Plant Physiol. (1999) 120: 637-644

UPDATE ON BIOCHEMISTRY

Sulfate Transport and Assimilation in Plants¹

Thomas Leustek^* and Kazuki Saito

Biotechnology Center for Agriculture and the Environment, Rutgers University, New Brunswick, New Jersey 08901-8520 (T.L.); and Faculty of Pharmaceutical Sciences, Chiba University, Yayoi-cho 1-33, Inage-ku, Chiba 263-8522, Japan (K.S.)

	INTRODUCTION

Top Introduction References

Sulfur is one of the six macronutrients required by plants and is found in the amino acids Cys and Met and in a variety of metabolites. When one considers that sulfur in plants is only 3% to 5% as abundant as nitrogen, it is perhaps understandable that sulfur assimilation has been less well studied. As a part of the Cys molecule, the sulfur group, called a thiol, is strongly nucleophilic (electron-donating), making it ideally suited for biological redox processes. When oxidized, two Cys molecules can form a covalent linkage called a disulfide bond, which is readily broken by reduction to form two thiol groups. Disulfide  dithiol interchange is so versatile that nearly all aerobic forms of life, including plants, have evolved to use this reaction as the dominant form of redox control. Redox control regulates enzymes and protects against oxidative damage.

	SULFUR IS THE FUNCTIONAL COMPONENT OF GSH AND OTHER REDOX FACTORS

Free Cys is not used for redox control. It is much too readily oxidized to cystine, the disulfide form, which is visible in the laboratory as a white precipitate that is formed within hours after preparing a solution of Cys. A variety of more stable thiol compounds are involved in redox regulation. The most abundant is glutathione, an enzymatically synthesized tripeptide in which Cys is linked via peptide bonds to the -carboxyl group of Glu and the -amino group of Gly. In plants glutathione is thought to be between 3 and 10 mM, and it is present in the major cellular compartments. The reduced form of glutathione is often referred to as GSH, whereas the disulfide form is GSSG. The balance between forms is overwhelmingly maintained in favor of GSH by the enzyme glutathione reductase, using NADPH as an electron source. The result is that the plant cytoplasm, chloroplast stroma, and mitochondrial matrix are highly buffered in the reducing state. Many intracellular enzymes require reducing conditions for activity, just as they require a specific pH or other properties of their chemical environment. The reason is that Cys residues in proteins can also form disulfide bonds, resulting in a disruption of structure and a loss of activity. There are special cases in which specific disulfide bonds are required for formation of tertiary and quaternary structure in a protein, but this is less common, especially for soluble intracellular proteins.

	REDOX FACTORS REGULATE PLANT METABOLIC PATHWAYS

Other factors that use the chemistry of disulfide  dithiol interchange to mediate redox reactions include the proteins thioredoxin, glutaredoxin, and protein disulfide isomerase. These proteins are nearly ubiquitous and play fundamental roles in many different types of regulation (Fig. 1). One of the first and best examples of the function of thioredoxin comes from Buchanan (1991). The dark reactions of CO₂ fixation must be strictly coordinated with the light reactions of photosynthesis. The coordination mechanism relies on the reductive activation of specific enzymes by thioredoxin, which is reduced by photosynthetically reduced Fd. New disulfide  dithiol redox regulated processes are being discovered each year, which attests to the prevalent roles that sulfur chemistry plays in biology.

View larger version (12K): [in this window] [in a new window]   Figure 1. Regulation of metabolism by disulfide  dithiol interchange. The diagram shows how thioredoxin functions as a regulation factor through reduction of a regulatory disulfide on a target enzyme. In the case of C assimilation, the source of electrons for thioredoxin reduction is Fd, reduced via the light reactions of photosynthesis. Thioredoxins also exist in the cytoplasm of plants where NADPH + H+ serves as an electron source. Recent evidence shows that thioredoxin has the potential to act as an oxidant mediating the formation of a disulfide bond on a target enzyme (Stewart et al., 1998). This activity could be important for activation of antioxidant enzymes during oxidative stress.

	GLUTATHIONE IS IMPORTANT IN STRESS MITIGATION

Because of its nucleophilic properties, glutathione serves as the first line of defense against the products of oxygen metabolism, reactive oxygen species, and other electrophilic compounds such as toxins (herbicides), xenobiotics, and heavy metals (May et al., 1998). When plants encounter reactive oxygen species, glutathione is a direct source of electrons for stress mitigation by the enzyme glutathione peroxidase or an indirect means to maintain a reduced pool of ascorbate, another antioxidant. Glutathione reacts directly with toxins in a reaction mediated by glutathione S-transferase. In this way the toxins are inactivated and tagged for transport into the vacuole and for degradation (Kreuz et al., 1996). In some plants heavy-metal detoxification is mediated by glutathione derivatives called phytochelatins, which have the general structure (-glutamylcysteine)_nGly (n = 2-11), and by Cys-rich proteins called metallothioneins. In both molecules thiol groups serve as the metal ion ligand.

