Sulfate formation via ATP sulfurylase in thiosulfate- and sulfite-disproportionating bacteria

Disproportionation of thiosulfate or sulfite to sulfate plus sulfide was found in several sulfate-reducing bacteria. Out of nineteen strains tested, eight disproportionated thiosulfate, and four sulfite. Growth with thiosulfate or sulfite as the sole energy source was obtained with three strains (Desulfovibrio sulfodismutans and the strains Bra02 and NTA3); additionally, D. desulfuricans strain CSN grew with sulfite but not with thiosulfate, although thiosulfate was disproportionated. Two sulfur-reducing bacteria, four phototrophic sulfur-oxidizing bacteria (incubated in the dark), and Thiobacillus denitrificans did not disproportionate thiosulfate or sulfite. Desulfovibrio sulfodismutans and D. desulfuricans CSN formed sulfate from thiosulfate or sulfite even when simultaneously oxidizing hydrogen or ethanol, or in the presence of 50 mM sulfate. The capacities of sulfate reduction and of thiosulfate and sulfite disproportionation were constitutively present. Enzyme activities required for sulfate reduction (ATP sulfurylase, pyrophosphatase, APS reductase, sulfite reductase, thiosulfate reductase, as well as adenylate kinase and hydrogenase) were detected in sufficient activities to account for the growth rates observed. ADP sulfurylase and sulfite oxidoreductase activities were not detected. Disproportionation was sensitive to the uncoupler carbonylcyanide m-chlorophenylhydrazone (CCCP) but not to the ATPase inhibitor dicyclohexylcarbodiimide (DCCD). It is proposed that during thiosulfate and sulfite disproportionation sulfate is formed via APS reductase and ATP sulfurylase, but not by sulfite oxidoreductase. Reversed electron transport must be assumed to explain the reduction of thiosulfate and sulfite by the electrons derived from APS reductase.Abbreviations CCCP Carbonylcyanide m-chlorophenylhydrazone - DCCD N,N-dicyclohexylcarbodiimide - APS adenosine 5-phosphosulfate (adenylylsulfate)     

The two-domain structure of 5'-adenylylsulfate (APS) reductase from Enteromorpha intestinalis is a requirement for efficient APS reductase activity

5'-Adenylylsulfate (APS) reductase from Enteromorpha intestinalis (EiAPR) is composed of two domains that function together to reduce APS to sulfite. The carboxyl-terminal domain functions as a glutaredoxin that mediates the transfer of electrons from glutathione to the APS reduction site on the amino-terminal domain. To study the basis for the interdomain interaction, a heterologous system was constructed in which the C domain of EiAPR was fused to the carboxyl terminus of the APS reductase from Pseudomonas aeruginosa (PaAPR), an enzyme that normally uses thioredoxin as an electron donor and is incapable of using glutathione for this function. The hybrid enzyme, which retains the [4Fe-4S] cluster from PaAPR, was found to use both thioredoxin and glutathione as an electron donor for APS reduction. The ability to use glutathione was enhanced by the addition of Na2SO4 to the reaction buffer, a property that the hybrid enzyme shares with EiAPR. When the C domain was added as a separate component, it was much less efficient in conferring PaAPR with the ability to use glutathione as an electron donor, despite the fact that the separately expressed C domain functioned in two activities that are typical for glutaredoxins, hydroxyethyl disulfide reduction and electron donation to ribonucleotide reductase. These results suggest that the physical connection of the reductase and C domain on a single polypeptide is critical for the electron-transfer reaction. Moreover, the effect of Na2SO4 suggests that a water-ordering component of the reaction milieu is critical for the catalytic function of plant-type APS reductases by promoting the interdomain interaction.     

