1 Emerging infectious diseases (EID) are continuing to pose threat to public health
In the past 50 years, the frequency and numbers of EID is growing rapidly all over the world and more than 70% of them are zoonoses, either from wild animals or transmitted by invertebrate vectors [
1,
2]. Generally, for wildlife-borne viruses, several steps are experienced before they are successfully transmitted to humans. Step 1, virus enters to animal populations (intermediate hosts) which have ecological overlapping niches with humans. Step 2, virus adapts to the new host through mutations in the genome and proteins. Step 3, the adapted virus enters to human populations which have close contact with the intermediate hosts. Step 4, virus adapts to humans through further mutations in the genome and proteins and becomes prevalent in human populations [
3,
4]. Virus can also enter directly from their natural reservoirs to human populations which are highly exposed to wild animals, which have occurred for Marburg virus in Uganda [
5] and Kenya [
6] and for Nipah virus in Bangladesh [
7], but most of them are abortive infection in this pathway.
In the past 20 years, three EID events caused by coronaviruses broke out in humans. They are severe acute respiratory syndrome coronavirus (SARS-CoV) [
8], Middle East respiratory syndrome coronavirus (MERS-CoV) [
9] and SARS-CoV-2 [
10–
12]. SARS-CoV and SARS-CoV-2 belong to the same species,
severe acute respiratory syndrome-related coronavirus (SARSr-CoV),
Sarbecovirus subgenus. SARS-CoV emerged at the end of November, 2002, in Guangdong Province, China, and appeared in 29 countries and regions in following three months [
13]. SARS was rapidly contained after 7 months in 2003 and caused 30–50 billion US dollars of economic losses in the world [
14]. The COVID-19 is the most severe EID in the past 100 years, its impact overwhelms all the world not only in public heath, but also in global economy, human behavior, culture and politics. Furthermore, SARS-CoV-2 is under rapid mutations due to large numbers of infections in the countries where a flexible quarantine policy is implemented. With the global efforts, great progress has been achieved in understanding the SARS-CoV-2, including the biological features, diagnostic technologies, vaccine and drug development, natural reservoirs and host ranges. However, how the virus was transmitted to humans is still unclear. In this brief review, we will revisit the SARS epidemic and its origins and focus on the epidemiology data of the early stage of COVID-19, the potential natural ancestor of SARS-CoV-2 and its possible transmission chain. We will also discuss the interspecies transmission mechanism in term of receptor utilization and call for the preparedness for the future prevention strategy and research directions.
2 Zoonotic origins of SARS-CoV
After SARS outbreak, the epidemiology data indicates that most of early cases had contact with wildlife directly or indirectly and were independently emerged in different regions of the province. It reminds the scientists to focus on the wildlife source in the wet markets and farms and collected the samples before the markets and farms were closed. With the identification of the causative agent SARS-CoV and their detection methods, scientists found that the SARS-CoV had circulated in the wet market animals for some time and infected some animal traders before the SARS outbreak [
15]. The comprehensive analyses based on the full-length SARS-CoV genome sequences detected from humans and animals further confirmed that the SARS-CoV was directly transmitted from wild animals traded in the wet markets to humans. Furthermore, it was observed that the viral spike proteins and an accessory protein ORF8 undergone rapid mutations during the virus transmission from animals to humans and between humans and humans [
16,
17]. It is worth of noting that the sample collection at very early stage of SARS outbreak play a key role in resolving the direct sources of SARS-CoV. However, it takes scientists 8 years to find the closest ancestor of SARS-CoV in
Rhinolophus species in Yunnan, a province of south-west China [
18]. During the search for the natural reservoirs of SARS-CoV, highly genetically diverse SARSr-CoVs were discovered in different
Rhinolophus species in Asian, European, and African countries (Fig.1) [
19–
22]. It was demonstrated that some bat SARSr-CoVs possess the capability to infect humans and animals based on the receptor analyses, cell and animal experimental infection and serological investigation [
18,
19,
23,
24]. It needs to note that bat SARSr-CoVs which use the same receptor, angiotensin-converting enzyme 2 (ACE2), have 10-fold less affinity to human ACE2 [
24]. It means that the SARS-CoV ancestor from bats must acquire further mutations to successfully adapt to humans. Those mutations were acquired during transmission in animals and humans observed by molecular epidemiology studies. Nevertheless, it is still a mystery how the SARS-CoV appeared in Guangdong Province as the SARSr-CoVs which use the human ACE2 have never been detected in other provinces except Yunnan through more than 15 years of surveillances. SARS-CoV disappeared in humans after the close-down of wet markets in 2004. No SARS-CoV was detected in either farmed or wild civets after 2004. Thus the most reasonable explanation is that its ancestor was accidentally transmitted to farmed animals whose location overlapped with the bat home range and which were then carried to wet markets in Guangdong Province (Fig.2).
