<<

INFORMATION TO USERS While the most advanced technology has been used to photograph and reproduce this manuscript, the quality of the reproduction is heavily dependent upon the quality of the material submitted. For example: • Manuscript pages may have indistinct print. In such cases, the best available copy has been filmed. • Manuscripts may not always be complete. In such cases, a note will indicate that it is not possible to obtain missing pages. • Copyrighted material may have been removed from the manuscript. In such cases, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, and charts) are photographed by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each oversize page is also filmed as one exposure and is available, for an additional charge, as a standard 35mm slide or as a 17”x 23” black and white photographic print. Most photographs reproduce acceptably on positive microfilm or microfiche but lack the clarity on xerographic copies made from the microfilm. For an additional charge, 35mm slides of 6”x 9” black and white photographic prints are available for any photographs or illustrations that cannot be reproduced satisfactorily by xerography. Order Number 8717696

The chemistry and pharmacological mechanism of

Park, Jeen-Woo, Ph.D.

The Ohio State University, 1987

UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106 PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .

1. Glossy photographs or pages _____

2. Colored illustrations, paper or print ______

3. Photographs with dark background _____

4. Illustrations are poor copy______

5. Pages with black marks, not original copy ______

6. Print shows through as there is text on both sides of p a g e ______

7. Indistinct, broken or small print on several pages

8. Print exceeds margin requirements______

9. Tightly bound copy with print lost in spine______

10. Computer printout pages with indistinct print ______

11. Page(s) ______lacking when material received, and not available from school or author.

12. Page(s) ______seem to be missing in numbering only as text follows.

13. Two pages num bered . Text follows.

14. Curling and wrinkled pages ______

15. Dissertation contains pages with print at a slant, filmed as received ______

16. Other______

University Microfilms International THE CHEMISTRY AND PHARMACOLOGICAL MECHANISM OF

SODIUM NITROPRUSSIDE

DISSERTATION

Presented in Partial Fullfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Jeen-Woo Park/ B.S., M.S.

★ * ★ * ★

The Ohio State University

1987

Reading Committee: Approved by Gary E. Means

David H. Ives

Richard P. Swenson

Department of Biochemistry To My Parents, Wife and Daughter

ii ACKNOWLEDGEMENTS

With my deepest gratitude to my adviser, Dr. Gary E.

Means, for his invaluable guidance, continuous encouragement and care.

I wish to express my gratitude to Dr. David H. Ives,

Dr. Edward J. Behrman and Dr. Richard P. Swenson for their encouragement and valuable help.

I would also like to express my gratitude to Mr. David

Chang of Campus Chemical Instrument Center for mass spectral data and Dr. George E. Billman of College of

Medicine for help in animal experiment.

Special thanks are due to my fellow graduate students for their helpful suggestion and assistance.

iii VITA

January 12, 1955 .... Born-Taegu, Korea

1978 ...... B.S., Seoul National University, Seoul, Korea

1980 ...... M.S. Biochemistry, Korea Advanced Institute of Science and Technology, Seoul, Korea

1980-1983 ...... Researcher, Korea and Tobacco Research Institute, Seoul, Korea

1983-1984 ...... Graduate School Fellow, The Ohio State University, Columbus, Ohio

1984-1987 ...... Teaching and Research Associate, Department of Biochemistry, The Ohio State University, Columbus, Ohio

PUBLICATIONS

Park, J.W. and Means, G.E. (1987) Inactivation of Angiotensin Converting by Sodium Nitroprusside. Biochem. Biophys. Res. Commun., in press.

Monera, 0., Chang, M.K., Park, J.W., and Means, G.E. (1986) The Deamination of Residues of . Fed. Proc., 45. 1540. Abstract of the ASBC/DBC-ACS Meeting, Washington D.C.

Park, J.W. and Means, G.E. (1985) Formation of N- from Sodium Nitroprusside and Secondary . New Engl. J. Med., 313, 1547.

Park, J.W., Kim, S.J., and Lee, H.J. (1983) Active Site Functional Residues of Urokinase as a plasminogen Activator. Korean Biochem. J. 16, 100.

iv Kwon, B.M. and Park, J.W. (1983) Study of Hydantoin III. Mass Spectrometric Behavior in the Electron Impact of 2- Imino-4-thiazolidinone Derivatives. J. Heterocyclic Chem., 20, 1725.

Kwon, B.M., Oh, D.Y., and Park, J.W. (1983) Mass Spectra of Organophosphorus Compounds II. Unimolecular Reactions of 2- Phenyl-l,3,2-dioxaphosphoran 2-oxide Derivatives in the Gas Phase. Phosphorus and Sulfur 16, 271.

Park, J.W. (1982) Analysis of Phenolics in Cigaret Smoke by GC/MS with Multiple Ion Selection Technique. Arch. Pharmacal. Res. 5., 71.

Kim, K.R., Zlatkis, A., Park, J.W., and Lee, U.C. (1982) Isolation of Essential Oils from Tobacco by Gas Co- Distillation/Solvent Extraction. Chromatographia 15, 559.

Park, J.W. (1982) Direct Analysis of Nicotelline from Tobacco by MS with DADI/MIKE Spectrometry. Arch. Pharmacal. Re3. 5., 25.

Park, J.W., Lee, H.J., and Kim, S.J. (1980) Studies on the Anticoagulation Process (Fibrinolysis) by Urokinase. I. Purification and Characterization of Urokinase from Human Urine. Korean Biochem. J. 13. 127.

FIELDS OF STUDY

Major Field: Biochemistry

Studies in Molecular Biology. Professors George A. Marzluf and Lee F. Johnson

Studies in Carcinogenesis. Professors George E. Milo and Dorothy E. Schumm

Studies in Biophysical Chemistry. Professor Alan G. Marshall

v TABLE OF CONTENTS

page

DEDICATION ...... ii

ACKNOWLEDGEMENTS ...... iii

VITA ...... iv

LIST OF TABLES ...... viii

LIST OF FIGURES ...... ix

INTRODUCTION ...... 1

MATERIALS ...... 14

METHODS ...... 16

Preparation of N- derivatives of amines ...... 16

Colorimetric method for the detection of N-nitrosamines ...... 18

Kinetic studies for reactions ...... 19

Effect of blocking agents on the N-nitrosation of piperazine by SNP ...... 19

Modification of angiotensin I, II and ACE by SNP ...... 20

Assay of ACE activity ...... 21

Reaction of SNP with ...... 21

Sulfhydryl group determination by Ellman reaction...... 22

Identification of GSNO by ion-exchange column ...... 22

vi Preparation of S-nitrosothiols ...... 23

Homololysis of S-nitrosothiols ...... 24

Nitrite determination ...... 24

Identification of gas product ...... 24

Heterolysis of S-nitrosothiols ...... 25

Transnitrosation of S-nitrosothiols ...... 25

Effect of GSNO on the ADH activity ...... 26

Reaction of GSNO with hemoglobin ...... 27

Reaction of GSNO with HSA ...... 27

Animal experiments ...... 28

Gas chromatography ...... 29

High-performance liquid chromatography ...... 29

Mass spectrometry ...... 30

Other analytical methods ...... 30

RESULTS ...... 32

N-Nitrosamines from the reaction of SNP with amines ...... 32

Modification of angiotensin I, II and ACE by SNP ...... 49

Reaction of SNP with thiols ...... 52

Preparation and characterization of S-nitrosothiols ...... 59

Chemical reactions of S-nitrosothiols ...... 70

Animal experiments ...... 89

DISCUSSION ...... 97

LIST OF REFERENCES ...... 129

vii LIST OF TABLES

Table page

1. Products obtained from the reactions of SNP with physiologically and pharmacologically important amines ...... 43

2. Pseudo-first order rate constants for the nitrosation of secondary amines by SNP ...... 46

3. Effects of glutathione, cysteine and ascorbate on the reaction of SNP with piperazine ...... 48

4. Effects of modification of ACE with SNP in the presence of competitive inhibitors...... 54

5. S-Nitrosothiols from the reaction of thiols with sodium at pH 1.5 ...... 63

6. Homolysis of 20 mM S-nitrosothiols at pH 7.4 and 37° ...... 74

7. Reaction of thiols with S-nitrosothiols ...... 82

8. Equilibria for the transnitrosation from GSNO to various thiols ...... 86

9. Thiols in human blood (extracellular) ...... Ill

viii LIST OF FIGURES

Figure page

1. GC chromatogram of N-nitrosamines from the reaction of SNP with secondary amines ...... 33

2. Mass spectra of N-nitrosamines ...... 34

3. IR spectrum of N-nitrosomorpholine obtained from the reaction of morpholine with SNP ...... 36

4. Isobutane-CI mass spectra of the product induced by interaction of SNP with synephrine and phenylephrine ...... 39

5. IR spectra of the product induced by interaction of SNP with synepherine and phenylephrine ...... 40

6. First-order plots for the nitrosation of ephedrine with various concentrations of SNP and a plot of the pseudo-first order rate constants versus the SNP concentration ...... 45

7. pH dependence for the reaction of SNP with ephedrine, creatine and creatinine ...... 47

8. HPLC of reaction mixtures from angiotensin I and II with SNP ...... 50

9. Dependence of pseudo-first order rate constant, k, for the deamination of angiotensin II against concentration of SNP ...... 51

10. Changes in activity on the modification of ACE with SNP ...... 53

11. Spectral changes with time in the reaction mixture of 100 mM SNP and 1 mM GSH ...... 55

12. HPLC of reaction mixture from 20 mM SNP and 10 mM GSH ...... 56

ix 13. Spectral changes with time in the reaction mixture of 0.5 mM SNP and 50 mM GSH ...... 57

14. HPLC profiles for the reaction of 5 mM SNP with 50 mM GSH ...... 58

15. Effect of pH on the formation of GSNO ...... 60

16. UV-VIS spectrum of GSNO ...... 62

17. Ellman test with time for the reaction of HSA with ...... 64

18. Changes in the absorbance at 330 nm and concomitant loss of groups in HSA, as determined by the Ellman test upon the reaction with HSA and sodium nitrite ...... 66

19. Stability of GSNO crystal under different conditions ...... 67

20. IR spectrum of GSNO crystal ...... 68

21. FAB mass spectrum of GSNO crystal ...... 69

22. HPLC of product from the homolysis of GSNO .... 71

23. GC/MS of gas product from the homolysis of GSNO ...... 72

24. Effect of temperature on the stability of GSNO ...... 75

25. Effect of pH on the decomposition of GSNO ..... 76

26. Effects of pH and time on the formation of NO^ upon the decomposition of GSNO ...... 77

27. GC/MS of gas products from the decomposition of GSNO ...... 78

28. Spectral changes with time in the reaction mixture of GSNO with GSH ...... 79

29. pH-rate profile of the reaction of GSNO with GSH ...... 81

x HPLC of reaction mixture from GSNO and cysteine and chromatographic analysis of the standards and reaction mixture on the column of amino acid analyzer ...... 83

HPLC profiles of transnitrosation reaction between GSNO and N-acetylcysteine ...... 85

VIS spectra of transnitrosation reaction between S-nitroso-N-acetylpenicillamine and GSH ...... 87

Dependence of pH on the degree of transnitrosation from GSNO to N-acetylcysteine ...... 88

Time courses and concentration dependence of inactivation of ADH by GSNO ...... 90

Increase of absorbance at 412 nm on addition of DTNB to HSA and HSA plus GSNO ...... 91

VIS spectra of deoxyHb and deojcyHb with GSNO ...... 92

VIS spectra of oxyHb and oxyHb plus excess dithionite with GSNO ...... 93

Decrease on the mean arterial blood pressure of monkeys elicited by infusion of SNP and GSNO ...... 95

Decrease on the mean arterial blood pressure of dogs elicited by infusion of SNP and GSNO ...... 96

Proposed mechanism for the deamination of £-amlno group of lysine by SNP ...... 106

Proposed mechanism for the reaction of GSH and SNP ...... 113

Proposed mechanism for the pharmacological action of SNP via GSNO and other S-nitrosothiols ...... 127

xi INTRODUCTION

Sodium nitroprusside, SNP, CNa2Fe(CN)^N0. 2 H 2 O], is a potent, fast acting, intravenous hypotensive agent. It is used to lower blood pressure during hypertensive and cardiac emergencies, for the clinical treatment of malignant hypertension, and to lower cardiac back pressure during periods of cardiac insufficiency (Kreye, 1980;

Gerber and Nies, 1983; Barnhart, 1985). As the only agent available to affect rapid, short-term reduction of blood pressure, it is also used for the induction of "controlled hypotension" during many kinds of surgery (Tinker and

Cucchiara, 1978; Donchin et al. , 1978) and for a variety of other purpose (Benitz et al.. , 1985). It is administered as a dilute aqueous solution directly into the blood stream and is said to be the fastest acting and most dependable of all known hypotensive agents. It acts specifically on the vascular smooth muscles, affecting their relaxation and thereby, vasodilation and lower cardiac back pressure.

Its full effects are expressed in less than a minute, persist until infusion is terminated and dissipate within five to ten minutes. Effective dosage ranges that have been reported in the literature for the treatment of

1 2 hypertensive crises are varied. In general/ anywhere from

15 to 800 ug/min may be required and use of the drug for periods of several hours up to 32 days (Tuzel, 1974). Its hypotensive effects were first reported in 1887 by

Davidsohn/ but it was not widely used until 1974 when it became available in a convenient to administer form. The delay in the development of the drug stemmed from the reluctance in marketing a cyanide-containing compounds

(Gerber and Nies, 1983). Its use has increased rapidly since that time.

SNP is an inorganic salt which, as a dihydrate

(molecular weight, 297.95), forms rhomboid bipyramidal crystals of ruby-red color. SNP dissolves rapidly in water to give clear, bright orange solutions that are stable for at least months at 25° in the dark. Nitroprusside

2 - dianion, CFe(CN)^N0] , consists of a ferrous center surrounded by five cyanide groups and a group.

It is very reactive, and aqueous solutions decompose rapidly under strong alkaline conditions or in bright light

(Swinehart, 1967).

Cyanide toxicity resulting from the breakdown of SNP and liberation of cyanide ion under certain conditions has been a serious concern for many years (Vesey et. aJL., 1974;

Nakamura et. al.., 1977). Cyanide relase as a consequence of SNP therapy has been documented in dogs (Michenenfelder,

1977), baboons (Posner et. aJL., 1976a), guinia pigs (Posner 3 et al. , 1976b)/ and man (Vesey et. aJL., 1974; Vesey et al.# 1976). In clinical use/ especially when SNP was administered in large doses or for a prolong period, patients develop severe metabolic acidosis, and death has been reported. In such cases, cyanide was detected in the blood or the tissues (MacRae and Owen, 1974; Jack, 1974;

Merrifield and Blundell, 1974). SNP molecule contains five cyanide groups and the nitroprusside ion was slowly decomposed to cyanide in the presence of erythrocytes, probably by the interaction of the ion with sulfhydryl groups of the erythrocytes (Hill, 1942; Page et. al.,

1955). Smith et. aT. (1974) and Nakamura et. al. (1977) suggested that hemoglobin and reduced glutathione (GSH) may involve metabolism of cyanide. Although cyanide detoxification occurs via the well known enzyme system termed rhodanese, by which cyanide is converted to thiocyanate which is then slowly cleared by the kidneys

(Lang, 1933), it is dose-related. Cyanide, especially, and thiocyanate are very toxic. The load capacity of the body's detoxifying system may be too low to cope with large amount of SNP (Ivankovich, 1978). Their toxicities limit the amounts of SNP that may be safely administered and almost completely preclude its administration to patients with impaired liver or kidney function. Because thiocyanate is a strong inhibitor of thyroxine biosynthesis, SNP is also precluded for patients with 4 thyroid dysfunction (Kreye, 1980).

Although N-nitrosation reactivity of SNP on the simple secondary amines under non-physiological conditions has been reported by a few authors (Maltz et al.. , 1971; Casado et al.. 1985), its potential reactivity under physiological conditions has not been considered seriously.

2 - The nitroprusside ion, [Fe(CN)5N0] , is very reactive and can react rapidly with many kinds of to give 3_ complexes having the general structure CFe(CN)^N0X]

(Swinehart, 1967; McCleverty, 1979). All of these reactions appear to involve addition of an anion, X , to the coordinated NO moiety. In each case, the nitrosyl ligand reacts, in effect, like a stabilized nitrosonium ion

(i.e. N0+). Maltz et. al. (1971) showed that SNP could be used to effect the nitrosation of amines under alkaline conditions. Short-lived adducts resulting from the addition of an to the coordinated NO group appear to undergo ligand exchange to give corresponding

N-nitrosamines as follows (Casado et al., 1985):

CFe(CN)5N0]2“ + RNH + 0H_ £ CFe(CN)5N0NR]3_ + H20

> CFe(CN)50H]4_ + RNNO + H+

Since many drugs contain secondary amine structures, oral and intravenous administration of these drugs might constitute a human health hazard if SNP is used as a hypotensive agent at the same time. Some physiological 5 components in human tissues and blood can also react with

SNP during intravenous infusion. In summary, many drugs and physiological components containing a secondary amino group are capable of reacting with SNP to form potentially carcinogenic N-nitrosamines. Reaction of some tertiary amines with SNP also may produce related N-nitrosamines.

In this research, N-nitroso compounds have been prepared the reaction of physiologically and pharmacologically important secondary amines with SNP under physiological conditions. The reaction products were characterized.

Kinetic studies on the N-nitrosation reactions and the effects of blocking agents on these reaction are also described.

Because SNP is capable of producing cyanide toxicity and plays potent N-nitrosating agent for the physiologically and pharmacologically important amines, the development of alternative drugs which have the similar pharmacological mechanism to SNP and are safe may be highly desirable. The main objective of the subsequent study has been to develop a better understanding of the molecular mechanism for the pharmacological effects of SNP and to explore the possibility of alternative drugs. It seems probable that if the mechanism by which SNP lowers blood pressure is understood, it will be possible to rationally design or select other compounds which effect the same m e c h a n i s m . 6

Considering -the wide popularity of SNP, its potential

for toxicity, and the fact that it has been available for over half a century, suprisingly little is known about its

molecular pharmacology. In 1929, Johnson reported that the pharmacological action of SNP is due to a nitroso (NO)

group. It may be the nitroso group that imparts activity to the drug, for if the nitroso group is replaced by another cyanide moiety, the molecule is hemodynamically

inactive (Gerber and Nies, 1983). Later, Page et al.

