization (in the epidermis), and inevitable exfoliation of cell remnants is the normal destiny of these

1663

terminally differentiated cells. Less is known about cell exfoliation in the epithelia of internal organs; however, studies of well-structured columnar colonic epithelium can be chosen as the most reliable source of information. Obvious links between cell differentiation and eventual shedding exist in this dynamic cell population with extremely high rates of cell proliferation and loss. The progeny of stem cells located at the base of the colonic crypt constantly migrate in the direction of the lumen. During this migration colonocytes gradually lose their proliferative capacity and undergo differentiation, thus terminally differentiated cells reaching surface (luminal) epithelium should be promptly eliminated to be replaced by future generations of their counterparts. Two ways of cell loss coexist in the colonic epithelium under normal conditions (Fig. 1). 1. Exfoliation of single colonocytes or colonocyte groups into the lumen of the gut. Strictly speaking, the shed cells enter the protective mucocellular layer covering colonic mucosa rather than the gut contents. The characterized protective mucus consists of two layers (dense inner and looser outer) formed by Mucin 2 molecules. 2. Apoptosis in situ followed by the engulfment of apoptotic cells by adjacent colonocytes or subepithelial macrophages. Physiological changes or development of pathological processes can signiﬁcantly shift proportions of cells eliminated by each of these mechanisms. There are two principal variants of postexfoliation cell fate (Fig. 1). Some colonocytes undergo immediate apoptosis (detachment-induced apoptosis or ▶ anoikis) following exfoliation, whereas other exfoliated cells and especially groups of cells maintain their structural integrity and possibly viability for minutes or even hours following exfoliation. In the human colon, exfoliated cells enter the mucocellular layer overlaying colonic mucosa and can be transferred distally, protected by the mucus, without being incorporated

1664

Exfoliation of Cells

Exfoliation of Cells, Fig. 1 Schematic representation of cell exfoliation in normal physiological conditions. Colon model, surface epithelium, and crypts are not shown. 1 Normal epithelium (two apoptotic cells shown). 2 Anoikis immediately following exfoliation. 3 Exfoliated normal epithelial cell preserving its structure. 4 Group of

exfoliated normal epithelial cells preserving their structure. 5 Cellular and nuclear debris. 6 Subepithelial macrophage. Semitransparent horizontal strips above the epithelium correspond to protective colorectal mucus layer (dense inner and looser outer layers)

into the feces. The movement of the cell-containing mucocellular layer is mostly driven by its close contact with moving fecal ﬂow as well as by constant peristaltic movements of the gut. There is no doubt that some proportion of exfoliated colonocytes can reach the gut contents, but these cells should be rapidly destroyed in the anaerobic bacteria-dominated fecal milieu rich in bile acids and other cytolytic agents. Observations of the presence of well-preserved epithelial cells in human fecal samples mostly concern colonocytes excreted together with the mucocellular layer fragments or exfoliated squamous cells of the anal epithelium sometimes misidentiﬁed as colonocytes. Similar exfoliation models with tissue-speciﬁc corrections can be applied to the epithelial populations of other internal organs. Preservation of cellular structure after exfoliation is described for cells of bronchial epithelium in sputum samples, bladder, urethral and prostate epithelium in urine samples, gastric epithelium in gastric lavage liquid, cervical, vaginal, and endometrial

epithelium in cervical smears, mammary gland epithelium in nipple aspirates, and pancreatic duct epithelium obtained endoscopically. Exfoliated cell migration or passive transport occurs in all internal passages or ducts, being facilitated by contents ﬂow, peristaltic movements, or celldriven mucus transport (cilia of the bronchial epithelium). Shed cells can be commonly found in human body excretions (cells of bladder, urethra, and occasionally prostate epithelium in urine, colonocytes and squamous anal epithelium in feces, bronchial, nasopharyngeal, and oral cavity epithelium in sputum, etc.). Exfoliated epithelial cells should be distinguished from nonepithelial free cells (leukocytes, macrophages, lymphocytes, etc.) that are also often present in the excreted materials but are unrelated to the process of cell exfoliation. Mechanisms Molecular mechanisms underlying cell exfoliation in normal conditions are poorly investigated and remain largely unknown. Extrapolation of the

