Involvement of a Rarely Used Splicing SD2b Site in the Regulation of HIV-1 vif mRNA Production as Revealed by a Growth-Adaptive Mutation

Abstract

We have previously reported an HIV-1 mutant designated NL-Y226tac that expresses Vif at an ultra-low level, being replication-defective in high-APOBEC3G cells, such as H9. It carries a synonymous mutation within the splicing SA1 site relative to its parental clone. In order to determine whether a certain mutant(s) emerges during multi-infection cycles, we maintained H9 cells infected with a relatively low or high input of NL-Y226tac for extended time periods. Unexpectedly, we reproducibly identified a g5061a mutation in the SD2b site in the two independent long-term culture experiments that partially increases Vif expression and replication ability. Importantly, the adaptive mutation g5061a was demonstrated to enhance vif mRNA production by activation of the SA1 site mediated through increasing usage of a rarely used SD2b site. In the long-term culture initiated by a high virus input, we additionally found a Y226Fttc mutation at the original Y226tac site in SA1 that fully restores Vif expression and replication ability. As expected, the adaptive mutation Y226Fttc enhances vif mRNA production through increasing the splicing site usage of SA1. Our results here revealed the importance of the SD2b nucleotide sequence in producing vif mRNA involved in the HIV-1 adaptation and of mutual antagonism between Vif and APOBEC3 proteins in HIV-1 adaptation/evolution and survival.

Keywords: APOBEC3; HIV-1; SA1; SD2b; Vif; adaptation; splicing.

Figure 1

HIV-1 genome organization. Of the splicing sites on the HIV-1 genome, only SD1, SA1, SD2, and SD2b sites important for the vif mRNA production are presented. The black square shows SA1D2prox that we previously identified as a regulatory region involved in the alteration of the vif mRNA/Vif protein expression levels. Restriction enzyme sites (SbfI and BsaBI) used to generate adaptive proviral clones are indicated. Nucleotide sequence around the SA1 and Y226tac mutation sites are highlighted.

Figure 2

Long-term cultures of H9 cells infected with NL4-3 or its ultra-low Vif-type mutant Y226tac to obtain adapted virus clones. (A,B) Virus samples were prepared from HEK293T cells transfected with the indicated clones and inoculated into H9 cells (10⁴, 10⁵, and 10⁶ RT units for NL4-3, Y226tac, and Y226tac(H), respectively, into 10⁶ cells). Culture media of infected cells were replaced every 3 days and fresh H9 cells were added to the cultures to promote viral growth when RT activity in the supernatants was decreased to the background level. To generate adapted virus clones, H9 cells inoculated with the culture supernatants indicated were collected and subjected to proviral DNA preparations as described.

Figure 3

Properties of various clones. Replication properties and Vif expression of control (WT NL4-3 and parental Y226tac) and adapted clones are shown. (A) Growth kinetics. Viruses prepared from HEK293T cells transfected with the indicated proviral clones were inoculated into H9 cells. Viral replication was monitored using virion-associated RT activity in the culture supernatants. Representative data from two independent experiments are shown. (B) Vif expression levels. HEK293T cells were transfected with the indicated clones, and on day 2 post-transfection, cell lysates for Western blotting analysis using anti-Vif and anti-ß-actin antibodies were prepared. An empty vector pUC19 was used as a negative control.

Figure 4

Identification of an adaptive mutation responsible for an increase in Vif expression level. (A) Mutations analyzed for an adapted clone (NL-Ad1-8). Region from vpu to env of NL-Ad1-8 carries four mutations (8env). The SA1 site is indicated. (B) Growth kinetics. Various clones constructed were transfected into HEK293T cells to prepare input virus samples for infection. H9 cells were infected with equal amounts (RT) of viruses and virus replication was monitored using RT activity in the culture supernatants. (C) Vif expression levels. HEK293T cells were transfected with the indicated clones and the cell lysates were used for Western blotting analysis. Clones analyzed are shown as indicated. Plasmids pUC19 and pNL4-3 were used as negative and positive controls, respectively.

Figure 5

Effect of an adaptive mutation g5061a within SD2b on Vif expression and viral early infectivity. (A) Schematic representation and sequences of parental mutation Y226tac (SA1 site) and adaptive mutation g5061a (SD2b site). Mutations within the sequence are highlighted by orange squares. The Vif start codon is indicated. (B) Vif expression levels. Cell lysates prepared from HEK293T cells transfected with the indicated proviral clones were analyzed for Vif and cellular ß-actin expression by immunoblotting. (C) Viral infectivity. Viruses prepared from HEK293T cells transfected with the indicated clones were inoculated into TZM-bl cells, and on day 2 post-inoculation, cells were lysed for luciferase assays. Infectivity is presented as luciferase activity relative to that exhibited by NL4-3. Mean values ± standard errors (SE) from three independent experiments are shown. Significance relative to NL4-3 was determined by Welch’s t-test. NS, not significant.

Figure 6

Growth ability and Vif expression level of adapted clones (NL-Ad3). (Left panel) For various clones, virus growth in H9 cells during the 21 days post-infection period as monitored using RT assays is indicated as positive (+) or non-detectable (−). In these clones, the mutations tac and g5061a are detected (Y) or not detected (N). (Right panel) Vif expression levels of the clones were analyzed as described in Figure 5.

Figure 7

Effect of an adaptive mutation Y226Fttc on viral replication ability, early infectivity, and Vif expression level. (A) Growth kinetics. Multi-cycle replication assays using H9 cells were performed as described in Figure 4. (B) Infectivity. Viral early infectivity was determined as described in Figure 5. (C) Vif expression levels. Vif expression levels of the indicated clones were determined as described in Figure 5.

Figure 8

Effect of various single-nucleotide mutations within the region around SA1 and SD2b on Vif expression levels of minigenome (pcNLmini-RI) constructs. (Left panel) Genome organization of the pcNLmini-RI vector [32] is presented along with various splicing sites, EcoRI site within Vpr-coding region, 5′ CMV promoter, and 3′ BGH poly A (pA). Nucleotide sequence of the region is shown. Mutations in low Vif-type, high Vif-type, and excessive Vif-type are indicated by blue, red, and green letters, respectively. Splicing sites and adaptive mutations (g5061a and Y226Fttc) are shown. (Right panel) Vif expression levels of minigenome constructs that carry the indicated mutations were analyzed as described in Figure 5. WT, wild type; Short, short exposure; Long, long exposure.

Figure 9

Effect of various mutations within the region around SA1 and SD2b on viral mRNA production. (A) Genome organization and splicing products of the minigenome (pcNLmini-RI). Splicing sites and exons are indicated. Primers used for semiquantitative RT-PCR analysis are shown by arrow heads and their names [32]. PCR products (Full, D1/A1, D1/A1-D2b/A2, D1/A1-D2/A2, and D1/A2) amplified using specific primer sets are presented along with the organization and length. (B,C) Effect of single-nucleotide mutations within the SA1D2prox region (B) and adaptive mutations (C) on the splicing pattern of the minigenome constructs. HEK293T cells were transfected with pcNLmini-RI vectors carrying the indicated mutations and on the next day, cell lysates were prepared for semiquantitative RT-PCR analysis using primer sets shown in (A). Signal intensities of PCR products were quantitated from three independent experiments. The intensities of the indicated mRNAs in each sample were normalized to those of all viral mRNAs (e1) and cellular gapdh mRNA. The normalized mRNA intensities in each sample relative to those of WT are presented.