CORE Metadata, citation and similar papers at core.ac.uk

Provided by Elsevier - Publisher Connector

Vahmc 282, rwnbcr 1, I-J PEls OS)hQO hprll l%Jl @31091 Fccicmrian of Eurspenn Biochcmieal Soclcricd 00145793/9l/si3,$0 AlR%‘lS OOl457939lQO309K

Status of the identity in copper oxidative

J.P. KlinnxuQ, D.M. Dooley2, J.A. Duinti3, P.F. Knowles4, B. MondoviS and J.J. Villafran&

Reccivcd29 January 1991

Much conflicting data have appeared in the litcraturc regarding the nature of the active site structures responsible for catalysis in thtcc classcr of copper cneymcs: the copper amine oxidascs, dopamine b-monooxygenasr and galactose oxidasc. Although pyrroloquinolinc quinonc has been proposed to be the active site cofactor in each instance, new findings indicate this is not the case. Instead. recently available data indicate a spectrum of strategies far substrate activation, which range from direct metal catalysis (dopamine /I-monooxygcnase) to the involvement of protein-derived radicals (galactosc oxidasc) and protein-derived quinoncs (copper amine oxidases).

Copper protein; Pyrroloquinolinc quinonc; Topa yuinonc

1. INTRODUCTION their reactivity toward carbonyl reagents such as phenyl- hydrazine, to yield chromophoric derivatives. In this A meeting on the subject of ‘Copper in Biological context, early investigations focused on a role for Systems’ was held on September 23-28, 1990 in as the reactive carbonyl [I]. Manziana, Italy. This meeting brought together a However, the presence of this cofactor has never been critical mass of investigators who have been pursuing directly demonstrated. Both NMR spectroscopy and cofactor structure in copper oxidative enzymes over the resonance Raman studies have been carried out. on past five to ten years. During this period, much con- several amine oxidases. ” P-NMR studies of porcine flicting data have appeared in the literature regarding serum amine oxidase failed to detect the expected signal the true nature of the active site structures responsible for the phosphate group of pyridoxal phosphate [2], for catalysis in three classes of copper enzymes: the cop- although such studies cannot rule out pyridoxal itself. per amine oxidases, dopamine ,&monooxygenase and Resonance Raman spectra of phenylhydrazine and galactose oxidase, At tlie present time considerable con- nitrophenylhydrazine derivatives of amine oxidases fusion exists among investigators not directly involved from porcine and bovine sera, pea seedlings, porcine in studies of these enzymes. The goal of this paper is a kidney and Arthrobacter P1 indicate different spectra clarification of the status of our knowledge regarding from the analogous derivatives of pyridoxal phosphate each system. [3]. Chemical evidence also argues strongly against a simple carbonyl cofactor. Amine substrates have been shown to be competitive with regard to hydrazine in- 2. COPPER AMINE OXIDASES hibitors of amine oxidases and, as such, are expected to Copper amine oxidases are widespread and have been form Schiff base intermediates. Yet no tritium transfer studied from bacteria, plants and mammals. Historical- to could be dernonstrated upon incubation of ly, three different cofactor structures have besn propos- bovine serum amine oxidase with substrate and NaBT4 ed for these enzymes. [4]. More recently, it has been shown that NaCNBH3 treatment of bovine serum amine oxidase with substrate 2. I, Pyridoxal phosphate leads to the incorporation of one mol of ‘“C-labeled A prominent feature of the copper amine oxidases is substrate per mol of enzyme subunit ‘concomitant with enzyme inactivation. Once again, no iransfer of tritium Correspondence uddress: J.P. Klinman, Department of Chemistry, from NaCNk3Ts to enzyme could be detected, leading to University of California, Berkeley, CA 94920, USA the proposal of a dicarbonyl-type cofactor structure

