Red Blood Cell Stimulation of Platelet Nitric Oxide Production

Jun 20, 2007 - the Communication between Cells in the. Bloodstream. Jamie S. Carroll, Chia-Jui Ku, ... is released from red blood cells (RBCs) upon de...
1 downloads 12 Views 217KB Size
Download PDF
Anal. Chem. 2007, 79, 5133-5138

Red Blood Cell Stimulation of Platelet Nitric Oxide Production Indicated by Quantitative Monitoring of the Communication between Cells in the Bloodstream Jamie S. Carroll, Chia-Jui Ku, Welivitiya Karunarathne, and Dana M. Spence*

Department of Chemistry, Wayne State University, Detroit, Michigan 48202

ATP is a recognized stimulus of nitric oxide synthase and is released from red blood cells (RBCs) upon deformation. The objective of this work is to demonstrate that RBCs stimulate nitric oxide production in platelets by employing a continuous flow analysis system in which the stream contains both RBCs and platelets. Here, two drugs known to improve blood flow in vivo (pentoxyfilline and iloprost) are shown to increase both the release of RBC-derived ATP and the production of platelet-derived NO. A flowbased chemiluminescence assay (in vitro) was employed to quantitatively determine the amount of ATP released from erythrocytes subjected to flow-induced deformation. Prior to being subjected to flow, erythrocytes were incubated in the absence or presence of 4.8 µM pentoxyfilline or 80 nM iloprost. Erythrocytes obtained from rabbits (n ) 22) that were subjected to flow released 239 ( 29 nM ATP. When treated with pentoxyfilline, the ATP released from the flowing RBCs increased to 450 ( 94 nM ATP. An increase in RBC-derived ATP was also measured for iloprost-incubated RBCs in flow (362 ( 45 nM ATP). Importantly, platelets that were loaded with diaminofluorofluorescein diacetate, an intracellular fluorescence probe for NO, exhibited increases in fluorescence intensity by 16% in the presence of RBCs treated with pentoxyfilline and a 10% increase when treated with iloprost. When the ATP release from the RBCs was inhibited with glybenclamide, the platelet fluorescence intensity decreased by 25 and 51% for RBCs incubated with pentoxyfilline and iloprost, respectively. In an experiment not involving the RBC, inhibition of the P2x receptor on the platelets (an ATP receptor) resulted in no increase in platelet NO production, suggesting that the NO production in the activated platelet is due to ATP.

In addition to its well-known ability as a vasodilator, NO is also able to inhibit platelet activation and aggregation. Platelets have also been shown to produce NO upon activation.1 Freedman et * To whom correspondence should be addressed. Phone: 313.577.8660. Fax: 313.577.2942. E-mail: [email protected]. (1) Lantoine, F.; Brunet, A.; Bedioui, F.; Devynck, J.; Devynck, M.-A. Biochem. Biophys. Res. Commun. 1995, 215, 842-848. 10.1021/ac0706271 CCC: $37.00 Published on Web 06/20/2007

