- Research article
- Open Access
Adaptive evolution of the spike gene of SARS coronavirus: changes in positively selected sites in different epidemic groups
© Zhang et al; licensee BioMed Central Ltd. 2006
- Received: 16 April 2006
- Accepted: 04 October 2006
- Published: 04 October 2006
It is believed that animal-to-human transmission of severe acute respiratory syndrome (SARS) coronavirus (CoV) is the cause of the SARS outbreak worldwide. The spike (S) protein is one of the best characterized proteins of SARS-CoV, which plays a key role in SARS-CoV overcoming species barrier and accomplishing interspecies transmission from animals to humans, suggesting that it may be the major target of selective pressure. However, the process of adaptive evolution of S protein and the exact positively selected sites associated with this process remain unknown.
By investigating the adaptive evolution of S protein, we identified twelve amino acid sites (75, 239, 244, 311, 479, 609, 613, 743, 765, 778, 1148, and 1163) in the S protein under positive selective pressure. Based on phylogenetic tree and epidemiological investigation, SARS outbreak was divided into three epidemic groups: 02–04 interspecies, 03-early-mid, and 03-late epidemic groups in the present study. Positive selection was detected in the first two groups, which represent the course of SARS-CoV interspecies transmission and of viral adaptation to human host, respectively. In contrast, purifying selection was detected in 03-late group. These indicate that S protein experiences variable positive selective pressures before reaching stabilization. A total of 25 sites in 02–04 interspecies epidemic group and 16 sites in 03-early-mid epidemic group were identified under positive selection. The identified sites were different between these two groups except for site 239, which suggests that positively selected sites are changeable between groups. Moreover, it was showed that a larger proportion (24%) of positively selected sites was located in receptor-binding domain (RBD) than in heptad repeat (HR)1-HR2 region in 02–04 interspecies epidemic group (p = 0.0208), and a greater percentage (25%) of these sites occurred in HR1–HR2 region than in RBD in 03-early-mid epidemic group (p = 0.0721). These suggest that functionally different domains of S protein may not experience same positive selection in each epidemic group. In addition, three specific replacements (F360S, T487S and L665S) were only found between 03-human SARS-CoVs and strains from 02–04 interspecies epidemic group, which reveals that selective sweep may also force the evolution of S genes before the jump of SARS-CoVs into human hosts. Since certain residues at these positively selected sites are associated with receptor recognition and/or membrane fusion, they are likely to be the crucial residues for animal-to-human transmission of SARS-CoVs, and subsequent adaptation to human hosts.
The variation of positive selective pressures and positively selected sites are likely to contribute to the adaptive evolution of S protein from animals to humans.
- Severe Acute Respiratory Syndrome
- Amino Acid Site
- Severe Acute Respiratory Syndrome
- Interspecies Transmission
- Codeml Program
SARS is a new infectious disease that emerged in the Guangdong province of China in November 2002. It caused 8,096 infection cases including 774 deaths worldwide during its epidemic . The causative pathogen of SARS was identified as a novel strain of human coronavirus, named as SARS-CoV, and its complete genome was sequenced in March 2003 [2–5]. In May 2003, SARS-CoVs were also isolated from a few Himalayan palm civets (Paguma larvata) and a raccoon dog (Nyctereutes procyonoides) in a food market in Shenzhen (Guangdong, China) . These isolations provided the first evidence that wild animals could be reservoirs for SARS-CoV, and that the virus might be transmitted from animals to humans. The re-emergence of SARS in 2003–2004 in Guangdong, China confirmed that SARS-CoV was independently transmitted from animals to humans .
The S protein of SARS-CoV is composed of 1,255 amino acids, and is responsible for viral attachment and entry into host cells [4, 5]. It is also a major antigenic determinant that induces generation of neutralizing antibodies and protective immunity at least in human host . Unlike some coronaviruses, in which S protein can be cleaved into two functional subunits, S1 and S2, the S protein of SARS-CoV is not cleavable due to the absence of the proteolytic cleavage site. However, two domains, S1 (residues 14–680) and S2 (residues 681–1,255) were identified in SARS-CoV S protein in the light of their homology with the S1 and S2 subunits . Domain S1 is responsible for binding to angiotensin-converting enzyme-2 (ACE2), which serves as the functional receptor of SARS-CoV [10, 11]. Domain S2 mediates viral entry into host cells [12, 13]. Previous works indicated that interspecies transmission may be due to the acquisition of mutations in S protein which allows human infection, suggesting that S protein ought to be a major target of selective pressure [6, 7, 14].
A criterion for the determination of selective pressure is to compare nonsynonymous (amino acid-changing; d N ) with synonymous (silent; d S ) substitution rates in protein-coding genes. The nonsynonymous/synonymous rate ratio (ω = d N /d S ) provides a straightforward measurement of selective pressure at the protein level. The ω values of > 1, 1 and < 1 indicate positive (diversifying) selection, random drift and negative (purifying) selection, respectively. The pairwise analysis showed that S protein, which has an average ω ratio for all amino acid sites greater than 1, is under overall positive selective pressure [7, 14, 15]. However, the process of adaptation of S protein to human receptors and the positively selected sites associated with this process remain unclear. Because previous epidemic phases [7, 14] are unable to adequately reflect the process of animal-to-human transmission and of viral adaptation to human host, we therefore reclassified the SARS groups in the present study. We found that the S protein of SARS-CoV experiences variable positive selective pressures and the positively selected sites are changeable in different epidemic groups. These observations provide a good evidence for understanding the molecular adaptation of SARS-CoV from animals to humans.
Positive selection on S genes of SARS-CoV during the whole outbreak from 2002 to 2004
Likelihood values and parameter estimates for 45 S gene sequences from the whole epidemic of SARS from 2002 to 2004.
