Báo cáo khoa học: Purification and partial characterization of seven glutathione S -transferase isoforms from the clam Ruditapes decussatus pptx

Purification and partial characterization of seven glutathione S -transferase isoforms from the clam Ruditapes decussatus Pascal Hoarau, Ginette Garello, Mauricette Gnassia-Barelli, Miche ` le Romeo and Jean-Pierre Girard UMR 1112 INRA-UNSA, Laboratoire Re ´ ponse des Organismes aux Stress Environnementaux, Faculte ´ des Sciences, Universite ´ de Nice-Sophia Antipolis, Nice, France This paper deals with the purification and the partial char- acterization of glutathione S-transferase (GST) isoforms from the clam Ruditapes decussatus. For the first step of purification, two affinity columns, reduced glutathione (GSH)–agarose and S-hexyl GSH–agarose, were mounted in series. Four affinity fractions were thus recovered. Further purification was performed using anion exchange chroma- tography. Seven fractions, which present a GST activity with 1-chloro-2,4-dinitrobenzene (CDNB) as substrate, were collected and analyzed by RP-HPLC. Seven distinct GST isoforms were purified, six of them were homodimers, the last one was a heterodimer consisting of the subunits 3 and 6. Kinetic parameters were studied. Results showed that isoforms have distinct affinity and V max for GSH and CDNB as substrates. The catalytic activity of the heterodimer isoform appeared to be a combination of the ability of each subunit. The immunological properties of each purified isoform were investigated using three antisera anti-pi, anti- mu and anti-alpha mammalian GST classes. Three isoforms (3-3,6-6and3-6)seemtobecloselyrelatedtothepi-class GST. Both isoforms 1-1 and 2-2 cross-reacted with antisera to pi and alpha classes and the isoform 5-5 cross-reacted with the antisera to mu and pi classes. Subunit 4 was recognized by the three antisera used, and its N-terminal amino acid analysis showed high identity (53%) with a conserved sequence of an alpha/ml/pi GST from Fasciola hepatica. Keywords: clam; glutathione S-transferase; immunology; kinetics; N-terminal analysis. The glutathione S-transferases (GSTs; EC 2.5.1.18) bind lipophilic nonsubstrate ligands and compounds such as heme and bilirubin [1], and are implicated in the biosynthe- sis of prostaglandins [2]. Apart from their metabolic activities, they are involved in the detoxication of electro- philic and genotoxic compounds by both catalytic activity and direct binding. In fact, the effect of these reactions is to convert a reactive (lipophilic) molecule into a water-soluble (nonreactive conjugate) which can be excreted. These mechanisms play an important role in cellular protection against the toxicity of endogenous compounds and of a variety of xenobiotics [3]. The diversity of compounds metabolized by GSTs results from both the relatively nonspecific nature of the hydrophilic substrate binding site and the existence of numerous GST isoforms. GST activities are often assayed using 1-chloro- 2,4-dinitrobenzene (CDNB), a relatively nonspecific GST reference substrate [4]. GST-CDNB activity reflects the integration of GST isoenzyme activities. However, the use of GST substrates such as 1,2-dichloro-4-nitrobenzene, ethacrynic acid (ETHA), nitrobutyl chloride and D5-andro- stene-3,17-dione in conjunction with CDNB allows for a more complete biochemical characterization of GST iso- zyme activities [5]. GSTs are a multigene family of enzymes (isoforms), which have been grouped into seven classes based upon sequence homology and ability to catalyze the conjugation of glutathione to a broad range of electrophilic substrates in animal organisms. These classes are named alpha, mu, pi, theta, sigma, kappa, zeta and omega [1,6,7]. Some bacteria express a beta class GST [8] and higher invertebrates express a delta class GST [9]. The GST family is found in most aerobic eukaryotes and in some prokaryotes. In vertebrate organisms, particularly in rats (where 13 different isoforms have been found in the liver [10]) and in fish [11], GST isoenzymes are extensively studied and are distinct with regards to their subunit structure, isoelectric point, kinetics and immunological properties. In compari- son with what is known of vertebrates, relatively little information is available concerning GSTs from marine invertebrate organisms. Fitzpatrick and Sheehan [12] reported the purification of four GST isoforms in the digestive gland and gills of the blue mussel Mytilus edulis. In previous work [13], the differential induction of GSTs was