Biochimica et Biophysica Acta 1411 (1999) 310^322

Review

The reactions of copper proteins with nitric oxide

Jaume Torres *, Michael T. Wilson
Department of Biological Sciences, University of Essex, Wivenhoe Park, Colchester CO4 3SQ, Essex, UK Received 6 July 1998; received in revised form 10 December 1998; accepted 16 December 1998

Abstract Nitric oxide (NO) can act as a ligand for copper atoms and may also engage in redox chemistry with the metal once bound. Furthermore NO posses an unpaired electron which can couple with the unpaired electron on Cu2 . These properties have been exploited to probe the active sites of copper-containing enzymes and proteins. We review these studies. In addition to the use as a spectroscopic probe for the active site we draw attention to the rapid reactions of NO at the copper sites in Cytochrome c oxidase (CcO) and laccase. These reactions in CcO occur in the ms time range, at low NO concentrations and in the presence of oxygen and may therefore be of physiological relevance to the control of respiration. Finally we speculate on the wider role that NO may play in regulation of an important group of Type 2 copper containing enzymes. 1999 Elsevier Science B.V. All rights reserved.
Keywords: Nitric oxide; Copper; Inhibition; Cytochrome c oxidase; Oxidase ; EPR; Rapid reaction

Contents
1. 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Use of NO as a probe for copper proteins . . . 2.1. Haemocyanin and tyrosinase . . . . . . . . . . 2.2. Multicopper oxidases . . . . . . . . . . . . . . . 2.3. Cytochrome c oxidase . . . . . . . . . . . . . . . 2.4. Photolabile complexes of copper proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 312 312 313 314 315 316 316 317

3.

Fast redox reactions between NO and copper proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Cytochrome c oxidase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Laccase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Implications of fast electron transfer from NO to copper for other copper-containing proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Mononuclear (T2 containing) copper proteins: future directions . . . . . . . . . . . . . 4.2. Peptidylglycine K-amidating enzyme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Copper amino oxidases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Galactose oxidase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4.

. . . . .

. . . . .

. . . . .

. . . . .

318 319 319 320 320

* Corresponding author. Fax: +44-1223 766002; E-mail: jt248@cam.ac.uk 0005-2728 / 99 / $ ^ see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 5 - 2 7 2 8 ( 9 9 ) 0 0 0 2 2 - 5

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 320

311

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction In this article, although short and limited in scope, we have attempted to do two things. We review the literature in which NO is used as a probe, largely spectroscopic, to elucidate the nature of the copper sites of a number of proteins. We have limited the choice of proteins and have concentrated on those that are not generally considered to interact with NO; thus we have omitted discussion of copper based nitrite reductase which is reviewed elsewhere in this volume. In addition, we also touch upon some possible physiological implications of the reactions of NO with copper proteins. This section is more speculative, as relatively less is known, and we have relied more on data we have ourselves recently gathered on cytochrome c oxidase (CcO) and, to a lesser extent laccase and have extrapolated from these to other proteins. We have included this latter discussion as we believe such reactions, redox in nature, may prove to be an important future research area and our speculation may encourage experimentation, if only to prove us wrong. Perhaps the most widely known biological function of nitric oxide (NO) is in acting as a signal transducer in neurotransmission [1] and vasodilation [2]. In these instances, NO interacts with guanylyl cyclase by binding to the haem iron, which in turn probably releases its proximal histidine via a repulsive trans eect [3]. The enzyme is thus activated, producing cGMP. Furthermore, NO is also known to interact with other iron proteins, both in their reduced and in the oxidised states (see review by Cooper, this issue). In addition to interacting with iron, NO also interacts with copper. Thus, NO has been used extensively as a probe to study the binuclear copper centres in haemocyanin and tyrosinase [4^10], the iron/copper binuclear centre in CcO [11^15] and the multicopper oxidases, ceruloplasmin (Cp) [8,9,16^ 20], ascorbate oxidase (AO) [8,9,21] and laccase [22,23]. Interactions with mononuclear blue copper proteins, like azurin or halocyanin have also been

reported [9,24,25]. These experiments have in common that they were performed under strictly anaerobic conditions, as NO is known to combine with oxygen to produce NO2 and other unwanted products. Also, incubation with NO was generally in the presence of saturated NO solutions and the reactions observed were slow. Such high (V1 mM) concentrations of NO are, however, far from those found under physiological conditions where NO concentrations are, apart from pathological conditions, below 1 WM. For the reactions between copper proteins and NO to attain physiological relevance, rapid spectral changes should be observed at low NO concentration and in the presence of oxygen. Rapid reactions have in fact recently been reported between NO and CcO [26^28]. In these reactions, that were also observed at low NO concentration and in the presence of oxygen, a primary role was assigned to the copper atom (CuB ) that, together with an iron atom (Fea3 ), constitutes the binuclear centre. A general mechanism was suggested [28] where NO acts as a fast one-electron reductant at Cu2 B which, once reduced, equilibrates with other redox centres in the enzyme. The particular conformation (`pulsed') [29] of the binuclear centre required for this NO reactivity explains why these reactions had not been observed before. This `pulsed' species, although unstable, decaying in minutes to a `resting' state, is the form of the enzyme that is physiologically relevant. This redox mechanism has been suggested to play an important role in the control of the activity of CcO, and is possibly relevant to the control of a number of other copper-containing enzymes. For example, comparable rapid redox reactions (t1=2 6 1 s) have been observed on mixing NO with tree laccase, an oxidase like CcO, but containing copper as the only metal. In view of these studies, it may prove protable in the future to re-examine the interaction of NO with a number of other copper proteins. In particular, future investigation should perhaps be aimed at targets like Cp that, as laccase, contain a trinuclear copper centre. On the other hand, in higher eucaryotes, copper often forms an

