Mechanism of Heavy Metal Ion Activation of Phytochelatin (PC) Synthase

The dependence of phytochelatin synthase (γ-glutamylcysteine dipeptidyltranspeptidase (PCS), EC 2.3.2.15) on heavy metals for activity has invariably been interpreted in terms of direct metal binding to the enzyme. Here we show, through analyses of immunopurified, recombinant PCS1 from Arabidopsis thaliana(AtPCS1), that free metal ions are not essential for catalysis. Although AtPCS1 appears to be primarily activated posttranslationally in the intact plant and purified AtPCS1 is able to bind heavy metals directly, metal binding per se is not responsible for catalytic activation. As exemplified by Cd2+- and Zn2+-dependent AtPCS1-mediated catalysis, the kinetics of PC synthesis approximate a substituted enzyme mechanism in which micromolar heavy metal glutathione thiolate (e.g.Cd·GS2 or Zn·GS2) and free glutathione act as γ-Glu-Cys acceptor and donor. Further, as demonstrated by the facility of AtPCS1 for the net synthesis of S-alkyl-PCs from S-alkylglutathiones with biphasic kinetics, consistent with the sufficiency of S-alkylglutathiones as both γ-Glu-Cys donors and acceptors in media devoid of metals, even heavy metal thiolates are dispensable. It is concluded that the dependence of AtPCS1 on the provision of heavy metal ions for activity in media containing glutathione and other thiol peptides is a reflection of this enzyme's requirement for glutathione-like peptides containing blocked thiol groups for activity.

Phytochelatins (PCs) 1 are (␥-Glu-Cys) n -Xaa polymers whose synthesis from glutathione (GSH) is promoted by heavy metals (1,2). First identified in the fission yeast Schizosaccharomyces pombe and termed cadystins (3), PCs have since been found in some fungi, some marine diatoms, and all plant species investigated (4). PCs contain 2-11 ␥-Glu-Cys repeats, act as high affinity metal chelators, and facilitate the vacuolar sequestration of heavy metals, most notably Cd 2ϩ (2). PC-deficient Arabidopsis cad1 mutants are hypersensitive to Cd 2ϩ salts (5), Cd⅐PC complexes localize preferentially to the vacuole of intact plant cells (6), and in plant cell lines capable of tolerating high levels of Cd 2ϩ at least 90% of this metal is accumulated as Cd⅐PC complexes (2). In the organism for which the molecular basis of PC-dependent metal detoxification is best understood, S. pombe, vacuolar Cd 2ϩ sequestration is mediated by a 90.5-kDa vacuolar ATP-binding cassette transporter, heavy metal tolerance factor 1 (HMT1), that catalyzes the MgATP-energized uptake of Cd⅐PCs and apoPCs into the vacuoles of wild type but not hmt1 Ϫ cells (7,8). HMT1 homologs have not yet been isolated from plants, but an MgATP-energized transport pathway for PCs and Cd⅐PCs, analogous to that identified in S. pombe, has been characterized in vacuolar membrane vesicles isolated from oat roots (9).
Although it is more than a decade since the first report of the partial purification of heavy metal-, primarily Cd 2ϩ -, activated enzymes (PC synthases; ␥-glutamylcysteine dipeptidyltranspeptidases, EC 2.3.2.15) competent in the synthesis of PCs from GSH and related thiol tripeptides, by the net transfer of a ␥-Glu-Cys unit from one thiol peptide to another or to a previously synthesized PC molecule (10), it is only in the last year that three groups have simultaneously and independently cloned and characterized genes encoding this enzyme. Isolated from Arabidopsis, S. pombe, and wheat, these genes, designated AtPCS1, SpPCS, and TaPCS1, respectively, encode 40 -50% sequence-similar 50 -55-kDa polypeptides active in the synthesis of PCs from GSH (11)(12)(13). All known cad1 mutants are mutated in AtPCS1 (12), SpPCS disruptants are hypersensitive to heavy metals and deficient in cellular PCs (11), and heterologous expression of AtPCS1 in Saccharomyces cerevisiae, an organism that lacks PCS homologs and does not otherwise synthesize appreciable amounts of PCs, confers increased heavy metal tolerance and elicits Cd 2ϩ -dependent intracellular PC accumulation (13). As established by the capacity of cell-free extracts from AtPCS1-or SpPCS-transformed cells of Escherichia coli (12) and of purified FLAG epitope-tagged AtPCS1 (AtPCS1-FLAG) for the Cd 2ϩ -activated synthesis of short chain PCs from GSH in vitro (13), each of these gene products is not only necessary but also sufficient for the elaboration of PCs.
A physiologically crucial and biochemically intriguing property of PC synthase is its susceptibility to activation by heavy metals. It is by virtue of the activation of PC synthase-catalyzed PC biosynthesis by agents, heavy metal ions, that poison most enzymes that plants and fungi are able to mount a PCbased response to heavy metals.
