Coronaviruses possess the largest genomes of all RNA viruses at 26 – 32 kilobases (kb) in length (45). These are enveloped, single-stranded, positive-sense RNA viruses whose genomes contain at least 6 open reading frames (ORFs) that encode 16 non-structural proteins (NSP), 4 structural proteins, a 5’ cap structure and a 3’ poly (A) tail (46, 47). The family Coronaviridae belongs to the order Nidovirales (48). Within the family Coronaviridae is the subfamily Orthocoronavirinae which is subdivided into four genera, each containing a type species: Alphacoronavirus (alphacoronavirus I), Betacoronavirus (murine coronavirus), Gammacoronavirus (avian coronavirus), and Deltacoronavirus (bulbul coronavirus HKU11) ( 46 ). This review will be focused on the Betacoronavirus genera which is divided into five subgenera: Embecovirus, Sarbecovirus, Merbecovirus, Nobecovirus and Hibecovirus with Embecovirus, Sarbecovirus and Merbecovirus relevant to humans, as seen in Figure 1 (10, 26, 49–51).

Betacoronaviruses have a helical nucleocapsid contained within spherical envelopes which are coated with characteristic S glycoproteins protruding from the surface of the envelope (47). When observed via cryo-transmission electron microscopy, the S proteins of the virus have the appearance of a crown, which is where the name coronavirus originates (48). Each spike on the envelope of betacoronaviruses is a homotrimer of the glycosylated S protein. Each monomer S protein contains an S1 binding domain and an S2 fusion domain catalyzing anchoring to the membrane (52). The S protein, a class-1 fusion protein, is processed through the Golgi apparatus where it is heavily glycosylated via an encoded N-terminal signal tag. The M protein, the most abundant protein in the envelope, consists of three transmembrane domains, all of which give the virion its spherical form as well as binding to the nucleocapsid, serving as the scaffold for the virion (53). The E protein consists of 2 domains functioning to assist in the assembly and release of the viral particle from the host cell. Lastly, the N protein consists of two domains allowing binding to the genomic RNA and formation of the nucleocapsid (54). Both the N-terminal and C-terminal domains have RNA binding capacity, however the mechanism by which binding occurs varies. Due to large amounts of observed phosphorylation of the C-terminal domain of the N protein, it is hypothesized that phosphorylation catalyzes a conformational change, increasing the binding affinity between the N protein and viral RNA (55). Phosphorylation of the N protein also acts to tether the replicase-transcriptase complex (RTC) to the viral genome during virion assembly (48). The structural proteins play critical roles in virulence and establishment of infections within reservoirs and other hosts by facilitating viral attachment and entry into host cells.

Betacoronavirus genomes are positive sense and therefore in the same orientation as host mRNA. This allows for direct translation by the host to produce the two viral polyproteins (48) (pp1a and pp1ab) which are then cleaved into the 16 NSPs ( Figure 2 ). Some non-structural proteins exhibit host suppression, for example nsp3 and nsp16 seem to block the interferon-mediated immune response through inhibition of interferon (IFN) pathways (56). Through degradation of host mRNA and suppressing translation of host proteins, nsp1 contributes to the regulation of host cells and the immune response, while also promoting viral production (57). The use of 5’ caps on the viral genome is involved in immune evasion, by disguising as host mRNA which does not activate pattern recognition receptor pathways that would normally lead to the destruction of viral RNA (58, 59). The S protein is critical in viral infection as this is the protein responsible for binding and gaining entry into host cells (60).

The host specificity of betacoronaviruses is determined by their S protein, which binds host receptors that include aminopeptidase N, angiotensin converting enzyme 2 (ACE2) and dipeptidyl peptidase 4 (DPP4) in order to gain entry to host cells (61). Receptor-recognition of the S protein is facilitated by the S1 domain, which is composed of the N-terminal domain (NTD) and C-terminal domain (CTD) (62). The NTD is involved in receptor recognition whereas the CTD contains the receptor-binding domain (RBD) that binds the host cell receptor and determines specificity (63). In the case of SL-CoVs, there is a strong binding affinity between the viral S protein and the host cell receptor ACE2 (52), whereas MERS-like coronaviruses have a strong binding affinity to the DPP4 receptor in the host (64). Reservoir species, such as bats, have a protein homologous to the human ACE2 receptor which may enable transmission of SL-CoVs from bats to humans (65). As different SL-CoVs have varying binding affinity to ACE2 receptors, intermediate host infections can facilitate viral mutation of S protein allowing recognition and binding of ACE2 receptor in humans (66). Following binding of the S protein to its host cell receptor and endosomal uptake of the virus into the cell, the viral genome is subsequently released into the cytoplasm for transcription and replication (67).