Published by Efsevier Science J-ktblishersB. V. 1 which would release rritium from the initially reduecd rcncrivity of the two curbonyl groupr wilt be discussed complex [S], The numerous lines of investigation, below, HPLC purification of a thermolytie digestion outlined above, argue strongly against a role for indicated one major radioactive peptide peak in ca, pyridoxal phosphate in the copper amine oxidasea, 40% yield. Micro-Edman sequencing ur this peptidc Wowcver, reports of this cofactor in swum amine OX- gave: lau, asn, blank, asp, tyr. Subsequent charaetrrizn- idaae have appeared in the recent literature [6,7]. tion of this pentapeptide by mass spectromcrry and ‘H- NMR led to assignment of the blank at position 3 ta the oxidized form of a 2,4,S-rrihydroxyphenylalanine 2.2. Pyrroloquinoline qrtirione residue, referred to as topa [16), This assign- Two independent reports, published in 1984 (8,9], ment has been confirmed by chemical synthesis of a focused attention on a newly discovered bacterial co- topa quinonc model compound and by comparison of factor, pyrrolsquinolinc quinone, as the covnlenrly this model to the protein-derived peptidc using ‘H- bound cofactor in bovine serum amine oxidnse. The NMR [IG] and resonance Raman spectroscopy [IS]. A approach developed by the Delft group [IO] involved total of 75% of the radiolabel originally associated with prolonged incubation of enzyme with dinitrophesyl- protein could be recovered from the HPLC. A second hydrazinc in the presence of oxygen. Pronasc digestion radiolabeled pcptide, obtained in ea. 10% yield, was of derivatized enzyme was found to release a product scqucnced and found to be comparable to the original which showed the same retention time on HPLC and pcntapcptide (with the addition of two amino acids at ‘H-NMR spectrum as an authentic addact of pyrrolo- its C-terminus). Thus, a total of ca, 50% of prorein- quinone with dinitrophenylhydrazine. Extcn- associated radiolabel can be unambiguously assigned co sion of this methodology to a number of other amine the phcnylhydrazone of topa quinone [ 191. Since all of oxidases, which include kidney amine oxidase [IO], lysyl the protein-bound phcnylhydrazine shows similar oxidase [ 1 I], methylamine oxidasc from Arrhrobacm chromophoric properties, it seems unlikely that the re- Pl [ 121 and amine oxidase from pea seedlings [13], maining C-14 in derivatized enzyme is bound to a func- yielded similar results. Resonance Raman spectroscopy tional group different from topa quinone. In a recent of phenylhydrazone derivatives of various amine ox- study [18], resonance Raman spectra for the idases was also pursued, While protein-dinitrophenyl- phenylhydrazine and nitrophenylhydrazine derivatives hydrazone spectra showed a much greater similarity to of bovine serum amine oxidase were compared to the dinitrophenylhydrazone derivative of pyrrolo- spectra for the respective, protein-derived peptides. The quinoline quinone than pyridoxal phosphate, signifi- high degree of similarity observed between peptide and cant differences in the intensity and position of peaks protein spectra indicates that the topa quinone in were apparent [14]. Subsequent comparison of reso- isolated peptide is not an artifact of the isolation pro- nance Raman spectra of 2-hydrazinopyridine and p- cedure, Since the above described studies were carried nitrophenylhydrazine derivatives has indicated that pyr- out on a single amine oxidase (from bovine serum), it is roloquinoline is not the cofactor in amine oxidases [3]. important to question the generality of topa quinone as It is now believed that the combination of prolonged an active site structure in copper amine axidases. At the treatment of enzyme with phenylhydrazine, followed by current time, both resonance Raman spectroscopy [ 181 pronase digestion, leads to the formation of a cieriuative and sequencing of active site peptides [19] support a of a native structure and does not represent pyrrolo- structure similar to topa quinonc in the amine oxidases quinoline quinone itself at the enzyme active site. The from porcine serum and kidney, pea seedling, yeast and chemistry which produces pyrrol,oquinoline quinone is bacteria. Lysyl oxidase may be an important exception unknown and in need of further study. The requirement to this generalization, since no sequence homology to for oxygenation of reaction solutions to produce high active site peptides obtained from other amine oxidases yields of pyrroloquinoline quinone-like molecules [ 1O] can be discerned by examination of the lysyl amine ox- suggests the importance of redox reactions. Buechi et idase c-DNA derived protein sequence [20]. al. have already shown that can be converted to One unresolved issue concerns the relative reactivity pyrroloquinoline quinone in an oxidative manner [ 151. of the two prosthetic groups present in dimeric amine oxidase. While some of us find similar reactivity at both 2.3. Topa quinone sites [21J, others observe biphasic kinetics and/or varia- A recent study has focused on the proteolysis of a tions in spectral properties following derivatization of [‘%]phenylhydrazine derivative of bovine serum amine bovine [17,22] and porcine serum amine oxidase [23]. oxidase under mild conditions [16]. These studies used The structural basis for the observation of differential a highly purified preparation of enzyme, which shows a reactivities at the two active site cofactors is in need of titration end point with phenylhydrazine close to one clarification. A comparative study of the ratio of cop- mol incorporated per enzyme subunit. Others authors per to active site prosthetic groups, using enzyme had titrated only one mol of phenylhydrazine per dimer preparations from different laboratories, may help to using similar enzyme preparations [17]; the relative rationalize existing discrepancies (cf. [24]).

