Reduced Expression of Gi in Erythrocytes of Humans With Type 2 Diabete
http://www.100md.com
《糖尿病学杂志》
the Department of Pharmacological and Physiological Science, Saint Louis University, School of Medicine, St. Louis, Missouri
AC, adenylyl cyclase; IBMX, 3-isobutyl-1-methyl xanthine; PVDF, polyvinylidene difluoride
ABSTRACT
Human erythrocytes, by virtue of their ability to release ATP in response to physiological stimuli, have been proposed to participate in the regulation of local blood flow. A signal transduction pathway that relates these stimuli to ATP release has been described and includes the heterotrimeric G protein Gi and adenylyl cyclase (AC). In this cell, Gi activation results in increases in cAMP and, ultimately, ATP release. It has been reported that Gi expression is decreased in animal models of diabetes and in platelets of humans with type 2 diabetes. Here, we report that Gi2 expression is selectively decreased in erythrocytes of humans with type 2 diabetes and that this defect is associated with reductions in cAMP accumulation and ATP release in response to incubation of erythrocytes with mastoparan 7 (10 μmol/l), an activator of Gi. Importantly, this defect in ATP release correlates inversely with the adequacy of glycemic control as determined by levels of HbA1c (A1C). These results demonstrate that in erythrocytes of humans with type 2 diabetes, both Gi expression and ATP release in response to mastoparan 7 are impaired, which is consistent with the hypothesis that this defect in erythrocyte physiology could contribute to the vascular disease associated with this clinical condition.
Vascular disease associated with altered vascular reactivity is a major complication of diabetes. Several reports have shown that both endothelium-dependent and -independent vasodilation are impaired in this disease (1–6). This reduction in vascular reactivity has been attributed to decreased endothelial nitric oxide (NO) synthesis (3,4), increased NO degradation (2,6), and/or abnormalities in vascular smooth muscle (3). The mechanism notwithstanding, it is clear that NO-mediated vasorelaxation is reduced in humans with diabetes.
Previous studies have reported that erythrocyte physiology is altered in diabetes. For example, the oxidant stress to which erythrocytes are exposed is increased in a high-glucose environment and is associated with increased glycation of erythrocyte proteins (7,8). In addition, erythrocytes of humans with diabetes have been reported to be less deformable than those of healthy humans (9) and to release reduced amounts of ATP in response to osmotic stress, yet these cells were reported to contain increased amounts of intracellular ATP (10).
The erythrocyte, by virtue of its ability to release ATP in response to exposure to reduced oxygen tension (11–13) or to mechanical deformation (14–16), can participate in local control of vascular caliber (17–22). Erythrocyte-derived ATP has been shown to stimulate endogenous endothelial NO synthesis in circulation of the lung (12,23) as well as in striated muscle (17,18). Indeed, it has been proposed that in the vasculature of skeletal muscle, the erythrocyte plays an important role in the matching of oxygen delivery with metabolic need (20–22). A signal transduction pathway that relates both pharmacological and physiological stimuli to ATP release from erythrocytes has been described. Components of this pathway include the heterotrimeric G proteins Gs (24) and Gi (25,26), adenylyl cyclase (AC) (16), cyclic AMP-dependent protein kinase A (16), and the cystic fibrosis transmembrane conductance regulator (11,27).
Altered expression of one of the components of this pathway, Gi, has been reported in animal models of diabetes (28–32) as well as in platelets of humans with type 2 diabetes (33). Importantly, in human erythrocytes, activation of Gi results in stimulation of AC activity, leading to increased cAMP synthesis and release of ATP (25,26), a stimulus for endothelial NO synthesis. Here we investigate the hypothesis that expression of Gi is decreased in erythrocytes of humans with type 2 diabetes and that this defect is associated with reduced cAMP synthesis and ATP release in response to activation of this heterotrimeric G protein. Finally, we examine the relationship between impaired ATP release and the adequacy of glycemic control as determined by measurement of HbA1c (A1C) in humans with type 2 diabetes.
RESEARCH DESIGN AND METHODS
Blood was obtained by venipuncture from healthy humans and humans with type 2 diabetes, collected into heparinized syringes, and centrifuged at 500g at 4°C for 10 min. The plasma and buffy coat were discarded, and erythrocytes were resuspended and washed three times in buffer containing (in mmol/l): 4.7 KCl, 2.0 CaCl2, 140.5 NaCl, 1.2 MgSO4, 21.0 tris(hydroxymethyl)aminomethane, 5.5 glucose, and 0.5% BSA, final pH 7.4. This procedure results in a concentrated preparation of erythrocytes with a hematocrit of 65–70%. Wright stains of erythrocytes prepared in this fashion reveal less than one leukocyte per 50 high power fields. In addition, the platelet antigen CD41 was not detected in membrane preparations of erythrocytes, indicating there was no significant platelet contamination (data not shown). Cells were prepared on the day of use. The protocol for blood removal was approved by the institutional review board of Saint Louis University.
Measurement of ATP and hemoglobin.
ATP was measured by the luciferin-luciferase technique (11,12,14–19), which uses the ATP concentration dependence of light generated by the reaction of ATP with firefly tail extract. Sensitivity was augmented by the addition of synthetic D-luciferin to the crude firefly tail extract. A 200 μl sample of the red blood cell suspension was injected into a cuvette containing 100 μl crude firefly tail extract (10 mg/ml distilled water, FLE 250; Sigma) and 100 μl of a solution of synthetic D-luciferin (50 mg/100 ml distilled water; Sigma). The light emitted was detected using a luminometer (Turner Designs). A standard curve was obtained on the day of each experiment. To exclude the presence of significant hemolysis, free hemoglobin was determined at the time of ATP measurements. Samples were centrifuged at 500g for 10 min at 4°C, and the presence of hemoglobin in the supernatant was determined by light absorption at a wavelength of 405 nm. All data were excluded from experiments in which free hemoglobin increased. Mastoparan 7, at the concentrations used in this study, did not alter the sensitivity of the assay for authentic ATP (data not shown).
Measurement of cAMP.
For determination of cAMP, erythrocytes were diluted with wash buffer (see above) to achieve a hematocrit of 50%. Erythrocytes of healthy humans or humans with type 2 diabetes were incubated with either mastoparan 7 (activator of Gi, 10 μmol/l) (25) or its vehicle (PBS). Then, 1 ml of erythrocyte suspension was added to 4 ml of ice-cold absolute ethanol containing HCl (1 mmol/l), and the mixture was centrifuged at 14,000g for 10 min at 4°C. The supernatant was removed and stored overnight at –20°C to precipitate remaining proteins. Samples were then centrifuged a second time at 3,700g for 10 min at 4°C. The supernatant was removed and dried under vacuum centrifugation. Concentrations of cAMP were then determined with a cAMP Biotrak enzyme immunoassay system (Amersham Biosciences). Erythrocyte counts were obtained from blood at a hematocrit of 50%, and the amounts of cAMP measured were normalized to an erythrocyte red blood cell count of 1010 cells/ml.
Preparation of erythrocyte membranes.
Washed packed erythrocytes (2 ml, hematocrit 65–70%) were added to 200 ml of hypotonic buffer (5 mmol/l Tris-HCl, 2 mmol/l EDTA, with pH adjusted to 7.4) and stirred vigorously for 20 min. The mixture was centrifuged at 23,300g for 15 min. The supernatant was discarded and the membranes resuspended in hypotonic buffer. The mixture was centrifuged at 23,300g for 15 min and the supernatant discarded. The membranes were pooled, resuspended in buffer, and centrifuged again at 23,300g for 15 min. The protein concentration was determined with a bicinchoninic acid protein assay (Pierce). Membranes were aliquoted and stored at –80°C.
Identification of the subunit of heterotrimeric G proteins and ACII in erythrocyte membranes.