studied in the clam Ruditapes decussatus exposed to organic compounds. The present paper describes the purification of clam GST isoforms by affinity chromato- graphy using reduced glutathione (GSH)–agarose and S-hexyl GSH–agarose columns. The fraction recovered were then submitted to anion exchange chromatography, leading to the distinction of isoforms. Their characterization was performed by studying the kinetic parameters (V max , K m with the commonly used substrate CDNB and GSH) and by performing immunoblotting reactions. RP-HPLC analysis was used to identify the different constitutive Correspondence to P. Hoarau, UMR 1112 INRA-UNSA, Laboratoire Re ´ ponse des Organismes aux Stress Environnementaux, Faculte ´ des Sciences, Universite ´ de Nice-Sophia Antipolis, Parc Valrose, 06108 Nice Cedex 2, France. Tel.: +33 04 92 07 68 96, E-mail: hoarau@unice.fr Abbreviations:GST,glutathioneS-transferase; GSH, reduced gluta- thione; CDNB, 1-chloro-2, 4-dinitrobenzene; ETHA, ethacrynic acid. Enzyme: glutathione S-transferases (GST; EC 2.5.1.18). (Received 14 May 2002, revised 11 July 2002, accepted 25 July 2002) Eur. J. Biochem. 269, 4359–4366 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03141.x subunits of the purified isoforms. N-terminal amino acid analysis was carried out on a particular subunit in order to evaluate to which class of GST isoforms this subunit may belong. MATERIALS AND METHODS Animal maintenance Animals were collected from the lagoon of Thau (N.W. Mediterranean sea) and placed on ice for transport to the laboratory. They were then transferred to a 200-L aquarium filled with natural aerated sea water, in a closed circuit, at 18 °C with a photoperiod of 12 : 12 h for 2 days. Preparation of the postmitochondrial fraction All procedures were carried out at 0–4 °C. A pool of 12 animals was homogenized in a buffer (20% w/v, 10 m M Hepes, 250 m M sucrose, 1 m M phenylmethanesulfonyl fluoride, 1 m M dithiothreitol pH 7.4) using an ultra-Turrax homogenizer. The homogenate was centrifuged at 9000 g for 20 min to remove cell debris, nuclei and mitochondria. All analyses were performed on the supernatants which were used immediately or stored at )80 °C. Enzyme assays GST activity was measured spectrophotometrically at 37 °C using two substrates, CDNB and ETHA, according to the method of Habig et al. [14]. Specific activity was expressed as lmol productÆmin )1 Æmg )1 protein. The protein concentration was determined according to Bradford [15]. Purification of GSTs All procedures were performed at 0–4 °C. Fifty milliliters of the supernatant were dialyzed with a cut-off of 10 kDa against a solution of 20 m M Tris/HCl, 0.2 M NaCl, 1 m M dithiothreitol pH 7.9 to eliminate GSH. The supernatants were passed through two affinity columns (Sigma) moun- ted in series. The first was a GSH–agarose column (C1) and the second a S-hexyl GSH–agarose column (C2). Each column was then separately eluted with two solutions. The first solution (E1) was 200 m M Tris/HCl containing 0.2 m M GSH pH 9, and the second (E2) was 200 m M Tris/HCl containing 5 m M GSH pH 9. Four fractions were thus obtained, they were called: C1E1, C1E2, C2E1 and C2E2. The fractions were then dialyzed (cut-off 10 kDa) against 20 m M Tris/HCl, 1 m M dithio- threitol pH 9. GSTs from the two fractions of each column (C1 and C2) were fractionated further by anion exchange chromatogra- phy using a 5-mL Q cartridge column (Bio-Rad). The column was equilibrated with 20 m M Tris/HCl, 1 m M dithiothreitol pH 9, at a flow rate of 1.0 mLÆmin )1 .The absorbance was measured at 280 nm during the experiment. The elution conditions are shown in Table 1. One-milliliter fractions were collected during the experiment and their activity with CDNB as a substrate was measured. Further analyses were performed only in the seven fractions where CDNB activity was measurable. Kinetic parameters The K m and V max determinations with CDNB as a substrate for GST assays were performed in triplicate at 25 °Cwith varying concentrations of CDNB (0.04–5.12 m M )anda constant GSH concentration (50 m M ). The K m and V max with GSH as a substrate were performed with varying concentrations of GSH (0.026–15.6 m M ) and a constant CDNB concentration (5 m M ) according to the method of Gallagher et al. [4]. Kinetic constants were calculated using nonlinear regression iterative program that gave a best fit of the experimentally measured activities to the Michaelis– Menten equation. k cat was calculated as follows according to the Michaelis–Menten equation for