Structural features and biochemical characterization of SARS-CoV-2 Nsp13.

(A) Domain architecture and structure of Nsp13 from PDB (7nio). (B) SDS-PAGE analysis of purified Nsp13. (C) Size-exclusion chromatography (SEC) profile of Nsp13, indicating a monomeric state in solution. (D) Equilibrium dissociation constants (Kd) of Nsp13 with various nucleic acid substrates. (E) Binding affinity of the ATPase-deficient mutant K288R compared to wild-type Nsp13. (F) ATPase activities of wild-type Nsp13 and K288R mutant in the presence of ssRNA. Data are presented as mean ± SD (n=3). Statistical significance was determined by unpaired Student’s t-test for comparisons between two groups, or one-way ANOVA followed by Tukey’s HSD test for comparisons involving more than two groups; ns, P>0.05, P < 0.05, p < 0.01, P<0.001, P<0.0001 (apply to all figures).

Divalent cations activate ATP-independent DNA unwinding and allosterically stabilize Nsp13.

(A) Schematic of the 5′-overhang DNA substrate with * denoting the FAM label (5′OhS14D16) and EMSA showing Nsp13-mediated, concentration-dependent unwinding in the presence of 5 mM Mg²⁺ without ATP (DNA substrate: 40 nM). Right panel: quantification of unwound DNA fraction. (B) EMSA and quantification illustrating the effect of Mg²⁺ concentration (0–20 mM) on ATP-independent DNA unwinding. (C) EMSA and quantification comparing the activation of DNA unwinding by different divalent cations (Mg²⁺, Ca²⁺, Mn²⁺). (D) CD spectra of Nsp13 in the absence and presence of 5 mM Mg²⁺. (E) Thermal denaturation curves of Nsp13 monitored by CD at 222 nm. (F) AlphaFold3-predicted structural models of the Nsp13–RNA fork complex, highlighting Mg²⁺-induced compaction between the RecA1 and RecA2 domains (RNA was selected for modeling due to its higher prediction confidence relative to DNA). Data are presented as mean ± SD (n=3).

Energy- and substrate-dependent DNA unwinding by Nsp13.

(A–B) EMSA and quantification comparing Nsp13-mediated duplex DNA unwinding under different nucleotide and Mg²⁺ conditions. K288R was included as a control to assess the contribution of ATP hydrolysis. (C) Schematic of the 24-bp DNA substrate unwound only in the presence of Mg2+. (D) Blunt-ended duplex DNA cannot be unwound by Nsp13 under either ATP-independent or ATP-dependent conditions. (E) Schematic of a 3′-overhang DNA substrate (3′OhD12S14) and EMSA showing Nsp13-mediated unwinding in the presence of 5 mM Mg²⁺, either in the absence or presence of 2 mM ATP. (F) Quantification of unwound DNA fraction. Data represent mean ± SD (n = 3).

Mg²⁺ activates ATP-independent and -dependent RNA unwinding with unique sensitivity and concentration dependence.

(A) Schematic of the RNA-fork substrate and EMSA showing RNA unwinding at 0.5 mM Mg²⁺ without ATP. (B) Quantification of unwinding efficiency by 2 µM Nsp13 at 0, 0.5, and 1 mM Mg²⁺. (C-D) EMSA and quantification of RNA unwinding under different ATP conditions. (E-F) Comparison of wild-type Nsp13 and K288R mutant unwinding at low protein concentration (80 nM). (G-H) EMSA and quantification showing triphasic concentration dependence of ATP-dependent RNA unwinding at 1 mM Mg²⁺. RNA substrate concentration: 40 nM. Data are presented as mean ± SD (n=3).

Nsp13 unwinds G4 structures through dual ATP-dependent and -independent pathways.

(A) Schematic of the G4 unfolding assay. (B–C) EMSA showing ATP-independent and ATP-dependent unfolding of blunt-ended DG4 (30 nM). (D) Quantification of DG4 unfolding efficiency. (E–F) EMSA showing direction-dependent unwinding of 5′- and 3′-overhang DG4 (25 nM) substrates. (G) Quantification confirming 5′→3′ directionality. (H–I) EMSA showing ATP-independent and ATP-dependent unfolding of SARS-CoV-2 RNA G4 (RG-1, 60 nM). (J) Quantification of RG-1 unfolding efficiency.

