INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ISCHEMIC PRECONDITIONING (IPC) refers to a phenomenon in which a tissue is rendered resistant to the deleterious effects of prolonged ischemia and reperfusion (I/R) by prior exposure to brief periods of vascular occlusion. This phenomenon was first demonstrated in the heart nearly a decade ago (21) and has been the subject of intensive investigation since that time. This work has led to the development of the concept that IPC-induced adenosine A₁-receptor stimulation during the period of preconditioning ischemia increases phospholipase C (PLC) activity, an event that is coupled by pertussis toxin-sensitive G proteins (3, 13, 15). Activation of PLC induces the formation of diacylglycerol, which in turn promotes the translocation and activation of protein kinase C (PKC) to cell membranes. Further activation of translocated PKC may also occur when adenosine receptors are repopulated during prolonged ischemia. Whereas this scheme indicates that adenosine serves as the initiator of IPC, other triggers may also contribute (including ₁-adrenergic receptor activation, bradykinin, opioids, oxidants, and nitric oxide), the relative importance of each depending on the model being studied (3, 13, 15).

Although it is clear that activation of adenosine receptors and PKC are critical to the development of the beneficial actions of preconditioning, the downstream effectors in the signaling cascade initiated by IPC are uncertain. A growing body of evidence suggests that activation of ATP-sensitive potassium (K_ATP) channels may serve as the cellular effectors in the heart (7). More recently, Akimitsu et al. (1), Gute et al. (8, 9), and others (16) have demonstrated that IPC prevents intestinal and skeletal muscle I/R injury by inhibiting postischemic leukocyte-endothelial cell interactions. These latter observations are important because they indicate that in addition to protecting against the deleterious effects of ischemia per se, the ability of IPC to induce cellular changes that also prevent leukocyte recruitment to ischemic tissues may limit the reperfusion component of I/R injury, which is primarily leukocyte dependent in the small intestine and other organs (6, 10, 11). Thus, in addition to K_ATP channels, IPC appears to target effector molecules that modulate the inflammatory response to I/R. A likely candidate effector molecule that may be targeted by the signaling cascade initiated by IPC is P-selectin. This notion is based on the following observations. First, I/R increases P-selectin expression in the small intestine (4). Second, postischemic leukocyte rolling in mesenteric venules is primarily mediated by P-selectin, a concept based on the observation that I/R-induced rolling is abolished by the administration of monoclonal antibodies (MAb) directed at functional epitopes on this adhesive ligand (17). Third, P-selectin-dependent leukocyte rolling is a prerequisite to establishing stationary adhesive interactions, subsequent leukocyte emigration, and microvascular barrier disruption after gut I/R (17). Finally, IPC completely prevents postischemic leukocyte rolling in rat mesenteric postcapillary venules, an effect that was abrogated by adenosine receptor blockade or PKC inhibition (8, 9). From these observations, we hypothesized that IPC abolishes I/R-induced leukocyte adhesion by preventing postischemic P-selectin expression via an adenosine-initiated, PKC-dependent pathway.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal procedures. Male Sprague-Dawley rats (n = 85) weighing 100-125 g were used in the radiolabeled antibody experiments. The rats were anesthetized intraperitoneally with thiobutabarbital (Inactin; 100 mg/kg body wt). After a surgical plane of anesthesia was attained, a tracheostomy was performed to maintain a patent airway during the experiment. Next, the left jugular vein and right carotid artery were cannulated with polyethelene tubing (PE-50). Thereafter, an abdominal incision was performed to expose the superior mesenteric artery (SMA). Ischemia of the small intestine was produced by occlusion of the SMA with a vascular artery clamp for a period specified by the animal protocol after which the tissue was permitted to reperfuse by removing the clamp.