double substrate enzymes such as GST: V i ¼ k cat E 0 Á GSH½ K mGSH þ GSH½  Á CDNB½ K mCDNB þ CDNB½  where V i is the initial velocity and E 0 is the initial concentration of enzyme. SDS/PAGE The seven anion exchange chromatography purified GSTs were submitted to reducing and nonreducing SDS/PAGE, using a Bio-Rad Mini-Protean II electrophoresis unit, with a 15% and 10% resolving gel, respectively, and a 4% stacking gel [16]. Western blotting Purified GSTs from ion exchange chromatography were separated by SDS/PAGE and then transferred electropho- retically to Immobilon-P membrane (Millipore). Nonspe- cific binding sites were saturated by incubation in 50 m M Tris/HCl, 150 m M NaCl, 0.1% (v/v) Tween 20 (buffer A), containing 5% (w/v) nonfat dried milk for 1 h at room temperature. The membrane was then incubated with primary antibody diluted with buffer A containing 1% Table 1. Elution conditions of the four affinity fractions during anion exchange chromatography. Fraction NaCl (m M ) Time (min) NaCl (m M ) Time (min) NaCl (m M ) Time (min) NaCl (m M ) Time (min) C1E1 0–450 3 450–640 12 650–750 5 C1E2 0–350 3 350–400 11 400–450 1 450–500 5 C2E1 0–200 3 200–550 12 550–750 5 C2E2 0–350 3 350–450 12 450–750 5 4360 P. Hoarau et al. (Eur. J. Biochem. 269) Ó FEBS 2002 (w/v) dried milk (buffer B) overnight under gentle agitation at 4 °C. The membrane was then rinsed three times with buffer B for 5 min and incubated with peroxidase-conjugated goat anti-rabbit IgG from Bio-Rad diluted 1 : 2500 with buffer B for 1 h at room temperature. The proteins which cross- reacted with the antibody were detected by luminescence (ECL, Amersham). Three antisera were used as primary antibody: a rabbit anti-(pi class GST) (polyclonal antibody; Chemicon) diluted 1 : 2000, a rabbit anti-(mu class GST) (polyclonal antibody; Interchim) diluted 1 : 500 and a rabbit anti-(alpha class GST) (polyclonal antibody; Inter- chim) diluted 1 : 2000. RP-HPLC HPLC analysis of each of the four affinity fractions and of all the anion exchange chromatography-purified GST isoforms was performed using a reverse phase column (250 · 4 mm) Licrosorb RP 18 Merck, particle size 5 lm, a Merck model L 500 and a D 2000 integrator. The mobile phase linear gradient (0–50% B over 40 min) of 0.05% (v/v) trifluoroacetic acid in water (solvent A) and 0.05% (v/v) trifluoroacetic acid in acetonitrile (solvent B) was used at a flow rate of 0.8 mLÆmin )1 . Peptide detection was performed at 214 nm [17]. Samples were injected in a 500-lL loop valve with bracketing, a method which improved the resolution and increased the peak heights according to Guinn and Hendricx [18]. N-Terminal amino acid sequencing One of the seven anion exchange purified isoforms, which is recognized by the three antisera used, was chosen for N-terminal amino acid sequencing. For this purpose, 30 pmol were applied to 15% SDS/PAGE and then transferred to a poly(vinylidene difluoride) membrane (Immobilon-P Millipore). This analysis was carried out by the Laboratory of Protein Microsequencing of Pasteur Institute (Paris, France). Statistical analyses Statistical comparison of the results was made using a nonparametric Mann–Whitney test with n ¼ 3ineachcase. RESULTS Purification of GSTs and HPLC separation of subunits The GST activities with CDNB and ETHA in each affinity purified fraction are shown in Fig. 1. The highest activities were found with CDNB (Fig. 1A), whereas the activities measured with ETHA were % 100 times lower than with CDNB. The GST/CDNB activity showed significant differences (% twofold; Fig. 1A) according to the affinity column used for the first elution carried out with 0.2 m M GSH. No differences were observed between C1 and C2 in the second elution performed with 5 m M GSH and the measured activity appeared to be much lower than with 0.2 m M GSH. The GST/ETHA activity showed significant differences (% fivefold; Fig. 1B) according to the column used only in the case of the second elution. Quantitatively 75% of the total CDNB activity loaded on the two columns were recovered in the four fractions. RP-HPLC analysis was performed on each of the four affinity fractions (Fig. 2). As all GST subunits were eluted before 20 min, results were thus presented within this period. Six subunits were identified and numbered from 1 to 6 as a function of their retention time. In the C1E1 affinity fraction, three peptides were eluted at 12.06, 14.09 and 16.40 min after the elution gradient starts. In the C2E1 fraction, two peptides were detected at 14.83 and 16.40 min. In