ACE2 is the receptor used for cell entry by many Sarbecoviruses, including SARS-CoV and SARS-CoV-2 (65). ACE2 is a type-I transmembrane protein found in epithelial cells of the lung, vascular endothelial cells, and renal tubular epithelium (68, 69). This receptor plays an important role in cardiac pathophysiology acting as a negative regulator of the Renin-angiotensin pathway in the lung, which regulates blood pressure and electrolyte levels (70). Numerous studies have been done to investigate the role of ACE2 in SARS-CoV infection. For example, viral loads in ACE2 knockout mice were far lower than those in the wild-type control mice, suggesting the virus was unable to enter host cells without ACE2 (71). The ability of the S protein to bind ACE2 is largely dependent on the affinity of the viral RBD - which may vary due to mutations in this area of the genome (39, 66, 72). Researchers were able to show that minor variations within the S protein in the RBD of Sarbecoviruses can lead to binding of ACE2 receptors in other animals (intermediate hosts) and humans (73). Investigating the S protein variability within different Sarbecoviruses and SL-CoVs has allowed for greater understanding of how cross-species transmission occurs.

The Sarbecoviruses consist of SARS-CoV and SARS-CoV-2, two of the most pathogenic coronaviruses identified to date (103). SARS-CoV was originally believed to have emerged from Paguma larvata (masked palm civets) after a case of SARS transmission from masked palm civets to humans (83, 104). However, this changed in 2005 when SL-CoVs were discovered in Chiroptera spp. bats, suggesting that bats may be the true reservoir for SARS-CoV (105) and that civets may instead be an intermediary host. This is not entirely a surprise as there are many coronaviruses which have been identified in various bat species, and SL-CoVs have been found in bat species such as Rhinolophus spp., Hipposideros spp. and Chaerophon spp. (5, 89, 104, 106). More specifically, SL-CoVs that bind ACE2 to mediate cell entry have been found in Rhinolophus sinicus (Chinese horseshoe bats) (65, 107). Further, in 2017 there were 11 new SL-CoVs identified in Rhinolophus sinicus bats from a cave in Yunnan province, China that shared 92-99% sequence homology to SARS-CoV (91). Today, Rhinolophus sinicus is considered the main reservoir of SL-CoVs and should likely be considered the origin point of SARS-CoV as well (108).

While SARS-CoV-2 has not been identified in bats to date bats are known to be an original source of alpha- and betacoronaviruses, with other Sarbecoviruses similar to SARS-CoV-2 known to be harboured in Rhinolophus spp. bats (109, 110). However there remain only speculations about the origins of SARS-CoV-2 with no direct evidence for the original source, leaving room unfortunately for wild hypotheses around the origins of this virus. The existence of a major virology laboratory (Wuhan Institute of Virology) with a program studying coronaviruses in the city where SARS-CoV-2 was first identified proved to be too large a coincidence for some, subsequently birthing several lab origin hypotheses (14). The first claim is of the virus being of manmade origin involves the observation of human immunodeficiency virus (HIV) sequences in the SARS-CoV-2 genome in a now retracted article by Pradhan et al. and again in another article (111). These findings were quickly refuted through bioinformatic analyses demonstrating the four short sequences occurred at different times and independent of each other and are insufficient evidence for a common ancestor (112, 113). Similarly there have been hypotheses around creation of a virus through gain of function experiments with both recombination and engineered mutations through serial passaging in animal models to obtain these changes to a SL-CoV suggested as possible routes of engineering this recombinant virus (114, 115). However, many of these proposed mutations may be present in other coronaviruses such as the furin cleavage site observed in the RBD of SARS-CoV-2 or a N501Y mutation that would have occurred for efficient replication in animal models (which was not observed in the early stages of the pandemic) and there remains no evidence of engineering within the RBD with the only explanation for their presence being viral evolution (14, 34, 116–121). There has been concern that the emergence of the virus into the human population may have resulted from a laboratory release. The two main hypotheses are, SARS-CoV-2 was created through gain of function experiments on related viruses via serial passaging though this has been refuted (34, 122, 123). The second hypothesis has posited that an accidental laboratory release of SARS-CoV-2 precipitated movement of the virus from the laboratory to the community though this has also been refuted (29, 122). Recent investigation of SARS-CoV-2 genomic diversity by Pekar and colleagues has provided evidence to support the emergence of SARS-CoV-2 through multiple zoonotic events (124). The analysis supports that SARS-CoV-2 lineages A and B resulted from at least two separate spillover events into humans in late 2019.

