Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Despite its clinical use for 150 years, the mechanism by which glyceryl trinitrate (GTN) induces vasorelaxation is not completely understood. It has been hypothesized that GTN is a prodrug that requires endothelium-independent biotransformation to the free radical, nitric oxide (NO) (Ignarro et al., 1989). This free radical can activate soluble guanylyl cyclase to increase cGMP formation and promote blood vessel relaxation (Murad et al., 1978; Ignarro et al., 1989). As such, it appears that GTN acts as a pharmacological replacement for endothelium-derived relaxing factor, which is also thought to be NO (Palmer et al., 1987; Furchgott, 1988).

In support of this NO-prodrug hypothesis, concentrations of the denitrated metabolite, glyceryl dinitrate, and cGMP have been found to increase during incubation of GTN with vascular smooth muscle at time points concurrent with vasodilation (Brien et al., 1988). Furthermore, using a chemiluminescence-headspace gas method, NO has been measured from the incubation of vascular smooth muscle with GTN (Marks et al., 1992; 1995). However, NO was not measurable until relaxation was nearly completed. In another set of experiments, it was observed that incubation of vascular smooth muscle with GTN served as a source of a diffusible relaxing factor that could relax a nonvascular smooth muscle bioassay tissue (Hussain et al., 1994). The relaxing factor was sensitive to hemoglobin but not sensitive to superoxide (Hussain et al., 1994, 1996). Taken together, these results suggest that a conjugate of NO, such as an S-nitrosothiol, rather than NO itself, is responsible for GTN-induced vasodilation.

S-Nitrosothiols of the generic structure RSNO, which are formed by reaction of NO and sulfhydryl-containing compounds, possess many of the pharmacological properties of GTN, including vasorelaxation (Ignarro et al., 1981; Mathews and Kerr, 1993) and inhibition of platelet aggregation (Mellion et al., 1983). Furthermore, an S-nitrosothiol was previously proposed to be an active metabolite produced by vascular biotransformation of GTN (Ignarro et al., 1981). Given the unstable chemical nature of S-nitrosothiols, the detection of such compounds has been elusive until recently, when investigators quantified endogenous S-nitrosothiols in blood plasma and pulmonary fluids (Stamler et al., 1992a; Gaston et al., 1993). The measurement of S-nitrosothiols has been achieved by using long-wave UV light photolysis, resulting in homolytic cleavage of the molecule yielding thiyl radical and NO, the latter of which is measured by chemiluminescence (Stamler et al., 1992b; Welch et al., 1996). From such observations, it follows that if an S-nitrosothiol or S-nitrosothiols are produced during the biotransformation of GTN by vascular smooth muscle, the exposure to UV light should increase GTN-derived NO production.

In the present study, the first objective was to determine whether light-sensitive NO conjugate or conjugates are produced during GTN biotransformation by bovine pulmonary artery (BPA) using a UV photolysis-chemiluminescence method. The second objective was to examine the effect of UV photolysis of a putative NO-conjugate on GTN-induced relaxation of vascular smooth muscle.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Drugs and Solutions. Krebs' solution contained 120 mM NaCl, 5.6 mM KCl, 1.2 mM MgSO₄, 1.2 mM NaH₂PO₄, 2.5 mM CaCl₂, 25 mM NaHCO₃, and 10 mM dextrose; 30 µM EDTA also was added. This solution was bubbled with 95% O₂/5% CO₂ (medical grade; Praxair Canada, Inc., Toronto, Ontario, Canada). Stock solutions of GTN (2.2 × 10² M) were obtained from Omega (Montreal, Quebec, Canada) as Nitroject. S-Nitroso-N-acetylpenicillamine (SNAP) was obtained from Colour Your Enzyme (Bath, Ontario, Canada). Stock solutions of SNAP were prepared in Krebs' solution. Working solutions of the above drugs were prepared on the day of experimentation using Krebs' solution; the stock and working solutions were kept on ice. A stock solution of 3.0 M KCl (BDH, Inc., Toronto, Ontario, Canada) was prepared in deionized water. NO calibration gases (3.2-512 ppm in N₂) and 5% NO/95% N₂ were obtained from Scott Specialty Gases (Troy, MI). All other chemicals used were of at least reagent grade and were obtained from BDH, Inc.