Erythrocyte membranes were solubilized in sample buffer (0.277 mol/l SDS, 60% glycerol, 0.4 mol/l dithiothreitol, 0.25 mol/l Tris HCl, and 0.004% bromophenol blue), heated (5 min, 100°C), and loaded onto a precast 4–20% gradient or 5% Tris-HCl Ready Gel (Bio-Rad). Gels were subjected to electrophoresis at 150 V for 90 min (4–20% gel) or 45 min (5% gel) with buffer containing 25 mmol/l Tris, 192 mmol/l glycine, and 0.1% wt/vol SDS, pH 8.3. After electrophoresis, proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane with transfer buffer (25 mmol/l Tris and 192 mmol/l glycine with 20% vol/vol methanol at pH 8.3) at 100 V for 1 h on ice. PVDF membranes were blocked with 5% nonfat dry milk in PBS (10 mmol/l sodium phosphate, pH 7.4, 150 mmol/l NaCl containing 0.1% Tween-20) and then incubated with antibodies directed against either the subunit of Gs (rabbit polyclonal, 1:1,000 dilution; Biomol), Gi1 (mouse monoclonal, 1:1,000 dilution; Exalpha), Gi2 (mouse monoclonal, 1:200 dilution; Biomol), or Gi3 (rabbit polyclonal, 1:1,000 dilution; Biomol). It was determined previously that the antibodies to the various subunits of the G proteins studied were both selective and specific (24,25). The presence of ACII was determined by incubation of membranes with antibodies directed against either an internal epitope of that protein (H-100, 1:600 dilution; Santa Cruz) or its COOH terminus (C-20, 1:600 dilution; Santa Cruz) (35). On separate gels, 5 μg of the same membrane preparations were treated in an identical manner but were incubated with antibody directed against -actin (mouse monoclonal, 1:5,000 dilution; Sigma) as a loading control. PVDF membranes were then incubated (1 h, 25°C) with either donkey anti-rabbit or sheep anti–mouse IgG linked to horseradish peroxidase (1:5,000 and 1:600 dilution for antibodies to G proteins/-actin and ACII, respectively; Amersham Biosciences) as secondary antibodies in 1% nonfat dry milk in PBS containing 0.1% Tween 20. After labeling with the secondary antibody, PVDF membranes were exposed to enhanced chemiluminescence using enhanced chemiluminescence (Amersham Biosciences).
Determination of relative amounts of G protein subunits and ACII present in erythrocyte membranes.
Amounts of membrane protein loaded on the gels were standardized by determination of the protein content of each preparation before loading and by determination of the amount of -actin present in each sample. Amounts of G protein or AC in individual membrane preparations were calculated as the ratio of the protein of interest to -actin, as determined by densitometry. Values for G proteins of healthy humans and humans with type 2 diabetes are expressed as the percentage of that G protein detected in the membrane of a healthy human run on the same gel. Values for ACII are expressed as the ratio of its density to that of -actin. All studies were performed in duplicate.
Incubation of erythrocytes with mastoparan 7.
Erythrocytes were diluted with wash buffer (see above) to achieve a hematocrit of 20%. Erythrocytes of healthy humans or humans with type 2 diabetes were incubated with either mastoparan 7 (activator of Gi, 10 μmol/l) (35) or its vehicle (PBS). Then, 1 ml of 20% erythrocytes was diluted 1:10 with wash buffer to achieve a hematocrit of 2%. The number of erythrocytes was then determined and, after an additional 1:50 dilution, ATP was measured. The amount of ATP released was corrected for dilution and normalized to a cell count of 4 x 108 erythrocytes/ml (2% hematocrit). ATP release was determined before (baseline) and at 5, 10, and 15 min after the addition of mastoparan 7 or its vehicle, PBS.
Statistical methods.
Statistical significance between experimental periods was determined with ANOVA. In the event that the F ratio indicated that changes had occurred, a Fischer’s least significant differences–protected Student’s t test was used to identify individual differences. Correlation between A1C levels and ATP release was determined by regression analysis. P 0.05 was considered to be statistically significant. Results are reported as the means ± SE.
RESULTS
Characteristics of subjects with type 2 diabetes.
Patients with type 2 diabetes were identified by physicians in the Endocrinology Clinic at Saint Louis University. A history form was completed for each individual that included a detailed listing of all medications, the subject’s age, as well as the most recent A1C level (values within 4 weeks of blood removal). The mean ages of healthy humans and humans with type 2 diabetes studied was 37 ± 5 (n = 11, range 24–63) and 51 ± 4 (n = 13, range 28–75), respectively. Patients with type 2 diabetes were treated with insulin (n = 6), lipid-lowering agents (n = 8), oral hypoglycemic agents (n = 10), low-dose aspirin (n = 9), diuretics (n = 5), -receptor antagonists (n = 7), ACE inhibitors or receptor antagonists (n = 9), oral nitrates (n = 5), and calcium channel blockers (n = 4). The mean A1C level for all subjects with type 2 diabetes was 8.7 ± 0.7 (n = 19, 5.5–17.1). All record keeping was in strict compliance with HIPAA (Health Insurance Portability and Accountability Act) regulations.
Expression of Gi in erythrocyte membranes.
Amounts of heterotrimeric G proteins, Gs, Gi1, Gi2, and Gi3, present in erythrocyte membranes of healthy humans and humans with type 2 diabetes were determined by Western analysis, quantified by densitometric scanning, and corrected for the amount of -actin in each sample. Among the G proteins measured, only Gi2 was significantly decreased in humans with type 2 diabetes (n = 9) when compared with amounts found in the erythrocyte membranes of healthy humans (n = 8) (Fig. 1).
Effect of an activator of Gi (mastoparan 7) on ATP release from erythrocytes of healthy humans and humans with type 2 diabetes.
Baseline (unstimulated) ATP release was not different between groups (Fig. 2). Incubation with mastoparan 7 (10 μmol/l) resulted in ATP release from erythrocytes of healthy humans (n = 8) as well as from those of humans with type 2 diabetes (n = 9) (Fig. 2). However, amounts of ATP released from erythrocytes of humans with type 2 diabetes were lower than amounts released by erythrocytes of healthy humans (Fig. 2). Moreover, maximum ATP release occurred at 7.5 ± 1.3 min in healthy humans and was delayed to 11.7 ± 1.3 min in humans with type 2 diabetes (P < 0.05).
Relationship between A1C and mastoparan 7–induced ATP release from erythrocytes of humans with type 2 diabetes.
Although ATP release from erythrocytes of humans with type 2 diabetes was decreased compared with that of healthy humans, some ATP release was detected. To determine whether the amount of ATP released correlated with glycemic control, we determined the relationship between A1C level and ATP release from erythrocytes of humans with type 2 diabetes. A1C has been reported to serve as a reliable marker of the adequacy of glycemic control, such that the greater the A1C, the poorer the glycemic control (34,36–39). As depicted in Fig. 3, there is a significant linear correlation between A1C and mastoparan 7–induced ATP release from erythrocytes (n = 9, P < 0.01).
Expression of ACII in erythrocyte membranes.
The presence of ACII in erythrocyte membranes of healthy humans and humans with type 2 diabetes was determined by Western analysis. Proteins were quantified by densitometric scanning and corrected for the amount of -actin in each sample. Using two antibodies directed against two distinct epitopes, it was determined that ACII is a component of the membranes of human erythrocytes. There was no difference in the levels of expression of ACII between healthy humans (n = 4) and humans with type 2 diabetes (n = 4) (Fig. 4). It is important to note that the molecular weight of ACII, based on analysis of amino acid composition, would be predicted to be 120 kDa. However, it has been reported that in membranes from several cell types, this protein may be found with an apparent molecular weight in the range of 200–250 kDa (40–42). These higher–molecular weight complexes have been suggested to result from dimerization of AC, association with other proteins in a signaling cascade, or glycosylation (35). We report that the ACII found in human erythrocytes, using two distinct antibodies, has an apparent molecular weight of 200 kDa (Fig. 4). This is consistent with reports demonstrating that ACII is found with a similar apparent molecular weight in human myometrium (42) and in rabbit erythrocytes (35).
Effect of mastoparan 7 on accumulation of cAMP in erythrocytes of healthy humans and humans with type 2 diabetes.
It was previously shown that AC activity is required for ATP release from human erythrocytes (16). Importantly, in the erythrocyte, activation of Gi has been shown to stimulate cAMP accumulation (25,26), presumably via the activity of the associated -subunit (43–46). If ATP release is decreased in erythrocytes of humans with type 2 diabetes in response to activation of Gi with mastoparan 7, then it would be anticipated that cAMP accumulation would be decreased as well.
Incubation of erythrocytes of healthy humans (n = 7) with mastoparan 7 resulted in an increase in cAMP (Fig. 5). In contrast, there was a small but not significant increase in cAMP in erythrocytes of humans with type 2 diabetes in response to mastoparan 7 (n = 7). Previously, it was reported that basal cAMP levels were reduced in vascular smooth muscle cells exposed to high glucose concentrations (30). Here, we found no decrease in baseline cAMP levels in erythrocytes of humans with type 2 diabetes compared with cells of healthy humans. To address this further, in separate studies, we incubated erythrocytes with 3-isobutyl-1-methyl xanthine (IBMX; 10 μmol/l, dissolved in ethanol and diluted with saline) to inhibit the degradation of cAMP. Under these conditions, cAMP levels were 2.9 ± 0.8 and 1.9 ± 0.2 pmol per 1010 erythrocytes for cells of healthy humans (n = 5) and humans with type 2 diabetes (n = 7), respectively. Although the cAMP value in the type 2 diabetes group was smaller than that in healthy subjects, it was not significantly different from that in erythrocytes of healthy individuals.
