Genetic diversity of spontaneously developed primary and metastatic mammary tumor cells in mice

Metastatic cells are often considered to be clonal derivatives of one of the primary tumor cell subpopulations. To determine if the cells of spontaneously developed lung nodules in a mammary tumor-bearing mouse represent the major or minor population of the cells in the primary tumor, comparisons were made of the pattern of their mouse mammary tumor virus (MMTV) proviral integrations, and their insertional mutations of the mouse mammary tumor proto-oncogenes, int-1 and int-2. Of the 78 tumor-bearing C3H/He mice sacrificed, seven mice showed metastatic lung nodules, but only four mice had nodule tissues adequate for the present analysis. Examination of the primary tumors and the corresponding lung nodules of these four mice revealed that the number of newly integrated MMTV proviruses and their sites of integration in the genomic DNAs of primary tumors and tumor nodules were variable, and that the int-1 gene was disrupted in three of the primary tumors but not in any of the metastasized tumor tissue. These results support the notion that, at least in some mice, the majority of cells constituting the primary mammary tumors and the corresponding metastases are of different genotypes and raise the possibility that the activation of different int-proto-oncogenes may be involved in the genesis of different cell subpopulations including metastatic cells in the same mouse.     

A congenic line of the BALB/c mouse strain with the endogenous mouse mammary tumor virus proviral gene Mtv-3: tissue-specific expression and correlation with resistance to mouse mammary tumor virus infection and tumorigenesis

Mouse mammary tumor virus (MMTV) expression and MMTV-induced tumorigenesis were studied in a congenic line of the BALB/cHeA strain, termed BALB/c-Mtv-3+, that carries the Mtv-3 proviral gene. BALB/c-Mtv-3+ mice were free of milk-transmitted MMTV and did not spontaneously develop mammary tumors. A specific Mtv-3 expression was observed in the mammary gland and spleen, but not in other lymphoid tissues, such as thymus and bone marrow. This expression was hormone dependent, as shown by the increase of MMTV mRNA during pregnancy. At the protein level, large amounts of p28, but only traces of gp52, the main MMTV core and envelope antigens, respectively, were observed, in agreement with the already described "partial" expression of the Mtv-3 gene products. The presence of the 24S (3.8 kilobases) mRNA encoding the MMTV env antigens in the spleen and the low gp52 reactivity in lactating mammary glands showed that this noncoordinate expression was probably due to a defect in translation or posttranslational processing of env proteins. The susceptibility of BALB/c-Mtv-3+ to experimental MMTV infection was studied. The presence of Mtv-3 conferred to BALB/c mice resistance to MMTV infection, as shown by measuring viral antigens released in the milk of infected mice and by recording the incidence of early mammary tumors. The presence of a nontumorigenic endogenous MMTV gene was therefore protective against exogenous MMTV infection.     

Characterization of Mammary Plaques in DDD Mice Congenic for Mtv-2 Gene, DDD/1-Mtv-2/Mtv-2

