Both tissue microenvironment and genetic changes are involved in development of cancer. We employed the Friend murine leukemia virus (F-MuLV)- induced erythroleukemia model to study the role of these parameters in induction of malignancy. The tissue microenvironment is composed of non-cellular and cellular components. In regards to the non-cellular part, we previously reported that vascular endothelial growth factor (VEGF), in combination with macrophage chemoattractant protein-5, contributes to leukemia progression in F-MuLV- infected mice. To study the influence of constitutively elevated VEGF levels on the progression of erythroleukemia, we inoculated VEGF hi/+ mice, which are heterozygous for a VEGF “hypermorphic” allele, with F-MuLV. Unexpectedly, a significant delay in erythroleukemia was observed in these mice when compared with wild-type controls. The VEGF hi/+ mice exhibited a higher natural killer (NK) cell activity, elevated B cells, and a decrease in T-cell number. Furthermore, higher erythroid progenitors (i.e. CD34+, CD36+, and TER119+ cells) were evident in the bone marrow, spleen, and peripheral blood of these mice. Also, the CFU-E levels were significantly elevated in VEGF hi/+ bone marrow cultures. We propose that a compensatory erythropoietic response combined with increased NK cell activity account for the extended survival of erythroleukemic, VEGF hi/+mice. In regards to the cellular component of tissue microenvironment we studied the role of B cells in response to F-MuLV. To test the hypothesis that virus- neutralizing antibodies are involved in providing sterilizing immunity to F-MuLV we inoculated adult female mice with F-MuLV so that their newborns are provided with anti-viral antibodies. F-MuLV challenge of these newborns did not lead to induction of erythroleukemia. Conversely, mice from a control group (newborns whose mother had not received viral inoculation) contracted erythroleukemia upon F-MuLV challenge, as shown by hepatosplenomegaly, anemia, and emergence of leukemic cells in the spleen. These results indicated the importance of anti-viral antibodies in immunity to F-MuLV and suggested that anti-F-MuLV antibodies were generated in mothers, transferred to their offspring and protected them from viral challenge. The key genetic event upon F-MuLV infection is viral integration at the Fli-1 locus. We set to identify F-MuLV integration sites in SCID mice following two observations that these mice show a delay in induction of leukemia and also they do not exhibit viral integration at the Fli-1 locus. We hypothesized that development of leukemia in these mice is due to F-MuLV integration at a region other than the Fli-1 locus. Using a GenomeWalking approach we identified a total of 15 viral integration sites in F-MuLV-infected SCID mice, with eight of them interrupting the following genes: Mex3d, Fam125b, Prdm16, Rhoq, Ahdc1, Zc3h4, Msh3, and Hcls1. Using PCR to amplify the virus- host DNA junction fragment we found that one of the viral insertion sites (chromosome 10; position 20,942,825) occurs with a frequency of 35 % and therefore is considered as a common integration site.

