INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

MESENTERIC ISCHEMIA activates a group of sympathetic afferent A- (mechanosensitive) and C-fiber (chemosensitive) nerves to reflexly evoke cardiovascular responses (14). These ischemically sensitive fibers can be stimulated by nociceptive substances such as bradykinin (BK), a potent nociceptive autacoid, and capsaicin, an extract of hot peppers (16, 19, 21). BK is a nonapeptide in the kinin family that contracts visceral smooth muscle and relaxes vascular smooth muscle (1). This peptide also directly and indirectly stimulates afferent nerve endings and has been implicated as a mediator of cardiovascular reflexes in several different organ systems, including the heart (8), testis (9), skeletal muscle (22), and abdominal visceral organs (13, 14, 16).

Brief (10-30 min) abdominal visceral ischemia induced by occlusion of the superior mesenteric and celiac arteries reflexly evokes cardiovascular responses and produces large (30-50 mmHg) and highly reproducible increases in systemic arterial pressure in the cat (7, 23, 25). Hemodynamic alterations resulting from this cardiovascular reflex include increases in systemic arterial pressure, heart rate, myocardial contractility, and systemic vascular resistance (7, 25). Although the reflex response to abdominal ischemia has been well defined, the possible mechanisms of afferent activation and reflex induction during an ischemic period in vivo have yet to be fully determined.

Results of previous studies from our group have indicated that exogenous BK applied to the serosal surface of a number of abdominal visceral organs produces marked pressor responses that are mediated by sympathetic afferents traveling in the splanchnic nerve (15, 16, 20). Although these studies support the notion that exogenously applied BK can produce a reflex cardiovascular response, the importance of endogenous BK produced during abdominal ischemia with regard to this reflex has yet to be determined. Specific BK receptor antagonists that competitively inhibit BK₂ receptors have been shown to effectively decrease receptor activation by BK (10, 21, 24, 26). We used two chemically unrelated specific BK₂ receptor antagonists, NPC-17731 (10) and HOE-140 (26), to test the hypothesis that BK produced endogenously during abdominal ischemia activates ischemically sensitive visceral afferents to induce a reflex cardiovascular response.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Preparation. Studies were performed in cats of either sex (1.9-6.0 kg). Anesthesia was initially induced by an injection of ketamine (40-50 mg/kg im) followed by a bolus injection of -chloralose (50 mg/kg iv). Subsequent injections of -chloralose were administered as needed to maintain a sufficient anesthetic state, as determined by response to paw pinch, eye reflexes, and abnormal respiration patterns. A cuffed endotracheal tube was inserted to artificially ventilate the animals (Harvard pump, model 661, Ealing, S. Natick, MA). Inspired gas was enriched with 100% oxygen, while arterial blood gases and pH were monitored frequently (model ABL3, Radiometer, Copenhagen, Denmark) and maintained within physiological limits (pH = 7.35-7.45, PCO₂ = 28-35 mmHg, PO₂ > 100 mmHg) by adjusting rate or depth of ventilation and/or administering sodium bicarbonate (1 M). A rectal probe was used to measure body temperature, which was maintained between 36.5 and 37.5°C by heating pads and heat lamps. This study was conducted in compliance with the "Guiding Principles in the Care and Use of Animals" endorsed by the American Physiological Society.

Protocols. After surgery the animals were allowed to recover for at least 30 min until blood pressure was stable and blood gases were within the normal range. Abdominal ischemia was induced for 20 min unless the reflex-induced increase in blood pressure attained a constant plateau for several minutes, in which case blood flow was restored to the ischemic region. Animals were given a bolus injection of NPC-17731 [30-40 µg/kg (22-29 nmol/kg) iv; Scios Nova, Mountain View, CA; n = 11 and 4 for groups 1 and 2, respectively] or HOE-140 [Icatibant; 30-40 µg/kg (23-31 nmol/kg) iv; Hoechst-Roussel Pharmaceuticals, Somerville, NJ; n = 8 and 4 for groups 1 and 2, respectively]. After injection, each animal was allowed to equilibrate for at least 30 min, at which time a second period of ischemia was induced. BK (1-10 µg/ml, 0.9-9 µM) and capsaicin (2-200 µg/ml, 6.5-650 µM) were applied to the surface of the gallbladder at the conclusion of the experiment to confirm BK₂ receptor blockade and the reflex responsiveness of the preparation, respectively.

