Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which triggered the Coronavirus Disease 2019 (COVID-19) outbreak, has caused enormous socioeconomic losses (Zhou et al., 2020). This virus is a member of the Nidovirales order, the Coronaviridae family and the Coronavirinae subfamily in the Betacoronavirus genus. SARS-CoV-2 is a spherical pleomorphic virus (50–140 nm in size). It is an enveloped virus with large, 20 nm-long, club-shaped peplomers that form a crown appearance. It has a helical nucleocapsid with the largest positive sense RNA (29,903 bases) (Lu et al., 2020). There are two protein groups in SARS-CoV-2 (structural and non-structural). The structural proteins include the spike glycoprotein (S), the nucleocapsid (N), the membrane protein (M) and the envelope protein (E) (Naqvi et al., 2020); whereas the non-structural proteins include the protease (NSP3 and NSP5), the RNA-dependent RNA polymerase protein (NSP12) (Begum et al., 2020), helicase (NSP13), endoribonuclease (NSP15) and other NSPs (NSP1, NSP2, NSP4, NSP6, NSP7, NSP8, NSP9, NSP10, NSP11, NSP14 and NSP16) that associated with immune suppression, viral replication and transcription (Chan et al., 2020).

The SARS CoV-2 spike glycoprotein initiates a viral infection through its attachment to angiotensin-converting enzyme 2 (ACE2) cell receptors, which are distributed widely in different organs, including oro/nasopharyngeal mucosae, the lungs, gastrointestinal tract (GIT) mucosae, the brain, kidneys and the liver (Hamming et al., 2004). The spike glycoprotein is a trimeric fusion molecule (as illustrated in Figure 1 ), and its corresponding gene consists of 3,822 nucleotides (21,563–25,384) that encode 1,273 amino acids. The spike glycoprotein is composed of two main subunits, an amino (N)-terminal S1 subunit and a carboxyl (C)-terminal S2 subunit, that are cleaved at the furin cleavage site (S1/S2 cleavage region) (Chan et al., 2020). The detailed molecular structure of the spike gene with detailed regions and locations is illustrated in Figure 1 and tabulated in Supplementary Table S1 , according to the Wuhan strain (Reference Sequence NC_045512.2, GeneID: 43740568) (Guruprasad, 2021; NCBI, 2021). The S1 subunit interacts with the ACE2 receptor subsequently promoting the viral infection (Hoffmann et al., 2020), and this interaction has an impact on both cross-species and human-to-human transmissions (Lu et al., 2015; Wan et al., 2020). The S1 subunit switches between a standing up receptor-binding domain (RBD) (e.g., one or two RBD-up conformers), amenable to ACE2 binding, and closed-down positions, undergoing spontaneous conformational transitions between ensembles of closed and open receptor accessible forms (Wan et al., 2020; Wang et al., 2020). In contrast with the buried S2 subunit, the S1 subunit domains are located on the surface of the spike glycoprotein, effectively protecting the fusion apparatus (Shang et al., 2020). Upon proteolytic activation at the S1/S2 region, the S1 subunit is dissociated from the S2 (Shang et al., 2020). Following the latter dissociation event, the S2 subunit undergoes a series of enormous structural rearrangements in order to orchestrate the fusion of cellular and viral membranes (Duan et al., 2020; Shang et al., 2020; Wan et al., 2020; Wang et al., 2020).

SARS-CoV-2 has evolved at a rate of approximately 1.5–3.3 × 10⁻³/per site/year (Chaw et al., 2020). Consequently, residue substitutions may affect viral infectivity, pathogenesis, transmission and viral virulence (Shafique et al., 2020). The high mutation rate and recombination frequency is a significant dilemma that may have a role in cross-species transmission (Elaswad et al., 2020). SARS-CoV-2, which has a bat precursor, evokes many questions regarding its natural origin, its transmissibility to different animal hosts and the possible incidence of animal viral reservoirs (Elaswad et al., 2020). Many experiments have investigated the susceptibility of different species to SARS-CoV-2 infection, including non-human primates, ferrets, cats, dogs, minks, tigers, mice and pigs (Hobbs and Reid, 2021). Furthermore, a case of zoonotic transmission from minks to farm workmen in Denmark was investigated (Prince et al., 2021).

