Alternative nitric oxide-producing substrates for no synthases

Free Radical Biology & Medicine, Vol. 37, No. 8, pp. 1105–1121, 2004 Copyright D 2004 Elsevier Inc. Printed in the USA. All rights reserved 0891-5849/$-see front matter

doi:10.1016/j.freeradbiomed.2004.06.031

Serial Review: Mechanisms and Novel Directions in the Biological Applications of Nitric Oxide Donors Serial Review Editor: S. Bruce King ALTERNATIVE NITRIC OXIDE-PRODUCING SUBSTRATES FOR NO SYNTHASES DANIEL MANSUY and JEAN-LUC BOUCHER UMR 8601–Universite´ Paris 5, 75270 Paris Cedex 06, France (Received 31 March 2004; Revised 22 June 2004; Accepted 24 June 2004) Available online 29 July 2004

Abstract—Nitric oxide (NO) is a key inter- and intracellular molecule involved in the maintenance of vascular tone, neuronal signaling, and host response to infection. The biosynthesis of NO in mammals involves a two-step oxidation of l-arginine (L-Arg) to citrulline and NO catalyzed by a particular class of heme-thiolate proteins, called NO-synthases (NOSs). The NOSs successively catalyze the N N-hydroxylation of the guanidine group of L-Arg with formation of N N-hydroxy-l-arginine (NOHA) and the oxidative cleavage of the CN(OH) bond of NOHA with formation of citrulline and NO. During the last decade, a great number of compounds bearing a CNH or CNOH function have been synthesized and studied as possible NO-producing substrates of recombinant NOSs. This includes derivatives of L-Arg and NOHA, N-alkyl (or aryl) guanidines, N,N’- or N,N-disubstituted guanidines, N-alkyl (or aryl) N’-hydroxyguanidines, N- (or O-) disubstituted N’-hydroxyguanidines, as well as amidoximes, ketoximes, and aldoximes. However, only those involving the NHC(NH2)jNH (or NOH) moiety have led to a significant formation of NO. All the N-monosubstituted N’-hydroxyguanidines that are well recognized by the NOS active site lead to NO with catalytic efficiences (k cat/K m) up to 50% of that of NOHA. This is the case of many N-aryl and N-alkyl N’-hydroxyguanidines, provided that the aryl or alkyl substituent is small enough to be accommodated by a NOS hydrophobic site located in close proximity of the NOS bguanidine binding site.Q As far as N-substituted guanidines are concerned, few compounds bearing a small alkyl group have been found to act as NO-producing substrates. The k cat value found for the best compound may reach 55% of the k cat of L-Arg oxidation. However, the best catalytic efficiency (k cat/K m) that was obtained with N-(4,4,4-trifluorobutyl) guanidine is only 100-fold lower than that of L-Arg. In a general manner, NOS II is a better catalyst that NOS I and III for the oxidation of exogenous guanidines and N-hydroxyguanidines to NO. This is particularly true for guanidines, as the ones acting as substrates for NOS II have been found to be almost inactive for NOS I and NOS III. Thus, a good NOproducing guanidine substrate for the two latter isozymes remains to be found. D 2004 Elsevier Inc. All rights reserved. Keywords—Nitric oxide, Nitric oxide synthase, Substrates, Guanidines, N-Hydroguanidines, Oxidation, Mechanism, Catalysis, Free radicals

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mechanism of NO Formation From NOS-Catalyzed Oxidation Of Structure of NOS. . . . . . . . . . . . . . . . . . . . . . . . NOS-catalyzed Nx -hydroxylation of L-Arg (Fig. 2) . . . . . . NOS-catalyzed oxidation of NOHA to citrulline and NO . . .

. . . . L-Arg . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

1106 1106 1106 1107 1108

This article is part of a series of reviews on bMechanisms and Novel Directions in the Biological Applications of Nitric Oxide Donors.Q The full list of papers may be found on the home page of the journal. Address correspondence to: Daniel Mansuy, UMR 8601–Universite´ Paris 5, 45 rue des Saints-Pe`res, 75270 Paris Cedex 06, France. Fax: +33 1 42 86 83 87; E-mail: [email protected]. 1105

1106

D. Mansuy and J.-L. Boucher

Formation of NO from NOS-catalyzed oxidation of l-Arg derivatives . . . . . . . . . . . Problems encountered in the search for NOS-dependent precursors of NO . . . . . . . NO formation from NOS-dependent oxidation of a-amino-acid analogs of l-Arg. . . . NO formation from NOS-dependent oxidation of non-a-amino-acid analogs of NOHA . Formation of NO from NOS-catalyzed oxidation of exogenous compounds . . . . . . . . Formation of NO from NOS-catalyzed oxidation of N-alkyl- (or aryl-) NV-Hydroxyguanidines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selectivity of N-Hydroxyguanidines as substrates for NOS I, II, and III. . . . . . . . . Formation of NO from NOS-catalyzed oxidation of guanidines . . . . . . . . . . . . . Relationship between the structure of NO-producing alternative substrates of NOS and their NO-formation efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . Relationship between the structure of N-substituted guanidines and N-Hydroxyguanidines and their affinity for NOS II . . . . . . . . . . . . . . . . . . . Relationship between the structure of N-substituted NV-hydroxyguanidines and their activity as NO-producing NOS substrates . . . . . . . . . . . . . . . . . . . Relationship between the structure of N-substituted guanidines and their activity as NO-producing NOS substrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

INTRODUCTION

Nitric oxide (NO) is a key inter- and intracellular messenger molecule involved in the maintenance of vascular tone, neuronal signaling, and host response to infection [1]. The biosynthesis of NO in mammals involves a two-step oxidation of l-arginine (L-Arg) to citrulline and NO [2,3]. This reaction is catalyzed by a particular class of heme-thiolate proteins, called NO synthases (NOSs) [4,5]. The NOSs successively catalyze the N N-hydroxylation of the guanidine group of L-Arg with formation of N N-hydroxy-l-arginine (NOHA) and the oxidative cleavage of the CjN(OH) bond of NOHA, with formation of citrulline and NO. The first step is a classical monooxygenation that consumes 1 mol of O2 and 1 mol of NADPH, whereas the second step is an atypical monooxygenation that results in a three-electron oxidation of NOHA and consumes 1 mol of O2 and 0.5 mol of NADPH (Eqs. (1) and (2)):

 þ Hþ Y RNHCðNH2 Þ RNHCðNH2 Þ ¼ NHþ 2 þ O2 þ 2e

1109 1109 1110 1111 1111 1112 1115 1116 1117 1117 1118 1118 1119 1119

deficit [7]. These NO donors are rapidly and, most often, nonselectively transformed to NO in mammals and have a short half-life in vivo. Another possible approach toward more stable NO donors would be to find more chemically and metabolically stable compounds that would be selectively oxidized by NOSs in situ, with formation of NO in a given tissue. This minireview describes the efforts that have been made during these past 5 years to find such alternative, exogenous NOproducing substrates of NOS. The design of such alternative substrates requires a good knowledge of the structure of the NOS active sites and of the mechanisms of the NOS-dependent formation of NO. It is why this minireview starts with a short introduction on the NOS structure and on the most recent mechanisms admitted for NO formation by NOS-catalyzed oxidation of its natural substrate, L-Arg. MECHANISM OF NO FORMATION FROM NOS-CATALYZED OXIDATION OF L-ARG

NOS

¼ NOH þ H2 O

Structure of NOS ð1Þ

NOS RNHCðNH2 Þ ¼ NOH þ O2 þ e þ Hþ Y RNHCONH2

þ NO þ H2 O;

ð2Þ

where R = CH2CH2CH2CH(NH2)COOH. Several pathophysiological situations are related to a deficit in NO biosynthesis in various tissues or organs [6]. NO itself or NO donors such as trinitroglycerin are used as therapeutic agents to counterbalance the NO

In mammals, three main NOS isozymes, the neuronal (nNOS or NOS I), inducible (iNOS or NOS II), and endothelial (eNOS or NOS III) NOSs, catalyze this reaction [4,5]. All of them are constituted by a single polypeptide chain comprising two domains. The Nterminal oxygenase domain contains the catalytic site involving the heme (iron protoporphyrin IX) and H4B cofactors and the L-Arg binding site. The C-terminal reductase domain contains two flavin cofactors, flavin adenine dinucleotide (FAD) and flavin mononucleotide

Alternative substrates of NO sythases

Fig. 1. Relative positioning of l-arginine, heme, and H4B in the NOS active site (reproduced, by permission of the publisher, from Ref. [9]).

(FMN), and a binding site for NADPH. The reductase domain catalyzes electron transfer from NADPH via FAD and FMN to the heme that is responsible for reductive dioxygen activation and L-Arg oxidation. A sequence of the protein located between the reductase and the oxygenase domains binds calmodulin and Ca2+; this binding regulates electron transfer between the two domains and makes nNOS and eNOS responsive to variations in cellular Ca2+ concentration. In fact, catalytically active NOSs are homodimers in which the electrons coming from the reductase of a given monomer are transferred to the oxygenase domain of the other monomer [8].

1107

Crystal structures that have been described for the oxygenase domain (NOSoxy) of several NOSs [9–15] have shown that the NOS heme, L-Arg binding site, and H4B binding site are very close together and structured by subunit dimerization. Fig. 1 shows that the three players involved in NOS-catalyzed reactions, H4B, heme, and LArg, can communicate as they are linked together via a complex network of hydrogen bond interactions. The guanidinium group of L-Arg, as well as the N-hydroxyguanidine group of NOHA [9,10,15], stack against the + distal face of the heme and have their CjNH2 or CjNOH function ideally located for reaction with the FejO or the Fe–OO active intermediates of the NOS catalytic cycle. H4B is bound too far away from the substrate to directly participate in the oxidation of larginine or NOHA. However, it establishes two direct hydrogen bond interactions with a heme propionate via its 3-NH and 2-NH2 functions and a third hydrogen bond with this propionate via its 4-keto function through a water molecule [16]. Interestingly, the same heme propionate forms a hydrogen bond with the a-amino group of L-Arg or NOHA [9,10,15]. NOS-catalyzed Nx -hydroxylation of L-Arg (Fig. 2) The mechanism of this reaction catalyzed by NOS is highly similar to that of cytochrome P450-dependent monooxygenations [17]. The successive iron species involved—i.e., the Fe(III) resting state and the Fe(II),  Fe(II)–O2, Fe(III)–OO , Fe(III)–OOH, and high-valent

Fig. 2. Different steps possibly involved in the N-hydroxylation of l-arginine by NOS (from Refs. [19–21]).

