当前位置:首页 >> 三年级英语 >>

Affinity purification of Schistosoma japonicum glutathione-S-transferase and its site-directed mutan

Journal of Chromatography A, 852 (1999) 151–159

Af?nity puri?cation of Schistosoma japonicum glutathione-Stransferase and its site-directed mutants with glutathione af?nity chromatography and immobilized metal af?nity chromatography
a, a a b Hsiu-Mei Chen *, San-Lin Luo , Kai-Ti Chen , Chong-Kuei Lii

Department of Chemical Engineering, National Taiwan University of Science and Technology, Taipei 106, Taiwan b Department of Nutrition, Chung Shan Medical College, Taichung 402, Taiwan

Abstract A C-terminally polyhistidine-tagged protein of Schistosoma japonicum glutathione-S-transferase, named as SjGST / His, and its Cys85→Ser, Cys138→Ser, and Cys178→Ser site-directed mutants were prepared and highly expressed in Escherichia coli. Both immobilized metal af?nity chromatography (IMAC) and glutathione (GSH) af?nity chromatography were used to purify these four enzymes. All of them were puri?ed with equal ef?ciency by Ni 21 -chelated nitrilotriacetic acid agarose gel, but not by GSH Sepharose 4B gel. The protein amounts of wild-type and Cys85→Ser enzymes puri?ed by the latter gel were three to seven-fold greater than those of the other two enzymes puri?ed by the same gel, while their speci?c activities were two-fold lower, presumably because of the occurrence of noncovalent aggregation. Both puri?cation methods yielded highly pure enzymes, while there were minor amounts of inter- and intra-disul?de forms in the IMAC puri?ed enzymes except for the Cys85→Ser mutant. Addition of dithiothreitol to GSH-af?nity puri?ed enzymes shifted all of their mass spectra of matrix-assisted laser desorption / ionization-time of ?ight mass spectrometry toward low molecular-mass regions, while addition of GSH to IMAC puri?ed enzymes shifted the spectra toward high molecular-mass regions. The shift values of wild-type enzyme were larger than those of the three mutants, indicating that the Cys85, Cys138, and Cys178 residues were S-thiolated by GSH during the GSH-af?nity puri?cation. This result was con?rmed by isoelectric focusing. These ?ndings suggest that IMAC is more ef?cient than the conventional GSH-af?nity system for the puri?cation of SjGST / His enzyme, especially for its mutants and fusion proteins. ? 1999 Elsevier Science B.V. All rights reserved. Keywords: Af?nity chromatography; Schistosoma japonicum; Immobilized metal af?nity chromatography; Gluatathione-Stransferase; Enzymes

1. Introduction Glutathione-S-transferases (GSTs) are a family of multifunctional proteins that can catalyze the nucleophilic addition of the thiol group of glutathione (GSH) to a variety of electrophiles or bind a range of hydrophobic ligands [1]. They are important thera*Corresponding author.

peutic targets in disease [2]. For example, Schistosoma japonicum GSTs, the major detoxi?cation enzymes in S. japonicum, have promising vaccine potential against Schistosomiasis [3,4], and their inhibitors are novel antischistosomal drugs [5]. The gene of Mr 26 000 GST from S. japonicum (SjGST) has been cloned [3,6], highly expressed in various host cells, and successfully used as a gene fusion system for high expression and af?nity puri?cation

0021-9673 / 99 / $ – see front matter ? 1999 Elsevier Science B.V. All rights reserved. PII: S0021-9673( 99 )00490-2


