Effects of carbon monoxide on trout and lamprey vessels - Regulatory ...

2 downloads 45 Views 499KB Size Report
Nov 12, 2008 - Zinc protopophyrin-IX (ZnPP-IX, 0.3–30 M), a HO inhibitor, elic- ited a small .... concentrations of the HO inhibitor, zinc protoporphyrin IX (ZnPP).
Am J Physiol Regul Integr Comp Physiol 296: R141–R149, 2009. First published November 12, 2008; doi:10.1152/ajpregu.90507.2008.

Effects of carbon monoxide on trout and lamprey vessels Ryan A. Dombkowski,1 Nathan L. Whitfield,2,3 Roberto Motterlini,4 Yan Gao,2 and Kenneth R. Olson2 1 Department of Biology, Saint Mary’s College, Notre Dame, Indiana; 2Indiana University School of Medicine–South Bend, South Bend, Indiana; 3Department of Biology, University of Notre Dame, Notre Dame, Indiana; and 4Vascular Biology Unit, Department of Surgical Research, Northwick Park Institute for Surgical Research, Harrow, United Kingdom

Submitted 17 June 2008; accepted in final form 4 November 2008

Dombkowski RA, Whitfield NL, Motterlini R, Gao Y, Olson KR. Effects of carbon monoxide on trout and lamprey vessels. Am J Physiol Regul Integr Comp Physiol 296: R141–R149, 2009. First published November 12, 2008; doi:10.1152/ajpregu.90507.2008.—Carbon monoxide (CO) is endogenously produced by heme oxygenase (HO) and is involved in vascular, neural, and inflammatory responses in mammals. However, the biological activities of CO in nonmammalian vertebrates is unknown. To this extent, we used smooth muscle myography to investigate the effects of exogenously applied CO (delivered via a water-soluble CO-releasing molecule, CORM-3) on isolated lamprey (Petromyzon marinus) dorsal aortas and examined its mechanisms of action on trout (Oncorhynchus mykiss) efferent branchial (EBA) and celiacomesenteric (CMA) arteries. CORM-3 dose-dependently relaxed all vessels examined. Trout EBA were twofold more sensitive to CORM-3 when precontracted with norepinephrine (NE) than KCl and CORM-3 relaxed five-fold more of the NE- than KCl-induced tension. Glybenclamide (10 ␮M), an ATPsensitive potassium channel inhibitor, inhibited NE-induced contraction, but did not affect CORM-3-induced relaxation. NS-2028 (10 ␮M), a soluble guanylyl cyclase inhibitor, had no effect on a NEcontraction, but inhibited a subsequent CORM-3-induced relaxation. Zinc protopophyrin-IX (ZnPP-IX, 0.3–30 ␮M), a HO inhibitor, elicited a small, yet dose-dependent and significant, increase in baseline tension but did not have any effect on subsequent NE-induced contractions or a nitric oxide-induced relaxation (via sodium nitroprusside). [ZnPP-IX] greater than 3 ␮M, however, significantly reduced the predominant vasodilatory response of trout EBA to hydrogen sulfide. These results implicate an active HO/CO pathway in trout vessels having an impact on resting vessel tone and CO-induced vasoactivity that is at least partially mediated by soluble guanylyl cyclase. gasotransmitters; guanylyl cyclase; hydrogen sulfide; vasodilator

of a group of three intensely studied “gasotransmitters,” small neuroactive and vasoactive gases produced endogenously during the catabolism of common biological molecules. Nitric oxide (NO), the first described gaseous mediator (23), relaxes vessels from most vertebrates but has little or no activity in cartilaginous fish vessels (13, 18) and contracts some vessels from jawless fish (19). The newest member of the group, hydrogen sulfide (H2S) (53), displays vasoactivity in all vertebrate classes but affects vessels in species-specific and even in a tissue-specific manner (11). While considerable research has been performed since the discovery of CO as a vasodilatory molecule (38), studies have focused solely on the cardiovascular effects of CO in mammalian species.

CARBON MONOXIDE (CO) IS ONE

Address for reprint requests and other correspondence: K. R. Olson, Indiana Univ. School of Medicine—South Bend, 1234 Notre Dame Ave., South Bend, IN 46617, USA (e-mail: [email protected]). http://www.ajpregu.org

CO is produced endogenously via the catabolism of heme by heme oxygenase (HO) enzymes, which also generates biliverdin and Fe2⫹. Heme oxygenase-2 (HO-2) is a constitutively expressed isoform, while heme oxygenase-1 (HO-1) is inducible (39). A third isoform, HO-3, has been described in multiple tissues, and while it bears structural similarity to HO-2, its physiological functions are unclear (42). HO-1 and HO-2 have been detected in mammalian vascular smooth muscle cells and CO generated from smooth muscle is thought to be a regulator of vascular tone (54). The vasodilatory effects of CO are predominantly via the activation of soluble guanylyl cyclase and calcium-activated potassium (KCa) channels (54). Even though it is widely accepted that CO is a vasodilator of mammalian blood vessels (15), this is not always the case. For example, CO has been demonstrated to dilate cerebral arterioles in piglets (35) but is ineffective at dilating cerebral vessels in dogs or rabbits (4). This demonstrates heterogeneity of the effects of CO within similar vascular tissues of species from the same vertebrate class. CO production by and regulation within fish vascular smooth muscle is of interest for several reasons. First, it has been reported that CO plays a more significant role in the regulation of mammalian vascular tone in the absence of endothelial NO (4, 33). Fish do not appear to possess endothelial nitric oxide synthase (30, 46, 50), with the possible exception being retinal tissue (25), and thus a greater role for CO might be anticipated. Second, environmental toxins and stress have been demonstrated to alter HO activity or induce HO-1 in both elasmobranch (17) and teleost (2, 7, 34, 43) fishes. Third, organic solvents, pesticides, and trace metals have a demonstrated ability to alter fish HO function (2, 7, 48, 51) and are far more common environmental toxicants in water than in air (28, 48, 59). Thus, aquatic vertebrates subjected to such pollutants may have diminished or altered CO-mediated vasoregulation. With this in mind, we investigated the role of CO in fish vascular smooth muscle by examining the effects of a watersoluble CO-releasing molecule (CORM-3) on efferent branchial (EBA) and celiacomesenteric (CMA) arteries of the rainbow trout and dorsal aorta of the sea lamprey. The potential mechanisms involved in the pharmacological action mediated by CO was also examined in the context of the interplay between CO and the other two “gasotransmitters”, NO and H2S. MATERIALS AND METHODS