(1955) suggested the possibility that SNP may penetrate the cell and interact with sulfhydryl groups in either

either membrane or contractile proteins. This possibility gained some credibility from the experiments of Needleman et al. (1973). They suggested the relaxation of vascular smooth muscle by SNP was due to the oxidation of key sulfhydryl groups but without specifying how this affects relaxation. Other workers have suggested a relationship exists between membrane sulfhydryl groups and calcium flux which may then affect contractile relaxation.

There is, however, no direct evidence to support such mechanism (Miletich and Ivankovich, 1978). Recently others have suggested that the drug elevates cyclic GMP and that may mediate vascular relaxation. At the present time there is considerable evidence supporting the hypothesis that relaxation by SNP and other , for example, (NO) gas, nitroglycerine, etc., is 7 mediated through activation of [GTP pyrophosphate-lyase (cyclizing), EC 4.6.1.23/ cyclic GMP accumulation/ activation of cyclic GMP-dependent protein kinase, altered protein phosphorylation, and dephosphorylation of myosin light chain (Rapport and Murad,

1983).

Guanylate cyclase activation by the nitroso group containing agents was first suggested by DeRubertis and

Craven (1976) in a report on the activation of that enzyme by nitrosamines and nitrite. There have been subsequent reports showing that guanylate cyclase can be activated by

SNP (Schultz et al.. , 1977), nitric oxide gas (Arnold et al.. 1977), 1-methyl-l-nitrosourea (Veseley et al., 1977), and nitroglycerine (Katsuki et. aT., 1977). Arnold et. al.

(1977) and Mittal and Murad (1977) demonstrated the activation of guanylate cyclase by nitric oxide and suggested that the latter moiety may be the common proximate species mediating the effects of azide, NH^OH, nitrite, SNP, and the nitroso compounds. Various inhibitors of guanylate cyclase have been useful in testing the hypothesis that -induced relaxation may be mediated through the formation of cyclic GMP. Reducing agents and free radical scavengers such as butylated hydroxianisole, hydroxiquinone, methylene blue, hemoglobin, methemoglobin, and myoglobin have been shown to inhibit guanylate cyclase activation by the nitrovasodilators 8

(Katsuki et. al.., 1977; Murad et al.., 1978; Mittal et. al., 1978; Gruetter et. al.., 1979; Ignarro et. al.., 1981).

Although the activation of guanylate cyclase by nitrovasodilators has been known for many years, the biochemical mechanism of activation has remained unsettled.

Craven and DeRubertis (1978) demonstrated that low concentrations of or hemoproteins are required for optimal expression of the responsiveness of a purified soluble guanylate cyclase to N-methyl-N'-nitrosoguanidine, nitric oxide, SNP, and nitrite. This change in enzyme responsiveness is associated with a concomitant loss of detectable heme in the enzyme preparation (Craven et al.,

1979). Preformed NO-heme complex was fully active as a stimulant of heme-deficient guanylate cyclase activity in the absence of added co-factors. These observations led to the proposal that activation of guanylate cyclase by nitric oxide and related agonists required the formation of

NO-heme (Craven and DeRubertis, 1978; DeRubertis et. al..

1978; Craven et al.., 1979). It also has been proposed that activation of guanylate cyclase by nitric oxide occurs by a free radical interaction with a heme moiety on the enzyme (Gerzer et al.. , 1981; Wolin et. al.., 1982).

These results suggest that the formation of a nitrosyl-heme complex is an obligate step in the pathway leading to activation of guanylate cyclase by N-nitroso compounds, NO, and related agents. Some controversy exists as to whether 9 or not forms other than the heme-containing enzyme are present in tissues. There have been many reports (Craven and DeRubertis, 1978; Gerzer et al., 1981; Lewicki et al.,

1982; Gerzer et al.., 1982) which supporting the view that heme is ordinarily bound to the enzyme, but removed during purification. Gerzer et al.. (1981) separated the holo- enzyme and apo form of soluble guanylate cyclase and showed that freshly prepared guanylate cyclase contained heme.

The apo form arose during the purification. These findings suggest that heme-free enzyme is not present in intact tissues.

Other reports (Ignarro et. al.. , 1980; Gruetter et: a l . ,

1980; Ignarro et. aT., 1981) have shown that

S-nitrosothiols, mainly S-nitrosocysteine, produced from the reaction of cysteine with NO gas, activate guanylate cyclase, although the chemical and pharmacological mechanism of this effect is far from clear. The authors of those reports have proposed a mechanism based on the instability of S-nitrosocysteine and subsequent effects of the NO formed upon its breakdown. This proposal lacks good chemical evidence, and has been recently challenged

(Craven and DeRubertis, 1983). The proposed activation of guanylate cyclase by S-nitrosothiols in the absence of heme was especially strongly criticized. The previous workers later showed that highly purifed, heme deficient preparation of liver and lung guanylate cyclase did not 10 respond to NO, S-nitrosothiols, SNP alone but are activated by NO-heme (Ignarro et. al . . , 1982; Ohlstein et al., 1982;

Ignarro et al.., 1986). They suggested that guanylate cyclase activation by NO and related agents is dependent in part on the presence of heme and thiols.

Protein sulfhydryl groups have been shown to be involved in modifying the responsiveness of the enzyme to nitric oxide, SNP and other related compounds (Braughler et. al., 1979a; 1979b). Agents with interact with sulfhydryl groups, such as N-ethylmaleimide, diamide, and cystamine inhibit guanylate cyclase activation (Katsuki et al..

1977), and the presence of free thiols influence both basal enzyme activity and activation of the purified enzyme by nitrovasodilators (Brandwein et aJL.., 1981). It has been suggested that redox reactions involving thiol group may participate both in the expression of basal guanylate cyclase activity and in the response of the enzyme to some antagonists and may be a mechanism of enzyme regulation.

The activation of guanylate cyclase without heme by nitrovasodilators without heme, however, is not significant

(Braugher et a l.., 1979a) as compared to that previously observed with heme or heme proteins. The oxidation and reduction of thiol in guanylate cyclase is likely responsible to reversal of activation instead of activation. 11

As part of an attempt to clarify how SNP lowers blood

pressure, we have investigated the effect of SNP on a

number of compounds normally found in the blood stream and tissues. There have been a few reports on the effect of

SNP on the renin-angiotensin system. Although several

reports suggest that SNP administration increases plasma

renin activity (Kaneko et al., 1967; Miller et. al.., 1977;

Lagerkranser et. a^.., 1985), there has been little

information on the influence of SNP on the angiotensin X

converting enzyme (EC 3.4.15.1; ACE) activity. SNP acts

as a deamination reagent for the primary amino groups at

physiological pH (Monera et. al.., 1986). ACE contains one

or more essential lysyl residue(s) at or near the active

site (Bunning et al.., 1978; Weare, 1982) and angiotensin I

and II have free 0(-amino group at N-terminal all of which

may be subject to modification by SNP.

Intravenously administered SNP can rapidly react with

thiol group-containing compounds under physiological

conditions, perhaps principally with GSH which is the most

abundant thiol compound in an in vivo system. The red

color which develops when SNP is added to thiols in

alkaline solution is one of the oldest methods to be used

for their detection (Swinehart, 1967). The color fades

fairly rapidly, and the overall reaction appears

kinetically complex (Mulvey and Waters, 1975; Morando et.

al., 1981). Although the main reaction products from the 12 reaction are disulfides (Morando et. al.. , 1981), Ignarro et al. (1981) have detected S-nitrosocysteine as a product of the reaction between SNP and cysteine under anaerobic / condition. Our assumption is that S-nitrosoglutathione

(GSNO) may be a common intermediate and serve as the nitroso group carrier in the mechanism by which SNP and several other nitrovasodilators lower blood pressure not only because it is the most abundant thiol in the blood stream and tissues but also its reaction with SNP is very fast. I will describe the reaction of SNP with GSH under physiological pH and detection of GSNO from this reaction.

S-Nitrosothiols are relatively simple compounds that have not received much study. Their alleged Instability has been a barrier to their study (Oae et. al.., 1977).

S-Nitrosothiols (RSNO) form from thiols in the presence of nitrite (Saville, 1958), N204 (Oae et al., 1977), and NO gas (Ignarro and Gruetter, 1980) under appropriate conditions and are structural analogs of

(H0N0) in which the hydroxyl group is replaced by an RS group. The primary aim of the remainder of this study was to characterize the chemistry of GSNO and related

S-nitrosothiols including, in particular, any reactions that may be responsible for its pharmacological properties.

On this basis the reaction of GSNO with hemeproteins and thiol group-containing proteins, were examined. To help demonstrate that GSNO is an intermediate in the 13 pharmacological mechanism of SNP, the effects of intravenous GSNO on the blood pressure of several animal species have been tested. Finally, the possibility that

GSNO and related S-nitrosothiols might be used as hypotensive drug to lower blood pressure in the same manner and under the same circumstances as SNP and advantages of

GSNO over SNP will be described. A molecular model for the pharmacological and biochemical mechanism of GSNO is also described. Along with an understanding of the chemistry and pharmacology of GSNO, other S-nitrosothiols and SNP this research also provide the possibility of developing of a new class of potent, effective, but relatively nontoxic hypotensive drugs. MATERIALS

SNP, piperazine, morpholine, propranolol, synephrine, phenylephrine.HC1, sarcosine, creatinine, creatine hydrate,

GSH, oxidized glutathione (GSSG), sodium ascorbate, cysteine hydrochloride hydrate, epinephrine, proline,

, palladium chloride, angiotensin I, angiotensin

II, L-aspartyl-0(({3-napthylamide) , hippuryl-L-histidyl-L-

leucine (HHL), N-hydroxyethyl-piperazine-N-2-ethane sulfonate (Hepes), L-Ala-L-Pro, Gly-L-Trp, bovine serum albumin (BSA), human serum albumin (HSA), 5,5'-dithiobis(2- nitrobenzoate) (DTNB), NAD, gelatin, N-l-napthylethylene diamine, cysteine ethyl ester, N-acetylcysteine, cysteamine, homocysteine, 0-mercaptoethanol, 3-mercapto- propionic acid, thioglycolic acid, dithiothreitol, penicillamine, N-acetylpenicillamine, methemoglobin, Tris base, angiotensin I converting enzyme, and yeast alcohol dehydrogenase were all obtained from Sigma Chemical Co.

Sodium nitrite and catechol were from Aldrich Chemical Co.

Ethyl acetate, ephedrine hydrochloride, sodium phosphate monobasic, sodium phosphate dibasic, ethanol, and D-glucose anhydrous were obtained from Mallinckrodt Chemicals Inc.

Sodium carbonate, sodium chloride, methanol, phosphoric

14 15 acid, diethyl ether, citric acid, sodium pyrophosphate, sulfanilic acid, ammonium acetate, acetonitrile, methylene chloride, boric acid, sodium dithionite, potassium carbonate, magnesium sulfate, and acetone were obtained from Baker Chemical Co. Diphenylamine, urea, and sodium hydroxide were obtained from Fisher Scientific Co.

Captopril (Capoten<§) ) was from Squibb Pharmaceutical Co. and phentolamine mesylate (Regitine mesylate) was from CIBA

Pharmaceutical Co. METHODS

Preparation of N-nitroso derivatives of amines.

N-Nitroso derivatives of amines were prepared according to both the procedure of Maltz et al.. (1971) and by reaction at pH 7.4 and 37° as described. N-Nitroso derivatives of morpholine, piperazine, and ephedrine were prepared by adding 0.1 M of these compounds to an aqueous solution (100 ml) containing SNP (0.4 M) and sodium carbonate (0.1 M) sufficient to bring the alkaline condition. The reaction mixture was stirred for ca. 12 hr at room temperature. Potassium carbonate (10 g) was then added and the mixture extracted with five 20-ml portions of methylene chloride. The methylene chloride extract was dried by adding magnesium sulfate and evaporated. The nitroso derivatives of propranolol, phentolamine mesylate, and epinephrine were prepared from each of those drugs

(10 mM) in 10 ml of 0.1 M phosphate buffer, pH 7.4, containing 40 mM of SNP. The reaction was carried out at

37° for 24 hr. Mixtures were subjected to ' high-performance liquid chromatography (HPLC). In the case of epinephrine, excess SNP was separated from the reaction mixture by passing the sample through an

16 17 anion exchange column (1.5 x 5 cm, Bio-Rad AG-2x8) which was pre-equilibrated with water. peaks were collected directly from the detector flow cell and dried.

The nitrosation of synephrine and phenylephrine was carried out by dissolving 0.1 M of drugs in 75 ml of 0.3 M phosphate buffer, pH 7.4 containing 0.5 M of SNP, and incubated for 12 hr at 37°. The products were extracted repeatedly with ethyl acetate to give 75 ml of total extract. The nitrosation of proline (0.1 M> with SNP (0.4

M) was carried out in 50 ml of 0.3 M phosphate buffer, pH

7.4. After reacting 24 hr at 37°, water was evaporated from the reaction mixture using a rotary evaporator and the residue was extracted with pure acetone. The acetone was removed by evaporation in a stream of nitrogen. The

N-nitroso derivative of creatine was prepared by adding creatine hydrate (0.1 M) to 50 ml of 0.3 M phosphate buffer, pH 7.4, containing SNP (0.5 M ) . After 24 hr incubation at 37° the reaction mixture was acidified and extracted three times with 20-ml portions of diethyl ether.

The diethyl ether solution was evaporated and the oily residue was allowed to crystallize. Repeated recrystallization from ethyl acetate gave a pale yellow sample. The N-nitroso derivative of sarcosine was prepared for use as authentic standard to identify the product from the reaction of creatine and SNP by dissolving sarcosine (0.2 M) in 50 ml of 0.3 M borate buffer, pH 9.0, 18 containing 0.5 M SNP. The reaction was continued for 6 hr at 37°. The resulting mixture was acidified and extracted with diethyl ether. Recrystallization from ethyl acetate gave a pale yellow product. The reaction of creatinine

<0.15 M) and SNP (0.6 M) was carried out at 37° for 24 hr.

The resulting solid was removed by filtration and washed with water. The product was recrystallized from hot water after treatment with charcoal. The nitrosation of nicotine was carried out by adding nicotine (0.02 M) to 50 ml aqueous solution of SNP (1 M>. The mixture was incubated for 24 hr at pH 7.4 and 37°. The product was extracted repeatedly with chloroform.

Identification of the products as nitrosamines was based on the data from gas chromatography/mass spectrometry

(GC/MS) and direct inlet/mass spectrometry (DI/MS) with electron impact (El) or chemical ionization (Cl) mode and was confirmed by infra-red (IR) spectrum, Preussmann test, and melting point.

Colorimetric method for the detection of N-nltrosamines.

The colorimetric detection of N-nitroso compounds was carried out by the method described by Preussmann et al.

(1964). Nitrosamines extracted with solvent or separated by HPLC were loaded on a TLC plate. Spray reagent consisting of 5 parts of a solution of diphenylamine, 1.5% in ethanol, and 1 part of a solution of pallidium chloride. 19

0.1% In 0.2% saline. A thin spray of this reagent was applied to the layer. Irradiation of the moist plate with ultraviolet (UV) light (A = 240 nm) for some minutes * max produced blue to violet spots in the case of N-nitroso compounds.

Kinetic studies for nitrosation reactions.

Pseudo-first order rate constants, k, for the reactions of SNP with each amine were calculated from plots of log (P -Poo)/Poo versus time, where P is the amount of L w N-nitrosamines at time t and Poo is the amount when the reaction is complete. Amounts of the various nitrosamines were determined at different reaction times by HPLC and GC.

Additionally, the initial rate method was also used to analyze the kinetic data to obtain kinetic data for reactions which were so slow as for monitoring of a large percentage of the reaction to be impractical.

Effect of blocking agents on the N-nitrosation of piperazine by SNP.

A solution of piperazine with or without blocking agents, ascorbate, cysteine, and GSH, in 20 ml of 0.1 M phosphate buffer, pH 7.4 was added to a solution of SNP in

30 ml of 0.1 M phosphate buffer, pH 7.4 (final concentration: SNP, 5, 10, 20 mM; piperazine, 5, 10 mM; blocking agent, 10, 20 mM). The mixture was stirred for

20 min at 37°. Potassium carbonate was then added and the 20 mixture was extracted with methylene chloride. The extract was dried by adding sodium sulfate and the methylene chloride was stripped off. The residue was dissolved in a small amount of methylene chloride and was applied to the GC.

Modification of angiotensin I, II and ACE by SNP

Modification of angiotensin I and II (0.25 mM each) with SNP were done in 2 ml of 0.05 M sodium phosphate o buffer, pH 7.5 at 23 for various time intervals.

Modified products were separated by HPLC and the collected

HPLC eluants were lyophilized and subjected to amino acid analyses after hydrolysis in 6 N HC1 for 22 hr. Reaction of the model compound, L-aspartyl-0((@-naphthylamide)

(10 mM) with SNP (40 mM) was carried out at pH 8.0 at 23° for 24 hr. The mixture was acidified and extracted with ether. Aliquots of the extract were applied to GC/MS with

Cl mode.

Chemical modification of ACE by SNP were carried out at 23° in 0.05 M Hepes buffer, pH 8.0, at an enzyme -7 concentration of 5 x 10 M. At indicated times, aliquots of the reaction mixtures were passed through a column of

Sephadex G-25 in order to remove excess SNP. Chemical modification of ACE by SNP in the presence of competitive inhibitors, L-Ala-L-Pro or Gly-L-Trp, was carried out by the same method. The modified enzyme was isolated by gel 21 filtration and then dialyzed extensively versus 0.01 M potassium phosphate buffer, pH 8.0, to remove inhibitors.