H.-M. Chen et al. / J. Chromatogr. A 852 (1999) 151 – 159

of recombinant proteins [7,8]. The protein contains four cycteine residues, with only the Cys169 residue buried inside the molecule and the other three, Cys85, Cys138, and Cys178, located on the surface of the protein. The structure of the recombinant protein produced from Escherichia coli indicates that neither intra- nor inter-disul?de linkage exists in the active dimeric enzyme [5]. GSTs are conventionally puri?ed with GSH af?nity chromatography [7] or other GSH-derivative af?nity systems where reduced GSH is routinely used as an eluting agent. However, there are several drawbacks with such puri?cation systems. First, most of the GST mutants, which are the subjects of a wide range of investigations involving the functional roles of the residues of GSTs, cannot be readily puri?ed by this method because their GSH binding af?nity may be affected by the mutations. Second, because recombinant SjGST loaded onto the GSH af?nity gels tends to form soluble aggregates as a result of the oxidation of the free thiol groups [9], high concentrations of reduced GSH or other reducing thiols are usually required for ef?cient elution. Protein S-glutathiolation, the formation of mixed disul?des on protein thiol groups with GSH, has been reported for a large variety of proteins [10,11]. Although the effect of the linkage of GSH to SjGST is not clear, the phosphatase activity of carbonic anhydrase III is known to be increased by monoglutathiolation but decreased by diglutathiolation [12]. In addition, it has been reported that the activities of Class Pi GSTs are sensitive to S-thiolation [13–15]. Finally, if there is glutathione disul?de (GSSG) contamination in the GSH elution buffer, the possibility of protein S-gluathiolation increases [10]. Recently, immobilized metal af?nity chromatography (IMAC), one of the most widely used puri?cation systems for recombinant proteins, was used to purify different recombinant GSTs [16,17]. Metal binding sites, such as spatially adjacent super?cial His residues [16] or a poly-His tag [17], were genetically engineered into human GST M1-1 and SjGST, respectively, to facilitate their puri?cation with IMAC. Both studies showed that GSTs could be readily puri?ed with this system as long as they contained accessible metal coordinating sites. In this study, we investigated the performance of

both the GSH af?nity chromatography and the IMAC on the puri?cation of SjGST and its sitedirected mutants. A six-His tag was engineered to the C-terminal end of SjGST to produce the metalbinding fusion protein, SjGST / His [18]. The three super?cial Cys residues of SjGST / His, Cys85, Cys138, and Cys178, were individually substituted with Ser. These mutants are particularly important in elucidating the functional roles of these Cys residues in enzyme activity and in probing the occurrence of oxidative aggregation and protein S-thiolation.

2. Experimental

2.1. Materials
E. coli BL21(DE3) was purchased from NovaGen (Madison, WI, USA). The plasmid pGSTH carrying the SjGST / His gene was constructed in this laboratory by cloning the SjGST gene fragment of pGEX5X-2 (Pharmacia Biotech, Uppsala, Sweden) into the multiple cloning sites of pET-30b (Novagen) [18], which are located before the gene of the six-His tag. Enzymes for DNA manipulation were either from New England BioLabs (Beverly, MA, USA) or from AGS (Heidelberg, Germany). GSH Sepharose 4B gels and Ni 21 -chelated nitrilotriacetic acid (Ni 21 NTA) agarose gels were from Pharmacia Biotech and QIAGEN (Santa Clarita, CA, USA), respectively. 1-Chloro-2,4-dinitrobenzene (CDNB) was from Wako (Osaka, Japan). GSH and imidazole were from Sigma (St. Louis, MO, USA). Ampholyte was from Pharmacia LKB (Piscataway, NJ, USA) and Gelbond was from Serva (Westbury, NY, USA). Sinapinic ¨ acid was from Hewlett-Packard (Boblinger, Germany).

2.2. Site-directed mutagenesis of SjGST /His
Site-directed mutagenesis of the SjGST / His gene on the pGSTH plasmid was performed following the method of Kunkel et al. [19]. The synthetic oligonucleotides used as primers to carry out the reactions are listed in Table 1. The primers were designed so that an extra restriction site (Table 1) was created in each mutation. The plasmids containing the mutated SjGST / His genes were screened by restriction diges-

H.-M. Chen et al. / J. Chromatogr. A 852 (1999) 151 – 159 Table 1 Synthetic oligonucleotides used as primers for site-directed mutagenesis SiGST / His mutants