All experiments were approved by the Institutional Animal Care and Use Committee. The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

0363-6119/09 $8.00 Copyright © 2009 the American Physiological Society

R141

R142

CARBON MONOXIDE AND FISH VESSELS

Animals Sea lamprey (Petromyzon marinus, 130 – 450 g) were trapped in streams feeding into the Great Lakes during the spring-summer spawning season and airlifted to the University of Notre Dame, where they were maintained in 500-liter rectangular tanks with aerated, flowing well water (15°C) and exposed to a 12:12-h light-dark photoperiod. They were not fed. Lampreys were anesthetized in benzocaine (1:5,000, wt/vol), and the dorsal aortas were dissected out and placed in HEPES buffer at 4°C until use. Rainbow trout (Oncorhynchus mykiss, kamloops strain, 0.3– 0.8 kg) of either sex were used for studies on isolated vessels. They were purchased from a local hatchery (Harrietta Hills Trout Farm, Harrietta, MI), kept in circulating 2,000-liter tanks containing throughflowing well water at 12–15°C, aerated with room air, and exposed to 12:12-h light-dark cycles. The fish were fed a maintenance diet of commercial trout pellets (Purina, St. Louis, MO). Trout were stunned by a blow to the head, and the EBA and CMA were removed and placed in HEPES buffer at 4°C until use. All procedures followed National Institutes of Health guidelines. Myography Vessels were cut into approximately one-half centimeter-long segments and mounted on 280-␮M-diameter stainless-steel wire hooks and suspended in 5-ml water-jacketed smooth muscle baths filled with the appropriate buffer at 14°C and aerated with room air. The bottom hooks were stationary; the upper ones were connected to Grass model FT03C force-displacement transducers (Grass Instruments, West Warwick, RI). Tension was measured on a Grass Model 7E or 7F polygraph (Grass Instruments, West Warwick, RI). Polygraph sensitivity was set to detect changes as small as 5 mg. Data were archived on a PC computer at 1 Hz using SoftWire (Measurement Computing, Middleboro, MA). The chart recorders were calibrated prior to each experiment. Length-tension relationships have been derived from KCl-contracted vessels from both species in other studies (unpublished) in this laboratory. These, plus general vessel size and physical characteristics were taken into account to apply a reasonable baseline (resting) tension (500 –700 mg for all vessels) for 0.5–1 h prior to experimentation. The vessels were then contracted with 80 mM KCl, washed twice, and resting tension re-established. The vessels were contracted a second time with 80 mM KCl, washed twice, and resting tension re-established for a minimum of 30 min before further experimentation. This procedure has been used to achieve optimum in vitro vessel activity (Olson KR, unpublished observations). Unless otherwise noted, vessels were used for only one experiment. In the lamprey dorsal aorta, two concentrations, 200 and 500 ␮M, of CORM-3 [CORM hereafter; tricarbonylchloroglycinato ruthenium(II), Ru(CO)3Cl(NH2CH2CO2); structure shown in Fig. 1D, inset], which when placed in buffer liberates 1 mol CO/mol CORM-3 (6, 20, 44), were examined on otherwise unstimulated dorsal aortas and vessels that had been maximally precontracted with either 1 ␮M norepinephrine (NE) or 100 nM U-46619, a synthetic thromboxane A2 mimetic. The cumulative dose-response characteristics of CORM-3 were examined in otherwise unstimulated rainbow trout vessels and in vessels precontracted with 80 mM KCl or 1 ␮M NE. All vessels precontracted with NE in this study were treated with 1 ␮M propranolol 10 min prior to NE addition to minimize the potential for NE-mediated vasodilation. To determine whether there is residual vasoactivity in the CORM remaining after CO is liberated, we dissolved CORM-3 in buffer and bubbled it overnight with a gentle stream of room air. The residual “inactive” CORM (300 ␮M) had no effect on KCl-contracted EBA (n ⫽ 3; not shown). Thus it was assumed that, similar to studies on mammalian vessels (20), all CORM-3 vasoactivity was due to CO. Because the addition of CORM predominantly resulted in a decrease in vessel tension, we examined the potential mechanism of AJP-Regul Integr Comp Physiol • VOL

CORM-induced relaxation in trout EBA precontracted with 1 ␮M NE. We used a single concentration of CORM (1 mM), which produced a near maximal (⬃85%) response. Briefly, EBA were contracted twice with KCl and then allowed to return to resting tension for at least 30 min as described above. The EBA were then incubated with either 10 ␮M NS-2028, an inhibitor of soluble guanylyl cyclase; 10 ␮M glybenclamide, an ATP-sensitive K channel (KATP) antagonist; 50 nM apamin, a small conductance KCa channel antagonist; or 50 nM charybdotoxin, a large-conductance KCa channel antagonist. Each agent was used independently and added to the bath 30 min prior to further experimentation. The EBA were then precontracted with 1 ␮M NE and allowed to reach a plateau phase of contraction, at which time 1 mM CORM was added. To confirm the inhibition of soluble guanylyl cyclase by NS-2028, the effect of 100 ␮M sodium nitroprusside (SNP), an NO donor, on NE contraction was also examined in trout EBA with and without NS-2028. Several previous studies have investigated and implicated a relationship between CO and both H2S (24, 31, 47, 64) and NO (3, 8, 21, 63) production pathways. To investigate the effects of endogenous CO production and its interplay with H2S, EBA were exposed to varying concentrations of the HO inhibitor, zinc protoporphyrin IX (ZnPP). Different groups of EBA were incubated with ZnPP ranging from 0.3 to 30 ␮M and were then precontracted with 1 ␮M NE; each vessel was exposed to only one ZnPP concentration. After the NE contraction plateaued, 300 ␮M Na2S (which generates H2S in solution) was added to the bath. Since 300 ␮M Na2S produces a triphasic response in the trout EBA (10), with the third dilatory phase predominating in NE-precontracted vessels, only the magnitude of the prolonged relaxation was examined. To examine any possible interaction of CO with NO, vessels were incubated with 30 ␮M ZnPP, contracted with 1 ␮M NE, and then exposed to 100 ␮M SNP, a known NO donating agent and vasodilator in trout (46). Polarographic Measurement of Sulfide Binding by Zinc Protoporphyrin The above experiments showed that ZnPP inhibited Na2S relaxations (see also Fig. 6). To determine whether this was due to ZnPP inhibiting a H2S-mediated increase in CO production, or a direct effect of ZnPP binding to or catalyzing degradation of H2S, we used a polarographic H2S sensor to measure the concentration of H2S gas in buffer in the presence and absence of ZnPP. This polarographic H2S sensor (9, 58) measures H2S gas in real time and has a resolution to 14 nM H2S or ⬃200 nM total sulfide, depending on pH (58). The sensor was connected to an Apollo 4000 Free Radical Analyzer (WPI, Sarasota FL). In one experiment, sulfide binding by zinc protoporphyrin-IX (ZnPP) was determined in a closed 1-ml chamber at 14°C by adding increasing concentrations of ZnPP (1, 3, 10, 30 ␮M) to pH 7.8 HEPES buffer spiked to 300 ␮M Na2S. H2S concentration was converted to total sulfide based on pH (58). In a second experiment, in an attempt to determine binding efficiency and duration, 10 ␮M ZnPP was first added to the chamber followed by three consecutive additions of 5 ␮M Na2S and then two subsequent additions of 5 ␮M ZnPP. Data Analysis Data were analyzed using BIOPAC software (BIOPAC Systems, Goleta, CA). The effective concentration producing half-maximal response (EC50) was determined from dose-response curves for individual vessels using Table Curve (Jandel, Chicago, IL). Student’s t-tests or one-way ANOVA repeated-measures followed by Bonferroni’s t-test for multiple comparisons (SigmaStat, Jandel Corp) were used for comparisons between groups of vessels. Results are expressed as means ⫾ SE. Significance was assumed at P ⱕ0.05. 296 • JANUARY 2009 •