However, despite these hypotheses there are historical patterns of zoonotic emergence and circulation for coronaviruses as well as the increasing identification of SARS-CoV-2 in numerous nonhuman animal species (14, 125). This is supported by the similarity in the route to human exposure through animal markets as this follows previous coronavirus outbreaks (39). There remains much to be understood about SARS-CoV-2 spillover into humans, and much of our current understanding has relied on epidemiological modeling. Molecular clock modeling of the genome for SARS-CoV-2 and the subsequent mutations suggest a recent emergence, some suggest however, this may not be true due to the highly mutated state of the genome and the effect this may have had on the linearity of the analysis (126–128). As a recent phylogenetic analysis suggests that the lineage of SL-CoVs that SARS-CoV-2 originates from diverged from ancestral bat CoVs sometime between 1948-1982 leaving the door open to the possibility of circulation under the radar (103). It is also unknown if a bat was the only animal involved in the evolution and emergence of SARS-CoV-2. Ongoing and future studies will continue to provide context and nuance for these questions.

In 2020, a group of researchers collected and sequenced samples from Rhinolophus spp. bats in Yunnan province, China, to better understand the zoonotic origins of SARS-CoV-2. Of interest, two bat CoVs had high nucleotide sequence homology to the full-length genome of SARS-CoV-2: RaTG13 (96.1%) and RmYN02 (93.3%) (129, 130). These viruses were collected from Rhinolophus affinis and Rhinolophus malayanus bats, respectively. While RaTG13 maintained high nucleotide sequence homology to the S gene (92.9%) and the RBD (85.3%), RmYN02 showed far lower sequence homology for the S gene (71.9%) and RBD (61.3%). In contrast, they also compared Pangolin viruses GD/2019 and GX/P5L/2017 to SARS-CoV-2 and found lower levels of nucleotide homology for the whole genome (GX/P5L/2017 = 85.2%) but high levels of amino acid sequence homology in the S gene (GX/P5L/2017 = 92.4% and GD/2019 = 90.7%) and RBD (GD/2019 = 97.4% and GX/P5L/2017 = 86.8%) (130, 131). It seems that while pangolins may have been involved in a recombination event affecting the RBD, the high level of homology between the genome sequences of SARS-CoV-2 and bat CoVs suggest that the virus originated from bats rather than pangolins (132, 133).

