Adenosine A3 Receptor Activation Modulates the Capillary Endothelial Glycocalyx

From the Department of Molecular Physiology and Biological Physics and the Cardiovascular Research Center, University of Virginia, Charlottesville, Va.

　

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Summary References

 
The endothelial glycocalyx is a dynamic extracellular matrix composed of cell surface proteoglycans, glycoproteins, and adsorbed serum proteins that has been implicated in the regulation and modulation of capillary tube hematocrit, permeability, and hemostasis. High tissue adenosine levels have been shown to adversely affect microvascular function and tissue survival after an ischemic episode, and previous work in this laboratory has shown that adenosine causes arteriolar constriction and degranulation of mast cells via the A₃ receptor (A3AR). We hypothesized that adenosine exerts at least part of its effect through modification of the glycocalyx via the A3AR. We used an in vivo cremaster model (hamster and mouse) in which circulating plasma was labeled with a 70-kDa FITC-dextran, and the capillaries were examined before and after superfusion with varying concentrations of adenosine (or other vasoactive molecules). Measurements of the dextran exclusion from an endothelial cell surface layer and red cell separation from the endothelial cell surface were made for up to 30 minutes. Our data indicate that adenosine causes a rapid and profound decrease in the ability of the glycocalyx to exclude dextran but only affects red blood cell exclusion at pharmacological levels. Knockout mice deficient in the A3AR were completely protected from glycocalyx changes attributable to adenosine. These data show a potential link between a known vasoactive tissue metabolite, adenosine, and regulation of the glycocalyx, which may be important during (patho)physiological changes in microvascular function during inflammatory insults.

Key Words: endothelium • inflammation • permeability • mast cells

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Summary References

 
Adenosine is a ubiquitous nucleoside that has been implicated in many facets of vascular function. There are four known adenosine receptors, A₁, A_2A, A_2B, and A₃,¹ each of which has a different pattern of tissue distribution and operates via different intracellular signaling mechanisms.² In addition to its well-known properties as a vasodilator, adenosine is intimately involved in a variety of other inflammatory reactions, with both proinflammatory³ and antiinflammatory⁴ roles. Previous work from this laboratory has shown that adenosine can, on the one hand, inhibit mast cell degranulation at low concentrations by activating the A_2A receptor⁵ and, on the other hand, cause mast cell degranulation via the A₃ receptor.⁶ Metabolites of adenosine may also play a key role in the inflammatory response, in that inosine accumulates in tissues in even greater concentrations than adenosine under inflammatory conditions² and is a highly selective agonist for A₃ receptors (A3AR). The foregoing suggests that the A3AR may be important in mediating pathophysiological changes in the vasculature under circumstances that increase tissue adenosine content.

Recent evidence has shown that ischemia/reperfusion injury, a common model of injury and inflammation, damages the endothelial cell glycocalyx and that this might contribute to the pathophysiological changes during reperfusion.⁷ The endothelial cell glycocalyx is an as-yet poorly understood matrix that lines the luminal surface of all blood vessels. This matrix is composed of glycosaminoglycans, proteoglycans, and glycoproteins originating from both the endothelial cells and adsorbed from circulating plasma. These molecules form a surface layer or glycocalyx, whose thickness is between 0.2 and 0.5 µm in various in vivo rodent models.^8–11 There is increasing evidence that the glycocalyx is important for physiological and pathophysiological regulation of microcirculatory function. It may play a role in the barrier properties of the capillary wall as a whole,^11–14 and evidence suggests that hematrocrit,^15–17 permeability,^12,13,18 blood flow,^19,20 and leukocyte adhesion²¹ may all be regulated by the glycocalyx.

