Differential effects of mouse APOBEC3 after incorporation into murine leukemia viruses

APOBEC3 proteins are cytidine deaminases that are potent inhibitors of retrovirus replication. These restriction factors induce G→A hypermutation as a result of deamination of cytidine in the single stranded transcripts generated by reverse transcription during replication. Mouse APOBEC3 (mA3) effectively inhibits the replication of endogenous retroelements whereas most exogenous murine leukemia viruses (MuLV) are largely resistant to the action(s) of mA3. However, several studies have reported significant inhibition of infectivity by virion-associated mA3 that did not appear to be the result of G→A hypermutation. In this report the influence of different levels of virion-associated mA3 on the exogenous MuLV, CasFrKP was examined. A correlation was observed between the level of incorporated mA3, the virion specific infectivity, the virion reverse transcriptase (RT) activity and the mutation frequency. Although G→A hypermutation of CasFrKP was not observed, the frequency of other transition mutations was significantly increased. G→A mutations induced by mA3 have been reported to exhibit a strong consensus of two bases flanking the mutation. No strong consensus was observed for mA3-induced mutations of CasFrKP suggesting a general decrease in the fidelity of the RT. The results of this study strongly suggest an interaction between mA3 and the RT of CasFrKP leading to a loss of activity as well as fidelity of the polymerase that ultimately results in a significant loss of infectivity. In contrast to CasFrKP, AKV, an ecotropic MuLV isolated from AKR mice, has been reported to undergo hypermutation mediated by mA3. In this report we also confirm that AKV does undergo G→A hypermutation upon infection of 3T3 cells, however hypermutation is not augmented in viruses that have incorporated HA-tagged mA3. Furthermore, the observed G→A transition mutations are not randomly distributed in that multiple mutations frequently occur in a single transcript. These observations suggest that the G→A mutations are the result of the incorporation of an unknown restriction factor in only a portion of the released virions.