To further investigate the bat origin hypothesis, investigations have assessed the selective pressures driving viral adaptation and evolution. MacLean and colleagues identified weak purifying selection among SARS-CoV-2 strains from the first 11 months of the pandemic (36). For the spike protein, diversifying selection occurred deeper in the phylogenetic branches of the Sarbecovirus clade, leading to a very generalist SARS-CoV-2 virus and is supported by the wide host range (36, 134). Others have suggested that the closest ancestral divergence of this virus is likely to be approximately four or five decades ago, based on similarities to bat CoVs RmYN02 and RaTG13, respectively (135, 136). Additionally, investigations have looked at the CG suppression within the viral genome (cytosine followed by a guanine in the 5’ to 3’ direction) due to their link to antiviral mechanisms in the host (137). Many vertebrate RNA viruses demonstrate similar patterns of 5’-CG-3’ dinucleotide suppression, where there is a lower number of CG dinucleotides than expected, as found within vertebrate genomes (137). Further, Takata and colleagues suggested that this suppression may highlight an adaptation with RNA viruses to evade host immunity through reduced discrimination of self- and non-self RNA (137). Analysis of the Sarbecovirus clades identified a phylogenetic shift towards CG suppression followed by an elevated substitution rate (36). This suggests an increase in selective pressure in the surrounding environment at the time (138, 139). These evolutionary factors taken together suggest that the virus evolved prior to the spillover event into humans, rather than through human-to-human infection during the pandemic (103). Thus, it is likely that SARS-CoV-2 was highly capable of infecting humans prior to the spillover event which led to the first COVID-19 case (140). Adding to this, the fact that SARS-CoV-2 can transmit readily to other animals (mink, cats, dogs, etc) - and in some cases transmit back to humans (mink) - supports the possibility of a generalist virus, where the virus already contained generalist ACE2 binding properties that could aid in efficient host switch across multiple intermediate animal species (139, 141).

We summarize the growing evidence of SARS-CoV-2 circulation and transmission patterns across various animal species in Figure 3 . Ferrets (Mustela putorius furo) are instrumental models for studying respiratory pathogenicity and transmission of viruses (181). Experimental studies show that ferrets are susceptible to SARS-CoV-2 infection of the upper respiratory tract early in the disease course (166). However, clinical signs of illness appear to be uncommon with transient fever, and mild respiratory symptoms reported (166, 182–185). There is still substantial viral shedding observed in the respiratory tract during infection making ferrets a useful model for studying transmission. During infection ferrets have been shown to infect healthy ferrets in close contact through the high degree of viral shedding in their feces, nasal secretions, urine and saliva (12). Direct and indirect transmission of SARS-CoV-2 has been demonstrated in ferrets and healthy ferrets become symptomatic following direct contact with infected ferrets, though separation of the animals with maintenance of shared airspace did result in some viral positivity in the absence of symptoms (139). The lack of clinical signs of illness in conjunction with the high amount of viral shedding suggest mostly asymptomatic infection and evidence that mustelids such as ferrets may have played a role as an intermediate host in SARS-CoV-2 emergence. While infection of ferret badgers has not been documented, they have a high degree of similarity to other animals which are permissive to SARS-CoV-2 infection and therefore are still an animal of interest.

Mink (Neovison vison) are also a potential SARS-CoV-2 intermediate host given the SARS-CoV-2 detection and onward transmission in mink from two farms in the Netherlands (186). The animals showed respiratory and gastrointestinal symptoms (187), and approximately 1.2-2.4% of animals succumbed to infection, the majority of which were pregnant females (188). Necropsies found signs of interstitial pneumonia and lung lesions (186). There is supportive evidence that SARS-CoV-2 was introduced to mink by farm workers with subsequent transmission between the animals (186). This was supported by sequencing of viral samples from both mink and humans which revealed significant homology between the viruses present in each sample (187). This became a cause for concern as it was also observed that variants of the virus had been transmitted from mink to humans (189), suggesting an intermediate host which could support viral recombination and rapid transmission (190). Transmission between humans and mink was observed in ten countries: Canada, Denmark, France, Greece, Italy, Lithuania, the Netherlands, Spain, Sweden and the USA (190, 191). It was determined early on that the mode of transmission was not direct as mink are housed separately. Viral RNA was also detected in early collection of inhalable dust samples indicating a potential route of exposure (186).

Raccoon dogs (Nyctereutes procyonoides) initially gained attention as possible intermediate hosts of SARS-CoV (21, 80, 151) for two main reasons: i) susceptibility to SARS-CoV infection (192); and ii) high ACE2 sequence similarity between raccoon dogs and humans (193, 194). Raccoon dogs can be productively infected with SARS-CoV-2 through experimental inoculation and transmit the virus to other healthy animals (162). However, clinical signs of illness such as increased body temperature or weight loss were not observed and virus isolated from infected animals had 100% sequence homology to the viral inoculum.

