Detection of Mammary Tumor Virus ENV Gene-like …routinely tested. A DNA and control primers provided by the manufacturer were used as control for polymerase activity. As an internal

ICANCER RESEARCH55, 5173-5179, November 15, 19951

Advances in Brief

Abstract

Mouse mammary tumor virus (MMTV) has been related to humanbreast cancer (BC) in previous studies. Although suggestive sequencehomology to MMTV has been described in BC DNA, the presence of

human endogenous retroviruses (HERs) confounded these results. Wehave selected a 660-bp sequence of the MMTV ens' gene with very lowhomology to HER or to any other human or viral gene. We have searched

for sequences homologous to it using the polymerase chain reaction. DNAwas extracted from fresh or frozen tissues using primers and probesconstructed to detect 660 bp; for paraffin-embedded tissues, we sought250-bp sequences by similar methodology.

The 660-bp sequence was detected in 121 (38.5%) of the 314 unselectedBC samples, In cultured BC cells, In 2 (6.9%) of 29 breast flbroadenomasand In 2 (1.8%) of 107 breast specimens from reduction mammoplastias.The sequence was not found in normal tissues including breast, lymphocytes from BC patients, nor in other human cancers or cell lines. The250-bp sequence was detected in 60 (39.7%) ofthe 151 BCs, and in 1 of 27normal breast samples assayed from paraffin-embedded sections. Oonin@

and sequencing of the 660 bp and 250 bp demonstrated that they are95-.99% homologous to MMTV env gene, but not to the known LEERsnorto other viral or human genes (<18%).

Southern blot analysis using labeled doned sequences showed that the660-bp sequences were present in low copy number as a 7—8-kbEcoRIfragment only In breast cancer samples and two breast cancer cell linesthat were positive by PCR.

These data indicate that 38—40%ofhuman breast cancers contain genesequences homologous to the MMTV env gene that are absent from othertumors and tissues. These MMTV env gene-like sequences may play a role

in the etiology of a large proportion of human breast cancer

Introduction

Studies of animal oncogenic retroviruses have been fundamental inrevealing the presence and properties of human cellular proto-oncogenes. MMTV,3 an agent associated with a high incidence of BC incertain strains of mice (over 90% among females), has been regardedas a potential model for human disease. Efforts to demonstrate thepresence of viruses in human BC through search for viral particles,immunological cross-reactivity, or sequence homology have yieldedcontradictory results. Several lines of evidence, however, associateMMTV with human BC. Detectable MMTV env gene-related antigenic reactivity has been found in tissue sections of BC (1—3),BC

cells in culture (4), human milk (5) in sera of patients (6), in cyst fluid(7), and in particles produced by a human breast carcinoma cell line(8). Sequence homology to MMTV has been found in human DNAunder low stringency conditions of hybridization (9), and RNA relatedto MMTV has been detected in human BC cells (10). The presence ofMMTV-related sequences in lymphocytes from patients with BC hasbeen reported (1 1) as well as detection of reverse transcriptase activityin their monocytes (12). These results have been difficult to interpretbecause they could not be distinguished from those potentially attributed to endogenous retroviral sequences (13—15)and the fact thatenv-gene-related antigenicity has been detected in epitopes of humanproteins (16). Nevertheless, May and Westley (17) have reported thepresence of MMTV-like sequences arranged as tandem repeats only inDNA from BC cells.

A large body of information has accumulated about the molecularbiology of MMTV (reviewed in Ref. 18). The virus does not carry atransforming oncogene, but rather acts as an insertional mutagen withseveral proviral insertion loci designated mt-i or wnt-1 (19) int-2 (20),int-3 (21), int-4 (22), and mt-S (23), which encode for growth factorsor other related proteins. These genes are not expressed in normalmammary tissue but become activated after integration of MMTVprovirus into the adjacent chromosomal DNA.

The human homologue of the int-2 locus has been located onchromosome 11 (24) and has been found amplified (in 15% of theBCs) and also expressed (25—28). It may be significant that in tumors

from Parsi women, who have a high incidence of breast tumors, theint-2 locus is amplified in 50% of the cases (29). The amplification ofint-2 and other genes in 11q13 is indicative of a poor prognosis (30,31). Both mouse and human int-2 have been sequenced (32). The geneencodes a protein of about Mr 27,000, which shows homology to bothbasic and acidic fibroblast growth factors (33).

The evidence discussed above suggests that a MMTV-like agentmight play a role in human BC. In this communication, we report theresults of a search for unique sequences of the MMTV env gene in apanel of randomly selected human BCs using DNA amplificationtechnology.

Materials and Methods

DNA from breast and other human cancers, human placentas, normalhuman tissues including breast, and from several human cell lines includingeight BC lines and two normal breast cell lines was extracted according to theprocedure of Delli Bovi et a!. (34). The DNA was resuspended in a solution

containing 0.05 MTris-HC1 buffer (jH 7.8) and 0.1 mM EDTA, and the amountrecovered was determined by microfluorometry using Hoechst 33258 dye (35).Plasmids containing the cloned genes of MMTV were obtained from theAmerican Type Culture Collection, propagated in Escherichia coli cultures,and purified using anion exchange minicolumns (Qiagen) or by precipitationwith polyethylene glycol(36). Oligonucleotide primers were synthesized at thecore facilities of the Brookdale Molecular Biology Center at Mount SinaiSchool of Medicine.

