As we enter the third year of the COVID-19 pandemic, there is a growing consensus that SARS-CoV-2 will eventually become an endemic infection, regularly circulating at relatively static or predictable levels. However, we remain unsure when the pandemic phase will end and what long-term coexistence with SARS-CoV-2 will entail. One key question – given the likely zoonotic origin of SARS-CoV-2 – is whether the virus can establish animal reservoirs, as pathogens with robust animal reservoirs are more challenging to control or eradicate. CAMB MVP student Andrew Marques recently contributed to a growing body of literature regarding potential SARS-CoV-2 animal reservoirs by publishing a study examining the prevalence of SARS-CoV-2 in wild white-tailed deer throughout Pennsylvania (1).
During a roughly three-month sampling period in late 2021, nasal swabs were collected from 93 white-tailed deer in 31 counties throughout Pennsylvania and tested for SARS-CoV-2 infection via PCR. Of the sampled population, 19.3% were found to be COVID positive. During the same sampling period, the test positivity rate for the whole United States ranged from an average of 4 – 5% to a high of 13.2% as case counts rose due to the ascendance of the omicron variant and the holiday travel season (2).
This is a surprisingly high rate of infection when compared to the human disease burden at the same time. Previous studies of wild deer populations further corroborate the animals bearing higher SARS-CoV-2 burden. Studies performed in Iowa and Ohio showed test positivity rates of 33% and 36%, respectively (3, 4). During each sampling period, the rates of infection in deer are substantially larger than those seen in people, suggesting that high rates of SARS-CoV-2 positivity in deer may be the norm.
Andrew and colleagues next examined the viral isolates obtained from the sampled deer; high quality viral genome sequences were successfully recovered from the seven deer samples with the highest concentrations of SARS-CoV-2 viral RNA. Using whole genome sequencing, two of the sequenced samples were shown to be derived from the Alpha variant of SARS-CoV-2 (B.1.1.7), and five were shown to be closely related to the Delta variant (AY.#).
The Alpha variant, first documented in the UK in September 2020 and designated a variant of concern by the WHO in December 2020, outcompeted more ancestral strains in some parts of the world. Later, the prevalence of the Alpha variant waned substantially as the more contagious Delta variant, which was first documented in India in October 2020 and designated a variant of concern in May 2021, began to comprise the overwhelming majority of observed cases (5).
The two Alpha sequences sampled from deer were shown to be significantly divergent both from each other and from nearly all Alpha variant strains collected from infected human subjects. The two isolates showed clearly distinct phylogeny and differed from each other by 45 nucleotide substitutions. For context, sequencing studies of SARS-CoV-2 infection in humans have shown that a typical virus accumulates roughly two single nucleotide mutations per month (6). Andrew and colleagues interpreted this substantial difference between recovered Alpha genomes as evidence that two separate human to deer transmission events gave rise to these cases, rather than a single jump to deer followed by rapid mutation of the virus in its new host.
The five recovered Delta sequences were annotated into three distinct clades within a broader Delta phylogenetic tree. Two of the genomes were assigned to one clade, AY.103, and two were assigned to a second, AY.88, and the final sequence was assigned to a third clade, AY.5. The two AY.88 sequences were identical, and the AY.5 sequence differed by 13 nucleotide substitutions, while the two AY.103 sequences differed by just 7 substitutions. Again, the most plausible interpretation of these data is at least two separate transmission events; one that introduced the AY.103 viruses into the deer population, and another that introduced the AY.88 viruses, followed by some slight divergence as mutations accumulated in the viral genome during replication in the deer.
Andrew and colleagues then moved on to further define the individual mutations that they observed in their recovered viral genomes. Interestingly, many of the substitutions observed in the deer viruses were rare or absent in databases of human clinical SARS-CoV-2 isolates but were highly enriched in a database of deer viral genome isolates. Multiple mutations observed in the 7 recovered deer genomes were found to be hundreds of times more prevalent in other deer isolates than in human-derived viral genomes. This is not surprising for viruses with broad enough tropism to infect multiple hosts, and suggests that SARS-CoV-2 in deer is subjected to considerable selective pressure based on species-specific restriction factors or other constraints.
These findings, currently available as a preprint on MedRXiv awaiting peer review, have serious implications for the future of the pandemic and our understanding of SARS-CoV-2 endemicity. First of all, widespread infection in wild white-tailed deer is a highly reproducible phenomenon in North America (1, 3, 4, 7). Given the high density of wild deer on many parts of the continent – Pennsylvania has roughly 30 per square mile or a statewide total of over 1 million, for example – they could certainly constitute a bonafide animal reservoir for SARS-CoV-2. Time will tell if these high levels of infection are stable. Given previous documentation of SARS-CoV-2 infection in other common mammals like hamsters, some mice, minks, and feral cats, the list of potential animal reservoirs appears to be growing (8). SARS-CoV-2 generally appears to show much broader tropism than other coronaviruses. The SARS coronavirus that caused the 2003 epidemic in east Asia has been shown to infect minks and wild boars, but at lower levels than those observed in deer (9).
