October 3, 2008

The dreaded words: "in mice"

Not a week goes by that someone doesn't email me with a news report about some food or supplement that cures diabetes. Invariably, the punchline of this report includes the words, "In mice." Trust me, folks. after ten years of following the diabetes racket all it takes to get me to stop reading any further are those dreaded words, "in mice."

Over this past decade I've seen diabetes dramatically improved in hundreds if not thousands of rodents. Mice have regained healthy blood sugars after being fed chamomile tea, red wine and curry spices. They've made dramatic recoveries after being dosed with amino acids. Hormones have made them slim and frisky. But those of us with blood sugar meters who have tested our blood sugar after testing substances that have done such wonders for mice continue to remain uncured.

There are reasons for this. One of the most important is that mice have bodies completely different from ours. They are tiny creatures adapted to living on a diet of seeds very different from the omnivore diet of us massive humans. They are also a prey species that uses the common prey species strategy of living fast, dying young, and leaving a lot of offspring. So their bodies that don't repair tissue the way ours do. Their metabolisms are much much faster, and not surprisingly, their pancreases, while sharing some similarities with ours, have major differences.

Finally, mice do not in any appreciable way, get old. They age quickly--a 2 year old mouse is a mouse methusalah. In the wild they rarely live for more than a few months. Any cat would be happy to explain why. This means that rodents cannot be a truly good research model for any of the long term diseases that emerge with aging , especially Type 2 diabetes.

And indeed, when we look more closely at the mice used in diabetes research we see how true this is. The researchers doing rodent diabetes research usually use special strains of lab mice that have been breds to exhibit some specific genetic makeup that is considered to be a "model" for diabetes.

If they are a model for Type 2 diabetes, these mice are invariably fat. For example, much Type 2 diabetes research has been done with the OB mouse. This is a monster of a mouse which is obese because it is unable to produce leptin. These mice breed true. Buy an OB mouse, and you can be sure it will grow extremely fat and develop diabetes. But that is where the similarity with diabetic people ends. Because while people do grow fat and develop diabetes, it is not because they are leptin deficient. The number of humans around the world who have been found to be obese (or diabetic) because they lack the gene needed to make leptin is less than ten.

Scientist are constantly breeding new mouse models for Type 2 diabetes--and patenting them--but as is with the case with the OB mouse, though these mice are indeed obese and prone to get diabetes, the gene defects that cause them to do this are not those that have been so far identified in people diagnosed with Type 2 diabetes.

Scientists also breed mice that have been given intentionally disabled genes--the so called "knockout mouse" because in these strains of mice a specific gene has been destroyed or "knocked out." Breeding knockout mice can be helpful way of examining what a specific gene does. DPP-4 knockout mice have been used to see what the impact of inhibiting DPP-4 would be. These mice do not produce DPP-4, they have low blood sugars, and they seem--as far as you can tell with a mouse--to go about their daily lives untroubled by the lack of DPP-4 which gave researchers the idea that inhibiting DPP-4 in humans might be okay. Since these mice don't live long enough to develop most human cancers, alas, the researchers also may have missed the fact that inhibiting DPP-4 would, over time, allow slow growing cancers to metastasize that would have not survived in a creature that was making DPP-4.

It is because studies with mice are done with these genetically selected or modified pedigreed mice that they often prove so disappointing when applied to people.

People with diabetes are not a single breed. Your diabetes is almost certainly NOT caused by the same gene defect that causes mine. Nor is it caused by the gene that predisposed your neighbor down the street to get diabetes. Furthermore, if you go to a distant country where people come from different stock than you do, the people there with diabetes are likely to have yet other kinds of genetic forms of diabetes than you and your neighbors.

So far, scientists have found specific genes that color the way that diabetes arises and behaves in Pacific Islanders, Pima Indians, Danes, Ashkenazi Jews, people from West Africa, and Japanese. While there are also genetic differences between individuals in these populations, each population has different sets of busted genes that are more likely to turn up in people with diabetes in that population than they are in others.