1108

D. Mansuy and J.-L. Boucher

FejO intermediates—are highly similar (Fig. 2). Moreover the NOS-catalyzed NN-hydroxylation of L-Arg and P450-catalyzed hydroxylations are believed to be performed by similar high-valent iron–oxo species. There are two main differences between the catalytic cycles of the two enzymes. The first one is that the electron transferred to the Fe(II)–O2 intermediate in NOS does not come from the reductase as may be the case in cytochrome P450, but from H4B, which is well positioned in the NOS active site to transfer an electron with intermediate formation of a H4B-derived radical [18–21]. The second difference is that one of the two protons necessary for the formation of the FejO hydroxylating species comes from the substrate in NOS, because the guanidine function of LArg (pK a ~13) should exist in its protonated form. A critical step for efficient hydroxylation of L-Arg is an electron transfer from H4B to heme Fe(II)–O2 (10–20 s1 from single turnover experiments) [22,23], which should be faster than the dissociation of Fe(II)–O2 to Fe(III) and  O2 . It has been recently proposed that H4B acts not only as an electron donor but also as a proton donor to the Fe(II)–O2 complex [24], and an iron porphyrin model system for this reaction has been described [21,25]. In this model system, reaction of a tetrahydropterin with a porphyrin Fe(II)–O2 complex leads to a Fe(III)–OOH intermediate and a tetrahydropterin-derived radical [25]. Efficient coupling between electron transfer from

S

NADPH and L-Arg hydroxylation, which results in the consumption of only 1 mol of NADPH for NOHA formation, should depend on the efficiency of the steps involved between Fe(II)–O2 formation and hydroxylation of L-Arg. These steps are in competition with the dissociation of the Fe–O bond of the Fe(II)–O2 and  Fe(III)–OOH intermediates leading to O2 and H2O2 and with the reduction of the iron–oxo complex, which results in a decoupling between NADPH consumption and L-Arg hydroxylation. In NOS containing both H4B and L-Arg, the formation of Fe(III)–OOH from Fe(II)–O2 is very efficient because of the network of bonds between H4B and L-Arg via a heme propionate that should facilitate electron (and maybe proton) transfer from H4B to Fe(II)–O2 (Fig. 1). Protonation of Fe(III)–OOH presumably by the guanidinium function of L-Arg and hydroxylation of the guanidine function of L-Arg are efficient because of the perfect positioning of this function relative to the heme plane due to an interaction with a well-conserved glutamate residue of the protein.

S

NOS-catalyzed oxidation of NOHA to citrulline and NO Most recent data based on stopped-flow and rapidfreeze EPR spectroscopy studies are in favor of the mechanism shown in Fig. 3 [20,21,26]. It also involves an electron transfer from H4B to NOSFe(II)–O2, followed by the addition of the Fe(III)–OO intermediate to the

Fig. 3. Different steps possibly involved in the NOS-catalyzed oxidation of NOHA into citrulline (Cit) and NO (from Refs. [19–21]).

Alternative substrates of NO sythases

CjNOH bond of NOHA. The cleavage of the O–O, C–N, and O–H bond of the ferric-alkylperoxo intermediate would have led to citrulline and NO. However, at this stage, a back electron transfer from the iron species (or  NO) to H4B+ would regenerate H4B and permit the exclusive formation of NO, instead of NO [26,27].

1109

[33–35]. However, all these oxidations of N-hydroxyguanidines catalyzed by cytochromes P450 or H4B-free NOSs are nonselective in terms of substrates, as all the studied N-hydroxyguanidines were found to be oxidized, and are nonselective in terms of products, as they lead not only to the corresponding urea, but also to the corresponding cyanamide (Eq. (3)) [33–35]:

S

S

FORMATION OF NO FROM NOS-CATALYZED OXIDATION

RNHCðNH2 Þ

OF L-ARG DERIVATIVES

P450 or H4 B

free NOS

¼ NOH Y NADPH;O2

Problems encountered in the search for NOS-dependent precursors of NO

RNHCONH2 þ RNHCN þ nitrogen oxides including NO: ð3Þ

Moreover, these reactions are inhibited to a great extent by superoxide dismutase (SOD) [32–34]. Thus, it seems that the observed oxidations of N-hydroxyguanidines by cytochromes P450 and H4B-free NOSs are  mainly due to O2 derived from the oxidase function  itself has of these hemeproteins. Accordingly, O2 been found to oxidize amidoximes and N-hydroxyguanidines with formation of nitrogen oxides [30,36]. In  these nonselective, O2 -dependent oxidations of Nhydroxyguanidines, both cyanamides and ureas are formed (Eq. (4)):

In the search for compounds susceptible to oxidization by NOSs or related monooxygenases with formation of NO, we have been rapidly faced with the problem of the nonselective formation of nitrogen oxides including NO from oxidation of N-hydroxyguanidines, amidoximes, and even ketoximes. Such molecules bearing a CjNOH function are easily oxidized by many chemical oxidizing agents such as H2O2, MnO2, Pb(OAc)4, tBuOOH, or peracids, with oxidative cleavage of their CjN(OH) bond and formation of nitrogen oxides [28–31]. Microsomal cytochromes P450 catalyze the oxidation by NADPH and O2 of ketoximes, amidoximes, and N-hydroxyguanidines with formation of NO2, NO3, and NO [32,33]. H4B-free NOS also catalyzes the oxidation of N-hydroxyguanidines, such as NOHA and N-aryl-NV-hydroxyguanidines, by NADPH and O2, with formation of nitrogen oxides including NO

S

S

S

S

 þO2 RCðNH2 ÞNOH Y RCONH2 þ RCN þ nitrogen oxides:

ð4Þ

Similar problems have not been encountered so far in the search for guanidines susceptible to be oxidization by

Table 1. Formation of NO from the Oxidation of L-Arg Analogs by NOS II (A) Changes in the a-amino-acid function: H2NC(=NH)NH-(CH2)3-CHR1R2 Compound L-Arg L-Arg methylester l-Argininamide N a-methyl-L-Arg N a-acetyl-L-Arg N a-benzoyl-L-Arg Argininic acid 2-Chloro 5-guanidino valeric acid

R1

R2

Activity (%)

Reference

CO2H CO2Me CONH2 CO2H CO2H CO2H CO2H CO2H

NH2 NH2 NH2 NHCH3 NHCOCH3 NHCOC6H5 OH Cl

100 b0.5 b0.5 2 F 0.5 b0.5 b0.5 1 F 0.5a b0.5

— [37] [37] [38] [38] [38] [38] [38]

Activity (%)

(B) Changes in the guanidine function: ZC(=NY)-X-(CH2)2CH(NH2)CO2H Compound

X

Y

Z

NN-hydroxy-L-Arg

NH-CH2

OH

NH2

N N-methyl-L-Arg Indospicine N N-hydroxyindospicine N q-iminoethyl-l-ornithine Canavanine 5-Keto-d,l-arginine

NH-CH2 CH2-CH2 CH2-CH2 CH2 NH-O C=O

H H OH H H H

NHCH3 NH2 NH2 CH3 NH2 NH2

Activities are expressed as a percentage of that found for L-Arg. a Samples may have been contaminated by low amounts of L-Arg [38].

140 180 5 0 b0.5 0 0 0

Reference [44] [3] [43] [39] [40] [41] [42] [45]

1110

D. Mansuy and J.-L. Boucher

Table 2. Effects of Changes in the Chain Joining the Guanidine and a-Amino-Acid Functions of L-Arg on the Rates of NO Formation by NOS II Compound H2NC(=NH)NH-X-CH(NH2)CO2H Homo-L-Arg Nor-L-Arg Dinor-L-Arg 3,4-Didehydro-D,L-Arg

3-Guanidino-d,l-phenylglycine 2-Guanidino-l-phenylalanine 3-Guanidino-l-phenylalanine trans 3,4-Cyclopropyl-L-Arg

X

Activity (%)

Reference

(CH2)4 (CH2)2 CH2 CH2CH=CH E-isomer Z-isomer 1,3-C6H4 1,2-C6H4-CH2 1,3-C6H4-CH2 (CH2)2C syn-Isomer anti-Isomer

73 F 5 b0.5 b0.5

[40] [38] [38]

4 b0.5 b0.5 b0.5 b0.5

[45] [45] [45] [46] [46]

a

[47] [47]

b0.5

Activities are expressed as a percentage relative to L-Arg. a Authors indicated that this compound was a substrate for NOS I, II, and III, but gave a K m value only for NOS I.

