Gail W. Sullivan

The role of inflammation in vascular diseases

When the body responds to an infectious insult, it initiates an immune response to eliminate the pathogen. The hallmark of the immune response is an inflammatory cascade that can also do extensive damage to host tissues. Inflammation is a major contributing factor to many vascular events, including atherosclerotic plaque development and rupture, aortic aneurysm formation, angiogenesis, and ischemia/reperfusion damage. The immune response is mediated by both circulating and resident leukocytes and the cells with which they interact (e.g., vascular endothelium and smooth muscle cells). The process is orchestrated by the activity of a changing series of released and displayed mediators. These include the expression of adhesion molecules on leukocytes and underlying vascular endothelium and the release of cytokines, chemokines, and tissue‐destructive metalloproteases and reactive oxygen species. This review focuses on the causes, the inflammatory processes involved, and possible strategies for decreasing vascular disease through regulation of the inflammatory response. J. Leukoc. Biol. 67: 591–602; 2000.

Cyclic AMP‐dependent inhibition of human neutrophil oxidative activity by substituted 2‐propynylcyclohexyl adenosine A2A receptor agonists

Novel 2‐propynylcyclohexyl‐5′‐N‐ehtylcarboxamidoadenosines, trans‐substituted in the 4‐position of the cyclohexyl ring, were evaluated in binding assays to the four subtypes of adenosine receptors (ARs). Two esters, 4‐{3‐[6‐amino‐9‐(5‐ethylcarbamoyl‐3,4‐dihydroxy‐tetrahydro‐furan‐2‐yl)‐9H‐purin‐2‐yl]‐prop‐2‐ynyl}‐cyclohexanecarboxylic acid methyl ester (ATL146e) and acetic acid 4‐{3‐[6‐amino‐9‐(5‐ethylcarbamoyl‐3, 4‐dihydroxy‐tetrahydro‐furan ‐2‐yl)‐9H‐purin‐2‐yl] ‐prop‐2‐ynyl}‐cyclohexylmethyl ester (ATL193) were >50×more potent than 2‐[4‐(2‐carboxyethyl)phenethylamino]‐5′‐N‐ethylcarboxamidoadenosine (CGS21680) for human A2A AR binding. Human A2A AR affinity for substituted cyclohexyl‐propynyladenosine analogues was inversely correlated with the polarity of the cyclohexyl side chain. There was a comparable order of potency for A2A AR agonist stimulation of human neutrophil [cyclic AMP]i, and inhibition of the neutrophil oxidative burst. ATL146e and CGS21680 were ∼equipotent agonists of human A3 ARs. We measured the effects of selective AR antagonists on agonist stimulated neutrophil [cyclic AMP]i and the effect of PKA inhibition on A2A AR agonist activity. ATL193‐stimulated neutrophil [cyclic AMP]i was blocked by antagonists with the potency order: ZM241385 (A2A‐selective)>MRS1220 (A3‐selective)>>N‐(4‐Cyano‐phenyl)‐2‐[4‐(2,6‐dioxo‐1,3‐dipropyl‐2,3,4,5,6,7‐hexahydro‐1H‐purin‐8‐yl)‐phenoxy]‐acetamide (MRS1754; A2B‐selective) ∼amp; 8‐(N‐methylisopropyl)amino‐N6‐(5′‐endohydroxy‐endonorbornyl)‐9‐methyladenine (WRC0571; A1‐selective). The type IV phosphodiesterase inhibitor, rolipram (100 nM) potentiated ATL193 inhibition of the oxidative burst, and inhibition by ATL193 was counteracted by the PKA inhibitor H‐89. The data indicate that activation of A2AARs inhibits neutrophil oxidative activity by activating [cyclic AMP]i/PKA.