Exfoliation of Cells

1665

Exfoliation of Cells, Fig. 2 Schematic representation of cell exfoliation from tumor surface. Colon model, surface epithelium, and crypts are not shown. Inset demonstrates secondary implantation of malignant cells distally from the primary tumor. 1 Tumor (malignant cells). 2 Exfoliated tumor cells and cell groups preserving their structure. 3 Occasional anoikis of a tumor cell immediately following

exfoliation. 4 Exfoliated normal epithelial cell preserving its structure. 5 Cellular and nuclear debris. 6 Subepithelial macrophage. 7 Neutrophilic leukocytes and necrotic tumor fragments (focal necrosis and inﬂammatory reaction). Disorderly oriented semitransparent strips of different length indicate disease-related deterioration of the protective mucus layer

information obtained in vitro might be misleading; therefore, only tentative suggestions can be made. At the cellular level, pathways leading to cell exfoliation should involve considerable structural and functional changes of the cytoskeleton, cell membrane, and membrane-bound subcellular structures responsible for the preservation of contacts between neighboring cells (▶ adherens junctions, ▶ tight junctions, ▶ gap junctions) as well as between cells and basal membranes (focal adhesions). A number of proteins associated with these structures (adherens junction–associated cadherins and catenins, tight junctionassociated claudins and occludin, gap junctionassociated connexins and focal adhesionassociated integrins and focal adhesion kinase) are likely to participate in exfoliation induction. In particular, the integrin system appears to be intimately involved in the homeostatic regulation of cell differentiation, providing communication between epithelial cells and the underlying

stromal extracellular matrix. The loss of contact with the extracellular matrix can trigger anoikis through integrin-mediated signaling, but it is evident that many exfoliated cells are able to evade this fate. Mechanisms of this phenomenon remain obscure and need profound investigation as well as other aspects of cell exfoliation in vivo. Cell Exfoliation in Neoplasia Cancer development is usually associated with a dramatic increase in cell exfoliation from tumor surface (Fig. 2). The increased exfoliation may reﬂect a compensatory response to the impairment of the apoptotic mechanism of cell elimination in situ commonly observed in malignant tumors. Anoikis induction appears to be the ultimate homeostatic goal of the enhanced exfoliation. Although this mechanism of cell elimination may be partially functional in early tumors, resistance to anoikis is now regarded as a hallmark of metastatic cancer cells, which tend to survive

1666

exfoliation. Loss of cell adhesion (often associated with E-cadherin function impairment) and disruption of normal interactions with underlying stroma associated with malignant progression also strongly contribute to the enhancement of exfoliation from tumors. Disease-related deterioration of the protective mucus layer can also facilitate cell exfoliation. Cancer cells are better adapted to survival in the conditions of oxygen deﬁciency; therefore their prolonged postexfoliation persistence on the surfaces of internal cavities, passages, and glandular ducts can be expected. These cells shed by tumors are often immature and retain proliferative capacity. Therefore their metastatic potential and ability for secondary implantation can be manifested. Reports describing the occurrence of secondary cancers or peritoneal carcinomatosis following surgical interventions on tumors of several sites (colorectum, ovary, gallbladder) attribute these secondary metastatic cancers to the direct reimplantation of malignant cells exfoliated from the primary tumors. There is a strong probability that cell exfoliation without immediate anoikis in physiological conditions presents a normal “prototype” of metastatic behavior. Clinical Aspects The possibility of using exfoliated epithelial cells for various clinical and research purposes is highly attractive because analysis of this material often allows avoiding highly invasive sampling (biopsy) of normally inaccessible tissues. Exfoliated cells can be collected either noninvasively (from sputum, urine, feces, etc.) or by simple and minimally invasive procedures (cervical smears, buccal swabs, direct cell collection from the surface of rectal mucosa during proctoscopy, etc.). Although exfoliated cells became widely employed for some speciﬁc medical, forensic, and research purposes (e.g., buccal mucosal cells are routinely used for DNA isolation for genotyping), clinical oncology remains the main area where their use is already common and looks even more promising with the introduction of modern molecular methodologies. Cytological diagnostic approaches based on the examination of exfoliated cells were

