Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Journal of Experimental Botany, Vol. 55, No. 404, Sulphur Metabolism in Plants Special Issue, pp. 1775–1783, August 2004 DOI: 10.1093/jxb/erh185 Advance Access publication 18 June, 2004 Plant adenosine 59-phosphosulphate reductase: the past, the present, and the future Stanislav Kopriva1,* and Anna Koprivova2 1 Albert-Ludwigs-University of Freiburg, Institute of Forest Botany and Tree Physiology, Georges-Köhler-Allee 053, D-79110 Freiburg, Germany 2 Albert-Ludwigs-University of Freiburg, Plant Biotechnology, Schänzlestr. 1, D-79104 Freiburg, Germany Received 26 January 2004; Accepted 27 April 2004 Abstract The sulphate assimilation pathway provides reduced sulphur for the synthesis of the amino acids cysteine and methionine. These are the essential building blocks of proteins and further sources of reduced sulphur for the synthesis of coenzymes and various secondary compounds. Several recent reports identified the adenosine 59-phosphosulphate reductase (APR) as the enzyme with the greatest control over the pathway. In this review, a short historical excursion into the investigations of sulphate assimilation is given with emphasis on the proposed alternative pathways to APR, via ‘bound sulphite’ or via PAPS reductase. The evolutionary past of APR is reviewed, based on phylogenetic analysis of APR and PAPS reductase sequences. Furthermore, recent biochemical analyses of APR that identified an iron–sulphur centre as a cofactor, proposed functions for different protein domains, and addressed the enzyme mechanism are summarized. Finally, questions that have to be addressed in order to improve understanding of the molecular mechanism and regulation of APR have been identified. Key words: Adenosine phosphosulphate reductase, cysteine biosynthesis, iron–sulphur centre, sulphate assimilation. lipids. In plants, in addition, a variety of S-containing secondary metabolites are synthesized, which often play an important role in defence against pathogens and herbivores. For incorporation into bio-organic compounds sulphate has to be reduced in the pathway of assimilatory sulphate reduction. This pathway is present in plants, algae, fungi, and autotrophic prokaryotes, but is missing in Metazoa and most parasitic prokaryotes and eukaryotes. Plant sulphate assimilation is the major source of reduced sulphur for animal and human diets. Nevertheless, the investigations of sulphur metabolism were always fewer compared with research on carbohydrates or nitrogen assimilation and, for a long time, a consensual view on the pathway was missing (Schmidt and Jäger, 1982). However, in the last few years substantial progress in understanding plant sulphate assimilation has been achieved (Leustek and Saito, 1999; Leustek et al., 2000; Hawkesford and Wray, 2000; Suter et al., 2000; Höfgen et al., 2001). Several recent reports identified adenosine 59-phosphosulphate reductase (APR) as the key enzyme in this pathway and demonstrated many important results on its biochemistry and regulation (Suter et al., 2000; Bick et al., 2001; Kopriva et al., 2001). Therefore, in this review, a short historical excursion into the investigations of sulphate assimilations is provided, with a focus on adenosine 59-phosphosulphate reductase, its biochemical characterization, and evolution, and finally questions on the understanding of the molecular mechanism and regulation of this enzyme are summarized. Introduction Historical overview Sulphur is found in the amino acids cysteine and methionine that are essential components of peptides and proteins, in iron–sulphur clusters and other cofactors, and as sulphonate group modifying proteins, polysaccharides, and The sulphate assimilation pathway was first resolved in the enteric bacteria Escherichia coli and Salmonella typhimurium using mutants auxotrophic for different sulphur compounds (Jones-Mortimer, 1968; Kredich, 1971). To be * To whom correspondence should be addressed. Fax: +49 761 2038302. E-mail: [email protected] Journal of Experimental Botany, Vol. 55, No. 404, ª Society for Experimental Biology 2004; all rights reserved 1776 Kopriva and Koprivova metabolized, the relatively inert sulphate must first be activated. The initial activation step, an adenylation to adenosine 59-phosphosulphate (APS) is catalysed by ATP sulphurylase (ATPS). In the second step APS is further phosphorylated by APS kinase to form adenosine 39phosphate 59-phosphosulphate (PAPS). PAPS is reduced in a thioredoxin-dependent reaction by PAPS reductase (CysH) to sulphite. In the second reduction step, sulphite is reduced by a NADPH-dependent sulphite reductase to sulphide. In bacteria, sulphide is finally incorporated into the amino acid skeleton of O-acetyl-L-serine (OAS) forming cysteine. It has long been known that plants are able to take up sulphate and convert it into cysteine (reviewed in Wilson, 1962). Sulphite and sulphide were identified as intermediates in sulphate reduction and the pathway was shown to be stimulated by light (Fromageot and Perez-Milan, 1959; Asahi, 1960). PAPS had been detected in plants (Schiff, 1959); therefore, in analogy with enteric bacteria, the same sequence of reactions was proposed for plant sulphate assimilation (Fig. 1). However, studies with the green alga Chlorella revealed that APS rather than PAPS was reduced (Tsang et al., 1971; Schmidt, 1972a). Among the products of this reaction, sulphite bound onto a thiol carrier, which could be replaced in vitro by glutathione (GSH), was most prominent (Schiff and Hodson, 1970; Schmidt, 1972b). The bound sulphite can be reduced further to bound sulphide by a ferredoxin-dependent thiosulphonate