3 COVID-19 emergence and the early epidemic in Wuhan
The emergence of COVID-19 was first noted in Huanan Seafood Wholesale Market, Wuhan, China, in late December 2019 when clusters of cases with unexplained pneumonia were reported by several health facilities [
25]. In early January 2020, by metagenomic sequencing and virus isolation from clinical samples, the causative agent of COVID-19 was rapidly identified to be a novel coronavirus belonging to the
Sarbecovirus subgenus, which shares 79% genome sequence identity with SARS-CoV and was later officially named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [
10,
11,
26,
27].
The Huanan Seafood Wholesale Market, which sold mainly aquatic products but also wild animal products, was initially regarded as a major early epicenter of the COVID-19 outbreak in Wuhan, as many cases during the first weeks of the epidemic were linked to this market including the vendors and dealers [
25,
28]. However, retrospective studies revealed that the number of early cases associated with other markets were similar to those associated with Huanan Market. Notably, the one reported with the earliest onset on December 8 had no link to the Huanan Market. Moreover, 44.6% of the cases with onset in December 2019 did not have any market exposure history [
29]. Analysis of the viral genomes sequenced from early COVID-19 patients suggested that there were two major SARS-CoV-2 lineages with different exposure history circulating contemporaneously during the early stage of outbreak in Wuhan, which may have emerged independently from a common ancestor in November 2019 [
28,
30]. According to the final report by the joint WHO–China study of SARS-CoV-2 origin, it is considered that an explosive outbreak started in Wuhan in early December 2019. While cases with more severe manifestation were identified by the healthcare system, other milder or asymptomatic cases were left unrecognized [
29]. The underestimated rate of asymptomatic infection and undocumented secondary transmission within wider community may account for those early cases without epidemiological link to the markets [
28].
4 Possible transmission chain of SARS-CoV-2
Genomic characterization of SARS-CoV-2 revealed that it is closely related to a coronavirus from
Rhinolophus bats, named RaTG13 [
10]. More recently, a coronavirus more closely related to SARS-CoV-2 than RaTG13 has been identified in a bat from Laos [
31]. Discovery of coronaviruses related to SARS-CoV-2 in pangolins were also reported [
32]. These findings suggest the likelihood of the SARS-CoV-2 emergence as a zoonotic event. However, unlike SARS-CoV, the possible source of which for human infection was identified to be civets in the markets [
15], the direct source of SARS-CoV-2 is still unclear.
4.1 SARS-CoV-2 detected in environmental samples
During the outbreak investigation following the closure of the Huanan Market, extensive testing of SARS-CoV-2 was conducted in environmental and animal product samples collected in the market. Over 400 animal-related samples were tested, including frozen farmed wildlife carcasses in the market and goods kept in the warehouses related to the market. None of those retrieved samples tested positive for SARS-CoV-2 RNA [
29,
33]. About 600 animal samples from the upstream supplier of the Huanan Market was surveyed as well, and all tested negative. There is no evidence for the presence of SARS-CoV-2 among animals traded in the Huanan Market [
29]. At the same time, testing of SARS-CoV-2 nucleic acid in 923 environmental samples in the Huanan Market showed a 7.9% positive rate, demonstrating widespread spread of SARS-CoV-2 in the market environment [
33]. However, it cannot be judged how the virus was introduced into the market [
29]. Furthermore, molecular epidemiological studies on early cases revealed that the virus lineage associated with the Huanan Market cluster was not the sole lineage existing in Wuhan then, implying the possibility of other transmission chains [
28–
30,
34]. All these findings, together with the epidemiological fact that a large part of early cases did not have exposure history to Huanan Market, suggest that the market may not be the source of the outbreak but an amplifier of the early epidemic.
4.2 SARS-CoV-2 detected in cold-chain products
SARS-CoV-2 can survive and maintain infectivity for long periods of time on the surfaces of frozen food such as meat, as well as on its outer packaging surface, under conditions of cold chain transportation [
35,
36]. After the outbreak in Wuhan, it has been proven that several more recent sporadic resurgences of COVID-19 in China, including those occurring in Beijing, Dalian, and Qingdao from April to October 2020, likely originated from the exposure to imported frozen food and food packaging contaminated by SARS-CoV-2 [
37–
39]. Live virus of SARS-CoV-2 was successfully isolated using the swab samples of the outside surface of imported frozen cod packages [
40]. These observations indicate that the risk of SARS-CoV-2 transmission caused by cold-chain products through long-distance transportation cannot be neglected [
40,
41]. Given that frozen seafood and meat products were commonly for sale at the Huanan Market and other markets, the possibility that SARS-CoV-2 was introduced into Wuhan via cold-chain products may not be ruled out.
