Thiol agents potentiate glyceryl trinitrate

mediated relaxation of rabbit taenia coli:

evidence for thiol-dependent biotransformation

Aman S. Hussain, Tawfic N. Abu-Zahra, James F. Brien, Gerald S. Marks, andKanji Nakatsu

Abstract: In the present study, the role of thiols on glyceryl trinitrate (GTN) induced relaxation of rabbit taenia coli strips(RTCS) was investigated. This study was designed to test the hypothesis that a deficiency in thiols is responsible for RTCSinsensitivity to GTN, and thus thiols play a key role in the enzymatic activation of GTN. Isolated RTCS bathed innormothermic, oxygenated Krebs solution were pretreated with the thiols L-cysteine (5 mM) and N-acetyl-L-cysteine (NAC,5 mM) for 30 min and washed. The effects of GTN were determined by changes in isometric tension of K+-precontractedRTCS. Both L-cysteine and NAC resulted in increased relaxations to GTN (0.1 nM – 10 µM) as the GTN relaxation EC50

decreased compared with that of the untreated RTCS (L-cysteine, 0.06 ± 0.12 µM and NAC, 0.08 ± 0.03 µM versus control0.25 ± 0.08 µM, n = 5, p < 0.05). In contrast, 5 mM D-cysteine had no significant effects on the RTCS GTN relaxation EC50

(0.16 ± 0.13 µM, n = 5). Similarly, the thiol donor L-methionine significantly increased RTCS sensitivity to GTN, as therelaxation EC50 decreased from the control value of 0.25 ± 0.08 µM to 10 ± 4 nM (n = 5, p < 0.001), whereas the D-isomer didnot. These results are consistent with the idea that thiols play a key stereospecific role in the metabolic activation of GTN inRTCS. However, RTCS treated with amino acids were still less sensitive to GTN compared with vascular tissue, and thissuggests that RTCS may be deficient in some other enzyme(s) relative to vascular tissue that is (are) responsible for theactivation of GTN.

Key words: glyceryl trinitrate, biotransformation, sulfhydryl, cysteine, methionine.

Résumé: Dans la présente étude, nous avons examiné le rôle des thiols dans la relaxation par le trinitrate de glycéryle (GTN)des bandelettes longitudinales du côlon (BLC) de lapins. Cette étude a eu pour but de vérifier l’hypothèse qu’une déficienceen thiols est responsable de l’insensibilité des BLC au GTN et que, par conséquent, les thiols jouent un rôle clé dansl’activation enzymatique du GTN. Des BLC isolées, baignant dans une solution Krebs oxygénée et normothermique, ont étéprétraitées avec les thiols, L-cystéine (5 mM) ou N-acétyl-L-cystéine (NAC, 5 mM), pendant 30 min, puis lavées. Les effets duGTN ont été déterminés d’après les variations de tension isométrique des BLC précontractées par du K+. Tant la L-cystéineque la NAC ont augmenté les relaxations induites par GTN (0,1 nM – 10 µM), alors que l’EC50 de la relaxation par GTN adiminué comparativement à celle des BLC non traitées (L-cystéine, 0,06 ± 0,12 µM et NAC, 0,08 ± 0,03 µM par rapport à lavaleur témoin 0,25 ± 0,08 µM, n = 5, p < 0,05). À l’opposé, une concentration de 5 mM de D-cystéine n’a pas eu d’effetsignificatif sur l’EC50 de la relaxation par GTN (0,16 ± 0,13 µM, n = 5). Similairement, le donneur de thiol, L-méthionine, aaugmenté significativement la sensibilité des BLC au GTN, alors que l’EC50 de la relaxation induite par GTN a diminué lavaleur témoin 0,25 ± 0,08 µM à 10 ± 4 nM par rapport à la valeur témoin (n = 5, p < 0,001), ce qui n’a pas été observé avec leD-isomère. Ces résultats sont en accord avec l’hypothèse que les thiols jouent un rôle stéréospécifique clé dans l’activationmétabolique du GTN dans les BLC. Toutefois, les BLC traitées par des acides aminés étaient encore moins sensibles au GTNque le tissu vasculaire, ce qui suggère que les BLC pourraient être déficientes en une enzyme ou plusieurs enzymesresponsable(s) de l’activation du GTN.