Neutrophil A2A Adenosine Receptor Inhibits Inflammation in a Rat Model of Meningitis: Synergy with the Type IV Phosphodiesterase Inhibitor, Rolipram

Bacterial meningitis is a disease worsened by neutrophil-induced damage in the subarachnoid space. In this study, the A2A adenosine receptors on human neutrophils were characterized, and the role of A2A receptors on the trafficking of leukocytes to the cerebrospinal fluid and on blood-brain barrier permeability (BBBP) was assessed in a rat meningitis model. Neutrophils bind the A2A selective antagonist, 125I-ZM241385 (Bmax=843 receptors/neutrophil; KD=0.125 nM). A selective A2A receptor agonist, WRC-0470 (2-cyclohexylmethylidene-hydrazinoadenosine; 0.03-1 microM), alone and synergistically with the type IV phosphodiesterase inhibitor, rolipram, increased neutrophil [cAMP]i and reduced cytokine-enhanced neutrophil adherence, superoxide release, and degranulation. These effects of WRC-0470 were reversed by ZM241385 (100 nM). In a lipopolysaccharide-induced rat meningitis model, WRC-0470 (0-0.9 microgram/kg/h), with or without rolipram (0-0.01 microgram/kg/h), inhibited pleocytosis and reduced the lipopolysaccharide-induced increase in BBBP, indicative of decreased neutrophil-induced damage.

Role of A2A adenosine receptors in inflammation

A2A adenosine receptors are expressed on immune cells including neutrophils, lymphocytes, eosinophils, monocytes/macrophages, and mast cells. Activation of A2A receptors on these cells stimulates an increase in [cyclic AMP]i and causes a diminution of inflammatory responses. In mast cells, degranulation is inhibited; in neutrophils, adherence is reduced and the release of reactive oxygen species is inhibited; in monocytes, differentiation to macrophages is inhibited and the release of tumor necrosis factor‐α is inhibited; and in lymphocytes, TCR‐triggered interleukin‐2 α chain (CD25) up‐regulation is reduced. In vivo, selective adenosine A2A agonists decrease inflammation in both infectious and noninfectious models. High concentrations (micromolar) of the A3 selective agonist, IB‐MECA, produce anti‐inflammatory responses that are mediated by A2A receptors. Selective activation of A2A adenosine receptors with pharmaceutical agents may be a useful strategy for ameliorating an inappropriate and/or an extensive inflammatory response. Drug Dev. Res. 45:103–112, 1998.

Adenosine A2A agonists in development for the treatment of inflammation

Extracellular adenosine binds specifically to a family of four G protein-coupled cell-surface adenosine receptors (ARs). As the activation of the A2AAR modulates the activity of multiple inflammatory cells including neutrophils, macrophages and T lymphocytes, the receptor is considered to be a promising pharmacological target for the treatment of inflammatory disorders. Although adenosine binds nonselectively to all four AR subtypes, A2AAR selective agonists have been developed and shown to inhibit multiple manifestations of inflammatory cell activation including superoxide anion generation, cytokine production and adhesion molecule expression. A2AAR agonists are also vasodilators, but the inhibition of inflammation occurs at low doses that produce few or no cardiovascular side effects. Therefore, the selective activation of the A2AAR by these compounds holds significant potential in the treatment of inflammation.

Both recombinant interleukin-1 (beta) and purified human monocyte interleukin-1 prime human neutrophils for increased oxidative activity and promote neutrophil spreading.

Both purified human monocyte interleukin‐1 and recombinant interleukin‐1 (beta) primed neutrophils for increased superoxide production and chemiluminescence in response to f‐met‐leu‐phe. In addition, purified human monocyte interleukin‐1 and recombinant interleukin‐1 (beta) altered neutrophil shape. Recombinant interleukin‐1 (alpha) used at the same concentration of interleukin‐1 (beta) did not prime neutrophils for increased superoxide production after stimulation with f‐met‐leu‐phe. interleukin‐1 expressed by monocytes in response to endotoxin stimulation could act as a modulator of neutrophil function.

