6 ms·
Anyone know why apparently so many things work on mice but not people? Apart from "cause mice aren't people" of course.
by andrewstuart 12y ago
Anyone know why apparently so many things work on mice but not people? Apart from "cause mice aren't people" of course.
- thinkcomp 12y agoThis page gives a good summary as to why mice are used in research: http://research.jax.org/mousegenetics/advantages/advantages-of-mouse.html http://research.jax.org/mousegenetics/advantages/advantages-... Humans have far more variation in environment, they think relatively independently (making sticking to a prescribed regimen less likely), and they are far more complex organisms in general.
- drmilsurp 12y agoLot's of things we think we know but don't and chemical space is super large? Why a molecule works within a mouse but not human, depends on what you mean by doesn't work. It might not be that it doesn't work but that the potential drug never makes it that far in the process for other reasons. Common reasons would be poor ADMET issues and difficulty in formulation for enteral route. edit for other route Al Mann spent almost a billion and 20 years on inhalable insulin and only recently found success. Also, works is a matter of interpretation. Better relative to other drugs on the market? Shows better outcomes than no treatment? There are some drugs out there that with time and more investigation are showing some lack of efficacy, eg gliptins
- ars 12y agoBecause you can experiment on them and keep iterating till you find what works. They also have short lives, so you don't have to wait long to see if some new treatment works.
- cperciva 12y agoOne reason is that "cures X in mice" doesn't actually mean "cures X in mice" -- it usually means "cures X in mice which have a genetic mutation which causes them to always develop X". (And sometimes it means "cures X in mice to which we have surgically performed X", e.g., X = spine severing.) For example, studies of type 1 diabetes usually take place in BB rats (http://en.wikipedia.org/wiki/Biobreeding_rat http://en.wikipedia.org/wiki/Biobreeding_rat) or NOD mice (http://en.wikipedia.org/wiki/NOD_mice http://en.wikipedia.org/wiki/NOD_mice). These animals are known to exhibit the symptoms of type 1 diabetes; but we don't know if their disease progression exactly matches that of humans with type 1 diabetes (and we can't know, because we don't know exactly how type 1 diabetes progresses in humans). It's entirely possible that a cure for NOD mice simply blocks a disease pathway which doesn't occur in humans anyway.
- tempestn 12y agoGood point, but a minor nitpick: using X to represent multiple things, especially in the same sentence, is confusing. You could always use other letters, like Y!
- Steuard 12y agoTo my eye, the previous comment was using X to represent the same thing, at least within each individual phrase. (Plugging in "X=spine severing" into both uses in that example seems to work just fine, for instance.)
- tempestn 12y agoAh, I suppose you're right. I guess my brain automatically rejected the notion of "cures spine severing", so assumed it must be something different. Apologies to parent!
- dnautics 12y agoI can give a hypothetical reason why specifically might not work (disclaimer: my PhD thesis is in this field). In most Type I diabetics, the autoantigen (as in the protein that is the target of immune attack) is this peptide called amylin. This peptide is a generally problematic peptide - it's metastable - and in Type II diabetics it forms plaques. Interestingly, mice and rats have amylin, but unlike most other mammals, it's not metastable[0]. I would not be surprised if in NOD (non-obese-diabetic) mice it's NOT the autoantigen. Based on different protein targetting, the mechanism of beta-cell killing in NOD mice might not have exactly the same autoimmune interactions, and a rejuvenation strategy might not work in humans quite the same way. It's not entirely clear to me when they plan on doing therapeutic intervention for children with Type I diabetes; by the time first symptoms show up it may be too late (unlike the mice where they knew when/that it would happen). Finally, it's a bit odd to me that they didn't look through health records of existing type I diabetes patients that have been incidentally prescribed Verapamil. There should be some usable data that hints at the possible usefulness of this strategy, unless the patients were treated for hypertension much later in life than getting diabetes (very likely), aka too late to do anything about it. [0]mice and rats are weird, they also have two copies of the insulin gene. One of the copies has an amino acid that is absolutely conserved across all vertebrates substituted for a substantially different one. [1] bonus: rats also don't have gall bladders. [2] bonus 2: While I was working on diabetes model systems in the lab, we had lots of verapamil floating around, because my boss had this crazy idea that verapimil would fix a whole category of diseases (calcium homeostasis is important for protein folding in the endoplasmic reticulum).
- bigtunacan 12y agoNo PhD here, but IAMA T1D. I noticed from the actual study that candidates for the trial must be within 3 months of diagnosis to even be considered; so it seems they are trying to screen and find only those people who have just recently developed diabetes and reverse it early in the process. In a sense this is sort of like catching a soon to be T2D in the pre-diabetic phase and telling them to eat less and run more. (There is no offense intended in that previous statement; with an obese pre-diabetic (type 2) it is possible to stop the disease before things are too late through exercise and diet, this is not the case in with a Type 1). This makes me question if this is an effective treatment ONLY when caught early; or if it could be useful for long time diabetics like myself. Also this treatment addresses beta cell replacement which is one side, but it doesn't address the autoimmune side which is actually causing the beta cell destruction in the first place. So would this be a lifetime pill that is taken to continuously build new beta cells? As I see it in the autoimmune T1D case as the beta cells remain under attack this would not be a "cure" per se, but rather an ongoing treatment that would result in an overall better outcome and quality of life for a diabetic (given it works at all).
- nmeofthestate 12y ago"apparently so many things work on mice but not people?" If we experimented on people (tho it would take longer) we would probably find lots of things that work on people but not mice.
- possibilistic 12y agoNobody complains about us keeping lines of various knockout mice you can order from a catalog. Nobody has a problem with us killing them when it's time to passage the cells. It's not an issue when we create stress responses by burning their feet. No problem when we inject them with toxins, bacteria, viruses, foreign cells, and a host of other things so we can watch how long it takes for them to die. I say this as someone who is a proponent of the murine model. I think animal models are incredibly valuable to science and we wouldn't be where we were without them. That said, imagine all the things I described being performed on humans... I don't think that would go over so well. A "nonfatal" lab setup wouldn't yield the same kind of results. You would be greatly limited to the scope of what you could investigate. Many experiments kill out of necessity. Also for the sake of throughput. Most of the experimental mice lines are doomed simply by the fact they lack critical immune function--we stock them that way for the sake of isolating experiments to a particular immune system function. Scientific abstraction. (And for reproducability. They're standardized equipment, if you will.)