DISCUSSION
In the work presented here, we demonstrate that Gi2 expression is decreased in erythrocytes of humans with type 2 diabetes (Fig. 1). Under the proposed signal-transduction pathway depicted in Fig. 6, activation of erythrocyte Gi would be anticipated to result in cAMP accumulation and release of ATP from human erythrocytes. We found that in erythrocytes of humans with type 2 diabetes, in association with reduced Gi2 expression, cAMP accumulation and ATP release are reduced in response to incubation of these cells with an agent, mastoparan 7, that activates Gi (Figs. 2 and 4). Our previous finding that Gi activation is a critical step for ATP release from human erythrocytes in response to exposure to reduced oxygen tension or mechanical deformation (25,26) suggests that a defect in ATP release from erythrocytes of humans with type 2 diabetes could impair the ability of that cell to participate in the matching of microvascular blood flow with metabolic need (20–22).
The finding that Gi2 expression is decreased in erythrocytes of humans with type 2 diabetes is consistent with reports of reduced expression of heterotrimeric G proteins of the Gi subclass in animal models of diabetes (28–32). An important role for Gi2 in the pathophysiology of diabetes was suggested by reports that mice deficient in Gi2 demonstrate insulin resistance, hyperinsulinemia, and impaired glucose tolerance (31) and that overexpression of Gi2 improves glucose tolerance in mice with streptozotocin-induced diabetes (33). The association between reduced Gi expression and diabetes is not limited to animal models. It was shown that expression of both Gi2 and Gi3 is reduced in platelets of humans with type 2 diabetes (24). Here we report that in erythrocytes of humans with type 2 diabetes, there is a selective decrease in Gi2 expression in the absence of changes in the expression Gs or other Gi subtypes (Fig. 1).
Decreased expression of Gi in erythrocyte membranes of humans with type 2 diabetes would be anticipated to have functional consequences. In the human erythrocyte, activation of Gi is required to demonstrate release of ATP in response to exposure to reduced oxygen tension and mechanical deformation, as demonstrated by studies in which pertussis toxin, an inhibitor of Gi, prevented ATP release from erythrocytes in response to mechanical deformation (25), exposure to reduced oxygen tension (26), as well as incubation with mastoparan 7, a direct activator of Gi (25).
Erythrocyte-derived ATP has been suggested to be an important determinant of microvascular perfusion. In isolated rat cerebral arterioles, a reduction in oxygen tension resulted in ATP release from the erythrocytes, accompanied by an increase in vascular caliber (12). This response was not seen when the vessels were perfused with salt solution in the absence of erythrocytes. In addition, ATP infused into arterioles and venules of the microcirculation of striated muscle of hamsters resulted in vasodilation and an increase in the erythrocyte supply rate (17,18). Importantly, application of ATP to the venular side of that microcirculation resulted in arteriolar vasodilation, i.e., the response was conducted upstream (18). It was shown that these effects of ATP were mediated, in large part, via the stimulation of NO synthesis (17,18). Taken together, these results provide strong support for the hypothesis that erythrocyte-derived ATP, released in the microcirculation, participates in the local control of vascular caliber via the stimulation of endogenous NO synthesis. Here, we show that mastoparan 7–induced ATP release from erythrocytes of humans with type 2 diabetes is reduced compared with erythrocytes of healthy humans (Fig. 2). This defect in ATP release in response to activation of Gi, which is required for ATP release in response to reduced oxygen tension, could limit the ability of the erythrocyte to participate in the regulation of microvascular blood flow distribution in humans with type 2 diabetes.
A1C, a measure of long-term glycemic control in diabetes, has been shown to correlate with diabetic complications, such that higher A1C levels are associated with increased number and severity of complications (34,36–39). We have determined that the degree of impairment of mastoparan 7–induced ATP release from erythrocytes of humans with type 2 diabetes correlates inversely with the adequacy of glycemic control, as determined by the concentration of A1C (Fig. 3). Although only correlative, one interpretation of this association is that the smaller ATP release associated with poorer glycemic control could contribute to the increased incidence of vascular complications in type 2 diabetes.
When heterotrimeric G proteins are activated, the subunit dissociates from the -complex. The subunit and the -complex can then regulate, either individually or synergistically, the catalytic activity of AC. The subunit of Gs activates all isoforms of AC (44–46). In addition to the subunit, it is now recognized that the -subunit of Gi is capable of activating at least three isoforms of AC (II, IV, and VII) (44–46). This ability has been shown to reside with the -component of the dimer, of which five types have been defined (42–44). Types 1, 2, 3, and 4 have been shown to activate AC (46). We determined previously that these -subunits are present in human erythrocyte membranes (35). Here, we report that erythrocytes of both healthy humans and humans with type 2 diabetes possess comparable amounts of ACII (Fig. 4), suggesting that the failure of mastoparan 7 to stimulate increases in cAMP in erythrocytes of humans with type 2 diabetes is not explained by a simple decrease in the amount of this AC isoform present. However, it is possible that ACII activity is reduced in erythrocytes of humans with type 2 diabetes. In addition, although poorly characterized, human erythrocytes do possess phosphodiesterase activity (47,48). Thus, it is also possible that the failure to demonstrate increases in cAMP in erythrocytes of humans with type 2 diabetes is related to increased phosphodiestrase activity. Although the results of the current study do not permit resolution of these issues, it is important to note that incubation of erythrocytes of humans with type 2 diabetes with the nonselective phosphodiesterase inhibitor IBMX did not result in cAMP levels that differed from those of healthy humans, suggesting the basal phosphodiesterase activity is not greatly different.
In summary, the finding that Gi is a necessary component of a signal transduction pathway that relates physiological stimuli to ATP release from erythrocytes, coupled with the finding that expression of Gi2 is reduced in the erythrocyte membranes of humans with type 2 diabetes, suggests that this defect in erythrocyte physiology could lead to a reduced stimulus for endogenous NO synthesis in the microvasculature. This hypothesis is supported by the findings that incubation of erythrocytes of humans with type 2 diabetes with an activator of Gi did not result in increased cAMP synthesis and that stimulated ATP release was greatly reduced compared with erythrocytes of healthy humans. Importantly, the impairment in ATP release correlates in an inverse fashion with the adequacy of gylcemic control. Taken together, the results presented here are consistent with the hypothesis that abnormal erythrocyte ATP release contributes to vascular disease in humans with type 2 diabetes.
ACKNOWLEDGMENTS
This work was supported by National Heart Lung and Blood Institute Grant HL-64180 and by a research grant from the American Diabetes Association.
