Substrate Specificities and Kinetic Parameters of Recombinant Drosophila melanogaster Glutathione S Transferases E6 and E7 Substrate Specificities and Kinetic Parameters of Recombinant Drosophila melanogaster Glutathione S -Transferases E6 and E7

D. melanogaster glutathione transferases E6 and E7 (DmGSTE6 and DmGSTE7) were suc- cessfully cloned, purified, and biochemically characterized. The recombinant proteins were readily purified using the combination of both anionic and BSP/GSH-agarose affinity chro - matography. Although both GSTs have significant identity in their amino acid sequence, each enzyme displayed unique biochemical characteristics. Both recombinant proteins were only active toward 1-chloro-2,4-dinitrobenzene (CDNB), 1,2-dichloro-4-nitrobenzene (DCNB), and p -nitrobenzyl chloride ( p- NBC) with significant difference in catalytic activi - ties. The findings have shown that neither GSTE6 nor GSTE7 was able to counter oxida tive stress. Comparatively, GSTE7 was a more efficient enzyme at turning over DCNB and p -NBC, based on its kcat/Km values which were of 0.183 and 2.25 min −1 mM −1 , respectively. Thin-layer chromatography analysis showed that both isoforms were not able to conjugate several tested insecticides. The inhibition kinetics of natural products and dyes toward GSTs in vitro revealed that phenol red possessed inhibition effects only on GSTE6 while rose ben - gal and cardiogreen inhibit significantly on both GSTE6 and GSTE7. In contrast, methylene blue dye and trans -chalcone have been shown to stimulate GSTE7 activity toward CDNB.


Introduction
One of the most popular classes of detoxification enzymes that constitute in almost all living organisms is the glutathione transferases (GSTs). GSTs conjugate the thiol groups of reduced glutathione (GSH) toward the negative charge center of lipid soluble compounds ( xenobiotics) to make it water soluble and easy to excrete out. These enzymes have wide distribution in nature and are found ubiquitously in almost all living things including plants, animals, and even bacteria. GSTs are expressed in sex, age, tissue, organ, species, and tumor-specific patterns of expression, and their composition differs significantly [1]. Insect GSTs can be categorized into six classes including Delta, Epsilon, Theta, Omega, Sigma, and Zeta, but it is the Delta and Epsilon classes that are most commonly associated with resistance [2,3]. Several studies also reported that Epsilon classes in Dipteran organisms are to confer insecticide resistance, and their catalytic diversity would likely promote their role in detoxification [4][5][6][7]. Another study [8] also suggested that the expression of the epsilon class GSTs, SlGSTE2, and SlGSTE3 genes in Spodoptera litura, a Lepidoptera detoxifies carbaryl, DDT, RH5992, malathion, and deltamethrin, which is a synthetic chemical insecticide.
A work in housefly isozymes suggested that MdGST6A and MdGST6B which belong to the epsilon class function as key enzymes in the detoxification of insecticides such as methyl parathion and lindane [9]. It was demonstrated that the expression of GSTE6 and GSTE7 in D. melanogaster significantly increased by more than 50% upon exposure to paraquat (1,1-dimethyl-4,4`-bipyridylium) and phenobarbital (PhB) [10]. Besides that, acute insecticide exposure of methyl parathion results in significant increase in expression of GSTE6 (100%) and GSTE7 (72%) [11]. This suggested their immediate involvement in insecticide metabolism.
Therefore, the current study wanted to investigate the behavior of two epsilon class D. melanogaster GSTs, namely, GSTE6 and GSTE7, against varieties of xenobiotics and therefore characterizes their biochemical contribution. This may give insight the potential role the GSTs could have played during xenobiotic metabolism. This could help anticipate the behavior of insecticides related GSTs in other dipterans of economic importance.

