| Joslin’s C. Ronald Kahn, M.D., on Where the Insulin Action Is |
 For the better part of a century, since its discovery in a Toronto
laboratory in 1921, insulin was known in the medical research community
as little more than the hormone that was lacking in
diabetes. But the
roles of insulin are many and diverse, and researchers are still trying
to make sense of it all. Insulin is the primary regulator of fat and
carbohydrate metabolism, and controls the storage of these nutrients as
well. It stimulates the synthesis of molecules involved in the function
and growth of cells, and of protein and even RNA and DNA molecules in
our cells.
|
C. Ronald Kahn, M.D., of the Joslin Diabetes Center, observes that his research points to
“one unifying hypothesis in which insulin resistance in different tissues can contribute different parts of the metabolic syndrome of diabetes.” |
The last decade has seen a
revolution in the understanding of insulin and the condition known as
insulin resistance; as is usually the case, this revolution has been
driven by technological innovations. No one has contributed more to
these breakthroughs than the Boston diabetologist C. Ronald Kahn, M.D.
Among the highly cited reports by Kahn and his colleagues in the last
two and a half decades are two papers with more than 1,000 citations
each, and an even 10 that have each garnered more than 500 citations.
The table
below features a selection of Kahn’s high-impact papers
published since 1994.
Kahn, 61, studied chemistry at the University of Louisville, where he
obtained his undergraduate degree in 1964, later attending the
University’s school of medicine to earn his M.D. degree in 1968. He
did his internship and residency at Barnes Hospital in St. Louis, and
then spent the 1970s working at the National Institutes of Health as a
researcher and attending physician. In 1981, he moved to Boston to join
the Harvard Medical School and become chief of the Division of Diabetes
and Metabolism at Brigham and Women’s Hospital and an associate staff
member of New England Deaconess Hospital. In 1985, Kahn joined the
Harvard-affiliated Joslin Diabetes Center and is now its president and
director. He is also the Mary K. Iacocca Professor of Medicine at
Harvard.
Kahn spoke to
Science Watch
from his office in Boston.
How would you describe your approach to deciphering the
insulin/insulin receptor system?
Kahn: We’ve tackled this problem using many approaches.
Over the past 10 years, we’ve focused on the use of genetic techniques
to understand how insulin works and what kind of insulin resistance
occurs when one specific step in this insulin-signaling network fails to
function properly. All such malfunctions would result in an
insulin-resistant state, but they may not all be the same state. Some
failures in the signaling network might predispose to one disease,
others to another. Our most extensive studies have involved use of
genetic techniques to knock out the insulin receptor and its substrates
in mice. Once the insulin receptor is triggered, it acts on a whole
family of substrates —these different proteins inside the cell.
The question is, does each substrate play a different role in insulin
action? One way to tell is to genetically knock them out, one after the
other. Do they affect diabetes, body weight, etc.? With this approach we
can begin to understand in a much more precise way where the defects
might be, so that we can then target them in therapeutics or envision
them diagnostically.
Why not just knock out the gene for the receptor in the
entire animal, which is a common procedure with other genes?
Kahn: When you knock out the insulin receptor in the whole
body, the animal can’t survive. It becomes completely insulin
resistant. In the first week of life, it goes into diabetic ketoacidosis
and dies. On the other hand, if you knock out the insulin receptor
substrate 1, known as IRS-1, which is what we did first, the animal is
very insulin resistant, but it’s still able to survive and grow. It’s
half its normal size and insulin resistant, but will live pretty much
its normal lifespan.
So what did these substrate knockouts tell you?
Kahn: First of all, it indicated that there must be another
substrate, another way that insulin can get its signal in. This led to
the notion that there was a second substrate called IRS-2. And Morris
White, who was initially a fellow in my lab and is now a faculty member
at Children’s Hospital in Boston, was eventually able to purify it and
clone it. Then Gus Leinhard, at Dartmouth, found IRS-3 and IRS-4, and
the family of substrates began to grow. A few other substrates have also
been identified by other techniques. All in all, this family of insulin
receptor substrates apparently serves to transduce signals, and when
each of these is knocked out, it is clear that each contributes to
different parts of the insulin-signaling pathways. So the knockout of
each of these animals has a somewhat different phenotype. That’s a
very important observation.
And this told us that the insulin-signaling pathway is a complex
network of signaling effects that diverge and converge. It’s not a
bunch of redundant pathways, but complementary pathways. This is a very,
very effective way to modulate a complex signal
You said that knocking out the substrates was the first
thing you did. Where did you go from there?
