The HIV-1 Nucleocapsid Regulates Its Own Condensation by Phase-Separated Activity-Enhancing Sequestration of the Viral Protease during Maturation

Abstract

A growing number of studies indicate that mRNAs and long ncRNAs can affect protein populations by assembling dynamic ribonucleoprotein (RNP) granules. These phase-separated molecular 'sponges', stabilized by quinary (transient and weak) interactions, control proteins involved in numerous biological functions. Retroviruses such as HIV-1 form by self-assembly when their genomic RNA (gRNA) traps Gag and GagPol polyprotein precursors. Infectivity requires extracellular budding of the particle followed by maturation, an ordered processing of ∼2400 Gag and ∼120 GagPol by the viral protease (PR). This leads to a condensed gRNA-NCp7 nucleocapsid and a CAp24-self-assembled capsid surrounding the RNP. The choreography by which all of these components dynamically interact during virus maturation is one of the missing milestones to fully depict the HIV life cycle. Here, we describe how HIV-1 has evolved a dynamic RNP granule with successive weak-strong-moderate quinary NC-gRNA networks during the sequential processing of the GagNC domain. We also reveal two palindromic RNA-binding triads on NC, KxxFxxQ and QxxFxxK, that provide quinary NC-gRNA interactions. Consequently, the nucleocapsid complex appears properly aggregated for capsid reassembly and reverse transcription, mandatory processes for viral infectivity. We show that PR is sequestered within this RNP and drives its maturation/condensation within minutes, this process being most effective at the end of budding. We anticipate such findings will stimulate further investigations of quinary interactions and emergent mechanisms in crowded environments throughout the wide and growing array of RNP granules.

Keywords: HIV-1; RNA; atomic-force microscopy; biomolecular condensates; enzyme catalysis; liquid–liquid phase separation; molecular dynamics; nucleocapsid; protease.

Figure 1

Quinary interactions and architectural behaviour of NC:NA NP complexes upon PR processing. (A) Sequence of NCp15 and PR cleavage sites (NC/SP2 in red, SP2/P6 in magenta). (B) Electrophoretic migration on agarose gel of M13 ssDNA:NCp7 NP complexes. Serial dilutions of both ssDNA (expressed in total [nt]) and NCp7, keeping a constant ratio RNC/nt = 1/10, were incubated for 30 min at 37 $${}^{\circ }$$C before electrophoresis at 4 $${}^{\circ }$$C and DNA staining. DNA smears; hallmarks of ssDNA/NCp7 aggregates (lanes 2–4) are converted into fast migrating ssDNA/NCp7 condensates (lanes 5–6) upon dilution. Position of circular ssDNA and traces of linear ssDNA are indicated. (C) Comparison of the electrophoretic mobility in 1% agarose of the circular M13 ssDNA in complex with NCp15, NCp9, and NCp7 upon increasing protein concentration. RNC/nt indicate the protein/nucleotide ratio. (D) AFM imaging of M13 ssDNA (0.4 nM) incubated from left to right without protein, with NCp15, NCp9, and NCp7 at RNC/nt = 1/10. (E,F) Sequential proteolysis of M13 ssDNA-bound NCp15 followed by EMSA, SDS-PAGE, and AFM. ssDNA (1 nM):NCp15 (750 nM) complexes were assembled at pH 6.25 and subsequently cleaved by PR (35 nM) at the indicated times at 37 $${}^{\circ }$$C. Each reaction was stopped by chilling the tubes on ice, and the tube contents were used to follow ssDNA:NC complex migration by EMSA (E), the maturation state of NC proteins by SDS-PAGE (inserts in F), and ssDNA:NC complex morphology by AFM (F) at t = 30 s, 8 min and 16 min. In (E), NC:NA complex migration controls for NCp15, NCp9, and NCp7 are shown at pH 7.4 and pH 6.25 (optimal for PR cleavage).