Few investigators have considered explicitly how heavy metals activate PC synthase but those that have considered it have assumed that activation is consequent on the direct binding of metal ions to the enzyme (2,4). Indeed in the most recent model for PC synthase action, it has been proposed that the strongly conserved N-terminal half of the enzyme is responsible for catalysis and that activation arises from the binding of metal ions to residues, possibly cysteine residues, within this domain (4). The presence of five conserved cysteine residues, two of which are vicinal, in the N-terminal halves of AtPCS1, SpPCS, and TaPCS1 is at least consistent with this notion, as is the observation that the three most extreme Arabidopsis cad1 alleles have amino acid substitutions in this region (12). An extension of this model, proposed to ascribe a role to the more sequence-divergent C-terminal half of the molecule and to account for the properties of the least extreme cad1 allele, cad1-5, a nonsense mutation causing premature termination and deletion of the C-terminal segment, is the concept of a C-terminal "metal-sensing domain" whose multiple cysteine residues bind heavy metals and bring them into contact with the putative "activation" site within the N-terminal, catalytic half of the molecule.
In the experiments described here we exploit the ease with which AtPCS1-FLAG can be purified to near homogeneity from AtPCS1::FLAG-transformed S. cerevisiae to yield high activity PC synthase preparations (13) to examine the mechanism by which this class of transpeptidase is activated by heavy metals. In so doing, we establish that although AtPCS1-FLAG confers tolerance to and is subject to posttranslational activation by a broad range of heavy metals, direct interaction of the enzyme with free metal ions is not the primary mode of activation. Instead, heavy metal ions are required for the formation of heavy metal peptide, GSH or PC, thiolates that serve as cosubstrates for catalysis via a substituted enzyme mechanism. On the basis of these findings and the efficacy of S-alkylglutathiones as substrates for the synthesis of S-alkyl-PCs in the complete absence of metal ions, it is inferred that AtPCS1 catalyzes the polymerization of GSH-derived thiol peptides containing blocked thiol groups, regardless of whether the substrate-active species is a heavy metal thiolate or thioether. If heavy metals do directly bind AtPCS1 in vivo it is to a limited extent and associated with only minor augmentation of synthetic activity.
Purification of AtPCS1-FLAG-The soluble fraction from pYES3-AtPCS1::FLAG-transformed DTY167 (DTY167/pYES3-AtPCS1 ::FLAG) cells was prepared by the disruption of spheroplasts as described (13), and AtPCS1-FLAG was purified on an anti-FLAG M2 affinity gel column (Sigma) according to the manufacturer's recommendations, except that the wash and elution buffers contained 10% (v/v) glycerol in addition to TBS (150 mM NaCl, 50 mM Tris-HCl, pH 7.4) and 0.1 M glycine-HCl (pH 3.5), respectively (13). For the experiments directed at determining if tightly bound metal ions retained during its extraction and purification might influence the activity or response of the enzyme to metal ions added to the reaction media, aliquots of purified AtPCS1-FLAG (30 g) were pretreated with 1 mM Tris-EGTA in Tris-buffered elution buffer (pH 8.0) on ice for 1 h before removal of the chelator by dialysis of the samples against 120 volumes of 10% (w/v) deionized glycerol in 50 mM HEPES-BTP buffer (pH 8.0) for 12 h at 4°C in Slide-A-Lyzer mini dialysis tubes (molecular weight cutoff of 10,000, Pierce). Control samples were manipulated in an identical manner except that EGTA was omitted from the pretreatment medium.
Measurement of PCs, S-Alkyl-PCs, and PC Synthase Activity-PC synthase activity was assayed in reaction media (200 mM HEPES-BTP buffer, pH 8.0) containing purified AtPCS1-FLAG (0.5 g) or no protein, the indicated concentrations of GSH or its S-alkyl derivatives, and/or heavy metal salt. For RP-HPLC, 500 -1000 l volumes of the extracts were made 5% (w/v) with 5-sulfosalicylic acid and centrifuged before aliquots (50 -100 l) of the supernatant were loaded onto an Econosphere C18, 150 ϫ 4.6-mm reverse-phase column (Alltech). The column was developed with a linear gradient of water, 0.05% (v/v) phosphoric acid, 17% (v/v) acetonitrile, 0.05% (v/v) phosphoric acid at a flow rate of 1 ml/min. For the quantitation of PCs, thiols were estimated spectrophotometrically at 412 nm by reacting aliquots (500 l) of the column fractions with 0.8 mM 5,5Ј-dithiobis(2-nitrobenzoic acid) (DTNB) (500 l) dissolved in 250 mM phosphate buffer, pH 7.6 (16). Calibration was with GSH. For the quantitation of S-alkyl-PCs, free amino groups were estimated fluorimetrically by reacting aliquots (500 l) of the column fractions with 0.4 M sodium borate (pH 9.7) (200 l) and fluorescamine (20 l of a 3 mg/ml solution dissolved in acetone) (17). Fluorescence was measured at excitation and emission wavelengths of 390 and 475 nm, respectively, after quenching unreacted fluorescamine by the addition of water (to a final volume of 1 ml). Calibration was with S-methylglutathione. Where indicated, thiol peptides were exhaustively reduced before RP-HPLC by incubating aliquots (500 l) of the reaction media with 0.4 M sodium borohydride at 37°C for 20 min before deproteinization and RP-HPLC. The kinetics of AtPCS1-FLAG-catalyzed incorporation of GSH or S-methylglutathione into PCs or S-methyl-PCs were determined by limiting the incubation times to 180 s and 90 s, respectively, to minimize substrate depletion, end product accumulation, and the formation of PCs or S-methyl-PCs containing more than two ␥-Glu-Cys or ␥-Glu-(methyl-Cys) repeats.
