Discovery of a subgenotype of human coronavirus NL63 associated with severe lower respiratory tract infection in China, 2018

Abstract

Human coronavirus NL63 (HCoV-NL63) is primarily associated with common cold in children, elderly and immunocompromised individuals. Outbreaks caused by HCoV-NL63 are rare. Here we report a cluster of HCoV-NL63 cases with severe lower respiratory tract infection that arose in Guangzhou, China, in 2018. Twenty-three hospitalized children were confirmed to be HCoV-NL63 positive, and most of whom were hospitalized with severe pneumonia or acute bronchitis. Whole genomes of HCoV-NL63 were obtained using next-generation sequencing. Phylogenetic and single amino acid polymorphism analyses showed that this outbreak was associated with two subgenotypes (C3 and B) of HCoV-NL63. Half of patients were identified to be related to a new subgenotype C3. One unique amino acid mutation at I507 L in spike protein receptor binding domain (RBD) was detected, which segregated this subgenotype C3 from other known subgenotypes. Pseudotyped virus bearing the I507 L mutation in RBD showed enhanced entry into host cells as compared to the prototype virus. This study proved that HCoV-NL63 was undergoing continuous mutation and has the potential to cause severe lower respiratory disease in humans.

Keywords: human coronavirus NL63; new subgenotype; phylogenetic analysis; pneumonia; viral entry; whole-genome sequencing.

Figure 1.

Identification of a cluster of patients with severe lower respiratory disease caused by HCoV-NL63. (A) Prevalence of HCoV-NL63 by year in the pediatric inpatient surveillance study at the First Affiliated Hospital of Guangzhou Medical University. (B) Case number of HCoV-NL63 infection by month in the pediatric inpatient surveillance study in 2018. (C) Clinical information of the HCoV-NL63 positive inpatients in 2018.

Figure 2.

Whole-genome sequencing of HCoV-NL63 identified in this study. (A) Whole genome coverage of HCoV-NL63 performed by CLC Genomics Workbench software version 11. (B) Summary of HCoV-NL63 sequences obtained in this study.

Figure 3.

Phylogenetic analyses based on partial spike genes of HCoV-NL63 identified in this study. Fourteen HCoV-NL63 spike gene partial sequences detected in this study were used for phylogenetic analysis by MEGA 7.0 software using Neighbour-joining method and further confirmed the presence of new subgenotype of HCoV-NL63. Bootstrap values greater than 60% were considered statistically significant for grouping.

Figure 4.

Phylogenetic analysis based on complete genome, S and ORF1ab genes of HCoV-NL63. All available HCoV-NL63 complete genomes (53 strains) from GenBank were collected and used for the evolutional analysis using MEGA 7.0 with 1000 bootstrap replications, Bootstrap values greater than 60% were considered statistically significant for grouping. (A). Five complete genomes derived from this study were in red. Nucleotide sequence alignments were created using MAFFT. Corresponding spike (B) and orf1ab (C) genes were used for the genotype identification.

Figure 5.

Time-resolved phylogenetic analysis of spike gene of HCoV-NL63 strains. The year of sampling, strain name and accession number are on the tip labels. Node labels indicate the posterior probabilities. Strains of subgenotypes C3 and B detected in Guangzhou were indicated in green and purple. BEAST software was used to estimate the most recent common ancestor (tMRCA) of the new subgenotype circulating in Guangzhou, based on nucleotide sequences of spike gene. Analyses were conducted under the best-fit nucleotide substitution model (GTR  +  I + G) and using a relaxed (uncorrelated lognormal) molecular clock model.

Figure 6.

Single amino acid polymorphism analysis of HCoV-NL63 spike protein. All available HCoV-NL63 complete genomes were aligned, and corresponding spike proteins were retrieved and used for single amino acid polymorphism analysis. Most of SAP lies in S1 domain, and one unique mutation I507L was identified in RBD of spike in yellow.

Figure 7.

I507L mutation in RBD promoted viral entry. The efficiency of viral entry was analysed using HCoV-NL63-S pseudotyped viruses bearing the wild-type (WT) or I507L (A, B) or E471D (C, D) mutant S proteins in Huh7 cells. Cell control serves as the negative control to determine background. Two different viral pseudotypes were employed in this study, including lentivirus system (A, C) and VSV system (B, D). Huh7 cells were infected with indicated pseudotyped viruses in triplicates. Luciferase activities were measured 72 h post infection. A Student's t test was used to analyse differences in mean values between groups. P values of <0.05 were considered statistically significant. Data are representative of three independent experiments.