Data analysis. Only animals with stable blood pressures and reflex responses of 10 mmHg to combined occlusion of the celiac and superior mesenteric arteries were used. Furthermore, animals in which the difference between control absolute blood pressures between the first and second ischemic periods exceeded 25 mmHg were excluded (n = 3).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Group 1. Occlusion of the celiac and superior mesenteric arteries produced an initial rise in blood pressure followed by a gradual decline in pressure during the next several minutes (Fig. 1A). After a variable time period, a gradual secondary increase in pressure became evident. This secondary response, which has been demonstrated to be reflex in nature (7), was analyzed. The blood pressure response was measured as the difference between the blood pressure nadir, which was typically attained several minutes after occlusion, and the peak secondary blood pressure response. In the repeatability group, the first and second combined occlusions of the superior mesenteric and celiac arteries resulted in similar changes in mean arterial blood pressure of 16 ± 2.7 and 16 ± 5.7% (P > 0.05), respectively (Fig. 2). Administration of the BK₂ receptor antagonists NPC-17731 and HOE-140 produced significant attenuation of the reflex rise of blood pressure in group 1 (Fig. 1B). For instance, NPC-17731 (n = 11) reduced the ischemia-induced increment of mesenteric arterial pressure from an average of 13 ± 1.2 to 6.2 ± 1.6% (P < 0.05; Fig. 3), whereas HOE-140 significantly attenuated the reflex blood pressure increase from 16 ± 3.7 to 8.4 ± 2.0% (P < 0.05; Figs. 1 and 3). There was no significant change in the control preischemia blood pressure before vs. after NPC-17731 or HOE-140 (Fig. 3). In the group of animals subjected to ganglionectomy after the first combination occlusion, the blood pressure response was virtually eliminated (Fig. 2). The blood pressure increase during the first occlusion (27 ± 5.8%) was reduced to 0.7 ± 4.0% during the second occlusion.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   A: chart recording of arterial blood pressure response to brief occlusion of superior mesenteric and celiac arteries. B: ischemic period was repeated after administration of HOE-140, a bradykinin (BK2) receptor antagonist. BK2 receptor blockade reduced reflex rise in arterial blood pressure by 75% in this animal.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Comparison of mean arterial blood pressure (MAP) responses to repeated brief occlusions of celiac and superior mesenteric arteries in cats. In 1 group, celiac and superior mesenteric ganglionectomies were performed after 1st ischemic period. In both groups, animals were allowed to recover for 30 min before 2nd ischemic period. Blood pressure increase was repeatable in ganglia-intact group. Ganglionectomy abolished blood pressure increment, confirming reflex nature of blood pressure response. Values in parentheses below bars indicate control (nadir) blood pressure.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Effect of 2 chemically dissimilar BK2 receptor antagonists, HOE-140 and NPC-17731, on visceral ischemia-induced reflex rise in arterial blood pressure. Both antagonists reduced blood pressure increment by ~50%. Values in parentheses below bars indicate control (nadir) blood pressure. * P < 0.05 vs. control.