Equilibrium Dialysis of AtPCS1-FLAG-Binding of Cd 2ϩ was determined by equilibrium dialysis of 400 -800 l (160 g) samples of purified AtPCS1-FLAG against 80-ml volumes of 10 mM Tris-HCl buffer, pH 7.8, containing 0.05-20 M 109 CdCl 2 (specific activity 22 Ci/mol) for 12 h at 4°C in 2-ml mini-collodion membrane tubes (molecular weight cutoff of 25,000, Schleicher & Schuell). Protein-bound 109 Cd was estimated by measuring the radioactivity of the bulk medium outside the dialysis tube and that of the solution within the dialysis tube and determining the increase in 109 Cd radioactivity consequent on AtPCS1-FLAG. Binding of Cu 2ϩ and Zn 2ϩ to AtPCS1-FLAG was estimated by measuring the decrease in equilibrium binding of a half-saturating (K L ) concentration of 109 Cd 2ϩ as the result of inclusion of a range of concentrations of Cu 2ϩ or Zn 2ϩ in the equilibrium dialysis buffer.
Northern Analyses-To assess the effects of treatment with heavy metal salts on the steady state levels of AtPCS1 transcripts, 21-day-old seedlings grown in Gamborg's B-5 medium were transferred into fresh medium containing 25 or 100 M CdSO 4 , CuSO 4 , or ZnSO 4 and incubated with shaking at 22°C for an additional 6 or 24 h before RNA extraction. Control seedlings were treated in an identical manner except that CdCl 2 , CuSO 4 , and ZnSO 4 were not added to the culture media.
Amino Acid Analyses-The chain lengths of the PCs and S-alkyl-PCs synthesized from GSH or S-alkylglutathiones, respectively, were determined by estimating their Glu/Gly or Glu/S-alkyl-Cys/Gly ratios (ratio ϭ n ϭ number of Glu-Cys or Glu-(alkyl-Cys) repeats per Gly) after acid hydrolysis of the appropriate HPLC fractions. Aliquots of the fractions were taken to dryness in pyrolyzed glass tubes, hydrolyzed in gas-phase 6 N HCl for 20 h at 110°C before ion-exchange chromatography, postcolumn derivatization with O-phthaldehyde, and fluorescence detection (17).

Calculation of Concentrations of Free Heavy Metal Ions and Their
Complexes-The concentrations of free heavy metal ions and their complexes with GSH and other ligands in the reaction media were calculated from their stability constants using the computer program SOLCON (from Dr. Y. E. Goldman, Dept. of Physiology, University of Pennsylvania). The stability constants used, which were obtained from Martell and Smith (18)  Other Computations-Kinetic parameters, AtPCS1-FLAG heavy metal-binding constants, and stoichiometries of binding were estimated by nonlinear least squares analysis (20) using the Ultrafit nonlinear curve-fitting package from BioSoft (Ferguson, MO).
Protein Estimations-Protein was estimated by the dye-binding method (21).

AtPCS1 Is Constitutively Expressed in the Intact Plant-High
stringency Northern analyses revealed a single 1.7-kb band of similar intensity after hybridization of random-primed 32 P-labeled AtPCS1 cDNA with total RNA extracted from roots and shoots of 21-day-old Arabidopsis seedlings, regardless of whether the seedlings had been exposed to 25 or 100 M Cd 2ϩ (CdSO 4 ), the most potent activator of AtPCS1-catalyzed PC synthesis (below), Cu 2ϩ (CuSO 4 ), an activator of intermediate potency, or Zn 2ϩ (ZnSO 4 ), an activator of weak to moderate potency, for 6 h (data not shown) or for 24 h before RNA extraction ( Fig. 1). From these results and those from earlier biochemical investigations, demonstrating that extractable PC synthase activity is not enhanced by the pretreatment of plant tissues or cell suspension cultures with heavy metal salts (2), modulation of AtPCS1 by heavy metals was inferred to be exerted at the enzyme level. All subsequent experiments were therefore directed at determining how metal ions interact with the enzyme to elicit PC synthetic activity and were performed on recombinant AtPCS1 (AtPCS1-FLAG). For this purpose, heterologously expressed AtPCS1-FLAG was 30-to 50-fold immunopurified from the soluble fraction of pYES3-AtPCS1::FLAG-transformed S. cerevisiae strain DTY167 to yield a single anti-FLAG antibodyreactive, M r ϭ 58,000 polypeptide species (13) whose activity consistently exceeded 35 mol/mg/min when assayed in standard reaction medium containing 25 M CdCl 2 , 3.3 mM GSH, and 200 mM HEPES-BTP (pH 8.0).

