Characterization of a low redox potential laccase from the basidiomycete C30 Agnieszka Klonowska 1 , Christian Gaudin 2, *, Andre ´ Fournel 3 , Marcel Asso 3 , Jean Le Petit 2 , Michel Giorgi 1 and Thierry Tron 1 1 Laboratoire de Bioinorganique Structurale CNRS UMR 6517 and 2 Laboratoire d’Ecologie Microbienne, CNRS UMR 6116, Faculte ´ des Sciences de St Je ´ ro ˆ me, Marseille, France; 3 Laboratoire de Bioe ´ nerge ´ tique et Bioinge ´ nierie des Prote ´ ines, CNRS UPR9036, Marseille, France A new exocellular laccase was purified from the basidio- mycete C30. LAC2 is an acidic protein (pI ¼ 3.2) prefer- entially produced upon a combined induction by copper and p-hydroxybenzoate. The spectroscopic signature (UV/visible and EPR) of this isoform is typical of multicopper oxidases, but its enzymatic and physico-chemical properties proved to be markedly different from those of LAC1, the constitutive laccase previously purified from the same organism. In particular, the LAC2 k cat values observed for the oxidation of the substrates syringaldazine (k cat ¼ 65 600 min )1 ), ABTS (2,2-azino-bis-[3-ethylthiazoline-6-sulfonate] (k cat ¼ 41 000 min )1 ) and guaiacol (k cat ¼ 75 680 min )1 ) are 10–40 times those obtained with LAC1 and the redox potential of its T1 copper is 0.17 V lower than that of LAC1 (E° ¼ 0.73 V). This is the first report on a single organism produ- cing simultaneously both a high and a low redox potential laccase. The cDNA, clac2, was cloned and sequenced. It encodes a protein of 528 amino acids that shares 69% identity (79% similarity) with LAC1 and 81% identity (95% similarity) with Lcc3-2 from Polyporus ciliatus (AF176321-1), its nearest neighbor in database. Possible reasons for why this basidiomycete produces, in vivo,enzyme forms with such different behaviors are discussed. Keywords: metalloenzyme; copper; redox potential; laccase; basidiomycete C30. Laccases (EC 1.10.3.2) catalyze the oxidation of a large spectrum of phenolic and nonphenolic substrates with a concomitant reduction of dioxygen to water [1,2]. They belong to the blue copper oxidase family and are charac- terized by the presence of a type 1 copper acting as a primary electron acceptor from reductant species, and a trinuclear copper site (one type 2 and two type 3) responsible for the four electron reduction of dioxygen [3]. These enzymes are common in plants, fungi, insects and bacteria [1]. In plants, they may mainly play a role in lignification [4] whereas in fungi they probably play the opposite role, i.e. delignifica- tion [5,6], among many others [1]. C30 is a white-rot basidiomycete that colonizes the evergreen oak (Quercus ilex L.) leaf litter in the Mediterra- nean area [7]. This fungus has previously been described as Marasmius quercophilus [7,8], but recent phenotypic and molecular evidence suggests that it may belong to another taxon [9,10]. Considering the biotope from which C30 was isolated, delignification is probably essential to this fungus and laccases may be key enzymes involved in this process [7]. Evidence for the production of at least three different laccases by C30 has previously been obtained [11]. The major one, LAC1, was purified, characterized and the correspond- ing gene sequenced [12]. However, under the conditions of culture used, the fungus did not produce other forms in sufficient quantities to allow their full characterization. Laccases are generally encoded by gene families [13,14]. Expression of these genes can be either constitutive or sensitive to the presence of inducers. Consequently, in different fungi, the production of minor laccase forms can be enhanced under appropriate culture conditions. This is the case in C3D for which addition of both copper and p-hydroxybenzoate to the culture broth greatly stimulated the production of three enzyme forms in addition to LAC1 [15]. Under these conditions, the fungus was able to syn- thesize the most active of these isoforms at a level similar to that of LAC1, the constitutive enzyme [15]. Therefore, using conditions of culture identical to those described in [15], our objectives in the present work were: (a) to produce enough material to purify and characterize this new isoform named LAC2; (b) to clone the corresponding coding sequence; and (c) to compare the properties of this inducible laccase with those of the constitutive LAC1. The great differences in properties observed between the two enzymes are discussed in terms of structure–function relationships. MATERIALS AND METHODS Enzyme production Precultures were carried out on malt extract/Tween 80 as previously described [12]. They were used to inoculate 3-L Correspondence to T. Tron, Laboratoire de Bioinorganique Structurale CNRS UMR 6517, case 432, Faculte ´ des sciences Saint Je ´ roˆ me, 13397 Marseille cedex 20, France. Tel.: 33 491 282856, Fax: + 33 491 983208, E-mail: thierry.tron@univ.u-3mrs.fr Abbreviations: ABTS, 2,2-azino-bis-[3-ethylthiazoline-6-sulfonate]. *Present address: Laboratoire de Bioinorganique Structurale CNRS UMR 6517. Note: a web site is available at http://www.lbs.fst.u-3mrs.fr (Received 25 June 2002, revised 4 October 2002, accepted 22 October 2002) Eur. J. Biochem. 