Elucidation of the Molecular Consequences of Two Unique p6Gag Mutations Derived from HIV-1 CRF07_BC-infected Patients

Background We previously observed that individuals infected with HIV-1 CRF07_BC showed slower disease progression than those infected with HIV-1 subtype B or CRF01_AE. CRF07_BC viruses carry two unique mutations in the p6 Gag protein: insertion of PTAPPE sequences downstream of the original Tsg101 binding domain, and deletion of a seven-amino-acid sequence ( 30 PIDKELY 36 ) that partially overlaps with the Alix binding domain. To further define the role of these mutations in virus release and replication, we introduced them into the HIV-1 proviral clone pNL4-3 for functional characterization. Results We found that the seven-amino-acid deletion, but not the PTAPPE insertion, significantly decreased virus release, Gag processing, and virus infectivity. The seven-amino-acid deletion also resulted in a virus replication defect in both T-cell lines and peripheral blood mononuclear cells. We found that these defects were caused by the seven-amino-acid deletion in p6 Gag , especially deletion of Tyr-36 of p6 Gag , not the deletion of the overlapping p6* sequence in the HIV-1 GagPol protein. The p6 Gag deletion mutant was resistant to a dominant-negative Alix fragment, suggesting a loss of binding between p6 Gag and Alix. Conclusions Our results indicate that the patient-derived seven-amino-acid deletion in p6 Gag of HIV-1 CRF07_BC virus affects virus release, infectivity and replication capacity by disrupting the interaction between HIV-1 p6 Gag and host protein Alix. These results may explain the slower disease progression observed in the subjects infected with HIV-1 CRF07_BC bearing this unique mutation.

capsid (CA), nucleocapsid (NC), p6, and two spacer peptides, spacer 1 (SP1) and spacer 2 (SP2). PR-mediated Gag processing triggers virus maturation and is essential for the conversion of the immature VLP to the infectious virion [11,12]. The p6 Gag domain is required for virion budding-off from the plasma membrane through the action of its two highly conserved late domain motifs: PT/SAP and YPXnL [11,12]. The PT/SAP motif plays a major role in HIV-1 release by binding to Tsg101, a component of the endosomal sorting complex required for transport I (ESCRT-I) [11,[13][14][15]. Mutation of the PT/SAP motif results in a severe defect in virus budding [16,17]. The YPXnL motif regulates virus release by directly binding to the ESCRT-associated host protein Alix [18,19]. Mutations that block YPXnL-Alix binding or overexpression of Alix disrupt virus replication and virion production [18][19][20][21][22] [26][27][28]. Thus, the 7-aa PIDKELY deletion in p6 Gag also results in deletion of amino acids DRQGTVS in p6 * (Fig. 1), which is 3 amino acids away from p6*-PR cleavage site. It has been reported that PR activation and/or PR-mediated maturation could be affected by the substitution of four amino acids, SFNF, near the p6*-PR cleavage site [29,30].
We hypothesize that the two p6 Gag late domain-related mutations described above, insPTAP and Δ7, may affect virus release by disrupting the interaction of p6 Gag with Tsg101 and Alix, respectively. In addition, the overlapping 7-aa deletion of DRQGTVS in p6 * may affect the activity of HIV-1 PR. In this study, we introduced these mutations into a full-length, infectious HIV-1 proviral clone and characterized their role in virus release and replication. We found that the 7-aa deletion in p6 Gag (Δ7), but not the PTAP insertion (insPTAP), significantly decreased virus release, Gag processing, virus infectivity and replication. We further demonstrated that the Δ7 mutation in p6 Gag , especially the deletion of tyrosine 36 (Y36), but not the corresponding deletion mutation in p6 * , caused the observed defects by disrupting the interaction between HIV-1 p6 Gag and Alix. These defects likely contribute to the slower disease progression observed in subjects infected with the HIV-1 CRF07_BC Δ7 variant.