Andrew’s data also raises the question of re-transmission from deer back into the human population. Based on another study in Canada, there was at least a single documented case of a re-transmission event in which a human patient was infected with a strain of SARS-CoV-2 that harbored mutations that were previously observed exclusively in viral samples recovered from infected wild deer (6). Again, time will tell if this observation is reproducible, but this is generally concerning news for pandemic mitigation efforts. If SARS-CoV-2 variants – such as the Alpha variants recovered in Andrew’s study – are capable of taking refuge in deer or other animal populations and eventually re-infecting humans, viruses that are no longer considered variants of concern could return to cause further damage in the future as population immunity wanes or changes.
Regarding the potential clinical relevance of transmission from infected deer back into the human population, Andrew states that “The worst-case scenario is that spillovers and spillbacks occur regularly with substantial changes to the viral genome while infections last in other species. This could have the potential to reintroduce new variants into the human population with [novel] mutations ... The best-case scenario is that white-tailed deer are a dead end in the context of human infections. I believe that the truth lies somewhere in the middle: deer get infected and can perhaps infect some humans, but the evolution is limited. Or at least as limited as evolution in humans”.
While the Delta samples were detected in October 2021, after the Delta variant had become the dominant strain throughout the US, the Alpha samples were recovered in November of 2021, long after the prevalence of Alpha lineage viruses in the human population had dropped precipitously in favor of Delta. This suggests that some variants of concern might be capable of long-term persistence in deer or potentially other animal reservoirs. Additionally, a longer period of incubation in the deer host could account for the comparatively large number of mutations that accumulated in the recovered Alpha sequences.
Andrew and colleagues then moved on to further define the individual mutations that they observed in their recovered viral genomes. Interestingly, many of the substitutions observed in the deer viruses were rare or absent in databases of human clinical SARS-CoV-2 isolates but were highly enriched in a database of deer viral genome isolates. Multiple mutations observed in the 7 recovered deer genomes were found to be hundreds of times more prevalent in other deer isolates than in human-derived viral genomes. This is not surprising for viruses with broad enough tropism to infect multiple hosts, and suggests that SARS-CoV-2 in deer is subjected to considerable selective pressure based on species-specific restriction factors or other constraints.
Another open question is how exactly wild deer are becoming infected at such high frequencies. Contact between deer and covid positive people could play a role, but it is also possible that another SARS-CoV-2 susceptible animal is acting as an intermediate host and ferrying the virus from human to deer. Recent in silico modeling studies have suggested that as many as several hundred mammal species may ultimately be susceptible to SARS-CoV-2, based on simulations of the viral spike protein’s ability to bind its entry receptor ACE2 in various animals (10). Interestingly, both wildlife observations and lab studies show that deer do not experience symptoms due to SARS-CoV-2 infection (7). However, this may not be the case for other animals that cross paths with deer. In particular, grazing livestock may be vulnerable to infection and disease, or could act as an intermediary in transmission of viruses back into the human population.
The high prevalence of SARS-CoV-2 in wild deer poses new challenges for pandemic control, and further complicates models of endemicity. Historically speaking, diseases with well-established animal reservoirs can be functionally impossible to eradicate and challenging to control. Going forward, people may need to exercise more caution around deer and other SARS-CoV-2 susceptible wildlife, including taking measures to keep pets and livestock protected. Indeed, these data suggest that broader screening for SARS-CoV-2 infection among potentially susceptible wildlife will be essential for understanding which species might be acting as viral reservoirs. Broadly, these findings are part of the growing body of evidence that suggests that we will all be dealing with SARS-CoV-2 for the foreseeable future and should continue to be vigilant and proactive with mitigation measures.
References
https://www.medrxiv.org/content/10.1101/2022.02.17.222 70679v2.full
https://coronavirus.jhu.edu/testing/individual-states
https://www.biorxiv.org/content/10.1101/2021.10.31.466 677v1
https://www.npr.org/sections/goatsandsoda/2021/11/10/1054224204/how-sars-cov-2-in-american-deer-could-alter-the-course-of-the-global-pandemic
https://www.who.int/en/activities/tracking-SARS-CoV-2- variants/
https://www.nature.com/articles/d41586-020-02544-6
https://www.biorxiv.org/content/10.1101/2022.02.22.4815 51v1.full
https://www.npr.org/sections/goatsandsoda/2022/03/09/1084440012/researcher-finds-stunning-rate-of-covid-among-deer-heres-what-it-means-for-human?live=1
https://www.sciencedirect.com/science/article/pii/ S0168170207001050?via%3Dihub
https://royalsocietypublishing.org/doi/pdf/10.1098/ rspb.2021.1651