4.3 SARS-CoV-2 detected in archived samples collected before the COVID-19 outbreak outside of China
Although Wuhan was the place where the outbreak of COVID-19 was first observed and reported, evidence for even earlier occurrence of SARS-CoV-2 outside China were also described by different studies. In separate retrospective surveys, SARS-CoV-2 RNA was found in sewage samples collected in March 2019 from Spain, and in late November 2019 from Brazil [
42,
43]. Reactivity to SARS-CoV-2 was detected by
in situ hybridization in skin biopsies from an Italian lady sampled in November 2019 [
44]. Serological surveys revealed prevalence of neutralizing antibodies of SARS-CoV-2 among populations in Italy and France before December 2019 [
45,
46]. These findings from different countries suggest the potential missed circulation of SARS-CoV-2 preceding its emergence in Wuhan. Another study analyzing the haplotypes of SARS-CoV-2 mutants in early 2020 revealed that a group of 4 mutations that occurred commonly in Europe from the beginning of the pandemic was absent in early Asian samples. The authors then hold the opinion that the European strains likely have acquired the mutations prior to the pandemic and spread in parallel with the Asian strains for a period [
47]. Based on the global SARS-CoV-2 genome data in public database, the time to the most recent ancestor of those SARS-CoV-2 genomic sequences were estimated to be between late September and early December 2019 using various modeling approaches [
48–
50]. However, these results shall not be interpreted as the definitive answer about the time of its origins. When SARS-CoV-2 first infected humans cannot be inferred with the information available at this point.
Being highly transmissible among humans, SARS-CoV-2 appears to have been well-adapted to humans even at the very beginning of the outbreak. It is argued by some evolution scientists that SARS-CoV-2 has undergone an adaptive shift from its animal hosts to humans in prior to the onset of the COVID-19 pandemic, which was a long-term and complex process involving step-by-step selection. Such adaptive evolution was unlikely to happen in an urban wet market in a brief period of time [
51]. In this view, the places where a virus was first detected or it caused the first epidemic is not necessarily the place where it originated [
52].
5 Animals susceptible to SARS-CoV-2
5.1 Domestic animals infected by SARS-CoV-2 from humans
The concern is growing over time about the host range of SARS-CoV-2, especially in light of the reported reverse zoonosis events, or transmission from humans to animals. The first case of SARS-CoV-2 natural infection in animals was reported in Hong Kong, which showed that 2 out of 15 dogs from the COVID-19 confirmed owners were PCR-positive for the SARS-CoV-2 [
53]. After that, other reports showed similar cases of canids being infected with SARS-CoV-2 after exposure to COVID-19 positive dog owners [
54–
56]. In addition, another popular pet animal, cat, has also been documented to be infected with SARS-CoV-2, some of the infected cats showed clinical signs, such as sneezing, nasal discharge, and anorexia so on, while some of them were asymptomatic [
57–
63]. A study also showed 15 out of 102 cats were seropositive for SARS-CoV-2 after the outbreak of COVID-19 in Wuhan, but the serum from the sera bank collected before the outbreak was negative, suggesting the cats might be infected naturally after exposure to COVID-19 patients [
64]. The captive wild animal such as lions, tigers, and gorillas have also been demonstrated to acquire infection of SARS-CoV-2 in the zoo [
65–
67]. Some of the tigers and lions showed signs of respiratory illness and one of the tigers was asymptomatic. The characterization of SARS-CoV-2 identified from tigers suggested the human–animal transmission between tigers and zookeepers. But the viral source of the lions and gorillas was unclear [
66,
68].
Other species of animals which have been reported infection by SARS-CoV-2 naturally was mink. There are several cases that have been reported in the farmed mink from Europe and America [
69–
73]. For example, in farms in the Netherlands, over 20 000 minks were tested positive for SARS-CoV-2 after suffering gastrointestinal and respiratory diseases. And at least one of the farmers was positive for COVID-19 in these farms. The viral sequences identified from minks suggested the cases were related to human-to-animal infection and the mink-to-mink transmission increased the spread of the virus. But the viral sequences of mink in different farms implied they were introduced by different sources. Besides that, researchers have identified some mutations in the viral sequences of minks, which suggested the virus has experienced adaptation to this new host [
69–
71]. It also provided a few examples currently that the natural infection animals maintained the viruses and experienced the adaptive evolution in their population, then spillback to humans.