Mots clés : trinitrate de glycéryle, biotransformation, sulfhydryle, cystéine, méthionine.

Introduction

It is widely accepted that the organic nitrate ester glyceryltrinitrate (GTN) acts to dilate blood vessels via enzymatic

biotransformation to nitric oxide (NO) or a related moleculethat is capable of activating guanylyl cyclase to increase intra-cellular cyclic guanosine monophosphate (cGMP) levels toeffect relaxation of vascular smooth muscle (Ignarro et al.1981). However, research has failed to definitively identify theenzyme(s) responsible for GTN biotransformation to its va-soactive drug. Candidate enzyme systems include the cyto-chrome P450 – P450 reductase system (Bennett et al. 1994),glutathione S-transferases (Tshushida et al. 1990; Hill et al.1992), and more recently, an unidentified membrane-associated enzyme (Seth and Fung 1993). Identification of theenzyme(s) responsible for biotransformation of GTN may provide

Received August 23, 1996.

A.S. Hussain, T.N. Abu-Zahra, J.F. Brien, G.S. Marks, andK. Nakatsu.1 Department of Pharmacology and Toxicology,Faculty of Medicine, Queen’s University, Kingston, ON K7L3N6, Canada.

1 Author for correspondence.

Can. J. Physiol. Pharmacol. 75: 807–811 (1997)

807

© 1997 NRC Canada

valuable insights into reversing clinical tolerance to thehemodynamic effects of GTN, which has been linked to de-creased biotransformation (Brien et al. 1988).

One aspect of GTN biotransformation that has garneredconsiderable attention is the involvement of sulfhydryl- orthiol-containing agents. Since biotransformation of GTN tonitric oxide requires a three-electron reduction, many havesuggested that thiols may play a key role in the biotransforma-tion of GTN (Harrison and Bates 1993). Manipulation of thiolsby oxidizing and alkylating agents has been shown to decreasethe effect of GTN in vitro and in vivo (Needleman andJohnson 1973; Greutter and Lemke 1985; Selke et al. 1991;Boesgaard et al. 1993; Wheatley et al. 1994) as well as to de-crease NO production from GTN (Bauer and Fung 1996).Some, but not all, studies have observed increased effects ofGTN following administration of thiol agents like L-cysteineor N-acetylcysteine (Horowitz et al. 1983; Winniford et al.1986; Selke et al. 1991; Wheatley et al. 1994). It has beenfound that the L-cysteine congener, N-acetyl-L-cysteine, canprevent in vivo tolerance, whereas the D-isomer could not(Newman et al. 1990). More recently, in support for the involve-ment of thiols in the enzymatic biotransformation of GTN, Kurzet al. (1991) observed that in vitro, small coronary arteries (less than100 µm in diameter), which are normally insensitive to the relaxanteffects of GTN, could be responsive to GTN in the presence ofL-cysteine but not the D-isomer. Since this stereospecific po-tentiation of GTN-induced relaxation by L-cysteine could beattenuated by inhibition of glutathione synthesis, it is likelythat these thiol agents increased relaxation to GTN by increas-ing glutathione production (Wheatley et al. 1992). In addition,it has been well established that increases in cellular cysteinecontent result in an increase in cellular glutathione (see review,Boesgaard 1995).

We have identified rabbit taenia coli strip (RTCS) to beanother smooth muscle type to be insensitive to the relaxanteffects of GTN in vitro; however, it is sensitive to biotransfor-mation products of GTN, namely NO and other liberators ofNO (Hussain et al. 1994, 1996). NO production from GTN inRTCS was also found to be lower relative to GTN-sensitivevascular tissue (Hussain et al. 1994); therefore, it is highlyprobable that lack of relaxation to GTN stems from decreasedbiotransformation rather than decreased levels or sensitivity ofguanylyl cyclase. To test the hypothesis that thiols play a keyrole in the activation of GTN, we have tested the effects of avariety of thiol agents on GTN-mediated relaxation of RTCS.In this report we present data indicating that thiols play a keyrole in the metabolic activation of GTN.

