- Email: [email protected]
Cadmium(II)-stimulated enzyme activation of Arabidopsis thaliana phytochelatin synthase 1
Journal of Inorganic Biochemistry 105 (2011) 111–117
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j i n o r g b i o
Cadmium(II)-stimulated enzyme activation of Arabidopsis thaliana phytochelatin synthase 1 Shinya Ogawa, Takahiro Yoshidomi, Etsuro Yoshimura ⁎ Department of Applied Biological Chemistry, School of Agriculture and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo, Tokyo 113-8657, Japan
a r t i c l e
i n f o
Article history: Received 3 August 2010 Received in revised form 25 September 2010 Accepted 27 September 2010 Available online 23 November 2010 Keywords: Arabidopsis thaliana Cadmium Phytochelatin Glutathion Phytochelatin synthase
a b s t r a c t Phytochelatin (PC), a class of heavy metal-binding peptides, is synthesized from the tripeptide glutathione (GSH) and/or previously synthesized PC in a reaction mediated by PC synthase (PCS). In the present study, the PC production rate catalyzed by recombinant Arabidopsis PCS1 (rAtPCS1) in the presence of a constant free Cd (II) level increased steadily and the kinetic parameters were approximated using a substituted-enzyme mechanism in which GSH and bis(glutathionato)cadmium acted as co-substrates. In contrast, the PC production rate as a function of GSH concentration at a constant total Cd(II) concentration reached a maximum, which shifted toward higher GSH concentrations as the concentration of Cd(II) was increased. These observations are consistent with the suggestion that rAtPCS1 possesses a Cd(II) binding site where Cd (II) binds to activate the enzyme. The affinity constant, optimized using a one-site mathematical model, successfully simulated the experimental data for the assay system using lower concentrations of Cd(II) (5 or 10 μM) but not for the assay using higher concentrations (50 or 500 μM), where a sigmoidal increase in PCS activity was evident. Furthermore, the PCS activity determined at a constant GSH concentration as a function of Cd(II) concentration also reached a maximum. These findings demonstrate that rAtPCS1 also possesses a second Cd(II) binding site where Cd(II) binds to induce an inhibitory effect. A two-site mathematical model was applied successfully to account for the observed phenomena, supporting the suggestion that rAtPCS1 possesses two Cd(II) binding sites. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Phytochelatins (PCs) are a family of peptides with the general structure (γ-Glu-Cys)n-Gly (PCn), where n ≥ 2. The peptides, designated as cadystin, were first identified in the fission yeast Schizosaccharomyces pombe grown in Cd(II)-containing medium [1]. Shortly after the initial identification, the peptides were detected in cells of the higher plant Rauvolfia serpentina in suspension culture [2]. To date, PCs have been identified in higher plants, algae, and some fungi exposed to toxic levels of heavy metal ions [3]. PCs play a role in detoxification by sequestering heavy metal ions, although other roles in Zn(II) homeostasis and the long-distance transport of Zn(II) have been suggested [4]. The synthesis of PC is catalyzed by phytochelatin synthase (PCS) using the tripeptide glutathione (γ-Glu-Cys-Gly; GSH) and/or previously synthesized PC as a substrate [5]. PCS, a γ-glutamylcysteinyl transpeptidase (EC 2.3.2.15), mediates a dipeptidyl transfer reaction in the presence of heavy metals, leading to the formation of PCn with a greater number of γ-glutamylcysteinyl units, which has increased affinity for heavy metals and the capacity to sequester them
⁎ Corresponding author. Tel.: + 81 3 5841 5153; fax: +81 3 5841 8027. E-mail address: [email protected] (E. Yoshimura). 0162-0134/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2010.09.011
more effectively [6]. Metals and metalloids belonging to groups 11–15 in the fourth, fifth, and sixth periods of the periodic table can activate PCS-mediated PC synthesis [7]. As PCS is expressed constitutively in cells, the metal ion dependency of PC synthesis represents a selfterminating reaction in which the activator metal ions are captured by the reaction product. The enzymatic mechanism of action for PCS in relation to activation by metal ions has yet to be resolved. Adding Cd(II) to a solution containing GSH and PCS from Silene cucubalus initiates PC synthesis, and the supplementing the solution with Cd(II) chelators, such as uncomplexed PCn mixtures or ethylenediaminetetraacetic acid, immediately terminates PC synthesis, suggesting that PCS is activated by metal ion binding [5]. Equilibrium dialysis of AtPCS1FLAG indicated seven Cd(II) binding sites in the protein with a dissociation constant of 0.54 μM [8]. However, under standard PCS assay conditions, in which GSH is present at a much higher level than Cd(II), the level of free Cd(II) ions is extremely low; using the reported equilibrium binding constants for Cd(II) and GSH, the free Cd(II) concentration can be predicted, which has been calculated as 6.638 × 10− 13 M when Cd(II) and GSH are present at total concentrations of 25 μM and 3.3 mM, respectively, at pH 8.0 [8]. Thus, the concentration of free Cd(II) ions is too low to account for the binding of Cd(II) ions to AtPCS1-FLAG [8]. Kinetic analysis of PC synthesis mediated by AtPCS1-FLAG has indicated a reaction process consistent
112
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
with a bisubstrate-substituted mechanism, whereby a GSH molecule and a Cd(II)-G2 complex act as co-substrates and the enzyme is already in an active form without binding to Cd(II) [8]. Furthermore, it has been inferred from the facility of AtPCS1 for the synthesis of Salkylated PCs from S-alkylated GSH in the absence of added heavy metal ions that GSH derivatives with thiol groups blocked either through heavy metal thiolate formation or S-alkylation are solely contingent on PC synthesis [8]. However, peptide scanning of PCS from wheat Triticum aestivum L. (TaPCS1) and the fission yeast S. pombe (SpPCS) revealed that Cd(II) binds to the proteins in the presence of GSH at a concentration of 1 mM [9]. In addition, Cd(II)dependent stabilization of PCS has been reported for the enzyme produced by the primitive red alga Cyanidioschyzon merolae, supporting the suggestion that Cd(II) binds to the enzyme [10]. Therefore, no definitive conclusion has been reached regarding the metal activation mechanism of PC synthesis. 2. Materials and methods 2.1. Heterologous expression and purification of N-terminal hexahistidinyl-tagged Arabidopsis thaliana PCS1
acid (TFA). The mixture was diluted with 0.1% TFA in which 0.5 μM desglycyl PC2 was added as an internal standard. PC concentrations were determined by HPLC using detection based on the dequenching of Cu(I)-BCS complexes [13,14]. The peptides were separated on an octadecylsilane column (4.6 mm ϕ × 150 mm, TSKgel ODS-80™; Tosoh, Tokyo, Japan) with a mobile phase consisting of 5% acetonitrile/0.1% TFA using a pump (PU-1580; Jasco, Tokyo, Japan) at a flow rate of 1 mL min− 1. Using another pump (PU-980; Jasco) at a flow rate of 1 mL min− 1, the eluent was merged with a post-column solution containing 50 mM CHES-NaOH (pH 10.0), 0.2 μM CuSO4, 0.5 μM BCS, and 5 μM ascorbic acid. The merged solution was passed through a mixing coil, and the fluorescence intensity at excitation and emission wavelengths of 280 and 395 nm, respectively, was monitored using an FP-920 fluorescence detector (Jasco). During HPLC analyses, the post-column solution was bubbled continuously with a stream of He. The integrated peak area was used to quantify the PC level after calibration with chemically synthesized PC (Hayashi Kasei, Osaka, Japan). PCS activity was expressed as the amount of PC2 synthesized by 1 mg of protein per minute, because the synthesis of PCn with n greater than 2 was negligible. 2.3. PCS activity assays in the presence of a constant level of free Cd(II)
We used a recombinant Arabidopsis thaliana PCS1 tagged with hexahistidine at its N-terminus (rAtPCS1). The AtPCS1 DNA sequence was amplified using PrimeSTAR HS DNA polymerase (TakaraBio, Shiga, Japan) and a plasmid containing the AtPCS1 gene (Riken BioResource Center, Saitama, Japan) [11,12] was used as a template. The forward and reverse PCR primers were 5′-GAGAGCTAGCATGGCTATGGCGAGTTTATATCG-3′ (NheI site underlined) and 5′-GAGAGAATTCCTAATAGGCAGGAGCAGCGAGATC-3′ (EcoRI site underlined), respectively. The resultant DNA fragment was transferred to the pET28b expression vector (Novagen, Madison, WI) as an in-frame fusion with a plasmid-encoded hexahistidine tag sequence, and the plasmid was used to transform Escherichia coli XL-1 Blue (Stratagene, La Jolla, CA). Following the confirmation of successful cloning by sequencing with T7 primers, the E. coli BL21 Rosetta (DE3) pLysS strain (Novagen) was transformed with the construct. E. coli cells carrying rAtPCS1 were pre-cultured in 200 mL of LuriaBertani broth at 25 °C. After reaching an optical density at 600 nm of 0.1–0.8, isopropyl 1-thio-β-D-galactoside was added to 40 μM, and the culture was incubated for a further 6 h to express the recombinant protein. The cells were collected by centrifugation at 8000× g for 15 min and then suspended in an extraction solution (25 mM Tris– HCl, pH 8.0, 400 mM NaCl, 10% glycerol, 1 mM 2-mercaptoethanol, 0.1% Tween 20, and 1 mM phenylmethylsulfonyl fluoride). The cells were disrupted by sonication, and the supernatant was obtained by centrifugation at 13,000× g for 10 min at 4 °C. The recombinant protein was purified on a Ni2+ affinity column by fast-performance liquid chromatography (Pharmacia LKB, Uppsala, Sweden). The supernatant (20 mL) was applied to a 1-mL HisTrap™ HP column (GE Healthcare Life Sciences, Piscataway, NJ) equilibrated with binding buffer (20 mM HEPES-NaOH, pH 7.4, 500 mM NaCl, 1 mM 2-mercaptoethanol, and 40 mM imidazole), the column was washed with 10 mL of binding buffer, and then the sample was eluted with a linear gradient of 40–250 mM imidazole in binding buffer over 20 min. The rAtPCS1 eluted at 150 mM imidazole. Analysis of the eluate by SDS-PAGE revealed a single band with a molecular mass of 56.5 kDa, confirming the homogeneity of the preparation.