Enzyme activities were measured by hydrolysis of HHL as described by Cushman and Cheung (1971).

Assay of ACE activity.

Incubations for the spectrophotometric assay of HHL hydrolysis by ACE were carried out at 37° in disposable tubes. Each 0.25-ml mixture contained the following components at the indicated final concentrations: potassium phosphate buffer, 100 mM; sodium chloride, 300 mM; HHL, 5 mM; and enzyme or modified enzyme, 0.15 ml. The enzyme was added last to initiate the reaction and tubes were incubated, usually for 30 min. The enzyme reactions were terminated by addition of 0.25 ml of 1 N HC1; the HC1 was added before the enzyme in the zero-time control assays.

The mixture was extracted with 1.5 ml of ethyl acetate by vortex mixing for 15 sec. After a brief centrifugation, a

1.0 ml aliquot of each ethyl acetate layer was transferred to a clean tube and then evaporated. The residual hippuric acid was redissolved in 1.0 ml water and the amount formed was determined from its absorbance at 228 nm.

Reaction of SNP with thiols.

The reactions of SNP and GSH at pH 7.4 and 23° at different molar ratios and in the presence and absence of air were followed by visible spectroscopy and HPLC. Air 22 was excluded by evacuation and equilibrium with gas.

The reaction of BSA and HSA with SNP at pH 7.4 and 23° with or without 8 M urea was also followed at 522 nm and and total free thiol groups were determined by the Ellman reaction at different times.

Sulfhvdrvl group determination by the Ellman reaction.

Sulfhydryl groups of simple thiols and thiol group- containing proteins were determined by reaction with

Ellman's reagent, DTNB (Ellman, 1959). For the determination of total sulfhydryl group content, thiols were dissolved in 2.5 ml of 0.1 M Tris buffer at pH 8.0.

A fresh solution of DTNB (0.01 M in 0.05 M sodium phosphate buffer, pH 7.0) was prepared. DTNB solution (0.1 ml) was added to thiol-containing sample and the absorbance at 412 nm (£ = 13,600) was recorded with time.

Identification of GSNO by anion exchange column.

One milliliter of 100 mM SNP, in the appropriate

0,,-free buffer, pH 7.4, was prepared under a continuous flow of N2 gas. To this solution, 1 ml of 200 mM GSH, prepared in 02~free H20, was added by means of a Hamilton gas-tight syringe, in a pyrex tube with a serum bottle stopper. A N2 gas flow was maintained through a 20-gauge needle, vented to the atmosphere. After incubation, the reaction mixture was applied to an anion exchange (Bio Rad,

AG 2-x8) column (1.5 x 5 cm) with a ultraviolet monitor at 23

254 nm. GSNO was eluted at low pH (^3) . The content of each peak was determined by UV-VIS spectroscopy and HPLC.

Preparation of S-nitrosothiols.

S-Nitrosothiols were prepared according to Saville

(1958) and Field et aT. (1978) by dropwise addition of HC1 to a solution containing equimolar amounts of thiol compound and sodium nitrite until a pH of 1.5 was attained.

After standing for 5 min at room temperature, the

S-nitrosothiol solution was neutralized with NaOH.

Prior to making S-nitrosocaptopril, 100 mg tablet of

Capoten® was extracted with ethanol and evaporated to prepare . For the preparation of GSNO crystal, minimum amount of water was added to the equimolar amount of GSH and sodium nitrite and then HC1 was added to make pH

1.5. The resulting solution was rapidly frozen in a dry ice-acetone bath, washed several times with cold absolute ethanol, and dried under vacuum. The solution of

S-nitroso-HSA was prepared by reaction of equimolar amounts of HSA and sodium nitrite at pH 3.0 at 23°. Quantitative conversion of HSA to S-nitroso-HSA was followed by

increased absorbance at 330 nm with time and by employing the Ellman reaction to determine the remaining free thiols.

Crystalline S-nitroso-HSA was prepared by gel-filtration over a Sephadex G-25 followed by lyophilization. 24

Homolvsis of S-nitrosothiols.

Kinetic experiments were performed spectrophoto- metrically under various conditions by following changes in absorbance at the A of each S-nitrosothiol. The max characterization of soluble and gaseous products was achieved by HPLC and GC/MS, respectively.

Nitrite determination.

2 - Concentrations of NO were determined by diazotization of sulfanilic acid and subsequent coupling with N-l-napthylethylene diamine by a procedure described by Bell et. ajl. (1963). A 50 pi of sample was added to the mixture consists of 1 ml of 0.2% sulfanilic acid, 1 ml of

0.1% N-l-napthylethylene diamine solution, 1 ml of 0.4 N

HC1, and diluted with water to 10 ml and let stand for 5 min. Absorbances were read at 550 nm and substracted from a reagent blank.

Identification of gas product.

Samples in septum-capped vials were flushed with He at

0° for 30 min. After the He flush, the septum-capped vials were incubated at 37° until the completion of reaction. After incubation, 1 ml of gas product was injected to GC/MS by a gas-tight Hamilton syringe.

Separations were carried out on a glass column (6' x 1/4"

1.D.) packed with Carbosieve S-II. Flow rate of carrier gas (helium) was 25 ml/min. The initial oven temperature 25 was 50°/ held Isothermal for 0.1 min after injection , and was then programmed to rise by 10°/min to 210°.

Heterolvsis of S-nitrosothiols.

The reaction of S-nitrosothiols and various thiols were normally carried out spectrophotometrically under different conditions by following changes in absorbance at absorption maximum of each S-nitrosothiols <^540 nm) due to the disappearance of S-nitrosothiol. Characterization of the main gaseous and soluble products of the reaction was achieved by GC/MS and HPLC, respectively. Pseudo-first order rate constants, k, for the heterolysis of

S-nitrosothiol and related thiol were measured by disappearance of S-nitrosothiols and calculated from first order plots. The disappearance of S-nitrosothiols was followed by measuring the change in absorbance at \ of max each S-nitrosothiol. The unsymmetrical disulfide from the reaction of GSNO with cysteine at pH 7.4 and 23° was characterized by HPLC and amino acid analysis.

Transnitrosation of S-nitrosothiols.

The transnitrosation of GSNO with mercaptoethanol,

N-acetylcysteine, N-acetylpenicillamine, and cysteine were normally followed by HPLC after mixing the two reactants at pH 7.4 and 23°. The transnitrosation of various S-nitrosothiols to GSH was also followed by same procedure. Direct observation of transnitrosation was 26 possible for some S-nitrosothiols. Transnitrosation of

S-nitroso-N-acetylpenicillamine to GSH gave a spectral change from A max ^88 nm (green) to 544 nm (red), the reaction was followed by VIS spectrophotometer at wavelength between 450 to 650 nm.

Effect of GSNO on the ADH activity.

The lyophilized enzyme was dissolved at 1 mg/ml in 0.1

M phosphate buffer, pH 7.5. Immediatedly prior to use, it was diluted (200-fold) with 0.01 M phosphate buffer, pH 7.5 containing 0.1% gelatin. Activities of ADH were determined by the method of Vallee and Hoch (1955) by monitoring the absorbance increases at 340 nm and 25°.

The following were pipetted into each cuvette: 0.032 M pyrophosphate buffer, pH 8.8, 1.5 ml; 2.0 M ethanol 0.5 ml;

0.01 M NAD, 0.1 ml. After incubation in the spectrophotometer for 3-4 min at 25° to achieve temperature equilibrium, assays were initiated by adding the enzyme or modified enzyme to the cuvette. Rates were determined from the AA340/min from the initial linear portion of the assay curve. For the modification reaction, the ADH was typically incubated at room temperature in 0.01 M phosphate buffer, pH 7.5 with an excess of GSNO. Samples (50 pi) were taken at various times and quenched by the rapid 60-fold dilution with assay mixture. Determined activities were subjected to 27 correction by comparison with reagent-free control.

Reaction of GSNO with hemoglobin.

For the preparation of hemoglobin, a methemoglobin solution (25 mg/ml) was equilibrated with nitrogen gas for

20 min on ice, sodium dithionite (Na^S^O^) added, and the resulting mixture passed through a Sephadex G-25 column

(1.5 x 5 cm). Deoxyhemoglobin was prepared from hemoglobin by evacuation and equilibrium with gas. The conversion of methemoglobin to hemoglobin and hemoglobin to deoxyhemoglobin were verified spectrometrically (Craven and

DeRubertis, 1983). Reaction of hemoglobin with GSNO (100 jjM) with excess sodium dithionite in 0.05 M phosphate buffer, pH 7.4 was carried out at room temperature.

Anaerobic incubation was conducted in pyrex glass tube sealed with a serum stopper and vented to the atmosphere through a 20 gauge needle. Deoxyhemoglobin and GSNO in

0.05 M phosphate buffer, pH 7.4 were equilibrated separately for 10 min at 0° with a continuous flow of N,,.

Transfer of GSNO solution was made with a gas-tight

Hamilton syringe.

Reaction of GSNO with HSA.

Reaction of GSNO and HSA in 0.1 M phosphate buffer, pH 7.4 at room temperature was followed by decrease of absorbance at 544 nm and by employing the Ellman test.

For the Ellman test, the mixture of HSA (2 mM) and GSNO

(10 mM) was incubated for 1 hr. The mixture was applied 28 to a Sephadex G-25 prior to the Ellman reaction.

Animal experiments.

Fifty milligrams of SNP (Nipride, Roche Laboratories) was reconstituted in 250 ml of 5% dextrose in water as a stock solution. The GSNO stock solution was diluted to

50 ug/ml of 5% dextrose. Seven mongrel dogs (18.1-31.8

Kg) and seven cynomologus monkeys (Macaca fascicularis;

4.4-9.0 Kg) were used in this study. The animals were given a pre-anesthetic (monkey; Ketamine HC1, 10 m g / K g

I.M.j dogs: 0.2 mg/kg fentanyl and 1 mg/kg droperidol,

Innovar Vet I.V.) and anesthetized with a mixture of c(-chloralose (50 mg/kg) and urethane (500 mg/kg) given as an intravenous injection. The left femoral artery and vein were cannulated with heparinized saline-filled catheters. Arterial pressure was obtained from the arterial catheter using an electromedics MS-20 solid state pressure transducer and recorded on a Gould 2400 chart recorder. Randomized venous infusions of GSNO (n=7) and

SNP (n=5) were made using a Harvard apparatus compact infusion pump model 976. The infusion was maintained until a new steady state value for mean arterial pressure had been obtained for each infusion rate. The infusion rate was varied over the following ranges: GSNO, 1.5-38.9 pg/kg/min (8 levels); SNP, 1.7-43.8 pg/kg/min (6 levels). 29

Gas chromatography.

A Varian Model 3300 GC equipped with a thermal conductivity detector (TCD) was used. Separations were carried out in stainless steel column (6' x 1/8" I.D.) packed with 3% OV-17 on Gas-chrom Q (100-120 mesh). The flow rate of helium carrier gas was typically maintained at

30 ml/min. TCD conditions: temperature, 250°; filament temperature, 300°; filament current, 178 mA. The injector port temperature was 260° and various column temperature programmings were used.

High-performance liquid chromatography.

Two different chromatographic solvent systems were employed, each involving gradient separations performed on a HPLC system consists of two Constametric Model IIIG dual piston pumps (LCD/Milton Roy), a gradient master

(LCD/Milton Roy), a dynamixer (LCD/Milton Roy), a Kratos spectroflow 773 variable wavelength detector, and a Linear single channel chart recorder Model 1200. Separations were achieved using reverse phase chromatography on a

Brownlee column (4.6 x 250 mm) of Lichrosorb RP-8 or RP-18

(10 jam particle size). Linear gradients from aqueous

0.1% phosphoric acid to a 1:1 (v/v) mixture of 0.1% phosphoric acid and acetonitrile or from aqueous 0.1 M ammonium acetate to a 1:1 (v/v) mixture of 0.1 M ammonium acetate and acetonitrile were developed at a flow rate of 30

1 ml/min. The effluent was monitored continuously at 215 or 220 nm.

Mass spectrometry.

The GC/MS computer system consists of a Finnigan Model

4021 GC/MS with a quadrupole type fitted with a combined

EI/CI ion source operating El mode with an ionization potential of 70 eV. A Nova 4 data system controlled the instrument and acquired, processed, and stored the data.

The ion source temperature was 250° for El and Cl mode.

For Cl mode, methane and isobutane (purity 99.999%) were used as a reagent gas. The samples were introduced into the ion source with a heated direct inlet probe or a gas chromatograph coupled via an open-split interface to a mass spectrometer. For GC/MS analyses, a glass capillary column (20 m x 0.25 mm I.D.) coated with 0V-17 or DB 1 was used. Fast atom bombardment (FAB) mass spectra for GSNO were obtained on a Kratos MS-30 mass spectrometer fitted with an Ion Tech B-ll NF saddle field atom gun. The FAB source was operated at room temperature with a Xenon gas 3 -1 flow of about 0.5 cm min at 10 psi, giving a beam at

B KeV of 1 mA of current. The crystal form of GSNO was mixed with glycerol and then coated on the probe.

Other analytical methods.

Melting points uncorrected were determined on a

Electrothermal melting point apparatus. UV-VIS spectra 31 were obtained on a Cary 118C spectrophotometer. IR spectra were obtained on a Beckman Model 4220 IR spectrophotometer with NaCl cell and Nujol as a mulling agent. Amino acid analyses were carried out on a Beckmann

Model 119CL amino acid analyzer. Samples were hydrolyzed in 6 N HC1 at 110° for 22 hr, in vacuo, using Kontes vacuum hydrolysis tubes. The hydrolysates were then dried under reduced pressure in a heated dessicator containing solid sodium hydroxide. RESULTS

N-Nitrosamines from the reaction of SNP with amines.

The determination of volatile N-nitroso compounds was

usually accomplished by a GC with a TCD and coupled to a capillary GC/MS. Figure 1 shows the GC chromatograms of

the N-nitrosamines products obtained from the reaction of several secondary amines with SNP. In the case of less volatile N-nitrosamines, GC/MS with Cl method was used to

obtain a quasi-molecular ion. For nonvolatile N-nitroso compounds, HPLC and DI/MS with El and Cl modes have proven to be valuable methods. For N-nitrosomorpholine (Figure

2a), the molecular ion was of notable abundance and loss of the NO, M - 30, at m/z 86 was observed. For several nitrosamines, a peak at M - 17 was observed in the upper mass range, which is certainly due to elimination of a hydroxy radical. In the Cl mode, the M + 1 ion, the protonated molecular ion, is almost always the most abundant ion in the high mass region of the spectrum, and the use of Cl in molecular weight determination is straightforward. By Cl, with isobutane as a reagent gas,

N-nitrosoproline (Figure 2b) shows an abundant quasi- molecular ion (M + H)+, m/z 145. Abundant quasi-molecular 33

0 (/> C o a

u 0 +* 0 Q

i!«U UL_

■ ■ ■ 0 2 4 0 2 4 0 2 4 Time, hr b

Figure 1. GC chromatogram of N-nitrosamines from the reaction of SNP with secondary amines: a) morpholine; b) piperazine; c) ephedrine. The peaks are marked as follows: 1. morpholine; 2. N-nitrosomorpholine;

3. N-nitrosopiperazine; 4. N#N '-dinitrosopiperazine;

5. ephedrine; 6. N-nitrosoephedrine. Relatlv* Intensity a) c Nntoopern wt G/SC (ehn as (methane gas). GC/MS-CI reagent with reagent as N-nitrosoephedrine (isobutane c) gas); DI/MS-CI with N-nitrosoproline b) N-nitrosamines. of spectra Mass 2. Figure ) -irsmrhln wt CM-I mode; GC/MS-EI with N-nitrosomorpholine a) — ».« © 188.8 n/i

1 i ,11 75.8 9 5 2 18 168 148 128 15 89 * 9.0 j 1L 117.1 l.i t 135.9 C m mi * iH.e 111

177.1 195.1 8 28 240 228 288 >• H. I 2S2.8 I . H *>?•» 223.9 235.0 235.0 223.9 268 T

34 35 ion (M + H> + , m/z 195, M + 29 (M + , m/z 223, and M + 41 (M + CgH^.)* are encountered from the methane-CI mass spectrum of N-nitrosoephedrine (Figure 2c).

Although the UV spectra for N-nitrosamines showed absorption maximum at about 350 nm (£ ca. 100) in alcohol, it is difficult to use UV spectroscopy for analytical purpose, because of its low extinction coefficient and interference due to the presence of starting material.

Because N-nitrosamines show relatively intense bands in the

IR spectrum, IR spectroscopy was used as additional evidence in the identification of the products.

N-Nitrosamines show three relatively intense bands in the

IR region at 1470-1350, 1316-1163 and 1093-1047 cm-1. The first two have been assigned to the vibrations of the N=0 bond and the last to the vibrations of the N-N bond

(Pensabene et. al.., 1972). Especially, the band at 1470-

1350 cm ^ was used additional evidence for N-nitrosamines detection. By using a Preussmann test, related

N-nitrosamines from the reaction of SNP with secondary amines showed blue to violet spots on a TLC plate.

N-Nitrosomorpholine from the reaction of morpholine with SNP was easily identified by GC/MS with El mode. In

+ + mass spectrum, m/z 116, (M) , and m/z 86, (M - NO) showed abundant intensity. In the IR spectrum,

N-nitrosomorpholine showed a prominent band around

1460 cm 1 (Figure 3). The nitrosation of piperazine by 36

100

» 60

40

20-

4000 3200 24001800 1400 1000 600 W ave Number (cm-1)

Figure 3. The IR spectrum of N-nitrosomorpholine obtained from the reaction of morpholine with SNP. 37

SNP produced both mono- and dinitroso derivatives. In an

El mass spectrum of N-nitrosopiperazine, m/z 115 [(M)+, 6% of base peak intensity)], m/z 85 [(M - N0) + , 100%], and m/z

56 (100%) were encountered. For N,N'-dinitrosopiperazine, m/z 144 [(M)+ , 28%], m/z 114 [

C

N-nitrosopiperazine to N,N'-nitrosopiperazine was thus observed after reaction for 1 hr with 20 mM SNP at pH 7.4 and 3 7 ° . The conversion of mono- into dinitroso derivatives thus appears to be slightly slower than the conversion of piperazine into N-nitrosopiperazine by SNP.