	PLANT SULFUR ASSIMILATION IMPACTS AGRICULTURE AND THE ENVIRONMENT

The preceding discussion illustrates sulfur's essential, general biological role. However, plants also incorporate sulfur into a wide range of secondary compounds that have an impact, in varied and subtle ways, on our use of plants and on the way that plants influence the environment. For example, the pungent odor and taste of onions, garlic, and cabbage are caused by sulfur-containing secondary compounds. The same compounds can impart an objectionable flavor to canola oil, diminishing its commercial value; but they also have been attributed to disease prevention in humans (Fahey et al., 1997). Some sulfur-containing phytoalexins such as camalexin may be important in combating plant pathogens (Zhao et al., 1998). Although sulfur was long thought not to limit plant productivity, the recent restrictions on emissions of sulfurous air pollutants, the ingredients of acid rain, have resulted in sulfur deficiency in some agricultural areas of the world. Another example is that sulfur assimilation by plants has been implicated as a potential factor in moderating climate. Marine algae are prodigious producers of dimethylsulfoniopropionate, a sulfur-containing analog of betaine (Cooper and Hanson, 1998). Dimethylsulfoniopropionate degradation releases dimethylsulfide, which volatilizes from the ocean and seeds the formation of clouds in the atmosphere. The global scale of this process is such that algal growth may actually influence climate.

	OVERVIEW OF THE SO42 ASSIMILATION PROCESS IN PLANTS

Sulfur is available to plants primarily in the form of anionic sulfate (SO₄²) present in soil. It is actively transported into roots and then distributed, mostly unmetabolized, throughout the plant. SO₄² is a major anionic component of vacuolar sap; therefore, it does not necessarily enter the assimilation stream. Gaseous sulfur dioxide (SO₂) is readily absorbed and assimilated by leaves, but it is significant as a nutrient source only in industrial areas with air pollution. Sulfur is assimilated in one of two oxidation states. SO₄² can be added to a hydroxyl group of an organic molecule. The reaction is referred to as sulfation and it is catalyzed by sulfotransferases. By contrast, Cys contains reduced sulfur, which is produced from SO₄² in a multistep pathway in which eight electrons are added to form sulfide (S²; Reaction 1). The reduction of SO₄² to S² consumes 732 kJ mol¹. By comparison, reduction of nitrate to NH₃ requires 347 kJ mol¹. The pathways of SO₄² assimilation in plants are depicted in Figure 2. The figure shows only those enzymes that are known with certainty by characterization of the defined activity of a purified enzyme and through gene cloning. Cys, the end product of the reductive pathway, is the starting material for production of Met, glutathione, and other metabolites containing reduced sulfur.

View larger version (21K): [in this window] [in a new window]   Figure 2. Plant sulfur assimilation pathways showing only those enzymes that have been conclusively demonstrated. The top line shows SO42 activation and reduction. The sulfation pathway is shown in the second line on the left and assimilation of reduced sulfur into Cys on the second line on the right. All enzymes are shown in bold above the reaction arrow, whereas intermediates are shown below the chemical structure or in isolation when the chemical structure is not shown. The R- group in sulfated metabolite refers to the metabolite that is sulfated. Fd indicates the reduced and oxidized forms of Fd.

In higher plants sulfation is a relatively minor fate for sulfur when compared with the reductive pathway. However, in marine algae, which produce large amounts of sulfated extracellular polysaccharides such as agar, sulfation accounts for a much greater proportion of the total assimilated sulfur.

	SO42 UPTAKE IS MEDIATED BY A FAMILY OF TRANSPORTERS WITH SPECIALIZED FUNCTIONS

The transport of SO₄² occurs across several membrane systems as it enters and is distributed throughout the plant and within cells. Transport across the plasma membrane occurs with protons at a ratio of 1 SO₄²:3 H⁺ (symport) and is driven by a proton gradient maintained by a proton ATPase. Transport across the tonoplast membrane is mediated by an unknown mechanism that is driven by the electrical gradient between the vacuole sap and cytoplasm. The phosphate/triose phosphate translocator of the inner chloroplast membrane or a proton/SO₄² symporter may mediate SO₄² transport into chloroplasts.

The plasma membrane transporters of plants have been characterized (Smith et al., 1997; Takahashi et al., 1997). The sequences of cDNAs cloned from Stylosanthes hamata, Arabidopsis, soybean, barley, maize, resurrection grass, and Indian mustard showed that the plasma membrane transporters of plants are most closely related to fungal and animal proton/SO₄² cotransporters. Hydropathy analysis revealed that the plant transporters may span the membrane 12 times, a structural feature that is typical of many types of solute symporters.