PCS activity was determined at various GSH concentrations while keeping the free Cd(II) concentration constant. For this purpose, the total Cd(II) and GSH concentrations were changed concomitantly from the initial values of 5 μM and 10 mM, respectively, in such a way that the ratio of [Cd(II)]t to [G]2t was maintained at 5 × 10− 2 M− 1. The concentration of free Cd(II) ions was calculated to be 1.22 × 10− 10 M under these conditions (Supplementary 1 and 2). 2.4. Determination of parameters Non-linear least squares analysis was performed using the Origin 8.1 program (OriginLab, Northampton, MA). 3. Results 3.1. PC synthesis activity of rAtPCS1 as a function of GSH concentration The PCS activity of rAtPCS1 as a function of the concentration of GSH supplemented in the assay solution is shown in Fig. 1. In the presence of 10 μM Cd(II) (Fig. 1, closed squares), activity increased and reached the maximum rate with an increase in GSH concentration
2.2. PCS activity assays The reaction mixture (100 μL) contained 100 ng/mL rAtPCS1, 10 mM 2-mercaptoethanol, 100 ng/mL bovine serum albumin, and various concentrations of GSH and CdSO4 in 200 mM HEPES-NaOH buffer (pH 8.0). The mixture was incubated at 35 °C for 15 min, and the reaction was terminated by adding 25 μL of 10% trifluoroacetic
Fig. 1. Effects of GSH concentration on the PCS activity of rAtPCS1 in assay solution containing total Cd(II) concentrations of 1 (closed triangles), 5 (closed circles), and 10 μM (closed squares). The solid lines (a, b, and c) represent simulated activity using the one-site model for the assay solutions containing 1, 5, and 10 μM Cd(II), respectively, where the Vmax and KE1 values listed in Table 1 were used.
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
113
up to 15 mM, and then decreased with further increases in GSH concentration. A similar trend was observed for the enzyme assay system in which 1 or 5 μM Cd(II) was added (Fig. 1, closed circles and triangles, respectively). It was also evident that the maximal activity occurred at higher GSH concentrations with an increase in Cd(II) concentration. Thus, the maximal activity was attained at GSH concentrations around 10 and 15 mM when the enzyme was assayed in the presence of 5 and 10 μM total Cd(II), respectively. Although the GSH concentration at which the maximal activity attained may be obscure in the enzyme assay system with 1 μM total Cd(II), it occurred at a concentration less than or equal to 5 mM. These phenomena still held for the assay solution containing higher levels of total Cd(II). As shown in Fig. 2, maximal PCS activity occurred at 30 and 60 mM GSH for the assay solution containing 50 μM (closed squares) and 500 μM Cd(II) (closed circles), respectively. In addition, a sigmoidal increase in activity was observed at GSH concentrations below 20 mM for the assay system containing 50 and 500 μM Cd(II), a prominent feature for the system at high levels of Cd(II). 3.2. PC synthesis activity of rAtPCS1 as a function of the total Cd(II) concentration The PCS activity of rAtPCS1 as a function of Cd(II) added to the assay solution is illustrated in Fig. 3. At a constant GSH concentration of 5 mM, PCS activity increased with increasing Cd(II) concentration up to 4 μM, and then decreased sharply (Fig. 3A). A similar activity profile was obtained for the solution containing 20 mM GSH, where the maximal activity was attained at 60 μM Cd(II) (Fig. 3B). These results indicated that rAtPCS1 has an optimum Cd(II) concentration for PCS activity, which shifts toward a higher Cd(II) concentration in the presence of higher levels of GSH. 3.3. PC synthesis activity of rAtPCS1 as a function of GSH concentration at a constant free Cd(II) concentration PCS activity at a constant free Cd(II) concentration of 1.22 × 10− 10 M was determined as a function of total GSH concentration. The activity increased steadily with total GSH concentration up to 50 mM (Fig. 4A). A Hanes plot of [G]t/activity versus [G]t showed a straight line at GSH concentrations greater than 25 mM, whereas the plot deviated upward at GSH concentrations below 25 mM (Fig. 4B). These findings indicated that the PCS synthesis reaction can display
Fig. 3. Effects of total Cd(II) concentration on the PCS activity of rAtPCS1 in the presence of 5 mM (A) and 20 mM (B) GSH. The inset in (A) is an expansion of the figure at Cd(II) concentrations from 0 to 40 μM. The solid lines represent the simulated activity, where the Vmax, KE1, and KE2 values listed in Table 2 were used.
Michaelis–Menten-type kinetics only at GSH concentrations above 25 mM. 4. Discussion 4.1. PCS activity of rAtPCS1 as a function of GSH
Fig. 2. Effects of GSH concentration on the PCS activity of rAtPCS1 in the assay solution containing total Cd(II) concentrations of 50 (closed squares) and 500 μM (closed circles). The solid lines (a and b) represent simulated activity using the two-site model for the assay solutions containing 50 and 500 μM Cd(II), respectively, where the Vmax, KE1, and KE2 values listed in Table 3 were used.
Although a definitive conclusion has not been reached regarding the mechanisms of metal ion activation of PC synthesis, enzyme acylation by a γ-glutamylcysteinyl group has been clearly demonstrated [15]. The thiol group of Cys-56 in AtPCS1 has been identified as an acylation site, which undergoes a reaction with GSH in a Cd(II)independent manner [15]. The γ-glutamylcysteinyl group on an acyl intermediate is transferred to an acceptor GSH or PCn − 1 molecule in PCn synthesis. Two distinct mechanisms have been proposed for the γ-glutamylcysteinyl dipeptide transfer. In the case of PC2 synthesis, for example, the γ-glutamylcysteinyl group on PCS is transferred to a Cd(II)-G2, in which the enzyme is already in an active form with no need for the direct binding of metals [8]. The alternative mechanism is that Cd(II) is required to bind PCS to activate it for the transfer of the dipeptide to an acceptor molecule such as GSH or PCn − 1 [5]. The PCS activity of rAtPCS1 determined at a constant total Cd(II) concentration demonstrated the maximum activity with respect to GSH concentration (Figs. 1 and 2). In addition, with an increase in the
114
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
concentration was attributable to a lower free Cd(II) level. These findings clearly demonstrated that AtPCS1 activation requires Cd(II) binding. A proposed substituted-enzyme mechanism [8] in which the enzyme is in an active form without being bound by Cd(II) and the GSH and Cd(II)-G2 molecules act as co-substrates is in conflict with the declining PCS activity at higher GSH concentrations at a constant total Cd(II) concentration, because an increase in GSH concentration at a constant total Cd(II) concentration will also elevate the free GSH levels and Cd-G2 levels. To make such a mechanism consistent with our observations, it would be necessary to hypothesize a GSH binding site on the enzyme that exerts an inhibitory effect. The steady increase in PCS activity with GSH concentration in the presence of a constant free Cd(II) level may be attributable to an increased concentration of Cd(II)-G2, the effect of which may surpass the inhibitory effect with regard to PCS activity, as the concentration of Cd(II)-G2 increased concomitant with that of GSH in the assay system (Fig. 4). However, this possibility can be excluded because the PCS activity at a constant free Cd(II) level approximately followed Michaelis–Menten-type kinetics at higher GSH concentrations. These observations indicated that the Cd(II)-G2 binding site was almost completely occupied by the complex and also suggested that there is no such GSH binding site on the enzyme, as discussed later. 4.2. Kinetic analysis of PC2 production mediated by rAtPCS1 at a constant free Cd(II) level The acylation of the enzyme by a γ-glutamylcysteinyl group occurs in a Cd(II)-independent manner [15], which leads to substitutedenzyme mechanisms as a PC synthesis mechanism. In general, using the maximal reaction rate, Vmax, and the enzyme reaction rate, v, a substituted-enzyme mechanism can be expressed as: v=
Fig. 4. PCS activity of rAtPCS1 as a function of GSH concentration at a constant free Cd (II) level of 1.22 × 10− 10 M (A). The simulated PCS activity indicated by a solid curve is based on a substituted-enzyme mechanism, in which the Michaelis constants for GSH and Cd(II)-G2 used were 18.0 mM and 5.14 μM, respectively. A Hanes plot, [G]t/activity versus [G]t, represents a linear relationship between [G]t/activity and [G]t at GSH concentrations greater than 25 mM, whereas the plots deviate upward from the linear relationship (B). This observation indicated that the PC synthesis reaction can be approximated by Michaelis–Menten-type kinetics only for GSH concentrations greater than 25 mM, as can be deduced from Eq. (2).
