- Email: [email protected]
Cyclosporine Inhibits Endotoxin-Induced Vasodilation of Isolated Rat Resistance Arterioles
Journal of Surgical Research 136, 112–115 (2006) doi:10.1016/j.jss.2006.04.026
Cyclosporine Inhibits Endotoxin-Induced Vasodilation of Isolated Rat Resistance Arterioles Vincent J. Obias, M.D.,* Gregory D. Rushing, M.D.,* Russell L. Prewitt, Ph.D.,† Darian Rice, M.D., Ph.D.,† and L.D. Britt, M.D., M.P.H.*,1 *Department of Surgery, †Department of Physiological Sciences, Eastern Virginia Medical School, Norfolk, Virginia Submitted for publication February 20, 2006
Background. An isolated arteriole fails to dilate in response to endotoxin unless a segment of aorta is included in the perfusion system. The unknown substance released by the aorta after exposure to endotoxin is dependent upon the NF-B pathway and induces inducible nitric oxide synthase (iNOS) in the arteriole. The purpose of this study was to determine if cyclosporine A (CSA) that inhibits both NF-B and iNOS would prevent the vasodilatory response to endotoxin. Materials and methods. Rats were injected with either 10 mg/kg of CSA or oil vehicle followed by the removal of a cremaster muscle. The feeding arteriole was isolated from the cremaster and mounted on micropipettes and pressurized to 70 mmHg in a superfused tissue bath. After an hour equilibration to develop spontaneous tone, a 1 cm segment of aorta was placed in the superfusion system upstream from the arteriole and Salmonella enteriditis endotoxin was added to the buffer at a concentration of 2.5 g/mL (ET) or continued infusion of buffer alone. Internal diameters of cannulated arterioles were measured with videomicroscopy and videocalipers for an additional hour. Results. Arterioles downstream from an aorta exposed to vehicle but not endotoxin developed 22.8 ⴞ 3.7% tone that remained unchanged over the following hour. Arterioles exposed to endotoxin started with 22.5 ⴞ 2.8% spontaneous tone and this fell over the following hour to 11.8 ⴞ 3.6%, P < 0.05. Pre-treatment of the rats with CSA tended to increase resting tone and completely prevented the loss of tone after endotoxin. Conclusions. Pre-treatment of the aortic segment with CSA resulted in the development of increased 1
To whom correspondence and reprint requests should be addressed at Department of Surgery, Eastern Virginia Medical School, 825 Fairfax Avenue, Norfolk, VA 23507-1912. E-mail: BrittLD@ evms.edu.
0022-4804/06 $32.00 © 2006 Elsevier Inc. All rights reserved.
tone in the downstream arteriole and completely blocked the vasodilatory response to endotoxin. These results suggest that CSA or a similar compound may be useful in the treatment of septic shock. © 2006 Elsevier Inc. All rights reserved.
Key Words: septic shock; cyclosporine A; endotoxin; nuclear factor kappa B; cremasteric arteriole. INTRODUCTION
Sepsis remains a major cause of morbidity and mortality despite the development of advanced antibiotics and complex algorithms in the management of septic shock. The hypotension characteristic of septic shock has been attributed to a loss of vascular tone in resistance arterioles. Current theories implicate bacterial products such as endotoxin as a major initiator of the septic shock cascade that leads to hypotension. Previous work from our laboratory has demonstrated that endotoxin alone does not reduce tone in isolated resistance arterioles unless a segment of aorta is also present in the bath superfusion system [1]. To elucidate the mechanism behind this phenomenon, segments of de-endothelialized aorta were used and blockade by inodmethacin were attempted. The vasodilation was not dependent on the endothelium nor could it be blocked by indomethacin. This led to the hypothesis that endotoxin causes the release of an unknown factor from the aorta that results in downstream vasodilation. Subsequently, we found that exposure of either the upstream segment of aorta alone or the downstream arteriole alone to pyridine-2,6-bis thiocarboxylate (PDTC), a potent nuclear factor kappa B (NF-B) inhibitor, prevented the vasodilatory response to endotoxin [2]. Thus, the unknown factor was not a direct vasodilator but depended upon NF-B for its production and response in the arteriole. NF-B is known to
112
OBIAS ET AL.: CYCLOSPORINE INHIBITS ENDOTOXIN-INDUCED VASODILATION
up-regulate the inducible form of nitric oxide synthase (iNOS) and we recently found that aminoguanidine, a specific inhibitor for iNOS, also blocked the vasodilatory response [3]. The immunosuppressive phosphatase inhibitor, cyclosporine A (CSA), protects against hemorrhagic shock and restores the reactivity of the aorta to phenylephrine by inhibiting the NF-B pathway [4]. In addition, CSA decreases vascular expression of iNOS in the rat aorta [5]. Thus, CSA appears to be an excellent candidate to block the mechanisms implicated in endotoxic vasodilation and the aim of this study was to determine if CSA can prevent the vasodilation of a rat cremasteric arteriole in response to upstream aortic endotoxin exposure. MATERIALS AND METHODS The Animal Care and Use Committee of Eastern Virginia Medical School approved