In vitro demonstration of transport and delivery of antibiotics by polymorphonuclear leukocytes.

Several antibiotics are concentrated inside polymorphonuclear leukocytes (PMN). To investigate whether PMN could act as vehicles for delivery of antibiotics, we combined an assay measuring PMN chemotaxis under agarose with a bioassay measuring levels of antibiotic in agar. Double-layer plates were made by pouring a layer of chemotaxis agarose into tissue culture plates and then adding a thin layer of Trypticase soy agar. Neutrophils were incubated with antibiotic for 1 h and then were washed and placed in wells made in the plates. After allowing PMN to migrate under the agar toward a chemoattractant well containing formyl-methionine-leucine-phenylalanine for 3 h, Streptococcus pyogenes was streaked on top of the agar and grown overnight. PMN migration and zones of inhibition of bacterial growth were measured. Neutrophils migrated 2.51 +/- 0.16 mm toward the chemoattractant well and 1.48 +/- 0.12 mm toward the medium well; migration was not significantly affected by any of the antibiotics used. Plates with PMN incubated without antibiotic showed insignificant inhibition of bacterial growth toward chemoattractant and medium wells (0.38 +/- 0.18 and 0.14 +/- 0.12 mm, respectively; for both, P > 0.05 for difference from 0). PMN incubated with oxacillin (3 micrograms/ml), a drug not concentrated in PMN, caused a similar lack of inhibition (0.28 +/- 0.09 mm toward chemoattractant; 0.14 +/- 0.03 mm toward medium). Incubation with 30 microns of ciprofloxacin per ml resulted in inhibition that was similar in both directions (1.40 +/- 0.16 versus 1.18 +/- 0.13 mm). However, for PMN incubated with azithromycin (3 micrograms/ml), an agent highly concentrated inside phagocytes, there was a large degree of inhibition which was significantly greater in the direction of chemoattractant than in the direction of medium (3.47 +/- 0.30 versus 1.89 +/- 0.25 mm; P < 0.001), indicating that release of bioactive azithromycin by neutrophils occurred after migration. Likewise, after incubation with rifampin (10 micrograms/ml), which is also concentrated by PMN, inhibition was significantly greater in the direction of chemoattractant than in the direction of medium (1.54 +/- 0.24 versus 0.81 +/- 0.28 mm; P = 0.001). We conclude that for certain antibiotics, PMN may act as vehicles for transport and delivery of active drug to sites of infection. Images

Adenosine modulation of tumor necrosis factor-α-induced neutrophil activation

We hypothesized that adenosine, known to be release from inflammatory sites, could lessen the potentially damaging activity of neutrophils (PMN) primed by tumor necrosis factor-alpha (TNF alpha) at sites of infection. We investigated the effect of adenosine on PMN primed with cell-free medium from mononuclear leukocytes (MNL) that had been treated with lipopolysaccharide (LPS) yielding a conditioned medium rich in TNF alpha and on PMN primed with recombinant human TNF alpha (rhTNF alpha). LPS (10 ng/mL) minimally primed PMN, but LPS-MNL-conditioned medium increased PMN chemiluminescence in response to f-Met-Leu-Phe (fMLP) 1242% compared with unprimed PMN. LPS-MNL-conditioned medium contained adenosine (approximately 30 nM). Converting the adenosine in the LPS-MNL-conditioned medium to inosine with adenosine deaminase (ADA) or blocking adenosine binding to PMN with the adenosine receptor antagonist 1,3-dipropyl-8-(phenyl-p-acrylate)-xanthine (BW A1433U) resulted in a near doubling of chemiluminescence. The LPS-MNL-conditioned medium contained TNF alpha (836 pg/mL; approximately 1 U/mL). Recombinant human TNF alpha (1 U/mL) primed PMN for a 1033% increase in chemiluminescence. Added adenosine decreased rhTNF alpha-primed PMN chemiluminescence (IC50 approximately 100 nM), and adenosine (100 nM) decreased both superoxide and myeloperoxidase release from rhTNF alpha-primed fMLP-stimulated PMN. The activity of adenosine was counteracted by ADA and BW A1433U, and the modulating effect of adenosine was on the primed response rather than on priming per se. Thus, physiological concentrations of adenosine reduce the effects of recombinant human TNF alpha and native human TNF alpha (released from LPS-treated MNL) on PMN activity. Endogenous adenosine may preclude or minimize damage to infected tissue by damping the TNF alpha-primed PMN oxidative response.

