expression of the c-erbb-3/her-3 and c-erbb-4/her-4 growth ... · function act as secondary...

Expression of the c-erbB-3/HER-3 and c-erbB-4/HER-4 GrowthFactor Receptors and Their Ligands, Neuregulin-1a,Neuregulin-1 b, and Betacellulin, in NormalEndometrium and Endometrial Cancer

Radhika Srinivasan, Eleanor Benton,Frances McCormick,1 Hillary Thomas, andWilliam J. Gullick 2

Receptor Biology Laboratory, Imperial Cancer Research FundMolecular Oncology Unit [R. S., E. B., W. J. G.], Division ofInvestigative Sciences [F. M.], Imperial College School of Medicine,Hammersmith Campus, London W12 0NN, and OncologyDepartment, The Royal Surrey County Hospital, Guildford, SurreyGU2 5XX [H. T.], United Kingdom

ABSTRACTThe objective of this study was to determine the immu-

nohistochemical expression of the c-erbB-3 and c-erbB-4growth factor receptors and their principal ligands, theneuregulins and betacellulin, in normal endometrium anddetermine whether there was evidence of under- or overex-pression in endometrial adenocarcinoma. Immunohisto-chemistry was performed using well-characterized antibod-ies against each of the five proteins analyzed on formalin-fixed, paraffin-embedded archival material. Forty-threenormal endometrial samples (16 proliferative, 19 secretory,and 8 hyperplastic) and 41 endometrial adenocarcinomacases were analyzed. There was variable expression of thegrowth factor receptors and the ligands in the two principalphases of the menstrual cycle as well as in endometrialadenocarcinoma. In normal endometrium, the c-erbB-3 re-ceptor was weakly expressed in both phases. The c-erbB-4receptor and all of the ligands examined, neuregulina,neuregulin b, and betacellulin, were expressed at signifi-cantly higher levels in the secretory as compared with theproliferative phase of the menstrual cycle, suggesting a rolefor these proteins in endometrial maturation. In endometrialadenocarcinoma, overexpression of c-erbB-3, c-erbB-4, andbetacellulin with underexpression of neuregulina as com-pared with normal controls was observed. Neuregulinbexpression was not found to be significantly different in the

two groups. These results suggest that signaling through thec-erbB-3 and c-erbB-4 receptors and the ligands neuregulina, neuregulin b, and betacellulin are important in endome-trial carcinogenesis.

INTRODUCTIONGrowth factors regulate cellular growth and differentiation

by binding to cell surface receptors, several families of whichhave been defined based on structural and functional similari-ties. The type I growth factor receptor family consists of theprototype EGFR3 and the related members, c-erbB-2, c-erbB-3,and c-erbB-4 (also known as HER 1, HER 2, HER 3, and HER4, respectively). These transmembrane glycoprotein moleculesconsist of an extracellular ligand binding domain, a transmem-brane region, and an intracellular domain with tyrosine kinaseactivity. The ligands for these receptors are a large family ofproteins that bind to the extracellular domain, resulting in re-ceptor activation by homodimer and/or heterodimer formationand the subsequent transphosphorylation of tyrosine residues inthe cytoplasmic region (1). Interaction between the ligand andits cognate receptor/s may occur through juxtacrine, paracrine,or autocrine pathways. Broadly, these ligands/growth factorsmay be divided into three categories based on their recognitionof each receptor expressed alone: (a) those that primarily inter-act with EGFR; (b) the NRGs, also called heregulins whichprimarily interact with c-erbB-3 or c-erbB-4; and (c) thoseligands that interact equally with both EGFR and c-erbB-4.

The NRGs are encoded by three distinct genes calledNRG1 (2), NRG2 (3–5), andNRG3 (6). The NRG1 gene hasbeen studied in most detail and has been shown to encode acomplex family of proteins that includes more than a dozenisoforms derived from alternative splicing (7). Each NRG var-iant, however, contains an EGF-like motif responsible for re-ceptor binding, although the COOH-terminal third may either bederived from thea or b exon, which appears to determine thenature of the second receptor recruited into the dimer (8, 9).c-erbB-3 and c-erbB-4 act, respectively, as the low and highaffinity receptors of these ligands (10), and EGFR and c-erbB-2function act as secondary receptors, the recruitment into het-erodimers of which is hierarchically controlled (11).

Betacellulin, initially identified as a growth factor from aninsulinoma cell line, is an 80-amino acid protein derived from a177-amino acid precursor that contains an EGF-like element(12). It has been shown to function as a ligand for EGFR (13)

Received 3/17/99; revised 7/15/99; accepted 7/30/99.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisementin accordance with 18 U.S.C. Section 1734 solely toindicate this fact.1 Present address: Department of Histopathology, Derriford Hospital,Plymouth, Devon, PL6 8DH.2 To whom requests for reprints should be addressed, at HammersmithHospital, MRC Cyclotron Building, Third Floor, Imperial Cancer Re-search Fund Oncology Group, Du Cane Road, London W12 OHS,United Kingdom.