N-Nitroso compounds from ephedrine, propranolol, and phentolamine with SNP were separated by HPLC and identified by DI/MS. Because molecular ions for these products were hardly detectable by MS with El mode, MS with Cl mode was

applied to obtain an abundant quasi-molecular ion

intensity. Protonated molecular ions, (M + H)+, for

N-nitrosoephedrine at m/z 195, N-nitrosopropranolol at m/z

289, and N-nitrosophentolamine at m/z 311 were observed

at higher intensity. 38

The sympathomimetic drugs, phenylephrine and its isomer, synephrine appear to undergo both N-nitro3ation and nitration of the aromatic ring upon reaction with excess

SNP as follows:

HO ?H ,CH, SNP Yn .CH3 ch-ch2-n Vh ------O J>-ch-ch2-nVno II

Phenylephrine

Isobutane-CI mass spectra of the product obtained in each case gave the same protonated molecular ion, (M + H) at m/z 242, and a prominent fragment ion, [ (M + H) - NO] , at m/z 212 (Figure 4). IR spectra of both products revealed a strong band at 1460 cm ^ (Figure 5), characteristic of the (N=0) group, and both gave a positive Preussmann test.

The products obtained from the reactions of SNP with both phenylephrine and synephrine thus reveal the anticipated

N-nitroso moiety but also an extra, unexpected, NO^ moiety, presumably, as a nitro-substituent on the respective aromatic rings.

The natural occurrence of amino acids which contain the secondary amine structure such as free proline and sarcosine in blood constitute a potential source of nitrosamines via interaction of SNP when it is administered y neato o SP ih ) yehie and synephrine a) with induced SNP of product the of interaction by spectra mass Isobutane-CI 4. Figure ) phenylephrine. b)

Relative Intensity Relative Intensity n/t 5I 51 ,111.1 95.1 T5.I 123,1 179.1 179.1 197.1 251

yw* | 39

40

100

.8 0 -

.2 60-

40-

20 -

100

.8 0 -

« 60 -

« 40-

20 -

4000 3200 2400 1800 1400 1000 600 Wave Number (cm-1)

Figure 5. IR spectra of product induced by interaction of

SNP with a) synephrine and b) phenylephrine. 41 intravenously. The product induced from the reaction of

SNP with proline was identified as N-nitrosoproline by isobutane-CI mass spectrum, protonated molecular ion at m/z

145, melting point (99-100°), and a positive result in

Preussmann test. N-Nitroso compound from the interaction of SNP with sarcosine was pale yellow with m.p. 66-67°.

The methane-CI mass spectrum of the product showed a protonated molecular ion at m/z 119, and characteristic fragment ions at m/z 103 (M - CHg)+ and m/z 88 (M - N0)+ .

The IR spectrum showed significant bands at 1720 cm ^ (C=0) along with 1455 cm *

The nitrosation of epinephrine by SNP produced a black, viscous solution, similar to that obtained upon reaction of SNP with catechol under the same conditions.

Because SNP can react as an oxidizing agent, it is possible to expect that catechol moiety of epinephrine might react with two moles of SNP to form an orthoquinone. However, the product induced by the reaction of epinephrine with SNP was separated by HPLC following the anion exchange chromatographic step. The collected HPLC peak showed a positive Preussmann test and the isobutane-CI mass spectrum confirmed the product as a N-nitrosoepinephrine. It is possible to postulate that the reaction mixture might contain N-nitrosoepinephrine and its catechol, semiquinone, and quinone derivatives. If the reaction was carried out 42 to completion/ this would have resulted in the loss o£ of

SNP present by the reaction between catechol moiety with

SNP, a factor which must be considered when comparing the extent of nitrosation of the structurally closely related compound, e.g. ephedrine.

The reaction of creatinine with SNP under physiological conditions for 2 hr incubation yielded 143 mg

(13.4% yield) of a white solid, m.p. 254-255° (with decomposition), which gave a negative result in a

Preussmann test for a N-nitroso group. In the isobutane-

CI mass spectrum, a quasi-molecular ion at m/z 143 as a base peak, and IR spectrum confirmed product as creatinine-

5-oxime.

The nitrosated product of creatine by SNP was a pale yellow sample with m.p. 66-67°, and the yield was 4.1% under physiological conditions for 24 hr. The methane-CI mass spectrum was the same as obtained with

N-nitrososarcosine. The reaction product was also characterized by GC and found to be identical to authentic

N-nitrososarcosine as synthesized directly from sarcosine.

N-Nitrosonornicotine obtained from the reaction of SNP with nicotine was identified by GC/MS with methane-CI which showed (M + H)+ at m/z 178, (M + H - 0H)+ at m/z 161, and

(M + H - N0)+ at m/z 148 as prominent peaks. The yield of

N '-nitrosonornicotine was quite low, less than 2%, under the given conditions. N-Nitrosamines identified from the 43

Table 1. Products Obtained from the Reactions of SNP with

Physiologically and Pharmacologically Important Amines.

. a Product . Amine m/z X yield Name"*

Morpholine 116

Ephedrine 195 (Cl) 1.5c'f N-Nitro3oephedrine a Phenylephrine 242 (Cl) o o N-Nltroso-nitrophenylephrine

Synephrine 242 (Cl) 11.7°'* N-Nitroso-nitrosynephrine

Propranolol 289 (Cl) 1.2c,f N-Nitrosopropranolol

Phentolamine 311 (Cl) 1.4C'f N-Nitrosophentolamine

Proline 145 (Cl) 1.2C/f N-Nitrosoproline

Sarcoslne 119 (Cl) 11.3d'9 N-Nitrososarcosine

Creatinine 143 (Cl) 13.4d '9 Creatinine-5-oxime

Creatine 119 (Cl) 4.1d 'h N-Nitrosoaarcosine

Nicotine 178 (CD 1.9b'A N '-Nitrosonornicotine

a) Molecular ions obtained in the electron impact (El) or chemical ionizationCCI) mode. b) Yields are based on the relative peak areas obtained by GC as compared to the starting material; c) HPLC with a 215 nm absorption monitor; d) from the weight of isolated product as compared to that of the starting amine. e) Reactions were conducted at 37° in the presence of 0.1 M phosphate buffer at pH 7.4 with e) 0.05 M amine and 0.01M SNP for 2 hr; f) 0.2 M amine and 0.5 M SNP for 2 hr; g) 6 hr; h) 24 hr; i) or 1.0 M SNP for 24 hr. j> All products except creatinine-5-oxime are positive in a Preussmann test. 44 reaction of pharmacologically and physiologically important secondary amines with SNP and the yield for each reaction are given in Table 1.

In order to obtain information on the reaction order for the nitrosation of ephedrine by SNP, pseudo-first order rate constants were determined as a function of SNP concentration. The result in Figure 6 shows clearly that there is a linear releationship between the rate constant and the SNP concentration. Table 2 lists the pseudo-first order rate constants, k, for the reaction of six secondary amines with SNP under physiological conditions.

The pH dependence for the nitrosation of ephedrine, creatine, and creatinine by SNP were studied over the range pH 7.4-10. Nitrosation rate for ephedrine and yield for creatine were strongly enhanced by an increase in pH in basic pH range (Figure 7a,b). The yield of creatinine-5- oxime from creatinine and SNP was also enhanced by an increase in pH (Figure 7c).

The effects of blocking agents known to inhibit reaction of nitrite with secondary amine were examined for the nitrosation of piperazine by SNP under physiological conditions. The blocking effects of GSH, cysteine, and ascorbate on the nitrosation of piperazine by SNP are compared in Table 3. The blocking action of ascorbate was slightly more pronounced than that of GSH and cysteine. 45

C o • I o 3 .E (0 a) E GC JiC X *o ^ o

0.8

0 60 120 0 0.2 0.4 Time, hr [SNP], M

Figure 6. a) First-order plots for the nitrosation of 5 mM ephedrine with 100 (o), 200 (□), 375 (•), and 500 CO) mM

SNP in 0.1 M borate buffer, pH 10 and 3 7 ° ; b) a plot of

the pseudo-first order rate constants versus the SNP

concentration. Table 2. Pseudo-first Order Rate Constants for the

Nitrosation of Secondary Amines by SNP .

Drug [SNP], M 102 x k, min 1

Piperazine 0.1 16b

Morpholine 0.1 2.2b

Ephedrine 0.5 0.012°

Propranolol 0.5 0.0091°

Phentolamine 0.5 0.013°

Proline 0.5 0.0082°

a) Reactions of 5 mM amine were conducted at 37°

in 0.1 M phosphate buffer, pH 7.4. b) Amounts of the nitrosamine products were

determined at different times by GC as

described in the text. c) Amounts of the nitrosamine derivatives were

determined by HPLC as described in the text. 47

30

4

Ic 20 x o

1

8 9 107 7 8 9 10

Figure 7. pH dependence for the reaction of 0.5 M SNP with a) ephedrine, 0.005 M, b) creatine, 0.1 M, and c) creatinine, 0.1 M at 37° in 0,1 M phosphate (pH 7.4),

Tris (pH 8.4) and borate buffers (pH 9.0 and 10.0). 48

Table 3. Effects of Glutathione, Cysteine and Ascorbate on the Reaction of SNP with Piperazine .

[Piperazine] [SNP] Blocking Yieldb (mM) (mM) Agent (m M ) (%)

10 10 - - 60.5

10 10 Glutathione 10 20.6

10 10 Glutathione 20 10.3

10 10 Cysteine 20 9.7

10 10 Ascorbate 20 7.3

a) Reactions were conducted in 0.1 M phosphate buffer at

pH 7.4 for 20 min at 37°.

b) Yields based on the peak areas obtained for

N-nitrosopiperazine and piperazine by gas chromatography. 49

Modification of angiotensin I, II# and ACE by SNP.

The time courses for the modification of angiotensin

I and II by SNP were examined by reversed phase HPLC

(Figure 8). The use of 0.1 M ammonium acetate/ acetonitrile as a solvent gradient system provided the best resolution. Addition of a 100-fold excess of SNP to angiotensin I and II at pH 7.5 at 23° resulted in a loss of the two vasopeptides in a pseudo-first order process with a -4 -4 -1 rate constant, k, of 7.6 x 10 and 3.7 x 10 min , respectively. Figure 9 shows that the degree of modification is associated with the concentration of SNP.

Modification resulted in three significant derivatives of each vasopeptide and was shown to involve deamination of the N-terminal amino acid residue (Asp) for each derivative was identified by amino acid analyses. Results of amino acid analyses of modified vasopeptides with SNP revealed a definite loss of a aspartic acid content. The structures of the modified N-terminal Asp of angiotensin I and II were determined by a combination of HPLC and GC/MS techniques by using L-aspartyl-ol( @-naphthylamide) as a model compound.

Like the HPLC pattern of angiotensin I and II, two large peaks and one small peak were obtained by GC/MS from the extract of reaction mixture from SNP and model compound.

Two large peaks showed identical m/z at 242 as a protonated molecular ion corresponds to the olefin derivatives of model compound. The peaks could be the cis and trans 50

a b

0.8 0.8

0 10 20 0 10 20 RT, min RT, min

Figure 8. HPLC of reaction mixtures from angiotensin I (a) and II (b) with 100-fold molar excess of SNP after 24 hr at

23°. A gradient from 0.1 M ammonium acetate to 0.1 M ammonium acetate/acetonitrile (1:1 by vol.) was used as the solvent system. Other conditions are shown in text.

Peaks: a) 1. alcohol derivative; 2 and 3. olefin derivatives; 4. angiotensin I b) 1. alcohol derivative

2 and 3. olefin derivatives; 4. angiotensin II. iue . eedne f suofrt re rt constant, rate order pseudo-first of Dependence 9. Figure , o te emnto o agoesn I against II angiotensin of deamination the for k, ocnrto o SNP. of concentration

103xk, min- 3 2 1 0 0.1 (SNP|,M 0.2

51 52 isomers of olefin derivatives. The small peak showed m/z at 260 as a protonated molecular ion that corresponds to alcohol derivative of model compound.

The time course of inactivation of ACE by SNP under indicated conditions is shown in Figure 10. Approximately

50% of the activity was lost in less than 4 min with 10 mM of SNP. The loss of ACE activity increased in rate with the concentration of SNP. As shown in Table 4, dipeptide inhibitors, L-Ala-L-Pro and Gly-L-Trp, significantly reduced the effects of the modification by SNP.

Recovery of activity was about 90% of the control.

Reaction of SNP with thiols.

The reaction of SNP with GSH at pH 7.4 and 23° gave rise to an intense red color which developed

"instantaneously" and then faded gradually. The spectrum of the colored intermediate showed a maximum at 522 nm.

The spectral changes with times are shown in Figure 11.

The primary stable product identified by HPLC after complete fading of color was identified GSSG when the reaction was carried out in the presence of air and molar excess of SNP (Figure 12). At a high value of GSH to SNP ratio, which is more likely to exist in an in. vivo system, the VIS spectrum showed additional absorbance around 400 nm

(Figure 13). Formation of S-nitrosoglutathione (GSN0) from SNP with GSH was difficult to monitor by visible 53

100

o or- X £ > 50

>

O <

0 30 60 Time, min

Figure 10. Changes in activity on the modification of ACE with 2 (o), 5 (•)/ and 10 (□) mM SNP in 0.05 M Hepes buffer, pH 8.0 at 23°. Activities are expressed as the ratio of that of the modified enzyme, V, and the unmodified control, V , times 100. c Table 4. Effects of Modification of ACE with SNP in the Presence of Competitive Inhibitors.

Modification Reaction3 Activity, V/Vc x 100b

ACE (control) 100

ACE + SNP 55

ACE + L-Ala-L-Pro 92

ACE + SNP + L-Ala-L-Pro 80

ACE + Gly-L-Trp 90

ACE + SNP + Gly-L-Trp 79

a) Reactions were carried out with ACE in 0.05 M

Hepes, pH 8.0, containing 0.3 M NaCl for 1 hr at

23° in the presence or absence of 10 mM SNP or

10 mM inhibitors as indicated. Modified enzyme was isolated by gel filtration and then dialysis. b) Activities are expressed as the ratio of the modified enzyme, V, and the unmodified control,

V , times 100. c 55

Omin

1.5

o oc

0.5

2)0

4 0 0 5 0 0 6 0 0 Wavelength, nm

Figure 11. Spectral changes with time in the reaction mixture of 100 mM SNP and 1 mM GSH at pH 7.4 and 23°. 56

2.0 GSSG

* «______»_____I_ 0 4 8 12 RT, min

Figure 12. HPLC of reaction mixture from 20 mM SNP and 10 mM GSH at pH 7.4, 23° after 1 hr. A gradient from 0.1%

HgPO^ to 0.1% HgPO^/methanol (1:1 by vol.) was used as the solvent system. Other conditions are given in text. 57

1.2

0.8

0.4 0 min

20

400 500 600 Wavelength, nm

Figure 13. Spectral changes with time in the reaction mixture of 0.5 mM SNP and 50 mM GSH at pH 7.4, 23 . 58

2.0

GSNO

0 4 8 RT, min

Figure 14. HPLC profile for the reaction of 5 mM SNP with

50 mM GSH at pH 7.4, 23° under nitrogen. A gradient from

0.1% HoP0. to 0.1% H^PO./methanol (1:1 by vol.) was used as 3 4 3 4 1 the solvent system. Other conditions are given in text. 59 spectroscopy because of the necessity of removing the

interferring red colored complex before spectral analysis of reaction mixture. Therefore, either HPLC or an ion exchange column was used to separate GSNO from the reaction mixture. Figure 14 shows the HPLC profile of the reaction of a 10-fold excess of GSH with SNP. Extra peak from the reaction conditions showed same retention time of GSNO prepared from the reaction of GSH with sodium nitrite at pH

1.5. With equimolar of GSH and SNP in the presence of air, no detectable GSNO was observed. When the reaction was carried out anaerobically, a small amount of GSNO was produced. This could be separated by ion exchange column from the anaerobic incubation of GSH and SNP. Its identity was confirmed on the basis of its very characteristic UV-VIS spectrum as compared to authentic

GSNO which was prepared by GSH and sodium nitrite as described in the methods section. In contrary to GSH, the thiol-containing protein, HSA, did not give coloration with

SNP. Although denaturated HSA as obtained in the presence of 8 M urea produced an absorbance increase at 522 nm and eliminates its reaction with DTNB, native HSA didn't react with SNP and thus didn't eliminate reaction with DTNB.

Preparation and characterization of S-nitrosothiols.

The formation of S-nitrosothiols was essentially completed instantaneous at pH 1.5 as described in the 60

0.3

= 0.2

0.1

5 pH

Figure 15. Effect of pH on the formation of GSNO from 20 mM

GSH plus 20 mM sodium nitrite at 23° for 10 min. 61 methods section. Figure 15 shows that GSNO formation from

GSH with sodium nitrite occurred mainly at acidic pH. At pH 7.4, insignificant amounts of GSNO were detected.

Exclusion of light and air had no noticeable effect on the formation of GSNO.

Solutions of S-nitrosothiols gave characteristic red color except those formed from penicillamine and

N-acetylpenicillamine/ which were green. All the

S-nitrosothiols have two significant X around 535-550 nm max and 325-340 nm except S-nitrosopenicillamine and

S-nitroso-N-acetylpenicillamine/ which showed A max at 592 and 588 nm, respectively, and at 325-340 nm. The UV-VIS spectrum of GSNO is shown in Figure 16. Extinction coefficient (£) for S-nitrosothiols at VIS range were not / very large. Even a large molar excess of sodium nitrite relative to the thiol compounds didn't affect the absorption obtained except in the case of dithiothreitol.

Because dithiothreitol has two thiol groups a 2:1 molar ratio of sodium nitrite and dithiothreitol produced approximately two-fold higher extinction coefficient value.