Exfoliation of Cells

developed ﬁrst. Exfoliative cytology analysis of smears prepared from cervical epithelium (examination of PAP smears) has become a standard screening procedure for cervical cancer. Diagnostic cytology is used for endometrial carcinoma, bronchogenic lung cancer, tumors of the bladder and prostate, and gastric cancer. The introduction of molecular assays targeting cancer biomarkers created another major direction, aiming to employ exfoliated cells as the material of choice for the molecular diagnosis and screening of oncological conditions. Approaches to colorectal cancer screening occupy a leading position among problems addressed in this area. Signiﬁcant efforts were concentrated on the analysis of exfoliated colonocytes (or DNA derived from these cells) present in human stool samples. Cancer-speciﬁc molecular changes are often detected by such analysis, especially when multimarker panels of PCR-based assays are applied. The sensitivity of these methods has considerably improved, but endoscopic conﬁrmation of the diagnosis is still required. The main reason for remaining problems is likely to be the relatively low presence of exfoliated colonocytes in the fecal samples. Human DNA found in this material can also derive from exfoliated anal epithelium or free cells (leukocytes, lymphocytes, macrophages, etc.), which certainly do not contain tumor markers. Moreover, strong fecal presence normally decreases PCR efﬁciency. Direct collection of colorectal mucus containing exfoliated cells from the surface of rectal mucosa or the anal area immediately following defecation appear to provide much more abundant and contamination-free material that can be used for a range of analytical procedures. Even simple detection of unusually high DNA yield in samples of exfoliated material directly collected at standardized conditions can be interpreted as a warning signal indicating a high probability of colorectal tumor presence. Multiple other applications of exfoliated cells for cancer diagnosis by molecular biomarker detection are being developed, but these studies remain research projects rather than proven clinical approaches. The present absence of a single biomarker allowing 100% positive identiﬁcation

Exosomal miRNA

of cancer presence is the most important obstacle interfering with the development of molecular diagnostic procedures based upon the use of exfoliated cells.

Cross-References ▶ Adherens Junctions ▶ Anoikis ▶ Gap Junctions ▶ Integrin Signaling ▶ Tight Junction

References Bogenrieder T, Herlyn M (2003) Axis of evil: molecular mechanisms of cancer metastasis. Oncogene 22:6524–6536 Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100:57–70 Johansson MEV, Sjövall H, Hansson GC (2013) The gastrointestinal mucus system in health and disease. Nat Rev Gastroenterol Hepatol 10:352–361 Loktionov A (2007) Cell exfoliation in the human colon: myth, reality and implications for colorectal cancer screening. Int J Cancer 120:2281–2289 Osborn NK, Ahlquist DA (2005) Stool screening for colorectal cancer: molecular approaches. Gastroenterology 128:192–206

See Also (2012) Biopsy. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 415. doi:10.1007/978-3-642-16483-5_644 (2012) Cilia. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 857. doi:10.1007/978-3-642-16483-5_1168 (2012) Focal adhesion. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, pp 1436–1437. doi:10.1007/978-3-642-16483-5_2227 (2012) Mucocellular layer. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2389. doi:10.1007/978-3-642-16483-5_3877 (2012) Multi-marker panels. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2394. doi:10.1007/978-3-64216483-5_3884

1667

Exobiotics ▶ Xenobiotics

Exosomal MicroRNA ▶ Exosomal miRNA

Exosomal miRNA Shuai Jiang1 and Wei Yan2 1 Department of Biology and Biological Engineering, California Institute of Technology, Pasadena, CA, USA 2 Department of Cancer Biology, Beckman Research Institute of City of Hope, Duarte, CA, USA

Synonyms Exosomal microRNA; Exosomal shuttle microRNA

Definition ▶ Exosomes, shuttling from donor cells to recipient cells, are cell-derived extracellular vesicles (EV) with 30–100 nm in diameter and dish- or classic cup-shaped morphology. Exosomes transport bioactive cargo carrying selective proteins, lipids, DNA, messenger RNAs (mRNAs), and small and large ▶ noncoding RNAs such as ▶ microRNAs (miRNAs). miRNA embedded in exosomes is termed as exosomal miRNA. Exosomal miRNAs regulate diverse biological processes in recipient cells.