all experimental protocols. Twenty male SpragueDawley rats between 150 and 270 g were used. They were housed with free access to food and drinking water and were exposed to a 12/12-h light/dark cycle. The rats were anesthetized with an intraperitoneal (i.p.) injection of sodium pentobarbital. A standard dose of 60 mg/kg was used with supplements administered if necessary to maintain anesthesia. The rats were then injected i.p. with either 10 mg/kg of CSA (n ⫽ 10) dissolved in lipid, or oil vehicle only (n ⫽ 10). The scrotal sac was then opened and the cremaster muscle dissected free from the surrounding tissues. The cremaster was then placed in cold (4°C) Krebsbicarbonate-HEPES buffer solution. The cremaster was pinned flat and under a dissecting microscope, a 2 mm segment of first order cremasteric arteriole was isolated, transferred to a tissue bath, and cannulated with micropipettes (Fig. 1). Vessels were secured with 10-0 monofilament suture. The vessel chamber was moved to a Zeiss inverted microscope coupled to a CCD video camera and a highresolution monitor. The arteriole was pressurized to 70 mmHg in the absence of intraluminal flow via an adjustable pressure servo. The vessel was warmed to 34°C, the in situ temperature of the rat cremaster, by continual infusion (4 mL/min) of KREBS-bicarbonate-
113
HEPES solution at pH 7.4. The buffer was bubbled with 95%N 2/5% CO 2 gas, resulting in a pO 2 of 60 mmHg at the level of the vessel. The arteriole was allowed to achieve spontaneous basal tone over 60 min (t ⫽ 0) and the vessels were tested to assure there were no pressure leaks. Internal diameters of cannulated arterioles were measured with videomicroscopy and videocalipers at a resolution of ⫾1 m. After the equilibration period, a 1 cm segment of abdominal aorta was removed from the rat and placed in a flow-through chamber connected in series to the arteriole so that superfused KREBSbicarbonate-HEPES buffer passed over the aorta and onto the cannulated arteriole. At this time (t ⫽ 60) Salmonella enteriditis endotoxin was added to the superfusion buffer at a concentration of 2.5 g/mL (ET) (n ⫽ 10) or continued infusion of buffer alone was maintained for an additional hour (t ⫽ 120) (n ⫽ 10). At the end of the study period, maximal vessel diameters and the confirmation of an intact cremasteric arteriole endothelium were determined by exposure to 3 M acetylcholine and viability was demonstrated by a contractile response to 3 M phenylephrine. Unresponsive arteries were judged to be damaged and excluded from the study.
Statistical Analysis Results are shown as a mean ⫾ SEM of the inside diameter and percent tone. Percent tone was calculated by expressing the diameter of the arteriole as a percent of the maximal diameter of the vessel after exposure to 3-M acetylcholine. Statistical significance was determined by the Kruskal-Wallis test (non-parametric ANOVA), and significance was assumed at P ⬍ 0.05.
RESULTS
The internal diameters of the arteries throughout the experimental period are shown in Fig. 2. All groups of vessels developed tone during the equilibration period. This intrinsic or myogenic tone is the hallmark of resistance arteries. Baseline tone of the arterioles was varied. Control, ET, and ET ⫹ CSA arteriolar tones were different, but not found to be statistically significant. The group treated with CSA alone, however, had
FIG. 1. Diagram of the experimental setup. The arteriole was mounted on micropipettes and suspended in a physiological buffer that flowing at 4 mL/min. The vessel, shown in the center, was transilluminated and viewed by video microscopy. The inside diameter of the arteriole was measured with a video caliper system that provides a digital readout free of errors because of parallax.
114
JOURNAL OF SURGICAL RESEARCH: VOL. 136, NO. 1, NOVEMBER 2006
remained unchanged at the end of the experiment (120 min). The two groups in which the aorta was exposed to CSA appeared to develop higher tone than the other groups but there were no statistically significant differences among any of the groups at the 60-min time point. Arterial percent tone fell from 22.5 ⫾ 2.8% to 11.8 ⫾ 3.6% in the group receiving endotoxin alone (ET) and this value was significantly lower than in the other three groups at 120 min. Maintenance of tone at 35.1 ⫾ 3.4% throughout the experimental period in the ET-CSA group indicated that pre-treatment with CSA prevented the fall in vascular tone induced by endotoxin. DISCUSSION
FIG. 2. Mean diameter of the arterioles in the experimental groups over the experimental period. The first 60-min is an equilibration period in which the vessels develop spontaneous tone. At 60 min a section of aorta was introduced into the superfusion line and either endotoxin or vehicle was applied for the next 60 min. The aortas and arterioles of the control group were exposed to vehicle only. The CSA group included an aorta from a rat treated with CSA (10 mg/kg). The ET group contained a vehicle treated aortic segment and was superfused with endotoxin at 2.5 g/mL. The ET-CSA group consisted of an aorta exposed to CSA and superfusion with endotoxin.