Pentoxifylline modulates activation of human neutrophils by amphotericin B in vitro.

The antifungal agent amphotericin B (AmB) alters neutrophil (polymorphonuclear leukocyte [PMN]) function, and this may be the mechanism for some of the adverse effects caused by AmB. AmB is a potent inhibitor of PMN migration, increases PMN adherence and aggregation, and primes PMN for increased oxidative activity in response to a second stimulus. AmB also stimulates mononuclear leukocytes (MNLs) to release inflammatory mediators which augment the effects of AmB on PMN function. In the present study, we observed that the methylxanthine derivative pentoxifylline decreased the effects of AmB on PMN function. AmB (2 micrograms/ml) priming doubled PMN chemiluminescence stimulated by fMet-Leu-Phe. In the presence of MNLs, AmB priming increased fMet-Leu-Phe-stimulated PMN chemiluminescence to 622% of unprimed PMN activity. Pentoxifylline (100 microM) blunted the rise in AmB-augmented PMN chemiluminescence in the presence of MNLs to 282% of unprimed PMN activity, and pentoxifylline metabolites were active at 10 microM. Pentoxifylline (100 microM) also blocked AmB-augmented PMN oxidative activity in whole blood, as measured by nitroblue tetrazolium reduction. In the presence of MNL, AmB (2 micrograms/ml) doubled the expression of the important PMN adherence factor Mac-1. Pentoxifylline (1 mM) decreased AmB-stimulated PMN Mac-1 expression back to unstimulated amounts. In the presence of MNLs, AmB (2 micrograms/ml) decreased PMN nondirected and directed migration to fMet-Leu-Phe to 40 and 38% of control PMN migration, respectively. Pentoxifylline (300 microM) counteracted AmB inhibition of nondirected and directed migration to fMet-Leu-Phe, resulting in migration that was 71 and 87% of control PMN migration, respectively. In contrast, the methylxanthine caffeine (100 muM) increased AmB-enhanced chemiluminescence but did not affect AmB-inhibited PMN migration. Pentoxifylline should be evaluated as adjunctive therapy to lessen the inflammatory damage caused by AmB.

Methylxanthines with adenosine alter TNFα-primed PMN activation

Abstract Methylxanthines are best known as phosphodiesterase inhibitors that cause a rise in intracellular cAMP. One would expect the two methylxanthines, caffeine and pentoxifylline, to have similar actions on neutrophils (PMN). However, caffeine stimulated and pentoxifylline inhibited PMN oxidative activity. Micromolar concentrations of pentoxifylline decreased native and recombinant tumor necrosis factor-α (TNFα)-primed formyl met-leu-phe (fMLP)-stimulated PMN chemiluminescence, superoxide production and myeloperoxidase (MPO) release. In contrast, equal concentrations of caffeine increased chemiluminescence and MPO release with no effect on superoxide production. These activities of the methylxanthines were only observed in the presence of physiological concentrations of adenosine, and were abolished by the treatment of the PMN with adenosine deaminase. The activities of adenosine, pentoxifylline and caffeine on PMN activity could not be readily explained by changes in PMN [cAMP]. Thus for TNFα-primed PMN, pentoxifylline decreases PMN activity by enhancing the effect of adenosine on degranulation and superoxide production; whereas caffeine increases PMN activity by counteracting the effect of adenosine on degranulation.