© 2007 American Chemical Society

al. simultaneously measured NO production and aggregation of platelets upon activation with ATP.2 These authors placed an electrode for NO into the cell of an aggregometer in order to simultaneously monitor the NO release from platelets and the aggregation of the platelets, thus demonstrating that NO was produced and released by these cells upon activation. From these studies, the authors concluded that NO released by the platelets was key to preventing further platelet recruitment to the activated platelets, but only mildly prevented their adhesion to an endothelium. In question concerning the production and release of plateletderived NO is the mechanism by which the NO production is stimulated. For example, ATP is a recognized stimulus of nitric oxide synthase (NOS) and subsequent production of platelet NO; moreover, it is also known that ATP is released from platelets upon activation.3 However, it has not been established if ATP released from activated platelets has the ability to stimulate the production of NO in platelets. Here, results from experiments provide evidence that stimulation of NO production in platelets activated with ADP and thrombin, two known activators of platelets, is actually stimulated by ATP that is released from these cells upon activation. Although iloprost has already been shown to increase the ATP release from red blood cells (RBCs),4 pentoxyfilline (Trental), which has been reported to have beneficial effects in the diabetic kidney,5-7 has not been investigated as a stimulus of ATP release. Previous reports have shown that a pentoxyfilline derivative may act as an anti-inflammatory agent by suppressing oxygen radical production or behave as a scavenger of reactive oxygen species.5 More recently, reports have suggested that adding pentoxyfilline to a diabetic patient’s daily medicinal intake improved some of the complications associated with the disease, perhaps by reducing (2) Freedman, J. E.; Loscalzo, J.; Barnard, M. R.; Alpert, C.; Keaney, J. F.; Michelson, A. D. J. Clin. Invest. 1997, 100, 350-356. (3) Beigi, R.; Kobatake, E.; Aizawa, M.; Dubyak, G. R. Am. J. Physiol. 1999, 276, C267-278. (4) Olearczyk, J. J.; Stephenson, A. H.; Lonigro, A. J.; Sprague, R. S. Am. J. Physiol. 2004, 287, H748-H754. (5) Davila-Esqueda, M. E.; Martinez-Morales, F. Exp. Diabesity Res. 2004, 5, 245-251. (6) Gunduz, Z.; Canoz, O.; Per, H.; Dusunsel, R.; Poyrazoglu, M. H.; Tez, C.; Saraymen, R. Renal Failure 2004, 26, 597-605. (7) Navarro, J. F.; Mora, C.; Muros, M.; Macia, M.; Garcia, J. Am. J. Kidney Dis. 2003, 42, 264-270.

Analytical Chemistry, Vol. 79, No. 14, July 15, 2007 5133

the amount of lipid peroxidation in the cell.8 However, several reports have also demonstrated that pentoxyfilline acts specifically on improving RBC deformability, ultimately reducing fibrinogen concentration and platelet adhesion and improving blood viscosity.8 Here, data are provided suggesting that the efficacy of pentoxyfilline in vivo, especially in relationship to platelet adhesion, may be RBC-mediated via its ability to increase the ATP released from erythrocytes that are subjected to deformation. Moreover, data are also presented showing that pentoxyfilline and iloprost, two drugs known to improve blood flow in vivo, have the ability to increase the concentrations of ATP released from RBCs that are subjected to deformation while flowing through microbore tubing. Previous reports concerning RBC-derived ATP and ATPstimulated NO production suggests that, in vivo, the ability of RBCs to release ATP may act as a stimulus for NO production in platelets. This is an extension of the RBCs’ role to stimulate endothelium NO (via the RBCs ability to release ATP). To determine whether this communication between RBCs and platelets exists, a microflow system was used in conjunction with fluorescence detection to measure increases in platelet-derived NO in the presence of various levels of RBC-derived ATP. Collectively, the system reported here establishes the existence of molecular-mediated communication between RBCs and platelets in the circulation. EXPERIMENTAL SECTION Experimental Animals. All surgical procedures involving animals used in this study were performed under protocols approved by the Animal Investigation Committee at Wayne State University. For obtaining rabbit RBCs, male New Zealand White rabbits (2.0-2.5 kg) were anesthetized with ketamine (8.0 mg/ kg) and xylazine (1.0 mg/kg) followed by pentobarbital sodium (15 mg/kg iv). After tracheotomy, the rabbits were mechanically ventilated (tidal volume 20 mL/kg, rate 20 breaths/min; Harvard ventilator). A catheter was placed into a carotid artery, heparin (500 units, iv) was administered, and after 10 min, animals were exsanguinated. Isolation and Purification of Platelets and RBCs. Prior to purifying the RBCs from the whole blood of rabbits, the blood was centrifuged at 500g at 37 °C for 10 min. The platelet-rich plasma (PRP) and buffy coat were removed by aspiration; the PRP was saved for subsequent isolation of the platelets. Packed RBCs were resuspended and washed three times in physiological salt solution (PSS; in mM; 4.7 KCl, 2.0 CaCl2, 1.2 MgSO4, 140.5 NaCl, 21.0 tris(hydroxymethyl)aminomethane, and 11.1 dextrose with 5% bovine serum albumin, pH adjusted to 7.4). Platelets were isolated from the PRP by adding 1 mL of acid citrate dextrose (ACD) to 9 mL of the PRP and centrifuging at 1500g at 37 °C for 10 min. The platelets were then washed two times by centrifugation in a mixture containing Hank’s balanced salt solution (HBSS) and ACD (6:1 v/v). Our group has found that, if kept in the PRP, the platelets can be purified within 1-2 days after the surgical procedure and still produce reliable data. However, for results reported here, the platelets were consumed through experimental procedures on the day of harvesting them from the rabbit. (8) Radfar, M.; Larijani, B.; Hadjibabaie, M.; Rajabipour, B.; Mojtahedi, A.; Abdollhai, M. Biomed. Pharmacother. 2005, 59, 302-306.

