GUJHS. 2007 Sept; Vol. 4, No. 2
Chelsea Curtis, Michael DeLuca,
Kaitlin Mullen, Amy Smietana,
Samantha Tsang, Allan Angerio, PhD
MMTV, mouse mammary tumor virus, is a predominantly endogenous retrovirus whose long terminal repeat (LTR) encodes certain carcinogenic proteins and superantigens. Through reverse transcription, the retrovirus activates several oncogenes, and uses lymphocytes to infect mammary glands. Without these cells, the virus cannot be transmitted to the mammary glands. Aside from these lymphocytes, there are also various oncogenes and hormones that play a major role in the virus’s life cycle. MMTV uses the immune system of the mouse to establish a systemic infection and infiltrate the mammary tissue and insert itself into the cell as a provirus. There, through its insertion into the mammary epithelial stem cells, MMTV stimulates proto-oncogenes in the cell, causing tumorgenesis. MMTV, both an endogenous and exogenous virus, is acquired in mice through breast milk or the germ line. There is evidence of an MMTV-like virus in human breast cancer tumors, and the transmission of the virus is currently unknown. A theory exists that MMTV can be transmitted to humans, giving rise to an altered form of the virus. MMTV is one of many viruses that has been linked to cancer. HPV, which is now known as a causative agent for cervical cancer, can be used as a comparison to MMTV, although there are significant differences between HPV and MMTV.
Keywords: MMTV; mammary gland; retrovirus; SAg; HPV; Peyer’s Patches; gag; HERV; APCs; Wnt, Fgf, Notch; retroviral transcription
MMTV, the mouse mammary tumor virus, became publicized in the 1980s as a carcinogenic agent. This virus “is an endogenous murine retrovirus1,” with a long terminal repeat (LTR) that has been linked to the expression of MMTV in the form of mammary, as well as other organ, tumors in mice. The MMTV LTR found in transgenic mice, and a few wild-type mice, has been shown to be correlated with the development of mammary carcinomas1. This retrovirus, while remaining endogenous, can also be spread exogenously between the mouse species. A mouse can become infected with MMTV from an endogenous source, through chromosomal mouse DNA, or an exogenous source, such as milk contaminated by the infected mammary glands of the mother mouse. In 1987, Choi, Henrard, Lee, and Ross confirmed this statement, and mated large tumor antigen (LTag3 and LTag5) mice with normal ICR males. The offspring produced were 50 percent transgenic as well. These findings further indicate that MMTV is endogenously transmitted through a single chromosome integration2.
MMTV’s Long Terminal Repeat
To better understand MMTV expression, and subsequent mammary tumorgenesis, one must first examine it’s LTR. MMTV’s unique LTR closely mirrors that of the herpes simplex virus, namely its type 1 thymidine kinase gene (LTR-tk). It is also strikingly similar to the early tumor antigen (Tag) region of the human simian virus 40 (SV40)1. The MMTV retrovirus’s relationship to SV40 is one of regulation. Through various experiments with transgenic mice, researchers concluded that MMTV LTR functions as a transcriptional controller of SV40. Direct regulators are encoded on the LTR, namely the promoter and steroidal hormone glucocorticoid. However, though that region of the MMTV genome is a SV40 regulator, both viruses have tumor-promoting elements2. In an experiment conducted by Choi, Henrard, Lee, and Ross, a plasmid was produced that, under the regulation of MMTV’s LTR, the SV40’s early Tag region was produced and introduced into a set of transgenic mice2. During the experiment, the MMTV LTR supported the researchers’ hypothesis: “Transcription of the SV40 early region initiat[es] within the LTR2.” Therefore, MMTV’s LTR is a proven regulator of another virus that promotes tumor growth in various tissues of the body. The exact tissues affected by MMTV are part of the reason why researchers are interested in achieving a better understanding of the virus.
MMTV and Tumor Production
MMTV has become of particular interest to clinicians, though its exact effecter cells and life cycle are still under considerable scrutiny, due to its effect on the host’s immune system and its tumorgenic qualities2. B cells, the antibody-producers of the immune system, are the primary targets of MMTV. In B cells, as well as other vulnerable cells in infected organisms, significant amounts of viral MMTV RNA appear, amplify, and often cascade into tumor production2. While B cells are a target cell of MMTV, and could explain its dramatic effect on the immune system, the organs targeted by the virus attract the most scrutiny.