The cross-species transmission of coronaviruses and their ability to jump species barriers have been reported. Moreover, many research papers have investigated the factors that may strengthen the capability of coronaviruses to overcome species barriers and increase their virulence (Elaswad et al., 2020). These features are illustrated not only by the high mutation rates of coronaviruses (Su et al., 2016), but also by the large RNA genome of these viruses that allows the occurrence of recombination and mutation, resulting in the emergence of new coronaviruses (Ye et al., 2020). Despite the importance of these features, the interaction ability of coronaviruses with different ACE2 receptors has become the dominant and most important virulence factor of these viruses (Bolles et al., 2011). According to the fifth report of the Office International des Épizooties (OIE) regarding the SARS-CoV-2 situation up until the end of September 2021, 584 outbreaks in animals (mostly in minks, cats and dogs) were reported globally, affecting 12 species in 30 countries (as demonstrated in Figure 2 ) (OIE, 2021). Therefore, the OIE strongly encourages the reporting of SARS-CoV-2 occurrence in animals (OIE, 2021).

Furthermore, in our recent study, we reported SARS-CoV-2 spillover between humans and pet animals in Egypt in 30.3% of cats (10/33) and 24.2% of dogs (8/33), as evidenced by real-time reverse transcriptase polymerase chain reaction (rRT-PCR) (Mervat et al., 2021). Moreover, thrombocytopenia, lymphocytopenia and high levels of ferritin, C-reactive protein and D-dimers were reported in hematology and clinical studies (Mervat et al., 2021).

In order to further investigate such interspecies transmission, in this study, we conducted a full spike gene sequencing of SARS-CoV-2 detected in four cats, and their treating veterinarian. Samples selection was based on the severity of clinical signs. Sequencing was combined with computational analysis, including structural visualization and homology modeling of the glycoprotein to investigate the mutations and their effect on its structure and function. Structural modelling of the ACE2 receptor in different species was conducted to study the host susceptibility to SARS-CoV-2 infection.

Coronaviruses infect a wide variety of hosts (Shafique et al., 2020). In the present study, we aimed to investigate the mutations of the SARS-CoV-2 spike glycoprotein detected in cats and their effect on its structure and function by sequencing the full spike gene, followed by bioinformatics analysis. Furthermore, we highlighted the effect of residues substitution on viral infectivity, evolution and evasion from immune response. We found that cats with a history of direct contact with COVID-19-positive owners were also infected with COVID-19, revealing the possibility of SARS-CoV-2 transmission between humans and cats.

Generally, the high substitutions and mutations in coronaviruses revealed that the virus is adapting to new hosts (Duffy, 2018). The significance of amino acid substitution depends on the switching of properties and classes; for instance, a spike glycoprotein switching from leucine to methionine in L452M may have no significance as both are hydrophobic and neutral. Similarly, switching from tyrosine to phenylalanine in the Y453F mutation may have no significance as both are aromatic and hydrophobic. In contrast, substitution from glutamine to histidine in the Q498H mutation could significantly change the protein and side chain structure because they have different features, as histidine has a positive charge while glutamine has a neutral charge (Elaswad et al., 2020). Therefore, significant changes in a spike glycoprotein could change its viral infectivity, pathogenicity and virulence, and could increase cross-species transmission (Wan et al., 2020).

In the present study, SARS-CoV-2 incidence in cats was detected from four clinical samples, and was consistent with molecular results reported in previous studies (Newman et al., 2020). Furthermore, previous experimental studies have shown the clinical susceptibility of cats to SARS-CoV-2 (Hamer et al., 2021), which may be due to the similarity in ACE2 receptors between feline species and humans (Sailleau et al., 2020; Patterson et al., 2020; Hamer et al., 2021; Ruiz-Arrondo et al., 2021). Moreover, cats shed SARS-CoV-2 horizontally after infection for up to 6 days (Hamer et al., 2021), and the virus can persist in cats post-infection for up to 32 days (Bosco-Lauth et al., 2020; Shi et al., 2020). The high positive incidence in our study can be attributable to the fact that all tested cat samples came from homes with COVID-19-positive human cases.

SARS-CoV-2 transmission between humans and cats was reported in this study as the results of the phylogenetic analysis revealed a similarity between SARS-CoV-2 detected in cats (OK144251, OK144252 and OK144253 to OK144255) and in human cases reported in the same months (EPI_ISL1315064, EPI_ISL1936365 and EPI_ISL3274157), as presented in Figure 3 and Supplementary Table S4 . Additionally, the veterinarian who was in contact with the infected cats was asymptomatic; however, he tested positive for COVID-19 and his sequence result was identical to that of the young cat and her mother. Moreover, sequence similarity between the SARS-CoV-2 spike gene detected in the young cat and her mother confirms that the virus is the same. Thus, there is a potential support of SARS-CoV-2 transmission between humans and cats, but further studies are needed to investigate the direction of the viral transmission as the study’s design has not determined the infection direction whether animal-to-human or human-to-animal or animal-to-animal viral infection occurred.

