Mutations arising in SARS-CoV-2 spike on sustained human-to-human transmission and human-to-animal passage

Mutations arising in SARS-CoV-2 spike on sustained human-to-human transmission and human-to-animal passage

Robert F. Garry1,2

1Department of Microbiology and Immunology, Tulane University Medical Center, 1430 Tulane Avenue, New Orleans, Louisiana 70112 USA; E-Mail: [email protected]

2Zalgen Labs, LLC, Germantown, MD, USA

The proximal origins of SARS-CoV and MERS-CoV from civets and camels, respectively, are well documented. Few genetic changes in these viruses are required for the interspecies transfers to humans (Li, 2008). While precise details and timing of the evolutionary pathways remain to be elucidated, it is also apparent that SARS-CoV-2 emerged from the Sarbecovirus subgenus of the Betacoronaviruses via one or more interspecies transfers (Andersen et al., 2020; Boni et al., 2020). In contrast to SARS-CoV and MERS-CoV, SARS-CoV-2 has had an extended period of human-to-human transmission. While the evolutionary rate is not unusual for an RNA virus, mutations have occurred that appear to impact SARS-CoV-2 fitness (Kemp et al., 2020; Volz et al., 2020). For example, SARS-CoV-2 carrying G614 has replaced D614 as the predominant circulating variant (Volz et al., 2020). The D614G substitution abolishes a hydrogen-bond interaction with T859 of a neighboring monomer, which destabilizes the spike trimer and increases interaction of the receptor binding domain (RBD) with angiotension-converting enzyme 2ACE2 . By increasing viral load in the upper respiratory tract of COVID-19 patients, D614G may enhance SARS-CoV-2 transmission (Plante et al., 2020).

Recently, a SAR-CoV-2 variant emerged in the UK that has acquired 17 mutations, including 8 in spike (Rambaut et al., 2020). An apparently independent lineage emerged in South Africa that also has multiple spike mutations (Pond et al., 2020). Spike mutations have also occurred during interspecies transfers of SARS-CoV-2 from humans to animals, both during establishment of experimental models of COVID-19 and as an unintended consequence of human interactions with domestic, curated and commercial animals (Mahdy et al., 2020). Here, some of the mutations that have occurred to date in the SARS-CoV-2 spike during human-to-human transmission and following human-to-animal passage are compiled. This compilation highlights several commonly occurring natural features of coronavirus spike evolution that may be involved in interspecies transfers.

Homology modeling of the SARS-CoV-2 spike was performed in SWISS-MODEL (Waterhouse et al., 2018) using reference sequence QHD43416.1 and a closed prefusion configuration of the spike trimer pdb 6VXX (Walls et al., 2020) as template. The resulting model includes amino acids that are disordered in the spike cryoelectron microscopy structure and reverts the furin cleavage site and proline mutations used to stabilize the trimer. As in structures of all other CoV spikes, this model lacks most of the C-terminal helix (heptad repeat 2), membrane-proximal external region, transmembrane and intracellular domains.

Spike proteins carrying representative mutations that arise on human-to-human passage and human-to-animal passage included:

Spike amino acid sequences were aligned using Clustal Omega (Sievers et al., 2011).

Figure 1. Compilation of SARS-CoV-2 spike mutations occurring in humas and animals. Red spheres: United Kingdom (UK) variant, Blue spheres: South African (ZA) variant, Magenta: both UK/ZA variants, Yellow spheres: animals as indicated in the inset. NTD: Amino-terminal domain. RBD: Receptor binding domain.

Mutation of two amino acids (K479N and S487T) in palm civet SARS-CoV RBD allows this virus to infect humans (Li, 2008). These changes overcome the species barrier between civets and humans and enable favorable interactions between the RBD and contact residues on human ACE2. Q493 and N501 are the two SARS-CoV-2 residues that have similar interactions with ACE2 as SARS-CoV residues N479 and T487. The UK variant (Variant of Concern 202012/01, B.1.1.7) contains spike mutations N501Y, as well as delH69, delV70, delY145, A570D, P681H, T716I, S982A and D1118H and is estimated to have emerged in the UK during the Fall of 2020 (Fig. 1). N501Y likely enables a Pi-Pi interaction between Spike 501Y and ACE2 41Y. The South African (ZA) variant (501Y.V2, B.1.351) also includes N501Y plus variously D80A, D215G, K417N, E484K and A701V (Fig. 1). The UK and ZA variants have now been detected elsewhere around the world. The African Centre of Excellence for Genomics of Infectious Diseases (ACEGID), Redeemer’s University, Nigeria, identified two examples of an SARS-CoV-2 spike that shares the P681H in common with the UK variant. This mutation alters the out-of-frame insertion (relative to other sarbecoviruses) that generates the minimal furin cleavage site in SARS-CoV-2 spike (Happi et al., 2020).

