Figure 3

DND1 does not affect APOBEC3 function. (a) DND1 does not affect APOBEC3 antiretroviral activity. Infectivity of HIV-GFP produced in the presence of control vector (vector) (lane 1) or vector encoding mouse DND1 (lane 2) (P = 0.2082). Infectivity of HIV-GFP produced in the presence of mouse APOBEC3 (A3) (lane 3) and APOBEC3 plus DND1 (lane 4) (P = 0.1464). Empty vector was used to equalize the amount of construct used in each assay lane. Results from two independent experiments were averaged. Error bars indicate the difference in infectivity observed between the two experiments. (b) DND1 does not affect MusD restriction by APOBEC3. Effect of control vector (lane 1), mouse DND1 (lane 2), mouse APOBEC3 (lane 3), both mouse DND1 and APOBEC3 (lane 4) on MusD retrotransposition, relative to the vector control. Transposition was monitored by the number of G418-resistant colonies. Results from two independent experiments were averaged. P value comparing lanes 3 and 4 is 0.10980. (c) Histology section through testes of double homozygous male Ter/Ter ;A3−/− (or Ter/Ter; Apobec3−/−), (d) Ter/Ter and (e) wild-type (+/+) mice. Arrow points to lumen in seminiferous tubules showing lack of germ cells persist in Ter/Ter ; A3−/− testis similar to that in testis of Ter/Ter mice. Testes of Apobec3−/− (A3−/−) mice have normal wild-type germ cell histology similar to +/+ (not shown). (f) Proposed model how APOBEC3G regulates DND1 function. DND1 blocks miRNA activity (left panel). We observe decreased protein translation (as measured by luciferase activity) in the presence of APOBECG3 suggesting that APOBEC3G blocks DND1 function. When DND1 binding sites on P27-3′-UTR are inactivated, both DND1 and APOBEC3G fail to affect luciferase activity. This suggests that APOBEC3G functions through DND1, may be by removing or sequestering DND1 to restore miRNA access to the 3′-UTR of P27 (right panel).