269, 6119–6125 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03324.x conical flasks containing 600 mL of malt extract/Tween 80 containing 5 mgÆL )1 of CuSO 4 and 250 mgÆL )1 of p-hydroxybenzoate as inducers (malt extract/Tween 80/ Cu/p-hydroxybenzoate) [15]. The fungus was cultivated at 30 °C on a reciprocal shaker (50 r.p.m.) for 5 days [15]. Laccase activity The routine assay for laccase was based on syringaldazine oxidation in 0.1 M phosphate buffer (pH 5.7) at 30 °C. 2,2¢-Azinobis (3-ethylbenzothiazoline-6-sulfonate) (ABTS) and guaiacol were also used as substrates under the same conditions. Oxidation of syringaldazine, ABTS and guaia- col was monitored spectroscopically by absorbance meas- urements at 525 (e ¼ 6.5 · 10 4 M )1 Æcm )1 ), 420 (e ¼ 3.6 · 10 4 M )1 Æcm )1 ) and 418 nm (e ¼ 1.6 · 10 3 M )1 Æcm )1 ), respectively. One unit of laccase oxidizes 1 lmol of substrate per min. Kinetic parameters (k cat and K m ) were estimated using Lineweaver–Burk plots over a large range of substrate concentrations. LAC2 purification Culture supernatant (3.4 L) was filtered successively through gauze, paper filter and glass microfiber filters GFC and GFD (Whatman Ltd, Maidstone, UK), then concentrated 10-fold by ultrafiltration using YM10 mem- branes (Amicon, Millipore, Bedford, MA, USA), buffered with 20 m M phosphate, pH 6.0 (buffer A) and finally applied to an ion-exchange DEAE-Sepharose column (2.5 · 20 cm, Amersham Pharmacia Biotech Europe GmbH, Freiburg, Germany) equilibrated with the same buffer. Proteins were eluted at a flow rate of 4 mLÆmin )1 . withastepgradientofNaCl:0.1 M ,0.15 M ,0.2 M ,0.25 M , 0.3 M ,0.5 M and 1 M each for 25 min. More than 50% of the total laccase activity eluted during the 0.3 M step. The 0.3 M NaCl active fractions were then concentrated and loaded on a Sephacryl HR 100 column (1.2 · 80 cm, Amersham Pharmacia Biotech Europe GmbH, Freiburg, Germany) equilibrated with 20 m M phosphate buffer pH 6.0, 0.3 M NaCl and eluted at a flow rate of 0.5 mLÆmin )1 . Subse- quently, fractions containing laccase activity were pooled, concentrated and desalted. Enzyme purity was then con- firmed by SDS/PAGE. Enzyme characterization Determination of protein concentration, syringaldazine oxidation tests, native and denaturating PAGE and isoelec- tric focusing analysis were as previously described [12]. Laccase activities were detected by incubating the gels at 25 °Cin0.2 M acetate buffer containing 0.2% (w/v) p-phenylenediamine, at either pH 3.6 or pH 5.2. The purified protein was subjected to cyanogen bromide treat- ment as described in [16]; both N-terminal and internal CNBr peptide sequences were determined by stepwise Edman degradation. Dried gels were scanned with an Agfa SnapscanÒ 1236 piloted with FOTOLOOK Ò 2.09.6. Legends were added with CANVAS Ò 7. Laccase absorption was determined on a Uvikon 930 spectrophotometer (Kontron Instruments, Milan, Italy). X band EPR spectra were recorded on a Bruker (Wissem- bourg, France) ESP 300 spectrophotometer at 9.3 GHz and 16 K in 20 m M phosphate buffer, pH 6.0. The optimum pH for the enzyme was determined using 0.1 M phosphate buffer (pH 5.5–7.5), 0.1 M glycine/HCl solution (pH 3.0– 5.0) or 50 m M acetate buffer (pH 4.0–5.5). Its temperature tolerance was determined within the range of 30–85 °C using phosphate buffer. The redox potential (E°)oftheT1 copper of the two laccases was measured by anaerobic spectrophotometric titration. For both LAC1 and LAC2, 50–80 l M of enzyme in 0.1 M phosphate buffer pH 5.7 were titrated after adding 5–10 l M of the following mediators: ferrocenecarboxylic acid, ferrocenedicarboxylic acid and ferroin (Fluka). K 2 IrCl 6 and sodium dithionite were, respectively, used as oxidant and reductant. mRNA isolation and cDNA cloning The fungus was cultivated with inducers as described in [15]. Total RNA was extracted from frozen mycelium as des- cribed in [17]. Poly(A) + containing RNAs were purified with magnetic oligo d(T) beads (PolyA Tract, Promega). mRNA reverse