MMTV is linked to carcinogenic expressions in many organs of the body – but its most prominent and well studied effector-organ is the mammary gland. The virus’s tendency to be transcribed and expressed within breast tissue has lead virologists to question its high specificity.2 As of now, the developmental pathway of MMTV is attributed to reverse RNA transcription of the genome. The MMTV virus, containing RNA and not DNA, is introduced into a host organism. Then, by reverse RNA transcription, the viral RNA is transcribed into DNA and inserted into the host’s DNA. More often than not, the infectious DNA is inserted at a non-critical region of the host DNA, and may never be expressed. However, the mutated genome can sometimes carry that viral DNA in a critical region. The altered genetic sequence triggers a series of internal and external reactions that go on to cause either MMTV-linked tumor expression or non-expression, via activation of oncogenes2.
If the viral DNA does lead to mutated expression, several organs of the host’s body are affected. In fact, the expression of the MMTV transgene is not limited to lactating mammary glands, though that remains its primary target. In fact, the transgene appears chronologically in lactating mammary glands, lungs, kidneys, prostate, salivary glands, and lymphoid cells, but infects cells within the same organ simultaneously2. However, when MMTV is spread endogenously from the mother mouse to her offspring, its sequences only appear in the mammary tissue2. This finding also suggests that endogenous MMTV is somewhat tissue specific as well as its exogenous counterpart. At the same time, MMTV, when transmitted endogenously through the germ line, can become active and able to be transcribed in many different organs and tissues of the body2. The LTR, in this respect, is capable of affecting these cells due to a sequence encoded in the LTR that helps determine the tissue-specific expression2. However, other factors must be present to show the tissue specificity that leads to mammary carcinomas. All that is known so far with respect to MMTV’s tissue-specificity is that its LTR perhaps increases the extent of the virus’s expression. Research is still being conducted to determine if there are other areas of the viral genome that influence its replication and expression in other organs2.
Within lactating mammary glands, research has concluded that one sequence of proviral genes plays a key part in the development of mouse adenocarcinomas. Those sequences originate in “MMTV-c-myc hybrid-containing transgenic mice2.” The reasoning behind MMTV’s tissue-specific expression in lactating mammary glands may be the cause of, what Mok et al believe to be, “the production of virus particles that occurs during lactation3.” Without those virus particles, MMTV is not believed to have a significant enough amount of infected cells to later result in a tumor. In the same respect, the oncogenes most affected by MMTV infection are rarely found in mammary tissues. Therefore, the activation of the expression of such genes must be the result of direct MMTV stimulation, and not from the activation of the oncogenes3. These findings further suggest the direct correlation between MMTV and mammary tumorgenesis. They point to the viral particles themselves as the primary source, and not the oncogenes or other cofactors.
Transmission of MMTV
Mouse Mammary Tumor Virus (MMTV) is the leading cause of breast cancer in mice. 4 MMTV is both an endogenous and exogenous virus. Endogenously, mice inherit the virus during embryonic development. Exogenously, mice acquire the virus from the mother’s milk during breast feeding.
MMTV has been studied for many years in order to understand carcinogenic viruses and to determine any correlation with human breast cancer. With the emergence of DNA analysis in the 1990’s there has been evidence that MMTV may be transmitted to humans and shows to have an homologous strain apparent in human breast cancer tumors. While there is evidence supporting MMTV may be a cause of human breast cancer, there are also data that contradict this theory and it continues to be an issue of great controversy.
There is evidence that a human MMTV-like virus exists in humans that is 95% homologous to MMTV at the env gene sequence4. MMTV-like virus is distinguishable from Human Endogenous Retroviruses (HERV), which are viruses found in normal human DNA and accounts for 1% of an individual’s DNA.6 Studies have shown that MMTV-like env gene sequences were detected in 38% of breast cancer patients in the Wang study conducted in the United States, but an insignificant number in normal breast tissue (Figure 1). 8 This is consistent with the conclusion that MMTV-like virus cannot be categorized as a HERV because MMTV-like virus is not found in normal DNA. This also led scientists to believe that the MMTV-like virus in humans is acquired.