Nsp13 exhibits ATP/Mg²⁺-independent nucleic acid annealing activity.

(A) EMSA of DNA strand annealing by Nsp13 under four cofactor conditions: no Mg²⁺/ATP, Mg²⁺ only, ATP only, and Mg²⁺+ATP. (B) Concentration-response curve showing biphasic DNA annealing. (C) Quantification of DNA annealing efficiency at 200 nM Nsp13. (D) EMSA of RNA strand annealing under three cofactor conditions: no Mg²⁺/ATP, ATP only, and Mg²⁺+ATP. The Mg²⁺-only condition was omitted because 1 mM Mg²⁺ promotes spontaneous hybridization of the single-stranded RNA substrates. (E) Concentration-response curve of RNA annealing. (F) Quantification of RNA annealing efficiency at 200 nM Nsp13. Data are presented as mean ± SD (n=3).

An integrated functional model of Nsp13 as a tunable nucleic acid remodeler.

Nsp13 integrates key inputs—cofactors (ATP/Mg²⁺), enzyme concentration, and substrate topology—through a central regulatory hub where Mg²⁺ or Mg²⁺-nucleotide complex allosterically stabilizes its compact RecA1–RecA2 conformation. This tunable hub drives three distinct functional outputs: (1) ATP-dependent processive unwinding, (2) Mg²⁺-primed ATP-independent remodeling, and (3) ATP-independent strand annealing and chaperoning. The coordinated switching among these outputs enables Nsp13 to operate as a multifunctional, context-sensitive remodeler that supports diverse nucleic-acid transactions during viral replication and transcription.

DNA and RNA sequences used in different assays.

SARS-CoV-2 Nsp13 mass spectrometry.

Binding parameter of SARS-CoV-2 Nsp13.

Binding affinity of Nsp13 to various nucleic acid substrates measured by fluorescence anisotropy.

Binding isotherms of Nsp13 to (A) single-stranded DNA (ssDNA) and single-stranded RNA (ssRNA); (B) RNA hairpin; (C) DNA G-quadruplex (DG4); and (D) RNA G-quadruplex (RG4). Data are presented as mean ± SD from three independent experiments. Curves were fitted to the Hill equation to determine the equilibrium dissociation constant (Kd) and Hill coefficient (n).

Verification of ATP-independent DNA unwinding by His-tag-free Nsp13.

(A) EMSA showing ATP-independent DNA unwinding by His-tag-free Nsp13 (Nsp13-notag). (B) Quantification of unwound DNA fraction illustrating the effect of Mg²⁺ concentration on ATP-independent unwinding. Data are presented as mean ± SD (n=3). Statistical significance was determined by unpaired Student’s t-test.

Mechanism of Mg²⁺-mediated DNA unwinding by Nsp13.

(A)FRET-melting curves of duplex DNA in the presence of increasing Mg²⁺ concentrations (0–8 mM). (B) FRET-melting curves of duplex DNA in the presence of increasing Nsp13 concentrations (0–1.8 µM). (C) Bar graph showing that Nsp13 does not destabilize the duplex DNA substrate. (D) Binding curves of Nsp13 to duplex DNA in the presence of 0 mM and 0.5 mM Mg²⁺.

AlphaFold3-predicted Nsp13.

(A) The predicted structure of apo Nsp13 aligns well with the experimental structure 7nio from the PDB. (B) Nsp13-ss/dsRNA complex with Mg²⁺ and ADP. Mg²⁺ binds in the RecA1–RecA2 cleft and induces structural compaction. ADP binding further stabilizes the closed conformation.

AlphaFold3-predicted structures of Nsp13-ATP-Mg²⁺ and K288R-ATP-Mg²⁺ complexes.

Cartoon representation: Nsp13-ATP-Mg²⁺ in green, K288R-ATP-Mg²⁺ in orange. ATP molecules are shown as sticks, and Mg²⁺ ions as purple spheres.

The ATP/Mg²⁺ ratio fine-tunes Nsp13 unwinding of duplex DNA and Nsp13 exhibits 3′→5′ unwinding activity.