transcription and cDNA library construc- tion were performed under the experimental conditions described in the Marathon cDNA construction kit manual (Clontech). Specific cDNAs were amplified at annealing temperature of 54 °C, using a degenerate forward PCR primer (Eurogentec Seraing, Belgium) AK7 5¢CA(CT)TGG CA(CT)GGNTT(CT)TT(CT)CA3¢ (identical to the Primer I used in [18]). This primer is based on the consensus peptide HWHGFFQ found in copper-binding region I. The univer- sal Marathon cloning AP1 primer (Clontech) was used as the reverse primer; nucleotides in parenthese indicate minimal variations (degeneracy) for the same position. Two amplicons of 1.9 (cDNA20) and 1.8 kb (cDNA19) were separated and cloned. After sequencing of their 3¢ ends, specific primers were designed and used to clone full length cDNAs using the Marathon AP1 primer as forward primer. Their final sequencing (Genomexpress, Grenoble, France) confirmed that only cDNA20, amplified with the reverse primer AK20 5¢CAGAGAACGAACGTA TGTGCTGG3¢ under the conditions described in the Marathon cDNa cloning kit manual, encodes the peptides sequenced from LAC2. Nucleotide sequence accession no. The sequence of the C30 laccase cDNA clac2 reported in this paper has been submitted to GenBank under accession no. AF491761. Modeling of LAC1 and LAC2 enzymes 3D models for LAC1 and LAC2 were built on the basis of the reported crystal structures of Coprinus cinereus laccase (GenBank accession no. 1A65), using the MODELLER v4.0 [19]. Sequence identity with the templates was 57% for LAC1 and 53% for LAC2. Sequences were first aligned with CLUSTAL W [20]. Then, the alignments were manually adjusted in order to minimize the impact of insertions/ deletions in the final calculations. Geometric parameters for the copper coordinations were constrained according to those of C. cinereus laccase. The top set of calculated structures were finally analyzed with the PROCHECK v3.5 [21] 6120 A. Klonowska et al. (Eur. J. Biochem. 269) Ó FEBS 2002 and models visualized with the SWISS - PDBVIEWER v3.7b2 [22]. RESULTS Induction of laccase production When C30 was grown in shaken liquid culture amended with both copper sulfate and p-hydroxybenzoic acid, total laccase activity reached a value of 3.03 UÆmL )1 within 5 days. As expected from earlier studies [15], when checked on native gel electrophoresis, the pattern of laccase isoforms present in the extracellular fluid showed a high production of inducible laccases in addition to the constitutive LAC1 (data not shown). LAC2 purification Under these culture conditions, the most anionic laccase isoform is by far the most active [15]. The purification of this enzyme, named LAC2, was achieved in three steps, allowing us to recover 5.5 mg of enzyme with a specific activity of 934 UÆmg )1 , for a final yield of 50% (Table 1). The pure LAC2 produced a single band both on a SDS/PAGE gel, at a molecular mass of approximately 65 kDa (Fig. 1A), and on a native PAGE gel (Fig. 1B). Spectroscopic characterization The three types of copper usually present in multicopper oxidases were detected in purified LAC2. The intense blue of the enzyme reflected the presence of a T1 copper (k max ¼ 608 nm) whereas that of the binuclear T3 pair was indicated by a shoulder at 333 nm in the UV/visible spectrum (data not shown). Values of the constants extracted from the X band EPR spectrum for the T1 and the T2 coppers (Table 2) were found to be very similar to those previously obtained for LAC1 [12]. Redox potential of T1 copper The progressive reduction of the fully oxidized T1 copper was followed spectroscopically by the disappearance of absorption at 608 nm. Under our experimental conditions, the E° values determined for LAC1 and LAC2 were, respectively, 0.73 and 0.56 V vs. normal hydrogen electrode (Table 2). These values are consistent with those previously reported for fungal laccases. Physico-chemical properties and kinetic parameters When LAC2 syringaldazine oxidation was monitored as a function of pH (Table 2), the highest rates were obtained between pH 5.5 and 6, an optimum pH zone 1 unit higher than that previously determined for LAC1 [12]. The thermal dependence of this reaction was therefore subsequently tested in phosphate buffer pH 5.7 and maximal syringald- azine oxidation rate was reached at a temperature around 55 °C (Table 2). This temperature is identical to that found to be optimum for LAC1 when syringaldazine oxidation was tested at pH 4.5–5. Oxidation of syringaldazine, ABTS and guaiacol were monitored spectroscopically under different pH conditions for LAC1 