Kinetic analysis of human immunodeficiency virus type 1 assembly reveals the presence of sequential intermediates

Abstract

The assembly and budding of lentiviruses, such as human immunodeficiency virus type 1 (HIV-1), are mediated by the Gag protein precursor, but the molecular details of these processes remain poorly defined. In this study, we have combined pulse-chase techniques with density gradient centrifugation to identify, isolate, and characterize sequential kinetic intermediates in the lentivirus assembly process. We show that newly synthesized HIV-1 Gag rapidly forms cytoplasmic protein complexes that are resistant to detergent treatment, sensitive to protease digestion, and degraded intracellularly. A subpopulation of newly synthesized Gag binds membranes within 5 to 10 min and over several hours assembles into membrane-bound complexes of increasing size and/or density that can be resolved on Optiprep density gradients. These complexes likely represent assembly intermediates because they are not observed with assembly-defective Gag mutants and can be chased into extracellular viruslike particles. At steady state, nearly all of the Gag is present as membrane-bound complexes in various stages of assembly. The identification of sequential assembly intermediates provides the first demonstration that HIV-1 particle assembly proceeds via an ordered process. Assembly intermediates should serve as attractive targets for the design of antiviral agents that interfere with the process of particle production.

FIG. 1

Distribution of total and newly synthesized HIV-1 Gag on Optiprep velocity gradients. Transfected COS-1 cells were pulse-labeled for 7 min with Tran³⁵S-label, and a denucleated P100 fraction was centrifuged over a 0 to 18% Optiprep gradient. The fractions were analyzed for newly synthesized and total Gag. (A) Top, distribution of total Gag in a representative gradient. Gradient fractions were analyzed directly by anti-p24CA Western blotting. Bottom, newly synthesized Gag distribution in the same gradient. Gradient samples were immunoprecipitated with anti-CA antiserum, followed by SDS-PAGE and phosphorimaging. (B) Graphic depiction of the gradient in panel A. Bands of total and newly synthesized Gag were quantitated, and the values were normalized to the highest value in each gradient (arbitrarily set at 100) and plotted against the density of the fraction. (C) Graphic depiction of distribution of total and newly synthesized Gag31-GFP on an Optiprep gradient. Optiprep gradient fractionation was also performed with Gag31-eGFP and Gag69-eGFP, which express protein at levels comparable to pHXB2ΔBal, yielding the same results (not shown).

FIG. 2

Optiprep gradient migration of intracellular Gag at various times after synthesis and comparison to that of VLPs. (A) Migration of intracellular Gag. Transfected cells were pulse-labeled and incubated in chase medium for 0, 2, 4, or 6 h, and denucleated P100 fractions were prepared and fractionated on Optiprep gradients as described in the legend to Fig. 1. The total counts per minute of the Gag bands decreased dramatically by the first chase point (not shown). The counts per minute of radiolabeled Gag in each fraction was therefore normalized to the densest band in the gradient and plotted against density. A representative experiment is depicted. (B) Migration of Gag in VLPs. VLPs were isolated from the medium of transfected COS-1 cells as described in Materials and Methods and fractionated on an Optiprep gradient. The distribution of Gag in the gradient was determined by Western blotting and is shown superimposed upon the profiles of the 0- and 6-h chase points in the graph in panel A. Each experiment was performed twice.

FIG. 3

Time course of Gag localization to the P100 fraction. (A) Cells were transfected with pHXB2ΔBal, pulse-labeled for 2, 5, 10, or 20 min, and subjected to hypotonic lysis and P100 (P) and S100 (S) fractionation. The distribution of radiolabeled Gag at each time point was visualized by immunoprecipitation, SDS-PAGE, and autoradiography. A representative experiment is shown. (B) Graphic depiction of the time course of Gag localization to P100 fractions. Gag was expressed from two constructs: pHXB2ΔBal and pCMVGag, which does not encode other HIV-1 proteins. The quantitations of four experiments were averaged and are shown plotted alongside controls for cytosolic (β-Gal) and membrane-bound (gp120/160^env) proteins. The control proteins were detected by Western blotting. (C) The same analysis was performed on two constructs expressing 31 or 69 amino acids of N-terminal Gag sequence fused to GFP.

