The human genome contains multiple sequences of varying homology to mouse mammary tumour virus DNA

Oncogenic Viruses and Breast Cancer: Mouse Mammary Tumor Virus (MMTV), Bovine Leukemia Virus (BLV), Human Papilloma Virus (HPV), and Epstein-Barr Virus (EBV)

Background

Although the risk factors for breast cancer are well established, namely female gender, early menarche and late menopause plus the protective influence of early pregnancy, the underlying causes of breast cancer remain unknown. The development of substantial recent evidence indicates that a handful of viruses may have a role in breast cancer. These viruses are mouse mammary tumor virus (MMTV), bovine leukemia virus (BLV), human papilloma viruses (HPVs), and Epstein–Barr virus (EBV-also known as human herpes virus type 4). Each of these viruses has documented oncogenic potential. The aim of this review is to inform the scientific and general community about this recent evidence.

The evidence

MMTV and human breast cancer—the evidence is detailed and comprehensive but cannot be regarded as conclusive. BLV and human breast cancer—the evidence is limited. However, in view of the emerging information about BLV in human breast cancer, it is prudent to encourage the elimination of BLV in cattle, particularly in the dairy industry. HPVs and breast cancer—the evidence is substantial but not conclusive. The availability of effective preventive vaccines is a major advantage and their use should be encouraged. EBV and breast cancer—the evidence is also substantial but not conclusive. Currently, there are no practical means of either prevention or treatment. Although there is evidence of genetic predisposition, and cancer in general is a culmination of events, there is no evidence that inherited genetic traits are causal.

Conclusion

The influence of oncogenic viruses is currently the major plausible hypothesis for a direct cause of human breast cancer.

Breast cancer: is there a viral connection?

Viruses are known to be etiologically related to several types of human cancer. In several published studies, viruses such as human mammary tumor virus, human papillomaviruses, Epstein-Barr virus, human cytomegalovirus, herpes simplex virus, and measles virus have been implicated in the pathogenesis of a subset of breast cancers (BC). However, these studies have produced conflicting results, causing considerable controversy. In this context, recent demonstration of elevated levels of mutagenic antiviral enzyme APOBEC3B in a majority of BCs is a highly significant development, as it provides a possible mechanism for development of large numbers of mutations (kataegis) that characterize many of the BCs. It has also provided further impetus for revaluating the role of viruses in the pathogenesis of BC.

Human Mammary Tumor Virus (HMTV) sequences in human milk

Background

Retroviral sequences 90-95% homologous to the mouse mammary tumor virus (MMTV) were present in 38% of the breast cancers studied from American women and were not detectable in non-tumor breast tissue from the same patient. The entire proviral structure was described and viral particles were isolated from primary cultures of human breast cancer. This virus was designated as human mammary tumor virus (HMTV). Hormone response elements present in the HMTV Long-Terminal-Repeat (LTR) suggest a mechanism for association of HMTV with hormonally responding tissues. In fact, the incidence of HMTV sequences is higher in gestational breast cancers, which are associated with hormonal changes. Milk epithelial cells are also under hormonal regulation and therefore are excellent specimens for HMTV sequence detection.

Methods

The HMTV sequence was studied in milk samples from lactating women recruited with increased risk of breast cancer because they had undergone breast biopsies (Biopsy-Group) and lactating women without breast biopsies (Reference-Group).

Results

HMTV-env sequences were detected by PCR in milk of 7.61% of 92 women of the Reference-Group and in 20.55% of 73 women of the Biopsy-Group (p: 0.015). The sequences were 94-98% homologous to MMTV. HMTV-env and HMTV-env/LTR junction sequences were detected in high-speed pellet RNA, implying the presence of HMTV viral particles. PCR assays to detect the murine mitochondrial cytochrome oxidase gene and intracisternal-A-type particle sequences were performed to rule out mouse mitochondrial or genomic DNA contamination. Eight women of the 73 Biopsy-Group participants had breast cancer and the milk of only one of these eight women had HMTV-env sequences. In the remaining 65 women of the Biopsy-Group, under enough clinical suspicion to lead to biopsy, HMTV was detected in 14, nearly three times the number of milks as compared to the Reference-Group (21.54% versus 7.61%; p: 0.016).

