PF-3 Is nigerian breast cancer associated with endogenous retrovirus Xq21.33?

Abstract

Background: Wildschutte et al recently discovered a new HERV K HML-2 (HK2) endogenous retrovirus Xq21.33 (NCBI GenBank accession no. KU054272) to be uncommonly present in Pygmy populations and in Nigerian people in Africa (<4%). It has been estimated that this virus has been inserted in the human genome between ∼0.67–1.3 Mya and possibly earlier. Since this virus is fully intact and is similar to mouse mammary tumor virus, we looked to see if it is associated with breast cancer in Nigerian populations. Methods: Peripheral blood mononuclear cells were obtained with informed consent from Nigerian women with breast cancer (BCa) (237) and matched controls (225) without BCa. Samples from the United States included samples from a BCa study and HIV study carried out earlier. DNA was extracted using the DNeasy Blood and Tissue Kit (Qiagen). Samples were tested for the Xq21.33 virus using primer Xq21.33R2 ACGCTGTTTTGTCCCTTTGA and K1011 CGTCGACTTGTCCTCAATGACCACGCT to screen samples, and primers 1584R TTTGCGTTGACTGAGCCATTACCG, 2499R TTG AGC AAC ATC TTG GAG CCT TGC, 3460R ACTTGCCCAATATGCAGCCTTTCCTCC, 6115R TCAACTATGGCCTGTCCTTGC, F Flank TCAGTCTGAAGACAATGACATACT T and 5359F TCGGCTCAAAGAGCAGAGATGGTT to confirm positive samples. Samples were tested using the Long Amp PCR kit (New England Bio Labs) conditions suggested by the manufacturer using a total volume of 25ul. Hetero vs Homozygosity was determined using F Flank and Xq21.33R2 with Platinum taq Hi Fi Thermo Fisher also using conditions suggested by the manufacturer. Samples with adequate DNA were tested using primer K1011 and Xq21.1R2. Positive samples were all confirmed using Xq21.33R2 to 1584R, 2499R, 3460R and 5359F to F Flank. One batch of samples did not yield good quality DNA and some samples had such low yields of DNA that the data on these patients was not included in this study. Results: We detected 28/220 (12.7%) positive samples in women with BCa vs 23/236 (9.7%) in women without BCa. These differences were not statistically significantly different. 5/28 women with BCa (ages 38, 44, 50, 53, 82) vs 2/23 (ages 32, 58) normal women were homozygous for Xq21.33. As shown below women with BCa and virus were more likely to be post-menopausal (46%) than women without BCa and virus (8.7%). Similarly women with BCa and no virus were also more likely to be post-menopausal 35% vs women with no BCa and no virus 19.7%. There was not enough power in this study to determine if the virus is contributing to more cases of post-menopausal breast cancer in positive women. Among women who are older than 47 years, the proportion of virus xq21.33 positives is higher among women with BCa (14.9%) than women without BCa (5.13%), but it is not significant (P-value = 0.15) [see below] Virus xq21.33 in cases and controls in age groups >47 years n = 126. Among women who are 47 years or younger, the proportion of virus xq21.33 is lower among women with BCa (10.07%) than women without BCa (12.21%), but it is not significant (P-value = 0.60). Virus xq21.33 in cases and controls in age groups <47 years n = 321. We also tested for Xq21.33 in the following USA samples: Conclusions: Our sample size does not provide sufficient power to see if Xq21.33 presents an increased risk for breast cancer. However, the trend too increased risk in virus positive women 12.7% vs 9.7% in controls, and the increase in numbers of women >47 year with +BCa +virus (46.4%) vs the +BCa −virus (35.6%), and the striking low numbers of women in −BCa +virus (8.7%) vs (19.7%) in −BCa −virus and greater homozygosity in +Bca +virus women suggests differences which may be caused by carrying this provirus. We suggest designing trials with adequate sample sizes of virus+ and − young women without BCa to assess the future risk of cancer from this provirus. We are now sequencing the env gene to see if women with BCa have a more replicative competent virus. We are gathering pathological data to see if +women have different forms of BCa than –women. This virus may be still present in Nigerian rodent and primate populations and should be searched for in these animals. It may still infect people contributing to BCa without becoming endogenized. It could be detected by deep sequencing DNA from BCa tissue looking for integrated Xq21,33 in other sites than the X chromosome. Molecular clones and more cell lines, like DT22, should be established from BCa+ Virus+ to look for replicative virus with alternative integration sites. Virus should be searched for in HIV+ Nigerians to see if it contributes to other neoplasms. The discovery of a human BCa virus would present a new way of treating or preventing breast cancer using antiretrovirals.