and LAC2. Kinetic parameters, extracted from Lineweaver–Burk plots, are reported in Table 3. LAC2 exhibited much greater apparent k cat than those observed for LAC1 but with a lower affinity for all substrates tested. The catalytic efficiency (k cat /K m )ofLAC2proved to be 10 times that of LAC1 on syringaldazine, whereas the two laccases exhibited roughly the same efficiency on guaiacol. With the nonphenolic substrate ABTS, LAC1 efficiency was found to be two to three times that of LAC2. Although the k cat values decreased by factor 3, the ratio (k cat /K m ) was not much affected when LAC1 kinetics were recorded at pH 5.7 instead of pH 5.0, the optimum pH for LAC1. Azide inhibition Sodium azide inhibition of either syringaldazine or ABTS oxidation was measured for LAC1 and LAC2. In both cases, this inhibition was found of noncompetitive type. In the reaction conditions of optimum pH and at 30 °C, the observed I 50 for LAC2 (18 ± 5 l M ) is approximately 10 times higher than that for LAC1 (1.5 ± 0.2 l M ). N-Terminal and CNBr peptides sequence analysis Twenty lg of the purified protein were first reduced, then carboxymethylated and subjected to Edman degradation. Table 1. Purification of LAC2 from C30. Activities were measured using SGZ as substrate. Step Protein (mg) Total activity (U) Sp act (U/mg) Yield (%) Purification (fold) Supernatant 280 10314 37 100 1.0 Ultrafiltration 98 8505 87 83 2.4 DEAE-Sepharose 9 5580 642 54 17.5 Sephacryl S-100 5.5 5135 934 50 25.4 Fig. 1. Electrophoresis of purified LAC2. (A) Silver staining of a SDS/ 7.5% PAGE; lanes 1 and 5: molecular mass standards; lane 2: 0.5 lgof purified LAC1; lane 3: 0.5 lgofpurifiedLAC2;lane4:10lgoftotal extra cellular proteins. (B) Native PAGE stained with p-phenylene diamine; lane 1: total extra cellular proteins; lane 2: Lac1; lane 3, Lac2; proteins corresponding to 0.005 U were deposited per lane. Dried gels werescannedwithanAgfaSnapscanÒ 1236 using FOTOLOOK Ò 2.09.6. Legends were added with CANVAS Ò 7.0. Ó FEBS 2002 LAC2 from the basidiomycete C30 (Eur. J. Biochem. 269) 6121 The first 15 residues at the amino terminus are: AIGPKADLTISNANI. The first six amino acids of this sequence match perfectly the result obtained on the laccase contained in fraction D in a preliminary study on the enhancement of minor laccase production in C30 [15]. We also sequenced a 15-kDa internal peptide and found that the first 15 residues from this peptide are AIPNVGTINTDGGVN. A database search showed that these peptides are closely related to those found in laccase sequences from basidiomycete CECT 20197 (accession no. U65400), Trametes villosa (accession no. L49376 and L78077) and Trametes versicolor (accession no. Y18012). clac2 cDNA cloning The cDNA encoding LAC2 was cloned from a PCR amplified cDNA library. It contains an open reading frame 1584 bp long coding for a protein 528 residue long, a 76-bp 5¢-untranslated region, a 220-bp 3¢-untranslated region and a 31-bp poylA tail. The amino acid sequence deduced from the open reading frame contains the peptide sequences previously characterized from the LAC2 purified protein. The clac2 ORF is 36 bp longer than the lac1 ORF (AF162785) and the global identity between the two coding sequences is 67%. At the protein level, the two enzymes are 69% identical but LAC2 possesses 12 extra amino acids, seven of which constitute its carboxy terminus. The LAC2 nearest neighbors found in database are: Polyporus ciliatus Lcc3-2 (AF176321-1, identity/homology: 81/95%), Tra- metes versicolor LAC4 (Q12719, identity/homology: 71/ 92%), Trametes pubescens LAC1A (AF414808-1, identity/ homology: 70/91%) and Trametes villosa LAC5 (Q99056, identity/homology: 69/91). LAC1 and LAC2 homology models For each model, 10 calculated structures were fitted on the reference structure, the C. cinereus laccase [23]. The general organization of the two models was found to be very close to that of the reference structure. The copper coordination ligands are identical to those present in the reference and the geometry of the T1 copper is basically preserved. However, significant differences in folding were found in three regions all located within 10 A ˚ around the T1 copper. All of these differences are related to gaps present in the initial sequence alignment and corresponding to either insertion or deletion of one to four residues in the proteins considered (data not shown). 