NOS with NO formation, because of the much greater stability of guanidines by comparison to N-hydroxyguaS nidines toward O2 or other oxidants. NO formation from NOS-dependent oxidation of a-amino-acid analogs of L-Arg Small changes in the L-Arg or NOHA structure almost completely abolish NO formation upon NOS-dependent oxidation of the corresponding compounds. Thus, changes in the a-amino-acid moiety, such as replacement of the a-COOH function with a methyl ester or amide function, or acetylation or benzoylation of the a-NH2 group [37], result in an almost complete loss of NO formation (Table 1). Methylation of the a-NH2 function a leads to N -methyl-l-Arg that produces only low amounts of NO upon oxidation by NOS II, compared with L-Arg (~2% [38]). Replacement of the a-NH2 function of L-Arg with an OH or Cl group also gives inactive compounds [38]. The guanidine (or N-hydroxyguanidine) function of L-Arg (or of NOHA) is also crucial for NO formation, as

replacement of the y-NH group of L-Arg or NOHA with a N CH2 group leads to indospicine or N -hydroxyindospicine, which are unable to produce NO upon NOSdependent oxidation (Table 1) [39,40]. Replacement of the guanidine function of L-Arg with an amidine function, N as in N -iminoethyl-l-ornithine [41], or replacement of the NHCH2 moiety of L-Arg with a NH-O group, as in canavanine [42], also results in inactive compounds (Table N 1). N -methylation of L-Arg gives a compound that acts as a mechanism-based inhibitor of NOSs and leads to NO with low rates (5% relative to L-Arg) [43]. Small changes in the chain length of L-Arg are possible, as homo-l-Arg remains a good NO-producing substrate of NOS (Table 2) [40,44]. However, introduction of a double bond leads to an inactive compound if this double bond has a trans configuration and to a weakly active compound if it has a cis configuration (4% compared to L-Arg) [45]. Moreover, a decrease in the chain length, as in nor-l-Arg [38], nor-NOHA [40], or dinor-l-Arg [38], leads to inactive compounds. Finally,

Table 3. Formation of NO from the Oxidation by NOSs of Some N-Hydroxyguanidines, H2NC(=NOH)NH-X-CHR1R2, related to NOHA Activity (%) X

R1

R2

NOHA N x -hydroxyagmatine N x -hydroxyguanidino-pentanoic acid N-butyl-NV-hydroxyguanidine

(CH2)3 (CH2)3 (CH2)3 (CH2)3

CO2H H CO2H H

NH2 NH2 H H

N x -hydroxy-nor-L-Arg N x -hydroxyguanidino-propylamine N x -hydroxyguanidinobutyric acid N-propyl-NV-hydroxyguanidine

(CH2)2 (CH2)2 (CH2)2 (CH2)2

CO2H H CO2H H

NH2 NH2 H H

N x -hydroxydinor-L-Arg N x -hydroxyhomo-L-Arg(homo-NOHA) N-pentyl-NV-hydroxyguanidine

CH2 (CH2)4 (CH2)4

CO2H CO2H H

NH2 NH2 H

Compound

Activities are expressed as a percentage of those found for NOHA.

NOS I 100 2F1 b0.5 65 F 10 64 b0.5 b0.5 b0.5 36 F 5 70 b0.5 82 F 10 34 F 8 31

Reference

NOS II

NOS III

100 8F2 1F1 68 F 10 42 b0.5 b0.5 b0.5 15 F 3 15 b0.5 55 F 10 25 F 5 32

100 b0.5 2F1 41 F 5 20 b0.5 b0.5 b0.5 41 F 5 24 b0.5 70 F 10 b0.5 0

— [49] [49] [49] [50,51] [49,40] [49] [49] [49] [50] [40,49] [49] [49] [50]

Alternative substrates of NO sythases

replacement of the (CH2)3 chain of L-Arg with rigid cyclic moieties, such as a phenyl [45,46] or cyclopropyl [47] group, also gives compounds unable to produce NO upon NOS-dependent oxidation (Table 2). NO formation from NOS-dependent oxidation of non-a-amino-acid analogs of NOHA Because of the data indicated in the previous section, one could believe that only L-Arg and NOHA, or very close analogs, would act as NOS-dependent NO producers. Accordingly, desamino-l-Arg [38,48] and desamino-NOHA [48,49], as well as descarboxy-l-Arg [38,48] and descarboxy-NOHA [48,49], also called N agmatine and N -hydroxyagmatine, are very bad substrates of NOSs as they lead to rates of NO formation between 0 and 8% of those obtained with the natural substrates (Table 3). However, quite surprisingly, oxidations by NOSs of the NOHA analog in which both the aNH2 and the a-COOH functions of NOHA have been removed (N-butyl-NV-hydroxygunanidine) lead to a rate of NO formation very close to that found for NOHA itself (about 65, 55, and 30% for NOS I, NOS II, and NOS III, respectively; Table 3) [49–51]. In a similar manner, nor-NOHA, as well as its desamino and descarboxy derivatives, fails to be oxidized by NOSs with NO formation, whereas the nor-NOHA analog

1111

lacking both the a-amino and the a-carboxy functions of nor-NOHA (N-propyl-NV-hydroxyguanidine) is a relatively good substrate of NOSs, with NO formation rates between 15 and 70% of that of the natural substrate NOHA, as a function of the NOS isoform (Table 3) [49,50]. These data showed that the presence of the aamino-acid function of NOHA is not absolutely required for N-hydroxyguanidines to act as decent NOS substrates with high rates of NO formation. FORMATION OF NO FROM NOS-CATALYZED OXIDATION OF EXOGENOUS COMPOUNDS

The results described in the previous paragraph showed that simple, non-a-amino-acid N-alkyl-NVhydroxyguanidines, such as N-butyl- and N-propyl-NVhydroxyguanidines, were efficient NOS substrates for NO formation. This led us to test the oxidation by NOS of a great number of compounds involving a CjNH or CjNOH function, which are a priori the minimal functions necessary for oxidation with formation of NO according to the mechanism currently admitted for NOS. As far as compounds involving a CjNOH function are concerned, many N-alkyl- and N-aryl-NV-hydroxyguanidines have been shown to act as good NOS substrates with formation of NO, as this will be described

Fig. 4. Different series of compounds involving a CjNOH function tested as NO-producing NOS II substrates. Many N-monosubstituted NV-hydroxyguanidines, such as those shown here, are good NO-producing NOS II substrates. At the opposite end, all the indicated compounds deriving from N- or O-alkylation of N-aryl- (or alkyl-) NV-hydroxyguanidines, or from replacement of their RNH substituents with aryl, alkyl, or H groups, were found to be unable to produce NO upon oxidation by recombinant NOS II (see text).

1112

D. Mansuy and J.-L. Boucher

Thus far, only some N-alkylguanidines have been found to act as NOS substrates with NO formation [53] (vide infra), whereas N,N-disubstituted guanidines and all the tested amidines were not [52,53].

in detail in the following. However, the presence of an integer –NHC(NH2)jNOH moiety seems to be required for NO formation, as any supplementary N- or Oalkylation of N-alkyl- (or aryl-) NV-hydroxyguanidines leads to compounds that are unable to act as NOproducing NOS substrates. Such compounds are shown in Fig. 4. For instance, the compounds resulting either from the O-methylation or O-benzylation or from the Nmethylation of N-butyl-NV-hydroxyguanidine fail to produce NO upon oxidation by recombinant NOS II [51]. This is also the case for two other N,N-disubstituted NV-hydroxyguanidines shown in Fig. 4 [51]. In a similar manner, N,NV-diphenyl-NV-hydroxyguanidine is unable to generate NO [52]. Moreover, amidoximes resulting from the replacement of the aryl-NH or NH2 substituent of Naryl-NV-hydroxyguanidines with an aryl or methyl group, respectively (Fig. 4), are not substrates for NOS II. Finally several arylaldoximes and arylketoximes (Fig. 4) are also unable to act as NO-producing NOS II substrates [52]. Many compounds involving a CjNH function have also been synthesized and compared as NOS substrates.

Formation of NO from NOS-catalyzed oxidation of N-alkyl- (or aryl-) NV-hydroxyguanidines As mentioned above, several N-alkyl- (or aryl-) NVhydroxyguanidines are oxidized by NOSs with high rates of NO formation (Tables 3–6) [49-52,54]. Detailed kinetic studies of their oxidation by recombinant NOS II showed characteristics very similar to those of the NOS IIcatalyzed oxidation of NOHA [52,54]: (a) These oxidations require the presence of H4Bcontaining NOS II, NADPH, and O2. (b) They are strongly inhibited by usual NOS inhibitors such as N N-nitro-l-Arg. (c) They are not inhibited by SOD and catalase, contrary to the corresponding oxidations catalyzed by H4B-free NOS II, indicating that these oxidations are not  performed by O2 or H2O2 possibly derived from the NOS oxidase function.

S

Table 4. Formation of NO from the Oxidation Catalyzed by NOSs of Monosubstituted N-alkyl-NV-hydroxyguanidines, RNHC(=NOH)NH2 Activity (%) R Ethyl iso-Propyl Propyl 1-Methylpropyl 2-Methylpropyl Butyl ter-Butyl 2-Butenyl (E-isomer) 3-Butynyl 4-Hydroxybutyl 4,4,4-Trifluorobutyl 3-Methylbutyl Pentyl Hexyl Cyclopropyl Cyclopropylmethyl Cyclopentyl Cyclohexyl Benzyl 4-Methoxybenzyl 4-Chlorobenzyl 4-Nitrobenzyl 4-Methylbenzyl 2-Phenylethyl

NOS I

NOS II

NOS III

Reference

6 38 38 41 F 5 24 4 10 20 41 F 5 b0.5 b0.5 b0.5 b0.5 3F1 0 0 b0.5 0 b0.5 26 F 4 40 F 12 b0.5 b0.5 0 0

4-CH3O–C6H4–CH2 4-Cl–C6H4–CH2 4-NO2–C6H4–CH2 4-CH3–C6H4–CH2 C6H5–CH2–CH2

0 0 0 0 b0.5

3 32 15 15 F 5 15 12 5 42 68 F 10 b0.5 0.5 28 F 8 b0.5 95 F 8 24 32 25 F 5 3 1F1 b0.5 10 F 3 b0.5 b0.5 3 0 15 F 4 0 0 0 0 6F2

[50] [50] [51] [49] [50] [50] [50] [50] [49] [49,50] [51] [51]

C3H5–CH2 C5H9 C6H11 C6H5–CH2

1 76 70 36 F 5 70 18 2 64 65 F 10 b0.5 5 37 F 4 b0.5 41 F 8 6 31 34 F 8 4 2F1 70 F 8 62 F 20 b0.5 b0.5 3 0

CH3CH2 (CH3)2CH CH3CH2CH2 CH3CH2CH(CH3) CH3CH(CH3)CH2 CH3CH2CH2CH2 (CH3)3C CH3CH=CHCH2 CHuCH2CH2 HO–CH2CH2CH2CH2 CF3CH2CH2CH2 CH3CH(CH3)CH2CH2 CH3(CH2)3CH2 CH3(CH2)4CH2 C3H5

Activities are expressed as % of those found for NOHA. a Unpublished results from our lab.