A variety of interspecies transfers of SARS-CoV-2 from humans to animals have occurred. (Fig. 1) The N501Y mutation, observed in the UK and ZA variants was selected in 6 passages in aged mice and enables efficient replication of SARS-CoV-2 (Sun et al., 2020). Further passages of the N501Y mutant resulted in selection of Q493H and K417N, which increased pathogenicity the mouse model. K417N is also observed in the ZA variant. Q493K, a related mutation to Q493H, also increases both replication and pathogenesis in mice (Leist et al., 2020). Another mouse strain (EPI_ISL_459910) generated by six passages is also characterized by Q493K. This variant also carries a deletion of amino acids Q675 to N679 that appears to have occurred after a single passage in Vero cells (Liu et al., 2020).

Humans with SARS-CoV-2 have infected minks on fur farms in several countries (Koopmans, 2021). Recurrent mutations observed in mink include Y453F, F486L and N501T, which are all in the RBD (Lassaunière et al., 2020; Oude Munnink et al., 2020) (Fig. 1). As is the case with N501Y in humans and mice, the N501T change may enhance binding to mink ACE2. The same N501T mutation was observed in passage of SARS-Cov2 ferret, another mustelid (Richard et al., 2020). Two furin cleavage site mutations have occurred in mink and ferrets, I692 V and S686 respectively.

Several examples of transfer to domestic cats have been documented (Braun et al., 2020; Hamer et al., 2020; Neira et al., 2020; Wu et al., 2020). Felids in zoos have also been infected with SARS-CoV-2, presumably via contact with humans. Dogs are also permissive for infection by SARS-CoV-2 (Hamer et al., 2020; Sit et al., 2020). A variety of mutations in SARS-CoV-2 after interspecies transfers to felines or canines have been observed some of which are common to other interspecies transfers (Fig. 1). However, none of the mutations appear essential for replication in these species. A sequence from a zoo tiger did not show any mutations (Wang et al., 2020).

The mutations compiled above fall into four general classes:

  1. RBD mutations, which are of importance because some may provide both immune escape and a fitness advantage. As previously documented for SARS-CoV (Li, 2008), RBD mutations, as exemplified by N501Y/T are also important for interspecies transfer.
  2. The amino terminal domain (NTD), particularly the portion most exposed on the virion surface, represents another hotspot for spike mutations. There is evidence for immune selection in this region and preliminary evidence that at least one of these changes delH69/delV70 could improve fitness (Kemp et al., 2020).
  3. Variation in or near the FCS is a common natural occurrence in CoV evolution (Gallaher and Garry, 2020). The changes during transfer of SARS-CoV-2 to species provide additional examples of the importance of the spike cleavage sites at S1/S2 junction and S2’ interspecies transfer. The impact of specific changes such as a P681H on transmissibility are important to determine.
  4. Several spike mutations group with D614G, including Q613R found in lions. Mutations that occur in the metastable regions of spike could influence refolding to the 6-helix bundle during virus entry possibly affecting infection efficiency. Another mutation cluster in a metastable region is at the base of the structural model that involves the prefusion to postfusion transition of spike and is also of importance for neutralizing antibodies in other class I viral fusion proteins (Hastie et al., 2017).

Elucidation of the mechanisms by which viruses adapt to different hosts thereby crossing species barriers is important for identifying potential epizootic threats. Widespread transmission of an emerging pathogen, such as SARS-Cov-2 can potentially lead to further mutations that affect transmissibility or effectiveness of countermeasures. Infrastructure for continuous monitoring of infectious viral diseases, as implemented in the UK, should be enabled worldwide to response to such changes. Monoclonal antibody immunotherapeutics should be formulated as cocktails to prevent mutational escape to single antibodies (Baum et al., 2020). Vaccines should be designed to maximize polyclonal immune responses to multiple protective epitopes.