The susceptibility of white tailed deer (Odocoileus virginianus) to experimental infection was assessed in early 2021 (154). Intranasal inoculation of deer resulted in virus shedding through nasal secretions and transmission to naïve deer who later seroconverted. Indirect transmission was also observed as virus was identified in nasal swabs and transiently in fecal samples from naïve fawns housed in separate pens from infected animals, and which included plexiglass barriers (154). Additionally, RNA was detectable in infected animal tissues for up to 21 days post-infection. Other woodland animals such as bushy-tailed wood rats and skunks are capable of shedding virus in respiratory secretions (160). As these animals are shedding the virus and, in some cases, appear capable of transmitting to their surrounding environments, their role as intermediate hosts and potential future sources of spillback of novel SARS-CoV-2 variants to humans needs to be investigated.

Rodents are believed to have played a significant role in the emergence of human Embecoviruses (HCoVs OC43 and HKU1) and have therefore been hypothesized as potential intermediate hosts of SARS-CoV-2. Early studies suggested that mice were unlikely to be an intermediate host candidate for SARS-CoV-2 (130, 195, 196). However, subsequent work by Griffin and colleagues has demonstrated that deer mice are susceptible to infection resulting in asymptomatic or mild disease (197). Infected deer mice could also transmit virus to co-housed naïve mice. More recently, Stone and colleagues demonstrated that the SARS-CoV-2 Alpha and Beta variants could result in productive infection of wild-type C57BL/6 mice via intranasal inoculation (198). Alternatively, Syrian hamsters are susceptible to infection with SARS-CoV-2 presenting with clinical disease that resembles respiratory infection in humans as well as weight loss (159, 199–201). One study demonstrated that Syrian hamsters that had previous infection to SARS-CoV-2 had protection from re-infection with reduced replication in the upper respiratory tract and no observed transmission to naïve contact animals (202). Thus, hamsters have become widely used for investigations of SARS-CoV-2 infection (7, 161, 200). A recent preprint provides evidence for transmission of SARS-CoV-2 to humans from naturally infected hamsters, though the implications of this on the initiation of new human-to-human transmission chains remains to be determined (203).

ACE2’s ubiquitous presence within the animal kingdom and its high degree of similarity amongst mammalian species is a major contributor to the spread of betacoronaviruses around the world, most recently with SARS-CoV-2 (204). In silico modelling predicted the following species may exhibit binding affinity for the SARS-CoV-2 S protein: cats, cattle, monkeys, dogs, pigs, horses, sheep, and rabbits (in decreasing order) (205). Many other animals have been infected with SARS-CoV-2 experimentally or in nature and are able to host a productive infection. These cases will be discussed in the following paragraphs.

Initial attention was given to domestic animals such as dogs and cats as they would be high risk to transmit to humans if they could host a productive infection. Reports showed domestic cats had tested positive for SARS-CoV-2 in Europe, Asia, and North America (163, 205). Cats have also been shown to spread SARS-CoV-2 via respiratory droplets (166) though viral RNA has also been detected in nasal, oropharyngeal and rectal swabs (155, 206). This was concerning due to the proximity of cats to humans however, one study showed that cats exposed to the virus did not exhibit any symptoms (88, 207). There is conflicting data as to whether or not cats are asymptomatic throughout the infection (208), but if cats are in fact asymptomatic they may not be effective intermediate hosts. As for dogs, studies show that SARS-CoV-2 replicates poorly in these animals and that healthy dogs who come into contact with SARS-CoV-2-positive dogs remain seronegative (166). Overall, infected dogs showed no obvious signs of infection and most often did not have detectable levels of RNA present in biological samples (88, 166, 206). In an attempt to determine why this occurs, scientists analyzed the ACE2 receptor in dogs and found 5 amino acid substitutions but none within the RBD; therefore, this is believed to have minimal impact on binding of the S protein (196, 209). Based on the studies conducted so far it is unlikely cats or dogs played a role in the emergence of SARS-CoV-2.