Adenosine 5'-phosphosulfate sulfotransferase and adenosine 5'-phosphosulfate reductase are identical enzymes

Adenosine 5'-phosphosulfate (APS) sulfotransferase and APS reductase have been described as key enzymes of assimilatory sulfate reduction of plants catalyzing the reduction of APS to bound and free sulfite, respectively. APS sulfotransferase was purified to homogeneity from Lemna minor and compared with APS reductase previously obtained by functional complementation of a mutant strain of Escherichia coli with an Arabidopsis thaliana cDNA library. APS sulfotransferase was a homodimer with a monomer M(r) of 43,000. Its amino acid sequence was 73% identical with APS reductase. APS sulfotransferase purified from Lemna as well as the recombinant enzyme were yellow proteins, indicating the presence of a cofactor. Like recombinant APS reductase, recombinant APS sulfotransferase used APS (K(m) = 6.5 microM) and not adenosine 3'-phosphate 5'-phosphosulfate as sulfonyl donor. The V(max) of recombinant Lemna APS sulfotransferase (40 micromol min(-1) mg protein(-1)) was about 10 times higher than the previously published V(max) of APS reductase. The product of APS sulfotransferase from APS and GSH was almost exclusively SO(3)(2-). Bound sulfite in the form of S-sulfoglutathione was only appreciably formed when oxidized glutathione was added to the incubation mixture. Because SO(3)(2-) was the first reaction product of APS sulfotransferase, this enzyme should be renamed APS reductase.     

Sulfite-oxido-reductase is involved in the oxidation of sulfite in Desulfocapsa sulfoexigens during disproportionation of thiosulfate and elemental sulfur

The enzymatic pathways of elemental sulfur and thiosulfate disproportionation were investigated using cell-free extract of Desulfocapsa sulfoexigens. Sulfite was observed to be an intermediate in the metabolism of both compounds. Two distinct pathways for the oxidation of sulfite have been identified. One pathway involves APS reductase and ATP sulfurylase and can be described as the reversion of the initial steps of the dissimilatory sulfate reduction pathway. The second pathway is the direct oxidation of sulfite to sulfate by sulfite oxidoreductase. This enzyme has not been reported from sulfate reducers before. Thiosulfate reductase, which cleaves thiosulfate into sulfite and sulfide, was only present in cell-free extract from thiosulfate disproportionating cultures. We propose that this enzyme catalyzes the first step in thiosulfate disproportionation. The initial step in sulfur disproportionation was not identified. Dissimilatory sulfite reductase was present in sulfur and thiosulfate disproportionating cultures. The metabolic function of this enzyme in relation to elemental sulfur or thiosulfate disproportionation was not identified. The presence of the uncouplers HQNO and CCCP in growing cultures had negative effects on both thiosulfate and sulfur disproportionation. CCCP totally inhibited sulfur disproportionation and reduced thiosulfate disproportionation by 80% compared to an unamended control. HQNO reduced thiosulfate disproportionation by 80% and sulfur disproportionation by 90%.     

Sulfur assimilation and glutathione metabolism under cadmium stress in yeast, protists and plants

Glutathione (gamma-glu-cys-gly; GSH) is usually present at high concentrations in most living cells, being the major reservoir of non-protein reduced sulfur. Because of its unique redox and nucleophilic properties, GSH serves in bio-reductive reactions as an important line of defense against reactive oxygen species, xenobiotics and heavy metals. GSH is synthesized from its constituent amino acids by two ATP-dependent reactions catalyzed by gamma-glutamylcysteine synthetase and glutathione synthetase. In yeast, these enzymes are found in the cytosol, whereas in plants they are located in the cytosol and chloroplast. In protists, their location is not well established. In turn, the sulfur assimilation pathway, which leads to cysteine biosynthesis, involves high and low affinity sulfate transporters, and the enzymes ATP sulfurylase, APS kinase, PAPS reductase or APS reductase, sulfite reductase, serine acetyl transferase, O-acetylserine/O-acetylhomoserine sulfhydrylase and, in some organisms, also cystathionine beta-synthase and cystathionine gamma-lyase. The biochemical and genetic regulation of these pathways is affected by oxidative stress, sulfur deficiency and heavy metal exposure. Cells cope with heavy metal stress using different mechanisms, such as complexation and compartmentation. One of these mechanisms in some yeast, plants and protists is the enhanced synthesis of the heavy metal-chelating molecules GSH and phytochelatins, which are formed from GSH by phytochelatin synthase (PCS) in a heavy metal-dependent reaction; Cd(2+) is the most potent activator of PCS. In this work, we review the biochemical and genetic mechanisms involved in the regulation of sulfate assimilation-reduction and GSH metabolism when yeast, plants and protists are challenged by Cd(2+).     

