Background Bovine leukemia virus (BLV) infection is widespread in cattle globally and is present in marketed beef and dairy products. Human infection with BLV has been reported in breast and lung cancer tissues and was significantly associated with breast cancer in 3 case-control studies. The purpose of this current research was to determine if BLV is present in human blood cells and if antibodies to BLV are related to blood cell infection. Methods Standard liquid PCR and Sanger DNA sequencing were used to test for BLV in buffy coat cells (leukocytes and platelets) of blood specimens from 95 self-selected female subjects. Enzyme-linked immunosorbent assay (ELISA) for IgG, IgM, and IgA was used to detect antibodies to BLV in the plasma of the corresponding blood samples. Results BLV DNA was detected in the buffy coat cells of blood in 33/95 (38%) of the subjects by PCR and DNA sequencing. IgG antibodies were detected in 30/95(32%), IgM in 55/95(58%), and IgA in 30/95(32%) of the subjects. There was no significant correlation between presence of the antibodies and presence of BLV DNA. Conclusions This first report of BLV in human blood raises the question of whether infection of leukocytes could conceivably lead to leukemia as it does in infected cattle. Also, system wide circulation of infected blood cells could facilitate BLV transit to various internal tissues/organs with potential for their infection and subsequent development of cancer. The most likely route of BLV transmission to humans would be zoonotic, as a foodborne infection. Although eradicated from cattle in some countries, BLV still has a high rate of infection in the Americas, the Middle East, and parts of Europe and Asia. This report of BLV in the blood layer containing human leukocytes/platelets adds important information which could be useful to elucidate possible routes of transmission of BLV to humans and to prevent further human infection.

Purpose: To compare recent data sets on frequency of bovine leukemia virus (BLV) detection in breast tissue in different human populations, and association of BLV with a confirmed diagnosis of breast cancer and expression of common clinical biomarkers. Many risk factors are associated with breast cancer, but what causes initial molecular/cellular changes from normal to malignant is not well understood. Six types of human cancer are caused by viruses, and several viruses have been studied for a role in breast cancer. BLV is a common virus of US cattle (38% of beef herds, 89% of dairy herds, 90-100% of large dairy operations). BLV DNA and protein were previously identified in human tissues. Methods: Breast tissue specimens were archived formalin-fixed, paraffin-embedded tissue sections from sources mentioned below. In situ PCR (IS-PCR) was performed on intact deparaffinized sections on glass slides, using primers from the tax region (transforming gene), highly conserved and rarely deleted, as are most BLV genome regions as tumor progression occurs. BLAST sequence comparisons indicated primers had high homology only with BLV, and extremely low homology with human genomic sequences, including human endogenous retroviral sequences. Previous laboratory experiments confirmed this specificity and established appropriate controls for the assay. Sections were semiquantitatively scored for density of the digoxygenin-labeled IS-PCR product within mammary epithelial cells, upon microscopic examination by two independent examiners. Results: Previous data sets using samples from the Cooperative Human Tissue Network, USA (n=239) and New South Wales, Australia (n=96), showed BLV presence significantly more frequently in malignant tissue than in normal breast tissue from women with no breast cancer history. Odds ratios (OR) and confidence intervals (CI) were 3.07(1.66-5.69), p≤.001 and 4.72(1.71-13.05), p≤.003, respectively. New unpublished data indicate specimens from MD Anderson Cancer Center in Houston, TX (n=216) show an even higher OR (CI) of 5.59 (2.76-11.30), p≤.0001. Argentinian specimens showed lower frequency (22%) in malignant tissues than US (58%) and Australian specimens (80%). Cell proliferation markers, KI67 and HER2 overexpression were significantly associated with BLV presence in Argentinian malignant breast tissues. For 31 Australian subjects with breast cancer, archived normal breast tissues were available from surgery 3-10 years previous for an unrelated condition. For 23 (74%) of these, BLV was already present in normal tissue at least 3 years before cancer diagnosis, consistent with a causative temporal relationship between BLV presence and subsequent cancer development. Conclusions: New unpublished data indicate BLV in some Argentinian women, but the frequency in breast cancer tissue is lower than in US and Australian women. Tests for conventional prognosis/proliferation biomarkers (ER, PR, HER2, Ki67) in Argentinian specimens showed BLV presence significantly associated with moderate-high Ki67 level (p≤0.009) and HER2 overexpression (p≤0.0442), consistent with possible BLV role in enhanced cell proliferation. BLV’s mode of transformation is via the oncogenic protein (Tax), which inhibits host cell DNA repair via the base excision pathway. DNA damage observed in breast cancer cells, including driver mutations, could theoretically have been initiated due to DNA repair inhibition by BLV infection. Elucidating initiators for any cancer opens opportunities for primary and secondary prevention. For viruses, the reservoir (e.g., BLV-infected cattle) could be reduced, transmission to humans could be intercepted (e.g., education to avoid raw milk, raw beef appetizers), vaccines could be developed, and elimination of the virus might be achieved by developing BLV-targeting agents. Driver mutations already initiated by BLV might be targeted as part of personalized secondary prevention and/or therapy. Note: This abstract was not presented at the conference. Citation Format: Gertrude C. Buehring, HuaMIn Shen, Kimberly Baltzell, Jennette Sison, Savitri Krishnamurty, Pamela Lendez, Lucia Matinez-Cuesta, MariaVictoria Nieto-Farias, GuillerminaLaura Dolcini, MariaCarolina Ceriani. Bovine leukemia virus in breast tissue linked to increased cell proliferation and breast cancer risk [abstract]. In: Proceedings of the AACR Special Conference: Advances in Breast Cancer Research; 2017 Oct 7-10; Hollywood, CA. Philadelphia (PA): AACR; Mol Cancer Res 2018;16(8_Suppl):Abstract nr B30.