5134

Analytical Chemistry, Vol. 79, No. 14, July 15, 2007

Measurement of ATP from Erythrocytes. All reagents were purchased from Sigma Chemical (St. Louis, MO). The determination of ATP released from mechanically deformed RBCs was performed using microbore tubing (Polymicro Technologies, Phoenix, AZ) as a microcirculation mimic.9 A syringe pump (Harvard Apparatus, Holliston, MA) was used to propel two solutions; one solution contained luciferase with 2 mg of luciferin added per 5 mL of solution to enhance the chemiluminescence sensitivity. The other syringe contained a 7% solution of washed RBCs in either the presence or absence of pentoxyfilline or iloprost. The two solutions were pushed through microbore tubing having an inside diameter of 50 µm and combined at a mixing tee. The resultant mixed flow that eluted from this tee then traveled through a third section of tubing having a diameter of 75 µm. This last section of tubing was placed over a photomultiplier tube (Hamamatsu, Iwata City, Japan) in order to measure the chemiluminescence resulting from the mixing of the luciferase solution with ATP that was released from the RBCs while flowing through the microbore tubing. The system was calibrated each day with authentic ATP standards having concentrations from 0 to 1.5 µM. The correlation coefficient for this calibration is generally >0.99, thus establishing a linear relationship between the emission and ATP concentration. For ATP release studies that did not involve flow, 100 µL of a solution of RBCs (7% hematocrit) in the presence or absence of the release stimulants or inhibitors were combined with 300 µL of PSS. This mixture was then added to 50 µL of the luciferin/ luciferase mixture, and the resultant chemiluminescence emission was monitored by placing the cuvette over a photomultiplier tube. The system was also calibrated with authentic ATP. For inhibitor studies of ATP release, glybenclamide (Sigma) was prepared by adding 49 mg to 2 mL of 0.1 M NaOH and 7.94 mL of a dextrose solution (1 g of dextrose in 20 mL of deionized, distilled water). The solution was carefully heated to 52 °C until dissolved, resulting in a final stock concentration of 0.01 M. This stock was then diluted in the PSS to create a buffered solution of glybenclamide to which washed RBCs were added. For studies involving diamide, 0.0177 g of diamide was prepared in 50 mL of deionized water, creating a 2 mM stock solution. The stock solution was then diluted (1:10) in buffer after which 1 mL was added to 9 mL of 7% RBCs, creating a final diamide concentration of 20 µM. Pentoxyfilline was prepared in buffer containing 5.4 µM pentoxyfilline in the PSS prior to the addition of washed RBCs. After addition of the RBCs, the final concentration of the pentoxyfilline was 4.8 µM. Finally, for studies involving iloprost, a procedure identical to that involving pentoxyfilline was followed except that the final concentration in the RBCs was 100 nM. Determination of Platelet-Derived NO in a Static System To Verify NOS Stimulation via ATP. Fluorescence emission spectra were obtained using 1 µM diaminofluorofluorescein diacetate (DAF-FM DA) in a quartz cuvette with excitation wavelength at 495 nm and emission wavelength at 515 nm. Both excitation and emission slit widths of 3 nm were used in all experiments. Aliquots of 10 µM DAF-FM DA (100 µL) were added into each vial and allowed to incubate with varying volumes of (9) Sprung, R. J.; Sprague, R. S.; Spence, D. M. Anal. Chem. 2002, 74, 22742278.