a basal tone much more constricted than the other control or experimental groups (P ⬍ 0.05) (Fig. 2). The control group which received neither CSA nor endotoxin maintained the tone without change through the next 60 min. Upon application of acetylcholine at the end of the experimental period, the vessels dilated as shown as the last point on the figure. The arteries exposed to endotoxin had a similar diameter at 60 min but after introduction of endotoxin the diameter gradually increased and was significantly greater than the control diameter at 120 min. The endotoxin exposed arteries (ET) dilated further upon application of acetylcholine indicating that they had not lost all of their tone after the 60-min exposure period. The group with an aortic section from a CSA treated rat constricted to a smaller diameter than the control and the ET group of arteries during the equilibration period (P ⬍ 0.05) and they maintained that reduced diameter until dilated with acetylcholine. The group exposed to both ET and CSA in the aorta (ET-CSA) also constricted to a smaller diameter but were statistically different only from the ET group of arteries at 60 min. Despite the exposure to endotoxin, these arteries downstream from a CSA treated aorta maintained their constricted state throughout the following 60-min and then dilated in response to the acetylcholine. Because of differences in relaxed diameter of the arteries, the amount of vascular tone was calculated as a percent of the relaxed diameter and expressed as a percent. Percent tone in the control group was 22.8 ⫾ 3.7 at the end of the equilibration period (60 min) and
Previous studies from our laboratory demonstrated that endotoxin alone was not sufficient to cause vasodilation in isolated rat cremaster arterioles. Glembot et al. discovered that a section of rat aorta was required upstream of an isolated arteriole in order for the arteriole to dilate in response to endotoxin [6]. As in the present experiments, Glembot et al. placed an isolated segment of aorta in the superfusion line where it was exposed to the buffer solution mixed with endotoxin. The resultant mixture was then allowed to flow downstream to the isolated rat arteriole and the arteriole was noted to vasodilate. It was hypothesized that some substance was released from the upstream section of aorta after endotoxin exposure that caused a loss of tone in the downstream resistance vessel. Our model required slight modification, because of the lipophylic nature of CSA. While addition of CSA to the isolated aortic segment in buffer solution would have been preferred, this was not possible. Vasodilators known to be produced by the endothelium are NO and PGI 2 but removal of the aortic endothelium or application of indomethacin did not prevent the vasodilatory response. Additional experiments by Snyder et al. demonstrated that pretreatment of rats with PDTC or MG132 (both blockers of the NF-B pathway), prevented the endotoxin-induced vasodilatory response [2]. In addition, the response was prevented if either the aorta or the arteriole alone was from a rat receiving PDTC. Thus, the substance released by the aorta was not a direct vasodilator but was a product associated with the NF-B pathway and the response by the arteriole also was dependent upon the NF-B pathway. A major contributor to the loss of tone in septic shock is the up-regulation iNOS that is capable of producing large amounts of NO. NF-B is a DNA binding protein that is required for transcription of many proinflammatory genes including iNOS. Viol et al. recently showed that aminoguanidine, a specific inhibitor for iNOS, prevented the vasodilatory response to endotoxin by the downstream arteriole [3]. Because CSA has been shown to inhibit NF-B as well as iNOS [7], it
OBIAS ET AL.: CYCLOSPORINE INHIBITS ENDOTOXIN-INDUCED VASODILATION
promised to be efficacious in preventing the response to endotoxin in our isolated arteriolar model. The results of these experiments clearly show that CSA completely prevented the loss of tone after exposure to endotoxin suggesting that it may be useful in therapy for septic shock. In support of this conclusion, Altavilla et al. demonstrated that CSA increased the survival rate in a rat model of hemorrhagic shock and decreased the amount of NF-B in the nucleus by preventing the loss of IB that keeps NF-B bound and inactive [4]. In shock induced by splanchnic artery occlusion, CSA was also effective to revert hypotension, restore vascular reactivity to phenylephrine, and blunt iNOS induction and activity resulting in lower plasma nitrite/nitrate concentrations [8]. In our study, CSA also had an effect on basal arteriolar tone. CSA is thought to block NF-B activation by inhibiting the action of calcineurin that correlates with its immunosuppressive activity [9]. Regardless of the mechanism, the inhibition of iNOS is the major beneficial effect for preventing loss of vascular tone