0 0 0 0 b0.5

a a

[50] [50] [49] [50] [49] [51] a

[51] [51] [50] [50] [52] [50] [50] [50] [50] [52]

Alternative substrates of NO sythases

(d) They selectively lead to NO and the corresponding urea in a 1:1 molar ratio (Eq. (5)), as demonstrated in the particular case of N-para-chlorophenyl-NV-hydroxyguanidine [54]. Moreover, the corresponding cyanamide, N-parachlorophenylcyanamide, that is formed in large amounts in the oxidation catalyzed by H4B-free NOS II [34] is only a very minor product in NOS II-catalyzed oxidation of N-para-chlorophenyl-NV-hydroxyguanidine [54] (Eq. (5)):

1113

Moreover, NOS II-catalyzed oxidations of N-hydroxyguanidines exhibit classical saturation kinetics and Lineweaver–Burk plots. Table 6 compares the K m and k cat values measured for several N-hydroxyguanidine substrates and shows that many of them are oxidized with k cat values similar to that of NOHA. Particularly striking are the data obtained for N-butyl-NV-hydroxyguanidine, which is oxidized by NOS II with formation of NO with a great catalytic efficiency and a k cat/K m value only two times lower than in the case of NOHA

ð5Þ

[49]. Furthermore, it is noteworthy that the NOS IIdependent oxidation of this substrate (at saturating concentrations) occurs without decoupling between NADPH consumption and NO formation. Thus, formation of 1 mol of NO upon NOS II-catalyzed oxidation of this N-hydroxyguanidine requires the consumption of

only 0.5 mol of NADPH, as in the case of NOHA and in agreement with Eq. (2) [49]. From the data obtained so far on a great number of N-alkyl- or N-aryl-NV-hydroxyguanidines, it seems that many monosubstituted N-hydroxyguanidines are efficiently oxidized with formation of NO. Actually,

Table 5. Formation of NO from the Oxidation Catalyzed by NOSs of Monosubstituted N-Aryl-NV-hydroxyguanidines, ArNHC(=NOH)NH2 Activity (%) Ar C6H5 3-F–C6H4 4-F–C6H4 2-Cl–C6H4 3-Cl–C6H4 4-Cl–C6H4 4-Br–C6H4 4-CH3–C6H4 4-tert-Butyl-C6H4 4-OH–C6H4 2-CH3O–C6H4 4-CH3O–C6H4 3-NH2–C6H4 4-NH2–C6H4 4-NO2–C6H4 3-CF3–C6H4 4-CF3–C6H4 3-CH2OH–C6H4 3-CH2NH2–C6H4 4-(CH2CO2H)–C6H4 4-(O–CH2CO2H)–C6H4 4-[CH(CH3)CO2H]–C6H4 1-Naphthyl 6-Indazolyl 3-Thienyl

NOS I 1.5 F 0.5 0 b0.5 4.5 F 1 b0.5 b0.5 b0.5 0 b0.5 0 b0.5 0 b0.5 8.5 F 1 b0.5 3.5 F 1 0 2.5 F 1 4.5 F 1 b0.5 b0.5 b0.5 n.d. n.d. n.d. n.d. n.d. 0 b0.5 b0.5

NOS II 20 F 32 51 F 41 F 5F 5F 13 F 13 7.5 F 15 17.5 F 24 b0.5 16.5 F 5F 8F 29 3.5 F 8F 2F 2.5 F 0.5 F 6F 1.5 F b0.5 b0.5 b0.5 0 2.5 F 9.5 F

Activities are expressed as % of those found for NOHA. n.d. = not determined.

NOS III

Reference

b0.5

[52] [50] [52] [52] [52] [52] [52] [50] [52] [50] [52] [50] [52] [52] [52] [52] [50] [52] [52] [52] [52] [52] [52] [52] [52] [52] [52] [50] [52] [52]

3 0 8 6 1 1 3

n.d. b0.5 b0.5 b0.5 b0.5 0

2

b0.5 0

4

b0.5 0

3 2 2 1 2 1 1 0.2 1 0.5

0.5 2

b0.5 b0.5 b0.5 b0.5 0 6.5 F 2 b0.5 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 0 b0.5 b0.5

1114

D. Mansuy and J.-L. Boucher

the size of the alkyl or aryl group is crucial for recognition, binding, and oxidation of N-hydroxyguanidines by NOSs. Thus, Table 4 shows that alkyl Nhydroxyguanidines with a propyl or butyl chain are good NOS substrates, whereas those bearing either a shorter (ethyl) or a longer (hexyl) chain almost completely fail to produce NO. N-pentyl-NV-hydroxyguanidine is an intermediate case, as it is still active to produce NO upon oxidation by NOS I and II, although with a lower activity, but it is not a substrate of NOS III. Introduction of a methyl substituent into the propyl or butyl chain generally results in a dramatic decrease in the activity (compare R = propyl with R = 1- or 2-methylpropyl or R = butyl with R = 3-methylbutyl). In a more general manner, introduction of a function into the butyl chain (a cis double bond, a terminal triple bond, a 4-OH or 4-NH2 function; Tables 3 and 4) leads to a large decrease in the NO formation rate. More bulky compounds with R = t-butyl, cyclopentyl, cyclohexyl, or para-substituted benzyl are completely inactive, whereas those with R = benzyl and 2phenylethyl are only slightly active with NOS II. Finally, the best alkyl substrates are those for which R = butyl, propyl, isopropyl, or cyclopropyl. Table 6 shows that two of them (R = butyl and isopropyl) are very efficiently oxidized by NOS II with k cat/K m values only 2 times lower than that of NOHA, whereas the k cat/K m values found for those with R = cyclopropyl and pentyl are at least 10 times lower. However, it is noteworthy that the K m values measured for the butyl, isopropyl, and cyclopropyl NV-hydroxyguanidines are almost equal to that of NOHA.

Many N-aryl-NV-hydroxyguanidines are oxidized with NO formation by NOS II; only a few of them are substrates of NOS I and III (Table 5) [50,52,54]. Here too, the size of the aryl group is critical as only compounds bearing a small aryl group, such as Ar = phenyl or 3- or 4-fluorophenyl, are good substrates for NOS II. Those bearing a too bulky aryl group, such as Ar = 4-t-butyl-C6H4, 4-NO2-C6H4, 4-(CH2COOH)C6H4, 4-(OCH2COOH)-C6H4, or 1-naphthyl, are not substrates of NOS II. The k cat found for the oxidation of the N-aryl-NVhydroxyguanidines may reach values very close to that of NOHA, as is the case for compounds with Ar = 4-fluoroor 4-methylphenyl (Table 6). However, the K m values are generally much higher than those found for the best N-alkyl-NV-hydroxyguanidines. The K m for the smallest phenyl derivatives are already around 300 AM and introduction of a more bulky para-substituent on the phenyl ring leads to dramatic increases in K m (K m = 2600 AM for Ar = 4-methoxyphenyl). At the opposite end, the lowest K m found so far for a N-hydroxyguanidine was obtained for a small N-aryl derivative, with Ar = 3-thienyl (Table 6) [52]. Attempts have been made to find a correlation between the K m and the k cat values determined for NOS II-catalyzed oxidation of N-aryl-NV-hydroxyguanidines and many classical descriptors for geometric, electronic, and hydrophobic properties of the parasubstituent of the phenyl ring, such as the Hansch constant k; the Hammett constants j, j, or j+; or the dipole moment [52]. The k cat values did not lead to any simple correlation with any of the descriptors used, and

Table 6. Kinetic Constants Measured for the Formation of NO from Oxidation of NOHA, Homo-NOHA, and Some N-Alkyl- and N-Aryl-NV-hydroxyguanidines Catalyzed by NOS II Compound

K m (AM)

k cat (min1)

k cat/K m (min1 AM1)

Reference

NOHA

40 F 10 40 F 10 146 F 20

425 F 75 480 F 60 410 F 50

10.60 12.00 2.80

[52] [49] [49]

5.8 0.90 0.9 5.25 1.33 0.48 1.17 0.20 0.27 0.26 0.06

[49] [49]

Homo-NOHA R or Ar = Butyl Pentyl 4,4,4-Trifluorobutyl Isopropyl Cyclopropyl Phenyl 4-F-Phenyl 4-Cl-Phenyl 4-CH3-Phenyl 4-OH-Phenyl 4-CH3O-Phenyl a b c d

55 F 10a 310 F 50 840 F 100 38 F 5c 60 F 10d 270 F 90 300 F 40 500 F 50 1100 F 300 425 F 50 2600 F 600

K m value for NOS I: 67 AM [51]. Unpublished results from our laboratory. K m value for NOS I: 56 AM [51]. K m value for NOS I: 40 F 6 AM [51].