Kahn: The next step after creating all these single
knockouts was to create more complex genetic models. We’re fairly
certain that human type 2 diabetes is a complex genetic disease, a
polygenic disease resulting from interaction of the environment with
multiple different partial genetic defects. So our method was to create
polygenic models from these monogenic models. For example, we could
alter one of the two alleles of a gene to create a heterozygous defect,
and by combining these partial defects ask to what extent there was some
resemblance between these complex partial defects and human type 2
diabetes. One of our most interesting models of this type was a
double-heterogeneous knockout of the insulin receptor and insulin
receptor substrate 1. When mice had 50% of both of these genes, about
half those animals developed type 2 diabetes-type syndrome; the other
half didn’t. And this seemed to be an excellent polygenic model of
type 2 diabetes.
Was it around this time that you began knocking out
insulin signaling in specific tissues? What were you hoping to learn?
High-Impact Papers by C. Ronald Kahn et al.,
Published Since 1994
(Ranked by total citations)
|
| Rank |
Paper |
Citations |
| 1 |
M.F.
White, C.R. Kahn, "The insulin signaling
system," J. Biol. Chem., 269(1): 1-4,
1994. |
1,032 |
| 2 |
B.
Cheatham, et al., "Phosphatidylinositol
3-kinase activation is required for insulin stimulation of
pp70 S6 kinase, DNA synthesis, and glucose-transporter
translocation," Mol. Cell. Biology, 14(7):
4902-11, 1994. |
824 |
| 3 |
E.
Araki, et al., "Alternative pathway of
insulin signaling in mice with targeted disruption of the
IRS-1 gene," Nature, 372(6502): 186-90,
1994. |
590 |
| 4 |
B.
Cheatham, C.R. Kahn, "Insulin action and the
insulin signaling network," Endocrine Rev.,
16(2): 117-42, 1995. |
539 |
| 5 |
C.R.
Kahn, "Insulin action, diabetogenes, and the cause
of Type-II diabetes," Diabetes, 43(8):
1066-84, 1994. |
459 |
| 6 |
R.N.
Kulkarni, et al., Tissue-specific knockout of
the insulin receptor in pancreatic beta cells creates an
insulin secretory defect similar to that in type 2
diabetes, Cell, 96(3): 329-39, 1999. |
290 |
| SOURCE: Thomson
Scientific
Web of Science |
|
Kahn: Well, diabetes affects the liver, muscle, and fat
tissue, but it also affects insulin receptors on virtually every tissue
in the body. So the question was, could we learn about the role of the
insulin receptor and insulin action in each tissue without ending up
with a lethal phenotype? We started with a muscle knockout of the
insulin receptor, and then went on to do liver, fat, brain, vascular
endothelial cells, brown adipocyte cells, beta cells in the pancreatic
islets of Langerhans, cardiac cells, and we have others ongoing.
Why did you start with muscle tissue?
Kahn: There’s a lot of evidence that insulin resistance in
muscle is the earliest defect in type 2 diabetes. So we thought that if
we made the muscle resistant, maybe eventually the animal would evolve
to become diabetic as it ages or becomes obese. But that didn’t
happen. Even though muscle is a major site of glucose metabolism, it
turns out that when you knock out the insulin receptor in muscle, you
don’t create diabetes. And that’s because the muscle can still take
up glucose through insulin-independent mechanisms like exercise. The
other thing that happens is that the animals will take glucose into
other tissues more avidly if the muscle is not using as much. For
instance, glucose uptake into fat will increase. So the animals become
slightly obese, but not diabetic.
What happened when you knocked out the insulin receptor
in fat tissues?
Kahn: That animal has a number of interesting phenotypes,
but it didn’t get diabetes either. In fact, the fat-specific insulin
receptor knockout mice are lean and protected from diabetes and obesity.
So which tissue knockouts did lead to diabetes?
Kahn: Three tissues seem to be most important in creating
the diabetic phenotype in terms of insulin resistance. First is the
liver. Insulin resistance at the liver makes mice hyperglycemic and
makes them show other characteristics of insulin resistance. They become
hyperglycemic because they can’t turn off hepatic glucose output. The
second tissue that turns out to be a major player in producing diabetes
is the beta cell —the insulin-secreting cell.
Why the beta cell? That seems surprising.
Kahn: This was really a surprise. We knew there might be
insulin receptors on the beta cell, but most researchers thought the
role of insulin on the beta cell would be to turn off secretion, like a
positive-feedback loop. In fact, however, insulin on the beta cell is
important for normal function, especially for the ability of the cell to
sense glucose level, i.e., getting glucose to stimulate insulin. And
when we knock out the insulin receptor in beta cells, the beta cells don’t
sense glucose normally and therefore don’t release insulin normally,
and so the mice become hyperglycemic and diabetic. So the beta cell and
the liver are the two most important. We expected the liver. We didn’t
expect the beta cell.