DDD/1 mice free from exogenous mouse mammary tumor virus (MMTV) do not develop any neoplastic mammary lesions. In GR mice, the expression of Mtv-2 , an endogenous proviral MMTV, leads to 100% incidence of mammary ductal hyperplasias and tumors. An Mtv-2 congenic line, DDD/1- Mtv-2 / Mtv-2 , was established by introducing Mtv-2 from GR into DDD/1 to elucidate its function. Development of mammary plaques (MPQ) characterized by ductal hyperplasias was investigated in 152 congenic females on day 17 to 19 of the first pregnancy. The incidence of MPQ was 48.0% and most MPQ-positive mice (75.3%) had only one MPQ. Generally, MPQ were small in size: the diameter was as small as ≦3 mm in 77.6% of them. Of 84 MPQ implanted into intact fat pads, 43 (51.2%), 38 (45.2%) and 3 (3.6%) showed undetectable, pregnancy-dependent and autonomous growths; respectively when the hosts underwent pregnancy. Almost all MPQ produced normal-appearing ductal-alveolar outgrowths in mammary epithelium-divested or cleared fat pads of virgins. MPQ implanted into cleared fat pads were very similar to normal mammary glands in the responses to progesterone (P) and estradiol (E) alone or in combination except for association of ductal hyperplasias in 4 of 12 MPQ under E+P treatment. These findings revealed the preneoplastic nature of MPQ. Exogenous MMTV proviruses were demonstrated in all MPQ. The int-2 DNA rearrangement was found in 2 of 10 MPQ but in none of 9 mammary carcinomas and the int-1 DNA rearrangement in none of 10 MPQ but in 5 of 10 carcinomas. It is thus likely that the Mtv-2 gene participates in a very early stage of mammary tumorigenesis not directly but indirectly through insertion mutation of host genes, while the cellular oncogenes, int-2 and int-1 , may contribute to preneoplastic transformation of mammary epithelium and progression from preneoplastic to more malignant states, respectively.     

The effect of parity, tumor latency and transplantation on the activation of int loci in MMTV-induced, transplanted C3H mammary pre-neoplasias and their tumors.

Mouse mammary tumor virus (MMTV) infection of mammary glands results in proviral insertion into host DNA and activation of cellular genes. Clonal expansion of cells bearing insertional mutations results in hyperplastic alveolar nodules (HAN) and tumors. HAN, transplanted into epithelium-cleared mammary fat pads, form hyperplastic alveolar outgrowths (HOGs). Previous work indicates the commonly MMTV-activated genes wnt-1 and int-2 are rarely affected in HOGs and HOG-derived tumors. To determine the basis of the dichotomy between the frequency of wnt/int gene activation in HOG-derived tumors and tumors from breeders of the identical inbred mouse strain, we compared the activation of wnt-1, int-2 and int-3 in tumors from virgin and breeding C3H mice, in consecutive primary tumors arising in individual C3H breeders and in C3H HOGs at early passages. Activation of wnt-1 or int-2 was rare in HOG-derived tumors (6% and 0%) compared with primary tumors in breeders (52% and 14%). int-3 was never found to be activated. wnt-1 was activated in the same percentage of primary tumors from virgins as from breeders. int-2 was activated only in tumors from breeders. wnt-1 activation also did not correlate with shorter tumor latency in multiple tumors from individual breeders. wnt-1 RNA was not detected in HOGs at early transplant generations, however, low levels of wnt-1 RNA were variably found in the epithelium of virgin mammary glands. We cannot explain why C3H HOGs and their derivative tumors develop without wnt-1 expression when the majority of C3H primary mammary tumors possess an MMTV-activated wnt-1 gene.     

Activation of endogenous MMTV proviruses in murine mammary cancer induced by chemical carcinogen

A study was undertaken to determine whether activation of expression of silent endogenous mouse mammary tumor virus (MMTV) proviruses may occur during tumor induction by a chemical carcinogen. A series of transplantable mammary tumors induced in BALB/c mice by treatment with dimethylbenz(alpha)anthracene (DMBA), pituitary isograft, or both was examined. The results obtained suggest that chemical carcinogens may induce mammary tumors through more than one pathway. Two of 9 tumor lines produced virus-specific products at levels above those observed during the course of normal mammary gland development. One tumor contained high levels of MMTV-specific envelope [3.8 kilobase (kb)] and genomic length (8.9 kb) RNAs. This tumor expressed core- and envelope-related proteins detectable by immunoblotting (including p28, gp52, and gp36), displayed an acquired provirus with a restriction map different from those of described exogenous MMTV strains, and contained abundant virus particles. The other tumor that expressed high levels of MMTV gene products contained envelope-specific (3.8 kb) and long-terminal-repeat-specific (1.6 kb) messages but no full-length RNA. It exhibited an aberrant 39 kDa, envelope-related protein, but no virus particles. Methylation data implicated the usually silent endogenous Mtv-8 provirus as the source of the abnormal envelope protein. None of the tumors expressed RNA from the putative mammary oncogenes, int-1 or int-2. We propose that chemical carcinogens may activate different cellular genes by mutation and that, in a subset of DMBA-induced mammary tumors, the target genes include endogenous MMTV proviruses that are normally not expressed. The effect on provirus expression varies from tumor to tumor, but is stable over passage of a given tumor. MMTV may be of etiological importance in the genesis of those DMBA-induced tumors which contain high levels of MMTV-specific products, but its action in the BALB/c system is not mediated through enhanced expression of the int-1 or int-2 preferred integration regions.     