Chemicals and source of insect
Unless otherwise stipulated, chemicals employed were of the highest grade obtainable. Buffer components, pesticides, dyes, and GST substrates used were purchased from Sigma-Aldrich. D. melanogaster obtained from the Genetic Department of the University of Malaya was reared on oats and glucose-based diet at room temperature. Only 5 days post emerged, flies were used for the experiments.

DNA and protein analysis
Nucleotide and deduced amino acid sequences of DmGSTE6 and DmGSTE7 obtained from sequencing were compared to existing sequences in Gene Bank by BLAST searching (http:// www.ncbi.nlm.nih.gov). The software used for sequence analysis, matrix table of percentage amino acid identities sequence and alignment with CLUSTAL W, was BioEdit software version 7.2.0.

Cloning, expression, and purification of recombinant proteins
Total DNA was isolated from adult tissues using DNeasy Blood & Tissue Kit (Qiagen) according to the manufacturer's instruction. The primers for amplifying GSTE6 were forward primer containing Nde1 site: 5'-GGAATTC CATATG gtgaaattgactttatac-3′ and reverse primer containing EcoR1 site: 5'-CG GAATTC tcatgcttcgaatgtgaa-3′; the primers for amplifying GSTE7 were forward primer containing Nde1 site: 5'-GGAATTC CATATGcccaaattgatactgtac-3′ and reverse primer containing Xho1 site: 5'-CCG CTCGAGttaattcgatgcgaaagt-3′. Genomic DNA was denatured at 95°C for 3 min, followed by 32-cycle amplification (95°C for 30 seconds, 60°C for 30 seconds, 72°C for 1 min) with final extension at 72°C for 7 min. PCR products were analyzed using 1.0% agarose gel and stained with ethidium bromide. These two PCR products were cloned into pET 30a(+) expression vector (Novagen). The recombinant plasmids were analyzed by sequencing and transformed into E. coli BL21 (DE3) pLysS (Invitrogen) independently and grown at 37°C in 400 mL Luria-Bertani (LB) media containing 30 μg/mL kanamycin. Protein expression was induced by the addition of isopropyl β-D-thiogalactoside (IPTG) at final concentration of 1 mM. Incubation was continued for further 4 h at 37°C, after which the cells were harvested by centrifugation at 4°C (5000 rpm for 15 min) and resuspended in 5 mL binding buffer (25 mM sodium phosphate, pH 7.4). 100 μL of lysozyme (10 mg/mL) (Sigma-Aldrich) was added and incubated at room temperature for 1 h. The cell suspension was homogenized; the cell debris was palleted by centrifugation at 100,000× g for 1 h (4°C). The clear obtained supernatant was used as source for recombinant purification.
Chromatography was carried out using an ÄKTA Purifier FPLC equipped with UNICORN software Version 5.1 and a fraction collector (Frac900) for greater automation of the purification process. The recombinant proteins were purified using Hi-Trap™ Q HP column (5 ml, GE Healthcare) followed by BSP-GSH-Sepharose column (packed in TriconTM, 1 ml) (BSP-GSH-Sepharose matrix was a gift from Dr. AG Clark, Victoria University of Wellington). The BSP (2 mM) used to elute the proteins from affinity matrix was removed from eluate using Hi-Trap™ Desalting column (GE Healthcare).
The Hi-Trap™ Q HP column was equilibrated with 25 mM phosphate buffer and pH 7.4, and 5 mL lysate was applied to the column. Both GSTE6 and GSTE7 enzymes did not bind to the Hi-Trap™ Q HP column, so the flow through was collected and loaded to a BSP-GSH-Sepharose column which was pre-equilibrated with 25 mM phosphate buffer and pH 7.4. The column was washed with 1 M NaCl, and the proteins were eluted with 2 mM BSP in 25 mM phosphate buffer and pH 7.4. Purified enzymes were desalted using Hi-Trap™ Desalting column (15 ml) (GE Healthcare). Proteins were concentrated using Vivaspin 20 (10,000 MWCO, Sartorius). The pooled purified enzymes from subsequent purification were freeze-dried and stored in −20°C for further analysis.