Figure 1. Increased production of NO from platelets in the presence of 0.1 µM ATP. In (A), the traces represent the measured fluorescence intensities from platelets only (lower trace), platelets in the presence of DAF-FM DA (middle trace), and platelets with DAF-FM DA in the presence of ATP (upper trace). In (B), the bars represent the average of normalized results with error bars representing the standard error of the mean.

platelets (to establish the various DAF-FM DA concentrations) for 30 min before each fluorescence measurement was obtained. The final volume of each vial was 1.0 mL, diluted with HBSS. This system for measuring platelet NO with DAF-FM DA is described in detail elsewhere.10 For studies involving the inhibition of the P2x receptor, a 100 µM stock solution of NF449 octasodium salt ,4′,4′′,4′′′-xarbonylbis[imino-5,1,3-benzenetriyl bis(carbonylimino)]tetrakis(benzene-1,3disulfonic acid) octasodium salt (purchased from Sigma-Aldrich) was prepared by adding 1 mg of the NF449 to 6.6 mL of HBSS. NF449 is a selective P2x receptor antagonist, which inhibits ATP binding to the receptor. When using the inhibitor, 30 µL of the 100 µM stock solution was added to 3 mL of an apyrase-washed stock platelet suspension that contained ∼3 × 108 platelets/mL; the platelets already contained DAF-FM DA as described above. This solution was allowed to incubate for 30 min prior to addition of a NOS stimulus (ATP) or a platelet activator (ADP or thrombin). After addition of these final reagents, the solution was allowed to incubate for an additional 30 min prior to obtaining the emission at 515 nm. Preparation and use of L-NAME, a NOS inhibitor, has been described for platelets elsewhere.10 Determination of Platelet-Derived NO in a Flow System. Platelets were passed through microbore tubing in a manner similar to that in our previous studies involving RBCs.9,11 The system consists of three major components, namely, a dual syringe pump, the sections of microbore tubing, and a flow-through fluorescence detector (Jasco) housed with a capillary flow cell. DAF-FM DA-loaded platelets and RBCs were pumped through a section of microbore tubing using a syringe pump to propel the solution at a flow rate of 6.7 µL min-1. The pump is a conventional syringe pump where the syringe is easily accessible. Previous experience with this type of syringe has proven that platelet sedimentation during the measurement portion of the analysis is not problematic because the platelets are usually in solution as a suspension. The fluorescent signal that is produced in the platelets due to NO production was determined using the aforementioned flow-through fluorescence detector and monitored with a program written in-house with LabView (National Instruments). RESULTS Measurement of Platelet NO Production in a Microflow System. Recently, it was demonstrated that platelet-derived NO, (10) Ku, C.-J.; Karunarathne, W.; Kenyon, S.; Root, P.; Spence, D. Anal. Chem. 2007, 79, 2421-2426. (11) Carroll, J. S.; Subasinghe, W.; Raththagala, M.; Baguzis, S.; Oblak, T.; Root, P. D.; Spence, D. M. Mol. Biosyst. 2006, 2, 305-311.

Figure 2. Evidence that NO production in platelets is due to ATP stimulation of the P2x receptor. Platelet NO production was measured in the absence and presence of a stimulus of NO production (ATP) or activators of ATP release (ADP and thrombin), which lead to NO production. As shown, NO production increases in each case. However, identical measurements that were performed in the presence of NF449, a reagent that blocks the ATP receptor on platelets (P2x), result in a decrease in NO production. These data suggest that NO production in platelets is largely dependent upon ATP binding to the platelet receptor. L-NAME, a NOS inhibitor, decreased NO production with and without the P2x blocker, verifying the measured signal in each case was due to NO.