AtPCS1-FLAG Is Activated by a Broad Range of Heavy Metals and Is Sufficient for the Synthesis of PC 2-6 -AtPCS1-FLAG
retained all of the known characteristics of the PC synthetic activities of plant extracts. It was subject to activation by a broad range of heavy metal cations and oxyanions and was competent in the synthesis of both short chain and long chain PCs. AtPCS1-FLAG-catalyzed PC synthesis from GSH was obligatorily dependent on the provision of heavy metals. No activity was detectable when metals were omitted from the reaction medium, but the addition of Cd 2ϩ , Hg 2ϩ , As 3ϩ , AsO 2 Ϫ , Cu 2ϩ , Zn 2ϩ , Pb ϩ , AsO 4 3Ϫ , Mg 2ϩ , and Ni 2ϩ at total concentrations of 50 M increased the capacity of AtPCS1-FLAG for PC synthesis from GSH by 47.7, 40.3, 27.7, 27.2, 10.6, 8.3, 4.8, 3.9, and 3.6-fold, respectively, versus Co 2ϩ , the least stimulatory metal ion examined (Fig. 2). Because in no case, with the exception of Cu 2ϩ , did pretreatment of the terminated reaction media with sodium borohydride before RP-HPLC markedly change the estimated reduced thiol contents of the PCs synthesized or the apparent rank order with which the metal cations or oxyanions promoted PC synthesis (Fig. 2), it was concluded that the effects of most of the metal cations and oxyanions examined were exerted at the enzyme level, not at the level of the oxidation state and amenability of the thiol peptide reaction products to detection with DTNB. Cu 2ϩ was an exception in that prior reduction of the reaction products with sodium borohydride doubled DTNB reactivity (Fig. 2), suggesting an approximately 1:1 ratio of oxidized:reduced thiols in the PCs synthesized in media containing this metal ion.
As exemplified by the results obtained with standard reaction medium containing Cd 2ϩ , AtPCS1-FLAG was competent in the sequential synthesis of PC 2-6 from GSH (Fig. 3). For shorter term (0 -60 min) incubations, net synthesis of PC 2-4 was evident within 2, 5, and 20 min, respectively, of the addition of enzyme (Fig. 3, inset), indicating that whereas GSH, alone, was sufficient for the synthesis of PC 2 , the net synthesis of PC 3 and PC 4 required not only GSH but also PC 2 and PC 3 , respectively. For longer term incubations, the net synthesis of not only PC 2 , PC 3 , and PC 4 but also PC 5 and PC 6 was evident (Fig. 3). Qualitatively similar time dependences and ranges of chain length were observed when PC synthase activity was activated by Cu 2ϩ or Zn 2ϩ instead of Cd 2ϩ (data not shown). As would be predicted when GSH is the prevalent thiol peptide in the reaction medium and PC nϩ1 is derived from PC n ϩ GSH, the final amounts of thiol equivalents (ϭ ␥-Glu-Cys units) incorporated into PC 2-6 decreased exponentially with increase in chain length (Fig. 3). PC-associated thiols in the reaction media. Given that the complexes formed between heavy metals and thiol compounds are among the most stable known, it was decided to determine the likely concentrations of free metal ions and their complexes under conditions in which AtPCS1-FLAG-catalyzed PC synthesis was sustained.
The concentrations of free metal ions and their complexes were estimated by substitution of the stability constants of the complexes formed between the metal ions concerned and GSH into the SOLCON computer program. In the first instance, Cd 2ϩ and Zn 2ϩ were chosen as model metal ions, because of the ready availability of comprehensive compilations of the appropriate stability constants for these and their ligands, and the calculations were based on the composition of the standard reaction medium containing 25 M metal chloride, 3.3 mM GSH, and 200 mM HEPES-BTP buffer (pH 8.0).
Two crucial insights were gained from these analyses. The first was that under the conditions in which AtPCS1-FLAG catalyzed high rates of PC synthesis from GSH, the concentrations of free Cd 2ϩ and free Zn 2ϩ ([Cd 2ϩ ] free and [Zn 2ϩ ] free ) were only of the order of 10 Ϫ13 and 10 Ϫ9 M, respectively ( Table I). The second was that more than 98% of the total Cd 2ϩ and more than 80% of the total Zn 2ϩ added to the reaction medium were associated with GSH as their corresponding bidentate thiolates, bis(glutathionato)cadmium (Cd⅐GS 2 ) and bis(glutathionato)zinc (Zn⅐GS 2 ) ( Table I).
Recognition of the prevalence of metal thiolates and of the extremely low concentrations of free metal ions in media in which PC synthase activity was appreciable necessitated reconsideration of the form in which metal ions exert their effects on AtPCS1-FLAG-mediated catalysis. There were at least three explanations. (i) AtPCS1-FLAG has an extremely high inherent affinity for heavy metal ions, and subpicomolar or nanomolar concentrations of free Cd 2ϩ or Zn 2ϩ , for example, are sufficient for enzyme activation by direct binding. (ii) Heavy metal ions do not interact with AtPCS1-FLAG directly but instead do so as their corresponding thiolates. For instance, Cd 2ϩ and Zn 2ϩ must first associate with the enzyme as their Cd⅐GS 2 and Zn⅐GS 2 complexes before transfer of the metal ion to the putative activation site of AtPCS1-FLAG. (iii) The active substrates, or one of the active substrates, for AtPCS1-FLAGcatalyzed PC synthesis are heavy metal thiolates. Although direct interaction of the enzyme with heavy metals may not be a requirement for catalysis, there is a requirement for substrate containing thiol-associated heavy metal.