total Cd(II) concentration, the maximal activity shifted to a higher GSH concentration (Figs. 1 and 2). These observations were consistent with the notion that rAtPCS1 possesses a Cd binding site and that the binding of Cd(II) to this site activates the enzyme. The increase in PCS activity up to the GSH concentration at which the maximal rate is attained is due to the increasing concentration of the substrate GSH. In contrast, the suppression of enzyme activity at GSH concentrations greater than that at which the maximal activity is attained is attributable to a reduction in the level of activated (Cd(II)-bound) rAtPCS1. An increase in the concentration of GSH would reduce the level of free Cd(II), resulting in a reduction in the amount of rAtPCS1 bound by Cd(II). A further confirmation of the hypothesis was made using a PCS activity assay system at a constant free Cd(II) level. For this purpose, PCS activity was determined at various concentrations of total GSH while the ratio of [Cd(II)]t to [G]2t was maintained constant. As shown in Fig. 4, PCS activity increased steadily with GSH concentration, in contrast to the assay system with a constant total Cd(II) concentration, in which the PCS activity declined at higher GSH concentrations. These results demonstrated that the reduced PCS activity at higher GSH concentrations in the PCS assay system at a constant total Cd(II)
Vmax ½A½B KmB ½A + KmA ½B + ½A½B
ð1Þ
where A and B denote substrates that bound first and second to the enzyme, respectively, and KmA and KmB represent the Michaelis' constants for substrates A and B, respectively [16]. From Eq. (1), it appears that if the same substances act as substrates (i.e., A = B), then the reaction will follow simple Michaelis–Menten-type kinetics. However, this does not appear to be the case for PCS-catalyzed PC synthesis, because the Hanes plot of [G]t/activity versus [G]t yielded a non-linear relationship at GSH concentrations below 25 mM (Fig. 4B). From these results, together with the observation that enzyme acylation occurred even in the absence of Cd(II) addition, it is likely that a GSH molecule and a Cd(II)-G2 complex react as substrates binding to rAtPCS1 first and second, respectively, as suggested by Vatamaniuk et al. [8]. As the levels of Cd(II)-G2 complexes can be approximated by the total Cd(II) concentrations in the assay solution (Supplementary 2), the use of Eq. (S7) and a conditional association constant of Cd(II)-G2 complexes, β ′2 (Supplementary 1), provides the PC2 production rate at a constant level of free Cd(II), which can be expressed as: v=
V′max ½G2t
KmB + KmA Gt + ½G2t β′2 ½CdðIIÞ
ð2Þ
where V′max is nominal, because the fraction of Cd(II)-bound rAtPCS1 is unknown. Direct curve fitting of the PCS activity data to Eq. (2) was unsuccessful probably due to the lower contribution of KmB to the reaction rates at higher GSH concentrations, as revealed by linear relation at GSH concentrations above 25 mM in the Hanes plot (Fig. 4B). For these reasons, KmA and V′max were determined using activity data at GSH concentrations of 25–50 mM on the assumption that KmB was zero, followed by optimization of KmB for the whole set of activity data using the determined
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
KmA and V′max values. The obtained parameters, V′max = 149 ± 9, KmA =18.0±3.2 mM, and KmB =5.14±1.22 μM, are consistent with the experimentally determined activity because these parameters successfully simulated the PCS activity, as indicated by the solid line in Fig. 4A. Furthermore, the Michaelis constants for GSH and Cd-G2 complexes are in fair agreement with those reported previously [8], which supports the suggestion that PC production proceeds by a substituted-enzyme mechanism in which GSH donates a γ-glutamylcysteinyl group to the rAtPCS1 molecule and then the Cd(II)-G2 complex accepts the γglutamylcysteinyl group in PC2 synthesis. Our finding that the PCS synthesis rate followed Michaelis– Menten-type kinetics at a constant free Cd(II) level indicated that in Eq. (2) the KmB/β′2[Cd(II)][G]2t value becomes negligible with respect to (KmA/[G]t + 1) at GSH concentrations greater than 25 mM. This means that in the GSH concentration range, the Cd(II)-G2 site of the enzyme was almost fully occupied by the complex due to the elevation of Cd(II)-G2 concentration (from 31 to 120 μM at GSH concentrations from 25 to 50 mM). Nonetheless, no apparent inhibition of PCS activity was observed, indicating that there is no inhibitory GSH binding site on the enzyme. 4.3. Kinetic analysis of PC2 synthesis mediated by rAtPCS1 at a constant total Cd(II) concentration It is apparent that Cd(II) activates rAtPCS1 by binding to a site on the enzyme. Given the binding constant of Cd(II) to rAtPCS1, KE1, the amount of active enzyme can be expressed by the following equation: ½CdðIIÞ−rAtPCS1 = KE1 ½CdðIIÞ½rAtPCS1:
ð3Þ
Taking into account that only rAtPCS1 bound to Cd(II) has PCS activity, the reaction rate can be derived as a one-site model, as follows: v= KmA
f V ½G 1 max t K mB ½Gt + 1+
ð4Þ
½CdðIIÞt
where f1 represents the fraction of enzyme in an active form, which is expressed by: f1 =
1
ð5Þ
½G2t β′ 1+ 2 ⋅ K E1 ½CdðIIÞt
Here, the unit of Vmax is μmol min− 1 mg− 1 protein, in which the enzyme is supposed to be 100% active. The parameters Vmax and KE1 were optimized for the GSH concentration-dependent PCS activity shown in Fig. 1, where the KmA and KmB values obtained were employed. As shown in Table 1, PCS activity at 5 μM total Cd(II) concentration yielded a Vmax of 193 ± 17 μmol min− 1 mg− 1 protein
Table 1 Estimation of the affinity constants of Cd(II) to rAtPCS1 determined by enzyme activity as a function of the total GSH concentration using a one-site model.a) Cd(II) (μM)
Range of GSH (mM)
Vmax (μmol min mg− 1 protein)
1 5 10 50 500
0–50 0–50 0–50 0–50 0–80
152 ± 22 193 ± 17 302 ± 25 255 ± 27 205 ± 16
a)
−1
9
KE1/10 (M
−1
269 ± 102 28.4 ± 5.1 15.9 ± 2.7 45.5 ± 26.0 3.45 × 1018c)
)
115
and a KE1 of (28.4 ± 5.1) × 109 M− 1. The Vmax and KE1 values estimated from the activity determined at 10 μM total Cd(II) are fairly consistent with those obtained at 5 μM Cd(II), although a difference was apparent; the Vmax obtained at 10 μM total Cd(II) was 56% higher than that at 5 μM Cd(II) and the KE1 obtained at 10 μM total Cd(II) was 44% lower than that obtained at 5 μM. In both cases, the parameter obtained consistently simulated the PCS activity determined experimentally, as shown by the solid lines (b and c) in Fig. 1 for the total Cd (II) concentrations of 5 and 10 μM, respectively. In contrast, the PCS activity at 1 μM total Cd(II) yielded a much higher KE1 value than that obtained using 5 or 10 μM total Cd(II), with a correlation coefficient less than that at 5 or 10 μM total Cd(II) (Table 1). This may be a result of the poor data quality obtained at 1 μM total Cd(II), in which all of the activity plots except that of the origin revealed a decreasing profile; consequently, simulation of the activity using the estimated Vmax and KE1 did not adequately correlate with the experimentally determined activity, as shown by the solid line (a) in Fig. 1. Vmax and KE1 were optimized for the PCS activity of the assay solutions containing 50 μM and 500 μM total Cd(II). On the whole, the parameters obtained at 50 μM total Cd(II) were consistent with those obtained at 5 or 10 μM, although KE1 was 60% greater than that obtained at 5 μM total Cd(II) (Table 1). On the other hand, the optimization of the parameters at 500 μM total Cd(II) was unsuccessful due to the extraordinarily high KE1 value (Table 1). The Vmax and KE1 values obtained for PCS activity at higher levels of total Cd(II) (50 and 500 μM) yielded a lower correlation coefficient and a larger standard error in the KE1 value than those estimated from the activity determined at 5 or 10 μM total Cd(II). These differences may stem from the sigmoidal profile of the activity at lower concentrations of GSH in the presence of 50 or 500 μM total Cd(II) (Fig. 2). 4.4. Inhibitory second Cd(II) binding site of rAtPCS1 As shown in Fig. 3, the PC synthesis activity reached a maximum with respect to total Cd(II) concentration when the total GSH concentration was kept constant, and the maxima shifted toward higher Cd(II) concentrations with higher GSH concentrations. The optimum total Cd(II) concentration has also been demonstrated for AtPCS1 and PCS from soybean [7]. These findings suggest that rAtPCS1 possesses a second Cd(II) binding site with weaker affinity than the first site, and that the binding of Cd(II) to this site induces an inhibitory effect on PC synthesis. Given the binding constant of Cd(II) to Cd(II)-rAtPCS1 at the second site, KE2, the concentration of enzyme with two Cd(II) binding sites occupied by Cd(II) (Cd(II)2-rAtPCS1) can be expressed by the following equation: 2 CdðIIÞ2 −rAtPCS1 = KE1 KE2 ½CdðIIÞ ½rAtPCS1
ð6Þ
Assuming that rAtPCS1 with two Cd(II) binding sites fully occupied by Cd(II) ions completely loses its activity, the following equation of a two-site model can be obtained: v= KmA
f V ½G 2 max t K mB ½Gt + 1+
ð7Þ
½CdðIIÞt
2b)
R
0.681 0.926 0.938 0.870 0.771
Figures represent estimated value ± standard error. R represents the correlation coefficient between experimentally determined activity and that estimated using the one-site model. c) Standard error exceeded the limit that can be processed in the system.
where f2 represents the fraction of activated enzyme according to the following equation: f2 =
1 β0
½G2t ½CdðIIÞt K ⋅ + E20 ⋅ 1+ 2 K E1 ½CdðIIÞt ½Gt β2
ð8Þ
2
b)
The binding constants KE1 and KE2 were optimized to fit PCS activity according to Eq. (7) as a function of total Cd(II) concentration
116
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
Table 2 Estimation of the affinity constants of Cd(II) to rAtPCS1 determined under conditions with various concentrations of total Cd(II) with a constant total GSH concentration.a) GSH Range of Vmax (μmol min− 1 KE1/109 (M− 1) KE2/109 (M− 1) R2b) (mM) Cd(II) (μM) mg− 1 protein) 5 20
0–100 0–500
244 ± 17 194 ± 14
23.4 ± 9.8 25.3 ± 8.6
0.557 ± 0.081 0.403 ± 0.085
0.986 0.968
a)
Figures represent estimated value ± standard error. R represents the correlation coefficient between experimentally determined activity and that estimated using the two-site model. b)
(Table 2). Simulations of PCS activity using the Vmax, KE1, and KE2 values thus obtained for PCS activity at 5 and 20 mM GSH were in accordance with the experimentally obtained PCS activity in both cases, as shown by the solid lines in Fig. 3A and B. Furthermore, the parameters estimated from PCS activity at 5 mM GSH were consistent with those obtained at 20 mM GSH: the Vmax, KE1, and KE2 estimated from the activity determined at 5 mM GSH were 125%, 92.5%, and 138% of those at 20 mM GSH, respectively. These results demonstrated unequivocally that, in addition to the first Cd(II) binding site that is essential for activation of the enzyme, rAtPCS1 possesses a second Cd (II) binding site that is responsible for inactivation of the enzyme. A possible alternative interpretation for the reduced PCS activity in the presence of higher Cd(II) concentrations is partial denaturation of the enzyme. However, this contribution seemed to be excluded as the level of free Cd(II) was still low even at the highest total Cd(II) concentration in Fig. 3A, which was calculated to be 49 nM. A sigmoidal increase in activity was observed at GSH concentrations of 0–20 mM for the assay containing 50 and 500 μM total Cd(II) (Fig. 2). This is particularly true for the PCS assay performed in the presence of 500 μM total Cd(II). It is likely that an increase in free Cd (II) level at lower GSH concentrations is responsible for this phenomenon, because under such conditions, high levels of free Cd (II) may be present in the assay solution, and Cd(II) may occupy the second binding site of rAtPCS1, suppressing the activity. This type of inhibition of PCS activity is anticipated to be more severe at lower concentrations of total GSH. For these situations, Eq. (7) was employed again to follow the PCS activity determined as a function of total GSH concentration. Simulations of PCS activity using the optimized parameters, which are shown in Table 3, were consistent with the experimentally determined PCS activity in the presence of 5 and 20 mM GSH, as shown by the solid lines in Fig. 3A and B. An improvement in the correlation coefficient between simulated and experimentally determined PCS activity was apparent when the twosite model was employed. This also supports the suggestion that rAtPCS1 possesses two Cd(II) binding sites per molecule.
