Domain Organization of Phytochelatin Synthase

Abstract

Phytochelatin synthase (PCS) is a major determinant of heavy metal tolerance in plants and other organisms. No structural information on this enzyme is as yet available. It is generally believed, however, that the active site region is located in the more conserved N-terminal portion of PCS, whereas various, as yet unidentified (but supposedly less critical) roles have been proposed for the C-terminal region. To gain insight into the structural/functional organization of PCS, we have conducted a limited proteolysis analysis of the enzyme from Arabidopsis (AtPCS1), followed by functional characterization of the resulting polypeptide fragments. Two N-terminal fragments ending at positions 372 (PCS_Nt1) and 283 (PCS_Nt2) were produced sequentially upon V8 protease digestion, without any detectable accumulation of the corresponding C-terminal fragments. As revealed by the results of in vivo and in vitro functional assays, the core PCS_Nt2 fragment is biosynthetically active in the presence of cadmium ions and supports phytochelatin formation at a rate that is only ∼5-fold lower than that of full-length AtPCS1. The loss of the C-terminal region, however, substantially decreases the thermal stability of the enzyme and impairs phytochelatin formation in the presence of certain heavy metals (e.g. mercury and zinc, but not cadmium or copper). The latter phenotype was shared by PCS_Nt2 and by its precursor fragment PCS_Nt1, which, on the other hand, was almost as stable and biosynthetically active (in the presence of cadmium) as the full-length enzyme. AtPCS1 thus appears to be composed of a protease-resistant (and hence presumably highly structured) N-terminal domain, flanked by an intrinsically unstable C-terminal region. The most upstream part of such a region (positions 284–372) is important for enzyme stabilization, whereas its most terminal part (positions 373–485) appears to be required to determine enzyme responsiveness to a broader range of heavy metals. Phytochelatin synthase (PCS) is a major determinant of heavy metal tolerance in plants and other organisms. No structural information on this enzyme is as yet available. It is generally believed, however, that the active site region is located in the more conserved N-terminal portion of PCS, whereas various, as yet unidentified (but supposedly less critical) roles have been proposed for the C-terminal region. To gain insight into the structural/functional organization of PCS, we have conducted a limited proteolysis analysis of the enzyme from Arabidopsis (AtPCS1), followed by functional characterization of the resulting polypeptide fragments. Two N-terminal fragments ending at positions 372 (PCS_Nt1) and 283 (PCS_Nt2) were produced sequentially upon V8 protease digestion, without any detectable accumulation of the corresponding C-terminal fragments. As revealed by the results of in vivo and in vitro functional assays, the core PCS_Nt2 fragment is biosynthetically active in the presence of cadmium ions and supports phytochelatin formation at a rate that is only ∼5-fold lower than that of full-length AtPCS1. The loss of the C-terminal region, however, substantially decreases the thermal stability of the enzyme and impairs phytochelatin formation in the presence of certain heavy metals (e.g. mercury and zinc, but not cadmium or copper). The latter phenotype was shared by PCS_Nt2 and by its precursor fragment PCS_Nt1, which, on the other hand, was almost as stable and biosynthetically active (in the presence of cadmium) as the full-length enzyme. AtPCS1 thus appears to be composed of a protease-resistant (and hence presumably highly structured) N-terminal domain, flanked by an intrinsically unstable C-terminal region. The most upstream part of such a region (positions 284–372) is important for enzyme stabilization, whereas its most terminal part (positions 373–485) appears to be required to determine enzyme responsiveness to a broader range of heavy metals. Living organisms have developed a variety of mechanisms to counteract the toxic effects of heavy metal (HM) 1The abbreviations used are: HM, heavy metal; PCS, phytochelatin synthase; AtPCS1, phytochelatin synthase 1 from A. thaliana; PC, phytochelatin; GSH, reduced glutathione; MT, metallothionein; Nt, N-terminal; Ct, C-terminal; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight. ions. The most common is perhaps HM chelation and sequestration by metal binding (poly)peptides such as metallothioneins (MTs) and phytochelatins (PCs) (1Cobbett C. Goldsbrough P. Annu. Rev. Plant. Biol. 2002; 53: 159-182Crossref PubMed Scopus (2028) Google Scholar). The latter are (λ-Glu-Cys)n-X peptides, in which n usually ranges from 2 to 5 and X is commonly a Gly, but can also be a β-Ala, Glu, or Ser residue (2Rauser W.E. Cell Biochem. Biophys. 