Table 5 summarizes the color, X , and extinction max coefficient of a number of S-nitrothiols.

Reaction between HSA and sodium nitrite at pH 3.0 and room temperature were followed by the Ellman reaction which determines remaining free thiol groups of HSA. At high ratio of sodium nitrite to HSA, the rate of reaction 62

2.0 0.5

0) co (0 4a i. o VI JD <

300 400 500 600 Wavelength, nm

Figure 16. UV-VIS spectrum of GSNO. GSNO (2 mM) and GSNO

(20 mM) were followed in the range of 600-250 and 600-450 nm, respectively. 63

Table 5. S-Nitrosothiols from the Reaction of Thiols with

Sodium Nitrite at pH 1.5.

UV-VIS S-Nitrosothiol Color A ,n m (6 ) ' max S-Nitrosoglutathione red 544(15.0) 332(750)

S-Nitrosocysteine red 540(15.5) 334(760)

S-Nitrosocysteine ethyl ester red 542(15.5) 335(758)

S-Nitroso-N-acetylcysteine red 542(15.8) 338(875)

S-Nitrosocysteamine red 543(14.7) 331(650)

S-Nitrosohomocysteine red 545(18.7) 330(890)

S-Nitroso-(3-mercaptoethanol red 542(18.8) 328(990)

S-Nitroso-3-mercapto- red 544(17.9) propionic acid 327(970)

S-Nitrosothioglycolic acid red 538(9.1) 329(875)

S-Nitrosodithiothreitol red 541(18.4)a 332(870) 541(37.0) 332(1750)

S-Nitrosocaptopril red 542(19.8) 330(995)

S-Nitrosopenicillamine green 592(9.1) 342(450)

S-Nitroso-N-acetyl- green 588(12.1) penicillamine 328(635) a) [SNP]/CDithiothreitol3=1; b) CSNP]/[Dithiothreitol]=2 64

100

80

£ 6 0

4 0

10 20 10 20 TimeTime, min (NaNCU, mM

Figure 17. a) Ellman -test with time for the reaction of 100 pM HSA with 2.5 mM NaN02 at pH 3.0 and 23°; b) Ellman test for the reaction of 100 )iM HSA with various concentrations of NaN02 at pH 3.0 and 23°. 65

increased (Figure 17). Although detection of the

characteristic red color from the reaction of HSA and

sodium nitrite was impractical because of its low

extinction coefficient/ the Ellman reaction showed

a concomitant loss of free thiol group by the reaction.

The absorbance at 330 nm, characteristic for

S-nitrosothiols, increased with reaction time (Figure 18).

At pH 7.4/ S-nitroso-HSA slowly lost the NO group and

produced free thiol group with time, which was followed by the Ellman reaction. The crystal form of S-nitroso-HSA

was prepared by lyophilization with good yield.

The yield of crystalline GSNO, a pink powder, which

was prepared by the described method and determined by HPLC

and visible spectroscopy, was greater than 90%. The only

other soluble product from that procedure was GSSG. The

crystalline material was relatively stable in the air at 4° and, even at room temperature, 75% of original absorbance remained after 32 days. Exclusion of light and air had

virtually no effect on the stability of GSNO crystal

(Figure 19). Additional confirmation of the identity of

GSNO was obtained by IR spectrum and FAB/MS. From the IR spectrum of GSNO crystals, a peak characteristic of the sulfhydryl moiety in the GSH ( 2540 cm ^) was absent from the GSNO and the nitroso moiety was identified as a sharp peak at 1450 cm ^ (Figure 20). FAB is a soft ionization method that produces cationic molecular ion and fragment 66

100- '►0.6

80-

-0.4

60- o co SZ co CO n < 40*

- 0.2

2 0

Time, hr

Figure 18. Increase in the absorbance at 330 nm (---- ) and concomitant loss of thiol groups in HSA, as determined by

the Ellman test (— •— ) upon the reaction with 0.5 mM HSA

and 0.5 mM NaNO^ at pH 3.0 and 23 67

0-4

5 0.3 m w XJ < 02

0.1

0 10 20 30 days

0.4

^ 0.3 m (0 S3 < 0.2

0.1

0 2 4 6 8 days

Figure 19. a) Stability of GSNO crystal at 0° (■) and 23°

(□); b) Stability of GSNO crystal at 23° with the exclusion of light and air and then, N^ flushing (o), with the exclusion of light (□), and with light and air (•). 68

100

80

o 6 a

40 -SH

20

4000 32002 4 0 0 1800 1400 1000 600 Wave Number (cm *)

Figure 20. The IR spectrum of GSNO crystal. The spectrum of GSH around 2540 cm”1 (-SH) is shown for the absence of sulfhydryl moiety in GSNO. Relative Intensity Relative Intensity 100 100 x20 90. SO. 70. 60. 28. 60. 30. 40. 90. 10. 20. 30. 40. 50. 60. 70. 80. 10. Oil -. - TfffrTT 40 45 l'fV |S11 (‘I’*' n T | n m 11 t it nagmn i son n h lwr part. lower the in shown is enlargement oeua wih rgo fo mz 1 t 40 ih 20-fold with 400 high to The 210 m/z from region crystal. weight GSNO of molecular spectrum mass FAB 21. Figure 57 0 80 £0 1111 l n 11 i it f n 229 3 20 5 *0 7 *0 *90 *80 270 *60 250 240 230 ■i !inl 0 10 4 160 140 ii i i | ii i i i i i i 120 i | iin iT i i i 'i i t i 100 n i i i i | m n A11 i i*/ 'A 1111 in n n n ...... 115 .ll .ill i i.nlil .137 ? mt - iniiif rTi irriii triiiiii iiiitrn r t i i i ii i iiiM i i i i r |t i ri'rnrii,iTii iT irrT i ii i u iMi'fi i n i n i i 77 ’ ii nf.3 li.i 8 20 2 20 6 2! *60 240 *20 200 180 .. 185 ip m/z m/z . .ili 299 li.p.il *00 207 03 l l13 lJ1 1 1 .li1,3 „l J ll 1 *20 *10 *30 38 (MH l-H)* 137 P 1 4 350 340 7 2?38 2.3 359 329.337 308 29? 277 3 06 2 *0 360 *40 328 006 238 11111 < ■ I*! I I I■ l 111II i'l ■ 11 11 l i IT »111 I I 11*13 n 1 r f r iJ31 I :59 1 f f ! s 0 3 30 1 330 ^30 70

IIIIMITlp l T I M I I II I 0 410 400 400 «5 O' 70

ions from involatile and thermally labile substances. FAB permits the analysis nanomole amounts of GSNO without the need for prior chemical characterization which may destroy sensitive functional groups. The FAB mass spectrum of

GSNO crystal (Figure 21) dissolved in glycerol exhibited an abundant molecular ion (M + H)+ at m/z 337, sodium cation molecular ion (M + Na)+ at m/z 359, and (M + H - N0)+ at m/z 307 as a characteristic fragment ion. The spectrum also contained numerous intense peaks related to oligomers of glycerol and its related sodium cation ions, which are normal background for FAB. This is the first MS data for

a S-nitrosothiol.

Chemical reactions of S-nitrosothiols.

Homolvsis: GSNO and other S-nitrosothiols have been

shown to undergo several different reactions. Homolysis

of GSNO produced GSSG as a soluble product and NO and N^O

as gaseous products. The reaction was followed by reverse

phase HPLC (Figure 22). An increase in GSSG and

proportionate decrease in GSNO was observed during the time

courses of reaction. Gas products were analyzed by GC/MS

(Figure 23). The predominant peak at pH 7.4 was found at

m/z 30 and was assigned to nitric oxide. A small peak was

observed with a m/z of 44 corresponding to nitrous oxide.

The stability of S-nitrosothiols in solution varied

markedly depending on pH, temperature and concentration. 71

GSNO

GSSG

r V__ V LJ I 0 4 8 12 0 4 8 12 12 00 4 4 8 8 12 12 0 4 8 12

RT, min

0 hr 10 hr 24 hr 72 hr

Figure 22. HPLC of products from the homolysis of 100 mM ,o GSNO at pH 7.4 and 37 after 0, 10, 24, and 72 hr incubation. A gradient from 0.1% H^PO^ to 0.1%

HgPO^/acetonitrile (1:1 by vol.) was used as the solvent system. Other conditions are given in text. 72

pH7.4

O DC

scan 100 200 300 4 0 0 500 time 1:40 3:20 5:00 6:40 8:20

M* M

ft

17 I.. JL l^ TT" n a n

Figure 23. GC/MS of gas product from the homolysis of 50 mM

GSNO at pH 7.4 and 37° in the septum-capped vial filled with He gas. Mass spectra for peaks a and b are shown with m/z 30 and 44 as a molecular ion, respectively. 73

Although formation of a thlyl radical and nitric oxide has been reported during the photolysis of thionitrites

(S-nitrosothiols) (Barrett et aJL. , 1965; Barrett et a l . ,

1966) exclusion of ordinary light had no noticeable effect on the rate of reaction. The reaction proceeded with second order reaction kinetics under physiological conditions. Table 6 lists the second order rate constants, k2, for various S-nitrosothiols.

S-Nitrosocaptopril and GSNO were two most stable

S-nitrosothiols among the thirteen 13 S-nitrosothiols studied. The stability of 20 mM GSNO at pH 7.4 at different temperatures is shown in Figure 24. In the case of GSNO and moderately stable S-nitrosothiols, stabilities at pH 7.4 were similar to those at acidic pH but decreased at alkaline pH values. The reaction at alkaline pH as determined from the decrease in GSNO, was neither first order nor second order. The data in Figure 25 illustrates that the influence of pH on the decomposition of GSNO at the pH range of 7.4-9.3. At higher pH values, GSNO produced more N02 and N20 gas than lower pH (Figures 26 and 27). The ratio of NO/NgO from the decomposition of

GSNO at pH 3.0, 7.4 and 9.5 was 29.3, 9.0 and 0.2, respectively. Even S-nitrosocysteine was stable for several hours at pH 1.5. However, when the solution of

S-nitrosocysteine was adjusted to pH 7.4, a vigrous evolution of gas set in within several minutes, accompanied 74

Table 6. Homolysis of 20 mM S-Nitrosothiols at pH 7.4 and

S-Nitrosothiol k 2 (sec "*■)

S-Nitrosoglutathione 0.00030

S-Nitrosocysteine 0.11

S-Nitrosocysteine ethyl ester 0.043

S-Nitroso-N-acetylcysteine 0.0019

S-Nitrosocysteamine 0.30

S-Nitrosohomocysteine 0.0018

S-Nitroso-p-mercaptoethanol 0.0016

S-Nitroso-3-mercaptopropionic acid 0.0042

S-Nitrosothioglycolic acid 0.0036

S-Nitrosodithiothreitol 0 . 2 9 * 0.44

S-Nitrosocaptopril 0.000094

S-Nitrosopenicillamine 0.10

S-Nitroso-N-acetylpenicillamine 0.012

a) [SNP]/[Dithiothreitol]=1; b) [SNP]/CDithiothreitol3=2 75

. 2 -

0 10 20 Time, hr

Figure 24. Effect of temperature on the stability of GSNO at pH 7.4. Changes in the absorbance at 544 nm were determined at 4° (•), 23° (o), and 37° (□) with 20 mM GSNO. 76

100

80

E 60

CO 40

20

0 2 3 4 Time, hr

Figure 25. Effect of pH on the decomposition of GSNO at

37°. GSNO (50 mM) at pH 7.4 (o), 8.0 (•>, 8.5 (®), 9.0

(■), and 9.3 (— •--) was followed by changes in absorbance at 544 nm. 77

■D O O 3 ■o o 0 . dz 2 E

20 40 60 Time, min

Figure 26. Effects of pH and time on the formation of NO, upon the decomposition of GSNO, GSNO (50 mM) was incubated at 37 and pH 7.4 (•), 8.6 (o), 9.5 (□) and

N0^ formation was determined as described in methods. iue 7 G/S f a pout fo te cmoiin of ecomposition d the from products gas of GC/MS 27. Figure H ., ., n 9.5. and 7.4, 3.0, pH 0 M SO t 7 i sptum-capped va fle wt H at He with filled vial d e p p a c - m u t sep in 37° at GSNO mM 50

RIC RIC RIC 3.0 H p .4 7 H p scan .5 9 H p time 40 :4 1 100 tNO 20 :2 3 200 00 :0 5 0 0 3 40 :4 6 0 0 4 n 2 o 8:20 0 0 5 78

<3J O Js * 0! Im ©

-o

0.5,

JQmin

4OT §00 " 600 W avelength, nm Figure mi* t u r e 28. Spectral changes with time in the reaction

of 20 mM GSNO w i t h 100 mM GSH at pH 7.4, 23°. 80 by fading of red color.

Heterolvsis: The reaction of GSNO and GSH was accompanied by a decrease in at 544 nm (Figure 28).

This reaction followed the pseudo-first order kinetics in the presence of excess GSH. Figure 29 shows a pH dependence for the reaction. The reaction rate reached a maximum at pH values above 8.0. The soluble product of this reaction was identified as GSSG by reverse phase HPLC.

Identification of the gas product was conducted by GC/MS and only NgO was detected. Reaction rates for different

S-nitrosothiols with their related thiols varied considerably (Table 7).

The reaction of GSNO with cysteine gave GSSG, cystine and the mixed disulfide as products. The HPLC eluants corresponding to each were loaded onto the amino acid analyzer with and without hydrolysis. Three different peaks were observed similar to the HPLC chromatogram

(Figure 30). Standard GSH, GSSG, GSNO, cystine and the mixed disulfide emerged from the column of the amino acid analyzer as distinct peaks. The major products of the reaction of GSNO with cysteine eluted in the position of

GSSG, cystine and the mixed disulfide suggesting a complex reaction. The mixed disulfide which had median retention time between cystine and GSSG was collected and hydrolyzed and its identity confirmed by hydrolysis followed by amino acid analysis. 81

OJ T- 2

4 6 8 pH

Figure 29. pH-rate profile of the reaction of 10 mM GSNO with 100 mM GSH at 23°. 82

Table 7. The Reaction of Thiols with S-Nitrsothiolsa .

RSH S-Nitrosothiol k (min

Glutathione S-Nitrosoglutathione 0.0382

N-Acetylcysteine S-Nitroso-N-acetylcysteine 0.0801

Homocysteine S-Nitrosohomocysteine 0.461

(J-Mercaptoethanol S-Nitroso-(3-Mercaptoethanol 0. 245

3-Mercaptopropionic S-Nitroso-3-mercaptopropionic 0.333 acid acid

a) Reactions were conducted with 100 mM thiol and 5 mM

S-nitrosothiol at pH 7.4 and 23°. 83

a570 GSH

GSSG

GSScys

GSNO GSSG

cystine

cystine

GSNO + cysteine

GSNO

0 4 8 0 20 40 RT, min RT, min b

Figure 30. a) HPLC of reaction mixture from 2 mM GSNO and

2 mM cysteine at pH 7.4, 23°. A gradient from 0.1% HgPO^ to 0.1% HgPO^/acetonitrile (1:1 by vol.) was used as the solvent system. Other conditions are given in text. b) Chromatographic analysis of the standards and reaction mixture described above on the column of amino acid a n a l y z e r . 84

S-Nitroso-HSA also reacted with GSH relatively slowly.

With 0.5 mM S-nitroso-HSA and 10 mM GSH at pH 7.4 and 23°/ -4 the pseudo-first order rate constant, k, was 4.28 x 10 min *. The slow rates might be the result of certain degree of steric inaccessibility of GSH to the buried

S-nitrosothiol group on the HSA.

Transnitrosation: Transfer of the nitroso function from S-nitrosothiols to thiols proceeded very rapidly at pH

7.4 at 23°. The transnitrosation from GSNO to

N-acetylcysteine, cysteine, N-acetylpenicillamine and

@-mercaptoethanol clearly showed that rapid transfer of NO group to thiols. The degree of transnitrosation depended on the concentration of thiols. Figure 31 shows the extent of transnitrosation from 1 mM GSNO to

N-acetylcysteine in the range of 0.2-1.0 mM of

N-acetylcysteine. The reaction was reversible and the equilibrium constant, K , for each thiol varies considerably (Table 8). Transfer of the nitroso groups of

S-nitroso-N-acetylcysteine and S-nitroso-N- acetylpenicillamine to GSH was also observed. Figure 32 illustrates the shift of from 588 to 544 nm upon max transfer of the NO-moiety from S-nitroso-N-acetyl- penicillamine to GSH. The transnitrosation reaction showed pH dependence and at neutral and alkaline pH, the degree of transnitrosation was more significant than at acidic pH (Figure 33). 85

a b c

1

0 16 0 8 16 o 8 16 RT, min .

Figure 31. HPLC profiles of transnitrosation reaction between 1 mM GSNO and a) 0.2 mM, b) 0.5 mM, and c)l mM of

N-acetylcysteine at pH 7.4, 23°. A gradient from 0.1%

H„P0. to 0.1% H„P0./acetonitrile <1:1 by vol.) was used as 3 4 3 4 J the solvent system. Other conditions are given in text.

The peaks are marked as follows: 1. GSNO; 2. S-Nitroso-N- acetylcysteine. Table 8. Equilibria for the Transnitrosation from

GSNO to Various Thiols3 .

Thiol K ______? 9 .

N-Acetylcysteine 11.2

N-Acetylpenicillamine 0.0208

Cysteine 1.01

(J-Mercaptoethanol 3.05 87

Abs

0.2

0.1

0

0.2

0.1

5 0 0 6 0 0 5 0 0 600 Wavelength, nm

Figure 32. VIS spectra of 20 mM S-nitroso-N- acetylpenicillamine (a); transnitrosation reaction between

20 mM S-nitroso-N-acetylpenicillamine and 4 mM (b) 10 mM

GSH (c); 20 mM GSNO (d). 88

100 c o

c

c 2 40

20

pH

Figure 33. Dependence of pH on the degree of transnitrosation from the 0.5 mM GSN0 to 0.5 mM

N-acetylcysteine at 23° after mixing. Degree of transnitrosation was measured by amounts of S-nitroso-N- acetylcysteine produced by using HPLC. 89

Reaction with protein thiola: The reaction of GSNO with protein thiols was examined with ADH and HSA. Figure

34a shows time courses of inactivation of ADH at GSNO concentrations of 0.3 and 0.6 mM. Inactivation proceeds very rapidly with no apparent lag. Within 1 min, one- half of the enzyme activity was lost at 0.6 mM of GSNO, and little or no activity was apparent after 3 min. GSNO was also found to inactivate ADH in a concentration dependent manner (Figure 34b).