Conclusion

The significance of HMTV in milk from the Reference-Group, the greater frequency in the milk of women who had undergone a breast biopsy and its possible infectivity for infants are important questions under study. The similarity of HMTV to MMTV is striking and suggests one possible avenue for viral transmission in humans.

Lack of association of mouse mammary tumor virus-like sequences in Iranian breast cancer patients

OBJECTIVE

The present study aimed to detect mouse mammary tumor virus (MMTV)-like sequences in Iranian breast cancer patients in the city of Shiraz, located in southwest Iran.

SUBJECTS AND METHODS

We searched for two MMTV genetic regions in the peripheral blood leukocytes of 300 women with breast cancer, 300 age-matched healthy control subjects, and 50 breast tumor tissues. Two regions of MMTV, 660 bp and 250 bp, were searched by nested polymerase chain reactions.

RESULTS

None of the above two regions were detected. There were no differences between the control group and the breast cancer group.

CONCLUSION

Our findings did not show any association of MMTV-like sequences with breast cancer development in Iranian patients in Shiraz.

Human mammary tumor virus in inflammatory breast cancer

The authors have found that retroviral sequences with 85% to 95% homology to the mouse mammary tumor virus were present in 40% of the sporadic breast cancers of American women. These sequences were not found in normal breasts or other tumors. A whole proviral structure was detected in 2 tumors. Breast cancer cells in culture were shown to contain and shed betaretroviral particles. This virus was designated human mammary tumor virus (HMTV). The authors have investigated the presence of HMTV sequences in a variety of breast conditions and geographic locations. Here they report that inflammatory breast cancer from American women shows a higher incidence of viral sequences (71%) than sporadic breast cancers. Similar incidence has been found in inflammatory breast cancers from Tunisia, and in gestational breast cancers. Because these conditions represent highly invasive malignancies, it is concluded that HMTV is sometimes associated with a particularly malignant phenotype.

Mouse Mammary Tumor Biology: A Short History

For over a century, mouse mammary tumor biology and the associated Mouse mammary tumor virus (MMTV) have served as the foundation for experimental cancer research, in general, and, in particular, experimental breast cancer research. Spontaneous mouse mammary tumors were the basis for studies of the natural history of neoplasia, oncogenic viruses, host responses, endocrinology, and neoplastic progression. However, lacking formal proof of a human mammary tumor virus, the preeminence of the mouse model faded in the 1980s. Since the late 1980s, genetically engineered mice (GEM) have proven extremely useful for studying breast cancer and have become the animal model for human breast cancer. Hundreds of mouse models of human breast cancer have been developed since the first demonstration, in 1984, that the mouse mammary gland could be molecularly targeted and used to test the oncogenicity of candidate human genes. Now, very few scientists can avoid using a mouse model to test the biology of their favorite gene. The GEM have attracted a new generation of molecular and cellular biologists eager to apply their skills to these surrogates of the human disease. Newcomers often enter the field without an appreciation of the origins of mouse mammary tumor biology and the basis for many of the prevailing concepts. Our purpose in writing this short history of mouse mammary tumor biology is to provide a historical perspective for the benefit of the newcomers. If Einstein was correct in that "we stand on the shoulders of giants," the neophytes should meet their giants.

HIV-1 Rev can specifically interact with MMTV RNA and upregulate gene expression

We present evidence that the HIV-1 Rev protein can heterologously regulate expression of the simple beta retrovirus mouse mammary tumour virus (MMTV). Up to 10-fold upregulation was seen in a functional assay system when specific MMTV sequences were substituted for the HIV-1 Rev responsive element (RRE). RNA gel shift analysis showed that purified recombinant Rev could specifically bind to MMTV unique region 3 prime (U3) RNA and that these sequences could compete for wild-type Rev-RRE binding approximately 20-fold more efficiently than a non-specific competitor RNA. Using a combination of in silico and deletion mutation analyses, it was not possible to define any single specific secondary structure responsive to Rev, suggesting that a structure or combination of structures that only form in the context of the complete U3 transcript is/are required to interact with Rev. Taken together, these results suggest that HIV-1 Rev can directly bind to MMTV RNA as well as mediate upregulation of MMTV gene expression.