Preparation of BPA Strips and Rings. Bovine lungs were obtained from a local abattoir immediately after slaughter and immersed in ice-cold Krebs' solution during transport. When used for measurement of NO formation, the secondary branches of the BPA were removed, cleaned of connective tissue and blood, and cut longitudinally. To obviate the effects of endothelium-derived relaxing factor, the endothelial cell layer was removed by gently scraping with a razor blade, and blood vessels were washed with ice-cold Krebs' solution. In experiments in which BPA was used for isometric tension studies, 5-mm-wide rings were cut from the secondary branch of the pulmonary artery, and the endothelium was removed by gently rubbing of the intimal surface with a stainless steel wire for 30 s. Contraction of BPA rings in response to incubation with 1 µM acetylcholine was used as an indicator of endothelium removal.

Measurement of NO Formation. To measure NO formation from nitrovasodilators, a modification of the chemiluminescence-headspace gas method described by Brien et al. (1991, 1996) was used. To a micro-Fernbach flask (total volume, 7.5 ml) we added 3 ml of Krebs' solution, 100 U/ml superoxide dismutase, and a micro-stir bar to mix the sample continuously. A BPA strip that had been equilibrated for 1 h in Krebs' solution, bubbled with 95% O₂/5% CO₂ at 37°C, was added to samples requiring tissue. The flask was sealed with a rubber-sleeve septum (Aldrich Chemical Co., Inc., Milwaukee, WI), equilibrated at 37°C for 5 min, and then gassed with a stream of 20% O₂/5% CO₂/balance N₂ for 5 min at 37°C. With the use of a gas-tight syringe, SNAP (0.1 or 10 µM) or GTN (100 µM) was injected through the rubber-sleeve septum into flasks with or without BPA, and the samples were incubated at 37°C for 10 min. A 400-µl aliquot of headspace gas was taken from each sample at 0.5, 2, 5, and 10 min and analyzed for NO using a model 270 B Nitric Oxide Analyzer (Sievers Research, Inc., Boulder, CO).

Relaxation Response of BPA to SNAP and GTN. BPA rings were mounted in water-jacketed quartz-glass tissue baths containing 10 ml of Krebs' solution at 37°C and bubbled with 95% O₂/5% CO₂. Each BPA ring was connected to a model FT03D force displacement transducer (Grass Instruments, Quincy, MA) such that isometric changes in tension could be recorded using a Grass model 7 polygraph. BPA rings were subjected to 2.0g resting tension and equilibrated for 1 h during which the Krebs' solution in the tissue bath was changed every 15 min. After equilibration, the BPA rings were contracted maximally with 100 mM K⁺. After washout at 15-min intervals for 1 h, 20 mM K⁺ was added to obtain a submaximal contraction (60-80% of the maximal tone). After the submaximal contraction had stabilized, cumulative concentration-response curves were determined for SNAP (0.1 nM to 10 µM) or GTN (0.1 nM to 100 µM) incubated with BPA rings in the absence or presence of UV light.

Data Analysis. The data are presented as group ± S.D. mean values. NO formation data were statistically analyzed using a randomized-design, two-way ANOVA for incubation time and UV light exposure. For a significant F statistic (p < .05) for UV light exposure, a Student's t test for unpaired data (two-tailed) was conducted. BPA ring relaxation data, including individual nitrovasodilator concentrations and EC₅₀ values, were analyzed by Student's t test for unpaired data (two-tailed). Two groups of data were considered to be statistically different when p < .05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of UV Light on NO Formation From SNAP. For 10 µM SNAP in Krebs' solution, there was a time-dependent formation of NO that was increased (p < .05) when exposed to UV light (Fig. 1). The UV light exposure increased NO formation by 4.9-, 4.9-, 5.9-, and 8.2-fold, after 0.5, 2, 5, and 10 min of incubation, respectively. Similarly, when 0.1 µM SNAP in Krebs' solution was exposed to UV light, there were increases (p < .05) in NO formation at 0.5, 2, 5, and 10 min of 8.8-, 9.5-, 11.2-, and 12.9-fold, respectively (Fig. 1).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Time-dependent formation of NO from the incubation of 10 µM SNAP and 0.1 µM SNAP in Krebs' solution, in the presence () and absence () of UV light. The data are presented as group ± S.D. mean values of five experiments. *p < .05 compared with sample not exposed to UV light.