320 280 780 200 80 130 350 98 295 110 145

F F F F F F F F F F F

50 50 100 40 10 15 80 15 50 15 50

b b b

[52] [52] [52] [52] [52] [52]

Alternative substrates of NO sythases

1115

Table 7. Formation of NO from the NOS-Catalyzed Oxidation of Monosubstituted N-Alkyl- and N-Arylguanidines, RNHC(=NH)NH2 Activity (%) Compound Ethyl 2-Methoxyethyl 2-Ethylthioethyl Isopropyl Propyl 3-Methoxypropyl 3,3,3-Trifluoropropyl Butyl 3-Butenyl 2-Butenyl (E-isomer) 2-Methylbutyl 3-Methylbutyl 3,3-Dimethylbutyl 4-Hydroxybutyl 4,4,4-Trifluorobutyl 4-Aminobutyl Pentyl Hexyl Cyclopropyl Cyclobutyl Cyclopentyl Cyclohexyl Cyclopropylmethyl 2-Methyl-cyclopropylmethyl Benzyl 4-Fluorobenzyl 3-Hydroxybenzyl 3-Aminobenzyl Phenyl 2-Fluorophenyl 3-Fluorophenyl 4-Fluorophenyl 4-Chlorophenyl 4-Methylphenyl 4-Hydroxyphenyl 4-Methoxyphenyl 4-Methylphenyl 4-Nitrophenyl 2-Thienyl

R

NOS I

NOS II

NOS III

Reference

CH3CH2 CH3O–(CH2)2 C2H5S–(CH2)2 (CH3)2CH CH3(CH2)2 CH3O–(CH2)3 CF3(CH2)2 CH3(CH2)3

n.d. b0.5 b0.5 0 b0.5 b0.5 b0.5 b0.5 0 b0.5 b0.5 b0.5 b0.5 n.d. b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 n.d. b0.5 b0.5 b0.5 0 b0.5 b0.5 b0.5 n.d. 0 0 0 0 0 b0.5

b0.5 2 b0.5 0 b0.5 2 12 6 0 2 b0.5 b0.5 3 b0.5 b0.5 35 2 11 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 b0.5 0 b0.5 b0.5 b0.5 b0.5 0 0 b0.5 0 0 4

n.d. b0.5 n.d. 0 n.d. b0.5 n.d. b0.5 0 n.d. b0.5 n.d. n.d. n.d. n.d. b0.5 n.d. b0.5 b0.5 b0.5 n.d. n.d. n.d. b0.5 b0.5 n.d. n.d. n.d. n.d. 0 n.d. n.d. n.d. n.d. 0 0 0 0 0 n.d.

a

H2C=CH(CH2)2 CH3CH=CHCH2 CH3CH2CH(CH3)CH2 CH3CH(CH3)CH2CH2 CH3C(CH3)2CH2CH2– HO–(CH2)4 CF3(CH2)3 H2N–(CH2)4 CH3(CH2)4 CH3(CH2)5 C3H5 C4H7 C5H9 C6H11 C3H5–CH2 CH3–C3H4–CH2 C6H5–CH2 4-F–C6H4–CH2 3-HO–C6H4–CH2 3-H2N–C6H4–CH2 C6H5 2-F–C6H4 3-F–C6H4 4-F–C6H4 4-Cl–C6H4 4-CH3–C6H4 4-HO–C6H4 4-CH3O–C6H4 4-CH3–C6H4 4-NO2–C6H4 C4H3S

a a

[50] a a a

[53] [50] a

[51] [53] [53] [53] [53] [53] [53] a

[53] a a a

[53] [51,53] [51] [53] a a a

[50] a a

[53] [53] [50] [50] [50,53] [50] [50] a

Activities are expressed as % of those found for L-Arg. n.d. = not determined. a Unpublished results from our laboratory.

the K m values did not correlate with any of the descriptors for electronic or hydrophobic properties, taken either one by one or by pairs. However, a straightforward linear relationship was found between logK m and a geometric descriptor, called B5, which is related to the maximal width of the substituent [52]. Selectivity of N-hydroxyguanidines as substrates for NOS I, II, and III Constitutive NOS I and NOS III seem to be more active than NOS II for NO formation from the oxidation of the smallest N-alkyl-NV-hydroxyguanidines, with R = propyl, isopropyl, and cyclopropyl (Table 4). For R = butyl, NOS II is as active as NOS I, NOS III being slightly less active. A further increase in the size of the R

substituent leads to compounds that are no longer substrates of NOS III. For instance, N-pentyl-NV-hydroxyguanidine is not substrate for NOS III, whereas it remains a good substrate for NOS I and II. Even bigger compounds (with R = benzyl or 2-phenylethyl for instance) are not substrates for NOS I and III, but give small amounts of NO upon oxidation by NOS II. A similar difference in behavior of the three NOS isoforms is observed with N-aryl-NV-hydroxyguanidines, because NOS II gives the best results for this family of relatively large hydroxyguanidines and NOS III is the least active isozyme. Thus, from all the tested compounds in this series, only one, with Ar = 3-aminophenyl, was found to produce NO upon oxidation by NOS III. NOS I also is poorly active toward the N-aryl-NV-hydroxyguanidines in

1116

D. Mansuy and J.-L. Boucher Table 8. Kinetic Constants Measured for the Formation of NO from NOS II-Catalyzed Oxidation of L-Arg and Some Alkylguanidines, RNHC(=NH)NH2

Compound L-Arg R= Butyl3-Methylbutyl 4,4,4-Trifluorobutyl Pentyl a

CH3(CH2)3 CH3CH(CH3)CH2CH2 CF3(CH2)3 CH3(CH2)4

K m (AM)

k cat (min1)

k cat/K m (min1 AM1)

Reference

5F1

400 F 50

80

[53]

0.5 0.05 0.8 0.25

[53]

45 F 10 1630 F 200 275 F 50 240 F 50

23 80 220 60

F F F F

5 10 50 10

a

[53] a

Unpublished results from our laboratory.

general. At the opposite end, NOS II is able to oxidize a large number of such hydroxyguanidines, some of them (Ar = 3- or 4-fluorophenyl) with a remarkably high activity (Table 5). Thus, several compounds, particularly those with Ar = 3-fluoro-, 4-chloro-, 4-methylphenyl-, and 3-thienyl-, are selectively oxidized to NO by NOS II. These data suggest that, despite the great sequence and structure identity of NOS I, II, and III, it should be possible to find alternative NO-producing substrates that are selective for a given isoform. Formation of NO from NOS-catalyzed oxidation of guanidines A study of the oxidation by recombinant NOSs of about 50 monosubstituted guanidines bearing various alkyl or aryl groups has clearly shown that it is much more difficult to find guanidines than N-hydroxyguanidines as NOproducing substrates. This is illustrated in Table 7, which shows that none of the tested guanidines was substrate for NOS I and NOS III. Moreover, only very few alkylguanidines and only one arylguanidine (Ar = 2-thienyl) gave significant formation of NO upon oxidation by NOS II [53]. Thus, monosubstituted guanidines with R = propyl, isopropyl, cyclopropyl, benzyl, or substituted phenyl are inactive, contrary to the corresponding N-hydroxyguanidines, and butylguanidine leads to only a low rate of NO formation compared to L-Arg (6%; Table 7). Many substituents or functions have been introduced into the butyl chain of butylguanidine as shown in Table 7. However, only one of the resulting compounds acts as a markedly better substrate for NOS. The best guanidine substrate found so far is N-(4,4,4-trifluorobutyl)guanidine (TFBG) whose oxidation by NOS II provides NO with a remarkable rate (35% of that found for L-Arg). N-pentylguanidine also gives NO, but with lower rates (Table 7). More detailed kinetic studies performed on NOS IIcatalyzed oxidation of TFBG [53] have shown characteristics very similar to those of NOS II-catalyzed oxidation of L-Arg: (i) both activities absolutely require the presence of NOS II containing H4B, NADPH, and

O2; (ii) they are strongly inhibited by usual NOS inhibitors such as N N-nitro-l-Arg; and (iii) NO is formed in the reaction. Formation of NO itself has been measured by the usual hemoglobin assay [55] and has been confirmed [53] by EPR spectroscopy, after trapping by the ferrous Fe-DETC2 complex [56]. The k cat value determined for NOS II-catalyzed formation of NO from TFBG (220 min1) is only 2-fold lower than that of L-Arg (Table 8). However, the K m value is 55fold higher. At the opposite end, N-butylguanidine seems to have a better affinity (K m = 45 AM) and a much lower k cat (23 min1), and N-pentylguanidine exhibits an affinity similar to that of TFBG and also a much lower k cat (60 min1) (Table 8). NOS II-catalyzed oxidation of TFBG to NO consumes 2.9 mol of NADPH per mole of NO produced at saturating concentrations of TFBG. Thus, in terms of decoupling between NADPH consumption and NO formation, TFBG is not as perfect as L-Arg, the oxidation of which consumes 1.5 mol of NADPH per mole of NO produced (the theoretical value according to Eqs. (1) and (2)). Its NOS II-dependent oxidation leads to a 52% coupling, which is quite remarkable for a non-a-amino-acid guanidine, which requires two successive steps to produce NO [53]. NO formation from NOS II-dependent oxidation of TFBG also occurs in intact cells [53]. This has been shown on mouse macrophages previously treated with lipopolysaccharide and interferon-g, a mixture known to induce the expression of NOS II in these cells. Addition of 2.5 mM TFBG to activated macrophages in an L-Arg-free medium resulted in the accumulation of NO2 after 30 h. The amount of NO2 produced increases with an increase in the number of cells, the incubation time, and the TFBG concentration, whereas it strongly decreases in the presence of usual NOS inhibitors. From these experiments, the concentration of TFBG leading to the half-maximum rate of NO2 formation was found to be approximately 250 AM, a value that is similar to the K m measured for oxidation of TFBG by purified, recombinant NOS II (Table 8).