PCR was performedusing Taq polymerasefollowing the conditions rec

Received 8/3/95; accepted 9/26/95.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

I Supported by the T. J. Martell Foundation for Leukemia, Cancer, and AIDS Re

search, The Chemotherapy Foundation, the Charlotte Geyer Foundation, Derald H.Ruttenberg Fund, Ellen Block Memorial Fund, Faith Price-Rick Kash Fund, and MyraShaw Cancer Fund.

2 To whom requests for reprints should be addressed, at Division of NeoplasticDiseases, Mount Sinai School of Medicine, CUNY, New York, NY 10029.

3 The abbreviations used are: MMTV, mouse mammary tumor virus; BC, breast

cancer.

5173

Detection of Mammary Tumor Virus ENV Gene-like Sequences in Human BreastCancer'

Yue Wang, James F. Holland, Ira J Bleiweiss, Stella Melana, Xiangjun Liu, LsabellePelisson, Annette Cantarella,Kathleen Stellrecht, Sridhar Mani, and Beatriz G-T. Pogo2Department of Medicine, Division of Neoplastic Diseases (1'. W., J. F. H., S. M., X. L, I. P., K. S., S. M., B. G-T. P.], Departtnen:s of Pathology (1. J. B.] and Microbiology(A. C., B. G-T. P.], Mount Sinai School ofMedicine, CUNY, New York. New York 10029

Research. on January 25, 2021. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table 1 Primer and probe sequences and location in the MMTV envgeneDesignationSequence

(5—3')Location1CCFCACFGCCAGATC976-990laGGGAA1TCCFCACfGCCAGATC976-990iNCCTCACTGCCAGATCGCCT976-9932TACATCFGCCTGTGTrAC1388-14052NCCFACATCFGCCFGTGTrAC1386-14052aCCGCCATACGTGCTG1554-15683ATCI'GTGQCATACCI'1640-16263aGGGAATFCATCFGTGGCATACCI@1640-16263NATCFGTGGCATACCTAAAGG1640-16214GAATCGCflGGCFCG1661-16475LCCAGATCGCCITfAAGAAGG984-10033LTACAGGTAGCAGCACGTATG1558-1577

HUMAN BREAST CANCER AND RETROVIRAL SEQUENCES

ommended by the manufacturer (Perkin Elmer/Cetus) with regard to buffer,Mg2@,and nucleotide concentrations. Thermocycling was performed in a DNAcycler by denaturation at 94°Cfor 3 mm followed by either 35 or 50 cycles of94°Cfor 1.5 mm, 50°Cfor 2 mm, and 72°Cfor 3 mm. The ability of the PCRto amplify the selected regions of the MMTV env gene was tested by using aspositive templates the cloned MMTV env gene and the genomic DNA of theMCF-7 cell line, since it was shown to express gp52 immunological determinants (37). Optimal Mg2@,primer concentrations, and requirements for thedifferent cycling temperatures were determined with these templates. Themaster mix as recommended by the manufacturer was used. To detect possiblecontamination of the master mix components, a reaction without template was

routinely tested. A DNA and control primers provided by the manufacturerwere used as control for polymerase activity. As an internal control, amplification of a 120-bp sequence estrogen receptor gene was assayed using primersdesigned and generously provided by Dr. Beth Schachter (Mount Sinai Schoolof Medicine, New York, NY). In addition, primers for actin 5 gene amplification were also used.

The product of the PCR was analyzed by electrophoresis in a 2% agarosegel. A 1-kb DNA ladder (GIBCO-BRL) was used to identify the size of thePCR product. To determine whether the amplified sequences of the middleregion of the 660 bp faithfully reproduced the sequences of the env gene of

MMTV, an 18-mer sequence within the env gene was used as a probe for the660-bp amplified sequence. The 18-mer probe was 5' end labeled with [32P1ATP using T4 polynucleotide kinase and purified by the NENSORB nucleicacid purification cartridge (New England Nuclear). Southern blot hybridizationwas performed using the conditions described by Saiki et aL (38).

The product of the PCR (660 bp or 250 bp) was cloned directly from thereaction mixture into the TA cloning vector (Invitrogen) using the TA cloningkit and following the conditions recommended by the supplier. Direct cloningof the fragment isolated from the gel was also performed. Plasmid DNA waspurified by CsCl density gradient centrifugation or by precipitation withpolyethylene glycol (36), restricted with Hindlll and EcoRI, electrophoresed in

2% agarose gels, and transferred to nitrocellulose filters. Southern blot hybrid

ization was carried out using a 5 ‘-terminal-labeled internal probe as described

above. Cloning procedures were performed in laboratories totally separatefrom those where PCR was carried out. Automated DNA sequencing (usingApplied Technology Sequencer model 373A) was performed in the BrookdaleMolecular Biology Center. Sequence homology was determined using the IBIMacVector GenBank and GCG programs. To prevent contamination of thesamples, processing of human tissues was performed in a laminar flow hood.DNA extractions were done in a chemical hood located in a different roomfrom that where PCR was performed. PCR assays were assembled in abiological hood provided with UV. Aerosol-resistant tips and dedicated positive-displacement pipettes were used throughout. All equipment used for PCR(microcentrifuge, electrophoresis apparatus, micropipettes) was cleaned eachtime with 10% sodium hypochlorite to assure DNA decontamination (39).After the initial experiments were performed, the plasmid containing theMMTV env gene was frozen and never used again to avoid contamination.