Effect of UV Light on NO Formation From Interaction of GTN With L-Cysteine. Incubation of 100 µM GTN with 1 mM L-cysteine for 10 min resulted in time-dependent formation of NO (Fig. 2). When the GTN-L-cysteine samples were exposed to UV light, there was an increase (p < .05) in NO formation at all experimental time points. The increase in NO formation was 2.0-, 1.8-, 1.6-, and 1.9-fold at 0.5, 2, 5, and 10 min of incubation, respectively.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Time-dependent formation of NO from the incubation of 100 µM GTN with 1 mM L-cysteine in the presence () and absence () of UV light. Data are presented as group ± S.D. mean values of four experiments. *p < .05 compared with sample not exposed to UV light.

Effect of UV Light on NO Formation During GTN Incubation With BPA. Incubation of 100 µM GTN with BPA strips resulted in no detectable NO after 0.5 and 2 min (Fig. 3). However, in the presence of UV light, NO formation was measured after 0.5 and 2 min in the amount of 159 ± 45 and 171 ± 37 pmol NO/g tissue (n = 4), respectively. Longer incubation times of 5 and 10 min in the presence of UV light resulted in approximately 2.7- and 6.2-fold increases (p < .05) in NO formation, respectively, compared with no UV light exposure. The exposure of 100 µM GTN to UV light did not result in measurable formation of NO, thereby eliminating GTN itself as the source of UV light-derived NO in the GTN-BPA incubation system. This finding is consistent with previous observations that GTN per se is not photoactivated to form NO (Matsunaga and Furchgott, 1991; Venturini et al., 1993).

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Time-dependent formation of NO from the incubation of 100 µM GTN with BPA in the presence () and absence () of UV light. Data are presented as group ± S.D. mean values of four experiments. *p < .05 compared with sample not exposed to UV light. ND, no detectable NO formation at 0.5 and 2 min in the absence of UV light.

Effect of UV Light on Nitrovasodilator-Induced BPA Relaxation. The exposure to UV light resulted in photorelaxation of some BPA rings producing a 10% to 30% decrease in submaximal tone that plateaued after 5 min. After the BPA rings reached a stable tension, concentration-response curves were determined for SNAP or GTN. BPA rings also were incubated with these nitrovasodilator drugs in the absence of UV light. When BPA rings were incubated with SNAP in the presence of UV light, SNAP-induced relaxation was attenuated (p < .05) at nearly all concentrations of SNAP that were studied (Fig. 4). Furthermore, there was an increase (p < .05) in the EC₅₀ value of the SNAP concentration-response curve in the presence of UV light compared with the absence of UV light (2.7 ± 3.0 versus 0.24 ± 0.28 µM, respectively; n = 5). It was observed that not only was relaxation attenuated by UV light but also the ability of BPA rings to maintain SNAP-induced relaxation was impaired (Fig. 5). Incubation of BPA rings with GTN in the presence of UV light resulted in attenuation (p < .05) of GTN-induced relaxation at nearly all of the concentrations of GTN (Fig. 6). As well, the EC₅₀ value of the GTN concentration-response curve in the presence of UV light was increased (p < .05) compared with the absence of UV light (49 ± 23 versus 10 ± 6 nM, respectively; n = 5).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Attenuation of SNAP-induced relaxation of BPA rings by UV light. BPA rings contracted with 20 mM K+ were relaxed with cumulative doses of SNAP in the presence () or absence () of UV light. Data are presented as group ± S.D. mean values of five experiments. *p < .05 compared with sample not exposed to UV light.