3 The abbreviations used are: EGFR, epidermal growth factor receptor;NRG, neuregulin.

2877Vol. 5, 2877–2883, October 1999 Clinical Cancer Research

Research. on October 6, 2020. © 1999 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

and c-erbB-4 but does not interact with c-erbB-2 or c-erbB-3(14). As yet, there is little information on the physiological roleof betacellulin or its role in cancer, although it is expressed in anumber of normal tissues (15).

The human c-erbB-3 receptor, as its name suggests, is thethird member of the family and is widely expressed in normaltissues (16). The most significant difference between c-erbB-3and the other family members is that it has little or no tyrosinekinase activity (17, 18). Thus, although homodimers of c-erbB-3may be inactive, it has been shown to signal effectively bypairing with EGFR or c-erbB-4 in heterodimers and in particularwith c-erbB-2, for which it appears to be the preferred partner.The c-erbB-3 protein has also been shown to be expressed atelevated levels in a range of human malignancies (19).

c-erbB-4, the fourth and most recently described memberof the EGFR family (20), is aMr 180,000 glycosylated receptor,the pattern of expression of which in normal tissues has beendescribed in detail only recently (21). There have been only ahandful of reports on its expression in cancer (21), and bothunderexpression (22) and overexpression (23, 24) have beenreported.

The expression of EGFR and its ligands and c-erbB-2receptor in the normal and malignant endometrium have beenstudied previously (25–31). We have, in this pilot study, lookedfor the first time at the expression of c-erbB-3 and c-erbB-4growth factor receptors and their principal ligands, NRG-1a,NRG-1 b, and betacellulin, in normal endometrum and com-pared them with the expression in endometrial adenocarcino-mas. In normal endometrium, the expression of these proteinshas been compared in the two phases, proliferative and secre-tory, where the endometrium is under different hormonal mi-lieus and therefore may provide a clue regarding the regulationof these proteins by the sex steroid hormones, particularlyestrogen. Persistent and prolonged estrogenic stimulation is awell-known risk factor in the causation of endometrial carci-noma. In particular, in the breast, c-erbB-2 and c-erbB-3 mRNAand protein expression are down-regulated by estrogen andinduced by tamoxifen, but it is not established whether similar

regulation of expression occurs in endometrium. This may be ofmore than academic interest in the light of their role in cellproliferation and the apparent paradoxical effect of tamoxifen, awidely used antiestrogenic compound, on the incidence of breastand endometrial cancer (32).

MATERIALS AND METHODSAntibodies Used in this Study. The mouse monoclonal

antibodies, RTJ1 used for the detection of c-erbB-3 (33) andHFR-1 for the detection of c-erbB-4 (21), have been characterizedpreviously. Polyclonal antibodies were raised in rabbits against theligands NRG-1a, NRG-1 b, and betacellulin using syntheticpeptides derived from the COOH-terminal third of the EGF ho-mology region. The immunizing sequences were as follows:NRG-1 a, CQPGFTGARCTENVPMK (called NRG76); NRG-1b, PNEFTGDRCQNYVMAS (called NRG102); and betacellulin,CDEGYIGARCERVDLFY (called BTC 97).

Each purified peptide was coupled to keyhole limpet he-mocyanin and used to raise antibodies in rabbits. The serumtiters were determined by ELISA, and the antibodies werepurified by affinity chromatography as described by Rajkumar

Fig. 1 Western blot of recombinant NRG isoforms.Lane 1,rHNRG-a1(14–249);Lane 2,rHNRG1-a2; Lane 3,rHNRG-b1; Lane 4,rHNRG-b2.The blot was probed sequentially with affinity-purified antibodies to theaisoforms (NRG76) and theb isoforms (NRG102).

Fig. 2 The individual histograms were derived from plotting the num-ber of cases against the score for each of the proteins analyzed.M,proliferative endometrium;f, secretory endometrium;p, cases of en-dometrial hyperplasia.

2878c-erbB-3 and c-erbB-4 Receptors in Endometrial Cancer



and Gullick (34). The properties of antibodies raised againstNRG76 (35) and BTC97 (36) have been described previously.

The properties of the antibody raised to NRG1-b weredetermined by Western blotting. Purified, recombinant isoformsof NRG1 a and b variants were a generous gift from Dr. B.Ratzkin (Amgen). The forms used in this work are rH-NRG-a1(14–249), rH-NRG-a2 (14–241), rH-NRG-b1 (14–246), andrH-NRG-b2 (14–238) (Ref. 37). 0.5mg of each purified formwas run on a 10% polyacrylamide gel; this was transferred tonitrocellulose, and Western blotting was performed using 5mg/ml of affinity-purified NRG76 antibody, after which theblots were stripped and reprobed using the same concentrationof antibody NRG102.