Evidence of separate pathways for viral and chemical carcinogenesis in c3h/stwi mouse mammary glands

Mammary tumors in mice may arise as the result of exogenous infection with the mouse mammary tumor virus [MMTV(S)], usually via the milk; by the action of endogenous MMTV genes which are transmitted genetically and are sometimes expressed as infectious virus [MMTV(L)]; or by the action of chemical carcinogens. We have examined the etiological relationship between chemical and virus in the induction of mammary cancer in C3H/StWi mice. Mammary tumors were induced with 7.12-dimethylbenz(a)anthracene (DMBA) in virgin C3H/StWi mice infected with exogenous mouse mammary tumor virus (C3H/StMTV) and in uninfected C3H/StWi females. The percent tumor risk in females whose glands were infected with exogenous mouse mammary tumor virus [MMTV(S)] was not different from that of MMTV(S)-negative, C3H/StWi mice following treatment with DMBA. This result suggested that there was no synergistic effect between the two carcinogens, exogenous MMTV and Dmba. All the tumors arising in DMBA-treated C3H/StMTV virgins were positive for MMTV env gene product, gp52 and MMTV gag gene product, p²⁷ by radioimmune competition assay. MMTV(S)-induced hyperplastic alveolar nodules (HAN) were observed in 62 % of the untreated C3H/StMTV glands at 8 months of age. Therefore, MMTV(S) was present and active in the mammary gland during the experimental period but had no enhancing influence on the sensitivity of the gland to carcinogenesis by DMBA. Tumors appearing in DMBA-treated C3H/StWi virgin females were also tested for MMTV gp⁵² and p²⁷ antigens to determine the presence of endogenous MMTV gene activity. Only occasional C3H/StWi tumors were positive and in most of these, p²⁷, but not gp⁵² was detected, suggesting non-coordinate expression of these endogenous MMTV gene products. Both alveolar (HAN) and ductal hyperplasia (DH) were found in DMBA-treated C3H/StMTV glands, whereas only DH were found in DMBA-treated C3H/StWi mice. Nevertheless, the DMBA-induced C3H/StMTV tumor histopathology was remarkably indistinguishable from that in tumors produced by DMBA in C3H/StWi mice, implying that in both groups, tumors arose primarily from the chemically-induced mammary dysplasias. These data taken together appear to support the conclusion that DMBA and Mmtv follow separate pathways to the induction of cancer in the mouse mammary gland.     

Expression of int-2 mRNA in human tumors amplified at the int-2 locus

Gene amplification is a relatively frequent event in human malignant tumors and is believed to have an important function in neoplastic transformation and tumor progression. Our attention has been focused on the amplification and the expression of the int-2 gene for several reasons: (1) In the mouse mammary tumorigenesis int-2 is frequently activated by MMTV proviral integration. (2) The human homolog of int-2, located on chromosome 11q13, is frequently amplified in human primary tumors and is comprised in an amplification unit encompassing the hst gene, which is often coamplified; the amplification at the 11q13 locus in breast carcinomas correlates with a poor outcome of the disease. (3) int-2 and hst belong to the basic FGF gene family. All these observations raise the possibility that the human int-2 gene plays an active role in the neoplastic process, but this will prove to be true only if int-2 is expressed in human tumors. In the present study we used RNA:RNA in situ hybridization and Northern blot analysis to show that int-2 gene is expressed in a number of human carcinomas amplified at the same locus.     