In most of the species that have been analyzed, SO₄² transporters are encoded by a gene family. The situation in S. hamata is probably typical for most plants. In this species the individual transporters may have specialized functions, since they differ widely in affinity for SO₄², and they show distinct spatial and regulated patterns of expression (Smith et al., 1997). High-affinity forms with K_m for SO₄² of approximately 9 µM are expressed exclusively in roots, whereas the lower-affinity form with K_m for SO₄² of approximately 100 µM is expressed principally in leaves but also in roots. The steady-state level of mRNA for the high-affinity form increases rapidly after sulfur starvation, whereas the lower-affinity form is unresponsive or responds more slowly to changes in external SO₄² supply. These results imply that the increase in SO₄² transport activity observed in roots of sulfur-starved plants is due to an increase in the expression of specific transporters. One of the earliest observations of SO₄² transport into roots was that the uptake rate varies in relation to the [SO₄²] of the bathing solution. The results with S. hamata suggest that this multiphasic behavior may be due to the activities of separate transporters with different affinities for SO₄².

What is the function of multiple SO₄² transporters? The expression pattern of the high-affinity type suggests that it mediates uptake of SO₄² into the plant and is a way to adjust to variation in the external sulfur supply. By contrast, low-affinity transporters could function in SO₄² uptake, both from soil and from the apoplast solution that bathes internal cells. Evidence for specialization of function has been obtained from analysis of an Arabidopsis SO₄² transporter that is most closely related in sequence to the low-affinity type from S. hamata (Takahashi et al., 1997). It is expressed exclusively in the vascular parenchyma of roots and leaves and not in endodermal, cortical, or epidermal cells. The spatial expression pattern of this low-affinity-type transporter indicates that it must be responsible for uptake from the internal apoplastic pool of SO₄², not from the soil.

	SO42 IS AN INERT COMPOUND THAT MUST BE ACTIVATED BEFORE IT CAN BE METABOLIZED

The low reactivity of SO₄² is a barrier to assimilation that is overcome by formation of a phosphate-SO₄²-anhydride bond in the compound APS. The reaction is catalyzed by ATP sulfurylase (Reaction 2) and is the sole entry point for metabolism of SO₄². The equilibrium of the adenylation reaction favors the production of SO₄² and ATP, the reverse reaction. The K_eq is 10⁷ M in vitro, and the forward reaction can be measured only if enzymes that hydrolyze PPi and modify APS are included. Exactly how ATP sulfurylase operates in the forward direction in vivo has yet to be determined, since the conditions do not appear to be in equilibrium. The PPi concentration is approximately 0.3 mM in plant cells. Despite the theoretical difficulty, Arabidopsis ATP sulfurylase is able to produce APS in vivo under apparently nonequilibrium conditions, since it is fully able to substitute for Escherichia coli ATP sulfurylase even though the PPi concentration in the E. coli cytoplasm is approximately 0.5 mM (Murillo and Leustek, 1995). The Arabidopsis enzyme does not have an intrinsic ability to overcome PPi inhibition, which indicates that extrinsic mechanisms in E. coli facilitate the forward operation of plant ATP sulfurylase, mechanisms that may also function in plant cells.

There are two ATP sulfurylase isoforms in most plants: a major form localized in plastids and a minor form localized in the cytoplasm. Both enzymes have similar kinetic and structural properties. The isoenzymes are encoded by a gene family, and in Arabidopsis there are multiple genes for the plastid enzyme. Arabidopsis contains a cytosolic form of ATP sulfurylase, but the corresponding gene has not yet been identified (Rotte, 1998). The plastid enzyme exists in both leaves and roots and is responsible for initiating the reductive assimilation of SO₄^2-, since isolated chloroplasts can form Cys from SO₄² (Schürmann and Brunold, 1980). The cytoplasmic form probably functions by generating APS for sulfation reactions.

Whether plant ATP sulfurylase plays a role in regulating sulfur assimilation has been studied by a number of investigators (Logan et al., 1996; Lappartient et al., 1999). In general, the activity and steady-state mRNA levels increase when plants are starved for sulfur and decrease when plants are fed reduced forms of sulfur (Cys or glutathione). However, the changes in activity and mRNA, although reproducible, are relatively small; they increase or decrease by approximately 2-fold or less, and the regulation occurs mainly in roots. Two publications report the use of transgenic plants to explore whether ATP sulfurylase regulates SO₄² assimilation. Hatzfeld et al. (1998) concluded that it is not rate limiting, based on an analysis of a transgenic tobacco cell culture that overexpresses an Arabidopsis ATP sulfurylase but that does not show increased sulfur assimilation. By contrast, transgenic Indian mustard lines that overexpress a different ATP sulfurylase isoenzyme from Arabidopsis accumulate glutathione and show increased resistance to SeO₄² (Pilon-Smits et al., 1999). SeO₄² is a toxic analog of SO₄² that Indian mustard can reduce via the sulfur pathway to a nontoxic, volatile form. The opposite results could be due to differences in experimental systems.