312

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

essential part of catalytic sites in enzymes that synthesise neurotransmitters and peptide hormones. A pathophysiological role for NO in these systems would constitute an exciting new area of research. 2. Use of NO as a probe for copper proteins NO has been used extensively both as a spin and redox probe in copper proteins. The copper centres in copper proteins are generally divided into three main types according to their EPR and optical features [30]: Type 1 (T1) or blue, which has a small parallel hyperne coupling (Ae V40^70U1034 cm31 or 43^75 G), type 2 (T2) with normal EPR features, and type 3 (T3), a binuclear EPR-undetectable site. Optically, T1 has a relatively strong band in the visible region at V600 nm, T3 absorbs typically around 330 nm, whereas T2 is practically undetectable. The use of NO has been particularly extensive in the study of dioxygen reactive centres that are constituted either by copper pairs (e.g. T3) [31,32], or by mixed iron/copper pairs, as in the binuclear centre of CcO [33]. In these enzymes, two closely associated paramagnetic metal atoms form the oxygen-binding site (Cu/Cu in T3 centres or Fe/Cu in CcO), and antiferromagnetic exchange, facilitated by a bridging ligand, leaves them EPR silent. The rationale behind the use of NO in these cases resides in its characteristic electronic conguration, with an unpaired electron placed in an antibonding Z* orbital that can be used to transform an even spin state (EPR silent) into an odd spin state observable by EPR spectroscopy. This is accomplished by binding of NO to one of the metal atoms in the pair, disrupting this coupling and leaving the other metal EPR detectable. In addition, NO possesses a complex redox chemistry and can act as a one- or two-electron oxidant, yielding NO3 or N2 O, or as a reductant, yielding NO or NO3 2 . The nature of the redox reaction between NO and a given copper atom depends, therefore, on the relative redox potential of the copper site and an NO/NOx couple, and also on the availability of a nearby electron acceptor/donor, e.g. a second copper atom in a coupled pair. The redox potentials of the various couples involving NO are discussed by Hughes in this volume. The roles of NO as a ligand

or as a redox reagent can often be dicult to disentangle. 2.1. Haemocyanin and tyrosinase The rst example of a stable compound formed between NO and a copper protein was reported for haemocyanin, an oxygen carrier protein found in invertebrates [34]. The green complex observed was suggested to arise from a copper nitrosyl. In haemocyanin (Hc), neither the oxygenated nor deoxygenated forms exhibit EPR signals. However, addition of NO to deoxyHc under anaerobic conditions gave rise to an EPR signal (amounting to about 50% of the available copper) characteristic of magnetic dipole^dipole Cu2 pairs [4]. Another signal was also observed, corresponding to an isolated Cu2 ion, which was interpreted as arising from an NO molecule bound to one of the copper atoms in the pair. This established the presence of a copper pair in the oxygen-binding site of Hc. These results were extended to tyrosinase [5], the enzyme ultimately responsible for the production of melanin. As in Hc, two types of EPR signal were detected. One had components in the region of g = 4 which were assigned to the dipole^dipole Cu2 pairs (doubly oxidised enzyme) and the other (g = 2) corresponded to isolated Cu2 ions in an axially symmetric environment (singly oxidised enzyme). It was thus suggested that tyrosinase, like haemocyanin, possesses a pair of copper atoms in the oxygen binding site: Cu1 Cu1 in its active (oxygen binding) state and Cu2 Cu2 in the inactive (non-oxygen binding) state. Two dierent species, doubly and singly oxidised, were also found by Van der Deen and Hoving [6], although isotopic substitution experiments failed to detect NO binding to a copper atom in the singly oxidised form. Subsequently [7], the reaction between NO and deoxyHc was separated into two kinetically distinguishable processes: an initial reaction, which yields metHc and N2 O, and a slower reaction, in which metHc lost its triplet EPR signal at g = 4 and yielded a nitrosyl derivative with g = 2. As these authors found 1 equivalent of NO per equivalent of EPR detectable Cu2 , they suggested that NO was bound to Cu1 and that the adjacent cupric Cu gave rise to the EPR signal, i.e. Cu2 ...Cu1 ^NO species. Thus, under strictly anaerobic conditions, both de-

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

313

oxy and metHc give, eventually, the same derivative. In 1989, Salvato et al. [10] re-examined the reaction between NO and Hc, and proposed that the redox reactions observed were due, not to NO, but to NO2 . The suggestion that NO2 was the oxidant in this reaction arose from the observation that under strictly anaerobic conditions or low NO concentration, they failed to detect any change. Also, they proposed that the nal derivative consisted of a half-met centre [Cu1 Cu2 ] with no NO bound to it. 2.2. Multicopper oxidases 2.2.1. Ceruloplasmin (ascorbate oxidase) Ceruloplasmin, the major copper protein in blood plasma, is the most complex of the multicopper oxidases, containing six copper atoms (i.e. one trinuclear cluster and three T1 coppers). The physiologically relevant activity of Cp seems to be the oxidation of Fe2 (the redox form in which iron enters the bloodstream from liver stores) to Fe3 , which is taken up by transferrin. Another important function of Cp is copper transport (Cp is the copper provider for CcO and Cu/Zn SOD). In 1973, the interaction of NO with oxidised [16] and reduced [17] Cp was reported. Incubation of oxidised Cp with NO for 5 min. caused a decrease in the 615-nm band (corresponding to T1 copper) and the appearance of a broad band at 400 nm. The EPR band corresponding to T1 was bleached, but no eect was observed on the T2 EPR signal. These changes were enhanced at low temperature. As the optical changes were (at least partially) reversible upon illumination, these results were interpreted as the formation of a diamagnetic charge transfer (CT) complex between NO and T1 copper. Illumination, however, did not restore the EPR signal. As a result of these studies, a stoichiometry of one T2 per two or three T1 copper atoms in Cp was proposed. Similar observations were reported by Van Leeuwen et al. for oxidised AO [21]. Studies using reduced Cp [17] detected the presence of EPR signals attributed to magnetically coupled binuclear copper clusters (T3). Van Leeuwen and van Gelder [18] extended the study of the interaction of NO with oxidised Cp using CD, measuring the eect of NO for longer incubation times (20 min). They concluded that the T1 copper centres were not equivalent in their redox