the C1E2 fraction, three peptides were eluted at 7.2, 9.02 and 14.83 min and in the C2E2 fraction, only one peptide was eluted at 14.83 min. The first affinity column (C1) allowed binding of the six subunits while the second column (C2) bound only subunits 5 and 6 (Fig. 2). Proteins of each of the four affinity fractions were also separated using anion exchange chromatography. GST isoforms 1-1, 2-2 were isolated from the C1E2 affinity fraction; GST isoforms 3-3, 4-4 and 3-6 from the C1E1 frac- tion; and GST isoforms 5-5 and 6-6 from the C2E1 fraction. The purified fractions, which showed a significant GST/ CDNB activity, were further analyzed by RP-HPLC (Fig. 3) and by SDS/PAGE in nondenaturing conditions and denaturing conditions (data not shown). Results showed that six GST isoforms (1-1, 2-2, 3-3, 4-4, 5-5, and 6-6) appeared to be composed of two identical subunits. They are thus homodimers with native apparent molecular weight ranging from 44 to 52 kDa. The seventh purified isoform (3-6) is an heterodimer consisting of two subunits: subunits 3 and 6, as shown by the retention time (Fig. 3). Activities and kinetics specificities Table 2 summarizes the retention time (already given in Fig. 3), the apparent molecular mass and the GST-specific enzymatic activities of purified GST isoforms of R. decus- satus. All purified isoforms catalyzed the conjugation of GSH with CDNB, whereas isoforms 1-1, 2-2 and 4-4 had zero or low activity with ETHA as a substrate. Table 2 shows four groups according to their specific activities. The first group, composed of isoforms 1-1 and 2-2, had low Fig. 1. GST activities using (A) CDNB and (B) ETHA as substrates in the two eluted fractions (E1 and E2) from the two affinity columns. C1, GSH–agarose column; C2, S-hexyl GSH–agarose. Values are given as means ± SD. Ó FEBS 2002 Purification of R. decussatus GST isoforms (Eur. J. Biochem. 269) 4361 activity with CDNB (< 100 lmolÆmin )1 Æmg )1 ) and no activity with ETHA; the second group, represented only by the 4-4 isoform, had high activity with CDNB (% 300 lmolÆmin )1 Æmg )1 ) and low activity with ETHA (< 5 lmolÆmin )1 Æmg )1 )l the third group of isoforms (5-5, 6-6 and3-6) showedhighactivitywithboth CDNB(> 300 lmolÆ min )1 Æmg )1 ) and ETHA (> 15 lmolÆmin )1 Æmg )1 ); and the fourth group (isoform 3-3) showed the highest activity with ETHA and an intermediate activity with CDNB. In previous work [13], it was observed that GST isoforms induced by pesticides showed great activity with ETHA as a substrate. Thus, among the purified isoforms, those pre- senting high GST activity with this substrate were chosen to determine the kinetic parameters. Table 3 displays the kinetic constants of the purified isoforms (3-3, 5-5, 6-6 and 3-6). No statistical differences in V max (around 200 lmolÆ min )1 Æmg )1 ) were observed in the case of CDNB or GSH for any of the isoforms studied. On the contrary, the isoforms presented statistically different affinities for the two substrates GSH and CDNB, except in the case of the affinity for GSH of GST3-6 and GST 6-6. The 3-3 isoform showed the highest affinity for CDNB (K m ¼ 0.17 m M ) compared to the isoforms 3-6, 5-5 and 6-6 (in decreasing order). GSH K m values displayed a wide range from 0.37 m M (5-5 isoform) to 4.82 m M (3-3 isoform). The comparison of k cat demonstrated that all purified isoforms had rapid and distinct turnover in a narrow range (214.3– 250.2Æs )1 ). Differences were observed when comparing the efficiencies (expressed by the ratio k cat /K m )forthetwo substrates. The GST 3-3 isoform showed the best efficiency for CDNB followed by GSTs 3-6, 5-5 and 6-6 (in decreasing order). For GSH, the GST 5-5 presented the best efficiency compared to isoforms 6-6, 3-6 and 3-3 (in decreasing order). The 3-6 isoform, with kinetic parameters between those of GST isoforms 3-3 and 6-6, seemed to have the properties of each of its subunits. Immunoblotting Fig. 4A shows Western blotting analysis of the seven purified isoforms from R. decussatus performed with rabbit anti-(pi class GST) serum. Cross-reactivity was observed with all the purified isoforms (apparent molecular mass from 22 to 26 kDa). Purified isoforms 1-1 to 6-6 presented one single band whereas the heterodimer 3-6 displayed two bands located at 24 and 23 kDa thus confirming the HPLC results (Fig. 3). Moreover, isoforms 4-4, 5-5 and 6-6 also gave a positive reaction in immunoblotting (data not shown) when tested with a rabbit anti-(R. decussatus GST) serum, produced in our laboratory [13]. Fig. 2. RP-HPLC of each affinity fraction (C1E1, C2E1, C1E2 and C2E2). Protein samples (% 50 lg) were loaded on a Licrosorb RP 18 column. Elution conditions are given in Material and methods. Fig. 3. RP-HPLC of each ionically purified GST isoform. Protein samples (5 lg) were loaded under the same conditions as given in Fig. 2. 