	THE SULFATION ROUTE FOR ASSIMILATION: SO42 IS DIRECTLY INCORPORATED INTO PLANT METABOLITES

Plant sulfotransferases have been characterized that catalyze the sulfation of flavonol, desulfoglucosinolate, choline, and gallic acid glucoside (Varin et al., 1997). Sulfotransfer is the terminal step in the biosynthesis of these compounds. The function of sulfated flavonol and choline is unknown. Glucosinolates are the compounds responsible for the distinctive taste of mustards. Gallic acid glucoside, also known as turgorin or periodic leaf movement factor, is responsible for triggering nictinastic leaf movement in Mimosa pudica. Sulfation may regulate the process by activating the movement factor. The number of sulfated compounds in plants is not known. In contrast, sulfation plays a key role in the production of growth-regulating peptides in animals. However, recently, a sulfated regulator of cell proliferation, phytosulfokine-, was identified from plants (Matsubayashi et al., 1997).

Several common features of sulfotransferases have emerged from analysis of the enzymes and the encoding cDNAs. The sulfotransferases are strictly dependent upon the phosphorylated APS derivative, PAPS, as a SO₄² donor, and they all have remarkably high affinity for PAPS. They all contain two highly conserved, sulfotransferase signature sequences that may be involved in PAPS binding. Last, they are all localized in the cytoplasm or, in the case of the gallic acid glucoside sulfotransferase, to the inner surface of the plasma membrane.

Enzymes that synthesize PAPS have been described in plants. But how PAPS is supplied to the sulfotransferases is still to be determined. PAPS is formed through ATP-dependent phosphorylation of the 3-hydroxyl group of APS, catalyzed by APS kinase (Reaction 3). Cytoplasmic ATP sulfurylase exists in most plants, but a cytoplasmic form of APS kinase has not yet been specifically identified. However, such an enzyme could exist, or at least cannot be ruled out, because in Arabidopsis there are three different genes that encode APS kinase, and the localization of only one of them has been studied (Lee and Leustek, 1998; Schiffman and Schwenn, 1998). An alternative to in situ, cytoplasmic synthesis of APS and/or PAPS is a system in which the sulfonucleotides are exported from chloroplasts.

	SO42 IS REDUCED BEFORE INCORPORATION INTO Cys AND THE REDUCTION PATHWAY BEGINS WITH APS

Eight electrons are required to reduce SO₄² to S². The process occurs through the sequential action of two different enzymes. The exact mechanism in plants has been vigorously debated, resulting in a confusing proliferation of hypotheses (Hell, 1997). Rather than reiterating the debate, we discuss here only those enzymes that are known with certainty. Our intention is to simplify a confusing topic.

Several lines of evidence conclusively demonstrate that SO₄² reduction begins with APS in plants and eukaryotic algae. Schmidt (1972) identified an enzyme he called APS sulfotransferase, which has now been completely purified from a marine red macroalga (Kanno et al., 1996). cDNAs that encode APS sulfotransferase have been cloned from a marine green alga and from several higher plants, most notably Arabidopsis (Bick and Leustek, 1998). That the enzyme encoded by the cloned cDNAs was named APS reductase rather than APS sulfotransferase has inadvertently confounded the subject (Gutierrez-Marcos et al., 1996; Setya et al., 1996). Although there were reasonable arguments for the new name, none of these were conclusive enough to warrant abridgment of the original name. Here we submit to historical precedent and refer to the enzyme as APS sulfotransferase.

The two names derive from two possible catalytic mechanisms, neither of which has yet been confirmed. As proposed by Schmidt (1972), the sulfotransferase transfers SO₄² from APS to a thiol compound, generating a thiosulfonate. If GSH were used, the product would be S-sulfoglutathione (Reaction 4). By contrast, a reductase would be expected to transfer electrons from two GSH to APS, generating free sulfite (SO₃²) and GSSG (Reaction 5). It is clear from the amino acid sequence of APS sulfotransferase that, if it functions as a sulfotransferase, it belongs to a different class than the enzymes described earlier in relation to sulfation. They do not share any common sequence motifs. Rather, based on its amino acid sequence and its function, APS sulfotransferase appears to belong to a family of thiol-dependent reductases and thioltransferases.

	GSH PROVIDES ELECTRONS FOR THE FIRST REDUCTION STEP

Mounting evidence indicates that GSH is the most likely in vivo reactant. APS sulfotransferase shows a K_m [GSH] that is consistent with the in vivo concentration of GSH; it appears to be unable to efficiently use other common biological thiol compounds such as thioredoxin (Prior et al., 1999), and a domain of the enzyme is a GSH-dependent reductase that functions in a manner similar to that of glutaredoxin (Bick et al., 1998). Although it is functionally related to glutaredoxin, the amino acid sequence of the domain shows greater homology to thioredoxin. The distinction between thioredoxin and glutaredoxin is more a matter of the preferred electron source than of the sequence. For example, only glutaredoxin is able to use GSH. In this respect, it may be of some significance that glutaredoxin catalyzes reduction of disulfide substrates through a thioltransferase mechanism, i.e. the thiol is transferred with the formation of a glutathione-mixed disulfide intermediate. The analogous reaction for APS sulfotransferase would be the one depicted in Reaction 4. SO₃² can be produced under the reducing conditions in the chloroplast because S-sulfoglutathione is readily reduced nonenzymatically in the presence of excess GSH (Reaction 6) (Schürmann and Brunold, 1980). Based on these considerations we think it is very likely that APS sulfotransferase functions as a GSH:APS sulfotransferase. This hypothesis is depicted in Figure 2.