Investigations into agriculturally significant animals such as cattle were an important consideration as there are previous bovine links to coronaviruses in addition to an identified bovine coronavirus (210). For the most part, cattle did not seem to be able to host a productive SARS-CoV-2 infection with little to no viral replication or immune response detected in these animals. They also did not exhibit clinical signs of infection or transmit to nearby animals (164). Cattle and other farm animals are important to consider as they are commonly in contact with humans. One study in particular looked at susceptibility to SARS-CoV-2 in poultry and determined that chickens along with turkeys, geese, quail and ducks are not susceptible to infection (166, 211). Similarly, many studies have investigated SARS-CoV-2 infection in pigs, and it seems that they are resistant to infection as there are no signs of infection, no pathology and no viral RNA detected (212–214).

SARS-CoV-2 infections in a variety of zoo animals have been widely reported, including tigers, snow leopards, lions, gorillas and pumas. Animal testing followed signs of respiratory symptoms of disease and while transmission between animals in the same enclosures was reported, widespread transmission was observed in any of these cases (215). It is thought that the animals contracted the virus from asymptomatic workers.

Aside from these examples of natural infection, much of what we know of SARS-CoV-2 infection in animals has come from experimental infections. This has led to the discovery that rhesus monkeys can be experimentally infected (163). Two Rhesus macaque species (M. fascularis and M. mallata) and one common marmoset species (C. jacchus) were able to be infected with SARS-CoV-2 via the intratracheal and intranasal routes and demonstrated clinical signs of infection (216, 217). It has been reported old world monkeys are susceptible to infection and new world monkeys have much lower susceptibility when compared to human ACE2 there are 4 amino acid differences (218). Nonhuman primates have similar clinical presentation of COVID-19 to humans including viral replication in the upper and lower respiratory tracts, inflammation and focal edema among other less frequent signs and symptoms (13). Due to these clinical presentations, the phylogenetic relatedness of nonhuman primates to humans, and previous reports of HCoV-OC43 infections in nonhuman primates, concerns have been raised regarding the potential impacts of SARS-CoV-2 on endangered species, including great apes (88, 219).

Bank voles have been experimentally infected and had viral RNA present in nasal tissue for up to 21 days post infection, though no transmission to animals in direct contact was observed (220). Rabbits have been reported to be susceptible to infection with SARS-CoV-2 (164) however, there is conflicting data stating that in fact white cotton tail rabbits are resistant to infection (160). Additional studies have determined that squirrels (211), raccoons and black-tailed prairie dogs (160) are resistant to SARS-CoV-2 infection [Recommended further readings on the topic of experimental and natural infections of animals. (221–223)].

Malayan pangolins (Manis Javanica) and Chinese pangolins (Manis pentadactyla) are considered vulnerable or critically endangered as they are the most trafficked mammals in the world (224) due to the use of their meat and scales in traditional African and Chinese medicine (225). Pangolins are solitary, nocturnal mammals that dwell in remote sandy forests (226) away from humans (227), and thus, poaching provides the only real opportunity for human contact with wild populations (37). Pangolins were proposed as an intermediate host in the emergence of SARS-CoV-2 in the early days of the pandemic due to a high degree of sequence identity to pangolin-CoVs, their interaction with bat populations and their presence at the Hunan wet market during the time when the first documented cases of SARS-CoV-2 were identified (153).

