Biochimica et Biophysica Acta 1830 (2013) 3711–3718
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbagen
Crystal structure of a Bombyx mori sigma-class glutathione transferase exhibiting prostaglandin E synthase activity Kohji Yamamoto a,⁎, Akifumi Higashiura b, Mamoru Suzuki b, Kosuke Aritake c, Yoshihiro Urade c, Nobuko Uodome c, Atsushi Nakagawa b a b c
Faculty of Agriculture, Kyushu University Graduate School, 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan Institute for Protein Research, Osaka University, Suita 565-0871, Japan Department of Molecular Behavioral Biology, Osaka Bioscience Institute, 6-2-4, Furuedai, Suita, Osaka 565-0874, Japan
a r t i c l e
i n f o
Article history: Received 27 November 2012 Received in revised form 13 February 2013 Accepted 22 February 2013 Available online 1 March 2013 Keywords: Crystal structure Glutathione Lepidoptera Prostaglandin Prostaglandin synthase
a b s t r a c t Background: Glutathione transferases (GSTs) are members of a major family of detoxification enzymes. Here, we report the crystal structure of a sigma-class GST of Bombyx mori, bmGSTS1, to gain insight into the mechanism catalysis. Methods: The structure of bmGSTS1 and its complex with glutathione were determined at resolutions of 1.9 Å and 1.7 Å by synchrotron radiation and the molecular replacement method. Results: The three-dimensional structure of bmGSTS1 shows that it exists as a dimer and is similar in structure to other GSTs with respect to its secondary and tertiary structures. Although striking similarities to the structure of prostaglandin D synthase were also detected, we were surprised to find that bmGSTS1 can convert prostaglandin H2 into its E2 form. Comparison of bmGSTS1 with its glutathione complex showed that bound glutathione was localized to the glutathione-binding site (G-site). Site-directed mutagenesis of bmGSTS1 mutants indicated that amino acid residues Tyr8, Leu14, Trp39, Lys43, Gln50, Met51, Gln63, and Ser64 in the G-site contribute to catalytic activity. Conclusion: We determined the tertiary structure of bmGSTS1 exhibiting prostaglandin E synthase activity. General significance: These results are, to our knowledge, the first report of a prostaglandin synthase activity in insects. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Glutathione transferases [GSTs, EC 2.5.1.18] are ubiquitously expressed and are responsible for the cellular detoxification of diverse xenobiotics and endogenous substances by conjugation with reduced glutathione (GSH) [1,2]. Multiple classes of GSTs, including the alpha, mu, pi, omega, sigma, theta, and zeta classes in mammals, are defined based on differences in amino acid sequences [3]. Moreover, delta, epsilon, omega, sigma, theta, and zeta classes have been identified in dipteran insects such as Anopheles gambiae and Drosophila melanogaster [4]. GST catalyzes a major step in the xenobiotic detoxification pathway. Insect GSTs are of particular interest given their role in insecticide metabolism. In Lepidoptera, the sigma, omega, zeta, delta-class, and unclassified GSTs of the silkworm Bombyx mori have been reported [5–10]. Recently, the three-dimensional structure of an unclassified
Abbreviations: GSH, glutathione; GST, glutathione transferase; GSTS, sigma-class GST; PG, prostaglandin; PGDS, prostaglandin D synthase; PGES, prostaglandin E synthase; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis ⁎ Corresponding author at: 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan. Tel.: +81 92 621 4991; fax: +81 92 624 1011. E-mail address:
[email protected] (K. Yamamoto). 0304-4165/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbagen.2013.02.021
B. mori GST (bmGSTu) was determined [11]. Because the silkworm provides a model for studying Lepidopteran insects [12,13], comprehensive research on silkworm GSTs should provide insights into combating those species considered as agricultural pests. We reported the molecular cloning of a sigma-class GST produced by the B. mori and compared its catalytic properties with those of the fall webworm Hyphantria cunea [7]. During the course of our investigations on the structural properties, we discovered that the silkworm GST converted prostaglandin H2 (PGH2) to prostaglandin E2 (PGE2). Prostaglandins (PGs) exert numerous and diverse biological effects on a variety of physiological and pathological processes in mammalians, such as smooth muscle contraction, inflammation, and blood clotting. PGH2, derived from arachidonic acid, is an unstable intermediate and is converted efficiently into a more stable metabolite, PGE2. Prostaglandin E synthase [PGES, EC 5.3.99.3] and prostaglandin D synthase [PGDS, EC 5.3.99.2] catalyze the isomerizations of PGH2 to PGE2 and PGH2 to PGD2, respectively. Cytosolic PGES and hematopoietic PGDS of mammals are homologs of sigma-class GSTs [14,15]. However, to our knowledge, no structural data for this class of GST from Lepidoptera are available. Here, we report the threedimensional crystal structure of recombinant bmGSTS1 and the structure–function relationships involved in its catalytic action.