Methods and materials

Drugs and solutionsKrebs solution contained the following (mM): NaCl, 120; KCl, 5.6;MgSO4, 1.2; NaH2PO4, 1.2; CaCl2, 2.5; NaHCO3, 25; and dextrose,10; ethylenediaminetetraacetic acid was also added (30 µM). Theabove solution was aerated with 95% O2 – 5% CO2 (medical grade;Union Carbide Canada Ltd., Linde Division, Toronto, Ont.). Stocksolutions of GTN (22 mM) were obtained from DuPont (Toronto,Ont.) as Tridil®. L- and D-methionine, L- and D-cysteine, and N-acetyl-cysteine were obtained from Sigma Chemical Co. (St. Louis, Mo.).S-Nitroso-N-acetylpenicillamine (SNAP) was obtained from ColourYour Enzyme (Bath, Ont.).

Tissue strip preparationAnimals used in this experiment were cared for in accordance withthe principles and guidelines of the Canadian Council on AnimalCare, and the experimental protocol involving animals was approvedby the Queen’s University Animal Care Committee. Male NewZealand White rabbits (2.5–4.0 kg) were sacrificed by sodium pento-barbitone overdose (250 mg⋅kg-1) via rapid ear injection. Taenia coli,taken from the proximal portion of the large intestine, was cut intosmaller strips, 5 mm wide by 30 mm long, and intestinal contentswere washed away with ice-cold Krebs solution. The strips were sus-pended by use of polyester sutures in individual tissue baths contain-ing 10 mL Krebs solution at 37°C, through which 95% O2 – 5% CO2was continuously bubbled.

RTCS GTN concentration–response protocolRTCS were maintained at a preload tension of 2 g and equilibrated for1 h, during which time the Krebs solution was replaced every 15 min.Maximal contractions were elicited by a depolarizing stimulus of100 mM K+. After attaining steady maximal contractions, the stripswere washed and allowed to equilibrate for a further 60 min. Thiolagents or vehicle (Krebs solution) were administered and allowed toincubate for 30 min, and tissues were washed to remove any aminoacids not taken up by the tissue. To determine the relaxant effects ofGTN, the tissues were first contracted submaximally (50–60% ofmaximum) with 30 mM K+. Concentration–response curves werethen performed with GTN (0.1 nM – 10 µM). The isometric tensionof the RTCS was recorded by force-displacement transducers(Grass FT03D) coupled to a computer-based data acquisition system(MacLab®).

Data analysisConcentration–response curves for RTCS are presented with eachpoint representing the mean ± SEM. Analysis of the RTCS relaxationdata was conducted by one-way analysis of variance (ANOVA) fol-lowed by post hoc Newman–Keuls test; values were considered to bestatistically different at p < 0.05. Homogeneity of variance was con-firmed by Bartlett’s test prior to conducting parametric statisticalanalysis with ANOVA.