The reaction of 1 mM of HSA with 2.5 mM of GSNO at pH

7.4 generated decrease in absorbance at 544 nm. After

Sephadex G-25 chromatography of the reaction mixture, the free thiol content of HSA was determined by Ellman's reagent. During the reaction, the available thiol group decreased with time (Figure 35).

Reaction with hemeprotein: Figures 36 and 37 show the visible spectra for 25 pM of deoxyHb, 25 pM of deoxyHb with

200 pM of GSNO, 25 pM of oxyHb, and 25 pM of oxyHb, excess amount of sodium dithionite with 200 pM of GSNO at pH 7.4 at 23° in the region of 350-650 nm. The hemoglobin derivatives proceeded in both reactions displayed a characteristic shift in the Soret absorbance maxima from

409 nm for oxyHb and 430 nm for deoxyHb to 418 nm which is characteristic peak for nitrosyl hemoglobin (NO-Hb).

The reactions occurred almost immediately. 90

100 100

80 80

> > o 60 o 60 < < X X Q o < 40 < 40 8P

20 20

80 120 1.2 2.4 Time, sec GSNO , mM

Figure 34. a) Time courses of inactivation of ADH at GSNO concentrations of 0.3 (•) and 0.6

ADH was incubated with GSNO at 23°. for 45 sec. Activities were determined by the method described in text. 0.2

0.1

600 400 200 0 Time, sec

Figure 35. Increase of absorbance at 412 nm on addition

DTNB to 1 mM HSA <----- ) and 1 mM HSA plus 10 mM GSNO at

7.4 and 23° for 1 hr incubation (-----) after G-25 chromatography. 92

Abs

0.8

>

0.4

400 450 500 600 Wavelength, nm

Figure 36. VIS spectra of 25 jjM deoxyHb (---- ) and 25 jjM deoxyHb with 200 uM GSNO at 23° in 0.05 M phosphate buffer, pH 7.4 (----- ). For 450-350 nm range, samples were diluted

10 times. 93

Abs

0.8

0.4

400 450 500 600 Wavelength, nm

Figure 37. VIS spectra of 25 pM oxyHb (-----) and 25 pM oxyHb plus excess dithionite with 200 pM GSNO at 23° in

0.05 M phosphate buffer, pH 7.4 (----- ). For 450-350 nm range, samples were diluted 10 times. 94

Animal experiments.

Studies on the anesthesized dogs and monkeys showed

that GSNO significantly lowered blood pressure in each of

animals to about the same extent as SNP alone and that the onset and dissipation of its effect were essentially the same as obtained with SNP. Intravenous administration of

GSNO or SNP over a wide range of doses decreased systematic arterial pressure in a dose-related fashion. Figures 38

and 39 show the dose-response curves for dogs and monkeys

obtained upon administration of SNP and GSNO. The blood

pressure decrease in the cause of SNP and GSNO is greater

in monkeys than dogs. When 20pg/kg/min of SNP or GSNO

was administered, mean arterial blood pressure decreased

35-40 torr for dogs and around 45 torr for monkeys. 95

d) S3 so 0> t- Q.

cO

40 Dose (Mg/kg/min)

Figure 38. Decrease on the mean arterial blood pressure of

monkeys elicited by infusion of SNP (•) and GSNO (o) with

doses in the range of 0-40 jig/kg/min. © 1_ m 80 to 0 u CL

0 V. 0 60 •*-* i_ < c 0 0 2 40

50 Dose (pg/kg/min)

Figure 39. Decrease on the mean arterial blood pressure dogs elicited by infusion of SNP (•) and GSNO (O) with doses in the range of 0-50 pg/kg/min. DISCUSSION

Although cyanide toxicity has been considered a major problem in the clinical use of SNP as a hypotensive drug,

N-nitrosation reaction of SNP with secondary amines may impose on even more serious health hazard. N-Nitrosamines

(nitrosodialkylamines) have been recognized as hepatoxic agents in rats (Knowles et. al., 1974; Wishnok and Archer,

1976) and have subsequently been found to be potent in a wide range of organs and animal species

(Mirvish et al., 1980; Chu and Magee, 1981). Although evidence for the carcinogenic activity of N-nitroso compounds in man is lacking, it is generally assumed that this class of compounds is active in man as well, since some 80% of all the N-nitroso compounds tested in many animal species have been shown to induce tumors (Ohshima and Bartsch, 1981). The details of the biological

interactions through which nitrosamines initiate tumors are unknown, but the most widely quoted hypotheses are generally based on the modification of intracellular DNA through alkylation by an electrophilic metabolite of the nitrosamines (Mirvish et al.., 1980; Chu and Magee, 1981).

97 98

N-Nitrosomorpholine is acutely toxic even as a single oral dose to an adult male rat, with the of

N-nitrosomorpholine as 282 mg/kg (Druckrey et al,., 1961).

Blood vessels are the tumor site in the case of

N-nitrosomorpholine (Newberne and Shank, 1973). Because several widely used drugs contain morpholino moieties, it can be assumed that these drugs also might produce

N-nitrosamines upon reaction with SNP.

There has been a report (Greenblatt and Mirvish, 1973) showing that significant numbers of lung adenomas in mice are induced by nitrosopiperazine, and that

N,N'-dinitrosopiperazine is thought to be a stronger than the mononitroso derivative.

It is known that many less- and nonvolatile nitrosamines are as carcinogenic as the volatile nitrosamines (Mirvish, 1975).

The formation of C-nitroso compounds could arise by initial nitrosation of available phenolic group of synephrine and phenylephrine yielding C-nitroso derivatives, which react further with the decomposition products of SNP. Therefore, C-nitroso derivatives could arise by SNP oxidation of C-nitroso derivatives. An analogous reaction is the nitration of phenols by dilute nitric acid at acidic pH which is thought to involve a two- stage mechanism involving initial nitrosation of the activated nucleus followed by nitric acid oxidation to the 99 nitrophenol. Alternatively/ aerial or nitrogen dioxide oxidation of the C-nitroso derivatives could give rise to the same products (Knowles et al.. / 1974). Although the toxicity of aromatic C-nitroso and nitro compounds have not been widely examined, several C-nitroso and nitro compounds are regarded as highly toxic chemicals. Because the nitrosation product of synephrine and phenylephrine induced by SNP contain C-nitro as well as N-nitroso moieties, it can be easily assumed that these compounds could be toxic and carcinogenic.

Although the weak carcinogenicity Clog (l/Dj.^) = <0.6, where D5Q is the total molar dose required to induce tumors in 50% of the test animals] of N-nitrosoproline has been reported (Wishnok and Archer, 1976), N-nitrosoproline apparently does not undergo metabolism in mammals nor does it alkylate cellular macromolecules. Therefore, it is considered to be nonmutagenic and noncarcinogenic and is excreted nearly quantitatively in the urine (Mirvish et. al., 1980; Chu and Magee, 1981; Ohshima and Bartsch, 1981;

Brunneman et al., 1983). Determination of urinary

N-nitrosoproline could therefore be used to determine the extent of N-nitrosation of amines by SNP in humans, and current results on the N-nitrosation of proline under physiological condition might provide some information to these studies. The acute LD5Q value for oral administration of a single dose N-nitrososarcosine in BD 100

rats has been reported to be 5000 mg/kg and has been shown

to induce of the esophagus in the rats (Drukerey et

al, 1967).

Although the concentration of epinephrine in blood and the dose of epinephrine as a drug are quite low, the nitrosation, probably oxidation too, of epinephrine could be of significance due to its important physiological

effects as a neurotransmitter.

Creatinine, the end product of creatine metabolism,

is found together with creatine in muscle tissues, milk and

blood. In view of the abundance of creatinine and

creatine in muscle tissues and blood, the intravenous

administration of SNP may cause possible hazard in terms of

N-nitroso and related compound production. The mechanism

of the reaction to produce creatinine-5-oxime from

creatinine and SNP remains obscure, though the imino group

in creatinine does seem implicated in its reactivity

(Archer et al., 1971), and the secondary amino group in

creatinine appears to be unreactive.

creatinine creatinine-l-oxime

Although the toxicity of creatinine-5-oxime is as yet 101 unknown, oximes and their derivatives have often been reported to be toxic. Any toxicity of oximes has been due to the oximino group (=N-0H) which can be transformed into hydroxylamine.

A possible mechanism for the nitrosation of creatine by SNP could be similar to the mechanism proposed for the reaction of nitrous acid with creatine (Archer et al.,

1971). Reaction of creatine (1) with SNP to yield

N-carbamyl-N-methylglycine (2) and then, this product with more SNP to yield N-carboxyl-N-methylglycine (3), which gives N-methylglycine (sarcosine) (4) as a decarboxylated product. Because sarcosine contains secondary amino group it can be converted convert to N-nitrososarcosine (5) by

SNP. Although N-nitrososarcosine is thought to be a weak

1 IV

C02H SNP •COjH CHj 5 4 carcinogen, its significance in the incidence of human cancer must be evaluated because of the relatively high concentration of creatine in blood and muscle tissues.

A relatively nonvolatile nitrosamine, N'-nitroso- nornicotine induces multiple pulmonary adenomas in mice 102 with local invasion of the lung and the bronchi (Boyland et al., 1964). Both nornicotine (secondary amine) and nicotine (tertiary amine) could serve equally as precursors for N '-nitrosonornicotine, both being inhaled as components of cigarette smoke (Hoffmann et a l.., 1974). Although nornicotine also can convert to N '-nitrosonornicotine, probably faster than nicotine because it is secondary amine, nicotine, is in considerably higher concentration in tobacco, and probably plays a more significant role. The reaction of nicotine with SNP may to proceed via the cyclic iminium salt similar to the mechanism proposed for the reaction of sodium nitrite with nicotine (Smith, 1966).

Casado et. al.. (1985) have studied the reactions of SNP with several simple secondary amines under alkaline conditions and observed complex rate equations, including both first and second order terms in respect to amine concentration. At low concentrations of amine, like those existing during the pharmacological use of SNP, only the first order term should be significant. Observed rates of nitrosation varied markedly between amines (Table

2), the highest rates of nitrosation were generally associated with structually less hindered compounds such as piperazine and morpholine. It is possible to postulate that the large differences in their reactivities, presumably, reflect differences in basicity, nucleophilicity and steric constraints. Additional 103 studies on a series of sterically comparable amines will be needed to evaluate the Individual Importance of those factors on reactivities. Although the SNP concentrations for kinetic studies are unreallstlcally high for pharmacological purposes, since the nitrosation follows first-order kinetics it should be possible to predict that even very low concentration of drugs and SNP might produce enough N-nitrosamines to cause concern. Actually, the concentration of amines and SNP for this kind of study does not depend on the possibility of nitrosation by interaction of two reactants but depends on the analytical limitation of N-nitrosamine detection.

As in the case with nitrous acid, the less basic amines are the most reactive under physiological conditions and therefore potentially the most dangerous (Mirvish et. al.. 1973; Mirvish, 1975). The importance of basicity is indicated by the strong influence of pH on reactivities.

In contrast to the reactions of nitrous acid with amines, however, reaction rates of N-nitrosating by SNP are increased at higher pH values (Figure 7). Although formaldehyde catalyzes the conversion of various secondary amines to nitrosamines in neutral and basic medium under somewhat drastic conditions (Keefer and Roller, 1973), it is generally assumed that potentially hazardous quantities of carcinogenic N-nitroso compounds are mainly produced by reaction of nitrite with amines under mild conditions in 104 acidic medium, usually optimal at about pH 3.0 to 3.4.

Unlike nitrite, SNP preferentially acts as nitrosating agent in neutral and basic media. This fact holds important implications for the nitrosation mechanism of

SNP. For the nitrite reaction, investigations on mechanisms have suggested that "the protonation of nitrous acid appears necessary for initiating all nitrosation reaction" (Ridd, 1961; Hawksworth and Hill, 1971).

Despite a qualitative parallel between the reaction of amines with nitrite and SNP, as an explanation for the dependence of the rate constant and yield on pH, it can be assumed that the reactive form of SNP as a nitrosating agent is different from that of nitrite. Because of the nitrosation reactivity of SNP at physiological pH, it can act as a nitrosating agent in various organs, tissues and b l o o d .

Ascorbate and cysteine have been shown to block the nitrosation of secondary amine by nitrite under acidic conditions in an in. vivo system, the actual concentration of cysteine and ascorbate might be quite low. Therefore, the most significant blocking agent for nitrosation by SNP under physiological condition may be 6SH because of its considerably higher concentration. A reaction between ascorbate and nitrous acid has been described by Dahn et. al♦ (1960) and the mechanism by which ascorbic acid blocks the N-nitrosation of amines by sodium nitrite was discussed 105 by Mirvish et al.. (1975). The blocking effect of ascorbate on the N-nitrosation of piperazine by SNP can be explained by the role of ascorbate as a scavanger of

SNP as similar as in the case of nitrous acid.

The red colored complex developing when SNP is added to thiols in alkaline solution is one of the oldest methods for their detection (Swinehart, 1967). The blockage of

N-nitrosation by glutathione and cysteine may, therefore/ be due to the competition for available SNP between thiols and secondary amines. The amount of nitrosamine produced during the pharmacological administration of SNP under physiological conditions and at the concentration of amines usually present might well be variable due to the presence of blocking agents. However, exposure to even extremely small amounts of carcinogen can not be considered insignificant. It is now the consensus of opinion that

N-nitrosamines should be considered as potential health hazards at concentrations of part per billion (ppb, pg/kg).

The amines examined here are only a few example of the many hundreds that could react similary. Many amine drugs are administered orally at considerably higher dosage and several drugs are administered intravenously, sometimes at the same time. In the latter case, the combination of SNP amine drug might occur in the infusion device as well as in the blood stream. Furthermore, the blood stream normally contains low concentrations of several secondary amines 106

(Neurath et. a l.., 1977). Overall/ it is quite possible that many drugs containing secondary and tertiary amine structures/ which exist in blood and tissues, resulted in the oral and intravenous administration, and physiological amines can produce carcinogenic N-nitrosamines by SNP under in vivo conditions. Therefore, along with cyanide toxicity, the possible carcinogenicity of SNP as a nitrosating agent for amines should be seriously considered.

There is no question that SNP is a reactive molecule which can modify many cellular components. It is less clear which of these modifications actually contribute to the blood pressure lowering effect of SNP.

The effect of SNP on the angiotensin-renin system was examined as a part of an attempt to elucidate the mechanism of the hypotensive action of SNP. Evidence for the existence of an essential lysine residue in ACE has been obtained by reacting ACE with reagents known to modify this amino acid residue such as acetic anhydride, diethyl pyrocarbonate, diketene (Bunning et. al.., 1978). ACE is reported to be unaffected by sulfhydryl reagents such as iodoacetate, iodoacetamide, and p-hydroxymecuribenzoate

(Yang et aT., 1970; Cushman and Cheung, 1971). Except for the thiol group of cysteine, <*- and £- amino groups are most sensitive to the action of SNP. Since ACE does not have any reactive cysteinyl residues, the present results 107 point to deamination of an essential lysyl residue as the most likely possibility. Because competitive inhibitors of ACE partially protect the modification by SNP (Table 4), the lysine residue modified by SNP may be located at or near active site. Our earlier results showed that the deamination of £-amino group of lysine by SNP took place in neutral or weakly basic solutions and several alcohol and olefin derivatives of lysine were identified from the reaction mixture (Monera et al., 1986). Another implication of the reaction of SNP with an amino group in

ACE is involvement of a short-lived and reactive intermediate. Because the deamination of amino groups generates a diazonium ion (Maltz et. al., 1971) prior to the formation of stable olefin and alcohol derivatives (Figure

40)/ it could be possible that diazonium ion complex plays a role in the relatively fast inactivation of ACE (Figure

10) .

A striking feature of ACE is its activation by chloride and other monovalent anions (Skeggs et al., 1954).

Weare (1982) has suggested that ACE has two anion binding sites distinct from the active site which indirectly regulate the activity of enzyme; one of the sites contains a functional lysyl residues that is protected from the modification by high concentrations of chloride.

Concurrent changes in the activity of ACE result from modification of an additional lysyl residue near the active 108

h -c -c h 2 -c h 2 -c h 2 -c h 2 -o h

Lysine + SNP h -c -c h 2 -c h 2 -c h -c h 3 1 OH

[H-C-(CH„)-NEN ] I 2 4 I 2 2 diazonium ion complex H-C-CH--CH=CH-CH_ I 2 3

plus small amounts of one alcohol and two olefin derivatives

Figure 40. Proposed mechanism for the deamination of

£-amino group of lysine by SNP. 109 site. The relatively high enzyme activity of ACE retained after modification by SNP in the presence of 0.3 M NaCl

(Table 4) might be explained by this hypothesis.

There is a report showing that whereas plasma renin activity was increased by SNP, plasma ACE activity as measured in the collected arterial blood after SNP treatment, was reduced throughout the time that SNP was administered (Woodside et al.., 1984). Treatment of renin with an amino group (both c(- and £-) specific reagent, ethyl acetimidate, has no significant effect on renin activity. The result, therefore, indicates that amino groups are not essential for renin activity (Misono and

Inagami, 1980). The increased renin release was believed to be a physiological response to the hypotension per se owing to activation of intrarenal baroreceptors and increased activity of the renal sympathetic nerves

(Marshall et. al., 1981), rather than a specific response to the hypotensive drug itself. Another possible explanation for the increased plasma renin activity is the partial absence of a negative feedback normally exerted by angiotensin II on renin release. A similar effect has been observed upon administering the antihypertensive drug, captopril, a competitive inhibitor of ACE (Vidt et. al.,

1981; Jennings et. al.. , 1981).