	SO32+ REDUCTASE COMPLETES THE REDUCTION OF SULFUR WITH ELECTRONS FROM REDUCED Fd

In considering the next reduction step it is significant that S-sulfoglutathione and SO₃² could be available in plastids. Plant SO₃² reductase catalyzes the reduction of SO₃² using electrons donated from reduced Fd (Reaction 7). The enzyme has been convincingly demonstrated by purification and cloning of the corresponding gene and cDNA (Bork et al., 1998). SO₃² reductase shows a high affinity for SO₃² (K_m = approximately 10 µM), which would serve well for efficient metabolism of SO₃². An Fd-dependent enzyme that reduces thiosulonate to thiosulfide has been measured in cell lysates, but it has not been purified or unambiguously demonstrated. The hypothesis that sulfur is reduced as a thiol-bound form, called the "carrier-bound pathway," has been a tenet of the plant SO₄²-assimilation field. As indicated by the preceding discussion, the carrier-bound pathway need not be invoked because there is convincing evidence for an efficient SO₃² reductase. However, the existence of thiosulfonate reductase should not be dismissed as improbable, especially since there has not been a concerted effort to demonstrate it conclusively. Cloning of a thiosulfonate reductase cDNA would be definitive. It is our opinion that an excellent opportunity exists for devising a cloning strategy based on functional complementation of an E. coli SO₃² reductase mutant if the strain is engineered to express plant Fd and NADPH:Fd oxidoreductase.

	APS SULFOTRANSFERASE MAY BE A REGULATION POINT IN THE SULFUR-REDUCTION PATHWAY

APS sulfotransferase is encoded by a gene family in Arabidopsis (Bick and Leustek, 1998), all of whose members appear to encode plastid-localized enzymes. APS sulfotransferase is localized only in plastids and not in other cellular compartments (Rotte, 1998). No specialization of function has yet been ascribed to the APS sulfotransferase isoenzymes. SO₃² reductase is also plastid localized. In Arabidopsis there may be a single gene encoding this enzyme (Bork et al., 1998).

There is a great deal of evidence indicating that APS sulfotransferase is a prime regulation point in SO₄² assimilation (Brunold and Rennenberg, 1997). The activity of this enzyme changes rapidly in a variety of plant species after sulfur starvation, exposure to reduced sulfur compounds, heavy-metal stress, or other stresses. Heavy metals induce the synthesis of phytochelatins, and high concentrations of metal ions significantly increase the demand for Cys. Recent studies indicate that one potential mechanism for regulating APS sulfotransferase activity may involve changes in the steady-state mRNA level. Sulfur starvation (Gutierrez-Marcos et al., 1996; Takahashi et al., 1997) and heavy-metal treatment (Heiss et al., 1999; Lee and Leustek, 1999) induce the accumulation of APS sulfotransferase mRNA, but the response is limited to roots. By contrast, SO₃² reductase does not appear to be appreciably regulated at the mRNA level (Bork et al., 1998). The extent to which the regulation of mRNA abundance is responsible for changes in APS sulfotransferase activity has not yet been adequately explored.

	INCORPORATION OF REDUCED SULFUR INTO Cys: PROTEIN-PROTEIN INTERACTIONS MAY REGULATE THE PROCESS

The incorporation of S² into Cys is the last step in reductive SO₄² assimilation. The reaction is catalyzed by O-acetylserine(thiol)lyase from S² and OAS (Reaction 8). The synthesis of OAS is catalyzed by Ser acetyltransferase (Reaction 9). Ser acetyltransferase and OAS(thiol)lyase exist in an enzyme complex known as Cys synthase. The stability of the complex is affected by substrates (OAS disrupts it and S² stabilizes it), and it appears to form through specific protein-protein interactions (Bogdanova and Hell, 1997). Yet, the free form of each enzyme has catalytic activity, and the complex is not required for channeling of OAS (Droux et al., 1998). Moreover, in chloroplasts the ratio of OAS(thiol)lyase to Ser acetyltransferase is 300:1 (Droux et al., 1998, and refs. therein); therefore, only a fraction of the total OAS(thiol)lyase can be associated in the complex.