1999; 31: 19-48Crossref PubMed Scopus (558) Google Scholar). In plants and other organisms, PCs are synthesized from glutathione (GSH) by the enzyme phytochelatin synthase (PCS) (3Zenk M.H. Gene (Amst.). 1996; 179: 21-30Crossref PubMed Scopus (912) Google Scholar). As revealed by the pioneering studies of Zenk and coworkers (4Grill E. Loeffler S. Winnacker E.L. Zenk M.H. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 6838-6842Crossref PubMed Google Scholar), PCS is a cytosolic λ-Glu-Cys dipeptidyl transpeptidase (EC 2.3.2.15) that catalyzes the transpeptidation of the λ-Glu-Cys moiety of GSH onto a second GSH molecule to form PC2 or onto a preformed PCn molecule to form the corresponding n+1 derivative. In both fission yeast and plants, the in vivo synthesis of PCs can be induced by a variety of metal ions, the most effective of which is cadmium (2Rauser W.E. Cell Biochem. Biophys. 1999; 31: 19-48Crossref PubMed Scopus (558) Google Scholar). Our understanding of PCS function, expression, and mechanism of action has been greatly advanced by the identification of specific Arabidopsis PCS mutants (cad-1) (5Howden R. Goldsbrough P.B. Andersen C.R. Cobbett C.S. Plant Physiol. 1995; 107: 1059-1066Crossref PubMed Scopus (439) Google Scholar) and by the isolation of PCS-encoding genes (or cDNAs) from fission yeast (6Ha S.B. Smith A.P. Howden R. Dietrich W.M. Bugg S. O'Connell M.J. Goldsbrough P.B. Cobbett C.S. Plant Cell. 1999; 11: 1153-1164Crossref PubMed Scopus (558) Google Scholar, 7Clemens S. Kim E.J. Neumann D. Schroeder J.I. EMBO J. 1999; 18: 3325-3333Crossref PubMed Scopus (532) Google Scholar), Arabidopsis thaliana (8Vatamaniuk O.K. Mari S. Lu Y.P. Rea P.A. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7110-7115Crossref PubMed Scopus (340) Google Scholar), Triticum aestivum (7Clemens S. Kim E.J. Neumann D. Schroeder J.I. EMBO J. 1999; 18: 3325-3333Crossref PubMed Scopus (532) Google Scholar), Brassica juncea (9Heiss S. Wachter A. Bogs J. Cobbett C. Rausch T. J. Exp. Bot. 2003; 54: 1833-1839Crossref PubMed Scopus (173) Google Scholar), and the nematode Caenorhabditis elegans (10Vatamaniuk O.K. Bucher E.A. Ward J.T. Rea P.A. J. Biol. Chem. 2001; 276: 20817-20820Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar, 11Clemens S. Schroeder J.I. Degenkolb T. Eur. J. Biochem. 2001; 268: 3640-3643Crossref PubMed Scopus (118) Google Scholar). PCS mutants have invariably been found to be sensitive to heavy metals (especially cadmium), thus supporting the notion of a direct role of PCs in intracellular HM sequestration and detoxification (1Cobbett C. Goldsbrough P. Annu. Rev. Plant. Biol. 2002; 53: 159-182Crossref PubMed Scopus (2028) Google Scholar, 12Sanità Di Toppi L. Prasad M.N.V. Ottonello S. Prasad M. Strzalka K. Physiology and Biochemistry of Metal Toxicity and Tolerance in Plants. Kluwer Academic Publishers, Dordrecht, The Netherlands2002: 59-93Crossref Google Scholar). In most organisms, PCS is expressed constitutively (6Ha S.B. Smith A.P. Howden R. Dietrich W.M. Bugg S. O'Connell M.J. Goldsbrough P.B. Cobbett C.S. Plant Cell. 1999; 11: 1153-1164Crossref PubMed Scopus (558) Google Scholar, 8Vatamaniuk O.K. Mari S. Lu Y.P. Rea P.A. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7110-7115Crossref PubMed Scopus (340) Google Scholar) and leads to PC accumulation within a few minutes from exposure to cadmium or other HMs (most notably, copper, silver, arsenic, mercury, zinc, and lead) (13Grill E. Winnacker E.L. Zenk M.H. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 439-443Crossref PubMed Scopus (698) Google Scholar, 14Scheller H.V. Huang B. Hatch E. Goldsbrough P.B. Plant Physiol. 1987; 85: 1031-1035Crossref PubMed Google Scholar). This indicates that PCS is primarily regulated by HM-dependent enzyme activation. Despite the chemical diversity of metals that can function as enzyme activators, such an activation has long been thought to rely on a direct interaction between PCS and metal ions. Evidence for a substantially different mechanism has been provided by the results of a study conducted with recombinant PCS from Arabidopsis (AtPCS1), which implied that metal binding to the enzyme is not primarily responsible for catalytic activation (15Vatamaniuk O.K. Mari S. Lu Y.P. Rea P.A. J. Biol. Chem. 2000; 275: 31451-31459Abstract Full Text Full Text PDF PubMed Scopus (311) Google Scholar). More specifically, it has been reported that a Cd·GS2 thiolate complex (or a Cd·PCn complex) and free GSH can act, respectively, as the λ-Glu-Cys acceptor and donor of AtPCS1-catalyzed dipeptidyl transfer. Only a 2-fold enhancement of the transpeptidation rate seems to be conferred by