Estimates of parameters
Positively selected sites
ω = 0.64
p0 = 0.43340, p1 = 0.50712 (p2 = 0.05948), ω0 = 0.12, ω1 = 0.13, ω2 = 9.68
75T 239S 244T 311G 479R 609L 613E 743R 765V 778D 1148L 1163E
p0 = 0.63944 (p1 = 0.36056)
p0 = 0.78594, p1 = 0.16251 (p2 = 0.05155), ω2 = 10.29
75T 239S 244T 311G 479R 609L 613E 743R 765V 778D 1148L 1163E
p = 0.00568, q = 0.00856
p0 = 0.94614 (p1 = 0.05386)
p = 0.05787, q = 0.30879, ω = 10.11
75T 239S 244T 311G 479R 609L 613E 743R 765V 778D 1148L 1163E
Detection of recombination and positive selection on S genes of SARS-CoV in different epidemic groups
Phylogenetic analysis by ML estimation for SARS S gene sequences from different epidemic groups.
Estimates of parameters
02–04 interspecies epidemic group
ω = 0.64
p0 = 0.00000, = p1 = 0.99407 (p2 = 0.00593), ω0 = 0.00, ω1 = 0.38, ω2 = 66.02
p0 = 0.65391 (p1 = 0.34609)
p0 = 0.99407, p1 = 0.00000 (p2 = 0.00593), ω2 = 66.01
p = 0.01702, q = 0.03977
p0 = 0.99405 (p1 = 0.00595), p0 = 0.01217, q = 0.01714 ω = 67.24
03-early-mid epidemic group
ω = 0.94
p0 = 0.05053, p1 = 0.92251 (p2 = 0.02696), ω0 = ω1 = 0.00, ω2 = 40.88
p0 = 0.54981 (p1 = 0.45019)
p0 = 0.97303, p1 = 0.00000 (p2 = 0.02696), ω2 = 40.88
p = 0.00510, q = 0.00776
p0 = 0.97303 (p1 = 0.02697), p = 0.00500, q = 1.39962, ω = 40.87
03-late epidemic group
ω = 0.26
p0 = 0.43056, p1 = 0.37490 (p2 = 0.19455), ω0 = 0.25, ω1 = 0.26, ω2 = 0.26
p0 = 1.00000 (p1 = 0.00000)
p0 = 1.00000, p1 = 0. 00000 (p2 = 0.00000), ω0 = 0.26, ω1 = ω2 = 1.00
p = 33.88884, q = 99.00000
p0 = 1.00000 (p1 = 0.00000), p = 0.64576, q = 1.83663, ω = 2.08
Comparison of positively selected sites on S genes in different epidemic groups
Positively selected sites identified by Codeml program and REL in DataMonkey package.
Positively selected sites in different epidemic groups
78 113 139 147 227 239 261 336 425 462 472 479 480 558 607 608 609 613
49 75 77 144 239 244 311 344
701 714 743 754 765 856 894
778 860 861 1001 1148 1163 1179 1247
The number of positively selected sites in different functional domains of S protein.
02–04 interspecies epidemic group
03-early-mid epidemic group
Fisher's exact test (p)
Lineage fixation of positively selected sites on S genes for the adaptation of SARS-CoV to human host
The S protein of SARS-CoV is responsible for the receptor binding and membrane fusion . It is also a major antigen to stimulate humoral immunity of its host . The amino acid variation of S protein affects virus entry, tissue tropism and host range of SARS-CoV [11, 23]. Here, we confirmed that the S gene undergoes strong positive selection [7, 14–16], and identified twelve positively selected amino acid sites, including 75, 239, 244, 311, 479, 609, 613, 743, 765, 778, 1148, and 1163 during the whole SARS outbreak (Table 1). Among these sites, positions 239, 311, 479, 609, 743, 778, 1148, and 1163 appeared to be exposed on the surface of S protein [9, 24], suggesting that they are likely to play a key role in viral transmission and survival. In addition, it was worth pointing out that SARS-CoV is a rapidly evolving RNA virus with a mutation rate of 0.8–2.38 × 10-3 nucleotide substitution per site per year . The S gene sequences used in the present study were sampled during a year period, and some mutations might be accumulated in late-sampled sequences [26, 27]. However, whether the accumulation of these mutations influences the detection of positive selection and the identification of positively selected sites remains unclear . This requires further investigation to confirm.
Adaptation of an animal virus to a new human host usually faces two crucial bottlenecks: the receptor adaptation of viral surface protein to its new host, followed by the adaptation of key enzymes (e.g. viral replicases) associated with viral replication to new cellular components that possibly support poorly productive infection (e.g. non-permissive cells) [21, 28]. The latter is not always the step that limits host expansion and most viruses can establish productive infection after their entry of host cells . We found that two key replicases of SARS-CoV, RNA-dependent RNA polymerase (RdRp) and helicase, were not under positive selection (Zhang CY et al., unpublished data), which suggests that receptor adaptation of S protein to human host determines the animal-to-human transmission of SARS-CoV [11, 29]. The receptor adaptation of an animal virus to a new human host usually requires two key steps: initial breakthrough of receptor barrier (animal-to-human transmission), followed by the molecular adaptation to human cellular receptors (human-to-human transmission). The two steps together result in eventual establishment of stable infection necessary for efficient spread within human hosts.
In order to better reflect the course of viral trans-species transmission and subsequent adaptation to human hosts, the collection of SARS isolates was reclassified into three epidemic groups: 02–04 interspecies, 03-early-mid, and 03-late epidemic groups in the present study. The 02–04 interspecies epidemic group reflects the process of viral trans-species transmission, and 03-early-mid epidemic group represents the crucial phase of SARS-CoVs to adapt to human host. The two groups correspond with the two key steps described above for a virus to be adapted by a new cellular receptor.
