Japan is one of the few countries that actually publishes the number of people who die suddenly on the job. Their name for it is karoshi: death by overwork. In the later stages of type 2 diabetes, the endocrine pancreas undergoes karoshi and the patient is sentenced to a life of insulin shots. How does this happen? Are there factors that predispose islet cells to die in this fashion? Yet another piece of this puzzle was discovered and recently published in the Proceedings of the National Academy of Sciences (PNAS).
The Indiana University research group, led by Dr. Dorris Stoffers, began by considering an interesting protein called Pdx1. Pdx1 is something we call a transcription factor. Our genes code for proteins and some of these proteins, in turn, function to regulate large families of genes. Complex physiological processes then can be regulated by a very small number of these transcription factors which induce the production of different proteins in many different tissues all for some common purpose. Pdx1 is of interest to people studying diabetes because it sits on the insulin gene and promotes the production of insulin. This research group knew that it must do other things so they set about looking at its functions in a more global manner.
In the dim dark past (about 10 years ago) we used to laboriously analyze tissues, measuring the presence of specific proteins one by one. If an antibody for the protein was commercially available we could use the antibody in a variety of different formats to detect the presence of the protein. If we knew the sequence of the gene we could use that information to construct probes that would measure levels of messenger RNA (and thus estimate the level of gene transcription). Now, however, various companies have developed high throughput methods for globally analyzing virtually all of the messenger RNAs present at one time producing tens of thousands of data points in one experiment. We call these things “gene arrays”.
The authors combined this gene array technology with another technology which recently won its inventor with the Nobel Prize. Small interfering RNA (siRNA) is a kind of RNA that actually functions to destroy messenger RNAs. Because of the structure of nucleic acids such as RNA and DNA, they can do something called base pair. We have 4 bases: A, C, G, and T. A pairs with T while G pairs with C. If a nucleic acid has a perfectly complementary sequence it can base pair with its complementary strand to produce a double stranded nucleic acid. As you may recall, double stranded DNA forms a very beautiful helical coil and since it is made up of 2 strands we call it the “double helix”. At any rate, returning to siRNAs, the sequence of the RNA is complementary to a very specific messenger RNA and when it binds to it, there forms a new binding site for a protein complex with the delightful name: dicer. Dicer does just that; it cuts up the messenger RNA thus blocking the gene from expressing. This is both a fascinating mode of gene regulation and an incredible new tool for research as well as cutting edge therapies. No wonder that it garnered a Nobel Prize.
The authors created an siRNA against Pdx1 and used it to block the production of this master regulator. They did this in isolated cultured pancreatic cells. They then prepared total messenger RNA from these cells and compared it to messenger RNA from untreated (control) cells using the gene arrays I described above. What they saw was a very large number of changes in the genes being made. Remember Pdx1 is a transcription factor and it probably binds to lots of genes. Now, in one fell swoop, the authors had access to ALL of the genes that Pdx1 might regulate.
Cool. So, what did they find? They found that Pdx1 regulated a whole bunch of genes involved in the ability of the endoplasmic reticulum (ER) to handle stress. What does the ER have to be stressed about? Protein folding. Some proteins are happy floating about in the cytoplasm (usually as part of large macromolecular machines). Others, many others actually, are intimately associated with cell membranes. Receptors, channels, transporters, secreted proteins all need to be placed into or associate with the membrane or as they are synthesized. The ER is the place where this happens. ER stress is extraordinarily important in diabetes for a lot of reasons and explaining all of the connections will require more than one post. Suffice it to say here, ER stress will cause the cell to die. Remember the cell that we are talking about is the pancreatic beta cell.
In a parallel series of experiments the authors examined mice in which one copy of the Pdx1 gene was deleted. We will call them Pdx1+/- mice. Remember, we have 2 copies of each chromosome (except for the male sex chromosome) and so we have 2 copies of every gene. We cannot perform experiments on Pdx1-/- mice because the gene appears to be essential and Pdx1-/- mice do not survive. Pdx1+/- mice have about half the amount of Pdx1 protein as normal (Pdx+/+ mice) so removing one copy of the Pdx1 gene does have an effect. At any rate, the authors found that when Pdx1+/- mice were fed a high fat diet the ER from pancreatic beta cells looked terrible. Normal mice fed a high fat diet had only a slight change in the ER indicating increased insulin synthesis. Pdx1+/- mice were somewhat insulin resistant and after 1 month on the high fat diet had very high fasting blood glucose levels. By 4 months the poor things were diabetic. In comparison, the changes in normal mice fed a high fat diet were much smaller and these mice did not get diabetes. When they looked at the morphology of the pancreas in each population after being fed high fat and normal diets, the authors found that in normal mice the mass of pancreatic beta cells increased after several months on a high fat diet. In comparison Pdx1+/- mice fed a high fat diet showed little change in the number of beta cells. Also, when beta cell proliferation rates were examined there was no difference in the two populations. Thus if less cell birth cannot account for the lack of increase in beta cells, then we must invoke greater cell death and this brings us back to ER stress.
To sum up, as we punish our bodies with high fat diets we become resistant to insulin but since our pancreas did not get the memo, it ramps up production, in part, by growing. For those of us who do not get diabetes, the pancreas seems to be able to keep up. Since there are a great many obese people who do not have diabetes, it appears that some can keep this going for quite a bit of time. When things begin to break down, we now, thanks to the research just described, have a place to look. The process that Pdx1 regulates can now be examined in actual diabetics as well as prediabetic people. We await the result with excitement.