It is known that human breast feeding cannot transmit the MMTV-like virus to a child and that pregnancy actually reduces the risk of breast cancer.5 What may be a likely cause of MMTV transfer to humans giving rise to MMTV-like virus could be mouse to human contact. However, this is a highly controversial theory in which studies have had mixed results. There have been studies that show MMTV can infect human cells. One study done in 2005 shows successful infection of many different human cells by MMTV and could be an explanation for why MMTV-like env sequences are found in human breast cancer5. Contrary to these results, another study showed that only an insignificant amount of MMTV could adapt to human TfR1 receptors for transmission6. Another study indicated that MMTV is present in rodents, felines, and primates. Researchers hypothesize that cats may become infected with MMTV by mice, and cats could then infect humans7. In support of the theory that mice may infect humans with MMTV was a study that correlated the population of Mus domesticus, the common house mouse, in certain geographic areas versus the incidence of breast cancer. It was found that the species of mouse, Mus domesticus, is more susceptible to MMTV than other species of mice. In areas where Mus domesticus was most prevalent, there was an overall higher human breast cancer incidence rate8.
It is inconclusive if MMTV from mice or other animals can infect human beings and cause mammary gland tumors, but there is strong evidence of the presence of MMTV-like virus in human breast cancers.
As explained previously, mouse mammary tumor virus (MMTV) occurs in an endogenous and exogenous form9. The endogenous form, known as Mtv, is a provirus incorporated into the host DNA9. This form is generally not associated with secretion of viral particles9. Exogenous MMTV can be spread through infectious viral particles in secreted milk to mice offspring9. The uptake of the exogenous virus involves the gastrointestinal tract of the mouse upon ingestion of the milk9. The Peyer’s patches of the gastrointestinal tract contain B-cells and dendritic cells that act as antigen-presenting cells (APCs)9,10. MMTV initially infects these APCs, which then display the MMTV superantigen (SAg) with their major histocompatibility complex class II (MHCII). This expression of SAg is achieved through the reverse transcription of the MMTV RNA and its subsequent inclusion into the DNA of the host cell10. The Vβ chain of a receptor on CD4+ T-cells identifies the antigen being presented on the APC MHCII9. This recognition initiates an increase in the number of T-cells that are complimentary or cognates to the SAg on MMTV, SAg-cognate CD4+ T-cells9. This increase in SAg-cognate CD4+ cells induces a subsequent increase in the B-cell numbers9. (Refer to Figure 2 for a summary of this initial infection pathway).
These lymphocytes are responsible for the movement of the virus ultimately to the mammary gland9. The virus waits until mammary cells are actively dividing during puberty to infect the gland through their lymphocyte vehicles11. The MMTV SAg appears to be essential for infection of lymphocytes and therefore for infection of the mammary tissue, leading to the development of tumors9. There is some data, however, indicating the existence of a SAg ‘independent’ phase for initial lymphocyte infection12. B-cells and SAg-cognate T-cells are thought to be necessary for MMTV infection of the mammary gland though11. The mobility of the lymphocytes may facilitate infection of the mammary cells11. After initial infection and rapid increase in T-cells, the SAg-cognate T cells are deleted13. MMTV also down-regulates the immune response and facilitates infection of the mouse through an interaction with Toll-like receptor 4 (TLR4), which is associated with release of interleukin 10 (IL-10) by B-cells14.
Mammary Tissue Infection and Tumor Development Regulation
MMTV infection has been detected in other parts of the mouse besides the mammary glands: for instance, the lung and testes13. Replication apparently only occurs in the mammary glands, however13. Pluripotent stem cells in the epithelium of mammary glands are believed to be responsible for the infection of MMTV leading to tumor development13. Populations of cells in a single abnormal growth have similar insertions of MMTV into the host DNA and originate from a single cell15. A single stem cell on the epithelium, containing the inserted virus can produce pre-malignant and malignant populations after undergoing mutations15. These cells have lost their senescence and become ‘immortal15.’ The hyperplasia that develops is a result of the infected stem cell15.
Several oncogenes are upregulated by MMTV introduction into the host cells. The genes Wnt1(wingless family), Fgf3(fibroblast growth factor family), and Notch4 (Notch family) have been implicated in frequent upregulation in mammary tumors of infected mice9. Collaboration of these genes is important for tumor development13. For example, based on the locations of the insertion of MMTV in the host cell, the Wnt and Fgf gene groups likely work together to produce abnormal growth13. Although the activation of host oncogenes after viral insertion is thought to be the process by which MMTV causes tumor formation, there are other factors involved9. The MMTV encoded gag gene has been found to play a significant role in the development of mammary tumors9. It is believed that the gag gene works in conjunction with host proto-oncogenes9. Furthermore a gene on chromosome 13 of mice called mammary tumor susceptibility, mts, has been implicated in a relationship with the gag gene9. In addition to these factors, the mouse endocrine system induces MMTV replication through progesterone and glucocorticoids11. Since these hormones are secreted in greater quantity with pregnancy, virgin mice are less likely to develop mammary tumors11.