Group 2. The protocol to minimize blood volume shifts resulted in blood pressure patterns different from those of group 1 (Fig. 4). In this preparation, on induction of ischemia, no initial or minimal initial rise in blood pressure was observed (Fig. 4A). After a variable time a gradual rise in blood pressure was noted. Because the induction of ischemia in this preparation typically did not result in a transient blood pressure rise and subsequent nadir, the blood pressure response was measured from preocclusion control to peak blood pressure response. The reflex augmentation of the mean arterial pressure during the first and second ischemic periods was similar (36 ± 5.6 and 40 ± 2.2%, respectively, P > 0.05). The animals receiving NPC-17731 or HOE-140 displayed similar significant attenuation of the rise in blood pressure during ischemia (Fig. 4B). The reflex rise in blood pressure during the first ischemic period was 57 ± 15%, which was significantly attenuated to 20 ± 8.1% after administration of NPC-17731 (Fig. 4, Table 1). The reflex pressor increase of 43 ± 6.1% during the first ischemic period was reduced to 17 ± 8.3% by HOE-140. Neither antagonist significantly altered the control preischemia blood pressure (Fig. 4, Table 1).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   A: chart recording of arterial blood pressure response to brief diversion of blood flow from superior mesenteric and celiac vascular beds (group 2). B: period of ischemia was repeated after administration of NPC-17731, a BK2 receptor antagonist. BK2 receptor blockade reduced reflex rise in arterial blood pressure by 92% in this animal.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

This is the first study to show that endogenously produced BK is involved in cardiovascular reflexes evoked by abdominal visceral ischemia. Significant attenuation of the cardiovascular reflex was demonstrated for two disparate kinin B₂ receptor antagonists, NPC-17731 and HOE-140. Conversely, reflex responses were shown to be repeatable for both control groups 1 and 2 in the absence of the BK₂ receptor antagonists. Because celiac and superior mesenteric ganglionectomies eliminated the cardiovascular reflex responses elicited by visceral ischemia, the origin of the reflex pathway was confirmed. Thus endogenous BK, through a B₂ receptor mechanism, is capable of stimulating a sufficient number of abdominal visceral afferent endings to reflexly activate the cardiovascular system.

Ischemia alters the interstitial environment and can result in localized acidosis, increased proteolytic activity, and hypoxia, conditions that promote activation of the kinin system (11). Two major pathways are involved in the production of BK and Lys-BK (kallidin); the latter peptide can stimulate BK₂ receptors and can be converted to BK (32). In plasma, BK is produced by activated kallikrein; in tissues, activated kallikrein liberates kallidin. In addition, immune cells (e.g., mast cells and basophils) associated with inflammation can produce kinins by the release of cellular proteases, independent of kallikrein activity (32). Although the relative role of tissue or plasma BK or kallidin in the present study is unknown, kallikreins are present in tissues throughout the gastrointestinal tract (4). Furthermore, we previously reported that brief (5 min) abdominal visceral ischemia results in a three- to fourfold increase in portal venous BK (21). It is clear therefore that BK release is evoked by ischemia and contributes to the ischemia-induced pressor reflex, although the exact source of this liberated BK has not been determined.

Afferent nerve endings of C-fibers are fine, unmyelinated fibers located in the interstitial space (18). Their proximity to cells that contain sequestered tissue kallikrein, including the acinar and -cells of the pancreas and mucous cells of the colon (1), likely is an important factor in determining the concentration of BK at the nerve terminal. No information is available regarding the actual concentration of mediators at receptor sites of the nerve terminal, because the nerve endings cannot be selectively isolated by present techniques. However, BK produced by the various cell types must traverse the interstitial space to gain access to portal plasma, where we have documented an increase during brief ischemia as used in the present study (21).

The local increase in hydrogen ion concentration induced by regional ischemia plays a key role in the initiation of events leading to afferent nerve stimulation. We have reported that protons donated by lactic acid, in particular, contribute to the activation of ischemically sensitive visceral afferents (27). In cutaneous nociceptive afferent fibers, a positive interaction between acidic pH and inflammatory mediators, including BK, was reported with regard to the magnitude and prevalence of nociceptor activation (31). In addition, kinin activity is protected against destruction by kininases in an acidic environment (11). Our previous data show that the extent of acidosis during regional visceral ischemia and the magnitude of the cardiovascular response are correlated; i.e., occlusion of the superior mesenteric artery results in a significantly larger fall in pH in its downstream organs and a significantly greater reflex pressor response than celiac artery occlusion (25). The combined results of the present and past studies (25, 27, 31) suggest that the greater reflex response from the ischemic superior mesenteric vascular bed than from the celiac vascular distribution may be related to enhanced efficacy of liberated BK in a more acidic environment.