properties. Furthermore, at longer incubation times, the T1 copper was reduced, the eect was not entirely reversible upon NO removal and reversibility was only obtained after oxidation with oxygen. The T2 copper was also aected by incubation with NO as superhyperne patterns were observed superimposed on the EPR band corresponding to this centre. These patterns were not, however, assigned to NO binding. In addition, EPR signals compatible with the presence of magnetically coupled T3 coppers were also observed, as in Hc and tyrosinase. Evidence for the presence of such centres could previously only be provided from monitoring the absorbance the band at 330 nm. A further insight into the reactions of NO and Cp has been provided by using Cp lacking an EPR detectable T2 copper [20]. This Cp preparation is obtained by a rapid isolation procedure [35] and only ageing of this sample leads to the display of the T2 EPR band. These results support a much more important role for T2 in the interaction with NO. In this preparation NO is able to oxidise or reduce the enzyme depending on the latter's redox state. Upon ageing, Cp and the appearance of the T2 copper signal, the ability of the protein to be oxidised by NO is abolished. This suggests that the absence of oxidation previously reported was due to a defective preparation, i.e. when T2 copper becomes EPR detectable it no longer possesses its native structure. The structures of both AO [40] and Cp [41] are available and show each contains a trinuclear copper centre comprising T2 and T3 copper sites 2.2.2. Laccase Laccase catalyses the reduction of dioxygen to water and is able to take the four electrons necessary from a wide range of donors. The plant enzyme functions in lignin metabolism while the fungal protein is involved with pigment production, lignin degradation and detoxication. Laccase is a relatively simpler enzyme than CcO, which also reduces oxygen to water, in that it contains only copper. There are four copper atoms: T1, T2 and a binuclear T3 centre (see Scheme 3). T1 has a relatively strong band in the visible region (O615 V5700 M31 cm31 ), T3 absorbs at 330 nm (O330 V3600 M31 cm31 ) [36], whereas T2 is practically undetectable optically. Electron entry from the physiological substrate

314

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

into the enzyme is at T1, which transfers electrons to a trinuclear copper centre formed by T2 and T3 [37,38]. In this centre, analogous to the trinuclear centre found in AO [40] or Cp [41], oxygen binding and reduction take place. Two derivatives of this enzyme, where one of the copper atoms has been removed or substituted, have been extensively used in functional studies. In one of these, called T2 depleted (T2D), the T2 copper is removed by a chelator in the presence of reductants [42]. A second derivative is T1Hg, in which the copper atom in the T1 centre is substituted by the redoxinactive Hg2 [43]. The use of these derivatives has produced valuable information although, particularly in the case of T2D, not exempt from some controversy. For example, using T2D, Cole et al. showed that the T3 reduced copper atoms, unlike haemocyanin, do not bind oxygen [44]. This is in contrast with previous reports [45], in which oxygen was found to bind to T3 reduced in T2D laccase1 . Using T1Hg, on the other hand, it is possible to obtain an enzyme reduced with only three electrons, all of them in the trinuclear centre. The combination of fully reduced (three electrons) T1Hg with oxygen results in the formation of a species which is described as a hydroperoxide, bridged between an oxidised T3 copper and reduced T2 [44,47]. These authors have suggested that this intermediate precedes the formation of the native intermediate [38,39] in the normal catalytic cycle (Scheme 3). Rotilio et al. [22] reported that in tree laccase, T1
The product of these two T2D preparations appears to be dierent; whereas Cole et al. [44] report that the product after T2 removal contains the T3 pair reduced, Reinhammar and Oda [45] report that in their preparation, T2D contains T3 copper oxidised. Consequently, Reinhammar and Oda report that the oxygen reaction with T2D can only be observed following incubation with reductants which reduce T3. Cole et al. and Solomon et al. [49] have argued, however, that after such incubation, part of the T2 copper may be reconstituted by migration of reduced copper from other molecules, converting a fraction of the enzyme into the fully active form, which would explain the partial oxygen binding observed (40% of the molecules). The partial oxygen binding was attributed by Reinhammar and Oda to the observation that T3 copper is only partially reduced in T2D, even after 24 h incubation with the reductant. In the context of this discussion, it is worth mentioning that a crystal structure for T2D laccase has recently been obtained that supports the idea that the T3 coppers in T2D are reduced [46].
1

copper was reduced by NO. A systematic study was carried out by Martin et al. [23], who reported that NO could reduce both tree and fungal laccases, and reoxidise tree laccase. In these experiments, performed in the absence of oxygen and at 1 atm NO, T1 and T3 coppers were reduced simultaneously and reduction of T2 copper was slower. This reduction in fungal laccase was much faster than in tree laccase (t1=2 V2 min vs. t1=2 V70 min, respectively). This reduction was found to occur via a one-electron oxidation of NO to nitrite. From the small amount of nitrate that was also measured, it was concluded that a two-electron oxidation of NO to NO2 was not involved, as NO2 would decompose into both nitrite and nitrate. Also, the high nitrite concentration observed, relative to the enzyme concentration, indicated that a turnover existed in which the enzyme was alternatively reduced and reoxidised, in an analogous way to that observed for CcO [12] and also in Cp [20]. Oxidation was suggested to occur via the conversion of NO to N2 O, as the latter was detected in the gas phase by mass spectrometry. All these reactions were proposed to occur via the T2 copper. When a derivative lacking the T2 copper was used, T1 copper was reduced, but not T3, indicating probably that T3 reduction occurs as a concerted oxidation of T1 and T2 [23]. Furthermore, changes in the redox state of T1 were observed upon freezing in the presence of NO, reminiscent of the eects observed in Cp. Oxidation of T1 was detected when the fully reduced enzyme under NO was frozen, whilst this centre was reduced on freezing the partially reduced enzyme. Oxidation of T1 on freezing took place only when T2 was present, suggesting electron migration from T1 to an NO molecule bound to T2. In these experiments, the electron entry point from NO was postulated to be T2 (see below also). 2.3. Cytochrome c oxidase Cytochrome c oxidase is the terminal electron acceptor of the respiratory chain in the mitochondrion and in some bacteria. It contains two haem groups (Fea and Fea3 ) and two copper centres CuA and CuB . The site where oxygen binds and is reduced to water comprises Fea3 and CuB in close association and magnetically coupled. On addition of NO to the oxi-