However, to detect plasmid contamination from our own env gene clones,primers were designed to amplify plasmid sequences. All of the MMTVenv-positive samples were then tested for plasmid contamination.

Southern blotting and hybridization were performed as described (40) usingthe 660-bp cloned sequences labeled by the random primer procedure (41).Prehybridization and hybridization were performed in a solution containing6 X saline-sodium phosphate-EDTA, 5% Denhardt's solution, 0.5% SDS, 50%

formamide, and 100 @g/mldenaturated salmon testis DNA, incubated for 18 hat 42°C,followed by washings with 2 X SSC and 0.5% SDS at roomtemperature and at 37°C,and finally in 0.1 X SSC with 0.5% SDS at 68°Cfor30 mm (36). For paraffin-embedded tissue sections, the conditions describedby Wright and Manos (42) were followed using primers designed to detect a250-bp sequence.

Results

Selection of Specific MMTV env Gene Sequences. A computersearch for MMTV env gene homologous sequences was first performed, since sequence homology between the human endogenousretroviral sequences and MMTV had been described. The prototype ofthis group of human endogenous retroviruses is HERV-K10 (14). The

sequences of the env gene of MMTV (43) were aligned with se

quences of the env gene of the human endogenous retrovirus HERVK10 (14) using the IBI/Pustell Sequence Analysis Program. A regionof 660 bp of low homology (16%) was localized between MMTV envgene sequences 976 and 1640 (43). This internal domain of the outermembrane of the env gene has only one glycosylation site and ishighly conserved between strains. Two primers comprising 15-bpsequences at positions 976—990(primer 1) and 1626—1640(primer 3)were first synthesized. Later, longer primers were synthesized (iNand 3N). An 18-mer sequence in the middle ofthe 660-bp MMTV envregion (1388—1405 primer 2) was used as a probe to identify the660-bp sequence. A second oligomer probe was synthesized comprising the sequence 1554—1568(primer 2a) to be used for hybridizationwhen a sequence of around 250 bp (between positions 1388 and 1640)was amplified. For nested PCR reactions (44), another primer cornprising sequences 1647—1661(primer 4) was synthesized to be usedwith primer 1 in the first reaction and primers 2 and 3 in the second.Modified primers with GC clamps and extra sequences were alsosynthesized and used in the PCR (primers la and 3a). Another set ofprimers comprising sequences 974—1003 (5L) and 1558—i577 (3L)were subsequently developed because their Tm's matched and provided better amplification than the original primers. The sequences arerepresented in Table 1. All of them were able to amplify the expectedsequences.

Detection of MMTV-like env Gene Sequences in Human BreastTumor DNA. PCR was performed on DNA extracted from BCs,normal breast, and the plasmid containing the env gene of MMTVusing primers 1 and 3. Photographs of the ethidium bromide-stainedgels of the PCR product reveal the presence of an approximately660-bp sequence in some of the tumors (Fig. 1A Lanes 1 and 3), butnot in the normal tissue samples (Fig. 1A, Lanes 2 and 4). As apositive control the MMTV env gene was also amplified (Fig. 1A,Lane if). Similar results were obtained with modified primers la, 3a,3L, and 5L (data not shown). Southern blot hybridization of the gelwith 32P-labeled 18-mer oligonucleotide indicated that this internalsequence was present in the amplified material (Fig. 1B) and that thebands in the gel were not artifactual.

Our initial effort was to analyze a representative sample of BCspecimens as well as normal tissues and other tumors. To date, 343breast tumors have been processed, DNA extracted, and PCR performed. Of these 343 tumors, 314 were carcinomas and 29 werefibroadenomas. Amplification of sequences of 660 bp was observed in121 (38.5%) of the carcinomas and in 2 (6.9%) of the 29 fibroadenomas. These sequences were confirmed to be MMTV env gene-likesequences by hybridization with the labeled specific probe containingthe internal sequences. These sequences were not detected in theDNAS extracted from 20 normal organs, 23 cancers from other organs, and 26 samples of blood lymphocytes including 7 from BCpatients whose breast specimens were positive. From 107 samples ofnormal breast obtained from reduction mammoplasties, 2 (1.8%) were

5174

Research. on January 25, 2021. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table 2 Detection of MMTVenv gene-like sequences in human DNf reshorfrozen tissuesA

extractedfromMMTV

envgeneSamplensequences%PositiveBreast

carcinomas31412138.5Breastfibroadenomas2926.9Normal

breasts10721.8Normalbreasts―4negativeTumorsother than breast23negativeNormaltissues20negativeLymphocytes

Lymphocytes―26negative

7 negative

Human cell linesMMTV env genesequenceMCF-7(breast

carcinoma)PositiveT47-D(breastcarcinoma)NegativeBT-20(breastcarcinoma)NegativeMDA-MB-231(breastcarcinoma)NegativeZR-75-1(breastcarcinoma)NegativeSK-BR

3(breastcarcinoma)NegativeBT474(breastcarcinoma)NegativeED(breastcarcinoma)PositiveMCF-10(normalbreast)NegativeHB-447(normalbreast)NegativeHL-60(promyelocyticleukemia)Negative1(562(erythroleukemia)NegativeJurkat(T-cell

leukemia)NegativeHep6-2(hepatoma)Negative

HUMAN BREAST CANCER AND RETROVIRAL SEQUENCES

BA

Fig. 1. Amplification of 660 bp of MMTV-likeenv gene. DNA was extracted from frozen tissues.PCR was performed using primers 1 and 3. A, 2%agarose gel electrophoresis; B, Southern blot hybridization using 5' 32P end-labeled probe 2. Lanes1 and 3, BC; Lanes 2 and 4, normal breast; Lanes5, controlreaction(no DNA); lanesE, MM1'V envgene. M, molecular weight marker. Arrow, 510-bpband.