www.ajpregu.org

R143

CARBON MONOXIDE AND FISH VESSELS

Chemicals HEPES buffer contained (in mM): 145 NaCl, 3 KCl, 0.57 MgSO4 䡠 7H2O, 2 CaCl2 䡠 2H2O, 5 glucose, 3 HEPES acid, and 7 HEPES Na⫹ salt, pH 7.8. All buffers were stored in the refrigerator and used within 72 h of preparation. Unless otherwise stated, all chemicals were purchased from Sigma-Aldrich (St. Louis, MO). CORM-3 was synthesized and handled as previously described (6). RESULTS

Effect of CORM on Lamprey Vessels CORM at 200 and 500 ␮M relaxed lamprey dorsal aortas precontracted with NE (1 ␮M) and U-46619 (100 nM) but had only a small transient dilatory effect on otherwise unstimulated vessels (Fig. 1). The extent of relaxation mediated by CORM on U-46619-contracted vessels was significantly greater than the one observed in NE-contracted vessels. However, the force generated by U-46619 was significantly less than that produced by NE, and the absolute magnitude of the 200 and 500 ␮M CORM-mediated relaxation was similar between the groups. Effect of CORM on Trout Vessels CORM relaxed unstimulated trout EBA and CMA with similar EC50, 393 ⫾ 111 and 464 ⫾ 93 ␮M, respectively (data not shown). The maximal relaxation of unstimulated EBA and CMA was 27 ⫾ 14% and 32 ⫾ 17% of resting tension (500 –700 mg), respectively. Cumulative dose-response curves for CORM-induced relaxation of EBA precontracted with NE or KCl are shown in Fig. 2 and 3. EBA were ⬃2-fold more sensitive (P ⬍ 0.05) to CORM when precontracted with NE than with KCl (Fig. 2B; EC50, 352 ⫾ 77 and 698 ⫾ 21 ␮M CORM, respectively). CORM-mediated relaxation was also approximately five-fold more efficacious in EBA precontracted with NE (⬃80% relaxation) than in KCl-contracted EBA (⬃15% relaxation; Fig. 3). The threshold for CORM-induced relaxation was ⬃30 ␮M. NE (1 ␮M)-induced contractions in control EBA were ⬃80% of the maximal contractions elicited by 80 mM KCl. A bolus of 1 mM CORM applied to unstimulated (Fig. 4A) or NE-precontracted (Fig. 4B) vessels produced a biphasic response, a brief 1–2 min contraction (phase 1) followed by a

prolonged relaxation (phase 2). The magnitudes of both phases were significantly larger in vessels precontracted with NE than in unstimulated vessels. Glybenclamide (10 ␮M) was the only antagonist to have an effect on the NE-induced contraction, reducing it by ⬃50% (Fig. 4C). NE-induced contractions were unaffected by apamin (50 nM), charybdotoxin (50 nM), or NS2028 (10 ␮M). Incubation with apamin (50 nM), charybdotoxin (50 nM), or glybenclamide (10 ␮M) had no effect on either phase of the CORM-induced response of NE-precontracted EBA, while application of 10 ␮M NS-2028 resulted in nearly complete inhibition of both phases (Fig. 4D). ZnPP (0.1–30 ␮M) produced a dose-dependent increase in baseline tension in trout EBA (Fig. 5A). However, the presence of ZNPP did not affect a subsequent NE contraction (Fig. 5B). Na2S (300 ␮M) elicited a triphasic relaxation-contractionrelaxation in NE-precontracted EBA (Fig. 6A; cf. Ref 10). Pretreatment with ⱖ3 ␮M ZnPP significantly reduced the magnitude of the Na2S relaxation in an apparent dose-dependent relationship (Fig. 6, B and C). ZnPP (30 ␮M) did not affect the ability of SNP (100 ␮M) to relax NE-contracted EBA (Fig. 7, A and C), whereas in vessels treated with both NS-2028 (10 ␮M) and ZnPP, the SNP relaxation was abolished (Fig. 7, B and C). Polarographic Measurement of Sulfide Binding by Zinc Protoporphyrin ZnPP appeared to bind sulfide in a stable fashion (Fig. 8A). Na2S (300 ␮M) produced a large electrode current followed by a fall, indicating sulfide oxidation and/or escape from the chamber. One and three micromoles of ZnPP had little effect on the sulfide current, but 10 and 30 ␮M ZnPP produced incremental reductions in the electrode current. However, within a minute or two after ZnPP addition, the electrode current became stable, indicating that ZnPP does not catalyze H2S degradation. A second experiment to clarify the stoichiometry of sulfide-ZnPP binding was performed using lower ZnPP and Na2S concentrations. Prior addition of 10 ␮M ZnPP reduced the 5 ␮M sulfide current by ⬃80% but had little to no effect on the current increase generated by two subsequent 5 ␮M Na2S spikes (Fig. 8B). Two subsequent additions of 5 ␮M ZnPP produced reductions in electrode current approximately

Fig. 1. Typical traces of the effects of 200 and 500 ␮M CORM-3 on otherwise unstimulated (A), 1 ␮M norepinephrine-precontracted (B), and 100 nM U-46619-precontracted (C) lamprey dorsal aortas. Scale: 15 min and 200 mg. D: The effects of 200 ␮M (open bars) and 500 ␮M (solid bars) CORM-3 on 1 ␮M NEprecontracted (NE, n ⫽ 8) and 100 nM U-46619-precontracted (n ⫽ 4) lamprey dorsal aortas. Inset: structure of CORM-3. *Significant relaxations from the contraction plateau. **Significant differences from both plateau and from the 200 ␮M-induced contractions.

