Since the initial COVID-19 outbreak in Wuhan, China, SARS-CoV-2 has primarily spread among humans, with occasional human-animal transmission events. A wide range of animals have shown susceptibility, with natural infections documented in carnivores like dogs, cats, lions, tigers, otters, and ferrets, often through contact with infected humans. While most common livestock species haven't shown infection, several countries reported SARS-CoV-2 in farmed minks to the World Organisation for Animal Health (OIE). In the Netherlands, SARS-CoV-2 was first detected in farmed minks in late April 2020, showing respiratory symptoms and increased mortality. This prompted an in-depth One Health investigation combining whole-genome sequencing (WGS) with epidemiological data. Initial findings (April-June) revealed five distinct clusters of mink sequences, suggesting independent introductions from humans. Critically, farm workers were infected with mink strains, demonstrating animal-to-human transmission. Three of these clusters continued spreading, resulting in 68 out of 126 mink farms infected between April and November 2020. Subsequently, fur farming was banned in the Netherlands. This study delves deeper, utilizing Bayesian phylodynamic methods to understand SARS-CoV-2 transmission dynamics between farms and between minks and humans, aiming to identify transmission patterns and contributing factors.

Prior research on SARS-CoV-2 transmission has established its primary mode of spread as human-to-human, with documented instances of zoonotic spillover events into various animal species. Studies on mink farms in other countries highlighted the potential for SARS-CoV-2 adaptation and efficient transmission within these populations. Existing literature also emphasizes the importance of One Health approaches for investigating outbreaks involving animal-human interfaces, integrating epidemiological and genomic data for comprehensive analyses. Previous studies on the spread of other pathogens in confined animal populations like mink farms offered insights into the potential roles of factors such as animal movement, human contact, and environmental factors in transmission dynamics. The limited available research on SARS-CoV-2 transmission in farmed mink highlighted the need for deeper analysis, particularly in terms of the evolution of the virus within this specific host and the subsequent potential for spillover events.

This study analyzed SARS-CoV-2 infections in 68 mink farms in the Netherlands between April 24th and November 4th, 2020. Data included whole-genome sequences from 295 minks from 64 farms and 57 humans from 27 farms. Control measures, including culling, were implemented. The spatial distribution of infected farms was mapped. Phylogenetic analyses employed Bayesian methods in BEAST, including relaxed clock models and Skygrid for effective population size (Ne) and birth-death skyline (BDSKY) model for reproductive number (Rt). The analyses incorporated discrete trait models to infer transmission patterns between farms and between minks and humans, using Markov jumps and Bayesian stochastic search variable selection (BSSVS) for statistical significance. Generalized linear models (GLMs) in BEAST explored predictors of between-farm transmission, including distance, personnel links, feed suppliers, veterinary service providers, farm size, human population density, and time between sampling and culling. The analysis included a comparison of phylodynamics among three major clusters to identify differences in transmission dynamics and potential adaptations to the mink host. The data from the study was compared with existing data on SARS-CoV-2 lineages in the Netherlands, obtained from GISAID. Specific amino acid changes in the spike protein were analyzed to map their prevalence and potential links to transmission efficiency and adaptation to the mink host. The study also examined possible spillover events into the local human community. Throat swabs from escaped minks were included in the genomic sequencing and phylogenetic analysis to assess the potential role of escaped minks in farm-to-farm transmission.

The study identified five distinct SARS-CoV-2 clusters (A-E) in Dutch mink farms, originating from four lineages circulating in the human population. Cluster A was the largest, infecting 40 farms and persisting the longest. Three clusters (A, C, and D) showed sustained spread between farms after initial detection. The study found at least 43 zoonotic transmissions from minks to humans across multiple farms. The largest cluster (Cluster A) showed higher evolutionary rate and faster spatial spread than the other clusters. The analysis found that personnel links between farms significantly influenced transmission, especially in Cluster A, where distance between farms also played a significant negative role in transmission. Notably, the F486L substitution in the spike protein was prevalent in Cluster A and C, associated with longer and wider spread. Other amino acid substitutions (Y453F, L452M, and N501T) were also detected in multiple clusters, with varying prevalence and temporal patterns. Spillover events to the community were limited, with only three cases detected. The estimated evolutionary rate of SARS-CoV-2 in minks (7.9 × 10⁻⁴ subst/site/year) was similar to the early-pandemic rate in humans, with Cluster A exhibiting a faster rate, akin to rates seen in other studies of human SARS-CoV-2.

The findings demonstrate the adaptation of SARS-CoV-2 to farmed minks, with the emergence of specific spike protein mutations potentially enhancing transmissibility within mink populations. The limited community spillover suggests effective control measures, though the potential for future spillover events, particularly with novel variants, remains a concern. Personnel links were identified as a significant driver of transmission between farms, underscoring the importance of strict biosecurity protocols. The study's findings highlight the need for continuous surveillance and preventative measures within fur farming and other susceptible animal populations. While the ban on mink farming reduced this specific risk in the Netherlands, potential future outbreaks in other regions necessitate vigilance. The study also notes that the evolutionary rates observed are similar to those seen in human populations, further emphasizing the importance of surveillance and the potential for continued evolution and adaptation of the virus in mink and other animal reservoirs.

This study provides a detailed analysis of SARS-CoV-2 transmission dynamics within and from Dutch mink farms. The adaptation of the virus within the mink population, evidenced by the emergence of specific spike protein mutations, and the significant role of human movement in between-farm transmission, are key findings. While community spread remained limited, the potential for future spillover events necessitates ongoing surveillance and stringent biosecurity measures in mink farms and other at-risk animal populations. Further research is needed to fully understand the implications of these specific mutations on viral fitness and transmissibility.

The study acknowledges that the number of successfully sequenced human samples was lower than the total number of known human infections, which may have resulted in an underestimation of some transmission links. The study also notes that the impact of humans on transmission between farms may be underestimated given the challenges in identifying, locating, and sampling unregistered or moving workers. Additionally, the analysis relies on existing control measures, such as culling, and their effectiveness could have influenced the results. Further, the study acknowledges that some farms may have had more intensive sampling that others, which might have biased the results.