E 12345 E

shown in Fig. 3B. Using this procedure, we have analyzed 151 BCsamples and found that 60 (39.7%) possess the 250-bp sequence. Ofthe 27 normal breast samples obtained from reduction mammoplastiesassayed using this procedure, one was positive (3.7%). These results,in conjunction with those obtained from lymphocytes and normalbreast tissue of patients whose breast cancer was PCR positive,indicate that MM@1V-likesequences are present in a significant nurnber of human BC DNA which cannot be explained by DNA polymorphism.

Cloning and Sequencing of the MMTV-like env Gene Sc.quences. To find out whether there was homology to MMTV envgene throughout the whole 660-bp stretch, the product of the PCRfrom eight different tumors was cloned and sequenced. In Fig. 4 thesequence of different clones comprising around 600 bp are represented, as aligned to the MMTV env gene sequence of the GR andBR6 strains (45). This domain of the env gene in the OR strain is100% homologous to the C3H strain and 98% to the BR6 strain (43,46). Evaluation of the clones indicated that homology to MMTV envgene varied from 95 to 99%. Another seven clones comprising only250 bp were also sequenced. Homology to the MMTV env genevaried from 95 to 99% (data not shown). When compared to thehuman endogenous provirus HERV-KiO, the homology of all of theclones was <15%. When compared against all known viral and humangenes using the IBI MacVector GenBank and GCG programs, thehighest homology recorded was 18%.

Table 3 Detection of MMTVenv gene-like sequences in DNAfrom human cell linesin culture

a Histologically normal tissue from same breast as positive cancer.b Lymphocytes from BC patients who were positive for MMTV env gene sequences in

the tumor.

positive. In addition to DNA from lymphocytes from seven positivepatients, DNA from their normal homolateral breast tissue was testedin four cases. All were negative (Table 2). Finally, DNA ofthe MCF-7and ED@ breast cancer cell lines were shown to contain the 660-bpMMTV env gene-like sequences (Table 3), while four other breastcancer cell lines were positive only for the 250-bp sequence (T47-D,BT-474, BT-20, and MDA-MB-231).

The nested PCR was used in several instances to increase sensitivity and specificity, thus reducing the probability of false positives. In

Fig. 2, results of a representative nested reaction are shown usingprimers 1 and 4 in the first reaction (Fig. 14) and primers 2 and 3 forthe second reaction. The specificity of the reaction can be seen in thesecond amplification (Fig. 2B).

To study a large number of samples and to be able to performarchival studies, PCR of paraffin-embedded tissue sections was alsocarried out. Primers 2 and 3 were used to amplify a 250-bp sequencewithin the 660-bp stretch when DNA was extracted from paraffinembedded tissue sections, since larger size sequences are difficult toamplify after fixation. Tumor DNA was amplified (Fig. 3A, Lanes2—5),whereas normal breast DNA was not (Fig. 3A, Lane 1). Theidentification of this 250-bp sequence with the MMTV-like env genewas confirmed by hybridization with an internal probe (primer 2a) as

4 ED is a cell line developed in our laboratory from the pleural effusion of a patient

with an env-positive tumor.

5175

M12345

Research. on January 25, 2021. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

HUMAN BREAST CANCER AND RETROVIRAL SEQUENCES

A B

Fig. 2. Nested PCR. A, 2% agarose gel electrophoresis. 1, amplification of686 bp of MMTV-likeenv gene sequences from BC DNA using primers 1and 4. 2, amplification of 250 bp of MMTV-likeenv gene sequences using primers 2 and 3 and theproduct of reaction A 1 as template. B 1 and 2,Southern blot hybridization of the amplified products using probe 5' 32P end-labeled probe 2a.

@t'.@

.-

1 2 1 2

Genomic Southern Blot Analysis Using Cloned Sequences. Toinvestigate whether the env gene-like sequences were present inhuman DNA, Southern blot hybridization was performed using thecloned sequence as a probe. DNAS from normal breast normal tissues,env positive- or -negative breast tumors, tumors other than breast, and

breast cancer cell lines were restricted with EcoRI and in someinstances with PstI, BgIII, or KpnI. EcoRI is a frequent cutter restriction enzyme that digests MMTV proviral DNA between env and poigenes. Four different cloned 660-bp sequences were used as probesafter labeling with 32P by random prime labeling. Results of some ofthe Southern blot hybridization experiments are shown in Figs. 5 and6. They reveal the presence of a labeled restriction fragment migratingat approximately 7—8kb in BC DNA, ED, and two fragments inMCF-7 cells. Different restriction patterns were observed with theother three enzymes. The 660-bp sequence was absent in 10 normaltissues, 10 fibroadenomas, and 10 tumors from other tissues (data notshown).

It is important to emphasize that hybridization conditions for these

experiments were stringent to avoid interference with endogenoussequences that might interact with the probes.