One clue to the function of the complex is that association with OAS(thiol)lyase changes the kinetic behavior of Ser acetyltransferase from the Michaelis-Menten type to positive cooperativity with respect to its substrates, Ser and acetyl-CoA (Droux et al., 1998). This suggests that OAS (thiol)lyase functions as a regulatory subunit that regulates Ser acetyltransferase in response to OAS and S². Positive cooperativity is a form of allosteric regulation in which the velocity of a bisubstrate enzyme is highly sensitive to small changes in substrate concentration. One can think of the enzyme as having a hair-trigger control mechanism. The idea is appealing because Cys synthesis requires coordination of two converging pathways. If there is insufficient S² resulting from low activity of SO₄² reduction, the concentration of OAS will increase, causing dissolution of the Cys synthase complex. By contrast, overactivity of SO₄² reduction results in overabundance of S² and a shortage of OAS, a condition that would stabilize the complex. Ser acetyltransferase activity would be regulated, its velocity becoming less or more sensitive to its own substrates. Another possible form of regulation is an increase in the steady-state mRNA level for the plastid form of Ser acetyltransferase after sulfur starvation (Takahashi et al., 1997; Noji et al., 1998). Unlike the earlier steps in the pathway in which mRNA regulation occurs primarily in roots, plastid Ser acetyltransferase mRNA increases primarily in leaves. A third possible regulation mechanism is the feedback inhibition of Ser acetyltransferase by Cys. However, only the cytosolic isoform appears to be regulated in this way (Noji et al., 1998).

Ser acetyltransferase and OAS(thiol)lyase are the only sulfur assimilation enzymes localized in three compartments: the plastids, cytosol, and mitochondria. cDNAs have been cloned from a range of plant species encoding all of the different isoenzymes (Saito et al., 1994). The multifarious localization of Cys synthase enzymes presents a problem in that only plastids contain the full complement of enzymes needed for SO₄² reduction. It may be that S² is exported from plastids to supply the substrate required by cytosolic and mitochondrial Cys synthases. Evidence for a specialization of function was obtained by analysis of the cytosolic form of OAS(thiol)lyase from Arabidopsis. It is expressed predominantly in roots, in the vascular parenchyma, and in cortical cells (Gotor et al., 1997). Expression in leaves is concentrated in trichomes. Gotor et al. (1997) noted that the toxic heavy-metal Cd is known to accumulate in trichomes of exposed plants, so it could be that the cytosolic OAS(thiol)lyase has a specialized role, supplying Cys for a detoxification mechanism.

	NEGATIVE AND POSITIVE SIGNALS REGULATE SO42 ASSIMILATION

From the preceding discussion it is evident that SO₄² reduction and assimilation into Cys is regulated in plants by a range of mechanisms that include substrate availability, modulation of enzyme activity, and gene expression. Since the exclusive function of SO₄² reduction is to produce Cys, future studies must concentrate on the question of how the range of regulation mechanisms for individual enzymes is coordinated. Current thinking focuses on both negative and positive signals. It has been known for some time that reduced sulfur compounds such as Cys and glutathione, when applied to plants, repress the activity of the sulfur-assimilation enzymes. Conversely, SO₄² starvation induces the activity of certain enzymes. With the availability of DNA probes it was discovered that the steady-state levels of mRNA for SO₄² transporter, ATP sulfurylase, and APS sulfotransferase decline after the application of glutathione or Cys to plant roots (Smith et al., 1997; Lappartient et al., 1999; Lee and Leustek, 1999). In contrast, SO₄² starvation increases the steady-state level of mRNA for these proteins. Whether these responses are due to transcriptional or posttranscriptional mechanisms is not yet known.

One form of reduced sulfur, glutathione, could act as an endogenous signal because it is known to be transported through the phloem of plants and its level in phloem sap is markedly reduced after short-term sulfur starvation. Further support for this hypothesis comes from the "split-root" experiments of Lappartient et al. (1999) in which a portion of the root system was sulfur starved. In another portion of the root system fed normal levels of sulfate, the steady-state mRNA level and activity for SO₄² transporter and ATP sulfurylase increased at precisely the time that the level of transported glutathione declined. The implication is that glutathione acts as a negative signal or repressor. Positive regulation could be a derepression phenomenon caused by a decrease in repressor.

Recent experiments indicate, however, that a positive signal may also exist. When OAS was fed to roots of barley, the steady-state mRNA level for the high-affinity transporter increased coordinately with SO₄²-transport activity (Smith et al., 1997). The response was more rapid than when plants were starved for sulfur, but the magnitude of the increase was smaller. OAS feeding also caused the level of reduced sulfur compounds to increase, possibly explaining why the OAS-induced increase in transporter expression was attenuated compared with SO₄² starvation. Regulation is probably a balance between negative and positive signals. Other publications have noted that OAS stimulates SO₄²-reduction enzymes (Smith et al., 1997, and refs. therein). OAS is an appealing candidate as a positive regulator. It is known to influence the Cys synthase complex. A second effect on the expression of SO₄²-reduction enzymes could provide a means for coordination of the entire pathway.