HIV-1 p6 Gag deletions impair virus release, infectivity and replication
The deletion of the Tsg101-binding motif PTAP (ΔPTAP), insPTAP, Δ7, and PΔ7 p6 Gag mutations ( Fig. 1) were introduced into the full-length HIV-1 molecular clone pNL4-3. Wild type (WT) and mutant clones were transfected into 293T cells (Fig. 2a). Western blotting (WB) of cell and viral lysates was performed (Fig. 2a), and virus release and Gag processing efficiencies were quantified ( Fig. 2b and c). By quantifying the p24 levels in virions relative to total Gag, we determined that deletion of the Tsg101-binding motif PTAP (ΔPTAP) severely impaired virus particle production (Fig. 2b), as reported previously [16,17]. However, the PTAPPE insertion (insPTAP) downstream of the original PTAP motif did not significantly affect virus release (Fig. 2b) except for a slightly increased tolerance to overexpression of Tsg101 (data not shown). The 7-aa deletion (Δ7) and the double mutation PΔ7 moderately inhibited virus production, to ~ 77% and ~ 62% of the WT level, respectively (Table 1, Fig. 2b). Similar results were obtained when the reverse transcriptase (RT) activity of culture supernatants was measured to determine the efficiency of virus release (Table 1). In addition, the efficiency of Gag processing was ~ 73%, 73%, 52% and 37% for ΔPTAP, insPTAP, Δ7 and PΔ7, respectively (Fig. 2c). These results demonstrate that the 7-aa deletion and the double mutation PΔ7, but not the PTAP duplication in p6 Gag , cause defects in virus release and Gag processing. We further investigated the infectivity of these p6 Gag mutants and found that the insPTAP mutant showed levels of virus infectivity similar to those of WT in the TZM-bl system ( Table 1). The infectivity of the mutant Δ7 and the double mutant PΔ7 was ~ 56% and 69% WT level, respectively (Table 1). We next analyzed virus replication kinetics in the SupT1 and MT-4 T-cell lines and in primary human peripheral blood mononuclear cells (PBMCs) from two donors. As expected, the ΔPTAP mutant showed very low-level and delayed replication kinetics in the T-cell lines ( Fig. 3a and b), and no replication was observed in PBMCs ( Fig. 3c and d). The insPTAP mutant was replication competent in both T-cell lines and PBMCs with no major difference from WT (Fig. 3). However, the Δ7 and PΔ7 mutants were defective in virus replication in SupT1 and MT-4 T cells ( Fig. 3a and b, Table 1), while low-level replication was also observed in PBMCs from two different donors ( Fig. 3c and d, Table 1). The defects in viral replication were consistent with impairment of reverse transcription, in particular the initiation of reverse transcription (Fig. 4a), but not elongation (Fig. 4b), compared to WT. We found that the mutant insPTAP did not affect initiation of reverse transcription while the mutants PΔ7 and Δ7 reduced the initiation efficiency to ~ 88% and 57% of WT level, respectively (Fig. 4a). These results demonstrate that the 7-aa deletion and the double mutation PΔ7, but not the PTAP duplication mutation in p6 Gag , result in defects of virus infectivity and replication.
Deletion mutation in p6 Gag but not in p6* is responsible for the defects in virus release and Gag processing The 7-aa PIDKELY deletion in p6 Gag also results in deletion of amino acids DRQGTVS in p6* ( Fig. 1). To determine the effects of the deletion mutation in p6 Gag and p6* in virus release and Gag processing, we constructed several HIV-1 pNL4-3/KFS clones expressing p6 Gag with the 7-aa deletion (GagΔ7) and p6* with the 7-aa deletion (GagPolΔ7). The GagPol construct expresses GagPol but does not express Gag, as the result of a 1nucleotide insertion in the frameshift region that places gag and pol in the same ORF [31].
The Gag-and GagPol-expressing plasmids were co-transfected into 293T cells at a ratio of 15:1 to generate viral particles with a similar ratio of Gag to GagPol proteins as normal HIV-1 particles [31] (data not shown). We found that the deletion mutation in p6* did not significantly inhibit virus particle production or Gag processing (Fig. 5). In contrast, the deletion in p6 Gag resulted in a decrease in virus release and Gag processing efficiency tõ 47% of the WT level ( Fig. 5b and c). Furthermore, the deletions in p6 Gag and p6* did not affect the incorporation and processing of GagPol protein (data not shown). These results indicate that the 7-aa deletion in p6 Gag , but not the deletion in the p6* domain, impairs virus release and Gag processing.
Tyrosine 36 (Y 36 ) in p6 Gag is critical for virus release and Gag processing It has been reported that mutation Y36A in p6 Gag markedly impaired virus particle production and Gag processing [20], demonstrating an important role of Y36 in controlling HIV-1 release and Gag processing. Consistent with these results, we observed that deletion of Y 36 (ΔY) severely impaired virus production and Gag processing ( that contains a Phe-to-Asp substitution in Alix residue 676 that abrogates p6 binding [21], did not inhibit particle release ( Fig. 6a and b). Notably, overexpression of Alix V did not affect virus release for the three p6 Gag mutants analyzed: ΔY, Δ6, or Δ7 ( Fig. 6a and b), indicating that these mutations prevent the interaction between p6 Gag and Alix. These data suggest that the deletion mutations in p6 Gag inhibit virus release by disrupting the binding of p6 Gag and Alix protein.  [20]. Our study also indicated that PTAP duplication coupled with the 7aa deletion in p6 Gag may further interfere with the global folding of p6 and result in severe defects in particle release, Gag processing and virus replication.
It is noteworthy that deletion mutations in p6 Gag also result in amino-acid changes in p6* in the overlapping pol ORF. Several studies have reported that mutations upstream or downstream of the PR region potentially affect PR activity and Gag processing [47][48][49][50].
Wondrak et al. demonstrated that insertion of Alanine into the p6-PR cleavage site (from Phe-Pro to Phe-Ala-Pro) severely impaired the autoprocessing of PR [51]. Chiu et al.
reported that removal of the entire p6 * region did not affect incorporation of GagPol into virions, but abrogated viral infectivity [52]. In our study, the 7-aa deletion in p6* is 3-aa (Phe-Ser-Phe) away from the p6-PR cleavage site. Our results showed that the corresponding deletion mutation in p6* did not influence virus release, but moderately decreased Gag processing. In addition, the deletion mutations in p6 Gag or p6* did not containing both insPTAP and D7 (Fig. 1a, b). A PTAP motif deletion mutant (DPTAP) was used as a control for measuring defective virus release (Fig. 1b). or the RT activity of culture supernatants relative to WT level [17,57]. Virus maturation was measured by Gag processing and expressed as a ratio of virion-associated p24 over Pr55 Gag levels as described previously [6,30].