We found that S genes underwent strong positive selection in both groups of 02–04 interspecies epidemic and 03-early-mid epidemic, whereas no positive selection was observed in 03-late epidemic group (Table 2). It suggests that S protein experiences a step-by-step adaptation process to human cellular receptors. On the other hand, the amino acid sites under positive selection in 02–04 interspecies epidemic group differed clearly from those in 03-early-mid epidemic group, suggesting for the first time the changes in positively selected sites in different epidemic groups. It was reported previously that two functional domains S1 and S2 of SARS-CoV S protein are responsible for receptor recognition and membrane fusion, respectively . In domain S1, RBD has been demonstrated to concentrate in a 193-amino acid fragment (residues 318–510), which is adequate for binding to human ACE2 . In domain S2, two highly conserved regions HP1 (residues: 889–972) and 2 (residues: 1142–1185) are crucial for membrane fusion . Importantly, a larger proportion of positively selected sites was located in RBD than in HR1–HR2 or HR2 regions in 02–04 interspecies epidemic group, and a greater percentage of these sites occurred in HR1–HR2 region than in RBD in 03-early-mid epidemic group (Fisher's exact test: p = 0.045 for HR1–HR2 vs. RBD, and p = 0.033 for HR2 vs. RBD) (Table 4). These differences suggest that positive selection prefers influencing the receptor-binding function in 02–04 interspecies epidemic group, and then associates with the membrane fusion function in 03-early-mid epidemic group.
In 02–04 interspecies epidemic group, four positively selected sites 479, 609, 743 and 765 appeared to be fixed in 03-early-mid epidemic group. Comparing the amino acid sequences of S genes from 03-huSARS-CoV and 03-pcSARS-CoV, we discovered that two positively selected amino acid substitutions (N479K and T743A) are likely to play a role in viral receptor switch. Previous in vitro experiments proved that the residues at positions 479 and 487 are important determinants for SARS-CoV cell tropism and animal-to-human transmission [11, 23, 32, 33]. Residue 479N (asparagine) increases the affinity of S protein to human receptor by specifically interacting with the residue 34H (histidine) of ACE2, which is present in human but not in palm civet ACE2 [11, 32]. Recently, Zheng et al. reported that active peptide (residues 737–756) of S protein which contains residue 743T (threonine) effectively inhibited the entry of huSARS-CoV into human cells . The corresponding peptide with residue 743A (alanine) from 03-pcSARS-CoV appeared less potent against huSARS-CoV . These observations suggested that residue 743 may influence viral receptor tropism via a way different from ACE2 binding . Besides ACE2, human DC-SIGN and DC-SIGNL were also shown to be able to enhance SARS-CoV infection by a non-receptor mechanism [35, 36]. However, the influence of residue 743 on potential interaction between S protein and DC-SIGN or DC-SIGNL remains unknown. On the other hand, of the four positively selected sites, two variants L609A and V765A, were observed in most 04-pcSARS-CoVs, but not in 03-pcSARS-CoVs, suggesting that they may contribute to viral adaptive evolution from 03-pcSARS-CoV to 04-pcSARS-CoV.
The sequence comparison between groups of 02–04 interspecies epidemic and 03-early-mid epidemic allowed us to find additional three special substitutions (T1079C, C1460G and T1994C), which lead to three replacement (F360S, T487S and L665S) (Fig. 2). Of particular importance was that all three sites were monomorphic for serine in the 02–04 interspecies epidemic group, and utterly monomorphic for phenylalanine at site 360, threonine at 487 and leucine at 665 in the 03-human epidemic group (including 03-early-mid and 03-late phases) (Fig. 2). Previous studies revealed that the replacement T487S benefits the receptor switch of SARS-CoVs from palm civets to human [11, 23, 32]. However, both Codeml program and DataMonkey package did not detect positive selection on this site. Therefore, the best explanation should be that the selective sweep also drives the adaptive evolution of S gene from animals to humans, despite of little experimental evidence supporting the advantageous mutations in sites 360 and 665.
After breaking through the interspecies receptor barrier, SARS-CoV spread quickly among human hosts and formed the 03-early-mid epidemic group, a very important period for viral adaptation to human host. During this stage, four major sites (75, 239, 778 and 1163) of S protein were identified under positive selection. Among them, sites 778 and 1163 were located in the S2 domain of S protein, suggesting that they should associate with membrane fusion, less probability with receptor recognition [11, 23]. The S2 domain contains two highly conserved heptad repeats HP1 and HP2, both of them form a six-helix bundle structure via hydrophobic interaction, facilitating membrane fusion [12, 31]. At site 1163 of HR2 region, the positively charged residue lysine replaced the negatively charged residue glutamic acid in 03-early-mid epidemic group, suggesting that residue lysine may be advantageous for S protein adaptation to human cellular receptors. The antiviral research showed that active peptide (residues 1161–1180) with residue1163E derived from animal virus possesses less inhibitory activity against human SARS-CoVs than its human virus counterpart of 1163K , which supports the view that the substitution K1163E contributes to adaptation of SARS-CoV to human cellular receptors. On the other hand, a strong neutralizing epitope containing residues 1055–1192 also implicated the possible role of residue 1163 in induction of neutralizing antibodies despite of little evidence available for its action on antibody escape [37, 38]. As for residue 778, uncharged amino acid tyrosine was fixed in the 03-late epidemic group by replacing the negatively charged residue aspartic acid. The importance of residue 778 cannot be assessed until further site-directed mutagenesis research is conducted. With respect to residues 75 and 239, they are unlikely to participate in the viral adaptation process due to their location outside RBD and S2 domain regions.