The present study revealed changes in the spike glycoproteins and the unique amino acid substitutions of the cat samples in correspondence with the reference Wuhan isolate, as illustrated in Table 1 , Figure 4 and Supplementary Figure S1 . Seven critical amino acid substitutions were observed. The diverse and frequent single residue mutations that emerged independently and repeatedly suggests their possible adaptive role (Kubik et al., 2021). Different spike regions are subjected to distinct evolutionary changes as mutations at receptor-binding domains (RBDs) result in both changes in the binding affinity to ACE2 receptors as well as in viral evasion from neutralizing antibodies, whereas mutations in N-terminal domains (NTDs) are mostly related to the potential escape from neutralizing antibodies (Kubik et al., 2021). Interestingly, some of these mutations are unique to the Wuhan-Hu-1 reference isolate and are reported in cat samples for the first time in this study, as they were not previously detected in cat samples in Hong Kong, France, Italy or the USA. The latter mutations are (W152R), (A570V), (Q677H), (A899S) and (S1051Y). In this study, we focused on the spike gene as it the most variable part of the viral genome and it plays crucial role in the receptor binding. It is also the protein that carries the receptor binding domain (RBD) which is the main point of entry to the host (Fung and Liu, 2019). The spike is also the main protein that exhibit mutations due to immune escape and generally have a major immunological role (Harvey et al., 2021).

Structural visualization and homology models revealed that four residue substitutions were located on the viral surface, which may have appeared as viral antigenic evolution to evade neutralizing antibodies. These mutations, (W152R), (L452R), (A570V) and (A899S), have weakened critical interaction points with the neutralizing antibodies and are associated with low sensitivity to mAbs and vaccine sera (Kubik et al., 2021). The observed (W152R) mutation is critical and is frequently observed across geographical and phylogenetic contexts; thus, it is a hotspot for mutations (Kubik et al., 2021). The substitution of tryptophan to leucine (W152L) or arginine (W152R) is the most common mutation at this location based on the SOPHiA DDM database. In addition, the substitution of tryptophan to cysteine (W152C) or glycine (W152G) is rare (Sophiagenetics, 2021). During the second wave of the SARS-CoV-2 pandemic, between December 2020 and April 2021, the (W152R) mutation increased steadily and rapidly, with one of the most notable adaptive mutations (N501Y) in the receptor-binding domain (RBD) coexisting (Kubik et al., 2021). Similarly, (W152R)substitution affects viral evasion from neutralizing antibodies and escape from immunity, as demonstrated in structural studies in which tryptophan (W) removal has weakened critical interaction points with antibodies (Kubik et al., 2021). In particular, a residue in the antibody chain interacts with W152 in a hydrophobic pi-stacking interaction, wrapping W152 inside one of the complementarity-determining region (CDR) loops in the variable chains of antibodies (Kubik et al., 2021). Indeed, W152 is a key position that contributes to the latter interaction. As a result, changes in W152 impair not only the pi-stacking interaction, but also the pocket created by the antibody’s CDR loop, resulting in a significant reduction in binding affinity (Kubik et al., 2021). Furthermore, (W152R) substitution suggests viral evolution to maintain effective binding to the ACE2 receptor (Kubik et al., 2021). Therefore, monitoring the individual mutations that occur outside the receptor-binding domain (RBD) and inside the N-terminal domain (NTD) is critical.

