The capsid-spacer peptide 1 Gag processing intermediate is a dominant-negative inhibitor of HIV-1 maturation

Abstract

The human immunodeficiency virus type 1 (HIV-1) maturation inhibitor bevirimat disrupts virus replication by inhibiting the cleavage of the capsid-spacer peptide 1 (CA-SP1) Gag processing intermediate to mature CA. The observation that bevirimat delays but does not completely block CA-SP1 processing suggests that the presence of uncleaved CA-SP1 may disrupt the maturation process in trans. In this study, we validate this hypothesis by using a genetic approach to demonstrate that a non-cleavable CA-SP1 mutant exerts a dominant-negative effect on maturation of wild-type HIV-1. In contrast, a mutant in which cleavage can occur internally within SP1 is significantly less potent as a dominant-negative inhibitor. We also show that bevirimat blocks processing at both the major CA-SP1 cleavage site and the internal site. These data underscore the importance of full CA-SP1 processing for HIV-1 maturation and highlight the therapeutic potential of inhibitors that target this Gag cleavage event.

Fig. 1

Levels of uncleaved CA and CA-SP1 in WT and WT/CA5 mixed virions. (A, Upper panel). Proteolytic processing of the HIV-1 Gag precursor protein, Pr55^Gag. Major Gag domains – matrix (MA), capsid (CA), nucleocapsid (NC), and p6 - and spacer peptides SP1 and SP2 are indicated. Final Gag processing products are shown. (A, Lower panel). WT (NL4-3) and CA5 amino acid sequence spanning the CA-SP1 junction. Mutations in CA5 are highlighted and the Gag amino acid numbers are provided. PR cleavage sites flanking SP1 are shown with black arrowheads. The white arrowhead indicates the secondary, internal cleavage site within SP1. CA* represents CA plus a four-amino-acid extension into SP1, which accumulates in samples containing the L363I mutation. (B) Western blot of CA-SP1 and CA in particles generated by WT, CA5 or a mixture of these clones at the indicated DNA ratios. (C) Quantification of CA and CA-SP1 in virus particles. Percentages of CA-SP1 relative to total CA + CA-SP1 were determined using quantitative Western analysis. Error bars indicate standard deviation (n = 4).

Fig. 2

Dominant-negative effect of uncleaved CA-SP1 on HIV-1 infectivity. (A) TZM-bl cells were infected with virus stocks obtained by transfecting HeLa cells with WT (pNL4-3) or CA5 molecular clones, or WT/CA5 mixtures (at indicated DNA ratios). Virus stocks were normalized for RT activity and infection performed with 100,000, 50,000, 25,000, or 12,500 RT cpm. A representative graph indicating light units versus virus input used for each virus stock is shown. (B) Quantitative data showing levels of infectivity with a virus input of 50,000 RT cpm, n = 7. (C) Jurkat cells were infected with virus stocks normalized to 10,000 RT cpm. Supernatants were collected every 2–3 days and monitored for RT activity for a period of 45 days.

Fig. 3

Effect of uncleaved CA-SP1 on virion morphology. HeLa cells were transfected with WT (pNL4-3) or CA5 molecular clones, or WT/CA5 mixtures (at indicated DNA ratios). Bar graph shows the quantification of released virions (n = 128–354) based on their morphology and placed into the following categories: conical core, centric core, acentric core, acentric core + crescent, immature, and aberrant. A representative image for each morphology is shown. Percentages for each morphology are shown to the right of each bar.

Fig. 4

Comparison between CA5 and the L363I mutant in Gag processing patterns and dominant-negative effects on HIV-1 infectivity. (A) Western blot of CA, CA-SP1, and CA*, in WT, L363I, and CA5 particles. CA* represents CA + four amino acids of SP1. (B) TZM-bl cells were infected as described in Fig 2A with virus stocks obtained by transfecting HeLa cells with WT (pNL4-3), CA5, or L363I molecular clones, or WT/CA5 and WT/L363I mixtures at the indicated DNA ratios. Quantitative data showing levels of infectivity with a virus input of 50,000 RT cpm, n=3.

Fig 5

Effect of BVM on Gag processing and HIV-1 infectivity for the L363I and M367I mutants. (A–B) CA-SP1 processing in the presence or absence of 1 μg/ml BVM. HeLa cells were transfected with WT (pNL4-3), A1V (BVM-resistant mutant), L363I, or M367I molecular clones and incubated in the presence (+) or absence (−) of BVM. One day posttransfection, cells were metabolically radiolabeled with [³⁵S] Met/Cys for 2–4 hr and released virus particles were collected by ultracentrifugation. Gag proteins were detected from virus lysates by immunoprecipitation with HIV-Ig and (A) CA-SP1, CA*, and CA were resolved by SDS-PAGE. CA* is denoted by *. (B) Quantification of CA, CA-SP1, and CA* in virus particles. Percentages of CA-SP1 relative to CA + CA-SP1 [for WT (pNL4-3), A1V, or M367I mutants] or CA* + CA-SP1 (for L363I mutant) were determined by phosphorimager analysis. Error bars indicate standard deviations (n = 3). (C) TZM-bl cells were infected with virus obtained from HeLa cells transfected with WT (pNL4-3), L363I, or M367I molecular clones in the presence (+) or absence (−) of 1 μg/ml BVM. Virus produced one day posttransfection in the course of 4 hr with or without BVM was used for infection at different dilutions ranging from 63,000 to 1,250,000 RT cpm. Quantitative data showing levels of infectivity with a virus input of 250,000 RT cpm, n = 3.