3712
K. Yamamoto et al. / Biochimica et Biophysica Acta 1830 (2013) 3711–3718
2. Materials and methods 2.1. Protein crystallization and analysis A cDNA encoding bmGSTS1 [6] was inserted into the vector pET-11b (Novagen) and used to transform Escherichia coli strain BL21 (DE3) (TAKARA). Recombinant bmGSTS1 was purified according to published methods [6,7] using ammonium sulfate fractionation, ion-exchange chromatography, and gel filtration chromatography and then concentrated using a centrifugal filter (Millipore) to approximately 10 mg/ml in 20 mM Tris–HCl buffer, pH 8.5, containing 0.2 M NaCl. Crystallization was carried out using the sitting-drop vapor diffusion method at 20 °C using Crystal Screen Kits (Hampton Research), including PEG/Ion-1 and -2, Crystal Screen-1 and -2, and Cryo-1 and -2, as reservoir solutions. Each drop was formed by mixing an equal (0.2 μl) or two-fold greater volume (0.2:0.4 μl) of protein and reservoir solutions, respectively. Crystals suitable for X-ray analysis were grown for 1 week in 0.1 M HEPES, pH 7.5, 30% PEG400 (w/v), and 20% 1,2-propanediol (v/v). Crystals were soaked in the same reservoir solution containing 10 mM GSH before analysis. X-ray diffraction data were acquired with cryo-cooled crystals by synchrotron radiation at beamline BL44XU in SPring-8. Crystals were scooped with a nylon loop before they were frozen in liquid nitrogen. The diffraction data were collected from a single crystal at 100 K in a stream of nitrogen gas and were processed using the HKL2000 package [16]. 2.2. Determination of structure The crystal structure of bmGSTS1 was determined using the molecular replacement method aided by MOLREP software [17]. Unless otherwise indicated, the research model used for superposition was DmGSTS1-1, a sigma-class GST of D. melanogaster (PDB ID: 1M0U) [18]. The structures of bmGSTS1 and the complex with GSH were refined using PHENIX [19] with diffraction data at resolutions of 1.9 Å and 1.7 Å, respectively. Manual adjustments in the final structure were evaluated using Coot [20]. The stereochemical quality of the final model was assessed by the MolProbity program [21]. Crystallographic parameters and refinement statistics are summarized in Table 1. Figures were prepared using Coot and PyMOL software (http://pymol.sourceforge.net). The atomic coordinates and structure factors of native bmGSTS1 and bmGSTS1 complexed with GSH have been deposited in the Protein Data Bank (PDB ID: 3VPT for native bmGSTS1, PDB ID: 3VPQ for bmGSTS1 complexed with GSH). Alignment of deduced amino acid sequences was performed using ClustalW (ver. 1.83) with 10 and 0.2 as the gap creation and gap extension penalties, respectively.
Table 1 Data collection and refinement statistics. Parameter
bmGSTS1
Space group P6522 Unit cell parameters (Å) a = b = 56.1, c = 209.9 (°) α = β = 90, γ = 120 Beam line SPring-8 BL44XU Wavelength (Å) 0.9 Resolution range (Å) 27.1–1.90 (2.02–1.90) Total number of observation 307,637 reflections Total number of unique 22,394 (3165) reflections Multiplicity 13.7 (14.2) Rmergea (%) 4.9 (37.5) b 20.2 (4.1) bI>/bσ (I)> Completeness (%) 99.4 (100.0) Refinement statistics Resolution range (Å) 27.1–1.90 Number of reflections Working set/test set 16,203/804 Rworkc (%)/Rfreed (%) 20.4/25.3 Root-mean-square deviations Bond lengths (Å)/bond 0.019/1.7 angles (°) Average B-factors (Å2)/number of atoms Protein 48.2/3270 Small moleculese 71.7/156 Water 46.7/52 Ramachandran analysis Preferred regions (%) 97.5 Allowed regions (%) 1.5 Outliers (%) 1.0
GSH–bmGSTS1 P6522 a = b = 56.0, c = 209.3 α = β = 90, γ = 120 SPring-8 BL44XU 0.9 31.7–1.70 (1.78–1.70) 444,670 22,392 (1083) 19.9 (19.5) 6.0 (39.5) 72.3 (6.7) 99.8 (100.0) 31.7–1.70 22,289/1133 18.8/23.8 0.017/1.8
44.1/3283 64.6/251 44.2/78 99.0 0 1.0
a Rmerge = ∑(I − ‹I›) / ∑‹I›, where I is the intensity measurement for a given reflection and ‹I› is the average intensity for multiple measurements of this reflection. b Values in parentheses indicate the highest-resolution shell. c Rwork = ∑|Fobs − Fcal|∑Fobs, where Fobs and Fcal are observed and calculated structure factor amplitudes. d Rfree value was calculated for Rwork, using only an unrefined, randomly chosen subset of reflection data (5%) that were excluded from refinement. e Small molecules include polyethylene glycol and 1,2-propanediol.
a radioimaging plate system (Fuji Film, Tokyo, Japan). Kinetic parameters were obtained using nonlinear least-squares calculations. GST activity using 1-chloro-2,4-dinitrobenzene (CDNB) and GSH was measured spectrophotometrically [11]. Briefly, 0.01 ml of test solution was added to 1 ml of 50 mM sodium phosphate buffer, pH 6.5, containing 0.5 mM CDNB and 5 mM GSH as substrates. Changes in absorbance (340 nm/min) were monitored at 30 °C and converted into moles CDNB conjugated/min/mg protein by using the molar extinction coefficient of the resultant 2,4-dinitrophenyl-glutathione: ε340 = 9600 M−1 cm−1.