Antibody labeling procedure. P-selectin expression was assessed by the dual radiolabeled MAb approach described by Granger and co-workers (4, 20, 23) in which a binding radiolabeled MAb is administered to quantitate ligand expression together with a nonbinding MAb of the same class to correct for nonspecific antibody accumulation. The MAb used for the in vivo assessment of P-selectin expression were RMP-1, a mouse IgG1 against rat P-selectin (Pharmacia Upjohn, Kalamazoo, MI) and P23, a nonbinding murine IgG1 (Pharmacia Upjohn). The binding MAb directed against P-selectin was labeled with ¹²⁵I (DuPont NEN, Boston, MA), whereas the nonbinding MAb was labeled with ¹³¹I (DuPont NEN). In both instances the chloramine-T method was employed (2). Briefly, a 1-mg aliquot of MAb in 1.5 ml of sodium phosphate buffer (pH 7.4) was added to ¹²⁵I or ¹³¹I, with a total activity of 1.0 mCi, and 100 µg of chloramine-T. The mixture was incubated for 1 min at room temperature, and 62.5 µg of sodium metabisulfite were added. The total volume was brought to 2.5 ml with the addition of sodium phosphate buffer. The coupled MAb was then separated from free ¹²⁵I by gel filtration on a Sephadex PD-10 column (Pharmacia). The column was equilibrated with phosphate buffer containing 1% bovine serum albumin and was eluted with the same buffer. Two 2.5-ml fractions were collected, the second of which contained the labeled antibody. Absence of free ¹²⁵I or ¹³¹I was ensured by extensive dialysis of the protein-containing fraction. Less than 1% of the activity of the protein fraction was recovered from the dialysis fluid. SDS-polyacrylamide gel electrophoresis analysis showed normal heavy and light chain moieties of expected molecular weight. Labeled MAb were stored in 500-µl aliquots at 4°C and used within 3 wk after the labeling procedure. The specific activity of labeled MAb was 0.5 µCi/µg.

P-selectin expression was measured by injecting a mixture of 10 µg of ¹²⁵I-labeled P-selectin MAb, a dose that was determined in preliminary experiments, sufficient to saturate P-selectin expressed on vascular endothelium after exposure to lipopolysacharide (10 µg/kg), and 10⁶ counts ¹³¹I-labeled nonbinding MAb through the jugular vein catheter. Blood samples were obtained through the carotid artery catheter at 2.5 and 5 min after injection of the MAb mixture. The animals were heparinized (1 mg/kg heparin sodium) and rapidly exsanguinated by vascular perfusion of bicarbonate-buffered saline through the jugular catheter with simultaneous blood withdrawal through the blood carotid artery catheter. This was followed by perfusion of bicarbonate-buffered saline through the carotid artery catheter after the inferior vena cava was severed at the thoracic level. Tissue specimens were harvested and weighed.

Calculation of P-selectin expression. The method for calculating P-selectin expression has been previously described (4, 20, 23). Briefly, the ¹²⁵I-binding MAb and ¹³¹I-nonbinding MAb activities in different tissues and in 50-ml samples of cell-free plasma were counted in a 14800 Wizard 3 gamma counter (Wallac, Turku, Finland), with automatic correction for background activity and spillover. The total injected activity used in each experiment was calculated by counting a 2-µl sample of the radiolabeled MAb mixture. The radioactivity remaining in the tube used to mix the MAbs and the syringe used to inject the mixture were subtracted from the total injected activity. The accumulated activity of each MAb in an organ was expressed as the percentage of injected activity per gram of tissue. P-selectin expression was calculated by subtracting the accumulated activity per gram tissue of the nonbinding ¹³¹I MAb from the activity per gram tissue of the ¹²⁵I binding MAb.

Experimental protocols. The magnitude of P-selectin expression in the jejunum was assessed in rats subjected to I/R alone (20 min of ischemia followed by 60 min of reperfusion), IPC + I/R (5 min preconditioning ischemia then 10 min reperfusion before I/R), and in controls using the dual radiolabeled MAb technique described above. In addition to the jejunum, samples from a variety of other tissues (brain, heart, lungs, duodenum, and large intestine) were also obtained. The right and left testicles were harvested, and P-selectin expression in this paired organ was compared to ensure that the MAb mixture was adequately distributed. To assess the mechanism by which IPC prevents postischemic P-selectin expression, the roles of adenosine and PKC were addressed using specific antagonists and/or agonists, using doses shown to be efficacious in our previous work (1, 8, 9, 14, 26).