Created restriction site ]]] Cys85→Ser CACAACATGTTGGGTGGATCCCCAAAAGAGCGTGCA Bam HI ] ]]] Cys138→Ser AATGTTCGAAGATCGCTTAAGTCATAAAACATATTTAAATGGTGAT A? II ]] ]]] Cys178→Ser CGTTCCCAAAATTAGTAAGCTTTAAAAAACGTATTGAAGCT Hind III ]] a Bold letters: the substituted nucleotides; underlines: the mutant amino acid codons; upper lines: the created restriction site.



tions, sequenced, and transformed into E. coli BL21(DE3).

2.5. IMAC puri?cation of WT and mutant SjGST / His
Cell pellets harvested from 30 ml culture were suspended in 1.5 ml NTA loading buffer (0.1 M NaH 2 PO 4 , 0.1 M Tris–HCl, pH 8) and sonicated. Then, the clear cell lysate was mixed with 100 ml of 50% Ni 21 -NTA gel slurry, shaken gently at room temperature for 15 min, and centrifuged. The resulting pellet was washed three times with 10 bed volumes of NTA loading buffer containing 20 mM imidazole. To elute the protein, 100 ml NTA elution buffer (250 mM imidazole, 0.1 M NaH 2 PO 4 , 0.1 M Tris–HCl, pH 8) was used.

2.3. Enzyme Production
E. coli BL21(DE3) was the host strain used to produce wild-type (WT) and mutant SjGST / His enzymes. A single colony of this strain containing either WT or mutated pGSTH plasmids was used for inoculating Luria–Bertani (LB) broth containing 30 mg / ml kanamycin. After incubating overnight at 378C, the culture was diluted 25-fold with the same medium in a shaking ?ask. At an A 600 nm of 0.7–0.8, the expression was induced by the addition of isopropyl-b-D-thiogalactopyranoside (IPTG) to a ?nal concentration of 1 mM, and incubation was then continued for another 3 h. Cells were harvested by centrifugation followed by puri?cation as described below.

2.6. Assays
SjGST / His activities were determined with 1 mM GSH and 1 mM CDNB as substrates at pH 6.5 and 258C [1]. One unit (U) of WT or mutant SjGST / His was de?ned as the amount of enzyme required to produce 1 mmol of product per minute. Protein concentration was determined by the method of Bradford [20] with bovine serum albumin used as a standard.

2.4. GSH af?nity puri?cation of WT and mutant SjGST /His
Cell pellets harvested from 300 ml culture were suspended in 15 ml phosphate-buffered saline (PBS; 140 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , 1.8 mM KH 2 PO 4 , pH 7.4) and disrupted by sonication. The clear cell lysate was applied to a column packed with 200 ml GSH Sepharose 4B gels, followed by three washes with at least 10 bed volumes of PBS. Then, the protein was eluted with 400 ml GSH elution buffer (30 mM GSH, 50 mM Tris–HCl, pH 8). To remove GSH, the eluted protein was diluted four times with PBS and dialyzed against at least 2000 volumes of PBS with Mr cut off (MWCO) 3500 membranes.

2.7. Molecular mass measurements
The molecular masses of proteins were measured by matrix-assisted laser desorption / ionization-time of ?ight mass spectrometry (MALDI-TOF-MS), HP G2025A, which incorporated a 337 nm nitrogen laser with a desorption / ionization pulse width duration of 3 ns and an electron multiplier detector. The instrument was operated in positive ion re?ection mode with an acceleration potential of 128 kV. The mass scale was calibrated using peptide standard with a matrix of sinapinic acid. Sinapinic acid was also


H.-M. Chen et al. / J. Chromatogr. A 852 (1999) 151 – 159

used as the matrix for each sample, and an average of 155 laser shots were taken.