3D representations of region the closest to the T1 copper are given in Fig. 2. Polypeptide chain length variations of two residues for LAC1 (Fig. 2A) and three residues for LAC2 (Fig. 2B) in the loop containing the T1 copper proximal ligand (H396 in C. cinereus) induce noticeable structural changes around the copper. DISCUSSION The basidiomycete C30 secretes at least four different laccases, the proportion of which depends on culture conditions [12,15]. We have previously purified and Fig. 2. Comparison of the backbone superposition at the T1 copper site of the C. cinereus laccase and the C30 laccase models. The Ca trace of C. cinereus (1A65) laccase is shown in red. For clarity, only segments corresponding to loops L333–T341, V387–H399 and H451–A463 and coordinating residues H396, C452 and H457 are represented. The Ca traces of 10 calculated models of LAC1 (A) and LAC2 (B) are shown in grey (coordinating residues, nearly superposable to those of the C. cinereus laccase, have been omitted for clarity). Table 3. LAC1 and LAC2 kinetic parameters. ND, not determined. Substrate Enzymes pH k cat (min )1 ) K m (l M ) k cat /K m (min )1 Æl M )1 ) SGZ LAC1 5.0 1800 1.8 1000 5.7 600 0.9 670 LAC2 5.7 65 600 6.8 9650 GUA LAC1 5.0 ND ND – 5.7 2300 71 30 LAC2 5.7 75 680 1006 76 ABTS LAC1 5.0 3350 10.7 310 5.7 610 2.9 210 LAC2 5.7 41 000 536 80 Table 2. Physico-chemical and EPR parameters of copper sites of laccases 1 and 2 from C30. A // values are in 10 )4 cm )1 units; ND, not determined. EPR parameters T1 copper T2 copper Enzymes opt pH a T (°C) b E° (V) A // g // g ^ A // g // g ^ Ref. LAC1 4.5-5 55 0.73 96 ND ND > 140 ND ND 6 LAC2 5.5-6 55 0.56 88 2.165 2.025 172 2.25 2.027 This work a Values obtained with SGZ as substrate. b Temperature for which the main activity is reached with SGZ as substrate. 6122 A. Klonowska et al. (Eur. J. Biochem. 269) Ó FEBS 2002 characterized LAC1, the most abundant enzyme produced by C30 [12]. The purification of a second laccase (LAC2) from this fungus allows us to compare enzymes, the synthesis of which is regulated differently. Indeed, LAC1 is produced under all the conditions we have tested so far and thus is probably a constitutive form. On the other hand, LAC2, which is almost absent in noninduced cultures, becomes one of the most prominent laccases secreted when the growth medium is supplemented with copper and p-hydroxybenzoate; it can therefore be considered an inducible enzyme [15]. Such differences in their patterns of expression suggest a distinct physiological role for these two isoforms and, although they share basic properties, the large variation in catalytic activity for both phenolic and dyes supports this idea. The C30 laccase isoforms we have detected so far all have an apparent molecular mass close to 65 kDa and, except for a still-uncharacterized laccase, are all acidic proteins with pI ranging from 3.2 (LAC2) to 3.6 (LAC1) [15]. The optimum pH for syringaldazine oxidation is close to 4.5–5 for LAC1 and 5.5–6 for LAC2, values that, unlike the data obtained for several laccases [24], do not correlate with their pI. On the other hand, when tested at their respective optimum pH values, the two isoforms are the most active at the same temperature of 55 °C. The amino acid sequences deduced from lac1, the gene encoding LAC1, and clac2, the cDNA encoding LAC2, are 69% identical and the two proteins, LAC1 and LAC2, present very similar optical and EPR spectra. Generally speaking, all the above features of C30 enzymes are similar to those of other white-rot fungi. This is expected as laccases form a group of highly homologous proteins. However, several major physico-chemical features distinguish LAC1 from LAC2. One is the 170 mV differ- ence observed between their respective T1 redox poten- tial measured at pH 5.7. Indeed, with a potential E° ¼ 0.73 V, LAC1 belongs to the group of high redox potential laccases (T. versicolor, T. villosa, Rhizoctonia sol- ani, Pleurotus ostreatus POXC and POXA1b, Rigidosporus lignosus B and D) whereas LAC2, with a potential E° ¼ 0.56 V, belongs to the group of low redox potential laccases (C. cinereus, Myceliophthora thermophila, Scytali- dium thermophilum) [24–27]. To the best of our knowledge, this difference in potential is the largest so far reported between two laccases purified from the same organism. Moreover, even though the T1 potentials of R. lignosus laccases B and D and P. ostreatus POXC and POXA1b are, respectively, 40 and 90 mV different [24], all these four enzymes apparently belong to the high redox potential group. C30 is therefore the first organism for which a simultaneous production of high redox and low redox potential laccases is reported. Several attempts have been made to correlate the redox potential variations found in laccases to the nature