Fgf10 is an oncogene activated by MMTV insertional mutagenesis in mouse mammary tumors and overexpressed in a subset of human breast carcinomas

Mouse mammary tumor virus (MMTV) infection causes a high incidence of murine mammary carcinomas by insertion of its proviral DNA in the genome of mammary epithelial cells. Retroviral insertion can activate flanking proto-oncogenes by a process called insertional mutagenesis. By sequencing the DNA adjacent to MMTV proviral insertions in mammary tumors from BALB/c mice infected with C3H-MMTV, we have found a common MMTV insertion site in the Fgf10 locus. RT-PCR studies showed that Fgf10 is expressed only in those tumors harboring a MMTV proviral insertion in this locus, suggesting that Fgf10 is a proto-oncogene. The oncogenicity of Fgf10 was evaluated in vivo by subcutaneous transplantation of retrovirally transduced HC11 mammary epithelial cells into BALB/c mice. Highly vascularized invasive subcutaneous tumors developed indicating that Fgf10 can act as an oncogene. A survey of primary human breast carcinomas revealed strongly elevated Fgf10 mRNA levels in approximately 10% of the tumors tested, suggesting that Fgf10 may also be involved in oncogenicity of a subset of human breast cancers.     

Salivary gland tumors in transgenic mice with targeted PLAG1 proto-oncogene overexpression

Pleomorphic adenoma gene 1 (PLAG1) proto-oncogene overexpression is implicated in various human neoplasias, including salivary gland pleomorphic adenomas. To further assess the oncogenic capacity of PLAG1, two independent PLAG1 transgenic mouse strains were established, PTMS1 and PTMS2, in which activation of PLAG1 overexpression is Cre mediated. Crossbreeding of PTMS1 or PTMS2 mice with MMTV-Cre transgenic mice was done to target PLAG1 overexpression to salivary and mammary glands, in the P1-Mcre/P2-Mcre offspring. With a prevalence of 100% and 6%, respectively, P1-Mcre and P2-Mcre mice developed salivary gland tumors displaying various pleomorphic adenoma features. Moreover, histopathologic analysis of salivary glands of 1-week-old P1-Mcre mice pointed at early tumoral stages in epithelial structures. Malignant characteristics in the salivary gland tumors and frequent lung metastases were found in older tumor-bearing mice. PLAG1 overexpression was shown in all tumors, including early tumoral stages. The tumors revealed an up-regulation of the expression of two distinct, imprinted gene clusters (i.e., Igf2/H19 and Dlk1/Gtl2). With a latency period of about 1 year, 8% of the P2-Mcre mice developed mammary gland tumors displaying similar histopathologic features as the salivary gland tumors. In conclusion, our results establish the strong and apparently direct in vivo tumorigenic capacity of PLAG1 and indicate that the transgenic mice constitute a valuable model for pleomorphic salivary gland tumorigenesis and potentially for other glands as well.     

Adenomyoepitheliomatous lesions of the mammary glands in transgenic mice with targeted PLAG1 overexpression

