Of Mice and Men

By: Milo Molleson

From house mouse to laboratory mouse

The relationship between humans and mice has a long and captivating history. The house mouse, scientifically known as Mus musculus domesticus, has shared our narrative since its beginning; eating our grain and scurrying around our floors at night since before we left our nomadic lifestyles for settled farming^1,2. In much more recent history there has been a pivotal shift in our relationship with these little creatures, starting in 1664 with the experiments of Robert Hooke on the effects of changing air pressure on an organism, mice became not only our vermin housemates but valuable tools in medical experiments². Already accustomed to humans, these domesticated rodents were easy to handle and feed. Additionally, the house mouse is fast at breeding all year round and relatively robust for extensive inbreeding, allowing for the creation of inbred mouse lines with specifically desired genotypes and phenotypes^1–3. However, the decision to take mice from our pantry floors onto our laboratory benches was driven primarily by convenience rather than a deliberate assessment of their biological resemblance to humans². While we owe so much to our whiskered subjects in scientific research, there is a growing movement towards critical evaluation of the validity of mice as models of various human diseases and the implications this has on the interpretation of scientific results⁴.

Neuroinflammation in humans and mice

The utilization of laboratory mice in medical research has undoubtedly allowed us to shine a light onto previously enigmatic aspects of biology, including response to infections, successful tissue transplantation, and X-chromosome-linked genetic conditions^2–4. With transgenesis (the transfer of an isolated gene into an organism or cells) and genetic editing (such as CRISPR/Cas9 editing), researchers are able to accurately manipulate or introduce a selected gene, leading to discoveries that have advanced our knowledge of diseases like Huntington’s Disease^2,3. However, the divergence between mice and humans becomes an issue when investigating complex diseases that mice are naturally resistant to^4,5. For example, mice are particularly resistant to inflammation, hindering the creation of adequate mouse models for human inflammatory disease⁴. This has been validated by poor correlations between human and mouse model responses when treated with endotoxins, which should induce an inflammatory response^4,6. This is perhaps an especially pertinent issue in a scientific research that is increasingly identifying inflammation as a common feature for many human diseases, ranging from depression to Covid-19⁷.

Developing appropriate mouse models

The limitations of mouse models for human diseases, however, extends beyond neuroinflammation, encompassing a wide array of human diseases. Instead of a single comprehensive mouse model, a plethora of different mouse models has been created, each capturing different aspects of complex diseases to varying degrees⁵. For example, there are almost 200 rodent models of Alzheimer’s Disease (AD), yet none that successfully capture both the behavioral phenotype and underlying biological mechanisms of the disease⁸. The poor translational utility of this deluge of mouse models impedes progress towards a much-needed successful treatment for AD^8,9. While researchers aspire to develop adequate mouse models, the fundamental differences between mice and humans pose a substantial challenge^3,5,10.

Investigating psychological disorders

Further still, mice are often used to investigate human psychological disorders, such as depression and anxiety, with behavioral paradigms designed to induce psychiatric symptoms in mice^11–13. However, such models are known to be especially susceptible to poor translational validity to human psychological conditions, again hindering the development of novel treatments^11,12. For example, the Forced Swim Test, commonly employed as a model of depression and for screening anti-depressant drugs, forces the mouse to swim in a glass cylinder of water until it ceases to struggle and starts floating. A shorter duration of struggle-time in the water is interpreted as depression-like symptoms¹¹. There is controversy regarding the face, construct and translational validity of this paradigm, mainly because the test measures the acute effects of the antidepressant administered, whereas antidepressant drugs are known to take weeks to take effect in humans^13,14. Fortunately, such pitfalls in the design of behavioral models of psychological disorders are increasingly acknowledged, leading to substantial efforts to identify and standardize the most valid models^13,15. Nonetheless, it is important to consider the inherent challenges to mouse models of depression due to the fundamental differences between mice and humans^11,15.

“Raising the bar”

Evaluating the use of mouse models is not only a discussion of practicality and scientific rigor but also of morality^15,16.  In recent decades, there has (yet again) been a shift in our relationship with mice, as we begin to consider not only the utility of our whiskered companions, but also their suffering (The Lancet, 1985). This shift in moral regard for mice, and animals more generally, is reflected in our legal guidelines for ethical approval and efforts to minimize the number of animal lives used in experiments¹⁸. To this end, the Vrije Universiteit Amsterdam (VU) has been proactive in staying at the forefront of this movement. Upon interview with Dr. Oliver Stiedl, Chairman of the Animal Welfare Body of VU and VU Medical Center, it was clear that not only should there be a “blacklist of tests that are not really useful because their replicability value and translational validity are low”, but also an exploration of “alternative methods which avoid unnecessary stressors, allowing animals to have free choice and all the time they need”. For example, home cage-based behavioral phenotyping, where “the experiment comes to the animal”, leads to improved animal welfare and scientific validity. The sentiments of Dr. Stiedl are consistent with those of Prof. H. Shaw Warren of Harvard Medical School in his TedMed talk (2013), where he suggests that “rather than stopping work with mouse models, one way forward might be for the scientific community to raise the bar” by recognizing the species differences between mice and humans rather than assuming a uniform response.