These cases we discussed above suggested that the SARS-CoV-2 could be transmitted from humans to animals. While the large-scale and widespread distribution of SARS-CoV-2 infection in wildlife may be negligible because most of evidence found so far showed that they were in close contact with SARS-CoV-2 positive human beings. And the high population density of these animals in the farm or zoo may not happen in nature or the wild. But what we must take into account is, if the maintenance and sustained transmission were established in these species, the virus may experience natural selections in their population, like the mutations we found in minks and the different variants we have identified in humans. If the selected variants achieve a better fitness in these species, they may exist in their populations for the long term. And these species may become the intermediate hosts or a route of transmission to humans of SARS-CoV-2.
5.2 Animals susceptible to SARS-CoV-2 by experimental infection
In addition to the natural infection in animals we discussed above, the animal model is also an important tool to evaluate the host range of SARS-CoV-2. The experimental evidence of SARS-CoV-2 infection has been reported in a variety of mammals, such as monkeys, mice, ferrets, hamsters, cats, pigs, and so on [
74–
83]. The main purpose of these studies is to assess the pathogenesis of SARS-CoV-2 or find a suitable model for evaluating the vaccines and antivirals against SARS-CoV-2 infection. The animals used in current studies are mainly focused on classical experimental animal models, such as the nonhuman primates and rodents.
The nonhuman primates are closely related to humans and could reflect the similar clinical signs and pathology observed in humans, so they are widely used in human infectious disease research. The nonhuman primates used in the SARS-CoV-2 study include rhesus macaques (
Macaca mulatta), cynomolgus monkeys (
Macaca fascicularis), African green monkeys (
Chlorocebus sabaeus), and so on. They are all susceptible to SARS-CoV-2 infection and develop clinical symptoms such as fever, diarrhea, and pneumonitis like humans [
75,
76,
81,
82,
84].
Mice are the other animal models that are widely used in SARS-CoV-2 studies. Because of the inefficient receptor usage of SARS-CoV-2 in mouse ACE2, it often needs to be modified [
76,
79,
80]. Although the mice are not very susceptible to SARS-CoV-2 naturally, one study has reported that the mouse-adapted SARS-CoV-2 strains with a higher binding affinity to mouse ACE2 were obtained after several passaging in mice [
85]. Besides, natural selective SARS-CoV-2 variants that contain the N501Y mutation in the spike protein, such as the Alpha, Beta, and Gamma, have also been reported to infect wild mice [
86,
87]. Compared to mice, the hamsters are more permissive to SARS-CoV-2 infection. High levels of virus replication and histopathology have been reported in the nasal mucosa, respiratory, and small intestine of hamsters after inoculation with SARS-CoV-2 [
88–
91].
The ferrets are also susceptible to SARS-CoV-2 infection in laboratory studies. They could be infected by SARS-CoV-2 both via direct contact and the air [
74,
83]. So they might transmit the virus to other animals and could be used for the studies of SARS-CoV-2 transmission. Dogs and cats have also been used in experimental studies of SARS-CoV-2 [
74]. Although they were reported susceptible to SARS-CoV-2 naturally, the experimental infection results showed that the dogs have a lower susceptibility to SARS-CoV-2 than cats. The cats could be infected when they are contacted with other infected cats that shed viral RNA in the faces [
92]. The poultry such as chicken and ducks were not permissive to SARS-CoV-2 infection in experimental studies which is not surprising because of the incompatible primary receptor usage of SARS-CoV-2 in their ACE2 [
74,
93,
94]. In opposition, the ACE2 of pigs has been shown to mediate the entry of SARS-CoV-2 in cell level and bind to the RBD of SARS-CoV-2 spike in the protein–protein binding assay [
10,
94]. But one experimental infection study showed that there were no detected virus replication or antibodies in pigs after inoculating them with SARS-CoV-2 [
95]. And the other study displayed a conflicting result that viral RNA was detected from two pigs and the live virus was isolated from 1 pig after the orinasal inoculation [
95]. The latter study was also supported by the
in vitro infectivity studies and biochemical experiment [
10,
96]. Similar cases have also been shown in Chinese tree shrews. One study reported that the Chinese tree shrews were susceptible to SARS-CoV-2 and developed COVID-19-like symptoms after inoculation [
97], while another paper demonstrated that the Chinese tree shrews were not permissive to SARS-CoV-2 as other animal models [
98]. The inconsistent phenotype in experimental studies may be impacted by the dose, the strains, and the difference in the direct or indirect inoculation manner.