Virus replication and infectivity
PBMCs. After denaturation, proteins were subjected to SDS-PAGE, transferred to a polyvinylidene fluoride (PVDF) membrane, and incubated with HIV-Ig. The membrane was then incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies, and the chemiluminescence signal was detected by using Western Pico substrate (Thermo Scientific) or Western Femto substrate (Fdbio Science). Quantification of the protein band intensity was performed using ImageLab software (Bio-Rad).

Statistical analysis
Statistics were calculated using SPSS Statistics 20. Unpaired t tests were performed and two-tailed *P < 0.05, **P < 0.01, and ***P < 0.001 were considered statistically significant.  . c Gag processing. Gag processing was expressed as the ratio of p24 relative to Pr55Gag in virions. The data were plotted in bar graphs. The efficiency of virus release and Gag processing for WT was set as 100%. Error bars indicate the standard deviation from more than three independent experiments; ns, not significant. *P < 0.05, **P < 0.01, and ***P < 0.001.  The level of WT virus was set as 100%. b Relative efficiency of DNA elongation.
The level of R-U5 product was set as 100%, and the amounts of U3-U5 and R-5'UTR products were expressed relative to the R-U5 level. Results from three independent experiments were summarized; error bars indicate standard deviation (SD); ns, not significant. **P < 0.01.

Figure 5
The role of the seven-amino-acid deletion in p6Gag and the overlapping deletion in p6*. Detection of cell-and virion-associated proteins by WB analysis. a 293T cells were co-transfected with HIV-1 proviral clones that encode Gag and GagPol at a ratio of 15:1. Two days post-transfection, virus and cell lysates were harvested and measured by WB. Virus release efficiency (b) and Gag processing (c) were calculated as described in Figure 2. Virus production for WT was set as 100%. Standard deviation was obtained from more than three independent 32 experiments; ns, not significant. *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 6
The p6 mutants are resistant to inhibition by Alix V overexpression. a 293T cells were co-transfected with HIV-1 proviral DNA encoding WT Gag or the deletion of the 36th reside (ΔY), the 6-aa deletion (Δ6), or Δ7 mutant in p6Gag, together with control plasmid DNA (pcGNM2-Alix) or HA-Alix V or HA-Alix V/F676D expressing vector. At 24h post-transfection, cell-associated and virus-associated proteins were measured by WB. Pr55Gag, p41, p24 and HA-Alix V are indicated. The efficiency of virus release (b) and Gag processing (c) was calculated as described in Figure 2. The levels of WT virus produced in the absence of Alix V (-Alix) was set as 100%. Error bars show the standard deviation (SD) from 4 experiments. ns, not significant. *P < 0.05, **P < 0.01.