after endotoxin exposure. Although not statistically significant, the tendency of the arterioles downstream from CSA treated aortic segments to develop more tone is consistent with previous reports that CSA increases vasoconstriction both in vivo and in vitro. Rego et al. showed that CSA increases responsiveness to norepinephrine and potassium in the mesenteric vascular bed [10] and raises blood pressure in the rat independent of sympathetic nerve activity [11]. Lee et al. showed that acute application reduced the vasodilatory response to acetylcholine, an eNOS activator, in isolated rat aortas. In addition it lowered levels of cGMP that are elevated by NO produced by eNOS in response to acetylcholine. Thus, some inhibition of eNOS may contribute to the sustained tone of our isolated arterioles in the presence of CSA but as seen in Fig. 2, the vasodilatory response to acetylcholine was still present. The identity of the unknown factor released by the aorta exposed to endotoxin is of great interest. Its effectiveness when secreted into a buffer flowing at 4 mL/min suggests a powerful cytokine that works at extremely low concentrations such as TNF-␣ or one of the interleukins also downstream in the NF-B cascade. Previous work from our laboratory has shown that IL-1 and IL-6 do not dilate isolated arterioles although they are quite effective when applied to the same arteriole in situ where parenchymal tissue may provide an unknown factor [12]. TNF-␣ was also found to have no apparent affect on the diameter of isolated arterioles unless they were pre-treated with endotoxin [6]. Therefore, TNF-␣ may contribute to the response, but as it is ineffective when applied alone, some other factor induced in the arteriole by endotoxin must also be
115
present. It is likely that a combination of cytokines induced by endotoxin through the NF-B pathway will be found necessary to decrease vascular tone. CSA binds to CSA receptors found on T cells. Inhibition of T cells is why CSA is effective in transplant immunosuppression. Neutrophils and macrophages are the initial and primary cells involved in fighting sepsis. CSA is a strong, but not the only, chemotactic factor for neutrophils; abrogation of this signal will not eradicate neutrophil and macrophage function. In conclusion, CSA has been shown to be a hypertensive agent that is effective in treating hemorrhagic shock by blocking the action of NF-B and the induction of iNOS. It was extremely effective in preventing the loss of tone after endotoxin in our isolated resistance artery model suggesting that it can also be effective in prevention of vasodilation in other septic shock models. Although it is unlikely that CSA will be used in the current critical care setting for sepsis medications such as CSA with less immunosuppressive properties may become important in immunomodulating patients in the future. REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9. 10.
11.
12.
Glembot TM, Britt LD, Hill MA. Lack of direct endotoxininduced vasoactive effects on isolated skeletal muscle arterioles. Shock 1995;3:216. Snyder JG, Prewitt R, Campsen J, Britt LD. PDTC and Mg132, inhibitors of NF-kappaB, block endotoxin induced vasodilation of isolated rat skeletal muscle arterioles. Shock 2002;17:304. Viol AW, Prewitt R, Doviak M, Britt LD. Endotoxin releases a substance from the aorta that dilates an isolated arteriole by upregulating iNOS. J Surg Res 2005;127:106. Altavilla D, Saitta A, Guarini S, et al. Nuclear factor-kappaB as a target of cyclosporin in acute hypovolemic hemorrhagic shock. Cardiovasc Res 2001;52:143. Lee J, Kim SW, Kook H, Kang DG, Kim NH, Choi KC. Effects of L- arginine on cyclosporin-induced alterations of vascular NO/cGMP generation. Nephrol Dial Transplant 1999;14:2634. Glembot TM, Britt LD, Hill MA. Lack of direct endotoxininduced vasoactive effects on isolated skeletal muscle arterioles. Shock 1995;3:216. Mervaala E, Muller DN, Park JK, et al. Cyclosporin A protects against angiotensin II-induced end-organ damage in double transgenic rats harboring human renin and angiotensinogen genes. Hypertension 2000;35:360. Squadrito F, Altavilla D, Squadrito G, et al. Protective effects of cyclosporin-A in splanchnic artery occlusion shock. Br J Pharmacol 2000;130:339. Fruman DA, Burakoff SJ, Bierer BE. Immunophilins in protein folding and immunosuppression. FASEB J 1994;8:391. Rego A, Vargas R, Suarez KR, Foegh ML, Ramwell PW. Mechanism of cyclosporin potentiation of vasoconstriction of the isolated rat mesenteric arterial bed: Role of extracellular calcium. J Pharmacol Exp Ther 1990;254:799. Rego A, Vargas R, Cathapermal S, Kuwahara M, Foegh ML, Ramwell PW. Systemic vascular effects of cyclosporin A treatment in normotensive rats. J Pharmacol Exp Ther 1991;259:905. Minghini A, Britt LD, Hill MA. Interleukin-1 and interleukin-6 mediated skeletal muscle arteriolar vasodilation: In vitro versus in vivo studies. Shock 1998;9:210.