Given the potential contribution of the glycocalyx to the permeability barrier and the role of the A3AR in inducing adenosine-mediated edema, we hypothesized that topical application of adenosine in the appropriate concentration would cause changes in the barrier function of the glycocalyx and perhaps in the physical thickness of the layer. In addition, we hypothesized that this response would be mediated by the A3AR. To test our hypothesis, we used intravital microscopy of the cremaster muscle of mouse and hamster as a model of the microcirculation and measured the penetration of a 70-kDa, FITC-labeled dextran (dextran) into the glycocalyx as well as the spacing between the red cell column and the capillary wall. Here we present evidence that adenosine causes a dose-dependent decrease in ability of the glycocalyx to exclude dextran (increased porosity) and a reduction in the thickness of the layer, which is mediated via the A3AR.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Summary References

 
Animal Preparation
All animal experimentation was done in accordance with institutional guidelines under approved protocols. Cremaster muscles were studied in Syrian hamsters (Charles River Laboratories Inc, Wilmington, Mass) and mice (Hilltop Laboratories, Scottsdale, Ariz). The former provided comparison data with much of the prior work on this subject, and the latter species was used to allow us to make use of genetically modified strains. Male Syrian golden hamsters (n=22) weighing 110 to 150 g were anesthetized with intraperitoneal sodium pentobarbital (70 mg/kg body weight in saline). A tracheostomy was performed to ensure a patent airway, and the left femoral vein was cannulated for continuous infusion of 0.9% saline containing 10 mg/mL sodium pentobarbital (0.5 mL/h). The right femoral vein was also cannulated to allow infusion of FITC-dextran solutions. The cremaster muscle was prepared for experimentation as previously described.^8–11 After at least 45 minutes of stabilization, during which time arteriolar tone was evaluated as a measure of overall tissue viability, a bolus of 0.2 mL dextran (20 mg/mL in saline) was given via the right femoral vein cannula.

Male C57bl/6 mice (n=14), as well as A₃ knockout (KO) mice (n=5),²² were used. A3AR KO animals were generously provided by Marlene Jacobson (Merck Research Labs, West Point, Pa). These mice have been well characterized previously.^22–25 Mice were anesthetized with intraperitoneal sodium pentobarbital (40 mg/kg in saline) and surgically prepared as above, and a bolus of 0.05 mL FITC-dextran 70 (20 mg/mL in saline) was given. Mouse weights were between 22 and 27 g.

Intravital Microscopy
Capillaries in both hamsters (5.15±0.01 µm) and mice (5.23±0.01 µm) were observed with a x60 (NA 0.9) water immersion objective. Transillumination for bright field measurements was obtained with a 150-W xenon lamp. There was no statistical difference between the bright field capillary dimensions of hamsters or mice. Light from another xenon lamp equipped with a 450- to 490-nm excitation filter, a dichroic beam splitter (FT 510), and a barrier filter (LP 520) was used for epi-illumination of the fluorescent tracers. Epi-illumination of each capillary was limited to <10 seconds to prevent light-dye injury to the endothelium. Images of vessels were displayed on an in-line video monitor and recorded on S-VHS videotapes.

Glycocalyx Measurements
We used two separate measurements to assess the function of the glycocalyx before and after treatment with adenosine and other stimuli. For the first, we measured the space between the endothelial cell surface, as determined by bright field microscopy, and the edge of the dextran dye column. Under control conditions, a well-defined space could be measured, and the exclusion of, or penetration of dye into this region was used to determine the ability of the endothelial cell surface matrix to exclude macromolecules (porosity). The space between the endothelial cell surface and the average edge of the red cell column was also measured. The size of this space is determined by a combination of the thickness of the endothelial cell surface layer and the lubrication layer surrounding the moving red cells.⁸

Experimental Protocols
Baseline measurements of the dye exclusion and the red cell exclusion zones were obtained before addition of any stimulatory molecules. These data were averaged and are represented as T=0 for time curves. Test agents (adenosine, papaverine, and ATL-146e) were applied continuously for a period of 30 minutes to the cremaster via the superfusion solution. None of the agents used changed capillary diameter, although arteriolar diameter was affected by all. Because of the variability of the mice and preparations, it was difficult to take measurements at strictly defined time points; therefore, we collected data as often as possible and averaged them into 5-minute blocks to generate time curves. Additionally, we pooled the data after treatments to compare directly to pooled pretreatment values.