AtPCS1-FLAG Binds Heavy Metals at Only Low to Moderate Affinity-Of these three explanations, the first, direct binding, seemed the least capable of accounting for the activations measured in media containing millimolar concentrations of thiol peptides. Although AtPCS1-FLAG bound 109 Cd 2ϩ at high capacity (B max ϭ 7.09 Ϯ 0.94) as determined by equilibrium dialysis, the ligand-binding constant (K L ϭ 0.54 Ϯ 0.21 M) was 6 orders of magnitude greater than the value of [Cd 2ϩ ] free calculated for the standard reaction medium (Fig. 4A). A similar pattern was inferred for Cu 2ϩ and Zn 2ϩ . Inclusion of 1-20 M Cu 2ϩ in dialysis buffer containing a concentration of 109 Cd 2ϩ approximating its K L for binding to AtPCS1-FLAG (0.5 M) decreased equilibrium binding of 109 Cd 2ϩ to approximately 50% of the control (ϪCu 2ϩ ) level in a manner consistent with a K L for Cu 2ϩ binding of 5.6 Ϯ 1.5 M (Fig. 4B). Inclusion of the same concentrations of Zn 2ϩ in the dialysis buffer exerted little or no effect on the equilibrium binding of 0.5 M 109 Cd 2ϩ (Fig. 4B).
Heavy Metal Thiolates and Free GSH as Candidate Substrates for AtPCS1-FLAG-In agreement with the conclusions drawn from the equilibrium binding measurements, and as would be predicted from explanations (ii) and (iii), analyses of the steady state kinetics of AtPCS1-FLAG-catalyzed PC synthesis demonstrated that activity was strictly dependent on thiolate and free GSH concentration, not free metal ion concentration. Providing that the incubations were of sufficiently short duration (180 s) as to enable precise initial rate measurements and ensure exclusive synthesis of PC 2 , so precluding complications attending the synthesis of longer chain PCs, the kinetics of Cd 2ϩ -activated PC synthesis were uniform. When free GSH concentration was adjusted to values of 0.6 -6.6 mM and the concentrations of Cd⅐GS 2 were enumerated using the SOLCON program, the initial rates of AtPCS1-FLAG-catalyzed PC synthesis (v) approximated a series of Michaelis-Menten functions (Fig. 5A). In all cases, and in support of the notion that free metal ions are not essential for catalysis, other than through their interaction with substrate thiols, free GSH concentrations in excess of those required to complex Cd 2ϩ increased, rather than decreased, PC synthesis.
In strict agreement with the possibility that the reaction catalyzed by AtPCS1-FLAG proceeds via a substituted enzyme   (Fig. 5B). Enumeration of K mCd⅐GS 2 at limiting GSH concentration and of K m(GSH) at limiting Cd⅐GS 2 concentration yielded values of 9.2 Ϯ 2.3 M and 13.6 Ϯ 3.3 mM, respectively, suggesting that Cd⅐GS 2 is a high affinity substrate whereas free GSH is a low affinity substrate. The outcome of experiments in which an equivalent approach was applied to reaction media containing Zn 2ϩ instead of Cd 2ϩ was qualitatively the same, except that K mZn⅐GS 2 had a value of 4.5 Ϯ 0.9 M (data not shown). The simplest explanation for the high capacity of AtPCS1-FLAG for PC synthesis from GSH in media containing subpicomolar concentrations of free Cd 2ϩ or nanomolar concentrations of free Zn 2ϩ but micromolar concentrations of Cd⅐GS 2 or Zn⅐GS 2 , in conjunction with the adherence of the Cd⅐GS 2 or Zn⅐GS 2 and GSH concentration dependence of the rate of PC synthesis to ping-pong kinetics, is that at least one of the partial reactions catalyzed by AtPCS1 necessitates formation of a substituted enzyme intermediate. Nominally, AtPCS1-FLAG catalyzes a reaction of the form, AtPCS1 ϩ ␥-Glu-Cys-Gly ¡ AtPCS1-␥-Glu-Cys ϩ Gly (Eq. 1) AtPCS1-␥-Glu-Cys ϩ X ⅐ ͑␥-Glu-Cys-Gly) 2 ¡ AtPCS1 ϩ X ⅐ ͑␥-Glu-Cys) 2 -Gly ϩ ␥-Glu-Cys-Gly (Eq. 2) in which X is heavy metal and GSH (or X⅐GS 2 ) and X⅐GS 2 (or GSH) are ␥-Glu-Cys donor and acceptor, respectively.

AtPCS1-FLAG Is Not Obligatorily Dependent on Heavy Metal
Ions-Although the reaction scheme depicted in Equations 1 and 2 does not automatically preclude explanation (ii), the possibility that thiolates act to shuttle activatory metal ion to the enzyme, it does raise the question of whether thiol peptides containing blocked thiols might serve as substrates regardless of whether blocking is a consequence of heavy metal thiolate formation or some other thiol-specific modification.
The results summarized in Figs. 6 and 7 indeed demonstrate that explanation (ii) cannot be generally applicable in that substrate thiol-specific modifications other than those associated with heavy metal thiolate formation render thiol peptides amenable to transpeptidation by AtPCS1-FLAG.