reductase, and react with OAS to form cysteine (Schmidt, 1973). The enzyme catalysing the transfer of sulphate from APS to the thiol carrier was named APS sulphotransferase (APSST) (Schmidt, 1972b). APSST activity was detected in a variety of plants and photosynthetic bacteria (Schmidt, 1975; Schmidt and Trüper, 1977). Therefore, and because its activity was highly regulated, APSST was considered to represent the major sulphate-reducing enzyme in photosynthetic organisms, by contrast with enteric bacteria and yeast which seemed to utilize PAPS for reduction (Brunold, 1990; Schmidt and Jäger, 1992). Nevertheless, a PAPS- Fig. 1. Assimilatory sulphate reduction pathway in Escherichia coli. dependent pathway of plant sulphate assimilation was never excluded, especially when the purification of PAPS reductase from spinach had been reported (Schwenn, 1989). In attempts to clone plant APS and/or PAPS reducing enzyme(s) by complementation of E. coli mutants deficient in cysH, three different cDNA clones were obtained from Arabidopsis thaliana (Gutierrez-Marcos et al., 1996; Setya et al., 1996). These cDNAs encoded isoforms of a novel protein with three distinct domains: an N-terminal organelle targeting peptide, a central part homologous to E. coli PAPS reductase, and a C-terminal extension similar to thioredoxin. Since the enzyme produced sulphite from APS but not from PAPS, it was named APS reductase (APR) (Setya et al., 1996). In the following years several reviews tried to establish the role of APR in sulphate assimilation, but came to contradictory conclusions (Hell, 1997; Wray et al., 1998; Leustek and Saito, 1999). Whereas Wray et al. (1998) speculated that APS is directly reduced to free sulphite and further to free sulphide by SiR, Leustek and Saito (1999) revived the bound sulphite pathway, assuming that APR acts as a sulphotransferase forming S-sulphoglutathione as a reaction product. Sulphide may be formed in a reaction catalysed by SiR from free sulphite resulting from a nonenzymatic reaction of S-sulphoglutathione with GSH. The controversy about the enzyme catalysing the reduction of APS was resolved by Suter et al. (2000). The APSST from Lemna minor was isolated and the corresponding cDNA was cloned. The deduced amino acid sequence was 73% identical to that of APS reductase from A. thaliana, revealing that APS sulphotransferase and APS reductase were identical enzymes. This finding itself, however, could not clarify the controversy about the enzyme name and reaction mechanism. The method of measuring APR activity (Brunold and Suter, 1990) cannot distinguish between a free sulphite and sulphite bound to a thiol as the reaction products. Thus, it is compatible with both the free and the bound sulphite pathways and, correspondingly, with both reductase and sulphotransferase reaction mechanisms. From the published kinetic data neither mechanism could be excluded (Schmidt, 1972b; Abrams and Schiff, 1973; Schmidt and Jäger, 1992; Li and Schiff, 1992). The analysis of reaction products of the recombinant APR from L. minor, however, revealed that free sulphite was the only reaction product detected under non-oxidizing conditions, while S-sulphoglutathione was only formed when oxidized glutathione was present in the enzyme assay (Suter et al., 2000). This means that the enzyme is a reductase and that the bound sulphite pathway was most probably an artefact due to the presence of oxidized thiols in the reaction assay. These conclusions were corroborated by further biochemical characterization of APR (Weber et al., 2000; Kopriva et al., 2001). The major uncertainty in the pathway of sulphate assimilation in plants was, therefore, the presence or absence of PAPS reductase. The Arabidopsis and rice genomes Plant APS reductase do not contain any genes homologous to E. coli PAPS reductase, other than those encoding APR. However, since such plant PAPS reductase may have a structure completely divergent from that of the bacterial enzyme, only analysis of plants lacking APR activity would prove or exclude the PAPS-dependent sulphate assimilation in plants. In higher plants, neither mutants nor transgenic plants devoid of APR activity are known. For targeted gene disruption, the moss Physcomitrella patens is an ideal system because of high rates of homologous recombination (Schaefer, 2002). To produce and analyse plants lacking APR activity, the single copy gene encoding APR was cloned from P. patens and disrupted (Koprivova et al., 2002). This resulted in the complete loss of the correct transcript and enzymatic activity. However, surprisingly, the knockout plants were still able to grow on sulphate as the sole sulphur source. Although PAPS reductase activity could not be measured in the knockout plants, a cDNA and gene coding for this enzyme were detected in P. patens. The PAPS reductase from Physcomitrella does not contain the thioredoxin-like domain, the two cysteine pairs binding the FeS cofactor in APR are absent, and the protein seems to contain a chloroplast targeting peptide (Koprivova et al., 2002). The moss Physcomitrella patens is the first plant species where PAPS reductase was confirmed on the molecular level and also the first organism where both APS and PAPS-dependent sulphate assimilation co-exist. Figure 2 shows the new, now generally accepted, scheme of the sulphate assimilation pathway in plants. Sulphate is adenylated by ATP sulphurylase to APS, which is reduced by APS reductase to free sulphite; the electrons are derived from glutathione. APS can also be further phosphorylated to PAPS, which serves as a source of activated sulphate for a variety of sulphotransferases. At least in some lower