Comparison of wild type neuronal nitric oxide synthase and its Tyr588Phe mutant towards various l-arginine analogues

https://doi.org/10.1016/j.jinorgbio.2010.06.001Get rights and content

Abstract

Crystal structures of nitric oxide synthases (NOS) isoforms have shown the presence of a strongly conserved heme active-site residue, Tyr588 (numbering for rat neuronal NOS, nNOS). Preliminary biochemical studies have highlighted its importance in the binding and oxidation to NO of natural substrates L-Arg and Nω-hydroxy-l-arginine (NOHA) and suggested its involvement in mechanism. We have used UV–visible and EPR spectroscopy to investigate the effects of the Tyr588 to Phe mutation on the heme-distal environment, on the binding of a large series of guanidines and N-hydroxyguanidines that differ from L-Arg and NOHA by the nature of their alkyl- or aryl-side chain, and on the abilities of wild type (WT) and mutant to oxidize these analogues with formation of NO. Our EPR experiments show that the heme environment of the Tyr588Phe mutant differs from that of WT nNOS. However, the addition of L-Arg to this mutant results in EPR spectra similar to that of WT nNOS. Tyr588Phe mutant binds L-Arg and NOHA with much weaker affinities than WT nNOS but both proteins bind non α-amino acid guanidines and N-hydroxyguanidines with close affinities. WT nNOS and mutant do not form NO from the tested guanidines but oxidize several N-hydroxyguanidines with formation of NO in almost identical rates. Our results show that the Tyr588Phe mutation induces structural modifications of the H-bonds network in the heme-distal site that alter the reactivity of the heme. They support recent spectroscopic and mechanistic studies that involve two distinct heme-based active species in the two steps of NOS mechanism.

Graphical abstract

Comparison of spectroscopic and catalytic properties of wild type and Tyr588Phe mutant of neuronal nitric oxide synthase towards l-arginine analogues.

  1. Download : Download full-size image

Introduction

In mammals, nitric oxide, NO, is produced by three isoforms of nitric oxide synthases (NOSs) that can be distinguished according to their initial cellular identifications, primary sequences and modes of regulation [1], [2], [3]. All three NOSs are heme-thiolate proteins that display significant structural and mechanistic similarities. NOSs have a homodimeric bidomain structure consisting of an NH2-terminal oxygenase domain and a CO2H-terminal reductase domain. The oxygenase domain binds the heme prosthetic group, substrate l-arginine (L-Arg) and cofactor (6R)-5,6,7,8-tetrahydro-l-biopterin (H4B) in close proximity to the heme, whereas the reductase domain contains binding sites for flavins, FMN and FAD, and for substrate NADPH. These two domains are bound by a calmodulin (CaM) binding sequence that triggers the electron flow from the reductase domain to the oxygenase domain as a function of free Ca2+ concentration [4], [5]. All three NOSs catalyze the heme-dependent oxidation of L-Arg to l-citrulline and NO with intermediate formation of Nω-hydroxy-l-arginine (NOHA) (Scheme 1) [6], [7].

NO acts as a signalling agent in a huge range of physiological effects such as neurotransmission and vasodilatation. It is also involved in several pathologies associated to oxidative stress phenomena such as atherosclerosis and neurodegenerative diseases. Most of these diseases are linked to the concomitant formation of superoxide, hydrogen peroxide and peroxynitrite by NOSs [8], [9], [10], [11]. Since the discovery that NOSs are important sources for superoxide and hydrogen peroxide, the elucidation of NOSs mechanism constitutes a major objective for the biomedical community [12], [13], [14]. The development of treatments for these diseases constitutes an important pharmacological challenge and many studies are undertaken to discover potent and selective inhibitors of NO-synthases [15], [16].

In analogy with cytochromes P450, the following mechanism for NOS is presently proposed for NO formation from L-Arg. The first step of the NOS-dependent biosynthesis of NO (L-Arg hydroxylation to NOHA) would involve the formation of a ferric-peroxo complex upon one-electron reduction of the NOS–FeII–O2 complex by H4B. Double protonation of the peroxo complex leads to an oxo-ferryl complex ((Por°+)FeIV=O) believed to be responsible for the hydroxylation of the guanidine moiety of L-Arg [3], [17], [18], [19], [20], [21]. The second step of catalysis (NOHA oxidation to citrulline and NO) is believed to follow the same pathway up to the formation of the ferric-peroxo complex. Then, the catalytic sequence would involve either a direct nucleophilic attack of the hydroperoxo group [21] or of the peroxo species [22] on the hydroxyguanidinium carbon of NOHA. This reaction is followed by a rearrangement of the resulting tetrahedral complex with ultimate release of NO from an intermediate FeIII–NO complex [23], [24].