Alternative substrates of NO sythases RELATIONSHIP BETWEEN THE STRUCTURE OF NO-PRODUCING ALTERNATIVE SUBSTRATES OF NOS AND THEIR NO-FORMATION EFFICIENCY

As mentioned above, from all the compounds containing a CjNH or CjNOH moiety that have been tested so far, only monosubstituted N-hydroxyguanidines and guanidines have been found to act as significant NOS substrates with NO formation. Relationship between the structure of N-substituted guanidines and N-hydroxyguanidines and their affinity for NOS II In order to know the importance of the binding of these substrates to the NOS active site in their oxidation with formation of NO, we have determined the dissociation constants of their complexes with recombinant NOS II (D. Lefevre-Groboillot et al., manuscript in preparation). This has been done by difference visible spectroscopy according to a previously described method [57] based on the displacement of imidazole bound to NOS II–Fe(III) by the studied guanidine or Nhydroxyguanidine, which leads to the appearance of a peak at 395 nm corresponding to the high-spin pentacoordinated NOS II–Fe(III) complex having bound the substrate. The main results of this study are the following:

(i)

The dissociation constants found for a guanidine and the corresponding N-hydroxyguanidine are

1117

generally similar, and the variations in these constants as a function of the nature of the R substituent are very similar for the RNHC(NH2)j NH and RNHC(NH2)jNOH compounds. (ii) After L-Arg and NOHA, the compounds exhibiting the highest affinity for NOS II are small Nalkylguanidines or -NV-hydroxyguanidines with R = cyclopropyl, propyl, and butyl (K d between 2 and 30 AM). The compounds for which R = pentyl are still well recognized by the NOS II active site, with lower affinities. However, a further increase in the chain length (R = hexyl) leads to very low affinities for NOS II (K d N 1 mM). (iii) Guanidines and N-hydroxyguanidines bearing small aryl substituents (R = phenyl, 4-fluorophenyl, or 4methylphenyl) bind to NOS II with lower affinities than small N-alkyl derivatives (K d between 200 and 800 AM). (iv) In both series of compounds (N-alkyl- and N-arylsubstituted derivatives), any increase in the size of the R substituent leads to a dramatic decrease of the affinity. Thus, introduction of fluoro, hydroxy, methyl, or amino substituents into the butyl chain results in a 10-fold increase in K d. Moreover, the affinity N-para-substituted-phenyl derivatives decreases when the size of the para-substituent increases. The above data may be explained by considering the X-ray structures that have been published for NOS–Nhydroxyguanidine (or guanidine) complexes. Actually, in addition to the structures reported for NOSoxy com-

Fig. 5. Comparison of the active site structures of the NOS Ioxy complexes with N-butyl-NV-hydroxyguanidine (A, edge view; C, top view), NOHA (B, edge view), and N-isopropyl-NV-hydroxyguanidine (D, top view), from Ref. [12]. The heme is in red, the substrates are in green, and some amino acid residues are in gray.

1118

D. Mansuy and J.-L. Boucher

plexes with L-Arg and NOHA [9-11,15], three structures of NOSoxy with N-substituted NV-hydroxyguanidines are currently available. They concern the NOS Ioxy–Nbutyl-NV-hydroxyguanidine [12], NOS Ioxy–N-isopropylNV-hydroxyguanidine [12], and NOS IIIoxy–N-(4-chlorophenyl)-NV-hydroxyguanidine [58] complexes. A comparison of these structures shows that the substrates always interact with two NOS sites. All of them interact with the guanidine (or N-hydroxyguanidine) binding site in which highly conserved residues, a glutamate (E371 in NOS II, E592 in NOS I, and E363 in NOS III), a tryptophan (W366, 587, and 358 in NOS II, I, and III, respectively), and a glycine (G365, 586, and 357 in NOS II, I, and III, respectively) establish ionic and/or hydrogen bonds with the NHC(NH2)jNH (or NOH) function. In all cases except that of the NOS Ioxy– N-isopropyl-NV-hydroxyguanidine complex, the glutamate carboxylic acid residue interacts with the N y and N N atoms of the substrate and the (NH)CO residue of the conserved tryptophan with the N N atom of the substrate. In the case of all complexes with N-hydroxyguanidines except N-isopropyl-NV-hydroxyguanidine, there is an additional hydrogen bond interaction between the (N)– OH hydroxyguanidine function and the NH(CO) function of the conserved NOS glycine. The binding of Nisopropyl-NV-hydroxyguanidine to NOS Ioxy is atypical, as the hydroxyguanidine function has undergone a 1208 rotation, its (N)OH group now interacting with the (NH)CO group of W587 [12] (Fig. 5). In addition to this strong binding with the NOS bguanidine (or N-hydroxyguanidine) binding site,Q these five substrates establish interactions between their R substituent and another NOS binding site. For the natural substrates, L-Arg and NOHA, the second NOS binding site specifically recognizes the a-amino-acid function of the molecule via a series of highly conserved amino acid residues of the protein and a carboxylate group of the heme. This highly specific recognition of a-amino-acid substrates thus implies two protein sites and a great number of selective ionic and hydrogen bond interactions. It explains why small changes at the level of the a-amino-acid or guanidine (or N-hydroxyguanidine) function, or in the chain between these two functions, lead to a dramatic loss of the NO-producing activity (Tables 1 and 2). The hydroxyguanidino moiety of N-butyl- and N-4chlorophenyl-NV-hydroxyguanidines is bound in a manner similar to that of NOHA [12,58] (Fig. 5). This leads to an orientation of the molecules that allows the terminal part of the alkyl or aryl group to be accommodated in a small hydrophobic site involving a conserved valine residue (V567 and 338 in NOS I and III, respectively). In the NOS Ioxy–N-isopropyl-NVhydroxyguanidine complex, the isopropyl group fits in

another small hydrophobic pocket formed by P565, V567, and F584 [12]. This should explain why non-aamino-acid substrates bearing a small, hydrophobic substituent bind well to NOS, because of favorable hydrophobic interactions with the alternative hydrophobic protein binding sites. Small alkyl substituents, such as R = Et, are too short to interact with these hydrophobic sites of the protein (Table 4), whereas alkyl or aryl substituents that are too large (R = hexyl, t-butyl, para-t-butylphenyl, etc.) cannot fit into the space available in the small hydrophobic sites (Tables 4 and 5). This could explain why the compounds exhibiting the highest affinities (after L-Arg and NOHA) for NOS are those for which R = cyclopropyl, propyl, and butyl and why the affinity of aryl substrates greatly decreases when the size of the aryl group increases. Relationship between the structure of N-substituted NV-hydroxyguanidines and their activity as NO-producing NOS substrates A relatively great number of such hydroxyguanidines are good substrates for NOS, at least for NOS II (Tables 4 and 5), and it seems that any hydroxyguanidine sufficiently well recognized by NOS active sites is oxidized with formation of NO. Quite remarkably, the k cat values determined for the oxidation into NO of many of them are high (between 10 and 170% of that of NOHA; Table 6). This is true not only for those exhibiting high affinities for NOS II (R = butyl or isopropyl), but also for some lowaffinity substrates (R = 4-methylphenyl or 4-methoxyphenyl). The variation in the K m values is in agreement with that found for the K d values; the lowest K m values are obtained for the compounds exhibiting the highest affinity for NOS IIoxy, with the order (decreasing affinity): butyl, isopropyl, cyclopropyl N pentyl N phenyl, 4-F-phenyl N 4Cl-phenyl, 4-OH-phenyl N 4-methylphenyl, 4-methoxyphenyl. Interestingly, N-3-thienyl-NV-hydroxyguanidine, which bears a small aryl substituent, exhibits a very low K m value, lower than that of NOHA (25 instead of 40 AM; Table 6). Finally, the efficiency of NO formation by NOS II, measured by the k cat/K m values, is comparable for the natural substrate NOHA and for butyl- and isopropylN-hydroxyguanidine (11, 6, and 5 min1 AM1, respectively). Relationship between the structure of N-substituted guanidines and their activity as NO-producing NOS substrates Whereas all the N-hydroxyguanidines well recognized by the NOS II active site act as good substrates, all the guanidines well recognized by NOS II are not oxidized with formation of NO. For instance cyclopropylguanidine, which exhibits the highest affinity for NOS IIoxy,

Alternative substrates of NO sythases

as well as all the arylguanidines tested so far, completely fails to produce NO upon oxidation by recombinant NOS II (Table 7). The origin of this different behavior of aryland alkylguanidines remains to be established; it could be related to the different pK a of these two series of compounds (D. Lefevre-Groboillot et al., manuscript in preparation). Thus far, only guanidines bearing a small alkyl residue have been found to produce NO upon NOS II-dependent oxidation (Table 7) [53]. However, the efficiency of their NOS II-catalyzed oxidation to NO is much weaker than those of the corresponding N-hydroxyguanidines. For instance, the k cat/K m value measured for N-butylguanidine (0.5 min1 AM1) is 160-fold lower than that found for L-Arg, whereas the k cat/K m value measured for the corresponding N-hydroxyguanidine (6 min1 AM1) is only two times lower than that found for NOHA (Tables 6 and 8). Obviously this is related to the greater difficulty for NOS to efficiently catalyze two successive, different oxidations on exogenous substrates different from L-Arg. For R = butyl, the K m values measured for the oxidation of the guanidine and the hydroxyguanidine are similar; however, the k cat value found for the guanidine is much lower than the one determined for the corresponding Nhydroxyguanidine. Any introduction of a substituent on the butyl chain of N-butylguanidine results in a dramatic increase in the K m value; Table 8 illustrates this in the case of the introduction of a methyl group or of fluoro substituents into the butyl chain. However, this also leads to an increase in the k cat value, TFBG being characterized by a k cat closer to that of L-Arg (Table 8). The best compromise between such increases in K m and k cat was obtained for TFBG, whose NOS II-catalyzed oxidation to NO occurs with a k cat/K m value close to 1 min1 AM1 (100-fold lower than that found for L-Arg) [53]. The molecular origin of these effects of the structure of R on the efficiency of the NOS-catalyzed oxidation of guanidines to NO remains to be determined. However, a very recent study of the relationship between the k cat of NOS II-dependent oxidation of a series of N-alkylguanidines to NO and the kinetic constants, k on and k off, of binding of guanidines and corresponding N-hydroxyguanidines to NOS II has shown that the k cat of oxidation of guanidines increases when the k off of the corresponding hydroxyguanidines decreases (D. Lefevre-Groboillot et al., manuscript in preparation). This indicates that an important factor for an efficient NO-producing guanidine is that the intermediate N-hydroxyguanidine metabolite is not too rapidly released from the NOS active site. CONCLUSION