For tissue sections and immunohistochemistry, a total of 43normal endometria (16 proliferative, 19 secretory, and 8 hyper-plasia) and 41 endometrial adenocarcinoma (grading as perFederation Internationale Gynaecologists et Obstetristes classi-fication: 30 grade 1, 6 grade 2, and 5 grade 3) were analyzed byimmunohistochemistry of formalin-fixed, paraffin-embeddedtissues. The tissue blocks were retrieved from the archives of theHammersmith and Wexham Park Hospitals. No pretreatmentwas required for any of the antibodies. The streptavidin-biotin

immunoperoxidase technique was used. Briefly, sections werebrought to water, endogenous peroxidase was blocked, and theprimary antibodies were applied at 5mg/ml for the polyclonaland 1mg/ml for the monoclonal antibodies and incubated at 4°Covernight. Appropriate biotinylated secondary antibodies (swineantirabbit and rabbit antimouse) were used (Dako), followed bya third layer of horseradish peroxidase-labeled streptavidin(Dako). The color was developed using 3,39-diaminobenzidineand hydrogen peroxide. Negative controls included omission ofthe primary antibody and blocking of reactivity by preincuba-tion with a 50-fold excess of the corresponding immunizingpeptide. The positive control included was a section of thekidney wherein the tubular cells stained positive for the fiveantigens tested in this study.

For scoring of immunoreactivity, we used a scoring systemdescribed previously (33, 38, 39) wherein both the percentage ofcells positive and the overall intensity of staining were takeninto account. On the basis of the percentage of tumor cellspositive, a score of: 0, no cells positive; 1, up to 25% positive;2, 26–50% positive; 3, 51–75% positive; and 4,.75% positive,was assigned. A second score for intensity of staining wasassigned as follows: 0, negative; 1, weakly positive; 2, moder-ately positive; and 3, strongly positive.

The two individual scores were added, giving a final scorethat therefore could range from 0 to 7. Any evidence of mem-brane positivity was given an additional score of 1. In normalendometrium, the glands and stroma were evaluated separately.Scoring was performed by two independent observers (E. B. andR. S.). The rate of discordance was,5%, and all such caseswere reassessed together.

Statistical Analysis. The five proteins examined (c-erbB-3, c-erbB-4, NRG-1a, NRG-1 b, and betacellulin) wereanalyzed for association with the menstrual phase in normalendometrial samples using the Mann-Whitney and Kruskal-Wallis test. The expression in the endometrial glands was com-pared with the stromal expression using the Wilcoxon matchedpairs signed rank test. The endometrial adenocarcinoma sampleswere compared as a group with the normal endometrium for theexpression of each of the proteins and analyzed using theMann-WhitneyU test. Within the endometrial carcinoma group,association with the grade/differentiation was sought by theMann-Whitney and Kruskal-Wallis test.

RESULTSCharacterization of the Antibody NRG-102 Raised

against NRG-1 b. To examine expression of the NRG-1bform of NRG, antibodies were raised to a synthetic peptide fromthe unique COOH terminal sequence of the EGF-like domain.These were evaluated for their ability to recognize theb iso-forms and not to cross-react with thea variants by Westernblotting. The NRG102 antiserum reacted specifically with theb1 and 2 (Fig. 1,Lanes 3and4) proteins but did not recognize thea forms (Fig. 1,Lanes 1and2). Conversely, antibody NRG76recognized thea 1 and 2 forms and not theb variants (Fig. 1).The antibodies were then evaluated for their ability to react withformalin-fixed, parraffin-embedded tissues. The staining ob-tained with each antibody was blocked by the immunizingpeptide, confirming the specificity of their reaction. The char-

Fig. 3 The histograms were derived from plotting the number of casesagainst the score for each of the proteins analyzed.M, normal; f,endometrial adenocarcinoma cases.

2879Clinical Cancer Research



Fig. 4 A,proliferative endometrium;B, secretory endometrium showing c-erbB-4 (HFR-1) positivity. Note that the glands in the proliferative phaseshow weak reactivity, whereas the glands in the secretory phase are strongly positive.C–F,endometrial adenocarcinoma cases.C, overexpression ofc-erbB-3.D, mild overexpression of c-erbB-4.E, overexpression of betacellulin in grade 1 adenocarcinoma.F, underexpression of NRG-1a in a grade3 adenocarcinoma. Streptavidin-biotin immunoperoxidase with hematoxylin counterstaining was used.A andB, 3100; C–F, 3200.




acterization of the other antibodies used in this study has beenreferred to earlier in “Materials and Methods.”