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

315

dised binuclear centre of CcO the antiferromagnetic coupling between iron (S = 5/2) and copper (S = 1/2) is broken. The Cu2 B centre binds NO and remains EPR silent thus allowing the high spin ferric (g = 6) signal of Fea3 to be manifest [11] (Fig. 1). Both Fea3 and CuB , when reduced bind NO [12]. These authors reported complex redox reactions of NO and CcO; CcO was found to catalyse, under anaerobic conditions, reduction (to N2 O) or oxidation (to NO2 ) of NO, depending on the redox state of the enzyme. The same authors also observed production of NO when the reduced enzyme was incubated with nitrite. These reactions were similar to those reported using laccase (see above). Papers reporting the interaction between NO and CuB in CcO were published subsequently [13^15]. As for Cp, it was found that the complex formed between CuB and NO was light sen-

sitive [13^15], providing a way to investigate its optical absorption, which revealed bands at 640 nm and below 400 nm. 2.4. Photolabile complexes of copper proteins The susceptibility of copper^NO complexes to photodissociation has been studied in a number of other proteins [18]. These authors found that NO does not dissociate from T3 in Cp and Hc (although it is not clear whether NO is a ligand in these instances, see above). In azurin, a blue copper protein with only T1, similar changes to those reported for Cp (T1) were reported. Thus, upon incubation with NO at room temperature the optical band corresponding to T1 remains unchanged. When, however, the temperature was lowered to 200 K, the blue col-

Table 1 Summary of the interactions observed between NO and various copper types in copper proteins Enzyme Blue copper T1 enzymes Halocyanin Azurin Type 2 copper enzymes Superoxide dismutase Galactose oxidase Diamine oxidase Binuclear copper proteins Hemocyanin Tyrosinase Multicopper oxidases Ceruloplasmin Copper type Eect of NO Photodissociable Yes Yes Reference

Diamagnetic complex with T1 Diamagnetic complex with T1

[24,25] [18,24,25]

No eect Inhibition, but no spectral changes No eect

^ ^ ^

[18] [18] [18]

T3 T3

Type 3 oxidation and formation of half met derivative

No No

[2^10,13] [4^10]

T1

T3 T2 Ascorbate oxidase As in Cp Laccase (tree and fungal) T1 T2 T3 Cytochrome C oxidase

Diamagnetic complex with oxidized T1 T1 reduction at long incubation times Reoxidation of the fully reduced enzyme No eect, but see [18] Reduction and reoxidation Reduction and reoxidation NO binding Reduction and reoxidation

Yes

[8,9,16^20]

As in Cp ^ ^ ^

[18] [21] [22,23] [23] [23]

CuA CuB

No eect Binding and reduction

^ Yes

[11^15,26^28]

Table comprises data with the interpretation given by the original authors. In the case of ceruloplasmin, there is reason to believe [20] that some of the earlier work may have been performed on `damaged' protein. Consequently, some of the earlier conclusions may need to be reinterpreted in this light.

316

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

structural probe, the reactions between NO and the proteins taking many minutes to reach completion. At the time most of these studies were carried out, it was not appreciated that NO possesses a wide range of physiological functions in vivo. Fast redox reactions between NO and some copper enzymes have now been documented [26^28] and this may suggest that part of the physiological role of NO may be eected through reactions with copper proteins. It is to a consideration of these faster reactions that we now turn.
Fig. 1. EPR spectrum of 0.25 mM oxidized CcO under 1 atm NO and rapidly frozen. Temperature, 11 K; microwave power, 0.02 mW; modulation amplitude, 10 G; and microwave frequency, 9.25 GHz. Adapted from [12].

3. Fast redox reactions between NO and copper proteins 3.1. Cytochrome c oxidase Addition of NO to `pulsed' fully oxidised enzyme leads to a rapid (V100 s31 ) electron ejection from the binuclear centre to cytochrome a and CuA , and results in the reduction of V50% of the cytochrome a (Fig. 2), and a smaller percentage (V20%) of CuA [27]. The reactions of NO with the intermediates P (peroxy) and F (ferryl) present in the catalytic cycle of CcO have also been investigated. These species are depleted on reaction with NO in slower reactions, yielding the fully oxidised enzyme (O) [28]. A mechanism to account for these reactions has been proposed [28] in which the copper atom in the binuclear centre (CuB ) plays a central role. The initial events in these reactions are shown in Scheme 1, the rst step

our disappeared. By analogy with Cp, the bleaching of the blue colour was attributed to a diamagnetic CT complex between NO and Cu. This complex was found to be light sensitive, as the optical band was recovered after illumination (although not the EPR band, as in Cp and AO). A temperature-dependent electron distribution was suggested: Cu2 ^ NOHCu1 NO , in which the nitrosonium would be more populated at lower temperatures. The fact that photodissociation seemed to be associated with T1 copper (e.g. in AO, Cp or azurin) was used to suggest that CcO CuB might resemble a T1 copper [18], in contrast with other views comparing this centre to a T2 type site [11]. The observation that T1 copper sites bind NO in a photolabile complex has also been used to investigate the temperature dependence of the ligand binding equilibrium and the kinetics of the association reaction after photodissociation over a wide range of temperatures (80^280 K) in azurin and halocyanin [24,25]. These studies showed the presence of conformational substates in these enzymes. In contrast, the EPR and optical features of enzymes containing only T2 copper, e.g. superoxide dismutase (SOD), galactose oxidase (GO) or diamine oxidase (DO) were not aected by NO, even after 1 h incubation [18]. The eects observed on adding NO to copper proteins and the properties of the resulting products are summarised in Table 1. In the majority of the instances described, NO was used solely as a functional/

Fig. 2. Dierence spectra, relative to the oxidized CcO (species a) after mixing 1 mM NO with 7.5 WM pulsed CcO in a stopped-ow apparatus. The inset shows the time course of heme a reduction and a monoexponential t.

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

317

Scheme 1.

of which is NO binding to Cu2 B . Upon hydroxyla tion of the resulting nitrosonium (Cu1 B ^NO ) nitrous acid (HNO2 ) is formed, leaving CuB reduced. In CcO, the electron residing on CuB can then transfer to other redox centres within the enzyme, Fea , CuA or Fea3 [28]. In the latter case, oxygen intermediates would be reduced, leading to their depletion (Scheme 2). In agreement with this hypothesis, evidence was provided [28] showing that nitrite was rapidly formed within the binuclear centre following the addition of NO to the three species (O, P or F) (Fig. 3). This work suggested nitrosylation at Cu2 B instead of at 2 Fea3 is an early event in the fast inhibition of CcO by NO.