BK is unusual in its ability to stimulate many different cell types and activate a number of intracellular messenger systems, which lead directly and indirectly to production of inositol 1,4,5-trisphosphate and diacylglycerol and increased intracellular levels of protein kinase C, arachidonic acid products, superoxide free radicals, intracellular calcium, adenosine 3',5'-cyclic monophosphate, and guanosine 3',5'-cyclic monophosphate (3). These actions lead to many of the clinical manifestations of inflammation, including local pain, hyperemia, increased vascular permeability, and edema (11). The mechanism for stimulation of afferent nociceptive nerve endings, based largely on in vitro studies of hybridoma cell lines, has been proposed to involve a sequence of steps beginning with activation of phosphorylase C  diacylglycerol  protein kinase C  potassium channel phosphorylation  membrane depolarization  action potential generation (3). Furthermore, our previous data suggest that activation of phospholipase A₂ and the production of arachidonic acid metabolites via the cyclooxygenase or lipoxygenase pathway sensitize afferent nerve endings (15, 17, 29) and also may play a role in the production of oxygen-containing free radicals (3). In this latter regard, we have shown that reactive oxygen species, particularly H₂O₂ and · OH, activate ischemically sensitive afferent nerve fibers during visceral ischemia and reperfusion (28). Further studies are necessary to determine the subcellular mechanism(s) responsible for the effect of BK on visceral afferent nerve terminals.

We previously showed that blockade of BK₂ receptors in abdominal organs with HOE-140 or NPC-17731 significantly attenuates the activation of afferent C-fibers during ischemia (21). Furthermore, BK₁ receptors are not involved in this response, because ischemically sensitive C-fibers that responded to BK were not activated by a specific BK₁ receptor agonist (21). We also observed increased abdominal visceral afferent activity during ischemia after administration of captopril, an inhibitor of the BK degradation enzyme kininase II (21). The combination of present and past findings supports our overall hypothesis that endogenous BK released during abdominal visceral ischemia is responsible, at least in part, for stimulation of BK₂ receptors on afferent nerve fibers to induce reflex cardiovascular responses.

It was important in the present study to elicit reproducible reflex responses to visceral abdominal ischemia. With this thought in mind, we chose to study two types of preparations. Group 1 was utilized initially to establish a clear reflex response while minimizing surgically induced trauma to the preparation. However, the reflex responses were obscured partially in group 1 by the large initial hemodynamic changes at the onset of arterial ligation and by antagonist-induced changes in the subsequent blood pressure nadir. The more invasive preparation, group 2, however, demonstrated clear reflex responses due to elimination of several hemodynamic factors induced by a systemic ischemic period. In previous studies performed in our laboratory (7), we postulated that two mechanisms may contribute to the initial increase in systemic arterial pressure during abdominal ischemia. The first is a mechanical effect resulting from shunting more blood from a high- (mesenteric) to a low-compliance (hindlimb) arterial circuit. Second, the initial increase in blood pressure may be related to a sudden elevation of cardiac preload by short circuiting the long time constant mesenteric circulation during ischemia and placing its arterial inflow into the inferior vena cava while simultaneously allowing passive drainage from the mesenteric venous compartment. By utilizing two autoperfusion circuits to maintain a constant venous return to the inferior vena cava and to induce regional abdominal ischemia, we eliminated these direct hemodynamic effects.

The control blood pressures in group 2 animals were less than those in group 1. This likely was the result of greater blood loss associated with the more invasive surgery and the use of a venous return reservoir to maintain an ex vivo blood reserve. Conversely, the reflex increase in blood pressure was greater in group 2, which may have been related to the diminished resting blood pressure allowing a greater potential for blood pressure elevation (33). Furthermore, measurement of the blood pressure increment from nadir to peak response in group 1 may have underestimated the magnitude of the blood pressure increment. In particular, the true prereflex control blood pressure was partly obscured by the declining blood pressure that followed the initial occlusion-induced transient blood pressure elevation.