Human endogenous retroviruses in the primate lineage and their influence on host genomes

Primates emerged about 60 million years ago. Since that time various primate-targeting retroviruses have integrated in the germ line of primate species, and some drifted to fixation. After germ line fixation, continued activity of proviruses resulted in intragenomic spread of so-called endogenous retroviruses (ERVs). Variant ERVs emerged, amplified in the genome and profoundly altered genome structures and potentially functionality. Importantly, ERVs are genome modifiers of exogenous origin. The human genome contains about 8% of sequences of retroviral origin. The human ERVs (HERVs) comprise many distinct families that amplified to copy numbers of up to several thousand. We review here the evolution of several well-characterized HERV families in the human lineage since initial germ line fixation. It is apparent that endogenous retroviruses profoundly affected the genomes of species in the evolutionary lineage leading to Homo sapiens.

Of mice, cats, and men: Is human breast cancer a Zoonosis?

Mouse mammary tumor virus (MMTV), a member of the betaretroviridae, is the most common cause of breast cancer (BC) in mice. MMTV is transmitted in mice both in the germline as endogenous proviruses and exogenously as infectious virions. Here, we review a variety of evidence accumulated for six decades that has suggested that a human homologue of MMTV may exist. The findings include recent studies from several independent laboratories that have detected sequences very closely related to MMTV in DNA isolated from human BC tumors. Other laboratories, however, have failed to detect the MMTV-related sequences in human DNA samples, and conclusive evidence for a human mammary tumor virus has been elusive. We also reviewed additional studies, suggesting that betaretroviruses are present in a much wider range of species than previously known, including rodents, felines, and primates. The observation that a subset of cats may be infected with a close homologue of MMTV may be of epidemiological significance for human BC. Cats may become infected by MMTV from mice, and in turn may transmit the virus to humans, possibly after selection for variants with an expanded host range.

Transcription of human endogenous retroviral sequences related to mouse mammary tumor virus in human breast and placenta: similar pattern in most malignant and nonmalignant breast tissues

The human genome contains a large variety of sequences related to the mouse mammary tumor virus (MMTV). We have investigated the range of expression of human endogenous retroviral sequences (HERVs) related to MMTV (human MMTV-like; HML) as RNA in 60 breast cancers, 8 nonmalignant breast tissues, and 9 placentas. This was monitored using HML group-specific oligonucleotide probes in hybridizations toward PCR amplificates of HML pol sequences and internal control. The degree of expression of five HML groups varied between individuals and between tissues. On average, all HML groups were less expressed in breast tissues than in placenta. The hybridization signals of some HML RNAs were strongly correlated, indicating a nonstochastic mechanism and a concerted regulation of their expression. The PCR product from one breast cancer (BC 6), which gave an exceptionally high expression with probe hml-6, with a 20 times stronger signal than the rest of the cancers, was cloned and sequenced. The HML-6 transcript sequences were homogeneous in BC 6. The most predominant clone derived from the cancer was used as a probe in Southern hybridizations. The same restriction fragments were detected in human breast tissues, PBMCs (peripheral blood mononuclear cells), and breast cancer cell lines, except for one of the breast cancers and one of the nonmalignant breast tissues, which gave different banding patterns. A comparison of HML expression in normal and malignant breast tissue from the same individual would have been more precise than our comparison of samples from different persons. Bearing this limitation in mind, with a single exception, human MMTV-like sequences were not more actively expressed in malignant than in nonmalignant breast tissues. Nevertheless, an interesting diversity in their expression, especially between individuals, was found.