PLAG1 proto-oncogene overexpression has been causally linked to multiple tumors, highlighting its broad tumorigenic relevance. Here, the oncogenic potential of PLAG1 in mammary gland tumorigenesis was investigated in PLAG1 transgenic mice. To target mammary glands, mice of 2 independent PLAG1 transgenic strains, PTMS1 and PTMS2, in which PLAG1 expression can be modulated by Cre-mediation, were crossed with MMTV-Cre transgenic mice, resulting in P1-MCre and P2-MCre offspring, respectively. Hundred percentage of P1-MCre female mice showed mammary gland hyperplasia, caused by adenomyoepithelial adenosis, at 8 weeks. The tumorigenic process could not be studied further in P1-MCre mice, because concomitant fast-growing salivary gland tumors required euthanasia. Sixteen percentage of P2-MCre females developed mammary gland adenomyoepitheliomas within 30-45 weeks, and none displayed concomitant salivary gland tumors. To further study mammary gland tumorigenesis in PTMS1-derived mice, intercrossing with WAP-Cre transgenic mice, resulting in P1-WAPCre mice, was performed to target PLAG1 expression more specifically to mammary glands. Eighty percentage of such mice developed adenomyoepitheliomas within 53-88 weeks. All PLAG1-induced adenomyoepitheliomas revealed expression upregulation of Igf2/H19, Dlk1/Gtl2, Igfbps and Wnt signaling genes (Wnt6, Cyclin D1). Collectively, these results establish the oncogenic potential of PLAG1 in mammary glands of mice and point towards contributing roles of Igf and Wnt signaling.     

Hormones,chemicals and proviral gene expression as contributing factors during mammary carcinogenesis in C3H/StWi mice

C3H/StWi mice spontaneously lost their exogenous MMTV (MMTV-S) in 1958 and became a low mammary cancer subline. They do not ordinarily express their endogenous MMTV provirus as virions. Virigin C3H/StWi females were exposed to chemical carcinogens, 7-12 dimethyl(α)benzanthracene (2,6 mg) and urethane (200 mg) in the presence or absence of chronic hormonal stimulation of the mammary gland by pituitary isografts. By 10 months of age, mammary tumors developed in 40% of the females given DMBA, and in 59% of those given DMBA and carrying pituitary isografts. With urethane treatment alone, 14% of the mice developed mammary cancer during a 12-months period; however, 74% of the mice bore mammary tumors when pituitary isografts were present. None of the females given pituitary isografts alone developed mammary cancer during the experimental period. Mammary tumors from each group were evaluated by radioimmune competition assay, immunoperoxidase and electron microscopy to determine the extent of endogenous MMTV gene expression. In addition, the mammary glands of some of the non-tumor-bearing mice were studied by whole mount and by histology to determine the type, extent, and number of mammary dysplasias present in each group. In general, there was no correlation between tumor incidence and the presence of mammary tumor virus antigens. Of 32 mammary tumors tested in all groups, 10 were positive at low levels for MMTV antigens. In the positive tumors, the internal MMTV gag gene antigen, p28, was prevalent. Immunoperoxidase studies on these same tumors gave quantitatively similar results. It is apparent from these observations that chemical carcinogenesis of mouse mammary glands does not require or even favor the complete expression of endogenous MMTV genes. In this study, hormonal stimulation of the glands during and after exposure to the carcinogen increases the rate of tumorigenesis, but does not seem to favor MMTV gene expression. We conclude that C3H/StWi mice are susceptible to DMBA and urethane induction of mammary cancer and to an extent this process is positively influenced by the presence of hormonal stimulation. Our results support the view that a qualitative or quantitative shift in the expression of MMTV proviral sequences does not occur during tumor development.     

WNT and FGF gene clusters (review)

Mouse mammary tumor virus (MMTV) is a retrovirus, activating Wnt genes (Wnt1/int-1, Wnt3/int-4, Wnt10b), Fgf genes (Fgf3/int-2, Fgf4, Fgf8) and other genes (Notch4/int-3, Eif3s6/int-6) due to proviral integration. Among 19 WNT genes, WNT3 and WNT14B genes are clustered in human chromosome 17q21, WNT3A and WNT14 in human chromosome 1q42, WNT10A and WNT6 in human chromosome 2q35, and WNT10B and WNT1 in human chromosome 12q13. Among 22 FGF genes, FGF19, FGF4 and FGF3 genes are clustered in human chromosome 11q13, while FGF23 and FGF6 in human chromosome 12p13. WNT and FGF gene clusters are conserved between the human genome and the mouse genome. Activation of mouse Wnt or Fgf genes due to proviral integration of MMTV occurs in 5 out of 13 clustered genes, and in 1 out of 28 solitary genes (p=0.0033), which clearly indicates that mouse Wnt or Fgf gene clusters are recombination hot spots associated with carcinogenesis. Recombination results in retroviral integration as well as in chromosomal translocation, gene amplification and deletion during carcinogenesis. The CCND1-FGF19-FGF4-FGF3 gene cluster in human chromosome 11q13 is amplified in breast cancer, squamous cell carcinoma of head and neck, and bladder tumors, and is also translocated in parathyroid tumors and B-cell lymphoma. WNT gene clusters on human chromosome 1q42, 2q35, 12q13, and 17q21 as well as FGF gene cluster on human chromosome 12p13 might be amplified or translocated in human cancer just like FGF gene cluster on human chromosome 11q13.     

