Vascular Disease in Pre-Diabetes: New Insights Derived from Systems Biology

Abstract

In many cases vascular disease is present before the clinical onset of type 2 diabetes, that is, during the pre-diabetic period when insulin levels are markedly increased. In pre-diabetes, microvascular dysfunction correlates with plasma insulin levels and not blood glucose. Here we discuss the concept that insulin, at levels found in pre-diabetes, contributes to microvascular disease in skeletal muscle by inhibiting the release of the vasodilator, adenosine triphosphate (ATP), from erythrocytes.

Introduction

Pre-diabetes, a precursor to the development of type 2 diabetes, has been reported to be present in 25% of overweight adults aged 45–75 (12 million persons in the U.S.). The majority of these individuals can be anticipated to develop type 2 diabetes within 10 years¹. The American Diabetes Association defines pre-diabetes as the period during which blood glucose levels are higher than normal but not yet high enough to be classified as diabetes. The diagnosis of pre-diabetes is made by either measuring fasting plasma glucose or based on results of an oral glucose tolerance test (OGTT). A fasting blood sugar of 100 to 126 mg/dl or a two hour blood glucose of 140 to 200 mg/dl during an OGTT is defined as pre-diabetes. Recently, the measurement of the percentage of glycated hemoglobin (HbA1c) has been added to the diagnostic criteria for the diagnosis of type 2 diabetes². In pre-diabetes the HbA1c is typically 5.7 to 6.5%³. Although not a diagnostic criteria, another characteristic of pre-diabetes is the presence of markedly elevated plasma insulin that is associated with peripheral insulin resistance^4,5.

Recently, several studies have suggested that the vascular disease associated with type 2 diabetes may have its origins during this prediabetic period since, in many cases, the microvascular complications are present at the time of initial diagnosis of type 2 diabetes^5,6,7,8,9. The physiological mechanisms which contribute to the regulation of microvascular blood flow in skeletal muscle are complex and not as well understood as one might anticipate.

It is well recognized that, within skeletal muscle, blood flow (O₂ supply) must be regulated to precisely meet the metabolic requirements of the tissue (O₂ utilization). Although several mechanisms that could mediate this fundamental physiological response have been proposed and evaluated experimentally, none fully explains the observed sensitivity of the vasculature to changes in oxygen requirements within the physiological range¹⁰. If O₂ supply is to be adjusted to satisfy metabolic need in skeletal muscle, there must be a mechanism to sense and quantify the requirement of the tissue for O₂ and a means to communicate that information to the vascular smooth muscle responsible for altering vessel caliber and thus tissue perfusion. The vascular endothelium controls vascular caliber and coordinates the response of localized yet diverse stimuli initiated within the tissue. Thus, it is reasonable to suggest the one or more components of the oxygen transport pathway responsible for sensing oxygen need interacts with the endothelium resulting in the appropriate alterations in microvascular perfusion.

The Role of Oxygen

Oxygen is supplied to the tissue by diffusion from hemoglobin contained within the anucleated erythrocyte. Thus, the O₂ content of the erythrocyte falls as it traverses a tissue with the extent of the decrease directly linked to O₂ utilization by the tissue. Therefore, this mobile cell would be well positioned to function as a sensor of O₂ demand¹⁰. It has been demonstrated that when erythrocytes of most species, including humans, are exposed to reduced O₂ tension in the physiological range, they release both O₂ and adenosine tri-phosphate (ATP)¹¹, a potent vasodilator. ATP released from erythrocytes would be expected to bind to receptors on the endothelium of the blood vessel initiating the synthesis and release of vasodilators including nitric oxide, prostacyclin and/or other products of arachidonic acid metabolism, depending on the vessel in question¹⁰. Importantly, ATP instilled into in the lumen of arterioles and venules has been shown to induce a vasodilation which is conducted along the vasculature increasing blood flow to a specific tissue region^11,12. Thus, the concomitant release of O₂ and ATP by the erythrocyte would provide a mechanism by which oxygen need could be sensed and blood flow increased allowing for the precise distribution of blood flow to meet tissue oxygen requirements.