Characteristics

Exjade ® ▶ Deferasirox

Biogenesis of Exosomes and Exosomal miRNA Packaging Exosomes play a fundamental role in cell-cell communication and have been found in various

1668

Exosomal miRNA

Exosomal miRNA, Fig. 1 Exosomal miRNAs participate in tumorigenesis. Exosomes are secreted by donor cells through the fusion of multivesicular bodies (MVBs) with cell membrane. miRNAs are carried in exosomes and

functionally delivered to recipient cancer cells. Oncogenic exosomal miRNAs (a) and tumor-suppressive exosomal miRNAs (b) promote and inhibit tumorigenesis, respectively

cell types, especially in tumor cells. Cluster of differentiation (CD) 63, CD81, heat shock protein (HSP) 70, Alix (ALG-2-interacting protein X), and TSG101 (tumor susceptibility gene 101) serve as exosomal marker proteins. Exosomes are derived from the multivesicular bodies (MVBs), which are also known as late endosomes. The endosome origin is the hallmark of exosomes distinct from other larger kinds of extracellular vesicles. They are formed by endosomal membrane inward budding in MVB, and this process is controlled by endosomal sorting complex required for transport (ESCRT), ceramide, or tetraspanin complex (Miller and Grunewald 2015; Zhang et al. 2015). When MVBs fuse with plasma membrane, biologically active exosomes are secreted into extracellular environment (Fig. 1). However, the exact mechanism how exosomal contents, especially miRNAs, are loading into MVBs is still under investigation. In fact, miRNA sorted into exosome is a selective process. Proﬁling studies show that some miRNAs including miR-150, miR-142, miR-320, and miR-451 are preferably enriched in exosomes. Additionally, speciﬁc exosomal miRNAs express differently under different conditions. For instance, exosomal Let-7 miRNAs are much more expressed in gastric cell line AZ-P7a compared with other cancer cells, while exosomal miR-21 is abundant in the sera of glioblastoma patients compared with healthy donors. As for today, there are several models to clarify the mechanisms of the package of exosomal

miRNA. Firstly, the speciﬁc short sequence located on miRNAs can guide their loading into exosomes. For example, in B lymphocytes, 30 ends of uridylated endogenous miRNAs are secreted into B lymphocyte-derived exosomes, while 30 ends of adenylated endogenous miRNAs remain in B lymphocytes. Secondly, the package of miRNAs into exosomes requires the assistance of some intracellular functional proteins. For instance, the overexpression of neural sphingomyelinase 2 (nSMase2), the ﬁrst reported protein functioning in the miRNA selection into exosomes, can increase the exosomal miRNA number. In addition, argonaute 2 (AGO2) prefers to bind uracil or adenine on the 50 end of miRNAs. AGO2 deletion can decrease the expression of exosomal miRNAs including miR-150 and miR-451. Moreover, some modiﬁed RNA-binding protein contributes to the recognition of speciﬁc sequence on miRNAs. Sumoylated protein heterogeneous nuclear ribonucleoprotein A2B1 (hnRNPA2B1) binds to the speciﬁc 4 nt motifs (GGAG) in the 30 region of miRNA to mediate miRNA transporting to MVBs and then packaging into exosomes. KRAS is also engaged in the process of miRNA package into exosomes. ▶ KRAS mutant colorectal cancer (CRC) cells have more miR-100 sorted into exosomes, while wild type cells have more exosomal miR-10. Thus, speciﬁc motif located on miRNAs and intracellular functional molecules are both critical to the package of exosomal miRNAs. Apart from those two kinds of elements, the afﬁnity between MVB membrane and cellular miRNAs as well as cell-