Regulation of sulfate assimilation by nitrogen and sulfur nutrition in poplar trees

Plants cover their need for sulfur by taking up inorganic sulfate, reducing it to sulfide, and incorporating it into the amino acid cysteine. In herbaceous plants the pathway of assimilatory sulfate reduction is highly regulated by the availability of the nutrients sulfate and nitrate. To investigate the regulation of sulfate assimilation in deciduous trees we used the poplar hybrid Populus tremula × P. alba as a model. The enzymes of the pathway are present in several isoforms, except for sulfite reductase and -glutamylcysteine synthetase; the genomic organization of the pathway is thus similar to herbaceous plants. The mRNA level of APS reductase, the key enzyme of the pathway, was induced by 3 days of sulfur deficiency and reduced by nitrogen deficiency in the roots, whereas in the leaves it was affected only by the withdrawal of nitrogen. When both nutrients were absent, the mRNA levels did not differ from those in control plants. Four weeks of sulfur deficiency did not affect growth of the poplar plants, but the content of glutathione, the most abundant low molecular thiol, was reduced compared to control plants. Sulfur limitation resulted in an increase in mRNA levels of ATP sulfurylase, APS reductase, and sulfite reductase, probably as an adaptation mechanism to increase the efficiency of the sulfate assimilation pathway. Altogether, although distinct differences were found, e.g. no effect of sulfate deficiency on APR in poplar leaves, the regulation of sulfate assimilation by nutrient availability observed in poplar was similar to the regulation described for herbaceous plants.     

Biochemical studies on sulfate-reducing bacteria. XIV. Enzyme levels of adenylylsulfate reductase, inorganic pyrophosphatase, sulfite reductase, hydrogenase, and adenosine triphosphatase in cells grown on sulfate, sulfite, and thiosulfate.

Sulfate-reducing bacteria, Desulfovibrio vulgaris, strain Miyazaki, were grown on either sulfate, sulfite, or thiosulfate as the terminal electron acceptor. Better growth was observed on sulfite and less growth on thiosulfate than on sulfate. Enzyme levels of adenylylsulfate (APS) reductase [EC 1.8.99.2], reductant-activated inorganic pyrophosphatase [EC 3.6.1.1], sulfite reductase [EC 1.8.99.1] (desulfoviridin), hydrogenase [EC 1.12.2.1], and Mg2+-activated ATPase [EC 3.6.1.3] were compared in crude extracts of these cells at various stages of growth. 1) The specific activity of APS reductase in sulfite-grown cells was only one-fourth that in sulfate-grown cells throughout growth. Thiosulfate-grown cells had an activity intermediate between those of sulfate- and sulfite-grown cells. 2) Cells grown on sulfite had lower specific activity of reductant-activated inorganic pyrophosphatase than cells grown on sulfate or thiosulfate. 3) The specific activity of sulfite reductase (desulfoviridin) was highest in sulfite-grown cells. The sulfite medium gave the enzyme in high yield as well as with high specific activity. 4) The specific activities of hydrogenase and Mg2+-ATPase were not significantly altered by electron acceptors in the growth medium.     