To test for a role for adenosine as an initiator of IPC, two approaches were used. In the first, warmed (37°C) saline solutions (10 ml) containing either an A₁-receptor antagonist [8-cyclopentyl-1,3-dipropylxanthine (DPCPX), 10 nM], an A₂-receptor agonist [3,7-dimethyl-1-propylxanthine (DMPX), 10 nM], or adenosine deaminase (ADA 0.25 U/ml), an enzyme that degrades extracellular adenosine to its inactive metabolite, were placed in the abdominal cavity 5 min before and during preconditioning ischemia. As a second approach, pharmacological preconditioning with an adenosine A₁-receptor agonist [N⁶-cyclopentyladenosine (CPA), 1 nM] in lieu of IPC was accomplished by placing this compound in the peritoneal cavity for 5 min beginning 15 min before the onset of prolonged ischemia. To ensure adequate exposure of the small intestine to the drug solutions, the bowel was gently massaged during drug instillation. All drug solutions were removed from the peritoneal cavity at the end of the period of preconditioning ischemia, and the cavity was flushed three times with 15 ml of warmed saline.

The role of PKC in IPC was evaluated by utilizing antagonists of the enzyme with differing isoform specificities. Warmed (37°C) saline solutions (10 ml) containing chelerythrine (1 µM) or bisindolylmaleimide I (10 nM), which exhibit inhibitory activity toward all PKC isotypes (12, 25), or Go-6976 (10 nM), an inhibitor specific for the calcium-dependent (classical) PKC isoforms (18), were placed in the peritoneal cavity 5 min before induction of prolonged ischemia. The drugs were flushed from the peritoneal cavity at the end of period of prolonged ischemia as described above.

Saline was used as the vehicle for all drugs. To control for the effects of vehicle instillation, control animals and rats exposed to I/R alone were treated with warmed (37°C) saline (10 ml) alone, according to the protocol outlined above.

Statistical analysis. All values obtained in this study are expressed as means ± SE. The data were initially analyzed using a one-way analysis of variance. To identify which groups were statistically different, Bonferroni's multiple comparison test was employed. Statistical significance was defined at P  0.05. The number of experiments in each group is indicated in Table 1.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Figure 1 illustrates the effects of IPC alone, prolonged I/R, and IPC + I/R on jejunal P-selectin expression relative to that determined in nonischemic control experiments. I/R increased jejunal P-selectin expression by more than sevenfold compared with nonischemic control animals (0.116 ± 0.036 vs. 0.016 ± 0.004% ID/g, respectively, Fig. 1). This postischemic increase in jejunal P-selectin expression was completely prevented by preconditioning the small intestine with 5 min of ischemia and 10 min of reperfusion before prolonged I/R (0.023 ± 0.006 vs. 0.116 ± 0.036% ID/g) (Fig. 1). Exposing the jejunum to the IPC protocol alone (i.e., no subsequent I/R) did not alter P-selectin expression (0.023 ± 0.006% ID/g, Fig. 1).

View larger version (8K): [in this window] [in a new window]   Fig. 1.   P-selectin expression measured in samples of the jejunum obtained from control rats and from rats subjected to ischemic preconditioning (IPC) alone, ischemia-reperfusion (I/R), and IPC + I/R. * Values statistically different from control at P < 0.05.

