Mutations that adapt SARS-CoV-2 to mink or ferret do not increase fitness in the human airway

Abstract

SARS-CoV-2 has a broad mammalian species tropism infecting humans, cats, dogs, and farmed mink. Since the start of the 2019 pandemic, several reverse zoonotic outbreaks of SARS-CoV-2 have occurred in mink, one of which reinfected humans and caused a cluster of infections in Denmark. Here we investigate the molecular basis of mink and ferret adaptation and demonstrate the spike mutations Y453F, F486L, and N501T all specifically adapt SARS-CoV-2 to use mustelid ACE2. Furthermore, we risk assess these mutations and conclude mink-adapted viruses are unlikely to pose an increased threat to humans, as Y453F attenuates the virus replication in human cells and all three mink adaptations have minimal antigenic impact. Finally, we show that certain SARS-CoV-2 variants emerging from circulation in humans may naturally have a greater propensity to infect mustelid hosts and therefore these species should continue to be surveyed for reverse zoonotic infections.

Keywords: ACE2; COVID-19; SARS-CoV-2; antigenicity; coronavirus; ferret; mink; pandemic; zoonosis.

Graphical abstract

Figure 1

Passage of SARS-CoV-2 in ferrets results in spontaneous emergence of the mink-associated mutations Y453F and N501T (A) Ferret transmission chains from a previous study were deep sequenced to investigate any changes that occurred during infection and transmission of ferrets. Gray lines indicate previously described RNA shedding patterns seen in each ferret – pie charts indicate RBD mutations seen at each time point (as determined by deep sequencing) indicated by a black arrow. Sample from “Contact #1” marked with double dagger indicates virus that was used for subsequent experiments in Figure 2. (B) Maximum-likelihood phylogeny of SARS-CoV-2 genomes sampled from American mink (Neogale vison, formerly Neovison vison), highlighting the spike mutations ∆69-70, Y453F, F486L, or F486I, N501T, and D614G. Tip nodes are shown as points colored by sampling location, according to the color key. Columns to the right show the presence of either the WT amino acid(s) (light gray) or the mutations annotated above (colored bars). Major epidemiological lineages designated with the Pango nomenclature system are labeled. Black arrow indicates the branch that constitutes the Danish mink strain known as Cluster 5. At position 486, mutant viruses possessed 486L (leucine) except for a monophyletic clade formed of 20 sequences sampled in Latvia that possessed 486I (isoleucine) that are marked by a white asterisk.

Figure 2

The spike mutation, Y453F, enhances replication and morbidity in ferrets (A) Deep sequencing of RBD mutations of SARS-CoV-2 from ferret passage 2 swab (see Figure 1A) before and after isolation in Vero cells. (B–E) RNA (B) and infectious virus (C) shedding dynamics of ferrets directly infected with either WT (orange circles; as previously described in Peacock et al. [2021a]) or Y453F (i.e., ferret passage 2; black and white squares) SARS-CoV-2. n = 4 naive ferrets in each group were infected with 10⁵ p.f.u. of either virus. Percentage weight loss (D) and change in body temperature (E) were recorded daily. Data plotted as mean ± s.d. Statistics on (B) and (C) determined by multiple Mann-Whitney tests. ^∗0.05 ≥ P. (F and G) Spike RBD (F) and non-RBD (G) mutations seen in Vero-grown ferret passage 2 virus (time 0) from Figures 2A–2D and dynamics over time in directly infected ferrets. Hash sign marks samples with insufficient coverage to determine sequence proportions.