of the specific amino acids present in their active site as their oxidative capabilities appear tightly linked to this param- eter [24,28]. The replacement F463M in a T. villosa (accession no. AAC41686) high redox potential laccase provides a fourth coordinating axial ligand to the T1 copper resulting in a 100-mV drop of its potential [27]. On the other hand, the idea that the occurrence of a phenylalanine residue might correlate with a high redox potential in laccases was ruled out by the site directed replacement of L fi F in two laccases [26]. The presence of an F residue at the corresponding position in the sequence of both the C30 high redox potential LAC1 and low redox potential LAC2 sequences support this conclu- sion. Similarly, the replacement of the LEA amino acid triplet located immediately after the distal T1 coordinating histidine (H456 in C. cinereus, accession no. 1A65) in the high redox potential R. solani laccase (accession no. Q02081), by a SVG amino acid triplet found in the low redox potential M. thermophila (accession no. AAE35046) and vice versa did not affect significantly the E° of the recombinant enzymes [26]. In our study, the presence of a LEA tripeptide both in the C30 LAC1 and LAC2 enzymes correlates well with these results. In an effort to gain new insights into the factors influencing the potential of the T1 copper in laccases, data on eight high redox potential and four low redox potential enzymes may not be enough to design new targets for mutagenesis. On the other hand, models of our enzymes show substantial structural variations close to the metal center (Fig. 2) in a loop where main differences are found when 3D structures, including the recently published structures of T. versicolor [30,31], and Melanocarpus albomyces [32] laccases are compared (not shown). As folding of the backbone around the metal center was already proposed to be a major factor affecting the potential in iron–sulfur proteins [29], it seems to us that it would make sense to make mutants in this region. A second distinction between C30 LAC1 and LAC2 enzymes can clearly be made on the basis of their kinetic parameters for the oxidation of phenolic (syringaldazine and guaiacol) as well as nonphenolic (ABTS) substrates (Table 2). Depending on the pH conditions used, we found that LAC2 k cat values are one to two orders of magnitude higher than those of LAC1 whereas, the affinity for the three substrates of the former enzyme is lower as reflected by higher K m values. Strong differences in K m values are commonly observed for laccases. As discussed in previous studies [24,25], the differences in kinetics observed may be the consequence of the variability of certain of the amino acids involved in the substrates channel in specific enzymes. A substantial variation in the folding of the T1 copper pocket of the two enzymes, such as that mentioned above to explain their difference in T1 potential, could also account for their specific interaction with the substrates. In laccases, enzyme efficiency (k cat /K m ) correlates with the redox potential of the substrates [24] and the two C30 laccases behave more or less this way. In fact, like already observed by Garzillo et al. in their study on T. trogii, R. lignosus and P. ostreatus laccases [25], LAC1 and LAC2 activities on phenolic compounds seem only partly related to their specific redox capabilities. Indeed, LAC2 appears to be two to 10 times more efficient than LAC1 on phenolic substrates although with a 170-mV lower T1 copper redox potential. As phenol oxidation involves release of a proton, factors like hydrogen bonding or the extent of protonation of ionizable residues in the vicinity of the T1 copper probably have considerable effects on the overall efficiency. Differences between LAC1 and LAC2 are not restricted to the oxidation site as the two enzymes also react differently toward sodium azide, an inhibitor known to bind to the oxygen reduction site. It is likely that a channel governs the accessibility to the T2/T3 cluster. Therefore, a LAC2/LAC1 I 50 ratio of 10 probably reflects a significant variation in the Ó FEBS 2002 LAC2 from the basidiomycete C30 (Eur. J. Biochem. 