a direct interaction of the metal with AtPCS1, which binds approximately seven Cd2+ ions/monomer. One aspect of this model that has been challenged recently concerns the mechanism of termination. It has been argued that, because HM·PCn complexes are co-substrates for PCn+1 formation, it is not clear how PC synthesis can be terminated, at least under in vitro conditions, where there is an excess of free GSH, but no effective sink for product removal or sequestration (16Oven M. Page J.E. Zenk M.H. Kutchan T.M. J. Biol. Chem. 2002; 277: 4747-4754Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar). In addition, it has been found previously that even though PC synthesis can be induced in vivo by lead and mercury ions, the resulting HM·PC complexes mainly contain copper rather than the metal ions utilized for induction (17Maitani T. Kubota H. Sato K. Yamada T. Plant Physiol. 1996; 110: 1145-1150Crossref PubMed Scopus (198) Google Scholar). The availability of PCS coding sequences from a wide range of organisms has also led to a model for the functional organization of PCS, in which the more conserved N-terminal region is thought to correspond to the catalytic domain (18Cobbett C.S. Curr. Opin. Plant. Biol. 2000; 3: 211-216Crossref PubMed Scopus (435) Google Scholar). Although substantiated by the observation that mutations targeting the N-terminal portion of AtPCS1 (e.g. the cad1–3 mutant (5Howden R. Goldsbrough P.B. Andersen C.R. Cobbett C.S. Plant Physiol. 1995; 107: 1059-1066Crossref PubMed Scopus (439) Google Scholar)) dramatically reduce cadmium tolerance and in vivo PC production, this model is still largely hypothetical. In addition, the loss of part of the C-terminal region of AtPCS1, as in the cad1–5 mutant (5Howden R. Goldsbrough P.B. Andersen C.R. Cobbett C.S. Plant Physiol. 1995; 107: 1059-1066Crossref PubMed Scopus (439) Google Scholar), definitely leads to a defective cadmium tolerance (and impaired in vitro PC synthesis) phenotype. This argues that the putative C-terminal domain, which contains multiple cysteine residues, often arranged in pairs (or near pairs) as in MTs (18Cobbett C.S. Curr. Opin. Plant. Biol. 2000; 3: 211-216Crossref PubMed Scopus (435) Google Scholar), and for which diverse functions have been proposed in recent years, does somehow contribute to PCS functionality. Bacterial Expression and Purification AtPCS1—The AtPCS1 cDNA was isolated as the result of a multicopy suppression screening aimed at identifying Arabidopsis cDNAs (from a yeast expressible cDNA library cloned into the M. Plant J. Google Scholar)) of cadmium tolerance in the mutant a of S. of of and S. The coding region of AtPCS1 was the as a and the in the and digestion, the AtPCS1 was to the as an with a the was into was induced by 1 and to for at Bacterial were in 1 a protease and and by The was onto a in protease with and with the was with The and of were by on PubMed Scopus Google Scholar). was with the and as proteolysis of and V8 at a were in of and to at for of as in the of the were and by 5 in PubMed Scopus Google Scholar). or were to proteolysis was utilized for which were conducted in and at a were by by by with a to and by analysis on with the N-terminal and of AtPCS1 without any HM and of were as and the was onto AtPCS1 polypeptide fragments by were and to N-terminal analysis a and a were from in followed by with and were reduced with at followed by free cysteine by in the presence of at in the The from this was and the was by with and with In of polypeptide was by with in at The was and with a of the and the was for an at were by with and in at were with a of and in in was to the metal and at was were by the and were reported as of AtPCS1 fragments coding for were by the and as with the and The cDNA fragment was also under the the previously and the were cloned into the with and to the yeast digestion, the and cDNAs were into the the and previously to full-length AtPCS1 were utilized for and of both Tolerance in in vivo cadmium tolerance assays, the AtPCS1 cDNA and cloned in the yeast the full-length and the were into yeast the H. K. A. J. PubMed Google Scholar). were for to on yeast with and but and utilized for cadmium as previously L. A. Ottonello S. P. Plant Physiol. 