T Nation

Studies performed on Mice


So I came across this study:

Exercise training increases mitochondrial biogenesis in the brain


My question is, as a non-science background guy, how much can this sort of stuff actually be applied to humans?

We see these type of exercise studies performed on mice all the time... how much merit do they actually have?


We like mice as guinea pigs for various reasons, generally because:

  1. They are cheap (starting at $5 per mouse with ~$0.05 in daily maintenance)
  2. They are reliable
  3. They are efficient (easy to care for, quick to reproduce, with short generation times)
  4. They are standardized
  5. They are genetically similar (~99%, with onyl 300 or so genes lacking a counterpart in humans)
  6. They are almost entirely customizable (>1,000 strains to choose from, each with various knock-outs, knock-ins or other mutations)
  7. They are "humanizable" (able to carry/express human transgenes or be engrafted with functional human cells/tissue)
  8. Their use obviates the ethical concerns seen with humans, monkeys, etc.

We dislike mice as guinea pigs for various reasons, generally because:

  1. They are not human
  2. They tend to be used in their younger years for all diseases
    3a. Each murine model studied is genetically standardized with each sample studied under environmentally standardized conditions (we, in contrast, live a genetically and environmentally non-standardized existence)
    3b. These environmental standardizations often include a lack of exercise, over-nutrition, under-stimulation, and occasionally large-scale antibiotic use.
  3. They are often characterized by non-predictive results (pre-clinical trials validated by mice routinely flop when tried on humans for whatever reason)
  4. Despite their genetic similarities, genetic/proteomic responses to disease can differ drastically from that in humans on a case-by-caes basis (e.g., certain Huntington mice do not exhibit dyskinesia, so I hear)
  5. Tend to be best when used to obtain specific functional information (e.g., drug targets or dosage tailoring) rather than to map out disease mechanisms or models.
  6. It's not much fun to publish the studies that flop.


Mice are nice because they are flexible, inexpensive and efficient models; however, each study used tests genetically standardized samples in artificially standardized environments using animals commonly exhibiting non-human physiological responses to disease and, as such, should be taken with a shaker of salt when attempting to extrapolate the information to people.


Basically, it depends.


I just attended a mini presentation thing on the human brain (and a sheep and a rat brain) given by my last semesters biopsychology professor. She made it a point to explain that while there are some notable differences in the human and the rat brain (size, texture), they are actually fairly comparable when it comes to using them for studies.



Thanks for the clarification