Protein quantification (Bradford assay)
Protein concentration was determined using Coomassie Brilliant Blue G-250, and Bovine serum albumin was used as standard [12].

Molecular weight estimation
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out in a 12% polyacrylamide gel [13] using BioRad Mini-PROTEAN system. Mark 12™ unstained standard (Invitrogen, USA) was used as the protein marker, and the gels were stained with Colloidal Coomassie Blue G-250 [14]. Stained gels were scanned with Image Scanner III (GE Healthcare) and visualized and analyzed with Image Master Software.
For the determination of kinetic parameters, initial velocity data was obtained by varying the concentration of hydrophobic substrates (CDNB, NBC, and DCNB) at concentration range of 0.01 to 2.5 mM at fixed GSH concentration (1 mM is the total enzyme concentration. The effect of inhibitors on the catalytic activity was studied by analyzing the reaction rate (nmol/min) in the presence (concentration varied) and the absence of inhibitors. Experiments were repeated at least three times. The IC50 and EC50 were determined by fitting sigmoid concentrationresponse curves using GraphPad Prism 6.00 software.

Results
In the study both DmGSTE6 and DmGSTE7 were amplified using PCR where genomic DNA was used as the template. Their intron-free coding sequences [20] were used directly for expression of the proteins in E. coli under induction of IPTG. Proper orientation of the GSTE6 and GSTE7 expression constructs was confirmed by PCR analysis using the constructs as a template, DNA sequencing, and CDNB conjugation activity assay of the extracted bacteria lysate. The proteins were highly expressed and isolated using combination of an anion exchange and BSP/GSH-agarose matrix which has been shown to capture a number of Epsilon class GSTs from D. melanogaster [10,11]. The subunit size of GSTE6 and GSTE7 is predicted to
The work later proceeded into looking at the activities of the recombinants toward different common GST substrates as seen in Table 1. The comparison indicated that both recombinants reacted toward the same types of substrates but of different degree of specific activities. Our kinetic analysis as shown in Table 2 indicates the variation of kinetic parameters of both isoforms on DCNB and NBC. Comparatively GSTE6 has shown to have higher affinity toward DCNB than GSTE7.
To further investigate both isoforms' functional differences, several natural products and dyes were acted upon the enzymes as seen in Table 3. Each dye appeared to have different inhibitory effects on each GST isoform. Comparatively, cardiogreen and rose bengal had inhibited the activities of GSTE6 and GSTE7, respectively, at IC50 less than 10 nM.  The data are mean value of at least three independent experiments. *NE denotes no effect. Table 3. Effect of selected dyes and natural products on the activities of GSTE6 and GSTE7.

