Keck School of Medicine, University of Southern California, Los Angeles, California 90033
Address all correspondence and requests for reprints to: Richard N. Bergman, Ph.D., Professor and Chair, Keck School of Medicine, University of Southern California, MMR 630, 1333 San Pablo Street, Los Angeles, California 90033.
Insulin resistance has emerged as a critical risk factor for a plethora of chronic disorders, and it is a hallmark of the metabolic syndrome. Resistance is a risk factor not only for diabetes but also for dyslipidemia (1, 2), hypertension (3), cardiovascular disease (2, 4), colon cancer (5, 6), and possibly breast cancer (7). Individuals with functioning ß-cells respond to insulin resistance with elevated insulin secretion; whether resistance per se or compensatory hyperinsulinemia accounts for the pathogenesis of various chronic conditions remains under study (8).
Because of the central importance of insulin resistance in chronic disease, a monumental effort has been put forth over the last three decades to understand its mechanisms, resulting in the well-described insulin signaling pathways as well as glucose transporter mobilization (9). It has been assumed that defects in insulin action in vivo can be understood by identifying the specific step or steps in the signaling/mobilization cascade that are blocked in normally insulin-sensitive tissues, including skeletal muscle, adipose tissue, and liver. Particular progress has been made recently regarding the insulin resistance due to elevated free fatty acid concentration. Yu et al. (10) have suggested that free fatty acids increased intracellular fatty acyl-coenzyme A. The latter is proposed to activate protein kinase C- and phosphorylation of insulin receptor substrate 1. The resultant decrease in insulin receptor substrate 1 phosphorylation reduces activity of phosphatidylinositol 3-kinase activity and insulin signaling.
Focus on insulin resistance has emerged naturally from cellular studies. There are, however, important contrasts between insulin action in vitro and in the intact organism. Once secreted, insulin must traverse the convoluted pathway from the pancreatic islets to receptors on insulin-sensitive tissues. Due to a highly fenestrated endothelium, access to hepatocytes in vivo is virtually immediate. However, significant barriers exist to insulin action on skeletal muscle as well as adipose, the most abundant insulin-sensitive tissues. Those barriers relate to insulin access to receptors as well as access of glucose and other blood-borne nutrients to the consuming tissues. Glucose access to tissue metabolism can be limited by blood flow distribution pattern, if the rate of transport of glucose across the capillary endothelium is rapid. Because of its size, insulin access to receptors may be limited by transendothelial transport, which can be rate-limiting for stimulation of glucose disposal (11). Thus, in vivo, the overall effect of insulin can depend upon access of hormones and substrates to skeletal muscle cells, as well as the postreceptor events that follow insulin binding. The important but still unanswered question is the extent to which hemodynamic mechanisms may contribute to insulin resistance in pathological states.
Baron and his colleagues (12, 13, 14) pioneered the concept that insulin acts as a vasodilator and can thereby control access of glucose as well as insulin to skeletal muscle and fat. They demonstrated that the insulin effect is nitric oxide-dependent and the vasodilatory effect of insulin was reduced in insulin-resistant obese subjects, or those with diabetes. However, hemodynamic effects of insulin were most prominent at supraphysiological concentrations of the hormone, and skepticism remained regarding the importance of blood flow reductions per se mediating insulin resistance in vivo. An important advance was introduced by Clark and colleagues (15, 16). These authors suggested that insulin might control the bulk flow through tissues which consume glucosenutritive vs. nonnutritive pathways. Thus, the possibility was introduced that total flow might change little, but shifts could occur, increasing access of insulin and substrates to tissues that are metabolic consumers, as opposed to flow pathways that bypass insulin-sensitive glucose-consuming tissues. The same group has exploited the endothelium-dependent reaction whereby 1-methylxanthine is converted to 1-methylurate, a reaction that is catalyzed by the enzyme xanthine oxidase, which is localized in the capillary endothelium. Also, they exploited an approach in which contrast-enhanced ultrasound is used to detect labeled albumin containing microbubbles as a noninvasive method to identify perfusion through insulin-sensitive muscle tissue, and they have confirmed that supraphysiological insulin levels increase perfusion through capillaries in human volunteers (17, 18). Thus, it is clear that insulin can regulate the distribution of flow through insulin-sensitive tissues, and changes in distribution pattern could conceivably contribute to insulin resistance.