When assayed in media devoid of metal salts, AtPCS1-FLAG catalyzed the net synthesis of S-methyl-PCs from S-methylglutathione with a time dependence (Fig. 6) similar to that for the synthesis of unsubstituted PCs from the equivalent concentration of GSH in media containing heavy metals (Fig. 3). The sequence of appearance of S-methyl-PC 2 , S-methyl-PC 3 , and S-methyl-PC 4 in the reaction medium during shorter term (0 -60 min) incubations (Fig. 6, inset) was consistent with a precursor-product relationship analogous to that inferred for the synthesis of unsubstituted PCs from GSH (Fig. 3, inset), and longer term (6 h) incubations resulted in the net formation of S-methyl-PC 5 in addition to S-methyl-PC [2][3][4] (Fig. 6).
The facility of AtPCS1-FLAG for catalyzing the synthesis of S-alkyl-PCs from S-alkylglutathiones was not restricted to Smethyl derivatives. Not only S-methylglutathione but also Sethyl-, S-propyl-, S-butyl-, and S-hexylglutathiones were subject to transpeptidation by AtPCS1-FLAG (Table II). The initial rates of metal ion-independent S-alkyl-PC synthesis were similar for all of the S-alkylglutathione derivatives examined   (Fig. 7B): a high affinity, low capacity component (K m1 ϭ 1.0 Ϯ 0.2 mM; V max1 ϭ 38.1 Ϯ 2.2 mol/mg/min) evident at S-methylglutathione concentrations of 2 mM and less and a low affinity, high capacity component (K m2 ϭ 9.8 Ϯ 1.6 mM; V max2 ϭ 115.6 Ϯ 18.9 mol/ mg/min) evident at S-methylglutathione concentrations of 3 mM and greater. Behavior of this type would be expected if, as implied by the kinetics of Cd 2ϩ -dependent PC synthesis from GSH, S-methylglutathione must be capable of substituting for both the high affinity and low affinity substrates, Cd⅐GS 2 and free GSH, respectively. The near coincidence of the K m2 for S-methylglutathione-dependent S-methyl-PC 2 synthesis (Fig.  7B) with K m(GSH) for Cd 2ϩ -dependent PC 2 synthesis from Cd⅐GS 2 and GSH (13.6 Ϯ 3.3 mM, Fig. 5) but the approximately 110-fold greater value of K m1 for S-methyl-PC 2 synthesis (Fig.  7B) versus K mCd⅐GS 2 for PC 2 synthesis (Fig. 5) indicates that S-methylglutathione is a markedly more effective stereochemical analog of GSH than of Cd⅐GS 2 .
S-Alkyl-PC Synthesis Is Promoted by but Not Obligatorily Dependent on Heavy Metal Ions-On the one hand, the capacity of S-alkylglutathiones to serve as substrates for S-alkyl-PC synthesis in the complete absence of heavy metals established that blocked thiols on the substrate are sufficient for core catalysis. On the other hand, the sufficiency of S-alkylglutathiones as substrates despite their inability to form thiolates provided a unique opportunity to assess the influence of free heavy metal ions on AtPCS1-FLAG activity under conditions in which heavy metal-substrate interactions are minimized.
The effects of heavy metal ions on activity were examined by measuring the initial rates of AtPCS1-FLAG-catalyzed S-methyl-PC 2 synthesis from S-methylglutathione in reaction media containing different concentrations of Cd 2ϩ and by determining the effects of maximally activating concentrations of Cd 2ϩ on the S-methylglutathione concentration dependence of S-methyl-PC 2 synthesis.  from their corresponding S-alkylglutathiones The reactions were performed as described in the legend to Fig. 3 except that GSH was replaced by 3 mM concentrations of S-methylglutathione, S-ethylglutathione, S-propylglutathione, S-butylglutathione, or S-hexylglutathione, and heavy metals were omitted from the reaction media. S-Alkyl-PC 2 derivatives were the sole products from these short (90 s) duration incubations. Note that the rates shown here are expressed in units of mol of S-alkyl-PC 2 synthesized whereas the rates of PC synthesis from unsubstituted GSH are expressed in units of mol of thiol incorporated. The results of these experiments were extremely informative in three respects. With respect to core catalysis it was evident that although Cd 2ϩ promoted the synthesis of S-methyl-PC 2 , the promotions were moderate in that approximately 50% of synthetic activity was sustained in the complete absence of metal ions (Figs. 7 and 8). With respect to the kinetics of S-methyl-PC 2 synthesis from S-methylglutathione, the effects of stimulatory concentrations of Cd 2ϩ were exerted primarily at the V max level. The biphasic substrate concentration dependence of S-methyl-PC 2 synthesis was retained in reaction media containing a maximally activating concentration of Cd 2ϩ (0.5 M) and both V max1 and V max2 were increased by 1.7-and 1.9-fold versus control media lacking metal ions (Fig. 7). By contrast, K m2 was unaffected (10.2 Ϯ 0.1 mM with and 9.8 Ϯ 1.6 mM without Cd 2ϩ , Fig. 7), and K m1 was increased from 1.0 Ϯ 0.2 mM to 1.4 Ϯ 0.1 mM (Fig. 7). With respect to the facility of AtPCS1-FLAG for binding heavy metals, the concentrations of Cd 2ϩ required for activation of S-methyl-PC 2 synthesis (0.025-1.00 M, Fig. 8), although commensurate with the concentrations required for direct binding to the enzyme as determined by equilibrium dialysis (Fig. 4), were more than 5 orders of magnitude greater than those prevailing in reaction media containing unsubstituted GSH (Table I).