AJP-Regul Integr Comp Physiol • VOL

296 • JANUARY 2009 •

www.ajpregu.org

R144

CARBON MONOXIDE AND FISH VESSELS

Fig. 2. A: typical trace of the CORM-3 cumulative dose-response curve in a trout efferent branchial artery (EBA) precontracted with 1 ␮M NE. Numbers ⫽ ⫺log M CORM-3 concentration; scale, 15 min and 200 mg tension. B: cumulative dose-response curve of CORM3-induced relaxations of 1 ␮M NE- (open squares, n ⫽ 8) and 80 mM KCl- (diamonds, n ⫽ 8) precontracted trout EBA. Values expressed as a percentage of total (100%) relaxation, means ⫾ SE. EBA precontracted with NE (EC50 ⫽ 352 ⫾ 77 ␮M) were approximately twofold more sensitive to CORM compared with those precontracted with KCl (EC50 ⫽ 698 ⫾ 21 ␮M). *Significantly different from the other prestimulus at same CORM-3 concentration.

equivalent to the 5 ␮M Na2S-induced increases in current. These results suggest that in a cell-free environment, ZnPP binds sulfide in an approximate 1:1 ratio.

vasoactivity has been well conserved throughout vertebrate evolution. CORM-3-Induced Relaxation and Mechanisms of Action

DISCUSSION

The importance of CO as a physiologically relevant signaling molecule in a variety of mammalian systems has been well documented (61). Here, we show that CO produces vasorelaxation in vessels from lamprey and trout. Otherwise, unstimulated trout vessels appear to respond only slightly to CO and CO-induced relaxation in precontracted trout vessels is partially dependent on activation of guanylyl cyclase (GC). The decreased efficacy of CO in KCl-depolarized vessels also suggests that a portion of the CO response involves either potassium or chloride channel activation, but small-conductance or large-conductance KCa channels do not appear to be involved. To our knowledge, this is the first report of COmediated relaxation of vascular smooth muscle from nonmammalian vertebrates. Furthermore, these studies suggest that CO

Fig. 3. CORM-3-induced relaxation of 1 ␮M norepinephrine (NE, squares; n ⫽ 8) or 80 mM KCl (diamonds; n ⫽ 8) precontracted trout efferent branchial arteries. Values are expressed as a percentage of the precontraction force relaxed by CORM-3 (means ⫾ SE). CORM ⬎100 ␮M produced a significantly (*) greater relaxation of the NE contraction. At 10 mM, the CORM relaxation of a NE contraction was approximately five-fold greater than that of the CORM relaxation of KCl-contracted vessels. AJP-Regul Integr Comp Physiol • VOL

The vasorelaxation produced in lamprey and trout vessels by CORM is consistent with CO-mediated relaxation of mammalian blood vessels (6, 20, 61). Both GC and cGMP-independent activation of big-conductance calcium-activated potassium channels (BKCa channels) mediate the endothelium-independent aspect of this response in mammals (20, 54). In the present study, we found that activation of GC also appears to be involved in CO relaxation of trout EBA, but evidence for a contribution of potassium channels to this response is more tenuous. GC and cGMP. CO binds ferrous iron of heme proteins, which generally inhibits their function (61). The notable exception to this is GC (61), and the basis of CO activation of soluble GC has been described (36). In this study, NS-2028, a specific inhibitor of soluble GC in mammalian tissue (45) almost completely inhibited CORM-mediated relaxation of trout EBA (Fig. 4D) and completely inhibited the vasodilatory effect of SNP in trout EBA (Fig. 7). Because SNP releases another soluble GC activator, NO, it seems reasonable to conclude that NS-2028 is also an effective inhibitor of trout CG and that much, if not all, of the vasodilatory effect of CO in trout is mediated through cGMP formation. The brief vasoconstriction that we observed shortly after the addition of CORM to trout EBA (Fig. 4) has not, to our knowledge, been observed in mammalian vessels. This constriction is significantly greater in prestimulated vessels, which suggests that an increased level of cytosolic Ca2⫹, and possibly Ca2⫹ cycling, during cell activation may enhance the contractile response. We think it is unlikely that this contractile response is mediated by anything other than CO because the degassed inactive CORM (iCORM) was not vasoactive in EBA, nor has it been found to be vasoactive in mammalian vessels (20). The failure to observe a similar contraction in mammalian vessels, however, suggests that this is more likely a unique response of trout vessels to CO. Interestingly, both the constriction and relaxation were inhibited by NS-2028. This suggests that the contraction is also cGMP dependent. The CO-mediated contraction could be due to the presence of a sensitive cGMP-dependent Cl- channel (41, 40), but this remains to be determined. 296 • JANUARY 2009 •

www.ajpregu.org

R145

CARBON MONOXIDE AND FISH VESSELS

Fig. 4. Typical traces of the biphasic contraction-relaxation of 1 mM CORM-3 on otherwise unstimulated (A) and 1 ␮M norepinephrine (NE)-precontracted (B) trout efferent branchial arteries. The horizontal scale ⫽ 15 min. C: effects of inhibitors on 1 ␮M NE contraction of EBA. Values of NE-induced contractions are expressed as a percent of 80 mM KCl control contraction in otherwise untreated controls (C, n ⫽ 8) or vessels incubated 30 min prior to NE addition with NS-2028 (NS, 10 ␮M, n ⫽ 4), glybenclamide (GLY; 10 ␮M, n ⫽ 4), apamin (APA, 50 nM, n ⫽ 4), or charybdotoxin (CH, 50 nM, n ⫽ 4). *Significantly different from the control (C) group. D: effects inhibitors on CORM-3 biphasic response in NE-precontracted vessels as a percentage of the NE-induced tension. *Significantly different from control (C) group.

