The hantavirus nucleocapsid protein recognizes specific features of the viral RNA panhandle and is altered in conformation upon RNA binding

Abstract

Hantaviruses are tripartite negative-sense RNA viruses and members of the Bunyaviridae family. The nucleocapsid (N) protein is the principal structural component of the viral capsid. N forms a stable trimer that specifically recognizes the panhandle structure formed by the viral RNA termini. We used trimeric glutathione S-transferase (GST)-N protein and small RNA panhandles to examine the requirements for specific recognition by Sin Nombre hantavirus N. Trimeric GST-N recognizes the panhandles of the three viral RNAs (S, M, and L) with high affinity, whereas the corresponding plus-strand panhandles of the complementary RNA are recognized with lower affinity. Based on analysis of nucleotide substitutions that alter either the higher-order structure of the panhandle or the primary sequence of the panhandle, both secondary structure and primary sequence are necessary for stable interaction with N. A panhandle 23 nucleotides long is necessary and sufficient for high-affinity binding by N, and stoichiometry calculations indicate that a single N trimer interacts with a single panhandle. Surprisingly, displacement of the panhandle structure away from the terminus does not eliminate recognition by N. The binding of N to the panhandle is an entropy-driven process resulting in initial stable N-RNA interaction followed by a conformational change in N. Taken together, these data provide insight into the molecular events that take place during interaction of N with the panhandle and suggest that specific high-affinity interaction between an RNA binding domain of trimeric N and the panhandle is required for encapsidation of the three viral RNAs.

FIG.1.

RNA binding profile of trimeric GST-N with the viral and complementary panhandles corresponding to the S, M, and L segments. (A) The nucleotide sequence and highly probable secondary structure of each of the RNAs is shown at the left. For each of the panhandle sequences, the nucleotides common to the both the complementary RNA (cRNA) and its antecedent viral RNA (vRNA) are shown in grey, while the nucleotides of the complementary RNA that vary relative to the viral RNA are shown in black. To generate viral RNA (randomized), the 5′ nucleotides of the S segment RNA were randomized and the 3′ nucleotides were chosen such that the secondary structure of the RNA would be the same as that of the viral RNA panhandle. Nucleotides that vary relative to those of the viral RNA panhandle are shown in black. (B) The RNA binding profiles as a function of increasing purified trimeric GST-N are shown in the right part of the figure. Increasing concentrations of purified trimeric GST-N were incubated with a constant concentration of ³²P-labeled RNA and protein-RNA complexes were isolated by filter binding as described in the Materials and Methods section. The data are plotted to show the binding profile at relatively high concentrations of trimeric GST-N and at 80 mM salt to display binding to complementary RNA and viral RNA (randomized). To determine the dissociation constants for the viral RNAs, the data were replotted to optimize visualization and analysis at lower trimeric GST-N concentrations. The data shown are from one experiment. However, each of the RNAs was analyzed in three separate experiments, and the measured K_ds with standard deviations are shown in Table 1.

FIG.2.

Nucleotide variants in the S segment RNA panhandle. RNAs were synthesized as described in the Materials and Methods section. (A) Probable secondary structures of a set of RNAs containing nucleotide substitutions. Substitutions are shown in black. The RNAs shown in B are secondary structures of S segment viral RNA (vRNA) panhandles of decreasing length, as indicated. Panhandle RNAs containing additional nucleotides at the 5′ or 3′ terminus or both are shown in C. Additional nucleotides are indicated in black. The likely secondary structure of each of the RNAs was assessed by both P-num analysis and examination of alternative structures (Materials and Methods). The secondary structures of all of the RNAs except RNA 14 are highly probable. The three substitutions in RNA 14 would be expected to substantially disrupt base pairing such that the terminal two nucleotides would be unpaired in a substantial fraction of the RNA population.

FIG. 3.

RNA binding competition analysis of S segment viral RNA variants. As described in detail in the Materials and Methods section and the text, a constant amount of purified trimeric GST-N protein was incubated with a constant amount of labeled RNA corresponding to the wild-type viral RNA panhandle and increasing concentrations of unlabeled competitor RNA. The quantitative reduction of binding of trimeric GST-N to labeled RNA as a function of increasing competitor RNA was measured for each RNA. The figure depicts the competition curves for three of the variant RNAs and for the wild type. The competition curves for the other RNAs in Fig. 2 are not shown. However, the maximum extent of competition for all of the S segment panhandle variants is presented in Table 2. The data in the figure and in Table 2 are the averages of three separate experiments. The standard deviation for the measured values of all of the RNAs is presented in Table 2.

FIG. 4.

Binding stoichiometry of trimeric N-panhandle interaction. The results from binding RNA experiments were replotted to determine stoichiometry as described in the Materials and Methods section. Data for RNA 5, which is bound by trimeric GST-N at high affinity, and RNA 2, which is bound at lower affinity, are shown. The data indicate that three monomers are bound per panhandle. Since the binding experiments were carried out with purified trimeric GST-N, this corresponds to 1 trimer/panhandle. The binding data for all of the RNAs analyzed were similarly plotted and indicated similar stoichiometry independent of binding affinity (not shown).

FIG. 5.

Fluorescence spectroscopic and thermodynamic analysis of N protein trimer and S segment viral RNA panhandle association. (A) As described in the Materials and Methods section, a fixed concentration of N protein trimer in binding buffer was excited at 295 nm, and fluorescence intensity at 330 nm was recorded. The experiment was carried out at 25°C. The fluorescence signal from free binding buffer in the absence of GST-N protein trimer was subtracted where appropriate. The fluorescence intensity of the N protein trimer at 330 nm was monitored at different input concentrations of S segment Sin Nombre virus RNA panhandle. A plot of fluorescence intensity (at 330 nm) as a function of Sin Nombre virus S segment RNA panhandle concentration was then plotted in panel A. The inset shows the fluorescence emission spectra (λ_ex = 295 nm) of N protein trimer alone (▪) and in the presence of S segment viral RNA panhandle at concentrations of 50 nM (○) and 110 nM (□). (B) The N protein bound at different input concentrations of S segment Sin Nombre virus RNA panhandle was calculated from panel A. A plot of the N protein bound at each input concentration of S segment Sin Nombre virus RNA panhandle is shown in panel B. The dissociation constant, corresponding to the concentration of S segment Sin Nombre virus RNA panhandle at which half of the N protein trimer is bound, was calculated from this binding profile (panel B). ln K_app (1/K_d) at three different temperatures was calculated. A plot of ln K_app versus 1/T (van't Hoff plot) is shown in the inset. ΔH and ΔS values were calculated from the van't Hoff plot as described in the Materials and Methods section. a.u., arbitrary units.

FIG. 6.

Stern-Volmer plots of tryptophan fluorescence quenching of N protein trimer by acrylamide. A fixed concentration of trimeric GST-N in binding buffer was excited at 295 nm, and emission was recorded at 330 nm. The fluorescence studies were carried out at room temperature. The fluorescence signal from free binding buffer in absence of GST-N protein trimer was subtracted whenever required. The fluorescence intensity at 330 nm was determined at different input concentrations of acrylamide. Plots of F₀/F as a function of acrylamide concentration (Stern-Volmer plot) for free GST-N protein trimer in absence (□) and presence (○) of RNA of S segment Sin Nombre virus RNA panhandle (panel A), RNA 6 (panel B), and RNA2 (panel C) was used to calculate the Ksv values as described in the Materials and Methods section. The data shown are for viral RNA, RNA 6, and RNA 2 in graphs A, B, and C, respectively.