Results and discussion

Previous studies in our laboratory have found that rabbit taeniacoli strips (RTCS) are insensitive to the relaxant effects ofGTN yet sensitive to its biotransformation products, namelyNO (Hussain et al. 1994). It has been found that RTCS poorlybiotransform GTN to NO. One explanation for such insensitiv-ity could be that increased oxidative metabolism in gastroin-testinal smooth muscle may accelerate oxidation of thiols. It ishypothesized that such thiols are key cofactors in the biotrans-formation of GTN, and a deficiency of reduced thiols mayaccount for the decreased ability of RTCS to biotransformGTN, relative to sensitive vascular tissue. The hypothesis pre-dicts that supplementation of RTCS with reduced thiols orthiol donors should reverse pseudotolerance to the relaxanteffects of GTN. Incubation of RTCS with 5 mM L-cysteineresulted in a significant leftward shift in the GTNconcentration–response profile (Fig. 1), as the EC50 of relaxa-tion decreased from 0.25 ± 0.08 to 0.06 ± 0.12 µM. In addition,L-cysteine increased maximum RTCS relaxation mediated byGTN from 34.8 ± 1.1 to 56.5 ± 2.2%. NAC (5 mM) also in-creased the sensitivity of RTCS to GTN as the EC50 decreasedto 0.08 ± 0.03 µM, and maximal relaxation to GTN was alsoincreased compared with the control (83.6 ± 4.3 versus34.8 ± 1.1%, respectively). The results also suggest that NAC

Can. J. Physiol. Pharmacol. Vol. 75, 1997808

© 1997 NRC Canada

is more effective than L-cysteine in potentiating the effects ofGTN, and this is most likely due to the fact that NAC increasesintracellular thiol levels more effectively than L-cysteine as itis more cell permeable (Greutter and Lemke 1985). Such ob-servations were taken to mean that L-cysteine is involved in theenzymatic activation of GTN; however, it is also possible thatL-cysteine interacted with GTN nonenzymatically to producea nitrosothiol or NO that could relax smooth muscle (Feelischand Noack 1987).

To determine whether increased RTCS relaxation was dueto nonenzymatic interactions between GTN and L-cysteine, theeffects of D-cysteine were also investigated. Inclusion of5 mM D-cysteine in the RTCS bathing medium resulted in nosignificant effect on EC50 for relaxation compared with con-trols (0.25 ± 0.08 versus 0.16 ± 0.13 µM) but did effect a sig-nificant decrease in maximal relaxation compared with theuntreated control (24.8 ± 1.7 versus 34.8 ± 1.1%, respec-tively). If increased RTCS relaxation was due to direct inter-actions of L-cysteine and GTN, D-cysteine pretreatment ofRTCS should also increase its sensitivity to GTN. Thestereospecific requirement for L-cysteine to potentiate RTCSrelaxation by GTN indicates that the involvement may be en-zymatic in nature rather than nonspecific. These data are con-sistent with the work of others who found small porcinecoronary vessels, which are normally insensitive to GTN,could be made sensitive by pretreating vascular tissue withL-cysteine, but not with its D-isomer (Selke et al. 1990, 1991;Kurz et al. 1991).

Another possible explanation is that both enantiomers

formed nitrosothiols through nonenzymatic interactions withGTN. The uptake of these nitrosothiols (S-nitroso-L-cysteineand S-nitroso-D-cysteine) may have been stereospecific, result-ing in only the L-nitrosothiol being taken up into the cell, thusexplaining the different effect of the enantiomers. However,Harrison and Bates (1993) found that in small coronary ar-tery, both S-nitroso-L-cysteine and S-nitroso-D-cysteine wereequally potent at relaxing the tissue. Given that the majorityof cysteine is stored intracellularly as the tripeptide glu-tathione, it may be that L-cysteine increases GTN relaxationby increasing cellular glutathione, which could be critical forGTN biotransformation. This proposal is consistent with stud-ies that have demonstrated that the effects of L-cysteine andN-acetylcysteine in coronary vessels in vitro can be inhibitedby preventing production of glutathione (Wheatley et al.1994). Further support for the involvement of glutathione wasreported by Boesgaard et al. (1993), who observed decreasedeffectiveness of GTN in anesthetized rats after administrationof buthionine sulfoximine, an inhibitor of glutathione synthe-sis. Preliminary results in our laboratory suggest that glu-tathione is key to GTN metabolic activation in the RTCS asL-cysteine potentiation of GTN relaxation can be inhibited bybuthionine sulfoximine, an inhibitor of glutathione synthesis(H. Lei and K. Nakatsu, unpublished observations).