plants PAPS can be directly reduced to sulphite by PAPS reductase. The significance of this reaction in higher plants is uncertain, which is demonstrated by a dashed line in the Fig. 2. Assimilatory sulphate reduction pathway in higher plants, current view. 1777 scheme. Sulphite is reduced by a ferredoxin-dependent sulphite reductase to sulphide and incorporated into the amino acid skeleton of OAS to form cysteine. Biochemical characterization of APR APR catalyses a thiol-dependent two-electron reduction of APS to sulphite (Suter et al., 2000). APR activity is measured as acid volatile radioactivity after incubation with [35S]APS and thiols. The sulphite evolved after the addition of H2SO4 is trapped in triethanolamine and the radioactivity is determined in the scintillation counter (Brunold and Suter, 1990). The reaction has a pH optimum of 8.5–9, is stimulated by concentrations of sulphate higher than 0.6 M, and is inhibited by 59-AMP (Brunold and Suter, 1990; Setya et al., 1996). The enzyme was first purified from the green alga Chlorella as a protein of molecular weight greater than 300 kDa (Schmidt, 1972b). However, because of its limited amount in cells it has not often been studied in detail. Li and Schiff (1991) extracted APR from Euglena gracilis and purified to homogeneity a tetramer of 25 kDa subunits held together by disulphide bonds. The enzymes from Chlorella and Euglena differed not only in their molecular weights, but also in the efficiency of APS reduction using different thiol compounds as electron donors. While APR from Chlorella utilized GSH and dithiothreitol (DTT) with the same efficiency, the activity of Euglena APR with GSH reached only 23% of the activity with DTT (Schmidt, 1972b; Li and Schiff, 1991). A further source for APR purification was the marine macroalga Porphyra yezoensis where the subunit Mr was estimated to be 43 000 (Kanno et al., 1996). The molecular weight calculated from the deduced amino acid sequence of the APR cDNAs from Arabidopsis and Lemna was also 43 000 (Gutierrez-Marcos et al., 1996; Setya et al., 1996; Suter et al., 2000). The molecular mass of native Lemna APR (Suter et al., 2000) and recombinant APR2 from A. thaliana (M Weber and S Kopriva, unpublished data) was estimated to be 91 kDa, revealing that, in higher plants, the enzyme is a dimer of 43 kDa subunits. The APR monomers seem to be joined by a disulphide bond, because after treatment with thiols APR elutes from the column as monomer. This disulphide is most probably formed by the only Cys conserved among APS and PAPS reductases (Cys248 of the ECGLH motif in APR2), since the molecular weight of modified APR where this Cys was exchanged with Ser also corresponded with that of an APR monomer (M Weber and S Kopriva, unpublished results). Accordingly, after pre-incubation with thiols (DTT, GSH) in the absence of APS or AMP, APR loses its activity (S Kopriva, unpublished results; Bick et al., 2001). On the other hand, APR expressed in E. coli lacking a thioredoxin reductase (trxB) was 11–45-fold more active than the enzyme expressed in the wild-type strain (Bick et al., 2001). Only the APR1 isoform of A. thaliana, but not APR2 and APR3, 1778 Kopriva and Koprivova was activated in the trxB strain, revealing possible isoformspecific differences. Bick et al. (2001) concluded from their data that a regulatory Cys pair is present in APR1, which enables rapid post-translational regulation of APR activity in vivo. However, a change between an active dimer and inactive monomer might be an alternative explanation. Localization of APR Sulphate reduction was demonstrated in isolated chloroplasts (Schmidt and Trebst, 1969) and, correspondingly, in spinach the APR activity was localized in chloroplasts (Schmidt, 1976; Fankhauser and Brunold, 1978). Western analysis of pea chloroplast fractions detected APR protein in the stroma, but not in any of the membrane fractions (Prior et al., 1999) and immunogold labelling revealed APR in chloroplasts of three Flaveria species with different types of photosynthesis (Koprivova et al., 2001). The comparison of the N-terminal sequence of purified APR from Lemna with the amino acid sequence deduced from the cDNA confirmed the presence of a chloroplast targeting peptide encoded by the full-length cDNA (Suter et al., 2000). Indeed, recombinant APR from Catharanthus roseus could be imported into intact pea chloroplasts and correctly processed there (Prior et al., 1999). APR is composed from two distinct domains The mature APR consists of two distinct domains, a PAPS reductase-like one and a thioredoxin-like one. When a truncated APR lacking the thioredoxin-like domain was expressed in E. coli and purified, low or no APR activity was measured (Bick et al., 1998; Prior et al., 1999; Weber et al., 2000). Adding the corresponding, separately expressed, thioredoxin-like domain reconstituted APR activity (Bick et al., 1998; Prior et al., 1999). The activity could be recovered after the addition of thioredoxin (Prior et al., 1999; Weber et al., 2000). Only thioredoxin m, but not thioredoxin f, was able to reconstitute the activity of the A. thaliana APR2 isoform (Weber et al., 2000). The CXXC thioredoxin active site motif found in the C-terminal domain of APR, however, also occurs in glutaredoxin, another class of thiol:disulphide oxidoreductases. Glutaredoxin does not require an enzymatic reduction, like thioredoxin by thioredoxin reductase, but is reduced by glutathione, whose oxidized form must, in turn, be reduced by glutathione reductase (Rietsch and Beckwith, 1998). Glutaredoxin was identified as an alternative electron donor for sulphate reduction in an E. coli mutant lacking