2.8. Isoelectric focusing ( IEF)
For IEF, an equal amount of protein from each sample was applied onto a 0.8 mm thin gel containing 4% acrylamide and 2% LKB preblended pH 3.5–10 or pH 5–8 ampholyte. The gel was prefocused for 10 min at 48C before sample loading and samples were focused for 60 min at the same temperature with a voltage limited to 1500 V and a current and power limited to 2.75 mA / cm and 1.125 W/ cm gel width, respectively. After IEF, the gel was stained with Commassie blue.

3. Results and discussion

3.1. Af?nity puri?cation of WT and mutant SjGST / His enzymes
The same puri?cation conditions were used for all enzymes in both the GSH and the Ni 21 -NTA af?nity puri?cation methods. The gel amounts used in both methods were much less than those required for the optimal binding, and signi?cant amounts of enzymes were therefore still observed in the ?ow-through and the wash fractions (results not shown). 30 mM GSH and 250 mM imidazole were used to elute proteins

from the GSH and the Ni 21 -NTA gels, respectively, because they yielded the best elution ef?ciency. Table 2 shows the values of total protein and speci?c activities of WT and mutant SjGST / His enzymes puri?ed from the GSH and the Ni 21 -NTA af?nity gels. The protein amounts and speci?c activities recovered from the Ni 21 -NTA gels were similar for the four enzymes, indicating that the production, activity, and metal binding af?nity of SjGST / His were not affected by the mutations. The presence of high imidazole concentration (250 mM) or the addition of 20 mM b-mercaptoethanol in the NTA elution buffer had no effect on enzyme activity. On the other hand, the values in the GSH af?nity puri?cation were dramatically different among the four different enzymes. The protein amounts of Cys138→Ser and Cys178→Ser mutants recovered from the GSH gels were much less than those of the other two enzymes, implying that the ability of SjGST / His to bind with GSH af?nity gels could be affected by these mutations. However, the speci?c activities of these enzymes were higher than those of the other two and close to the results in Ni 21 -NTA. The four-fold dilution and subsequent dialysis of WT and Cys85→Ser SjGST / His enzymes increased their speci?c activities to 11.1 and 11.6 U / mg, respectively, suggesting that a portion of the eluted enzymes were in inactive aggregated forms. The aggregation might have resulted from a high density of adsorbed proteins on the GSH af?nity gels, which were subsequently eluted in high concentrations (67–82 mg / ml).

Table 2 Total protein and speci?c activity of WT and mutant SjGST / His puri?ed by GSH af?nity chromatography and IMAC a SjGST / His form Crude extracts: Speci?c activity (U / mg) 3.2 2.1 5.3 4.7 GSH gels b Total protein (mg) 26.8 32.9 4.4 8.3 Speci?c activity (U / mg) 6.3 6.9 16.1 12.1 Ni 21 -NTA gels c Total protein (mg) 3.6 4.6 3.7 4.2 Speci?c activity (U / mg) 14.0 12.5 12.1 12.6

WT Cys85→Ser Cys138→Ser Cys178→Ser
a b

Data are reported as an average of three runs. The enzyme was puri?ed from 300 ml cell cultures with 200 ml GSH Sepharose 4B gels. c The enzyme was puri?ed from 30 ml cultures with 50 ml Ni 21 -NTA gels.

H.-M. Chen et al. / J. Chromatogr. A 852 (1999) 151 – 159


3.2. Sodium dodecylsulfate–polyacrylamide gel electrophoresis ( SDS–PAGE) of WT and mutant SjGST /His enzymes
Fig. 1 shows the SDS–PAGE electrophoretic patterns of WT and mutant SjGST / His enzymes puri?ed by using the two different af?nity puri?cation methods. All the enzymes puri?ed from the GSH adsorbents showed only a single band (G1) under both the reducing and the nonreducing conditions (Fig. 1a), suggesting that the formation of inactive oxidative aggregates between the SjGST / His monomers was eliminated at 30 mM GSH elution. This implies that noncovalent aggregation was responsible for the decrease in the activities of eluted WT and Cys138→Ser SjGST / His enzymes from the GSH af?nity adsorbents. For the IMAC puri?ed enzymes (Fig. 1b), the patterns under reducing conditions were the same as those in Fig. 1a. However, several minor forms of oxidative aggregates at higher molecular masses (G3) were observed under nonreducing conditions, suggesting that intermolecular disul?de linkages formed between SjGST / His monomers when the enzymes were highly expressed in E. coli. The extra minor bands (G2) located below the major bands in WT, Cys138→Ser, and Cys178→Ser SjGST / His enzymes represented intramolecularly oxidized mono-

