Báo cáo khoa học: Different roles of functional residues in the hydrophobic binding site of two sweet orange tau glutathione S-transferases pdf

In this study, we show that two naturally occurring tau GSTs GSTUs exhibit distinctive kinetic parameters towards 1-chloro-2,4-dini-trobenzene CDNB, although they differ only in three am

binding site of two sweet orange tau glutathione

S-transferases

Angela R Lo Piero, Valeria Mercurio, Ivana Puglisi and Goffredo Petrone

Dipartimento di Scienze Agronomiche, Agrochimiche e delle Produzioni Animali (DACPA), Universita` di Catania, Italy

Introduction

The glutathione S-transferases (GSTs; EC 2.5.1.18) are

members of a multifunctional superfamily of enzymes

catalyzing the conjugation of glutathione (GSH) to the

electrophilic groups of hydrophobic and usually

cytotoxic molecules of either endogenous or exogenous

origin [1–4] GSTs are widely distributed in nature from humans to bacteria [5–7] In plants, the GSH addition reaction is coupled to the vacuolar compart-mentation of the GSH conjugates because of the lack

of an effective excretion pathway, which is active in

Keywords

glutathione S-transferase; H-site;

site-directed mutagenesis; sweet orange;

tau class glutathione S-transferase

Correspondence

A R Lo Piero, Dipartimento di Scienze

Agronomiche, Agrochimiche e delle

Produzioni Animali (DACPA), Universita` di

Catania, Via S Sofia 98, 95123, Catania,

Italy

Fax: +39 95 7141581

Tel: +39 95 7580238

E-mail: rlopiero@unict.it

(Received 27 July 2009, revised 7 October

2009, accepted 5 November 2009)

doi:10.1111/j.1742-4658.2009.07481.x

Glutathione S-transferases (GSTs) catalyze the conjugation of glutathione

to hydrophobic compounds, contributing to the metabolism of toxic chemicals In this study, we show that two naturally occurring tau GSTs (GSTUs) exhibit distinctive kinetic parameters towards 1-chloro-2,4-dini-trobenzene (CDNB), although they differ only in three amino acids (Arg89, Glu117 and Ile172 in GSTU1 are replaced by Pro89, Lys117 and Val172 in GSTU2) In order to understand the effects of the single mis-matched residues, several mutant GSTs were generated through site-direc-ted mutagenesis The analysis of the kinetic parameters of the mutants led to the conclusion that Glu117 provides a critical contribution to the maintenance of a high-affinity CDNB-binding site However, the substitu-tion E117K gives rise to mutants showing increased kcat values for CDNB, suggesting that Lys117 might positively influence the formation

of the transition state during catalysis No changes in the Km values towards glutathione were found between the naturally occurring GSTs and mutants, except for the mutant caused by the substitution R89P in GSTU1, which showed a sharp increase in Km Moreover, the analysis of enzyme reactivation after denaturation showed that this R89P substitution leads to a two-fold enhancement of the refolded enzyme yield, suggesting that the insertion of proline might induce critical structural modifications

In contrast, the substitution P89R in GSTU2 does not modify the reacti-vation yield and does not impair the affinity of the mutant for glutathi-one, suggesting that all three residues investigated in this work are fundamental in the creation of enzymes characterized by unique biochem-ical properties

Abbreviations

4-NPB, 4-nitrophenethyl bromide; CDNB, 1-chloro-2,4-dinitrobenzene; ECA, ethacrynic acid; GmGSTU-4-4, Glycine max tau glutathione S-transferase-4-4; GSH, glutathione; G-site, glutathione-binding site; GST, glutathione S-transferase; GSTU, tau glutathione S-transferase; H-site, hydrophobic binding site; NBD-Cl, 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole.

animals [8,9] GSTs, in addition to transfer of GSH to

toxic compounds, act as GSH-dependent peroxidase,

isomerase and oxidoreductases, playing pivotal roles in

plant cell protection, as reviewed by Moons [10] Plant

GSTs are abundantly expressed, and show major

tran-scriptional regulation It has been reported that GST

transcript levels can markedly increase in response to

a wide variety of stressful conditions, such as

herbi-cides [1,11], chilling [11,12], hypoxic stress [13],

dehy-dration [14], wounding [15], and pathogen attack

[16,17] Most GSTs are active as dimers, either

homo-dimers or heterohomo-dimers of subunits ranging from 23 to

30 kDa in size Subunits of all known GST structures

exhibit a two-domain fold, the N-terminal domain and

C-terminal domain, including the highly conserved

GSH-binding site (G-site) and the more divergent

cosubstrate-binding domain or hydrophobic binding

site (H-site) [7,18] The G-site includes both a-helices

and b-strands as secondary structure elements The

topological arrangement of these elements is usually

bababba, similar to the thioredoxin fold of other

GSH-binding or cysteine-binding proteins The H-site

is entirely helical, with a variable number of a-helices,

depending on the specific enzymes [19]