Fig. 3. Spectra relative to oxidized CcO after the addition of 10 mM sodium nitrite to pulsed CcO (999) and after the addition of 40 WM NO to species F (- - -). The visible region has been expanded (U3). Concentration of CcO, 4.7 WM. The buffer was HEPES 0.1 M, pH 7.4, containing 0.5% Tween 80.

3.2. Laccase In contrast with the early experiments, in which NO was used, under anaerobic conditions, as a probe (see above) we have recently obtained results in the presence of oxygen and using relatively low NO concentrations (40 WM versus 2 mM in previous experiments). Surprisingly, addition of NO to resting oxidised (as prepared) tree laccase from Rhus vernicifera in the presence of oxygen resulted in very rapid optical changes to produce a spectrum (Fig. 4) similar to that which has been assigned to a peroxide intermediate by Solomon and colleagues [48]. This interpretation is open to question because similar spectra have been observed by Reinhammar (personal communication) after the addition of reductants to laccase under anaerobic conditions, thus excluding peroxide formation. It remains clear, however, that NO in the presence of oxygen reacts rapidly with tree laccase as prepared. This reaction has been further investigated by EPR spectroscopy, which can monitor the redox states of the T1 and T2 centres. The

changes in the EPR spectrum on addition of NO to laccase are shown in Fig. 5. This gure shows that the intensity of the low eld component (MI = 33/2), corresponding to the oxidised T2 copper, drops V30^40% after the addition of NO (Fig. 5, insert), although the percentage of T2 reduction could be even higher (see gure legend). This is followed by slow reoxidation. One possible explanation for the optical and EPR changes may be postulated in Scheme 4, in which the enzyme as prepared is partially reduced and reduction of T2 by

Scheme 2.

Fig. 4. Dierence spectra relative to the resting oxidized enzyme (as prepared) after the addition of an aliquot of 2 mM NO (nal concentration, V100 WM) to an aerobic ([O2 ]V240 WM) solution containing V70 WM laccase in HEPES 100 mM pH 6.0. The spectrum was collected immediately after the addition of NO. No spectral change was observed on addition of NO to the anaerobic enzyme.

318

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

Scheme 3. Main species present in the catalytic cycle of laccase. For the sake of clarity, the steps from the native intermediate to the fully reduced enzyme have been simplied. Also, most of the hydroxide groups that bridge either between the two T3 coppers, or between T2 and T3, have not been represented. According to [49], these bridging groups are present only when the two bridged copper atoms are simultaneously oxidized.

NO leads to the formation of the peroxide intermediate under oxygen. This interpretation depends on fully reduced T3 not binding O2 and on the presence

Fig. 5. EPR spectra of oxidized resting laccase at pH 6 (65 WM) measured at 77 K. The low eld component of the T2 copper signal (MI = 33/2) is indicated by an arrow. Inset: intensity of this component changing during the experiment and normalized respect to the intensity obtained after 1 h. The rst two points (before addition of 100 WM NO) were taken as a control, and the following measurements were taken at dierent time points (0.5, 1, 10, 11, 60 and 68 min) after the addition of NO. The line represents an exponential t to the data.

of fully reduced T3 in a fraction of the enzyme as prepared [36], but is in contradiction with results of Reinhammar and Oda [45] and must presently be considered tentative. The reaction in Scheme 1 should not necessarily inhibit laccase if it only leads to reduction of the T2 copper. However, production of nitrite within the trinuclear cluster could generate some inhibitory effect. An inhibitory eect could also be expected if, for example, NO and reduced copper formed a relatively stable complex Cu1 ^NO. Clearly, more experiments are needed to clarify any inhibitory eects of NO on laccase and, by extrapolation, on other enzymes containing a trinuclear centre (e.g. AO and Cp) and to elucidate the mechanism of this inhibition. 4. Implications of fast electron transfer from NO to copper for other copper-containing proteins We suggest that the reactions reported for CcO

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

319

Scheme 4. This scheme provides a tentative interpretation of the results obtained on mixing NO with laccase as prepared. Copper Types 1, 2 and 3 are represented as in Scheme 3 (center). The scheme incorporates the conclusions drawn by Solomon and coworkers (see text), namely that in a fraction of the enzyme as prepared (here V25%) the T3 coppers are reduced and that reduced T3 copper does not bind oxygen unless the T2 copper is also reduced. Where T2 and T3 copper centres are reduced, oxygen binding and electron transfer can occur leading to the formation of a bound peroxide intermediate. Addition of NO leads to reduction of T2 in all molecules, but only in that fraction of the enzyme in which T3 is reduced does this lead to reaction with oxygen. Although this scheme is consistent with experimental observations, the non-reactivity of reduced T3 is not in agreement with the conclusions of Reinhammer (see text).

and laccase may hold implications for the regulation of proteins that utilise copper for catalysis, i.e. copper proteins that can use dioxygen either as an electron acceptor or to add functional groups to organic substrates. These include the relatively well-known Cp (vertebrates) (particularly in the T2 EPR-non detectable form) and AO (plants), but also recent additions to this list, such as the bacterial phenoxazinone synthase, the fungal bilirubin oxidase and sulochrin oxidase or FET3, a ferroxidase found in the yeast Saccharomyces cerevisae (see [49] for review). If NO were to react rapidly with Cp, as it does with CcO and laccase, then NO may have a profound inuence on the ferroxidase activity of Cp. In this respect, Cp has been found to inhibit Fenton chemistry-induced oxidative damage to deoxyribose, lipids and DNA [50], possibly via Fe2 depletion. If NO inhibited the ferroxidase activity, this would hold implications for the pathological effects of NO in sepsis. 4.1. Mononuclear (T2 containing) copper proteins: future directions The possible interaction of NO with mononuclear copper proteins [51] has not received much attention. In fact, no eect on the EPR parameters of the copper was found when NO was reacted with SOD or amine oxidase [18]. We believe, nevertheless, that it may be protable to re-examine the reactions of T2 copper proteins with NO, especially considering the extreme variability in reactivity found in other cop-

