SARS-CoV-2's ability to infect multiple mammalian species is a key characteristic, similar to other coronaviruses. The virus emerged in humans in late 2019, likely after zoonotic spillover from bats, though the exact origin remains unclear. The evolutionary events preceding the initial Wuhan outbreak are poorly understood due to limited genomic data from that period. The virus likely circulated in humans before detection, given the similarity of its presentation to other respiratory infections and the high rate of asymptomatic infections (~40%). A successful host jump requires several traits, including the ability to infect the novel host's cells and efficient transmission within the new population. SARS-CoV-2's ability to infect multiple mammalian species is primarily due to the conservation of the angiotensin-converting enzyme 2 (ACE2) receptor across mammals. Efficient human-to-human transmission is rapid, with early estimates of R0 ranging from 1.5-6.5. Evolutionary analyses suggest that efficient human-to-human transmission and ACE2 usage were present in ancestral bat lineages, indicating pre-adaptation to humans. However, the emergence of more transmissible variants of concern (VoCs) like Alpha, Delta, and Omicron demonstrates ongoing adaptation to its human host. Following the initial zoonotic jump, multiple secondary host jumps from humans to animals have been reported. As of March 17, 2022, 1282 high-quality SARS-CoV-2 genomes from 25 animal species were on GISAID. Outbreaks in mink farms (Netherlands and Denmark) demonstrated rapid transmission, with mink-to-mink transmission and spillback to humans. In the USA, outbreaks in white-tailed deer also showed significant onward transmission. This study aims to determine whether SARS-CoV-2 required host-adaptive mutations to jump into animal hosts, the extent of host-specific adaptation, and the impact of animal introduction on the virus's evolutionary trajectory. The rapid testing and sequencing in these outbreaks provide valuable insights into evolutionary events surrounding spillovers.

The introduction extensively reviews the existing literature on SARS-CoV-2 transmission, host range, and adaptation. It cites numerous studies on cross-species transmission, the role of ACE2 receptor, the reproductive number (R0) of SARS-CoV-2 and its variants, the evolutionary history of the virus, and outbreaks in specific animal populations like mink and deer. The literature review highlights the pre-existing knowledge about SARS-CoV-2's potential for host switching and its ongoing adaptation to humans, establishing a framework for the current study's focus on animal adaptation.

The study utilized published and publicly available SARS-CoV-2 sequences from mink and deer, comparing them to carefully curated subsamples of human sequences. The authors focused on identifying changes in mutational biases, genomic composition, and mutation rates in animal-associated clusters. They screened for mutations potentially arising due to host-specific adaptation, assessed the bioinformatic impact of these mutations, and built phylogenetic trees using IQ-Tree with a GTR+I substitution model. To identify candidate mutations for host adaptation, they compared mink and deer sequences to human isolates with matching PANGO lineage, sampling dates, and country of origin. Criteria for candidate mutations included: a two-fold higher allele frequency in animals than humans; animal allele frequency >0.1; independent emergence at least three times in each animal phylogeny; not inherited from the parent human lineage; and presence in at least three independent clusters. The authors also conducted analyses of nucleotide-nucleotide transitions and dinucleotide frequencies to investigate changes in the genomic landscape, performed tip-calibration of animal-human phylogenies to estimate substitution rates, used PROVEAN scores to assess the impact of non-synonymous mutations on protein function, and conducted structural analyses using HADDOCK and mCSM-PPI2 to predict changes in the stability of Spike:ACE2 complexes.

Phylogenetic analysis revealed multiple independent spillover events of SARS-CoV-2 into mink and deer. The study identified five candidate mutations for mink-specific adaptation (NSP9_G37E, Spike_F486L, Spike_N501T, Spike_Y453F, ORF3a_L219V) and one for deer (NSP3a_L1035F). These mutations, however, appeared to confer minimal advantage for human-to-human transmission. No substantial changes to the mutation rate or evolutionary trajectory resulted from circulation in mink and deer. Analysis of nucleotide-nucleotide transitions showed differences between mink and deer compared to humans, but overall mutational profiles were similar, dominated by CU transitions. Principal components analysis of dinucleotide frequencies indicated minimal differences in genome composition across hosts. Tip-calibration of phylogenies estimated substitution rates consistent with previous estimates, suggesting no significant host-specific rate variation except possibly in Latvian minks. PROVEAN analysis predicted most candidate mutations to be functionally neutral, even the Spike mutations. Structural analysis of Spike:ACE2 complexes predicted marginal changes in stability due to the candidate mutations. The emergence of some candidate animal-adaptive mutations predated documented human-to-animal spillovers, suggesting these mutations weren't strictly required for the host jumps.

The findings suggest that efficient animal-to-animal transmission of SARS-CoV-2 in mink and deer required minimal adaptation, supporting the virus's 'generalist' nature as a mammalian pathogen. The absence of fixed mutations in animal populations suggests host-specific adaptation wasn't necessary for spillover. The study highlights the importance of considering mutations in non-Spike proteins, as some strong candidate deer adaptive mutations were found in ORF1ab and ORF3a. The conflicting results from in vitro and bioinformatic analyses underscore the complex relationship between mutations and viral fitness. The study emphasizes white-tailed deer as important animal models for understanding SARS-CoV-2 transmission, given their high prevalence and wide distribution. The authors discuss the challenges of controlling SARS-CoV-2 in animal reservoirs and the risk of spillback to humans, emphasizing the need for continued surveillance.

This study demonstrates that SARS-CoV-2's transmission to novel hosts like mink and deer requires minimal adaptation, highlighting its generalist nature. While some animal-adaptive mutations were identified, these did not significantly impact human transmission. The lack of significant changes in mutation rates or genomic landscapes in animal populations further supports this conclusion. The research underscores the need for ongoing surveillance of SARS-CoV-2 in both human and animal populations to assess its adaptive potential and mitigate future risks.

The study acknowledges limitations in disentangling complex transmission patterns solely based on sequence data and the possibility of undetected early outbreaks. The method for identifying putatively adaptive alleles relies on comparisons to human backgrounds, which could potentially miss certain mutations. The use of PROVEAN scores may not perfectly capture the functional impact of mutations on viral proteins, and the bioinformatic analyses do not fully elucidate the complex relationship between mutations and viral fitness. Further in vitro and in vivo studies are needed for better understanding.