Possibilities of a viral etiology for human breast cancer. A review

Previous studies related mouse mammary tumor virus (MMTV) to human breast cancer. However, the presence of human endogenous retroviruses (HERs) confounded these results. We selected a 660-bp sequence of the MMTV env gene with low homology to HER (or any other known gene) and searched for a sequence homologous to it, using the polymerase chain reaction (PCR). The 660-bp sequence was detected in 131 (39%) of 335 unselected breast cancers, in 2 (6.9%) of 29 fibroadenomas, and in 2 (1.65%) of 121 normal breast specimens. The sequence was not present in normal tissues, or in other human cancers or cell lines. Cloning and sequencing of the 660-bp sequence revealed that it is 95-98% homologous to MMTV env gene, but not the known HERs or other viral or human gene. Southern blot hybridization using labeled cloned sequences demonstrated that the 660-bp sequence was present in very low copy number as a 6-8 kb EcoRI fragment only in breast cancer samples and in some of the human breast cancer cell lines that were positive by PCR. Preliminary experiments using reverse transcriptase (RT)-PCR indicated that expression of the 660-bp sequence can be detected in 65% of the positive tumors. We were also able to identify in breast cancer DNA a segment of 1.6 kb comprising LTR and env gene sequences, which are homologous to MMTV, but not to the HERs. The origin of the MMTV-like sequences in tumor DNA could be the result of integrated MMTV-like sequences derived from a human mammary virus.

Mouse mammary tumor virus and mammary tumorigenesis in wild mice

The current knowledge of the distribution of the mouse mammary tumor virus (MMTV) proviral genomes and the mechanism of mammary tumorigenesis by MMTV in mice, with the main emphasis on Asian feral mice, is reviewed. The relevant earlier discoveries on the mode of MMTV transmission are summarized to provide an outline of the biology of MMTV. Finally, the viral etiology of human breast cancer will be discussed.

Differential expression of human endogenous retroviral sequences similar to mouse mammary tumor virus in normal peripheral blood mononuclear cells

Mouse mammary tumor virus (MMTV) is a retrovirus that causes breast cancer in certain strains of mice. In a previous study we identified, by sequencing clones from human lymphocytes, six groups with similarities to MMTV. Using a primer pair derived from pol sequences conserved within types A, B, and D retroviruses and probes from the six human MMTV-like (HML-1 to HML-6) groups in an internally controlled hybridization assay we investigated the normal variation of expression in PBMCs. Variations occurred within all groups but was most significant within group HML-1, where hybridization signals differed by more than 500-fold between individuals. Groups HML-2 and HML-3 showed consistently stronger hybridization signals than groups HML-1 and HML-5, while group HML-6 resulted in weak signals for all individuals. Stringent hybridization of the amplified cDNA to 20 individual HML clones also demonstrated a marked heterogeneity of expression. Hybridization signals from some groups and sequences were found to be correlated, either in a positive or negative fashion. RNA isolated from PBMCs collected from two donors at four different time points (in the morning and in the afternoon on the same day, repeated 1 week later) was also analyzed using the six hml probes. A small variation in hybridization signals was seen in samples collected on the same day, but a larger difference was observed in samples taken 1 week later. The correlations and the differences in the expression of HMLs between individuals implicate a complex transcriptional regulation system of these sequences.

Identification of three human sequences with viral superantigen-specific primers

The open reading frame (ORF) in the long terminal repeat (LTR) of mouse mammary tumor virus (MMTV) has recently been shown to encode multiple products including a negative acting factor (Naf) and a superantigen (Sag). Expression of superantigens from endogenous MMTV loci in the mouse results in the deletion of whole classes of T cells. In a PCR approach, with primers to the MMTV ORF and hybridization to MMTV specific probes, we have identified three human sequences. Direct sequencing of PCR products revealed that one of these products is related to a human autoantigen that is conserved among many species and is expressed in testes and sperm. The second sequence that we have identified is novel, and no evidence for expression of this sequence could be obtained. Finally, the third ORF-like sequence is a new member of a previously described family of human endogenous retroviruses (RTVL-I). This sequence is transcribed in several human cell lines, including B lymphoblastoid cells, and is thus the first demonstration that an RTVL-I-related sequence can be expressed. Taken together, these findings raise the intriguing possibility that the human genome contains superantigen-like sequences, some of which are also related to endogenous retroviruses, that may influence the T cell repertoire.