Data and Statistical Analysis
Recorded video images of the experiments were analyzed using the Image-1 software package as previously described.^8–10 For each capillary, the anatomical diameter, width of dextran column, and red cell diameter were measured using calibrated video calipers. All data are expressed as mean±SE. Mean data were compared using ANOVA and paired Student’s t tests where appropriate. Differences were considered significant at P<0.05. Unless otherwise noted, all reagents were purchased from Sigma. The ATL-146e was generously provided by Dr Joel Linden (University of Virginia, Charlottesville, Va).

	Results

Top Abstract Introduction Materials and Methods Results Discussion Summary References

 
The baseline exclusion zone for dextran was 0.31±0.02 µm in hamsters (56 control vessels) and 0.22±0.01 µm in mice (102 vessels,. Baseline red cell exclusion was 0.49±0.01 µm in hamsters and 0.47±0.01 µm in mice

fig.ommitted Figure 1. Comparison of baseline and adenosine data between hamsters and wild-type mice. A, Comparison of hamster baseline glycocalyx dye exclusion with that of the mouse. The mouse glycocalyx is slightly less able to exclude 70-kDa FITC-dextran. Both species respond similarly to 1 µmol/L adenosine (B). The red cell exclusion of both species is identical (C). *Different from hamster baseline; $Different from mouse baseline, P<0.05.

Adenosine treatment (1 µmol/L) caused a significant decrease in the dye exclusion zone in both hamsters and mice . The dye exclusion zone in hamsters was reduced by 70% to 0.10±0.01 µm and in mice was reduced by 57% to 0.10±0.01 µm. In hamsters, additional experiments were conducted to elucidate the dose dependence of the glycocalyx effect . A low dose of 0.01 µmol/L adenosine elicited no response from the glycocalyx, whereas the highest dose of 100 µmol/L decreased the dextran exclusion zone 83% to 0.05±0.01 µm. Only the highest dose, however, had an effect on the red cell exclusion zone, which was reduced 23% to 0.38±0.01 µm from a control of 0.5±0.01 µm

fig.ommitted Figure 2. Adenosine causes a dose-dependent effect in the glycocalyx. Topical application of adenosine in hamsters causes a dose-dependent decrease in the dye exclusion properties (A), whereas only the highest dose of adenosine has any effect on the ability of the glycocalyx to exclude red cells (B). *Different from control (no adenosine), P<0.05.

Dextran exclusion data from mice treated with adenosine (10^-6 mol/L) were additionally analyzed to determine the time dependence of this response.  shows that adenosine decreases the dye exclusion zone as early as we can measure it (6 minutes). This decrease is sustained over the course of the 25-minute measurement period.

fig.ommitted   Figure 3. Reaction to adenosine is rapid and sustained. Time course of adenosine (1 µmol/L) effect on the dye exclusion properties in wild-type mice. Time 0 represents pooled control measurements. Note the rapid response of the glycocalyx to adenosine. The kinetics of this response are similar to those seen in hamsters.

Adenosine has effects on many cell types, including smooth muscle, and thus it produces substantial hemodynamic change in addition to effects mediated via the A₃ receptor. To assess the effects of vasodilation (and thus a change in capillary shear stress) per se, we applied papaverine, which causes arteriolar vasodilation without endothelial cell activation.²⁶ Papaverine produced dilations similar to those seen with adenosine but did not change either the dextran or the red cell exclusion zones

fig.ommitted   Figure 4. Papaverine has no effect on glycocalyx measurements. Papaverine was used to induce a similar vasomotor response in wild-type mice but did not cause any change in the glycocalyx properties.

To determine which adenosine receptor was responsible for the effect on the glycocalyx, we treated mice deficient in the A3AR with 1 µmol/L adenosine as described above. As seen in Figure 5, the endothelial glycocalyx of A₃ KO mice exhibited normal baseline dimensions but failed to respond to adenosine. The dye exclusion zone was 0.22±0.01 µm compared with 0.23±0.02 µm for control c57bl/6 mice. We also confirmed the lack of involvement of the A_2A receptor by treating one group of mice with the A_2A agonist ATL-146e (100 nmol/L). Activation of the A_2A receptor had no effect on the glycocalyx, and the data closely resemble those for papaverine (0.21±0.02 µm control versus 0.21±0.02 µm ATL for both dyes and 0.47±0.02 µm control versus 0.46±0.01 µm ATL for red blood cell–free zone). In fact, these data support the papaverine findings, because this dose of ATL elicits a vasodilation but no glycocalyx change.