The authors thank J.L. Sprague for inspiration.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
McVeigh GE, Brennan GM, Johnston GD, McDermott BJ, McGrath LT, Henry WR, Andrews JW, Hayes JR: Impaired endothelium-dependent and independent vasodilation in patients with type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 35:771–776, 1992
Williams SB, Cusco JA, Roddy MA, Johnstone MT, Creager MA: Impaired nitric oxide-mediated vasodilation in patients with non-insulin-dependent diabetes mellitus. J Am Col Cardiol 27:567–574, 1996
Yugar-Toledo JC, Tanus-Santos JE, Sabha M, Sousa MG, Cittadino M, Tácito LHB, Moreno H: Uncontrolled hypertension, uncompensated type II diabetes, and smoking have different patterns of vascular dysfunction. Chest 125:823–830, 2004
van Etten RW, de Koning EJP, Verhaar MC, Gaillard CA, Rabelink TJ: Impaired NO-dependent vasodilation in patients with type II (non-insulin-dependent) diabetes mellitus is restored by acute administration of folate. Diabetologia 45:1004–1010, 2002
Jonathan Goodfellow J, Ramsey MW, Luddington LA, Jones CJH, Coates PA, Dunstan F, Lewis MJ, Owens DR, Andrew H, Henderson AH: Endothelium and inelastic arteries: an early marker of vascular dysfunction in non-insulin dependent diabetes. Br Med J 312:744–745, 1996
Bagi Z, Koller A, Kaley G: Superoxide-NO interaction decreases flow- and agonist-induced dilations of coronary arterioles in type 2 diabetes mellitus. Am J Physiol 285:H1404–H1410, 2003
McCance DR, Dyer DG, Dunn JA, Bailie KE, Thorpe SR, Baynes JW, Lyons TJ: Maillard reaction products and their relation to complications in insulin-dependent diabetes mellitus. J Clin Invest 91:2470–2478, 1993
Brownlee M, Cerami A, Vlassara H: Advanced glycosylation end products in tissues and the biochemical basis of diabetic complications. N Engl J Med 318:1315–1321, 1988
Petruzzi E, Orlando C, Pinzani P, Sestini R, Del Rosso A, Dini G, Tanganelli E, Buggiani A, Pazzagli A: ATP release by osmotic shock and HbA1c in diabetic subjects’ erythrocytes. Metabolism 43:435–440, 1994
McMillan DE, Utterback NG, La Puma J: Reduced erythrocyte deformability in diabetes. Diabetes 27:895–901, 1978
Bergfeld GR, Forrester T: Release of ATP from human erythrocytes in response to a brief period of hypoxia and hypercapnea. Cardiovasc Res 26:40–47, 1992
Dietrich HH, Ellsworth ML, Sprague RS, Dacey RG: Red blood cell regulation of microvascular tone through adenosine triphosphate. Am J Physiol 278:H1294–H1298, 2000
Ellsworth ML, Forrester T, Ellis CG, Dietrich HH: The erythrocyte as a regulator of vascular tone. Am J Physiol 269:H2155–H2161, 1995
Sprague RS, Ellsworth ML, Stephenson AH, Lonigro AJ: ATP: the red blood cell link to NO and local control of the pulmonary circulation. Am J Physiol 271:H2717–H2722, 1996
Sprague RS, Ellsworth ML, Stephenson AH, Kleinhenz ME, LonigroAJ: Deformation-induced ATP release from red blood cells requires cystic fibrosis transmembrane conductance regulator activity. Am J Physiol 275:H1726–H1732, 1998
Sprague RS, Ellsworth ML, Stephenson AH, Lonigro AJ: Participation of cAMP in a signal-transduction pathway relating erythrocyte deformation to ATP release. Am J Physiol 281:C1158–C1164, 2001
McCullough WT, Collins DM, Ellsworth ML: Arteriolar responses to extracellular ATP in striated muscle. Am J Physiol 272:H1886–H1891, 1997
Collins DM, McCullough WT, Ellsworth ML: Conducted vascular responses: communication across the capillary bed. Microvas Res 56:43–53, 1998
Dietrich HH, Kajita Y, Dacey RG: Local and conducted vasomotor responses in isolated rat cerebral arterioles. Am J Physiol 271:H1109–H1116, 1996
Ellsworth ML: The red blood cell as an oxygen sensor: What is the evidence Acta Physiol Scand 168:551–559, 2000
Gonzalez-Alonzo J, Olsen DB, Saltin B: Erythrocyte and the regulation of human skeletal muscle blood flow and oxygen delivery. Circ Res 91:1046–1055, 2002
Ellsworth ML: Red blood cell derived ATP as a regulator of skeletal muscle perfusion. Med Sci Sports Exerc 36:35–41, 2004
Sprague RS, Olearczyk JJ, Spence DM, Stephenson AH, Sprung RW, Lonigro AJ: Extracellular ATP signaling in the rabbit lung: erythrocytes as determinants of vascular resistance. Am J Physiol 285:H693–H700, 2003
Olearczyk JJ, Stephenson AH, Lonigro AJ, Sprague RS: Receptor-mediated activation of the heterotrimeric G protein Gs results in ATP release from erythrocytes. Med Sci Monit 7:669–674, 2001
Olearczyk JJ, Stephenson AH, Lonigro AJ, Sprague RS: Heterotrimeric G protein Gi is involved in a signal transduction pathway for ATP release from erythrocytes. Am J Physiol 286:H940–H945, 2004
Olearczyk JJ, Stephenson AH, Lonigro AJ, Sprague RS: NO inhibits signal transduction pathway for ATP release from erythrocytes via its action on heterotrimeric G protein Gi. Am J Physiol 287:H748–H754, 2004
Liang G, Stephenson AH, Lonigro AJ, Sprague RS: Erythrocytes of humans with cystic fibrosis fail to stimulate nitric oxide synthesis in isolated rabbit lungs. Am J Physiol 288:H1580–H1585, 2004
Hashim S, Liu YY, Wang R, Anand-Srivastava MB: Streptozotocin-induced diabetes impairs G protein linked signal transduction in vascular smooth muscle. Mol Cell Biochem 240:57–63, 2000
Matsuda N, Hattori Y, Gando S, Watanuki S, Kemmotsu O, Kanno M: Differential gene transcription regulation of Gi isoforms and Gs protein expression in diabetic rat hearts. Naunyn Schmiedebergs Arch Pharmacol 361:53–60, 2000
Hashim S, Li Y, Nagakura TS, Anand-Sirvastava MB: Modulation of G protein expression and adenylyl cyclase signaling by high glucose in vascular smooth muscle. Cardiovasc Res 63:709–718, 2004
Moxham CM, Malbon CC: Insulin action impaired by deficiency of the g-protein subunit Gi2. Nature 379:840–844, 1996
Grawer D, Milligan G, Spiegel AM, Unson CG, Houslay MD: Abolition of the expression of inhibitory guanine nucleotide regulatory protein Gi activity in diabetes. Nature 327:229–232, 1987
Livingstone C, McLellan AR, McGregor MA, Wilson A, Connell JMC, Small M, Milligan G, Paterson KR, Houslay MD: Altered G protein expression and adenylate cyclase activity in platelets of non-insulin-dependent diabetes (NIDDM) male subjects. Biochem Biophys Acta 1096:127–133, 1991
Wolffenbuttel BHR, Giordano D, Founds HW, Bucala R: Long-term assessment of glucose control by haemoglobin-AGE measurement. Lancet 347:513–515, 1996
Sprague RS, Bowles EA, Stumpf MS, Rickettes G, Freidman A, Hou WH, Stephenson AH, Lonigro AJ: Rabbit erythrocytes possess adenylyl cyclase II that is activated by the heterotrimeric G proteins Gs and Gi. Pharmacol Rep 57 (Suppl.):222–228, 2005
Sowers JR, Haffner S: Treatment of cardiovascular and renal risk factors in the diabetic hypertensive. Hypertension 40:781–788, 2002
The Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications of insulin-dependent diabetes mellitus. N Engl J Med 329:977–986, 1993
Nathan DM, Singer DE, Hurxthal K, Goodson JD: The clinical information value of the glycosylated hemoglobin assay. N Engl J Med 310:341–346, 1984
Larsen ML, Horder M, Mogensen EF: Effect of long-term monitoring of glycosylated hemoglobin levels in insulin-dependent diabetes mellitus N Engl J Med 323:1021–1025, 1990
Tang WJ, Gilman AG: Adenylyl cyclases. Cell 70:869–872, 1992
Neer EJ, Echeverria D, Knox S; Increase in the size of brain adenylate cyclase with activation by guanosine 5'-(beta, gamma-imino)triphosphate. J Biol Chem 255:9782–9789, 1980
Price SA, Pochun I, Phaneuf S, Bernal AL: Adenylyl cyclase isoforms in pregnant and non-pregnant human myometrium. J Endocrinology 164:21–30, 2000
Sprague RS, Bowles EA, Olearczyk JJ, Stephenson AH, Lonigro AJ: The role of G protein subunits in the release of ATP from human erythrocytes. J Physiol Pharmacol 53:667–674, 2002
Tang W, Gilman AG: Type-specific regulation of adenylyl cyclase by G protein subunits. Science 254:1500–1503, 1991
Federman AD, Conklin BR, Schrader KA, Reed RR, Bourne HR: Hormonal stimulation of adenylyl cyclase through Gi-protein beta gamma subunits. Nature 356:159–161, 1992
Bayewitch ML, Avidor-Reiss T, Levy R, Pfeuffer T, Nevo I, Simonds WF, Vogel Z: Differential modulation of adenylyl cyclase I and II by various G subunits. J Biol Chem 273:2273–2276, 1998
Dreiling CE: Localization of 2',3'-cyclic nucleotide 3'-phosphodiesterase in human erythrocyte membranes. Biochem Biophys Acta 649:587–594, 1981
Petrov V, Fagard R, Lijene P: Human erythrocytes contain Ca2+, calmodulin-dependent cyclic nucleotide phosphodiesterase which is involved in the hydrolysis of cAMP. Meth Find Exp Clin Pharmacol 20:387–393, 1998(Randy S. Sprague, Alan H. Stephenson, El)
AC, adenylyl cyclase; IBMX, 3-isobutyl-1-methyl xanthine; PVDF, polyvinylidene difluoride
ABSTRACT
Human erythrocytes, by virtue of their ability to release ATP in response to physiological stimuli, have been proposed to participate in the regulation of local blood flow. A signal transduction pathway that relates these stimuli to ATP release has been described and includes the heterotrimeric G protein Gi and adenylyl cyclase (AC). In this cell, Gi activation results in increases in cAMP and, ultimately, ATP release. It has been reported that Gi expression is decreased in animal models of diabetes and in platelets of humans with type 2 diabetes. Here, we report that Gi2 expression is selectively decreased in erythrocytes of humans with type 2 diabetes and that this defect is associated with reductions in cAMP accumulation and ATP release in response to incubation of erythrocytes with mastoparan 7 (10 μmol/l), an activator of Gi. Importantly, this defect in ATP release correlates inversely with the adequacy of glycemic control as determined by levels of HbA1c (A1C). These results demonstrate that in erythrocytes of humans with type 2 diabetes, both Gi expression and ATP release in response to mastoparan 7 are impaired, which is consistent with the hypothesis that this defect in erythrocyte physiology could contribute to the vascular disease associated with this clinical condition.
Vascular disease associated with altered vascular reactivity is a major complication of diabetes. Several reports have shown that both endothelium-dependent and -independent vasodilation are impaired in this disease (1–6). This reduction in vascular reactivity has been attributed to decreased endothelial nitric oxide (NO) synthesis (3,4), increased NO degradation (2,6), and/or abnormalities in vascular smooth muscle (3). The mechanism notwithstanding, it is clear that NO-mediated vasorelaxation is reduced in humans with diabetes.
Previous studies have reported that erythrocyte physiology is altered in diabetes. For example, the oxidant stress to which erythrocytes are exposed is increased in a high-glucose environment and is associated with increased glycation of erythrocyte proteins (7,8). In addition, erythrocytes of humans with diabetes have been reported to be less deformable than those of healthy humans (9) and to release reduced amounts of ATP in response to osmotic stress, yet these cells were reported to contain increased amounts of intracellular ATP (10).
The erythrocyte, by virtue of its ability to release ATP in response to exposure to reduced oxygen tension (11–13) or to mechanical deformation (14–16), can participate in local control of vascular caliber (17–22). Erythrocyte-derived ATP has been shown to stimulate endogenous endothelial NO synthesis in circulation of the lung (12,23) as well as in striated muscle (17,18). Indeed, it has been proposed that in the vasculature of skeletal muscle, the erythrocyte plays an important role in the matching of oxygen delivery with metabolic need (20–22). A signal transduction pathway that relates both pharmacological and physiological stimuli to ATP release from erythrocytes has been described. Components of this pathway include the heterotrimeric G proteins Gs (24) and Gi (25,26), adenylyl cyclase (AC) (16), cyclic AMP-dependent protein kinase A (16), and the cystic fibrosis transmembrane conductance regulator (11,27).
Altered expression of one of the components of this pathway, Gi, has been reported in animal models of diabetes (28–32) as well as in platelets of humans with type 2 diabetes (33). Importantly, in human erythrocytes, activation of Gi results in stimulation of AC activity, leading to increased cAMP synthesis and release of ATP (25,26), a stimulus for endothelial NO synthesis. Here we investigate the hypothesis that expression of Gi is decreased in erythrocytes of humans with type 2 diabetes and that this defect is associated with reduced cAMP synthesis and ATP release in response to activation of this heterotrimeric G protein. Finally, we examine the relationship between impaired ATP release and the adequacy of glycemic control as determined by measurement of HbA1c (A1C) in humans with type 2 diabetes.
RESEARCH DESIGN AND METHODS
Blood was obtained by venipuncture from healthy humans and humans with type 2 diabetes, collected into heparinized syringes, and centrifuged at 500g at 4°C for 10 min. The plasma and buffy coat were discarded, and erythrocytes were resuspended and washed three times in buffer containing (in mmol/l): 4.7 KCl, 2.0 CaCl2, 140.5 NaCl, 1.2 MgSO4, 21.0 tris(hydroxymethyl)aminomethane, 5.5 glucose, and 0.5% BSA, final pH 7.4. This procedure results in a concentrated preparation of erythrocytes with a hematocrit of 65–70%. Wright stains of erythrocytes prepared in this fashion reveal less than one leukocyte per 50 high power fields. In addition, the platelet antigen CD41 was not detected in membrane preparations of erythrocytes, indicating there was no significant platelet contamination (data not shown). Cells were prepared on the day of use. The protocol for blood removal was approved by the institutional review board of Saint Louis University.
Measurement of ATP and hemoglobin.
ATP was measured by the luciferin-luciferase technique (11,12,14–19), which uses the ATP concentration dependence of light generated by the reaction of ATP with firefly tail extract. Sensitivity was augmented by the addition of synthetic D-luciferin to the crude firefly tail extract. A 200 μl sample of the red blood cell suspension was injected into a cuvette containing 100 μl crude firefly tail extract (10 mg/ml distilled water, FLE 250; Sigma) and 100 μl of a solution of synthetic D-luciferin (50 mg/100 ml distilled water; Sigma). The light emitted was detected using a luminometer (Turner Designs). A standard curve was obtained on the day of each experiment. To exclude the presence of significant hemolysis, free hemoglobin was determined at the time of ATP measurements. Samples were centrifuged at 500g for 10 min at 4°C, and the presence of hemoglobin in the supernatant was determined by light absorption at a wavelength of 405 nm. All data were excluded from experiments in which free hemoglobin increased. Mastoparan 7, at the concentrations used in this study, did not alter the sensitivity of the assay for authentic ATP (data not shown).
Measurement of cAMP.
For determination of cAMP, erythrocytes were diluted with wash buffer (see above) to achieve a hematocrit of 50%. Erythrocytes of healthy humans or humans with type 2 diabetes were incubated with either mastoparan 7 (activator of Gi, 10 μmol/l) (25) or its vehicle (PBS). Then, 1 ml of erythrocyte suspension was added to 4 ml of ice-cold absolute ethanol containing HCl (1 mmol/l), and the mixture was centrifuged at 14,000g for 10 min at 4°C. The supernatant was removed and stored overnight at –20°C to precipitate remaining proteins. Samples were then centrifuged a second time at 3,700g for 10 min at 4°C. The supernatant was removed and dried under vacuum centrifugation. Concentrations of cAMP were then determined with a cAMP Biotrak enzyme immunoassay system (Amersham Biosciences). Erythrocyte counts were obtained from blood at a hematocrit of 50%, and the amounts of cAMP measured were normalized to an erythrocyte red blood cell count of 1010 cells/ml.
Preparation of erythrocyte membranes.
Washed packed erythrocytes (2 ml, hematocrit 65–70%) were added to 200 ml of hypotonic buffer (5 mmol/l Tris-HCl, 2 mmol/l EDTA, with pH adjusted to 7.4) and stirred vigorously for 20 min. The mixture was centrifuged at 23,300g for 15 min. The supernatant was discarded and the membranes resuspended in hypotonic buffer. The mixture was centrifuged at 23,300g for 15 min and the supernatant discarded. The membranes were pooled, resuspended in buffer, and centrifuged again at 23,300g for 15 min. The protein concentration was determined with a bicinchoninic acid protein assay (Pierce). Membranes were aliquoted and stored at –80°C.
Identification of the subunit of heterotrimeric G proteins and ACII in erythrocyte membranes.
Erythrocyte membranes were solubilized in sample buffer (0.277 mol/l SDS, 60% glycerol, 0.4 mol/l dithiothreitol, 0.25 mol/l Tris HCl, and 0.004% bromophenol blue), heated (5 min, 100°C), and loaded onto a precast 4–20% gradient or 5% Tris-HCl Ready Gel (Bio-Rad). Gels were subjected to electrophoresis at 150 V for 90 min (4–20% gel) or 45 min (5% gel) with buffer containing 25 mmol/l Tris, 192 mmol/l glycine, and 0.1% wt/vol SDS, pH 8.3. After electrophoresis, proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane with transfer buffer (25 mmol/l Tris and 192 mmol/l glycine with 20% vol/vol methanol at pH 8.3) at 100 V for 1 h on ice. PVDF membranes were blocked with 5% nonfat dry milk in PBS (10 mmol/l sodium phosphate, pH 7.4, 150 mmol/l NaCl containing 0.1% Tween-20) and then incubated with antibodies directed against either the subunit of Gs (rabbit polyclonal, 1:1,000 dilution; Biomol), Gi1 (mouse monoclonal, 1:1,000 dilution; Exalpha), Gi2 (mouse monoclonal, 1:200 dilution; Biomol), or Gi3 (rabbit polyclonal, 1:1,000 dilution; Biomol). It was determined previously that the antibodies to the various subunits of the G proteins studied were both selective and specific (24,25). The presence of ACII was determined by incubation of membranes with antibodies directed against either an internal epitope of that protein (H-100, 1:600 dilution; Santa Cruz) or its COOH terminus (C-20, 1:600 dilution; Santa Cruz) (35). On separate gels, 5 μg of the same membrane preparations were treated in an identical manner but were incubated with antibody directed against -actin (mouse monoclonal, 1:5,000 dilution; Sigma) as a loading control. PVDF membranes were then incubated (1 h, 25°C) with either donkey anti-rabbit or sheep anti–mouse IgG linked to horseradish peroxidase (1:5,000 and 1:600 dilution for antibodies to G proteins/-actin and ACII, respectively; Amersham Biosciences) as secondary antibodies in 1% nonfat dry milk in PBS containing 0.1% Tween 20. After labeling with the secondary antibody, PVDF membranes were exposed to enhanced chemiluminescence using enhanced chemiluminescence (Amersham Biosciences).
Determination of relative amounts of G protein subunits and ACII present in erythrocyte membranes.
Amounts of membrane protein loaded on the gels were standardized by determination of the protein content of each preparation before loading and by determination of the amount of -actin present in each sample. Amounts of G protein or AC in individual membrane preparations were calculated as the ratio of the protein of interest to -actin, as determined by densitometry. Values for G proteins of healthy humans and humans with type 2 diabetes are expressed as the percentage of that G protein detected in the membrane of a healthy human run on the same gel. Values for ACII are expressed as the ratio of its density to that of -actin. All studies were performed in duplicate.
Incubation of erythrocytes with mastoparan 7.
Erythrocytes were diluted with wash buffer (see above) to achieve a hematocrit of 20%. Erythrocytes of healthy humans or humans with type 2 diabetes were incubated with either mastoparan 7 (activator of Gi, 10 μmol/l) (35) or its vehicle (PBS). Then, 1 ml of 20% erythrocytes was diluted 1:10 with wash buffer to achieve a hematocrit of 2%. The number of erythrocytes was then determined and, after an additional 1:50 dilution, ATP was measured. The amount of ATP released was corrected for dilution and normalized to a cell count of 4 x 108 erythrocytes/ml (2% hematocrit). ATP release was determined before (baseline) and at 5, 10, and 15 min after the addition of mastoparan 7 or its vehicle, PBS.
Statistical methods.
Statistical significance between experimental periods was determined with ANOVA. In the event that the F ratio indicated that changes had occurred, a Fischer’s least significant differences–protected Student’s t test was used to identify individual differences. Correlation between A1C levels and ATP release was determined by regression analysis. P 0.05 was considered to be statistically significant. Results are reported as the means ± SE.
RESULTS
Characteristics of subjects with type 2 diabetes.
Patients with type 2 diabetes were identified by physicians in the Endocrinology Clinic at Saint Louis University. A history form was completed for each individual that included a detailed listing of all medications, the subject’s age, as well as the most recent A1C level (values within 4 weeks of blood removal). The mean ages of healthy humans and humans with type 2 diabetes studied was 37 ± 5 (n = 11, range 24–63) and 51 ± 4 (n = 13, range 28–75), respectively. Patients with type 2 diabetes were treated with insulin (n = 6), lipid-lowering agents (n = 8), oral hypoglycemic agents (n = 10), low-dose aspirin (n = 9), diuretics (n = 5), -receptor antagonists (n = 7), ACE inhibitors or receptor antagonists (n = 9), oral nitrates (n = 5), and calcium channel blockers (n = 4). The mean A1C level for all subjects with type 2 diabetes was 8.7 ± 0.7 (n = 19, 5.5–17.1). All record keeping was in strict compliance with HIPAA (Health Insurance Portability and Accountability Act) regulations.
Expression of Gi in erythrocyte membranes.
Amounts of heterotrimeric G proteins, Gs, Gi1, Gi2, and Gi3, present in erythrocyte membranes of healthy humans and humans with type 2 diabetes were determined by Western analysis, quantified by densitometric scanning, and corrected for the amount of -actin in each sample. Among the G proteins measured, only Gi2 was significantly decreased in humans with type 2 diabetes (n = 9) when compared with amounts found in the erythrocyte membranes of healthy humans (n = 8) (Fig. 1).
Effect of an activator of Gi (mastoparan 7) on ATP release from erythrocytes of healthy humans and humans with type 2 diabetes.
Baseline (unstimulated) ATP release was not different between groups (Fig. 2). Incubation with mastoparan 7 (10 μmol/l) resulted in ATP release from erythrocytes of healthy humans (n = 8) as well as from those of humans with type 2 diabetes (n = 9) (Fig. 2). However, amounts of ATP released from erythrocytes of humans with type 2 diabetes were lower than amounts released by erythrocytes of healthy humans (Fig. 2). Moreover, maximum ATP release occurred at 7.5 ± 1.3 min in healthy humans and was delayed to 11.7 ± 1.3 min in humans with type 2 diabetes (P < 0.05).
Relationship between A1C and mastoparan 7–induced ATP release from erythrocytes of humans with type 2 diabetes.
Although ATP release from erythrocytes of humans with type 2 diabetes was decreased compared with that of healthy humans, some ATP release was detected. To determine whether the amount of ATP released correlated with glycemic control, we determined the relationship between A1C level and ATP release from erythrocytes of humans with type 2 diabetes. A1C has been reported to serve as a reliable marker of the adequacy of glycemic control, such that the greater the A1C, the poorer the glycemic control (34,36–39). As depicted in Fig. 3, there is a significant linear correlation between A1C and mastoparan 7–induced ATP release from erythrocytes (n = 9, P < 0.01).
Expression of ACII in erythrocyte membranes.
The presence of ACII in erythrocyte membranes of healthy humans and humans with type 2 diabetes was determined by Western analysis. Proteins were quantified by densitometric scanning and corrected for the amount of -actin in each sample. Using two antibodies directed against two distinct epitopes, it was determined that ACII is a component of the membranes of human erythrocytes. There was no difference in the levels of expression of ACII between healthy humans (n = 4) and humans with type 2 diabetes (n = 4) (Fig. 4). It is important to note that the molecular weight of ACII, based on analysis of amino acid composition, would be predicted to be 120 kDa. However, it has been reported that in membranes from several cell types, this protein may be found with an apparent molecular weight in the range of 200–250 kDa (40–42). These higher–molecular weight complexes have been suggested to result from dimerization of AC, association with other proteins in a signaling cascade, or glycosylation (35). We report that the ACII found in human erythrocytes, using two distinct antibodies, has an apparent molecular weight of 200 kDa (Fig. 4). This is consistent with reports demonstrating that ACII is found with a similar apparent molecular weight in human myometrium (42) and in rabbit erythrocytes (35).
Effect of mastoparan 7 on accumulation of cAMP in erythrocytes of healthy humans and humans with type 2 diabetes.
It was previously shown that AC activity is required for ATP release from human erythrocytes (16). Importantly, in the erythrocyte, activation of Gi has been shown to stimulate cAMP accumulation (25,26), presumably via the activity of the associated -subunit (43–46). If ATP release is decreased in erythrocytes of humans with type 2 diabetes in response to activation of Gi with mastoparan 7, then it would be anticipated that cAMP accumulation would be decreased as well.
Incubation of erythrocytes of healthy humans (n = 7) with mastoparan 7 resulted in an increase in cAMP (Fig. 5). In contrast, there was a small but not significant increase in cAMP in erythrocytes of humans with type 2 diabetes in response to mastoparan 7 (n = 7). Previously, it was reported that basal cAMP levels were reduced in vascular smooth muscle cells exposed to high glucose concentrations (30). Here, we found no decrease in baseline cAMP levels in erythrocytes of humans with type 2 diabetes compared with cells of healthy humans. To address this further, in separate studies, we incubated erythrocytes with 3-isobutyl-1-methyl xanthine (IBMX; 10 μmol/l, dissolved in ethanol and diluted with saline) to inhibit the degradation of cAMP. Under these conditions, cAMP levels were 2.9 ± 0.8 and 1.9 ± 0.2 pmol per 1010 erythrocytes for cells of healthy humans (n = 5) and humans with type 2 diabetes (n = 7), respectively. Although the cAMP value in the type 2 diabetes group was smaller than that in healthy subjects, it was not significantly different from that in erythrocytes of healthy individuals.