View larger version (8K): [in this window] [in a new window]   Fig. 5.   Representative polygraph tracings showing the response of BPA rings contracted with 20 mM K+ and relaxed with cumulative doses of SNAP. A, the molar concentration of SNAP is expressed as the negative logarithm. B, in the presence of UV light, the ability of BPA rings to maintain SNAP-induced relaxation was impaired.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Attenuation of GTN-induced relaxation of BPA rings by UV light. BPA rings contracted with 20 mM K+ were relaxed with cumulative doses of GTN in the presence () or absence () of UV light. Data are presented as group ± S.D. mean values of five experiments. *p < .05 compared with sample not exposed to UV light.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The purpose of the present study was to test the hypothesis that vascular smooth muscle, such as BPA, biotransforms organic nitrates to a vasoactive S-nitrosothiol. The measurement of NO subsequent to UV photolysis of the S-NO bond has been used to detect S-nitrosothiols in blood plasma (Stamler et al., 1992a), and we used a modification of this method in our studies. SNAP was used as a model S-nitrosothiol to test a modification of the method that was suitable for our requirements. The molar yield of NO from 10 µM SNAP after a 0.5-min incubation in Krebs' solution increased from 0.48% to 2.3% in the presence of UV light (Fig. 1). In contrast, the molar yield of NO from 0.1 µM SNAP increased from 6.6% to 59% in the presence of UV light (Fig. 1). Because many earlier studies involved the use of millimolar rather than micromolar concentrations of SNAP, the extent of formation of NO may have been underestimated with respect to lower, physiologically relevant concentrations of S-nitrosothiols. Thus, based on the long-wavelength, UV-light-stimulated formation of NO from SNAP, it appears that the UV photolysis-chemiluminescence method is an appropriate technique for detecting putative S-nitrosothiols produced by GTN biotransformation during incubation with vascular smooth muscle.

Feelisch and Noack (1987) proposed that GTN elicits its vasodilator effects through a nonenzymatic interaction with endogenous thiol-containing compounds, resulting in the formation of a vasorelaxant S-nitrosothiol. Our results showed a time-dependent increase in NO formation from the interaction of 100 µM GTN with 1 mM L-cysteine (Fig. 2). The interaction of GTN and L-cysteine is postulated to result in the formation of S-nitroso-L-cysteine, but conclusive evidence for formation of this S-nitrosothiol has not been provided. Because S-nitrosothiols undergo UV photolysis to form NO and because UV light increases NO formation from the interaction of GTN and L-cysteine, the results are therefore consistent with formation of an S-nitrosothiol during this reaction.

In the absence of UV light, chemiluminescence-headspace-gas measurement of NO formation from the incubation of 100 µM GTN with BPA was possible only after 5 min (Fig. 3). This result is consistent with previous results obtained in our laboratory (Marks et al., 1992, 1995). In a previous study, the inability to measure NO from GTN on interaction with vascular smooth muscle at time points concurrent with vasorelaxation (less than 5 min) led to the proposal that a conjugate of NO such as an S-nitrosothiol, rather than NO per se, may be the vasoactive metabolite derived from GTN (Marks et al., 1995). In the present study, when GTN was incubated with BPA in the presence of UV light, NO was measured as early as 0.5 min (Fig. 3). Furthermore, at 5 and 10 min of incubation, NO formation was greater in the presence of UV light compared with no UV light exposure. Because the wavelength output of the UV source (320-380 nm) includes the range of 330 to 350 nm, which is capable of S-nitrosothiol homolysis (Butler and Rhodes, 1997), it appears therefore that UV light is forming NO from the photolysis of GTN-derived S-nitrosothiol or S-nitrosothiols in the BPA incubates. Furthermore, the fact that NO formation was measurable at 0.5 min, an experimental time point that coincides with the onset of GTN-induced vasodilation of BPA (Kawamoto et al., 1990), supports the concept that an S-nitrosothiol is the vasoactive metabolite produced during the biotransformation of GTN by BPA.