The human cathepsin D-encoding gene is transcribed from an estrogen-regulated and a constitutive start point

The 5' flanking sequences, exon 1 and part of intron 1 of the human cathepsin D (CTD)-encoding gene (CTD) have been cloned and sequenced. RNase protection experiments identified two major transcription start points (tsp) located 14 and 63 nucleotides upstream of the start codon. The proximal -14, but not the distal -63 tsp has upstream near-concensus TATAAA and CCAAT sequences. Estrogens increase transcription from the -14 tsp, but not the -63 tsp and CTD is therefore unique among estrogen-regulated genes in having estrogen-regulated and constitutive transcription. Sequencing approximately 800 bp upstream and 600 bp downstream of the tsp failed to identify a consensus 13-bp palindromic estrogen-response element (ERE); however, four half-palindrome GGTCA motifs were located within 340 bp upstream of the -14 bp tsp. Thus, estrogen regulation of CTD may not be mediated by a consensus ERE.

Characterization of novel reverse transcriptase encoding human endogenous retroviral sequences similar to type A and type B retroviruses: differential transcription in normal human tissues

The polymerase chain reaction was used to amplify genomic DNA and reverse-transcribed RNA from human lymphocytes, using primers derived from conserved regions within the retroviral reverse transcriptase. Sequencing of 33 cloned amplification products revealed that a variety of sequences with similarity to mouse mammary tumor virus, mouse intracisternal A particle, and human endogenous retrovirus K10 were detected with this primer pair. The sequences were divided into six subgroups, with a nucleotide sequence dissimilarity of about 25% between the subgroups. Members within five of the subgroups were most closely related to human endogenous retrovirus K10 and mouse mammary tumor virus, whereas sequences of the sixth subgroup also showed similarity to mouse intracisternal A particle. Ten of the sequences had open reading frames with preference for silent mutations at conserved sites. Southern blot analysis showed that some HML (human endogenous MMTV-like) subgroups (HML-4 and HML-5) were present in a few copies (about 5), whereas others (HML-1 to HML-3 and HML-6) were present in at least 10 to 20 copies per genome. Northern (RNA) blot analysis revealed that several of the subgroups are differentially expressed in human normal tissues. A complex pattern of transcripts from about 12 to 1.4 kb was found in several of the tissues tested. However, the most abundant expression was detected in lung (all subgroups), skeletal muscle (HML-4 and HML-5), placenta (HML-2 and HML-5), and kidney (HML-2, HML-3 and HML-5). Expression of reverse transcriptase sequences in human tissues may have biological consequences. The described sequences are similar to elements which cause carcinoma and are immunoregulatory in mice. It remains to be seen whether human sequences also have such functions.

Evolution of mouse mammary tumor virus-related sequences in the human genome

The human genome contains numerous copies of elements with sequence homology to the mouse mammary tumor virus (MMTV). We have been interested in whether these elements are still actively transposing in the germ lines, and investigated this question by looking at restriction fragment patterns. Genomic DNA from humans and selected animals were digested with three different restriction enzymes and hybridized with five probes representative of five subgroups of the human MMTV-related elements. Two polymorphisms as well as two cases of sexual dimorphism were discovered in human DNA. The latter indicate the presence of at least two elements of the MMTV family on the Y chromosome. DNA samples from non-primate mammals were all unreactive. Both chimpanzee and rhesus monkey had restriction patterns of roughly the same number of bands and total intensity as humans. The number of differences in patterns between humans, and between humans and chimpanzee or rhesus monkey, were compatible with being caused by point mutations. The data indicate that these elements were actively spreading in the genome some time before the split between apes and Old World monkeys, while being relatively inactive, at least as to transpositions in the germ line, in later ages.

Antibodies in human sera recognizing a recombinant outer membrane protein encoded by the envelope gene of the human endogenous retrovirus K

Human endogenous retrovirus (HERV) K is present in about 50 copies in the human genome and transcription of HERV-K has previously been detected in several tumor cell lines as well as in peripheral blood lymphocytes from healthy donors. By means of immunoblotting we investigated the presence of antibodies recognizing recombinant HERV-K-envelope outer membrane constructs in different serum collectives. A total of 12.6% of sera obtained from normal blood donors was found positive. In serum collectives from breast carcinoma patients, HIV-1-positive individuals, and persons with cytomegalovirus infections no significant difference from the normal blood donor serum collective could be observed. Only a group of persons with a repeatedly raised serum neopterin concentration (> 10 nmol/liter) of unknown cause (HIV and hepatitis B and C virus infections were excluded) showed a significant higher percentage of HERV-K-outer membrane envelope-positive sera (21%). Furthermore we could observe a parallel HIV-1/HERV-K seroconversion, which probably is not due to an HIV-1/HERV-K-outer membrane envelope cross-reactivity. Possible implications of these findings are discussed.