Discussion

The search for virus-related sequences in human breast cancer hasbeen hampered by great variation reported in previous studies, by thepresence of endogenous retroviral sequences in human DNA, and bythe lack of sensitivity of the methods used. The studies reported hereincircumvent these deficiencies by focusing on sequences with lowhomology to human endogenous retroviruses, by investigating a largenumber of tumors and several types of controls, and by using the mostsensitive technology presently available.

The results indicate that unique MMTV env gene sequences arepresent in 38.5% of the BC samples analyzed and 39.7% of archivalsamples of BC, and that these sequences are absent in normal tissues,including lymphocytes from patients with positive BC and in cancersother than breast. Normal breast tissue and fibroadenomas have a low

A B

12345

Fig. 3. Amplification of 250 bp of MMTV-likeenv gene. DNA was extracted from paraffin-embedded tissue sections. PCR was performed usingprimers 2 and 3. A, 2% agarose gel electrophoresis;B, Southern blot hybridization using 5' 32P-labeledprobe 2a. Lane 1, normal breast; Lanes 2—5,BC;Lanes E: MMTV env gene. M, molecular weightmarker.

E5176

Research. on January 25, 2021. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

HUMAN BREAST CANCER AND RETROVIRAL SEQUENCES

*

0 100 200 300I I I I I I I

I I I I

I I I

I I I•III 11111 II

I I I I

I II I

I I III II

I I I I

I Ill II

*

0 100 200 300 400 500 600MMTENVGR I I I I I I I I I I 1 I I I I I I I I I I I I I I

CIonel66 • I I I I I I 1 Ill I

CIone33O • I I I I

I U I I@ I@I 1 I I

I I I I II II I@ I I @I

I I I I I II•I @II I

I I IU I I I I I

M A B C D E F 6 1—1 Furthermore, experiments in which the cloned amplified sequenceswere used for hybridization with DNA from BCs or normal tissues

@@ *@, i revealed that homologous DNA was only present in BC DNA.

. The detection of MMTV env gene sequences in 2 fibroadenomas of

, .@@ 29 and in 2 normal breasts from 107 samples is of uncertain signifi

@@ cance. Although extreme precautions were taken during extraction of

- DNA, performance of the PCR and cloning procedures, the possibility

of cross-contamination with positive samples cannot entirely be cx. cluded. Alternatively, the sequences may represent histologically un

@ . recognized cells that were or will be neoplastic.

j BCs were 90% invasive ductal carcinomas, which reflects the

prevalence of this type of neoplasm. Most patients were node positive,. Fig. 5. Southern blot hybridization of genomic DNA. DNA was extracted from frozen which is probably artifactual since it was necessary that tumor size be

tissues or cell lines, digested with EcoRI, and transferred to nitrocellulose paper. Hybrid- . . .ization with 32P-labeled clone 166. DNA from A, B, and G, env gene-positive BC, C and sufficiently large to provide an aliquot for research, and tumor size

D, env-negative BC, E and F, normal breast; H, MCF-7 cells. M, molecular weight correlates with node positivity. Clinical characteristics of patients withmarker, Arrowhead, 9-kb band. . . .

or without viral sequences will be published elsewhere.It is unlikely that differences in homology between the MMTV env

frequency (1 .8—6.9%) of positive results. When cloned and se- gene and the cloned human sequences are generated by errors cornquenced, the sequences were found to be highly homologous to the milled by the Taq polymerase. It has been estimated that the rate ofMMTV env gene, but not to the endogenous retroviral sequences (14). nucleotide misincorporation is 1 X i05/cycle (47), and, therefore,

5177

400500600III II I I

MMTENVBR

Clonel 66

CIone33O

CIone332

CIone49O

CIone6Ol

CIone6O8

CloneT6

CloneEDFig. 4. Nucleotide sequence of the cloned

MMTV env gene-like sequences as compared tothe env sequences of the GR and BR6 strains ofMMTV using the GCG program. , potential glycosylation site; I, mismatch to MMTV.

C1one332

CIone49O

CIone6O 1

Clone6O8

CloneT6

CloneED

Research. on January 25, 2021. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

HUMAN BREAST CANCER AND RETROVIRAL SEQUENCES

clinical disease status, may provide another potential molecularmarker, and may distinguish a subset of human breast cancer forwhich viral etiology is tenable. This has implications for epidemiology and potentially for therapy and prevention.

Acknowledgments

14 We are indebted to Drs. Beth Schacter, James Strauchen, Thomas Fasy,Swan Thung, Michael Shafir, Rafael Mira Lopez, and Micsunica Platica, allfrom Mount Sinai School of Medicine, for their generosity in providing us withvaluable materials and to Dolores Klaft for excellent secretarial help.

References

1. Mesa-Tejada, R., Keydar, I., Ramanarayanan, M., Ohno, T., Fenogiio, C., andSpiegelman, S. Detection in human breast carcinomas of an antigen immunologicallyrelated to a group-specific antigen of mouse mammary tumor virus. Proc. NatI. Aced.Sci.USA, 75:1529—1533,1978.

2. Levine, P., Mourali, N., Tabbave, F., Costa, J., Mesa-Tejada, R., Spiegelman, S.,Muenz, R., and Bekesi, Y. G. Immunopathologic features of rapidly progressingbreast cancer (RPBC) in Tunisia. Proc. Am. Assoc. Cancer Res, 21: 170, 1980.