Figure 2

NC:NA quinary interactions. (A) The N-terminal 3$${}_{10}$$-helix and the NC-SP2 cleavage site contain DNA strand bridging motifs. The Wenxiang diagram shows the residues of the N-terminal 3$${}_{10}$$-helix in contact with SL3-RNA (red) or SL2-RNA (blue). The diagram of the 53–60 sequence shows the Q53-F56-G58-K59 motif mirroring the K3-G4-F6-Q9 motif in contact with SL2. The schematics indicate possible NA-NC-NA network assemblies provided by such NA binding domains. (B) Superposition of the N-terminal domain backbone of NCp7 in complex with SL3 (PDB ID: 1A1T) and SL2 (PDB ID: 1F6U). (C) Potential of mean force (PMF) plot in the 2D reaction coordinate space of the K59-Q53 C_α distance (d_KQ) and K59-F56-Q53 C_α angle (θ_FQK) order parameters. The NC-SP2 apo-octapeptide exhibits substantial flexibility and is energetically dominated (PMF ∼ 5–7 kcal/mol) by a turn-like structure (d_KQ ∼ 4–6 Å, θ_KFQ ∼ 20°–40°). The conformational ensemble also samples a region (d_KQ ∼ 10–12 Å, θ_KFQ ∼ 140°–180°) with less frequency (PMF ∼1–3 kcal/mol) that is compatible with a 3₁₀ helical structure and where NMR conformers of NCp7 from 1F6U (grey circles) are located. For NMR conformers, d_KQ corresponds to the K3-Q10 C_α distance and θ_FQK to the K3-F6-Q9 C_α angle. (D) Structure of best-fitting conformer of the R52–K59 segment superimposed head-to-tail with the N-terminal domain of NCp7. (E) Probability distribution of the conformer ensemble C_α root-mean-squared deviation (RMSD) of the K59-R52 residues aligned to the K3-R10 residues of a single structure of the N-terminal 3₁₀ helix in 1F6U. Three conformer sub-populations within an RMSD of 2 Å, 1.5 Å and 1 Å were extracted. Mapping the conformers onto the d_KQ–θ_KFQ order parameters (blue circles) show they occupy the same conformational sub-space. (F) Electrophoretic migration in agarose of M13 ssDNA (circular and linear forms) or MS2 RNA in complex with increasing concentrations of a K3AF6AQ9A NCp7 mutant. (G,H) AFM imaging of wt-NCp7 and K3AF6AQ9A NCp7 mutant incubated with M13 ssDNA in the absence of magnesium. (I) DLS analysis of NCp7 (green), NCp9 (blue), and K3AF6AQ9A NCp7 mutant (orange) in complex with M13 ssDNA (black) for increasing NC:nt ratio in the absence of magnesium. Experiments with free and bound-ssDNA are reported on the same graph for clarity.

Figure 3

Nucleocapsid maturation is promoted by NA through sequestration of PR. (A) In vitro proteolysis of SP1-NC, SP2-p6, and NCp7-SP2 cleavage sites by PR in absence (black) or presence (red) of M13 ssDNA or MS2 RNA under unfavourable conditions (pH 6.25 and 0.1 M NaCl). NC concentrations were fixed at 6 $$\mathsf{\mu }$$M, PR at 600 nM, and NA at 120 $$\mathsf{\mu }$$M (nucleotide), unless otherwise indicated. In absence of NA, PR optimum is found at pH 5.0 in presence of 1.5 M NaCl. The assays, which used peptide fluorescence quantification after SDS-PAGE separation of the cleavage products, are described in Supplementary Figure S3a,b. (B) Comparison of NCp15 cleavage by PR in presence of MS2 RNA, NC-specific HIV RNA stem-loops (SL3, TAR), HIV gRNA fragments (1–615, 1–61), and a d(A)13 oligonucleotide, a weak NC substrate. The RNC/nt was fixed at 1/20 for each experiment. NCp15 cleavage rate was calculated as a function of RNA length (SL1 (17 nt), SL2 (23 nt), SL3 (14 nt), SL4 (24 nt), and HIV gRNAs fragments 1–61, 1–102, 1–152, 1–278, 1–311, 1–615, 1–1333, and 1–4001 are reported in the right panel). (C) Cleavage of a DABCYL-MA-CA-EDANS peptide probe (5.2 $$\mathsf{\mu }$$M) by PR (50 nM) at 30 $${}^{\circ }$$C followed by FRET. The reactions were performed with PR alone at pH 5.5 in presence of 0.1 M NaCl, PR and M13 ssDNA (13.4 nM), and PR + M13 ssDNA (13.4 nM) + NCp15 (5 $$\mathsf{\mu }$$M). (D) Delay of the FRET probe cleavage extrapolated from (C) as a function of M13 ssDNA:NCp15 complex concentration, with RNCp15/nt = 1/20.