2.3. Measurements of enzyme activity 2.4. Site-directed mutagenesis Recombinant wild-type bmGSTS1 was overexpressed and purified as described above. Final protein preparations were subjected to SDS-PAGE using a 15% polyacrylamide slab gel containing 0.1% SDS [22] followed by visualization with Coomassie Brilliant Blue R250. Protein concentration was measured using a Protein Assay Kit (Bio-Rad Laboratories) with bovine serum albumin as a standard. PGES activity was measured using 10 μM [ 14C]PGH2 as substrate in a reaction mixture (50 μl) containing 100 mM Tris–HCl, pH 8.0, 1 mM GSH, and 0.1 mg/ml IgG in the presence or absence of 2 mM MgCl2 [23]. After incubation at 25 °C for 0.5 min, 300 μl of stop solution (diethylether:methanol:1 M citric acid = 30:4:1 (by volume)) was added to the reaction mixture, and the resulting organic solvent containing fatty acids was dehydrated by the addition of Na2SO4 powder. The products were separated by thin layer chromatography at −20 °C with mobile phase (diethylether: methanol:acetic acid = 90:2:1 by volume). The conversion rate from 14 C-labeled substrate to 14C-labeled products was calculated by using
Amino acid-substitution mutants of bmGSTS1 were constructed using a plasmid containing the coding sequences for the bmGSTS1 wild-type enzyme and the Quick-Change Site-Directed Mutagenesis Kit according to the manufacturer's recommendations (Agilent Technologies, Wilmington, USA). The nucleotide sequences of the full-length mutant cDNA were determined. Sequence alignments were performed using ClustalW (ver. 1.83), with 10 and 0.2 as the gap creation and gap extension penalties, respectively. 3. Results 3.1. Structural determination and refinement The crystallographic space group is P6522 with unit cell dimensions of a = b = 56.13 Å, and c = 209.89 Å. The final X-ray diffraction data
K. Yamamoto et al. / Biochimica et Biophysica Acta 1830 (2013) 3711–3718
and structural refinement statistics are presented in Table 1. The refined model had an Rwork of 20.4% and an Rfree of 25.3% for data with a resolution between 27.1 Å and 1.9 Å. Its root-mean-square deviations for bond lengths and angles were 0.019 Å and 1.7°, respectively.
3713
Ramachandran analysis showed that 97.5% of the dihedral angles were found to be in the most preferred regions, 1.5% in the allowed regions, and 1.0% in the outlier regions. The final X-ray data and structural refinement statistics are shown in Table 1. Rwork and Rfree values
A
B
C
2
2
PMG
8
2
2
9
1
1
9
1
3
4
5
3 4
5 7
7
1
6
8
3
3 4
6 4 Fig. 1. Primary and tertiary structure alignments. (A) Primary structure alignments. Amino acid sequences of bmGSTS1 (in this study), DmGSTS1-1 (PDB ID: 1M0U), rat PGDS (PDB ID: 1PD2), mPGES1 (PDB ID: 3DWW) and cPGES (PDB ID: 1EJF) were used. The α-helices and β-strands are indicated in red and green, and labeled with α and β. Based on the coordinates generated here, amino acid residues of the G- and H-sites are indicated by orange and blue circles, respectively, on top of the alignment. A green circle represents a residue interacting with both the G- and H-sites. (B) Superposition of the tertiary structures of bmGSTS1 with GSH–bmGSTS1. Green and blue colors indicate bmGSTS1 and GSH–bmGSTS1. GSH is shown in purple. (C) Tertiary structure superposition of bmGSTS1 with DmGSTS1-1. Green and purple colors indicate bmGSTS1 and DmGSTS1-1, respectively. Starting points of α-helices and β-strands are shown by α and β.
3714
K. Yamamoto et al. / Biochimica et Biophysica Acta 1830 (2013) 3711–3718
were 18.8% and 23.8%, respectively, with a resolution between 31.7 Å and 1.7 Å, and the root mean square deviations for bond lengths and angles were 0.017 Å and 1.8°, respectively. Average B-factors, especially for small molecules were relatively high. Ramachandran analysis showed that 99.0% of the dihedral angles were found to be in the most preferred regions and 1.0% in the outlier regions. 3.2. Structure description of bmGSTS1 The crystal structure of bmGSTS1 was solved by the molecular replacement method. This B. mori protein shares 44.5% sequence identity with the query model DmGSTS1-1, a GST from D. melanogaster [PDB ID: 1M0U]. When secondary structural elements, defined by the DSSP program [24], were incorporated into the model they were conserved in a structure-based alignment (Fig. 1A,B and C). Each monomer of bmGSTS1 includes 9 α-helices and 4 β-strands (Fig. 1A and B). The structure possesses 2 distinct domains as follows: an N-terminal domain (residues 1–74) and a C-terminal domain (residues 82–204) (Fig. 1B) connected by a linker loop (residues 75–81). The former domain contains 4 βstrands—β1 (residues 4–8), β2 (residues 30–33), β3 (residues 53–56), and β4 (residues 59–62)—and 3 α-helices—α1 (residues 16–24), α2 (residues 39–42), and α3 (residues 64–74). The latter domain comprises the α-helices α4 (residues 82–105), α5 (residues 111–139), α6 (residues 151–167), α7 (residues 172–175), α8 (residues 181–187), and α9 (residues 190–198). 3.3. Amino acid residues in the active site A monomer structure was obtained for bmGSTS1, and after comparison with this model, a prominent cleft between N-terminal and C-terminal domains was apparent (Fig. 1B and C). Comparison of unliganded and liganded bmGSTS1 structures did not reveal any changes in the main and side chains of bmGSTS1 (Fig. 1B). The Superpose program [25] showed structural homology between bmGSTS1 and GSH– bmGSTS1 with a root-mean-square deviation of 0.445 Å/195 residues. The GSH-binding site (G-site) was located in the deep cleft between the 2 domains (Fig. 1B). According to its gel-filtration elution profile, bmGSTS1 exists as a dimer. One molecule of GSH was bound by the