Ion channels. Trout EBA precontracted with a ligand such as NE were approximately twofold more sensitive to the COreleasing agent than vessels precontracted with KCl (Fig. 2). In addition, CORM was fivefold more potent in NE- than in KCl-contracted vessels (Fig. 3). This suggests that either a portion of the CO-induced vasodilation is mediated by ion channels that are rendered ineffective when the cell is depolarized or that the Ca2⫹ influx associated with KCl depolarization cannot be overcome by CO. The two most likely candidates for CO-activated channels are K⫹ and/or Cl⫺ channels. Although one study has suggested that CO activates KATP channels (20), most (54 –56, 60) have focused more on CO binding and activation of large-conductance Ca2⫹-activated K⫹ (BKCa) channels. Previous studies have demonstrated the apparent presence of KCa channels in isolated retinal ganglion cells from trout (26, 27), and even charybdotoxin-sensitive gill epithelial cells from sea bass, Dicentrarchus labrax (16). However, these channels have yet

to be characterized in trout vascular smooth muscle. The ineffectiveness of glibenclamide, apamin, and charybdotoxin in our experiments (Fig. 4D) suggest that either 1) KCa and KATP channels are not present in trout EBA, or 2) KCa and KATP channels are not a molecular target of CO in these fish. It should be noted that our experience with glibenclamide as an inhibitor of KATP in trout tissues has been generally inconclusive (Olson KR, unpublished observations), which leads us to question whether these channels are even functional in trout vessels. The inhibitory effect of glibenclamide on the NE-induced contraction of trout EBA (Fig. 5) may shed some light on another potential mechanism of cell activation in trout vessels. In addition to its inhibitory effect on mammalian KATP channels, glibenclamide is also an inhibitor of cystic fibrosis transmembrane conductance regulator (CFTR) Cl⫺ channels and of swelling- and Ca2⫹-activated Cl⫺ channels (49, 62). Furthermore, NE-induced contraction in mammalian vessels is par-

Fig. 5. Effects of zinc protoporphyrin-IX (ZnPP) on trout efferent branchial arteries. A: ZnPP produced a dose-dependent contraction of unstimulated vessels. B: norepinephrine (1 ␮M) contractions were unaffected by pretreatment with ZnPP (means expressed as a % of 80 mM KCl control contraction). *Significantly different from control; control, n ⫽ 24; all other, n ⫽ 16.

AJP-Regul Integr Comp Physiol • VOL

296 • JANUARY 2009 •

www.ajpregu.org

R146

CARBON MONOXIDE AND FISH VESSELS

Fig. 6. Effects of ZnPP on the H2S-induced relaxation of trout efferent branchial arteries. A: typical trace showing the triphasic response to H2S (300 ␮M Na2S) in norepinephrine- (NE; 1 ␮M) precontracted EBA. B: typical trace of the same protocol after treatment with ZnPP (30 ␮M). C: overall relaxation after 30 min of H2S exposure represented as a percent relaxation of the NE-induced precontraction. Control, n ⫽ 24; all other, n ⫽ 16. Scale, 30 min and 1 g. *Significantly different from the control group. **Significantly different from all other treatment groups.

tially dependent on an increase in Cl⫺ permeability and subsequent Cl⫺ efflux (37). Collectively, these studies suggest that Cl⫺ channels are also involved in NE-mediated contraction of trout EBA.

tonic level of CO on vessel tonus. Since HO activity has been demonstrated in both bony (1, 2, 7, 51, 52) and cartilaginous (17) fishes, this activity is not unexpected.

Endogenous CO Production

Interactions of CO With Other Vasoactive Gases, NO, and H2S

Mammalian blood vessels possess a complete enzymatic pathway for de novo heme biosynthesis, and it appears that local heme production is sufficient to provide substantial substrate for CO production (29). The nonspecific heme-oxygenase (HO) inhibitor zinc protoporphyrin-IX (ZnPP) elicited a small, yet dose-dependent increase in resting trout EBA tension (Fig. 6A). Although we cannot rule out a potential stimulatory effect of ZnPP, these studies suggest that there is a

In addition to its direct effects, CO may also exert activity indirectly through interactions with mechanisms mediated by nitric oxide (NO) and hydrogen sulfide (H2S). For example, NO forms a pentacoordinate complex with the GC heme, whereas CO forms a hexacoordinate complex of CO with the GC heme, resulting in decreased potency (36). Thus, depending on the relative degree of GC activation by NO, CO may either exacerbate or ameliorate the NO-induced dilation. A

Fig. 7. The effects of sodium nitroprusside (SNP, 100 ␮M) on norepinephrine (NE, 1 ␮M)-precontracted trout efferent branchial arteries. A: typical trace of SNP-induced relaxation of NE-precontracted vessels pretreated with ZnPP (30 ␮M). B: typical trace of SNP-induced relaxation of NE-precontracted vessels pretreated with ZnPP (30 ␮M) and NS-2028 (10 ␮M). C: effects of SNP (100 ␮M) on NE (1 ␮M)-precontracted vessels that were otherwise untreated (control, C), incubated with 30 ␮M ZnPP, or incubated with both 10 ␮M NS-2028 and 30 ␮M ZnPP. Scale 15 min and 1 gram of tension; *Significantly different from both other groups.

AJP-Regul Integr Comp Physiol • VOL

296 • JANUARY 2009 •

www.ajpregu.org

R147

CARBON MONOXIDE AND FISH VESSELS

Fig. 8. Tracings demonstrating the polarographic measurement of H2S binding by ZnPP in a cell-free environment. H2S concentration was converted to total sulfide. A: increasing concentrations of ZnPP (1, 3, 10, and 30 ␮M; closed arrows) proportionally reduced the 300 ␮M Na2S-generated current (open arrow). B: 10 ␮M ZnPP (closed arrow) essentially eliminated one 5 ␮M Na2S spike (open arrow) but had little effect on subsequent 5-␮M Na2S spikes. Two 5-␮M ZnPP spikes produced reductions in the Na2Sgenerated signal, approximately equivalent to the 5 ␮M Na2S spikes.