fig.ommitted Figure 5. Adenosine effect on dye exclusion is absent in A3 KO mice. Genetically modified mice lacking the A3AR were treated with 1 µmol/L adenosine using the same protocol as wild-type mice. Whereas control mice exhibited the expected decrease in glycocalyx exclusion to 70-kDa FITC-dextran, the effect was completely absent in KO mice. *Different from wild-type control, P<0.05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Summary References

 
There is mounting evidence that the endothelial glycocalyx influences several microvascular parameters, including capillary tube hematocrit,^15,17,27 blood flow,²⁰ permeability,^12,18 and leukocyte adhesion.²¹ More recently, it has become apparent that the glycocalyx is not a static anatomical barrier but is a dynamic, regulated structure that is sensitive to modulation by cytokines,¹⁰ free radicals,^8,15,28 and several physiologically relevant enzymes.^11,12,17,18

Adenosine is a ubiquitous metabolic byproduct that mediates a variety of microvascular functions. The most relevant to this study is its ability, as well as that of its metabolite inosine, to inhibit (low concentrations) or stimulate (higher concentrations) mast cell degranulation.^5,6,22,29 The inhibitory effect at low concentrations of adenosine (100 nmol/L) is mediated by the mast cell A_2A receptor,⁵ and the degranulation at higher concentrations (>1 µmol/L) is caused by binding to the mast cell A3AR.^6,29,30It has also been found that adenosine-mediated edema formation is eliminated in both A₃ KO mice and in mast cell–deficient mice.²³ Thus, we hypothesized that increased tissue adenosine would lead to an alteration of glycocalyx exclusion characteristics or a decrease in glycocalyx thickness. Additionally, we hypothesized that this response would be mediated via the A3AR, possibly leading to mast cell degranulation.

There are presently four recognized adenosine receptors, A₁, A_2A, A_2B, and A₃.^1,2 Of these, we focused on the A₃ receptors, because they are found on a variety of vascular cells and have been shown to be involved in proinflammatory responses. The A₃ receptor has also been found to mediate adenosine-induced edema in mice via mast cell degranulation,²³ and we have previously shown that stimulation of the A₃ receptor causes mast cell-mediated vasomotor changes.⁶ Thus, we tested the hypothesis that the adenosine effects on the glycocalyx were also attributable to A₃ receptor activation by using A3AR KO mice. These mice exhibited no response of the glycocalyx after treatment with adenosine (Figure 5). Baseline glycocalyx values preceding treatment are normal. Combined with our data showing that the A_2A agonist ATL-146 has no effect on the glycocalyx, these data show that adenosine-mediated changes in the glycocalyx are produced by A₃ receptor activation in the mouse cremaster microcirculation.

We have summarized some possible factors linking adenosine to the glycocalyx in Figure 6. Adenosine receptors are found on several different cell types that are intimately involved with microvascular regulation. Although vascular smooth muscle cells express the A₃ receptor,^25,31 we believe that the papaverine data combined with the ATL-146e data make it unlikely that vascular smooth muscle regulates the glycocalyx through changes in intravascular pressure, blood flow, or shear stress. Likewise, leukocytes are not known to express the A₃ receptor, although they do express A_2A receptors. The ATL-146e data argue against a role of leukocytes in adenosine-mediated modification of the glycocalyx, because stimulation of the A_2A receptor was without effect.

fig.ommitted   Figure 6. Possible mechanisms involved in adenosine damage to glycocalyx. The papaverine data suggest that the route through changes in shear stress is not involved. Likewise, the lack of A3 receptor expression of leukocytes makes it unlikely that they play a role. Of the 2 remaining cell types, previous work supports the mast cells as the primary cell type involved in adenosine-mediated modulation of the glycocalyx.

Endothelial cells express both the A_2A and the A₃ receptors.^2,25 It is possible that A₃ stimulation leads to production of an autocrine factor that leads to the glycocalyx changes we observed. There is evidence that the endothelium produces reactive oxygen species, and there is substantial evidence that reactive oxygen species (ROS) lead to degradation of the glycocalyx.^8,15,28,32 However, evidence produced by this (and other) laboratories points to another cell type as the primary initiator of tissue injury via adenosine treatment, the mast cell.

