Glucose, that innocuous sweet molecule, is reactive. Chemically, what happens is that glucose has a slight reducing capacity (i.e. the capacity to accept electrons) and it is quite happy to accept electrons from the epsilon amino group of the amino acid, lysine. Lysine is found in virtually every protein and when that electron is shared it creates an irreversible chemical bond between the glucose and the protein. The protein has become glycosylated and your diabetes complications have begun.
Clinically, glycosylation is a very useful way to assess blood glucose levels over time. Hemoglobin, being a protein, is as vulnerable to glycosylation as anything else. The major form of glycosylated hemoglobin is called hemoglobin A1c (hA1c). Now, the half life of hemoglobin (really the half life of the red blood cell) is a little short of 2 months. The amount of hA1c in the blood is a very good estimate of the average amount of blood sugar present over those two months. Thus while you may be very good just before your doctor visit and have a low blood sugar reading, you cannot fake the hA1c measure. Indeed, this measure is so useful that the American Diabetes Association has decided to promote the use of it for the diagnosis of diabetes. A hA1c level below 6.5 (i.e. 6.5% of your hemoglobin is in this form) is correlated with the absence of complications developing over time. As your hA1c level rises, so does the likelihood of complications. So, obviously this is a pretty useful number.
The glycosylation of hemoglobin is benign. It does nothing to hemoglobin’s ability to carry oxygen. However, that cannot be said for other proteins. Indeed, glycosylation has been directly implicated in the development of several diabetic complications. Glycosylation in and of itself is not the problem. Many proteins are glycosylated by enzymatic reactions and the various sugar chains placed on the surface of a protein plays an essential role in that protein’s function. Rather, it is the specific properties of this non-enzymatic attachment of glucose that appears to be pathological. Each of these proteins can undergo cyclization and rearrangement or crosslinking to create something called an advanced glycosylation end product (AGE).
AGEs can directly affect tissue flexibility by accumulating in extracellular matrix. The stiffening of tissues leads to decreased kidney function as well as an increased work load for the heart. The crosslinked sugars trap plasma proteins and low-density lipoproteins (LDL) leading to atherosclerosis.
Additionally, AGEs bind to a receptor. The receptor is interesting as it is capable of binding a wide variety of glycosylated proteins. We call the receptor; RAGE, as in the receptor for AGE. RAGE, while being well defined genetically, remains a bit elusive as a signaling protein. In part this is because, RAGE can associate with other proteins, modifying its function. The binding of AGEs results in a number of interesting changes inside the cell including the generation of reactive oxygen species and the activation of the important inflammatory transcription factor: NF-kappa B. Some consequences of RAGE activation include increased synthesis of matrix proteins, the secretion of inflammatory mediators, and the increase in cellular stress via reactive oxygen generation. Why would we have a system to do this? It seems counterintuitive. The thing to keep in mind here is that these are pathological events. RAGE really is designed to interact with other proteins as opposed to AGEs to do various important things. For example, RAGE signaling has been shown to promote survival of tissues under stress. AGEs should not be present in appreciable amounts and it is their presence that sends this receptor down this destructive path.
One very powerful way in which we can establish the importance of this (or any other) receptor system is through genetic disruption. Colloquially we refer to these as “knock out” mice. We have the technology to grow mouse embryonic stem cells and, through a process called homologous recombination, introduce a large piece of DNA that contains a relatively small mutation. By carefully constructing the DNA such that the mutation disrupts the production of a particular gene we then have created a situation where one allele on the mouse chromosome is now mutated. We need to screen from thousands to millions of cells to find the one cell that has the right integration event. Once we find that cell and grow it up into millions of cells we can start including small numbers of these stem cells in mouse embryos and eventually, after what seems like forever, we get a mouse that has incorporated the mutated stem cell into its germ line tissues. This is the magic event that we have been waiting for. Now we have the mutant allele fixed in a living mouse in such a way that it will be transmitted via Mendelian genetics to its offspring. This has been done for hundreds of different genes and many of them are commercially available through venders such as Jackson Labs in Bar Harbor Maine.
So, what happens with the RAGE knockout mouse? Well, when the RAGE knockout mouse (no function copies of RAGE) is made diabetic, and is compared to diabetic heterozygous littermates (littermates that have one functional copy of RAGE) it was observed that the degree of complications for the RAGE knockout mouse was much less. In particular, kidney disease was decreased. This is strong evidence that AGE engagement of RAGE plays a big role in diabetes associated kidney failure.
People with RAGE mutations have been studied as well. Another word for this genetic variation is called polymorphism. Indeed, polymorphisms involving certain amino acid substitutions have been associated with increased kidney disease in individuals. Again, this is strong evidence associating RAGE with kidney disease.
Interestingly, AGE – RAGE interactions can be blocked. Antibodies have been generated as well as soluble forms of the receptor which would be predicted to slurp up AGEs and keep them from binding to the real receptors. Clinical trials are in the formative stages so we do not yet know whether intervention at this level will be beneficial. All we can say is that inappropriate RAGE is bad on so many levels.