mers which had higher mobility in SDS-PAGE than the reduced forms [21–23]. Therefore, the disappearance of this band in the Cys85→Ser mutant suggests that the intramolecular disul?de linkage may occur between Cys85 and other Cys residues.

3.3. MALDI-TOF-MS measurements of WT and mutant SjGST /His enzymes
Fig. 2 shows the MALDI-TOF-MS spectra of WT SjGST / His enzyme from different preparations. Only the spectra of the monomer forms are presented here because the signal intensities of dimeric forms were very low. The spectrum of the protein monomer eluted from the GSH af?nity adsorbents showed a peak value around 29 001 m /z. The addition of 30 mM dithiothreitol (DTT) to the protein sample shifted the spectrum to the left, reducing the molecular mass of the peak value to 28 487. In contrast, for the WT enzyme puri?ed from the Ni 21 -NTA gels, the spectrum showed a peak value of around 28 425 and the addition of 30 mM GSH shifted the spectrum to the right with a peak value at 29 031. We routinely observed similar spectral shifts for WT enzyme. In addition, we examined the molecular masses of WT enzyme eluted with only 10 mM GSH from GSHaf?nity gels and the dialyzed WT enzyme and obtained spectra similar to Fig. 2a. These results

Fig. 1. SDS–PAGE of WT and mutant SjGST / His puri?ed by (a) GSH Sepharose-4B gels and (b) Ni 21 -NTA gels under reducing (lane 1–4) and nonreducing (lane 5–8) conditions. (a) 2 mg and (b) 15 mg of protein was applied to each lane on 0.75 mm thick 12.5% polyacrylamide gels. For reducing electrophoresis, the protein was analyzed in the presence of 1% b-mercaptoethanol. Lanes 1, 55WT, 2, 65Cys85→Ser, 3, 75Cys138→Ser, 4, 85Cys178→Ser. The reduced, intra-disul?de linked, and inter-disul?de linked forms of SjGST / His are designated as G1, G2, and G3, respectively.


H.-M. Chen et al. / J. Chromatogr. A 852 (1999) 151 – 159

Fig. 2. MALDI-TOF mass spectra of WT SjGST / His. (a) Enzyme eluted from GSH gels. (b) 30 mM DTT 1 enzyme eluted from GSH gels. (c) Enzyme eluted from Ni 21 -NTA gels. (d) 30 mM GSH 1 enzyme eluted from Ni 21 -NTA gels. Arrows indicate the monomers of the enzyme at different states.

imply that the WT SjGST / His enzyme puri?ed from GSH-af?nity gels was S-thiolated with eluting GSH, while that from Ni 21 -NTA gels was not. The spectra obtained in this study were not narrow and sharp at all, presumably because of the existence of several different mixed-disul?de forms. The shift-patterns of the MALDI-TOF-MS spectra of the three mutant SjGST / His enzymes were similar to those of the WT form. Table 3 summarizes the peak values of their monomer forms in each spectrum. As described above, the addition of DTT to all of the GSH-af?nity puri?ed enzymes decreased their measured molecular masses, while the addition of GSH to all the IMAC puri?ed enzymes increased the values. The shift values of the WT enzyme were larger than those of the mutant enzymes. In addition,

for all of the three mutant enzymes, the molecular masses of the DTT-treated GSH-af?nity puri?ed forms and IMAC puri?ed forms were close to those of WT enzyme of the same preparations, while their oxidized forms were smaller. These results suggest that the Cys85, Cys138, and Cys178 residues were located on the enzyme surface and were susceptible to oxidative modi?cation by GSH. The theoretical molecular masses of the reduced monomers of WT and any one of the mutant SjGST / His enzymes are 28 217 and 28 201, respectively, based on the amino acid sequences of the cloned protein monomers. That is, the measured molecular masses of the enzymes eluted from Ni 21 -NTA adsorbents were higher than the theoretic values. This suggests that it is possible that the crude enzymes produced by E. coli