Discussions
The molecular weights of purified enzymes were approximately 25 kDA, respectively, which corresponded to the calculated molecular masses (Figure 1) and were in agreement with data previously reported [6].
It was observable ( Table 1) that both recombinants have the same affinity toward certain substrates notably of compound containing aromatic ring. However GSTE7 seemed to have higher specific activities toward each conjugated substrate as compared to GSTE6. GSTE7 was reacted 9-, 4-, and 68-fold higher toward CDNB, DCNB, and NBC, respectively, as compared to GSTE6. Both were unable to conjugate lipid peroxidation products, and neither could it have shown peroxidase activities. This implied that neither GSTE6 nor GSTE7 could have directly participated in countering oxidative stress in fruit flies. In another instance, substrate specificity variation was observed when other substrates are used. It was previously reported that GSTE6 reacted toward 4-hydroxynonenal, adrenochrome, phenethyl isothiocyanate, 5-hydroperoxyeicosatetraenoic acid, and hydroxyethyldisulfide, while GSTE7 reacted only to 4-hydroxynonenal, 5-hydroperoxyeicosatetraenoic acid, and 2-hydroxyethyldisulfide [6].
In comparison to Musca domestica, DmGSTE6 is 79 and 77% similar and also 62 and 59% identical to MdGST6A and MdGST6B, respectively. Both MdGST6A and MdGST6B showed more or less 40% identity with other Drosophila Epsilon class proteins. MdGST6A and MdGST6B were both known involved in insecticide metabolism in houseflies [9]. Further assumption on both isoforms (DmGSTE6 and DmGSTE7) that could have been participating in insecticide conjugation was much supported by the induction study performed previously [10]. Through proteomic analysis, the expression of GSTE6 and GSTE7 was shown to increase in the methyl parathion-challenged adult flies. The authors however cautiously suggested any direct relationship of the isoforms increase expression to metabolism of the tested insecticide. There were reports suggested that only Epsilon class GSTs were able to react with DCNB [21] and the detoxification ability of GSTs against insecticides is correlated to its ability to react with DCNB [9].There was also instances of which GSTE1-1z of a reportedly high specific activity toward DCNB and yet does not confer insecticide resistance in Anopheles gambiae [22].
The rate of conjugation reaction, V max , for GSTE7 toward DCNB was however ten times faster than GSTE6. The behavior could have implied the differences in the stabilization of the GSH in the hydrophilic pocket that lead to the conjugation of DCNB. This has remarkably affected the Kcat and Kcat/Km values of both isoforms. The turnover and the catalytic efficiency of GSTE7 were, respectively, sixfold and twofold higher than those of GSTE6. GSTE7 has lower K m value to suggest its higher affinity toward NBC. The behavior of GSH conjugation to NBC could probably differ to what has been with DCNB where the speed of reaction was very much higher in GSTE7 when NBC was the second substrate. The catalytic efficiency, Kcat/Km, of the reaction catalyzed by GSTE7 was shown 12 times higher than of the GSTE6. These observations proposed that GSTE7 was a better isoform for conjugation of the selected substrates than GSTE6.
These differences indicated that GSTE6 and GSTE7 have considerable variations in their secondary structural organization. Such variations in structure may form the basis of differences in their corresponding substrate specificities and in catalytic efficiency [23] although both originated from same cluster and located next to each other on the genomic DNA.
Our preliminary investigative attempt to show that both isoforms could conjugate insecticides was performed using thin-layer chromatography. Pesticides such as temephos, malathion, DDT, fenthion, fenitrothion, permethrin, bromophos, chlorpyrifos, clodinafoppropargyl, fenoxaprop-ethyl, propoxur, isoproturon, and methyl parathion were used in the test. None of the pesticides appeared conjugated by both isoforms. The test suggested both recombinant GSTE6 and GSTE7 did not react and involve in detoxification of insecticides and herbicides directly. Thus so far, the role of GSTE6 and GSTE7 in detoxification of insecticides remains unclear as there is no any promising evidence to prove its involvement in detoxification process.
A comprehensive microarray-based atlas of adult gene expression in multiple Drosophila tissues available (http://flyatlas.org) reported that GSTE6 expressed in adult crop, midgut, tubule, hindgut, ovary, and larval hindgut while GSTE7 expressed in adult crop, midgut, tubule, hindgut, virgin spermatheca and larval midgut, and hindgut and fat body. Other reports also indicated that GSTE6 was found abundant in hindgut, while GSTE7 was found abundant in Malpighian tubules [24]. The co-expression of GSTE6 and GSTE5 was suggested to play a role in male reproductive fitness and success [25]. It was identified that a potential DNA transcription factor binding motifs (TFBMs) of cytochrome P 450 , GSTs, and carboxylesterases is expressed in the D. melanogaster third instar larval midguts [26].