Cyclosporine Inhibits Endotoxin-Induced Vasodilation of Isolated Rat Resistance Arterioles Vincent J. Obias, M.D.,* Gregory D. Rushing, M.D.,* Russell L. Prewitt, Ph.D.,† Darian Rice, M.D., Ph.D.,† and L.D. Britt, M.D., M.P.H.*,1 *Department of Surgery, †Department of Physiological Sciences, Eastern Virginia Medical School, Norfolk, Virginia Submitted for publication February 20, 2006
Background. An isolated arteriole fails to dilate in response to endotoxin unless a segment of aorta is included in the perfusion system. The unknown substance released by the aorta after exposure to endotoxin is dependent upon the NF-B pathway and induces inducible nitric oxide synthase (iNOS) in the arteriole. The purpose of this study was to determine if cyclosporine A (CSA) that inhibits both NF-B and iNOS would prevent the vasodilatory response to endotoxin. Materials and methods. Rats were injected with either 10 mg/kg of CSA or oil vehicle followed by the removal of a cremaster muscle. The feeding arteriole was isolated from the cremaster and mounted on micropipettes and pressurized to 70 mmHg in a superfused tissue bath. After an hour equilibration to develop spontaneous tone, a 1 cm segment of aorta was placed in the superfusion system upstream from the arteriole and Salmonella enteriditis endotoxin was added to the buffer at a concentration of 2.5 g/mL (ET) or continued infusion of buffer alone. Internal diameters of cannulated arterioles were measured with videomicroscopy and videocalipers for an additional hour. Results. Arterioles downstream from an aorta exposed to vehicle but not endotoxin developed 22.8 ⴞ 3.7% tone that remained unchanged over the following hour. Arterioles exposed to endotoxin started with 22.5 ⴞ 2.8% spontaneous tone and this fell over the following hour to 11.8 ⴞ 3.6%, P < 0.05. Pre-treatment of the rats with CSA tended to increase resting tone and completely prevented the loss of tone after endotoxin. Conclusions. Pre-treatment of the aortic segment with CSA resulted in the development of increased 1
To whom correspondence and reprint requests should be addressed at Department of Surgery, Eastern Virginia Medical School, 825 Fairfax Avenue, Norfolk, VA 23507-1912. E-mail: BrittLD@ evms.edu.
0022-4804/06 $32.00 © 2006 Elsevier Inc. All rights reserved.
tone in the downstream arteriole and completely blocked the vasodilatory response to endotoxin. These results suggest that CSA or a similar compound may be useful in the treatment of septic shock. © 2006 Elsevier Inc. All rights reserved.
Key Words: septic shock; cyclosporine A; endotoxin; nuclear factor kappa B; cremasteric arteriole. INTRODUCTION
Sepsis remains a major cause of morbidity and mortality despite the development of advanced antibiotics and complex algorithms in the management of septic shock. The hypotension characteristic of septic shock has been attributed to a loss of vascular tone in resistance arterioles. Current theories implicate bacterial products such as endotoxin as a major initiator of the septic shock cascade that leads to hypotension. Previous work from our laboratory has demonstrated that endotoxin alone does not reduce tone in isolated resistance arterioles unless a segment of aorta is also present in the bath superfusion system [1]. To elucidate the mechanism behind this phenomenon, segments of de-endothelialized aorta were used and blockade by inodmethacin were attempted. The vasodilation was not dependent on the endothelium nor could it be blocked by indomethacin. This led to the hypothesis that endotoxin causes the release of an unknown factor from the aorta that results in downstream vasodilation. Subsequently, we found that exposure of either the upstream segment of aorta alone or the downstream arteriole alone to pyridine-2,6-bis thiocarboxylate (PDTC), a potent nuclear factor kappa B (NF-B) inhibitor, prevented the vasodilatory response to endotoxin [2]. Thus, the unknown factor was not a direct vasodilator but depended upon NF-B for its production and response in the arteriole. NF-B is known to
112
OBIAS ET AL.: CYCLOSPORINE INHIBITS ENDOTOXIN-INDUCED VASODILATION
up-regulate the inducible form of nitric oxide synthase (iNOS) and we recently found that aminoguanidine, a specific inhibitor for iNOS, also blocked the vasodilatory response [3]. The immunosuppressive phosphatase inhibitor, cyclosporine A (CSA), protects against hemorrhagic shock and restores the reactivity of the aorta to phenylephrine by inhibiting the NF-B pathway [4]. In addition, CSA decreases vascular expression of iNOS in the rat aorta [5]. Thus, CSA appears to be an excellent candidate to block the mechanisms implicated in endotoxic vasodilation and the aim of this study was to determine if CSA can prevent the vasodilation of a rat cremasteric arteriole in response to upstream aortic endotoxin exposure. MATERIALS AND METHODS The Animal Care and Use Committee of Eastern Virginia Medical School approved all experimental protocols. Twenty male SpragueDawley rats between 150 and 270 g were used. They