5.3 Animal susceptible to SARS-CoV-2 analyzed by receptor utilization in vitro
In addition to the natural and experimental infection by SARS-CoV-2, protein–protein interaction assays and pseudovirus entry tests were also used for predicting the host ranges of SARS-CoV-2. The results are consistent with those observed above that SARS-CoV-2 can infect wide-ranged animals covering the families Primates, Lagomorpha, Carnivora, Perissodactyla, Pholidota, Artiodactyla [
93,
94,
96,
99] (Tab.1). The binding assay between the SARS-CoV-2 spike receptor binding domain (RBD) and ACE2 derived from different hosts showed that SARS-CoV-2 has a relatively high binding affinity to ACE2 from cat, dog, rabbit, goat, sheep, cattle, bovine, pig, and so on [
93,
94]. In sum, the domestic species which are susceptible to SARS-CoV-2 infection highlight the necessity to monitor the livestock. Despite that the experimental data and biochemical analyses may not reflect the actual dynamics of the SARS-CoV-2 in natural conditions, the large population numbers and close contact with humans make them pose potential risks in the maintenance and spread of SARS-CoV-2. Furthermore, the current circulating SARS-CoV-2 variants which maintain multiple mutations in the spike protein such as Delta and Omicron were selected in the human populations. The higher transmissibility and immune escape are the major concerns of those variants right now. Given that SARS-CoV-2 is keeping on mutating, if those variants experience natural selection in the animals and obtain new mutations, then the reverse transmission to humans will be another concern. Therefore, the extensive surveillances for wildlife or domestic animals are required to prevent the SARS-CoV-2 circulation between animals and humans.
6 Detection of SARS-CoV-2-like viruses in wildlife
Since early 2020, the heating up hunt for the ancestor of SARS-CoV-2 has driven the discovery of an increasing number of SARS-CoV-2-like coronaviruses in diverse
Rhinolophus species from China and South-east Asian countries including Thailand, Cambodia, and Laos (Fig.1) [
100–
104]. Most of these bat coronaviruses shared higher than 91% genome sequence identities with SARS-CoV-2. The bat coronavirus RaTG13 from Yunnan Province of China, whose genome sequence was 96.2% identical to SARS-CoV-2, was once recognized as the closest known relative of SARS-CoV-2 [
10]. Nevertheless, this record has been updated by novel bat sarbecoviruses recently discovered in North Laos (Fig.1) [
31]. One of the Lao bat coronaviruses detected in
Rhinolophus malayanus, named BANAL-20-52, had 96.8% genome sequence identity to SARS-CoV-2. The S protein of BANAL-20-52 and another SARS-CoV-2-like strain termed BANAL-20-236 showed strikingly high similarity to that of SARS-CoV-2, differing from SARS-CoV-2 by one and two residues respectively in the receptor binding motif (RBM). Studies have demonstrated its capability of efficiently binding to the human ACE2 and mediating the hACE2-dependent entry into human cells, which is not possessed by RaTG13 [
31]. Further pseudovirus-based ACE2 utilization assays suggested that the bat BANAL coronaviruses from Laos may have broad host range and potential of zoonotic transmission [
105]. Although the furin-like cleavage site of SARS-CoV-2 is absent in the spike of both RaTG13 and BANAL coronaviruses, similar insertion was observed in some SARS-CoV-2-like coronavirus of bat origin such as RmYN02 from China and RacCS203 from Thailand, suggesting that acquisition of the furin cleavage site may take place in nature [
100,
102]. In addition to these strains circulating in Yunnan and South-east Asia, there are a few bat SARSr-CoVs more closely related to SARS-CoV in the RNA-dependent RNA polymerase (RdRp) gene but more similar to SARS-CoV-2 on the full-length genome level, such as Rc-o319 and ZC45 reported from Japan and Zhejiang, China, respectively (Fig.1). Collectively, these findings indicate a wide distribution of SARS-CoV-2-lineage sarbecoviruses among bat populations in Asia. In addition, mutation signature analysis found that the mutation spectrum of SARS-CoV-2 was similar to several bat coronaviruses, including RaTG13, BANAL-20-52, as well as RshSTT182 from Cambodia, but it was distinct from that evolving in non-bat hosts, suggesting that SARS-CoV-2 and these bat viruses may have shared the same host environment before the outbreak. Such host environment, however, is difficult to duplicate by an arbitrary cell culture condition set in a laboratory. These results provide further proof of the natural origin of SARS-CoV-2 [
106]. In sum, more and more evidence show that bats are important natural reservoirs of the progenitor of SARS-CoV-2.