Crystal structures of the oxygenase domains of the three mammalian NOSs have shown very strong active-site conservation [25], [26], [27], [28], [29]. These structures identified key amino acid residues involved in the binding of substrates L-Arg and NOHA, and of H4B cofactor. They showed that these two substrates bind in an almost identical manner with their α-amino acid function as a key factor for binding at the NOSs active sites and a highly conserved crucial Glu residue (Glu592 in rat nNOS) H-bonded with the α-NH2 and the guanidino- (or N-hydroxyguanidino) group of L-Arg (or NOHA) [29]. Furthermore, the crystallographic studies revealed a H-bond network involving L-Arg, diatomic ligands, and an active-site water molecule that could be involved in proton shuttling during catalysis [30], [31]. They also identified H-bonds between the α-CO2H group of L-Arg (or NOHA) and the OH-group of a highly conserved Tyr residue (Tyr588 in rat nNOS) to be of importance for recognition of both substrates at the heme active site (Fig. 1A). The Tyr588 aryl ring is close to the heme and to the aromatic rings of Trp587 and H4B, suggesting that this residue could play important roles in substrate binding and/or catalysis.

NOSs not only catalyze the oxidation of L-Arg and NOHA but also the oxidation of some non α-amino acid alkylguanidines (Gua) [32], [33], and N-alkyl- and N-aryl-N′-hydroxyguanidines (NOHG) with formation of NO in reactions very similar to the oxidation of L-Arg and NOHA to citrulline and NO [33], [34], [35], [36]. Different binding modes are reported for some aryl- and alkyl-NOHG in the substrate-binding pocket of NOSs [29], [37]. In the case of n-butyl-NOHG, as observed with L-Arg and NOHA, the critical H-bonds established between the terminal NH2 group and the δ-NH with Glu592 are conserved but the end of the butyl group curls towards Gln478 and Val567, away from the binding site for L-Arg or NOHA α-amino acid moiety (Fig. 1B) [29]. Different orientations are observed for close analogues. In the case of i-propyl-NOHG, the terminal NH2 and NHOH groups of the N-hydroxyguanidine moiety always establish H-bond with Glu592 of nNOS and the methyl groups of the i-propyl moiety make van der Waals contacts with side chains of both Val567 and Phe584 but a clearly distinct orientation is observed for the OH group in comparison with n-butyl-NOHG [29]. In the case of 4-chlorophenyl-NOHG, the aryl-group is directed towards the heme-propionate of eNOS [37]. Anyway, the NOHG moiety of these small compounds remains in close proximity of the heme and reacts with the heme-active species to form NO [33], [34], [35], [36].

In a preliminary study, we have studied the effects of Tyr588 mutations in nNOS on the binding affinity of L-Arg and NOHA and on the formation of NO from these two substrates. The Tyr588Phe mutation increased the dissociation constants for L-Arg and NOHA (8- and 2.8-fold, respectively), slightly reduced the formation of NO from L-Arg but stimulated the formation of NO from NOHA [38]. Similar effects were observed when the corresponding mutation was introduced into eNOS: the Tyr357Phe mutant of eNOS exhibited increased dissociation constants for L-Arg and NOHA (62- and 36-fold, respectively), a strongly reduced rate of NO formation from L-Arg but an almost identical rate of NO formation from NOHA [39]. The origins of these effects remained unclear. It was proposed that they could be linked to structural rearrangements in the distal site with changes in the substrate-binding orientation and in the H-bonds network implied in electron and proton transfers. The involvement of distinct heme-based active species for the two steps of catalysis was also proposed [38], [39].

In order to better understand the role(s) of the OH-group of Tyr588 at the active site of nNOS, we have investigated the impact of its deletion on the structural and catalytic properties of the Tyr588Phe mutant. We have compared for the first time the EPR spectra of heme–FeIII complexes of wild type (WT) nNOS and of the mutant as isolated and after the addition of L-Arg. We have studied the ability for WT and Tyr588Phe nNOSs to bind a large series of guanidines and N-hydroxyguanidines that do not bear an α-amino acid function. Finally, we have compared the ability for WT nNOS and Tyr588Phe to oxidize this series of Gua and NOHG with formation of NO. Our study underlines strong differences in the distal environment of the heme–FeIII complexes of these proteins. It demonstrates the role of the Tyr588 OH-group in the binding of α-amino acids and in the H-bond network crucial to the first step of catalysis. These data obtained with a large series of compounds supports a mechanism for NOS that involves two distinct heme-active species that was previously based on experiments using L-Arg and NOHA only.