were housed with free access to food and drinking water and were exposed to a 12/12-h light/dark cycle. The rats were anesthetized with an intraperitoneal (i.p.) injection of sodium pentobarbital. A standard dose of 60 mg/kg was used with supplements administered if necessary to maintain anesthesia. The rats were then injected i.p. with either 10 mg/kg of CSA (n ⫽ 10) dissolved in lipid, or oil vehicle only (n ⫽ 10). The scrotal sac was then opened and the cremaster muscle dissected free from the surrounding tissues. The cremaster was then placed in cold (4°C) Krebsbicarbonate-HEPES buffer solution. The cremaster was pinned flat and under a dissecting microscope, a 2 mm segment of first order cremasteric arteriole was isolated, transferred to a tissue bath, and cannulated with micropipettes (Fig. 1). Vessels were secured with 10-0 monofilament suture. The vessel chamber was moved to a Zeiss inverted microscope coupled to a CCD video camera and a highresolution monitor. The arteriole was pressurized to 70 mmHg in the absence of intraluminal flow via an adjustable pressure servo. The vessel was warmed to 34°C, the in situ temperature of the rat cremaster, by continual infusion (4 mL/min) of KREBS-bicarbonate-
113
HEPES solution at pH 7.4. The buffer was bubbled with 95%N 2/5% CO 2 gas, resulting in a pO 2 of 60 mmHg at the level of the vessel. The arteriole was allowed to achieve spontaneous basal tone over 60 min (t ⫽ 0) and the vessels were tested to assure there were no pressure leaks. Internal diameters of cannulated arterioles were measured with videomicroscopy and videocalipers at a resolution of ⫾1 m. After the equilibration period, a 1 cm segment of abdominal aorta was removed from the rat and placed in a flow-through chamber connected in series to the arteriole so that superfused KREBSbicarbonate-HEPES buffer passed over the aorta and onto the cannulated arteriole. At this time (t ⫽ 60) Salmonella enteriditis endotoxin was added to the superfusion buffer at a concentration of 2.5 g/mL (ET) (n ⫽ 10) or continued infusion of buffer alone was maintained for an additional hour (t ⫽ 120) (n ⫽ 10). At the end of the study period, maximal vessel diameters and the confirmation of an intact cremasteric arteriole endothelium were determined by exposure to 3 M acetylcholine and viability was demonstrated by a contractile response to 3 M phenylephrine. Unresponsive arteries were judged to be damaged and excluded from the study.
Statistical Analysis Results are shown as a mean ⫾ SEM of the inside diameter and percent tone. Percent tone was calculated by expressing the diameter of the arteriole as a percent of the maximal diameter of the vessel after exposure to 3-M acetylcholine. Statistical significance was determined by the Kruskal-Wallis test (non-parametric ANOVA), and significance was assumed at P ⬍ 0.05.
RESULTS
The internal diameters of the arteries throughout the experimental period are shown in Fig. 2. All groups of vessels developed tone during the equilibration period. This intrinsic or myogenic tone is the hallmark of resistance arteries. Baseline tone of the arterioles was varied. Control, ET, and ET ⫹ CSA arteriolar tones were different, but not found to be statistically significant. The group treated with CSA alone, however, had
FIG. 1. Diagram of the experimental setup. The arteriole was mounted on micropipettes and suspended in a physiological buffer that flowing at 4 mL/min. The vessel, shown in the center, was transilluminated and viewed by video microscopy. The inside diameter of the arteriole was measured with a video caliper system that provides a digital readout free of errors because of parallax.
114
JOURNAL OF SURGICAL RESEARCH: VOL. 136, NO. 1, NOVEMBER 2006
remained unchanged at the end of the experiment (120 min). The two groups in which the aorta was exposed to CSA appeared to develop higher tone than the other groups but there were no statistically significant differences among any of the groups at the 60-min time point. Arterial percent tone fell from 22.5 ⫾ 2.8% to 11.8 ⫾ 3.6% in the group receiving endotoxin alone (ET) and this value was significantly lower than in the other three groups at 120 min. Maintenance of tone at 35.1 ⫾ 3.4% throughout the experimental period in the ET-CSA group indicated that pre-treatment with CSA prevented the fall in vascular tone induced by endotoxin. DISCUSSION
FIG. 2. Mean diameter of the arterioles in the experimental groups over the experimental period. The first 60-min is an equilibration period in which the vessels develop spontaneous tone. At 60 min a section of aorta was introduced into the superfusion line and either endotoxin or vehicle was applied for the next 60 min. The aortas and arterioles of the control group were exposed to vehicle only. The CSA group included an aorta from a rat treated with CSA (10 mg/kg). The ET group contained a vehicle treated aortic segment and was superfused with endotoxin at 2.5 g/mL. The ET-CSA group consisted of an aorta exposed to CSA and superfusion with endotoxin.