I am grateful to scientists worldwide who have made SARS-CoV-2 genome sequences available to the research community prior to publication. Kristian G. Andersen, Edward C. Holmes and Andrew Rambaut provided essential input and discussion. Work on emerging viruses in the Garry Laboratory is supported by the National Institutes of Health, the Coalition for Epidemic Preparedness Innovations, the Burroughs Wellcome Fund, the Wellcome Trust, the Center for Disease Prevention and Control, and the European & Developing Countries Clinical Trials Partnership.


Andersen, K.G., Rambaut, A., Lipkin, W.I., Holmes, E.C., and Garry, R.F. (2020). The proximal origin of SARS-CoV-2. Nat Med 26, 450-452.

Baum, A., Fulton, B.O., Wloga, E., Copin, R., Pascal, K.E., Russo, V., Giordano, S., Lanza, K., Negron, N., Ni, M., et al. (2020). Antibody cocktail to SARS-CoV-2 spike protein prevents rapid mutational escape seen with individual antibodies. Science 369, 1014-1018.

Boni, M.F., Lemey, P., Jiang, X., Lam, T.T., Perry, B.W., Castoe, T.A., Rambaut, A., and Robertson, D.L. (2020). Evolutionary origins of the SARS-CoV-2 sarbecovirus lineage responsible for the COVID-19 pandemic. Nature microbiology 5, 1408-1417.

Braun, K.M., Moreno, G.K., Halfmann, P.J., Baker, D.A., Boehm, E.C., Weiler, A.M., Haj, A.K., Hatta, M., Chiba, S., Maemura, T., et al. (2020). Transmission of SARS-CoV-2 in domestic cats imposes a narrow bottleneck. Preprint.

Gallaher, W.R., and Garry, R.F. (2020). Naturally occurring indels in multiple coronavirus spikes.

Hamer, S.A., Pauvolid-Corrêa, A., Zecca, I.B., Davila, E., Auckland, L.D., Roundy, C.M., Tang, W., Torchetti, M., Killian, M.L., Jenkins-Moore, M., et al. (2020). Natural SARS-CoV-2 infections, including virus isolation, among serially tested cats and dogs in households with confirmed human COVID-19 cases in Texas, USA. bioRxiv :

Happi, C., Ihekweazu, C., Nkengasong, J., Oluniyi, P.E., and Olawoye, I. (2020). Detection of SARS-CoV-2 P681H Spike Protein Variant in Nigeria.

Hastie, K.M., Zandonatti, M.A., Kleinfelter, L.M., Heinrich, M.L., Rowland, M.M., Chandran, K., Branco, L.M., Robinson, J.E., Garry, R.F., and Saphire, E.O. (2017). Structural basis for antibody-mediated neutralization of Lassa virus. Science 356, 923-928.

Kemp, S.A., Harvey, W.T., Datir, R.P., Collier, D.A., Ferreira, I., Carabelli, A.M., Robertson, D.L., and Gupta, R.K. (2020). Recurrent emergence and transmission of a SARS-CoV-2 Spike deletion ΔH69/V70. bioRxiv 2020.12.14.422555; doi:

Koopmans, M. (2021). SARS-CoV-2 and the human-animal interface: outbreaks on mink farms. The Lancet Infectious diseases 21, 18-19.

Lassaunière, R., Fonager, J., Rasmussen, M., Frische, A., Strandh, C.P., Rasmussen, T.B., Bøtner, A., and Fomsgaard, A. (2020). Working paper on SARS-CoV-2 spike mutations arising in Danish mink, their spread to humans and neutralization data.

Leist, S.R., Dinnon, K.H., 3rd, Schäfer, A., Tse, L.V., Okuda, K., Hou, Y.J., West, A., Edwards, C.E., Sanders, W., Fritch, E.J., et al. (2020). A Mouse-Adapted SARS-CoV-2 Induces Acute Lung Injury and Mortality in Standard Laboratory Mice. Cell 183, 1070-1085.e1012.

Li, F. (2008). Structural analysis of major species barriers between humans and palm civets for severe acute respiratory syndrome coronavirus infections. J Virol 82, 6984-6991.