GSTE6 was reported to have GRE-like, Fox-like, NF-kappaB-like, and E47-like TFBMs while GSTE7 to have GRE-like and E47-like TFBMs. The four mentioned TFBMs are known to have mammalian function and were observed to be linked to the oxidative stress response [26]. The author reported that GSTE6 and GSTE7 responded different levels of dietary hydrogen peroxide. However, the author concluded that there is no solid evidence to prove if some or all of the potential TFBMs are functional or response of the midgut-associated GSTs to the oxidative stressor, dietary H2O2. They may simply be associated with these genes with limited or no role in response to this oxidative stressor. In another study, GSTE7 also appeared to be involved in activation of survival program through immune deficiency (IMD) pathway as it was reported being expressed in a strongly infected airway epithelium of D. melanogaster [27].
Exposure of Drosophila to toxins evokes coordinated response by the Malpighian tubules, involving both alterations in detoxification pathways as well as enhanced transport through DHR96, the Drosophila ortholog of the vertebrate PXR/CAR family of nuclear receptors [21]. In relation with that statement, a study [28] stated that in insects, either two distinct receptors have evolved the ability to regulate a very similar set of genes. More than one receptor pathway exists to regulate similar sets of genes. This suggests the possibilities of induction of GSTE6 and GSTE7 together with other genes. Apart from that, basal expression and induction were detected in the key metabolic tissues, namely, sections of the midgut and the Malpighian tubules.
However, the difference in the expression of both GSTE6 and GSTE7 and its inability to detoxify possibly due to cis-regulatory elements controlling the expression of genes may not be acting independently whereby the substrate models may be acting solely to increase the transcriptional output of the tissue-specific modules [28], and the fact that these two genes are found sequentially on the chromosome may support a model of coordinated regulation [7]. In another instance, reports have indicated that DmGSTE6 was strongly co-expressed with DmGSTE7, DmGSTE8, DmGSTE5, DmGSTE3, DmGSTE9, and DmGSTD1, while DmGSTE7 was strongly co-expressed with DmGSTE6, DmGSTE8, DmGSTE3, and DmGSTE9 [29]. These gave insights of possible role of a selective protein to be the key regulator of sets of genes.
Rose bengal was known to inhibit several drug-metabolizing enzymes such as cytochrome P450 and UDP-glucuronosyltransferase [30]. Phenol red however inhibited GSTE6 much effective than on GSTE7.
Methylene blue dye and trans-chalcone had shown significant enhancement of GSTE7 activity toward CDNB with EC50 ranging from 1 to 2 × 105 nM. Chalcones are open chain flavonoids that are widely biosynthesized in plants. A study [31] revealed the pharmacological properties of natural and synthetic chalcones as antioxidant, cytotoxic, anticancer, antimicrobial, antiprotozoal, antiulcer, antihistaminic, and anti-inflammatory activities but mechanism of action of trans-chalcone as an inhibitor to GSTE6 while as a stimulator for GSTE7 is remaining unclear. It was reported that certain haloalka(e)nes including ethylene bromide and methylene chloride form a highly reactive episulfonium ion intermediates that catalyze GST activation reactions [32]. The intermediate effect as an inhibitor to GSTE6 while as a stimulator for GSTE7 activity is remaining unclear. Basic triphenylmethane dyes such as crystal violet have been shown to inhibit glutathione S-transferases from both insect sources [33] and rat liver [34]. The mode of inhibition of crystal violet appeared to involve competition by the free dye with the electrophilic substrate [35].

Conclusions
In conclusion, the study has highlighted that both DmGSTDE6 and DmGSTDE7 behave catalytically different toward same substrates, despite their high sequence similarity. DmGSTDE7 appeared to demonstrate high specific activities toward mentioned substrates and catalytically stimulated by trans-chalcone and methylene blue. None of the isoforms appeared to be able to conjugate pesticides in vitro and hence contradictory to previous work that may cautiously suggest their involvement in methyl parathion metabolism. Many other related findings may lead to assumption that the increase in expression of both isoforms could have been a result of co-expression with other related genes, and yet they have not involved directly in pesticide metabolism. The work has also indicated that both isoforms were not participating in oxidative stress and its functional role thus far was of the normal detoxification xenobiotic taken by the organism. This may thus generate further interest in their functional roles, and a more suitable approach could be adapted to realize the goal.