Dorrance H. Hamilton Research Laboratories Division of Endocrinology, Diabetes & Metabolic Diseases Department of Medicine Jefferson Medical College Thomas Jefferson University Philadelphia, Pennsylvania 19107
Address all correspondence and requests for reprints to: Kevin Jon Williams, Dorrance H. Hamilton Research Laboratories, Division of Endocrinology, Diabetes & Metabolic Diseases, Department of Medicine, Jefferson Medical College, Thomas Jefferson University, 1020 Locust Street, Suite 348, Philadelphia, Pennsylvania 19107. E-mail: K_Williams{at}Lac.jci.tju.edu
Here is a good clinical definition of diabetes mellitus: "a state of premature cardiovascular death which is associated with chronic hyperglycaemia and may also be associated with blindness and renal failure " (1). The clinical definition was quite different before the introduction of insulin and oral agents, but now diabetics die overwhelmingly from atherosclerotic vascular disease (1, 2). And we do not really know why.
But it is even worse than that. We are failing to provide diabetics as a group with the same extent of cardiovascular benefit that the rest of the population in the developed world has experienced over the past several decades (3, 4). Age-adjusted incidence of cardiovascular disease in the United States has declined overall during the past 30 yr, largely owing to successful management of conventional cardiovascular risk factors. Nevertheless, while nondiabetic men experienced a decline of 36.4% in age-adjusted heart disease mortality over roughly a 10-yr period, the decline among diabetic men was only 13.1%. Nondiabetic women benefited from a 27% decline over the same period, whereas diabetic women have experienced an increase of 23% (3). In other words, conventional risk factor managementalthough clearly beneficial in diabeteshas failed to translate into the same magnitude of improvement for the overall diabetic population as it has for nondiabetics.
Moreover, whatever pathophysiology is occurring in individuals with diabetes seems to affect nondiabetics to an important extent. In particular, subclinical impairments of glucose tolerance have a substantial impact on the health of nondiabeticsthe blood concentration of hemoglobin A1c is a strong predictor of cardiovascular events even among nondiabetics (5, 6). In fact, the vast majority of excess cardiovascular event associated with hemoglobin A1c concentrations above 5% occurs in nondiabetics, simply because there are so many more nondiabetics than diabetics (5). In other words, the enhanced cardiovascular risk in diabetes is an unsolved problem of tremendously broad importance.
The study by Tannock et al. (7) addresses a crucial issue and provides important guidance for future research. Many lines of evidence support the Response-to-Retention hypothesis of early atherogenesis (8, 9, 10, 11, 12, 13, 14, 15). This hypothesis holds that the key pathogenic event that starts an otherwise-normal artery on a path toward atherosclerosis is the extracellular retentionor trappingof cholesterol-rich, atherogenic lipoproteins within the subendothelial region of the arterial wall. Arterial-wall retention of lipoproteins has been documented within hours after acute induction of hypercholesterolemia, well before any other known changes (16, 17). Once retained, these lipoprotein undergo enzymatic and oxidative modifications that generate biologically active by-products that provoke local responses, including chemotaxis of monocyte/macrophages and other inflammatory cells, thereby resulting in lesion progression. Presumably, diabetes enhances something in the response-to-retention cascade.
Tannock et al. (7) examined the possibility that triglyceride-rich lipoproteins in type 2 diabetes might show enhanced affinity for biglycan, an arterial-wall chondroitin sulfate proteoglycan that members of this research group had previously shown to colocalize with lipoprotein deposits in human atherosclerotic lesions (18). Based on some prior reports of abnormal apoprotein composition of these particles, the expected result would have been enhanced binding, but these authors found otherwise. From a purely theoretical standpoint, one potential criticism can be drawn from their evidence that the triglyceride-rich lipoproteins in their study did not show an elevated content of apolipoprotein E, a protein that can mediate binding to biglycan and other proteoglycans. Their isolation procedure is not to blamethe authors carefully compared a physically gentle chromatographic method with high-salt ultracentrifugation, which is known to tear some apoproteins off lipoprotein particles. The unremarkable apoprotein composition of the particles reported by Tannock et al. (7) might instead reflect the degree of diabetic control of their patients, but such patients still show enhanced atherosclerosis. Thus, their results appear clinically relevant.
Where to look now? Lipoproteins are typically the easiest item to examine, because we can get them from peripheral blood. The higher concentrations of atherogenic lipoproteins in diabetic blood, the lower concentrations of antiatherogenic particles, and structural or chemical alterations in diabetic low-density lipoprotein or high-density lipoprotein remain possibilities. As Tannock et al. (7) point out, however, part of the answer may also lie in the nature of the arterial wall, which is not so easy to sample. The biglycan used in their study came from cultured human vascular smooth muscle cells, which is a perfectly reasonable source, but it is not the diabetic vascular wall. Several prior studies have concluded that the matrix of human diabetic arteries exhibits key alterations, including reductions in heparan sulfate and increases in the content of chondroitin sulfate, particularly chondroitin sulfate-B (also known as dermatan sulfate) (19). Nondiabetic atherosclerotic lesions exhibit a similar pattern (19, 20, 21, 22), and diabetic atherosclerotic lesions show the greatest decreases in the heparan to chondroitin ratio (19). Although arterial-wall chondroitin sulfate proteoglycans, such as biglycan, can trap lipoproteins and are, therefore, considered to be generally atherogenic (15), heparan sulfate proteoglycans have a variety of potentially antiatherogenic functions. These include maintenance of the endothelial permeability barrier, which could block lipoprotein penetration into the vessel wall (23), and inhibition of important later events, such as migration of macrophages (24), proliferation of smooth muscle cells (25, 26, 27), and activation of the coagulation cascade (28, 29). Angiotensin II can play a role in these alterations of the diabetic vascular matrix, which might explain some cardio- and reno-protection from converting enzyme inhibitors (30). Other candidate mediators and targets require careful consideration as well.
The goal always remains to improve peoples lives. A clear pathophysiological explanation for the most deadly modern feature of diabetes would be a very large step to aid the design of rational therapies.
![]() |
Acknowledgments |
---|
Received November 9, 2001.
Accepted November 19, 2001.
![]() |
References |
---|
![]() ![]() |
---|