HIV-1 protease and reverse transcriptase control the architecture of their nucleocapsid partner

Abstract

The HIV-1 nucleocapsid is formed during protease (PR)-directed viral maturation, and is transformed into pre-integration complexes following reverse transcription in the cytoplasm of the infected cell. Here, we report a detailed transmission electron microscopy analysis of the impact of HIV-1 PR and reverse transcriptase (RT) on nucleocapsid plasticity, using in vitro reconstitutions. After binding to nucleic acids, NCp15, a proteolytic intermediate of nucleocapsid protein (NC), was processed at its C-terminus by PR, yielding premature NC (NCp9) followed by mature NC (NCp7), through the consecutive removal of p6 and p1. This allowed NC co-aggregation with its single-stranded nucleic-acid substrate. Examination of these co-aggregates for the ability of RT to catalyse reverse transcription showed an effective synthesis of double-stranded DNA that, remarkably, escaped from the aggregates more efficiently with NCp7 than with NCp9. These data offer a compelling explanation for results from previous virological studies that focused on i) Gag processing leading to nucleocapsid condensation, and ii) the disappearance of NCp7 from the HIV-1 pre-integration complexes. We propose that HIV-1 PR and RT, by controlling the nucleocapsid architecture during the steps of condensation and dismantling, engage in a successive nucleoprotein-remodelling process that spatiotemporally coordinates the pre-integration steps of HIV-1. Finally we suggest that nucleoprotein remodelling mechanisms are common features developed by mobile genetic elements to ensure successful replication.

Figure 1. Proteolysis of NCp15 by PR.

(A) Sequential and ordered proteolysis of the HIV-1 NC domain. Digestion of the Gag polyprotein by PR leads to the mature NCp7 via the NCp15 and NCp9 intermediate forms . (B) Proteolysis of NCp15 (1 µM) at 30°C and pH 6.6 for 2 h. The products shown were generated using three concentrations of recombinant PR (50, 100 and 200 nM) and incubated in the presence of a circular ssDNA (3,352 nt, 1.25 nM), a G4-DNA (250 nM), a 415-mer RNA fragment corresponding to the HIV-1 leader region (20 nM) or a 3,352 dsDNA plasmid (1 nM). Controls: p6, NCp9 and NCp7 are shown in the three far left lanes, whereas products of NCp15 incubation with two concentrations of PR (50 and 200 nM) in the absence of nucleic acid are shown to their right.

Figure 2. Differences in DNA binding between NCp15 and NCp7 and the effect of NCp15 proteolysis.

(A,B) EMSA using G4-DNA (5 nM) and increasing amounts of NCp7 or NCp15. Complexes were formed for 15 min. at 37°C before electrophoresis. Note that in the case of NCp15 there is only one shifted G4-DNA species (A), whereas quantification of the DNA bound fraction as a function of the total NCp concentration (B) shows a 6-fold drop in affinity for G4-DNA with NCp15 (solid line) compared to NCp7 (dashed line). (C,D,E) TEM visualization of ssDNA (5 nM) without protein (C) or with saturating amounts (3.4 µM) of NCp7 (D) or NCp15 (E). (F,G,H) TEM visualization of NCp15 proteolysis in the presence of ssDNA after 15 min. (F and G) and after 40 min. (H) incubation with saturating amounts of NCp bound to ssDNA and a NCp15-to-PR ratio of 30 corresponding to [NCp15] = 3.4 µM, [PR] = 100 nM and [ssDNA] = 5 nM. The nascent aggregates revealed in (F) are shown with a higher magnification in (G). Scale bars in each panel correspond to 150 nm.

Figure 3. Comparison of DNA synthesis by RT in the absence and presence of NCp7.

(A) The progression of DNA synthesis is shown with or without NCp7 for a reaction time of 30 min. as a function of RT concentration. (B) The reaction is shown as a function of time with or without NCp7 for a fixed concentration of RT. The DNA products (B) have been sedimented before electrophoresis. Lower (L) and upper (U) half-volumes collected after sedimentation are indicated underneath. Concentrations of ssDNA and NCp were, respectively, 5 nM, 3.4 µM (A and B) and RT was 50 nM (B). DNA and NCp7 were premixed for 4 min. at 37°C, following incubation with RT for 2 min. before addition of dNTPs (100 µM each) to start the reaction.

Figure 4. Extrusion of dsDNA produced by RT from ssDNA-NCp7 co-aggregates.

Progression of DNA synthesis analyzed by TEM after 2 min. (A), 10 min. (B) and 40 min. (C) from reactions with ssDNA (5 nM), RT (50 nM) and NCp7 (3.4 µM). Disaggregation after 10 min. (B) appeared both at the periphery and within the aggregates. A few individual molecules were visible close to the aggregate after 40 min. (C). (D) A typical DNA product visualized by TEM after 40 min. of DNA synthesis with subsequent incubation for 15 min. at 70°C in the presence of 0.4 M NaCl. (E) Band shift analysis of DNA flap synthesis within the dsDNA produced by RT after 40 min. at 37°C with 50 nM RT, with or without 3.4 µM NCp7. An excess of DraI and AlwnI enzymes that digest this dsDNA into two fragments (1800 and 1500 bp) was added. The DraI-AlwnI digestion products of the plasmid DNA are shown as a control on the left. When the HIV-1 central DNA flap is fully synthesized (i. e. with NCp7), the 1800 bp fragment is shifted to a slower migrating band. Magnification is identical for panels A,B,C. The scale bars correspond to 250 nm.

Figure 5. Hypothetical model of HIV-1 nucleocapsid assembly controlled by PR-directed NCp15 proteolysis.

The HIV-1 immature (A) and mature (E) capsid are depicted, focusing on the nucleoprotein region. This model was inspired from L. Henderson's (NCI-Frederick) illustration, taking into account the ratios of Gag, Gag-pol, NCp, PR, RT, IN and Vpr. (B) Gag cleavage between p2 and NCp15 by PR resulting in RNA-bound NCp15 removal from the capsid domains. (C) and (D) digestion of RNA-bound NCp15 yielding NCp9 (C), followed by NCp7 (D) which yields to an overall aggregation/condensation of the nucleocapsid that leads to (E).

Figure 6. Hypothetical model for HIV-1 nucleocapsid dismantling controlled by RT directed dsDNA synthesis.

(A) translocation of RT on a minus DNA template coated with NCp7. RT displaces NCp7 during this translocation reaction while converting ssDNA into dsDNA, which does not allow NCp7 to re-associate in an aggregative mode. The dismantling force is driven by dNTP hydrolysis that allows continuous RT translocation. Magnesium is critical as it weakens dsDNA-binding by NCp7. (B) macroscopic consequence of the NCp7 dismantling activity of RT. A reverse transcription complex is depicted from a perspective of the nucleocapsid architecture used in figure 5. Once reverse transcription is finished, resulting in dsDNA, the aggregative nucleic acid binding mode of NCp7 is drastically reduced, facilitating the binding of Vpr and IN (as well as other available cellular partners) and thus generating a PIC (represented here in a simplified way according to Nermut et al.[27]).