Besides bats, pangolins are another wildlife that intrigued concern of scientists searching for the natural origin of SARS-CoV-2. Two sub-lineages of SARS-CoV-2-like coronaviruses have been found in pangolins that were trafficked from South-east Asia and confiscated by Chinese customs at ports of Guangdong and Guangxi provinces [
32,
107,
108]. One sub-lineage of pangolin coronaviruses was highly similar to SARS-CoV-2 in the RBD, but the overall genome of both pangolin coronaviruses were phylogenetically more distant from SARS-CoV-2 than most of the bat SARS-CoV-2-like viruses. Like SARS-CoV and MERS-CoV, SARS-CoV-2 may have jumped from its natural hosts into humans via an intermediate host. Normally, the virus carried by an intermediate host is a more direct progenitor of the pandemic virus, and it is closer to the human virus than its original version in natural hosts. Therefore, the pangolin coronaviruses known to date, which share 85.5% to 90.3% genome sequence identities with SARS-CoV-2, do not firmly support the idea that pangolins are the intermediate host of SARS-CoV-2, but suggest that these animals may be another natural host of SARS-CoV-2-like coronaviruses. In China, a metagenomic study on 1725 game animals from 16 mammalian species across 19 provinces, including pangolins, civets, badgers, porcupines, etc., did not detect any SARS-CoV-2-like virus sequences [
109]. Large-scale retrospective surveys using both molecular and serological testing on livestock, poultry, and captive wildlife sampled before 2020 in China gained no evidence of SARS-CoV-2 infection [
29]. So far, the potential intermediate hosts of SARS-CoV-2 remain elusive.
7 Hypothesis of SARS-CoV-2 emergence
Although we are still far away from drawing a very clear picture about how SARS-CoV-2 emerged, we can make a hypothesis based on what we have learnt. The genetic distance between SARS-CoV-2 and its closest relatives in the bat reservoirs represent decades of evolution, suggesting a missing link of a more direct progenitor in an unidentified intermediate host. The ancestor of SARS-CoV-2 may first spill over to the intermediate hosts, spread within their populations during which it acquired mutations that made it better fit humans and evolved into the precursor of the pandemic virus, and was then transmitted to humans (Fig.2). Such intermediary step may involve more than one animal species. Those domesticated wild animals farmed in high density and with sufficient chances of human–animal interface have a higher likelihood to be the intermediate hosts [
29]. At present, we are unable to determine whether the first zoonotic transmission of SARS-CoV-2 to humans was associated with the markets. There is also a viewpoint that SARS-CoV-2 could have got into humans much earlier than the onset of its outbreak [
51]. It may have accumulated adaptive mutations by multi-step evolution in human populations and shuttling between humans and animal hosts for long time, and finally developed its extreme adaptiveness to humans. It may have then spread to Wuhan from elsewhere by chance, and the superspreading events occurring at the market caused an epidemic. Lastly, the possibility also exists that the virus was introduced to Wuhan through contaminated cold-chain food products (Fig.2).
8 SARSr-CoVs in wildlife are capable of utilizing human ACE2 and have potential spillover risk in the future
Virus binding to the host cell surface is the initial and critical step for successful infection and interspecies transmission. SARSr-CoVs are highly divergent in spike proteins which are responsible for receptor binding and virus entry. They can be divided into two major clades compared with the SARS-CoV spike RBD (Fig.3). Clade 1 includes four subclades: 1a (SARS-CoV), 1b (SARS-CoV-2), 1c, and 1d. Both subclades 1c and 1d have one deletion (deletion 1), while 1a and 1b subclades do not contain any deletions except for the RshSTT182 and Rc-o319 strains in the SARS-CoV-2 subclade, which contain a similar deletion 1 and deletion 2, respectively. Clade 2 has two deletions (deletion 1 and deletion 2). Previous studies revealed that the viruses in clade 1 without any deletions all utilize human ACE2 but with different efficiency (Tab.1) [
24,
31,
103,
110–
112]. SARS-CoV-2 has the highest binding affinity among all tested SARSr-CoVs, followed by bat SARSr-CoVs found in Laos, pangolin CoVs and SARS-CoV. The SARS-CoV lineage found in bats in China showed 10-fold lower binding affinity to human ACE2 than SARS-CoV [
24]. Although bat SARSr-CoV RaTG13 shares 96.2% identify with SARS-CoV-2 in genome level, it displayed much lower binding affinity to and entry efficiency in human ACE2 [
103]. In addition, the spike RBDs of some bat and pangolin SARSr-CoVs can bind other mammalian ACE2, suggesting a wide host range of these viruses [
103,
111–
115].
The bat SARSr-CoVs, which contain one deletion in the RBD, such as the RaTG15 identified from China, the Rc-o319 from Japan, the BtKY72 from Africa, have been reported to utilize the ACE2 from bat but not human for cell entry [
99,
103]. Zhou
et al. also reported that some novel bat SARSr-CoVs identified from China, such as RsYN04 and RmYN08, which share similar deletion with RaTG15 in the RBD, could bind to human ACE2 but with a very weak affinity [
101]. The bat SARSr-CoVs that contain two deletions in their RBDs cannot bind or utilize ACE2 from humans or other species [
19,
99,
102,
110].