During the past decade, a great number of compounds bearing a CjNH or CjNOH function have

1119

been synthesized and studied as possible NO-producing substrates of recombinant NOSs. This includes derivatives of L-Arg and NOHA, N-alkyl- (or aryl-) guanidines, N,NV- or N,N-disubstituted guanidines, N-alkyl(or aryl-) NV-hydroxyguanidines, and N (or O)-disubstituted NV-hydroxyguanidines, as well as amidoximes, ketoximes, and aldoximes. However, so far, only those involving the NHC(NH2)jNH (or NOH) moiety have led to a significant formation of NO. All the Nmonosubstituted NV-hydroxyguanidines that are well recognized by the NOS active site lead to NO with catalytic efficiencies (k cat/K m) up to 50% of that of NOHA. This is the case of many N-aryl- and N-alkyl-NVhydroxyguanidines, provided that the aryl or alkyl substituent is small enough to be accommodated by a NOS hydrophobic site located in close proximity to the NOS bguanidine binding site.Q As far as N-substituted guanidines are concerned, only a few compounds bearing a small alkyl group have been found to act as NO-producing substrates for NOS II. The k cat value found for the best compound may reach 55% of the k cat of L-Arg oxidation. However, the best catalytic efficiency, which was obtained with N-(4,4,4trifluorobutyl)guanidine, is only 100-fold lower than that of L-Arg. In a general manner, NOS II is a better catalyst than NOS I and III for the oxidation of exogenous guanidines and N-hydroxyguanidines to NO. This is particularly true for guanidines as the ones acting as substrates for NOS II have been found to be almost inactive for NOS I and NOS III. Thus, a good NO-producing guanidine substrate for the two latter isozymes remains to be found. Some Naryl-NV-hydroxyguanidines, with Ar = 4-chloro or 4methylphenyl, are substrates only for NOS II. However, selective N-hydroxyguanidines for NOS I and NOS II have not been found so far, even though compounds with Ar = 4-hydroxyphenyl and 3-aminophenyl are better substrates for NOS I and NOS III, respectively. Acknowledgment—The authors thank Sylvie Dijols, Catherine Moali, Axelle Renodon-Corniere, David Lefevre-Groboillot, and Magali Moreau for their important contributions to the results obtained in the authors’ laboratory that are described in this article. REFERENCES [1] Ignarro, L.; Murad, F., eds. Nitric oxide: biochemistry, molecular biology, and therapeutic implications. San Diego: Academic Press; 1995. [2] Stuehr, D. J.; Kwon, N. S.; Nathan, C. F.; Griffith, O. W.; Feldman, P. L.; Wiseman, J. N N-hydroxy-l-arginine is an intermediate in the biosynthesis of nitric oxide from l-arginine. J. Biol. Chem. 266:6259 – 6263; 1991. [3] Klatt, P.; Schmidt, K.; Uray, G.; Mayer, B. Multiple catalytic functions of brain nitric oxide synthase: biochemical characterization, cofactor-requirement, and the role of N N-hydroxy-larginine as an intermediate. J. Biol. Chem. 268:14781 – 14787; 1993.

1120

D. Mansuy and J.-L. Boucher

[4] Alderton, W. K.; Cooper, C. E.; Knowles, R. G. Nitric oxide synthases: structure, function and inhibition. Biochem. J. 357: 593 – 615; 2001. [5] Pfeiffer, S.; Mayer, B.; Hemmens, B. Nitric oxide: chemical puzzles posed by a biological messenger. Angew. Chem. Int. Ed. 38:1714 – 1731; 1999. [6] Moncada, S.; Higgs, E. A. Molecular mechanisms and therapeutic strategies related to nitric oxide. FASEB J. 9:1319 – 1330; 1995. [7] Wang, P. G.; Xian, M.; Tang, X.; Wu, X.; Wen, Z.; Cai, T.; Janczuk, A. J. Nitric oxide donors: chemical activities and biological applications. Chem. Rev. 102:1091 – 1134; 2002. [8] Siddhanta, U.; Presta, A.; Fan, B.; Wolan, D.; Rousseau, D. L.; Stuehr, D. J. Domain swapping in inducible nitric-oxide synthase: electron transfer occurs between flavin and heme groups located on adjacent subunits in the dimer. J. Biol. Chem. 273:18950 – 18958; 1998. [9] Crane, B. R.; Arvai, A. S.; Ghosh, D. K.; Wu, C.; Getzoff, E. D.; Stuehr, D. J.; Tainer, J. A. Structure of nitric oxide synthase oxygenase dimer with pterin and substrate. Science 279:2121 – 2126; 1998. [10] Raman, C. S.; Li, H.; Martasek, P.; Kral, V.; Masters, B. S.; Poulos, T. L. Crystal structure of constitutive endothelial nitric oxide synthase: a paradigm for pterin function involving a novel metal center. Cell 95:939 – 950; 1998. [11] Fischmann, T. O.; Hruza, A.; Niu, X. D.; Fossetta, J. D.; Lunn, C. A.; Dolphin, E.; Prongay, A. J.; Reichert, P.; Lundell, D. J.; Narula, S. K.; Weber, P. C. Structural characterization of nitric oxide synthase isoforms reveals striking active-site conservation. Nat. Struct. Biol. 6:233 – 242; 1999. [12] Li, H.; Shimizu, H.; Flinspach, M.; Jamal, J.; Yang, W.; Xian, M.; Cai, T.; Wen, E. Z.; Jia, Q.; Wang, P. G.; Poulos, T. L. The novel binding mode of N-alkyl-NV-hydroxyguanidine to neuronal nitric oxide synthase provides mechanistic insights into NO biosynthesis. Biochemistry 41:13868 – 13875; 2002. [13] Crane, B. R.; Rosenfeld, R. J.; Arvai, A. S.; Ghosh, D. K.; Ghosh, S.; Tainer, J. A.; Stuehr, D. J.; Getzoff, E. D. Nterminal domain swapping and metal ion binding in nitric oxide synthase dimerization. EMBO J. 18:6271 – 6281; 1999. [14] Ghosh, D. K.; Crane, B. R.; Ghosh, S.; Wolan, D.; Gachhui, R.; Crooks, C.; Presta, A.; Tainer, J. A.; Getzoff, E. D.; Stuehr, D. J. Inducible nitric oxide synthase: role of the N-terminal beta-hairpin hook and pterin-binding segment in dimerization and tetrahydrobiopterin interaction. EMBO J. 18:6260 – 6270; 1999. [15] Crane, B. R.; Arvai, A. S.; Ghosh, S.; Getzoff, E. D.; Stuehr, D. J.; Tainer, J. A. Structures of the N N-hydroxy-l-arginine complex of inducible nitric oxide synthase oxygenase dimer with active and inactive pterins. Biochemistry 39:4608 – 4621; 2000. [16] Raman, C. S.; Martasek, P.; Masters, B. S. S. The porphyrin handbook, Vol. 4. New York: Academic Press; 293 – 339; 2000. [17] Ortiz de Montellano, P. R., ed. Cytochromes P450: structure, mechanism, and biochemistry. New York: Plenum; 1995. [18] Hurshman, A. R.; Krebs, C.; Edmondson, D. E.; Huynh, B. H.; Marletta, M. A. Formation of a pterin radical in the reaction of the heme domain of inducible nitric oxide synthase with oxygen. Biochemistry 38:15689 – 15696; 1999. [19] Gorren, A. C.; Mayer, B. Tetrahydrobiopterin in nitric oxide synthesis: a novel biological role for pteridines. Curr. Drug Metab. 3:133 – 157; 2002. [20] Wei, C. C.; Crane, B. R.; Stuehr, D. J. Tetrahydrobiopterin radical enzymology. Chem. Rev. 103:2365 – 2383; 2003. [21] Mathieu, D.; Mathieu, D.; Battioni, P.; Boucher, J.-L.; Mansuy, D. Reaction between iron porphyrins and tetrahydropterins. J. Porph. Phthal. 8:265 – 278; 2004. [22] Wei, C. C.; Wang, Z. Q.; Wang, Q.; Meade, A. L.; Heman, C.; Hille, R.; Stuehr, D. J. Rapid kinetic studies link tetrahydrobiopterin radical formation to heme-dioxy reduction and arginine