Mechanisti-cally, the catalysis of the nucleophilic aromatic

substi-tution reactions comprises the substrate binding to the

enzyme’s active site, ionization of GSH to form the

highly reactive thiolate anion, and nucleophilic attack

by the thiolate at the substrate electrophilic center

A hydroxyl group, provided in most plant GSTs by a

serine, is considered to be required for the correct

ori-entation and stabilization of the deprotonated thiolate

anion in the active site of the enzyme [20] On the basis

of protein sequence similarity, active site residue and

gene organization plant GSTs are grouped in four

main classes (phi, tau, zeta, and theta) [18,21,22] The

majority of the plant GSTs belongs to the tau (GSTU)

and phi classes, which are plant-specific Among the

plant GSTs, GSTUs are the most numerous, and

members of this class overlap in their function of

enhancing crop stress tolerance Despite the important

roles of the GSTUs, extensive analysis of critical

resi-dues in both domain sites is needed, although the

highly conserved nature of the G-site [23] and the

recent crystallographic characterization of two GSTUs

[24–26] have allowed researchers to formulate some

general considerations about the N-terminal domains

of the enzymes Moreover, the existence of a conserved

electron-sharing network that helps the glutamyl

c-car-boxylate of GSH to function as a catalytic base,

accepting the proton from the thiol group to form an

anionic GSH, has been reported [27] This network is

characterized by an electrostatic interaction between

negatively and positively charged amino acids stabi-lized by an array of hydrogen bonds, and appears to

be a functionally conserved motif in all GST classes [27] In the Glycine max GSTU-4-4 (GmGSTU4-4)– GSH complex, the strictly conserved residues Arg18, Glu66, Ser67 and Asp103 appear to form the proposed electron-sharing network [26] Studies of the catalytic properties of Pinus tabulaeformis GSTU1 by site-direc-ted mutagenesis, which demonstrate the crucial role of both Ser67 and Glu66 in GSH binding, corroborate these findings [28] In contrast, the H-sites of GSTs exhibit a low degree of sequence identity, and hence unique structures that reflect different functions in vivo [18] In a previous study, we isolated from sweet orange leaves two distinct GSTU genes sharing 98.6% homology at the nucleotide levels and containing a

651 bp ORF The encoded proteins differ in only three amino acids: the triad Arg89, Glu117 and Ile172 found

in the isoform GSTU1 is replaced by the triad Pro89, Lys117 and Val172 in the GSTU2 isoform, all of the mismatches being located in the H-sites of the enzymes [11] As long as the enzymes were expressed in vitro and purified, they exhibited different specific activity with 1-chloro-2,4-dinitrobenzene (CDNB) as substrate, GSTU1 showing a value three-fold lower than that observed for GSTU2 [11] In the present work, site-directed mutagenesis was used to evaluate the effects

of the single mismatched residues on the kinetic prop-erties of isoforms GSTU1 and GSTU2, and also to address questions regarding the functional roles of these H-site residues The results have both academic relevance and practical importance, as they will form the basis for the design of new engineered GSTs show-ing altered substrate specificity and enhanced activity towards xenobiotics

Results and Discussion

Wild-type (GSTU1 and GSTU2) and mutant GSTs were expressed and purified as described in Experimen-tal procedures After purification, all recombinant pro-teins showed a single band in SDS⁄ PAGE, and an identical molecular mass of about 26.0 kDa, corre-sponding to the calculated molecular mass of the recombinant sweet orange GST subunits (Fig 1) CDNB is generally considered to be the classic GST substrate, because most GST isoenzymes display high activity towards it It has been shown that the sweet orange GST isoforms exhibit quite different specific activities with CDNB as substrate, GSTU2 being three-fold more active than GSTU1 [11] Conse-quently, steady-state kinetic characterization of both isoforms with respect to CDNB was performed prior