direct CO-mediated inhibition of NO synthesis has also been reported (32). Inhibition of the CO/HO pathway by H2S and of H2S synthesis by CO/HO has been proposed in aortic smooth muscle (31), whereas the CO/HO pathway may be upregulated by H2S in pulmonary vessels (47). There is considerable debate on the presence of a NO system in fish, although recent evidence supports the presence of nitrigeric perivascular nerves and absence of an endothelial source of NO production (12). Certainly, there is little evidence to support endothelial NO production in trout EBA (46). Thus CO-NO interactions are not likely to occur in isolated trout vessels. On the other hand, our studies suggest that there may be an interaction between CO and H2S because ZnPP had a substantial inhibitory effect on the third phase of H2S relaxation of trout EBA (Fig. 6). To determine whether the inhibitory effect of ZnPP was indirect, due to either ZnPP binding with H2S or catalytically degrading H2S, we measured the effect of ZnPP on H2S in a cell-free solution in real time using a polarographic H2S sensor. As shown in Fig. 8, ZnPP appears to bind to H2S in a 1:1 ratio but does not appear to catalytically degrade it. Because we found that the vasorelaxation produced by 300 ␮moles/l was substantially inhibited by only 30 ␮moles/l ZnPP, which would be expected to reduce the H2S concentration to 270 ␮moles/l, which is still vasoactive (10), it is possible that at least a portion of the complex triphasic H2S response in trout EBA is mediated by sulfide-induced activation of the HO/CO pathway. Along these lines, Christodoulides et al. (5) demonstrated that while ZnPP did not bind to endogenous NO, it still inhibited soluble GC. Our results, however, do not suggest that ZnPP inhibits soluble GC, as ZnPP-treated vessels relaxed in a similar fashion as controls when exposed to the NO-donor SNP (Fig. 7C). With the absence of endothelial NO, the interaction of H2S and CO is perhaps even more intriguing in fish. Both H2S (10) and CO (Fig. 6A) appear to serve as endogenously produced gases involved in the modulation of vascular smooth muscle tone in trout. While the trout EBA response to H2S is complex (10), this study demonstrates that the activity of heme oxygenase is necessary for the predominant third dilatory phase of the H2S-mediated response (Fig. 7). Evolutionary conservation of CO as a gasotransmitter. It is interesting that all three “gasotransmitters”, NO, CO, and H2S, were originally thought to be merely toxic molecules to living organisms, and now, all three are recognized in mammals for their wide range of biological activities. The present study also suggests that important biological activities of all three gasoAJP-Regul Integr Comp Physiol • VOL

transmitters are relatively ancient and common traits in vertebrates. The dilatory response to exogenous CO is similar in conductance vessels from sea lamprey (Fig. 1), trout (Figs. 2– 4), and in multiple mammalian models (see Refs. 22 and 61). To date, CO is the only one of the three gasotransmitters that is both produced in the vasculature and, as these initial studies suggest, appears to have identical vasoactive properties across the vertebrate phylum. While NO dilates most vertebrate blood vessels, it constricts cyclostome vessels and a few vessels in elasmobranchs (14). Vessel responses to H2S have both species- and intravascular-related variability and range from monophasic relaxation, to monophasic vasoconstriction or complex multiphasic responses (11). Clearly, further examination of CO responses in other vertebrate vessels, and especially in small-resistance vessels in fish where tissue perfusion is regulated, is needed to confirm this observation. Environmental Impact of CO in Fish CO production by and regulation within fish vascular smooth muscle is of interest for several reasons. First, it has been reported that CO plays a more significant role in the regulation of mammalian vascular tone in the absence of endothelial NO (4, 33). Fish do not appear to possess endothelial nitric oxide synthase (30, 46, 50), with the possible exception being retinal tissue (25), and thus a greater role for CO might be anticipated. Second, environmental toxins and stress have been demonstrated to alter HO activity or induce HO-1 in both elasmobranch (17) and teleost (2, 7, 34, 43) fishes. Third, organic solvents, pesticides, and trace metals have the ability to alter fish HO function (2, 7, 48, 51) and are far more common environmental toxicants in water than in air (28, 48, 59). Thus, aquatic vertebrates subjected to such pollutants may have diminished or altered CO-mediated vasoregulation. Perspectives and Significance CO, like the other gasotransmitters NO and H2S, appears to be a phylogenetically ancient vascular signaling molecule. However, unlike NO and H2S, the effects of CO appear to be well conserved across vertebrate phylogeny, at least at this preliminary stage of investigation. Although it remains to be determined how CO production is regulated, as well as its overall impact in the piscine cardiovascular system, it is clear that the environmental and physiological diversity of fishes provides unique and ample opportunities to explore the forces that forged the exploitation of this regulatory mechanism. The 296 • JANUARY 2009 •

www.ajpregu.org

R148

CARBON MONOXIDE AND FISH VESSELS

physiological impact of exogenous gases also deserves consideration given the potential ability of these gases to readily cross the respiratory epithelium. Finally, these studies point out the need for a greater awareness of the environmental impact of an ever-increasing number of molecules with adverse affects on heme oxygenase and a demonstrated propensity for them to accumulate in aquatic environments. ACKNOWLEDGMENTS The authors are indebted to Dr. W. Swink, U.S. Geological Survey Biological Research Station for providing lamprey. This work was supported in part by NSF Grant IOS 0641436 (to K. R. Olson) and a SISTAR grant (Student Independent Study and Teaching Research) from Saint Mary’s College (R. A. Dombkowski). The authors also thank Prof. Brian Mann from the University of Sheffield (UK) for the synthesis CORM-3. REFERENCES 1. Ariyoshi T, Shiiba S, Hasegawa H, Arizono K. Effects of the environmental pollutants on heme oxygenase activity and cytochrome P-450 content in fish. Bull Environ Contam Toxicol 44: 189 –196, 1990. 2. Ariyoshi T, Shiiba S, Hasegawa H, Arizono K. Profile of metal-binding proteins and heme oxygenase in red carp treated with heavy metals, pesticides and surfactants. Bull Environ Contam Toxicol 44: 643– 649, 1990. 3. Artinian LR, Ding JM, Gillette MU. Carbon monoxide and nitric oxide: interacting messengers in muscarinic signaling to the brain’s circadian clock. Exp Neurol 171: 293–300, 2001. 4. Brian JE Jr, Heistad DD, Faraci FM. Effect of carbon monoxide on rabbit cerebral arteries. Stroke 25: 639 – 643, 1994. 5. Christodoulides N, Durante W, Kroll MH, Schafer AI. Vascular smooth muscle cell heme oxygenases generate guanylyl cyclase-stimulatory carbon monoxide. Circulation 91: 2306 –2309, 1995. 6. Clark JE, Naughton P, Shurey S, Green CJ, Johnson TR, Mann BE, Foresti R, Motterlini R. Cardioprotective actions by a water-soluble carbon monoxide-releasing molecule. Circ Res 93: e2– e8, 2003. 7. Dalwani R, Dave JM, Datta K. Alterations in hepatic heme metabolism in fish exposed to sublethal cadmium levels. Biochem Int 10: 33– 42, 1985. 8. Di GC, Grilli A, Ciocca I, Macri MA, Daniele F, Sabatino G, Cacchio M, De Lutiis MA, Da PR, Di NF, Felaco M. Carotid body NO-CO interaction and chronic hypoxia. Adv Exp Med Biol 475: 685– 690, 2000. 9. Doeller JE, Isbell TS, Benavides G, Koenitzer J, Patel H, Patel RP, Lancaster JR, Darley-Usmar VM, Kraus DW. Polarographic measurement of hydrogen sulfide production and consumption by mammalian tissues. Anal Biochem 341: 40 –51, 2005. 10. Dombkowski RA, Russell MJ, Olson KR. Hydrogen sulfide as an endogenous regulator of vascular smooth muscle tone in trout. Am J Physiol Regul Integr Comp Physiol 286: R678 –R685, 2004. 11. Dombkowski RA, Russell MJ, Schulman AA, Doellman MM, Olson KR. Vertebrate phylogeny of hydrogen sulfide vasoactivity. Am J Physiol Regul Integr Comp Physiol 288: R243–R252, 2005. 12. Donald JA, Broughton BR. Nitric oxide control of lower vertebrate blood vessels by vasomotor nerves. Comp Biochem Physiol A Mol Integr Physiol 142: 188 –197, 2005. 13. Donald JA, Broughton BR, Bennett MB. Vasodilator mechanisms in the dorsal aorta of the giant shovelnose ray, Rhinobatus typus (Rajiformes; Rhinobatidae). Comp Biochem Physiol A Mol Integr Physiol 137: 21–31, 2004. 14. Donald JA, Olson KR. Nervous control of circulation—the role of gasotransmitters, NO, CO, and H2S. Acta Histochem 313: 362–368, 2008. 15. Durante W, Schafer AI. Carbon monoxide and vascular cell function. Int J Mol Med 2: 255–262, 1998. 16. Duranton C, Mikulovic E, Tauc M, Avella M, Poujeol P. Potassium channels in primary cultures of seawater fish gill cells. II. Channel activation by hypotonic shock. Am J Physiol Regul Integr Comp Physiol 279: R1659 –R1670, 2000. 17. Dwivedi J, Trombetta LD. Acute toxicity and bioaccumulation of tributyltin in tissues of Urolophus jamaicensis (yellow stingray). J Toxicol Environ Health A 69: 1311–1323, 2006. 18. Evans DH, Gunderson MP. A prostaglandin, not NO, mediates endothelium-dependent dilation in ventral aorta of shark (Squalus acanthias). Am J Physiol Regul Integr Comp Physiol 274: R1050 –R1057, 1998. AJP-Regul Integr Comp Physiol • VOL

