The search for a COVID-19 animal model

See allHide authors and affiliations

Science  29 May 2020:
Vol. 368, Issue 6494, pp. 942-943
DOI: 10.1126/science.abc6141

As the pandemic caused by severe acute respiratory syndrome–coronavirus 2 (SARS-CoV-2) continues to cause worldwide upheaval, scientists are racing to find appropriate animal models to study the coronavirus disease 2019 (COVID-19) attributed to the virus. The optimal animal model will depend on the scientific question. On page 1016 of this issue, Shi et al. (1) describe severe viral burden and airborne transmission of SARS-CoV-2 between cats and ferrets, highlighting an important animal model for SARS-CoV-2 transmission. Additionally, on page 1012 of this issue, Rockx et al. (2) found that young and aged cynomolgus macaques infected with SARS-CoV-2 shed virus in the upper and lower respiratory tract, but failed to develop severe clinical symptoms. These animal models offer distinct platforms to ask specific questions about SARS-CoV-2 infection, induction of disease, and transmission.

Humans with COVID-19 display a wide range of disease symptoms, from asymptomatic to severe pneumonia (3). Translating data from a single animal model to the varied disease outcomes in humans is not only challenging, but potentially misleading. Studies examining the efficacy of vaccines and antiviral drugs traditionally use models of severe disease, which may not mimic the common pathology in the majority of COVID-19 patients and could limit understanding of other important questions, including infection dynamics and transmission. Previous work on other emerging coronaviruses, such as Middle East respiratory syndrome–coronavirus (MERS-CoV) and SARS-CoV-1, have included mice, hamsters, ferrets, monkeys, and camels as animal models (4). Although mice are the preferred research animal model because of their cost, reproduction rate, and wealth of reagents available for studying this species, early reports indicate that they are unsuitable for SARS-CoV-2 infection, likely due to receptor incompatibility (5). Therefore, transgenic mice that express the human angiotensin-converting enzyme 2 (hACE2), which is the host cell receptor for SARS-CoV-2 entry, will be useful to examine COVID-19 (57). Preliminary work with hACE2 mice demonstrates susceptibility but limited disease severity (8) (see the table).

Shi et al. administered a large dose of SARS-CoV-2 intranasally to a wide range of animals—ferrets, cats, dogs, pigs, chickens, and ducks—to test replication, pathogenesis, and transmission. The virus replicated efficiently in the upper respiratory tract of cats and ferrets, suggesting that these animals are permissive to SARS-CoV-2. However, no severe clinical symptoms such as weight loss or respiratory distress were noted in ferrets or cats. In the other models, virus was not detected in nasal or rectal swabs, with the exception of two dogs in which SARS-CoV-2 RNA was detected in rectal swabs and the animals produced antibodies against the virus. These data suggest that pigs, chickens, and ducks are not permissive to SARS-CoV-2 infection.

Searching for the best animal model to study COVID-19

Comparison of currently available animal models for SARS-CoV-2 infection and COVID-19.

Rockx et al. found that cynomolgus macaques infected with SARS-CoV-2 using a combined intranasal and intratracheal administration shed virus in the upper and lower respiratory tract, but clinical symptoms were mild. Similarly, Gao et al. (9) tested the efficacy of an inactivated vaccine in rhesus macaques. In this study, vaccination with inactivated SARS-CoV-2 produced antibodies against the viral spike protein and nucleoprotein in mice, rats, and macaques. Macaques vaccinated with 3 or 6 µg of inactivated virus in alum were challenged 22 days later and demonstrated reduced viral RNA in nasal and anal swabs.

Together, these studies, along with preliminary studies in hamsters (10), suggest that there are only a handful of susceptible SARS-CoV-2 animal models (hamsters, ferrets, cats, and nonhuman primates). Susceptibility is likely driven by either binding affinity to the host ACE2 receptor or by differences in host protease activity on the spike protein, both key factors in coronavirus cell entry (11). In addition to release of infectious virus in the respiratory tract, viral RNA was found in rectal swabs of cats, ferrets, and macaques (1, 2), which is consistent with initial clinical observations of patients with COVID-19 (12) and suggests a potential role for the fecal–oral transmission route.