in either extracellular or intracellular form, could be quantitatively determined using variations of the DAF family of probes.10 However, those measurements were performed in a static system that did not involve flow. In vivo, platelets are part of the flowing bloodstream, typically occupying the space closest to the vessel wall due to the non-Newtonian characteristics of whole blood (i.e., the platelets are displaced from the center of the flow by the larger RBCs). Therefore, to more closely mimic the physical conditions that the platelet would be subjected to in vivo, and to prepare for subsequent studies involving the RBC, it was imperative that the NO production by the platelets be measured in a flowing stream. The data shown in Figure 1 demonstrate the ability to measure both basal levels of NO in the platelets and the increase in NO production when these platelets are stimulated with ATP (a recognized stimulus of NO production in platelets). The bottom trace in Figure 1A is essentially the background due to platelets flowing through the microflow system described above. The middle trace is the emission resulting from platelets incubated in Analytical Chemistry, Vol. 79, No. 14, July 15, 2007

5135

Figure 3. Improved release of deformation-induced release of ATP from red cells incubated in pentoxyfilline or iloprost under conditions of flow. The RBCs were prepared as a 7% hematocrit prior to incubation with reagents and subsequent deformation in the microflow system composed of microbore tubing having an inside diameter that approximates resistance vessels in vivo (50 µm). The measured chemiluminescence intensities of ATP release from (a) RBCs in buffer (232 ( 29 nM, n ) 22), (b) RBCs in buffer incubated with diamide (60 ( 12 nM, n ) 12), or (c) RBCs with glybenclamide (97 ( 15 nM, n ) 13). The RBC-derived ATP was also measured for RBCs (d) incubated in pentoxyfilline (450 ( 94 nM, n ) 6) and pentoxyfilline in the presence of (e) diamide (157 ( 39 nM, n ) 6) and (f) glybenclamide (179 ( 54 nM, n ) 6). Similar results are also reported for RBCs (g) incubated in iloprost (362 ( 45 nM, n ) 7) and iloprost in the presence of (h) diamide (80 ( 24 nM, n ) 8) and (i) glybenclamide (133 ( 21 nM, n ) 9).

1 µM DAF-FM DA for 30 min prior to being pumped through the system. The upper trace is the fluorescence intensity of another aliquot of the DAF-FM DA-loaded platelets moving through the system; however, these platelets were stimulated with 0.1 µM ATP. As evident by the intensity increase in the traces, the system enables the detection of basal and stimulated levels of NO production in platelets in a flowing stream. Previous work by other groups2 provides evidence that platelets have the ability to release ATP upon activation with ADP and other agonists that result in activation of the platelets. In addition, the data in Figure 1B and elsewhere10 show that these platelets can also produce their own NO (intracellularly through NOS) via stimulation with ATP. Such a scheme suggests that platelets can self-stimulate their own NO; that is, upon activation, the platelets release ATP which may, in turn, stimulate NOS and subsequent NO production. To date, there has been one example of selfstimulated production of NO in platelets.6 This measurement was performed using a carbon electrode and showed a transient of NO production upon activation with ADP. Moreover, that same measurement was not clear on whether the activation of the NOS was due to the ADP or the released ATP. Substances such as collagen, thrombin, fibrinogen, and ADP all activate platelets and induce an ATP release. We hypothesized that ATP release, and not the activators themselves, stimulate the NO production in platelets. To provide evidence for this theory, experiments were performed in which platelets were stimulated with ADP or thrombin to induce ATP release. Measurements of this ATP release, via the chemiluminescence assay for ATP described above, upon activation were verified (data not shown). Moreover, these agents were shown to increase the production of NO by these platelets as shown in Figure 2. However, when the ATP receptor on platelets (the P2x receptor) was blocked, the NO production decreased for every activator studied (second bar for each set in Figure 2). The importance of this finding is that only ATP is able to stimulate NO production in platelets; the platelet activators (which results in a shape change of the platelets) do result in NO production, but perhaps through an ATP-mediated mechanism. This is potentially very important because it is known that some patient groups known to have low NO production or 5136 Analytical Chemistry, Vol. 79, No. 14, July 15, 2007