A
active site of each bmGSTS1 monomer (Figs. 1B and 2A). Fig. 2A shows bound GSH displayed as a 2Fo-Fc electron density map. The GSH– bmGSTS1 structure indicated that the GSH was secured tightly inside the active site pocket formed by the residues Tyr8, Leu14, Trp39, Lys43, Gln50, Met51, Gln63, and Ser64 of bmGSTS1 (Fig. 2B); these were superimposed on the corresponding residues of DmGSTS1-1 as follows: Tyr54, Leu60, Trp85, Lys89, Gln96, Met97, Gln109, and Ser110. The active site can be divided into 2 subsites, the G-site and the hydrophobic-binding site (H-site). GSH binds to this site, with its γ-glutamyl region forming hydrogen bonds with the side chain of Gln63 (Gln109 in DmGSTS1-1), the hydroxyl group, and the main chain of Ser64 (Ser110) (Fig. 3A). The cysteinyl moiety of GSH appeared to form hydrogen bonds with the main chain of Met51 (Met97 in DmGSTS1-1) and with the hydroxyl group of Tyr8 (Tyr54 in DmGSTS1-1) (Fig. 3B). The glycyl portion of GSH, interacting with the side chain of Lys43 (Lys89 in DmGSTS1-1), is in close contact with the side chain of Trp39 (Trp85 in agGST1-6) (Fig. 3C).
3.4. Comparison of the structures of bmGSTS1 and prostaglandin D and E synthases The bmGSTS1 amino acid sequence was 32.4%, 12.7%, and 8.3% identical to those of rat PGDS (PDB ID: 1PD2), microsomal human PGES (mPGES1) (PDB ID: 3DWW), and cytosolic human PGES (cPGES) (PDB ID: 1EJF), respectively (Fig. 1A). We found that the α-helices and β-strands of bmGSTS1 were highly conserved in the sequence of rat PGDS (Figs. 1A and 5A). There are 6 α-helices and no β-strands in mPGES1, whereas cPGES possesses 8 β-strands but no α-helices (Fig. 1A). The structure of PGDS can be superposed on that of bmGSTS1, although the α5 regions do not overlap (Fig. 4). The GSH-binding site amino acid residues in rat PGDS are Tyr8, Phe9, Arg14, Trp39, Lys43, Lys50, Ile51, Pro52, Gln63, Ser64, and Asp97 [14]; and 8 out of 10 residues are superposed to those in bmGSTS1. Residues equivalent to GSH-binding sites in mPGES1 (Arg38, Arg70, Tyr117, Arg126, and Tyr130) were not observed in bmGSTS1. To our knowledge, no data have been published regarding the GSH-binding sites of cPGES.
B Tyr8 Leu14
Ser64 GSH
Met51 Gln63 Gln50
Lys43
Trp39
Fig. 2. GSH-binding to bmGSTS1. Carbon atoms in GSH and bmGSTS1 are shown in magenta and green, respectively. Colors of the atoms are as follows: oxygen (red), nitrogen (blue), and sulfur (yellow). (A) The σΑ-weighted 2FO-FC electron density map contoured at 1.5 σ. (B) Tertiary structure of bmGSTS1 showing amino acid residues near the G-site.
K. Yamamoto et al. / Biochimica et Biophysica Acta 1830 (2013) 3711–3718
A
3715
B Ser64
Tyr8 Met51 2.62
2.95
2.96
2.67
2.87
GSH
Gln63 GSH 3.23
C Trp39
2.95
Lys43 GSH 2.70
Fig. 3. Tertiary structure superposition of bmGSTS1 showing amino acid residues near the G-site. Coloring schemes for displays are as described in the caption of Fig. 2. Hydrogen bonds are indicated by broken lines. (A) Residues interacting with the γ-glutamyl region of GSH. (B) Residues interacting with the cysteinyl moiety of GSH. (C) Residues interacting with the glycyl portion of GSH.
To determine additional activities, we investigated whether bmGSTS1 is involved in the metabolism of prostaglandin. Purified bmGSTS1 was incubated with PGH2 in the presence of GSH. Thin-layer chromatography (TLC) (Fig. 5A) showed that bmGSTS1 did not convert PGH2 to PGD2 but converted PGH2 to PGE2. Nonenzymatic conversion of PGH2 to PGE2 occurred in the lane without enzyme (Fig. 5A). If heat-denatured (boiled) bmGSTS1 was incubated in the presence of GSH, no increased formation of PGE2 was observed (Fig. 5A, lane labeled “heat”). After incubation with human PGDS, PGD2 was formed (Fig. 5A, lane labeled by HPGDS). Purified bmGSTS1 showed Michaelis–Menten rate behavior toward GSH. The Km values with and without Mg2+ were 0.18 and 0.19 mM, respectively, whereas no change in Vmax was observed with or without Mg2+ (Fig. 5B).
of GSH was bound inside the pocket formed by residues Tyr8, Leu14, Trp39, Lys43, Gln50, Met51, Gln63 and Ser64 of bmGSTS1 (Fig. 2). To determine which residues contribute to the catalytic activity of bmGSTS1, 8 residues were each converted to Ala by using site-directed mutagenesis, and the resulting mutants were named Y8A, L14A, W39A, K43A, Q50A, M51A, Q63A, and S64A. After purification from E. coli, each preparation migrated as a single band upon SDS-PAGE. The specific activities of the bmGSTS1 mutants were compared with those of the wild-type enzyme using PGH2 (Fig. 6A) and 1-chloro-24-dinitrobenzene (CDNB) (Fig. 6B). The specific activities of all the mutants were significantly reduced. The most prominent change was the reduction in the activities in W39A and Q63A.