Although SNP can inactivate ACE to some extent, and can modify angiotensin I and II in. vitro at a relatively 110 slow rate, It Is unlikely that the renln-anglotensln system

is the major mechanism for the hypotensive action of SNP.

Because the reaction between SNP and ACE is relatively fast there is still some possibility that this reaction is a supplementary mechanism for the pharmacological action of

SNP. Since the actual hypotensive action of SNP is almost

instantaneous, the mechanism must involve a very rapid reaction. Because physiological systems contain large amounts of thiols, e.g. GSH, HSA, cysteine (Table 9) and the rate of reaction between some thiols and SNP has been known to be very fast (Johnson and Wilkins, 1984), this reaction could be the most important reaction in. vivo.

Ignarro et al.. (1980) have suggested that

S-nitrosocysteine may be formed from SNP and cysteine.

They have suggested that NO may be spontaneously released from the SNP and that NO may then react with cysteine to give S-nitrosocysteine. They have not, however, presented any evidence for the production of NO from SNP. SNP has been considered stable up to pH 12.7 (Swinehart, 1967), and it has been reported that a small amount of NO is not reactive enough to produce S-nitrosocysteine from cysteine

(Prior et al.., 1982). The red colored complex formed upon the reaction of SNP with GSH appears to be

3 - CFe(CN) J.N0SG3 . The mechanism of GSNO production may involve simple dissociation and ligand exchange as follows: Table 9. Thiols in Human Blood (Extracellular)3 .

Thiol mM/liter

Glutathione (reduced) 4.5 + 0.7

Serum Albumin 0.57 - 0.79

Cysteine 0.09 - 0.12

Other <0.1

a) Geigy Scientific Tables, vol 3, 8th ed. (1984). 112

CFe(CN) NOSG] + HO a V ------> [Fe(CN)5OH3 + GSNO + H+

Figure 41 for the reaction of SNP and GSH can be proposed on the basis the data of Morando et. al.. (1981) and our results. When the [GSH]/[SNP] ratio is low and in the presense of air, the formation of GSNO is not noticeable

(Figure 12), but when excess amount of GSH is present, GSNO formation is significant (Figure 14). At equimolar amounts of GSH and SNP, detectable amounts of GSNO were produced under anaerobic condition. At very high ratios of CGSH3/[SNP3 as in an in. vivo system, traces of oxygen are rapidly consumed by the systems and relatively anaerobic conditions are estabilished. Therefore, it can be easily imagined that the yield of GSNO from the reaction of SNP with GSH under in vivo condition may be much higher.

Monomeric serum albumin is 60-70% mercaptoalbumin containing 1 SH per mol and 30-40% nonmercaptoalbumin, which is a mixed disulfide between cysteine or GSH and albumin (Hughes et. al., 1949; Anderson, 1966). On the basis of the effect of urea on the reaction of SNP and HSA, it is quite clear that the reaction of SNP with a protein thiol is strongly influenced by the site in which the thiol resides. This appears to indicate the inaccessibility of

SNP for the thiol site of HSA which is not exposed outside. GSH, OH' [Fe( CN) 5( NO)] [Fe( CN) 5( NO) ( SG)] 3 ( red)

d eca y d eca y OH' iSSG SNO

4- [Fe( CN) 5( NOz)] [Fe( CN) 5( NO)] [Fe( CN) 5OH] 4‘

Figure 41. Proposed mechanism for the reaction of GSH with 114

Because the reaction between GSH and SNP is extremely fast, and GSH is the most abundant thiol in most physiological systems, it can be postulated that GSNO could be an active species for the SNP as a hypotensive drug.

Furthermore, there has been a report suggesting that another major decay product of the red colored complex from the reaction of SNP and thiols, [Fe(CN)gNO]^ and disulfide failed to affect activity of guanylate cyclase, (Craven and

DeRubertis, 1983), a key enzyme for the blood pressure lowering mechanism of nitrovasodilators.

Solutions of S-nitrosothiols have been easily prepared by the reactions of thiols and sodium nitrite at acidic pH.

Until now, only S-nitroso-N-acetylpenicillamine crystals have been isolated and had their X-ray structure established (Field et. al.., 1978). Crystals of other

S-nitrosothiols have been reported to be highly unstable in air (Ignarro et al, 1980), however, crystalline of

GSNO is much more stable than aqueous solution of GSNO

(Figure 19). Because self-decomposition (homolysis) of

S-nitrosothiols follows second order kinetics, highly concentrated solutions, as such may be obtained by evaporation of solvent, undergo rapid decomposition.

However, the crystalline form of GSNO was recovered with more than a 90% yield by the method described.

Contrary to other reports, some S-nitrosothiols are quite stable. As many S-nitrosothiols are unstable, and 115 all of them are unstable under some conditions, they have generally been considered to be very unstable. This, however, is not correct. Our results have shown that GSNO and some other S-nitrosothiols are quite stable under physiological conditions. S-Nitrosocaptopril and GSNO are two of the most stable compounds and S-nitrosocysteine and some related derivatives are the least stable of the thirteen S-nitrosothiols studied under physiological conditions (Table 6). Generally, a free amino group near the decomposing RS-N=0 bond accelerates the decomposition markedly, as shown by the values of k ^ for

S-nitrosocysteine, S-nitrosocysteamine, and

S-nitrosocysteine ethyl ester. In the case of S-nitroso-

N-acetylcysteine, where the amino group is blocked by an acetyl moiety, the rate of homolysis is one or two orders of magnitude slower than for S-nitrosocysteine and its amino group free derivatives. With GSNO, where the amino group is probably some distance away in the tripeptide, the free amino group seems to have little effect on the rate of homolysis. The reaction rate for most S-nitrosothiols without a free amino group is moderately slow, and that might support the above idea. On the basis of kinetic data (Table 6) and product analysis (Figures 22 and 23), the reaction mechanism of homolysis at acidic and neutral pH can be postulated as follows: 116

2 GSNO ------* GSSG + 2NO VI

Although previous reports (Schultz et. al.. , 1969; Prior et.

al., 1982) for S-nitrosocysteine did not detect NO as a gaseous product, NO was positively identified as a

homolytic product of GSNO.

Based on the results from the N02 determination

(Figure 26) and gas analysis (Figure 27) the reaction

mechanism for the breakdown of GSNO at different pH is

postulated as follows:

acidic and neutral pH GSNO ------► 1/2 GSSG + 1/2 NO

0 H ' - GSNO ---- » GS + N0„ + H

GSNO,

VII

[ GSS GNO” ] -- » GSSG + HNO H

HNO — (1/2 H2N202) 1/2 N20 + 1/2 H20

Salts of hyponitrous acid, HON=NOH, are known to decompose

to N20 and H20 (Addison and Lewis, 1955) . At acidic and

neutral pH the predominant reaction for the breakdown of

GSNO is homolysis, but both homolysis and heterolysis

reactions occur at alkaline pH.

Although there have been some reports of

S-nitosothiols as intermediate in the oxidation of some 117 aromatic and thiols by nitrogen oxides (Oae et al.,

1977a; 1977b), there are no reports on the reaction of GSNO and other S-nitrosothiols with GSH or other thiols. The formation of symmetric or unsymmetric disulfides from the heterolysis of S-nitrosothiols has been known to proceed, via the nucleophilic attacks of thiols, to S-nitrosothiols, as shown below (Oae et. a l.., 1977b);

+ R-S-NO + R'SH — [R-S-S-R'N0~ 3 —* RSSR' + HNO VIII H

Because HNO is unstable it readily converts to the gaseous product, N20. In the reaction of GSNO with GSH, nitrous oxide and GSSG were identified as the only products, consistent with the proposed mechanism.

The rates of heterolysis for several S-nitrosothiols and related thiols (Table 7) varied due to differing structural restrictions on these reactions. The reaction of GSNO with cysteine produced both symmetrical and unsymmetrical disulfides as products, i.e., GSSG, GSSCys and cystine. Examination of this reaction showed that

GSNO rapidly exchanges the NO moiety with cysteine to give

S-nitrosocysteine and GSH, to account for the its observed mixture of products.

S-Nitrosothiols can act as a source of the nitrosyl moiety by transferring their nitroso-group directly to other thiols (Transnitrosation). Although this reaction 118 has never been reported, this reaction Is similar to the direct transnltrosatlon reactions between nltrosamlne with secondary amine. Challls et al.. (1973) proposed that

N-nitrosodiphenylamine may transfer the nitroso function directly to other amines. Secondary amines, like other other bases, can act as carriers of the nitroso moiety.

Transnltrosatlon of the nitroso function from GSNO to GSH

and other thiols may occur as follows:

GSNO + RSH , > GSH + RSNO IX

The degree of transnltrosatlon depends on the structure of

the S-nitrosothiols and thiols (Table 8). In conclusion,

GSNO may be involved in several different reactions.

Among three reactions, transnitrosation is the fastest and

homolysis is the slowest.

On the basis of the chemical reactions of GSNO, the

possibility of reaction between GSNO and thiol group-

containing and proteins, along with hemoprotein,

has been studied. Yeast ADH (EC 1.1.1.1) is a tetramer

made up of four identical or very similar subunits. Each

subunit has two active site sulfhydryl groups (Harris,

1964). The first event of inactivation of ADH by GSNO may

be of a sulfhydryl of GSNO through

transnitrosation reaction to give a protein-bound nitroso

group. Another posslbilty is that the protein-SNO is

capable of reacting further with GSH to form mixed 119 disulfides. The thiol group in albumin is in a sterically restricted environment that has hydrophobic character

(Cornell and Kaplan, 1978a; 1978b), and there is also evidence for the existence of a nearby negatively charged group (Wilson et. al.., 1980). The reaction of GSNO but not

SNP with HSA may be an indication of the different accessibility of these compounds in a hydrophobic environment. It can be easily postulated that many proteins which contain active site thiols may also react with GSNO with similar manner.

The reaction of hemoglobin with NO has been known for many years. Solutions of oxyHb, when converted to deoxyHb by evacuation followed by mixing with nitric oxide, turn distinctly red and show two wide and somewhat diffuse absorption bands at 574.5 and 536 nm. The same result is obtained in the presence of NO and sodium dithionite. The chemical species obtained in each case is referred to as nitrosyl hemoglobin (NO-Hb) and is considered to consist of homoglobin coordinated to an NO moiety (Keilin and Hartree,

1937). There are reports showing that a mixture of oxyHb, dithionite and either nitrite or SNP also result in spectral shifts to those typical for NO-Hb. The reaction with SNP required longer incubation times (Smith and

Kruszyna, 1974). We have observed a similar reactions of hemoglobin with GSNO. GSNO almost immediately transfers its NO moiety to or oxyHb in the presence of dithionite 120

(Figures 36 and 37).

On the basis o£ chemical reactions of GSNO and other

S-nitrosothiols/ a biochemical and pharmacological mechanisms for the pharmacological effects of these compounds can be postulated. In physiological systems, transnitrosation is the most rapid chemical reaction.

Although homolysis of GSNO to produce NO gas, which has been considered the active species involved in lowering blood pressure and in converting the heme moiety of guanylate cyclase to NO-heme, as a major product this reaction may be slow to account in terms of fast onset of

SNP. Whereas S-nitrosocysteine is one of the most unstable S-nitrosothiols at higher concentration, it is much more stable at the low concentration existing during its pharmacological use and its decomposition by that process is too slow to account for the rapid pharmacological effect of SNP. A higher ratio of

CGSH]/CGSN0] may be produced under in. vivo condition, therefore, heterolysis and transnitrosation have better possibility to involve in an in. vivo system. But heterolysis produces N£0 gas as product, and this gas has has been known to be chemically inert (Cotton and

Wilkinson, 1983) and has no significant effect on blood pressure. Therefore, transnitrosation may be the most important reaction in terms of the pharmacological action of SNP via GSNO. Transnitrosation from GSNO to other 121 physiological acceptors (e.g. thiol group-containing proteins and enzymes, heme moiety of guanylate cyclase and secondary amines) may be also possible. There is a possibility that S-nitrosothiols also can act as nitrosating agents. Although Oae et a^. (1977) described the possibility of this reaction, they did not pursue that reaction further.

R'R"-NH + RSNO J=± R'R"-N-NO RSH X

The reaction of piperazine and GSNO is one or two magnitudes slower than that of piperazine and SNP. And transnitrosation from GSNO to piperazine was much slower than to GSH. Because of the excess amount of GSH in physiological systems, N-nitrosamine formation may only be a secondary possibility in this case, a fact which substantially decreases the likelihood of nitrosamine formation during administration of SNP.

Direct evidence for a role of the NO-heme complex in the activation of guanylate cyclase was obtained from studies with preformed NO-Hb generated by reaction of NO gas with hemoglobin (Craven and DeRubertis, 1978; 1979;

1983). Therefore, the effect of GSNO on the formation of

NO-Hb from hemoglobin was used as a model for reaction of

GSNO with the heme-moiety of guanylate cyclase under physiological conditions. If GSNO reacts with hemoglobin to produce NO-Hb, this may be taken as indirect evidence 122 for the biochemical mechanism of guanylate cyclase activation. The addition of GSNO to hemoglobin instantaneously produced NO-Hb (Figures 36 and 37) therefore, S-nitrosothiols might transfer NO group to enzyme-bound heme moiety to generate the NO-heme adduct of guanylate cyclase. Studies by Ignarro et al (1982) have suggested that the binding of NO to heme, which results in displacement of the iron out of the plane of prophyrin ring

(Scheidl and Frisse, 1975; Perutz et. al., 1976; John and

Waterman, 1980) resulted in a moiety which resembles protoporphyrin IX, a recently identified activator of guanylate cyclase. Heme was not required for enzyme activation by protoporphyrin IX. Activation of guanylate cyclase by NO-heme may, in turn, be the result of the formation of protoporphyrin IX or a protoporphyrin IX-like binding interaction from the NO-heme complex (Ignarro et al., 1982b).

Several reports show that modification of the thiol group of guanylate cyclase induces inhibition of enzyme activity (Brandwein et al., 1981; Kamisaki et al.., 1986).

Therefore, it is quite possible that GSNO can modify the same site. But, practically, GSNO may modify the heme moiety, which will activate the enzyme, or thiol groups, which results in inhibition of enzyme. Therefore, the reaction of guanylate cyclase with GSNO for the modification of thiol group will complicate the 123 interpretation of the exact effect of GSNO on the thiol group. Thus another thiol group-containing enzyme or protein might be model for the thiol modification effect of

GSNO. The observation of GSNO on ADH and HSA suggested that the accessibility of GSNO to the thiol site of the protein may be an important factor. Because of reports that modification of a thiol group on guanylate cyclase almost always results in inactivation of this enzyme, it is possible that this reaction may involve reversal of guanylate cyclase activation by nitrovasodilators. Barry et al. (1979) revealed that large amounts of NO elicited less activation of guanylate cyclase than did smaller amounts, and dithiothreitol prevented the decrease in activation by the larger amount of NO. This may indicate that the heme moiety of guanylate cyclase is much more sensitive than the thiol site, and that the thiol site may be buried inside the enzyme in the native state, and then at high concentrations of NO, the thiol moiety is also modified to a larger extent. Activation of soluble guanylate cyclase by SNP results in an increased availability for mixed disulfide formation (Kamisaki et al., 1986) which indirectly shows possible structural change by SNP via NO-heme. Activation of soluble guanylate cyclase results in a change in active site thiols, rendering them more available for interaction.

This alteration might involve a conformational change in 124

protein structure, making these thiols more physically

accessible or altering their chemical characteristics, thus

making them more highly reactive. It is possible that the

predominant role of active site thiol of guanylate cyclase

is responsible for basal enzyme activity, while the heme moiety, which does not effect basal activity, markedly

enhances the degree of activation. Therefore, activation of guanylate cyclase by NO-heme may be due primarily to

reaction of GSNO, and inhibition by reaction with thiol

groups may be a secondary reaction, which is accelerated by

structural alteration induced by NO-heme, and involves

reversal of activation.

As the rapidly formed, first intermediate in the pharmacological mechanism of SNP, GSNO should be at least

as effective as SNP, and should have the same rapid onset

and brief duration. A remarkable similarity between the

SNP and GSNO was observed with respect to the hemodynamic

effects of two compounds in the dogs and monkeys (Figures

38 and 39). These are the first and only studies showing

the effects of GSNO on blood pressure in animals. Ignarro

et al.. (1981) have described a similar effect of three

other S-nitrosothiols including S-nitrosocysteine on the

blood pressure of anesthesized cats. Although their

pharmacological mechanism is based on the instability of

S-nitrosocysteine, the other moderately stable

S-nitrosothiols showed similar effect on blood pressure. 125

Furthermore, our results show that GSNO Is probably even more potent than S-nitrosocysteine to decrease blood pressure. Neverthless, GSNO is around 400-fold more stable than S-nitrosocysteine at physiological conditions.

The concentration of GSNO used for animal experiment is actually quite low, and at that concentration, GSNO may be very stable. The calculated half-life at that concentration of GSNO at pH 7.4 and 37° is several hundred days, and it will be much longer in the refrigerator.

Therefore, this result indirectly shows that homolysis of

S-nitrosothiols is not a major pharmacological mechanism and transnitrosation may be involved for the pharmacological action of S-nitrosothiols. Under physiological conditions, the transfer of NO moiety from

GSNO and other S-nitrosothiols to GSH would be extremely fast. The blood stream contains relatively high concentrations of GSH and even higher levels are present in most tissues. The introduction of other S-nitrosothiols into the blood stream might result in a transfer of the NO moiety to the heme of guanylate cyclase. Most of the NO moieties would probably be transfered to GSH to give GSNO and a corresponding thiol, and the resulting GSNO will then undergo reaction with the heme of guanylate cyclase. The administration of S-nitrosothiols other than GSNO should therefore lower blood pressure, very similar to GSNO.