Liu, Z., Zheng, H., Lin, H., Li, M., Yuan, R., Peng, J., Xiong, Q., Sun, J., Li, B., Wu, J., et al. (2020). Identification of Common Deletions in the Spike Protein of Severe Acute Respiratory Syndrome Coronavirus 2. J Virol 94.:e00790-20. doi: 10.1128/JVI.00790-20.

Mahdy, M.A.A., Younis, W., and Ewaida, Z. (2020). An Overview of SARS-CoV-2 and Animal Infection. Frontiers in veterinary science 7, 596391.

Neira, V., Brito, B., Agüero, B., Berrios, F., Valdés, V., Gutierrez, A., Ariyama, N., Espinoza, P., Retamal, P., Holmes, E.C., et al. (2020). A household case evidences shorter shedding of SARS-CoV-2 in naturally infected cats compared to their human owners. Emerging microbes & infections, 1-22.

Oude Munnink, B.B., Sikkema, R.S., Nieuwenhuijse, D.F., Molenaar, R.J., Munger, E., Molenkamp, R., van der Spek, A., Tolsma, P., Rietveld, A., Brouwer, M., et al. (2020). Transmission of SARS-CoV-2 on mink farms between humans and mink and back to humans. Science. epublication 10.1126/science.abe5901.

Plante, J.A., Liu, Y., Liu, J., Xia, H., Johnson, B.A., Lokugamage, K.G., Zhang, X., Muruato, A.E., Zou, J., Fontes-Garfias, C.R., et al. (2020). Spike mutation D614G alters SARS-CoV-2 fitness. Nature. epublication 10.1038/s41586-020-2895-3.

Pond, S.L., K., Wilkison, E., Weaver, S., Jame, S.E., Tegally, H., de Oliveira, T., and Martin, D. (2020). A preliminary selection analysis of the South African V501.V2 SARS-CoV-2 clade.

Rambaut, A., Loman, N., Pybus, O., Barclay, W., Barrett, J., Carabelli, A., Connor, T.R., Peacock, T., Robertson, D.L., and Volz, E. (2020). Preliminary genomic characterisation of an emergent SARS-CoV-2 lineage in the UK defined by a novel set of spike mutations.

Richard, M., Kok, A., de Meulder, D., Bestebroer, T.M., Lamers, M.M., Okba, N.M.A., Fentener van Vlissingen, M., Rockx, B., Haagmans, B.L., Koopmans, M.P.G., et al. (2020). SARS-CoV-2 is transmitted via contact and via the air between ferrets. Nature communications 11, 3496.

Sievers, F., Wilm, A., Dineen, D., Gibson, T.J., Karplus, K., Li, W., Lopez, R., McWilliam, H., Remmert, M., Söding, J., et al. (2011). Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Molecular systems biology 7, 539.

Sit, T.H.C., Brackman, C.J., Ip, S.M., Tam, K.W.S., Law, P.Y.T., To, E.M.W., Yu, V.Y.T., Sims, L.D., Tsang, D.N.C., Chu, D.K.W., et al. (2020). Infection of dogs with SARS-CoV-2. Nature 586, 776-778.

Sun, S., Gu, H., Cao, L., Chen, Q., Yang, G., Li, R.-T., Fan, H., Ye, Q., Deng, Y.-Q., Song, X., et al. (2020). Characterization and structural basis of a lethal mouse-adapted SARS-CoV-2. bioRxiv doi:

Volz, E., Hill, V., McCrone, J.T., Price, A., Jorgensen, D., O’Toole, Á., Southgate, J., Johnson, R., Jackson, B., Nascimento, F.F., et al. (2020). Evaluating the Effects of SARS-CoV-2 Spike Mutation D614G on Transmissibility and Pathogenicity. Cell.
doi: 10.1016/j.cell.2020.11.020. Online ahead of print.

Walls, A.C., Park, Y.J., Tortorici, M.A., Wall, A., McGuire, A.T., and Veesler, D. (2020). Structure, Function, and Antigenicity of the SARS-CoV-2 Spike Glycoprotein. Cell 181, 281-292.e286.