4362 P. Hoarau et al. (Eur. J. Biochem. 269) Ó FEBS 2002 GST isoforms 1-1, 2-2 and 4-4 were also recognized in Western blotting analysis performed with rabbit anti-(alpha class GST) serum (Fig. 4B), whereas the rabbit anti-(mu class GST) serum only gave a positive reaction with isoforms 4-4 and 5-5 (Fig. 4C). N-Terminal amino acid sequence analysis As subunit 4 is the only one which cross-reacts with the three classes of mammalian GST antibodies, N-terminal amino acid analysis was performed to determine to which class this subunit belongs. Table 4 displays the results of the amino acid sequence compared to published results. There was 60% identity with GSH binding protein from the mussel M. edulis and with a GST from the nematode Clonorchis sinensis, and 53% with a GST from the platyhelminth Fasciola hepatica. N-Terminal sequence of R. decussatus presented 47, 26 and 20% of identity with the mu, alpha and pi class consensus sequences from mammals, respectively. The N-terminal sequence of subunit 4 of R. decussatus exhibits a high similarity with the corresponding consensus mu GST class from mammals. DISCUSSION In studies carried out on the blue mussel M. edulis [12], Fitzpatrick and Sheehan (1993) have shown that four different GST isoforms are induced in this animal in response to pollutants. In previous work [13] we demon- strated the same phenomenon in the clam R. decussatus with chemical pollutants especially with 2,2-bis-(p-chloro- phenyl)-1,1-dichloroethylene. Nevertheless, few studies have concerned the purification and characterization of GST isoforms in mollusks. The interest in using the clam R. decussatus is a result of the fact that it is a very abundant species commercialized for human consumption around the Mediterranean. In this paper, an attempt was made to purify the different GST isoforms present in R. decussatus. For this purpose, the use of two affinity columns (GSH–agarose and S-hexyl GSH–agarose) demonstrated that R. decussatus GSTs have different affinities for the chromatography matrices. These results are in agreement with those published for nonverte- brate GSTs [19]. Two elution concentrations were similarly used by Kunze [20] for the purification of GSTs from porcine liver. The originality of the method used in the present study is the combination of two elution concentrations and the use of two affinity columns mounted in series, which allowed the GSTs to be separated into four distinct groups. Their analysis by RP-HPLC showed that the first column, GSH–agarose, binds seven isoforms, whereas the S-hexyl GSH column allowed improvement of the purification yield only. In this study, reduced GSH was used as affinity matrix competitor for the two columns because all described GST isoforms have good affinity for free GSH. Moreover, the GSH is nonin- hibitory and may stabilize the enzyme [19], allowing further purifications and assays. S-hexyl was used as a competitor for the second column after elution with free GSH, no CDNB activity was measurable in the recovered fractions. Most of the GST affinity purifications reported [21,22] were performed using only the GSH–agarose column and showed arecoveryof% 50% of the initial CDNB activity. In the present study, 75% of the total GST activity measured with CDNB as a substrate was recovered in all the four affinity fractions. When measuring the activity in the front fraction, we observed that 25% of the GST activity was lost during the charge step. Some GST isoforms may not be bound to the two affinity matrices. Other types of substrate-linked mat- rices are now investigated in the laboratory. Clark [19] Table 2. Biochemical characteristics of R. decussatus GST isoforms. Results are expressed as means ± SD. Reverse phase HPLC retention time (Rt) of subunits expressed in minutes; apparent molecular mass of each subunit expressed in kDa (see Fig. 4); GST specific activities (lmol substrate per min per mg protein) of purified isoforms with CDNB and ETHA as substrates. ND, not detected. GST isoforms 1-1 2-2 3-3 4-4 5-5 6-6 3-6 HPLC retention time (Rt) (min) 7.20 9.02 12.06 14.09 14.83 16.40 12.06/16.40 Subunit molecular mass (kDa) 26 24 24 25 25 23 24/23 Specific activity (lmol substrateÆmin )1 mg protein )1 ) CDNB 38 ± 30 467 ± 41 290 ± 17 270 ± 35 86 ± 4 361 ± 21 91 ± 10.1 ETHA 59 ± 12 30 ± 3 15 ± 5 4.4 ± 2.4 ND 50 ± 10 0.8 ± 0.1 Table 3. Kinetic constants of purified GST isoforms (3-3; 5-5; 6-6; 3-6). V max and K m , k cat and ratio k cat /K m with GSH and CDNB as substrates. Results are expressed as means ± SD. Constant GST 3-3 GST 5-5 GST 6-6 GST 3-6 V max (CDNB) (lmolÆmin )1 Æmg )1 ) 204.1 ± 4.2 242.1 ± 25.3 222.4 ± 7.9 200.2 ± 9.3 V max (GSH) (lmolÆmin )1 Æmg )1 ) 181.6 ± 5.4 208.2 ± 16.3 199.6 ± 7.8 192.1 ± 10.4 K m (CDNB) (m M ) 0.17 ± 0.02 1.65 ± 0.49 2.88 ± 0.37 0.86 ± 0.19 K m (GSH) (m M ) 4.82 ± 0.33 0.37 ± 0.08 1.75 ± 0.43 2.20 ± 0.27 k cat (s )1 ) 214.3 ± 0.7 250.2 ± 6.4 234.4 ± 2.4 217.9 ± 4.8 k cat /K m (CDNB) (m M )1 Æs )1 ) 1244.97 151.72 81.34 252.69 k cat /K m (GSH) (m M )1 Æs )1 ) 44.417 672.95 133.82 99 Ó FEBS 2002 Purification of R. decussatus GST isoforms (Eur. J. Biochem. 269) 4363 reports that some GST isoforms are not linked to GSH affinity matrices. The same phenomenon may occur in R. decussatus. Moreover, some GSTs, particularly those belonging to mammalian theta class, are not able to metabolize CDNB and thus the activity of some GSTs may be underestimated or even undetected when using this assay. The GST activities recorded for the four affinity fractions varied according to the substrate used, the strongest activity being obtained with CDNB and the lowest with ETHA. ETHA activity, which was too low to be detected in the original post-mitochondrial fraction, was detectable after the separation of GST isoforms into several fractions. The seven isoforms, separated by anion exchange chro- matography, were partially characterized. K m , evaluated using CDNB and GSH, gave significantly different results for the four selected GST isoforms. K m using CDNB as a substrate in nonvertebrate organisms, showed a range of affinity between 0.05 and 1.8 m M [19]. This is in agreement with K m values from R. decussatus. K m values for GSH covered a wide range (0.3–2.9 m M ) like those reported in the literature for GST from nonvertebrate organisms (0.06– 2.7 m M ). This allows one of the isoforms to catalyze the conjugation in optimal conditions, whatever the concentra- tion of endogenous GSH. This phenomenon is of great interest as GSH is involved in many cellular processes, especially in redox cycles [23,24]. The conjugation turnover number, k cat , and the catalytic efficiency, k cat/ K m(substrate) , allowed us to distinguish the four GSTs tested. As few studies gave such kinetics constants for GST and none for GSTs from mollusc organisms, the comparison of the kinetic parameters from R. decussatus GSTs remained limited. Nevertheless, in the marsupial Antechinus stuartii, Bolton and Ahokas [22] found k cat values for CDNB rangingfrom40to290Æs )1 . Prapanthadara et al.[25] showed that the turnover of two GST isoforms called GST-4 and GST-4c from Anopholes dirus were 149.85 and 18.69Æs )1 for k cat , 172.24 and 62.30 m M )1 s )1 for the ratio k cat/ K m(GSH) and 713.57 and 186.9 m M )1 s )1 for the ratio k cat /K m(CDNB) .Asregardsk cat , evaluated in R. decussatus, our results are in good agreement with the data in the literature; nevertheless the range of variation is narrower. The same conclusion can be made for k cat /K m(substrate) . Some of the commercial antibodies are very specific to one class GST and are used for distinguishing GSTs from a large range of organisms. In the present study, all isoforms purified from R. decussatus were recognized by rabbit antisera specific of mammalian pi-class GST. Studies of aquatic species such as salmonids [26], Bufo bufo [27] and M. edulis [28], showed that their GSTs belong to the pi class. The GST isoforms 1-1 and 2-2 from R. decussatus were also recognized by rabbit antisera Fig. 4. Immunoblotting of each anion exchange purified isoform (100 ng per well). (A) Rabbit anti-pi class GST antisera (Chemicon). (B) Rabbit anti-(alpha class GST) serum (Interchim). (C) Rabbit anti-(mu class GST) serum (Interchim). MW, prestained low range SDS/PAGE standards from Sigma (phosphorylase b, BSA, ovalbumin, carbonic anhydrase, soybean trypsin inhibitor and lysozyme, respectively, 111, 73, 47.5, 33.9, 28.8 and 20.5 kDa) were not revealed by ECL but visualized by superimposition of the X-ray film on the membrane. Lane 1, purified 1-1 isoform; lane 2, purified 2-2 isoform; lane 3, purified 3-3 isoform; lane 4, purified 4-4 isoform; lane 5, purified 5-5 isoform; lane 6, purified 6-6 isoform; lane 7, purified 3-6 isoform. Table 4. Comparison