The mast cell route seems to be the most likely mode of action of adenosine because of the convincing evidence by Tilley et al²³ that adenosine-mediated edema is absent in both A₃ KO mice and in mast cell–deficient mice (WBB6F₁/J-W/W^v). We have also shown that A3AR stimulation causes vasomotor changes that are mediated by the mast cell.⁶ The mast cell degranulation product responsible for opening the endothelial cell surface matrix and allowing the dextran to penetrate is not yet known, although there is evidence that ROS,^8,28 tumor necrosis factor- (TNF-),¹⁰ and numerous enzymes^11,17 modulate the porosity and the red cell exclusion zone of the glycocalyx.

Degradation of either protein or carbohydrate chains comprising the matrix might in part be the result of a direct effect of ROS on the matrix components. However, the ultimate effects of TNF- and ROS may be much more complex. TNF- has been shown to activate several endothelial cell signaling cascades, leading to the activation of mitogen-activated protein kinases, AKT, c-Jun NH₂ terminal kinase, and phosphatidylinositol 3 kinase and possibly src.³³ In addition, TNF- also increases [Ca²⁺], and IP₃ and triggers ROS production.³⁴ Many of these signals may be involved in active regulation of the glycocalyx by the endothelial cell. In addition, Murphy et al³⁴ have reported that treatment of endothelial cells with TNF- leads to production of O₂^-, and we have shown that ROS affect the glycocalyx.⁸ Moreover, there is abundant evidence that ROS can directly affect glycosaminoglycan and proteoglycan structure,^35–37 all of which suggests that a direct effect of ROS may be key in the modification of the glycocalyx that we observe.

We have shown previously^7,10and in this study that the ability of the glycocalyx to exclude macromolecules can be differentially altered compared with the red cell exclusion component. Enzymatic evidence^11,17 has lead us to hypothesize that this difference reflects the complex composition of the glycocalyx, where hyaluronan (HA) principally regulates porosity and proteoglycans regulate thickness of the layer. This concept fits well with data from Moseley et al,³⁵ who showed that O₂^- generated by neutrophils degraded HA as well as chondroitin sulfate and dermatan sulfate. Additionally, they found that the HA was more sensitive to ROS degradation than were the sulfated glycosaminoglycans. This could explain how low doses of adenosine caused a change in porosity without affecting red cell exclusion whereas high doses affected both, although to different degrees. This difference in sensitivity may also be responsible for the differences seen between the baseline glycocalyx porosity of mice and hamsters in Figure 1, although there are several possible explanations. Additional experiments with a variety of molecular probes of porosity and a deeper understanding of the synthesis will be required to fully understand these issues.

	Summary

Top Abstract Introduction Materials and Methods Results Discussion Summary References

 
These data show that the capillary endothelial glycocalyx is modulated by stimulation of the A₃ adenosine receptor. Although we do not show the target cell responsible for this effect, adenosine-induced increases in permeability have been shown to be absent in mast cell KO mice.²³ We believe this is attributable, at least in part, to modification of glycocalyx function. The implications of this are of considerable interest because of the number of pathological conditions that lead to changes in microvascular permeability. In addition, modification of the glycocalyx has been shown to play a role in the adhesion of leukocytes²¹ and in ischemia/reperfusion injury.⁷ Although the function of the glycocalyx has not been evaluated in a sufficient number of experimental models, it is tempting to hypothesize that an early response during inflammatory vascular injury is modification of the glycocalyx. This provides a therapeutic target where there is potential for many pathological processes to converge and may allow for significant progress in ameliorating the consequences.

　

　

	Acknowledgments

 
This work was supported by US Public Health Service Grant HL12792 to B.R.D. and HL 69640 to S.H.P. We wish to thank David N. Damon and Kathleen H. Day for their excellent technical assistance. Some of these data were previously published in abstract form.

	Footnotes

 
Some of these data were previously published in abstract form (FASEB J. 2001;15:A44).