per enzymes depending on the preparation (in Cp [20,35] and in CcO [26^28]) and on the variability between dierent species (as in fungal or plant laccases [23]). Dopamine L-monooxygenase (DLM) and peptidylglycine K-amidating enzyme (PAM), found in neurosecretory vesicles, are included here because, although containing two copper atoms (both T2), only one is involved with catalysis. DLM catalyses the conversion of dopamine (DA) to norepinephrine (NE), inserting an atom of oxygen into the benzylic position of the ethylamine side chain of dopamine (see [52], for review). DLM contains two typical T2 cupric ions (Cu2 ) which are reduced by ascorbate. These two Cu atoms, however, have dierent functions. One (CuA ) interacts with the reductant (ascorbate) whereas the other (CuB ) binds substrate and product (DA, NE and oxygen) and catalyses the activation of the C^H bond in DA. Either of these atoms could bind (and be reduced by) NO, as in Scheme 1. However, CO is known to bind only to one of them (CuB ), in a manner competitive with oxygen [53]. It is normally assumed that the rst species formed is a peroxide intermediate [54], followed by formation of Cu(II)^OW as hydroxylating agent via formation of a tyrosyl radical. In this case, NO could also inhibit this enzyme by interacting with either radical. 4.2. Peptidylglycine K-amidating enzyme As DLM, PAM also contains two T2 copper

320

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322

atoms. This enzyme intervenes in the synthesis of peptide hormones, catalysing the incorporation of an amide linkage at the C-terminus (e.g. thyrotropin, calcitonin, oxytocin or substance P). In fact, half of the peptide hormones are synthesised in this way. Both the mechanism of catalysis and coordination of the copper atoms are assumed to be similar in DLM and PAM. In both enzymes, the copper sites are reduced by ascorbate and oxidised in the presence of the substrate. Reduction of these metal sites has been found to lead to dramatic structural changes [55], which could possibly be mimicked by the reaction depicted in Scheme 1. The similarity in the reaction mechanism also suggests, as in DLM, the possibility of interaction of NO with a radical. 4.3. Copper amino oxidases The ubiquitous copper amino oxidases (CAO) can be found in bacteria, yeast, plants and mammals, and contain only one copper (T2). These enzymes catalyse the two-electron oxidation of amines to aldehydes, reducing oxygen to H2 O2 . In doing so, they most probably regulate the levels of important amines, such as dopamine or histamine. As they have only one copper, they use an organic cofactor (topaquinone, TPQ), covalently bound to the peptide backbone, to provide the second reducing equivalent. As in DLM or PAM, the copper sites suer a change in conguration on reduction, which results in lower co-ordination number around the metals. These changes are expected to create a site for oxygen interaction with the reduced copper and, as for DLM or PAM, they could be induced by such reactions with NO as depicted in Scheme 1. 4.4. Galactose oxidase Another mononuclear copper enzyme is fungal galactose oxidase, GO, which catalyses the oxidation of alcohols to aldehydes. Copper is directly complexed to an oxygen atom donated by tyrosine residue (Tyr-272) which, in turn, is cross-linked via a thioether to a cysteine [56]. NO binds specically to this enzyme with a 1:1 stoichiometry, but to a site other than the copper atom as the copper EPR spectrum remains largely (90%) unchanged [57]. It has been shown that NO inhibits GO (C.E. Cooper, per-

sonal communication) and thus this inhibition is possibly mediated via interaction of NO with this site, which may coincide with the tyrosyl radical. Another possible role of NO in the regulation of copper proteins could be related to the post-translational modications of some of these enzymes. In CAO and GO, for example, this modication produces the cofactor TPQ or the modied tyrosyl radical, respectively, via the modication of a tyrosine residue. Furthermore, in the case of CAO, there is convincing evidence that copper is responsible for TPQ biogenesis [58,59], and it is also likely to be the case for the modied tyrosyl radical in GO. Although the mechanism for the formation of these cofactors is still unclear, one may speculate that at least in CAO, which is found in mammals, NO could play a role in the regulation or impeding the formation of the cofactors, as copper has been found to be involved in these processes. 5. Conclusion As we have shown, perhaps the most characteristic feature of the interaction between NO and copper in biological systems is the dual, and sometimes dicult to distinguish, nature of NO as a ligand and as a redox reagent. There are, in addition, several examples of a subtle, temperature-dependent, equilibrium between these two extremes. Recent studies indicate that redox reactions can be very fast, even at low NO concentrations in the presence of oxygen, and hence, of biological importance. In connection with these ndings, the recent synthesis of a complex between mononuclear copper and NO [60], constitutes an important step in the characterisation of the interaction that takes place between NO and copper in biological systems. References
[1] J. Garthwaite, C.L. Boulton, Nitric oxide signaling in the central nervous system, Annu. Rev. Physiol. 57 (1995) 683^706. [2] S. Moncada, A. Higgs, The L-arginine nitric oxide pathway, New Engl. J. Med. 329 (1993) 2002^2012. [3] T.G. Traylor, V.S. Sharma, Why NO?, Biochemistry 31 (1992) 2847^2849.