2002; PubMed Scopus Google Scholar). the of PC the were to an of in with and for an at PCS and PC phytochelatin synthase were conducted at in GSH, and of the different PCS as in the the and of HMs or were also as in the aimed at formation, the as preformed PC2 by the of at a of of were with an of and by on an of was utilized for were by with and were GSH The was utilized for PC and in yeast which were by and in as previously J. M. P. H. J. Plant Physiol. PubMed Scopus Google Scholar). were from of PCS polypeptide with PubMed Scopus Google Scholar), by the of were of with an of 5 residue Only without in the AtPCS1 were for which the A. thaliana B. juncea T. aestivum and C. elegans The PCS coding from was by of the corresponding a B. juncea PCS of T. of Plant and were utilized with to the The that PCS be composed of an N-terminal catalytic domain and a C-terminal domain, role is but most was on the of a limited and of PCS from organisms as diverse as a plants, and a nematode (1Cobbett C. Goldsbrough P. Annu. Rev. Plant. Biol. 2002; 53: 159-182Crossref PubMed Scopus (2028) Google Scholar, C.S. Curr. Opin. Plant. Biol. 2000; 3: 211-216Crossref PubMed Scopus (435) Google Scholar). The from such an is but still largely and of the availability of a of PCS from plants, we within the PCS in a more and As revealed by this analysis organisms are the of is with the of even though the between is not with a of in from to of the Arabidopsis enzyme. It also be that the of different PCS ranges from to thus that a rather and perhaps C-terminal region is in PCS to This is by the observation that the of in the C-terminal region is to that of the N-terminal region, only PCS sequences are and even more such analysis is to plants and to important and conserved function of the C-terminal region, at least in specific of the domain, the an of the structural and functional organization of of AtPCS1 a of at the the was to near from an by for PCS was with the specific as the corresponding from which the been by not but To the structural organization of PCS, was to limited proteolysis with V8 both in the and in the presence of Cd2+ ions fragment in a in the presence of V8 Although this fragment was by the major product of AtPCS1 proteolysis at this of other were also One such with an of and a less fragment an which the limited proteolysis stable to at least 1 As in 2 and a was in the presence of The most with the conducted in the of was a reduced rate of full-length AtPCS1, which a and was still detectable even This was not to of the V8 because at the not of an such as not AtPCS1 is to Cd2+ ions at even in the of GSH (15Vatamaniuk O.K. Mari S. Lu Y.P. Rea P.A. J. Biol. Chem. 2000; 275: 31451-31459Abstract Full Text Full Text PDF PubMed Scopus (311) Google Scholar, T. C. S. 2003; PubMed Scopus Google Scholar), this most a of the enzyme. This is with the observation that the rate of formation of the fragment was than in 2 and and that effects to by cadmium were in the presence of copper, HM T. C. S. 2003; PubMed Scopus Google Scholar). To the of the protease-resistant of AtPCS1, the V8 and as as other less from a limited proteolysis were by and both to N-terminal and to in followed by for were to AtPCS1 on the of and of the of the was by into fragment from and the or of V8 fragments were found to the as the As in the polypeptide as at 372 to the AtPCS1 whereas the other major proteolysis the fragment as at No polypeptide corresponding to the C-terminal fragments to and PCS_Nt2 be even upon and analysis of proteolysis less polypeptide that were by It thus appears that region of AtPCS1, even in the presence of HM ions, is to and a N-terminal domain (PCS_Nt1) from a and unstable C-terminal region. at a second which upon of the the N-terminal core domain which at proteolysis 2 and In of AtPCS1 in a the functional of the major polypeptide fragments by limited we to of limited of however, in a of which a thus to of the AtPCS1 cDNA coding for the and PCS_Nt2 This the in E. and of both with no of (especially by the of an in vivo functional analysis of both N-terminal in of S. an E. that is of yeast mutant phenotype is by AtPCS1 (8Vatamaniuk O.K. Mari S. Lu Y.P. Rea P.A. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7110-7115Crossref PubMed Scopus (340) Google was used for aimed to insight into the cadmium detoxification and PC of AtPCS1 the of mutant with the or the PCS_Nt2 cDNA cloned into the multicopy yeast as as of full-length AtPCS1 and to on Cd2+ of different cDNA coding for the C-terminal region of the PCS_Nt2 site corresponding to the of the AtPCS1 was also in this As in conferred a Cd2+ tolerance only lower than that of full-length substantially lower Cd2+ tolerance was in PCS_Nt2 which, were from the in the presence of a Cd2+ as as a but in Cd2+ tolerance was the were on a Cd2+ As in were by the of PCs PC2 and produced by in the presence of This that cadmium by the fragments is to PC and not to Cd2+ sequestration by the (and T. C. S. 2003; PubMed Scopus Google Scholar)) of The latter for which a in Cd2+ yet to any detectable of PC of the the of the of fragments to an Cd2+ tolerance and to PC synthesis in both cDNAs were into the into E. and used for the and of the corresponding The and were the as utilized to full-length