3.5. Active-site residues
4. Discussion
As shown in Fig. 2, the GSH-binding site (G-site) was located in the N-terminal domain of each monomer. In the G-site, 1 molecule
In the present study, we elucidated the tertiary structure of bmGSTS1, a B. mori sigma-class GST. Its globular shape is similar to
3716
K. Yamamoto et al. / Biochimica et Biophysica Acta 1830 (2013) 3711–3718
2
2 1
9
3
4
5 7
1
8
3 6 4 Fig. 4. Structure superposition of bmGSTS1 with rat PGDS. Tertiary structure superposition of bmGSTS1 with rat PGDS. Green and orange colors indicate bmGSTS1 and rat PGDS.
those of other known sigma-class GSTs of rat, nematode, and fruit fly [14,18,26]. In the N-terminal domain, which is composed of 3 α-helices and 4 β-strands that form a βαβαββα motif, the β3 and β4 strands are antiparallel (Fig. 1A and B) as are the β3 and β4 strands of DmGSTS1-1. However, comparison with DmGSTS1-1 revealed that the α7 helix of bmGSTS1 is not present in DmGSTS1-1 and rat sigma-class GST. One small helix (93PMG95) is specific to DmGSTS1-1 but is not present in rat and silkworm sigma class GSTs. The lock-and-key motif is crucial for stabilizing hydrophobic interactions of the GST monomers [18,26]. In DmGSTS1-1, Met94 of 1 subunit interacts with a region created by residues Thr140, Phe174, Tyr175, Lys178, and Leu179 of the other subunit [18]. We show here the presence of overlapping residues (Phe48, Phe94, Phe128, Phe129, Lys132, and Leu133) in bmGSTS1. This further suggests that the similarity of this B. mori enzyme to the sigma-class GST of D. melanogaster. Other amino acid residues have been proposed to participate in the intersubunit interaction of GSTs of the alpha, mu, and pi classes [27] and in the delta-class GST of Anopheles dirus GST [28]. These studies indicate that the “key” residue Phe104 of 1 subunit interacts with a hydrophobic region formed by “lock” residues Ala68, Leu103, Phe104, and Val107 of the other subunit in A. dirus delta-class GST [28]. In adGSTD4-4, Leu6, Thr31, Leu33, Ala35, Glu37, Lys40, and Glu42 form a small hydrophobic core and an ionic bridge contributing to stabilization of the α2 helix [29]. It should be noted that there are 2 of 6 corresponding residues (Glu37 and Glu41) in bmGSTS1 compared to that of A. dirus. In vitro mutagenesis studies will be required to help verify these interactions. The electron density for GSH contoured at 1.5 σ was clear (Fig. 2A). Our result revealed that the GSH tightly bound to G-site of bmGSTS1. We compared the structures of bmGSTS1 and the monomer of DmGSTS1-1. The superposition of the backbones of bmGSTS1 and DmGSTS1-1 revealed that both possess an N-terminal domain, providing the G-site and a C-terminal domain, which includes the H-site. The secondary structures of these GSTs were comparable, particularly at the G-site, thus providing a starting model for GSH binding in bmGSTS1. Variations in H-site structures of GSTs are responsible for their different substrate specificities. The H-site of DmGSTS1-1 contains residues that are mostly hydrophobic: Val57,
Ala59, Leu60, Arg145, Ala149, Tyr153, Tyr208, Tyr211, and Val249 [18]. In the sequence of bmGSTS1, we found that 7 of the 9 residues, Val11, Ala13, Leu14, Arg99, Ala103, Tyr107, Tyr162, Ala165, and Leu204, are identical to those in DmGSTS1-1. The remaining residue, (Leu204), is structurally similar to the corresponding DmGSTS1-1 residue, whereas Ala165 of bmGSTS1 is replaced by Tyr211 in DmGSTS1-1. Recently, the electron-sharing network that contributes to the catalytic activity of GST has been proposed [30]. This motif can be divided into types I and II [30]. The type I electron-sharing networks exemplified by the GSTs of delta, theta, omega and tau classes contain an acidic amino acid residue at position 64, whereas the type II networks (GSTs of alpha, mu and pi classes) have a polar amino acid residue (glutamate) capable of interacting with the γ-glutamyl portion of GSH. Gln63 is conserved in the sequence of bmGSTS1, which resembles a member of the type II network. In bmGSTS1, Asp97 could participate in an ionic interaction, which is a characteristic of type II networks. We previously determined the tertiary-structure of bmGSTu [11]. Comparison of bmGSTS1 to bmGSTu showed that secondary structures including α-helices and β-strands were conserved, and there are also similarities in tertiary-structures in both GSTs. The identity between bmGSTS1 and bmGSTu sequences is 27.5% with diversity of H- and G-sites in both GSTs. Thus, the higher-order features among GSTs are similar, whereas the G- and H-sites are GST-class specific, which likely affects their substrate specificity. We found that bmGSTS1 exhibited PGES activity. Although PGs are present in insects [31–33], little information is available on related
A ( g/ml)
Without MgCl2 0
1.2
12
120
With MgCl2 1200 1.2
12
120 1200 HPGDS
heat
PGH2
PGD2
PGE2 PGF 2
B
Fig. 5. Production of [14C]PGE2 by bmGSTS1. (A) Various concentrations of bmGSTS1s (μg/ml), described on the TLC image, were incubated with [14C]PGH2 and GSH in the presence or absence of MgCl2. [14C]PGH2 and its metabolites were extracted, separated by TLC, and analyzed by autoradiography. Positions of PGH2, PGD2, PGE2, and PGF2α are shown by arrows on the left. HPGDS (0.2 μg/ml) stands for human prostaglandin D synthase. Heat-denatured bmGSTS1 (50 μg/ml) was used in the lane labeled “heat”. (B) Michaelis–Menten plots of the reactions performed in the presence (open circle) or absence (closed circle) of MgCl2.