Both tissue microenvironment and genetic changes are involved in development of cancer. We employed the Friend murine leukemia virus (F-MuLV)- induced erythroleukemia model to study the role of these parameters in induction of malignancy. The tissue microenvironment is composed of non-cellular and cellular components. In regards to the non-cellular part, we previously reported that vascular endothelial growth factor (VEGF), in combination with macrophage chemoattractant protein-5, contributes to leukemia progression in F-MuLV- infected mice. To study the influence of constitutively elevated VEGF levels on the progression of erythroleukemia, we inoculated VEGF hi/+ mice, which are heterozygous for a VEGF “hypermorphic” allele, with F-MuLV. Unexpectedly, a significant delay in erythroleukemia was observed in these mice when compared with wild-type controls. The VEGF hi/+ mice exhibited a higher natural killer (NK) cell activity, elevated B cells, and a decrease in T-cell number. Furthermore, higher erythroid progenitors (i.e. CD34+, CD36+, and TER119+ cells) were evident in the bone marrow, spleen, and peripheral blood of these mice. Also, the CFU-E levels were significantly elevated in VEGF hi/+ bone marrow cultures. We propose that a compensatory erythropoietic response combined with increased NK cell activity account for the extended survival of erythroleukemic, VEGF hi/+mice. In regards to the cellular component of tissue microenvironment we studied the role of B cells in response to F-MuLV. To test the hypothesis that virus- neutralizing antibodies are involved in providing sterilizing immunity to F-MuLV we inoculated adult female mice with F-MuLV so that their newborns are provided with anti-viral antibodies. F-MuLV challenge of these newborns did not lead to induction of erythroleukemia. Conversely, mice from a control group (newborns whose mother had not received viral inoculation) contracted erythroleukemia upon F-MuLV challenge, as shown by hepatosplenomegaly, anemia, and emergence of leukemic cells in the spleen. These results indicated the importance of anti-viral antibodies in immunity to F-MuLV and suggested that anti-F-MuLV antibodies were generated in mothers, transferred to their offspring and protected them from viral challenge. The key genetic event upon F-MuLV infection is viral integration at the Fli-1 locus. We set to identify F-MuLV integration sites in SCID mice following two observations that these mice show a delay in induction of leukemia and also they do not exhibit viral integration at the Fli-1 locus. We hypothesized that development of leukemia in these mice is due to F-MuLV integration at a region other than the Fli-1 locus. Using a GenomeWalking approach we identified a total of 15 viral integration sites in F-MuLV-infected SCID mice, with eight of them interrupting the following genes: Mex3d, Fam125b, Prdm16, Rhoq, Ahdc1, Zc3h4, Msh3, and Hcls1. Using PCR to amplify the virus- host DNA junction fragment we found that one of the viral insertion sites (chromosome 10; position 20,942,825) occurs with a frequency of 35 % and therefore is considered as a common integration site.

APOBEC3 (A3) proteins are a family of host-encoded cytidine deaminases that protect against retroviruses and other viral intruders. Retroviruses, unlike other viruses, are able to integrate their genomic proviral DNA within hours of entering host cells. A3 proteins hinder retroviral infectivity by editing retroviral replication intermediates, as well as by inhibiting retroviral replication and integration through deamination-independent methods. These proteins thus constitute the first line of immune defense against endogenous and exogenous retroviral pathogens. The overall goal of my Master's project was to better understand the critical role A3 proteins play in restricting inter- and intra-host transmission of retroviruses. There are two specific aspects that I focused on: first, investigating the role of mouse APOBEC3 (mA3) in limiting the zoonotic transmission of murine leukemia retroviruses (MLVs) in a rural environment; second, to identify the molecular features in MLVs that confer susceptibility or resistance to deamination by mA3. For the first part of my project, we collected blood samples from dairy and production cattle from four different geographical locations across Canada. We then designed a novel PCR screening strategy targeting conserved genetic regions in MLVs and Mouse Mammary Tumor Virus (MMTV) and MMTV-like betaretroviruses. Our results indicate that 4% of animals were positive for MLV and 2% were positive for MMTV. Despite crossing the species barrier by gaining entry into bovine cells, our study also demonstrates that the bovine A3 protein is able to potently inhibit the spread of these murine retroviruses in vitro. The next question we asked was whether mA3 could also mutate and restrict murine endogenous retroviruses and thereby partake in limiting zoonotic transmission. Moloney MLV and AKV MLV are two highly homologous murine gammaretroviruses with opposite sensitivities to restriction by mA3: MoMLV is resistant to restriction and deamination while AKV is sensitive to both. Design of MoMLV/AKV hybrid viruses enabled us to map the region of mA3 resistance to the region encoding the glyco-Gag accessory protein. Site-directed mutagenesis then allowed us to correlate the number of N-linked glycosylation sites with the level of resistance to deamination by mA3. Our results suggest that Gag glycosylation is a possible viral defence mechanism that arose to counteract the evolutionary pressure imposed by mA3. Overall, my projects show the important role A3 proteins play in intrinsic immunity, whether defending the host from foreign retroviral invaders or endogenous retroviral foes.