269) 6123 size of the channel from LAC1 to LAC2. On the other hand, in a recent study on laccases reactivity with dioxygen [34], it is speculated from steady state analysis that laccases have a conserved O 2 binding domain and that the rate of O 2 reduction is dependent on that of substrate oxidation. In our case, this means that LAC2 should reduce dioxygen much faster than LAC1 and a similar investigation on oxygen reduction rates must be undertaken on C30 laccases to further the description of their reactivity. Generally speaking, when compared to kinetic data on laccases from other fungi, it appears that while LAC1 activity is similar to other laccases for both phenolic and nonphenolic substrates, LAC2 is a remarkably efficient enzyme at least on the three substrates tested. A comparison of enzyme efficiency restricted to the group of low redox potential laccases reveals that, depending on the substrate, LAC2 appears to be 2–100 times more efficient than its C. cinereus, M. thermophila and S. thermophilum counter- parts [24,33]. Again, as for their differences in T1 potential, this is the first report on enzymes produced by a single organism with such markedly different catalytic efficiency. From a physiological point of view, the constitutive LAC1, abundant in the culture supernatant [12], exhibits a relat- ively high affinity for phenols but a relatively low capacity for oxidation when compared to the inducible LAC2. The consequences of these differences is not yet known but, by analogy with permeation systems [34] for which such a contrast between affinity and rate is often observed, we could interpret the differences in laccase properties as a need for the organism to maintain both a low capacity/high specificity system when substrate level is low and a high capacity/low specificity system when the substrate is abun- dant. In conclusion, we have demonstrated that LAC2, a laccase produced by the basidiomycete C30 following copper and p-hydroxybenzoate induction, is a low redox potential enzyme with unusually high oxidative capabilities. The kinetic data obtained both on phenolic and nonphenolic substrates indicate that LAC2 might be a good catalyst for the transformation of different substrates. As laccases are generally produced as a number of isoenzymes encoded by multigene families, the expression of which varies from fungus to fungus, it is highly probable that other fungi contain the equivalent of LAC2. A search for the appropri- ate conditions of expression of a given activity being empirical and time consuming, it will probably be more efficient to use a heterologous expression system for laccase activities to find other enzymes with high oxidative capacities. ACKNOWLEDGEMENTS A. K. is the recipient of an Agence de l’Environement et de la Maıˆtrise de l’Energie (ADEME) fellowship. This work was in part supported by a grant from the Conseil Ge ´ ne ´ ral 13. We thank Gilles Iacazio, Marius Re ´ glier, Jalila Simaan and Marjorie Sweetko for their critical reading of the manuscript. REFERENCES 1. Gianfreda, L.F., Xu & Bollag, J M. (1999) Laccases: a useful group of oxidoreductive enzymes. Bioremediation J. 3, 1–25. 2. Thurston, C.F. (1994) The structure and function of fungal lac- cases. Microbiology 140, 19–26. 3. Solomon, E.I., Sundaram, U.M. & Machonkin, T.E. (1996) Multicopper oxidases and oxygenases. Chem. Rev. 96, 2563–2605. 4. Mayer, A.M. (1987) Polyphenol oxidases in plants. Recent pro- gress. Phytochem. 26, 11–20. 5. Hattaka, A. (1994) Lignin-modifying enzymes from selected white-rot fungi: production and role in lignin degradation. FEMS Microbiol. Rev. 13, 125–135. 6. Youn, H D., Hah, Y.C. & Kang, S O. (1995) Role of laccase in lignin degradation by white-rot fungi. FEMS Microb. Lett. 132, 183–188. 7. Tagger, S., Pe ´ rissol,C.,Gil,G.,Vogt,G.&LePetit,J.(1998) Phenoloxidases of the white-rot fungus Marasmius quercophilus isolated from an evergreen oak litter (Quercus ilex L.). Enz. Microb. Technol. 23, 372–379. 8. Farnet, A M., Tagger, S. & LePetit, J. (1999) Effects of copper and aromatic inducers on the laccases of the white-rot fungus Marasmius quercophilus. C. R. Acad. Sci. Paris/ Life Sci. 322, 1–5. 9. Farnet, A.M. (1998) Variabilite ´ phe ´ notypique et ge ´ ne ´ tique chez Marasmius quercophilus basidiomyce ` te colonisant une litie ` re de che ˆ ne vert. PhD Thesis. University of Aix-Marseille III, Marseille. 10. Klonowska, A., Gaudin, C., Ruzzi, M., Colao, M.C. & Tron, T. (2002) Ribosomal DNA sequence analysis shows that the basi- diomycete C30 belongs to the genus Trametes. Res. Microbiol,in press. 11. Dedeyan, P. (1996) Purification et caracte ´ risation des laccases de Marasmius quercophilus. PhD Thesis. University of Aix-Marseille III, Marseille. 12. Dedeyan, B., Klonowska, A., Tagger, S., Tron, T., Iacazio, G., Gil, G. & LePetit, J. (2000) Biochemical and molecular char- acterization of a laccase from Marasmius quercophilus. Appl. Env. Microbiol. 66, 925–929. 