thioredoxin (Tsang and Schiff, 1978). Glutaredoxin, but not thioredoxin, also catalyses the reduction of dehydroascorbate (Washburn and Wells, 1999). Although the sequence of the C-terminal domain of APR is more similar to thioredoxin than to glutaredoxin, it possesses the enzymatic activity of glutaredoxin, since (i) glutathione is an efficient hydrogen donor for APR in vitro (KM=0.6–3 mM) (Schmidt, 1972b; Bick et al., 1998; Prior et al., 1999), (ii) APR and the C-terminal domain catalyse the glutathione-dependent reduction of dehydroascorbate and hydroxyethyldisulphide and possesses ribonucleotide reductase and protein disulphide isomerase activities (Bick et al., 1998; Prior et al., 1999), (iii) the C-terminal domain can complement E. coli mutants lacking glutaredoxin, but not those defective in thioredoxin (Bick et al., 1998), and (iv) in the C-terminal domain of APR from the moss Physcomitrella patens the CXXC motif is changed to CXXS, an alteration tolerated by glutaredoxin but not thioredoxin (Koprivova et al., 2002). The two domains of APR can not only be expressed separately and possess independent enzyme activities; they also seem to catalyse distinct steps in APS reduction. The first approach to study the molecular mechanism of the APR reaction led to the identification of a reaction intermediate, with sulphite covalently bound to a cysteine residue conserved between APS and PAPS reductases (Weber et al., 2000). The [35S]sulphite was lost from the protein by reaction with reduced thiols or sulphite, but also by incubation at 37 8C without the addition of thiols. As truncated APR2, missing the C-terminal domain, remained labelled at 37 8C, the thioredoxin-like domain seems to be required to release the bound SO2ÿ 3 from the reaction intermediate. In analogy with the S-sulphoglutathione discussion, it has to be noted that the reaction intermediate was formed even in conditions where APR was inactive (pH 6) and, thus, it could not result from a chemical reaction between free sulphite and an enzyme dimer (Weber et al., 2000). The APR reaction can thus be divided into three independent steps: a transfer of sulphate from APS to the active cysteine residue, the release of the sulphite by the C-terminal domain, probably by forming a intramolecular disulphide bridge, and recovery of the active enzyme dimer by reaction with thiols (Fig. 3). For the first step, which results in binding of sulphite to APR, neither the thioredoxin-like domain nor an external electron source is required and also the electrons necessary for the sulphite release are provided by the APR protein. The molecular mechanisms of these reaction steps are, however, far from being understood. The identification of this reaction intermediate thus corroborated the reductase mechanism of APR. Similar, but not identical, intermediates as found in the APR reaction, with sulphite covalently bound to a protein were described previously (Abrams and Schiff, 1973; Schmidt, 1973; Li and Schiff, 1992). Schmidt (1973) and Abrams and Schiff (1973) obtained a low molecular weight protein with bound sulphite in Chlorella extracts only in the presence of thiols, giving the bound sulphite pathway its name. In addition, APR from Euglena gracilis was shown to form a stable reaction intermediate. The enzyme, however, was not characterized further; but it differs in composition from the higher plant APR (Li and Schiff, 1992). By contrast, the reaction mechanism of yeast Plant APS reductase 1779 Fig. 3. Reaction steps in APS reduction by APR. The squares and circles represent the N-terminal and C-terminal domains, respectively. The active site cysteine residue, as well as the Cys possibly forming an intramolecular disulphide, are indicated by S. Fig. 4. Neighbor–Joining tree of APR and PAPS reductase protein sequences. The sequences were retrieved from GenBank, aligned with Clustal W, and the tree was constructed with the MEGA 2.1 software. The accession numbers for the protein sequences used are as follows (from the top of the tree): T49106, P92979, P92981, AAB05871, AAF18999, CAB65911, AY594279, CAD22096, AAM18118, T52251, O05927, AAF23174, AAR35093, CAD16130, BX640413 (DNA sequence), P56891, AAK86625, D90470, AAL64292, P71752, CAC33945, CAD75117, AAO05823, P94498, AAF28889, P18408, CAE76190, CAD32963, AY594280, AAP99141, CAE08679, P72794, Q8P607, P57501, P17853, P17854. The horizontal line divides the tree in the APR clade and the PAPS reductase clade. Asterisk marks the bispecific sulphonucleotide reductase from Bacillus subtilis (Berndt et al., 2004). PAPS reductase has the following sequence of reactions: an interaction of the enzyme with reduced thioredoxin, storage of two electrons probably at the conserved Cys residues of the enzyme dimer, interaction with PAPS, and release of sulphite and adenosine 39,59-bisphosphate (Schwenn et al., 1988; Berendt et al., 1995). APR contains a FeS cluster APRs from A. thaliana and C. roseus which had been expressed in E. coli, were described as proteins without prosthetic groups or cofactors (Gutierrez-Marcos et al., 1996; Setya et al., 1996; Prior et al., 1999). On the other hand, APR was purified from L. minor as a yellow-brown protein (Suter et al., 2000) indicating the presence of a cofactor, possibly FAD or/and an FeS cluster. UV/visible spectra of recombinant APR from Lemna and all three isoforms from A. thaliana indicated the presence of an iron–sulphur centre and, indeed, iron and acid-labile sulphide were found to bind to the APR protein (Kopriva et al., 2001). Electron paramagnetic resonance and Mössbauer spectroscopy confirmed the presence of a diamagnetic [4Fe–4S]2+ cluster. This cluster is unusual since