The members of these discrete populations have different patterns of onset of Type 2 diabetes and they also usually have different likelihoods of developing specific diabetic complications. Some populations are more prone to develop kidney failure where others are more apt to go blind or need amputations. People with these different underlying genetic profiles also respond differently to various diabetes medications.

Given this diversity among people with Type 2, the usefulness of the purebred "obese diabetic" mouse is limited. It may be able to answer some questions for researchers, but the question they can't answer with any authority is, "Did I just find a substance that cures human diabetes?"

When we move from Type 2 diabetes to Type 1, we run into another mouse model. Research on autoimmune Type 1 diabetes is almost always done with the the NOD (non obese diabetic) mouse, which is a strain of mice that breeds true for a susceptibility to a mousy kind of autoimmune diabetes that shares some antibodies with human autoimmune diabetes.

Dr. Denise Faustman's current research that has raised such hopes in the Type 1 community is based on her results curing NOD mice. But while we hope that the techniques which were so helpful to these mice work for people, that they will make the transition from mice to humans is not guaranteed. Because while the mouse model is similar to human Type 1 diabetes, it is still a mouse, and there are still significant differences between a Type 1 mouse and a Type 1 person.

Beyond that, the genetics underlying Type 1 are not all that uniform either. While there are certain genetic markers that have been identified as being more common among people with Type 1 diabetes, genes alone do not explain the disorder and there are significant genetic differences between different people with Type 1.

Sometimes rodent research is done with "streptozotocin-treated" rats or mice. Streptozotocin is a poison that is specific to the beta cell. Give it to a normal rat and you end up with a rat with a dead pancreas. This is useful if you want to see what life is like for a critter with a dead pancreas, and these animals can be useful if all you want is an animal that cannot control its blood sugar without help. But again, a poisoned mouse, like any mouse is only of limited value in studying the impact of treatments on humans with diabetes.

Because when you give a mouse that healing substance --no matter what its impact on blood sugar--you can't ask the mouse if its tummy hurts. You can't ask if it has a headache, if its little nerves are tingling, if it can still remember where it put its car keys. When you give a mouse a healing substance, you also can't discover whether ten years later the mouse will develop a serious cancer of a type that takes ten years to develop or if it will end up with a damaged heart muscle. Mice cannot not really model any disease that takes a long time to ripen in a long living animal designed to preside over the top of the food chain--as we are.

The strength of mouse studies is that they are are a quick way of testing an idea to see if it has any plausability at all. A lot of medical ideas turn out to kill mice or leave them with such obvious problems that you wouldn't want to waste any more time on them--though it is not impossible that there are things that might work very well in people that get lost because they don't work in the very different physiologies of mice. But if something does work in a mouse, there is a lot of work ahead to see if it works in other larger mammals, and eventually in people. Most things that look good in rodents, don't.

Which is why people with diabetes are drinking red wine, slurping down chamomile tea, supplementing with hundreds of expensive minerals, amino acids, and herbal extracts, each one of which has done great things for rodents, but they are still running blood sugars much higher than normal.

And why I really wish the media would stop publishing these articles with those dreaded words, "In mice."

8 comments:

Lee Ann Thill said...

I usually don't read any further when I see "in mice" either. Even if it's something that will eventually have promising applications for people, it's so far in the future that it's really of no consequence to me. Not to say I'm not grateful for the research because they have to start somewhere, but I'd ridden the "cure in 5-10 years" roller coaster for so long that I had to get off for my own peace of mind.

Stephan Guyenet said...

Hi Jenny,

I totally agree that evidence in mice is often over-extrapolated. I cringe every day when I read the health section of Google news.

But I think mouse studies are still useful. For example, the Ob mice. They are leptin deficient. As you said, very few people are leptin deficient. But many people are leptin resistant. The two are similar.

So I think the Ob and especially the Zucher (leptin receptor KO) are useful because they model leptin resistance. They get something that looks a lot like metabolic syndrome, suggesting that leptin resistance may be a path to it.

So while those studies don't definitively show that leptin resistance/deficiency cause met syn in humans, they do suggest viable hypotheses that can be cross-checked with human data (for example, the fact that leptin resistance typically precedes insulin resistance and overweight).

meanderings said...