13. Mansur, M., Suarez, T., Fernandez-Larrea, J.B., Brizuela, M.A. & Gonzalez, A.E. (1997) Identification of a laccase gene family in the new lignin-degrading basidiomycete CECT 20197. Appl. Env. Microbiol. 63, 2637–2646. 14. Yaver, D.S. & Golightly, E.G. (1996) Cloning and characteriza- tion of three laccase genes from the white-rot basidiomycete Trametes villosa: genomic organization of the laccase gene family. Gene 181, 95–102. 15. Klonowska, A., LePetit, J. & Tron, T. (2001) Enhancement of minor laccases production in the basidiomycete Marasmius quer- cophilus C30. FEMS Lett. 200, 25–30. 16. Matsudaira, P. (1990) Limited N-terminal sequence analysis. Meth Enzymol. 182, 602–613. 17. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. & Rutter, W.J. (1979) Isolation of biologically active ribonucleic acid from sour- ces enriched in ribonuclease. Biochemistry 18, 5294–5299. 18. D’souza, T.M., Boominathan, K. & Reddy, C.A. (1996) Isolation of laccase gene-specific sequences from white-rot and brown-rot fungi by PCR. Appl. Env. Microbiol. 62, 3739–3744. 19. Sali, A. & Blundell, T.L. (1993) Comparative protein modeling by satisfaction of spatial restraints. J. Mol. Biol. 234, 779–815. 20. Thompson, J.D., Higgins, D.G. & Gibson, T.J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucl Acids Res. 22, 4673–4680. 21. Laskowski, A.R., MacArthur, M.W., Moss, D.S. & Thornton (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Cryst. 26, 283–291. 22. Guex, N. & Peitsch, M.C. (1997) SWISS-MODEL and the Swiss- PdbViewer: An environment for comparative protein modeling. Electrophoresis 18, 2714–2723. 23. Ducros, V., Brzozowski, A.M., Wilson, K.S., Brown, S.H., Ost- ergaard, P., Schneider, P., Yaver, D.S., Pedersen, A.H. & Davies, G.J. (1998) Crystal structure of the type-2 Cu depleted laccase from Coprinus cinereus at 2.2 A ˚ resolution. Nat. Struct. Biol. 5, 310–316. 6124 A. Klonowska et al. (Eur. J. Biochem. 269) Ó FEBS 2002 24. Xu, F., Shin, W., Brown, S.H., Wahleithner, J.A., Sundaram, U.M. & Solomon, E.I. (1996) A study of a series of recombinant fungal laccases and bilirubin oxidase that exhibit significant dif- ferences in redox potential, substrate specificity, and stability. Biochim. Biophys. Acta. 1292, 303–311. 25. Garzillo, A.M., Colao, M.C., Buonocore, V., Oliva, R., Falcigno, L., Saviano, M., Santoro, A.M., Zappala, R., Bonomo, R.P., Bianco, C., Giardina, P., Palmieri, G. & Sannia, G. (2001) Structural and kinetic characterization of native laccases from Pleurotus ostreatus, Rigidoporus lignosus,andTrametes trogii. J. Protein Chem. 20, 191–201. 26. Xu, F., Berka, R.M., Wahleitner, J.A., Nelson, B.A., Schuster, J.R., Brown, S.H., Palmer, A.E. & Solomon, E.I. (1998) Site- directed mutations in fungal laccases: effect on redox potential, activity and pH profile. Biochem. J. 334, 63–70. 27. Xu,F.,Palmer,A.E.,Yaver,D.S.,Berka,R.M.,Gambetta,G.A., Brown, S.H. & Solomon, E.I. (1999) Targeted mutations in a Trametes villosa laccase. Axial perturbations of the T1 copper. J. Biol. Chem. 274, 12372–12375. 28. Messerschmidt, A. & Huber, R. (1990) The blue oxidases, ascor- bate oxidases, laccase and ceruloplasmin. Eur. J. Biochem. 187, 341–352. 29. Stephens, P.J., Jollie, D.J. & Warshel, A. (1996) Protein control of redox potentials of iron-sulfur proteins. Chem. Rev. 96, 2491– 2513. 30. Piontek, K., Antorini, M. & Choinowski, T. (2002) Crystal Structure of a laccase from the fungus Trametes versicolor at 1.90- A ˚ resolution containing a full complement of coppers. J. Biol. Chem. 277, 37663–31669. 31. Bertrand, T., Jolivalt, C., Briozzo, P., Caminade, E., Joly, N., Madzak, C. & Mougin, C. (2002) Crystal structure of a four- copper laccase complexed with an arylamine: insights into sub- strate recognition and correlation with kinetics. Biochemistry 2002, 7325–7333. 32. Hakulinen, N., Kiiskinen, L.L., Kruus, K., Saloheimo, M., Paananen, A., Koivula, A. & Rouvinen, J. (2002) Crystal structure of a laccase from Melanocarpus albomyces with an intact trinuclear copper site. Nat Struct. Biol. 9, 601–605. 33. Xu, F. (2001) Dioxygen reactivity of laccase. Appl. Biochem. Biotechn 95, 125–133. 34. Eide, D.J. (1998) The molecular biology of metal ion transport in Saccharomyces cerevisiae. Annu. Rev. Nutr. 18, 441–469. Ó FEBS 2002 LAC2 from the basidiomycete C30 (Eur. J. Biochem. 269) 6125 . Characterization of a low redox potential laccase from the basidiomycete C30 Agnieszka Klonowska 1 , Christian Gaudin 2, *, Andre ´ Fournel 3 , Marcel. idea. The C30 laccase isoforms we have detected so far all have an apparent molecular mass close to 65 kDa and, except for a still-uncharacterized laccase,