2 Vulumc 281. nunrbcr I

procedure (U.W, Green nnd SA. Buinc, unpublished results), Thus, all available evidence paints tawnrrl the The kinetic mcrhnninm af”dupeminc &msnooxygen- pnrricipzxtian of n tyrosyl rndical in rha catulytie asc has been rlgarously pursued through a combination meelinniam of galnctosr oxidasr. of kinetics and spcctrsseopy [25], Results from these studies show that rcdox cycling at the two capper ntoms 5. FINAL COMMENTS per subunit can account for the catnlycic mechanism. A possible ralc for nn organic cafnctor wns sug~esred The authors wish to point out the exciting new through the use OF phcnylhydrazine derivatives as car- dcvclopmcnts which have emerged concerning the use banyt renpcntx [26], In these studies, Duine ancl co- of amino acid derivatives as redox cofactors. With workers [2G] isolated a pyrroloquinolinc quinonc-like regard to copper containing enzymes, a spectrum of mo1eculc which was shown to have the same retention strategies exist for substrate activation: these range time on MPLC as an autlienric pyrroloquinoline from direct metal catalysis (dapomine fl~monooxy- quinone derivnlivc, Phcnylhydrazine had previously gcnasc) to the involvement of protein-dcrivcd radicals been shown to act as a stoichiomrtric, m@ehnnism- (galnctosc sxidasc) and prorein.derivcd quinoncs (cop- bnscd inhibitor, leading to covalent labeling of protein per amine oxidascs). (271. Subsequently, Villafrnnca and co-workers [28] focused on the isolation and characterization of labeled active site pcptidcs. These stuc,ies demonstrated label- tibn GJsn, lnr:-bO02878j, thr National Research Council, Univrr- ing of tyrosinc and histidine by phenylhydrazinc; no shies of Rome ‘La Snpicnza’ and ‘Tor Vcrgsra’, It?lian Society of pyrroloquinolinc quinone could be detected, Variances Biocticmisrry (truly), in experimental results between the laboratories of Duine and Villafranca are actribured to the very different experimental conditions used by the two REFERENCES laboratories for the derivatization of doRamine B- [I] Snell, E.E., Braunstcin, A.E., Scvcrin, ES, and Torrhinsky, monooxygenase by phcnylhydrazine. lr is r&v belicvcd Yu+M, (cds.) (1068) Pyridosal Calnlysis: Enzymes and Model that pyrroloquinoline quinone, as such, is absent, but Systems, Vol. 35, Wiley Intcrscience, New York. [2] Knowles,P.F., Pnndcye, K.B., Ruis, F.X., Spencer, CM., that an analogous structure may be formed following Moog, RX, McGulrl, M.A. nnd Do&y, D.M. (1987) Biochem. treatment of enzyme under the prolonged conditions J. 241, GO3-G08. employed by Duine and co-workers. [3] Doolcy, D.M., Brown, D.E. and McGuirl, M.A. (lQQl), in prcpamtion. [4] Suva. R.H. and Abelcs. K.H. (1978) Biochemistry 17, 4, CALACTOSE OXIDASE 3538-3545. [5] Hartmann, C. and Klinman, J,P. (1987) J. Biol. Chcm. 262,

