Do the 3Rs need to evolve for modern translational research with mice?

Shows how research mice are intrinsically different than wild. Creating a problem for translatability in medical research.

by Christopher S. Keator, Ph.D.
Assistant Dean, Laboratory Animal Resources
Founding Chair, WMed IACUC
Assistant Professor, Department of Biomedical Sciences
WMU Homer Stryker M.D. School of Medicine
Kalamazoo, Michigan, USA

History and recent controversies surrounding the 3Rs.
Why haven’t the 3Rs evolved to match the progress of modern scientific research? 
The first detailed description of the principles of the 3Rs – replacement, reduction, and refinement – is credited to The Principles of Humane Experimental Technique, published by Russel and Burch in 1959. The 3Rs have remained tenets in the animal care and use research world for the last 60 years – but are the original 3Rs still adequate for the translational research conducted in today’s modern world? Do we need to start considering a 4th R or even a 5th R?  I’m not the first to ask these questions; others before me have been asking similar questions for at least the past decade.  Francis Collins, the past director of the NIH, in many ways, re-energized this discussion when he intimated concerns regarding problems with ‘replication’ and ‘reproducibility’ of experiments, calling into question how we conduct modern science. 

Reproducibility and Scientific Rigor.
Are lab mice causing some of our issues?
The problem of reproducibility is coupled with questions about ‘scientific rigor’, as a scientific community, perhaps we need to start discussing the possibility that the beloved lab mouse may be the cause of some of our issues. I recognize this may seem like a shocking ‘accusation’, but based on numbers alone, the connection to the use of mice for research with the concerns about reproducibility must be scrutinized. 

How many mice do we use for research?
An analytical study published in 2021, which generated quite a discussion and some controversy, suggested more than 100 million mice and rats are used for research in the US every year.  The actual number may be lower, but even 20 years ago an older study (published ~2001) noted that mice were the experimental animal model used in more than 3,000 research manuscripts published annually. A simple search in October 2022 for ‘mouse’ on PubMed reveals more than 1.97 million manuscripts that mention this animal. Regardless of the actual number of lab mice used for research in the US and other countries, there is little doubt the mighty mouse is the most common animal research model used for translational research around the world.

Why does research love mice?
Admittedly, mice can be of great benefit to research. The genome and proteome (nucleic acids and proteins) of mice are comparable with humans, and that’s often the main justification most principal investigators provide for using the little rodents in animal use protocols.  They are mammals, and therefore mice and humans share much of the same basic organ function and systems physiology.  Mice can develop (or be induced) to get many of the same malignant cancers as humans, and many pathophysiologic processes and disorders have been studied in these animals. From a management perspective, mice are easy to breed, relatively easy to handle, and easy to manipulate. Moreover, when you group house four or five animals in a single cage, the mouse is by far the cheapest mammalian animal model used for research.

Disadvantages of mice for translational research.
Should we pause a bit longer when comparing mice with humans?
Mice are cheap and pretty good models, but the disadvantages of using mice for translational research are almost as numerous as the benefits. They are a prey species, and we are the predator species arguably at the top of the evolutionary food web; therefore, there are definite differences between the species.  Mice are quadrupeds and nocturnal, exhibiting most activity 1-2 hours after ‘lights out’.  Mice have a higher thermoneutral set point and require less water than humans, resulting in more concentrated urine.  They are coprophagic (i.e., they eat their own poop), exhibit relatively poor eyesight, and rely mostly on olfactory cues (sense of smell). Reproductively, the animals exhibit a 4-day estrous cycle and are proliferous litter-bearing animals who will abandon their young at the first hint of danger to survive. Finally, older zoological studies with wild mus musculus suggest a male mouse in the wild normally occupies a home range of 5,000 to 25,000 square feet, whereas the females giving birth and tending to young pups utilize approximately 2,500 to 10,000 square feet. Putting this into perspective, a wild mouse typically owns an area the size of most ‘large’ homes, akin to the mouse who co-starred with Nathan Lane and Christopher Walken in the 1997 movie Mouse Hunt

How many transgenic strains of mice can we create until we think there’s enough?
In addition to some of the major natural differences between humans and mice mentioned thus far, modern science has genetically manipulated the animals via elaborate gene knock-in and knock-out strategies to the point that we can almost create and/or order a ‘transgenic mouse on demand’. Coupled with CRISPR technology, the number of current ‘strains’ of lab mice is anyone’s best guess, but JAX labs advertises it has the largest mouse repository in the world boasting more than 13,000 different strains. Collectively, modern science has pushed the lab mouse well beyond the ‘fancy mouse’ bred by Miss Abby Lathrop, whose mice were used in the early 1900’s for animal research studies at Harvard.

Extrinsic factors in research influence mouse physiology and behavior.
Have we over-manipulated their habitat?
It’s not just the different strains that stretches our comparability amongst studies, but it’s also how we control their living environment.  Most research institutions around the world house 4-5 lab mice in a cage that’s ~75 square inches, which is noncomparable with the home ranges of wild mice.   Most facilities carefully manipulate the animal’s circadian rhythm by placing them in a tightly controlled vivarium with defined light/dark cycles, something most wild mice never experience. We regulate their room temperature, we feed them special diets, we give them different types of enrichment, and we change their nesting & bedding materials weekly. We often disrupt their sleep to conduct daily health checks and monitor their exposure to pathogens (via a variety of methods) to keep them free of disease(s). 

The effects of extrinsic factors on lab mice have been evaluated for years and were recently considered in a workshop held by NIH ORIP in September 2022. What’s clear is that multiple studies have shown how some of these extrinsic factors alter mouse physiology and behavior in both strain and sex-specific ways that are difficult to fully evaluate and appreciate; suggesting our attempts to create an ‘ideal environment’ may potentially contribute to our problems with reproducibility.

Could expanding the 3R’s fix the problem?
Attend PRIMR22 Annual Conference sessions to learn more.
In sum, I’m not overly surprised we have reproducibility issues with modern research using mice; with more than 13,000 different strains, in many ways I’m more surprised at the amount of useful data we produce, especially when one considers how different the extrinsic factors change from one institution to the next.  Collectively, due to the growing shortage of nonhuman primates for research, I think we should all pause to consider what it will mean if even more scientists start using the readily available lab mouse for translational research studies.

Circling back to the 3Rs, if 90% or more of most research studies are already using mice, should we expand the 3Rs and ask the IACUC to more judiciously interpret the use of mice for research?  Reproducibility, repeatability, rigor, relevance and rationalization all start with ‘R’ and in today’s world may potentially be more important factors to consider than replacement, reduction, or refinement.  As an animal resources director, IACUC Chair and research investigator, I fully recognize these questions are ‘hot button topics’ that are both sensitive and often controversial.  This is especially true if the science using mice isn’t peer-reviewed by an external agency and the IACUC is also charged with (in some manner) evaluating the science behind the use of animals.  Smaller institutions, which have fewer resources and scientists conducting large-scale experiments with mice, are probably at an even bigger disadvantage when faced with expanding the 3Rs to determine if the use of animals is justified in an unfunded scientific proposal. 

I’m not intentionally picking on lab mice – I enjoy working with them – but if your program uses large numbers of mice for research and something I said touches a nerve (either positively or negatively), you may wish to check out one of the many sessions at 2022 PRIM&R’s Annual Conference discussing the 3Rs or one of the sessions devoted to small programs.