to any other investigation concerning their catalytic

mechanism The values of kinetic parameters obtained

by non-linear regression analysis are listed in Table 1

The results showed that GSTU1 had a lower apparent

Km for CDNB (0.75 mm) than that exhibited by

GSTU2 (1 mm), this finding being consistent with a

higher affinity of GSTU1 for this substrate The Km

values of GSTU1 and GSTU2 for CDNB are in the

range of those of plant GSTs, whose values vary from

0.12 to 4.43 mm [24,29,30] As expected on the basis of

the complete identity of the G-site sequences, the same

Km for GSH was registered for both isoforms

(0.5 mm), which is in general agreement with other

published Km values for GSH [31] However, in the case of both substrates, GSTU2 showed higher cata-lytic efficiency (kcat⁄ Km) as well as higher kcat values (Table 1) The alignment of the H-site sequences of 20 GSTUs from different plant sources showed that Pro89 and the Lys117 found in GSTU2 are strictly conserved residues, whereas, at the 172 position, the isoleucine is well conserved but valine or threonine can also be found (Fig S1) Therefore, the naturally occur-ring GSTU1 (Arg89, Glu117, and Ile172) probably has unique features, and might perform a different func-tional role in vivo than GSTU2 This assumption is also supported by the analysis of gene expression, which showed that the GSTU1 gene is induced by cad-mium sulfate, CDNB, cyhalothrin, and cold stress, whereas GSTU2 is constitutively expressed [11] Given that these naturally occurring GSTUs provided a good opportunity to understand how specific amino acids might contribute to the enzyme’s biochemical behav-ior, several mutant GSTs, hereafter referred to by their distinctive amino acid triad, were generated by site-directed mutagenesis In particular, the RKI and RKV mutants were obtained by individually replacing the Glu117 of GSTU1 (REI) with Lys117, and the Pro89

of GSTU2 (PKV) with Arg89, respectively Moreover, the PEI and PKI mutants were obtained by replacing Arg89 of GSTU1 (REI) with Pro89, and Glu117 of the PEI mutant with Lys117, respectively Steady-state kinetic analysis of purified RKI and RKV mutants showed sharp increases of approximately six-fold and five-fold in the Km value for CDNB as compared with GSTU1 (REI) (Table 1) Consequently, the results sug-gest that Lys117-containing enzymes, either wild type

or mutant, do not easily accommodate CDNB in their active site, and also highlight the critical contribution

of Glu117 to CDNB recognition Accordingly, the PEI mutant, in which the Glu117 is restored as compared with the PKI mutant, shows a Km value for CDNB similar to that of GSTU1 (REI) (Table 1) However, the kcatvalues for both GSTU2 (PKV) and the E117K

36 kDa

29 kDa

Fig 1 SDS ⁄ PAGE of the purified wild-type and mutant GSTs Lane

1: unpurified E coli extract Lane 2: purified wild-type GSTU1 (REI).

Lane 3: purified wild-type GSTU2 (PKV) Lane 4: purified PEI

mutant Lane 5: RKV mutant Lane 6: purified PKI mutant Lane 7:

molecular mass marker The mutant enzymes are referred to by

their distinctive triad of amino acids located, respectively, at

posi-tions 89, 117, and 172.

Table 1 Steady-state kinetic constants of wild-type and mutant GSTs The kinetic parameters were calculated by using nonlinear regression analysis with the HYPER 32 program Each value represents the mean ± standard deviation of three replicates.

GST

Km

(m M )

Vmax (l M Æmin)1)

kcat (s)1· 10 3 )

kcat⁄ K m

( M )1Æs)1)

Km (m M )

Vmax (l M Æmin)1)

kcat (s)1· 10 3 )

kcat⁄ K m

( M )1Æs)1)

GSTU1 (REI) 0.75 ± 0.03 2.0 ± 0.10 23.8 ± 1 31.7 ± 1.45 0.5 ± 0.02 1.0 ± 0.1 13.8 ± 0.24 27.7 ± 1.15 GSTU2 (PKV) 1.0 ± 0.07 4.0 ± 0.06 108.1 ± 1 108.1 ± 5.4 0.5 ± 0.03 4.0 ± 0.15 76.9 ± 0.4 153.8 ± 6.94 RKI 5.0 ± 0.2 9.0 ± 0.5 257.1 ± 2 51.4 ± 2.0 0.6 ± 0.03 5.0 ± 0.20 147.0 ± 0.9 245.0 ± 4.26 RKV 4.0 ± 0.4 8.0 ± 0.25 119.4 ± 1 29.8 ± 0.6 0.35 ± 0.02 3.3 ± 0.10 129.4 ± 0.5 369.7 ± 21.12 PKI 4.0 ± 0.4 4.5 ± 0.15 160.7 ± 1.6 40.0 ± 1.5 0.7 ± 0.04 4.0 ± 0.15 64.0 ± 0.63 91.4 ± 2.16 PEI 0.85 ± 0.05 3.0 ± 0.07 28.8 ± 0.07 33.9 ± 2.4 4.0 ± 0.02 7.0 ± 0.3 70.7 ± 0.7 17.6 ± 0.24