Exposure of the small bowel to ADA during the period of preconditioning ischemia reversed the protective effect of IPC on the expression of P-selectin (0.078 ± 0.003 vs. 0.016 ± 0.004% ID/g, respectively, Fig. 2). The role of distinct adenosine receptor subtypes was evaluated by exposing the intestine to specific receptor antagonists or agonists during the period of preconditioning ischemia. As shown in Fig. 2, blockade of adenosine A₁ receptors with DPCPX abolished the effect of IPC to prevent P-selectin expression (0.120 ± 0.037 vs. 0.016% ID/g, respectively). However, exposing the small intestine to the adenosine A₂-receptor antagonist DMPX failed to reverse the effects of IPC to reduce P-selectin expression (0.052 ± 0.013% ID/g, respectively, Fig. 2). The latter observations support a role for adenosine A₁-receptor activation as a trigger for the effect of IPC to prevent postischemic P-selectin expression. This notion is supported by the observation that the jejunum could be pharmacologically preconditioned to resist the effect of I/R to increase jejunal P-selectin expression by exposing the bowel to an adenosine A₁-receptor agonist (N⁶-cyclopentyladenosine) over a time frame similar to the period of preconditioning ischemia (0.044 ± 0.007% ID/g, Fig. 2).

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Effect of adenosine deaminase (ADA), adenosine A1-receptor blockade (A1 Block), or A2-receptor blockade (A2 Block) on reductions in P-selectin expression induced by IPC. Pharmacological preconditioning was produced by administration of an adenosine A1-receptor agonist (A1-PC) in lieu of IPC. * Values statistically different from control; + and # statistical difference from I/R and IPC, respectively, at P < 0.05.

The effect of exposing the intestine to PKC antagonists during the period of prolonged ischemia is depicted in Fig. 3. Treatment with the PKC antagonists chelerythrine or bisindolylmaleimide I (0.120 ± 0.043 and 0.151 ± 0.039% ID/g) prevented the effect of IPC (0.016 ± 0.004% ID/g) to reduce jejunal P-selectin expression induced by I/R (0.116 ± 0.036% ID/g). Whereas these observations support a role for PKC in the signaling cascade induced by IPC, they do not provide insight regarding the class of PKC isoforms involved in the prevention of P-selectin expression because both chelerythrine and bisindolylmaleimide block the activity of all PKC isotypes. To determine whether the classical or novel isoform(s) of PKC were activated over the prolonged ischemic period, we treated the intestine with Go-6976, a highly specific antagonist of the classical isotypes PKC- and PKC-1 that exhibits no activity toward the novel isotypes PKC- or PKC-, even at millimolar concentrations (18). In contrast to the marked effect of chelerythrine and bisindolylmaleimide, treatment with Go-6976 did not alter the effect of IPC to prevent postischemic P-selectin expression (0.025 ± 0.007% ID/g) (Fig. 3). This latter result suggests that the beneficial effects of IPC on postischemic P-selectin expression are mediated by activation of one or more of the novel PKC isoforms.

View larger version (11K): [in this window] [in a new window]   Fig. 3.   Inhibition of protein kinase C with bisindolylmaleimide I (Bos) or chelerythrine (Chel), but not Go-6976, abolished the effect of IPC to prevent increased P-selectin expression induced by I/R. * Values statistically different from control; + and # statistical difference from I/R and IPC, respectively, at P < 0.05.