Immunohistochemistry. A total of 43 samples of nor-mal endometrium and 41 endometrial adenocarcinomas wereanalyzed using these reagents for the immunohistochemicalexpression of the c-erbB-3 and c-erbB-4 growth factor ligandsand receptors NRG-1a, NRG-1 b, and betacellulin by immu-nocytochemical staining. Predominantly cytoplasmic stainingwas seen with all of the antibodies used, which was diffuse withthe exception of NRG1-a, which showed both a diffuse and agranular pattern of staining. Membrane staining for thec-erbB-3, c-erbB-4 receptors and the ligand betacellulin wasseen only in the context of endometrial cancers in 12, 4, and 11cases (30, 8, and 28% of the total). Nuclear immunoreactivitywas observed in the normal endometrial stromal nuclei in asmall number of cases for NRG1-aand b, betacellulin, andc-erbB4. No nuclear staining was seen in endometrial cancers.

The levels of expression of each of the receptor or ligandwere compared among: (a) the glandularversusstromal com-partments in normal endometrial samples; (b) glands in theproliferative versussecretory phases; (c) the entire group ofnormal endometrial samplesversusendometrial adenocarcino-mas, and (d) in the endometrial carcinoma group, correlation tohistological tumor grade,i.e., grade 1versusgrades 2 and 3combined (in view of the small number of cases in each group).The results obtained for each of the proteins examined aredescribed below. The hyperplastic endometrium, in general,showed expression of each of the proteins analyzed at levelssimilar to that seen in the proliferative endometrium. The his-tograms obtained by plotting the glandular reactivity scoreagainst the number of cases in the two phases of the menstrualcycle and for hyperplastic endometrium for each protein ana-lyzed are shown in Fig. 2. The histograms derived similarly bycomparing the normal groupversusthe endometrial carcinomaare displayed in Fig. 3.

c-erbB-3. Normal endometrial glands expressed low lev-els of c-erbB-3 in both phases of the menstrual cycle, with thescores ranging from 0 to 4, which were not significantly differ-ent in the two phases. The stroma also showed weak expression.In endometrial cancers, the levels of c-erbB-3 was variable andranged from 0 to 8, and using an arbitrary cutoff score of 7, therewas evidence of overexpression (Fig. 4C) in 30% (12 of 40)cases. When analyzed as a group, there was significant overex-pression in the carcinomas as compared with the normals (MannWhitney U test;P 5 0.00001). In addition, there was a corre-lation with tumor grade, with the well-differentiated tumorsshowing higher levels of c-erbB-3 as compared with the mod-erately and poorly differentiated tumors analyzed as a group(Kruskal-Wallis test,P 5 0.0001).

c-erbB-4. In normal endometria, both the glands and thestroma showed levels ranging from weak to moderate; the scoresranged from 0 to 6 with significantly higher levels in the endome-trial stroma as compared with the glands (Wilcoxon signed ranktest,P , 0.0001). There was a significant difference in the level ofc-erbB-4 with respect to the menstrual phase, with the glands in thesecretory phase showing higher levels of expression as comparedwith the proliferative phase (P 5 0.0001; Fig. 4,A andB). Theendometrial carcinomas generally showed modest levels, and usinga cutoff score of 7 and above, there was evidence for overexpres-

sion in six cases (15%). Furthermore, when the carcinomas wereanalyzed as a group and compared with the normals, significantoverexpression was observed in the endometrial carcinomas (Fig.4D; Mann-WhitneyU test,P 5 0.0002).

NRG-1 a. This ligand was expressed in both the glandsand stroma with scores ranging from 2 to 7, with significantlyhigher levels of expression in the stroma than in the glands (Wil-coxon signed rank test,P 5 0.0007). Incidentally, the vascularsmooth muscle and the uterine smooth muscle also expressedmodest levels of NRG-1a. There was also a weak association withthe menstrual phase, with the secretory glands showing higherlevels of expression than the proliferative glands (P 5 0.0064). Inendometrial carcinomas, variable levels were noted; the scoresranged from 0 to 7. Overall, there was a decrease in the level ofexpression in endometrial cancers (Fig. 4F) as compared withnormals, which was statistically significant (Mann Whitney test,P 5 0.0035). Furthermore, a significant decrease in the levels ofexpression in the moderately and poorly differentiated tumors ascompared with the well-differentiated tumors was also observed(Mann-Whitney test,P 5 0.0001).

NRG-1 b. In normal endometria, both the glands andstroma showed weak to moderate levels of expression, with thescores ranging from 0 to 7. The glands in the secretory phaseshowed significantly higher levels of expression as comparedwith the proliferative phase (P5 0.0065). Endometrial cancersshowed variable levels of expression, with the scores rangingfrom 0 to 8, and these were not significantly different fromnormal levels. There was no correlation of NRG-1b expressionto the tumor grade.

Betacellulin. Both the endometrial glands and stromashowed levels of expression ranging from weak to high with thescores from 0 to 6. The glands in the secretory phase showedsignificantly higher levels of expression than those in the pro-liferative phase (P5 0.0011). In endometrial cancers, the scoresranged from 2 to 8, and using a cutoff score of 7, there was clearevidence of overexpression in 33% of cases (13 of 40). Further-more, overexpression was observed when endometrial carcino-mas (Fig. 4E) were analyzed as a group and compared withnormal controls (Mann-Whitney test,P , 0.00001).