Long terminal repeats of dwarf hamster endogenous retrovirus are highly diverged and do not maintain efficient transcription

Recently we described a new endogenous proretrovirus of dwarf hamster Phodopus sungorus (MRS-Ps). Its sequence possesses evident homology with the endonuclease domain of the mouse mammary tumor virus pol gene. Here we present nucleotide sequence data on three clones of retroviral long terminal repeats. As many as 15% of substituted, deleted, and inserted base pairs were found while comparing these sequences. Hence, MRS-Ps seems to be rather an old genetic element which originated about 30 million years ago. One LTR is 877 bp long and contains numerous elements that control its transcriptional activity: TATA-box, glucocorticoid responsive element, NF1-binding site, etc. Nevertheless, this LTR does not govern efficient transcription of adjacent genes in a transient expression assay. In addition, we failed to find MRS-specific mRNA in adult, embryonic, and mammary tumor cells.

Distribution of mouse mammary tumor virus-related sequences does not correlate with the taxonomic position of their hosts

Sequences (MRS) distantly related to mouse mammary tumor virus (MMTV) were found in genomes of a wide range of mammalian species using blot hybridization. The number of MRS copies and the degree of their homology with the hybridization probe varied and did not correlate with the taxonomic position of the species. Nevertheless, within a genus the set of MRS was species specific and reflected the taxonomic relation between the species. MRS were also found in avian genomes and the degree of their homology did not correlate with the taxonomic position of the species either. The origin and distribution of MRS is discussed on the basis of the authors' and published data.

NCL-CB11, a new monoclonal antibody recognizing the internal domain of the c-erbB-2 oncogene protein effective for use on formalin-fixed, paraffin-embedded tissue

The c-erbB-2 proto-oncogene encodes a 190 k Mr protein representing a putative growth factor receptor with considerable homology to the EGF receptor. Gene amplification and overexpression of the oncogene protein have been demonstrated in a variety of tumours including breast cancer, where it is associated with a poor prognosis. In this study we have produced and characterized a mouse monoclonal antibody, designated NCL-CB11, to the c-erbB-2 protein. NCL-CB11 was generated to a synthetic peptide sequence corresponding to a site of predicted antigenicity near the C-terminus of the internal domain of the protein. NCL-CB11 produces intense membrane-associated immunohistochemical staining in a proportion of human cancer cells. The specificity of the antibody is supported by Western blotting and immunoprecipitation studies. Reactivity with an internal site of the protein is confirmed by the necessity of cell permeabilization for reactivity in fluorescence-activated cell sorter (FACS) analysis. A high degree of correlation between immunohistochemical staining using NCL-CB11, and c-erbB-2 gene amplification has been observed. NCL-CB11 should prove to be a valuable reagent for investigations into the pathological significance of c-erbB-2 protein expression.

Expression of the oestrogen regulated pNR-2 mRNA in human breast cancer: relation to oestrogen receptor mRNA levels and response to tamoxifen therapy

The pNR-2 mRNA is regulated by oestrogens in cell lines established from metastatic human breast cancer cells. The levels of the pNR-2 and oestrogen receptor RNAs have been measured in 96 tumour samples from patients undergoing surgery for breast cancer. Oestrogen receptor mRNA was detected in 90% of the 60 primary breast tumour samples from patients not receiving endocrine therapy at the time of surgery, whereas the pNR-2 RNA was detected in 57%. In primary tumours the expression of pNR-2 was entirely dependent upon oestrogen receptor RNA expression. When the 60 primary tumours were considered, pNR-2 and oestrogen receptor mRNA levels were significantly correlated. There was no significant correlation for pNR-2 positive tumours. pNR-2 mRNA levels were similar in tumours of pre- and post-menopausal patients and were independent of tumour differentiation and nodal status. Oestrogen receptor and pNR-2 mRNA levels were also measured in 21 tumour samples from patients receiving primary tamoxifen therapy. Eleven of these had shown an objective response and a significantly larger number of tumours from these patients contained pNR-2 mRNA than from patients who did not respond (chi 2 = 6.08, P less than 0.025).