In addition to SARSr-CoVs, attention should also be paid to those coronaviruses that utilize receptors other than ACE2 for cell entry. For example, diverse coronaviruses belonging to the subgenus of MERS-CoV have been identified and some of them have been demonstrated to utilize the same DPP4 receptor as MERS-CoV, suggesting the potential of cross-species infection [
116,
117].
9 Continuing evolution and emergence of SARS-CoV-2 variants
While SARS-CoV-2 has been circulating for 4 years within the human populations, it is keeping evolving and new variants of concern (VOCs) are continuing to emerge. Since first emergence in late 2021, variants of SARS-CoV-2 Omicron lineage with altering immune escape properties such as BA.5, XBB, EG.5 have continued to cause infection surge on a global scale. More recently, the variant BA.2.86 and its descendant JN.1 are rising globally and becoming new dominant SARS-CoV-2 strains [
118,
119]. The wide spreading through human populations, together with the rapid evolution driven by its relatively high rate of mutation, recombination, insertion, and deletions, has contributed to the diversification of novel SARS-CoV-2 variants [
120]. Some mutations may change the intrinsic functional properties of the virus including infectivity, virulence, and transmissibility. It may also alter virus antigenicity and thus confer the ability to evade immune response. Such evolution of antigenic novelty can be a key determinant of the reproductive success of the variants. In a word, SARS-CoV-2 is further gaining fitness as a result of a complex interplay of the evolving virological characteristics in context of changing human immunity [
121].
10 Strategies of preparedness for the next disease X caused by SARSr-CoVs
The capability of human and animal ACE2 utilization by the bat or pangolin-derived SARSr-CoVs suggests their potential risk to public health. Concerning the wide distribution and spillover risk of SARSr-CoV, it is necessary to prepare in advance for the future prevention and control against the EIDs caused by this groups of viruses. Due to the emergence of variants and the uncertainty of novel coronavirus emerging in the future, development of universal diagnostic methods, pan-vaccines, and broad-spectrum antivirals should be strengthened.
10.1 Diagnostic methods
Efficient diagnostic methods can help the early detection and surveillance of virus transmission. SARSr-CoVs are relatively conserved in some regions and their encoding proteins, such as ORF1b which encodes core complex responsible for virus genome transcription and replication, structural protein genes encoding the envelope protein, membrane protein, and nucleocapsid (N) protein. These genes are ideal targets for designing the universal primers for nucleic acid detection of viruses covering the same species, genus or family. Similar strategy can be used for the development of virus antigen and antibody detection. Previously, we have developed a pan-qPCR covering the SARSr-CoV RdRp gene and demonstrated its efficiency for the early detection of SARS-CoV-2 in COVID-19 patients at the beginning of the outbreak [
10,
122]. We also developed an ELISA targeting the N protein of a bat SARSr-CoV and found 2.7% of SARSr-CoV antibody in humans living around the bat caves where we discovered the gene pool of SARSr-CoV [
19,
23]. We used this method for screening the antibody in patients who were hospitalized at the early outbreak [
123]. These methods can be optimized and developed for the routine surveillances of virus transmission in domestic animals and humans at the early stage.
10.2 Pan-vaccine
Vaccine is the ultimate strategy to eliminate or eradicate a human or animal pathogen. The only successful eradication is smallpox vaccine after its prevalence for more than 10 thousand years. But it takes hundreds of years to prove the concept and develop the efficacy commercial vaccine. In recent 50 years, human beings experienced the EIDs at a higher frequency and yearly based. The process to develop the vaccine cannot catch up with the speed of disease emerging and transmission, particularly the newly emerged diseases or highly mutated viruses, such as SARS-related CoVs, SARS-CoV and SARS-CoV-2. With the current knowledge of viruses and techniques such as mRNA delivery, it is possible to prepare the pan-vaccine or chimeric vaccine against a group of viruses which share common epitopes inducing neutralization antibody or simulating the T cell response [
124,
125]. Although SARSr-CoV are highly diverse in the S proteins, they show similar conformational structures revealed by X-ray, cryo-EM or model simulation, suggesting that there have potential conformational epitopes which induce cross-protection antibodies [
22,
112,
113]. In addition, the SARSr-CoV spikes are highly identical in the domains responsible for membrane fusion, which may serve as ideal targets for the development of pan-vaccine [
126]. This hypothesis has been proved in recent publications aiming to find neutralization antibody therapeutics for COVID-19. Several groups have identified monoclonal antibodies which cannot only neutralize the SARS-CoV-2 variants but also bat SARSr-CoVs [
127–
131]. In addition, Tan
et al. found that the SARS survivors who have been immunized with the BNT162b2 mRNA vaccine have a high-level and broad-spectrum neutralization antibody which are capable of neutralizing not only known variants of concern of SARS-CoV-2, but also SARSr-CoV identified in bats and pangolins [
132].