A total of 12 sites (75, 239, 244, 311, 479, 609, 613, 743, 765, 778, 1148 and 1163) in S protein were detected under positive selective pressure. Among them, 8 sites are exposed on the surface of S protein. It was also found that the S protein of SARS-CoV experiences variable positive selective pressures before reaching stabilization, and the positively selected sites are changeable in different epidemic groups. More importantly, a larger proportion of positively selected sites identified in 02–04 interspecies epidemic group was located in RBD region of S protein, suggesting that receptor binding function is predominant at this stage. On the other hand, more positively selected sites were located in HR1–HR2 region in 03-early-mid epidemic group, suggesting that the membrane fusion function becomes a major task in association with positive selection at this period. The variation of positive selective pressures and positively selected sites of S protein provide a valuable evidence for understanding the molecular adaptation of S protein from animals to humans.
A total of 102 Spike (S) gene sequences of SARS-CoVs and SARS-like-CoVs from human and animals were retrieved from the GenBank (Table S1, see additional file 2). Among them, 100 S gene sequences were obtained from five previously classified epidemic phases: 02–03, 03-early, 03-middle, 03-late and 03–04 epidemic phases, which represent the SARS outbreak sequence from later 2002 to early 2004 [7, 14]. The 02–03 epidemic phase contained SARS-like-CoV sequences from animals (03-pcSARS-CoV) . The strains of 03-early, 03-middle and 03-late phases were isolated from human (03-huSARS-CoV). The 03–04 epidemic phase represented the re-emergence of SARS in human patients (04-huSARS-CoV) and palm civets (04-pcSARS-CoV) . The other two S sequences of SARS-like-CoVs isolated from bats were used as the outgroup of phylogenetic tree [39, 40].
Phylogenetic analysis, tree construction and recombination analysis
After deletion of identical sequences, only 45 distinctive sequences out of 100 S sequences were used for phylogenetic analysis (Table S1, see additional file 2). They were aligned together with two outgroup sequences using CLUSTAL × (Ver. 1.83) . The phylogenetic tree of S gene was obtained by using ML (maximum likelihood) (PHYML v2.4.4)  and NJ (neighbor-joining) (MEGA 3.0)  methods, and the reliability of the trees was evaluated by the bootstrap method with 1,000 replications. The d N /d S value was used to detect positive selection. Since recombination of genes can result in artificially high d N /d S values and a false detection of positive selection [18, 19], the SimPlot Version 3.5.1  was applied to determine whether the recombination occurs in S gene of SARS-CoV.
Re-classification of SARS epidemic phases
The outbreak of SARS was previously divided into five epidemic phases: 02–03 epidemic, 03-early, 03-middle, 03-late, and 03–04 epidemic phases based on the epidemiological investigation [7, 14]. The tree topology showed that isolates from the 02–03 epidemic, 03-human epidemic and 03–04 epidemic phases formed a monophyletic clade, respectively (Fig. 1) . The clades formed by 03-pcSARS-CoVs and 04-pcSARS-CoVs clustered together, diverging from the clade formed by 03-huSARS-CoVs, which have been demonstrated being responsible for transmission of SARS-CoVs from animals such as palm civets to human [6, 7, 14]. These suggested that the origin of 03-huSARS-CoVs should be a virus strain, which was prevalence in palm civets or other animals before November 2002, but different from both 03-pcSARS-CoV and 04-pcSARS-CoV [6, 7, 25].
Because the 02–03 epidemic phase included SARS-CoV sequences from animals (palm civets and a raccoon dog) and the 03–04 epidemic phase contained 04-huSARS-CoV and 04-pcSARS-CoV, they should be at least partially reflect the course of viral interspecies transmission. In order to more realistically detect the adaptive evolution of S genes in interspecies transmission, we merged 02–03 and 03–04 epidemic phases into a unique epidemic group: 02–04 interspecies epidemic group, representing the course of SARS-CoV interspecies transmission. With respect to 03-human epidemic, it was divided into three phases in the previous studies. In the current study, it formed a monophyletic clade in phylogenetic tree with a low bootstrap value (41%) support (Fig. 1). In fact, the sequences from 03-early and 03-middle phases clustered together, and did not split each other. In addition, pairwise comparison of S gene sequences demonstrated that 03-early and 03-middle phases under positive selection, suggesting that both phases associate with the adaptation to human receptor . On the other hand, the 03-late epidemic phase experienced a longer epidemic time (more than four months) than the sum of 03-early and 03-middle epidemic phases (about three months), and showed a lower sequence divergence . We therefore divided 03-human epidemic into two groups: 03-early-mid and 03-late epidemic groups according to the epidemiological investigation and their epidemic time. Thus, the evolutionary process of S proteins during the whole SARS epidemic was simplified into three epidemic groups: 02–04 interspecies, 03-early-mid and 03-late epidemic groups in the present study (Fig. 1). Each group might represent a unique epidemic phase in the whole SARS outbreak including initial animal-to-human transmission phase, adaptation to human receptor phase and subsequent lineage fixation phase, respectively.