Measurement of oestrogen receptor mRNA levels in human breast tumours

A sensitive single-stranded hybridisation probe for the oestrogen receptor mRNA was synthesised using T3 polymerase from oestrogen receptor cDNA cloned in the Bluescript vector. This probe was used to measure oestrogen receptor mRNA in total RNA extracted from breast tumours. Oestrogen receptor mRNA was detected in 41 of 47 (87%) tumours whereas cytosolic oestrogen receptor protein was detected in only 18 out of 39 (46%). There was a significant correlation between the levels of the oestrogen receptor, as measured by 3H-oestradiol binding, and the oestrogen receptor mRNA. Oestrogen receptor mRNA was detected in 15 of the 21 tumours that did not contain detectable oestrogen receptor protein. This suggests that detection of the mRNA is a more sensitive means for establishing oestrogen receptor status than the radioligand oestrogen receptor assay. Oestrogen receptor mRNA was found in all histological tumour types examined. Its level was related to tumour differentiation. All the tumours that did not contain oestrogen receptor mRNA and most of the tumours that contained only low levels of oestrogen receptor mRNA were classified as grade III according to Bloom and Richardson: the median of the oestrogen receptor mRNA levels was significantly lower in this group.

Expression of human sequences related to those of mouse mammary tumor virus

Sequences related to those of the mouse mammary tumor virus (MuMTV) genome have been cloned from human DNA by screening a library prepared from the DNA of a human breast cancer cell line with MuMTV gag-pol DNA. Nine distinct groups of (MuMTV-related) sequences were identified among 100 lambda recombinants by cross-hybridization experiments with subcloned fragments containing gag-pol-related DNA. The largest group, of 64 recombinants, contains the MuMTV-related sequences cloned by others. The other eight groups contain MuMTV-related sequences that have not been described previously. The gag-pol regions of one recombinant from each of the nine groups were hybridized to RNA prepared from five human breast cancer cell lines, from placenta, and from two cell lines derived from other malignancies. RNAs were detected by probes for seven of the groups. The RNAs ranged in size from 1.2 to 12 kilobases. Probes for six of the groups detected large RNAs that could represent transcripts of full-length proviral DNA. Two of the probes detected RNA in one breast cancer cell line only. Most of the RNAs were detected in more than one cell line.

A novel human nonviral retroposon derived from an endogenous retrovirus

In a human genome, we found dispersed repetitive sequences homologous to part of a human endogenous retrovirus termed HERV-K which resembled mouse mammary tumor virus. For elucidation of their structure and organization, we cloned some of these sequences from a human gene library. The sequence common to the cloned DNA was ca. 630 base-pairs (bp) in length with an A-rich tail at the 3' end and was found to be a SINE (short interspersed repeated sequence) type nonviral retroposon. In this retroposon, the 5' end had multiple copies of a 40 bp direct repeat very rich in GC content and about the next 510 nucleotides were homologous to the 3' long terminal repeat and its upstream flanking region of the HERV-K genome. This retroposon was thus given the name, SINE-R element since most of it derived from a retrovirus. SINE-R elements were present at 4,000 to 5,000 copies per haploid human genome. The nucleotide sequence was ca. 90% homologous among the cloned elements.

Stimulation of expression of the human endogenous retrovirus genome by female steroid hormones in human breast cancer cell line T47D

Expression of the human endogenous retrovirus genome HERV-K, homologous to mouse mammary tumor virus, was investigated in cultured human tumor cells. In several cell lines, the HERV-K genome was expressed as an 8.8-kilobase poly(A)+ RNA which appeared to be a full-size transcript of this genome. In the human breast cancer cell line T47D, stimulation of HERV-K genome expression by progesterone was observed after estradiol treatment.