mutants (RKI, RKV, and PKI) were increased as

com-pared with those of the Glu117-containing enzymes

(REI and PEI) (Table 1) This finding suggests a

nega-tive influence of Glu117 on the catalytic event It has

been shown that the addition of GSH to CDNB

occurs via an addition–elimination sequence involving

a short-lived r-complex intermediate (Meisenheimer

complex) as transition state [32] The structure of the

transition state [32] indicates that the enzyme might

provide electrophilic assistance interactions to the

developing charge of the r-complex o-nitro group [33]

The comparison of the crystal structure of the rat

M1-1 GST in complex with the transition state analog

and with the product reveals two completely different

binding modes for the intermediate and the product,

suggesting that a specific motion is associated with

the collapse of the intermediate into the product

[34] More recently, Axarli et al [26] have also shown

that the catalytic reaction of GmGSTU4-4 is barely

sensitive to the nature of the leaving group, as the

sub-stitution of the chlorine atom with the more

electro-negative fluorine in the CDNB molecule did not affect

the kcat values Altogether, these results are consistent

with the idea that the rate-limiting step of the CDNB–

GSH catalytic reaction is the physical event of product

release, probably involving structural motions or

con-formational changes of the ternary complex In this

context, we propose that the substitution in the orange

GSTU1 of a nucleophilic residue by an electrophilic

one (E117K) could function in strongly favoring the

formation of the transition state during GSH addition

to CDNB by providing the required electrophilic

assis-tance to the developing transition state

As regards the analysis of kinetic parameters

regard-ing the natural in vivo substrate GSH, the mutants

containing Lys117 showed apparent Km values similar

to those reported for the wild-type GSTs (Table 1)

However, the RK and PK enzymes, both wild type

and mutants, showed a strong increase in the kcat⁄ Km

values with respect to the Glu117-containing enzymes

(REI and PEI) (Table 1), indicating that the

substitu-tion E117K is crucial to the formasubstitu-tion of enzymes with

higher catalytic efficiency Interestingly, the PEI

mutant showed an eight-fold increase in the Km value

for GSH as compared with GSTU1 (REI) (substitution

R89P) (Table 1), suggesting that such mutation of the

H-site might exert its effects upon the G-site kinetic

properties Recently, Axarli et al [25] reported the

crystal structure of the GSTU4-4 from soybean, which

shares 71% sequence similarity with the orange

GSTUs [11] The consistency in the fold of GSTs and

the availability of the structure of GmGSTU4-4

prompted us to construct molecular homology models

for Citrus sinensis GSTU1 (REI) and the PEI mutant,

as well as for Citrus sinensis GSTU2 (PKV) and the RKV mutant, by submitting the amino acid sequences

to swiss-model [35] Then, the alignments of the 3D structural homology models were performed between the wild-type GSTs and their respective mutants at the 89 position (REI–PEI, and PKV–RKV) by the web-based program matras [36] The analysis of the superimposed REI–PEI models revealed relevant con-formational differences between the enzyme structures, mainly involving a-helix1 (from Ser13 to Gly27) and a-helix3 (from Ser67 to Asp80), which represent the catalytic ‘core’ of the G-site [26] (numbering of helices

is consistent with that of other GSTs) (Fig 2A) In contrast, minimal structural perturbations of the afore-mentioned a-helix1 and a-helix3 were observed in the case of the superimposed PKV–RKV models, whose major nonoverlapping regions are, instead, localized in the H-site of the enzymes (Fig 2B) These findings are

in agreement with the different values of apparent Km

for GSH observed between GSTU1 (REI) and the PEI mutant, and overall indicate that the substitution R89P in GSTU1 might modify the architecture of the G-site, thus negatively influencing the enzyme’s affinity for GSH Furthermore, Table 2 shows the recovered activity following denaturation and refolding of wild-type GSTUs and the PEI and RKV mutants The reac-tivation yield of the PEI mutant was two-fold higher than that observed in GSTU1 (REI) (Table 2), thus supporting our hypothesis that Pro89 might have a structural role in the PEI mutant However, the recov-ered activities of both GSTU2 (PKV) and the RKV mutant were similar, suggesting that the substitution P89R in GSTU2 (PKV) does not affect the refolding process Consequently, the putative structural role assigned to the Pro89 of the PEI mutant cannot be attributed to the Pro89 of GSTU2 (PKV), as the RKV mutant, arising from the P89R substitution in GSTU2 (PKV), has both similar Km values (Table 1) and similar reactivation yields after denaturation (Table 2) as GSTU2 (PKV) Therefore, the results lead to the conclusion that all three amino acids inves-tigated in this work take part in the creation of enzymes showing unique structures and, consequently, functions