Unlike dopamine @-monooxygcnase, galactose 962-965 a oxidase is a monomeric enzyme containing a single cop- [G] Buffoni, F, and Cambi, S. (1990) Anal. Biochem. IS?, 44-50. per atom per polypeptide chain [29,30]. In this context, .[71 . Buffoni. F. 0990) Biochim. Bionhys. Acra 1040, 77-83. (81 Lobe&in-Verb&, C.L., Jongejan, J,A,, Frank, J. and it has been difficult to explain an enzyme-catalyzed, Duine, J.A. (1984) FEBS Lett. 170, 305-309, two electron oxidation of substrate, It has been known [Y] Amcyama, M., Hayashi, U., Matsushita, K., Shinagawa, E. and for some time that galactose oxidase can be oxidatively Adachi, 1, (1984) Agric. Biol, Chem. 48, 561-565. activated, through a process originally suggested to in- [IO] Vander Mccr, R,A., Jongejan. J.A., Frank, J. and Duine, J,A. volve Cu(III) formation [31]. Because the redox poten- (1986) FEBS Lett, 206, 11 l-114. (I 11 Van der Mcer, B.A. and Duine, J.A. (1986) Biochem. J. 239, tial for Cu(I1) --t Cu(II1) is expected to be very high [32], 789-791. the presence of such a species in a protein active site has 1121 Van lersel, J., Vandcr Meer, R,A. and Duine. J.A. (1986)Eur. been controversial. Independent reports in 1988 sug- J. Biochem. 161, 415-419. gested the involvement of an additional redox cofactor, [13] Glatz, Z,, Kovar, _I., Machola, 1.. and Pet, P. (1987) Biochem. J. 242, 603-606. either pyrroloquinoline quinone (Duine and co-workers [14] Moog, RS., McCuirl, M.A., Cot&, C,E. and Dooley, D.M. [33]) or a protein derived radical (Whittaker and co- (1986) Proc. Natl. Acad. Sci. USA 83, 8435-8439. workers [34]). A recent, high-resolution x-ray structure [15] Buechi, G., Botkin, J.H., Lee, G,C.M. and Yakushijin (1985) J. for galactose oxidase unambiguously eliminates the Am. Chem. Sot. 107, 5555-5556. possibility of an active site quinone, revealing, instead, [I61 Janes, S.M., Mu, D., Wemmer, D., Smith. A.J., Kaur, S., Maltby, D., Burlingame, A.L, and Klinman, J.P. (1990) Science active site t.yrosines [35]. This is consistent with a recent 248, 981-987. EPR characterization of the copper-free form of galac- [17] Morpurgo, I-., Agostinelli, E., Mondovi, B,, Avigliano, L., tose oxidase, which implicates a tyrosyl radical [36]. Us- Silvestrini, R., Stefancich, G. and Artico, M. J. Med. Chem., ing ribonucieotide reductase as a modei system for submitted. [IS] Brown, D.E,, McGuirl, M.A., Dooley, D.M., Janes,S.M., Mu, tyrosyl radical containing proteins, Duine and co- D. and Klinman. J.P. (1991) J. Biol. Chem., in press. workers have been able to produce pyrroloquinoline [19] Janes, S&l,, (1490) Ph,D..Thesis, University of California, quinone-like derivatives using their hexanol extraction Berkeley, USA:

3 Volume 282, number 1 April lW9sfr

120) Traekmun, P.C., Wolanski, A,, Pr?w, A.M., Tang, S&., 129) Amirral, D., Betnsfein, L., Morse, D, and Horeeker, B,L. (i963) Qt’lncr, G,D,, Tarxler, R.F. and Kapn, H. (lOsO) Dioehemtrrry J. Ilioi. Chcm. 2311,2281-2284. 29.4863-4870. [30) Kasman, D,J,, l%ln#cr, M-J,, Wrtner, R.E. knd Masarro. t&f, (21) Jane& S-M, and Klinman, J,P. (1991) Bioehcmlsrry, In press, (I9741 Arch. B&horn. Dlophyr* 163, 456-467, (22) MOndovi, B,, Avipliano, L. and Morpurgo, L. (1988) [St) Hamilron, G,A,, Adolf, P.K,, de Jersey, J., Dtr Bolr, O.C., Phnrmacot. Rcr. Comm, 20, suppl., 39-43, Dyrkrex, G,R, and Libby, RD. (1978) f. Am. Chem, $a$. lt)(l, (23) Collison, D., Knowles, P.F., Mnbbs, F,E., Ruir, F.X., Sin&, 189%1912. I., Doolcy, D.M.. Co& C.E. and McGuirt, M-A. (1989) (321 Kusl, R,N, (1979) In: Kirk.Blhmer Encyetopedia of Chcmlesl Biochcm. J. 264,653’-669, Technology (Grnyson, M. and Etkroth, B., eds,) 3rd cd., Vol. ,[24) Klinman, 3.P. and Mondovi, 8. (1990) Discussion In the 7, pp, 97-109, Wiley, New York. Symposium on ‘Copper in Biological Systems’, Manxiana, Italy, 1331 Van der Mcer, R.A. Jongejan, J,A. rnd Uuine, J.A. (1989) J. Sept, 23-28. Biot, Chem, 264, 7792-7794. 125) Stewart, LC, and Klinman, J,P, (1988) Annu. Rev. B&hem, 134) Whittaker, M,M, and Whirlaker, J-W, (1988) 5, Diet. Chcm. 57, 551-592. 263. 6574-6080. 1261 Van dcr Mccr, B.A., Jongcjan, J.A. and Duine, J.A. (1988) [351 Iro,‘N., Phillips, S.E.V., S~evenr, C., Ggct, ZB., McPherson, FEBS Lctt, 231, 303-307. M.J,, Kern, J,N,, Yadav, K,D,S, and Knowles, P.F. Nature, in [27) Fitzpatrick, P.F. and Vitlafranca, J,J, (lY66) J. Biol. Chcm, preSs. 261, 4510-4516. [361 Whit[oker, MUM, and Whitlaker, J.W. (1990) J. Biol, Chain, 1281. . Robertson. J,G.. Kumar. A.. Manccwicz, J0Aqand Villnfmnca, 265, 9610-9613, 3.3, (1989jJ. Biot. Chcm, 2&, i9916-19921.