FIG. 4

Trypsin protection assay of Gag at various times after synthesis. Denucleated S1 fractions were prepared from transfected cells after pulse labeling and incubation in chase medium for 0, 10, or 60 min. The samples were either untreated (lanes 1 and 2) or treated with trypsin in the absence (−) (lanes 3 and 4) or presence (+) (lanes 5 and 6) of a protease inhibitor cocktail (prot. inhib's.). Samples were treated with trypsin in parallel in the presence of 0.2% NP-40 (lanes 7 and 8). The amounts of total Gag (top panel) and radiolabeled Gag (bottom three panels) in the products were determined by Western blotting (W.B.) and autoradiography, respectively.

FIG. 5

Flotation analysis of Gag protein. (A) Transfected cells were subjected to pulse-labeling for 7 min and chase in the presence of excess Met-Cys. Denucleated P100 fractions were prepared, resuspended in 80% sucrose, overlaid with 65 and 10% sucrose, and centrifuged at 200,000 × g for 2 h. The distribution of labeled Gag was analyzed by autoradiography of the gradient fractions. The positions of the bottom of the tube and the 65%/10% interface are indicated. (B) The distribution of labeled Gag in the flotation is shown alongside the distribution of total Gag, as determined by Western blotting. Internal controls for cytosolic (β-Gal) and membrane-bound (gp120/160^env) proteins are also shown. Analyses of the newly synthesized and total proteins were performed at least four times. The 10- and 30-min chases were performed once.

FIG. 6

Migration of Gag in the absence or presence of detergent in sucrose and Optiprep gradients. (A) Denucleated S1 fractions were prepared from transfected cells, layered over 16 to 60% sucrose gradients after no treatment (top) or treatment with 1% NP-40 (bottom), and centrifuged for 16 h at 100,000 × g. Gag was detected in the gradient fractions by Western blotting. The far left band is the gradient pellet. (B) Denucleated P100 fractions were layered over 0 to 18% Optiprep gradients after no treatment (top) or addition of 1% NP-40 (bottom), centrifuged for 3 h at 100,000 × g, and analyzed for Gag protein. (C) The experiment described for panel B was performed on Jurkat T cells transfected with pHXB2ΔBal.

FIG. 7

Migration of Gag on Optiprep gradients in the absence or presence of detergent. (A) (Top two rows) Transfected cells were pulse-labeled (7 min) and fractionated into S100 and denucleated P100 fractions. Each fraction was fractionated on Optiprep gradients as described in the legend to Fig. 6B in the absence or presence of 1% NP-40. Labeled protein in the gradient fractions was visualized by autoradiography. (Bottom row) The cells were pulse labeled and incubated for 4 h in chase medium. A denucleated P100 fraction was prepared and analyzed on Optiprep gradients. (B) Optiprep gradient analysis was performed on a denucleated P100 fraction from transfected cells as described in the legend to Fig. 1. The fractions corresponding to a density of 1.098 to 1.106 g/ml (brackets) were pooled, treated with 0.1% SDS (left) or 1% NP-40 (right), and fractionated on Optiprep gradients. Gag was detected in the gradient fractions by Western blotting.

FIG. 8

Kinetics of disappearance of Gag from cells and its appearance in VLPs. COS-1 cells were pulse-labeled and incubated in chase medium for various lengths of time. The amount of Gag remaining in the cells and in extracellular VLPs was determined by immunoprecipitation and autoradiography. (A) A representative experiment. Medium from twice the number of cells was harvested when producing VLPs. (B) Quantitation of the experiment shown in panel A. (C) Graph of the 2- to 6-h time points from panel B plotted in isolation. (D) The same experiment was repeated with Jurkat T cells transfected with pHXB2ΔBal. (E) The experiment described in panels A to C was repeated, this time retaining the RIPA-insoluble pellet from the clarification step. The pellet was extensively sonicated and immunoprecipitated with anti-CA antibody in parallel with the cell-associated RIPA-soluble fraction.

FIG. 8

Kinetics of disappearance of Gag from cells and its appearance in VLPs. COS-1 cells were pulse-labeled and incubated in chase medium for various lengths of time. The amount of Gag remaining in the cells and in extracellular VLPs was determined by immunoprecipitation and autoradiography. (A) A representative experiment. Medium from twice the number of cells was harvested when producing VLPs. (B) Quantitation of the experiment shown in panel A. (C) Graph of the 2- to 6-h time points from panel B plotted in isolation. (D) The same experiment was repeated with Jurkat T cells transfected with pHXB2ΔBal. (E) The experiment described in panels A to C was repeated, this time retaining the RIPA-insoluble pellet from the clarification step. The pellet was extensively sonicated and immunoprecipitated with anti-CA antibody in parallel with the cell-associated RIPA-soluble fraction.