Sequence similarity was identified by two independent studies that identified SL-CoV in Malayan pangolins confiscated from illegal wildlife traders (77, 228). The identified SL-CoVs had 85.5-92.4% sequence identity to SARS-CoV-2; however, these pangolin-CoVs had 97.4% sequence identity to the RBD of the SARS-CoV-2 S protein (specifically pangolin CoVs GD/P1L and GD/P2S) (77, 131, 132, 228). Specifically, the six amino acid residues in the RBD of the S1 protein identified as critical for binding the host ACE2 receptor are conserved between pangolin-CoVs and SARS-CoV-2 (34, 131, 229). This high degree of sequence identity in the RBDs of SARS-CoV-2 and pangolin-CoVs suggests either a recombination event occurred or these viruses have highly similar RBDs due to convergent evolution (77, 228). For the recombination hypothesis it was posited to have occurred between pangolin-CoVs from Malayan pangolins and RaTG13 from R. affinis bats due to the high degree of similarity in the RBD of pangolin-CoVs to SARS-CoV-2, while SL-CoV RaTG13 is the closest relative to SARS-CoV-2 albeit with a distinct RBD site of the S protein (77, 131, 132, 228). However, when analyzed through alignment there is low nucleotide similarity compared to the high amino acid similarity and many misalignments within the sequence, suggesting recombination is unlikely and merely convergently evolved features of these distinct SARS-CoV-2 and pangolin CoV viruses (77, 103, 172, 229, 230). Although recombination has been observed in CoVs there is no evidence in the S protein of the SARS-CoV-2 lineage and instead appears to be an artifact of the metagenomic analysis that detected recombination initially (103, 123). Supporting the idea of convergent evolution, a recent analysis of SL-CoV RaTG13, pangolin-CoVs and SARS-CoV-2 finding that pangolin-CoVs diverged from the SL-CoVs approximately 150-180 years ago (103). If recombination occurred it would be expected for these viruses to have a higher degree of similarity in specific regions, which has not been observed (29, 132, 174). For recombination to occur it also implies the viruses, in this case SL-CoV RaTG13 and pangolin-CoV GD/P1L or GD/P2S, would co-infect the same cell; however, R. affinis bats (Species SL-CoV RaTG13 was found in) do occupy the same natural range as M. javanica pangolins (122, 231, 232).

To date, pangolin-CoVs have not been found in Chinese pangolins, which share the same habitat range as R. affinis bats (228). There is limited evidence for cohabitation of bats with pangolins of any species and does not appear very common, with one study finding bats and pangolins in Gabon within the same burrows (233). Follow up studies to detect pangolin-CoVs in Malayan pangolins have been unsuccessful with a study of 334 confiscated pangolins finding no sarbecoviruses raising doubts around the role of pangolins in the emergence of SARS-CoV-2 (123, 228, 234). Further to this point, a study of wild Malayan pangolins in Malaysia determined that there is no observed circulation for betacoronaviruses, filoviruses and flaviviruses, suggesting that any detected sarbecoviruses are most likely acquired through contact within the smuggling trade of these animals (234). The animals that pangolin-CoVs have been isolated were found to be either severely ill or already dead (82, 228). Animals becoming severely ill is not what is expected for intermediate hosts as this would greatly limit viral amplification through host immune response and ultimately death, as well as interactions with other species limiting ability to spread the virus to other species (110). Some have presented the possibility for contributing factors to the severe illness and death of the tested animals, including due to the stressful environment they find themselves in in close contact with other animals and humans as well as other viruses that are commonly found in pangolins including Sendai virus (235, 236). While some studies have shown the presence of coronaviruses in captivity (82), there remains no evidence of SARS-CoV-2 or SL-CoVs in wild pangolins. More to that point, Yuan et al. claim that pangolins were not present at the Huanan Wildlife Market during the initial identification of SARS-CoV-2 in 2019 (237). In addition to the lack of circulating coronaviruses, the solitary lifestyle of pangolins makes it difficult for pangolin populations to amplify pangolin-CoVs, which suggests they would be a poor intermediate or reservoir host candidate from an epidemiological standpoint (103). Considering the studies conducted to date on the role of Malayan pangolins in the emergence of SARS-CoV-2 as an intermediate host does not line up with the hypothesis and the initial phylogenetic reports have been refuted.

There is continuing evaluation of the susceptibility and onward transmission potential of various animal species to SARS-CoV-2. However, it has also been hypothesized that SARS-CoV-2 emergence may have occurred in the absence of an intermediate host (36). Here, the authors suggest that diversifying positive selection was limited to the early phase of the pandemic and that SARS-CoV-2 has much weaker purifying selection as compared to related sarbecoviruses. The authors further suggest that the transmission of SARS-CoV-2 to additional nonhuman animal species supports the generation of a generalist virus in a bat reservoir.

Ongoing research continues to actively investigate the hypotheses for the emergence and transmission of SARS-CoV-2 demonstrating the collective effort of the scientific community to determine the origins of this pandemic, there is still much to be known and continued collaboration will be essential.