only three of the iron sites exhibited the same Mössbauer parameters, one of these being either tetragonally FeS3X co-ordinated, with X representing a non-sulphur ligand (C, N, O), or trigonally sulphur-co-ordinated. There is no signature for binding of an FeS cluster in the sequence of plant APR. However, the major difference between the N-terminal part of APR and PAPS reductases is the presence of two additional cysteine pairs in the plant enzyme. Only three from these four additional cysteine residues can bind the FeS cluster, thus explaining its extraordinary characteristics (Kopriva et al., 2001). It has to be noted that dissimilatory APS reductase, an enzyme catalysing the reduction of APS in anaerobic, sulphate-reducing bacteria such as Desulphovibrio sp., also carries two [4Fe–4S] centres in addition to FAD. The structure of dissimilatory APR is, however, completely different from that of plant APR. The protein is a dimer of a large FAD-containing subunit with a small one possessing two [4Fe–4S] clusters (Verhagen et al., 1994); there is no sequence homology among these proteins. Thus, plant APR represents a novel type of APS reductase with a [4Fe–4S] centre as the sole cofactor. Bacterial assimilatory APR Recently, another type of APS reductase was characterized in several bacteria. Although, as discussed above, it was long believed that heterotrophic bacteria reduce sulphate via PAPS, an APS-dependent sulphate assimilation was confirmed in several species of Rhizobiaceae, Pseudomonas, Burkholderia, Ralstonia, Mycobacterium, and Bacillus (Abola et al., 1999; Bick et al., 2000; Kopriva et al., 2002; Williams et al., 2002). The corresponding enzymes, encoded by CysH genes, are more similar to the N-terminal part of plant APR than to PAPS reductase of E. coli and are lacking the thioredoxin-like domain. Accordingly, the APR activity of these proteins is highly stimulated by the addition of thioredoxin. The bacterial assimilatory APR 1780 Kopriva and Koprivova also possesses the two Cys pairs conserved with the plant enzyme which are not found in PAPS reductase from E. coli. As these Cys residues participate in binding of the iron–sulphur cluster in plant APR (Kopriva et al., 2001), it is not surprising that the APRs from Sinorhizobium meliloti and Pseudomonas aeruginosa bind an FeS centre with the same characteristics as the FeS cluster in plant APR (Kopriva et al., 2002). Very recently, the CysH gene product from Bacillus subtilis was characterized as a bifunctional APS/PAPS reductase (Berndt et al., 2004). This protein has higher affinity for PAPS than for APS and requires thioredoxin or glutaredoxin for catalytic activity. Most importantly, this sulphonucleotide reductase also binds the FeS cofactor, which is rapidly lost upon exposure to air. After chemical removal of the cofactor APS reduction was not detectable, whereas residual PAPS reductase activity could still be measured (Berndt et al., 2004). Reduction of APS versus PAPS thus seems to be dependent on the FeS cluster and the two conserved Cys pairs seem to represent a marker to distinguish CysH homologues capable of APS reduction. The ability of the B. subtilis APS/PAPS reductase to react with both sulphonucleotides seems to be an exception, since the recombinant APRs from P. aeruginosa and S. meliloti were clearly monospecific for APS (Abola et al., 1999; Bick et al., 2000; Kopriva et al., 2002). If any of the other bacterial assimilatory APRs are bispecific like the B. subtilis enzyme, what structural features are the determinants of this bispecificity, and which sulphonucleotide is actually reduced in vivo remains to be elucidated. Evolution of APR The evolutionary past of APR is anything but straightforward, especially after cloning a PAPS reductase from the moss P. patens (Koprivova et al., 2002). The plant APR gene most probably originated from a fusion between prokaryotic genes for APS or PAPS reductase and thioredoxin. Since all APRs isolated or cloned from higher and lower plants, and the APR from the green algae Chlamydomonas rheinhardtii (Ravina et al., 2002) and Enteromorpha intestinalis (Gao et al., 2000), have the same structure, this fusion must have occurred early in the evolution of plants. However, the origin of the prokaryotic (P)APS reductase gene involved in this fusion, the existence of at least four different enzymes catalysing reduction of activated sulphate (plant APR, assimilatory bacterial APR, dissimilatory APR, and PAPS reductase), and the distribution of these enzymes in different organisms are not clear, yet. Sulphate assimilation is present in plants, algae, fungi, and many autotrophic prokaryotes. Among the 123 species of the Archae and Eubacteria superkingdoms, whose genomes have been completely sequenced to date, only 64 possess homologues of APR or PAPS reductase. Among these 64 species, an assimilatory APR (including the bispecific APS/PAPS reductase from B. subtilis) is present in 38 species whereas PAPS reductase can be found in 26 species, when the presence of the two Cys pairs serves as a marker to distinguish between these two proteins (Kopriva et al., 2002; see above). Utilization of APS for sulphate reduction is thus much more widely distributed than it was originally believed. During the search of sequence databases for homologues of plant APR and PAPS reductase a putative PAPS reductase was identified in the EST collection from the spike moss Selaginella lepidophylla, a vascular plant (S Kopriva, unpublished results). This is of great importance because it proves that the PAPS reductase found in P. patens (Koprivova et al., 2002) did not originate from a late horizontal gene transfer to