SARS-CoV-2 variants with NSP12 P323L/G671S mutations display enhanced virus replication in ferret upper airways and higher transmissibility

Abstract

With the emergence of multiple predominant SARS-CoV-2 variants, it becomes important to have a comprehensive assessment of their viral fitness and transmissibility. Here, we demonstrate that natural temperature differences between the upper (33°C) and lower (37°C) respiratory tract have profound effects on SARS-CoV-2 replication and transmissibility. Specifically, SARS-CoV-2 variants containing the NSP12 mutations P323L or P323L/G671S exhibit enhanced RNA-dependent RNA polymerase (RdRp) activity at 33°C compared with 37°C and high transmissibility. Molecular dynamics simulations and microscale thermophoresis demonstrate that the NSP12 P323L and P323L/G671S mutations stabilize the NSP12-NSP7-NSP8 complex through hydrophobic effects, leading to increased viral RdRp activity. Furthermore, competitive transmissibility assay reveals that reverse genetic (RG)-P323L or RG-P323L/G671S NSP12 outcompetes RG-WT (wild-type) NSP12 for replication in the upper respiratory tract, allowing markedly rapid transmissibility. This suggests that NSP12 P323L or P323L/G671S mutation of SARS-CoV-2 is associated with increased RdRp complex stability and enzymatic activity, promoting efficient transmissibility.

Keywords: CP: Immunology; NSP12 mutation; RNA-dependent RNA polymerase; SARS-CoV-2; ferret; transmissibility.

Figure 1.. In vitro growth kinetics and in vivo transmission characteristics of SARS-CoV-2 strains and variants in ferrets.

Comparison of differential growth kinetics among the reference strain (Wuhan-Hu-1-like, Clade L), D614G strain, and variant viruses (Alpha, Beta, and Delta) in (a) Vero E6, (b) Vero^ACE2/TMPRSS2, (c) Huh-7, and (d) Calu-3 cells. The cells were infected with viruses at an MOI of 0.01. Ferrets (n = 9/group) were inoculated intranasally with 5.0 log₁₀ TCID₅₀/mL of each representative SARS-CoV-2 strain. Nasal wash specimens were collected at a 2-day interval from directly infected ferrets, and daily from the first-round transmission and second-round transmission ferrets. Ferrets (n = 3/group) were sacrificed at 3 and 5 dpi. Infectious virus titers in the (e) nasal wash specimen, (f) nasal turbinates, and (g) lung tissue samples were measured by determining the TCID₅₀ per unit of measurement (g or mL) using Vero E6 cells. In Fig. 1e and f, black * indicates statistical results compared to the reference strain, and red * indicates statistical values versus D614G/P323. The limit of detection (1.3 log₁₀ TCID₅₀/ml or log₁₀TCID₅₀/g) is indicated by a dotted line for each representation. The p values were calculated using a two-way ANOVA, with Turkey’s post-test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 2.. SARS-CoV-2 RdRp reporter gene assay in fold change.

(a) The results of amino acid genetic variance analysis are summarized as a graph, showing information regarding genetic variations in the NSP12 among D614G and SARS-CoV-2 variants in comparison with the Wuhan-hu-1 (Clade L) reference sequence. (b-d) 293T cells were transiently transfected with SARS-CoV-2 NSP12 wild-type (WT), P323L, G671S, P323L/G671S, P323D, or D760N/D761N enzyme-dead mutant. NSP12 was co-transfected with the NSP7 and NSP8 cofactors, followed by incubation at 33°C or 37°C. (b) NSP12 only. (c) NSP7:NSP8:NSP12 ratio of 1:1:1. (d) NSP7:NSP8:NSP12 ratio of 1:2:1. NLuc activities were measured after 24 hours post-transfection. Black * indicates statistical results compared to the reference NSP12-only, and red * indicates statistical values versus NSP12-WT in complex formation with NSP7-NSP8. The p values were calculated using a one-way ANOVA, with the Kruskal-Wallis test. *p<0.033, ** p<0.002, *** p<0.001

Figure 3.. Structural analysis of the effects of the P323L and G671S variations on the NSP12 and NSP8 complex and the Molecular dynamics (MD) simulations for each single-point amino acid substitution variant.