Wang, L., Mitchell, P.K., Calle, P.P., Bartlett, S.L., McAloose, D., Killian, M.L., Yuan, F., Fang, Y., Goodman, L.B., Fredrickson, R., et al. (2020). Complete Genome Sequence of SARS-CoV-2 in a Tiger from a U.S. Zoological Collection. Microbiol Resour Announc 9. epublication 10.1128/mra.00468-20

Waterhouse, A., Bertoni, M., Bienert, S., Studer, G., Tauriello, G., Gumienny, R., Heer, F.T., de Beer, T.A.P., Rempfer, C., Bordoli, L., et al. (2018). SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res 46, W296-w303.

Wu, L., Chen, Q., Liu, K., Wang, J., Han, P., Zhang, Y., Hu, Y., Meng, Y., Pan, X., Qiao, C., et al. (2020). Broad host range of SARS-CoV-2 and the molecular basis for SARS-CoV-2 binding to cat ACE2. Cell discovery 6, 68.

1 Like


William R. Gallaher

Emeritus Faculty, LSU Health, New Orleans, LA, USA
and Mockingbird Nature Research Group, Pearl River LA, USA
Email: [email protected]

The foregoing post by Bob Garry brings together a lot of disparate information on the accumulation of SARS-CoV-2 mutations observed thus far in humans and animals. A master illustrator, his superimposition of mutations on the first 1147 amino acids of the spike glycoprotein is likely to already adorn the walls of virologists around the world as a useful reference. I certainly concur with his analysis, most especially the four major points he makes in his conclusions. This will come as no surprise to those familiar with the close professional relationship between Bob and myself over four decades.

I would like to add a few points to his post from my perspective.

First of all, not all amino acid substitutions are equal, even when they appear to affect, as in the case of S1147L, a region of potential significance (as Bob aptly pointed out). Medical educators have long taught young medical officers the adage: “When you hear hoofbeats, think horses, not zebras.” Focus first on the more common and significant etiologies of what you observe, without being distracted by the possibility of the more exotic. Of the 40 mutations on Bob’s chart, there are a lot of zebras, mutations observed only in animals and a number of them only once months ago. We need to maintain greater attention on those like N501Y, D614G and P681H, horses rapidly galloping through the human population that are likely to significantly affect the efficiency of human to human transmission.

The horse analogy is apt in another way. When a new major variant like B.1.1.7 is detected, and determined to be significant, we are using travel restrictions too late to be of any good. As detection of the variant in several parts of the US clearly shows, that horse left the barn door of the UK months ago. Repeatedly now, the virus has shown itself much swifter of foot than we are. Greatly increased sequence surveillance on an ongoing basis is a far better weapon than imposing travel restrictions. Similarly, US states test for COVID at four-fold different rates, even in adjacent states (32 per 100 people in PA; 139 per 100 in NY). Jurisdictions run by the four horsemen, See No Evil, Hear No Evil, Speak No Evil, and Smell No Evil, are not pulling their weight, only prolonging the agony of the pandemic. We need to know far more about the virus in the human population, far more quickly. There is doubtlessly far more out there than we are seeing reflected in Bob’s chart. What we don’t know does hurt us.

However, we should not let the number of mutations charted, or the hazards presented by the likes of B.1.1.7, to leave us with the impression that the virus is rapidly mutating, or threatening the efficacy of the vaccine. The global rate of daily viral replication is truly massive now, in the quadrillions of genomes per day, yet we see very little mutation of any significance in the human population. The vaccine produces a robust polyclonal immune response that is not likely to be circumvented by mutation at the observed rate any time soon.

One specific observation I would make is that, of the 40 mutations on the chart, 9, nearly a quarter of them, involve changes removing (3) or adding (6) a histidine. In the reference SARS CoV 2 Hu-1 sequence, histidine constitutes only 1.2% of the protein. That so many observed changes involve a relatively rare amino acid is interesting, particularly since the zwitterionic and nucleophilic properties of histidine add a specific functionality wherever it appears.

Finally, I focus attention on what is not seen in the post, the base of the spike protein from amino acids 1148 to 1273, that includes nearly the entire Heptad Repeat 2 (HR2) region, membrane spanning region, and internal region of the protein. In Figure 1, I have constructed a model of the base regions. The model is based on the same methodology I used previously to model the homologous protein regions of the retroviruses, filoviruses and arenaviruses based on the known propensities of each amino acid to be found in either an alpha helix, beta pleated sheet, turn or random coil (1-4).