Bryophytes, but must have been present in the common ancestor of Bryophytes and vascular plants. The APR amino acid sequences from different plant and algae species are well conserved, containing 60–80% identical amino acid residues. The PAPS reductase-like domains of plant APR are 22–27% identical to the PAPS reductase from E. coli or yeast and 40–50% to the assimilatory bacterial APR. Figure 4 shows a Neighbor–Joining tree constructed from representative APR and PAPS reductase sequences, retrieved from the GenBank. As described in Kopriva et al. (2002), using a larger but less diverse set of sequences, the tree is divided into two major clades. The first clade contains plant and algae APRs, together with many putative and confirmed assimilatory bacterial APRs and the bispecific B. subtilis enzyme, the other clade comprises PAPS reductases from the two plant species, fungi, enteric bacteria, and cyanobacteria. In all organisms of the ‘APR’ clade tested to date, APR activity was detected in protein extracts or in corresponding recombinant proteins, whereas the organisms of the ‘PAPS reductase’ clade seem to possess the PAPS reductase activity exclusively (Abola et al., 1999; Bick et al., 2000; Kopriva et al., 2002; Williams et al., 2002; Berndt et al., 2004). In a reciprocal approach, a partial APR gene was cloned from a cyanobacteria Plectonema 73110, which, in contrast to other cyanobacteria, possess an APS reducing enzyme (Schmidt, 1977) that clustered with the APS reductases from plants and several c-proteobacteria (Kopriva et al., 2002). As expected from the discussion above, all proteins from the ‘APR’ clade possess the two Cys pairs, which are not present in the enzymes of the ‘PAPS reductase’ clade. Although previously considered to be the major bacterial enzyme for the reduction of activated sulphate, PAPS reductase seems to be restricted to only few groups of cproteobacteria and cyanobacteria. As APS reducing species are found also in these phyla and since the APR tree topology (Fig. 4) does not correspond to the evolutionary relationship based on rRNA, a horizontal gene transfer must have played an important role in today’s distribution of the two enzyme activities. This is interesting, because Plant APS reductase also the evolution of dissimilatory APS reductase was affected by frequent horizontal gene transfers (Friedrich, 2002). From the two enzymes, APR seems to be evolutionarily older than PAPS reductase since (i) the PAPS reductase only occurs in two Eubacterial phyla, whereas APR was confirmed in most taxonomical groups of Eubacteria, (ii) dissimilatory sulphate-reducing bacteria and Archae possess APS reductase which contains the FeS cluster as a cofactor; and (iii) the activation to APS requires less energy than to PAPS (see discussion in Kopriva et al., 2002). The evolution of PAPS reductase, which does not need the iron–sulphur cluster, might have been an adaptation to an iron- and/or sulphur-poor environment. Open questions and future research needs Although much progress in the characterization of APR has been made in recent years, there are still many important questions to answer. To date, the exact reaction mechanism and the role of the FeS cofactor in catalysis are not known. For this purpose, a crystallization of the plant enzyme and a structure analysis would certainly deliver the necessary data. The evolutionary studies of APR and PAPS reductase would profit from an extended analysis of the distribution of these enzymes in lower plants, algae, and primitive Eukaryotes. Further characterization of the bifunctional B. subtilis enzyme will be very important for understanding the evolution of the two enzymes in bacteria. Although APR is highly regulated in plants (for a review see Leustek et al., 2000; Brunold et al., 2003; Kopriva and Koprivova, 2003), no immediate signals and transcription factors responsible for this regulation are known. Therefore, molecular analysis of APR regulation in plants and the biochemical analysis of the role of APR in the control of sulphate assimilation would be most important. Here, a genetic approach promises the most efficient way to address the APR regulation at the molecular level. Studying the latter question Vauclare et al. (2002) revealed that APR possesses a very high control over the flux through sulphate assimilation. APR would, therefore, be a good target for genetic engineering in order to obtain plants with a higher content of reduced sulphur for nutrition or increased stress tolerance. Indeed, it seems that increased APR activity results in greater flux through the pathway. Tsakraklides et al. (2002) showed that overexpression of APR in Arabidopsis led to higher contents of reduced sulphur compounds. The inorganic compounds sulphite and thiosulphate were increased to a greater extent than the thiols cysteine and glutathione. These results revealed that, for the synthesis of thiols, the availability of organic acceptors is as important as the production of sulphide by the sulphate assimilation pathway (Tsakraklides et al., 2002). Therefore, more extended flux control analysis using plant material with increased APR activity is needed to evaluate the contribution of sulphate reduction and the availability of 1781 carbohydrates in the control of cysteine synthesis and to suggest the most efficient way to produce plants with a maximal capacity for the synthesis of thiols. Thus, considering the central role of APR in sulphate assimilation and in the primary metabolism of plants, more about this interesting enzyme will surely be heard in future and the open questions will hopefully move from substantial ones to comparatively fine details. References