Simultaneous Overexpression. In endometrial carcino-mas, there was evidence for the simultaneous overexpression ofmore than one factor analyzed as follows: c-erbB-3 and c-erbB-4 in three cases; c-erbB-3 and c-erbB-4 and betacellulin intwo cases; c-erbB4 and betacellulin in two cases; and for c-erbB-3 and betacellulin in seven cases.

DISCUSSIONThe human endometrium is perhaps the most dynamic

tissue in the body that undergoes cyclical proliferation, differ-entiation, and shedding in response to the female sex hormones,wherein one might expect a differential expression of receptortyrosine kinases in the various phases. In this study, we exam-ined the expression of c-erbB-3 and c-erbB-4 and their principalligands, the NRG-1a, NRG-1b, and betacellulin in the normalendometrium during the two principal phases of the menstrualcycle, the proliferative and the secretory phases, which served asa baseline against which the expression in endometrial adeno-carcinoma was assessed.




The heterogeneity of immunoreactivity for the proteins testedin this study necessitated the use of the scoring system detailed in“Materials and Methods.” This system has been used in previousreports from this laboratory (33, 38, 39). The secretory endometrialglands showed significantly increased levels relative to the prolif-erative phase of the c-erbB-4 receptor and all its ligands examinedincluding NRG-1a, NRG-1 b, and betacellulin. In contrast, c-erbB-3 was expressed weakly in both phases without any differ-ential distribution. This would imply that c-erbB-4 and its ligandsplay a role in the differentiation and maturation of the endome-trium. It is also possible that this group of ligands and the c-erbB-4receptor may be regulated by estrogen, and future experiments onendometrial cancer-derived cell lines should clarify this issue. Theendometrial stroma showed immunoreactivity for c-erbB-3,c-erbB-4, and all of the ligands; however, the levels of expressionwere variable. When the levels of glandularversusstromal expres-sion were compared, c-erbB-4 and NRG-1a showed relativelyhigher levels in the stroma. In previous studies, EGFR has beenreported to be expressed in the normal endometrial glands (25, 26),but a correlation to the menstrual phase was not observed (25). Inanother study, EGFR was reported to be present in the endometrialstromal cells but not in the glands with increasing levels in thesecretory phase (28). The c-erbB-2 receptor, on the contrary, isreported to be expressed only in the glands and not in the stromaand is also apparently up-regulated in the secretory phase (28).

Both the receptors c-erbB-3 and c-erbB-4 showed evidenceof overexpression when the group of endometrial carcinomaswas compared with the normal as a whole. Such an analysis ofthe two groups using the Mann-Whitney test is, perhaps, morerealistic rather than giving arbitrary cutoff values to determineoverexpression. For instance, using a cutoff score of 6, theproportion of c-erbB-3 and c-erbB-4 “overexpressing” tumorswas 39 and 72%, respectively. If a cutoff score of 7 were used,the rates would be 32 and 17%, respectively. The situation withthese two receptors is complicated because of heterogeneity ofexpression, unlike c-erbB-2 positivity where “overexpression”is generally homogeneous and thus more easily defined. How-ever, from the frequency histograms depicted in Fig. 3, it is seenthat a score of 5 or more in the case of c-erbB-3 and a score of7 or more in the case of c-erbB-4 are indicative of expressionabove normal levels. There have been earlier studies on theexpression of EGFR and c-erbB-2 in endometrial carcinomas. Itis worth mentioning here that whereas some studies report“expression,” others report “overexpression.” Thus, EGFR isexpressed in 24% (29) to 67% (26) of endometrial carcinomasand is overexpressed in 49% of cases in a study where acorrelation to the occurrence of metastases was noted (27).c-erbB-2 overexpression is variably reported in different seriesand ranges from as low as 12% (30) to 52% of cases (31). TheERBB2gene is also amplified in about 21% of cases (31). Ingeneral, c-erbB-2 expression has correlated to more aggressivedisease and shortened overall survival (29–31).

Among the ligands, betacellulin showed evidence of over-expression. Betacellulin is a ligand for c-erbB-4 and EGFR andin the light of this finding, the relatively mild overexpression ofc-erbB-4 noted in this study assumes greater importance. Withrespect to the NRGs, underexpression of NRG-1a in endome-trial cancers was significant. A similar observation of loss ofNRG-1 a expression in prostatic adenocarcinoma has been

made in one study (22), whereas in another study on prostaticcarcinoma, up-regulation of the protein was noted (35). We didnot observe significant difference in the levels of NRG-1bexpression in malignant and normal endometria. The role ofNRGs in cancer is very complex, not only because of thenumerous isoforms that are present that may possess differentfunctions, but also in the wide range of receptor combinationsthat they may stimulate. Experiments with breast cancer-derivedcell lines have suggested that this large family of proteins playsa dual role, promoting mitogenesis as well as a differentiation indifferent settings (40, 41)

The occasional staining of the nuclei of endometrial stromalcells with the antibodies against NRG-1a, NRG-1b, betacellulin,and c-erbB-4 is intriguing and merits a more detailed analysis. TheNRGs possess nuclear localization signals in their sequence andhave been shown to undergo rapid nuclear translocation in thebreast cancer-derived SK-BR-3 cell line (42). The c-erbB-4 recep-tor also possesses nuclear localization signals in its sequence, andpreviously we have reported the observation of occasional nuclearstaining seen in the epithelial lining of the distal tubules of thekidney and the ducts of the breast (21).