Table 3 MALDI-TOF-MS measurements of WT and mutant SjGST / His puri?ed with either GSH Sepharose 4B gels or Ni 21 -NTA gels Puri?cation gels GSH States Monomer of SjGST / His form (M r ) WT Puri?ed form Puri?ed form 130 mM DTT Puri?ed form Puri?ed form 130 mM GSH 29 001 28 487 Cys85→Ser 28 841 28 416 Cys138→Ser 28 848 28 474 Cys178→Ser 28 835 28 522

Ni 21 -NTA

28 425 29 031

28 503 28 730

28 451 28 658

28 472 28 690

H.-M. Chen et al. / J. Chromatogr. A 852 (1999) 151 – 159


BL21(DE3) were somewhat S-thiolated by molecules smaller than GSH. The same phenomena were also observed for in vivo S-thiolation by GSH for a variety of proteins produced by E. coli [24], Saccharomyces cerevisiae [25], and other cells [26,27], as well as the S-thiolation by other intracellular low-molecular-mass thiols [26,28]. We also analyzed the amino acid compositions of dialyzed WT SjGST and SjGST / His enzymes eluted from GSH-af?nity adsorbents and obtained higher amounts of glutamic acid and glycine than the theoretical values predicted from the sequences of unmodi?ed enzymes in both enzymes (data not shown). The amount of cysteine obtained was not higher than the theoretical value because it was unstable in direct acid hydrolysis. However, these results were suf?cient to imply that the enzymes had been S-thiolated by GSH during the puri?cation. For the IMAC-puri?ed enzymes, the results of amino acid analysis were not much different from the values predicted from the unmodi?ed enzyme sequences. Peptide digestions, more accurate molecular mass measurement, and amino acid analyses may be helpful in further identifying the small thiol molecules which are linked with crude SjGST / His enzymes in E. coli BL21(DE3) in vivo.

Fig. 3. IEF of WT (lanes 1, 2), Cys85→Ser (lanes 3, 4), Cys138→Ser (lanes 5, 6), and Cys178→Ser (lanes 7, 8) SjGST / His puri?ed by GSH Sepharose-4B gels. 20 mg of protein was applied to each lane. Lanes 1, 3, 5, 75puri?ed native enzymes; 2, 4, 6, 85puri?ed enzymes treated with 30 mM DTT at 378C for 30 min.

3.4. IEF of WT and mutant SjGST /His enzymes
Recently, IEF analysis has been applied to study the occurrence of thiol–disul?de exchanges between protein thiol and GSH of several proteins [27,29]. Based on the presence of one negative charge on GSH (an NH 1 and a COO 2 group from the gluta3 mate and a COO 2 from the glycine residue of GSH), the isoelectric point (pI) of a protein can be changed once it reacts with GSH. Otherwise, the presence of acidity in the a-carboxyl group in the glutamate residue of GSH (pKa ?2.8) may also manifest a change to GSH modi?ed protein on IEF gels. Therefore, we used IEF to verify the S-glutathiolation of SjGST / His enzymes. DTT and GSSG were each used to react with the enzyme samples before loading on IEF gels to con?rm the modi?cation. Figs. 3 and 4 show the IEF results of WT and mutant SjGST / His enzymes puri?ed from GSH and Ni 21 NTA gels, respectively. All of the GSH-af?nity puri?ed enzymes displayed many isoforms over a

wide pH range (pI?6.0–6.8) under nonreducing conditions on the IEF gels. However, all of them shifted to a narrower and more basic pH range (pI?6.5–6.85) after the DTT treatment. This suggests that the SjGST / His enzymes were S-thiolated by GSH during the GSH-af?nity puri?cation. The IEF patterns of the mutant enzymes were not clearly different from that of the WT enzyme because of the extreme multiplicity of the mixed-disul?de forms. On the other hand, the IEF isoforms of SjGST / His enzymes puri?ed from Ni 21 -NTA adsorbents (Fig.

