A highly conserved residue in the C-terminal helix of HIV-1 matrix is required for envelope incorporation into virus particles

Abstract

The incorporation of viral envelope (Env) glycoproteins into nascent particles is an essential step in the production of infectious human immunodeficiency virus type 1 (HIV-1). This process has been shown to require interactions between Env and the matrix (MA) domain of the Gag polyprotein. Previous studies indicate that several residues in the N-terminal region of MA are required for Env incorporation. However, the precise mechanism by which Env proteins are acquired during virus assembly has yet to be fully defined. Here, we examine whether a highly conserved glutamate at position 99 in the C-terminal helix is required for MA function and HIV-1 replication. We analyze a panel of mutant viruses that contain different amino acid substitutions at this position using viral infectivity studies, virus-cell fusion assays, and immunoblotting. We find that E99V mutant viruses are defective for fusion with cell membranes and thus are noninfectious. We show that E99V mutant particles of HIV-1 strains LAI and NL4.3 lack wild-type levels of Env proteins. We identify a compensatory substitution in MA residue 84 and show that it can reverse the E99V-associated defects. Taken together, these results indicate that the C-terminal hydrophobic pocket of MA, which encompasses both residues 84 and 99, has a previously unsuspected and key role in HIV-1 Env incorporation.

Fig 1

Functional domains and single amino acid substitutions in the HIV-1 MA protein. A schematic representation of previously defined functional domains in HIV-1 MA and the relative locations of previously characterized single amino acid substitutions. The E99V MA substitution analyzed in this study is located at the beginning of the C-terminal helix of the protein.

Fig 2

E99V or E99K MA mutant particles have marked defects in virus infectivity and replication. (A) Virus stocks generated in 293T cells by transfection with the indicated pLAI molecular clones were normalized on the basis of RT activity and used to infect TZM-bl cells. Single-cycle infectivity was determined at 24 h postinfection using the Beta-Glo infectivity assay and a luminometer. Infectivity is based on the level of LAI-WT, which represents 100% infectivity. The data shown are representative of four independent experiments performed in duplicate. (B) Virus stocks generated in 293T cells by transfection with the indicated pLAI clones were normalized on the basis of RT activity and used to infect various human T-cell lines: M8166 (top left), C8166 (top right), or MT-4 cells (bottom). Fresh cells and medium were added to each culture on day 7. RT activities in cell culture supernatants were monitored every other day using a standard enzymatic assay. The data shown are representative of two independent experiments performed in duplicate.

Fig 3

HIV-1 Env proteins are inefficiently incorporated into E99V or E99K MA mutant particles. Transfected HeLa cells (top) or virus particles (bottom) were harvested, and cellular or viral lysates were prepared as previously described (13). The levels of Env proteins in lysates were analyzed by SDS-PAGE and immunoblotting using monoclonal antibodies directed against HIV-1 Env, HIV-1 Gag, and human β-actin or against HIV-1 Env and HIV-1 Gag, respectively.

Fig 4

E99V or E99K MA mutant particles have a marked defect in membrane fusion. Equal amounts of virus were added to TZM-bl cells grown in 96-well plates. The plates were centrifuged at room temperature for 30 min and then incubated further at 37°C for 50, 100, or 150 min. Cells were washed with 1× PBS at the indicated times, and the BLaM substrate was added to the culture medium. Plates were incubated overnight at room temperature to allow cleavage of the loaded substrate. Cellular BLaM activity was monitored as an indicator of virus-cell fusion events using a fluorimeter.

Fig 5

The E99V MA-associated defect is mitigated in NL4.3-derived virus particles. (A) Relative infectivity of each mutant virus and the corresponding wild-type virus was assayed at 24 or 48 h postinfection as described in the legend to Fig. 2A. The assessment of infectivity is based on the level of LAI-WT determined at 24 h postinfection. The data are representative of three independent experiments performed in duplicate. (B) Equal RT units of the indicated viruses were used to infect MT-4 cells, and virus replication kinetics were monitored every other day for 11 days, as described in the legend to Fig. 2B. Fresh cells and medium were added to each culture on day 7. The data shown are representative of two independent experiments performed in duplicate. (C) Lysates of transfected HeLa cells (top) or virus particles (bottom) were prepared and analyzed as described in the legend to Fig. 3.

Fig 6

(A) Amino acid differences between the LAI and NL4.3 MA proteins. Alignment of the LAI and NL4.3 MA protein sequences showing four nonconservative amino acid differences (highlighted in red). (B) Relative locations of various residues in the HIV-1 MA structure. Different views of the HIV-1 MA protein (PDB accession number 1UPH [63]) are shown: the top row shows the membrane-binding surface on top, and the bottom row shows MA structures with the membrane-binding surface on the right. Three different depictions of each view are shown: left, stick models; middle, ribbon models; right, space-filling models. Residue E99 is shown in green. The C-terminal hydrophobic pocket consists of V84 (orange), Y86, I92, V94, A100, and K103 (blue) (7). Amino acids 28, 124, and 125 are shown in red; these positions and residue 84 differ between the LAI and NL4.3 MA sequences. Residue 84, a threonine in LAI and a valine in NL4.3, has close proximity to residue 99 compared to the positions of the other three nonconservative changes.

Fig 7

Amino acid substitution at residue 84 rescues the E99V-associated defect in LAI-derived particles but exacerbates its effect in NL4.3-derived particles. (A) Single-cycle infectivity was monitored at 48 h postinfection using the Beta-Glo infectivity assay, as described in the legend to Fig. 2A. The assessment of infectivity for each mutant virus is based on the level for LAI-WT (top) or NL4.3-WT (bottom), which represents 100% infectivity. The amino acid identities at positions 84 and 99 in each virus are indicated above the graphs. The data shown are representative of two independent experiments performed in duplicate. (B) Viral lysates were prepared and analyzed as described in the legend to Fig. 3. (C) Virus-cell membrane fusion events were assayed in TZM-bl cells as described in the legend to Fig. 4. The data shown are representative of three independent experiments performed in duplicate.

Fig 8

Infectivity of LAI-derived viruses with E99V or E99K MA substitutions is specifically enhanced by a T84V MA substitution. The data depict the fold improvement between viruses with and without the T84V substitution at four serial dilutions. Data shown are representative of four independent Beta-Glo infectivity assays performed in duplicate and analyzed at 24 h postinfection.

Fig 9

Truncated Env proteins can rescue membrane fusion but not the infectivity defect of LAI-derived particles with an E99V or E99K MA substitution. (A) Single-cycle infectivities of LAI-derived viruses with either full-length (blue bars, inoculated at 5K RTU/well) or truncated (red bars, inoculated at 2.5K RTU/well) HIV-1 Env were monitored in TZM-bl cells at 24 h postinfection using the Beta-Glo reporter assay. Infectivity of each mutant virus is based on the levels in LAI-WT or LAI-Δ144 infections, which represent 100% infectivity. The data shown are representative of four independent experiments performed in duplicate. (B) Equal RT units of the indicated viruses were used to infect MT-4 cells, and virus replication kinetics were monitored every other day for 11 days as described in the legend to Fig. 2B. Fresh cells and medium were added to each culture on day 7. The data shown are representative of two independent experiments performed in duplicate. (C) Virus-cell membrane fusion events were assayed in TZM-bl cells as described in the legend to Fig. 4. The data shown are representative of three independent experiments performed in duplicate.