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322 [4] A.J.M. Schoot Uiterkamp, Monomer and magnetic dipolecoupled Cu2+ EPR signals in nitrosylhemocyanin, FEBS Lett. 20 (1972) 93^96. [5] A.J.M. Schoot Uiterkamp, H.S. Mason, Magnetic dipolecoupled Cu2+ pairs in nitric oxide treated tyrosinase: a structural relationship between the active sites of tyrosinase and hemocyanin, Proc. Natl. Acad. Sci. USA 70 (1973) 993^ 996. [6] H. Van der Deen, H. Hoving, Nitrite and nitric oxide treatment of Helix pomatia hemocyanin single and double oxidation of the active site, Biochemistry 16 (1977) 3519^ 3525. [7] J. Verplaetse, P. Van Tornout, G. Defreyn, R. Witters, R. Lontie, The reaction of nitrogen monoxide and of nitrite with deoxyhemcyanin and methaemocyanin of Helix pomatia, Eur. J. Biochem. 95 (1979) 327^331. [8] R. Wever, R. Boelens, E. De Boer, B.F. Van Gelder, A.C.F. Gorren, H. Rademaker, J. Inorg. Biochem. 23 (1985) 227^ 232. [9] A.C.F. Gorren, E. De Boer, R. Wever, The reaction of nitric oxide with copper proteins and the photodissociation of copper^nitric oxide complexes, Biochim. Biophys. Acta 916 (1987) 38^47. [10] B. Salvato, G.M. Giacometti, M. Beltramini, F. Zilio, G. Giacometti, R.S. Magliozzo, J. Peisach, The oxidation of octopus-vulgaris hemocyanin by nitrogen-oxides, Biochemistry 28 (1989) 680^684. [11] T.H. Stevens, G.W. Brudwig, D.F. Bocian, S.I. Chan, Structure of cytochrome a3^Cua3 couple in cytochrome c oxidase as revealed by nitric oxide binding studies, Proc. Natl. Acad. Sci. USA 76 (1979) 3320^3325. [12] G.W. Brudvig, T.H. Stevens, S.I. Chan, Reactions of nitric oxide with cytochrome c oxidase, Biochemistry 19 (1980) 5275^5285. [13] R. Boelens, H. Rademaker, R. Pel, R. Wever, EPR studies of the photodissociation reactions of cytochrome c oxidase^ nitric oxide complexes, Biochim. Biophys. Acta 679 (1982) 84^94. [14] R. Boelens, R. Wever, B.F. Van Gelder, H. Rademaker, An EPR study of the photodissociation reactions of oxidized cytochrome c oxidase^nitric oxide complexes, Biochim. Biophys. Acta 724 (1983) 176^183. [15] R. Boelens, H. Rademaker, R. Wever, B.F. Van Gelder, The cytochrome c oxidase-azide-nitric oxide complex as a model for the oxygen binding site, Biochim. Biophys. Acta 765 (1984) 196^209. [16] R. Wever, F.X.R. Van Leeuwen, B.F. van Gelder, The reaction of nitric oxide with ceruloplasmin, Biochim. Biophys. Acta 302 (1973) 236^239. [17] F.X.R. Van Leeuwen, R. Wever, B.F. van Gelder, EPR study of nitric oxide-treated ceruloplasmin, Biochim. Biophys. Acta 315 (1973) 200^203. [18] F.X.R. Van Leeuwen, B.F. van Gelder, A spectroscopic study of nitric oxide-treated ceruloplasmin, Eur. J. Biochem. 87 (1978) 305^312. [19] L. Calabrese, F. Malatesta, D. Barra, Purication and prop-

321

[20]

[21]

[22]

[23]

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34] [35]

[36]

erties of bovine ceruloplasmin, Biochem. J. 199 (1981) 667^ 673. G. Musci, S. Di Marco, M.C. Bonaccorsi di Patti, L. Calabrese, Interaction of nitric oxide with ceruloplasmin lacking an EPR detectable type 2 copper, Biochemistry 30 (1991) 9866^9872. F.X.R. Van Leeuwen, R. Wever, B.F. van Gelder, L. Avigliano, B. Mondovi, The interaction of nitric oxide with ascorbate oxidase, Biochim. Biophys. Acta 403 (1975) 285^ 291. G. Rotilio, L. Morpurgo, M.T. Graziani, M. Brunori, The reaction of nitric oxide with Rhus vernicifera laccase, FEBS Lett. 54 (1975) 163^166. C.T. Martin, R.H. Morse, R.M. Kanne, H.B. Gray, B.G. Malstrom, S.I. Chan, Reactions of nitric oxide with tree and fungal laccase, Biochemistry 20 (1981) 5147^5155. D. Ehrenstein, G.U. Niehaus, Conformational substates in azurin, Proc. Natl. Acad. Sci. USA 89 (1992) 9681^9685. D. Ehrenstein, M. Filiaci, B. Scharf, M. Engelhard, P.J. Steinbach, G.U. Niehaus, Ligand binding and protein dynamics of cupredoxins, Biochemistry 34 (1995) 12170^12177. J. Torres, C.E. Cooper, M.T. Wilson, Possible interaction of nitric oxide with the ferryl intermediate of cytochrome c oxidase, Biochem. Soc. Trans. 450S (1996) 24. C.E. Cooper, J. Torres, M. Sharpe, M.T. Wilson, Nitric oxide ejects electrons from the binuclear centre of cytochrome c oxidase by reacting with oxidised copper : a general mechanism for the interaction of copper proteins with nitric oxide?, FEBS Lett. 414 (1997) 281^284. J. Torres, C.E. Cooper, M.T. Wilson, A common mechanism for the interaction of nitric oxide with the oxidized binuclear centre and oxygen intermediates of cytochrome c oxidase, J. Biol. Chem. 273 (1998) 8756^8766. E. Antonini, M. Brunori, A. Colosimo, C. Greenwood, M.T. Wilson, Oxygen `pulsed' cytochrome c oxidase: functional properties and catalytic relevance, Proc. Natl. Acad. Sci. USA 74 (1977) 3128^3132. B. Reinhammar, B.G. Malmstrom, in: T.G. Spiro (Ed.), Copper Proteins, Wiley Interscience, New York, 1981, chapter 3. ngstro L. Petersson, J. A m, A. Ehrenberg, Magnetic susceptibility of laccases and ceruloplasmin, Biochim. Biophys. Acta 526 (1978) 311^317. D.M. Dooley, R.A. Scott, J. Ellinghaus, E.I. Solomon, H.B. Gray, Magnetic susceptibility studies of laccase and oxyhemocyanin, Proc. Natl. Acad. Sci. USA 75 (1978) 3019^3022. B.F. Van Gelder, H. Beinert, Studies of the heme components of cytochrome c oxidase by EPR spectroscopy, Biochim. Biophys. Acta 189 (1969) 1^24. re , A. Schneider, C.R. Se ances Soc. Biol. Fil. (1919) Ch. Dhe 1041^1043. L. Calabrese, M. Carbonaro, G. Musci, Chicken ceruloplasmin. Evidence in support of a trinuclear cluster involving type 2 and 3 copper centers, J. Biol. Chem. 263 (1988) 6480^6483. L.-S. Kau, D.J. Spira-Solomon, J.E. Penner-Hann, K.O.