Understanding the cellular events of MMTV induced tumors in mice is important for gaining insight into human breast cancer16. For example, resistance to systemic MMTV infection in mice was achieved through the use of a rat antibody gp52, an envelop protein17. In mice with the antibody present, lymphocytes were still infected at first; however, SAg dependent proliferation did not occur and there was no mammary infection17. A similar antibody approach may be explored for human mammary tumors of viral origin.
Steroidal Control of MMTV
Expression of MMTV involves cofactors active at the molecular level. Two such cofactors are glucocorticoids and progesterone. As previously stated, the LTR is an integral region for MMTV transcription and expression. Within that LTR resides a hormone response element (HRE)18.
Once the LTR contacts certain steroidal hormones, such as glucocorticoids and progesterone, the viral chromatin undergoes chemical and mechanical changes. It has been demonstrated that hormone receptors are the primary targets of this change. These changes in its structure allow DNase to break down the chromatim with greater ease during transcription, and help the DNase reach the HRE within the LTR18.
At the same time, the chromatin alterations also stimulate transcription factors to bind to the viral genome. For instance, nuclear factor I (NFI) is normally unable to bind to the HRE. However, the presence of steroidal hormones makes the connection possible. When these cofactors, steroidal hormones and NFI, are present MMTV becomes transcriptionally active. HRE-NFI binding combined with MMTV promoter specifically triggers MMTV transcription in the host’s cells18.
Progesterone, in the transcriptional process, also acts as an important cofactor. Once the HRE is exposed to DNase and the hormone receptor sites are within contact, progesterone becomes active in the process. The hormone instigates the binding of a different transcriptional factor, OTF-1, to a specific region within the LTR. However, rather than progesterone regulating the joining of OTF-1 to the LTR, it occurs in reverse. The transcription factor regulates the activity of progesterone in the presence of the MMTV promoter, and thus also regulates the MMTV’s ability to bind DNA to progesterone receptors. This, in turn, affects the entire transcription process. Without OTF-1 and progesterone as cofactors, MMTV would be rendered incapable of replicating and causing tumorgenesis18.
HPV Relevance to MMTV
Like breast cancer, cervical cancer has also been associated with a virus—the human papillomavirus—and both have been found to be two of the most common cancers among women. The first research that linked HPV with head and neck cancer was published in 2001— even though it had been suspected, since 1983, that HPV type 16 had an association with cervical cancer16. Today, scientists have found that, there are over seventy types of the human papillomavirus17. The four most prominent types of HPV, which have been viewed as causative agents for cervical cancer, are HPV type 16, HPV type 18, HPV type 31, and HPV type 3316.
There are three stages to HPV-linked cervical cancer: cervical dysplasia carcinoma in situ, and invasive cancer of the cervix. Cervical dysplasia—the first stage where there are either mild, moderate, or severe amounts of abnormal cells— has been found to frequently be regressive, meaning that many people who have cervical dysplasia may not necessarily reach the stage of invasive cancer. Carcinoma in situ (CIS) is pre-invasive cancer, meaning that the cancer has only engaged the surface cells of the neck and head. Ninety percent of patients who have experienced the last stage of cervical cancer, invasive cancer of the cervix, have been found to have some strain of HPV19.
Not only does HPV lead to cervical cancer, it has also been linked to breast cancer; although researchers can not provide sufficient evidence supporting this relationship. Unlike HPV, MMTV is believed to be found primarily in mice. Researchers continue to study MMTV because they speculate that there is a human homologue of MMTV, as mentioned earlier. Although the majority of those studying the virus believe that MMTV is solely found in mice, there are some who argue that MMTV is also found in humans. What these researchers fail to demonstrate, however, is how MMTV in mice has crossed the xenographic barrier into humans17.
MMTV, unlike HPV, lacks an oncogene. Instead, when positioned close to cellular protooncogenes, the inserted virus magnifies the protooncogenes’ activity and causes irregularity in the cell cycle. As is evident, HPV and MMTV are similar in that they are both retroviruses, although they behave somewhat differently—MMTV is a beta-retrovirus and HPV is not19. HPV and MMTV also share a similarity in that they both have various, distinct strains, although the prominent breast cancer-linked strains for MMTV are still being speculated upon.