19. Evans DH, Harrie AC. Vasoactivity of the ventral aorta of the American eel (Anguilla rostrata), Atlantic hagfish (Myxine glutinosa), and sea lamprey (Petromyzon marinus). J Exp Zool 289: 273–284, 2001. 20. Foresti R, Hammad J, Clark JE, Johnson TR, Mann BE, Friebe A, Green CJ, Motterlini R. Vasoactive properties of CORM-3, a novel water-soluble carbon monoxide-releasing molecule. Br J Pharmacol 142: 453– 460, 2004. 21. Foresti R, Motterlini R. The heme oxygenase pathway and its interaction with nitric oxide in the control of cellular homeostasis. Free Radic Res 31: 459 – 475, 1999. 22. Foresti R, Shurey C, Ansari T, Sibbons P, Mann BE, Johnson TR, Green CJ, Motterlini R. Reviewing the use of carbon monoxide-releasing molecules (CO-RMs) in biology: implications in endotoxin-mediated vascular dysfunction. Cell Mol Biol (Noisy -le-grand) 51: 409 – 423, 2005. 23. Furchgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle cells by acetylcholine. Nature 288: 373–376, 1980. 24. Han Y, Qin J, Chang X, Yang Z, Du J. Hydrogen sulfide and carbon monoxide are in synergy with each other in the pathogenesis of recurrent febrile seizures. Cell Mol Neurobiol 26: 101–107, 2006. 25. Haverkamp S, Kolb H, Cuenca N. Endothelial nitric oxide synthase (eNOS) is localized to Muller cells in all vertebrate retinas. Vision Res 39: 2299 –2303, 1999. 26. Henne J, Jeserich G. Maturation of spiking activity in trout retinal ganglion cells coincides with upregulation of Kv3.1- and BK-related potassium channels. J Neurosci Res 75: 44 –54, 2004. 27. Henne J, Pottering S, Jeserich G. Voltage-gated potassium channels in retinal ganglion cells of trout: a combined biophysical, pharmacological, and single-cell RT-PCR approach. J Neurosci Res 62: 629 – 637, 2000. 28. Jaffe R, Gardinali PR, Cai Y, Sudburry A, Fernandez A, Hay BJ. Organic compounds and trace metals of anthropogenic origin in sediments from Montego Bay, Jamaica: assessment of sources and distribution pathways. Environ Pollut 123: 291–299, 2003. 29. Jaronczyk K, Bui L, Soong JM, McLaughlin BE, Marks GS, Brien JF, Nakatsu K. The source of heme for vascular heme oxygenase II: de novo heme biosynthesis in rat aorta. Can J Physiol Pharmacol 82: 218 –224, 2004. 30. Jennings BL, Broughton BR, Donald JA. Nitric oxide control of the dorsal aorta and the intestinal vein of the Australian short-finned eel Anguilla australis. J Exp Biol 207: 1295–1303, 2004. 31. Jin HF, Du JB, Li XH, Wang YF, Liang YF, Tang CS. Interaction between hydrogen sulfide/cystathionine gamma-lyase and carbon monoxide/heme oxygenase pathways in aortic smooth muscle cells. Acta Pharmacol Sin 27: 1561–1566, 2006. 32. Johnson FK, Johnson RA. Carbon monoxide promotes endotheliumdependent constriction of isolated gracilis muscle arterioles. Am J Physiol Regul Integr Comp Physiol 285: R536 –R541, 2003. 33. Johnson FK, Teran FJ, Prieto-Carrasquero M, Johnson RA. Vascular effects of a heme oxygenase inhibitor are enhanced in the absence of nitric oxide. Am J Hypertens 15: 1074 –1080, 2002. 34. Kaminsky L. The role of trace metals in cytochrome P4501 regulation. Drug Metab Rev 38: 227–234, 2006. 35. Kanu A, Leffler CW. Carbon monoxide and Ca2⫹-activated K⫹ channels in cerebral arteriolar responses to glutamate and hypoxia in newborn pigs. Am J Physiol Heart Circ Physiol 293: H3193–H3200, 2007. 36. Kharitonov VG, Sharma VS, Pilz RB, Magde D, Koesling D. Basis of guanylate cyclase activation by carbon monoxide. Proc Natl Acad Sci USA 92: 2568 –2571, 1995. 37. Lamb FS, Barna TJ. Chloride ion currents contribute functionally to norepinephrine-induced vascular contraction. Am J Physiol Heart Circ Physiol 275: H151–H160, 1998. 38. Lin H, McGrath JJ. Vasodilating effects of carbon monoxide. Drug Chem Toxicol 11: 371–385, 1988. 39. Maines MD. The heme oxygenase system: a regulator of second messenger gases. Annu Rev Pharmacol Toxicol 37: 517–554, 1997. 40. Matchkov VV, Aalkjaer C, Nilsson H. Distribution of cGMP-dependent and cGMP-independent Ca2⫹-activated Cl⫺ conductances in smooth muscle cells from different vascular beds and colon. Pflu¨gers Arch 451: 371–379, 2005. 41. Matchkov VV, Aalkjær C, Nilsson H. A cyclic GMP-dependent calciumactivated chloride current in smooth-muscle cells from rat mesenteric resistance arteries. J Gen Physiol 123: 121–134, 2004. 296 • JANUARY 2009 •