The blood pressure response to abdominal ischemia was reduced by an average of 53% by two specific BK₂ receptor antagonists in the present study. Interestingly, we recently reported that NPC-17731 reduced the discharge frequency of ischemically sensitive afferent nerve fibers by 46% during abdominal visceral ischemia (21). The similar magnitude of these two related effects suggests that changes in the firing frequencies of this subgroup of C-fibers can alter arterial blood pressure in a proportional fashion. Conversely, the inability of BK₂ receptor blockade to fully eliminate the reflex pressor response to abdominal ischemia suggests that other mediators are involved in the stimulation of splanchnic afferent nerves during visceral ischemia.

In addition to BK, we previously postulated that several other factors contribute to the activation of ischemically sensitive afferent fibers. These include reactive oxygen species such as · OH (28), arachidonic acid products (15, 17), hypoxemia (in an indirect fashion) (6), and decreased pH through production of lactic acid (27). Other potential stimuli include histamine (30) and serotonin (12). In this regard, we recently reported significant increases in histamine and serotonin concentrations in portal venous plasma and intestinal lymph during brief abdominal visceral ischemia (5). It therefore seems likely that the stimulation of afferent nociceptive fibers is a multifactorial process, which may ensure that nerve endings are activated during an ischemic event regardless of possible tachyphylaxis to one or more mediators (31).

Two potential limitations of the present study should be addressed. First, it is possible that inhibition of the ischemia-induced pressor response by a BK antagonist was related to a nonspecific action of the agent. This explanation appears unlikely, however, because a second, structurally dissimilar BK antagonist produced a similar degree of inhibition.

A second potential limitation was the period of ischemia of up to 20 min, which was longer than the shorter (5-10 min) durations used in previous studies of ischemia-induced BK production (21) and activation of ischemically sensitive C-fibers (15, 17). In this regard, although afferent activity can occur within the first few minutes of ischemia, we recently reported that spatial and temporal summation of sympathetic afferent activity in response to visceral ischemia actually continues to increase beyond 5-10 min and reaches a maximum by 17-20 min (23). Therefore, the rationale for the longer time period in the current study was to allow full development of the cardiovascular response by maintaining the ischemic period for up to 20 min.

In summary, two dissimilar BK₂ receptor antagonists, HOE-140 and NPC-17731, reduced the magnitude of the visceral ischemia-induced pressor reflex by ~50%. Control studies verified the repeatability of the pressor response, and the reflex nature of the response was confirmed by combined superior mesenteric and celiac ganglionectomies. These findings support our hypothesis that endogenous BK produced during abdominal ischemia plays a major role in the activation of afferent sympathetic nerve fibers to elicit reflex cardiovascular responses.

Perspectives. A reflex cardiovascular sympathetic response to abdominal visceral ischemia may serve to provide compensatory flow to the ischemic region through collateral blood vessels to maintain local perfusion. The reflexly increased systemic arterial blood pressure, coupled with diminished blood pressure distal to the site of occlusion, would produce an augmented blood pressure differential that would promote blood flow through open collateral vessels to the ischemic region. Although this mechanism may provide a degree of nutrient and oxygen supplementation to the compromised tissue, it also is possible that the cellular injury produced by reperfusion (i.e., ischemia-reperfusion injury) is exacerbated by augmented arterial blood pressure and by elevated sympathetic stimulation in general. We believe that an increased sympathetic outflow response to local occlusive ischemia, as utilized in the present study, may be one important aspect of a generalized cardiovascular response to ischemia, whether it is induced by hemorrhagic shock, septic shock, embolic occlusion of a normal or stenotic blood vessel, or, in the clinical setting, arterial cross clamping during surgical repair or organ replacement. The degree to which the cardiovascular response is beneficial or detrimental would depend on a variety of factors, including the specific organ(s) at risk, duration and level of ischemia, body temperature, and tissue metabolic status.