Exosomal miRNA

activation-dependent miRNA-targeted transcript level changes in donor cells also contribute to sorting of miRNAs into exosomes (Squadrito et al. 2014). Exosomal miRNA is secreted into extracellular environment by employing exosome as vesicle. To date, the investigation of the underlying secretion mechanism is still going on. The secretion of exosomes in parent cells is dependent on important regulatory molecules such as small ▶ GTPase Rab family including Rab27, Rab28, Rab31, and Rab11 and Rab effector molecules (SYTL4 and SLAC2B). The tumor suppressor protein p53 modulates transcription of the downstream genes such as tsap6 and chmp4c that regulate the endosomal compartment and lead to elevated exosome secretion. Additionally, nSMase2 and calcium ionophore also regulate exosome secretion. Bio-functions of Exosomal miRNA in Cancer Exosomes utilize surface receptor such as MHC interaction or plasma membrane fusion to transfer contents in recipient cells, where exosomal miRNAs can bind to the 30 untranslated region (30 UTR) of target mRNAs. By repressing the expression of direct targets in recipient cells at posttranscription level, exosomal miRNAs carry out their functions in acceptor cells, most of which are closely linked to human cancers. Exosomal miRNAs can function as oncogenic exosomal miRNAs in tumor invasion, metastases, tumor angiogenesis, and immune suppression in various cancers. In breast cancer, exosomal miRNAs can promote breast cancer cell metastasis by direct targeting downstream genes in recipient cells. For example, exosomal miR-105, characteristically overexpressed and produced by metastatic breast cancer cell line MDA-MB-231, can directly repress downstream target tight junction-related gene (ZO-1) in endothelial cells, damaging the natural barrier integrity against breast cancer metastasis. IL4-activated macrophages-derived exosomal miR-223 can promote invasion of human breast cancer cell line SKBR3. In addition, exosomal miRNAs can also modulate tumor microenvironment. Stromal cell, an important component in tumor microenvironment, can

1669

produce exosomal miRNAs to assist breast cancer cells to be resistant to cancer therapy through activating STAT1 signaling pathway and NOTCH3 on breast cancer cells (Miller and Grunewald 2015). Also, exosomal miRNAs can promote angiogenesis. Blood vessel formation is essential for tumor cell growth. Melanoma cellderived miR-9 induces the migration of endothelial cells and promotes cancer angiogenesis by regulating JAK-STAT pathway. Exosomal miR-135b derived from hypoxic multiple myeloma cells can block its target factor-inhibiting hypoxia-inducible factor 1 (HIF-FIH-1) when delivered to human umbilical vein endothelial cells (HUVECs). Hypoxic exosomal miRNAs enhance angiogenesis via the HIF-FIH signaling pathway under the condition of hypoxia. Besides, human endothelial cell line HMEC-1 s-derived exosomal miR-214 stimulates migration program and angiogenesis in recipient HMEC-1 cells. Exosomal miRNAs might also assist cancer cells to escape from immune cell detection. For instance, TW03-derived exosomes can block T-cell proliferation and T helper cell differentiation, leading to inhibit the function of T cells in human nasopharyngeal carcinoma (NPC) (Ye et al. 2014). Furthermore, ﬁve exosomal miRNAs have been identiﬁed to be abundant in the patient sera, and they modulate the MARK1 signaling pathway to affect cell proliferation. By contrast, some exosomal miRNAs are underexpressed in multiple cancers. These are known as tumor-suppressive exosomal miRNAs. Tumorsuppressive exosomal miRNAs have characteristics similar with tumor-suppressive genes that can inhibit cancer cell proliferation, metastasis, and induce apoptosis. For instance, it is well known that Let-7 miRNAs usually play a tumorsuppressive role by repressing oncogenes such as RAS and HMGA2, a metastatic gastric cancer cell line AZ-P7a could keep their oncogenic capacity through releasing tumor-suppressive Let-7 miRNAs via exosomes into the extracellular environment. In addition, tumor-suppressive exosomal miR-143 that derived from noncancerous cells can suppress the growth of prostate cancer cells by inhibiting its target genes including KRAS and ERK5 both in vitro and

1670

in vivo. Thus, exosomal miRNAs participate in tumorigenesis. Oncogenic exosomal miRNAs and tumor-suppressive exosomal miRNAs can promote and inhibit process of tumorigenesis, respectively (Fig. 1).

Exosomal miRNAs in Cancer Diagnostic and Clinical Use Exosomal miRNAs play fundamental characters during cancer progression. So far, more and more evidence shows exosomal miRNAs could be used as diagnostic biomarkers for various cancers. For colorectal cancer, exosomal levels of seven miRNAs including miR-21 and miR-223 are signiﬁcantly hyperactivated in primary cancer patients compared with healthy controls, with signiﬁcantly downregulation after surgical resection of colorectal tumors. Thus, exosomal miRNA signatures could be utilized to mirror pathological process of colorectal cancer. For lung cancer, the expression of some exosomal miRNAs in patients’ sera is much higher than non-lung cancer tissues, indicating circulating exosomal miRNAs might be useful as biomarker for lung adenocarcinoma as well. For breast cancer, exosomal miRNAs can also be used as a diagnostic biomarker. Certain serum miRNAs are also highly correlated with breast cancer tissues. For instance, oncogenic miR-21 and miR-155 are signiﬁcantly abundant in breast cancer specimens, whereas miR-126 is dramatically under-expressed. In ovarian cancer, tumor-derived exosomal miRNA signatures exhibit dramatically different proﬁles compared with that from benign tissues. So, exosomal miRNAs can be applied as diagnostic markers of ovarian cancer and references for tumor stages. Additionally, expression of exosomal miR-21 is correlated with tumor progression stages in esophageal squamous carcinoma, indicating that it may be a useful target for cancer therapy. Exosomal miR-1290 and miR-375 can be used as prognostic marker in castration-resistant prostate cancer as well. Exosomes have already been employed by human virus to transfer miRNAs to noninfected