The second mutation observed in this study was (L452R) in the receptor-binding motif (RBM), as illustrated in Supplementary Figures S2B, C . This mutation was first reported in January 2021 in California, USA in the 20C/S:452R variant, and its frequency has increased over time (Zhang W et al., 2021). Since the first observation, novel variants of SARS-CoV-2, including this amino acid substitution, have been seen in the Delta/21A/B.1.617.2, Kappa/21B/B.1.617.1 and Epsilon/21C/B.1.427.9 viruses with increased virulence and infectivity (Lauring and Hodcroft, 2021). Additionally, in this mutation, a factor of a 3- to 6-fold reduction in antibody neutralization titers in vaccinated individuals was observed (McCallum et al., 2021). Furthermore, this substitution reduced the neutralizing activity of monoclonal antibodies against RBD by 14- to 35-fold (McCallum et al., 2021). Moreover, a 2-fold increase in viral shedding was reported in the involved variants (Deng et al., 2021). It was also found that (L452R) could strengthen ACE2 receptor-binding affinity (Chen et al., 2020) and increase viral infectivity (Liu et al., 2021). A mutation in the C-terminal domain 1 (CTD1) at position 570 was also observed. In addition, (A570V) was detected in North America and Asia (Guruprasad, 2021). Recently, biophysical and Cryo-EM studies demonstrated that some residue substitutions in the spike have affected the conformational dynamic equilibrium between the open RBD-up and closed RBD-down forms of the spike (Henderson et al., 2020). In particular, the authors found that certain residue substitutions in the regions of contact between domains could affect the equilibrium between the open and closed states. For example, the combined mutations in the A570, T572, F855 and N856 positions of SD1 and S2 inter-protomer regions have led to the dominance of the open RBD-up receptor-accessible form of the spike. Therefore, these structural rearrangements have resulted in the enhancement of the binding affinity of the spike to the ACE2 host receptor (Henderson et al., 2020). More specifically, the substitution of A570 with a more hydrophobic one, such as L570, has aided in such an equilibrium shift. Hence, the hydrophobic (A570V) mutation ( Supplementary Figure S1 ) observed in our samples may contribute to a similar equilibrium shift. The fourth mutation on the surface was (A899S) in the fusion-peptide proximal region (FPPR). This mutation was recently observed in Jordan (Alanagreh et al., 2021), Saudi Arabia (Alghoribi et al., 2021) and Italy (Castelli et al., 2021). This substitution is recorded in lineage C.36 and is associated with high transmissibility and low sensitivity to mAbs and vaccine sera (Castelli et al., 2021).

The fifth mutation in our study (D614G) was found downstream receptor-binding domain in the C-terminal domains 2 (CTD2). Such a mutation became dominant during the emergence of the COVID-19 pandemic (Korber et al., 2020a; Yurkovetskiy et al., 2020). It is not only able to enhance the viral/host interaction on the cellular level (Ortega et al., 2020), but it is also involved in the upward increase in SARS-CoV-2 pathogenicity, increasing viral transmission and evasion from immune response and vaccine (Korber et al., 2020b). This mutation appears to enhance the spread of SARS-CoV-2 and increases the viral infectivity that resulted from increasing the stability of the spike trimer. Certainly, in such a mutation, there is loop misplacement, structural rearrangements and interaction between downstream receptor-binding domains, which act as wedges within the spike trimer that strengthen the stability and prevent pre-maturation of the spike, and improve the strong assembly of the virus, resulting in the enhancement of the viral binding with the ACE2 receptor, as presented in Figure 4 . This could facilitate viral entry, infectivity and virulence. It is unsurprising that all newly emerged variants contain this mutation and that they are more infectious than the Wuhan reference virus (Zhang J et al., 2021). The (D614G) substitution was reported to improve viral replication in human lung epithelial cells and primary human airway tissues by improving viral infectivity and stability (Plante et al., 2021). Furthermore, this mutation lowered the effectiveness of COVID-19 vaccinations in clinical studies (Plante et al., 2021).

The (Q677H) mutation in the S1/S2 cleavage site was detected in both the two cat samples and the veterinarian sample in this study. The (Q677H) mutation was firstly identified in clade 20C in the USA (Pater et al., 2021). It was also detected in the (Eta/21D/B.1.525) variant (Hodcroft et al., 2021). As the position of such a mutation is found in a critical polybasic region of the furin-binding site and it could influence S1/S2 cleavage, SARS-CoV-2 infectivity and pathogenicity could be subsequently affected (Hodcroft et al., 2021). The seventh mutation observed in this study is (S1051Y) in the β-hairpin region of subunit 2 (S2). The three-dimensional organization of HR1 (residues 910−985), CH regions (residues 986−1035), β-hairpin (1035–1068) and UH (residues 736−781) is a significant feature of the spike glycoprotein structure (Walls et al., 2016). Structural modeling and analysis suggest that the (S1051Y) mutation ( Figures 4C–E ) structure. Indeed, this mutation could play a role in viral entry during the replication cycle. The latter hypothesis requires further experimental investigation.

Despite the multiple mutations found in the sequences presented in this study, all these mutations had been previously reported in humans. Hence, although some of the latter mutations may be important for the virus strain survival within cats, these mutations are not unique to this particular host, and it is likely that a cross-infection between humans and cats has occurred. These findings raise several questions, including: What factors control the SARS-CoV-2 host ranges? Do variations in the host receptor play a role in the susceptibility of the host to SARS-CoV-2 infection? Can we obtain insights into the susceptibility of a species by comparing the species receptor with the human receptor? To answer these questions, we analyzed the ACE2 receptors from various animal species using sequence and structural bioinformatics tools.