Why a model? Because, depending on imaging method, we do not yet have a consensus of the overall structure of this region. Bob’s illustration reflects the general finding from X-ray crystallography of the extramembranal spike structure that reports this region as “disordered”, i.e. not any consistent pattern of electron density (5). On the other hand, protein fragments, without glycosylation, show the classic post-fusion pattern of a six-helix bundle, anything but disordered, even though the “fusion core” interacting the helices only runs 19 amino acids, or 5.5 turns (6). A third view is provided by cryogenic electron microscopy of virions (7), showing a substantial base that is reflected in the CDC cartoon of the virus seen nightly on every TV screen in the world.

What we have with regard to aa1147 to 1273 is an example of “observer effect”, common in particle physics but not so much recognized in structural biology. The premise is that the act of observation, in this case the method of fixation, changes what is observed. Whether it is crystallization, desiccation, deglycosylation, fragmentation, chemical fixation or flash-freezing, a conformationally dynamic protein is driven to a constant structure set at a minimum of free energy to produce an image of maximum atomic resolution.

The conformational and polymorphic dynamism of fusion/entry proteins is lost in their imaging, for the same reason Abraham Lincoln (or Queen Victoria) was never photographed smiling. The conditions of imaging with long exposure times precluded the contraction of facial muscles that could not be held long enough to keep the image in focus. So proteins never breathe or flex, and Lincoln never smiles, when their picture is being taken.

Clearly the overall image is accurate in most cases, just possibly more compact than in real life, with zero of the conformational undulation that is central to its polymorphic potential. The existence of the six-helix bundle and the fusion core in the post-fusion form is confirmed by the extraordinary antiviral activity of fusion inhibitors like Fuzeon for HIV-1 in the 1990s (8) and the EK1C4 inhibitor being developed by Shibo Jiang’s group for SARS-CoV-2 (6), both effective in the nanomolar range. What is far from clear is the pre-fusion structure of this region.

A model is one way to examine the several peptide regions of the aa1143 through aa1273 region of the S2 protein, consider the effect of carbohydrate that comprises 25% of its total molecular weight, and recognize several important features. Whether the model correctly predicts the structure is of minor importance. Careful study of a comprehensive image of the peptide sequence is the point.

The first observation to be made from the model is its truly incredible constancy in amino acid sequence. Over more than 125 amino acids, there is only a single amino acid substitution, L1224F, in a single dog sequence from Texas last July, to add to the 40 in Bob’s chart. None observed in humans at all. Indeed, between SARS-CoV Urbani of 2003, and SARS-CoV-2 Hu1, there are only 3 amino acid substitutions, 2 of them I/V and M/L that are trivial. The otherwise quite disparate SARS-CoV of 2003 and SARS-CoV-2 of 2019 differ in this region only 2.3%, despite a tidal wave of 67 (17%) overwhelmingly synonymous mutations as genetic background. Indeed, the overall restriction of mutation to the wobble base is so obvious that one can tell the reading frame by simply looking at the nucleotide sequence, without reference to the start or stop codons or the amino acid sequence. This is an amino acid sequence, and glycosylation pattern, cast in stone during incessant viral replication, likely for well over a century, and throughout a mind-boggling degree of replication in humans over the last 10 months.

There is a lot of potential functionality packed into the 131 amino acids shown in the model. It divides rather obviously into different regions.
At the top, overlapping Bob’s figure, is the sequence 1143PELDSFKEELDKYFK. This region has a high propensity to form an amphipathic alpha helix (its content of E, L, K and F). It is almost a direct tandem repeat of a sequence similar to the HIV sequence ELDKWA that defines the broadly neutralizing 2F5 monoclonal determinant (9). The portion in bold type constitutes a Cholesterol-Recognition Amino acid Consensus (CRAC) sequence (10), also reminiscent of the LWYIK motif that closely follows the ELDKWA motif in HIV-1 (11). There are two other CRAC motifs in SARS-CoV-2, one just prior to the fusion peptide region, and the second just prior to membrane insertion that we will get to here in a bit. S2 is rather loaded with motifs with a known affinity for cholesterol concentrated in target membranes.