Section snippets

Materials

H4B was purchased from Alexis (Coger, Paris, France). Calmodulin, 2′,5′-ADP-Sepharose, CaM-Sepharose and Sephadex G25 were products of Amersham-Pharmacia Biotech Inc. l-Arginine, NADPH, NOHA, superoxide dismutase (SOD), catalase, hemoglobin, and all other reagents were obtained from Sigma and were of the highest purity commercially available.

The synthesis and the physico-chemical characteristics of Gua 4–17 and NOHG 19–30 have been described elsewhere [32], [34], [40]. Their structures are

Spectroscopic properties of WT and Tyr588Phe nNOS

Using a previously described strategy [38], we have expressed and purified WT nNOS and its Tyr588Phe mutant from transformed E. coli in identical yields (4–5 mg prot/L of culture). Native WT nNOS and the Tyr588Phe mutant purified in the absence of L-Arg and in the presence of H4B exhibited broad Soret peaks around 400 nm that are characteristic of a mixture of HS- and LS-heme–FeIII complexes with a majority of heme–FeIII complex in the pentacoordinated (HS) state [42]. After reduction with sodium

Discussion

In previous studies, it has been observed that the Tyr588Phe mutation in nNOS increased the dissociation constants for L-Arg and NOHA, reduced the formation of NO from L-Arg but stimulated the formation of NO from NOHA [38]. The corresponding mutation in eNOS led to almost identical conclusions [39]. In the present work, we have compared the spectroscopic and catalytic properties of WT and Tyr588Phe nNOS in the presence of a large series of L-Arg and NOHA analogues that differ by modification

Abbreviations

    H4B

    (6R)-5,6,7,8-tetrahydro-l-biopterin

    CaM

    calmodulin

    DETC

    diethyldithiocarbamate

    DTT

    dithiothreitol

    Gua

    guanidine

    Hepes

    N-(2-hydroxyethyl)piperazine-N′-2-ethane sulfonic acid

    HS

    high spin

    ImH

    imidazole

    NOHA

    Nω-hydroxy-L-Arg

    NOHG

    N-hydroxyguanidine

    n, i, and eNOS

    neuronal, inducible, and endothelial nitric oxide synthase respectively

    LS

    low spin

    SOD

    superoxide dismutase

    WT

    wild type

Acknowledgments

The authors thank M.-A. Sari, B. Ramassamy, and M. Jaouen (UMR 8601) for their help in the preparation of NOS and determination of NOS activities.

References (49)

  • A.C.F. Gorren et al.

    Biochim. Biophys. Acta

    (2007)
  • H.M. Abu-Soud et al.

    J. Biol. Chem.

    (1994)
  • D.J. Stuehr et al.

    J. Biol. Chem.

    (1991)
  • P. Klatt et al.

    J. Biol. Chem.

    (1993)
  • S. Pou et al.

    J. Biol. Chem.

    (1999)
  • P. Tsai et al.

    Biochem. Pharmacol.

    (2005)
  • B.R. Babu et al.

    Curr. Opin. Chem. Biol.

    (1998)
  • N. Bec et al.

    J. Biol. Chem.

    (1998)
  • J. Tejero et al.

    J. Biol. Chem.

    (2008)
  • C.C. Wei et al.

    J. Biol. Chem.

    (2003)
  • D.J. Stuehr et al.

    J. Biol. Chem.

    (2004)
  • C.S. Raman et al.

    Cell

    (1998)
  • D. Mansuy et al.

    Free Radic. Biol. Med.

    (2004)
  • Y. Sato et al.

    Biochem. Biophys. Res. Commun.

    (2001)
  • P.F. Chen et al.

    Arch. Biochem. Biophys.

    (2003)
  • D.J. Stuehr et al.

    J. Biol. Chem.

    (1992)
  • M.E. Murphy et al.

    Methods Enzymol.

    (1994)
  • J.C. Salerno et al.

    J. Biol. Chem.

    (1995)
  • W.K. Alderton et al.

    Biochem. J.

    (2001)
  • D.K. Ghosh et al.

    Front. Biosci.

    (2003)
  • B.S. Masters et al.

    FASEB J.

    (1996)
  • S. Moncada et al.

    Pharmacol. Rev.

    (1991)
  • B. Heinzel et al.

    Biochem. J.

    (1992)
  • W.A. Pryor et al.

    Am. J. Physiol.

    (1995)
  • Cited by (0)

    View full text