4.5. Affinity constants of rAtPCS1 to Cd(II) ions It has been demonstrated unequivocally that rAtPCS1 possesses two Cd(II) binding sites, although the affinity constants varied slightly
Table 3 Estimation of the affinity constants of Cd(II) to rAtPCS1 determined by the enzyme activity as a function of the total GSH concentration using a two-site model.a) Cd(II) (μM)
Range of GSH (mM)
Vmax (μmol min− 1 mg− 1 protein)
KE1/109 (M− 1)
KE2/109 (M− 1)
R2b)
50 500
0–50 0–80
399 ± 62 472 ± 94
14.2 ± 4.6 4.55 ± 1.88
0.961 ± 0.442 0.924 ± 0.333
0.971 0.984
a)
Figures represent estimated value ± standard error. R represents the correlation coefficient between experimentally determined activity and that estimated using the two-site model. b)
between determinations. Among the Vmax and KE1 values obtained by applying the one-site model, the parameters estimated from the PCS activity determined at 5 and 10 μM total Cd(II) are likely reasonable estimates (Table 1). The parameters obtained at 1 μM Cd(II) seemed to be erroneous due to poor data quality. The parameters obtained at 50 and 500 μM Cd(II) also seemed to be incomplete, because the plots exhibited sigmoidal profiles at lower GSH concentrations. The parameters obtained using the two-site model are likely to be consistent, although the KE1 at 500 μM total Cd(II) represented approximately 20% of that obtained from the data sets at 5 or 20 mM total GSH with varying total Cd(II) concentration (Table 2 versus Table 3). Averaging these values amounted to Vmax = 301 ± 48 μmol min− 1 mg− 1 protein, KE1 = (1.86 ± 0.61) × 1010 M− 1, and KE2 = (7.11 ± 2.83) × 108 M− 1, where the parameters obtained using the one-site model with PCS activity at 1, 50, and 500 μM total Cd(II) were omitted. The KE2 value may be greater than the value presented in this study, because an assumption was made in the optimization that PCS activity is completely lost when the second Cd(II) binding site of rAtPCS1 is occupied by Cd(II). However, the observation that the PCS activity of rAtPCS1 determined at 500 μM Cd(II) represented only 8% of that at 4 μM Cd(II) (Fig. 3A) indicates that the enzyme bound by two Cd(II) ions seemed to have very limited PCS activity. This was supported by the optimization practice, in which consistent parameters can be obtained only when the PCS activity of the enzyme bound by two Cd(II) ions supposedly had less than 3% enzyme activity. The optimization of PC activity at a constant free Cd(II) concentration gave a nominal Vmax value (V′max). Using a calculated free Cd(II) level of 1.22 × 10− 10 M, the association constant of Cd(II) to rAtPCS1 yielded fractional concentrations of rAtPCS1, Cd(II)-rAtPCS1, and Cd(II)2-rAtPCS1 of 28.8%, 65.5%, and 5.7%, respectively. From these values, Vmax was calculated to be 227 μmol min− 1 mg− 1 protein, in reasonable agreement with the averaged value.
4.6. Comparison to other models A kinetic analysis of PC synthesis catalyzed by AtPCS1-FLAG indicated a substituted-enzyme mechanism where GSH and Cd(II)-G2 play roles as co-substrates, with the enzyme already in an active form without Cd(II) binding [8]. Although supportive of a substitutedenzyme mechanism, our results demonstrated that the enzyme requires Cd(II) binding for activation. According to the model proposed by Vatamaniuk et al., the enzyme is active without bound Cd(II), as indicated by the much lower level of estimated free Cd(II) compared to the dissociation constant of Cd(II) to AtPCS1-FLAG [8]. In their study, equilibrium dialysis was performed to estimate the dissociation constant of Cd(II) to AtPCS1-FLAG, which demonstrated that the protein possessed seven Cd(II) binding sites with a dissociation constant of 5.6 μM. The free Cd(II) level was calculated to be 6.638 × 10− 13 M in the reaction medium containing total Cd(II) and GSH at concentrations of 25 μM and 3.3 mM, respectively, buffered at pH 8.0, using the computer program SOLCON. However, determination of the dissociation constant was performed by fitting a single hyperbolic curve as a function of Cd(II) concentration, although the protein apparently had seven Cd(II) binding sites [8]. In this context, this value represents an overall average dissociation constant and one or more of the sites may exhibit much lower dissociation constants. Peptide scanning analyses for TaPCS1 and SpPCS indicated that Cd(II) can still bind these PCSs in the presence of 1 mM GSH [9]. Furthermore, the stability constants employed by Vatamaniuk et al. [8] were so high that the estimated free Cd(II) level was three orders of magnitude less than those calculated with the constants from other sources [17]. From the conditional stability constants obtained in the current study, the free
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
Cd(II) level was demonstrated to be 5.6 × 10− 9 M. Therefore, it can be concluded that Cd(II) binds rAtPCS1 at two sites of the enzyme. 4.7. Potential physiological relevance of the Cd(II) binding sites of rAtPCS1 As suggested previously [5], the first Cd(II) binding site on rAtPCS1 may play a role in the regulation of PC synthesis in vivo. The enzyme is activated by the intrusion of Cd(II) ions in cells through binding to the site and is deactivated by removal of the ion from the activated enzyme through complexation of the product PCs with Cd(II). In contrast, the physiological role of the second site is obscure. Oxidative stress is a symptom induced by Cd(II) [18]. At a high dose of Cd(II), cells may attempt to cope with the oxidative stress, where GSH is an indispensable component of the defense system through the ascorbate-GSH cycle [19], rather than decrease the activity of Cd(II). As long as the synthesis of PCs is prompt, however, free Cd(II) may not reach a level that would allow the ion to bind to the second site.
Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10.1016/j.jinorgbio.2010.09.011.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
Abbreviations AtPCS1 Arabidopsis thaliana PCS1 AtPCS1-FLAG AtPCS1 with a C-terminal FLAG tag BCS bathocuproinedisulfonate Cd(II)-G2 1:2 Cd(II) GSH complex [Cd(II)]t total concentration of Cd(II) CHES N-cyclohexyl-2-aminoethanesulfonic acid desglycyl PC2 PC2 devoid of the C-terminal glycine; G GSH with various ionized states total concentration of G [G]t, GSH glutathione HEPES 2-[4-(2-hydroxyl)-1-piperazinyl]ethanesulfonic acid rAtPCS1 recombinant AtPCS1 TFA trifluoroacetic acid PC phytochelatin phytochelatin with n γ-Glu-Cys units PCn PCS phytochelatin synthase
117
[12]
[13] [14] [15] [16] [17] [18] [19]
N. Kondo, K. Imai, M. Isobe, T. Goto, Tetrahedron Lett. 25 (1984) 3869–3872. E. Grill, E.-L. Winnacker, M.H. Zenk, Science 230 (1985) 674–676. M.H. Zenk, Gene 179 (1996) 21–30. S. Clemens, D. Peršoh, Plant Sci. 177 (2009) 266–271. E. Grill, S. Löffler, E.-L. Winnacker, M.H. Zenk, Proc. Natl Acad. Sci. USA 86 (1989) 6838–6842. S. Löffler, A. Hochberger, E. Grill, E.-L. Winnacker, M.H. Zenk, FEBS Lett. 258 (1989) 42–46. M. Oven, J.F. Page, M.H. Zenk, T.M. Kutchan, J. Biol. Chem. 277 (2002) 4747–4754. O.K. Vatamaniuk, S. Mari, Y.-P. Lu, P.A. Rea, J. Biol. Chem. 275 (2000) 31451–31459. T. Maier, C. Yu, G. Küllertz, S. Clemens, Planta 218 (2003) 300–308. Y. Osaki, T. Shirabe, H. Nakanishi, T. Wakagi, E. Yoshimura, Metallomics 1 (2009) 353–358. M. Seki, P. Carninci, Y. Nishiyama, Y. Hayashizaki, K. Shinozaki, Plant J. 15 (1998) 707–720. M. Seki, M. Narusaka, A. Kamiya, J. Ishida, M. Satou, T. Sakurai, M. Nakajima, A. Enju, K. Akiyama, Y. Oono, M. Muramatsu, Y. Hayashizaki, J. Kawai, P. Carninci, M. Itoh, Y. Ishii, T. Arakawa, K. Shibata, A. Shinagawa, K. Shinozaki, Science 296 (2002) 141–145. S. Ogawa, R. Ichiki, M. Abo, E. Yoshimura, Anal. Chem. 81 (2009) 9199–9200. S. Ogawa, T. Yoshidomi, T. Shirabe, E. Yoshimura, J. Inorg. Biochem. 104 (2010) 442–445. O.K. Vatamaniuk, S. Mari, A. Lang, S. Chalasani, L.O. Demkiv, P.A. Rea, J. Biol. Chem. 279 (2004) 22449–22460. H. Bisswanger, Enzyme Kinetics, Second, Revised and Updated Edition, WILEYVCH Verlag GmbH & Co., Weinheim, Germany, 2008. S. Ogawa, E. Yoshimura, Anal. Biochem. 402 (2010) 200–202. A.S. Molina, C. Nievas, M.V.P. Chaca, F. Garibotto, U. González, S.V. Marsá, C. Luna, M.S. Giménez, F. Zirulnik, Plant Growth Regul. 56 (2008) 285–295. L.A. del Río, L.M. Sandalio, F.J. Corpas, J.M. Palma, J.B. Barroso, Plant Physiol. 141 (2006) 330–335.