To summarize, this pilot study highlights that c-erbB-4receptor and its ligands, the NRGsa and b, and betacellulinplay a role in normal endometrial maturation. In endometrialadenocarcinoma, overexpression of the receptors c-erbB-3 andc-erbB-4 and the ligand betacellulin and underexpression ofNRG-1a occur. It is hoped that this study will form the basis foranalyses of these proteins in a larger series of endometrialcancers, which will enable us to identify useful molecular mark-ers and identify targets for new therapies.

REFERENCES1. Alroy, I., and Yarden, Y. The ErbB signaling network in embryo-genesis and oncogenesis: signal diversification through combinatorialligand-receptor interactions. FEBS Lett.,410: 83–86, 1997.2. Wen, D., Peles, E., Cupples, R., Suggs, S. V., Bacus, S. S., Luo, Y.,Trail, G., Hu, S., Silbiger, S. M., Benhery, R., Koski, R. A., Lu, H. S.,and Yarden, Y. Neu differentiation factor: a transmembrane glycopro-tein containing an EGF domain and an immunoglobulin homology unit.Cell, 69: 559–572, 1992.3. Carraway, K. L., III, Weber, J. L., Unger, M. J., Ledesma, J., Yu, N.,Gassman, M., and Lai, C. Neuregulin-2, a new ligand of ErbB-3/ErbB-4-receptor tyrosine kinases. Nature (Lond.),387: 512–516, 1997.4. Chang, H., Riese, D. J., II, Gilbert, W., Stern, D. F., and McMahan,U. J. Ligands for ErbB-family receptors encoded by a neuregulin-likegene. Nature (Lond.),387: 509–512, 1997.5. Busfield, S. J., Michnick, D. A., Chickering, T. W., Revett, T. L.,Ma, J., Woolf, E. A., Comrade, C. A., Dussault, B. J., Woolf, J.,Goodearl, A. D. J., and Gearing, D. P. Characterisation of a neuregulin-related gene,Don-1,that is highly expressed in restricted regions of thecerebellum and hippocampus. Mol. Cell Biol.,17: 4007–4014, 1997.6. Zhang, D., Sliwkowski, M. X., Mark, M., Frantz, G., Akita, R., Sun, Y.,Hilton, K., Crowley, C., Brash, J., and Godowski, P. J. Neuregulin-3(NRG3): a novel neural tissue-enriched protein that binds and activatesErbB4. Proc. Natl. Acad. Sci. USA,94: 9562–9567, 1997.7. Ben-Baruch, N., and Yarden, Y. Neu differentiation factors: a familyof alternatively spliced neuronal and mesenchymal factors. Proc. Soc.Exp. Biol. Med.,206,221–227, 1994.8. Tzahar, E., Pinkas-Kramarski, R., Moyer, J. D., Yanigahara, D., Koski,R. A., Yarden, Y., Klapper, L. N., Alroy, I., Levkowitz, G., Shelly, M.,Henis, S., Eisenstein, M., Ratzkin, B. J., Sela, M., Andrews, G. C., andYarden, Y. Bivalence of EGF-like ligands drives the ErbB signallingnetwork. EMBO J.,16: 4938–4950, 1997.