10.3 Pan-drugs
RNA viruses, especially for the viruses from the same family or genus, are relatively conserved in their RNA-dependent RNA polymerase and main protease (M
pro), which can serve as the common targets for the development of antivirals. Some compounds and peptides have been designed or repurposed as promising pan-drugs for SARS-CoV-2 and other human coronaviruses. Based on their targeted proteins, these pan-drugs can be mainly categorized as viral replication inhibitors and entry inhibitors [
133,
134].
The RdRp-targeting nucleotide analogs act as replication inhibitors by introducing disruptive and lethal mutations during the replication of virus genome. Remdesivir and favipiravir, two previously developed nucleotide analogs, have been reported to have broad-spectrum antiviral effects on various emerging viruses including SARS-CoV-2, but their clinical effectiveness against COVID-19 is controversial [
133]. Another novel nucleotide analog licensed by Merck, molnupiravir, was reported to have reduced the hospitalization of COVID-19 by about 50% and it became the first oral drug approved for treating COVID-19 [
135,
136]. As molnupiravir was shown to broadly inhibit propagation of SARS-CoV-2, SARS-CoV, and MERS-CoV, and it is resistant to the proofreading exoribonuclease encoded by coronaviruses, it can be a potential candidate of pan-CoV therapeutic agent for future use [
136]. The M
pro of coronavirus plays an important role in the proteolytic processing of polyproteins responsible for viral replication and transcription. Its substrate binding pocket is highly conserved among various coronaviruses, making it plausible to develop wide-spectrum antivirals against coronaviruses with this target. It was announced by Pfizer that its novel protease inhibitor, paxlovid (PF-07321332 and ritonavir), showed reduced risk of hospitalization or death by 89% in patients with COVID-19 [
137]. The clinical trials also demonstrated that China’s domestically developed oral M
pro inhibitor, simnotrelvir can effectively shorten the time to the resolution of symptoms among mild-to-moderate adult patients with COVID-19 [
138].
Prior to RNA replication, receptor binding and membrane fusion are the initial step for coronavirus infection. As a conserved 6-helix bundle (6-HB) structure composed of heptad repeat 1 and 2 (HR1 and HR2) trimers is critical for membrane fusion, the infection can be inhibited by peptides that interact with HR1 and block the 6-HB formation [
139]. A cholesterol-conjugated lipopeptide named EK1C4, which was derived from the HR2 domain of human coronavirus OC43, exhibited high inhibitory activity against SARS-CoV-2 and other human coronaviruses as well as different bat SARSr-CoVs. Preclinical data suggest that it has the potential to be developed as a pan-CoV fusion inhibitor for prevention and therapeutics of SARS-CoV-2 and future emerging coronaviruses [
139,
140].
Corresponding to different antiviral mechanisms, we will have multiple directions for preparation of pan-CoV or pan-virus drugs for future pandemic. Since each single drug has its own shortcomings, a reliable strategy would be developing cocktail therapy that combine two or more wide-spectrum antivirals inhibiting various steps of the viral life cycle. The stock of these pan-drugs would be a powerful defense in the combats against any newly emerging outbreaks in the future.
10.4 Long-term surveillances
While it takes time to develop the vaccine and antiviral drugs which are highly costly, the most efficient and cost-effective strategy is the long-term surveillance if we know the pathway and key factors involved in the EID emergence. Previous studies show that
Rhinolophus species carry highly diverse SARSr-CoV and some of them have the capability of interspecies transmission in term of human ACE2 utilization. These viruses have evolved within their natural reservoirs over evolutionary time and spillover to other animals and humans when anthropogenic environmental changes alter population structure of their reservoir hosts, and bring wildlife, livestock, and people into contact. While many interspecies infections of viruses are abortive, in small or regional scale, some EIDs become pandemics when zoonotic pathogens transmit easily among people, and spread in rapidly urbanizing landscapes, megacities, and travel and trade networks [
1,
2]. Thus the early detection of viruses in the transmission chains such as wildlife, domestic animals, and high risk populations will help to identify their interspecies transmission at the early stage and call for rapid response to prevent the further transmission. This prophylactic strategy applies for all wildlife-borne diseases and should be implemented by the One Health approach, which aims to attain the harmony in human, animal, and environment health via interdisciplinary collaborations including the microbiology, veterinary, epidemiology, ecology, and sociology, as well as cross-sectoral and cross-regional coordination from different government departments and different countries at global level [
141].