The third tissue involved is the brain, and this was also unexpected.
Insulin plays at least two important roles in the brain. For one,
insulin has a kind of leptin-like effect to suppress appetite. But, even
more surprisingly, it also turns off the liver production of glucose.
This probably happens through the nervous connections between the brain
and liver, the vagus nerve. All in all, then, these three tissues turn
out to be the most important tissues, controlling glucose metabolism in
which insulin resistance occurs. The only one that we expected would be
critical was the liver. We hadn’t thought about the beta cell and the
brain.
So how has this informed your understanding of diabetes?
Kahn: We can put a lot of this together into one unifying
hypothesis in which insulin resistance in different tissues can
contribute different parts of the metabolic syndrome of diabetes.
Insulin resistance at the level of the beta cell creates a
glucose-secretion defect; insulin resistance at the liver creates a
defect in hepatic glucose output; and insulin resistance in the brain
contributes to the obesity problem, as well as to the control of hepatic
glucose output. And even insulin resistance at the muscle contributes,
causing glucose to be shunted into fat, causing a little bit more
obesity and increased triglyceride levels. So it all kind of fits
together, but in a way we didn’t expect.
One of the most surprising things you found when you
knocked out insulin in the fat tissue was that these mice had remarkable
longevity. Tell us about that.
Kahn: Well, the first thing we noticed with these animals
was that they were leaner. They had about a 50% reduction in fat mass.
That wasn’t completely unexpected. After all, insulin is a hormone
that stimulates fat synthesis. So if there’s no insulin action in fat,
you would expect less fat synthesis. What was striking in these animals
was that they were not only leaner, but they really resisted becoming
obese. We put them on a high-fat diet, and they didn’t become obese.
We created a chemical injury to the centers in the hypothalamus that
control appetite, but this didn’t cause obesity even though it caused
the mice to overeat. As animals age, they all tend to gain weight, and
mice are no exception. However, these mice didn’t. So this created an
opportunity to look at a very important question about the relationship
between body weight, food intake, and longevity. It had been known for
some years that if you calorically restrict animals they will live
longer. That’s true in C. elegans, drosophila, mammals.
It’s also well known that lean animals live longer than obese animals.
But here’s a real chicken-and-egg question: Obviously if you eat less,
you’re leaner. The question is, do you live longer because you’re
leaner, or because you’re eating less? It’s not clear which is the
most important determinant of increased life span.
So which one is it?
Kahn: Well, these mice, which we dubbed FIRKO, for
"fat-specific insulin receptor knockouts," were leaner but
they didn’t eat less. They ate a normal amount of food. And it turns
out they lived considerably longer than normal mice. So being leaner
will allow mice to live longer even if they don’t eat less. Maybe food
restriction also helps, but if animals are not food-restricted they can
still live longer if they’re leaner.
If it’s eating more, how does it stay lean?
Kahn: Presumably it’s using up energy in other ways. That’s
one of the things we’re still studying: Where is the energy going? How
is it utilizing energy? We’re now doing studies looking at where the
energy is being expended. At the moment we don’t know.
What can this tell you about insulin resistance and
diabetes that might be important clinically?
Kahn: There are a number of factors we think are
particularly important. One is to understand, when we treat diabetes
with insulin-sensitizing agents, are we improving the insulin action in
all tissues or only some tissues? If the beta cell is really insulin
resistant, how can we improve its sensitivity? Maybe the treatment we
use is not as effective on the beta cell. Maybe if we could improve that
action of the drugs we use, it would be very valuable. Also, how do we
modify insulin action in the brain? The other very important clinical
notion is the unique role of insulin resistance in fat. If we can find a
way to block insulin action in fat, but not in other tissues, we could
find a way to create the beneficial effects we see in these FIRKO mice.
People could eat more, not gain weight, not get diabetes, and maybe even
live longer.
 |
| •
|
View the Special Topic on Diabetes. |
| •
|
C. Ronald Kahn answers a few questions about his fast breaking
paper in field of Multidisciplinary. |
| • |
C. Ronald Kahn answers a few questions about his new hot paper in
the field of Multidisciplinary. |
Science
Watch®, May/June 2005, Vol. 16, No. 3
Citing URL:
http://www.sciencewatch.com/may-june2005/sw_may-june2005_page3.htm |
|
.