Abola AP, Willits MG, Wang RC, Long SR. 1999. Reduction of adenosine-59-phosphosulfate instead of 39-phosphoadenosine-59phosphosulfate in cysteine biosynthesis by Rhizobium meliloti and other members of the family Rhizobiaceae. Journal of Bacteriology 181, 5280–5287. Abrams WR, Schiff JA. 1973. Studies of sulfate utilization by algae. II. An enzyme-bound intermediate in the reduction of adenosine59-phosphosulfate (APS) by cell-free extracts of wild-type Chlorella and mutants blocked for sulphate reduction. Archives of Mikrobiology 94, 1–10. Asahi T. 1960. Reduction of sulfite to sulfide in mung bean leaf. Journal of Biochemistry 48, 772–773. Berendt U, Haverkamp T, Prior A, Schwenn JD. 1995. Reaction mechanism of thioredoxin: 39-phosphoadenylylsulfate reductase investigated by site-directed mutagenesis. European Journal of Biochemistry 233, 356–437. Berndt C, Lillig CH, Wollenberg M, Bill E, Mansilla MC, de Mendoza D, Seidler A, Schwenn JD. 2004. Characterization and reconstitution of a 4Fe–4S adenylyl sulfate/phosphoadenylyl sulfate reductase from Bacillus subtilis. Journal of Biological Chemistry 279, 7850–7855. Bick JA, Åslund F, Chen Y, Leustek T. 1998. Glutaredoxin function for the carboxyl-terminal domain of the plant-type 59adenylylsulfate reductase. Proceedings of the National Academy of Sciences, USA 95, 8404–8409. Bick JA, Dennis JJ, Zylstra GJ, Nowack J, Leustek T. 2000. Identification of a new class of 59-adenylylsulfate (APS) reductases from sulfate-assimilating bacteria. Journal of Bacteriology 182, 135–142. Bick JA, Setterdahl AT, Knaff DB, Chen Y, Pitcher LH, Zilinskas BA, Leustek T. 2001. Regulation of the plant-type 59-adenylyl sulfate reductase by oxidative stress. Biochemistry 40, 9040–9048. Brunold C. 1990. Reduction of sulfate to sulfide. In: Rennenberg H, Brunold C, De Kok LJ, Stulen I, eds. Sulfur nutrition and sulfur assimilation in higher plants. The Hague, The Netherlands: SPB Academic Publishing, 13–31. Brunold C, Suter M. 1990. Adenosine 59-phosphosulfate sulphotransferase. In: Lea P, ed. Methods in plant biochemistry, Vol. 3. London: Academic Press, 339–343. Brunold C, Von Ballmoos P, Hesse H, Fell D, Kopriva S. 2003. Interactions between sulfur, nitrogen and carbon metabolisms. In: Davidian JC, Grill D, De Kok LJ, Stulen I, Hawkesford MJ, Schnug E, Rennenberg H, eds. Sulfur transport and assimilation in plants: regulation, interaction, and signaling. Leiden, The Netherlands: Backhuys Publishers, 45–56. Fankhauser H, Brunold C. 1978. Localization of adenosine 59phosphosulfate sulfotransferase in spinach leaves. Planta 143, 285–289. Friedrich MW. 2002. Phylogenetic analysis reveals multiple lateral transfers of adenosine-59-phosphosulfate reductase genes among sulfate-reducing microorganisms. Journal of Bacteriology 184, 278–289. 1782 Kopriva and Koprivova Fromageot P, Perez-Milan H. 1959. Reduction of sulfate to sulfite by tobacco leaf. Biochimica et Biophysica Acta 32, 457–464. Gao Y, Schofield OM, Leustek T. 2000. Characterization of sulfate assimilation in marine algae focusing on the enzyme 59-adenylylsulfate reductase. Plant Physiology 123, 1087–1096. Gutierrez-Marcos JF, Roberts MA, Campbell EI, Wray JL. 1996. Three members of a novel small gene-family from Arabidopsis thaliana able to complement functionally an Escherichia coli mutant defective in PAPS reductase activity encode proteins with a thioredoxin-like domain and ‘APS reductase’ activity. Proceedings of the National Academy of Sciences, USA 93, 13377–13382. Hawkesford M, Wray J. 2000. Molecular genetics of sulphate assimilation. Advances in Botanical Research 33, 159–223. Hell R. 1997. Molecular physiology of plant sulfur metabolism. Planta 202, 138–148. Höfgen R, Kreft O, Willmitzer L, Hesse H. 2001. Manipulation of thiol contents in plants. Amino Acids 20, 291–299. Jones-Mortimer MC. 1968. Positive control of sulphate reduction in Escherichia coli. Biochemical Journal 110, 589–595. Kanno N, Nagahaisa E, Sato M, Sato Y. 1996. Adenosine 59phosphosulfate sulfotransferase from the marine macroalga Porphyra yezoensis Ueda (Rhodophyta): stabilization, purification, and properties. Planta 198, 440–446. Kopriva S, Büchert T, Fritz G, et al. 2001. Plant adenosine 59phosphosulfate reductase is a novel iron–sulfur protein. Journal of Biological Chemistry 276, 42881–42886. Kopriva S, Büchert T, Fritz G, et al. 2002. The presence of an iron– sulfur cluster in adenosine 59-phosphosulfate reductase separates organisms utilizing adenosine 59-phosphosulfate and phosphoadenosine 59-phosphosulfate for sulfate assimilation. Journal of Biological Chemistry 277, 21786–21791. Kopriva S, Koprivova A. 2003. Sulphate assimilation: a pathway which likes to surprise. In: Abrol YP, Ahmad A, eds. Sulphur in higher plants. Dordrecht: Kluwer Academic Publishers, 87–112. Koprivova A, Melzer M, von Ballmoos P, Mandel T, Brunold C, Kopriva S. 2001. Assimilatory sulfate reduction in C3, C3–C4, and C4 species of Flaveria. Plant Physiology 127, 543–550. Koprivova A, Meyer A, Schween G, Herschbach C, Reski R, Kopriva S. 2002. Functional knockout of the adenosine 59phosphosulfate reductase gene in Physcomitrella patens revives an old route of sulfate assimilation. Journal of Biological Chemistry 277, 32195–32201. Kredich NM. 1971. Regulation of L-cysteine biosynthesis in Salmonella typhimurium. Journal of Biological Chemistry 246, 3474–3484. Leustek T, Martin MN, Bick JA, Davies JP. 2000. Pathways and regulation of sulfur metabolism revealed through molecular and genetic studies. Annual Review of Plant Physiology and Plant Molecular Biology 51, 141–165. Leustek T, Saito K. 1999. Sulfate transport and assimilation in plants. Plant Physiology 