An alternative approach to examining the effect of insulin on blood flow distribution is to exploit access to interstitial fluid. Renkin (19) originally defined the relationship between flux of solute from blood to tissue. In the present issue, Gudbjörnsdóttir et al. (20) have exploited the Renkin analysis to examine the effects of hyperinsulinemia on what fraction of the glucose or insulin that enters the tissue exits radiallypresumably to interstitial fluid bathing insulin-sensitive cellsand what fraction passes through to the venous blood. The calculation of this fraction required measurements of blood flow through the deep muscles of the forearm, which they measured with plethysmography, as well as measurements of interstitial fluid concentrations of glucose and insulin. What is impressive is that these measurements were made in conscious human volunteers during several protocols, including the oral glucose tolerance test and the hyperinsulinemic clamp. The authors calculate that the so called permeability fraction, a measure of the fraction of substance entering the tissue to exit radially through the endothelium, increased substantially during the oral glucose tolerance test. They reported a very low permeability fraction under basal conditions, but a 10-fold increase during a one-dose hyperinsulinemic clamp. At elevated insulin, the apparent extraction fraction was not further increased when insulin was further increased, suggesting a saturation of the insulin effect at pharmacological levels. Thus, the studies of Gudbjörnsdóttir are important in that they suggest that changes in blood flow distribution stimulated by hyperinsulinemia appear to mirror changes in glucose utilization (20). This suggests that insulin-induced alteration in blood flow patterns could be as important as direct signaling of cells by insulin in establishing the rate of glucose utilization in vivo.
One challenge in applying the use of the Renkin analysis to measure permeability fraction is the need to estimate interstitial concentrations of small and large molecules. The calculations are very sensitive to the steady-state interstitial levels. The Gothenburg group of Lonnroth and colleagues that published the study in JCEM have used the microdialysis technique, as have several other groups working in man (21, 22). Briefly, fluid is pumped at low flow rates into and simultaneously out of a demarcated segment of muscle tissue, and the rates of equilibration of solutes of interest (e.g. glucose, lactate) or proteins of interest (insulin) between inflow and outflow are compared. Extrapolation is used to estimate the interstitial concentration (where net flux of solute across the dialysis membrane is zero). These methods are important because they allow for estimation of interstitial concentration in manthey are problematic in that interstitial fluid levels are measured by extrapolation and confirmation is not usually available. The method is limited because it is difficult to measure time-dependent changes in solutes or proteins because equilibration of concentrations across the dialysis membrane is approached slowly. For example, in the studies of Gudbjörnsdóttir et al., measurements of insulin transport could not be made at low insulin levels because equilibration was not approached.
An alternative approach to measure interstitial levels of solute is to sample lymph. The latter approach, which is amenable to studies in experimental animals, does not require extrapolation and can be used to make dynamic measurements. Unfortunately, it is difficult to sample lymph at frequent intervals in man, although it has been done (23). It will be of interest to compare interstitial levels measured with the microdialysis technique with lymph-based measurements as an alternative confirmation and application of the Renkin approach to assess permeability fraction in vivo.
It is important to remember that shifts in blood flow distribution within a limb not only will alter the availability of nutrient to the tissues, but also of insulin itself. Thus, changes in the time course and effectiveness of insulin to stimulate glucose uptake could be limited by availability of insulin at the level of the receptor. Relevant to this, in our laboratory Ellmerer et al. (24) have recently demonstrated an effect of insulin per se on the volume of distribution of insulin in the tissues, a result that would be consistent with the present demonstration of insulin stimulation of distribution of blood flow from nonnutritive to nutritive tissues.
The overwhelming majority of studies of insulin resistance have been done on isolated cells or tissues, and our perception of insulin resistance is that it is related to defects in the insulin signaling pathway. But, insulin resistance is closely tied to changes in vascular function. The effect of insulin to stimulate glucose utilization is remarkably slow. Note that in normal man it takes 6 h to achieve a steady-state rate of glucose uptake when insulin is elevated even to physiologically normal levels (Hamilton-Wessler, M., personal communication). Although there may well be effects of insulin on the signaling pathway as well as transporter mobilization that proceeds at that rate, it appears equally likely that there are rapid and/or slower changes in blood flow distribution that affect the access of insulin to sensitive tissues. It is clear that not all of the events that are responsible for the painfully slow increase in glucose uptake which occurs under hyperinsulinemic conditions have been elucidated. The paper in the present issue of JCEM by Gudbjörnsdóttir and colleagues is an important contribution to elucidate the role of the vascular system in the overall effect of insulin. The putative role in the pathogenesis of insulin resistance requires further study.
Footnotes
This work was supported by National Institutes of Health Grants DK27619 and DK29867 (to R.N.B.).
Received August 18, 2003.
Accepted August 18, 2003.
References
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |