A selective phosphodiesterase 3 inhibitor rescues low PO2-induced ATP release from erythrocytes of humans with type 2 diabetes: implication for vascular control

Abstract

Erythrocytes, via release of ATP in areas of low oxygen (O(2)) tension, are components of a regulatory system for the distribution of perfusion in skeletal muscle ensuring optimal O(2) delivery to meet tissue needs. In type 2 diabetes (DM2), there are defects in O(2) supply to muscle as well as a failure of erythrocytes to release ATP. The goal of this study was to ascertain if a phosphodiesterase 3 (PDE3) inhibitor, cilostazol, would rescue low O(2)-induced ATP release from DM2 erythrocytes and, thereby, enable these cells to dilate isolated erythrocyte-perfused skeletal muscle arterioles exposed to decreased extraluminal O(2). Erythrocytes were obtained from healthy humans (HH; n = 12) and humans with DM2 (n = 17). We determined that 1) PDE3B is similarly expressed in both groups, 2) mastoparan 7 (G(i) activation) stimulates increases in cAMP in HH but not in DM2 erythrocytes, and 3) pretreatment of DM2 erythrocytes with cilostazol resulted in mastoparan 7-induced increases in cAMP not different from those in HH cells. Most importantly, cilostazol restored the ability of DM2 erythrocytes to release ATP in response to low O(2). In contrast with perfusion with HH erythrocytes, isolated hamster retractor muscle arterioles perfused with DM2 erythrocytes constricted in response to low extraluminal PO(2). However, in the presence of cilostazol (100 μM), DM2 erythrocytes induced vessel dilation not different from that seen with HH erythrocytes. Thus rescue of low O(2)-induced ATP release from DM2 erythrocytes by cilostazol restored the ability of erythrocytes to participate in the regulation of perfusion distribution in skeletal muscle.

Fig. 1.

A: identification of phosphodiesterase (PDE)3B in human erythrocyte membranes. Solubilized membranes of healthy human (HH) and type 2 diabetes (DM2) erythrocytes were probed with an antibody generated against the first 300 amino acid residues of human PDE3B or β-actin (representative of 7 and 8 individual membrane preparations from healthy human and type 2 diabetes erythrocytes, respectively). B: determination of PDE3B content in erythrocyte membranes. Quantification by densitometric scanning corrected for the amount of β-actin in each sample. Values are means ± SE. NS, no significant difference.

Fig. 2.

Effect of mastoparan 7 (MAS 7; 10 μM) on cAMP levels in healthy human (A) and DM2 erythrocytes (B) in the absence and presence of cilostazol (Cilo; 100 μM). Erythrocytes were incubated with Cilo for 30 min before addition of MAS 7, and the reaction was stopped after an additional 15 min. Values are means ± SE. *Different from respective control (P < 0.05); †different from respective control (P < 0.01).

Fig. 3.

Effect of exposure to reduced O₂ tension on ATP release from erythrocytes of healthy humans (n = 7) and humans with DM2 (n = 10). In a tonometer, isolated erythrocytes (hematocrit 20%) were exposed to gas mixtures containing either 15% O₂, 6% CO₂, and balance nitrogen (Po₂ = 107 ± 5 mmHg) or 0% O₂, 6% CO₂, balance nitrogen (Po₂ = 10 ± 1 mmHg). ATP release was determined after a 30-min equilibration with 15% O₂ and 10 min after exposure to 0% O₂. In separate studies, type 2 diabetes erythrocytes were exposed to identical gas compositions in the absence and presence of the PDE3 inhibitor cilostazol (100 μM). Values are means ± SE. *Greater than respective normoxia (15% O₂) value (P < 0.05); †different from respective normoxia value and from DM2 erythrocytes in the absence of cilostazol. RBCs, red blood cells.

Fig. 4.

Effect of reduced extraluminal O₂ tension on dilation of isolated skeletal muscle arterioles perfused with erythrocytes (RBCs). Isolated arterioles were exposed to either extraluminal normoxia (room air, Po₂ = 147 ± 2 mmHg) or reduced O₂ tension (Po₂ = 17 ± 2 mmHg) and perfused with buffer containing well-oxygenated erythrocytes from either healthy humans (n = 5) or humans with DM2 (n = 10, HbA1c = 8.3 ± 0.8). In some studies type 2 diabetes erythrocytes were pretreated with cilostazol (Cilo) at a concentration of either 10 μM (n = 3) or 100 μM (n = 7). Values are means ± SE. *Different from value during extraluminal normoxia (P < 0.05); †different from vessels perfused with type 2 diabetes erythrocytes in the absence of Cilo (P < 0.05).

Fig. 5.

Proposed model of the mechanism by which cilostazol increases cAMP accumulation and ATP release from type 2 diabetes erythrocytes. Exposure of healthy human erythrocytes to reduced extracellular oxygen (O₂) tension results in release of O₂ from hemoglobin (desaturation) and activation of a signaling pathway for ATP release. This pathway requires an increase in cAMP that is regulated by PDE3 activity. The conduit for ATP release is pannexin 1 (33). In erythrocytes of humans with type 2 diabetes, expression of G_i and ATP release in response to exposure to reduced O₂ are decreased. Cilostazol, by inhibiting the degradation of cAMP in this pathway, enhances low O₂-induced ATP release. AC, adenylyl cyclase; CFTR, cystic fibrosis transmembrane conductance regulator; (+)activation; (−)inhibition.