You're amazing and I thank you for taking the time (lots of it!) to help us understand diabetes - all of the types.

Anonymous said...

Please read this. Dr Faustman's research validated using HUMAN blood samples.
http://www.diabeteshealth.com/read/2008/09/11/5906.html

Anonymous said...

"The dreaded words: in people with a generic drug"


Yes, it is indeed true that in 2001 in mice, we cured end stage diabetes. The work represented disease reversal, not just disease prevention in truly end stage diabetic animals. The pancreas regenerated after the bad T cells were removed in part with a cheap generic drug called BCG. Now almost 8 years later, there are indeed a lot of healthy happy and formerly diabetic mice around the world as researchers verify, that at least in end stage diabetic mice, we have them well treated.

We are now in human clinical trials and human diabetic blood now shows that the way we killed the bad T cells in mice, is similarly occurring in human blood. Please log onto our web site - faustmanlab.org Your presented information is not accurate and out of date.

We are in human clinical trials, not mice. The trials are unique in trying to verse diabetes in people with long standing disease, the reason the mouse data drove us to try to design trials not just in new onset diabetes, the topic of most mouse cures.

The latest data just published two weeks ago now show in 100s of type 1 diabetic blood samples, the bad T cells that cause this disease can be targeted and directly killed, not just immunosuppressed

It is rare to have mechanistic data correlating mouse cures with human trials i.e. humans like mice have a similar defect in bad T cells that allow them to be killed.

You have a right to be skeptical of the vast numbers of mouse cures. You have a right to be skeptical of vast numbers of mouse cures in not yet diabetic mice, the majority of the studies that obtain headline news. We could not agree more. But you should distinguish research work in mice that reverse established diabetes. You should be aware that although the happy mouse data was published in 2001, we are deeply in human trials, not mouse trials. Indeed this is a rare human clinical trial in that drug mechanism is known and verified for the effect we desire.

As many know this is also a unique trial in using a cheap generic drug called BCG. Criticisms of moving this forward to humans and misinformation that leads all to believe that only mice show this defects is puzzling. One cannot help but think of a central question, money.

Why would anyone disparage the testing of a cheap safe generic drug in people with long standing disease to reverse disease? Why would anyone publish misinformation dated to 2001 time period of research findings instead of research updates of the 2008 time frame?

We cannot predict and we cannot promise any human cures. What we can promise is that we are the first group in the world taking on the touch task of trying to reverse with a well-known safe immune intervention diabetes of many years duration. WE could fail but we will try. We can also promise you we are also unique in conducting these trials solely with the support of public donations and without any pharmaceutical money. Maybe this is why the public is sometimes mislead to believe we are still only curing mice. You just might not want something cheap to work or even be tested.

Dr Denise Faustman

Jenny said...

Dr. Faustman,

I think you misinterpreted what I wrote. I did not suggest that you were working with mice now, only that your current research with humans has not been completed and published so the scientific world does not yet know its outcome.

I am thrilled you are doing it and have contributed to the fund raising effort that is making it possible.

Anonymous said...

Aren't you ever tempted to eat some diabetic mice, see if you can catch the cure off them?

Anonymous said...

I agree with the comments about hearing about a huge variety of things that cure type 1 but here in Ireland, we are all watching Dr Faustman's trials with bated breath because they are different in their bases and their outcomes to date. I think we should support Dr Faustman and my family, friends and I are all praying hard for her success. Instead of knocking this genius, why don't we in the diabetic community pull together to support her. I read a suggestion from a lady in the States that if all of us in the community sent $25 to the fund for this research, it would only take 1 million of us to cover Phase 2. And there are far more than 1 million people with type 1 across the world! Let us be positive. TB, polio, etc etc were once incurable until a brilliant scientist on a mission made it their life's search. Let us thank God that a brilliant scientist of Dr Faustman's calibre has taken on our cause to see what she can do. So far, she has given me hope and we all need hope... God bless her,
Aine Maire in Cork, Ireland