Properties of the cysteine residues and the iron-sulfur cluster of the assimilatory 5'-adenylyl sulfate reductase from Enteromorpha intestinalis

The 5'-adenylyl sulfate (APS) reductase from the marine macrophytic green alga Enteromorpha intestinalis uses reduced glutathione as the electron donor for the reduction of APS to 5'-AMP and sulfite. The E. intestinalis enzyme (EiAPR) is composed of a reductase domain and a glutaredoxin-like C-terminal domain. The enzyme contains a single [4Fe-4S] cluster as its sole prosthetic group. Three of the enzyme's eight cysteine residues (Cys166, Cys257, and Cys260) serve as ligands to the iron-sulfur cluster. Site-directed mutagenesis experiments and resonance Raman spectroscopy are consistent with the presence of a cluster in which only three of the four ligands to the cluster irons contributed by the protein are cysteine residues. Site-directed mutagenesis experiments suggest that the thiol group of Cys250, a residue found only in algal APS reductases, is not an absolute requirement for activity. The other four cysteines that do not serve as cluster ligands, all of which are required for activity, are involved in the formation of two redox-active disulfide/dithiol couples. The couple involving Cys342 and Cys345 has an E(m) value at pH 7.0 of -140 mV, and the one involving Cys165 and Cys285 has an E(m) value at pH 7.0 of -290 mV. The C-terminal portion of EiAPR, expressed separately, exhibits the cystine reductase activity characteristic of glutaredoxins. It is proposed that the Cys342-Cys345 disulfide provides the site for entry of electrons from reduced glutathione and that the Cys166-Cys285 disulfide may serve as a structural element that is essential for keeping the enzyme in the catalytically active conformation.     

Kinetic Assays for Determining In Vitro APS Reductase Activity in Plants without the Use of Radioactive Substances

Adenosine 5'-phosphosulfate (APS) reductase (APR; EC 1.8.4.9) catalyzes the two-electron reduction of APS to sulfite and AMP, a key step in the sulfate assimilation pathway in higher plants. In spite of the importance of this enzyme, methods currently available for detection of APR activity rely on radioactive labeling and can only be performed in a very few specially equipped laboratories. Here we present two novel kinetic assays for detecting in vitro APR activity that do not require radioactive labeling. In the first assay, APS is used as substrate and reduced glutathione (GSH) as electron donor, while in the second assay APS is replaced by an APS-regenerating system in which ATP sulfurylase catalyzes APS in the reaction medium, which employs sulfate and ATP as substrates. Both kinetic assays rely on fuchsin colorimetric detection of sulfite, the final product of APR activity. Incubation of the desalted protein extract, prior to assay initiation, with tungstate that inhibits the oxidation of sulfite by sulfite oxidase activity, resulted in enhancement of the actual APR activity. The reliability of the two methods was confirmed by assaying leaf extract from Arabidopsis wild-type and APR mutants with impaired or overexpressed APR2 protein, the former lacking APR activity and the latter exhibiting much higher activity than the wild type. The assays were further tested on tomato leaves, which revealed a higher APR activity than Arabidopsis. The proposed APR assays are highly specific, technically simple and readily performed in any laboratory.     

Regulation of sulphate assimilation by glutathione in poplars (Populus tremula x P. alba) of wild type and overexpressing gamma-glutamylcysteine synthetase in the cytosol

Glutathione (GSH) is the major low molecular weight thiol in plants with different functions in stress defence and the transport and storage of sulphur. Its synthesis is dependent on the supply of its constituent amino acids cysteine, glutamate, and glycine. GSH is a feedback inhibitor of the sulphate assimilation pathway, the primary source of cysteine synthesis. Sulphate assimilation has been analysed in transgenic poplars (Populus tremula x P. alba) overexpressing gamma-glutamylcysteine synthetase, the key enzyme of GSH synthesis, and the results compared with the effects of exogenously added GSH. Although foliar GSH levels were 3-4-fold increased in the transgenic plants, the activities of enzymes of sulphate assimilation, namely ATP sulphurylase, adenosine 5'-phosphosulphate reductase (APR), sulphite reductase, serine acetyltransferase, and O-acetylserine (thiol)lyase were not affected in three transgenic lines compared with the wild type. Also the mRNA levels of these enzymes were not altered by the increased GSH levels. By contrast, an increase in GSH content due to exogenously supplied GSH resulted in a strong reduction in APR activity and mRNA accumulation. This feedback regulation was reverted by simultaneous addition of O-acetylserine (OAS). However, OAS measurements revealed that OAS cannot be the only signal responsible for the lack of feedback regulation of APR by GSH in the transgenic poplars.     