complications involving blood flow are also known to have RBCs that release lower than normal amounts of ATP upon pharmacological activation or physical deformation of the erythrocyte. Patients with primary pulmonary hypertension,12 cystic fibrosis,13 and diabetes14,15 are example patient groups with these lower ATP release values. Effects of Some Drugs on ATP Release from RBCs. The data shown in Figure 1 demonstrate that platelet NO can be measured via fluorescence spectrophotometry in tubing having a diameter that approximates an arteriole in vivo. Moreover, in Figure 2, the data suggest that the increase in NO production via platelet activation is due to an ATP stimulus alone. These results thus render the RBC a possible determinant of platelet physiology in vivo because it has been shown that the RBC is capable of releasing ATP via either mechanical deformation9,14,16 or pharmacologically.15 To demonstrate the ability of certain pharmaceuticals to increase the amount of ATP released from RBCs that are subjected to deformation, aliquots of a 7% hematocrit of RBCs were pumped through a 30-cm section of the microbore tubing having an inside diameter of 50 µm in the presence and absence of pentoxyfilline or iloprost. Pentoxyfilline is thought to improve blood flow via its ability to make the RBC more deformable through a radical scavenging mechanism. An increase in deformability would then be anticipated to lead to an increase in deformation-induced release of ATP from the RBC. As shown in Figure 3, a significant increase in ATP release (p < 0.001) from the RBCs is measured for those cells incubated with pentoxyfilline (450 ( 94 nM ATP, bar d) when compared to those cells in buffer alone (232 ( 29 nM ATP, bar a). Also shown in Figure 3 is the ATP release from an aliquot of the RBCs that were incubated in 20 µM diamide. (12) Sprague, R. S.; Stephenson, A. H.; Ellsworth, M. L.; Keller, C. Lonigro, A. J. Exp. Biol. Med. 2001, 226, 434-439. (13) Sprague, R. S.; Ellsworth, M. L.; Stephenson, A. H.; Kleinhenz, M. E.; Lonigro, A. J. Am. J. Physiol. 1998, 275, H1726-H1732. (14) Carroll, J.; Raththagala, M.; Subasinghe, W.; Baguzis, S.; Oblak, T. D. a.; Root, P.; Spence, D. Mol. BioSyst. 2006, 2, 305-311. (15) Sprague, R. S.; Stephenson, A. H.; Bowles, E. A.; Stumpf, M. S.; Lonigro, A. J. Diabetes 2006, 55, 3588-3593. (16) Sprague, R. S.; Ellsworth, M. L.; Stephenson, A. H.; Lonigro, A. J. Am. J. Physiol. 1996, 271, H2717-H2722.

Figure 4. Incubation of RBCs with buffer (a) or pentoxyfilline (b). The resultant ATP release was measured in the absence of flow. In (a), the release was determined to be 81 ( 29 nM ATP, while in the presence of pentoxyfilline, the release was 53 ( 22 nM ATP (n ) 6 rabbits). As expected, the increase in ATP release from RBCs (c) incubated in iloprost (d) increased from 43 ( 8 nM ATP to 176 ( 11 nM ATP (n ) 5 rabbits). The pentoxyfilline and iloprost studies were performed with different rabbits on different days.

Figure 5. Percent change in fluorescence due to platelet NO production stimulated by RBCs in the presence and absence of stimulators and inhibitors of ATP release. The percent changes are reported relative to RBCs flowing with the DAF-FM DA-loaded platelets alone. In (a), RBCs incubated with pentoxyfilline prior to flowing with platelets in microbore tubing resulted in a 16% increase in emission intensity; in (b), the RBCs were incubated with pentoxyfilline and diamide, resulting in a 37% decrease in platelet-derived NO; RBCs were treated with glybenclamide and pentoxyfilline in (c) and the platelet-derived NO decreased by 25%; in (d), RBCs were incubated with iloprost, resulting in an increase in platelet NO production of 10%; the iloprost-induced increase in NO production was reduced in (e) where RBCs treated with glybenclamide and iloprost resulted in a decrease in platelet-derived NO of 51%. The data represent the average of n ) 3 rabbits and demonstrate the ability to monitor RBC/platelet communication in the presence of flow.