This is followed by a glycosylated region that does not have high helical potential, with only a single L and single A with such propensity, 1158NHTSPDVDLGDISGINASVVN with the N-glycosylation sites in bold type. Notably, these sites are likely occupied by high-mannose type glycosyl moieties, adding about 6000 daltons of hydroxyl-rich carbohydrate. X-ray crystals of the six-helix bundle include this region in the extended alpha helix, but lack any effect of this glycosylation on the actual structure.

Next is the fusion core region 1179IGKEIDRLNEVAKNLNESSL (6), with the third glycosylation site at its base. This and the remainder of the extramembranal region, through a second CRAC sequence 1203LGKYEQYIK, model as a strongly amphipathic alpha helix covering 35 amino acids and 10 helical turns – clearly a defining feature of the HR2 domain. This is immediately followed by an intensely hydrophobic sequence, here 1212WPWYIWLGF (5/9 aromatic, one-third tryptophane) uniformly found in Class I Viral Fusion/Entry Proteins (1,2). We term this the Juxtamembrane Aromatic Rich region (JAR) that clearly functions as a critical partner to the fusion peptide in inducing membrane fusion. It is projected here as continuing the alpha helix due to the potential stacking of the two tryptophanes (W) adjacent in the helical configuration. The combination of a CRAC motif followed by an especially potent JAR motif is a powerful region for cholesterol-targeted membrane perturbation. It is among the most powerful among any of the Class I Fusion Proteins, including HIV-1 and Ebola. The efficiency of the 1179-1220 region of SARS-CoV-2, or the virtually identical equivalent in SARS-CoV Urbani, as a fusion machine cannot be overestimated.

The functionality of S2 does not end there. The membrane-spanning region begins routinely enough, but is followed on the cytoplasmic side by an unusual clustering of no less than 10 cysteines. Within and just below the membrane, these are likely acylated by fatty acids, to help anchor the large spike protein to the bilayer with greater avidity. Beyond that point, they constitute a phalanx of free sulfhydryl groups for intermolecular interaction.

This is followed by a charged region, principally a cluster of acidic residues, with significant alpha helical potential, before the protein terminates in the 1271HYT triplet. This histidine can combine with any of the available sulfhydryls to bind divalent cations. Ca2+ is shown, but the open configuration would accommodate any divalent cation.

In conclusion, the portion of S2 below the most widely published structure of the spike glycoprotein, from aa1143 on, can be visually subdivided into 9 discernible structural and functional regions, each perfected to a virtually constant sequence over an immeasureable replication history covering much more than a century of Sarbecovirus evolution. I, for one, regard it in awe. The phrase “fearfully and wonderfully made” comes to mind.

Yes, all of what I have described is in the vaccine, but expressed to a far lower degree than due to active infection. Also, all of the many other weapons of SARS-CoV-2 have been excluded from the vaccine. Both my wife and I have appointments to be inoculated with the vaccine in a few days, and approach being immunized with enthusiasm.

ACKNOWLEDGMENTS: I thank Bob Garry for graciously forwarding me his ALN file for the sequences he used, to permit the quick check of aa1148-1273 for additional mutations. I also dedicate this post to my former wife Betty Jean Burton Grier Gallaher RN, a heroic ER nurse, who passed away January 10, 2021 from COVID subsequent to her coming out of retirement to help treat COVID patients. In her late 70s, at high risk to herself, she went in anyway, to do what she could in the COVID crisis that would take her own life from her.