Analysis of the adaptive evolution and identification of positively selected sites
The program Codeml implemented in the PAML 3.14 b software package was used to investigate the adaptive evolution of S protein . A total of 45 aligned S gene sequences, isolated from the different epidemic groups, were selected to test whether they were under positive selection in the whole outbreak. Six models of codon substitution, M0 (one-ratio), M1a (NearlyNeutral), M2a (PositiveSelection), M3 (discrete), M7 (beta), and M8 (beta and ω) were used in the analysis . M0 assumes that all sites have the same ω ratio. M1a assumes two classes of sites in proteins in proportions p0 and p1 (1 – p0) with 0 < ω0 < 1 (purifying selection) and ω1 = 1 (neutral sites). M2a adds a proportion (p2) to account for a class of sites where w2 is estimated from the data and can be > 1. M3 uses a general discrete distribution with three site classes, with the proportions (p0, p1 and p2) and the ω ratios (ω0, ω1 and ω2) estimated from the data. M7 assumes a beta distribution (p, q) for 10 different ω ratios in the interval (0, 1). M8 adds an extra class of sites with positive selection (ω > 1) to the beta (M7) model [46, 47]. Therefore, the null models M0, M1a and M7 fix the ω ratios between 0 and 1, and do not allow the presence of positively selected sites. The alternative models M2a, M3 and M8 account for positive selection by using parameters, which estimate ω greater than 1, and allow for the variable ω along codon sequence.
Likelihood ratio test (LRT)  was performed for detecting the presence of positively selected sites by comparing the models which do not allow for positive selection with the models which allow for positive selection. The LRT was performed by taking twice the difference in log likelihood between nested models, and testing for significance using the χ2 distribution with the degrees of freedom (d.f.) equivalent to the difference in the number of parameters between models. If the LRT is significant, positive selection is inferred. In the present study, three LRTs (M0 vs. M3, M1a vs. M2a, and M7 vs. M8) were used to detect positive selection. The Bayes empirical Bayes (BEB) approach implemented in M2a and M8 was used to determine the positively selected sites by calculating the posterior probabilities (p) of ω classes for each site . The sites with high posterior probabilities (p > 0.95) coming from the class with ω > 1 were believed to be under positive selection .
Because positive selection on S gene was detected during the whole SARS outbreak, the sequences from three different epidemic groups were also analyzed using Codeml program to further specify the epidemic group in which positive selection occurred and the exact location of positively selected sites in S proteins. To further confirm positively selected sites identified by using program Codeml, three additional methods including single likelihood ancestor counting (SLAC), fixed effects likelihood (FEL), and random effects (REL) as implemented in the on-line DataMonkey package were employed [49, 50]. To detect positively selected sites, the 0.1 level of significance was used for both SLAC and FEL, and default significant level of Bayes factor > 50 was used for REL. In the process of analysis, the results of three epidemic groups were obtained. However, as a result of the large amount of data, which exceed the acceptability of DataMonkey package, we failed to obtain the results of 45 sequences, which represent the whole SARS outbreak. In order to reduce the influence of type-I error rate (false discovery rate) , the sites identified under positive selection at significant level by both Codeml program and REL methods in DataMonkey package were used to investigate their lineage fixation for the adaptation of SARS-CoV S protein from animals to human hosts.
The statistical analysis for the comparison of positively selected sites between different groups was performed using one-sided χ2 test or one-sided Fisher's exact test in GraphPad Prism version 4.03 for Windows demo.
We would like to express our thanks to several anonymous reviewers for their helpful comments on this work. This work was supported by the National Natural Science Foundation of China (No. 30600352), the Natural Science Foundation of Jiangsu Province, China (No. BK2006550), and the Startup Fund from Jiangsu University for Advanced Scholars (No. 2281270002).
- The World Health Organization (WHO). http://www.who.int/csr/sars/country/table2004_04_21/en/index.html
- Drosten C, Gunther S, Preiser W, van der Werf S, Brodt HR, Becker S, Rabenau H, Panning M, Kolesnikova L, Fouchier RA, Berger A, Burguiere AM, Cinatl J, Eickmann M, Escriou N, Grywna K, Kramme S, Manuguerra JC, Muller S, Rickerts V, Sturmer M, Vieth S, Klenk HD, Osterhaus AD, Schmitz H, Doerr HW: Identification of a novel coronavirus in patients with severe acute respiratory syndrome. N Engl J Med. 2003, 348: 1967-1976. 10.1056/NEJMoa030747.View ArticlePubMedGoogle Scholar