Diamide is a recognized oxidant that results in a stiffened membrane and reduced RBC deformability.17 As expected, RBCs incubated with diamide resulted in a decrease in released ATP (60 ( 12 nM ATP, bar b); however, in the presence of pentoxyfilline and diamide, the RBCs still released ATP at levels that were statistically greater (157 ( 39 nM ATP, bar e, p < 0.001) than RBCs that were exposed to diamide alone. These data provide evidence suggesting that pentoxyfilline has the ability to increase the release of ATP from RBCs whose deformability is diminished. Figure 3 also contains data indicating that the increase in ATP release from RBCs incubated with pentoxyfilline is not due to cell lysis. Specifically, the RBCs were incubated with glybenclamide, an anion transport inhibitor that will inhibit ATP release from RBCs. In the presence of the glybenclamide, the ATP release from the RBCs is decreased to 97 ( 15 nM ATP (bar c). If the ATP release from the pentoxyfilline-incubated RBCs was due to cell lysis, the inhibitory effect of the glybenclamide would not be measured. Finally, when the RBCs were subjected to both glybenclamide and pentoxyfilline, the value (179 ( 54 nM ATP, bar f, p < 0.005) is statistically equivalent to the results for incubating the RBCs in glybenclamide alone. There have been reports of other pharmaceutical agents having the ability to stimulate ATP release from RBCs. Specifically, iloprost, a stable analogue of prostacyclin, has been reported to increase the release of RBC-derived ATP. Figure 3 shows that iloprost’s effect on ATP release from RBCs can be determined quantitatively in a flowing stream. When incubated in 80 nM iloprost, the ATP release from the RBCs was determined to be 362 ( 45 nM ATP (bar g). Not unlike the effect of pentoxyfilline, the effect of the iloprost could also be inhibited with diamide (80 ( 24 nM ATP (bar h)) or glybenclamide (133 ( 21 nM ATP (bar i)). While the data in Figure 3 suggest that certain drugs known to improve blood flow in the circulation may, at least in part, may

be RBC-mediated (due to the RBC’s ability to release ATP, a recognized stimulus of NO production), it does not prove whether the drug efficacy is related to flow. That is, pentoxyfilline is known to increase RBC deformability; therefore, due to flow-induced shear, the RBC may become more deformed in the presence of pentoxyfilline, resulting in an increase in ATP release from the cell. However, iloprost has been shown to be effective independent of flow, working through an IP-receptor/Gs protein stimulated pathway.18 To verify if flow is a requirement for either of these drugs’ efficacy, the ability of the RBCs to release ATP in the presence of pentoxyfilline or iloprost was examined in the absence of flow. The data in Figure 4 suggest that the ability of pentoxyfilline to increase RBC-derived ATP is indeed dependent upon flow. The concentration of extracellular ATP (81 ( 29 nM ATP) did not show any appreciable statistical increase in the presence of the pentoxyfilline (53 ( 22 nM ATP). However, consistent with previous reports by Olearczyk,18 the amount of extracellular ATP in the presence of iloprost did increase from 43 ( 8 nM ATP to 176 ( 11 nM ATP, even in the absence of flow. Erythrocyte-Stimulated NO Production in Platelets. Collectively, the data in Figures 1-4 suggest a possible unique relationship between RBCs and platelets in the circulation. Specifically, the RBC may be able to communicate with the platelet through RBC-derived ATP, stimulating the important NO molecule. Data verifying this relationship are summarized in Figure 5. Here, the ability to measure platelet-derived NO upon stimula-

(17) Kosower, N.; Kosower, E.; Wertheim, B.; Correa, W. Biochem. Biophys. Res. Commun. 1969, 37, 593-596.

(18) Olearczyk, J. J.; Stephenson, A. H.; Lonigro, A. J.; Sprague, R. S. Med. Sci. Monit. 2001, 7, 669-674.