www.ajpregu.org

CARBON MONOXIDE AND FISH VESSELS 42. McCoubrey WK Jr, Huang TJ, Maines MD. Isolation and characterization of a cDNA from the rat brain that encodes hemoprotein heme oxygenase-3. Eur J Biochem 247: 725–732, 1997. 43. Mori T, Hiraka I, Kurata Y, Kawachi H, Mano N, Devlin RH, Nagoya H, Araki K. Changes in hepatic gene expression related to innate immunity, growth and iron metabolism in GH-transgenic amago salmon (Oncorhynchus masou) by cDNA subtraction and microarray analysis, and serum lysozyme activity. Gen Comp Endocrinol 151: 42–54, 2007. 44. Motterlini R. Carbon monoxide-releasing molecules (CO-RMs): vasodilatory, anti-ischaemic and anti-inflammatory activities. Biochem Soc Trans 35: 1142–1146, 2007. 45. Olesen SP, Drejer J, Axelsson O, Moldt P, Bang L, Nielsen-Kudsk JE, Busse R, Mulsch A. Characterization of NS 2028 as a specific inhibitor of soluble guanylyl cyclase. Br J Pharmacol 123: 299 –309, 1998. 46. Olson KR, Villa J. Evidence against non-prostanoid endothelial-derived relaxing factor(s) in trout vessels. Am J Physiol Regul Integr Comp Physiol 260: R925–R933, 1991. 47. Qingyou Z, Junbao D, Weijin Z, Hui Y, Chaoshu T, Chunyu Z. Impact of hydrogen sulfide on carbon monoxide/heme oxygenase pathway in the pathogenesis of hypoxic pulmonary hypertension. Biochem Biophys Res Commun 317: 30 –37, 2004. 48. Schlenk D, Sapozhnikova Y, Baquirian JP, Mason A. Predicting chemical contaminants in freshwater sediments through the use of historical biochemical endpoints in resident fish species. Environ Toxicol Chem 21: 2138 –2145, 2002. 49. Schultz BD, DeRoos AGD, Venglarik CJ, Singh AK, Frizzell RA, Bridges RJ. Glibenclamide blockade of CFTR chloride channels. Am J Physiol Lung Cell Mol Physiol 271: L192–L200, 1996. 50. Toda N, Ayajiki K. Phylogenesis of constitutively formed nitric oxide in non-mammals. Rev Physiol Biochem Pharmacol 157: 31– 80, 2006. 51. van den Hurk P. Bile fluorescence, heme oxygenase induction, and increased biliverdin excretion by mixtures of environmental toxicants. Aquat Toxicol 77: 202–209, 2006. 52. van den Hurk P. Heme oxygenase induction and biliverdin excretion: implications for the bile fluorescence biomarker. Mar Environ Res 62 Suppl: S16 –S19, 2006.

AJP-Regul Integr Comp Physiol • VOL

R149

53. Wang R. The gasotransmitter role of hydrogen sulfide. Antioxid Redox Signal 5: 493–501, 2003. 54. Wang R, Wang Z, Wu L. Carbon monoxide-induced vasorelaxation and the underlying mechanisms. Br J Pharmacol 121: 927–934, 1997. 55. Wang R, Wu L. Interaction of selective amino acid residues of Kca channels with carbon monoxide. Exp Biol Med (Maywood ) 228: 474 – 480, 2003. 56. Wang R, Wu L. The chemical modification of KCa channels by carbon monoxide in vascular smooth muscle cells. J Biol Chem 272: 8222– 8226, 1997. 57. Wang R, Wu L, Wang Z. The direct effect of carbon monoxide on KCa channels in vascular smooth muscle cells. Pflu¨gers Arch 434: 285–291, 1997. 58. Whitfield NL, Kreimier EL, Verdial FC, Skovgaard N, Olson KR. A reappraisal of H2S/sulfide concentration in vertebrate blood and its potential significance in ischemic preconditioning and vascular signaling. Am J Physiol Regul Integr Comp Physiol 294: R1930 –R1937, 2008. 59. Willey JD, Kieber RJ, Avery GB Jr. Changing chemical composition of precipitation in Wilmington, North Carolina, USA: implications for the continental USA. Environ Sci Technol 40: 5675–5680, 2006. 60. Wu L, Cao K, Lu Y, Wang R. Different mechanisms underlying the stimulation of KCa channels by nitric oxide and carbon monoxide. J Clin Invest 110: 691–700, 2002. 61. Wu L, Wang R. Carbon monoxide: endogenous production, physiological functions, and pharmacological applications. Pharmacol Rev 57: 585– 630, 2005. 62. Yamazaki J, Hume JR. Inhibitory effects of glibenclamide on cystic fibrosis transmembrane regulator, swelling-activated, and Ca2⫹-activated Cl⫺ channels in mammalian cardiac myocytes. Circ Res 81: 101–109, 1997. 63. Yang ZX, Qin J. Interaction between endogenous nitric oxide and carbon monoxide in the pathogenesis of recurrent febrile seizures. Biochem Biophys Res Commun 315: 349 –355, 2004. 64. Zhang QY, Du JB, Zhang CY, Tang CS. The regulation of carbon monoxide/heme oxygenase system by hydrogen sulfide in rats with hypoxic pulmonary hypertension. Zhonghua Jie He He Hu Xi Za Zhi 27: 659 – 663, 2004.

296 • JANUARY 2009 •

www.ajpregu.org