Exosomal miRNA

cells, thereby assisting virus spread, which tells us exosomes may be applied as therapeutic vesicles and function as a good delivery system for tumorsuppressive exosomal miRNAs in cancer therapy. For example, delivering tumor-suppressive exosomal miR-143 leads to the shrink of development of prostate cancer cells in nude mice (Kosaka et al. 2013). Tumor-suppressive miRNA Let-7a can inhibit breast cancer cell growth when introduced into EGFR-expressing cells. Moreover, exosomes can be able to cross the blood–brain barrier, which could enhance the efﬁciency of delivery of medications to cancer cells, whereas there are still some questions need to be addressed when employing tumor-suppressive exosomal miRNAs in cancer therapy. For instance, how to facilitate more speciﬁc tumorsuppressive exosomal miRNA loading and packaging into exosomes? How to enhance the uptake efﬁciency of exosome by recipient cells? Secondly, apart from tumor-suppressive exosomal miRNAs, oncogenic exosomal miRNAs can also be applied in tumor therapy. Due to their functions in tumor angiogenesis and tumor invasion, oncogenic exosomal miRNAs might be utilized as cancer vaccines, whereas their cancer-promoting effects should be monitored during cancer research trials. Lastly, the small size of exosomal miRNAs also beneﬁts them as an ideal target for drug designing. Collectively, exosomal miRNAs could potentially be served as clinical tools for cancer diagnostic and clinical use for various cancers. However, any side effects when using bioengineered or naturally occurring exosomal miRNAs need to be considered. In a word, using exosomal miRNAs might be an exciting but challenging application in cancer therapy in the near future.

Cross-References ▶ Exosome ▶ GTPase ▶ KRAS ▶ MicroRNA ▶ Noncoding RNA

Exosome

References Kosaka N et al (2013) Exosomal tumor-suppressive microRNAs as novel cancer therapy “exocure” is another choice for cancer treatment. Adv Drug Deliv Rev 65(3):376–382 Miller IV, Grunewald TG (2015) Tumor-derived exosomes: tiny envelopes for big stories. Biol Cell 107:1–19 Squadrito ML et al (2014) Endogenous RNAs modulate microRNA sorting to exosomes and transfer to accept cells. Cell Rep 8(5):1432–1446 Ye SB et al (2014) Tumor-derived exosomes promote tumor progression and T-cell dysfunction through the regulation of enriched exosomal microRNAs in human nasopharyngeal carcinoma. Oncotarget 5(14):5439–5452 Zhang J et al (2015) Exosome and exosomal microRNA: trafﬁcking, sorting, and function. Genomics Proteomics Bioinformatics 13(1):17–24

See Also (2012) P53. In: Schwab M (ed) Encyclopedia of cancer, 3rd edn. Springer, Berlin/Heidelberg, p 2747. doi:10.1007/978-3-642-16483-5_4331

Exosomal Shuttle MicroRNA ▶ Exosomal miRNA

Exosome Peter Kurre1 and Ben Doron2 1 Department of Pediatrics, Oregon Health and Science University, Portland, OR, USA 2 Oregon Health and Science University, Portland, OR, USA

Synonyms Extracellular vesicles

Definition Exosomes are membrane-enclosed vesicles that are derived from the endocytic compartment and

1671

released at the plasma membrane into the extracellular space. The plasma membrane is not the source of the lipid bilayer of exosomes; rather, exosomes originate from luminal pinocytosis of early endosomes. These vesicles range from 30 nm to 100 nm in diameter and trafﬁc cargo in an autocrine, paracrine, and endocrine fashion. Exosomes contain a subset of biologically active macromolecules present in the cell: protein, lipids, and multiple RNA species. The composition of exosomal cargo is unique to the producing cells, giving healthy and diseased cells a speciﬁc exosomal signature. The trafﬁcking of these molecules into neighboring cells alters their behavior, a process involved in neuronal signaling, fetal development, tissue homeostasis and repair, adaptive immunity, and cancer progression.