As revealed by analysis of the corresponding were upon yet the of was substantially lower for than for full-length AtPCS1. of this in was to the of the and to to In as in a was revealed by an analysis conducted on a were in the of PCS_Nt1, and a of was in the of PCS_Nt2 This which is of the produced by limited to the of the polypeptide and to its to be into the PCS_Nt2 even upon in a and in the presence of protease This also the of the of recombinant which in was as the of The and with full-length AtPCS1, were for phytochelatin synthase in the presence of Cd2+ and As in both were of PCs to at least and the with full-length AtPCS1 to in the of the latter to substantially to at least To a more functional between the PCS the from the formation of multiple that in function as we on the rate of PC2 This was from PC by PCS at from to on the and of enzyme at which the of PC2 with without any accumulation of PCs and specific for PC2 formation that were only and ∼5-fold lower than the of full-length AtPCS1. In addition, as in also the of of the was found to be with that of the full-length of PC2 synthesis by AtPCS1 and its were at the of and GSH, in the presence of full-length AtPCS1 or PCS_Nt2 The of PC2 formation was at from to on the and of enzyme aimed to the accumulation of PC and HM of PCS PC2 of the core polypeptide is only and 2-fold lower than that of full-length AtPCS1 and yet it is less than both of enzyme in Cd2+ tolerance to yeast As by the reported in a for this diverse functional is that of supporting the in vitro synthesis of PC2 with an to that of enzyme be less active in the of this by the rate of formation by PCS in GSH and preformed Cd2+ ions under in vitro PC synthesis and the of PC2 as a As in however, the of PCS_Nt2 was found to be reduced by the as PC2 with the corresponding of full-length AtPCS1 or that in also for the of PCS_Nt2 to PC than is that have a reduced stability (and thus a in vivo and in vitro with the full-length enzyme. To this full-length AtPCS1 and its were at for of in an Cd2+ and followed by the of Cd2+ and GSH, and by the of PC under As in PCS_Nt2 to be the least stable of the PCS it more than of its a at and it was almost an the conditions, was only less stable than the full-length enzyme and of its even a at thus to an important role of the C-terminal region of AtPCS1, the between positions and HM binding is the only other function of the C-terminal portion of PCS that has been In in a recent T. C. S. 2003; PubMed Scopus Google Scholar) seven HM binding in PCS with of which are located in the corresponding C-terminal portion of AtPCS1 that is in also that of such HM binding has no major on in vivo HM and thus presumably on PC It was however, that of have a for cadmium than for thus to AtPCS1 and the loss of HM binding does perhaps catalytic in the presence of HM ions other than different heavy copper, zinc, and mercury, to PC were for this As in AtPCS1 and its were in the presence of PC was in of the in the presence of Cd2+ ions. however, PC by and was impaired with full-length AtPCS1, in or as the HM ions and the of PCS_Nt2 in the presence of was only of the in the presence of as to an of full-length AtPCS1 in the presence of ions that was only lower than that by Cd2+ ions. results were from conducted in the presence of a HM of not The of with and in the presence of or were almost by the of the PCS were at with or in the of GSH, followed by the of GSH and PC was the as that in conducted in the presence of Cd2+ as the HM and without any This indicates that impaired in the presence of or is not to of by The HM of PCS revealed by does not with the chemical of the different In the HMs we have are largely metal ions. and are also in the of is to that of whereas the of is almost the as the and are metal ions, with in that different to PCS be to different it is that in of interaction with GSH with specific metal binding on the In the of the full-length are by no more than were with that the C-terminal region of PCS is somehow for supporting a range to HM ions. Despite in and different thermal the the HM more in this is the that, PCS_Nt2 the C-terminal HM binding by analysis T. C. S. 2003; PubMed Scopus Google Scholar), of are in the however, that an putative HM binding which is in AtPCS1 positions and is conserved in most PCS sequences (especially from but not in the enzyme by T. C. S. 2003; PubMed Scopus Google Scholar), is from both and It thus be to this site is responsible for the HM responsiveness of The isolation of PCS-encoding genes and cDNAs has the to studies aimed at understanding the structural organization and mechanism of action of this enzyme at both the and The of PCS, however, is not yet and to enzyme with have on mainly from studies and