The substrate specificity of the sweet orange GSTs was also investigated in order to identify catalytic activities that may be related to their biological func-tion To this end, some substrates in addition to CDNB were examined, including 7-chloro-4-nitro-benzo-2-oxa-1,3-diazole (NBD-Cl), with which mam-malian a-class GSTs show high activity [37], the alkyl halide 4-nitrophenethyl bromide (4-NPB), related to

the role of GSTs in detoxification processes, and

ethac-rynic acid (ECA), a phenylacetic derivative that

con-tains an electrophilic group, similar to the cytotoxic

a,b-alkenals, which may be formed during oxidative stress [29] (Table 3) Overall, CDNB is the preferred substrate of both naturally occurring (REI and PKV) and mutant GSTs, with the RKV and PKV enzymes showing the highest specific activity Relatively lower activity was detected with the mammalian GST sub-strate NBD-Cl (Table 3) Interestingly, the mutant enzyme RKV is able to conjugate 4-NPB to GSH, whereas wild types, as well as other mutant GSTs, are not, thus suggesting the crucial role of Val172, but not

of Ile172, in creating the active site architecture that is suitable to accommodate the aforesaid substrate This finding is of particular interest, as alkyl halides have toxicological interest in view of their occurrence as environmental pollutants [38] Therefore, the RKV mutant, owing to the distinguishing ability to conju-gate 4-NPB to GSH (Table 3) and to the extremely high catalytic efficiency towards GSH (Table 1), exhib-its great potential for the development of enzymes with novel properties, e.g the ability to reclaim strongly contaminated environments through nucleophilic sub-stitution reactions involving either aryl halide or alkyl halide xenobiotics None of the enzymes is active with ECA, suggesting that wild-type and mutant GSTs might not be directly involved in the removal of harm-ful oxidative stress byproducts (Table 3) The data showing that recombinant orange GSTs do not exhibit

in vitro glutathione peroxidase activity (not shown) support this last hypothesis

Experimental procedures

Molecular cloning of sweet orange GSTU1 and GSTU2

Cloning of sweet orange GSTU1 and GSTU2 genes and their transfer into the expression vector pEXP1–DEST (Invitrogen, Carlsbad, CA, USA) was as described by Lo Table 2 Recovered activity after denaturation and refolding of

wild-type and mutant GSTs The recovered GST activities were

measured in standard conditions after dilution of the denaturating

agent to an ineffective concentration Each value represents the

mean ± standard deviation of three replicates.

GST Recovered activity after denaturation (%)

A

B

Fig 2 Representation of the 3D homology models of the wild-type

GSTs and their respective mutants at position 89 (A) Superposition

of wild-type GSTU1 (REI) and the PEI mutant (R89P) (B)

Superposi-tion of the wild-type GSTU2 (PKV) and the RKV mutant (P89R).

Wild-type GSTs are shown in yellow, and mutants are shown in

red The nonoverlapping regions between the superimposed 3D

models appear as red areas.

Table 3 Specific activity of the wild-type and mutant GSTs towards different substrates The GST assay was performed in standard conditions in the presence of 1 m M of different sub-strates Each value represents the mean ± standard deviation of three replicates ND, not detectable.

Activity [nmol (minÆmg)1)]

Piero et al [11] The nucleotide sequences of wild-type

GSTs were submitted to GenBank under the following

accession numbers: EF597102 (GSTU1-coding sequence)

and FJ184997 (GSTU2-coding sequence)