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j i n o r g b i o
Cadmium(II)-stimulated enzyme activation of Arabidopsis thaliana phytochelatin synthase 1 Shinya Ogawa, Takahiro Yoshidomi, Etsuro Yoshimura ⁎ Department of Applied Biological Chemistry, School of Agriculture and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo, Tokyo 113-8657, Japan
a r t i c l e
i n f o
Article history: Received 3 August 2010 Received in revised form 25 September 2010 Accepted 27 September 2010 Available online 23 November 2010 Keywords: Arabidopsis thaliana Cadmium Phytochelatin Glutathion Phytochelatin synthase
a b s t r a c t Phytochelatin (PC), a class of heavy metal-binding peptides, is synthesized from the tripeptide glutathione (GSH) and/or previously synthesized PC in a reaction mediated by PC synthase (PCS). In the present study, the PC production rate catalyzed by recombinant Arabidopsis PCS1 (rAtPCS1) in the presence of a constant free Cd (II) level increased steadily and the kinetic parameters were approximated using a substituted-enzyme mechanism in which GSH and bis(glutathionato)cadmium acted as co-substrates. In contrast, the PC production rate as a function of GSH concentration at a constant total Cd(II) concentration reached a maximum, which shifted toward higher GSH concentrations as the concentration of Cd(II) was increased. These observations are consistent with the suggestion that rAtPCS1 possesses a Cd(II) binding site where Cd (II) binds to activate the enzyme. The affinity constant, optimized using a one-site mathematical model, successfully simulated the experimental data for the assay system using lower concentrations of Cd(II) (5 or 10 μM) but not for the assay using higher concentrations (50 or 500 μM), where a sigmoidal increase in PCS activity was evident. Furthermore, the PCS activity determined at a constant GSH concentration as a function of Cd(II) concentration also reached a maximum. These findings demonstrate that rAtPCS1 also possesses a second Cd(II) binding site where Cd(II) binds to induce an inhibitory effect. A two-site mathematical model was applied successfully to account for the observed phenomena, supporting the suggestion that rAtPCS1 possesses two Cd(II) binding sites. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Phytochelatins (PCs) are a family of peptides with the general structure (γ-Glu-Cys)n-Gly (PCn), where n ≥ 2. The peptides, designated as cadystin, were first identified in the fission yeast Schizosaccharomyces pombe grown in Cd(II)-containing medium [1]. Shortly after the initial identification, the peptides were detected in cells of the higher plant Rauvolfia serpentina in suspension culture [2]. To date, PCs have been identified in higher plants, algae, and some fungi exposed to toxic levels of heavy metal ions [3]. PCs play a role in detoxification by sequestering heavy metal ions, although other roles in Zn(II) homeostasis and the long-distance transport of Zn(II) have been suggested [4]. The synthesis of PC is catalyzed by phytochelatin synthase (PCS) using the tripeptide glutathione (γ-Glu-Cys-Gly; GSH) and/or previously synthesized PC as a substrate [5]. PCS, a γ-glutamylcysteinyl transpeptidase (EC 2.3.2.15), mediates a dipeptidyl transfer reaction in the presence of heavy metals, leading to the formation of PCn with a greater number of γ-glutamylcysteinyl units, which has increased affinity for heavy metals and the capacity to sequester them
⁎ Corresponding author. Tel.: + 81 3 5841 5153; fax: +81 3 5841 8027. E-mail address: [email protected] (E. Yoshimura). 0162-0134/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2010.09.011
more effectively [6]. Metals and metalloids belonging to groups 11–15 in the fourth, fifth, and sixth periods of the periodic table can activate PCS-mediated PC synthesis [7]. As PCS is expressed constitutively in cells, the metal ion dependency of PC synthesis represents a selfterminating reaction in which the activator metal ions are captured by the reaction product. The enzymatic mechanism of action for PCS in relation to activation by metal ions has yet to be resolved. Adding Cd(II) to a solution containing GSH and PCS from Silene cucubalus initiates PC synthesis, and the supplementing the solution with Cd(II) chelators, such as uncomplexed PCn mixtures or ethylenediaminetetraacetic acid, immediately terminates PC synthesis, suggesting that PCS is activated by metal ion binding [5]. Equilibrium dialysis of AtPCS1FLAG indicated seven Cd(II) binding sites in the protein with a dissociation constant of 0.54 μM [8]. However, under standard PCS assay conditions, in which GSH is present at a much higher level than Cd(II), the level of free Cd(II) ions is extremely low; using the reported equilibrium binding constants for Cd(II) and GSH, the free Cd(II) concentration can be predicted, which has been calculated as 6.638 × 10− 13 M when Cd(II) and GSH are present at total concentrations of 25 μM and 3.3 mM, respectively, at pH 8.0 [8]. Thus, the concentration of free Cd(II) ions is too low to account for the binding of Cd(II) ions to AtPCS1-FLAG [8]. Kinetic analysis of PC synthesis mediated by AtPCS1-FLAG has indicated a reaction process consistent
112
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
with a bisubstrate-substituted mechanism, whereby a GSH molecule and a Cd(II)-G2 complex act as co-substrates and the enzyme is already in an active form without binding to Cd(II) [8]. Furthermore, it has been inferred from the facility of AtPCS1 for the synthesis of Salkylated PCs from S-alkylated GSH in the absence of added heavy metal ions that GSH derivatives with thiol groups blocked either through heavy metal thiolate formation or S-alkylation are solely contingent on PC synthesis [8]. However, peptide scanning of PCS from wheat Triticum aestivum L. (TaPCS1) and the fission yeast S. pombe (SpPCS) revealed that Cd(II) binds to the proteins in the presence of GSH at a concentration of 1 mM [9]. In addition, Cd(II)dependent stabilization of PCS has been reported for the enzyme produced by the primitive red alga Cyanidioschyzon merolae, supporting the suggestion that Cd(II) binds to the enzyme [10]. Therefore, no definitive conclusion has been reached regarding the metal activation mechanism of PC synthesis. 2. Materials and methods 2.1. Heterologous expression and purification of N-terminal hexahistidinyl-tagged Arabidopsis thaliana PCS1
acid (TFA). The mixture was diluted with 0.1% TFA in which 0.5 μM desglycyl PC2 was added as an internal standard. PC concentrations were determined by HPLC using detection based on the dequenching of Cu(I)-BCS complexes [13,14]. The peptides were separated on an octadecylsilane column (4.6 mm ϕ × 150 mm, TSKgel ODS-80™; Tosoh, Tokyo, Japan) with a mobile phase consisting of 5% acetonitrile/0.1% TFA using a pump (PU-1580; Jasco, Tokyo, Japan) at a flow rate of 1 mL min− 1. Using another pump (PU-980; Jasco) at a flow rate of 1 mL min− 1, the eluent was merged with a post-column solution containing 50 mM CHES-NaOH (pH 10.0), 0.2 μM CuSO4, 0.5 μM BCS, and 5 μM ascorbic acid. The merged solution was passed through a mixing coil, and the fluorescence intensity at excitation and emission wavelengths of 280 and 395 nm, respectively, was monitored using an FP-920 fluorescence detector (Jasco). During HPLC analyses, the post-column solution was bubbled continuously with a stream of He. The integrated peak area was used to quantify the PC level after calibration with chemically synthesized PC (Hayashi Kasei, Osaka, Japan). PCS activity was expressed as the amount of PC2 synthesized by 1 mg of protein per minute, because the synthesis of PCn with n greater than 2 was negligible. 2.3. PCS activity assays in the presence of a constant level of free Cd(II)
We used a recombinant Arabidopsis thaliana PCS1 tagged with hexahistidine at its N-terminus (rAtPCS1). The AtPCS1 DNA sequence was amplified using PrimeSTAR HS DNA polymerase (TakaraBio, Shiga, Japan) and a plasmid containing the AtPCS1 gene (Riken BioResource Center, Saitama, Japan) [11,12] was used as a template. The forward and reverse PCR primers were 5′-GAGAGCTAGCATGGCTATGGCGAGTTTATATCG-3′ (NheI site underlined) and 5′-GAGAGAATTCCTAATAGGCAGGAGCAGCGAGATC-3′ (EcoRI site underlined), respectively. The resultant DNA fragment was transferred to the pET28b expression vector (Novagen, Madison, WI) as an in-frame fusion with a plasmid-encoded hexahistidine tag sequence, and the plasmid was used to transform Escherichia coli XL-1 Blue (Stratagene, La Jolla, CA). Following the confirmation of successful cloning by sequencing with T7 primers, the E. coli BL21 Rosetta (DE3) pLysS strain (Novagen) was transformed with the construct. E. coli cells carrying rAtPCS1 were pre-cultured in 200 mL of LuriaBertani broth at 25 °C. After reaching an optical density at 600 nm of 0.1–0.8, isopropyl 1-thio-β-D-galactoside was added to 40 μM, and the culture was incubated for a further 6 h to express the recombinant protein. The cells were collected by centrifugation at 8000× g for 15 min and then suspended in an extraction solution (25 mM Tris– HCl, pH 8.0, 400 mM NaCl, 10% glycerol, 1 mM 2-mercaptoethanol, 0.1% Tween 20, and 1 mM phenylmethylsulfonyl fluoride). The cells were disrupted by sonication, and the supernatant was obtained by centrifugation at 13,000× g for 10 min at 4 °C. The recombinant protein was purified on a Ni2+ affinity column by fast-performance liquid chromatography (Pharmacia LKB, Uppsala, Sweden). The supernatant (20 mL) was applied to a 1-mL HisTrap™ HP column (GE Healthcare Life Sciences, Piscataway, NJ) equilibrated with binding buffer (20 mM HEPES-NaOH, pH 7.4, 500 mM NaCl, 1 mM 2-mercaptoethanol, and 40 mM imidazole), the column was washed with 10 mL of binding buffer, and then the sample was eluted with a linear gradient of 40–250 mM imidazole in binding buffer over 20 min. The rAtPCS1 eluted at 150 mM imidazole. Analysis of the eluate by SDS-PAGE revealed a single band with a molecular mass of 56.5 kDa, confirming the homogeneity of the preparation.