GTN biotransformation in vascular tissue results in the production of intracellular nitrite anion (NO₂) in addition to NO (Bennett and Marks, 1984); therefore, it is possible that enhanced NO formation from the incubation of GTN with BPA in the presence of UV light may have resulted from photolytic decomposition of NO₂ to NO (Matsunaga and Furchgott, 1989, 1991). Several factors make this interpretation unlikely. It has been estimated that vascular smooth muscle contains micromolar concentration of endogenous NO₂ (Bennett and Marks, 1984), and in the present study, BPA irradiated with UV light in the absence of GTN resulted in an apparent NO chemiluminescence signal that was not different from the background signal produced by Krebs' solution exposed to UV light (data not shown). Also, it is likely that any NO derived from intracellular NO₂, either from endogenous stores or from GTN biotransformation, is rapidly degraded in situ by superoxide radical anion, which is generated simultaneously during the photolysis of NO₂ (Matsunaga and Furchgott, 1989).

An assumption underlying the term "nitrovasodilator" is that compounds such as SNAP and GTN function through the formation of NO. Based on this assumption and the observation that UV light photolyzed SNAP to form NO, it was anticipated that relaxation of BPA by SNAP would be potentiated by UV light. Instead, it was found that SNAP-induced relaxation of BPA was attenuated in the presence of UV light (Fig. 4), with an 11-fold increase in the EC₅₀ value for SNAP-induced relaxation. A possible explanation for this observation is that SNAP acts as a carrier of NO to the intracellular target for NO. As the nitrosyl moiety of a thionitrite compound, rather than as free NO, there would be less vulnerability of the vasoactive moiety to the inactivating effects of compounds such as oxygen and superoxide radical anion (Robak et al., 1992). If this were the case, then the formation of NO from SNAP by UV light, before SNAP reached the intracellular target, would result in diminished vasorelaxation due to increased exposure of NO to inactivating compounds. This interpretation is supported by the observation in the present study that SNAP-induced relaxation of BPA rings in the absence of UV light reached a stable tension after the addition of each concentration of SNAP, whereas in the presence of UV light, SNAP-induced relaxation of BPA rings could not be sustained (Fig. 5). The inability of the UV-irradiated BPA rings to maintain SNAP-induced relaxation is consistent with the idea that NO has been formed from SNAP in that the transient nature of the relaxation is characteristic of that for NO-induced relaxation of vascular smooth muscle in tissue bath studies (Furchgott et al., 1992; Hussain et al., 1997). These results are consistent with previous studies in which it was found that the formation of NO from S-nitrosothiols did not correlate with S-nitrosothiol-induced relaxation of vascular smooth muscle (Kowaluk and Fung, 1990; Mathews and Kerr, 1993). It is possible that the NO moiety of S-nitrosothiols never exists as free NO but instead is transferred to a macromolecule causing S-nitrosylation and activation of the enzymic activity of the macromolecule. Macromolecule S-nitrosylations have been studied extensively by investigators who have demonstrated that this reaction occurs in vitro and in vivo (Lipton et al., 1993; Scharfstein et al., 1994).

Incubation of GTN with BPA rings in the presence of UV light resulted in the attenuation of GTN-induced relaxation (Fig. 6) in a similar manner to that observed for SNAP. There was a 5-fold increase in the EC₅₀ value for GTN-induced relaxation of BPA rings exposed to UV light. Based on the observations for SNAP-induced relaxation of BPA rings in the presence of UV light, it is proposed that attenuation of GTN-induced vasodilation may be the result of homolytic cleavage of an S-nitrosothiol produced during GTN biotransformation by BPA. As discussed previously, this S-nitrosothiol may serve as a carrier of NO to its intracellular target. UV light was less effective in attenuating GTN-induced relaxation of BPA rings compared with SNAP. A possible explanation is that the biotransformation of GTN to a putative S-nitrosothiol would occur intracellularly and be less susceptible to photolysis by UV light than SNAP, an S-nitrosothiol that was administered extracellularly to the BPA.

In summary, when GTN was incubated with BPA in the presence of UV light, there was measurable NO formation at a time that coincided with the onset of vasorelaxation. However, despite enhanced NO formation in the presence of UV light, for GTN incubated with BPA and for SNAP incubated in Krebs' solution, the vasodilator response to these nitrovasodilators was attenuated. In view of the chemical reactivity of NO and the potential for S-nitrosothiols to be transported and converted to NO at a target site of action, our results are more consistent with the idea that a stabilized form of NO such as an S-nitrosothiol, rather than NO itself, is the vasoactive metabolite formed when GTN interacts with vascular smooth muscle.