3. lloyd, R., Rosen, P. P., Sarkar, N. H., Jimenez, D., Kinne, D. W., Menendez-Botet,C., and Schwartz, M. K. Murine mammary tumor virus related antigen in human malemammary carcinoma. Cancer (Phila.), 51: 654—661, 1983.

4. Litvinov, S. V., and Golovkina, T. V. Expression of proteins immunologically relatedto murine mammary tumour virus (MMTV) core proteins in the cells of breast cancercontinuous lines MCF-7, T47D, MDA-231 and cells from human milk. Ada Virol.,33: 137—142,1989.

5. Zofler, S., Kemmer, C., Lossnitzer, A., Grossmann, H., and Johannsen, B. A. Mousemammary tumour virus-related antigens in core-like density fractions from largesamples of women's milk. Eur. J. Cancer, 16: 455—467, 1980.

6. Day, N. K., Witkin, S. S., Sarkar, N. H., Kinne, D., Jussawalla, D. J., et a!. Antibodiesreactive with murine mammary tumor virus in sera of patients with breast cancer:Geographic and family studies. Proc. NatI. Aced. Sci. USA, 78: 2483—2487, 1981.

7. Witkin, S. S., Sarkar, N. H., Kinne, D. W., Breed, C. N., Good, R. A., and Day, N. K.Antigens and antibodies cross-reactive to the murine mammary tumor virus in humanbreast cyst fluids. J. Clin. Invest., 67: 216—222, 1981.

8. Keydar, I., Ohno, T., Nayak, R., Sweet, R., Simoni, F., Weiss, F., Karby, S.,Mesa-Tejada, R., and Spiegelman, S. Properties of retrovirus-like particles producedby a human breast carcinoma cell line: immunological relationship with mousemammary tumor virus proteins. Proc. Natl. Acad. Sci. USA, 81: 4188—4192, 1984.

9. Callahan, R., Drohan, W., Tronick, S., and Schlom, J. Detection and cloning ofhuman DNA sequences related to the mouse mammary tumor virus genome. Proc.NatI. Aced. Sci. USA, 79: 5503—5507, 1982.

10. Axel, R., Schiom, J., and Spiegelman, S. Presence in human breast cancer of RNAhomologous to mouse mammary tumor virus RNA. Nature (Land.), 235: 32—36,1972.

11. Crepin, M., Lidereau, R., Chermann, J. C., Pouillart, P., Magdamenat, H., andMontagnier, L. Sequences related to mouse mammary tumor virus genome in tumorcells and lymphocytes from patients with breast cancer. Biochem. Biophys. Res.Commun., 118: 324—331, 1984.

12. Ai-Sumidaie, A. M., Hart, C. A., Leinster, S. J., and Green, C. D. Particles withproperties of retroviruses in monocytes from patients with breast cancer. Lancet, 1:5—8,1988.

13. Westley, B., and May, F. E. B. The human genome contains multiple sequencesof varying homology to mouse mammary tumour virus DNA. Gene, 28: 221—227,1984.

14. Ono, M., Yasunaga, T., Miyata, T., and Ushikubo, H. Nucleotide sequence of humanendogenous retrovirus genome related to the mouse mammary tumor virus genome.J. ViM., 60: 589—598, 1986.

15. Fail, 0., Murray, A. B., Schmidt, J., Leib-Mosch, C., Erfie, V., and Hehlmann, R.Retrovirus-like particles from the human T47D cell line are related to mousemammary tumor virus and are of endogenous origin. J. Gen. Virol., 73: 1087—1097, 1992.

16. Hareuveni, M., and Lathe, R. Breast cancer sequences identified by mouse mammarytumor virus (MMTV) antiserum are unrelated to MMTV. mt. j. Cancer, 46: 1134—1135, 1990.

17. May, F. E. B., and Westley, B. R. Characterization of sequences related to the mousemammary tumor virus that are specific to MCF-7 breast cancer cells. Cancer Rca.,49:3879—3883, 1989.

18. Slagle, B. L., and Butel, J. S. Exogenous and endogenous mouse mammary tumorviruses: replication and cell transformation. In: D. Medina, W. Kidwell, G. Heppner,and E. Anderson (eds.), Cellular and Molecular Biology of Mammary Cancer, pp.275—306.New York: Plenum Press, 1987.

19. Nusse, R., and Varmus, H. E. Many tumors induced by the mouse mammary tumorvirus contain a provirus integrated in the same region of the host genome. Cell, 31:99—109,1982.

20. Peters, G., Brookes, S., Smith, R., and Dickson, C. Tumorigenesis by mouse mammary tumor virus: evidence for a common region for provirus integration in mammarytumors. Cell, 33: 369—377, 1983.

21. Gallahan, D., Kozak, C., and Callahan, R. A new common integration region (int-3)for mouse mammary tumor virus on mouse chromosome 17. J. Virol., 61: 218—220,1987.

5178

ABCDEFGH

Fig. 6. Southern blot hybridization of genomic DNA. Experimental conditions as inFig. 5. DNA from A and B; env-negative BC; C and D, env-positive breast cancer; E,molecular weight marker (nonlabeled), F—H,normal breast. Arrowhead, position of 9-kbmarker.

only a total of 0.32 nucleotides misincorporated should be expected in660 bp after 50 cycles.The differencesin homologybetweenclonesfrom different patients is likely to represent heterogeneicity of the envgene.