Figure 3

Mink- and ferret-associated spike mutations allow more efficient entry into cells expressing the ferret ACE2 receptor (A) Pseudovirus entry in human or ferret ACE2-expressing cells. Mutant SARS-CoV-2 spike-containing pseudovirus entry into HEK 293Ts expressing human or ferret ACE2 or empty vector. Entry normalized to signals from human ACE2 expressing cells. Each data point indicates mean value taken from a completely independent repeat (n ≥ 3). Data plotted as mean ± s.d. Statistics were determined by comparing log-transformed values of ferret ACE2 entry using a one-way ANOVA with multiple comparisons against the WT. ^∗0.05 ≥ p > 0.01; ^∗∗0.01 ≥ p > 0.001; ^∗∗∗0.001 ≥ p > 0.0001; ^∗∗∗∗p ≤ 0.0001. (B and C) Entry of SARS-CoV-2 spike mutant-expressing lentiviral pseudotypes into BHK-21 cells expressing different mammalian ACE2 proteins. Pseudovirus shown contain either D614G (B) or D614 (C). Representative repeat shown from n ≥ 3 repeats. Data plotted as mean ± s.d. (D) Structure of ACE2/Spike RBD interface showing key mink-adaptation residues and nearby residues that differ in mustelid and human ACE2. Figure made using PyMOL (Schrödinger) and PDB: 7A94(Benton et al., 2020).

Figure 4

The common mink and ferret adaptation, Y453F, attenuates virus replication in primary human airway cells (A–C) Human primary airway epithelial cells cultured at air-liquid interface were infected at an MOI of approximately 0.1 with (A) a mixture of parental and ferret-adapted England/2 virus, (B) a mixture of mink-adapted “Cluster 5” virus and a D614G control, or (C) either isogenic WT (D614G) or D614G + Y453F -containing reverse genetics-derived virus isolates. Virus titers were measured by TCID₅₀ (C) E gene qPCR (A, B). Statistics for competition assays were determined by one-way ANOVA with multiple comparisons against time 0. Statistics for the head-to-head growth curve (C) were determined by multiple unpaired t tests on log-transformed data. All infections were performed on triplicate wells from matched donors (n = 3). Data plotted as mean ± s.d. ^∗0.05 ≥ p > 0.01; ^∗∗p ≤ 0.01.

Figure 5

Mink- and ferret-associated mutations have a minimal impact on SARS-CoV-2 antigenicity (A and B) Live virus neutralization comparing WT or Y453F-containg ferret passage 2 (A) or the isogenic reverse genetics-derived WT (D614G) and D614G + Y453F-containing SARS-CoV-2 isolates (B) using N = 6 human convalescent antisera from the first UK wave (∼April-June 2020); (A) or n = 10 double-dose BNT162b2 (Pfizer-BioNTech mRNA vaccine) human antisera (B). Fold differences annotated on graph indicate differences in geometric means of NT₅₀. Statistics were determined by two-tailed Wilcoxon test with matched pairs. ^∗0.05 ≥ P. (C) Pseudovirus neutralization of different mink adaptations containing mutants using N = 8 human convalescent antisera from the first UK wave (∼April-June 2020). Fold differences annotated on graph indicate differences in geometric means of NT₅₀. Statistics determined by matched pair Friedman nonparametric test with multiple comparisons against WT. ^∗0.05 ≥ P.

Figure 6

Several variants of concern show enhanced entry into ferret ACE2 expressing cells Mutant SARS-CoV-2 spike-containing pseudovirus entry into HEK 293Ts expressing human or ferret ACE2, or empty vector. Entry normalized to signals from human ACE2 expressing cells. Each data point indicates data from a completely independent repeat (n ≥ 3). Data plotted as mean ± s.d. Statistics were determined by comparing log-transformed values of ferret ACE2 entry using a one-way ANOVA with multiple comparisons against the WT. ^∗0.05 ≥ p > 0.01; ^∗∗0.01 ≥ p > 0.001; ^∗∗∗0.001 ≥ p > 0.0001; ^∗∗∗∗p ≤ 0.0001. RBD mutational profile of the different spike proteins is shown below. Cells in orange indicate changes from WT/D614G. Alpha also known as B.1.1.7; Beta also known as B.1.351; Gamma also known as P.1; Eta also known as B.1.525; Iota also known as B.1.526 + E484K.