The data presented in Table 1 summarizes P-selectin expression in the lungs, heart, brain, duodenum, colon, and testicles. Duodenal and colonic P-selectin expression was not modified by prolonged occlusion of the SMA followed by reperfusion, IPC alone, or IPC + I/R, indicating that the effect of IPC was specific for the intestinal tissue perfused by the SMA (i.e., jejunum). Similar results were noted in the brain, heart, and lungs. P-selectin expression in these organs was not affected by exposing preconditioned small intestines to ADA, DPCPX, DMPX, CPA, chelerythrine, bisindolylmaleimide, or Go-6976. The similarities in P-selectin expression noted for the right and left testicles in the different groups indicate that the antibodies were uniformly distributed.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We recently demonstrated that IPC abrogates postischemic microvascular barrier disruption and completely prevents I/R-induced leukocyte rolling, adhesion, and emigration by a mechanism that is initiated by stimulation of adenosine receptors (1, 8) and involves PKC activation (9). These are important observations because they indicate that IPC may target cellular processes involved in leukocyte recruitment as end effectors of this phenomenon. From the growing body of evidence suggesting that activation of sarcolemmal and/or mitochondrial K_ATP channels serve as an effector of IPC in the myocardium (7) and the fact that leukocyte activation can be prevented by administration of exogenous K_ATP channel agonists (7), it is tempting to postulate that K_ATP channel activation underlies the antiadhesive actions of IPC. However, our earlier work failed to support a role for K_ATP channels in the anti-inflammatory effects of IPC in the mesentery (8). Thus identification of the end effector(s) of the antiadhesive effects of preconditioning remains unclear. However, because postischemic leukocyte rolling (and thus subsequent stationary adhesion and emigration) is critically dependent on the expression of P-selectin on venular endothelium (17) and IPC prevents I/R-induced leukocyte rolling in the mesentery (8, 9), we hypothesized that IPC may act to modulate the expression of this adhesive ligand in the postischemic small bowel. Our results are consistent with this concept in that the sevenfold increase in jejunal P-selectin expression noted after 60 min of reperfusion following 20 min of ischemia was completely prevented by prior exposure of the intestine to a very brief period (5 min) of vascular occlusion (Fig. 1).

P-selectin is also expressed by activated platelets (5), a fact that raises the possibility that the increased P-selectin noted in jejunal samples was due to accumulation of these cells rather than by endothelial expression in postcapillary venules. However, this explanation is unlikely because we did not observe platelets (which can be distinguished from leukocytes by their much smaller diameter in our intravital microscopic observations) adhering to the vascular wall in any of the groups. In addition, no platelet-leukocyte aggregates were observed during these studies. For these reasons, our data most likely represents changes in P-selectin expression by the vascular endothelium. It is also important to note that P-selectin expression was not modified in the other organs sampled as part of these studies (and not subjected to I/R) by any of the experimental perturbations or pharmacologicl agents used. Thus the alterations in P-selectin expression were confined to the regions of the bowel subjected to I/R.

A large body of evidence indicates that the cardioprotective effects of IPC are triggered by the release of adenosine during the period of preconditioning ischemia (3, 13, 15). We have recently extended these observations to the anti-inflammatory effects of IPC in the microcirculation (1, 8). In these studies, we showed that postischemic microvascular barrier disruption and leukocyte-endothelial cell adhesive interactions in single postcapillary venules of the rat mesentery were completely prevented by IPC, using the same protocol described in the present study. These anti-inflammatory effects of IPC were prevented in mesenteries superfused with ADA or adenosine A₁ (but not A₂)-receptor antagonists during the period of preconditioning ischemia (8). Moreover, we were able to pharmacologically precondition the mesentery to resist I/R-induced leukocyte rolling, adhesion, and emigration by topical application of adenosine to the mesentery over a time frame similar to that for preconditioning ischemia (8). An identical pattern of response was noted in the present study with regard to P-selectin expression in that intraperitoneal instillation of ADA or adenosine A₁-, but not A₂-, receptor antagonists during the period of preconditioning ischemia prevented postischemic P-selectin expression. Moreover, pharmacological preconditioning with an A₁-receptor agonist mimicked this effect of IPC. Thus our data suggests that adenosine A₁-receptor activation during IPC triggers downstream molecular signaling events that act to limit postischemic P-selectin expression.

Whereas our earlier (1, 8) and present results indicate that adenosine is the major trigger of the anti-inflammatory effects of IPC in the mouse cremaster muscle and rat small intestine, respectively, Kubes and co-workers (16) have presented evidence which suggests that adenosine plays a much less prominent role as an initiator of the antiadhesive effects of IPC in the cat mesentery. Species differences may represent one explanation for these divergent effects, a notion supported by the fact that ₁-adrenergic receptor activation, rather than adenosine, appears to initiate the cardioprotective effects of IPC in rat hearts (19). However, it is also possible that drug accessibility may be more limited in the thicker cat mesentery than in the rat mesentery.