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

hydroxylation in inducible nitric oxide synthase. J. Biol. Chem. 276:315 – 319; 2001. Hurshman, A. R.; Krebs, C.; Edmondson, D. E.; Marletta, M. A. Ability of tetrahydrobiopterin analogues to support catalysis by inducible nitric oxide synthase: formation of a pterin radical is required for enzyme activity. Biochemistry 42:13287 – 13303; 2003. Sorlie, M.; Gorren, A. C.; Marchal, S.; Shimizu, T.; Lange, R.; Andersson, K. K.; Mayer, B. Single-turnover of nitric-oxide synthase in the presence of 4-amino-tetrahydrobiopterin: proposed role for tetrahydrobiopterin as a proton donor. J. Biol. Chem. 278:48602 – 48610; 2003. Mathieu, D.; Frapart, Y.-M.; Bartoli, J.-F.; Boucher, J.-L.; Battioni, P.; Mansuy, M. Very general formation of tetrahydropterin cation radicals during reaction of iron porphyrins with tetrahydropterins: model for the corresponding NO-synthase reaction. J. Chem. Soc. Chem. Comm.:54 – 55; 2004. Wei, C. C.; Wang, Z. Q.; Hemann, C.; Hille, R.; Stuehr, D. J. A tetrahydrobiopterin radical forms and then becomes reduced during N N-hydroxyarginine oxidation by nitric oxide synthase. J. Biol. Chem. 278:46668 – 46673; 2003. Adak, S.; Wang, Q.; Stuehr, D. J. Arginine conversion to nitroxide by tetrahydrobiopterin-free neuronal nitric oxide synthase: implications for mechanism. J. Biol. Chem. 275:33554 – 33561; 2000. Fukuto, J. M.; Wallace, G. C.; Hszieh, R.; Chaudhuri, G. Chemical oxidation of N-hydroxyguanidine compounds: release of nitric oxide, nitroxyl and possible relationship to the mechanism of biological nitric oxide generation. Biochem. Pharmacol. 43:607 – 613; 1992. Fukuto, J. M.; Stuehr, D. J.; Feldman, P. L.; Bova, M. P.; Wong, P. Peracid oxidation of an N-hydroxyguanidine compound: a chemical model for the oxidation of N N-hydroxy-larginine by nitric oxide synthase. J. Med. Chem. 36:2666 – 2670; 1993. Sennequier, N.; Boucher, J.-L.; Battioni, P.; Mansuy, D. Superoxide anion efficiently performs the oxidative cleavage of CjNjOH bonds of amidoximes and N-hydroxyguanidines with formation of nitrogen oxides. Tetrahedron Lett. 36:6059 – 6062; 1995. Vadon LeGoff, S.; Boucher, J.-L.; Mansuy, D. Oxidation of arylamidoximes by various chemical and biomimetic systems: comparison with their oxidation by hemeproteins. C.R. Acad. Sci. Paris, Ser. IIc 3:785 – 792; 2000. Jousserandot, A.; Boucher, J.-L.; Henry, Y.; Niklaus, B.; Clement, B.; Mansuy, D. Microsomal cytochrome P450 dependent oxidation of N-hydroxyguanidines, amidoximes and ketoximes: mechanism of the oxidative cleavage of their CjN(OH) bond with formation of nitrogen oxides. Biochemistry 37: 17179 – 17191; 1998. Mansuy, D.; Boucher, J. L. Oxidation of N-hydroxyguanidines by cytochromes P450 and NO-synthases and formation of nitric oxide. Drug Metab. Rev. 34:593 – 606; 2002. Moali, C.; Boucher, J. L.; Renodon-Corniere, A.; Stuehr, D. J.; N Mansuy, D. Oxidations of N -hydroxyarginine analogues and various N-hydroxyguanidines by NO synthase II: key role of tetrahydrobiopterin in the reaction mechanism and substrate selectivity. Chem. Res. Toxicol. 14:202 – 210; 2001. Clague, M. J.; Wishnok, J. S.; Marletta, M. A. Formation of G N y-cyanoornithine from N -hydroxy-l-arginine and hydrogen peroxide by neuronal nitric oxide synthase: implication for mechanism. Biochemistry 36:14465 – 14473; 1997. Modolell, M.; Eichmann, K.; Soler, G. Oxidation of NG-hydroxyl-arginine to nitric oxide mediated by respiratory burst: an alternative pathway to NO synthesis. FEBS Lett. 401:123 – 126; 1997.

Alternative substrates of NO sythases [37] Stuehr, D. J.; Griffith, O. W. Mammalian nitric oxide synthases. Adv. Enzymol. Relat. Areas Mol. Biol. 65:287 – 346; 1992. [38] Grant, S. K.; Green, B. G.; Stiffey-Wilusz, J.; Durette, P. L.; Shah, S. K.; Kozarich, J. W. Structural requirements for human inducible nitric oxide synthase substrates and substrate analogue inhibitors. Biochemistry 37:4174 – 4180; 1998. [39] Feldman, P. L.; Chi, S.; Sennequier, N.; Stuehr, D. J. Synthesis of the l-arginine congener l-indospicine and evaluation of its interaction with nitric oxide synthase. Bioorg. Med. Chem. Lett. 6:111 – 114; 1996. [40] Moali, C.; Brollo, M.; Custot, J.; Sari, M. A.; Boucher, J. L.; Stuehr, D. J.; Mansuy, D. Recognition of alpha-amino acids bearing various CjNOH functions by nitric oxide synthase and arginase involves very different structural determinants. Biochemistry 39:8208 – 8218; 2000. [41] Fast, W.; Nikolic, D.; VanBreemen, R. B.; Silverman, R. B. Mechanistic studies of the inactivation of inducible nitric oxide synthase by N 5-(1-iminoethyl)-l-ornithine (L-NIO). J. Am. Chem. Soc. 121:903 – 916; 1999. [42] Li, X.; Atkinson, R. N.; King, S. B. Preparation and evaluation of new l-canavanine derivatives as nitric oxide synthase inhibitors. Tetrahedron 57:6557 – 6565; 2001. [43] Olken, N. M.; Marletta, M. A. N G-methyl-l-arginine functions as an alternate substrate and mechanism-based inhibitor of nitric oxide synthase. Biochemistry 32:9677 – 9685; 1993. [44] Moali, C.; Boucher, J.-L.; Sari, M.-A.; Stuehr, D. J.; Mansuy, D. Substrate specificity of NO synthases: detailed comparison of larginine, homo-l-arginine, their N N-hydroxy derivatives, and N Nhydroxy-nor-l-arginine. Biochemistry 37:10453 – 10460; 1998. [45] Lee, Y.; Marletta, M. A.; Martasek, P.; Roman, L. J.; Masters, B. S.; Silverman, R. B. Conformationally-restricted arginine analogues as alternative substrates and inhibitors of nitric oxide synthases. Bioorg. Med. Chem. 7:1097 – 1104; 1999. [46] Shearer, B. G.; Lee, S.; Franzmann, K. W.; White, H. A. R.; Sanders, D. C. J.; Kiff, R. J.; Garvey, E. P.; Furfine, E. S. Conformationally restricted arginine as inhibitors of human nitric oxide synthase. Bioorg. Med. Chem. Lett. 7:1717 – 1763; 1997. [47] Fishlock, D.; Perdicakis, B.; Montgomery, H. J.; Guillemette, J. G.; Lajoie, G. A. Synthesis and evaluation of trans 3,4-cyclopropyl larginine analogues as isoform selective inhibitors of nitric oxide synthase. Bioorg. Med. Chem. 11:869 – 873; 2003. [48] Nishida, C. R.; Knudsen, G.; Straub, W.; Ortiz de Montellano, P. R. Electron supply and catalytic oxidation of nitrogen by cytochrome P450 and nitric oxide synthase. Drug Metab. Rev. 34:479 – 501; 2002. [49] Dijols, S.; Perollier, C.; Lefevre-Groboillot, D.; Pethe, S.; Attias, R.; Boucher, J. L.; Stuehr, D. J.; Mansuy, D. Oxidation of N Nhydroxyarginine analogues by NO-synthase: the simple, non amino acid N-butyl-NV-hydroxyguanidine is almost as efficient an NO precursor as N N-hydroxyarginine. J. Med. Chem. 44:3199 – 3202; 2001.

1121

[50] Xian, M.; Fujiwara, N.; Wen, Z.; Cai, T.; Kazuma, S.; Janczuk, A. J.; Tang, X.; Telyatnikov, V. V.; Zhang, Y.; Chen, X.; Miyamoto, Y.; Taniguchi, N.; Wang, P. G. Novel substrates for nitric oxide synthases. Bioorg. Med. Chem. 10:3049 – 3055; 2002. [51] Jia, Q.; Cai, T.; Huang, M.; Li, H.; Xian, M.; Poulos, T. L.; Wang, P. G. Isoform-selective substrates of nitric oxide synthase. J. Med. Chem. 46:2271 – 2274; 2003. [52] Renodon-Corniere, A.; Dijols, S.; Perollier, C.; Lefevre-Groboillot, J. L.; Boucher, J. L.; Attias, R.; Sari, M. A.; Stuehr, D.; Mansuy, D. N-aryl-NV-hydroxyguanidines, a new class of NO-donors after selective oxidation by nitric oxide synthases: structure-activity relationship. J. Med. Chem. 45:944 – 954; 2002. [53] Dijols, S.; Boucher, J. L.; Lepoivre, M.; Lefevre-Groboillot, D.; Moreau, M.; Frapart, Y.; Rekka, E.; Meade, A. L.; Stuehr, D. J.; Mansuy, D. First non-alpha-amino acid guanidines acting as efficient NO precursors upon oxidation by NO-synthase II or activated mouse macrophages. Biochemistry 41:9286 – 9292; 2002. [54] Renodon-Corniere, A.; Boucher, J. L.; Dijols, S.; Stuehr, D. J.; Mansuy, D. Efficient formation of nitric oxide from selective oxidation of N-aryl-NV-hydroxyguanidines by inducible nitric oxide synthase. Biochemistry 38:4663 – 4668; 1999. [55] Murphy, M. E.; Noack, E. Nitric oxide assay using hemoglobin method. Methods Enzymol. 233:240 – 250; 1994. [56] Mulsch, A.; Vanin, A.; Mordvintcev, P.; Hauschildt, S.; Busse, R. NO accounts completely for the oxygenated nitrogen species generated by enzymic l-arginine oxygenation. Biochem. J. 288:597 – 603; 1992. [57] McMillan, K.; Masters, B. S. S. Optical difference spectrophotometry as a probe of rat brain nitric oxide synthase heme– substrate interaction. Biochemistry 32:9875 – 9880; 1993. [58] Raman, C. S.; Li, H.; Martasek, P.; Southan, G.; Masters, B. S.; Poulos, T. L. Crystal structure of nitric oxide synthase bound to nitro indazole reveals a novel inactivation mechanism. Biochemistry 40:13448 – 13455; 2001.

ABBREVIATIONS

FAD — flavin adenine dinucleotide FMN — flavin mononucleotide H4B — tetrahydrobiopterin NADPH — nicotinamide adenine dinucleotide phosphate hydrogen NOHA — NN-hydroxy-l-arginine NOS — nitric oxide synthase NOSoxy — NOS oxygenase domain TFBG — N-4,4,4,-trifluorobutylguanidine