In conclusion, GSNO may be an intermediate in the mechanism 126 by which SNP and probably several other nitrovasodilators lower blood pressure. The proposed mechanism is shown in

Figure 42.

The fast and brief action of SNP is of interest because the actual metabolism of the drug seems to be much slower. Smith and Kruszyna (1974) reported that the drug is metabolized relatively slowly, with a half-life of 1 to

2 hr in. vitro. It can be explained by that SNP can rapidly form complexes with thiols and produce GSNO and 4- CFe(CN)(.0H3 , which has no hypotensive effect and is subject to slow mechanism to produce cyanide ion.

The yield of GSNO from from reactions of GSH and sodium nitrite at different pH values (Figure 15) is consistent with weak hypotensive effects of sodium nitrite as compared to SNP. Whereas SNP markedly activates guanylate cyclase under physiological conditions, sodium nitrite causes little or no activation of guanylate cyclase at less than 10 mM (Katsuki et al. , 1977; Arnold et. al..

1977). This provides indirect evidence of involvement of

GSNO in the activation of guanylate cyclase by these nitrovasodilators. At pH 7.4, very little GSNO was produced from GSH and sodium nitrite.

Among the several S-nitrosothiols, S-nitrosocaptopril may have quite important clinical implications. The combination of captopril plus SNP to achieve a desireable therapeutic effect and to extend its administration period OH' -3 SNP + GSH ± [Fe( CN) 5( NO) ( SG)]

GSH -4 RSNO— *GSNO + [Fe( CN) 5OHj RSH

l_GC jhemej -> NO-lhernel I

GTP —*-»I cGMP + PPi

cGMP Kinase

P ro tein ------> (p)-Protein

(?)-M yosin I M yosin light chain light chain I

ContractionI R elaxation

Figure 42. Proposed mechanism for the pharmacological action of SNP via GSNO and other S-nitrosothiols. 128 has been tried. Several reports suggest slmultanous use of captopril and SNP (Jennings et aJL_., 1981; Woodside et al., 1984) may be used to reduce the dose requirement for

SNP induced hypotension, which is desireable in order to reduce the toxicity of SNP. If S-nitrosocaptopril is used as hypotensive drug it will play dual roles. Administered

S-nitrosocaptopril may act as a carrier for NO moiety and captopril which induced denitrosation of S-nitrosocaptopril by GSH may act as an inhibitor to ACE.

This study provides a molecular model for the physiological role of GSNO and other S-nitrosothiols along with an understanding of the possible carcinogenicity and pharmacological mechanism of SNP. The most significant aspect of this study is that it may lead to the development of a new class of potent, effective, but relatively nontoxic hypotensive drugs. The use of GSNO and other

S-nitrosothiols in place of SNP to eliminate or reduce side effects and risks associated with toxicity and carcinogenicity of SNP. LIST OF REFERENCES

Addison, C.C. and Lewis, J. (1955) Qurt. Rev. 9, 115.

Andersson, L.O. (1966) Biochim. Biophys. Acta 117, 115.

Archer, M.C., Clark, S.D., Thilly, J.E., and Tannenbaum, S.R. (1971) Science 174. 1341.

Arnold, W.P., Mittal, C.K., Katsuki, S., and Murad, F. (1977) Proc. Natl. Acad. Sci. USA 74., 3203.

Barnhart, E.R. (1985) Physicians Desk Reference, 39th Ed., Medical Economics Co. Inc., Ordall, New Jersey.

Barrett, J., Debenham, D.F., and Glauser, J. (1965) Chem. Commun. 248.

Barrett, J., Fitzgibbons, L.J., Glauser, J., Still, R.H., and Young, P.N.W. (1966) Nature 211, 848.

Barry, B.K., Gruetter, D.Y., Ohlstein, E.H., and Ignarro, L.J. (1979) Pharmacologist 21, 252.

Bell, F.K., O'Neill, J.J., and Bargison, R.M. (1963) J. Pharm. Sci. 52, 637.

Benitz, W.E., Malachowski, N., Cohen, B.A., Stevenson, D.K., Ariange, R.L., and Sunshine, P. (1985) J. Pediatr. 1 0 6 . 102.

Boyland, E., Roe, F.J.C., and Gorrod, J.W. (1964) Nature 2 0 2 , 1126.

Brandwein, H.J., Lewicki, J.A., and Murad, F. (1981) J. Biol. Chem. 256, 2958.

Braughler, J.M., Mittal, C.K., and Murad, F. (1979a) Proc. Natl. Acad. Sci. USA 76, 219.

Braughler, J.M., Mittal, C.K., and Murad, F. (1979b) J. Biol. Chem.254, 2450.

129 130

Brunneman, K.D., Scott, J.C., Hoffmann, D. (1983) J. Agr. Chem. 31. 905.

Bunning, P., Holmquist, B., and Riordan, J.F. (1978) Blochem. Blophys. Res. Commun. 83, 1442.

Casado, J., Mosquera, M., Rodriguez prieto, M.F., and Vasquez Tato, J. (1985) Ber. Bunsenges. Phys. Chem. 89. 735.

Challis, B.C. and Osborne, M.R. (1973) J. Chem. Soc. Perkin II 1526.

Chu, C. and Magee, P.N. (1981) Cancer Res. 41, 3653.

Cornell, C.N. and Kaplan, L.J. (1978a) Biochemistry 17, 1750.

Cornell, C.N. and Kaplan, L.J. (1978b) Biochemistry 17. 1755.

Cotton, F.A. and Wilkinson, G. (1983) Advanced Inorganic Chemistry, 4th Ed. Wiley Sons., New York.

Craven, P.A. and DeRubertis, F.R. (1978) J. Biol. Chem. 2 5 3 , 8433.

Craven, P.A., DeRubertis, F.R., and Pratt, D.W. (1979) J. Biol. Chem. 254, 8213.

Craven, P.A. and DeRubertis, F.R. (1983) Biochim. Biophys. Acta 745. 310.

Cushman, D.W. and Cheung, H.S. (1971) Biochem. Pharmacol. 2 0 . 1637.

Dahn, H., Loewe, L., and Burton, C.A. (1960) Helv. Chem. Acta 43. 320.

Davidsohn, K. (1887) Thesis, Albertus-University, Konigsberg, Prussia, as quoted in Kreye (1980).

DeRubertis, F.R. and Craven, P.A. (1976) Science 193, 897.

DeRubertis, F.R., Craven, P.A., and Pratt, D.W. (1978) Biochem. Biophys. Res. Commun. 83, 158.

Donchin, Y., Amirav, B., Sahar, A., and Yarkoni, S. (1978) Br. J. Anaesth. 50. 849. 131

Druckrey, H., Preussmann, R., Schmahl, D., and Muller, M. (1961) Naturwissenschaften 48, 134.

Druckrey, H., Preussmann, R., Schmahl, D., and Ivankovic, S. (1967) Z. Krebsforsch. 69, 103.

Ellman, G.L. (1959) Arch. Biochem. Biophys. 82. 70.

Field, L., Dilts, R.V., Ravichandran, B., Lenhert, P.G., and Carnahan, G.E. (1978) J. Chem. Soc. Perkin I 249.

Gerber, J.S. and Nies, A.S. (1983) Pharmacology of Antihypertensive Drugs. In Genest, J., Kuchel, 0., Hamet, P., and Cantin., (eds) Hypertension; Physiopathology and Treatment, 2nd Ed., McGraw-Hill, New York.

Gerzer, R., Bohme, E., Hofmann, F., and Schultz, G. (1981) FEBS Lett. 132, 71.

Gerzer, R., Radany, E.W., Garber, D.L. (1982) Biochem. Biophys. Res. Commun. 108. 678.

Greenblatt, M. and Mirvish, S.S. (1973) J. Natl. Cancer Inst. 50, 119.

Gruetter, C.A., Barry, B.K., McNamara, D.B., Gruetter, D.Y., Kadowitz, P.J., and Ignarro, L.J. (1979) J. Cyclic Nucl. Res. 5., 211.

Gruetter, C.A., Barry, B.K., McNamara, D.B., Kadowitz, P.J., and Ignarro, L.J. (1980) J. Pharmacol. Exp. Ther. 214, 9.

Harris, J.I. (1964) Nature 203, 30.

Hawksworth, G.M. and Hill, M.J. (1971) Br. J. Cancer 25, 520.

Hill, H.E. (1942) Aust. Chem. Inst. J. Proc. 9, 89.

Hoffmann, D., Hecht, S.S., Ornaf, R.M., and Wynder, E.L. (1974) Science 186, 265.

Hughes, W.L., Saroff, H.A., and Carney, A.L. (1949) J. Amer. Chem. Soc. 71. 2476.

Ignarro, L.J., Edward, J.C., Gruetter, D.Y., Barry, B.K., and Gruetter, C.A. (1980) FEBS Lett. 110. 275.

Ignarro, L.J. and Gruetter, C.A. (1980) Biochim. Biophys. Acta 631. 221. 132

Ignarro, L.J., Lippton, H., Edwards, J.C., Baricos, W.H., Hyman, A.L., Kadowitz, P.J., and Gruetter, C.A. (1981) J. Pharmacol. Exp. Ther. 218, 739.

Ignarro, L.J., Degnan, J.N., Baricos, W.H., Kadowitz, P.J., Wolin, M.S. (1982a) Biochim. Biophys. Acta 718. 49.

Ignarro, L.J., Wood, K.S., and Wolin, M.S. (1982b) Proc. Natl. Acad. Sci. USA 79/ 2870.

Ignarro, L.J., Adams, J.B., Horwitz, P.M., and Wood, K.S. (1986) J. Biol. Chem. 261, 4997.

Ivankovich, A.P., Miletich, D.J., and Tinker, J.H. (1978) Int. Anesthes. Clin. 16, 1.

Jack, R.D. (1974) Br. J. Anaesth. 46, 952.

Jennings, G.L., Gelman, J.S., Stockigt, J.R., and Korner, P.I. (1981) Clin. Sci. 61., 521.

John, M.E. and Waterman, M.R. (1980) J. Biol. Chem. 255. 4501.

Johnson, C.C. (1929) Arch Int. Pharmacodyn. Ther. 35, 489.

Johnson, M.D. and Wilkins, R.P. (1984) Inorganic Chem. 23, 231.

Kamisaki, Y., Waldman, S.A., and Murad, F. (1986) Arch. Biochem. Biophys. 251, 709.

Kaneko, Y., Ikeda, T., and Ueda, H. (1967) J. Clin. Invest. 4 6 , 105.

Katsuki, S., Arnold, W., Mittal, C., and Murad, F. (1977) J. Cyclic Nucl. Res. 3., 23.

Keefer, L.K. and Roller, P.P. (1973) Science 181. 1245.

Keilin, D. and Hartree, E.F. (1937) Nature 139. 548.

Knowles, M.E., McWeeny, D.J., Couchman, L., and Thorogood, M. (1974) Nature 2 4 7 . 288.

Kreye, V.A.W. (1980) Sodium Nitroprusside. In Scriabine, A. (ed). Pharmacology of Antihypersentive Drugs, Raven Press, New York.

Lagerkranser, M., Solleri, A., Irestedt, L., Tidgren, B., and Andrew, M. (1985) Acta Anaesthesiol. Scand. 29, 45. 133

Lang, K. (1933) Biochem. Zeitschr. 259, 243.

Lewicki, J.A., Brandwein, H.J., Mittal, C.K., Arnold, W.P., and Murad, F. (1982) J. Cyclic Nucl. Res. 8, 17.

MacRae, W.R. and Owen, M. (1974) Br. J. Anaesth. 46, 324.

McCleverty, J.A. (1979) Chem. Rev. 79. 53.

Maltz, H., Grant, M.A., and Navaroli, M.C. (1971) J. Org. Chem. 36. 363.

Marshall, W.K., Bedford, R.F., Arnold, W.P., Miller, E.D., Longnecker, D.E., Sussman, M.D., and Hakala, M.W. (1981) Anesthesiology 55, 277.

Merrifield, A.J. and Blundell, M.D. (1974) Br. J. Anaesth. 4 6 , 324.

Michenfelder, J.D. (1977) Anesthesiology 46., 196.

Miletich, D.J. and Ivankovich, A.D. (1978) Int. Anesthes. Clin. 16, 89.

Miller, E.D., Ackerly, J.A., Vaughan, D., Peach, M.J., and Epstein, R.M. (1977) Anesthesiology 47, 257.

Mirvish, S.S., Sams, J. Fan, T.Y., and Tannenbaum, S.R. (1973) J. Natl. Cancer Inst. 51, 1833.

Mirvish, S.S. (1975) Toxic. Appl. Pharmacol. 31, 325.

Mirvish, S.S., Cardesa, A., Wallcave, L., Shubik, P. (1975) J. Natl. Cancer Inst. 55, 633.

Mirvish, S.S., Bulay, 0, Runge, R.G., and Patil, K. (1980) J. Natl. Cancer Inst. 64. 1435.

Misono, K.S. and Inagami, T. (1980) Biochemistry 19, 2616.

Mittal, C.K. and Murad, F. (1977) Proc. Natl. Acad. Sci. USA 74, 4360.

Mittal, C.K., Arnold, W.P., Murad, F. (1978) J. Biol. Chem. 2 5 3 , 1266.

Monera, 0., Chang, M.K., Park, J.W., and Means, G.E. (1986) Fed. Proc. 45. 1540.

Mulvey, D. and Waters, W.A. (1975) J. Chem. Soc. Dalton Trans. 951. 134

Murad, F., Mittal, C.K., Arnold, W.P., Katsuki, S., and Kimura, H. (1978) Adv. Cyclic Nucl. Res. 9, 145.

Nakamura, S., Shin, T., Hirokata, Y., and Shigematsu, A. (1977) Br. J. Anaesth. 49, 1239.

Needleman, P., Jakschik, B., Johnson, E.M.Jr. (1973) J. Pharmacol. Exp. Ther. 187. 324.

Neurath, G.B., Dunger, M., Pein, F.M., Ambrosius, D., and Schreiber, 0. (1977) Fd. Cosmet. Toxicol. 15. 275.

Newberne, P.M. and Shank, R.C. (1973) Fd. Cosmet. Toxicol. 11, 819.

Oae, S., Fukushima, D., and Kim, Y.H. (1977a) J. Chem. Soc. Chem. Commun. 407.

Oae, S., Kim, Y.H., Fukushima, D., and Takata, T. (1977b) Chem. Lett. 893.

Ohlstein, E.H., Wood, K.S., and Ignarro, L.J. (1982) Arch. Biochem. Biophys. 218. 187.

Ohshima, H. and Bartsch, H. (1981) Cancer Res. 41, 3658.

Page, I.H., Corcoran, A.C., and Dustan, H.P. (1955) Circulation 11. 118.

Pensabene, J.W., Fiddler, W., Dooley, C.J., Doerr, R.C., and Wasserman, A.E. (1972) J. Agr. Food Chem. 20. 274.

Perutz, M.F., Kilmartin, J.V., Nagai, K., Szabo, A., and Simon, S.R. (1976) Biochemistry 15, 378.

Posner, M.A., Rodkey, F.L., and Tobey, R.E. (1976a) Anesthesiology 44, 330.

Posner, M.A., Tobey, R.E., and McElroy, H. (1976b) Anesthesiology 44, 157.

Preussmann, R., Daiber, D., and Hengy, H. (1964) Nature 2 0 1 , 502.

Prior, W.A., Church, D.F., Govindan, C.K., and Crank, G. (1982) J. Org. Chem. 47, 156.

Rapport, R.M., and Murad, F. (1983) J. Cyclic Nucl. Pro. Phos. Res. 9, 281.

Ridd, J.H. (1961) Qurt. Rev. 15, 418. 135

Saville, B. (1958) Analyst 83, 670.

Scheidl, W.R. and Frlsse, M.E. (1975) J. Am. Chem. Soc. 97./ 17.

Schultz, K.D., Schultz, K., and Schultz, G. (1977) Nature 265. 750.

Schulz, U. and McCalla, D.R. (1969) Can. J. Chem. 47, 2021.

Skeggs, L.T., Marsh, W.H., Kahn, J.R., and Shumway, N.P. (1954) J. Exp. Med. 99., 275.

Smith, D.W.E. (1966) Science 152. 1273.

Smith, R.P. and Kruszyma, H. (1974) J. Pharmacol. Exp. Ther. 191, 557.

Swinehart, J.H. (1967) Coordin. Chem. Rev. 2, 385.

Tinker, J.H. and Cucchiara, R.F. (1978) Int. Anesthes. Clin. 16., 89.

Tuzel, I.H. (1974) J. Clin. Pharmacol. 14, 494.

Vallee, B.L. and Hoch, F.L. (1955) Proc. Natl. Acad. Sci. USA 41, 327.

Veseley, D.C., Rovere, L.E., and Levey, G.S. (1977) Cancer Res. 3 7 , 28.

Vesey, L.J., Cole, P.V., Linnell, J.C., and Wilson, J. (1974) Br. J. Med. 20, 140.

Vesey, C.J., Cole, P.V., and Simpson, P.J. (1976) Br. J. Anaesth. 48., 651.

Vidt, D.G., Bravo, E.L., and Fouad, F.M. (1981) New Engl. J. Med. 306. 214.

Weare, J.A. (1982) Biochem. Biophys. Res. Commun. 104, 1319.

Wilson, J.M., Wu, D., DeGrood, R.M., and Hupe, D.T. (1980) J. Am. Chem. Soc. 102. 359.

Wishnok, J.S. and Archer, M.C. (1976) Br. J. Cancer 33. 307.

Wolin, M.S., Wood, K.S., and Ignarro, L.J. (1982) J. Biol. Chem. 257, 13312. 136

Woodside, T.Jr., Garner/ L., Bedford, R.F., Sussman, M.D., Miller, E.D.Jr., Longnecker, D.E., and Epstein, R.M. (1984) Anesthesiology 60, 413.

Yang, H.Y.T., Erdos, E.G., and Levin, Y. (1970) Biochim. Biophys. Acta 214, 374.