Analytical Chemistry, Vol. 79, No. 14, July 15, 2007

5137

tion with ATP secreted from deformed RBCs in a microflow system is demonstrated. The percent increase or decrease in platelet-derived NO (measured spectrofluorometrically with the DAF-FM DA probe) is presented for RBCs in the presence of either pentoxyfilline or iloprost. In the presence of pentoxyfilline, the fluorescence from the platelet NO production increased by 16%. However, this increase in platelet NO production due to pentoxyfilline was reduced by 37 and 25% in the presence of diamide or glybenclamide, respectively. Iloprost resulted in a 10% increase in platelet NO production, a value that was reduced by 51% in the presence of glybenclamide. Moreover, neither pentoxyfilline nor iloprost stimulated NO production in the platelets in the absence of the RBC. CONCLUSIONS In the isolated perfused rabbit lung, Sprague has demonstrated that RBC-derived ATP was a requirement in the perfusate in order to measure a flow-induced decrease in mean arterial pressure.19 It has also been demonstrated that RBCs release ATP upon mechanical deformation as they traverse microbore tubing20 or pass through pores that have diameters approximating those of resistance vessels in vivo.21 The release of ATP from deformed RBCs increases as the tubing or pore diameter decreases. Collectively, the aforementioned reports suggest that the RBC may be a determinant of vascular caliber in the microcirculation. The role of the RBC in the maintenance of vascular caliber is due to the ability of ATP to stimulate NOS, thus leading to subsequent production of nitric oxide that stimulates cGMP-dependent relaxation of the smooth muscle cells surrounding the vessel. Studies have shown that patients with certain diseases, or complications arising from those diseases, have RBCs that release less ATP than the RBCs obtained from healthy controls. For example, patients with pulmonary hypertension12 and cystic fibrosis13 release less RBC-derived ATP upon mechanical deforma(19) Sprague, R. S.; Olearczyk, J. J.; Spence, D. M.; Stephenson, A. H.; Sprung, R. W.; Lonigro, A. J. Am. J. Physiol.: Heart Circulatory Physiol. 2003, 285, H693-H700. (20) Fischer, D. J.; Torrence, N. J.; Sprung, R. J.; Spence, D. M. Analyst 2003, 128, 1163-1168. (21) Sprague, R. S.; Ellsworth, M. L.; Stephenson, A. H.; Lonigro, Andrew, J. Am. J. Physiol. Heart Circulatory Physiol. 1996, 271, H2717-H2722. (22) Garnier, M.; Attali, J. R.; Valensi, P.; Delatour-Hanss, E.; Gaudey, F.; Koutsouris, D. Metab., Clin. Exp. 1990, 39, 794-798.

5138

Analytical Chemistry, Vol. 79, No. 14, July 15, 2007

tion than the RBCs from healthy controls. Recently, in a separate study, it was reported that patients with type II diabetes mellitus released 91 ( 10 nM ATP versus 190 ( 10 nM ATP release in control RBCs.11 In the latter study, experiments provided evidence suggesting that a weakened oxidant defense system within the RBCs rendered the cells less deformable, a trait of diabetic RBCs, and unable to release normal levels of ATP upon deformation.11 To date, these lowered values of ATP release from the RBC have been discussed in terms of the endothelial cell and its inability to produce the proper levels of NO needed for vasodilatory purposes. Here, the data presented provide evidence that the decreased levels of RBC-derived ATP may be more of an important factor of platelet function in the circulation, especially considering the results from Figure 2 that suggest platelet NO is stimulated through ATP. The results shown here demonstrate that the ATP released by the RBCs is able to stimulate NO production in platelets. This is, the decrease in NO production in the platelets decreases substantially when the RBCs are in the presence of a cell-stiffening agent (diamide) or an inhibitor of ATP release (glybenclamide). It should also be noted that neither of these agents has an affect on platelet NO production in the absence of the RBCs. While in vivo studies will be necessary to truly verify any conclusions drawn here from in vitro studies, the data presented suggest that RBCs are able to stimulate NO production in platelets via their ability to release ATP. The results provided here also demonstrate that the efficacy of pentoxyfilline and iloprost to improve blood flow may be due to their ability to facilitate increased levels of ATP released from RBCs. Thus, the efficacy of pentoxyfilline may be pronounced in diabetic patients, especially in light of the fact that it is recognized that this patient group has now been reported to have less deformable erythrocytes22 and to release low amounts of ATP11 compared to healthy, nondiabetic patients. ACKNOWLEDGMENT This work was funded by the National Institutes of Health (HL 073942) and by an American Heart Association (Greater Midwest Affiliate) Predoctoral Fellowship to JSC.

Received for review March 29, 2007. Accepted June 1, 2007. AC0706271