DISCUSSION
In the work presented here, we demonstrate that Gi2 expression is decreased in erythrocytes of humans with type 2 diabetes (Fig. 1). Under the proposed signal-transduction pathway depicted in Fig. 6, activation of erythrocyte Gi would be anticipated to result in cAMP accumulation and release of ATP from human erythrocytes. We found that in erythrocytes of humans with type 2 diabetes, in association with reduced Gi2 expression, cAMP accumulation and ATP release are reduced in response to incubation of these cells with an agent, mastoparan 7, that activates Gi (Figs. 2 and 4). Our previous finding that Gi activation is a critical step for ATP release from human erythrocytes in response to exposure to reduced oxygen tension or mechanical deformation (25,26) suggests that a defect in ATP release from erythrocytes of humans with type 2 diabetes could impair the ability of that cell to participate in the matching of microvascular blood flow with metabolic need (20–22).
The finding that Gi2 expression is decreased in erythrocytes of humans with type 2 diabetes is consistent with reports of reduced expression of heterotrimeric G proteins of the Gi subclass in animal models of diabetes (28–32). An important role for Gi2 in the pathophysiology of diabetes was suggested by reports that mice deficient in Gi2 demonstrate insulin resistance, hyperinsulinemia, and impaired glucose tolerance (31) and that overexpression of Gi2 improves glucose tolerance in mice with streptozotocin-induced diabetes (33). The association between reduced Gi expression and diabetes is not limited to animal models. It was shown that expression of both Gi2 and Gi3 is reduced in platelets of humans with type 2 diabetes (24). Here we report that in erythrocytes of humans with type 2 diabetes, there is a selective decrease in Gi2 expression in the absence of changes in the expression Gs or other Gi subtypes (Fig. 1).
Decreased expression of Gi in erythrocyte membranes of humans with type 2 diabetes would be anticipated to have functional consequences. In the human erythrocyte, activation of Gi is required to demonstrate release of ATP in response to exposure to reduced oxygen tension and mechanical deformation, as demonstrated by studies in which pertussis toxin, an inhibitor of Gi, prevented ATP release from erythrocytes in response to mechanical deformation (25), exposure to reduced oxygen tension (26), as well as incubation with mastoparan 7, a direct activator of Gi (25).
Erythrocyte-derived ATP has been suggested to be an important determinant of microvascular perfusion. In isolated rat cerebral arterioles, a reduction in oxygen tension resulted in ATP release from the erythrocytes, accompanied by an increase in vascular caliber (12). This response was not seen when the vessels were perfused with salt solution in the absence of erythrocytes. In addition, ATP infused into arterioles and venules of the microcirculation of striated muscle of hamsters resulted in vasodilation and an increase in the erythrocyte supply rate (17,18). Importantly, application of ATP to the venular side of that microcirculation resulted in arteriolar vasodilation, i.e., the response was conducted upstream (18). It was shown that these effects of ATP were mediated, in large part, via the stimulation of NO synthesis (17,18). Taken together, these results provide strong support for the hypothesis that erythrocyte-derived ATP, released in the microcirculation, participates in the local control of vascular caliber via the stimulation of endogenous NO synthesis. Here, we show that mastoparan 7–induced ATP release from erythrocytes of humans with type 2 diabetes is reduced compared with erythrocytes of healthy humans (Fig. 2). This defect in ATP release in response to activation of Gi, which is required for ATP release in response to reduced oxygen tension, could limit the ability of the erythrocyte to participate in the regulation of microvascular blood flow distribution in humans with type 2 diabetes.
A1C, a measure of long-term glycemic control in diabetes, has been shown to correlate with diabetic complications, such that higher A1C levels are associated with increased number and severity of complications (34,36–39). We have determined that the degree of impairment of mastoparan 7–induced ATP release from erythrocytes of humans with type 2 diabetes correlates inversely with the adequacy of glycemic control, as determined by the concentration of A1C (Fig. 3). Although only correlative, one interpretation of this association is that the smaller ATP release associated with poorer glycemic control could contribute to the increased incidence of vascular complications in type 2 diabetes.
When heterotrimeric G proteins are activated, the subunit dissociates from the -complex. The subunit and the -complex can then regulate, either individually or synergistically, the catalytic activity of AC. The subunit of Gs activates all isoforms of AC (44–46). In addition to the subunit, it is now recognized that the -subunit of Gi is capable of activating at least three isoforms of AC (II, IV, and VII) (44–46). This ability has been shown to reside with the -component of the dimer, of which five types have been defined (42–44). Types 1, 2, 3, and 4 have been shown to activate AC (46). We determined previously that these -subunits are present in human erythrocyte membranes (35). Here, we report that erythrocytes of both healthy humans and humans with type 2 diabetes possess comparable amounts of ACII (Fig. 4), suggesting that the failure of mastoparan 7 to stimulate increases in cAMP in erythrocytes of humans with type 2 diabetes is not explained by a simple decrease in the amount of this AC isoform present. However, it is possible that ACII activity is reduced in erythrocytes of humans with type 2 diabetes. In addition, although poorly characterized, human erythrocytes do possess phosphodiesterase activity (47,48). Thus, it is also possible that the failure to demonstrate increases in cAMP in erythrocytes of humans with type 2 diabetes is related to increased phosphodiestrase activity. Although the results of the current study do not permit resolution of these issues, it is important to note that incubation of erythrocytes of humans with type 2 diabetes with the nonselective phosphodiesterase inhibitor IBMX did not result in cAMP levels that differed from those of healthy humans, suggesting the basal phosphodiesterase activity is not greatly different.
In summary, the finding that Gi is a necessary component of a signal transduction pathway that relates physiological stimuli to ATP release from erythrocytes, coupled with the finding that expression of Gi2 is reduced in the erythrocyte membranes of humans with type 2 diabetes, suggests that this defect in erythrocyte physiology could lead to a reduced stimulus for endogenous NO synthesis in the microvasculature. This hypothesis is supported by the findings that incubation of erythrocytes of humans with type 2 diabetes with an activator of Gi did not result in increased cAMP synthesis and that stimulated ATP release was greatly reduced compared with erythrocytes of healthy humans. Importantly, the impairment in ATP release correlates in an inverse fashion with the adequacy of gylcemic control. Taken together, the results presented here are consistent with the hypothesis that abnormal erythrocyte ATP release contributes to vascular disease in humans with type 2 diabetes.
ACKNOWLEDGMENTS
This work was supported by National Heart Lung and Blood Institute Grant HL-64180 and by a research grant from the American Diabetes Association.