- Ksiazek TG, Erdman D, Goldsmith CS, Zaki SR, Peret T, Emery S, Tong S, Urbani C, Comer JA, Lim W, Rollin PE, Dowell SF, Ling AE, Humphrey CD, Shieh WJ, Guarner J, Paddock CD, Rota P, Fields B, DeRisi J, Yang JY, Cox N, Hughes JM, LeDuc JW, Bellini WJ, Anderson LJ: A novel coronavirus associated with severe acute respiratory syndrome. N Engl J Med. 2003, 348: 1953-1966. 10.1056/NEJMoa030781.View ArticlePubMedGoogle Scholar
- Marra MA, Jones SJ, Astell CR, Holt RA, Brooks-Wilson A, Butterfield YS, Khattra J, Asano JK, Barber SA, Chan SY, Cloutier A, Coughlin SM, Freeman D, Girn N, Griffith OL, Leach SR, Mayo M, McDonald H, Montgomery SB, Pandoh PK, Petrescu AS, Robertson AG, Schein JE, Siddiqui A, Smailus DE, Stott JM, Yang GS, Plummer F, Andonov A, Artsob H, Bastien N, Bernard K, Booth TF, Bowness D, Czub M, Drebot M, Fernando L, Flick R, Garbutt M, Gray M, Grolla A, Jones S, Feldmann H, Meyers A, Kabani A, Li Y, Normand S, Stroher U, Tipples GA, Tyler S, Vogrig R, Ward D, Watson B, Brunham RC, Krajden M, Petric M, Skowronski DM, Upton C, Roper RL: The Genome sequence of the SARS-associated coronavirus. Science. 2003, 300: 1399-1404. 10.1126/science.1085953.View ArticlePubMedGoogle Scholar
- Rota PA, Oberste MS, Monroe SS, Nix WA, Campagnoli R, Icenogle JP, Penaranda S, Bankamp B, Maher K, Chen MH, Tong S, Tamin A, Lowe L, Frace M, DeRisi JL, Chen Q, Wang D, Erdman DD, Peret TC, Burns C, Ksiazek TG, Rollin PE, Sanchez A, Liffick S, Holloway B, Limor J, McCaustland K, Olsen-Rasmussen M, Fouchier R, Gunther S, Osterhaus AD, Drosten C, Pallansch MA, Anderson LJ, Bellini WJ: Characterization of a novel coronavirus associated with severe acute respiratory syndrome. Science. 2003, 300: 1394-1399. 10.1126/science.1085952.View ArticlePubMedGoogle Scholar
- Guan Y, Zheng BJ, He YQ, Liu XL, Zhuang ZX, Cheung CL, Luo SW, Li PH, Zhang LJ, Guan YJ, Butt KM, Wong KL, Chan KW, Lim W, Shortridge KF, Yuen KY, Peiris JS, Poon LL: Isolation and characterization of viruses related to the SARS coronavirus from animals in southern China. Science. 2003, 302: 276-278. 10.1126/science.1087139.View ArticlePubMedGoogle Scholar
- Song HD, Tu CC, Zhang GW, Wang SY, Zheng K, Lei LC, Chen QX, Gao YW, Zhou HQ, Xiang H, Zheng HJ, Chern SW, Cheng F, Pan CM, Xuan H, Chen SJ, Luo HM, Zhou DH, Liu YF, He JF, Qin PZ, Li LH, Ren YQ, Liang WJ, Yu YD, Anderson L, Wang M, Xu RH, Wu XW, Zheng HY, Chen JD, Liang G, Gao Y, Liao M, Fang L, Jiang LY, Li H, Chen F, Di B, He LJ, Lin JY, Tong S, Kong X, Du L, Hao P, Tang H, Bernini A, Yu XJ, Spiga O, Guo ZM, Pan HY, He WZ, Manuguerra JC, Fontanet A, Danchin A, Niccolai N, Li YX, Wu CI, Zhao GP: Cross-host evolution of severe acute respiratory syndrome coronavirus in palm civet and human. Proc Natl Acad Sci U S A. 2005, 102: 2430-2435. 10.1073/pnas.0409608102.PubMed CentralView ArticlePubMedGoogle Scholar
- Buchholz UJ, Bukreyev A, Yang L, Lamirande EW, Murphy BR, Subbarao K, Collins PL: Contributions of the structural proteins of severe acute respiratory syndrome coronavirus to protective immunity. Proc Natl Acad Sci U S A. 2004, 101: 9804-9809. 10.1073/pnas.0403492101.PubMed CentralView ArticlePubMedGoogle Scholar
- Spiga O, Bernini A, Ciutti A, Chiellini S, Menciassi N, Finetti F, Causarono V, Anselmi F, Prischi F, Niccolai N: Molecular modelling of S1 and S2 subunits of SARS coronavirus spike glycoprotein. Biochem Biophys Res Commun. 2003, 310: 78-83. 10.1016/j.bbrc.2003.08.122.View ArticlePubMedGoogle Scholar
- Li W, Moore MJ, Vasilieva N, Sui J, Wong SK, Berne MA, Somasundaran M, Sullivan JL, Luzuriaga K, Greenough TC, Choe H, Farzan M: Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus. Nature. 2003, 426: 450-454. 10.1038/nature02145.View ArticlePubMedGoogle Scholar
- Li W, Zhang C, Sui J, Kuhn JH, Moore MJ, Luo S, Wong SK, Huang IC, Xu K, Vasilieva N, Murakami A, He Y, Marasco WA, Guan Y, Choe H, Farzan M: Receptor and viral determinants of SARS-coronavirus adaptation to human ACE2. EMBO J. 2005, 24: 1634-1643. 10.1038/sj.emboj.7600640.PubMed CentralView ArticlePubMedGoogle Scholar
- Bosch BJ, van der Zee R, de Haan CA, Rottier PJ: The coronavirus spike protein is a class I virus fusion protein: structural and functional characterization of the fusion core complex. J Virol. 2003, 77: 8801-8811. 10.1128/JVI.77.16.8801-8811.2003.PubMed CentralView ArticlePubMedGoogle Scholar
- Xiao X, Chakraborti S, Dimitrov AS, Gramatikoff K, Dimitrov DS: The SARS-CoV S glycoprotein: expression and functional characterization. Biochem Biophys Res Commun. 2003, 312: 1159-1164. 10.1016/j.bbrc.2003.11.054.View ArticlePubMedGoogle Scholar
