||Scott King Thinks That the Complete Human Genome Won't Help Find a Cure for Diabetes
|"It's good news that we're moving so fast. But it's even better news that researchers throughout the world are using this data now to investigate the genetic underpinnings of health and diseases ranging from Alzheimer's to diabetes," said Dr Francis Collins, director of the US National Institute of Health's Human Genome Research Institute. ( BBC interview, 3/30/00)
Imagine that you repair televisions for a living. A strange customer arrives with a television model you've never seen before.
"I think I can fix it," you say. "But I'll need the manual."
"No problem, buddy. I'll just go get it from the truck."
In a moment he returns with a pile of 46 phone book sized manuals. "Here it is!"
You open the first book. The page is covered with tiny zero's and one's. 01000101101110101010110101000101010010100010100100111110101011110010001....
The next page is too. And the next. You open another volume - more zero's and one's.
"Hey! This is useless! I need a wiring diagram!"
"But, buddy, this is all the information needed to make the television set!"
Lately I have felt like our television repair man when I hear that the newly complete sequence of the human genetic code will make it possible to cure diseases like diabetes. Of course the publication of 85% of the human genome (and eventually 100%) is a great accomplishment, but it helps to keep the value of that information in perspective.
The computer code of zero's and one's above is used to drive the machines that can make the components of a television and assemble them. In biology the sequence of ATCG letters drives the protein synthesis machinery, and everything in biology is either protein or made by proteins. The problems is that the information is in code. It is not a diagram. (It is, after all, called the genetic code!)
The proteins we understand best are enzymes. They catalyze chemical reactions. For instance, the enzyme that turns the sugar glucose into the sugar fructose is called glucose isomerase.
Biochemists write the reaction this way:
glucose <-------------> fructose
This is as simple as a protein gets. Yet, if I gave you the sequence of ATCG's that encodes this protein, you could not predict what the protein does. You would actually have to make it and study it in a laboratory.
A genetic disease is one caused by a known genetic defect. If a hunk of DNA is missing you can get developmental diseases. If a single DNA base is wrong (A instead of T) you can get molecular diseases such as sickle-cell anemia. The brilliant Linus Pauling worked out the nature of genetic diseases in 1949. ( Click here to read more about Pauling's work on genetic diseases.)
Diabetes is not a genetic disease. It is a disease of metabolic control. Excess sugar in the blood is caused by the activities of liver cells, muscle cells, fat cells, and of course islet cells making and secreting insulin.
The islets alone have scores of genes that directly affect insulin secretion. The insulin gene is single gene, but there is a special enzyme to process proinsulin into insulin; a special transporter to bring glucose into the inside of the beta cell; and many enzymes that detect the changes in glucose level and cause insulin secretion to suddenly increase. We know the final step is a sudden rush of calcium into the cells, which seems to cause the storage vesicles filled with insulin to merge with the beta cell membrane and release insulin into the blood. The steps between glucose metabolism and calcium are very poorly understood. We certainly do not understand much bout which components of the system fail in type 2 diabetes. In fact, it is possible that the failure is not a failure of a component, but a failure of the system. All the component proteins might be just fine!
Even when science has identified a specific molecular mechanism the information is usually not enough to help. We have known the molecular basis of sickle-cell anemia for 50 years. People with this genetic diseases are still treated with diet and drugs.
Type 1 diabetes has a strong genetic component. The probability of an identical twin developing this disease is over 50% if the other twin developed type 1 diabetes. Pursuit of this clue and many others lead to an amazing specific molecular medicine results about ten years ago. The key protein has the name HLA-DQ, and the most important amino acid residue is at position 57 of the beta chain. (This protein is involved in immune system function, and type 1 diabetes is an autoimmune disease.) If you have an aspartic acid (found in all proteins and in aspartame sweetener) at HLA-DQ57 you have a very low chance of developing diabetes - less than 0.1%! If you have certain other amino acids in that precise position you have an elevated chance of diabetes.
The different HLA-DQ proteins are encoded by the slightly different variants of these proteins present in the population naturally. If you are unlucky enough to have the worst combination of several such genes, your probability of developing diabetes is over 50%. So we have known a lot about the molecular basis of autoimmune diabetes for ten years.
This knowledge has not yet resulted in any new therapeutics. Even though to a large degree the genetic basis of type 1 diabetes is understood, turning that knowledge into a new therapy is not possible with the technology we currently have.
Scott R. King
Take a look at the human genome at:
UC Santa Cruz Genome Project You can actually download your very own copy of the human genome. Amaze your friends.
Also, Dr. Collins thinks that researchers will use genome information to cure diabetes
Discuss this letter in the Open Forum