Epidemiological data have clearly defined that COVID-19 has a higher case fatality rate in individuals over 60 years of age (3). Yet, most in vivo research models use young, healthy animals. Models that recapitulate this aging phenotype would be incredibly useful to examine the underlying biology behind this phenomenon. To address this, Shi et al. examined COVID-19 symptoms in kittens versus cats aged 6 to 9 months, but found the opposite phenotype. COVID-19 was more severe in kittens; one kitten died on day 3 after infection and had a larger viral distribution compared to older cats. This variation in viral dissemination in kittens may be reminiscent of the vast tissue tropism and variation in disease severity exhibited by SARS-CoV-2 in humans (12). By contrast, Rockx et al. tested SARS-CoV-2 in young and aged macaques and did not observe any age-dependent differences. Of note, the virus is not consistently lethal in any of the animals tested thus far, nor does SARS-CoV-2 infection in these animals recapitulate the severe clinical symptoms observed in humans. As these animal models continue to be developed, attention should be paid to the role of age and other health conditions; these factors may be critical parameters that are necessary to fully evaluate human disease.

Transmission of viruses between people, either through contact (direct or indirect) or virus-containing aerosols, is a key determinant of viral disease burden globally. An important aspect of the SARS-CoV-2 pandemic, and previous influenza pandemics, is efficient airborne transmission of the virus. Airborne transmission can include a wide range of aerosol sizes. At close contact ranges, the exposure to large and small aerosols containing viruses is high. Shi et al. examined airborne transmission of SARS-CoV-2 between kittens and aged cats and observed that in both scenarios, the virus could transmit through the air to 33% of naïve recipient cats or kittens. In these studies, animals were separated by perforated barriers that limit physical contact but allow for air to be shared between the experimentally infected donor and the susceptible recipient. Studies of influenza virus transmission have indicated that viral replication in the upper respiratory tract, and specifically the soft palate, play an important role in airborne transmission (13). Shi et al. found that SARS-CoV-2 replicated in the soft palate of cats, kittens, and ferrets. Although ferret transmission was not examined in this study, a report suggested a similar airborne transmission rate of 30% for SARS-CoV-2 in ferrets (14). No contact transmission between dogs and other animals (pigs, ducks, and chickens) was observed (1).

The transmission of SARS-CoV-2 between cats highlights the susceptibility of this animal model to infection. Consistent with this observation, transmission of SARS-CoV-2 from humans to tigers was recently documented, as was virus spread among big cat units in the Bronx Zoo (15). On the basis of data from Shi et al., infected cats appear asymptomatic, so infections in cats may go undetected. Additional studies are needed into the seroprevalence of SARS-CoV-2–specific antibodies in cats and identification of coronaviruses from this animal source to ascertain the potential for cats to be an intermediate host for SARS-CoV-2.

As the pursuit of SARS-CoV-2 vaccines and antivirals surges on, animal models play the most important role to determine the effectiveness of potential therapeutic strategies. The available studies suggest that hamsters, ferrets, and cats may serve as attractive alternatives to nonhuman primate and transgenic mouse studies. Because hamsters and transgenic mice display the most severe clinical symptoms, such as weight loss, they may provide robust small-animal models for studying efficacy of various vaccine platforms. By contrast, cats and ferrets may provide a useful model system for studying transmissibility of the virus and the effectiveness of antivirals to limit spread. With robust reduction in viral load as presented by Gao et al. (9), nonhuman primates may offer the most relevant model to assess vaccine and antiviral effectiveness before rapid deployment to humans. Therefore, continued evaluation of mice to nonhuman primate models will provide critical data on the animals best suited to study the many open questions about COVID-19.

References and Notes

Acknowledgments: The authors are funded by NIH CEIRS HHSN272201400007C (S.S.L.) and R00 AG049092 (V.D.M.).
View Abstract

Stay Connected to Science

Navigate This Article