- Chinese SARS Molecular Epidemiology Consortium: Molecular evolution of the SARS coronavirus during the course of the SARS epidemic in China. Science. 2004, 303: 1666-1669. 10.1126/science.1092002.View ArticleGoogle Scholar
- Yeh SH, Wang HY, Tsai CY, Kao CL, Yang JY, Liu HW, Su IJ, Tsai SF, Chen DS, Chen PJ: Characterization of severe acute respiratory syndrome coronavirus genomes in Taiwan: molecular epidemiology and genome evolution. Proc Natl Acad Sci U S A. 2004, 101: 2542-2547. 10.1073/pnas.0307904100.PubMed CentralView ArticlePubMedGoogle Scholar
- Shi P, Yu L, Fu YX, Huang JF, Zhang KQ, Zhang YP: Evolutionary implications of Avian Infectious Bronchitis Virus (AIBV) analysis. Cell Res. 2006, 16: 323-327. 10.1038/sj.cr.7310041.View ArticlePubMedGoogle Scholar
- Merl D, Escalante A, Prado R: Comparison of Bayesian, maximum likelihood and parsimony methods for detecting positive selection. 2005, http://www.ams.ucsc.edu/reports/ams2005-03.pdfGoogle Scholar
- Anisimova M, Nielsen R, Yang Z: Effect of recombination on the accuracy of the likelihood method for detecting positive selection at amino acid sites. Genetics. 2003, 164: 1229-1236.PubMed CentralPubMedGoogle Scholar
- Shriner D, Nickle DC, Jensen MA, Mullins JI: Potential impact of recombination on sitewise approaches for detecting positive natural selection. Genet Res. 2003, 81: 115-121. 10.1017/S0016672303006128.View ArticlePubMedGoogle Scholar
- Stavrinides J, Guttman DS: Mosaic evolution of the severe acute respiratory syndrome coronavirus. J Virol. 2004, 78: 76-82. 10.1128/JVI.78.1.76-82.2004.PubMed CentralView ArticlePubMedGoogle Scholar
- Holmes EC, Rambaut A: Viral evolution and the emergence of SARS coronavirus. Philos Trans R Soc Lond B Biol Sci. 2004, 359: 1059-1065. 10.1098/rstb.2004.1478.PubMed CentralView ArticlePubMedGoogle Scholar
- Galetto R, Negroni M: Mechanistic features of recombination in HIV. AIDS Rev. 2005, 7: 92-102.PubMedGoogle Scholar
- Qu XX, Hao P, Song XJ, Jiang SM, Liu YX, Wang PG, Rao X, Song HD, Wang SY, Zuo Y, Zheng AH, Luo M, Wang HL, Deng F, Wang HZ, Hu ZH, Ding MX, Zhao GP, Deng HK: Identification of two critical amino acid residues of the severe acute respiratory syndrome coronavirus spike protein for its variation in zoonotic tropism transition via a double substitution strategy. J Biol Chem. 2005, 280: 29588-29595. 10.1074/jbc.M500662200.View ArticlePubMedGoogle Scholar
- Bernini A, Spiga O, Ciutti A, Chiellini S, Bracci L, Yan X, Zheng B, Huang J, He ML, Song HD, Hao P, Zhao G, Niccolai N: Prediction of quaternary assembly of SARS coronavirus peplomer. Biochem Biophys Res Commun. 2004, 325: 1210-1214. 10.1016/j.bbrc.2004.10.156.View ArticlePubMedGoogle Scholar
- Zhao Z, Li H, Wu X, Zhong Y, Zhang K, Zhang YP, Boerwinkle E, Fu YX: Moderate mutation rate in the SARS coronavirus genome and its implications. BMC Evol Biol. 2004, 4: 21-10.1186/1471-2148-4-21.PubMed CentralView ArticlePubMedGoogle Scholar
- Drummond AJ, Pybus OG, Rambaut A, Forsberg R, Rodrigo AG: Measurably evolving populations. Trends Ecol Evol. 2003, 18: 481-488. 10.1016/S0169-5347(03)00216-7.View ArticleGoogle Scholar
- Drummond A, Forsberg R, Rodrigo AG: The inference of stepwise changes in substitution rates using serial sequence samples. Mol Biol Evol. 2001, 18: 1365-1371.View ArticlePubMedGoogle Scholar
- Baranowski E, Ruiz-Jarabo CM, Domingo E: Evolution of cell recognition by viruses. Science. 2001, 292: 1102-1105. 10.1126/science.1058613.View ArticlePubMedGoogle Scholar
- Kaye M: SARS-associated coronavirus replication in cell lines. Emerg Infect Dis. 2006, 12: 128-133.PubMed CentralView ArticlePubMedGoogle Scholar
- Wong SK, Li W, Moore MJ, Choe H, Farzan M: A 193-amino acid fragment of the SARS coronavirus S protein efficiently binds angiotensin-converting enzyme 2. J Biol Chem. 2004, 279: 3197-3201. 10.1074/jbc.C300520200.View ArticlePubMedGoogle Scholar
- Xu Y, Lou Z, Liu Y, Pang H, Tien P, Gao GF, Rao Z: Crystal structure of severe acute respiratory syndrome coronavirus spike protein fusion core. J Biol Chem. 2004, 279: 49414-49419. 10.1074/jbc.M408782200.View ArticlePubMedGoogle Scholar
- Li F, Li W, Farzan M, Harrison SC: Structure of SARS coronavirus spike receptor-binding domain complexed with receptor. Science. 2005, 309: 1864-1868. 10.1126/science.1116480.View ArticlePubMedGoogle Scholar
- Sui J, Li W, Murakami A, Tamin A, Matthews LJ, Wong SK, Moore MJ, Tallarico AS, Olurinde M, Choe H, Anderson LJ, Bellini WJ, Farzan M, Marasco WA: Potent neutralization of severe acute respiratory syndrome (SARS) coronavirus by a human mAb to S1 protein that blocks receptor association. Proc Natl Acad Sci U S A. 2004, 101: 2536-2541. 10.1073/pnas.0307140101.PubMed CentralView ArticlePubMedGoogle Scholar
