Mechanistic differences between nucleic acid chaperone activities of the Gag proteins of Rous sarcoma virus and human immunodeficiency virus type 1 are attributed to the MA domain

Abstract

Host cell tRNAs are recruited for use as primers to initiate reverse transcription in retroviruses. Human immunodeficiency virus type 1 (HIV-1) uses tRNA(Lys3) as the replication primer, whereas Rous sarcoma virus (RSV) uses tRNA(Trp). The nucleic acid (NA) chaperone function of the nucleocapsid (NC) domain of HIV-1 Gag is responsible for annealing tRNA(Lys3) to the genomic RNA (gRNA) primer binding site (PBS). Compared to HIV-1, little is known about the chaperone activity of RSV Gag. In this work, using purified RSV Gag containing an N-terminal His tag and a deletion of the majority of the protease domain (H6.Gag.3h), gel shift assays were used to monitor the annealing of tRNA(Trp) to a PBS-containing RSV RNA. Here, we show that similar to HIV-1 Gag lacking the p6 domain (GagΔp6), RSV H6.Gag.3h is a more efficient chaperone on a molar basis than NC; however, in contrast to the HIV-1 system, both RSV H6.Gag.3h and NC have comparable annealing rates at protein saturation. The NC domain of RSV H6.Gag.3h is required for annealing, whereas deletion of the matrix (MA) domain, which stimulates the rate of HIV-1 GagΔp6 annealing, has little effect on RSV H6.Gag.3h chaperone function. Competition assays confirmed that RSV MA binds inositol phosphates (IPs), but in contrast to HIV-1 GagΔp6, IPs do not stimulate RSV H6.Gag.3h chaperone activity unless the MA domain is replaced with HIV-1 MA. We conclude that differences in the MA domains are primarily responsible for mechanistic differences in RSV and HIV-1 Gag NA chaperone function. Importance: Mounting evidence suggests that the Gag polyprotein is responsible for annealing primer tRNAs to the PBS to initiate reverse transcription in retroviruses, but only HIV-1 Gag chaperone activity has been demonstrated in vitro. Understanding RSV Gag's NA chaperone function will allow us to determine whether there is a common mechanism among retroviruses. This report shows for the first time that full-length RSV Gag lacking the protease domain is a highly efficient NA chaperone in vitro, and NC is required for this activity. In contrast to results obtained for HIV-1 Gag, due to the weak nucleic acid binding affinity of the RSV MA domain, inositol phosphates do not regulate RSV Gag-facilitated tRNA annealing despite the fact that they bind to MA. These studies provide insight into the viral regulation of tRNA primer annealing, which is a potential target for antiretroviral therapy.

FIG 1

(A) Predicted secondary structures of unmodified bovine tRNA^Trp, RSV PBS, and RSV 60-mer. The circled nucleotides differ between bovine (U16, U47) and chicken tRNA^Trp (C16, C47). The PBS-containing structure shown is derived from positions 56 to 130 of the RSV genome, which is part of the 180-mer PBS construct used in this work (see Materials and Methods). Shown in red and blue are two sets of complementary sequences that base pair during tRNA primer annealing as described previously (6). The RSV 60-mer construct is derived from nt 1300 to 1360 of the RSV genome, with nt 1300 changed from T to G to facilitate in vitro transcription. (B) WT RSV and HIV-1 Gag constructs and variants used in this work. All RSV Gag variants are derived from the previously described genomic sequence of pRC.V8 (45). All HIV-1 Gag variants are derived from HIV-1 isolate BH10. Amino acid residue numbers are shown above constructs. In chimeras, the color indicates whether the domain is from RSV (red, green, blue, and black) or HIV-1 Gag (orange). H6.H32R.3h is a previously described chimera in which the first 10 amino acids of RSV MA are replaced with the first 32 amino acids of HIV-1 MA (42). H6.H132R.3h consists of the entire HIV-1 MA domain fused to RSV ΔMA.3h. The amino acids flanking the junction region of each chimera are shown explicitly below each construct. In the core of H6.H132R.3h, an extra D residue is present at the junction (black).

FIG 2

(A) Single-time-point (30 min) gel-shift annealing assays comparing RSV H6.Gag.3h and NCp12 in the presence of the indicated concentration of NaCl. (B) Annealing time course assays for RSV H6.Gag.3h and NCp12 using 3 μM protein in 50 mM NaCl. The curves are single-exponential fits to the averages from three or more trials with standard deviations indicated.

FIG 3

(A) Single-time-point (30 min) gel-shift annealing assays comparing RSV H6.Gag.3h, H6.GagΔNCΔSP, and MA. (B) Annealing time course assays comparing H6.Gag.3h and H6.ΔMA.3h using 3 μM protein. The curves are single-exponential fits to the averages from three or more trials with standard deviations indicated.

FIG 4

Annealing time course assays for HIV-1 GagΔp6 (A), RSV H6.Gag.3h (B), H6.H32R.3h (C), and H6.H132R.3h (D) using 0.8 μM protein in the absence or presence of 5 μM IP6. The curves are single-exponential fits to the averages from three or more trials with standard deviations indicated.

FIG 5

Fluorescence measurements to monitor binding of 3′-FITC-RSV 60 to RSV and HIV-1 Gag variants. (A) FA binding assays for RSV Gag variants. (B) Binding of RSV MA performed at higher protein concentrations. The inset shows the same data plotted on a linear rather than log 10 scale. FA binding assays were performed with HIV-1 GagΔp6 (C), HIV-1 H6.MA (D), and RSV/HIV-1 Gag chimeras (E and F). The FA data for RSV Gag variants, HIV-1 GagΔp6, and the chimeras were fit to 1:1 binding curves as described previously (35). The fluorescence intensity change of HIV-1 H6.MA was fit as previously described (20). All graphs represent the averages from three or more trials with standard deviations indicated.

FIG 6

FA measurements to monitor competition of IP6 with 3′-FITC-RNA 60 for binding to RSV H6.Gag.3h and RSV MA. Graphs represent the averages from three or more trials with standard deviations indicated.