Figure 4

Theoretical model of RNP-modulated enzyme-substrate reaction kinetics. (A) Reaction rate is governed by a combination of effective concentrations of enzyme (E) and two substrates (S1 and S2) in each of two volume domains, pervaded (P) and unpervaded (U) by NA, as well as the absorption equilibrium for different species. S1 is a competitive substrate, S2 is a component of the RNP. For K$${}_E$$≫ 1, E (in this case, PR) is sequestered into the NA-pervaded volume, whilst for K$${}_E$$≪ 1, its pervaded concentration is reduced. (B) A one-substrate rate model (S2 only) is fit to experimental data to determine non-linear K$${}_E$$-dependence (exponent $$\xi$$) on the contiguous number c of S2 (in this case, NCp15) molecules bound per NA, when above a critical threshold, $$c_{crit}$$. (C) Fitted two-substrate kinetic model of NCp15 and competitive substrate (MA-CA) with NA (black), MA-CA only without NA (red) and MA-CA only with NA (brown) data. For the NCp15-containing system, the early reaction is dominated by high contiguity (c > $$c_{crit}$$, K$${}_E$$≫ 1), inducing enzyme (PR) sequestration. This effect dissipates upon processing (c < $$c_{crit}$$, K$${}_E$$≪ 1). A total of 90% of NCp15 cleavage (dashed black line) is calculated from the model to occur after ∼400 s. (D) Differential enzyme decay: The presence of NA stabilizes the PR dimer (black and brown); in the NA-absent reaction, PR decays in the model with an experimental half-life (black). (E) Scaling to in virio conditions yields 90% (dotted purple line) NCp15 processing within 260 s (solid black line). The dotted black line represents the 100% processing level.

Figure 5

Nucleocapsid condensation within HIV-1 particles depends on NCp15 processing and is detectable in membrane-attached particles. (A) TEM images of purified HIV-1 NL4-3 virions accumulating NCp15 (uncleavable p6 and SP2 sites), NCp9 (uncleavable SP2), or wt-NCp7. NCp15-containing particles present defects in nucleocapsid condensation, while NCp9- and NCp7-containing viruses show correct core condensation into an electron-dense dark spot. Quantitation was done for 180 counted particles. (B,C) TEM images of latently infected ACH-2 cells producing viral particles at the plasma membrane after 48 h activation by Vorinostat. The majority of membrane-attached HIV-1 particles produced by latently infected ACH-2 cells are immature particles in the presence of LPV, a PR inhibitor (B). In the absence of LPV, the particles contain an electron-dense dark spot indicative of nucleocapsid condensation (C). Bottlenecks characterizing budding intermediates are pointed to by arrows. (D) Quantitation of attached and free particles (top) and particles containing a condensed RNP (bottom), as noted by a dark spot, in the presence or absence of LPV. Counting was performed for 200 particles for LPV-treated ACH2 cells and 500 particles for non-treated ACH2 cells.

Figure 6

Nucleocapsid condensation, quinary interactions and WSM transition of the nucleocapsid during HIV maturation. (A) Weak-strong-moderate (WSM) quinary properties of NC proteins with RNA (white) throughout virus maturation. p6 (orange)-NCp7 (green) contacts assemble NCp15 networks upon the RNA, blocking exposed residues engaged in NCp9 or NCp7: NA contacts. Cleavage of p6 unmasks the NCp9 domain, and the NCp7-SP2 (yellow) domain engages NC:NA:NC and NA:NC:NA networks favouring strong RNA-mediated quinary interactions and RNP granulation. Cleavage of SP2 removes one “K/F/Q” NA binding patch (see Figure 2A), which reduces the quinary network between RNA and NCp7. (B) Nucleocapsid WSM transition in the context of a virus particle. (1) Virus particles at the plasma membrane bud from self-assembly of Gag and GagPol on gRNA, with the NC domain as the RNA binder. The gRNA is shown in white within the layer of NC domains, assembled as bundles of six at the base of Gag hexamers. (2) The dimerization of GagPol self-activates PR (pink) and initiates maturation. (3) The SP1-NC site is cleaved, liberating the gRNA:NCp15 RNP, which in turns sequesters PR. (4) Rapid cleavage of NCp15 into NCp9, which liberates p6, unlocks the strong quinary properties of NCp9. This quickly compacts the gRNA and favours viral budding. gRNA condensation allows internal reorganization of RT (red) and IN (gold), self-assembly of the conical capsid after separation of MA from CA-SP1, and maturation of CA-SP1, while NCp9 are matured to NCp7 in (5 and 6).