And beyond mouse models

In light of the growing critical evaluation of mouse models and the advent of new technological possibilities, the field of science is also opening up to promising solutions beyond mouse models^15,18. Among these alternatives, human-derived cell cultures, such as induced Pluripotent Stem Cells (iPSCs) or “human organs on a chip”, present exciting new avenues^18–20. Such technologies, being human-derived, are promising for overcoming the limitations posed by genetic and biological differences between mice and humans¹⁹. Alternatively, the rise of in silico computer models offers another compelling perspective on human pathology research and drug development. Already, computer models and machine learning paradigms are being utilized in a wide variety of medical research contexts, from accurately predicting cardiotoxic hERG potassium channel blockade to the simulation of magnetoencephalography (MEG) data based on input network parameters^21–23. Although iPSCs and computer models are very unlikely to entirely replace in vivo models, they can complement and build upon mouse model findings²⁴, aiding in the diversification of scientific approaches to overcome the above described challenges.

In our long journey alongside mice, we have come to owe a lot to these animals for the crucial role played in advancing our understanding of human diseases and disorders. However, their limitations in translational validity and ethical considerations have prompted critical reflection within the scientific community and an exploration of alternative approaches. Perhaps it is high time we settle our differences with mice, by honing our methodological rigor and looking beyond mouse models to better advance our understanding and treatment of human diseases.

About the author

Further reading

1. Bittel, J. Humans ‘Domesticated’ Mice 15,000 Years Ago. National Geographic (2017).
2. The laboratory mouse. (Elsevier Academic Press, 2004).
3. Masopust, D., Sivula, C. P. & Jameson, S. C. Of Mice, Dirty Mice, and Men: Using Mice To Understand Human Immunology. J. Immunol. 199, 383–388 (2017).
4. Shaw Warren, H. Why do we use mice to study human diseases? (2013).
5. Yokoyama, M., Kobayashi, H., Tatsumi, L. & Tomita, T. Mouse Models of Alzheimer’s Disease. Front. Mol. Neurosci. 15, 912995 (2022).
6. Seok, J. et al. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc. Natl. Acad. Sci. 110, 3507–3512 (2013).
7. Sagris, M. et al. Inflammatory Mechanisms in COVID-19 and Atherosclerosis: Current Pharmaceutical Perspectives. Int. J. Mol. Sci. 22, 6607 (2021).
8. Götz, J., Bodea, L.-G. & Goedert, M. Rodent models for Alzheimer disease. Nat. Rev. Neurosci. 19, 583–598 (2018).
9. Myers, A. & McGonigle, P. Overview of Transgenic Mouse Models for Alzheimer’s Disease. Curr. Protoc. Neurosci. 89, (2019).
10. Lux, A. & Nimmerjahn, F. Of Mice and Men: The Need for Humanized Mouse Models to Study Human IgG Activity in Vivo. J. Clin. Immunol. 33, 4–8 (2013).
11. Commons, K. G., Cholanians, A. B., Babb, J. A. & Ehlinger, D. G. The Rodent Forced Swim Test Measures Stress-Coping Strategy, Not Depression-like Behavior. ACS Chem. Neurosci. 8, 955–960 (2017).
12. Cryan, J. F. & Holmes, A. The ascent of mouse: advances in modelling human depression and anxiety. Nat. Rev. Drug Discov. 4, 775–790 (2005).
13. Dzirasa, K. & Covington, H. E. Increasing the validity of experimental models for depression: Mouse affective syndrome. Ann. N. Y. Acad. Sci. 1265, 36–45 (2012).
14. Yankelevitch-Yahav, R., Franko, M., Huly, A. & Doron, R. The Forced Swim Test as a Model of Depressive-like Behavior. J. Vis. Exp. 52587 (2015) doi:10.3791/52587.
15. Kafkafi, N. et al. Reproducibility and replicability of rodent phenotyping in preclinical studies. Neurosci. Biobehav. Rev. 87, 218–232 (2018).
16. Langley, G., Evans, T., Holgate, S. T. & Jones, A. Replacing animal experiments: choices, chances and challenges. BioEssays 29, 918–926 (2007).
17. Reduction of the use of animals in the development and control of biological products. Lancet Lond. Engl. 2, 900–902 (1985).
18. Mak, I. W., Evaniew, N. & Ghert, M. Lost in translation: animal models and clinical trials in cancer treatment. Am. J. Transl. Res. 6, 114–118 (2014).
19. Deng, B. Mouse models and induced pluripotent stem cells in researching psychiatric disorders. Stem Cell Investig. 4, 62–62 (2017).
20. Horánszky, A., Becker, J. L., Zana, M., Ferguson-Smith, A. C. & Dinnyés, A. Epigenetic Mechanisms of ART-Related Imprinting Disorders: Lessons From iPSC and Mouse Models. Genes 12, 1704 (2021).
21. Aronov, A. Predictive in silico modeling for hERG channel blockers. Drug Discov. Today 10, 149–155 (2005).
22. de Haan, W., Mott, K., van Straaten, E. C. W., Scheltens, P. & Stam, C. J. Activity Dependent Degeneration Explains Hub Vulnerability in Alzheimer’s Disease. PLoS Comput. Biol. 8, e1002582 (2012).
23. Tulloch, J., Zamani, R. & Akrami, M. Machine Learning in the Prevention, Diagnosis and Management of Diabetic Foot Ulcers: A Systematic Review. IEEE Access 8, 198977–199000 (2020).24. Patton, E. E. et al. Melanoma models for the next generation of therapies. Cancer Cell 39, 610–631 (2021).