9. Gullick, W. J., and Srinivasan, R. The type 1 growth factor receptorfamily: new ligands and their receptors and their role in breast cancer.Breast Cancer Res. Treat.,52: 43–53, 1998.10. Tzahar, E., Levkowitz, G., Karunagaran, D., Yi, L., Peles, E., Lavi, S.,Chang, D., Liu, N., Yayon, A., Wen, D., and Yarden, Y. ErbB-3 andErbB-4 function as the respective low and high affinity receptors of all neudifferentiation factor/heregulin isoforms. J. Biol. Chem.,269:25226–25233, 1994.11. Tzahar, E., Waterman, H., Chen, X., Levkowitz, G., Karunagaran, D.,Lavi, S., Ratzkin, B. J., and Yarden, Y. A hierarchical network of interre-ceptor interactions determines signal transduction by neu differentiationfactor/neuregulin and epidermal growth factor. Mol. Cell. Biol.,16: 5276–5287, 1996.12. Shing, Y., Christofori, G., Hanahan, D., Ono, Y., Sasada, R., Igarashi,K., and Folkman, J. Betacellulin: a mitogen from pancreatic beta celltumors. Science (Washington DC),259: 1604–1607, 1993.13. Watanabe, T., Shintani, A., Nakata, M., Shing, Y., Folkman, J., Iga-rashi, K., and Sasada, R. Recombinanat human betacellulin. Molecularstructure, biological activities, and receptor interaction. J. Biol. Chem.,269:9966–9973, 1994.14. Riese, D. J., II, Bermingham, Y., van Raaij, T. M., Buckley, S.,Plowman, G. D., and Stern, D. F. Betacellulin activates the epidermalgrowth factor receptor and erbB-4 and induces cellular response patternsdistinct from those stimulated by epidermal growth factor or neuregulin-b.Oncogene,12: 345–353, 1996.15. Seno, M., Tada, H., Kosaka, M., Sasada, R., Igarashi, K., Folkman, J.,Ueda, M., and Yamada, H. Human betacellulin, a member of the EGFfamily dominantly expressed in the pancreas and small intestine, is fullyactive in a monomeric form. Growth Factors,13: 181–191, 1996.16. Prigent, S. A., Lemoine, N. R., Hughes, C. M., Plowman, G. D.,Selden, C., and Gullick, W. J. Expression of the c-erbB-3 protein in normalhuman adult and foetal tissues. Oncogene,7: 1273–1278, 1992.17. Prigent, S. A., and Gullick, W. J. Identification of c-erbB-3 bindingsites for phosphatidyl 39 kinase and SHC using an EGF receptor/c-erbB-3chimera. EMBO J.,13: 2831–2841, 1994.18. Guy, P. M., Platko, J. V., Cantley L. C., Cerione R. A., and Carraway,K. L. Insect cell expressed p180 erbB3 possesses an impaired tyrosinekinase activity. Proc. Natl. Acad. Sci. USA,91: 8132–8136, 1994.19. Gullick, W. J. The c-erbB-3/HER3 receptor in human cancer. CancerSurv.,27: 339–349, 1996.20. Plowman, G. D., Culouscou, J. M., Whitney, G. S., Green, J. M.,Carlton, G. W., Fox, L., Neubauer, M. G., and Shoyab, M. Ligand specificactivation of HER4/p180erbB4, a fourth member of the epidermal growthfactor receptor family. Proc. Natl. Acad. Sci. USA,90: 1746–1750, 1993.21. Srinivasan, R., Poulsom, R., Hurst, H. C., and Gullick, W. J. Expres-sion of the HER4/c-erbB-4 protein and mRNA in normal human fetal andadult tissues and in a survey of nine solid tumour types. J. Pathol.,185:236–245, 1998.22. Lyne, J. C., Melhem, M. F., Finley, G. G., Wen, D., Lui, N., Deng,D. H., and Salup, R. Tissue expression of neu differentiation factor/heregu-lin and its receptor complex in prostate cancer and its biologic effects onprostate cancer cellsin vitro. Cancer J. Sci. Am.,3: 21–30, 1997.23. Gilbertson, R. J., Perry, R. H., Kelly, P. J., Pearson, A. D. J., and Lunec,J. Prognostic significance of HER2 and HER4 co-expression in childhoodmedulloblastoma. Cancer Res.,57: 3272–3280, 1997.24. Faksvag Haugen, D. R., Akslen, L. A., Varhaug, J. E., and Lillehaug,J. R. Expression of c-erbB-3 and c-erbB-4 proteins in papillary thyroidcarcinomas. Cancer Res.,56: 1184–1188, 1996.25. Berchuck, A., Soisson, A. P., Olt, G. J., Soper, J. T., Clarke-Pearson,D. L., Bast, R. C., and McCarty, K. S. Epidermal growth factor receptorexpression in normal and malignant endometrium. Am. J. Obstet. Gynecol.,161: 1247–1252, 1989.26. Niikura, H., Sasano, H., Kaga, K., Sato, S., and Yajima, A. Expressionof epidermal growth factor family proteins and epidermal growth factorreceptor in human endometrium. Hum. Pathol.,27: 282–289, 1996.

27. Khalifa, M. A., Abdoh, A. A., Mannel, R. S., Haraway, S. D.,Walker, J. L., and Min, K. Prognostic utility of epidermal growth factorreceptor overexpression in endometrial adenocarcinoma. Cancer(Phila.),73: 370–376, 1994.

28. Wang, D., Fujii, S., Konishi, I., Nanbu, Y., Iwai, T., and Nonogaki, H.Expression of c-erbB-2 protein and epidermal growth factor receptor innormal tissues of the female genital tract and in the placenta. VirchowsArchiv. A Pathol. Anat.,420: 385–393, 1992.

29. Wang, D., Konishi, I., Koshiyama, M., Mandai, M., Nanbu, Y., Ish-ikawa, Y., Mori, T., and Fujii, S. Expression of c-erbB-2 protein andepidermal growth factor receptor in endometrial carcinomas. Cancer(Phila.),72: 2628–2637, 1993.