Obviously, many host factors, may play a crucial role in susceptibility to SARS-CoV-2 infection. One of the possible factors likely to influence the virus host range is the variation in the viral receptor itself from one species to another. In this work, we showed that simple sequence comparisons ( Table 2 and Figure 5 ), phylogenetic analysis ( Figure 6 ) and the structural modeling ( Figure 7 ) of residues at the binding interface between ACE2 receptors from different hosts and the viral RBD may be used as an initial effective indicator of the susceptibility of the host to SARS-CoV-2 cross-infection. More specifically, the amino acid composition of this region (residues 19–90 and 324–393) may provide an indication of the binding capacity of the virus to the specific species host cell, and hence a determinant for the susceptibility of the host. This study, therefore, reveals that a simple phylogenetic analysis and identity matrix of the aligned residues could provide a quantitative measure of the likelihood of virus entry into the cells of various hosts.

The phylogenetic analysis and identity matrix of the aligned residues at the binding interface between the ACE2 and the viral RBD ( Figure 6B and Supplementary Table S5 ) revealed that SARS-CoV-2 has the ability to interact with a broad range of ACE2 cell receptors in different species, which could increase its virulence and could enhance its inter-species transmission (Su et al., 2016) (Bolles et al., 2011). The sequence identity of the latter region between humans and cats (83.8%) is higher than that between humans and dogs (79.5%), which suggests that the potential recognition of the RBD-spike glycoprotein in cat ACE receptors is stronger than that in dogs, and these results are in accordance with a previous study in China (Luan et al., 2020). This high sequence identity in cats compared to human could explain the relatively high SARS-CoV-2 infection rate in the feline species (Hamer et al., 2021). Hence, regular monitoring of companion animals for SARS-CoV-2 infection is highly recommended. The similarity percentage in ermines (minks) (78.1%) is lower than that in cats and dogs, however the reported infection rate of SARS-CoV-2 in minks is high, according to the fifth report of the OIE (2021). This may be an indicator that there are other effective factors in minks which make this species more susceptible to SARS-CoV-2 than others, and these factors need further investigation. The sequence identity was 74.6% in greater horseshoe bats and 78.8% in Chinese pangolins. Previous studies have suggested that SARS-CoV-2 from bat CoV RaTG13 and pangolin isolates, have sequence identities of 96.3% and 89%, respectively. These studies concluded that, bats could be the natural origin of SARS-CoV-2 (Bosco-Lauth et al., 2020; Shi et al., 2020), and although pangolins cannot transmit SARS-CoV-2 directly due to the percentage of divergence, they may play an intermediate role (Calvet et al., 2021). Indeed, there is a need for further investigation to discover other factors that may point to bats being the potential natural host for SARS-CoV-2, rather than the receptor homology and binding affinity. Overall, the presented work paves the way for binding affinity predictions and free energy calculations through molecular dynamics studies, and opens the gates for further investigation into species susceptibility to SARS-CoV-2 infection.

This study displays critical mutations in the spike glycoprotein of SARS-CoV-2 detected in cats and in their treating veterinarian, through full spike gene sequencing, structural visualization and homology modeling. The analysis of the detected mutations revealed potential evidence for spillover of SARS-CoV-2 between humans and cats. Furthermore, the reported mutations could affect the viral evasion from immune response, the viral infectivity and pathogenicity. Hence, regular complete genome sequencing of SARS-CoV-2 in humans and nearby companion animals, as well as studying the recombination events in such viruses, are recommended in order to track newly emerging mutations and investigate their effects on viral adaptation, transmission, and their impact on therapeutics and vaccine design.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The study was conducted according to ethical guidelines and approved by the Ethics Committee of the Animal Health Research Institute (protocol code 11429 and date of approval 12/2020). The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the owners for the participation of their animals in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Conceptualization, MH, AE-D, NH, MS, and HH; methodology, MH, AE-D, NH, MS, and HH; software, MH and OA; validation, MH, AE-D, NH, MS, OA, and HH; formal analysis, MH, OA, and HH; investigation, MH, AE-D, NH, MS, OA, and HH; resources, MH, AE-D, NH, MS, OA, and HH; data curation, MH, AE-D, NH, OA, and HH; writing—original draft preparation, MH, OA, and HH; writing—review and editing, MH, OA, and HH; visualization, MH and OA; supervision, AE-D, NH, MS, and HH; project administration, MH, NH, and HH; funding acquisition, MH. All authors have read and agreed to the published version of the manuscript.

This research was funded by the Science, Technology & Innovation Funding Authority (STDFA), grant number 44585.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.