The effects of Cd 2ϩ were not attributable to activation consequent on the removal of endogenous heavy metal from At-PCS1-FLAG during extraction and/or purification. Pretreatment of purified AtPCS1-FLAG with 1 mM EGTA and subsequent dialytic removal of the chelator before the measurement of S-methyl-PC 2 synthesis neither decreased the activity of the enzyme versus control samples pretreated in an identical manner in media lacking chelator nor influenced the concentration dependence of or degree to which the enzyme was activated by the direct addition of Cd 2ϩ to the reaction medium (Fig. 8). The concentrations of Cd 2ϩ required for halfmaximal and maximal activation of AtPCS1-FLAG, 0.025 and 0.50 M, respectively, were the same regardless of whether the enzyme had or had not been pretreated with EGTA (Fig. 8). Both EGTA-pretreated and control enzyme were less stimu-lated by Cd 2ϩ concentrations in excess of 1 M and inhibited by concentrations in excess of 5 M (Fig. 8). DISCUSSION The results of these investigations reveal that AtPCS1, and by implication other PC synthases, are almost exclusively regulated by heavy metals at the posttranslational level and catalyze a bisubstrate transpeptidation reaction in which both free GSH and its corresponding heavy metal thiolate are cosubstrates. Further, it is shown that although both free GSH and its heavy metal thiolate are ordinarily required for maximal activity, other compounds, for instance S-substituted glutathione derivatives, can substitute for both in such a way as to overcome the enzyme's otherwise obligatory requirement for heavy metals for activity.
The facility with which S-alkyl-PCs can be synthesized from S-alkylglutathiones in the complete absence of added heavy metal ions is significant in two respects. (i) In the context of the finding that in reaction media containing concentrations of GSH optimal for heavy metal-dependent PC synthesis, most of the heavy metal present is complexed with GSH, the high activity of S-alkylglutathiones as substrates in the absence of heavy metals implies that heavy metal ions do not activate catalysis in media containing free thiols through direct interaction with the enzyme but instead do so through interaction with the substrate. As would be expected if this were the case, the activity of AtPCS1-FLAG at a given concentration of free GSH increases as a simple Michaelian function of Cd⅐GS 2 or Zn-GS 2 concentration, and AtPCS1-FLAG although able to bind heavy metal ions directly does so at too low an affinity for direct binding to be appreciable in media containing thiol peptides. That the capacity of AtPCS1-FLAG for S-methyl-PC 2 synthesis from S-methylglutathione in media lacking added metal ions is retained despite exhaustive pretreatment with metal chelator excludes the possibility of very high affinity substoichiometric heavy metal binding and/or retention of bound metal throughout the extraction and purification procedures used for preparation of the enzyme used in these experiments. (ii) It demonstrates that at least some glutathione derivatives containing blocked thiol groups are sufficient for recognition by and transpeptidation by AtPCS1-FLAG. With specific regard to S-alkylglutathiones it suggests that this class of compounds bears sufficient resemblance to free GSH and its heavy metal thiolates to serve as both substrate and cosubstrate. S-Alkylglutathiones can act as both donor and acceptor in the transpeptidation reaction in so far as neither thiolassociated heavy metal on the substrate (or cosubstrate) or a free thiol on the cosubstrate (or substrate) are absolute prerequisites for the transpeptidation reaction. The biphasic substrate concentration dependence of S-methyl-PC 2 synthesis from S-methylglutathione, the fact that a high affinity component can be clearly resolved from a low affinity component, is consistent with this explanation if it is assumed that the former corresponds to "metal thiolate-like" binding and the latter to "free GSH-like" binding. The 100-fold lower affinity of At-PCS1-FLAG for S-methylglutathione versus metal thiolates but its approximately equivalent affinity for S-methylglutathione and GSH is explicable in terms of the closer stereochemical resemblance of S-methylglutathione to GSH than to, for example, Cd⅐GS 2 or Zn⅐GS 2 .
Previous investigations of partially purified preparations of PC synthase from Silene cubulatus cell suspension cultures have shown that another S-substituted glutathione, S-bimaneglutathione, can serve as a substrate (10). However, in the studies of the enzyme from this source neither incorporation of S-bimaneglutathione into PCs nor alleviation of the dependence of activity on heavy metals was determined. Substrate activity was assessed by monitoring a partial reaction, Gly release, not S-bimane-PC formation, and all of the assays were performed in media containing 100 M Cd 2ϩ (10).
The sufficiency of blocked thiol groups on at least one of the two substrate molecules required for core catalysis by AtPCS1-FLAG does not necessarily preclude the augmentation of activity by direct metal ion binding to the enzyme. Indeed, when the reaction conditions are designed so as to be compatible with the availability of not only sufficient substrate but also adequate concentrations of free metal ions, by exploiting the capacity of S-alkylglutathiones to act as substrates despite their inability to form heavy metal thiolates, promotion of S-alkyl-PC synthesis up and above that conferred by the provision of substrate containing blocked thiol groups is readily detectable. However, while undoubtedly of mechanistic interest and consistent with direct modulation of catalytic turnover by heavy metal binding, this effect is unlikely to be appreciable in vivo or in vitro when the dominant thiol peptide is unsubstituted GSH. The free Cd 2ϩ concentrations required for half-maximal stimulation of S-methyl-PC 2 synthesis are more than 5 orders of magnitude greater than those that prevail when the rates of synthesis of PCs from unsubstituted GSH are maximal.