30. Lukes, A. S., Kohler, M. F., Pieper, C. F., Kerns, B. J., Rodriguez,G. C., Soper, J. T., Clarke-Pearson, D. L., Bast, R. C., and Berchuck, A.Multivariable analysis of DNA ploidy, p53 and HER-2/neu as prognosticfactors in endometrial cancer. Cancer (Phila.),73: 2380–2385, 1994.

31. Saffari, B., Jones, L. A., el-Naggar, A., Felix, J. C., George, J., andPress, M. F. Amplification and overexpression of HER-2/neu (c-erbB-2) inendometrial cancers: correlation with overall survival. Cancer Res.,55:5693–5698, 1995.

32. Assikis, V. J., Neven, P., Jordan, J. C., and Vergote, I. A realisticclinical perspective of tamoxifen and endometrial carcinogenesis. Eur. J.Cancer,32A: 1464–1476, 1996.

33. Rajkumar, T., Goden, C. S. R., Lemoine, N. R., Gullick, W. J. Expres-sion of the c-erbB-3 protein in gastrointestinal tumours determined bymonoclonal antibody RTJ1. J. Pathol.,170: 271–278, 1993.

34. Rajkumar, T., and Gullick, W. J. The production of antibodies usingsynthetic peptides as immunogens.In: J. A. Z. Jankowski and J. M. Polak(eds.), Clinical Gene Analysis and Manipulation, pp. 99–110. Melbourne,Australia: Cambridge University Press, 1996.

35. Leung, H. Y., Weston, J., Gullick, W. J., and Williams, G. A potentialautocrine loop between heregulin-a and erbB-3 receptor in human prostaticadenocarcinoma. Br. J. Urol.,79: 212–216, 1997.

36. Tada, H., Sasada, R., Kawaguchi, Y., Kojima, I., Gullick, W. J.,Salomon, D. S., Igarashi, K., Seno, M., and Yamada, H. Membrane an-chored betacellulin: isoforms, juxtacrine activity, and regulated cleavage.J. Cell. Biochem.,72: 423–434, 1999.

37. Wen, D., Suggs, S. V., Karunagaran, D., Liu, N., Cupples, R. L.,Luo, Y., Janssen, A. M., Ben-Baruch, N., Trollinger, D. B., Jacobsen, V.L., Meng, S. Y., Lu, H. S., Hu, S., Chang, D., Yang, W., Yanigahara, D.,Koski, R. A., and Yarden, Y. Structural and functional aspects of themultiplicity of neu differentiation factors. Mol. Cell. Biol.,14: 1909–1919, 1994.

38. Rajkumar, T., Stamp, G. W. H., Pandha, H. S., Waxman, J., andGullick, W. J. Expression of the type 1 tyrosine kinase growth factorreceptors EGF receptor, c-erbB2 and c-erbB3 in bladder cancer. J. Pathol.,179: 381–385, 1996.

39. Rajkumar, T., Stamp, G. W. H., Hughes C. M., and Gullick W. J.c-erbB-3 protein expression in ovarian cancer. J. Clin. Path. Mol. Pathol.,49: 199–202, 1996.

40. Lupu, R., Cardillo, C., Cho, C., Harris, L., Hijazi, M., Perez, K.,Rosenberg, K., Yang, D., and Tang, C. The significance of heregulin inbreast cancer tumor progression and drug resistance. Br. Cancer Res. Treat.,38: 57–66, 1996.

41. Krane, I. M., and Leder, P. NDF/heregulin induces persistence ofterminal end buds and adenocarcinomas in the mammary glands of trans-genic mice. Oncogene,12: 1781–1788, 1996.

42. Li, W., Park, J. W., Nuijens, A., Sliwkowski, M. X., and Keller, G. A.Heregulin is rapidly translocated to the nucleus and its transport is corre-lated with c-myc induction in breast cancer cells. Oncogene,12: 2473–2477, 1996.




1999;5:2877-2883. Clin Cancer Res Radhika Srinivasan, Eleanor Benton, Frances McCormick, et al. Endometrial Cancer

, and Betacellulin, in Normal Endometrium andβNeuregulin-1 ,αFactor Receptors and Their Ligands, Neuregulin-1

Expression of the c-erbB-3/HER-3 and c-erbB-4/HER-4 Growth

Updated version

http://clincancerres.aacrjournals.org/content/5/10/2877

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/5/10/2877.full#ref-list-1

This article cites 39 articles, 13 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/5/10/2877.full#related-urls

This article has been cited by 15 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/5/10/2877To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/content/5/10/2877.full#ref-list-1

http://clincancerres.aacrjournals.org/content/5/10/2877.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



expression of the c-erbb-3/her-3 and c-erbb-4/her-4 growth ... · function act as secondary...

Documents