Original received May 12, 2003; first resubmission received August 15, 2003; second resubmission received November 5, 2003; accepted November 11, 2003.

	References

Top Abstract Introduction Materials and Methods Results Discussion Summary References

 

1. Fredholm BB, IJzerman AP, Jacobson KA, Klotz KN, Linden J. International Union of Pharmacology, XXV: nomenclature and classification of adenosine receptors. Pharmacol Rev. 2001; 53: 527–552.

2. Linden J. Molecular approach to adenosine receptors: receptor-mediated mechanisms of tissue protection. Annu Rev Pharmacol Toxicol. 2001; 41: 775–787.

3. Blackburn MR. Too much of a good thing: adenosine overload in adenosine-deaminase-deficient mice. Trends Pharmacol Sci. 2003; 24: 66–70.

4. Cronstein BN. Adenosine, an endogenous anti-inflammatory agent. J Appl Physiol. 1994; 76: 1:5–13.

5. Fenster MS, Shepherd RK, Linden J, Duling BR. Activation of adenosine A₂ receptors inhibits mast cell degranulation and mast cell-dependent vasoconstriction. Microcirculation. 2000; 7: 129–135.

6. Shepherd RK, Linden J, Duling BR. Adenosine-induced vasoconstriction in vivo: role of the mast cell and A₃ adenosine receptor. Circ Res. 1996; 78: 627–634.

7. Platts SH, Linden J, Duling BR. Rapid modification of the glycocalyx caused by ischemia-reperfusion is inhibited by adenosine A_2A receptor activation. Am J Physiol Heart Circ Physiol. 2003; 284: H2360–H2367.

8. Vink H, Duling BR. Identification of distinct luminal domains for macromolecules, erythrocytes, and leukocytes within mammalian capillaries. Circ Res. 1996; 79: 581–589.

9. Vink H, Duling BR. Capillary endothelial surface layer selectively reduces plasma solute distribution volume. Am J Physiol Heart Circ Physiol. 2000; 278: H285–H289.

10. Henry CB, Duling BR. TNF- increases entry of macromolecules into luminal endothelial cell glycocalyx. Am J Physiol Heart Circ Physiol. 2000; 279: H2815–H2823.

11. Henry CBS, Duling BR. Permeation of the luminal capillary glycocalyx is determined by hyaluronan. Am J Physiol Heart Circ Physiol. 1999; 46: H508–H514.

12. Huxley VH, Williams DA. Role of a glycocalyx on coronary arteriole permeability to proteins: evidence from enzyme treatments. Am J Physiol Heart Circ Physiol. 2000; 278: H1177–H1185.

13. Adamson RH. Increased permeability of mesenteric capillaries of pithed frogs associated with mild protease digestion of the endothelial glycocalyx. J Physiol. 1988; 407: 27P. Abstract.

14. Adamson RH, Clough G. Plasma proteins modify the endothelial cell glycocalyx of frog mesenteric microvessels. J Physiol. 1992; 445: 473–486.

15. Constantinescu AA, Vink H, Spaan JA. Elevated capillary tube hematocrit reflects degradation of endothelial cell glycocalyx by oxidized LDL. Am J Physiol Heart Circ Physiol. 2001; 280: H1051–H1057.

16. Secomb TW, Hsu R, Pries AR. Blood flow and red blood cell deformation in nonuniform capillaries: effects of the endothelial surface layer. Microcirculation. 2002; 9: 189–196.

17. Desjardins C, Duling BR. Heparinase treatment suggests a role for the endothelial cell glycocalyx in regulation of capillary hematocrit. Am J Physiol. 1990; 258: H647–H654.

18. Adamson RH. Permeability of frog mesenteric capillaries after partial pronase digestion of the endothelial glycocalyx. J Physiol. 1990; 428: 1–13.

19. Vogel J, Sperandio M, Pries AR, Linderkamp O, Gaehtgens P, Kuschinsky W. Influence of the endothelial glycocalyx on cerebral blood flow in mice. J Cereb Blood Flow Metab. 2000; 20: 1571–1578.

20. Pries AR, Secomb TW, Jacobs H, Sperandio MB, Osterloh K, Gaehtgens PAL. Microvascular blood flow resistance: role of endothelial surface layer. Am J Physiol Heart Circ Physiol. 1997; 273: H2272–H2279.