Characteristics Mechanism In contrast to the shedding of microvesicles which bud at the plasma membrane, exosomes are contained as vesicles within endosomal compartments, termed multivesicular bodies (MVB). Their unique biogenesis is also reﬂected in their lipid composition resembling that of the early endosomes. Exosomes originate as inward buddings of the endosomal membrane to create intraluminal vesicles (ILVs). ILVs accrue during the transition from early to late endosomes, also known as multivesicular bodies. Movement toward the plasma membrane is controlled by the cytoskeleton and small GTPases. Endosomes move along microtubule tracks via molecular motors dictated by Rab GTPases and the phosphoinositide proﬁle on the outer lipid leaﬂet of the vesicles. Secretion occurs when endosomes fuse with the plasma membrane and release their exosomes. This allows the cell to manage exosomal output in a temporally and spatially controlled fashion by using multiple cytoskeletal and membrane proteins to mediate fusion and secretion. Subject to an active process, cells change both the output of exosomes and the cargo within them under various stress-inducing conditions.

1672

Cargo Protein

Exosomes contain internal maturation proteins and the proteins bound for recipient cells within the lumenal compartment. To date, close to 4,600 different proteins have been shown to associate with exosomes. The cellular origin of exosomes accounts for the fact that many exosomal proteins are involved in endosomal pathways. The most common proteins include the tetraspanins CD9, CD63, and CD81, which act as protein scaffolds; ﬂotillin, which aids in vesicle formation; Alix, an adaptor protein required for endosomal trafﬁcking; and TSG101, a regulator of vesicular trafﬁcking. These proteins are frequently used as markers for the classiﬁcation as exosomes. Exosomes also contain cytoplasmic proteins that act in recipient cells. These include metabolic enzymes, signal transducers, and transcription factors. One mechanism by which proteins are sorted into exosomes relies upon ESCRT (endosomal sorting complexes required for transport), a complex of proteins that coordinates both budding and the sorting of proteins into ILVs. This complex was initially described as being required for the sorting of ubiquitinated proteins destined for the lysosomal degradation, but it has been shown to sort proteins into exosomes. The posttranslational modiﬁcations that ESCRT recognizes for sorting are still unclear, but seem to be primarily orchestrated by a combination of the ubiquitin proﬁle and association of other “guide” proteins. This work on sorting was performed on viral proteins and may vary for endogenous protein exportation. Lipid

Exosomes are enriched in cholesterol, ceramides, sphingomyelin, and saturated species of phosphatidylcholine and phosphatidylethanolamine. The lipid composition of exosomes contributes to both ILV formation and trafﬁcking within the cell of origin. In studies where components of the ESCRT complex are knocked out, ILVs are still generated in a mechanism that seems to be aided by the increased incorporation of ceramides, which increase membrane curvature.

Exosome

Sphingosine-1-phosphate was also shown to contribute to ILV formation, as a reduction in ﬂotillin+ exosomes was observed in sphingosine kinase 2 knockout cell lines. Importantly, in contrast to microvesicles whose bilayer reﬂects that of the plasma membrane, exosomes contain additional lipid moieties of endosomal origin. RNA