K. Yamamoto et al. / Biochimica et Biophysica Acta 1830 (2013) 3711–3718
B Specific activity (µmol/min/mg protein)
Specific activity (µmol/min/mg protein)
A
3717
0.10 0.08 0.06 0.04 0.02 0
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0
mutants
mutants
Fig. 6. Comparison of specific activities toward PGH2 (A) and CDNB (B) by bmGSTS1 mutants. The activities for wild-type and each mutant were determined. Data represent the mean values of 3 independent experiments.
enzymes involved in the production of PG. The tertiary and secondary structures of bmGSTS1 are similar to those of rat PGDS. Moreover, the GSH-binding sites of PGDS are conserved in bmGSTS1. Interestingly, bmGSTS1 isomerizes PGH2 to PGE2 and the affinity for GSH was about 4-fold higher than for human microsomal PGES [34]. The activity of bmGSTS1 was not stimulated by Mg2+ in a concentration-dependent manner. The Mg 2+ increased the activity of human PGDS with a decrease in the Km for the substrate or GSH [35]. We found that bmGSTS1 catalyzed the conversion of prostaglandin H2 to its E2 form. To verify the participation of specific amino acid residues in activity and possibly GSH binding, site-directed mutagenesis of 8 residues was performed. The results suggest that all of these residues in bmGSTS1 play some roles in its catalytic mechanism. Tyr8 of bmGSTS1 appears to be the critical active site equivalent to Tyr, Ser and Cys in pi, delta, and omega classes of GSTs, respectively. As seen in Figs. 2 and 3, Tyr8 binds to GSH with other residues in the G-site of bmGSTS1 cooperatively. Assuming that Tyr8 is not entirely crucial for binding of GSH, it thus appears that the effect on enzyme activities (Fig. 6) also has other explanations. The PGH2-binding sites in the amino acid sequence of PGES are poorly characterized, and the identity of the binding sites of rat PGDS is a matter of speculation [15]. The putative binding sites in PGDS were not conserved in bmGSTS1 when we aligned their sequences. A similar result has been reported for the enzyme from the pathogenic parasite of humans, Onchocerca volvulus. A different binding mode for PGH2 was proposed [36], indicating that variations in PGH2 binding mechanisms exist between species. Because PGH2 is unstable, a stable analogue of PGH2 is required to elucidate the binding sites. PGE2 was first discovered in the seminal vesicles of sheep. We detected PGE2 as the product of bmGSTS1 in B. mori using liquid chromatography/mass spectrometry (Yamamoto, K., unpublished data), suggesting that bmGSTS1 is active in vivo as a PGES. Our result showing the presence of PGE2 in B. mori agrees with a previous study [37]. In insects, PGs influence several aspects of immunity, including cellular defense reactions to bacterial, fungal, parasitic, and viral infections [38]. Thus, bmGSTS1 could play important roles in the cellular defense reactions in B. mori. A cytosolic PGES that requires GSH for its activity and belongs to the GST family based on its amino acid sequence has been described in rat [14]. A membrane-associated PGES partially purified from microsomal fractions of bovine and sheep vesicular glands has also been found to require GSH for its activity and was expressed in E. coli [39]. Membrane-associated PGES that does not require GSH for activity is present in heart, spleen, and uterine microsomes [40]. The development of selective and nonselective inhibitors of bmGSTS1
should help establish the physiological role of bmGSTS1 in the silkworm life cycle. In summary, we describe here the high-resolution crystal structures of a bmGSTS1 that exhibits PGES activity. We were able to reconstruct its entire architecture by comparing it to the three-dimensional structure determined for the enzyme from D. melanogaster. By preparing bmGSTS1–GSH complexes and site-directed mutagenesis, we identified the amino acid residues involved in catalysis. We are currently pursuing co-crystallization of bmGSTS1 with a suitable inhibitor–GSH conjugate to aid in the rational design of more effective pesticides.
Acknowledgements This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan. The work was also supported in part by the Research Grant for Young Investigators of Faculty of Agriculture, Kyushu University. This work was performed under the Cooperative Research Program of Institute for Protein Research, Osaka University. The synchrotron radiation experiments were carried out at the BL44XU of SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No. 2011A6651, and 2011B6651).