21. Mulivor AW, Lipowsky HH. Role of glycocalyx in leukocyte-endothelial cell adhesion. Am J Physiol Heart Circ Physiol. 2002; 283: H1282–H1291.

22. Salvatore CA, Tilley SL, Latour AM, Fletcher DS, Koller BH, Jacobson MA. Disruption of the A₃ adenosine receptor gene in mice and its effect on stimulated inflammatory cells. J Biol Chem. 2000; 275: 4429–4434.

23. Tilley SL, Wagoner VA, Salvatore CA, Jacobson MA, Koller BH. Adenosine and inosine increase cutaneous vasopermeability by activating A₃ receptors on mast cells. J Clin Invest. 2000; 105: 361–367.

24. Guo Y, Bolli R, Bao W, Wu WJ, Black RG Jr, Murphree SS, Salvatore CA, Jacobson MA, Auchampach JA. Targeted deletion of the A₃ adenosine receptor confers resistance to myocardial ischemic injury and does not prevent early preconditioning. J Mol Cell Cardiol. 2001; 33: 825–830.

25. Zhao Z, Makaritsis K, Francis CE, Gavras H, Ravid K. A role for the A₃ adenosine receptor in determining tissue levels of cAMP and blood pressure: studies in knock-out mice. Biochim Biophys Acta. 2000; 1500: 280–290.

26. Iguchi M, Nakajima T, Hisada T, Sugimoto T, Kurachi Y. On the mechanism of papaverine inhibition of the voltage-dependent Ca²⁺ current in isolated smooth muscle cells from the guinea pig trachea. J Pharmacol Exp Ther. 1992; 263: 194–200.

27. Klitzman B, Duling BR. Microvascular hematocrit and red cell flow in resting and contracting striated muscle. Am J Physiol. 1979; 237: H481–H490.

28. Vink H, Constantinescu AA, Spaan JA. Oxidized lipoproteins degrade the endothelial surface layer: implications for platelet-endothelial cell adhesion. Circulation. 2000; 101: 1500–1502.

29. Doyle MP, Linden J, Duling BR. Nucleoside-induced arteriolar constriction: a mast cell-dependent response. Am J Physiol. 1994; 266: H2042–H2050.

30. Reeves JJ, Jones CA, Sheehan MJ, Vardey CJ, Whelan CJ. Adenosine A₃ receptors promote degranulation of rat mast cells both in vitro and in vivo. Inflamm Res. 1997; 46: 180–184.

31. Deguchi H, Takeya H, Urano H, Gabazza EC, Zhou H, Suzuki K. Adenosine regulates tissue factor expression on endothelial cells. Thromb Res. 1998; 91: 57–64.

32. Beresewicz A, Czarnowska E, Maczewski M. Ischemic preconditioning and superoxide dismutase protect against endothelial dysfunction and endothelium glycocalyx disruption in the postischemic guinea-pig hearts. Mol Cell Biochem. 1998; 186: 87–97.

33. Madge LA, Pober JS. TNF signaling in vascular endothelial cells. Exp Mol Pathol. 2001; 70: 317–325.

34. Murphy HS, Shayman JA, Till GO, Mahrougui M, Owens CB, Ryan US, Ward PA. Superoxide responses of endothelial cells to C5a and TNF-: divergent signal transduction pathways. Am J Physiol. 1992; 263: L51–L59.

35. Moseley R, Waddington RJ, Embery G. Degradation of glycosaminoglycans by reactive oxygen species derived from stimulated polymorphonuclear leukocytes. Biochim Biophys Acta. 1997; 1362: 221–231.

36. Hawkins CL, Davies MJ. Direct detection and identification of radicals generated during the hydroxyl radical-induced degradation of hyaluronic acid and related materials. Free Radic Biol Med. 1996; 21: 275–290.

37. Hawkins CL, Davies MJ. Degradation of hyaluronic acid, poly- and monosaccharides, and model compounds by hypochlorite: evidence for radical intermediates and fragmentation. Free Radic Biol Med. 1998; 24: 1396–1410.