322

J. Torres, M.T. Wilson / Biochimica et Biophysica Acta 1411 (1999) 310^322 Hodgson, E.I. Solomon, X-ray absorption-edge determination of the oxidation-state and coordination-number of copper ^ application to the type-3 site in Rhus-vernicifera laccase and its reaction with oxygen, J. Am. Chem. Soc. 109 (1987) 6433^6442. M.D. Allendorf, D.J. Spira, E.I. Solomon, Low temperature magnetic circular dichroism studies of native laccase: spectroscopic evidence for exogenous ligand bridging at a trinuclear copper active site, Proc. Natl. Acad. Sci. USA 82 (1985) 3063^3067. B. Reinhammar, Chem. Scripta 25 (1985) 172^176. L.-E. Andreasson, R. Bra nden, B. Reinhammar, Kinetic studies of Rhus vernicifera laccase. Evidence for a multielectron transfer and an oxygen intermediate in the reoxidation reaction, Biochim. Biophys. Acta 438 (1976) 370^379. A. Messerschmidt, H. Luecke, R. Huber, X-ray structures and mechanistic implications of 3 functional-derivatives of ascorbate oxidase from zucchini-reduced, peroxide and azide forms, J. Mol. Biol. 230 (1993) 997^1014. I. Zaitseva, V. Zaitsev, G. Card, K. Moshov, B. Bax, A. Ralph, P. Lindley, The X-ray structure of human serum ceruloplasmin at 3.1 angstrom: nature of the copper centres, J. Biol. Inorg. Chem. 1 (1996) 15^23. M.T. Graziani, L. Morpurgo, G. Rotilio, B. Mondovi, Selective removal of type 2 copper from Rhus vernicifera laccase, FEBS Lett. 70 (1976) 87^90. M.M. Morie-Bebel, M.C. Morris, J.L. Menzie, D.R. McMillin, Mixed-metal derivative of laccase containing mercury (II) in the type-1 binding-site, J. Am. Chem. Soc. 106 (1984) 3677^3678. J.L. Cole, G.O. Tan, E.K. Yang, K.O. Hodgson, E.I. Solomon, Reactivity of the laccase trinuclear copper active-site with dioxygen- an x-ray absorption-edge study, J. Am. Chem. Soc. 112 (1990) 2243^2249. B. Reinhammar, Y. Oda, J. Inorg. Biochem. 11 (1979) 115^ 117. V. Ducros, A.M. Brzozowski, K.S. Wilson, S.H. Brown, P. stergaard, P. Schneider, D.S. Yaver, A.H. Pedersen, G.J. Davies, Crystal structure of the type-2 Cu depleted laccase from Coprinus cinereus at 2.2 angstrom resolution, Nat. Struct. Biol. 5 (1998) 310^316. J.L. Cole, D.P. Ballou, E.I. Solomon, Spectroscopic characterization of the peroxide intermediate in the reduction of dioxygen catalyzed by the multicopper oxidases, J. Am. Chem. Soc. 113 (1991) 8544^8546. W. Shin, V.M. Sundaram, J.L. Cole, H.H. Zhang, B. Hedman, K.O. Hodgson, E.I. Solomon, Chemical and spectroscopic denition of the peroxide-level intermediate in the multicopper oxidases: relevance to the catalytic mechanism of dioxygen reduction to water, J. Am. Chem. Soc. 118 (1996) 3202^3215. [49] E.I. Solomon, U.M. Sundaram, T.E. Machonkin, Multicopper oxidases and oxygenases, Chem. Rev. 96 (1996) 2563^ 2605. [50] J.M.C. Gutteridge, G.J. Quinlan, Antioxidant protection against organic and inorganic oxygen radicals by normal human plasma: the important primary role for iron binding and iron-oxidizing proteins, Biochim. Biophys. Acta 1156 (1992) 144^150. [51] J.P. Klinman, Mechanisms whereby mononuclear copper proteins functionalize organic substrates, Chem. Rev. 96 (1996) 2541^2561. [52] L.C. Stewart, J.P. Klinman, Dopamine L-hydroxylase of adrenal chroman granules: structure and function, Annu. Rev. Biochem. 57 (1988) 551^592. [53] N.J. Blackburn, T.M. Pettingill, K.S. Seagraves, R.T. Shigeta, Characterization of a carbon-monoxide complex of reduced dopamine L-hydroxylase ^ evidence for inequivalence of the Cu(I) centers, J. Biol. Chem. 265 (1990) 15383^15386. [54] N. Ahn, J.P. Klinman, Mechanism and modulation of dopamine L-monooxygenase by pH and fumarate as deduced from initial rate and primary deuterium isotope eect studies, Biochemistry 22 (1983) 3096^3106. [55] J.S. Boswell, B.J. Reedy, R. Kulathila, D.J. Merkler, N.J. Blackburn, Structural investigations of the coordination environment of the active site copper centres of recombinant bifunctional peptidylglycine K-amidating enzyme, Biochemistry 35 (1996) 12241^12250. [56] N. Ito, S.E.V. Phillips, C. Stevens, Z.B. Ogel, M.J. McPherson, J.N. Keen, K.D.S. Yadav, P.F. Knowles, 3-Dimensional structure of galactose-oxidase ^ an enzyme with a built-in secondary cofactor, Faraday Discuss. 93 (1992) 75^84. [57] M.M. Whittaker, J.W. Whittaker, Ligand interactions with galactose oxidase: mechanistic insights, Biophys. J. 64 (1993) 762^772. [58] R. Matsuzaki, T. Fukui, H. Sata, Y. Ozaki, K. Tanizawa, Generation of the topa quinone cofactor in bacterial monoamino oxidase by cupric ion-independent autooxidation of a specic tyrosyl residue, FEBS Lett. 351 (1994) 360^364. [59] Y.H. Choi, R. Matsuzaki, T. Fukui, E. Shimizu, T. Yorifuji, H. Sata, Y. Ozaki, K. Tanizawa, Copper/topa quinone-containing histamine oxidase from Arthrobacter globiformis. Molecular cloning and sequencing, overproduction of precursor enzyme and generation of topa quinone cofactor, J. Biol. Chem. 270 (1995) 4712^4720. [60] C.E. Ruggiero, S.M. Carrier, W.E. Antholine, J.W. Whittaker, C.J. Cramer, W.B. Tolman, Synthesis and structural and spectroscopic characterization of mononuclear copper nitrosyl complexes ^ models for nitric-oxide adducts of copper proteins and copper-exchanged zeolites, J. Am. Chem. Soc. 115 (1993) 11285^11298.

[37]

[38] [39]

[40]

[41]

[42]

[43]

[44]

[45] [46]

[47]

[48]