Deep sequencing of exosomes has revealed the existence of a vast array of RNA species in exosomes. Importantly, the RNA content of exosomes is not a proportional reﬂection of the transcriptome of the cell, but exhibits enrichment and exclusion of speciﬁc transcripts. The mechanism by which selection is accomplished remains under investigation, but may utilize miRNA sequence motifs. This “EXOmotif,” in tandem with a sumoylated heterogenous ribonucleoprotein A2B1 (hnRNPA2B1), selectively sorts miRNA into exosomes. Similarly, cis elements on mRNA transcripts have also been identiﬁed to be correlated with exosomal sorting. One study demonstrated interplay between miRNA and mRNA in the sorting efﬁciency of transcripts, and another study showed the presence of miRNA/RISC complex within exosomes. Currently, more research needs to be performed to elucidate these sorting mechanisms, but it is clear that RNA sorting is dependent upon both cis- and trans-elements. Trafficking Exosomes are relatively resistant to strong shearing forces and enzymatic degradation, making them suitable candidates for the delivery of fragile or cell impermeable molecules. Alternatively, exosomes in the extracellular space can trafﬁc back to the cell of origin, neighboring cells, or into the bodily ﬂuids for transmission to other organs and tissues. Several mechanisms have been described for cellular entry of exosomes. Recipient cells take up exosomes in a fashion similar to cell-cell adhesion, through the use of ICAMs and integrins. Surface receptors on the recipient cells recognize exosome proteins, glycoproteins, and lipid moieties, leading to cell-type speciﬁc trafﬁcking. This is of particular importance in

Exosome

signaling within the immune system. For example, CD169+ macrophages recognize B-cell-derived exosomes particularly by the decoration of 2–3 linked sialic acid and glycoproteins containing speciﬁc mannose residues. This is important for communication between cells within the spleen and lymph nodes. Lectins on the surface of recipient cells contribute to preferential exosome uptake, as do clathrins, dynamin, and caveolae. Expression of these surface receptors modulates the speciﬁcity and afﬁnity for exosome reception. Further, exosomes can either fuse with the recipient cell’s plasma membrane, releasing their cargo into the cytosol, or become engulfed and enter the endocytic pathway. Release of cargo into the cellular cytosol affects recipient cell behavior. Of interest within the exosome ﬁeld is the trafﬁcking of functional RNA molecules between cells. Messenger RNA transcripts generated in a donor cell have been shown to be translated to functional protein in recipient cells. Furthermore, miRNA generated in a producer cell can exhibit suppression on recipient cell gene expression. This type of cell-cell communication contributes to tissue development and homeostasis and is especially important in the cross talk between the stromal and parenchymal components within organs. Exosomes in Cancer The development of cancer is a dynamic process, and exosomes contribute to a variety of events that enable cancer to progress. Almost all tumor types appear to exhibit an increase in exosomal output upon transformation. The cell-autonomous accumulation of mutations and the subsequent signaling dysregulation allow tumors to overcome cellular checkpoints and growth restrictions, whereas intercellular signaling shapes their microenvironment by manipulating neighboring cells and tissue. Exosomal-mediated cell signaling appears to contribute to a tumor-proliferative microenvironment through interactions with neighboring stroma and via immune evasion. Microenvironment

Tumors are capable of transforming their surroundings into a state that supports malignancy.

1673

This includes modulation of the extracellular matrix (ECM) and induction of angiogenesis. ECM is altered in part by the manipulation of ﬁbroblasts near the tumor. These cells are referred to as cancer-associated ﬁbroblasts (CAF) and remodel the architecture and composition of surrounding ECM to promote metastasis and vascularization. Exosome-mediated transport of the signaling molecules TGF-b1 and FGF-2 contributes to the altered ﬁbroblast phenotype. Multiple studies have also shown that exosomal output of stromal cells is altered in a tumor environment, which reinforces tumor proliferation by providing growth factors and signals back to the tumor cells. Angiogenesis is the process by which vascularization is introduced to a tumor. This process is required for solid tumors, as their size and growth rate require the delivery of oxygen to counteract their hypoxic environment. Hypoxia and nutrient depletion provoke the release of exosomes containing angiogenic miRNA and stimulatory signaling molecules that induce the neovascular formation of blood vessels in the tumor. Immune System

While the tumor microenvironment contains cells from both the innate and adaptive arms of the immune system, most tumors are able to suppress the local immune response. The communication between immune cells, mediated by cellular contact or the release of chemokines and cytokines, establishes the balance between pro- or antiinﬂammatory responses. Exosomes inﬂuence this equilibrium through several mechanisms. Exosome-mediated immune suppression was shown in experiments demonstrating increased tumor proliferation and decreased immune response when transplanted tumors were accompanied by injections of exosomes derived from the same tumor. Candidate mechanisms in support of this observation include delivery of proapoptotic ligands to reduce tumor inﬁltrating lymphocytes in the microenvironment, delivery of antiinﬂammatory cytokines, and induction of regulatory T cells that increase the immune tolerance of the tumor.

1674

Exosome

Biomarkers