An alternative explanation for these observations may bethat NAC or L-cysteine may be acting as antioxidants. Giventhat the experimental setting (95% O2 – 5% CO2) may bea source of reactive oxygen species, such as hydroxyl free radi-cal and superoxide, which can scavenge NO, it is possible that

Fig. 1. The effects of thiol donors, L-cysteine (5 mM) and NAC(5 mM), on relaxation of RTCS by GTN (0.1 nM – 10 µM). Levelsof relaxation are expressed as the mean ± SEM of fivedeterminations. Both L-cysteine and NAC (L-isomer) pretreatmentsfor 30 min significantly increased RTCS relaxations compared withuntreated control, whereas D-cysteine pretreatment had nosignificant effects, except at the highest concentration of GTNtested (10 µM). Significant differences between relaxations weredetermined by one-way ANOVA followed by a Newman–Keulspost hoc test. *p < 0.05.

Fig. 2. The effects of methionine (5 mM) on relaxation of RTCS byGTN (0.1 nM – 10 µM). Relaxations are expressed as themean ± SEM of five determinations. L-Methionine pretreatment for30 min resulted in a significant increase in RTCS relaxation toGTN compared with control tissues; however, pretreatment ofRTCS with the D-isomer of methionine resulted in no significantchanges in relaxation, except at the highest concentration of GTNtested (10 µM). Significant differences between relaxations weredetermined by one-way ANOVA followed by a Newman–Keulspost-hoc test. *p < 0.05.

Hussain et al. 809

© 1997 NRC Canada

thiol donors may reduce these oxygen species and thus in-crease the potency of GTN. This is unlikely in light of earlierwork in which we observed that superoxide generation in themuscle bath did not affect the relaxant effects of GTN in RTCS(Hussain et al. 1996). In addition, while RTCS is insensitiveto GTN, it is extremely sensitive to spontaneous NO donorssuch as SIN-1, SNAP, and SNP, as well as NO gas itself(Hussain et al. 1994). Thus, the GTN RTCS pseudotolerancephenomenon could not be explained simply by the presence ofhigh oxygen radical species. If thiol donors used in this studywere acting as antioxidants and reducing oxygen radical spe-cies, one would expect that D-cysteine would also potentiateRTCS relaxation to GTN. This was not the case, as D-cysteineactually caused an attenuation of the maximal RTCS dilationto GTN.

The effects of L-cysteine pretreatment were also determinedfor another nitrovasodilator, SNAP, which does not requiremetabolic conversion. The hypothesis predicts that if the ef-fects of L-cysteine are specific for biotransformation of GTN,L-cysteine pretreatment would not influence RTCS relaxationto SNAP. In three determinations (Fig. 2) SNAP (0.1 nM –10 µM) mediated relaxation of RTCS was not significantlyaffected by pretreatment with 5 mM L-cysteine.

The effects of the amino acid methionine on RTCS relaxa-tion by GTN were also studied (Fig. 3). Unlike cysteine,methionine does not possess a free thiol group (and thus wouldobviate nonenzymatic interactions with GTN) and requires in-tracellular processing to form L-cysteine, via the transsulfura-tion pathway (Mudd et al. 1965). Addition of 5 mML-methionine resulted in a significant increase in RTCS relaxa-tions to GTN, as evidenced by a lowering of the relaxationEC50 compared with the untreated control (10 ± 4 nM versus0.25 ± 0.08 µM, respectively) and the increased level of