	FOOTNOTES

   Received March 2, 1999; accepted March 23, 1999.

	ABBREVIATIONS

Abbreviations: APS, 5-adenylylsulfate. OAS, O-acetylserine. PAPS, 3-phosphoadenosine-5-phosphosulfate.

	LITERATURE  CITED

Top Introduction References

Bick JA, Åslund F, Chen Y, Leustek T (1998) Glutaredoxin function for the carboxyl terminal domain of the plant-type 5-adenylylsulfate (APS) reductase. Proc Natl Acad Sci USA 95: 8404-8409 [Abstract/Free Full Text]

Bick JA, Leustek T (1998) Plant sulfur metabolism: the reduction of sulfate to sulfite. Curr Opin Plant Biol 1: 240-244 [CrossRef][ISI][Medline]

Bogdanova N, Hell R (1997) Cysteine synthesis in plants: protein-protein interactions of serine acetyltransferase from Arabidopsis thaliana. Plant J 11: 251-262 [CrossRef][ISI][Medline]

Bork C, Schwenn JD, Hell R (1998) Isolation and characterization of a gene for assimilatory sulfite reductase from Arabidopsis thaliana. Gene 212: 147-153 [CrossRef][ISI][Medline]

Brunold C, Rennenberg H (1997) Regulation of sulfur metabolism in plants: first molecular approaches. Prog Bot 58: 164-186

Buchanan BB (1991) Regulation of CO₂ assimilation in oxygenic photosynthesis: the ferredoxin/thioredoxin system. Perspectives on its discovery, present status, and future development. Arch Biochem Biophys 288: 1-9 [CrossRef][ISI][Medline]

Cooper AJL, Hanson AD (1998) Advances in enzymology of the biogeochemical sulfur cycle. Chemtracts-Biochem Mol Biol 11: 729-747

Droux M, Ruffet ML, Douce R, Job D (1998) Interactions between serine acetyltransferase and O-acetylserine(thiol)lyase in higher plants: structural and kinetic properties of the free and bound enzymes. Eur J Biochem 255: 235-245 [ISI][Medline]

Fahey JW, Zhang YS, Talalay P (1997) Broccoli sprouts: an exceptionally rich source of inducers of enzymes that protect against chemical carcinogens. Proc Natl Acad Sci USA 94: 10367-10372 [Abstract/Free Full Text]

Gotor C, Cejudo FJ, Barroso C, Vega JM (1997) Tissue-specific expression of ATCYS-3A, a gene encoding the cytosolic isoform of O-acetylserine(thiol)lyase in Arabidopsis. Plant J 11: 347-352 [CrossRef][ISI][Medline]

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. Proc Natl Acad Sci USA 93: 13377-13382 [Abstract/Free Full Text]

Hatzfeld Y, Cathala N, Grignon C, Davidian JC (1998) Effect of ATP sulfurylase overexpression in bright yellow 2 tobacco cells. Regulation of ATP sulfurylase and SO₄² transport activities. Plant Physiol 116: 1307-1313 [Abstract/Free Full Text]

Heiss S, Schäfer H, Haag-Kerwer A, Rausch T (1999) Cloning sulfur assimilation genes of Brassica juncea L.: cadmium differentially affects the expression of a putative low affinity sulfate transporter and isoforms of ATP sulfurylase and APS reductase. Plant Mol Biol 39: 847-857 [CrossRef][ISI][Medline]

Hell R (1997) Molecular physiology of plant sulfur metabolism. Planta 202: 138-148 [CrossRef][ISI][Medline]

Kanno N, Nagahisa E, Sato M, Sato Y (1996) Adenosine 5-phosphosulfate sulfotransferase from the marine macroalga Porphyra yezoensis Ueda (Rhodophyta): stabilization, purification, and properties. Planta 198: 440-446 [CrossRef]

Kreuz K, Tommasini R, Martinoia E (1996) Old enzymes for a new job. Herbicide detoxification in plants. Plant Physiol 111: 349-353 [ISI][Medline]

Lappartient AG, Vidmar JJ, Leustek T, Glass ADM, Touraine B (1999) Inter-organ signaling in plants: regulation of ATP sulfurylase and sulfate transporter genes expression in roots mediated by phloem-translocated compounds. Plant J 18: 89-95 [CrossRef][ISI][Medline]

Lee S, Leustek T (1998) APS kinase from Arabidopsis thaliana: genomic organization, expression, and kinetic analysis of the recombinant enzyme. Biochem Biophys Res Commun 247: 171-175 [Medline]

Lee S, Leustek T (1999) The effect of cadmium on sulfate assimilation enzymes in Brassica juncea. Plant Sci 141: 201-207 [CrossRef]

Logan HM, Cathala N, Grignon C, Davidian J-C (1996) Cloning of a cDNA encoded by a member of the Arabidopsis thaliana ATP sulfurylase multigene family. J Biol Chem 271: 12227-12233 [Abstract/Free Full Text]