Differentiation-dependent expression of provirus-activated int-1 oncogene in clonal cell lines derived from a mouse mammary tumor

The int-1 mammary oncogene is frequently activated by proviral insertion in mouse mammary tumors. To characterize the target cell for the oncogenic action of int-1, we have isolated permanent cell lines with distinct morphologies and differentiation characteristics, starting from a tumor with a rearranged int-1 gene. Polygonal cells had retained many differentiation markers of epithelial cells and produced adenocarcinomas upon transplantation in syngenic mice. Sphere-forming-cuboidal cells are poorly differentiated and produced anaplastic tumors. Cuboidal and elongated cells were negative for epithelial markers. Cuboidal cells were poorly tumorigenic, but elongated cells produced highly malignant sarcoma-like tumors. In all lines, the int-1 gene was identically rearranged due to insertion of proviral DNA of the Mouse Mammary Tumor Virus, but the expression of int-1 varied with the state of differentiation of the cells. Polygonal cells contained relatively high levels of int-1 RNA, which were not influenced by steroid hormones. In the sphere-forming-cuboidal cells, expression of int-1 was low but inducible by dexamethasone. In the cuboidal and elongated cells no expression of int-1 was detectable, showing that the continued expression of int-1 was not required for progression to more malignant cells. By immunoprecipitation, two int-1 protein species, of 42 and 40 kD were identified in polygonal and in sphere-forming-cells but not in the culture media.     

Acquired Expression of hst-1 in an Autonomous Subline (Chiba Subline 2) Derived from Androgen-responsive Mouse Mammary Tumor (Shionogi Carcinoma 115)

Since growth of Shionogi Carcinoma 115 (SC 115) and its autonomous subline (CS 2) were regulated by fibroblast growth factor-like peptide, expression of int-2 and hst-1 was examined in these cell lines. Hybridization of genomic DNA with long terminal repeat of mouse mammary tumor virus (MMTV) revealed the same pattern of restriction fragments, showing the same integration of MMTV. Although weak expression of int-2 was noticed in the two cells, clear expression of hst-1 was seen only in CS 2 cultured with/without testosterone. It is suggested that autonomous growth of androgen-unresponsive CS 2 is connected with expression of hst-1 .     

Acquired expression of hst-1 in an autonomous subline (Chiba subline 2) derived from androgen-responsive mouse mammary tumor (Shionogi carcinoma 115)

Since growth of Shionogi Carcinoma 115 (SC 115) and its autonomous subline (CS 2) were regulated by fibroblast growth factor-like peptide, expression of int-2 and hst-1 was examined in these cell lines. Hybridization of genomic DNA with long terminal repeat of mouse mammary tumor virus (MMTV) revealed the same pattern of restriction fragments, showing the same integration of MMTV. Although weak expression of int-2 was noticed in the two cells, clear expression of hst-1 was seen only in CS 2 cultured with/without testosterone. It is suggested that autonomous growth of androgen-unresponsive CS 2 is connected with expression of hst-1.     

A biochemical approach to the study of the transmission of mouse mammary tumor viruses in mouse strains RIII and C3H.