PCS activity was determined at various GSH concentrations while keeping the free Cd(II) concentration constant. For this purpose, the total Cd(II) and GSH concentrations were changed concomitantly from the initial values of 5 μM and 10 mM, respectively, in such a way that the ratio of [Cd(II)]t to [G]2t was maintained at 5 × 10− 2 M− 1. The concentration of free Cd(II) ions was calculated to be 1.22 × 10− 10 M under these conditions (Supplementary 1 and 2). 2.4. Determination of parameters Non-linear least squares analysis was performed using the Origin 8.1 program (OriginLab, Northampton, MA). 3. Results 3.1. PC synthesis activity of rAtPCS1 as a function of GSH concentration The PCS activity of rAtPCS1 as a function of the concentration of GSH supplemented in the assay solution is shown in Fig. 1. In the presence of 10 μM Cd(II) (Fig. 1, closed squares), activity increased and reached the maximum rate with an increase in GSH concentration
2.2. PCS activity assays The reaction mixture (100 μL) contained 100 ng/mL rAtPCS1, 10 mM 2-mercaptoethanol, 100 ng/mL bovine serum albumin, and various concentrations of GSH and CdSO4 in 200 mM HEPES-NaOH buffer (pH 8.0). The mixture was incubated at 35 °C for 15 min, and the reaction was terminated by adding 25 μL of 10% trifluoroacetic
Fig. 1. Effects of GSH concentration on the PCS activity of rAtPCS1 in assay solution containing total Cd(II) concentrations of 1 (closed triangles), 5 (closed circles), and 10 μM (closed squares). The solid lines (a, b, and c) represent simulated activity using the one-site model for the assay solutions containing 1, 5, and 10 μM Cd(II), respectively, where the Vmax and KE1 values listed in Table 1 were used.
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
113
up to 15 mM, and then decreased with further increases in GSH concentration. A similar trend was observed for the enzyme assay system in which 1 or 5 μM Cd(II) was added (Fig. 1, closed circles and triangles, respectively). It was also evident that the maximal activity occurred at higher GSH concentrations with an increase in Cd(II) concentration. Thus, the maximal activity was attained at GSH concentrations around 10 and 15 mM when the enzyme was assayed in the presence of 5 and 10 μM total Cd(II), respectively. Although the GSH concentration at which the maximal activity attained may be obscure in the enzyme assay system with 1 μM total Cd(II), it occurred at a concentration less than or equal to 5 mM. These phenomena still held for the assay solution containing higher levels of total Cd(II). As shown in Fig. 2, maximal PCS activity occurred at 30 and 60 mM GSH for the assay solution containing 50 μM (closed squares) and 500 μM Cd(II) (closed circles), respectively. In addition, a sigmoidal increase in activity was observed at GSH concentrations below 20 mM for the assay system containing 50 and 500 μM Cd(II), a prominent feature for the system at high levels of Cd(II). 3.2. PC synthesis activity of rAtPCS1 as a function of the total Cd(II) concentration The PCS activity of rAtPCS1 as a function of Cd(II) added to the assay solution is illustrated in Fig. 3. At a constant GSH concentration of 5 mM, PCS activity increased with increasing Cd(II) concentration up to 4 μM, and then decreased sharply (Fig. 3A). A similar activity profile was obtained for the solution containing 20 mM GSH, where the maximal activity was attained at 60 μM Cd(II) (Fig. 3B). These results indicated that rAtPCS1 has an optimum Cd(II) concentration for PCS activity, which shifts toward a higher Cd(II) concentration in the presence of higher levels of GSH. 3.3. PC synthesis activity of rAtPCS1 as a function of GSH concentration at a constant free Cd(II) concentration PCS activity at a constant free Cd(II) concentration of 1.22 × 10− 10 M was determined as a function of total GSH concentration. The activity increased steadily with total GSH concentration up to 50 mM (Fig. 4A). A Hanes plot of [G]t/activity versus [G]t showed a straight line at GSH concentrations greater than 25 mM, whereas the plot deviated upward at GSH concentrations below 25 mM (Fig. 4B). These findings indicated that the PCS synthesis reaction can display
Fig. 3. Effects of total Cd(II) concentration on the PCS activity of rAtPCS1 in the presence of 5 mM (A) and 20 mM (B) GSH. The inset in (A) is an expansion of the figure at Cd(II) concentrations from 0 to 40 μM. The solid lines represent the simulated activity, where the Vmax, KE1, and KE2 values listed in Table 2 were used.
Michaelis–Menten-type kinetics only at GSH concentrations above 25 mM. 4. Discussion 4.1. PCS activity of rAtPCS1 as a function of GSH
Fig. 2. Effects of GSH concentration on the PCS activity of rAtPCS1 in the assay solution containing total Cd(II) concentrations of 50 (closed squares) and 500 μM (closed circles). The solid lines (a and b) represent simulated activity using the two-site model for the assay solutions containing 50 and 500 μM Cd(II), respectively, where the Vmax, KE1, and KE2 values listed in Table 3 were used.
Although a definitive conclusion has not been reached regarding the mechanisms of metal ion activation of PC synthesis, enzyme acylation by a γ-glutamylcysteinyl group has been clearly demonstrated [15]. The thiol group of Cys-56 in AtPCS1 has been identified as an acylation site, which undergoes a reaction with GSH in a Cd(II)independent manner [15]. The γ-glutamylcysteinyl group on an acyl intermediate is transferred to an acceptor GSH or PCn − 1 molecule in PCn synthesis. Two distinct mechanisms have been proposed for the γ-glutamylcysteinyl dipeptide transfer. In the case of PC2 synthesis, for example, the γ-glutamylcysteinyl group on PCS is transferred to a Cd(II)-G2, in which the enzyme is already in an active form with no need for the direct binding of metals [8]. The alternative mechanism is that Cd(II) is required to bind PCS to activate it for the transfer of the dipeptide to an acceptor molecule such as GSH or PCn − 1 [5]. The PCS activity of rAtPCS1 determined at a constant total Cd(II) concentration demonstrated the maximum activity with respect to GSH concentration (Figs. 1 and 2). In addition, with an increase in the
114
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
concentration was attributable to a lower free Cd(II) level. These findings clearly demonstrated that AtPCS1 activation requires Cd(II) binding. A proposed substituted-enzyme mechanism [8] in which the enzyme is in an active form without being bound by Cd(II) and the GSH and Cd(II)-G2 molecules act as co-substrates is in conflict with the declining PCS activity at higher GSH concentrations at a constant total Cd(II) concentration, because an increase in GSH concentration at a constant total Cd(II) concentration will also elevate the free GSH levels and Cd-G2 levels. To make such a mechanism consistent with our observations, it would be necessary to hypothesize a GSH binding site on the enzyme that exerts an inhibitory effect. The steady increase in PCS activity with GSH concentration in the presence of a constant free Cd(II) level may be attributable to an increased concentration of Cd(II)-G2, the effect of which may surpass the inhibitory effect with regard to PCS activity, as the concentration of Cd(II)-G2 increased concomitant with that of GSH in the assay system (Fig. 4). However, this possibility can be excluded because the PCS activity at a constant free Cd(II) level approximately followed Michaelis–Menten-type kinetics at higher GSH concentrations. These observations indicated that the Cd(II)-G2 binding site was almost completely occupied by the complex and also suggested that there is no such GSH binding site on the enzyme, as discussed later. 4.2. Kinetic analysis of PC2 production mediated by rAtPCS1 at a constant free Cd(II) level The acylation of the enzyme by a γ-glutamylcysteinyl group occurs in a Cd(II)-independent manner [15], which leads to substitutedenzyme mechanisms as a PC synthesis mechanism. In general, using the maximal reaction rate, Vmax, and the enzyme reaction rate, v, a substituted-enzyme mechanism can be expressed as: v=
Fig. 4. PCS activity of rAtPCS1 as a function of GSH concentration at a constant free Cd (II) level of 1.22 × 10− 10 M (A). The simulated PCS activity indicated by a solid curve is based on a substituted-enzyme mechanism, in which the Michaelis constants for GSH and Cd(II)-G2 used were 18.0 mM and 5.14 μM, respectively. A Hanes plot, [G]t/activity versus [G]t, represents a linear relationship between [G]t/activity and [G]t at GSH concentrations greater than 25 mM, whereas the plots deviate upward from the linear relationship (B). This observation indicated that the PC synthesis reaction can be approximated by Michaelis–Menten-type kinetics only for GSH concentrations greater than 25 mM, as can be deduced from Eq. (2).