Matsubayashi Y, Takagi L, Sakagami (1997) Phytosulfokine- a sulfated pentapeptide, stimulates the proliferation of rice cells by means of specific high- and low-affinity binding sites. Proc Natl Acad Sci USA 94: 13357-13362 [Abstract/Free Full Text]

May MJ, Vernoux T, Leaver C, Van Montagu M, Inzé D (1998) Glutathione homeostasis in plants: implications for environmental sensing and plant development. J Exp Bot 49: 649-667 [Abstract/Free Full Text]

Murillo M, Leustek T (1995) ATP sulfurylase from Arabidopsis thaliana and Escherichia coli are functionally equivalent but structurally and kinetically divergent: nucleotide sequence of two ATP sulfurylase cDNAs from Arabidopsis thaliana and analysis of a recombinant enzyme. Arch Biochem Biophys 323: 195-204 [CrossRef][Medline]

Noji M, Inoue K, Kimura N, Gouda A, Saito K (1998) Isoform-dependent differences in feedback regulation and subcellular localization of serine acetyltransferase involved in cysteine biosynthesis from Arabidopsis thaliana. J Biol Chem 273: 32739-32745 [Abstract/Free Full Text]

Pilon-Smits EAH, Hwang SB, Lytle CM, Zhu YL, Tai JC, Bravo RC, Chen YC, Leustek T, Terry N (1999) Overexpression of ATP sulfurylase in Indian mustard leads to increased selenate uptake, reduction, and tolerance. Plant Physiol 119: 123-132 [Abstract/Free Full Text]

Prior A, Uhrig JF, Heins L, Wiesmann A, Lilliq CH, Stolz C, Soll J, Schwenn JD (1999) Structural and kinetic properties of adenylylsulfate reductase from Catheranthus roseus cell cultures. Biochim Biophys Acta 1430: 25-38 [CrossRef][Medline]

Rotte C (1998) Subcellular localization of sulfur assimilation enzymes in Arabidopsis thaliana (L.) HEYNH. Diplomarbeit thesis, Carl von Ossietzky Universität Oldenburg, Oldenburg, Germany

Saito K, Kurosawa M, Tatsuguchi K, Takagi Y, Murakoshi I (1994) Modulation of cysteine biosynthesis in chloroplasts of transgenic tobacco overexpressing cysteine synthase [O-acetylserine(thiol)lyase]. Plant Physiol 106: 887-895 [Abstract]

Schiffmann S, Schwenn JD (1998) Isolation of cDNA clones encoding adenosine-5-phosphosulfate kinase (EC 2.7.1.25) from Catharanthus roseus (accession no. AF044285) and an isoform (akn2) from Arabidopsis (accession no. AF043351) (PGR 98-116). Plant Physiol 117: 1125 [Free Full Text]

Schmidt A (1972) An APS sulfotransferase from Chlorella. Arch Mikrobiol 84: 77-86 [CrossRef][Medline]

Schürmann P, Brunold C (1980) Formation of cysteine from adenosine 5-phosphosulfate (APS) in extracts from spinach chloroplasts. Z Pflanzenphysiol 100: 257-268

Setya A, Murillo M, Leustek T (1996) Sulfate reduction in higher plants: molecular evidence for a novel 5-adenylylphosphosulfate (APS) reductase. Proc Natl Acad Sci USA 93: 13383-13388 [Abstract/Free Full Text]

Smith FW, Hawkesford MJ, Ealing PM, Clarkson DT, Vanden Berg PJ, Belcher AR, Warrilow AG (1997) Regulation of expression of a cDNA from barley roots encoding a high affinity sulphate transporter. Plant J 12: 875-884 [CrossRef][ISI][Medline]

Stewart EJ, Åslund F, Beckwith J (1998) Disulfide bond formation in the Escherichia coli cytoplasm: an in vivo role reversal for the thioredoxins. EMBO J 17: 5543-5550 [CrossRef][ISI][Medline]

Takahashi H, Yamazaki M, Sasakura N, Watanabe A, Leustek T, de Almeida-Engler J, Engler G, Van Montagu M, Saito K (1997) Regulation of cysteine biosynthesis in higher plants: a sulfate transporter induced in sulfate-starved roots plays a central role in Arabidopsis thaliana. Proc Natl Acad Sci USA 94: 11102-11107 [Abstract/Free Full Text]

Varin L, Marsolais F, Richard M, Rouleau M (1997) Biochemistry and molecular biology of plant sulfotransferases. FASEB J 11: 517-525 [Abstract]

Zhao JM, Williams CC, Last RL (1998) Induction of Arabidopsis tryptophan pathway enzymes and camalexin by amino acid starvation, oxidative stress, and an abiotic elicitor. Plant Cell 10: 359-370 [Abstract/Free Full Text]

Copyright Clearance Center:   0032-0889/99/120//08
 
© 1999 American Society of Plant Physiologists