Site-directed mutagenesis

Site-directed mutagenesis of GSTU1 and GSTU2 was

per-formed by PCR using sweet orange pEXP–GSTU1 and

pEXP–GSTU2 as templates (Gene Tailor Site-directed

mutagenesis system; Invitrogen) The PCR reaction

mix-tures contained 10 lm each primer, 1 U of Accuprime Pfx

DNA polymerase (Invitrogen), 0.3 mm each dNTP, 1 mm

MgSO4, and 18 lg of methylated plasmid DNA, in final

volume of 50 lL PCR conditions were optimized to the

following: 94C for 2 min (one cycle), 94 C for 30 s,

55C for 30 s, and 68 C for 5 min (20 cycles), and 68 C

for 10 min (one cycle) All the mutagenesis primers are

shown in Table 4 The PCR products were analyzed on a

1% agarose gel containing 0.5 lgÆmL)1 ethidium bromide

and purified with the Qiaquick gel extraction kit (Qiagen,

Hilden, Germany) All the mutants were confirmed by

sequencing the plasmid DNA with the T7 promoter and T7

reverse primers

In vitro expression and purification of sweet

orange wild-type and mutant GSTs

In vitro expression of functionally active GSTs, both wild

types and mutants, was performed according to the method

described in Lo Piero et al [11] Briefly, protein expression

was achieved in a cell-free system (Expressway Plus;

Invitro-gen) by incubating the plasmids (15 lg), in IVPS Plus

Escherichia coliextract for 6 h at 25C, to promote proper

protein folding The recombinant proteins were purified by

loading the cell-free extract onto a His-graviTrap column

prepacked with Ni2+–Sepharose 6 fast flow (GE Healthcare,

Milwaukee, WI, USA) The unspecific bound proteins were

removed by washing the column with 20 mm phosphate

buffer (pH 8.0), 500 mm NaCl, and 20 mm imidazole The

His-tagged protein was eluted with 500 mm imidazole by

recovering 0.2 mL fractions Fractions were tested for

pro-tein content using the Bradford method [39], and those

included in the peak core were collected and assayed for GST activity (see below) A negative control of the recombi-nant protein expression was also performed by incubating the E coli extract with empty plasmid SDS⁄ PAGE was car-ried out according to the method described in Laemmli [40]

GST enzyme assay

The GST assay was routinely performed as described in Lo Piero et al [3] In the substrate specificity experiment, the reaction mixture (final volume 0.5 mL), containing 1· NaCl⁄ Pi (pH 7.4), 1 mm glutathione, 1 mm different sub-strates, and purified recombinant enzymes (10–20 lg), was incubated at 30C for 15 min GSH conjugates were detected by measuring the absorbance of samples at

340 nm Molar extinction coefficients of 9600 m)1Æcm)1 (CDNB), 13 000 m)1Æcm)1 (NBD-Cl) and 1200 m)1Æcm)1 (4-NPB) were used All measurements were adjusted by subtracting the absorbance values obtained for the nonen-zymatic conjugation of substrates The apparent Km and

Vmaxvalues for GSH of both wild-type and mutant GSTs were determined in the presence of GSH in the concentra-tion range 0.1–1.0 mm and a fixed CDNB concentraconcentra-tion of

1 mm Alternatively, for the determination of CDNB apparent Km and Vmax values, GSH was used at a fixed concentration of 1 mm and the CDNB concentration was varied in the range 0.1–1 mm The kinetic parameters were derived using nonlinear regression analysis with the hyper32 program, available at http://homepage.ntlworld com/john.easterby/hyper32.html The experiments were repeated three times on independent enzyme preparations Kinetic comparison of the His-tagged and untagged enzymes showed that the extra six histidines on the N-ter-minus did not interfere with the activity or function of the enzymes (data not shown)

Refolding studies

Refolding experiments were performed according to a slight modification of the method described by Zeng et al [28] All enzymes (70 lg) were incubated in a denaturation buf-fer (4 m guanidinium chloride, 0.1 m phosphate, and 1 mm EDTA, pH 6.5) at 25C for 30 min At the end of the Table 4 Primers used in site-directed mutagenesis.

