Mice Exposed to Microbes ('Dirty' Mice) More Accurately Predict the Real-World Immune Response to mRNA Vaccines
Background
mRNA vaccines have demonstrated their potential as a platform technology by dramatically reducing hospitalization and mortality rates during the COVID-19 pandemic. However, clinical trials have revealed the need for booster doses, and a decline in antibody levels over time has been repeatedly observed. The challenge lies in the fact that these immune response dynamics are not adequately predicted in preclinical studies.
Current preclinical vaccine trials rely on specific-pathogen-free (SPF) mice. SPF mice are raised in nearly sterile environments, which enhances experimental reproducibility. However, this contrasts sharply with the human immune system, which is constantly exposed to a diverse range of pathogens throughout life. As vaccine candidates that showed strong efficacy in SPF mice have sometimes exhibited lower-than-expected efficacy in actual clinical trials, there is a growing need for animal models that more accurately reflect human immune responses.
Key Findings
A joint research team from the University of Minnesota and Emory University (Beatriz Praena, Frances K Shepherd, and Cera A McDonald, co-first authors) evaluated the response to SARS-CoV-2 mRNA vaccines using a 'dirty' mouse model, in which laboratory mice were co-housed with pet store mice to induce broad microbial exposure. The results were published online in mBio on July 6, 2026.
After the primary vaccination, the serum spike-binding antibody titers in dirty mice were significantly lower than those in SPF mice. A second booster dose was required to reach SPF-level antibody titers, which is consistent with the pattern observed in humans. After tracking the mice for five months after the primary vaccination, the antibody titers in dirty mice decreased more rapidly than those in SPF mice. The neutralizing activity against the Omicron variant was also more significantly reduced in dirty mice, which is also consistent with the trend reported in vaccinated human populations.
The research team also verified the experimental reproducibility of this model. The pattern of pathogen exposure and the level of T cell activation in the co-housed dirty mice remained consistent over time, and there was little seasonal variation. A single co-housing event was sufficient to achieve a sufficient level of broad microbial exposure, allowing for the establishment of a stable experimental system without repeated additional exposures.
Significance and Prospects
This study experimentally demonstrates a key limitation of SPF mouse-based preclinical models: the potential for overestimation of vaccine efficacy. The dirty mouse model was able to reproduce key characteristics observed in human clinical trials, such as the need for booster doses, antibody decay, and reduced protection against variants.
However, further research is needed to determine to what extent the microbial exposure profile of dirty mice represents the actual infection history of humans. The pathogen profile of pet store mice used for co-housing may vary depending on the region or source, so model standardization remains a challenge. Nevertheless, the fact that a reproducible immune background can be established with a single co-housing event is a strength that enhances the practicality of this model.
This model can most directly contribute to the preclinical screening of mRNA vaccine candidates. It can reduce the number of cases in which candidates that show high antibody titers in existing SPF mice fail to meet expectations in clinical trials. It can be used to more realistically validate the optimization of booster vaccination strategies, the prediction of antibody duration, and the evaluation of the range of protection against variants before human clinical trials.
If this co-housing model can be applied to the development of mRNA vaccines for other infectious diseases, such as influenza and RSV, in addition to COVID-19, it has the potential to become a universal platform for systematically reducing the gap between preclinical and clinical studies. Reducing vaccine development failure rates leads to reduced development costs and timelines, which in turn affects the overall pandemic preparedness system.
Although specific-pathogen-free (SPF) mice have traditionally been used to test candidate vaccines, recent work has demonstrated that "dirty" mice with broad microbial exposure more appropriately recapitulate human immune responses. Using a dirty mouse model in which lab mice are co-housed with pet store mice, we modeled SARS-CoV-2 mRNA vaccine responses in dirty and SPF mice. In this study, dirty mice showed reduced serum spike-binding antibody titers after prime vaccination and required a second booster dose to reach SPF-level spike antibody titers. Additionally, spike antibody titers waned faster in dirty mice in the 5 months following prime vaccination, while the neutralizing activity of these antibodies was reduced against Omicron variants, an effect that has also been observed in vaccinated humans. We further investigated the seasonality and consistency of pathogens in co-housed dirty mice, as well as the impact of serial microbial exposure on our animal model system. We found that pathogen exposure and T cell activation remained consistent over time and that a single co-housing event was sufficient to provide broad microbial exposure. This work demonstrates that the dirty mouse co-housing system is a promising, translationally representative approach to screen candidate mRNA vaccines for efficacy and durability prior to human clinical trials. IMPORTANCE: The development of mRNA vaccines during the COVID-19 pandemic dramatically reduced hospitalization and death rates for infected individuals. However, booster vaccinations were required to achieve high antibody titers, and protection waned over time. Our research leveraged a "dirty" mouse model to test whether SARS-CoV-2 mRNA vaccinations in animals with previous microbial exposure better modeled human immune responses. We found that, unlike standard SPF mice, dirty mice also require a booster vaccination to reach maximum antibody titers and experience waning serum antibody titers over time. We pro
This model can most directly contribute to the preclinical screening of mRNA vaccine candidates. It can reduce the number of cases in which candidates that show high antibody titers in existing SPF mice fail to meet expectations in clinical trials. It can be used to more realistically validate the optimization of booster vaccination strategies, the prediction of antibody duration, and the evaluation of the range of protection against variants before human clinical trials.
Beyond COVID-19, if this co-housing model can be applied to the development of mRNA vaccines for other infectious diseases, such as influenza and RSV, it has the potential to become a universal platform for systematically reducing the gap between preclinical and clinical studies. Reducing vaccine development failure rates leads to reduced development costs and timelines, which in turn affects the overall pandemic preparedness system.