To demonstrate more clearly that erythrocytes can contribute to the local vasodilation that occurs in areas of low oxygen tension, we isolated arterioles (~ 50 μm diameter) from the skeletal muscle of hamsters, placed them in a buffer containing chamber on an inverted microscope and perfused these resistance vessels with either buffer with a normal oxygen tension (~ 135 mm Hg) or the same buffer containing erythrocytes. In the absence of erythrocytes, when the O₂ tension on the extra-luminal side of these isolated arterioles was reduced to ~20 mmHg, we observed no significant change in vascular caliber. However, when these same vessels were subsequently perfused with buffer containing human erythrocytes, the same decrease in extra-luminal O₂ tension resulted in a significant increase in vessel diameter¹³. Since the buffer or buffer containing erythrocytes entering the blood vessel had a normal O₂ tension this experimental approach reflects the effects of decreases in O₂ outside of the blood vessel on vascular caliber, mimicking what occurs in tissues as the erythrocyte enters regions of tissue with increased O₂ demand. If we accept that this mechanism of blood flow control contributes to normal matching of oxygen supply with need in skeletal muscle, then it becomes of interest to determine if such a mechanism is altered in pre-diabetes, a condition in which the control of blood flow in that tissue has been reported to be compromised.

Although microvascular dysfunction in pre-diabetes has not been found to correlate with plasma glucose, age, body mass index, serum lipids or blood pressure, a direct correlation with plasma insulin has been reported ^6,8. One conclusion is that the elevated insulin somehow contributes to the microvascular disease in pre-diabetes, either via an effect on either the vasculature or the oxygen-carrying erythrocyte.

Increased Insulin Levels

Increased insulin levels have been shown to be associated with impaired vascular function in humans. It was demonstrated that the there was a negative correlation between the impairment of the hyperemic response (increase in blood flow in response to local warming of a limb) and plasma insulin levels⁸ in individuals with pre-diabetes as well as in healthy individuals¹⁴ suggesting that insulin has adverse effects on the circulation although no mechanism has been forthcoming.

It is of interest that, although insulin is not required for glucose entry into human erythrocytes, these cells possess well characterized insulin receptors^15,16 but no role for insulin signaling in that cell has been defined. Is it possible that insulin binding to its receptor on erythrocytes could impact the capacity of these cells to contribute to normal microvascular perfusion in skeletal muscle by interfering with the release of ATP?

Recently, we defined a signaling pathway in human erythrocytes that couples exposure to reduced O₂ tension and ATP release (See Figure 1)^10,17. This pathway includes the heterotrimeric G protein, Gi, and adenylyl cyclase (AC). Increases in cAMP are required for ATP release. The magnitude and duration of increases in cAMP in signaling pathways are precisely regulated by phosphodiesterases (PDEs). We have shown that a specific PDE, PDE3, is associated with the hydrolysis of the cAMP produced when erythrocytes are exposed to reduced O₂ tension¹⁸. This PDE is also present in adipose tissue and its activity is enhanced by insulin¹⁹. Recently, we determined that insulin also activates the PDE3 in human erythrocytes¹⁸ and impairs ATP release in response to exposure of these cells to low O₂ tension¹³. Thus, via this mechanism, the elevated insulin present in pre-diabetes would impair the ability of these cells to stimulate vasodilation of blood vessels in areas of increased O₂ need in skeletal muscle.

Proposed mechanism by which erythrocytes participate in the matching of oxygen (O2) supply with demand in skeletal muscle. When O2 demand in the tissue increases, oxygen is released from erythrocytes. The fall in oxygen saturation (SO2) of hemoglobin results in activation of the heterotrimeric G-protein, Gi. This initiates a signal transduction pathway leading to activation of adenylyl cyclase, increases in 3′5′-adenosine monophosphate (cAMP), activation of protein kinase A (PKA) and the cystic fibrosis transmembrane conductance regulator (CFTR) and, ultimately, adenosine triphosphate (ATP) release. The released ATP binds to purinergic receptors (PR) on the endothelium (Endo) which induces the production of vasodilators that relax smooth muscle (SMC) inducing a vasodilation which is conducted upstream. Insulin inhibits ATP release by binding to receptors on the erythrocyte membrane (IR) activating a phosphodiestease (PDE3) that enhances the breakdown of cAMP (see text for additional detail).