Implicit in the finding that the steady state kinetics of At-PCS1-FLAG-catalyzed PC 2 synthesis from GSH in media containing heavy metal ions approximate a scheme in which heavy metal thiolate, as exemplified by Cd⅐GS 2 or Zn⅐GS 2 , and free GSH interact via a substituted enzyme intermediate, not via a ternary complex, to form PC 2 is the concept of formation of an enzyme covalent intermediate during catalysis. Specifically, given that PC synthase is a dipeptidyltranspeptidase (1, 2, 10), the kinetics of heavy metal-dependent PC 2 synthesis from GSH implicate the formation, coincident with cleavage of the Cys-Gly peptide bond of the first substrate (GSH or Cd⅐GS 2 ), of an enzyme ␥-Glu-Cys acyl intermediate, which in turn plays the role of activated donor for transpeptidation of the second substrate (Cd⅐GS 2 or GSH). If correct, an important corollary follows from this interpretation: the likelihood that the initial nucleophilic attack on the scissile bond of the first substrate is by an enzyme hydroxyl-derived oxyanion or thiol-derived thiolate anion and results in the formation of a ␥-Glu-Cys-enzyme oxyester or thioester, respectively. A mechanism analogous to that of serine proteases (23), cysteine proteases (24), and cysteine hydrolases (25-27) may therefore be invoked, in which case at least some of the energy required for condensation of the ␥-Glu-Cys unit from the first substrate with the ␣-amino group of the second substrate during PC synthesis is derived from an enzyme oxyester of intermediate energy or an enzyme thioester of high energy formed during the first phase of the catalytic cycle.
A scheme summarizing these conclusions is shown in Fig. 9. According to this scheme, a substantially modified version of that proposed by Cobbett (4, see "Introduction"), AtPCS1 is considered to catalyze a dipeptidyltranspeptidation reaction in which the ␥-Glu-Cys donor acylates the enzyme, concomitant with the release of Gly. The activated ␥-Glu-Cys-AtPCS1 acyl intermediate so formed then transfers the ␥-Glu-Cys unit to the second substrate to generate a product extended by the condensation of one new ␥-Glu-Cys repeat with the N terminus of the acceptor. The minimum condition that must be satisfied for this reaction to proceed is that at least one of the thiol groups on one of the substrate molecules is blocked either through heavy metal thiolate formation or S-alkylation. Although heavy metals are not crucial for core catalysis, which is presumably mediated by the conserved N-terminal half of the enzyme, other than through substrate thiolate formation, they are capable of augmenting activity in the presence of substrateactive S-alkyl derivatives. However, unlike heavy metal-mediated catalytic activation in media containing unsubstituted thiols, heavy metal-mediated augmentation requires relatively high concentrations of free metal ions and appears to derive from the direct binding of metal ions to AtPCS1. In light of the dispensability of this binding reaction for core catalysis it is inferred to be at a site distinct from the active site, possibly via the multiple Cys residues found in the more sequence-divergent C-terminal half of the molecule.
In addition to these enzymological insights, two informative physiological implications follow from the results presented. The first is that, contrary to the prevailing model (2,28), termination of the reactions catalyzed by PC synthases cannot be solely contingent on the chelation of heavy metals because GSH and PC complexes containing heavy metal are active substrate species. Instead, termination more likely results from exhaustion of the heavy metal pool such that free thiols (GSH and apo-PCs) compete with thiolates for the high affinity site of the synthase. Diminution of the substrate-active thiolate pool, whether it be by the incorporation of heavy metals into higher order, substrate-inactive metal⅐PC complexes or by the removal of metal⅐PC complexes from the cytosolic pool into the Step 2 is transfer of the ␥EC unit from the substituted enzyme intermediate to the second substrate to generate a product containing one additional ␥EC repeat.
Step 3 is transport of the product from the cytosol into the vacuole. Solid arrows denote the core catalytic pathway. Dashed arrows denote an auxiliary catalytic pathway in which heavy metals, such as Cd 2ϩ , accelerate catalysis by binding to the enzyme at a site distinct from but coupled with the substrate-binding site(s). The sequence-conserved N-terminal and sequence-divergent C-terminal halves of AtPCS1 are depicted in black and white, respectively. Steps 1 and 2 are inferred to have different substrate requirements in that R is H or CH 3 through C 6 H 11 and R' is a heavy metal or CH 3 through C 6 H 11 . The R-and R'-substituted forms of glutathione are considered to participate in Steps 1 and 2, respectively (or vice versa) but not both.
vacuole is probably the determining factor for ensuring that PC synthesis meets but does not exceed demand.
The second physiological implication of the utilization by AtPCS1-FLAG of heavy metal thiolates as substrates is that the cytosolic concentration of free heavy metal ions need not increase even transitorily for net PC synthesis. Given the high values of the stability constants of heavy metal⅐GSH complexes and the fact that the prevailing concentration of GSH, the most abundant intracellular thiol, is between 1 and 10 mM (29), any metal that gains access to the cytosol would be expected to be rapidly converted to its corresponding thiolate. The GSH thiolates so formed, because of the moderately high and constitutive expression of PCS genes would, in turn, be incorporated into derivatives, PCs, that also bind heavy metals but at higher affinity (2).