a basal tone much more constricted than the other control or experimental groups (P ⬍ 0.05) (Fig. 2). The control group which received neither CSA nor endotoxin maintained the tone without change through the next 60 min. Upon application of acetylcholine at the end of the experimental period, the vessels dilated as shown as the last point on the figure. The arteries exposed to endotoxin had a similar diameter at 60 min but after introduction of endotoxin the diameter gradually increased and was significantly greater than the control diameter at 120 min. The endotoxin exposed arteries (ET) dilated further upon application of acetylcholine indicating that they had not lost all of their tone after the 60-min exposure period. The group with an aortic section from a CSA treated rat constricted to a smaller diameter than the control and the ET group of arteries during the equilibration period (P ⬍ 0.05) and they maintained that reduced diameter until dilated with acetylcholine. The group exposed to both ET and CSA in the aorta (ET-CSA) also constricted to a smaller diameter but were statistically different only from the ET group of arteries at 60 min. Despite the exposure to endotoxin, these arteries downstream from a CSA treated aorta maintained their constricted state throughout the following 60-min and then dilated in response to the acetylcholine. Because of differences in relaxed diameter of the arteries, the amount of vascular tone was calculated as a percent of the relaxed diameter and expressed as a percent. Percent tone in the control group was 22.8 ⫾ 3.7 at the end of the equilibration period (60 min) and
Previous studies from our laboratory demonstrated that endotoxin alone was not sufficient to cause vasodilation in isolated rat cremaster arterioles. Glembot et al. discovered that a section of rat aorta was required upstream of an isolated arteriole in order for the arteriole to dilate in response to endotoxin [6]. As in the present experiments, Glembot et al. placed an isolated segment of aorta in the superfusion line where it was exposed to the buffer solution mixed with endotoxin. The resultant mixture was then allowed to flow downstream to the isolated rat arteriole and the arteriole was noted to vasodilate. It was hypothesized that some substance was released from the upstream section of aorta after endotoxin exposure that caused a loss of tone in the downstream resistance vessel. Our model required slight modification, because of the lipophylic nature of CSA. While addition of CSA to the isolated aortic segment in buffer solution would have been preferred, this was not possible. Vasodilators known to be produced by the endothelium are NO and PGI 2 but removal of the aortic endothelium or application of indomethacin did not prevent the vasodilatory response. Additional experiments by Snyder et al. demonstrated that pretreatment of rats with PDTC or MG132 (both blockers of the NF-B pathway), prevented the endotoxin-induced vasodilatory response [2]. In addition, the response was prevented if either the aorta or the arteriole alone was from a rat receiving PDTC. Thus, the substance released by the aorta was not a direct vasodilator but was a product associated with the NF-B pathway and the response by the arteriole also was dependent upon the NF-B pathway. A major contributor to the loss of tone in septic shock is the up-regulation iNOS that is capable of producing large amounts of NO. NF-B is a DNA binding protein that is required for transcription of many proinflammatory genes including iNOS. Viol et al. recently showed that aminoguanidine, a specific inhibitor for iNOS, prevented the vasodilatory response to endotoxin by the downstream arteriole [3]. Because CSA has been shown to inhibit NF-B as well as iNOS [7], it
OBIAS ET AL.: CYCLOSPORINE INHIBITS ENDOTOXIN-INDUCED VASODILATION
promised to be efficacious in preventing the response to endotoxin in our isolated arteriolar model. The results of these experiments clearly show that CSA completely prevented the loss of tone after exposure to endotoxin suggesting that it may be useful in therapy for septic shock. In support of this conclusion, Altavilla et al. demonstrated that CSA increased the survival rate in a rat model of hemorrhagic shock and decreased the amount of NF-B in the nucleus by preventing the loss of IB that keeps NF-B bound and inactive [4]. In shock induced by splanchnic artery occlusion, CSA was also effective to revert hypotension, restore vascular reactivity to phenylephrine, and blunt iNOS induction and activity resulting in lower plasma nitrite/nitrate concentrations [8]. In our study, CSA also had an effect on basal arteriolar tone. CSA is thought to block NF-B activation by inhibiting the action of calcineurin that correlates with its immunosuppressive activity [9]. Regardless of the mechanism, the inhibition of iNOS is the major beneficial effect for preventing loss of vascular tone