Role of spike compensatory mutations in the interspecies transmission of SARS-CoV-2

Abstract

SARS-CoV-2, the virus responsible for COVID-19 in humans, can efficiently infect a large number of animal species. Like any virus, and particularly RNA viruses, SARS-CoV-2 undergoes mutations during its life cycle some of which bring a selective advantage, leading to the selection of a given lineage. Minks are very susceptible to SARS-CoV-2 and owing to their presence in mass rearing, they make a good model for studying the relative importance of mutations in viral adaptation to host species. Variants, such as the mink-selected SARS-CoV-2 Y453F and D614G or H69del/V70del, Y453F, I692V and M1229I were identified in humans after spreading through densely caged minks. However, not all mink-specific mutations are conserved when the virus infects human populations back. Many questions remain regarding the interspecies evolution of SARS-CoV-2 and the dynamics of transmission leading to the emergence of new variant strains. We compared the human and mink ACE2 receptor structures and their interactions with SARS-CVoV-2 variants. In minks, ACE2 presents a Y34 amino acid instead of the H34 amino acid found in the human ACE2. H34 is essential for the interaction with the Y453 residue of the SARS-CoV-2 Spike protein. The Y453F mink mutation abolishes this conflict. A series of 18 mutations not involved in the direct ACE2 interaction was observed in addition to the Y453F and D614G in 16 different SARS-CoV-2 strains following bidirectional infections between humans and minks. These mutations were not random and were distributed into five different functional groups having an effect on the kinetics of ACE2-RD interaction. The interspecies transmission of SARS-CoV-2 from humans to minks and back to humans, generated specific mutations in each species which improved the affinity for the ACE2 receptor either by direct mutation of the core 453 residue or by associated compensatory mutations.

Keywords: ACE2; COVID-19 vaccine; Coronavirus; Mink; SARS-CoV-2 variants; hamster.