The authors thank J.L. Sprague for inspiration.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
McVeigh GE, Brennan GM, Johnston GD, McDermott BJ, McGrath LT, Henry WR, Andrews JW, Hayes JR: Impaired endothelium-dependent and independent vasodilation in patients with type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 35:771–776, 1992
Williams SB, Cusco JA, Roddy MA, Johnstone MT, Creager MA: Impaired nitric oxide-mediated vasodilation in patients with non-insulin-dependent diabetes mellitus. J Am Col Cardiol 27:567–574, 1996
Yugar-Toledo JC, Tanus-Santos JE, Sabha M, Sousa MG, Cittadino M, Tácito LHB, Moreno H: Uncontrolled hypertension, uncompensated type II diabetes, and smoking have different patterns of vascular dysfunction. Chest 125:823–830, 2004
van Etten RW, de Koning EJP, Verhaar MC, Gaillard CA, Rabelink TJ: Impaired NO-dependent vasodilation in patients with type II (non-insulin-dependent) diabetes mellitus is restored by acute administration of folate. Diabetologia 45:1004–1010, 2002
Jonathan Goodfellow J, Ramsey MW, Luddington LA, Jones CJH, Coates PA, Dunstan F, Lewis MJ, Owens DR, Andrew H, Henderson AH: Endothelium and inelastic arteries: an early marker of vascular dysfunction in non-insulin dependent diabetes. Br Med J 312:744–745, 1996
Bagi Z, Koller A, Kaley G: Superoxide-NO interaction decreases flow- and agonist-induced dilations of coronary arterioles in type 2 diabetes mellitus. Am J Physiol 285:H1404–H1410, 2003
McCance DR, Dyer DG, Dunn JA, Bailie KE, Thorpe SR, Baynes JW, Lyons TJ: Maillard reaction products and their relation to complications in insulin-dependent diabetes mellitus. J Clin Invest 91:2470–2478, 1993
Brownlee M, Cerami A, Vlassara H: Advanced glycosylation end products in tissues and the biochemical basis of diabetic complications. N Engl J Med 318:1315–1321, 1988
Petruzzi E, Orlando C, Pinzani P, Sestini R, Del Rosso A, Dini G, Tanganelli E, Buggiani A, Pazzagli A: ATP release by osmotic shock and HbA1c in diabetic subjects’ erythrocytes. Metabolism 43:435–440, 1994
McMillan DE, Utterback NG, La Puma J: Reduced erythrocyte deformability in diabetes. Diabetes 27:895–901, 1978
Bergfeld GR, Forrester T: Release of ATP from human erythrocytes in response to a brief period of hypoxia and hypercapnea. Cardiovasc Res 26:40–47, 1992
Dietrich HH, Ellsworth ML, Sprague RS, Dacey RG: Red blood cell regulation of microvascular tone through adenosine triphosphate. Am J Physiol 278:H1294–H1298, 2000
Ellsworth ML, Forrester T, Ellis CG, Dietrich HH: The erythrocyte as a regulator of vascular tone. Am J Physiol 269:H2155–H2161, 1995
Sprague RS, Ellsworth ML, Stephenson AH, Lonigro AJ: ATP: the red blood cell link to NO and local control of the pulmonary circulation. Am J Physiol 271:H2717–H2722, 1996
Sprague RS, Ellsworth ML, Stephenson AH, Kleinhenz ME, LonigroAJ: Deformation-induced ATP release from red blood cells requires cystic fibrosis transmembrane conductance regulator activity. Am J Physiol 275:H1726–H1732, 1998
Sprague RS, Ellsworth ML, Stephenson AH, Lonigro AJ: Participation of cAMP in a signal-transduction pathway relating erythrocyte deformation to ATP release. Am J Physiol 281:C1158–C1164, 2001
McCullough WT, Collins DM, Ellsworth ML: Arteriolar responses to extracellular ATP in striated muscle. Am J Physiol 272:H1886–H1891, 1997
Collins DM, McCullough WT, Ellsworth ML: Conducted vascular responses: communication across the capillary bed. Microvas Res 56:43–53, 1998
Dietrich HH, Kajita Y, Dacey RG: Local and conducted vasomotor responses in isolated rat cerebral arterioles. Am J Physiol 271:H1109–H1116, 1996
Ellsworth ML: The red blood cell as an oxygen sensor: What is the evidence Acta Physiol Scand 168:551–559, 2000
Gonzalez-Alonzo J, Olsen DB, Saltin B: Erythrocyte and the regulation of human skeletal muscle blood flow and oxygen delivery. Circ Res 91:1046–1055, 2002
Ellsworth ML: Red blood cell derived ATP as a regulator of skeletal muscle perfusion. Med Sci Sports Exerc 36:35–41, 2004
Sprague RS, Olearczyk JJ, Spence DM, Stephenson AH, Sprung RW, Lonigro AJ: Extracellular ATP signaling in the rabbit lung: erythrocytes as determinants of vascular resistance. Am J Physiol 285:H693–H700, 2003
Olearczyk JJ, Stephenson AH, Lonigro AJ, Sprague RS: Receptor-mediated activation of the heterotrimeric G protein Gs results in ATP release from erythrocytes. Med Sci Monit 7:669–674, 2001
Olearczyk JJ, Stephenson AH, Lonigro AJ, Sprague RS: Heterotrimeric G protein Gi is involved in a signal transduction pathway for ATP release from erythrocytes. Am J Physiol 286:H940–H945, 2004
Olearczyk JJ, Stephenson AH, Lonigro AJ, Sprague RS: NO inhibits signal transduction pathway for ATP release from erythrocytes via its action on heterotrimeric G protein Gi. Am J Physiol 287:H748–H754, 2004
Liang G, Stephenson AH, Lonigro AJ, Sprague RS: Erythrocytes of humans with cystic fibrosis fail to stimulate nitric oxide synthesis in isolated rabbit lungs. Am J Physiol 288:H1580–H1585, 2004
Hashim S, Liu YY, Wang R, Anand-Srivastava MB: Streptozotocin-induced diabetes impairs G protein linked signal transduction in vascular smooth muscle. Mol Cell Biochem 240:57–63, 2000
Matsuda N, Hattori Y, Gando S, Watanuki S, Kemmotsu O, Kanno M: Differential gene transcription regulation of Gi isoforms and Gs protein expression in diabetic rat hearts. Naunyn Schmiedebergs Arch Pharmacol 361:53–60, 2000
Hashim S, Li Y, Nagakura TS, Anand-Sirvastava MB: Modulation of G protein expression and adenylyl cyclase signaling by high glucose in vascular smooth muscle. Cardiovasc Res 63:709–718, 2004
Moxham CM, Malbon CC: Insulin action impaired by deficiency of the g-protein subunit Gi2. Nature 379:840–844, 1996
Grawer D, Milligan G, Spiegel AM, Unson CG, Houslay MD: Abolition of the expression of inhibitory guanine nucleotide regulatory protein Gi activity in diabetes. Nature 327:229–232, 1987
Livingstone C, McLellan AR, McGregor MA, Wilson A, Connell JMC, Small M, Milligan G, Paterson KR, Houslay MD: Altered G protein expression and adenylate cyclase activity in platelets of non-insulin-dependent diabetes (NIDDM) male subjects. Biochem Biophys Acta 1096:127–133, 1991
Wolffenbuttel BHR, Giordano D, Founds HW, Bucala R: Long-term assessment of glucose control by haemoglobin-AGE measurement. Lancet 347:513–515, 1996
Sprague RS, Bowles EA, Stumpf MS, Rickettes G, Freidman A, Hou WH, Stephenson AH, Lonigro AJ: Rabbit erythrocytes possess adenylyl cyclase II that is activated by the heterotrimeric G proteins Gs and Gi. Pharmacol Rep 57 (Suppl.):222–228, 2005
Sowers JR, Haffner S: Treatment of cardiovascular and renal risk factors in the diabetic hypertensive. Hypertension 40:781–788, 2002
The Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications of insulin-dependent diabetes mellitus. N Engl J Med 329:977–986, 1993
Nathan DM, Singer DE, Hurxthal K, Goodson JD: The clinical information value of the glycosylated hemoglobin assay. N Engl J Med 310:341–346, 1984
Larsen ML, Horder M, Mogensen EF: Effect of long-term monitoring of glycosylated hemoglobin levels in insulin-dependent diabetes mellitus N Engl J Med 323:1021–1025, 1990
Tang WJ, Gilman AG: Adenylyl cyclases. Cell 70:869–872, 1992
Neer EJ, Echeverria D, Knox S; Increase in the size of brain adenylate cyclase with activation by guanosine 5'-(beta, gamma-imino)triphosphate. J Biol Chem 255:9782–9789, 1980
Price SA, Pochun I, Phaneuf S, Bernal AL: Adenylyl cyclase isoforms in pregnant and non-pregnant human myometrium. J Endocrinology 164:21–30, 2000
Sprague RS, Bowles EA, Olearczyk JJ, Stephenson AH, Lonigro AJ: The role of G protein subunits in the release of ATP from human erythrocytes. J Physiol Pharmacol 53:667–674, 2002
Tang W, Gilman AG: Type-specific regulation of adenylyl cyclase by G protein subunits. Science 254:1500–1503, 1991
Federman AD, Conklin BR, Schrader KA, Reed RR, Bourne HR: Hormonal stimulation of adenylyl cyclase through Gi-protein beta gamma subunits. Nature 356:159–161, 1992
Bayewitch ML, Avidor-Reiss T, Levy R, Pfeuffer T, Nevo I, Simonds WF, Vogel Z: Differential modulation of adenylyl cyclase I and II by various G subunits. J Biol Chem 273:2273–2276, 1998
Dreiling CE: Localization of 2',3'-cyclic nucleotide 3'-phosphodiesterase in human erythrocyte membranes. Biochem Biophys Acta 649:587–594, 1981
Petrov V, Fagard R, Lijene P: Human erythrocytes contain Ca2+, calmodulin-dependent cyclic nucleotide phosphodiesterase which is involved in the hydrolysis of cAMP. Meth Find Exp Clin Pharmacol 20:387–393, 1998(Randy S. Sprague, Alan H. Stephenson, El)