after endotoxin exposure. Although not statistically significant, the tendency of the arterioles downstream from CSA treated aortic segments to develop more tone is consistent with previous reports that CSA increases vasoconstriction both in vivo and in vitro. Rego et al. showed that CSA increases responsiveness to norepinephrine and potassium in the mesenteric vascular bed [10] and raises blood pressure in the rat independent of sympathetic nerve activity [11]. Lee et al. showed that acute application reduced the vasodilatory response to acetylcholine, an eNOS activator, in isolated rat aortas. In addition it lowered levels of cGMP that are elevated by NO produced by eNOS in response to acetylcholine. Thus, some inhibition of eNOS may contribute to the sustained tone of our isolated arterioles in the presence of CSA but as seen in Fig. 2, the vasodilatory response to acetylcholine was still present. The identity of the unknown factor released by the aorta exposed to endotoxin is of great interest. Its effectiveness when secreted into a buffer flowing at 4 mL/min suggests a powerful cytokine that works at extremely low concentrations such as TNF-␣ or one of the interleukins also downstream in the NF-B cascade. Previous work from our laboratory has shown that IL-1 and IL-6 do not dilate isolated arterioles although they are quite effective when applied to the same arteriole in situ where parenchymal tissue may provide an unknown factor [12]. TNF-␣ was also found to have no apparent affect on the diameter of isolated arterioles unless they were pre-treated with endotoxin [6]. Therefore, TNF-␣ may contribute to the response, but as it is ineffective when applied alone, some other factor induced in the arteriole by endotoxin must also be
115
present. It is likely that a combination of cytokines induced by endotoxin through the NF-B pathway will be found necessary to decrease vascular tone. CSA binds to CSA receptors found on T cells. Inhibition of T cells is why CSA is effective in transplant immunosuppression. Neutrophils and macrophages are the initial and primary cells involved in fighting sepsis. CSA is a strong, but not the only, chemotactic factor for neutrophils; abrogation of this signal will not eradicate neutrophil and macrophage function. In conclusion, CSA has been shown to be a hypertensive agent that is effective in treating hemorrhagic shock by blocking the action of NF-B and the induction of iNOS. It was extremely effective in preventing the loss of tone after endotoxin in our isolated resistance artery model suggesting that it can also be effective in prevention of vasodilation in other septic shock models. Although it is unlikely that CSA will be used in the current critical care setting for sepsis medications such as CSA with less immunosuppressive properties may become important in immunomodulating patients in the future. REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9. 10.
11.
12.
Glembot TM, Britt LD, Hill MA. Lack of direct endotoxininduced vasoactive effects on isolated skeletal muscle arterioles. Shock 1995;3:216. Snyder JG, Prewitt R, Campsen J, Britt LD. PDTC and Mg132, inhibitors of NF-kappaB, block endotoxin induced vasodilation of isolated rat skeletal muscle arterioles. Shock 2002;17:304. Viol AW, Prewitt R, Doviak M, Britt LD. Endotoxin releases a substance from the aorta that dilates an isolated arteriole by upregulating iNOS. J Surg Res 2005;127:106. Altavilla D, Saitta A, Guarini S, et al. Nuclear factor-kappaB as a target of cyclosporin in acute hypovolemic hemorrhagic shock. Cardiovasc Res 2001;52:143. Lee J, Kim SW, Kook H, Kang DG, Kim NH, Choi KC. Effects of L- arginine on cyclosporin-induced alterations of vascular NO/cGMP generation. Nephrol Dial Transplant 1999;14:2634. Glembot TM, Britt LD, Hill MA. Lack of direct endotoxininduced vasoactive effects on isolated skeletal muscle arterioles. Shock 1995;3:216. Mervaala E, Muller DN, Park JK, et al. Cyclosporin A protects against angiotensin II-induced end-organ damage in double transgenic rats harboring human renin and angiotensinogen genes. Hypertension 2000;35:360. Squadrito F, Altavilla D, Squadrito G, et al. Protective effects of cyclosporin-A in splanchnic artery occlusion shock. Br J Pharmacol 2000;130:339. Fruman DA, Burakoff SJ, Bierer BE. Immunophilins in protein folding and immunosuppression. FASEB J 1994;8:391. Rego A, Vargas R, Suarez KR, Foegh ML, Ramwell PW. Mechanism of cyclosporin potentiation of vasoconstriction of the isolated rat mesenteric arterial bed: Role of extracellular calcium. J Pharmacol Exp Ther 1990;254:799. Rego A, Vargas R, Cathapermal S, Kuwahara M, Foegh ML, Ramwell PW. Systemic vascular effects of cyclosporin A treatment in normotensive rats. J Pharmacol Exp Ther 1991;259:905. Minghini A, Britt LD, Hill MA. Interleukin-1 and interleukin-6 mediated skeletal muscle arteriolar vasodilation: In vitro versus in vivo studies. Shock 1998;9:210.