Fig. 4. IEF of WT (lanes 1–3), Cys85→Ser (lanes 4–6), Cys138→Ser (lanes 7–9), and Cys178→Ser (lanes 10–12) SjGST / His puri?ed by Ni 21 -NTA gels. 10 mg of protein was applied to each lane. Lanes 1, 4, 7, 105puri?ed enzymes treated with 30 mM GSSG at 378C for 30 min; 2, 5, 8, 115puri?ed enzymes treated with 30 mM DTT at 378C for 30 min; 3, 6, 9, 125puri?ed native enzymes.


H.-M. Chen et al. / J. Chromatogr. A 852 (1999) 151 – 159

4) were less than those in Fig. 3. All of the untreated enzymes were unexpectedly focused at slightly more acidic regions than the enzymes treated with DTT, suggesting that the enzymes were S-thiolated in vivo by negatively charged small unknown molecules. All of the DTT-treated enzymes shifted to two distinct pI values (pI?6.2 and 6.3), indicating that all the enzymes had two reduced isoforms. One of them was the native form, and the other might have resulted from the deamidation of Asn144 residue on the Asn144–Gly145 bond in the enzymes. Deamidation of an Asn residue that is followed by a Gly residue is a common non-catalytic reaction in aged or heat-treated proteins under neutral or alkaline conditions [30]. It results in the formation of either an Asp–Gly or isoAsp–Gly bond and consequently increases the acidity of the pI value of the enzymes [31]. Treatment with GSSG clearly shifted the pI value of the IMAC-puri?ed WT enzyme by about 0.2 but did not shift the pI value of the mutant enzymes. However, the IEF patterns of the treated mutant enzymes were somewhat different from those of the untreated ones, suggesting the occurrence of disul?de exchanges.

enzymes. The results of this study also suggest that for the GSH-af?nity puri?cation of GST fusion proteins, S-thiolation by GSH or oxidative aggregation should be cautiously examined, especially in cases where the target protein needs structurally correct disul?de linkages or unlinked reduced Cys residues for activity.

Acknowledgements The research was supported by the National Science Council of Taiwan under Grant NSC 872214-E-011-019.

[1] W.H. Habig, M.J. Pabst, W.B. Jakoby, J. Biol. Chem. 249 (1974) 7130. [2] T.H. Rushmore, C.B. Pickett, J. Biol. Chem. 268 (1993) 11475. [3] D.B. Smith, K.M. Davern, P.G. Board, W.U. Tiu, E.G. Garcia, G.F. Mitchell, Proc. Natl. Acad. Sci. 83 (1986) 8703. [4] A. Sher, S.L. James, R. Correa-Oliveira, S. Hieny, E. Pearce, Parasitology 98 (1989) S61. [5] M.A. McTigue, D.R. Williams, J.A. Tainer, J. Mol. Biol. 246 (1995) 21. [6] R.B. Saint, J.A. Beall, R.J. Grumont, G.F. Mitchell, E.G. Garcia, Mol. Biochem. Parasitol. 18 (1986) 333. [7] D.B. Smith, K.S. Johnson, Gene 67 (1988) 31. [8] A.C. Ward, I.G. Macreadie, Yeast 10 (1994) 441. [9] H. Abeliovich, J. Shlomai, Anal. Biochem. 228 (1995) 351. [10] J.A. Thomas, B. Poland, R. Honzatko, Arch. Biochem. Biophys. 319 (1995) 1. ? [11] H. Sies, A.L. Dafre, Y. Ji, T.P.M. Akerboom, Chemico-Biol. Interactions 111–112 (1998) 177. [12] E. Cabiscol, R.L. Levine, Proc. Natl. Acad. Sci. 93 (1996) 4170. [13] K. Tamai, K. Satoh, S. Tsuchida, I. Hatayama, T. Maki, K. Sato, Biochem. Biophys. Res. Commun. 167 (1990) 331. [14] H.W. Dirr, K. Mann, R. Huber, R. Ladenstein, P. Reinemer, Eur. J. Biochem. 196 (1991) 693. [15] J. Nishihira, T. Ishibashi, M. Sakai, S. Nishi, T. Kumazaki, Y. Hatanaka, S. Tsuda, K. Hikichi, Biochem. Biophys. Res. Commun. 188 (1992) 424. [16] G. Chaga, M. Widersten, L. Andersson, J. Porath, U.H. Danielson, B. Mannervik, Protein Eng. 7 (1994) 1115. [17] C.A. Panagiotidis, S.J. Silverstein, Gene 164 (1995) 45. [18] H.-M. Chen, K.-T. Chen, S.-L. Luo (1999) Manuscript in preparation. [19] T.A. Kunkel, J.D. Roberts, R.A. Zakour, Methods Enzymol. 154 (1987) 367.