Mouse mammary tumor virus (MMTV) proviral sequences were detected in the cellular DNA of mammary tumors and livers of RIII and C3H mice by molecular hybridization with radioactively labelled MMTV 60-70S RNA or tritiated MMTV complementary DNA (cDNA). By means of DNA:DNA reassociation kinetics, the DNA of the mammary tumor cells of these two mouse strains were found to contain more MMTV proviral sequences than the DNA of liver cells of these same tumor-bearing mice. Evidence is also presented that the DNA of the liver cells lacks a part (approximately 25%) of the MMTV proviral sequences found in the mammary tumor cells of these mouse strains. The relationship of the extra MMTV proviral sequences found in mammary tumor cells to the early mammary tumor-igenesis seen in these mouse strains is discussed.     

A congenic line of the DDD mouse strain, DDD/1-Mtv-2/Mtv-2: establishment and mammary tumorigenesis

A single dominant gene on chromosome 18, Mtv-2, controls both the early appearance of mammary tumors and expression of mouse mammary tumor virus (MMTV) in the milk. A congenic DDD mouse strain, DDD/1-Mtv-2/Mtv-2 (DDD-Mtv-2), was developed by introducing this gene from GRS/AJms (GR) into DDD/1 mice by repeating 12 backcrosses and subsequent inbreeding using mammary tumors as a marker for selection. Southern blot analysis of the liver DNA from the resulting congenic mice with EcoRI and MMTV-U3 prove revealed that two DNA fragments corresponding to Mtv-2 were specifically transferred from GR to congenic mice. Detection of MMTV-gp52 antigen in the mammary gland and mammary tumor development in DDDfDDD-Mtv-2 mice demonstrated the production of infectious mature MMTV by Mtv-2 in congenic mice. About 80% of breeding DDD-Mtv-2 females developed mammary tumors in the course of one-year follow-up. The tumor incidence was lower and the tumor age higher than those in GR mice, suggesting less active functioning of the gene on the DDD genetic background. About 70% of these tumors were morphologically classified as pale cell and type P carcinomas peculiar to GR mice. The gene seemed to control the histologic features of mammary tumors. Congenic mice carried an MMTV provirus in an incomplete form on Y chromosome. The DDD-Mtv-2-strain will provide a new model for biological and molecular researches into mouse mammary tumorigenesis.     

Tissue Distribution of Mouse Mammary Tumor Virus (MMTV) Antigens and Endogenous MMTV Loci in Japanese Laboratory Mouse Strains

The distribution of mouse mammary tumor virus (MMTV) antigens was studied by the immunoper-oxidase method in the II-TES and I-TES mouse strains as well as their progenitors, CS and DBA/2 strains. In the II-TES, I-TES and CS strains, and BALB/c mice foster-nursed with these strains, MMTV antigens were found not only in epithelial cells of the mammary glands but also in those of other tissues including the seminal vesicle, vas deferens, epididymis, prostate, parotid, submandibular, lacrimal, sebaceous, and urethral glands. In DBA/2 and BALB/cfDBA/2 mice, however, the MMTV antigens were found only in the mammary glands. Electron microscopic examination showed MMTV particles in these organs. When we examined the presence of Mtv- 1 and 2 proviruses, which are known to be responsible for MMTV expression, in the genomes of the II-TES, I-TES, CS and DBA/2 strains by Southern blotting, Mtv- 2 was not found in any of the mice and Mtv- 1 was found in the II-TES and DBA/2 mice but not in the I-TES and CS mice. Instead, four new endogenous MMTV loci, which have never previously been reported in laboratory mouse strains, were detected in the genomes of the II-TES, I-TES and CS strains. One (designated Mtv -42) was common in the three strains and the other three (designated Mtv- 43, 44 and 45) were common in the II-TES and I-TES strains or the II-TES and CS strains. These results thus suggest that new endogenous MMTV loci may be responsible for MMTV expression in a variety of tissues of these three strains.