Fig. 1

A comparison of ACE2 sequences from 24 different species. The species considered bling to belonging to the Hominidae, Cercopithecidae, Felidae, Lustelidae, Viverridae, Suidae, Rhinolophidae, Manidae, Muridae, Phasianidae, Trinychidae, Pipidae and Elapidae families. The multialignment was performed using Clustal Omega multiple sequence alignment. All sequences were obtained from the NCBI reference sequence database (see Materials and Methods) and were numbered according to amino acid position on the Homo sapiens ACE2 protein. In the schematic representation the comparison of the Homo sapiens ACE2 consensus protein and ACE2 orthologs was focused on sequence region 30–41, particularly the amino acid residue 34. The amino acids (single letter code) are in black. The amino acids at position 34 are highlighted in blue. Some of the amino acids known to be important for viral tropism are highlighted in red. The peptide sequence comparison includes Homo sapiens (human), Macaca mulatta (monkey), Felis catus (cat), Panthera tigris (tiger), Neovison vison (American mink), Mustela nigripes (black-footed ferret), Mustela putorius furo (ferret), Mustela lutreola (European mink), Mustela erminae (ermine), Paguma larvata (palm civet), Sus scrofa domestic (pig), Sus scrofa (boar), Rhinolophus affinis (bat), Rhinolophus macrotis (bat), Rhinolophus sinicus (bat), Rhinolophus pearsonii (bat), Manis javanica (pangolin), Rattus rattus (rat), Mus musculus (mouse), Gallus gallus (hen), Pelodiscus sinensis (turtle), Xenopus tropicalis (frog), and Ophiophagus hannah (snake). For complete sequences comparison of the ACE2 polymorphism among Chiroptera see [17]. For complete sequences comparison of the ACE2 polymorphism among Mustelidae see [31]. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Structural homology between human and mink (EU and US) ACE-2. a-b. Superimposition of human and mink ACE2 structures. The models in panels a and b show two opposite views of superimposed structures of the N-terminal domain of human, European mink (Eu Mink)/Mustela lutreola and American mink (US Mink)/Neovison vison ACE2 with the location of H34 and Y34, as well as Y41 and Y83. H34 amino acid (human ACE2) is colored in green. Mink Y34 amino acid is colored in yellow (Eu Mink) and blue (US Mink). c. Representation of the conformation of RBD in interaction with the N-terminal domain of ACE2. In this representation ACE2 is above the SARS-CoV-2 RBD. H34 amino acid (human ACE2) is colored in green. Mink Y34 amino acid is colored in purple (Eu Mink) and cyan (US Mink). The potential clash between Y453 and Y34 (EU and US minks) is indicated by an arrow. A mutation of Y453 in the spike RBD is essential to avoid this clash. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Interaction of human and mink ACE2 with the RDB in the viral spike of the Wuhan HU1 and selected variants of SARS-CoV-2. a. Human ACE2-RBD Wuhan HU1 complex. Left panel: The electrostatic surface potential of human ACE2 is mostly electronegative (red areas), with sparse electropositive spots (blue areas). The surface recognized by the RBD is indicated by a yellow frame. Middle panel: molecular complex between the Wuhan HU1 RBD and human ACE2. The N501 in the viral spike interacts with residue Y41 of ACE2. Y453 interacts with Y34, and F486 interacts with Y83. Right panel: electrostatic surface potential of the RBD. The electropositive areas are complementary to the electronegative interface of ACE2. b. EU mink AC2 interaction with the Wuhan HU1 RBD. Left panel: electrostatic surface potential of EU mink ACE2. It is clearly distinct from human ACE2 and more electropositive. Right panel: EU mink ACE2 interaction with the Wuhan HU1. c. EU mink ACE2 interaction with the Y453F variant. d. US mink AC2 interaction with the Wuhan HU1 RBD. Left panel: electrostatic surface potential of US mink ACE2. It is clearly distinct from both human ACE2 and EU mink ACE2. Right panel: EU mink ACE2 interaction with the Wuhan HU1. e. US mink ACE2 interaction with the Y453F/F486L/N501T variant. The H34 in human ACE2 is essential for the interaction with the Y453 of the SARS-CoV-2 Wuhan HU-1 reference strain (a). The molecular models (b-e) Y34 in minks ACE2 causes a conflict when approaching the viral spike RBD. Specifically, the OH group of Y453 is too close from Y34 (arrows in b and d), which is obviously not the case when Y453 faces H34 in the human ACE2 (a). The Y453F found in mink SARS-CoV-2 eliminates this conflict (c and e). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Maximum likelihood phylogenetic trees of the SARS-CoV-2 spike gene and protein from different isolates. Sample names are built with the Genbank of GISAID accession number followed by a four-letter code (Mlut for Mustella lutreola, Nvis for Neovison vison and Hsap for Homo sapiens) identifying the species and a country code (DK for Denmark, NL for The Netherlands and USA for United States of America) indicating the geographical origin of the sample except for the sequences from China in which the city of origin is indicated. The tree was rooted using the spike gene sequence of the Wuhan Hu1 SARS-CoV-2 strain. The colour code corresponds to the nature of the 453 and 654 residues. Blue: Outgroup rooting the tree. Purple: Y453 + D614. Yellow: Y453 + D614G. Green: Y453F + D614. Brown: Y453F + D614G. a) Maximum likelihood (HKY) phylogenetic tree of the SARS-CoV-2 spike genes. b) Maximum likelihood (Q.plant + G + F) phylogenetic tree of SARS-CoV-2 spike. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Position of the principal mutations in the SARS-CoV-2 spike protein. For clarity, two views of the spike protein are shown, one in ribbons (left panel), the other one in ribbon + surface rendition (right panel). A colour code indicates the localization of the mutations: green, signal sequence; red, RBD; orange, cleavage site (*); cyan, amino acid residues involved in the conformational change that demasks the RBD; purple, amino acid residues involved in the fusion machinery. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Supplementary Fig. 1

Schematic representation of the quasispecies evolutionary process.