References [1] I. Listowsky, M. Abramovitz, H. Homma, Y. Niitsu, Intracellular binding and transport of hormones and xenobiotics by glutathione S-transferase, Drug Metab. Rev. 19 (1998) 305–318. [2] R.N. Armstrong, Structure, catalytic mechanism, and evolution of the glutathione transferases, Chem. Res. Toxicol. 10 (1997) 2–18. [3] B. Mannervik, P.G. Board, J.D. Hayes, I. Listowsky, W.R. Pearson, Nomenclature for mammalian soluble glutathione transferases, Methods Enzymol. 401 (2005) 1–8. [4] C.P. Tu, B. Akgül, Drosophila glutathione S-transferases, Methods Enzymol. 401 (2005) 204–226. [5] K. Yamamoto, P. Zhang, F. Miake, N. Kashige, Y. Aso, Y. Banno, H. Fujii, Cloning, expression and characterization of theta-class glutathione S-transferase from the silkworm, Bombyx mori, Comp. Biochem. Physiol. 141B (2005) 340–346. [6] K. Yamamoto, P. Zhang, Y. Banno, H. Fujii, Identification of a sigma-class glutathione S-transferase from the silkworm, Bombyx mori, J. Appl. Entomol. 130 (2006) 515–522. [7] K. Yamamoto, H. Fujii, Y. Aso, Y. Banno, K. Koga, Expression and characterization of a sigma-class glutathione S-transferase of the fall webworm, Hyphantria cunea, Biosci. Biotechnol. Biochem. 71 (2007) 553–560. [8] K. Yamamoto, S. Nagaoka, Y. Banno, Y. Aso, Biochemical properties of an omega-class glutathione S-transferase of the silkmoth, Bombyx mori, Comp. Biochem. Physiol. 149C (2009) 461–467. [9] K. Yamamoto, Y. Shigeoka, Y. Aso, Y. Banno, M. Kimura, T. Nakashima, Molecular and biochemical characterization of a Zeta-class glutathione S-transferase of the silkmoth, Pestic. Biochem. Physiol. 95 (2009) 25–128.
3718
K. Yamamoto et al. / Biochimica et Biophysica Acta 1830 (2013) 3711–3718
[10] K. Yamamoto, H. Ichinose, Y. Aso, Y. Banno, M. Kimura, T. Nakashima, Molecular characterization of an insecticide-induced novel glutathione transferase in silkworm, Biochim. Biophys. Acta 1810 (2011) 420–426. [11] Y. Kakuta, K. Usuda, T. Nakashima, M. Kimura, Y. Aso, K. Yamamoto, Crystallographic survey of active sites of an unclassified glutathione transferase from Bombyx mori, Biochim. Biophys. Acta 1810 (2011) 1355–1360. [12] K. Mita, M. Morimyo, K. Okano, Y. Koike, J. Nohata, H. Kawasaki, K. Kadono-Okuda, K. Yamamoto, M.G. Suzuki, T. Shimada, M.R. Goldsmith, S. Maeda, The construction of an EST database for Bombyx mori and its application, Proc. Natl. Acad. Sci. U. S. A. 100 (24) (2003) 14121–14126. [13] K. Mita, M. Kasahara, S. Sasaki, Y. Nagayasu, T. Yamada, H. Kanamori, N. Namiki, M. Kitagawa, H. Yamashita, Y. Yasukochi, K. Kadono-Okuda, K. Yamamoto, M. Ajimura, G. Ravikumar, M. Shimomura, Y. Nagamura, T. Shin-I, H. Abe, T. Shimada, S. Morishita, T. Sasaki, The genome sequence of silkworm, Bombyx mori, DNA Res. 11 (1) (2004) 27–35. [14] T. Tanioka, Y. Nakatani, N. Semmyo, M. Murakami, I. Kudo, Molecular identification of cytosolic prostaglandin E2 synthase that is functionally coupled with cyclooxygenase-1 in immediate prostaglandin E2 biosynthesis, J. Biol. Chem. 275 (2000) 32775–32782. [15] Y. Kanaoka, H. Ago, E. Inagaki, T. Nanayama, M. Miyano, R. Kikuno, Y. Fujii, N. Eguchi, H. Toh, Y. Urade, O. Hayaishi, Cloning and crystal structure of hematopoietic prostaglandin D synthase, Cell 90 (1997) 1085–1095. [16] Z. Otwinowski, W. Minor, Processing of X-ray diffraction data collected in oscillation mode, Methods Enzymol. 276 (1997) 307–326. [17] A. Vagin, A. Teplyakov, An approach to multi-copy search in molecular replacement, Acta Crystallogr. D: Biol. Crystallogr. 56 (2000) 1622–1624. [18] B. Agianian, P.A. Tucker, A. Schouten, K. Leonard, B. Bullard, P. Gros, Structure of a Drosophila sigma class glutathione S-transferase reveals a novel active site topography suited for lipid peroxidation products, J. Mol. Biol. 326 (2003) 151–165. [19] P.D. Adams, P.V. Afonine, G. Bunkòczi, V.B. Chen, N. Echols, J.J. Headd, L. Hung, S. Jain, G.J. Kapral, R.W.G. Kunstleve, A.J. McCoy, N.W. Moriarty, R.D. Oeffner, R.J. Read, D.C. Richardson, J.S. Richardson, T.C. Terwilliger, P.H. Zwart, The Phenix software for automated determination of macromolecular structures, Methods 55 (2011) 94–106. [20] P. Emsley, K. Cowtan, Coot: model-building tools for molecular graphics, Acta Crystallogr. D: Biol. Crystallogr. 60 (2004) 2126–2132. [21] V.B. Chen, W.B. Arendall III, J.J. Headd, D.A. Keedy, R.M. Immormino, G.J. Kapral, L.W. Murray, J.S. Richardson, D.C. Richardson, MolProbity: all-atom structure validation for macromolecular crystallography, Acta Crystallogr. D: Biol. Crystallogr. 66 (1) (2010) 12–21. [22] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. [23] M. Lazarus, B.K. Kubata, N. Eguchi, Y. Fujitani, Y. Urade, O. Hayaishi, Biochemical characterization of mouse microsomal prostaglandin E synthase-1 and its colocalization with cyclooxygenase-2 in peritoneal macrophages, Arch. Biochem. Biophys. 397 (2002) 336–341. [24] R. Maiti, G.H. Van Domselaar, H. Zhang, D.S. Wishart, SuperPose: a simple server for sophisticated structural superposition, Nucleic Acids Res. 1 (2004) W590–W594. [25] W. Kabsch, C. Sander, Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features, Biopolymers 22 (1983) 2577–2637.