(a) Pro323 (red dashed circle) and neighboring residues in the NSP12 replication and transcription complex (RTC) structure (PDB entry 7CYQ, left panel). The mutation of Pro323 to leucine is shown in the red dashed circle (right panel). The estimated interaction residues of NSP12 and NSP8 are illustrated with light blue and light brown, respectively. (b) Gly671 (red dashed circle, left panel) and the mutation of Gly671 to serine (red dashed circle, right panel) are shown in the same SARS-CoV-2 RTC structure. The structure models showing the mutations were generated using PyMOL (Schrödinger, L. & De Lano, W., 2020. PyMOL, Available at: http://www.pymol.org/pymol.). NSP12 and NSP8 are represented in light blue and light brown, respectively. Hydrogen bonds are indicated as black dashed lines and the hydrophobic interactions with neighboring residues are illustrated with dash lines with arrowheads on both ends.

Figure 4.. SARS-CoV-2 NSP12 bearing the P323L or P323L/G671S mutations enhance RdRp interaction with NSP7/NSP8/RNA.

(a) SDS-PAGE of purified SARS-CoV-2 NSP7 and NSP8 confirmed by Coomassie blue staining. (b) SDS-PAGE of purified SARS-CoV-2 NSP12 WT, P323L, G671S, or P323L/G671S indicated by Coomassie blue staining (left) or western blot using an anti-NSP12 antibody (right). (c) In vitro transcribed SARS-CoV-2 3ʹ-end 55-nt RNA in 3% agarose gel, right panel. The predicted secondary structure of the SAR-CoV-2 3ʹ-end 55-nt RNA (left). (d, e, and f) Microscale thermophoresis assay was performed to analyze the ability of RdRp to form a complex with experimental sets of (d) WT versus P323L, (e) WT versus P323L, (f) or WT versus P323L/G671S NSP12. Binding affinity was examined by monitoring the thermophoretic traces, and the dissociation constant (K_D) was determined by measuring the change in fluorescence (ΔFnorm). Data are presented as the mean ± s.d. of three independent analyses.

Figure 5.. Growth kinetics of RG-derived SARS-CoV-2 in vitro.

Comparison of differential growth kinetics among RG SARS-CoV-2 viruses (WT, P323L, G671S, and P323L/G671S mutant) in (a-b) Vero E6, (c-d) Vero^ACE2/TMPRSS2, (e-f) Huh-7, and (g-h) Calu-3 cells at 33°C or 37°C. Cells were infected with each RG virus at an MOI of 0.01. The supernatants of infected cells were harvested at the indicated time points, and virus titers were determined using a tissue culture infection dose 50% (TCID₅₀) in Vero cells. The p values were calculated using a two-way ANOVA, with Turkey’s post-test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 6.. Replication and transmission of RG-derived SARS-CoV-2 in ferrets.

Ferrets (n = 9/group) were inoculated intranasally with 10⁵ TCID₅₀/ml of RG-derived SARS-CoV-2. Nasal wash specimens were collected at a 2-day interval from directly infected ferrets, and daily from first-round and second-round transmission ferrets. Ferrets (n = 3/group) were sacrificed at 3 and 5 dpi. The infectious virus titers in (a) nasal wash specimen, (b) nasal turbinates, and (c) lung tissue samples were measured by determining the TCID₅₀ per unit of measurement (g or mL) using Vero E6 cells. (d-e) Competitive transmission assays of RG-WT and RG-P323L viruses (d) or RG-WT and RG-P323L/G671S viruses (e) in donor ferrets (n=3) and cohoused direct contact (DC) naïve ferrets (n=3). RG-WT (P323/G671) and RG-P323L viruses or RG-WT (P323/G671) and RG-P323L/G671S viruses were mixed at a 1:1 ratio based on their infectious virus titers, the mixture was intranasally inoculated into ferrets. Nasal washes were collected from mixed virus-infected ferrets (days 2, 4, and 6) or cohoused DC ferrets (3, 4, 5, and 6), and then analyzed by NGS. Virus titers (RNA copy number) were determined by qPCR on the left y-axis and the proportions of viruses were shown on the right y-axis. All results indicate triplicate repeats. The limit of detection (1.3 log₁₀ TCID₅₀/ml or log₁₀ TCID₅₀/g) is indicated by a dotted line for each representation. The p values were calculated using a two-way ANOVA, with Turkey’s post-test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.