relaxations at the highest concentration of GTN tested(79.9 ± 2.7 versus 34.8 ± 1.1%, respectively). To determinewhether the potentiating effects of L-methionine were due toincreased intracellular L-cysteine, which could nonenzymati-cally interact with GTN to produce nitrosothiol or NO, theeffects of D-methionine were also studied. Unlike the L-isomer,D-methionine had no significant effects on the relaxant actionsof GTN, except at the 10 µM concentration of GTN, whichproduced only 20.4 ± 2.6% relaxation compared with 34.8 ±1.1% in the untreated RTCS. This observation differs fromothers, in which 1 mM L-methionine incubated with rat aorticsmooth muscle cells resulted in no change in intracellular sulf-hydryl content (Münzel et al. 1992) and no effect on in vitrorelaxation of canine coronary blood vessels. These divergentobservations could be reconciled by the possibility that vascu-lar smooth muscle cannot convert L-methionine to L-cysteinevia the transsulfuration pathway because of low activity orabsence of the required enzymes, whereas RTCS is able toconvert methionine to cysteine. The effects of L-methioninewere similar to those of NAC in that both of these cysteinedelivery systems were more effective in increasing the relaxanteffects of GTN in RTCS than L-cysteine itself. BothL-methionine and NAC would have to enter the cell before theamino acids could be converted to L-cysteine by transsulfura-tion of methionine and deacetylation of NAC, respectively. Incontrast, L-cysteine itself may be partially oxidized before itcan enter the cell, thereby resulting in decreased efficiency ofintracellular thiol delivery.

In both instances where RTCS was treated with theD-isomer of cysteine or methionine, there was an attenuationof the maximal relaxation; there is no obvious explanation ofthis observation. It is theoretically possible that D-cysteinecompeted with L-cysteine in the glutathione biosynthetic path-way; however, we are not aware of any experimental evidenceaddressing this possibility.

At first glance, increase of cellular cysteine content and thusglutathione content would implicate the glutathione-dependent enzyme family, glutathione S-transferases (GSTs)as an enzyme capable of biotransforming GTN to active me-tabolites capable of smooth muscle relaxation. However, GSTinhibitors had no effect on NO formation from GTN in intact(Chong and Fung 1993) and broken cell vascular tissue prepa-rations (Kurz et al. 1993). Other studies using denitrated GTNand vascular relaxation as indices for biotransformation haveproduced mixed results with respect to the effects of variousGST inhibitors (Lau et al. 1992; Chong and Fung 1993; Nigamet al. 1993). In addition, human vascular responses to GTNwere unaffected in those individuals with deficiencies in µ-class GSTs (Haefeli et al. 1993).

In spite of supplementation of the RTCS with high concen-trations of the potentiating L-amino acids, the sensitivity of theRTCS to GTN could not be increased to match that observedin rabbit aortic strips (Hussain et al. 1994). This could be at-tributed to an absence or deficiency of an enzyme system(s) inRTCS for the metabolic activation of GTN. Alternatively, theRTCS and rabbit aorta strip may involve different pathwaysfor activation of GTN.

In summary, NAC, L-cysteine, or the L-cysteine donor,L-methionine, was able to potentiate RTCS relaxation to GTN,whereas the D-isomers of cysteine and methionine had no ap-preciable effect on relaxation of RTCS. Such results implicate

Fig. 3. The effects of L-cysteine (5 mM) on relaxation of RTCS bySNAP (0.1 nM – 10 µM). Relaxations are expressed as themean ± SEM of three determinations. L-Cysteine pretreatment for30 min resulted in no significant changes in RTCS relaxation toSNAP compared with control tissues.

Can. J. Physiol. Pharmacol. Vol. 75, 1997810

© 1997 NRC Canada

a stereospecific role of L-cysteine or more likely its storageform, glutathione, in the metabolic activation of GTN in non-vascular smooth muscle, which is distinct from the GST sys-tem. The observations from this study warrant further study ofRTCS, in particular, levels of thiols present in this tissue andhow cysteine and methionine affect these levels. A substantialdifference still exists in the GTN sensitivity between thiol-treated RTCS and vascular tissue, which may be indicative ofan unidentified deficient enzyme in RTCS.

Acknowledgements

This study is supported by a grant from the Heart and StrokeFoundation of Ontario. Aman S. Hussain is a recipient of aHeart and Stroke Foundation of Canada Traineeship. Theauthors also acknowledge Mr. Brian McLaughlin andMr. Mark Kearney for helpful discussion of this work.

