Molecular epidemiology, evolution and phylogeny of SARS coronavirus

Abstract

Shortly after its emergence in southern China in 2002/2003, Severe Acute Respiratory Syndrome coronavirus (SARS-CoV) was confirmed to be the cause of SARS. Subsequently, SARS-related CoVs (SARSr-CoVs) were found in palm civets from live animal markets in Guangdong and in various horseshoe bat species, which were believed to be the ultimate reservoir of SARSr-CoV. Till November 2018, 339 SARSr-CoV genomes have been sequenced, including 274 from human, 18 from civets and 47 from bats [mostly from Chinese horseshoe bats (Rhinolophus sinicus), n = 30; and greater horseshoe bats (Rhinolophus ferrumequinum), n = 9]. The human SARS-CoVs and civet SARSr-CoVs were collected in 2003/2004, while bat SARSr-CoVs were continuously isolated in the past 13 years even after the cessation of the SARS epidemic. SARSr-CoVs belong to the subgenus Sarbecovirus (previously lineage B) of genus Betacoronavirus and occupy a unique phylogenetic position. Overall, it is observed that the SARSr-CoV genomes from bats in Yunnan province of China possess the highest nucleotide identity to those from civets. It is evident from both multiple alignment and phylogenetic analyses that some genes of a particular SARSr-CoV from bats may possess higher while other genes possess much lower nucleotide identity to the corresponding genes of SARSr-CoV from human/civets, resulting in the shift of phylogenetic position in different phylogenetic trees. Our current model on the origin of SARS is that the human SARS-CoV that caused the epidemic in 2002/2003 was probably a result of multiple recombination events from a number of SARSr-CoV ancestors in different horseshoe bat species.

Keywords: Evolution; Molecular epidemiology; Phylogeny; SARS coronavirus.

Fig. 1

Genome organization of SARS-CoV. ORF1ab with nsp1–16 are colored in blue. Structural proteins including S, E, M and N are in pink. Accessory proteins were numbered and in yellow. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

The diversity of CoVs as demonstrated with a phylogenetic tree targeting the 303 bp partial RdRp sequence (position 15,293–15,596 with respect to Human SARS-CoV TOR2). The neighbor-joining phylogenetic tree was constructed with maximum composite likelihood method by MEGA 7.0. The test of phylogeny was statistically supported by the bootstraps value calculated from 1000 trees. Bat CoVs were labeled with black triangles. The branches of Gammacoronavirus and Deltacoronavirus were compressed.

Fig. 3

(A) Nucleotide identity and (B) Genetic Distance along the SARS-CoV genome for all available SARSr-CoVs with complete genomes. A comparison is made with reference to Civet SARSr-CoV SZ3. Strains are listed in (A) descending order from the top according to the whole genome identity (B) ascending order from the top according to the whole genome genetic distance at the last column. Red and blue boxes represent the highest and lowest end of the identity, respectively. White boxes represent deletions. The titles of SARSr-Rf-BatCoVs from greater horseshoe bats are highlighted in pink. The data were generated by Matrix Global Alignment Tool (MatGAT) (Campanella et al., 2003). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

(A) Nucleotide identity and (B) Genetic Distance along the SARS-CoV genome for all available SARSr-CoVs with complete genomes. A comparison is made with reference to Civet SARSr-CoV SZ3. Strains are listed in (A) descending order from the top according to the whole genome identity (B) ascending order from the top according to the whole genome genetic distance at the last column. Red and blue boxes represent the highest and lowest end of the identity, respectively. White boxes represent deletions. The titles of SARSr-Rf-BatCoVs from greater horseshoe bats are highlighted in pink. The data were generated by Matrix Global Alignment Tool (MatGAT) (Campanella et al., 2003). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Phylogenetic tree constructed based on the nt sequences of the RBD of the S protein of SARS-CoVs and SARSr-CoV. SARSr-Rf-BatCoVs are labeled with dots. Brackets in red, orange, yellow and blue represented the descending order of nt identity with respect to civet SARSr-CoV SZ16. The phylogenetic tree was constructed by Maximum Likelihood method with T92 + G as the substitution model by MEGA 6.0. The test of phylogeny was statistically supported by the bootstraps value calculated from 1000 trees. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Multiple alignment of type I ORF8 of SARSr-Rs-BatCoV YN2013, GX2013, civet SARSr-CoV and SARSr-Rf-BatCoV strain Rf4092 against the type II ORF8 from other SARSr-Rf-BatCoVs. Host-specific residues were highlighted with red boxes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6

Phylogenetic tree constructed based on the nt sequences of ORF8 of SARS-CoVs. SARSr-Rf-BatCoVs are labeled with dots. Brackets in red, orange, yellow and blue represented descending order of nt identity with respect to civet SARSr-CoV SZ16. The phylogenetic tree was constructed by Maximum Likelihood method with T92 + G as the substitution model by MEGA 6.0. The test of phylogeny was statistically supported by the bootstraps value calculated from 1000 trees. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)