(A) EMSA (left) and quantification (right) demonstrating that increasing Mg²⁺ concentrations inhibit Nsp13-mediated unwinding at a fixed ATP level (1 mM). The 5′-overhang substrate 5′OhS14D24 (40 nM) was used. (B) Balanced ATP/Mg²⁺ sustains unwinding efficiency. EMSA (left) and quantification (right) showing that maintaining a 1:1 stoichiometry between ATP and Mg²⁺ preserves robust unwinding across a range of ATP concentrations (1–5 mM). All data represent mean ± SD from three independent experiments. (C) Schematic of the 3′-overhang DNA substrate (3′OhD16S14). EMSA illustrates efficient unwinding by Nsp13 in the presence of 5–10 mM Mg²⁺, either without or with 2 mM ATP. Note: The FAM-labeled ssDNA strand migrates more slowly than the enzyme-released unwound product, likely due to secondary structure formation in the absence of Nsp13.

Mg²⁺ and Nsp13 concentration dependence of RNA fork unwinding.

(A) EMSA showing RNA unwinding by Nsp13 in the absence of ATP at 0 mM and 1 mM Mg²⁺. (B) EMSA analysis of RNA unwinding by the ATPase-deficient mutant K288R at low Nsp13 concentration (80 nM) under different ATP conditions. (C) Quantification of unwound RNA fractions at 0.5 mM Mg²⁺ as a function of Nsp13 concentration, reproduced from Figure 4A. Data represent mean ± SD from three independent replicates.

Analysis of RNA duplex unwinding under inhibitory Nsp13 conditions.

EMSA showing RNA duplex unwinding at the same Nsp13 concentration (0.4 μM) that inhibited RNA unwinding in Figure 4H in the presence of 1 mM ATP–Mg²⁺, but using a different RNA substrate concentration (40 nM vs. 80 nM).

ATP-dependent DNA unwinding by Nsp13 increases with protein concentration.

(A) EMSA analysis of helicase activity on duplex DNA with increasing Nsp13 concentrations (0.02–3.2 µM). Reaction conditions: 40 nM DNA, 1 mM Mg2+, 1 mM ATP, 37°C. (B) Quantification of unwound DNA fractions. Data represent mean ± SD from three replicates.

EMSA analysis of SARS-CoV-2 RNA G-quadruplex (RG-1) unfolding by varying concentrations of Nsp13 in 1 mM Mg²⁺.

Partial unfolding was observed at low to moderate protein concentrations, whereas higher Nsp13 levels inhibited G4 unfolding.

Nsp13 exhibits ATP- and Mg²⁺-independent strand annealing activity on diverse DNA and RNA substrates.

(A) Schematic of DNA and RNA substrate structures used in annealing assays. (B) EMSA results showing Nsp13-mediated annealing of DNA substrates with 5′-overhang, 3′-overhang, and fork structures, respectively. (C) Concentration–response curve of DNA annealing activity in the absence of Mg2+ and ATP. (D) Annealing efficiency for three DNA configurations at 200 nM Nsp13. (E) EMSA results showing RNA annealing with 5′-overhang, 3′-overhang, fork, and blunt-end substrates. (F) Concentration–response curve of RNA annealing activity in the absence of Mg2+ and ATP. (G) Annealing efficiency for four RNA configurations at 200 nM Nsp13. Data are presented as mean ± SD (n=3). Statistical significance was determined by one-way ANOVA with Tukey’s HSD test; ***P < 0.001; ns, not significant. Note: DNA blunt-end substrates were excluded from the analysis due to instability under the annealing buffer conditions.

Chaperone activity of Nsp13.

(A) Schematic of the strand-exchange assay. Complementary 42-nt DNA and RNA stem-loops were used as substrates, with the FAM-labeled strands indicated by asterisks. (B–C) Time-course EMSA (B) and quantification (C) showing Nsp13-mediated strand exchange in the absence of ATP, demonstrating its intrinsic chaperone activity. Hybridization fraction was calculated from three independent replicates (mean ± SD). (D) Control experiment with ToPif1, a canonical 5′→3′ helicase. Pre-folded DNA stem-loops were incubated with increasing concentrations of ToPif1 (0–3 µM) under identical conditions. No strand exchange was observed, confirming that the chaperone activity is unique to Nsp13 and not a general property of helicases. All reactions were performed with 20 nM FAM-labeled and unlabeled strands.