To ascertain if insulin alters the ability of erythrocytes to stimulate dilation of skeletal muscle arterioles exposed to extra-luminal decreases in O₂, we again utilized isolated, perfused arterioles from the hamster¹³. Again, when these vessels were perfused with healthy human erythrocytes, exposure to reduced extraluminal O₂ resulted in significant vasodilation (See Figure 2). However, when the same vessels were perfused with erythrocytes pretreated with insulin (1 nM), vessel diameter did not change in response to a similar reduction in extra-luminal pO₂. This concentration of insulin did not directly interfere with the endothelium-dependent relaxation of the isolated vessels. One interpretation of these results is that insulin prevents ATP release from the erythrocyte when extra-luminal pO₂ is reduced leading to a decreased stimulus for the production of endothelium-derived vasodilators.

Figure 2.

Effect of insulin on the response of isolated arterioles perfused with erythrocytes (RBCs) to reduced oxygen tension. Isolated arterioles were exposed to either extra-luminal normoxia or reduced oxygen tension and perfused with buffer containing

Although it is clear that insulin impairs the ability of erythrocytes to alter the diameter of a single arteriole when extraluminal oxygen tension is reduced, it is important to ascertain if a similar effect of insulin can be demonstrated at the level of an intact skeletal muscle in vivo. To address this issue, we evaluated oxygen transport in the extensor digitorum longus (EDL) muscle in a well established animal model of pre-diabetes, the 7 week old Zucker Diabetic Fatty (ZDF) rat (Charles River). The homozygous dominant of this inbred strain of rat has a shortened leptin receptor protein which interferes with its ability to bind the adipose-derived hormone leptin which reportedly plays a key role in regulating energy intake and energy expenditure. When these animals are fed a high fat diet, they exhibit high insulin and near normal glucose levels at seven weeks and develop type 2 diabetes by 12 weeks, mimicking the progression seen in humans with pre-diabetes (See Figure 3). Using in vivo video microscopy and a functional imaging system, we measured oxygen transport parameters in pre-diabetic ZDF rats and their controls²⁰. We found that both erythrocyte supply rate and the oxygen saturation of erythrocytes in capillaries were significantly decreased in the ZDF pre-diabetic rats suggesting that the oxygen delivery system is impaired in these animals, possibly related to insulin-induced effects on erythrocytes.

Figure 3.

Blood sugar (panel A) and plasma insulin (panel B) in ZDF rats. Control = normal leptin receptor expression at 7 or 12 weeks of age; pre-diabetes = 7 weeks of age with defective leptin receptor expression; diabetes = 12 weeks of age with defective leptin receptor expression. Values are means ± SE, n=11. *; different from respective control (P<0.05). †; different from all other values (P<0.01).

Although in vitro and in vivo studies can reveal important insights and provide data on how individual oxygen regulatory mechanisms operate, this reductionist approach alone cannot establish the significance of those mechanisms in a fully functional microvascular bed. However, a systems biology approach using both in vitro and in vivo experimental data to develop and test a computational model which can then be used to interpret experimental data as well as providing a framework for understanding the regulatory system can markedly enhance our understanding of this complex physiological system. With such a computational model one can not only predict the response of the regulatory system to reduced oxygen supply but likewise explore the impact of an impaired regulatory system on maintaining tissue oxygenation in diseases such as pre-diabetes. A model, developed by Goldman²¹, integrates the geometric complexity of the microvasculature, the unique rheological properties of blood flow in microvascular networks and the coupling of oxygen convection within the microvasculature and O₂ diffusion/consumption within the surrounding 3D tissue volume. Using this model in conjunction with experimental data we were able to confirm the presence of a decrease in mean tissue PO₂ and an increase in PO₂ heterogeneity in pre-diabetic ZDF rats supporting the hypothesis that the elevated insulin levels present could have interfered with a control mechanism for the appropriate delivery of oxygen to meet tissue needs²⁰.

The multi-scale, systems biology approach used in these studies provides important information about the mechanisms that regulate oxygen supply to meet metabolic demand in skeletal muscle. Moreover, the development of a computational model of the circulation provides new insights into the microvascular disease associated with pre-diabetes and provides a framework for exploring possible therapeutic approaches. Although the focus here has been on pre-diabetes, the systems biology approach to the study of the microcirculation is extremely powerful and can provide new approaches to the understanding of those factors that regulate tissue perfusion in both health and disease.

Biography

Randy S. Sprague, MD, and Mary L. Ellsworth, PhD, are Professors in the Departments of Pharmacological and Physiological Science and Internal Medicine at the Saint Louis University School of Medicine.

Contact: spraguer@slu.edu