Our results are in agreement with earlier less stringent studiesthat suggested the involvement of the MMTV env gene in humanBC (1—8).Preliminary experiments also indicated that these sequences are expressed in some of the breast tumors and in BC cellsin culture5. We were unable to confirm reports, however, thatindicated that, as in the mouse, MMTV-like sequences were foundin lymphocytes from two patients with BC (11). The absence ofMMTV env-like sequences in lymphocytes could reflect the fate aunique lymphocyte subset over decades between initial encounterand the appearance of clinical breast cancer; contrawise, the humandisease may differ from the mouse model. Attempts to identifyunique MMTV-like pol gene sequences in human BC samples havebeen unsuccessful, because MMTV-pol gene sequences cannot bedistinguished from endogenous reverse transcriptase sequences(48).

The origin of the MMTV env gene-like sequences found in tumorDNA could be the result of integrated MMTV-like sequences from ahuman mammary tumor virus. Polymorphism of endogenous retroviral sequences is conceivable but can be ruled out because thesesequences were not detected in lymphocytes from the positive patients, in sections of breast with tumors in which abnormal cells wereabsent, nor in normal breast tissue. Recombination during tumorigenesis between endogenous sequences to resemble the MMTV env geneseems highly unlikely since no known human or viral sequence ismore than 18% homologous to the 660-bp sequence. Thus, the mostconservative interpretation is that our findings represent exogenoussequences from an agent similar to MMTV. Recombination between

endogenous and exogenous env gene sequences are known to accelcrate the development of malignancies in mice (49). Whether the660-bp sequences belong to an entire acquired provirus or to anexogenous fragment integrated into the endogenous sequences ispresently not known. Experiments are in progress to distinguishbetween these possibilities.

Several genetic alterations have been identified in human BC thatcan be useful as markers for prevention, detection, or prognosis(reviewed in Ref. 50). BRCa-1 and -2 genes have recently beendescribed. They account for 5% of the BC and are related to familialBC (51, 52). We have preliminaryevidence thatfamilial clusteringofthe MMTV env gene-like sequences occurs, accounting for an evenhigher percentage of cancers in affected families (53). The presence ofMMTV env gene-like sequences may be correlated with special

5 V. Go, Y. Wang, J. F. Holland, and B. G-T. Pogo. Detection of MMTV env gene-like

sequences expression by RT-PCR in human breast cancer and cell lines, manuscript inpreparation.

Research. on January 25, 2021. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

HUMAN BREAST CANCER AND RETROVIRAL SEQUENCES

22. Roelink, H., Wagenaar, E., Lopes da Silva, S., and Nusse, R. Wn:-3, a gene activatedby proviral insertion in mouse mammary tumors, is homolous to int-i/Wnt-1 and isnormally expressed in mouse embryos and adult brain. Proc. Nati. Acad. Sci. USA,87:4519—4523,1990.

23. Morris, V. L, Rao, 1. R., Kozak, C. A., Gray, D. A., Lee Chan, E. C. M., Cornell,1. J., Taylor, C. B., Jones, R. F., and McGrath, C. M. Characterization of Int-5, alocus associated with early events in mammary carcinogenesis. Oncogene Res., 6:53—63,1991.

24. Casey, G., Smith, R., McGillivray, D., Peters, G., and Dickson, C. Characterizationand chromosome assignment of the human homolog of int-2, a potential protooncogene. Mol. Cell Biol., 6: 502—510,1986.

25. Lidereau, R., Callahan, R., Dickson, C., Peters, G., Escot, C.. and Au, i. U. Amplification of the int.2 gene in primary human breast tumors. Oncogene Res., 2:285—291,1988.

26. Zhou, D. J., Casey, G., and Cline, M. J. Amplification of human int-2 in breastcancers and squamous carcinomas. Oncogene, 2: 279—282, 1988.

27. Liscia, D. S., Merlo, 0. R., Garrett, C., French, D., Marini-Constantini, R., andCallahan, R. Expression of int-2 mRNA in human tumors amplified at the int-2 locus.Oncogene,4: 1219—1224,1989.

28. Meyers, S. L., O'Brien, M. 1., Smith, T., and Dudley, J. P. Analysis ofthe mt-I, int-2,c-myc, and neu oncogenes in human breast carcinomas. Cancer Res., 50: 5911—5918,1990.

29. Bamabas-Sohi, N., Simha, M. R., Parikh, V. A., Feiulhade, F., Kurkure, A., Kouyoumdjian, J. C., Jussawalla, D. J., Doctor. V. M., and Therwath, A. Breast carcinomain a high-risk population: structural alterations in neu, int@2and p53 genes. BreastDis., 6: 13-26, 1993.

30. Schuwring, E., Verhoeven, E., van Tinteren, H., Peterse, L., Nunnink, B., Thunnissen,F. B. J. M., Devilee, P., Cornelisse, C. J., van de Vijver, M. J., and Mooi, W. J.Amplification of genes within chromosome 11q13 region is indicative of poorprognosis with operable breast cancer. Cancer Res., 52: 5229—5234, 1992.

31. Champeme, M-H., Bieche, I., Lizard, S., and Lidereau, R. ilql3 amplification inlocal recurrence of human primary breast cancer. Genes Chromosomes & Cancer, 12:128—133,1995.

32. Moore, R., Casey, G., Brookes, S., Dixon, M., Peters, 0., and Dickson, C. Sequence,topography and protein coding potential of mouse int-2: a putative oncogene activatedby mouse mammary tumor virus. EMBO J., 5: 919—924, 1986.