[26] J.D. Schuller, Q. Liu, I.A. Kriksunov, A.M. Campbell, J. Barrett, P.M. Brophy, Q. Hao, Crystal structure of a new class of glutathione transferase from the model human hookworm nematode Heligmosomoides polygyrus, Proteins 61 (2005) 1024–1031. [27] I. Sinning, G.J. Kleywegt, S.W. Cowan, P. Reinemer, H.W. Dirr, R. Huber, G.L. Gilliland, R.N. Armstrong, X. Ji, P.G. Board, B. Olin, B. Mannervik, T.A. Jones, Structure determination and refinement of human alpha class glutathione transferase A1-1, and a comparison with the Mu and Pi class enzymes, J. Mol. Biol. 232 (1993) 192–212. [28] J. Wongsantichon, A.J. Ketterman, An intersubunit lock-and-key ‘Clasp’ motif in the dimer interface of delta class glutathione transferase, Biochem. J. 394 (2006) 135–144. [29] A. Vararattanavech, P. Prommeenate, A.J. Ketterman, The structural roles of a conserved small hydrophobic core in the active site and an ionic bridge in domain I of delta class glutathione S-transferase, Biochem. J. 393 (2006) 89–95. [30] P. Winayanuwattikun, A.J. Ketterman, Glutamate-64, a newly identified residue of the functionally conserved electron-sharing network contributes to catalysis and structural integrity of glutathione transferases, Biochem. J. 402 (2007) 339–348. [31] H. Tunaz, S.M. Putnam, D.W. Stanley, Prostaglandin biosynthesis by fat body from larvae of the beetle Zophobas atratus, Arch. Insect Biochem. Physiol. 49 (2002) 80–93. [32] E.J. Wakayama, J.W. Dillwith, G.J. Blomquist, Characterization of prostaglandin synthesis in the housefly, Musca domestica (L.), Insect Biochem. 16 (1986) 903–909. [33] G.G. Gadelhak, V.K. Pedibhotla, D.W. Stanley-Samuelson, Eicosanoid biosynthesis by hemocytes from the tobacco hornworm, Manduca sexta, Insect Biochem. Mol. Biol. 25 (1995) 743–749. [34] S. Thorén, R. Weinander, S. Saha, C. Jegerschöld, P. Pettersson, B. Samuelsson, H. Hebert, M. Hamberg, R. Morgenstern, P. Jakobsson, Human microsomal prostaglandin E synthase-1, J. Biol. Chem. 278 (2003) 22199–22209. [35] T. Inoue, D. Irikura, N. Okazaki, S. Kinugasa, H. Matsumura, N. Uodome, M. Yamamoto, T. Kumasaka, M. Miyano, Y. Kai, Y. Urade, Mechanism of metal activation of human hematopoietic prostaglandin D synthase, Nat. Struct. Biol. 10 (2003) 291–296. [36] M. Perbandt, J. Höppner, C. Burmeister, K. Lüersen, C. Betzel, E. Liebau, Structure of the extracellular glutathione S-transferase OvGST1 from the human pathogenic parasite Onchocerca volvulus, J. Mol. Biol. 377 (2008) 501–511. [37] B.N. Yamaja, T. Setty, T.R. Ramaiah, Isolation and identification of prostaglandins from the reproductive organs of male silkworm, Bombyx mori L, Insect Biochem. 9 (1979) 613–617. [38] D.W. Stanley, C. Goodmana, S. An, A. McIntosh, Q. Song, Prostaglandins A1 and E1 influence gene expression in an established insect cell line (BCIRL-HzAM1 cells), Insect Biochem. Mol. Biol. 38 (2008) 275–284. [39] P.J. Jakobsson, S. Thoren, R. Morgenstern, B. Samelsson, Identification of human prostaglandin E synthase: a microsomal glutathione-dependent, inducible enzyme, constituting a potential novel drug target, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 7220–7225. [40] K. Watanabe, K. Kurihara, T. Suzuki, Purification and characterization of membrane bound prostaglandin E synthase from bovine heart, Biochim. Biophys. Acta 1439 (1999) 406–414.