between the N-terminal amino acid sequence of R. decussatus GST 4-4 isoform with other known GST sequences. Clonorchis sinensis and F.hepaticaGST sequences are from the SWISS-PROT database; M. edulis binding protein and mu, alpha, pi GST classes consensus sequences of mammals are from Fitzpatrick et al. [28]. Species N-terminal sequence Identity (%) GST 44 Ruditapes decussatus SELAYKKIRGLAQMN – GSH binding protein Mytilus edulis PTLGYWKIRGLAQPVR 60 GST Clonorchis sinensi MAPVLGYWKIRGLAQPIR 60 GST Fasciola hepatica MPAKLGY-KLRGLAQ 53 Consensus mu PMTLGYWDIRGLAHAIR 47 Consensus alpha AGKPVLHYFNARGRME 26 Consensus pi PPYTVVYFPVRGGCAAMR 20 4364 P. Hoarau et al. (Eur. J. Biochem. 269) Ó FEBS 2002 specific for the mammalian alpha class GSTs. These two GSTs, which are highly similar in their activities with CDNB and ETHA, may be related to the mammalian alpha class GST. The isoforms 3-3, 6-6, 3-6 and probably also 5-5 may belong to the mammalian pi class according to their immunological and biochemical properties. However, the immunological and activity properties of the 4-4 isoform are not sufficient to assign it to a GST class; this isoform is recognized by the three rabbit antisera (anti-alpha, anti-mu and anti-pi). It is now accepted that mu, alpha and pi mammalian GST classes have a common precursor the alpha/pi/mu class which probably arose from theta gene duplication [29,30]. Some GST isoforms of the alpha/mu/pi precursor class have been found in platyhelminths [29] and, more recently, in the nematode Caenorhabditis elegans [30]. The primary amino acid sequence of subunit 4 from R. decussatus showed 53% identity with GST from F.hepatica (Fh 47) which is clearly an isoform of the alpha/mu/pi precursor class. GST 4-4 shows similarity with a conserved isoform of the common alpha/mu/pi class; this hypothesis should be clarified by the determin- ation of its mRNA sequence. Immunological and biochemical properties are currently used to support the classification of individual GST isoforms [1]. In the case of the mollusk R. decussatus, direct primary amino acid sequence analysis was also performed. However, as in other invertebrates [28,31], the purified isoforms of R. decussatus could not readily be assigned to the present classification system which is based on mam- malian GSTs. ACKNOWLEDGEMENTS P. Hoarau was supported by a joint fellowship from PACA Region and SAFEGE-CETIIS, Aix-en-Provence. The authors thank Dr J. Trin- chant, Laboratoire de Physiologie Ve ´ ge ´ tale, Universite ´ de Nice Sophia- Antipolis for the use of an HPLC MERCK system. REFERENCES 1. Mannervick, B. (1985) The isoenzyme of the glutathione S-transferase. Adv. Enzymol. Relat. Areas. Mol. Biol. 57, 357–417. 2. Ketterer, B. (1988) Protective role of glutathione and glutathione transferases in mutagenesis and carcinogenesis. Mutat. Res. 2, 343–361. 3. Daniel, V. (1993) Glutathione S-transferases: gene structure and regulation of expression. CRCCrit.Rev.Biochem.Mol.Biol.28, 173–207. 4. Gallagher, E.P., Sheehy, K.M., Lame, M.W. & Segall, H.J. (2000) In vitro kinetics of hepatic glutathione S-transferase conjugaison in Largemouth Bass and Brown Bullheads. Environ. Toxicol. Chem. 19, 319–326. 5. Gallagher, E.P. & Sheehy, K.M. (2000) Altered glutathione S-transferase catalytic activities in female brown bullheads from a contaminated central Florida lake. Mar. Environ. Res. 50, 399– 403. 6. Board, P.G., Baker, R.T., Chelvanayagam, G. & Jermin, L.S. (1997) Zeta a novel class of glutathione transferases in a range of species from plants to humans. Biochem. J. 328, 928–935. 7. Edwards, R., Dixon, D.P. & Walbot, V. (2000) Plant glutathione S-transferases: enzymes with multiple functions in sickness and in health. Trends. Plant. Sci. 5, 193–198. 8. Vuilleumier, S. (1997) Bacterial glutathione S-transferases. What are they good for? J. Bacteriol. 179, 1431–1441. 9. Zhou, Z.H. & Syvanen, M. (1997) A complex glutathione S-transferase gene family in the housefly Musca domestica. Mol. Gen. Genet. 256, 187–194. 10. Satoh, K., Kitihara, A., Soma, Y., Inaba, Y., Hatayama, I. & Sato, K. (1985) Purification, induction and distribution of pla- cental glutathione transferase: a new marker enzyme for pre- neoplastic cells in the rat chemical hepatocarcinogenesis. Proc. Natl Acad. Sci. USA 82, 3964–3968. 11. Henson, K.L., Sheehy, K.M. & Gallagher, E.P. (2000) Con- servation of a glutathione S-transferase in marine and freshwater fish. Mar. Environ. Res. 50, 17–21. 