Oestrogen regulates cathepsin D mRNA levels in oestrogen responsive human breast cancer cells

A cDNA library was constructed from the mRNA of the ZR-75 oestrogen responsive human breast cancer cell line and screened for oestrogen regulated mRNA sequences. Of the recombinants isolated, 30 contained cDNA that corresponded to a single mRNA species of 2.1 kb that was induced between 10 and 15 fold by oestradiol treatment. The sequence of the entire open reading frame and 3' non-coding region of the mRNA was determined and shown to encode the aspartyl protease cathepsin D. The induction of cathepsin D mRNA is specific for oestrogen and is maximal at 3 X 10(-10) M. Cathepsin D mRNA levels were increased by oestrogen in 3 oestrogen responsive breast cancer cell lines. Cathepsin D mRNA was expressed but not regulated in an oestrogen receptor negative breast tumour cell line, BT 20, and in 2 other malignant cell lines, Hela and A431.

Isolation of an SSAV-related endogenous sequence from human DNA

We have found human DNA to contain a number of sequences related to simian sarcoma associated virus (SSAV). One of these sequences was isolated from a human genomic library. The molecular clone, termed S71, contains regions homologous to SSAV gag and pol fragments and SSAV LTR. Furthermore, hybridization experiments and DNA sequencing revealed distinct homologies to the reverse transcriptase coding region of several other retroviruses including baboon endogenous virus (BaEV) and murine leukemia viruses (MuLV) as well as retrovirus-like elements. Some sequence homology was also found with the C-type retrovirus-related multicopy human clone 4-1. S71 is present in only one copy per human genome equivalent and exhibits an EcoRI restriction fragment length polymorphism.

Structure of a human retroviral sequence related to mouse mammary tumor virus

Human sequences that are related to the mouse mammary tumor virus (MuMTV) genome were cloned from breast tumor cell DNA. Of 100 recombinants, only 1 hybridized with two different probes from separate regions of the MuMTV genome (gag-pol and long terminal repeat [LTR]). This sequence, NMWV 4, was shown to have a proviruslike structure. Hybridization to digests of normal and tumor cell DNA indicated that NMWV 4 and a few closely related sequences are endogenous to the human genome. The regions that contain homology to the MuMTV LTR were sequenced. Long repeated sequences with the hallmarks of retroviral LTRs were identified. The NMWV 4 LTR contains transcription initiation and termination signals and is flanked by a polypurine tract (5' LTR) and a primer-binding site (3' LTR). The primer-binding site is complementary to tRNA lysine, the primer used by MuMTV and HTLV-III. The polypurine tract is also similar to those of these two retroviruses.

Nucleotide sequence of human endogenous retrovirus genome related to the mouse mammary tumor virus genome

We determined the complete nucleotide sequence of the human endogenous retrovirus genome HERV-K10 isolated as the sequence homologous to the Syrian hamster intracisternal A-particle (type A retrovirus) genome. HERV-K10 is 9,179 base pairs long with long terminal repeats of 968 base pairs at both ends; a sequence 290 base pairs long, however, was found to be deleted. It was concluded that a composite genome having the 290-base-pair fragment is the prototype HERV-K provirus gag (666 codons), protease (334 codons), pol (937 codons), and env (618 codons) genes. The size of the protease gene product of HERV-K is essentially the same as that of A- and D-type oncoviruses but nearly twice that of other retroviruses. A comparison of the deduced amino acid sequences encoded by the pol region showed HERV-K to be closely related to types A and D retroviruses and even more so to type B retrovirus. It was noted that the env gene product of HERV-K structurally resembles the mouse mammary tumor virus (type B retrovirus) env protein, and the possible expression of the HERV-K env gene in human breast cancer cells is discussed.

The nucleotide sequence of the env gene from the human provirus ERV3 and isolation and characterization of an ERV3-specific cDNA

The nucleotide sequence of the env gene of a previously described human provirus (ERV3) has been determined beginning near the 3'-end of the pol gene and continuing through the 3'-LTR. Analysis of the nucleotide sequence revealed the presence of a long open reading frame of 1944 nucleotides that is capable of encoding a polypeptide that has characteristics of other retroviral glycoproteins and transmembrane proteins. These include the presence of seven potential glycosylation sites, a typical glycoprotein-transmembrane protein cleavage sequence, and amino acid homologies to the glycoproteins and transmembrane proteins of other retroviruses. Further, we have isolated an ERV3-specific cDNA clone from a library prepared from liver RNA of a 20-week human fetus. DNA sequence analysis of this clone revealed that it is identical to the ERV3 genomic clone in the 1110 nucleotides that were sequenced.