References

Bauer, J.A., and Fung, H.L. 1996. Specific binding of nitroglycerin tocoronary artery microsomes. Biochem. Pharmacol. 52(4): 619–625.

Bennett, B.M., McDonald, B.J., Nigam, R., and Simon, C.W. 1994.Biotransformation of organic nitrates and vascular smooth musclefunction. Trends Pharmacol. Sci. 15: 245–249.

Brien, J.F., McLaughlin, B.E., Kobus, S.M., Kawamoto, J.H.,Nakatsu, K., and Marks, G.S. 1988. Mechanism of glyceryl trini-trate induced vasodilation. I. Relationship between drug biotrans-formation, tissue cGMP elevation and relaxation of rabbit aorta.J. Pharmacol. Exp. Ther. 244: 322–327.

Boesgaard, S. 1995. Thiol compounds and organic nitrates. Dan.Med. Bull. 42(5): 473–484.

Boesgaard, S., Poulsen, H.E., Aldershvile, J., Loft, S.,Anderson, M.E., and Meister, A. 1993. Acute effects of nitroglyc-erin depend on both plasma and intracellular sulfhydryl compoundlevels in vivo Circulation, 87: 547–553.

Chong, S., and Fung, H. 1991. Biochemical and pharmacologicalinteractions between nitroglycerin and thiols. Biochem. Pharma-col. 42: 1433–1439.

Chong, S.J., and Fung, H.L. 1993. Relationship betweennitroglycerin-induced vascular relaxation and nitric oxide produc-tion. Probes with inhibitors and tolerance development. Biochem.Pharmacol. 45: 157–163.

Feelisch, M., and Noack, E. 1987. Nitric oxide (NO) formation fromnitrovasodilators occurs independently of hemoglobin or non-heme iron. Eur. J. Pharmacol. 142(3): 465–469.

Greutter, C.A., and Lemke, S.M. 1985. Dissociation of cysteine andglutathione levels from nitroglycerin-induced relaxation. Eur. J.Pharmacol. 111: 85–95.

Haefeli, W.E., Srivastava, N., Kelsey, K.T., Wiencke, J.K., Hoffman,B.B., and Blaschke, T.F. 1993. Glutathione S-transferase mu poly-morphism does not explain variation in nitroglycerin responsive-ness. Clin. Pharmacol. Ther. 53(4): 463–468.

Harrison, D.G., and Bates, J.N. 1993. The nitrovasodilators: newideas about old drugs. Circulation, 87(3): 1461–1467.

Hill, K.E., Hunt, R.W., Jr., Jones, R., Hoover, R.L., and Burk, R.F.1992. Metabolism of nitroglycerin by smooth muscle cells: in-volvement of glutathione and glutathione-S-transferase. Biochem.Pharmacol. 43: 561–566.

Horowitz, J.D., Antmann, E.M., Lorell, B.H., Barry, W.H., andSmith, T.W. 1983. Potentiation of the vascular effects of nitro-glycerin by N-acetylcysteine. Circulation, 68(6): 1247–1253.

Hussain, A.S., Poklewska-Koziell, M., Brien, J.F., Marks, G.S., andNakatsu, K. 1994. Nitric oxide formation from glyceryl trinitrate

by rabbit aortic strip: detection by rabbit taenia coli concurrentwith vasorelaxation. Br. J. Pharmacol. 113: 1508–1514.

Hussain, A.S., Brien, J.F., Marks, G.S., and Nakatsu, K. 1996. Super-oxide does not inhibit glyceryl trinitrate-rabbit aortic strip medi-ated relaxation of rabbit taenia coli: evidence for a protected formof nitric oxide? Drug Metab. Dispos. 24(7): 780–785.

Ignarro, L.J., Lippton, H., Edwards, J.C., Baricos, W.H.,Hyman, A.L., Kadowitz, P.J., and Greutter, C.A. 1981. Mecha-nisms of vascular smooth muscle relaxation by organic nitrates,nitrites, nitroprusside, and nitric oxide: evidence of the involve-ment of S-nitrosothiols as active intermediates. J. Pharmacol. Exp.Ther. 218: 739–749.