12. Fitzpatrick, P.J. & Sheehan, D. (1993) Separation of multiple forms of glutathione S-transferase from the blue mussel, Mytilus edulis. Xenobiotica 23, 851–861. 13. Hoarau, P., Gnassia-Barelli, M., Rome ´ o, M. & Girard, J.P. (2001) Differential induction of glutathione S-transferases in the clam Ruditapes decussatus exposed to organic compounds. Environ. Toxicol. Chem. 20, 523–529. 14. Habig, W.H., Pabst, M.J. & Jakobi, W.B. (1974) Glutathione S-transferases: The first enzymatic step in mercapturic acid for- mation. J. Biol. Chem. 249, 7130–7139. 15. Bradford, M. (1976) A rapid and sensitive method for the quan- tification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. 16. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage. Nature 227, 680–685. 17. Farrants, O., Meyer, D.J., Coles, B., Southan, C., Aitken, A., Johnson, P.J. & Ketterer, B. (1987) The separation of glutathione transferase subunits by reverse-phase high-pressure liquide chro- matography. Biochem. J. 245, 423–428. 18. Guinn, G. & Hendrix, D.L. (1985) Bracketing, a simple loading technique that increases sample recovery, improves resolution, and increases sensitivity in high performance liquid chromatography. J. Chrom. 348, 123–129. 19. Clark, A.G. (1989) The comparative enzymology of the glutathi- one S-transferases from non-vertebrate organisms. Comp. Bio- chem. Physiol. 92, 419–446. 20. Kunze, T. (1997) Purification and characterisation of class alpha andmuglutathioneS-transferases from porcine liver. Comp. Biochem. Physiol. 116 B, 397–406. 21. Bolton, R.M. & Ahokas, J.T. (1997) Purification and character- ization of hepatic glutathione transferase from an insectivorous marsupial, the brown antechinus (Antechinus stuartii). Xenobiotica 27, 573–586. 22. Yuen, W.K. & Ho, J.W. (2001) Purification and characterization of multiple glutathione S-transferase isozymes from Chironomidae larvae. Comp. Biochem. Physiol. 129A, 631–640. 23. Canesi, L., Viarengo, A., Leonzio, C., Filippelli, M. & Gallo, G. (1999) Heavy metals and glutathione metabolism in mussel tissues. Aquat. Toxicol. 46, 67–76. 24. Lu, S.C. (2000) Regulation of glutathione synthesis. Curr. Top. Cell Regul. 36, 95–116. 25. Prapanthadara, L., Promtet, N., Koottathep, S., Somboon, P. & Ketterman, A.J. (2000) Isoenzymes of glutathione S-transferase from the mosquito Anopheles dirus species B: the purification, partial characterization and interaction with various insecticides. Insect Biochem. Mol. Biol. 30, 395–403. 26. Dominey, R.J., Nimmo, I.A., Cronshaw, A.D. & Hayes, J.D. (1991) The major glutathione S-transferase in Salmonid fish livers is homologous to the mammalian pi class GST. Comp. Biochem. Physiol. 100B, 93–98. 27. Di Ilio, C., Aceto, A., Bucciarelli, T., Dragani, B., Angelucci, S., Miranda, M., Poma, A., Amicarelli, F., Barra, D. & Frederici, G. (1992) Glutathione transferase isoenzymes from Bufo bufo embryos at an early development stage. Biochem. J. 283, 217–222. 28. Fitzpatrick, P.J., Krag, T.O.B., Hojrup, P. & Sheehan, D. (1995) Characterization of glutathione S-transferase and a related Ó FEBS 2002 Purification of R. decussatus GST isoforms (Eur. J. Biochem. 269) 4365 glutathione-binding protein from gill of the blue mussel Mytilus edulis. Biochem. J. 305, 145–150. 29. Pemble, S.E. & Taylor, J.B. (1992) An evolutionary perspective on glutathione S transferase inferred from class-theta glutathione transferase cDNA sequences. Biochem. J. 287, 957–963. 30. Campbell, A.M., Teesdale-Spittle, P.H., Barrett, J., Liebau, E., Jefferies, J.R. & Brophy, P.M. (2001) A common class of nematode glutathione S-transferase revealed by the theoretical proteome of the model organism Caenorhabditis elegans. Comp. Biochem. Physiol. 128B, 701–708. 31. Sheehan, D., Meade, G., Foley, V.M. & Dowd, C.A. (2001) Structure, function and evolution of glutathione transferases: implications for classification of non mammalian members of an ancient enzyme superfamily. Biochem. J. 360, 1–16. 4366 P. Hoarau et al. (Eur. J. Biochem. 269) Ó FEBS 2002 . pi-class GST. Studies of aquatic species such as salmonids [26], Bufo bufo [27] and M. edulis [28], showed that their GSTs belong to the pi class. The GST isoforms 1-1 and 2-2 from R. decussatus. deals with the purification and the partial char- acterization of glutathione S- transferase (GST) isoforms from the clam Ruditapes decussatus. For the first step of purification, two affinity columns,. characteristics of R. decussatus GST isoforms. Results are expressed as means ± SD. Reverse phase HPLC retention time (Rt) of subunits expressed in minutes; apparent molecular mass of each subunit