33. Dickson, C., and Peters, G. Potential oncogene product related to growth factors.Nature (Land.), 326: 833, 1987.

34. Delli Bovi, P., Donti, E., Knowles, D. M., II, Friedman-Kien, A., Luciw, P. A., Dma,D., Dalla-Favera, R., and Basilico, C. Presence of chromosomal abnormalities andlack of AIDS retrovirus DNA sequences in AIDS-associated Kaposi's sarcoma.Cancer Res., 46: 6333—6338, 1986.

35. Cesarone, C., Bolognesi, C., and Santi, L. Improved microfluorometric DNA determination in biological material using 33258 Hoechst. Anal. Biochem., 100: 188—197,1979.

36. Sambrook, J., Fritsch, E. F., and Maniatis, T. Molecular Cloning/A LaboratoryManual. Cold Spring Harbor, NY: Cold Spring Harbor Press, 1989.

37. Yang, N. S., McGrath, C. M., and Furmanski, P. Presence of a mouse mammarytumor virus-related antigen in human breast carcinoma cells and its absence fromnormal epithelial cells. J. Nati. Cancer Inst., 61: 1205—1208,1975.

38. Saiki, R. K., Scarf, S., Faloona, F., et al. Enzymatic amplification of @-globingenomic sequence and restriction site analysis for diagnosis of sickle cell anaemia.Science (Washington DC), 230: 1350—1354, 1985.

39. Prince, A. M., and Andrus, L. PCR: how to kill unwanted DNA. Biotechniques, 12:358—360, 1992.

40. Southern, E. M. Detection of specific sequences among DNA fragments separated bygel electrophoresis. J. Mol. Biol., 98: 503—517,1975.

41. Feinberg, A. P., and Vogelstein, B. A technique for radiolabeling DNA restrictionendonuclease fragments to high specific activity. Anal. Biochem., 132: 6—13,1983.

42. Wright, D. K., and Manos, M. M. Sample preparation from paraffin-embedded tissuesin PCR protocols, pp. 153—158.In: A. M. Innis, D. H. Gelfand, J. J. Sninsky, and T. J.White (eds.), PCR Protocols: A Guide to Methods and Applications. New York:Academic Press, 1990.

43. Majors, J. E., and Varmus, H. E. Nucleotide sequencing of an apparent proviral copyof env mRNA defines determinants of expression of the mouse mammary tumor virusenv gene. J. Virol., 47: 495—504,1983.

44. Mullis, K. B., and Faloona, F. A. Specific synthesis of DNA in vitro via a polymerasecatalysed chain reaction. Methods Enzymol., 155: 335—350,1987.

45. Redmon, S. M. S., and Dickson, C. Sequence and expression of the mouse mammarytumor virus env gene. EMBO J, 2: 125—131,1983.

46. Moore, R., Dixon, M., Smith, R., Peters, G., and Dickson, C. Complete nucleotidesequence of a milk transmitted mouse mammary tumor virus: two frameshift suppression events are required for translation of gag and pol. J. Virol., 61: 480—490,1987.

47. Ehrlich, H. A.. Gelfand, D, and Sninsky, J. J. Recent advances in the polymerasechain reaction. Science (Washington DC), 252: 1643—1651,1991.

48. Deen, K. C., and Sweet, R. W. Murine mammary tumor virus pol-related sequencesin human DNA: characterization and sequence comparison with the complete mmmcmammary tumor virus pol gene. J. Virol., 57: 422—432, 1986.

49. DiFronzo, N. L., and Holland, C. A. A direct demonstration of recombinationbetween an injected virus and endogenous viral sequences, resulting in the generationof mink cell focus inducing viruses in AKR mice. J. Viral., 67: 3763—3770, 1993.

50. Runnenbaum, I. R., Nagarajan, M., Bowman, M., Soto, D., and Sukumar, S. Mutations in p53 as potential molecular markers for human breast cancer. Proc. Nail. Aced.Sci. USA, 88: 10657—10661,1991.

51. Miki, Y., Swensen, J., Shattuck-Eidens, D., Futreal, P. A.. Harshman, IC, et al. Astrong candidate for the breast and ovarian cancer susceptibility gene BRCAI. Science(Washington DC), 266: 66—71,1994.

52. Wooster, R., Neuhausen, S. L, Mangion, J., Quirk, Y., Ford, D., et al. Localizationof a breast cancer susceptibility gene BRCA2, chromosome 13q12—13. Science(Washington DC), 265: 2088—2090, 1994.

53. Holland, J. F., Pogo, B. G-T., Wang, Y., Bleiweiss, I., Shafir, M., Strauchen, J.,

Melana, S., and Invernale, A. Human familial breast cancers contain MMTV envgene-like sequences. Proc. Am. Assoc. Cancer Res., 35: 218, 1994.

5179

Research. on January 25, 2021. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

1995;55:5173-5179. Cancer Res   Yue Wang, James F. Holland, Ira J. Bleiweiss, et al.   in Human Breast Cancer

Gene-like SequencesENVDetection of Mammary Tumor Virus

  Updated version

  http://cancerres.aacrjournals.org/content/55/22/5173

Access the most recent version of this article 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://cancerres.aacrjournals.org/content/55/22/5173To request permission to re-use all or part of this article, use this link

Research. on January 25, 2021. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from