Kurz, M.A., Lamping, K.G., Bates, J.N., Eastham, C.L.,Marcus, M.L., and Harrison, D.G. 1991. Mechanisms responsiblefor the heterogeneous coronary microvascular response to nitro-glycerin. Circ. Res. 68: 847–855.

Kurz, M.A., Boyer, T.D., Whalen, R., Peterson, T.E., and Harrison,D.G. 1993. Nitroglycerin metabolism in vascular tissue: role ofglutathione S-transferases and relationship between NO and NO2

–

formation. Biochem. J. 292: 545–550.Lau, D.T., Chan, E.K., and Benet, L.Z. 1992. Glutathione

S-transferase-mediated metabolism of glyceryl trinitrate in subcel-lular fractions of bovine coronary arteries. Pharmaceut. Res.9(11): 1460–1464.

Mudd, S.H., Finkelstein, J.D., Irreverre, F., and Laster, L. 1965.Transsulfuration in mammals: microassays and tissue distributionof three enzymes of the pathway. J. Biol. Chem. 240: 4382–4392.

Münzel, T., Mülsch, A., Holtz, J., Just, H., Harrison, D.G., andBassenge, E. 1992. Mechanisms of interaction between the sulf-hydryl precursor L-methionine and glyceryl trinitrate. Circulation,86: 995–1003.

Needleman, P., and Johnson, E.M. 1973. Mechanism of tolerance toorganic nitrates. J. Pharmacol. Exp. Ther. 184: 700–715.

Newman, C.S., Warren, J.B., Taylor, G.W., Boobis, A.R., andDavie, D.S. 1990. Rapid tolerance to the hypotensive effects ofglyceryl trinitrate in the rat: Prevention by N-acetyl-L-cysteine butnot N-acetyl-D-cysteine. Br. J. Pharmacol. 99(4): 825–829.

Nigam, R., Whiting, T., and Bennett, B.M. 1993. Effect of inhibitorsof glutathione S-transferase on glyceryl trinitrate activity in iso-lated rat aorta. Can. J. Physiol. Pharmacol. 71(2): 179–184.

Selke, F.W., Myers, P., Bates, J.N., and Harrison, D.G. 1990. Influ-ence of vessel size on the sensitivity of porcine coronary microves-sels to nitroglycerin. Am. J. Physiol. 258: H515–H520.

Selke, F.W., Tomanek, R.J., and Harrion, D.G. 1991. L-Cysteine se-lectively potentiates nitroglycerin-induced dilation of coronarymicrovessels. J. Pharmacol. Exp. Ther. 258: 365–369.

Seth, P., and Fung, H.L. 1993. Biochemical characterization of amembrane-bound enzyme responsible for generating nitric oxidefrom nitroglycerin in the vascular smooth muscle cells. Biochem.Pharmacol. 46(8): 1481–1486.

Tsuchida, S., Maki, T., and Sato, K. 1990. Purification and charac-terization of glutathione transferases with an activity toward nitro-glycerin from human aorta and heart. Multiplicity of the humanclass mu forms. J. Biol. Chem. 265: 7150–7157.

Wheatley, R.M., Dockery, S.P., Kurz, M.A., and Harrison, D.G.1992. Interaction of nitroglycerin and sulfhydryl-donating com-pounds in coronary microcirculation. Circulation, 86(I): I509.(Abstr.)

Wheatley, R.M., Dockery, S.P., Kurz, M.A., Sayegh, H.S., andHarrison, D.G. 1994. Interaction of nitroglycerin and sulfhydryl-donating compounds in coronary microvessels. Am. J. Physiol.266: H291–H297.

Winniford, M.D., Kennedy, P.L., Wells, P.J., and Hillis, L.D. 1986.Potentiation of nitroglycerin induced coronary dilatation byN-acetylcysteine. Circulation, 73: 138–142.

Hussain et al. 811

© 1997 NRC Canada