total Cd(II) concentration, the maximal activity shifted to a higher GSH concentration (Figs. 1 and 2). These observations were consistent with the notion that rAtPCS1 possesses a Cd binding site and that the binding of Cd(II) to this site activates the enzyme. The increase in PCS activity up to the GSH concentration at which the maximal rate is attained is due to the increasing concentration of the substrate GSH. In contrast, the suppression of enzyme activity at GSH concentrations greater than that at which the maximal activity is attained is attributable to a reduction in the level of activated (Cd(II)-bound) rAtPCS1. An increase in the concentration of GSH would reduce the level of free Cd(II), resulting in a reduction in the amount of rAtPCS1 bound by Cd(II). A further confirmation of the hypothesis was made using a PCS activity assay system at a constant free Cd(II) level. For this purpose, PCS activity was determined at various concentrations of total GSH while the ratio of [Cd(II)]t to [G]2t was maintained constant. As shown in Fig. 4, PCS activity increased steadily with GSH concentration, in contrast to the assay system with a constant total Cd(II) concentration, in which the PCS activity declined at higher GSH concentrations. These results demonstrated that the reduced PCS activity at higher GSH concentrations in the PCS assay system at a constant total Cd(II)
Vmax ½A½B KmB ½A + KmA ½B + ½A½B
ð1Þ
where A and B denote substrates that bound first and second to the enzyme, respectively, and KmA and KmB represent the Michaelis' constants for substrates A and B, respectively [16]. From Eq. (1), it appears that if the same substances act as substrates (i.e., A = B), then the reaction will follow simple Michaelis–Menten-type kinetics. However, this does not appear to be the case for PCS-catalyzed PC synthesis, because the Hanes plot of [G]t/activity versus [G]t yielded a non-linear relationship at GSH concentrations below 25 mM (Fig. 4B). From these results, together with the observation that enzyme acylation occurred even in the absence of Cd(II) addition, it is likely that a GSH molecule and a Cd(II)-G2 complex react as substrates binding to rAtPCS1 first and second, respectively, as suggested by Vatamaniuk et al. [8]. As the levels of Cd(II)-G2 complexes can be approximated by the total Cd(II) concentrations in the assay solution (Supplementary 2), the use of Eq. (S7) and a conditional association constant of Cd(II)-G2 complexes, β ′2 (Supplementary 1), provides the PC2 production rate at a constant level of free Cd(II), which can be expressed as: v=
V′max ½G2t
KmB + KmA Gt + ½G2t β′2 ½CdðIIÞ
ð2Þ
where V′max is nominal, because the fraction of Cd(II)-bound rAtPCS1 is unknown. Direct curve fitting of the PCS activity data to Eq. (2) was unsuccessful probably due to the lower contribution of KmB to the reaction rates at higher GSH concentrations, as revealed by linear relation at GSH concentrations above 25 mM in the Hanes plot (Fig. 4B). For these reasons, KmA and V′max were determined using activity data at GSH concentrations of 25–50 mM on the assumption that KmB was zero, followed by optimization of KmB for the whole set of activity data using the determined
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
KmA and V′max values. The obtained parameters, V′max = 149 ± 9, KmA =18.0±3.2 mM, and KmB =5.14±1.22 μM, are consistent with the experimentally determined activity because these parameters successfully simulated the PCS activity, as indicated by the solid line in Fig. 4A. Furthermore, the Michaelis constants for GSH and Cd-G2 complexes are in fair agreement with those reported previously [8], which supports the suggestion that PC production proceeds by a substituted-enzyme mechanism in which GSH donates a γ-glutamylcysteinyl group to the rAtPCS1 molecule and then the Cd(II)-G2 complex accepts the γglutamylcysteinyl group in PC2 synthesis. Our finding that the PCS synthesis rate followed Michaelis– Menten-type kinetics at a constant free Cd(II) level indicated that in Eq. (2) the KmB/β′2[Cd(II)][G]2t value becomes negligible with respect to (KmA/[G]t + 1) at GSH concentrations greater than 25 mM. This means that in the GSH concentration range, the Cd(II)-G2 site of the enzyme was almost fully occupied by the complex due to the elevation of Cd(II)-G2 concentration (from 31 to 120 μM at GSH concentrations from 25 to 50 mM). Nonetheless, no apparent inhibition of PCS activity was observed, indicating that there is no inhibitory GSH binding site on the enzyme. 4.3. Kinetic analysis of PC2 synthesis mediated by rAtPCS1 at a constant total Cd(II) concentration It is apparent that Cd(II) activates rAtPCS1 by binding to a site on the enzyme. Given the binding constant of Cd(II) to rAtPCS1, KE1, the amount of active enzyme can be expressed by the following equation: ½CdðIIÞ−rAtPCS1 = KE1 ½CdðIIÞ½rAtPCS1:
ð3Þ
Taking into account that only rAtPCS1 bound to Cd(II) has PCS activity, the reaction rate can be derived as a one-site model, as follows: v= KmA
f V ½G 1 max t K mB ½Gt + 1+
ð4Þ
½CdðIIÞt
where f1 represents the fraction of enzyme in an active form, which is expressed by: f1 =
1
ð5Þ
½G2t β′ 1+ 2 ⋅ K E1 ½CdðIIÞt
Here, the unit of Vmax is μmol min− 1 mg− 1 protein, in which the enzyme is supposed to be 100% active. The parameters Vmax and KE1 were optimized for the GSH concentration-dependent PCS activity shown in Fig. 1, where the KmA and KmB values obtained were employed. As shown in Table 1, PCS activity at 5 μM total Cd(II) concentration yielded a Vmax of 193 ± 17 μmol min− 1 mg− 1 protein
Table 1 Estimation of the affinity constants of Cd(II) to rAtPCS1 determined by enzyme activity as a function of the total GSH concentration using a one-site model.a) Cd(II) (μM)
Range of GSH (mM)
Vmax (μmol min mg− 1 protein)
1 5 10 50 500
0–50 0–50 0–50 0–50 0–80
152 ± 22 193 ± 17 302 ± 25 255 ± 27 205 ± 16
a)
−1
9
KE1/10 (M
−1
269 ± 102 28.4 ± 5.1 15.9 ± 2.7 45.5 ± 26.0 3.45 × 1018c)
)
115
and a KE1 of (28.4 ± 5.1) × 109 M− 1. The Vmax and KE1 values estimated from the activity determined at 10 μM total Cd(II) are fairly consistent with those obtained at 5 μM Cd(II), although a difference was apparent; the Vmax obtained at 10 μM total Cd(II) was 56% higher than that at 5 μM Cd(II) and the KE1 obtained at 10 μM total Cd(II) was 44% lower than that obtained at 5 μM. In both cases, the parameter obtained consistently simulated the PCS activity determined experimentally, as shown by the solid lines (b and c) in Fig. 1 for the total Cd (II) concentrations of 5 and 10 μM, respectively. In contrast, the PCS activity at 1 μM total Cd(II) yielded a much higher KE1 value than that obtained using 5 or 10 μM total Cd(II), with a correlation coefficient less than that at 5 or 10 μM total Cd(II) (Table 1). This may be a result of the poor data quality obtained at 1 μM total Cd(II), in which all of the activity plots except that of the origin revealed a decreasing profile; consequently, simulation of the activity using the estimated Vmax and KE1 did not adequately correlate with the experimentally determined activity, as shown by the solid line (a) in Fig. 1. Vmax and KE1 were optimized for the PCS activity of the assay solutions containing 50 μM and 500 μM total Cd(II). On the whole, the parameters obtained at 50 μM total Cd(II) were consistent with those obtained at 5 or 10 μM, although KE1 was 60% greater than that obtained at 5 μM total Cd(II) (Table 1). On the other hand, the optimization of the parameters at 500 μM total Cd(II) was unsuccessful due to the extraordinarily high KE1 value (Table 1). The Vmax and KE1 values obtained for PCS activity at higher levels of total Cd(II) (50 and 500 μM) yielded a lower correlation coefficient and a larger standard error in the KE1 value than those estimated from the activity determined at 5 or 10 μM total Cd(II). These differences may stem from the sigmoidal profile of the activity at lower concentrations of GSH in the presence of 50 or 500 μM total Cd(II) (Fig. 2). 4.4. Inhibitory second Cd(II) binding site of rAtPCS1 As shown in Fig. 3, the PC synthesis activity reached a maximum with respect to total Cd(II) concentration when the total GSH concentration was kept constant, and the maxima shifted toward higher Cd(II) concentrations with higher GSH concentrations. The optimum total Cd(II) concentration has also been demonstrated for AtPCS1 and PCS from soybean [7]. These findings suggest that rAtPCS1 possesses a second Cd(II) binding site with weaker affinity than the first site, and that the binding of Cd(II) to this site induces an inhibitory effect on PC synthesis. Given the binding constant of Cd(II) to Cd(II)-rAtPCS1 at the second site, KE2, the concentration of enzyme with two Cd(II) binding sites occupied by Cd(II) (Cd(II)2-rAtPCS1) can be expressed by the following equation: 2 CdðIIÞ2 −rAtPCS1 = KE1 KE2 ½CdðIIÞ ½rAtPCS1
ð6Þ
Assuming that rAtPCS1 with two Cd(II) binding sites fully occupied by Cd(II) ions completely loses its activity, the following equation of a two-site model can be obtained: v= KmA
f V ½G 2 max t K mB ½Gt + 1+
ð7Þ
½CdðIIÞt
2b)
R
0.681 0.926 0.938 0.870 0.771
Figures represent estimated value ± standard error. R represents the correlation coefficient between experimentally determined activity and that estimated using the one-site model. c) Standard error exceeded the limit that can be processed in the system.
where f2 represents the fraction of activated enzyme according to the following equation: f2 =
1 β0
½G2t ½CdðIIÞt K ⋅ + E20 ⋅ 1+ 2 K E1 ½CdðIIÞt ½Gt β2
ð8Þ
2
b)
The binding constants KE1 and KE2 were optimized to fit PCS activity according to Eq. (7) as a function of total Cd(II) concentration
116
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
Table 2 Estimation of the affinity constants of Cd(II) to rAtPCS1 determined under conditions with various concentrations of total Cd(II) with a constant total GSH concentration.a) GSH Range of Vmax (μmol min− 1 KE1/109 (M− 1) KE2/109 (M− 1) R2b) (mM) Cd(II) (μM) mg− 1 protein) 5 20
0–100 0–500
244 ± 17 194 ± 14
23.4 ± 9.8 25.3 ± 8.6
0.557 ± 0.081 0.403 ± 0.085
0.986 0.968
a)
Figures represent estimated value ± standard error. R represents the correlation coefficient between experimentally determined activity and that estimated using the two-site model. b)
(Table 2). Simulations of PCS activity using the Vmax, KE1, and KE2 values thus obtained for PCS activity at 5 and 20 mM GSH were in accordance with the experimentally obtained PCS activity in both cases, as shown by the solid lines in Fig. 3A and B. Furthermore, the parameters estimated from PCS activity at 5 mM GSH were consistent with those obtained at 20 mM GSH: the Vmax, KE1, and KE2 estimated from the activity determined at 5 mM GSH were 125%, 92.5%, and 138% of those at 20 mM GSH, respectively. These results demonstrated unequivocally that, in addition to the first Cd(II) binding site that is essential for activation of the enzyme, rAtPCS1 possesses a second Cd (II) binding site that is responsible for inactivation of the enzyme. A possible alternative interpretation for the reduced PCS activity in the presence of higher Cd(II) concentrations is partial denaturation of the enzyme. However, this contribution seemed to be excluded as the level of free Cd(II) was still low even at the highest total Cd(II) concentration in Fig. 3A, which was calculated to be 49 nM. A sigmoidal increase in activity was observed at GSH concentrations of 0–20 mM for the assay containing 50 and 500 μM total Cd(II) (Fig. 2). This is particularly true for the PCS assay performed in the presence of 500 μM total Cd(II). It is likely that an increase in free Cd (II) level at lower GSH concentrations is responsible for this phenomenon, because under such conditions, high levels of free Cd (II) may be present in the assay solution, and Cd(II) may occupy the second binding site of rAtPCS1, suppressing the activity. This type of inhibition of PCS activity is anticipated to be more severe at lower concentrations of total GSH. For these situations, Eq. (7) was employed again to follow the PCS activity determined as a function of total GSH concentration. Simulations of PCS activity using the optimized parameters, which are shown in Table 3, were consistent with the experimentally determined PCS activity in the presence of 5 and 20 mM GSH, as shown by the solid lines in Fig. 3A and B. An improvement in the correlation coefficient between simulated and experimentally determined PCS activity was apparent when the twosite model was employed. This also supports the suggestion that rAtPCS1 possesses two Cd(II) binding sites per molecule.