Although HPV and MMTV can be compared, there are many significant differences between the two viruses. Thus, it is improbable that research done on HPV can directly help answer questions about MMTV. Research regarding the tumor growth process through HPV though, may yield invaluable insight into the function of MMTV in mammary tumor growth. Although there is now merely speculation as to whether MMTV is truly linked to human breast cancer, the virus may eventually follow the path of the human papillomavirus: in the near future, it may be possible to publish research demonstrating that MMTV, or a human version of MMTV, is indeed a prominent causative agent of breast cancer.
Figure 1. The graph shows six different studies that were conducted to test the presence of MMTV-like virus in human breast cancer tissue. The cases in red indicate human breast cancer tissue that did not contain the MMTV-like virus and the cases in blue indicate human breast cancer with MMTV-like virus. The percentages signify the number of MMTV-like virus cases found within the whole group tested.
- Henrard D, Ross, SR. Endogenous Mouse Mammary Tumor Virus Is Expressed in Several Organs in Addition to the Lactating Mammary Gland. Journal of Virology. 1988; 62:3046-3049.
- Choi, Y, Henrard, D, Lee, I, Ross, SR. The Mouse Mammary Tumor Virus Long Terminal Repeat Directs Expression in Epithelial and Lympoid Cells of Different Tissues in Transgenic Mice. Journal of Virology. 1987; 61:3013-3019.
- Mok, E, Golovkina, TV, Ross, SR. A Mouse Mammary Tumor Virus Mammary Gland Enhancer Confers Tissue-Specific but not Lactation-Dependent Expression in Transgenic Mice. Journal of Virology. 1992; 66:7529-7532.
- Pogo, B.G., Holland J.F. Possibilities of a Viral Etiology for Human Breast Cancer. Biology Trace Elem Res. 1997 Jan; 56(1):131-142.
- Indik, S., Gunzburg W.H. Mouse Mammary Tumor Virus Infects Human Cells American Association for Cancer Research. 2005 August; 65: 6651-6659.
- Zhang, Y., Rassa J.C. Identification of the Receptor Binding Domain of the Mouse Mammary Tumor Virus Envelope Protein. Journal of Virology. 2003 October; 75:10468-10478.
- Szabo, S., Haislip A.M., Garry R.F. Of Mice, Cats, and Men: is Human Breast Cancer a zoonosis? Microsc. Res. Tech. 2005 November; 68: 197-208.
- Stewart, T.H.M., Sage R.D. Breast Cancer Incidence Highest In The Range of One Species of House Mouse, Mus domesticus. British Journal of Cancer. 2000; 82: 446-451
- Swanson I, Jude BA, Zhang AR, Pucker A, Smith ZE, Golovkina TV. Sequences within the gag gene of mouse mammary tumor virus needed for mammary gland cell transformation. J Virol 2006;80(7):3215-24.
- Mpandi M, Otten LA, Lavanchy C, Acha-Orbea H, Finke D. Passive immunization with neutralizing antibodies interrupts the mouse mammary tumor virus life cycle.J Virol 2003;77(17):9369-77.
- Golovkina TV, Dudley JP, Ross SR. B and T cells are required for mouse mammary tumor virus spread within the mammary gland. J Immunol 1998;161(5):2375-82.
- Pobezinskaya Y, Chervonsky AV, Golovkina TV. Initial stages of mammary tumor virus infection are superantigen independent. J Immunol 2004;172(9):5582-7.
- Callahan R, Smith GH. MMTV-induced mammary tumorigenesis: gene discovery, progression to malignancy and cellular pathways. Oncogene 2000;19(8):992-1001.
- Jude BA, Pobezinskaya Y, Bishop J, Parke S, Medzhitov RM, Chervonsky AV, et al. Subversion of the innate immune system by a retrovirus. Nat Immunol2003;4(6):573-8.
- Smith GH. Stem cells and mammary cancer in mice. Stem Cell Rev 2005;1(3):215-23.
- Brower V, Accidental Passengers of Perpetrators? Current Virus—Cancer Research. Journal of the National Cancer Institute. 2004; 96: 257-258.
- Britannica Encyclopedia. The HPV page. Available at: http://britannica.com/eb/article-224774/cancer Accessed February 17, 2007.
- Beato, M, Chalepakis, G, Truss, M. Interplay of steroid hormone receptors and transcription factors on the mouse mammary tumor virus promoter. Journal of Steroid Biochemistry and Molecular Biology, 1992; 43(5):365-78.
- Goedert JJ, Rabkin CS, Ross SR. Prevalence of serologic reactivity against four strains of mouse mammary tumor virus among US with breast cancer. British Journal of Cancer. 2006; 94: 548-551