Maintenance of the Gag/Gag-Pol ratio is important for human immunodeficiency virus type 1 RNA dimerization and viral infectivity

Abstract

Production of the human immunodeficiency virus type 1 (HIV-1) Gag-Pol precursor protein results from a -1 ribosomal frameshifting event. In infected cells, this generates Gag and Gag-Pol in a ratio that is estimated to be 20:1, a ratio that is conserved among retroviruses. To examine the impact of this ratio on HIV-1 replication and viral assembly, we altered the Gag/Gag-Pol ratio in virus-producing cells by cotransfecting HIV-1 proviral DNA with an HIV-1 Gag-Pol expression vector. Two versions of the Gag-Pol expression vector were used; one contains an active protease [PR(+)], and the other contains an inactive protease [PR(-)]. In an attempt to produce viral particles with Gag/Gag-Pol ratios ranging from 20:21 to 20:1 (wild type), 293T cells were cotransfected with various ratios of wild-type proviral DNA and proviral DNA from either Gag-Pol expression vector. Viral particles derived from cells with altered Gag/Gag-Pol ratios via overexpression of PR(-) Gag-Pol showed a ratio-dependent defect in their virion protein profiles. However, the defects in virion infectivity were independent of the nature of the Gag-Pol expression vector, i.e., PR(+) or PR(-). Based on equivalent input of reverse transcriptase activity, we estimated that HIV-1 infectivity was reduced 250- to 1,000-fold when the Gag/Gag-Pol ratio in the virion-producing cells was altered from 20:1 to 20:21. Although virion RNA packaging was not affected by altering Gag/Gag-Pol ratios, changing the ratio from 20:1 to 20:21 progressively reduced virion RNA dimer stability. The impact of the Gag/Gag-Pol ratio on virion RNA dimerization was amplified when the Gag-Pol PR(-) expression vector was expressed in virion-producing cells. Virions produced from cells expressing Gag and Gag-Pol PR(-) in a 20:21 ratio contained mainly monomeric RNA. Our observations provide the first direct evidence that, in addition to proteolytic processing, the ratio of Gag/Gag-Pol proteins is also important for RNA dimerization and that stable RNA dimers are not required for encapsidation of genomic RNA in HIV-1.

FIG. 1

Schematic representation of proviral DNA constructs used in this study. The wild-type (WT) plasmid is HxB2-BH10 and has been previously described (43). To obtain the GP PR(+) construct, a PCR-amplified fragment was cloned into the HxB2-BH10 proviral DNA via ApaI and BclI restriction sites in order to eliminate the five-T heptanucleotide stretch which is responsible for the −1 ribosomal frameshifting during the translation of Gag and allow continuous expression of Gag-Pol by bypassing the Gag termination codon. The GP PR(−) plasmid was constructed in the same fashion as GP PR(+), with the exception that a PR-defective full-length HIV-1 PR(−) plasmid was used as the DNA template for GP PR(−) PCR mutagenesis (15).

FIG. 2

Impact of altered Gag/Gag-Pol ratios on HIV-1 protein profiles of infected cells and virions. Viral proteins were semiquantified by Western blot analysis. Cell lysates (A) and purified virions (see Materials and Methods) (B) were resolved by SDS–10% PAGE. Resolved proteins were probed using sera from HIV-1-infected individuals and monoclonal antibodies (Ab) to RT and p24 as described in Materials and Methods. Lane 1 (A and B) shows wild-type (WT) HIV-1 protein profiles in cell lysates and virions, respectively; lanes 2 to 5 (A and B) show cell lysate and virion protein profiles with estimated HIV-1 Gag/Gag-Pol ratios of 20:21, 20:11, 20:6, and 20:3.5, respectively. A GP PR(+) expression vector was used to produce excessive levels of Gag-Pol in lanes 2 to 5 (see Materials and Methods). Similarly, lanes 6 to 9 show cell lysate and virion protein profiles with estimated HIV-1 Gag/Gag-Pol ratios of 20:21, 20:11, 20:6, and 20:3.5, respectively. Expression vector GP PR(−) was used to produce an excess of Gag-Pol protein that is not processed by PR in lanes 6 to 9 (see Materials and Methods). MA, matrix.

FIG. 3

Infectivity of viral particles produced by cells with a wild-type or altered Gag/Gag-Pol ratio. Freshly isolated PBMCs were phytohemagglutinin stimulated for 3 days and then infected with either wild-type or mutant virus. Supernatants were collected 3, 7, 10, and 14 days (d) after infection, and the RT activity in each sample was measured. The results shown represent the mean and standard deviation of triplicate samples. Open symbols represent viruses derived from cells with an excess of PR(+) Gag-Pol, and closed symbols (except asterisks) represent viruses derived from cells with excess of PR(−) Gag-Pol. Symbols: ∗, wild-type virus; ⧫ and ◊, progeny viruses of cells with a predicted 20:3.5 Gag/Gag-Pol ratio; ▴ and ▵, predicted 20:6 Gag/Gag-Pol ratio; ■ and □, predicted 20:11 Gag/Gag-Pol ratio; ● and ○, predicted 20:21 Gag/Gag-Pol ratio.

FIG. 4

Effect of altered Gag/Gag-Pol ratios on virion RNA dimerization. The impact of various Gag/Gag-Pol ratios on genomic RNA dimerization was determined using melting curve and electrophoretic analysis of wild-type (WT) and mutant dimers. Virion RNA was resuspended in RNA dimerization buffer and heat denatured for 10 min at the indicated temperatures. Dimers and monomers were electrophoresed in a 1% native agarose gel and probed with an HIV-1 riboprobe as described in Materials and Methods. RNA dimerization analysis involved wild-type HIV-1 and mutant viruses derived from cells cotransfected with wild-type and GP PR(+) expression vectors to achieve Gag/Gag-Pol ratios of 20:1 to 20:21 (A), cells cotransfected with wild-type and GP PR(−) expression vectors to achieve Gag/Gag-Pol ratios of 20:3.5 to 20:21 (B), and cells transfected with wild-type proviral DNA, transfected with a PR-defective full-length HIV-1 Pr-defective plasmid [PR(−)] for the production of PR-defective immature HIV-1 particles, and cotransfected with wild-type GP PR(−) to achieve a Gag/Gag-Pol ratio of 20:21 (C). U, unheated.

FIG. 5

Effect of altered Gag/Gag-Pol ratios on packaging of genomic RNA. Three types of HIV-1 virions were used in this analysis; they were wild-type HIV-1 (WT), Pr-defective immature HIV-1 [PR(−)], and virions containing mainly monomeric RNA [WT:GP PR(−) (20:21)]. Supernatant collected from cells expressing only Gag-Pol precursor protein was also used as a control. Quantitation of pelleted viral protein was performed by protein dot immunoblot assay (A). Samples for RNA analysis (B and C) were standardized for protein concentration (A). Virion RNA encapsidation efficiency of wild-type and mutant viruses was analyzed by RNA dot blot hybridization assay (B). A series of 10-fold dilutions was done for quantitation. The stability of the virion-packaged RNA genome was determined by Northern blot analysis (see Materials and Methods for details of procedures) (C).