E117K-rev: TGTGGTCCATGTCTTCGTCGAAGCATC

RKI

P89R-rev: ATCAGAGGGAAGCAATGGAGCCTTGTC

RKV

R89P-rev: ATCAGAGGGAAGCAATGGAGCCTTGTC

PEI

E117K-rev: GCTGGACTTTCCAAATGTCTCATA

PKI

incubation period, they were rapidly diluted up to an

inef-fective guanidinium chloride concentration (1 : 30) in a

renaturation buffer (0.1 m phosphate, pH 6.5, 1 mm

EDTA), and the recovered activity towards the substrate

CDNB was immediately monitored

Sequence alignment and structural modeling

The comparison of the C-terminal GST protein sequences

was performed by using the multiple sequence alignment

program clustalw 1.8 To generate structural models,

amino acid sequences of GSTs, both wild type and

mutants, were submitted to swiss-model

(http://swissmod-el.expasy.org/) [35] Homology models were generated using

the known X-ray structure of GmGSTU4-4 (Protein Data

Bank code: 2vo4A) as template The 3D homology models

were compared with matras [36], and jmol version 2.7 was

used to generate 3D images

Acknowledgements

Financial support was provided by a grant to Dr

Angela Roberta Lo Piero by the University of Catania,

Fondi del Bilancio Universitario, Progetti di Ricerca di

Ateneo (PRA), 2006, Project title: ‘Isolamento e

carat-terizzazione di un gene codificante per la glutatione

transferasi coinvolta nei meccanismi di trasferimento

nel vacuolo dei pigmenti antociani nella polpa di

ara-nce rosse e bionde.’

References

1 Edwards R & Dixon DP (2000) The role of glutathione

transferases in herbicide metabolism In Herbicides and

Their Mechanisms of Action(Cobb AH & Kirkwood

RC eds), pp 38–71 Sheffield, Sheffield Academic Press

2 Axarli I, Rigden DJ & Labrou NE (2004)

Characteriza-tion of the ligandin site of maize glutathione transferase

I Biochem J 382, 885–893

3 Lo Piero AR, Puglisi I & Petrone G (2006) Gene isolation,

analysis of expression and in vitro synthesis of a

glutathi-one S-transferase from orange fruit [Citrus sinensis L

(Osbeck)] J Agric Food Chem 54, 9227–9233

4 Dixon DP, Lapthorn A, Madesis P, Mudd EA, Day A

& Edwards R (2008) Binding and glutathione

conjuga-tion of porphyrinogens by plant glutathione

transfer-ases J Biol Chem 283, 20268–20276

5 Hayes JD, Flanagan JU & Jowsey IR (2005) Glutathione

transferases Annu Rev Pharmacol Toxicol 45, 51–88

6 Allocati N, Federici L, Masulli M & Di Ilio C (2009)

Glutathione transferases in bacteria FEBS J, 276,

58–75

7 Sheehan D, Meade G, Foley VM & Dowd CA (2001)

Structure, function and evolution of glutathione

transferases: implications for classification of non-mam-malian members of an ancient enzyme superfamily Biochem J 360, 1–16

8 Sandermann H Jr (1992) Plant metabolism of xenobiot-ics Trends Biochem Sci 17, 82–84

9 Rea PA (1999) MRP sub family ABC transporters from plants and yeast J Exp Bot 50, 895–913

10 Moons A (2005) Regulatory and functional interactions

of plant growth regulators and plant glutathione trans-ferases (GSTs) Vitam Horm 72, 155–202

11 Lo Piero AR, Mercurio V, Puglisi I & Petrone G (2009) Gene isolation and expression analysis of two distinct sweet orange [Citrus sinensis L (Osbeck)] tau-type glutathione transferases Gene 443, 143–150

12 Lo Piero AR, Puglisi I, Rapisarda P & Petrone G (2005) Anthocyanin accumulation and related gene expression in red orange fruit induced by low tempera-ture storage J Agric Food Chem 53, 9083–9088

13 Moons A (2003) Osgstu3 and osgtu4, encoding tau class glutathione S-transferases, are heavy metal and hypoxic stress-induced and differentially salt stress-responsive in rice roots FEBS Lett 553, 427–432

14 Bianchi MW, Roux C & Vartanian N (2002) Drought regulation of GST8, encoding the Arabidopsis homo-logue of ParC⁄ Nt107 glutathione transferase ⁄ peroxi-dase Physiol Plant 116, 96–105

15 Vollenweider S, Weber H, Stolz S, Chetelat A & Farmer EE (2000) Fatty acid ketodienes and fatty acid ketotrienes: Michael addition acceptors that accumulate

in wounded and diseased Arabidopsis leaves Plant J 24, 467–476

16 Mauch F & Dudler R (1993) Differential induction of distinct glutathione transferases of wheat by xenobiotics and by pathogen attack Plant Physiol 102, 1193–1201

17 Dean JD, Goodwin PH & Hsiang T (2005) Induction

of glutathione S-transferase genes of Nicotiana benth-amianafollowing infection by Colletotrichum destructi-vumand C orbiculare and involvement of one in resistance J Exp Bot 56, 1525–1533

18 Edwards R, Dixon DP & Walbot V (2000) Plant gluta-thione transferases: enzymes with multiple functions in sickness and in health Trends Plant Sci 5, 193–198