Bovine leukemia virus (BLV), is an oncogenic virus that infects cattle worldwide and is the causative agent of leukemias, lymphomas and persistent lymphocytosis. BLV has also been found in humans and has recently been proposed as a risk factor for developing breast cancer in the USA and Australia. In Colombia, there is evidence of infection in women but no correlation with breast cancer. This study was aimed at comparing the presence of the virus in breast tissue from different sources: necropsies of women without tumor development (normal breast tissue), and surgeries of benign and malignant tumors, to better understand the role of BLV in Colombia. A cross-sectional study was designed in which 315 participants were included. Paired samples of breast tissue and blood were obtained from surgeries and necropsies in Bogotá city. The presence of BLV was determined by nested PCR and in situPCR targeting different viral genes. For the nested PCR, DNA was extracted from fresh tissues and blood samples, and human GAPDH amplification was carried out to assess DNA quality for the PCRs. Afterwards, different BLV genes were detected by nested PCR. In situ PCR was directed to the tax region of the virus and was done to FFPE sections of the same samples. Positive samples were considered when at least one of the techniques was positive for the virus and were confirmed by Sanger sequencing. Correlation with other risk factors related with breast cancer (HER2, PR, ER and KI67) and the presence of the virus were determined. From the overall population, 40% of the samples were positive for BLV. In the breast cancer population, 37% of the samples were infected with the virus; similar distributions were found in the other groups (pre-malignant, 33%; benign tumors, 27%). Interestingly, almost 60% of the samples were infected in the no-tumor group. 22% of the samples were positive for both breast tissue and blood from the same patients. All the positive samples were confirmed to have BLV by Sanger sequencing of different gene regions of the virus. When correlating viral presence with other risk factors of breast cancer, it was found that most of the positive BLV samples were also positive for progesterone (79%) and estrogen (93%) receptors and were negative for the HER2 mutation. It could thus be possible that the hormonal profile could be linked with the presence of the virus. This study shows that irrespective of the breast pathology, BLV can be found in Colombian women both in breast tissue and blood. Sanger sequencing showed that the virus found in humans is similar to previously reported sequences obtained from cattle with high identity percentages. Even when the biology of the virus remains unknown in humans, it is a big concern to find the presence of an oncogenic virus coming from animals. Further studies are needed to understand the viral mechanisms related with cancer in humans. Citation Format: Nury N. Olaya-Galán, Sandra P. Salas-Cárdenas, Adriana P. Corredor-Figueroa, Gertrude C. Buehring, HuaMin Shen, Manuel A. Patarroyo, Ma, Fernanda Gutierrez. Evidence of bovine leukemia virus genes detected in Colombian women with and without breast cancer: A zoonotic infection [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4220.

Evidence of the presence of bovine leukemia virus (BLV) in human beings and its association with breast cancer has been published in the literature, proposing it as a zoonotic infection. However, not enough evidence exists about transmission pathways nor biological mechanisms in human beings. This study was aimed at gathering experimental evidence about susceptibility of human cell lines to BLV infection. Malignant and non-malignant human cell lines were co-cultured with BLV-infected FLK cells using a cell-to-cell model of infection. Infected human cell lines were harvested and cultured for 3 to 6 months to determine stability of infection. BLV detection was performed through liquid-phase PCR and visualized through in situ PCR. Seven out of nine cell lines were susceptible to BLV infection as determined by at least one positive liquid-phase PCR result in the 3-month culture period. iSLK and MCF7 cell lines were able to produce a stable infection throughout the 3-month period, with both cytoplasmic and/or nuclear BLV-DNA visualized by IS-PCR. Our results support experimental evidence of BLV infection in humans by demonstrating the susceptibility of human cells to BLV infection, supporting the hypothesis of a natural transmission from cattle to humans.

No abstract