120, 637–644. Li JY, Schiff JA. 1991. Purification and properties of adenosine 59phosphosulphate sulphotransferase from Euglena. Biochemical Journal 274, 355–360. Li J, Schiff JA. 1992. Adenosine 59-phosphosulfate sulfotransferase from Euglena: enzyme-bound intermediates. Plant Cell Physiology 33, 63–72. Prior A, Uhrig JF, Heins L, Wiesmann A, Lillig CH, Stoltze C, Soll J, Schwenn JD. 1999. Structural and kinetic properties of adenylyl sulphate reductase from Catharanthus roseus cell cultures. Biochimica et Biophysica Acta 1430, 25–38. Ravina CG, Chang CI, Tsakraklides GP, McDermott JP, Vega JM, Leustek T, Gotor C, Davies JP. 2002. The sac mutants of Chlamydomonas reinhardtii reveal transcriptional and post- transcriptional control of cysteine biosynthesis. Plant Physiology 130, 2076–2084. Rietsch A, Beckwith J. 1998. The genetics of disulphide bond metabolism. Annual Review of Genetics 32, 163–184. Schaefer DG. 2002. A new moss genetics: targeted mutagenesis in Physcomitrella patens. Annual Review of Plant Biology 53, 477– 501. Schiff JA. 1959. Studies on sulfate utilization by Chlorella pyrenoidosa using sulfate-S-35: the occurrence of S-adenosyl methionine. Plant Physiology 34, 73–80. Schiff JA, Hodson RC. 1970. Pathway of sulfate reduction in algae. Annals of the New York Academy of Sciences 175, 555–576. Schmidt A. 1972a. Über Teilreaktionen der photosynthetischen Sulfatreduktion in zellfreien Systemen aus Spinatchloroplasten und Chlorella. Zeitschrift für Naturforschung 27b, 183– 192. Schmidt A. 1972b. On the mechanism of photosynthetic sulfate reduction. An APS-sulfotransferase from Chlorella. Archives of Microbiology 84, 77–86. Schmidt A. 1973. Sulfate reduction in a cell-free system of Chlorella. The ferredoxin-dependent reduction of a protein-bound intermediate by a thiosulfonate reductase. Archives of Microbiology 93, 29–52. Schmidt A. 1975. Distribution of the APS-sulfotransferase activity among higher plants. Plant Science Letters 5, 407–415. Schmidt A. 1976. The adenosine-59-phosphosulfate sulfotransferase from spinach (Spinacea oleracea L.). Stabilization, partial purification and properties. Planta 130, 257–263. Schmidt A. 1977. Assimilatory sulfate reduction via 39-phosphoadenosine-59-phosphosulfate (PAPS) and adenosine-59-phosphosulfate (APS) in blue-green algae. FEMS Microbiology Letters 1, 137–140. Schmidt A, Jäger K. 1992. Open questions about sulfur metabolism in plants. Annual Review of Plant Physiology and Plant Molecular Biology 43, 325–349. Schmidt A, Trebst A. 1969. The mechanism of photosynthetic sulfate reduction by isolated chloroplasts. Biochimica et Biophysica Acta 180, 529–535. Schmidt A, Trüper HG. 1977. Reduction of adenylylsulfate and 39phosphoadenylylsulfate in phototrophic bacteria. Experientia 33, 1008–1009. Schwenn JD. 1989. Sulfate assimilation in higher plants: a thioredoxin-dependent PAPS-reductase from spinach leaves. Zeitschrift für Naturforschung 44c, 504–508. Schwenn JD, Krone FA, Husmann K. 1988. Yeast PAPS reductase: properties and requirements of the purified enzyme. Archives of Microbiology 150, 313–319. Setya A, Murillo M, Leustek T. 1996. Sulphate reduction in higher plants—molecular evidence for a novel 59-adenylylsulfate reductase. Proceedings of the National Academy of Sciences, USA 93, 13383–13388. Suter M, von Ballmoos P, Kopriva S, Op den Camp R, Schaller J, Kuhlemeier C, Schürmann P, Brunold C. 2000. Adenosine 59phosphosulfate sulfotransferase and adenosine 59-phosphosulfate reductase are identical enzymes. Journal of Biological Chemistry 275, 930–936. Tsang ML-S, Goldschmidt EE, Schiff JA. 1971. Adenosine-59phosphosulfate (APS35) as an intermediate in the conversion of adenosine-39-phosphate-59-phosphosulfate (PAPS35) to acid volatile radioactivity. Plant Physiology 47, 20. Tsang ML-S, Schiff JA. 1978. Assimilatory sulfate reduction in an Escherichia coli mutant lacking thioredoxin activity. Journal of Bacteriology 134, 131–138. Tsakraklides G, Martin M, Chalam R, Tarczynski MC, Schmidt A, Leustek T. 2002. Sulphate reduction is increased Plant APS reductase in transgenic Arabidopsis thaliana expressing 59-adenylylsulphate reductase from Pseudomonas aeruginosa. The Plant Journal 32, 879–889. Vauclare P, Kopriva S, Fell D, Suter M, Sticher L, von Ballmoos P, Krähenbühl U, Op den Camp R, Brunold C. 2002. Flux control of sulphate assimilation in Arabidopsis thaliana: adenosine 59phosphosulphate reductase is more susceptible to negative control by thiols than ATP sulphurylase. The Plant Journal 31, 729–740. Verhagen MF, Kooter IM, Wolbert RB, Hagen WR. 1994. On the iron–sulfur cluster of adenosine phosphosulfate reductase from Desulfovibrio vulgaris (Hildenborough). European Journal of Biochemistry 221, 831–837. Washburn MP, Wells WW. 1999. The catalytic mechanism of the glutathione-dependent dehydroascorbate reductase activity of thioltransferase (glutaredoxin). Biochemistry 38, 268–274. 1783 Weber M, Suter M, Brunold C, Kopriva S. 2000. Sulfate assimilation in higher plants: characterization of a stable intermediate in the adenosine 59-phosphosulfate reductase reaction. European Journal of Biochemistry 267, 3647–3653. Wilson LG. 1962. Metabolism of sulfate: sulfate reduction. Annual Review of Plant Physiology 13, 201–224. Williams SJ, Senaratne RH, Mougous JD, Riley LW, Bertozzi CR. 2002. 59-adenosinephosphosulfate lies at a metabolic branch point in mycobacteria. Journal of Biological Chemistry 277, 32606–32615. Wray JL, Campbell EI, Roberts MA, Gutierrez-Marcos JF. 1998. Redefining reductive sulfate assimilation in higher plants: a role for APS reductase, a new member of the thioredoxin superfamily? Chemico-Biological Interactions 109, 153– 167.