19 Frova C (2006) Glutathione transferases in the genom-ics era: new insight and perspective Biomol Eng 23, 149–169

20 Labrou NE, Mello LV & Clonis YD (2001) Functional and structural role of the glutathione binding residues

in maize (Zea mays) glutathione S-transferase I Biochem J 358, 101–110

21 Dixon DP, Lapthorn A & Edwards R (2002) Plant glutathione transferases Genome Biol 3, 3004.1–3004.10

22 Frova C (2003) The plant glutathione gene family: genomic structure, functions, expression and evolution Physiol Plant 119, 469–479

23 Droog F (1997) Plant glutathione S-transferases, a tale

of theta and tau J Plant Growth Regul 16, 95–107

24 Thom R, Cummins I, Dixon DP, Edwards R, Cole DJ

& Lapthorn AJ (2002) Structure of a tau class

gluta-thione S-transferase from wheat active in herbicide

detoxification Biochemistry 41, 7008–7020

25 Axarli I, Dhavala P, Papageorgiou AC & Labrou NE

(2009) Crystallographic and functional characterization

of the fluorodifen-inducible glutathione transferase from

Glycine maxreveals an active site topography suited for

diphenylether herbicides and a novel L-site J Mol Biol

385, 984–1002

26 Axarli I, Dhavala P, Papageorgiou AC & Labrou NE

(2009) Crystal structure of Glycine max glutathione

transferase in complex with glutathione: investigation of

the mechanism operating by the tau class glutathione

transferases Biochem J 422, 247–256

27 Winayanuwattikun P & Ketterman AJ (2005) An

elec-tron-sharing network involved in the catalytic

mecha-nism is functionally conserved in different glutathione

transferase classes J Biol Chem 280, 31776–31782

28 Zeng QY & Wang XR (2005) Catalytic properties of

glutathione-binding residues in a s class glutathione

transferase (PtGSTU1) from Pinus tabulaeformis FEBS

Lett 579, 2657–2662

29 Kilili KG, Atanassova N, Vardanyan N, Clatot N,

Al-Sabarna K, Kanellopoulos PN, Makris AM &

Kampranis SC (2004) Differential roles of tau class

glutathione S-transferase in oxidative stress J Biol

Chem 279, 24540–24551

30 Dixon DP, Cole DJ & Edwards R (1999) Dimerisation

of maize glutathione transferases in recombinant

bacte-ria Plant Mol Biol 40, 997–1008

31 Clark AG (1989) The comparative enzymology of the

glutathione S-transferases from non-vertebrate

organ-ism Comp Biochem Physiol B 92, 419–446

32 Armstrong RN (1997) Structure, catalytic mechanism,

and evolution of glutathione transferase Chem Res

Toxicol 10, 2–18

33 Johnson WW, Liu S, Ji X, Gilliland GL & Armstrong

RN (1993) Tyrosine 115 participates both in chemical

and physical steps of the catalytic mechanism of a

glu-tathione transferase J Biol Chem 268, 11508–11511

34 Ji X, Armstrong RN & Gilliland GL (1993) Snapshots along the reaction coordinate of an SNAr reaction cata-lyzed by glutathione transferase Biochemistry 32, 12949–12954

35 Schwede T, Kopp J, Guex N & Peitsch MC (2003) SWISS-MODEL: an automated protein homology-mod-eling server Nucleic Acids Res 31, 3381–3385

36 Kawabata T (2003) MATRAS: a program for protein 3D structure comparison Nucleic Acids Res 31, 3367–3369

37 Ricci C, Caccuri AM, Lo Bello M, Pastore A, Piemonte

F & Federici G (1994) Colorimetric and fluorometric assays of glutathione transferase based on 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole Anal Biochem 218, 463–465

38 Wheeler JB, Stourman NV, Thier R, Dommermuth A, Vuilleumier S, Rose JA, Armstrong R & Guengerich

FP (2001) Conjugation of haloalkanes by bacterial and mammalian glutathione transferases: mono- and dihalo-methanes Chem Res Toxicol 14, 1118–1127

39 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding Anal Biochem 72, 248–254

40 Laemmli UK (1970) Cleavage of structural proteins during the assembly of head of bacteriophage T4 Nature 227, 680–685

Supporting information

The following supplementary material is available: Fig S1 Alignment of the sweet orange GST H-site sequences with those of 18 representative members of tau class GSTs

This supplementary material can be found in the online version of this article

Please note: As a service to our authors and readers, this journal provides supporting information supplied

by the authors Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset Technical support issues arising from supporting information (other than missing files) should be addressed to the authors