  1.               Gallaher, W. R., Ball, J.M., Garry, R.F., Griffin, M.C., and Montelaro, R.C 1989  “A general model for the transmembrane proteins of HIV and other retroviruses,”  AIDS Research and Human Retroviruses 5, 431-440.
  2.               Gallaher, W. R. 1996 “Similar structural models of the transmembrane proteins of Ebola and Avian sarcoma viruses,” Cell 85: 477-478.
  3.               Gallaher, W.R., de Simone, C. and Buchmeier, M. (2001). The viral transmembrane superfamily:  possible divergence of arenavirus and filovirus glycoproteins from a common RNA virus ancestor. BMC Microbiol 1, 1.
  4.              Gallaher, W.R. and Garry, R.F. (2015).  Modeling of the Ebola virus delta peptide reveals a potential lytic sequence motif. Viruses 7(1):285-305. doi: 10.3390/v7010285.
  5.             Walls, A.C., Park, Y.J., Tortorici, M.A., Wall, A., McGuire, A.T., and Veesler, D. (2020). Structure, Function, and Antigenicity of the SARS-CoV-2 Spike Glycoprotein. Cell 181, 281-292.e286.
  6.             Xia S, Liu M, Wang C, Xu W, Lan Q, Feng S, Qi F, Bao L, Du L, Liu S, Qin C, Sun F, Shi Z, Zhu Y, Jiang S, Lu L. Inhibition of SARS-CoV-2 (previously 2019-nCoV) infection by a highly potent pan-coronavirus fusion inhibitor targeting its spike protein that harbors a high capacity to mediate membrane fusion. Cell Res. 2020 Apr;30(4):343-355. doi: 10.1038/s41422-020-0305-x.
  7.             Wrapp D, Wang N, Corbett KS, Goldsmith JA, Hsieh CL, Abiona O, Graham BS, McLellan JS. Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation. Science. 2020 Mar 13;367(6483):1260-1263. doi: 10.1126/science.abb2507.
  8.             Wild CT, Shugars DC, Greenwell TK, McDanal CB, Matthews TJ. Peptides corresponding to a predictive alpha-helical domain of human immunodeficiency virus type 1 gp41 are potent inhibitors of virus infection. Proc Natl Acad Sci U S A. 1994 Oct 11;91(21):9770-4. doi: 10.1073/pnas.91.21.9770.
  9.             Muster T, Steindl F, Purtscher M, Trkola A, Klima A, Himmler G, Rüker F, Katinger H. A conserved neutralizing epitope on gp41 of human immunodeficiency virus type 1. J Virol. 1993 Nov;67(11):6642-7. doi: 10.1128/JVI.67.11.6642-6647.1993.
  10. Li H, Papadopoulos V. Peripheral-type benzodiazepine receptor function in cholesterol transport. Identification of a putative cholesterol recognition/interaction amino acid sequence and consensus pattern. Endocrinology. 1998 Dec;139(12):4991-7. doi: 10.1210/endo.139.12.6390.
  11. Epand, R. F., Thomas, A., Brasseur, R., Vishwanathan, S. A., Hunter, E., & Epand, R. M. (2006). Juxtamembrane protein segments that contribute to recruitment of cholesterol into domains. Biochemistry, 45(19), 6105–6114.

This is an updated figure to include P.1 mutations.

zoo6a plus P1 copy.pdf (2.2 MB)

zoo plus california copy.pdf (2.5 MB)

Update/correction includes B.1.429.

zoo plus california copy.pdf (2.5 MB) Update to clarify that the amino acids deleted in some B.1.351 are either 241-3 or 242-4. The deletion is out of frame.

Updated model to add amino terminal sequences, including L18F, T20N and P26S zoo plus cali humans and animals copy.pdf (2.4 MB)

zoo plus cali humans and animals 4-26-21 copy.pdf (2.5 MB) Adding B.1.526 and B.1.617

I had hoped not to have to update this.

Prior version waspublished in Telenti et al., Nature (2021) Aug;596(7873):495-504. doi: 10.1038/s41586-021-03792-w.

This version compares Alpha and Delta to Omicron.

zoo plus Omicron 11-27-21a.pdf (1.8 MB)

This version makes corrects mutations sustained at Q493 and Q498 in humans (Omicron) and mice plus adds a few mink mutations. zoo plus Omicron 12-3-21 copy.pdf (1.8 MB)

BA.1-3 copy.pdf (1.5 MB)

Mutations arising in Omicron subvariants included those shared with prior variants and those arising upon animal-to-human transfer.

BA.1-3 Venn.pdf (274.7 KB)

Venn diagram comparing Omicron subvariant mutations in Spike. This figure is based on:

Figure 1 of Desingu PA, Nagarajan K, and Dhama K. 2022. Emergence of Omicron third lineage BA.3 and its importance. J Med Virol. Online prior to print.

William Haseltine has also published a Venn diagram including all the Omicron subvariant mutations: Birth Of The Omicron Family: BA.1, BA.2, BA.3. Each As Different As Alpha Is From Delta.

Residues in which mutations have arisen upon passage in rodents are highlighted.