- Zheng BJ, Guan Y, Hez ML, Sun H, Du L, Zheng Y, Wong KL, Chen H, Chen Y, Lu L, Tanner JA, Watt RM, Niccolai N, Bernini A, Spiga O, Woo PC, Kung HF, Yuen KY, Huang JD: Synthetic peptides outside the spike protein heptad repeat regions as potent inhibitors of SARS-associated coronavirus. Antivir Ther. 2005, 10: 393-403.PubMedGoogle Scholar
- Jeffers SA, Tusell SM, Gillim-Ross L, Hemmila EM, Achenbach JE, Babcock GJ, Thomas WDJ, Thackray LB, Young MD, Mason RJ, Ambrosino DM, Wentworth DE, Demartini JC, Holmes KV: CD209L (L-SIGN) is a receptor for severe acute respiratory syndrome coronavirus. Proc Natl Acad Sci U S A. 2004, 101: 15748-15753. 10.1073/pnas.0403812101.PubMed CentralView ArticlePubMedGoogle Scholar
- Marzi A, Gramberg T, Simmons G, Moller P, Rennekamp AJ, Krumbiegel M, Geier M, Eisemann J, Turza N, Saunier B, Steinkasserer A, Becker S, Bates P, Hofmann H, Pohlmann S: DC-SIGN and DC-SIGNR interact with the glycoprotein of Marburg virus and the S protein of severe acute respiratory syndrome coronavirus. J Virol. 2004, 78: 12090-12095. 10.1128/JVI.78.21.12090-12095.2004.PubMed CentralView ArticlePubMedGoogle Scholar
- Zhang H, Wang G, Li J, Nie Y, Shi X, Lian G, Wang W, Yin X, Zhao Y, Qu X, Ding M, Deng H: Identification of an antigenic determinant on the S2 domain of the severe acute respiratory syndrome coronavirus spike glycoprotein capable of inducing neutralizing antibodies. J Virol. 2004, 78: 6938-6945. 10.1128/JVI.78.13.6938-6945.2004.PubMed CentralView ArticlePubMedGoogle Scholar
- Keng CT, Zhang A, Shen S, Lip KM, Fielding BC, Tan TH, Chou CF, Loh CB, Wang S, Fu J, Yang X, Lim SG, Hong W, Tan YJ: Amino acids 1055 to 1192 in the S2 region of severe acute respiratory syndrome coronavirus S protein induce neutralizing antibodies: implications for the development of vaccines and antiviral agents. J Virol. 2005, 79: 3289-3296. 10.1128/JVI.79.6.3289-3296.2005.PubMed CentralView ArticlePubMedGoogle Scholar
- Lau SK, Woo PC, Li KS, Huang Y, Tsoi HW, Wong BH, Wong SS, Leung SY, Chan KH, Yuen KY: Severe acute respiratory syndrome coronavirus-like virus in Chinese horseshoe bats. Proc Natl Acad Sci U S A. 2005, 102: 14040-14045. 10.1073/pnas.0506735102.PubMed CentralView ArticlePubMedGoogle Scholar
- Li W, Shi Z, Yu M, Ren W, Smith C, Epstein JH, Wang H, Crameri G, Hu Z, Zhang H, Zhang J, McEachern J, Field H, Daszak P, Eaton BT, Zhang S, Wang LF: Bats are natural reservoirs of SARS-like coronaviruses. Science. 2005, 310: 676-679. 10.1126/science.1118391.View ArticlePubMedGoogle Scholar
- Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG: The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997, 25: 4876-4882. 10.1093/nar/25.24.4876.PubMed CentralView ArticlePubMedGoogle Scholar
- Guindon S, Lethiec F, Duroux P, Gascuel O: PHYML Online--a web server for fast maximum likelihood-based phylogenetic inference. Nucleic Acids Res. 2005, 33: W557-9. 10.1093/nar/gki352.PubMed CentralView ArticlePubMedGoogle Scholar
- Kumar S, Tamura K, Nei M: MEGA3: Integrated software for Molecular Evolutionary Genetics Analysis and sequence alignment. Brief Bioinform. 2004, 5: 150-163. 10.1093/bib/5.2.150.View ArticlePubMedGoogle Scholar
- Lole KS, Bollinger RC, Paranjape RS, Gadkari D, Kulkarni SS, Novak NG, Ingersoll R, Sheppard HW, Ray SC: Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination. J Virol. 1999, 73: 152-160.PubMed CentralPubMedGoogle Scholar
- Yang Z: PAML: a program package for phylogenetic analysis by maximum likelihood. Comput Appl Biosci. 1997, 13: 555-556.PubMedGoogle Scholar
- Yang Z, Nielsen R, Goldman N, Pedersen AM: Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics. 2000, 155: 431-449.PubMed CentralPubMedGoogle Scholar
- Yang Z, Wong WS, Nielsen R: Bayes empirical bayes inference of amino acid sites under positive selection. Mol Biol Evol. 2005, 22: 1107-1118. 10.1093/molbev/msi097.View ArticlePubMedGoogle Scholar
- Anisimova M, Bielawski JP, Yang Z: Accuracy and power of the likelihood ratio test in detecting adaptive molecular evolution. Mol Biol Evol. 2001, 18: 1585-1592.View ArticlePubMedGoogle Scholar
- Kosakovsky Pond SL, Frost SD: Not so different after all: a comparison of methods for detecting amino acid sites under selection. Mol Biol Evol. 2005, 22: 1208-1222. 10.1093/molbev/msi105.View ArticlePubMedGoogle Scholar
- Kosakovsky Pond SL, Frost SD: Datamonkey: rapid detection of selective pressure on individual sites of codon alignments. Bioinformatics. 2005, 21: 2531-2533. 10.1093/bioinformatics/bti320.View ArticleGoogle Scholar
- Guindon S, Black M, Rodrigo A: Proceedings of the SMBE Tri-National Young Investigators' Workshop 2005. Control of the false discovery rate applied to the detection of positively selected amino acid sites. Mol Biol Evol. 2006, 23: 919-926. 10.1093/molbev/msj095.View ArticlePubMedGoogle Scholar