4. Conclusion In conclusion, we have used the WT and mutant SjGST / His enzymes to compare the GSH af?nity and the IMAC puri?cation systems. The IMAC system puri?ed the mutant enzymes as ef?ciently as the WT enzyme, while the ef?ciency of the GSHaf?nity system was greatly dependent on the enzyme structure. Mutant SjGST / His enzymes whose GSH binding sites have been affected by the mutations cannot be readily puri?ed with the GSH-af?nity gels. The noncovalent aggregation of the puri?ed enzymes was not as prevalent in the IMAC system as in the GSH af?nity system. In addition, use of GSH as an eluting agent in the GSH af?nity puri?cation system covalently modi?ed the super?cial Cys residues, Cys85, Cys138, and Cys178, of SjGST / His through S-thiolation or thiol–disul?de exchange. Such modi?cations may affect on the enzyme activity or stability because of the existences of a bulky charged GSH group. Therefore, the Ni 21 -NTA puri?cation system is better for the puri?cation of SjGST / His

H.-M. Chen et al. / J. Chromatogr. A 852 (1999) 151 – 159 [20] M.M. Bradford, Anal. Biochem. 72 (1976) 248. [21] D.P. Goldenberg, T.E. Creighton, Anal. Biochem. 138 (1984) 1. [22] H. Shen, K. Tamai, K. Satoh, I. Hatayama, S. Tsuchida, K. Sato, Arch. Biochem. Biophys. 286 (1991) 178. [23] K. Tamai, H. Shen, S. Tsuchida, I. Hatayama, K. Satoh, A. Yasui, A. Oikawa, K. Sato, Biochem. Biophys. Res. Commun. 179 (1991) 790. [24] T. Takao, M. Kobayashi, O. Nishimura, Y. Shimonishi, J. Biol. Chem. 262 (1987) 3541. [25] Y. Taniyama, C. Seko, M. Kikuchi, J. Biol. Chem. 265 (1990) 16767.


[26] R. Brigelius, Oxidative Stress, in: H. Sies (Ed.), Academic Press, New York, 1985, p. 243. [27] C.-K. Lii, Y.-H. Chai, W. Zhao, J.A. Thomas, S. Hendrich, Arch. Biochem. Biophys. 308 (1994) 231. [28] M. Kikuchi, Y. Taniyama, S. Kanaya, T. Takao, Y. Shimonishi, Eur. J. Biochem. 187 (1990) 315. [29] J.A. Thomas, D. Beidler, Anal. Biochem. 157 (1986) 32. [30] H.T. Wright, Protein Eng. 4 (1991) 283. [31] N.P. Bhatt, K. Patel, R.T. Borchardt, Pharm. Res. 7 (1990) 593.