Strong evidence has accumulated indicating that activation of specific PKC isoforms by preconditioning is an essential downstream signaling event in IPC in virtually every model of I/R, species, and organ tested to date (3, 9, 13, 15, 19, 24). Our results indicate that the protective action of IPC to limit postischemic P-selectin expression also involves PKC activation (Fig. 3). Superfusion of the mesentery during prolonged ischemia with two highly specific, structurally unrelated, and mechanistically different PKC inhibitors chelerythrine (blocks the substrate recognition domain, 12) and bisindolylmaleimide (interferes with the ATP binding site, 25) completely abolished the beneficial action of IPC to prevent P-selectin expression. Moreover, PKC activation by IPC appears to occur downstream in the signaling cascade initiated by adenosine A₁-receptor activation during preconditioning ischemia, because treatment with the PKC antagonists was effective when administered during prolonged ischemia. This concept is consistent with the adenosine hypothesis for cardiac preconditioning that has been put forward by Downey and co-workers (3), wherein adenosine A₁-receptor activation during IPC promotes the translocation and activation of PKC, events that are coupled by phospholipase C-dependent production of diacylglycerol.

Whereas our data with chelerythrine and bisindolylmaleimide support a role for PKC in the anti-inflammatory effects of IPC, these inhibitors are nonspecific with regard to PKC isotypes. At least 11 different subtypes of PKC have been identified, which have been subdivided into three major classes (22, 24). Of these, the isotypes in the classical and novel classes have received the most attention in terms of their regulation and role in different physiological processes. The conventional or classical isotypes (, 1, 2, ) require calcium for their activity (22). On the other hand, the novel isoforms (, , , ) lack the calcium-binding domain and are thus differentially regulated compared with the classical isotypes (22). A number of recent reports indicate that IPC induces selective translocation and activation of the novel isotypes PKC- and PKC- in the heart (19, 24), which then act to phosphorylate downstream molecular targets that mediate IPC. One limitation in establishing the role for a particular PKC isotype(s) relates to the lack of inhibitors that are specific for the different isoforms. However, Go-6976 is a recently described PKC antagonist that exhibits a high degree of specificity for the classical isotypes PKC- and PKC-1 but demonstrates no inhibitory activity toward PKC- or PKC-, even at millimolar concentrations (18). On the basis of this specificity and the suggested role for the novel isoforms in cardiac preconditioning, we hypothesized that treatment with Go-6976 would not influence the effectiveness of IPC in our model. As shown in Fig. 3, Go-6976 failed to influence the effectiveness of IPC in preventing postischemic leukocyte adhesion. Thus our results are consistent with the concept that the beneficial actions of IPC involve isotypes other than the classical isotypes (PKC- and PKC-1), most probably the novel isoforms.

We do not believe that the lack of effect of Go-6976 was due to use of an insufficient dose because we have demonstrated that this compound completely prevents hydrogen peroxide-induced microvascular barrier disruption (14) and the increased leukocyte adhesion induced by I/R alone (no preconditioning) (26). It is important to note that in the latter study, we used the same experimental model (rat mesentery subjected to 20 min of ischemia and 60 min of reperfusion) and dose as employed in the present study. Taken together, these results suggest that the classical PKC isotypes mediate the proinflammatory effects of I/R alone (no preconditioning), whereas the novel isotypes play a critical role in the prevention of postischemic leukocyte-endothelial cell interactions in preconditioned mesenteries.

In summary, our data indicate that IPC completely prevents postischemic P-selectin expression, an observation that provides the molecular basis for the fact that preconditioning abolishes I/R-induced leukocyte rolling (8, 9), which is largely P-selectin dependent (17). These powerful anti-inflammatory actions of IPC are triggered by stimulation of adenosine A₁ receptors during the period of preconditioning ischemia and involve subsequent activation of nonconventional PKC isotypes during prolonged ischemia. Thus these studies have identified P-selectin expression as a novel downstream effector target of the adenosine-initiated, PKC-dependent, anti-inflammatory signaling cascade that is activated in IPC.