4.5. Affinity constants of rAtPCS1 to Cd(II) ions It has been demonstrated unequivocally that rAtPCS1 possesses two Cd(II) binding sites, although the affinity constants varied slightly
Table 3 Estimation of the affinity constants of Cd(II) to rAtPCS1 determined by the enzyme activity as a function of the total GSH concentration using a two-site model.a) Cd(II) (μM)
Range of GSH (mM)
Vmax (μmol min− 1 mg− 1 protein)
KE1/109 (M− 1)
KE2/109 (M− 1)
R2b)
50 500
0–50 0–80
399 ± 62 472 ± 94
14.2 ± 4.6 4.55 ± 1.88
0.961 ± 0.442 0.924 ± 0.333
0.971 0.984
a)
Figures represent estimated value ± standard error. R represents the correlation coefficient between experimentally determined activity and that estimated using the two-site model. b)
between determinations. Among the Vmax and KE1 values obtained by applying the one-site model, the parameters estimated from the PCS activity determined at 5 and 10 μM total Cd(II) are likely reasonable estimates (Table 1). The parameters obtained at 1 μM Cd(II) seemed to be erroneous due to poor data quality. The parameters obtained at 50 and 500 μM Cd(II) also seemed to be incomplete, because the plots exhibited sigmoidal profiles at lower GSH concentrations. The parameters obtained using the two-site model are likely to be consistent, although the KE1 at 500 μM total Cd(II) represented approximately 20% of that obtained from the data sets at 5 or 20 mM total GSH with varying total Cd(II) concentration (Table 2 versus Table 3). Averaging these values amounted to Vmax = 301 ± 48 μmol min− 1 mg− 1 protein, KE1 = (1.86 ± 0.61) × 1010 M− 1, and KE2 = (7.11 ± 2.83) × 108 M− 1, where the parameters obtained using the one-site model with PCS activity at 1, 50, and 500 μM total Cd(II) were omitted. The KE2 value may be greater than the value presented in this study, because an assumption was made in the optimization that PCS activity is completely lost when the second Cd(II) binding site of rAtPCS1 is occupied by Cd(II). However, the observation that the PCS activity of rAtPCS1 determined at 500 μM Cd(II) represented only 8% of that at 4 μM Cd(II) (Fig. 3A) indicates that the enzyme bound by two Cd(II) ions seemed to have very limited PCS activity. This was supported by the optimization practice, in which consistent parameters can be obtained only when the PCS activity of the enzyme bound by two Cd(II) ions supposedly had less than 3% enzyme activity. The optimization of PC activity at a constant free Cd(II) concentration gave a nominal Vmax value (V′max). Using a calculated free Cd(II) level of 1.22 × 10− 10 M, the association constant of Cd(II) to rAtPCS1 yielded fractional concentrations of rAtPCS1, Cd(II)-rAtPCS1, and Cd(II)2-rAtPCS1 of 28.8%, 65.5%, and 5.7%, respectively. From these values, Vmax was calculated to be 227 μmol min− 1 mg− 1 protein, in reasonable agreement with the averaged value.
4.6. Comparison to other models A kinetic analysis of PC synthesis catalyzed by AtPCS1-FLAG indicated a substituted-enzyme mechanism where GSH and Cd(II)-G2 play roles as co-substrates, with the enzyme already in an active form without Cd(II) binding [8]. Although supportive of a substitutedenzyme mechanism, our results demonstrated that the enzyme requires Cd(II) binding for activation. According to the model proposed by Vatamaniuk et al., the enzyme is active without bound Cd(II), as indicated by the much lower level of estimated free Cd(II) compared to the dissociation constant of Cd(II) to AtPCS1-FLAG [8]. In their study, equilibrium dialysis was performed to estimate the dissociation constant of Cd(II) to AtPCS1-FLAG, which demonstrated that the protein possessed seven Cd(II) binding sites with a dissociation constant of 5.6 μM. The free Cd(II) level was calculated to be 6.638 × 10− 13 M in the reaction medium containing total Cd(II) and GSH at concentrations of 25 μM and 3.3 mM, respectively, buffered at pH 8.0, using the computer program SOLCON. However, determination of the dissociation constant was performed by fitting a single hyperbolic curve as a function of Cd(II) concentration, although the protein apparently had seven Cd(II) binding sites [8]. In this context, this value represents an overall average dissociation constant and one or more of the sites may exhibit much lower dissociation constants. Peptide scanning analyses for TaPCS1 and SpPCS indicated that Cd(II) can still bind these PCSs in the presence of 1 mM GSH [9]. Furthermore, the stability constants employed by Vatamaniuk et al. [8] were so high that the estimated free Cd(II) level was three orders of magnitude less than those calculated with the constants from other sources [17]. From the conditional stability constants obtained in the current study, the free
S. Ogawa et al. / Journal of Inorganic Biochemistry 105 (2011) 111–117
Cd(II) level was demonstrated to be 5.6 × 10− 9 M. Therefore, it can be concluded that Cd(II) binds rAtPCS1 at two sites of the enzyme. 4.7. Potential physiological relevance of the Cd(II) binding sites of rAtPCS1 As suggested previously [5], the first Cd(II) binding site on rAtPCS1 may play a role in the regulation of PC synthesis in vivo. The enzyme is activated by the intrusion of Cd(II) ions in cells through binding to the site and is deactivated by removal of the ion from the activated enzyme through complexation of the product PCs with Cd(II). In contrast, the physiological role of the second site is obscure. Oxidative stress is a symptom induced by Cd(II) [18]. At a high dose of Cd(II), cells may attempt to cope with the oxidative stress, where GSH is an indispensable component of the defense system through the ascorbate-GSH cycle [19], rather than decrease the activity of Cd(II). As long as the synthesis of PCs is prompt, however, free Cd(II) may not reach a level that would allow the ion to bind to the second site.
Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10.1016/j.jinorgbio.2010.09.011.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
Abbreviations AtPCS1 Arabidopsis thaliana PCS1 AtPCS1-FLAG AtPCS1 with a C-terminal FLAG tag BCS bathocuproinedisulfonate Cd(II)-G2 1:2 Cd(II) GSH complex [Cd(II)]t total concentration of Cd(II) CHES N-cyclohexyl-2-aminoethanesulfonic acid desglycyl PC2 PC2 devoid of the C-terminal glycine; G GSH with various ionized states total concentration of G [G]t, GSH glutathione HEPES 2-[4-(2-hydroxyl)-1-piperazinyl]ethanesulfonic acid rAtPCS1 recombinant AtPCS1 TFA trifluoroacetic acid PC phytochelatin phytochelatin with n γ-Glu-Cys units PCn PCS phytochelatin synthase
117
[12]
[13] [14] [15] [16] [17] [18] [19]
N. Kondo, K. Imai, M. Isobe, T. Goto, Tetrahedron Lett. 25 (1984) 3869–3872. E. Grill, E.-L. Winnacker, M.H. Zenk, Science 230 (1985) 674–676. M.H. Zenk, Gene 179 (1996) 21–30. S. Clemens, D. Peršoh, Plant Sci. 177 (2009) 266–271. E. Grill, S. Löffler, E.-L. Winnacker, M.H. Zenk, Proc. Natl Acad. Sci. USA 86 (1989) 6838–6842. S. Löffler, A. Hochberger, E. Grill, E.-L. Winnacker, M.H. Zenk, FEBS Lett. 258 (1989) 42–46. M. Oven, J.F. Page, M.H. Zenk, T.M. Kutchan, J. Biol. Chem. 277 (2002) 4747–4754. O.K. Vatamaniuk, S. Mari, Y.-P. Lu, P.A. Rea, J. Biol. Chem. 275 (2000) 31451–31459. T. Maier, C. Yu, G. Küllertz, S. Clemens, Planta 218 (2003) 300–308. Y. Osaki, T. Shirabe, H. Nakanishi, T. Wakagi, E. Yoshimura, Metallomics 1 (2009) 353–358. M. Seki, P. Carninci, Y. Nishiyama, Y. Hayashizaki, K. Shinozaki, Plant J. 15 (1998) 707–720. M. Seki, M. Narusaka, A. Kamiya, J. Ishida, M. Satou, T. Sakurai, M. Nakajima, A. Enju, K. Akiyama, Y. Oono, M. Muramatsu, Y. Hayashizaki, J. Kawai, P. Carninci, M. Itoh, Y. Ishii, T. Arakawa, K. Shibata, A. Shinagawa, K. Shinozaki, Science 296 (2002) 141–145. S. Ogawa, R. Ichiki, M. Abo, E. Yoshimura, Anal. Chem. 81 (2009) 9199–9200. S. Ogawa, T. Yoshidomi, T. Shirabe, E. Yoshimura, J. Inorg. Biochem. 104 (2010) 442–445. O.K. Vatamaniuk, S. Mari, A. Lang, S. Chalasani, L.O. Demkiv, P.A. Rea, J. Biol. Chem. 279 (2004) 22449–22460. H. Bisswanger, Enzyme Kinetics, Second, Revised and Updated Edition, WILEYVCH Verlag GmbH & Co., Weinheim, Germany, 2008. S. Ogawa, E. Yoshimura, Anal. Biochem. 402 (2010) 200–202. A.S. Molina, C. Nievas, M.V.P. Chaca, F. Garibotto, U. González, S.V. Marsá, C. Luna, M.S. Giménez, F. Zirulnik, Plant Growth Regul. 56 (2008) 285–295. L.A. del Río, L.M. Sandalio, F.J. Corpas, J.M. Palma, J.B. Barroso, Plant Physiol. 141 (2006) 330–335.