Subcellular distribution of glutathione precursors in Arabidopsis thaliana

Glutathione is an important antioxidant and has many important functions in plant development, growth and defense. Glutathione synthesis and degradation is highly compartment-specific and relies on the subcellular availability of its precursors, cysteine, glutamate, glycine and γ-glutamylcysteine especially in plastids and the cytosol which are considered as the main centers for glutathione synthesis. The availability of glutathione precursors within these cell compartments is therefore of great importance for successful plant development and defense. The aim of this study was to investigate the compartment-specific importance of glutathione precursors in Arabidopsis thaliana. The subcellular distribution was compared between wild type plants (Col-0), plants with impaired glutathione synthesis (glutathione deficient pad2-1 mutant, wild type plants treated with buthionine sulfoximine), and one complemented line (OE3) with restored glutathione synthesis. Immunocytohistochemistry revealed that the inhibition of glutathione synthesis induced the accumulation of the glutathione precursors cysteine, glutamate and glycine in most cell compartments including plastids and the cytosol. A strong decrease could be observed in γ-glutamylcysteine (γ-EC) contents in these cell compartments. These experiments demonstrated that the inhibition of γ-glutamylcysteine synthetase (GSH1) - the first enzyme of glutathione synthesis - causes a reduction of γ-EC levels and an accumulation of all other glutathione precursors within the cells.     

Engineering tolerance and hyperaccumulation of arsenic in plants by combining arsenate reductase and gamma-glutamylcysteine synthetase expression

We have developed a genetics-based phytoremediation strategy for arsenic in which the oxyanion arsenate is transported aboveground, reduced to arsenite, and sequestered in thiol-peptide complexes. The Escherichia coli arsC gene encodes arsenate reductase (ArsC), which catalyzes the glutathione (GSH)-coupled electrochemical reduction of arsenate to the more toxic arsenite. Arabidopsis thaliana plants transformed with the arsC gene expressed from a light-induced soybean rubisco promoter (SRS1p) strongly express ArsC protein in leaves, but not roots, and were consequently hypersensitive to arsenate. Arabidopsis plants expressing the E. coli gene encoding gamma-glutamylcysteine synthetase (gamma-ECS) from a strong constitutive actin promoter (ACT2p) were moderately tolerant to arsenic compared with wild type. However, plants expressing SRS1p/ArsC and ACT2p/gamma-ECS together showed substantially greater arsenic tolerance than gamma-ECS or wild-type plants. When grown on arsenic, these plants accumulated 4- to 17-fold greater fresh shoot weight and accumulated 2- to 3-fold more arsenic per gram of tissue than wild type or plants expressing gamma-ECS or ArsC alone. This arsenic remediation strategy should be applicable to a wide variety of plant species.     

The role of the novel adenosine 5′-phosphosulfate reductase in regulation of sulfate assimilation of <Emphasis Type="Italic">Physcomitrella patens</Emphasis>

Sulfate assimilation provides reduced sulfur for the synthesis of the amino acids cysteine and methionine and for a range of other metabolites. The key step in control of plant sulfate assimilation is the reduction of adenosine 5′-phosphosulfate to sulfite. The enzyme catalyzing this reaction, adenosine 5′phosphosulfate reductase (APR), is found as an iron sulfur protein in plants, algae, and many bacteria. In the moss Physcomitrella patens, however, a novel isoform of the enzyme, APR-B, has recently been discovered lacking the co-factor. To assess the function of the novel APR-B we used homologous recombination to disrupt the corresponding gene in P. patens. The knock-out plants were able to grow on sulfate as a sole sulfur source and the content of low molecular weight thiols was not different from wild type plants or plants where APR was disrupted. However, when treated with low concentrations of cadmium the APR-B knockout plants were more sensitive than both wild type and APR knockouts. In wild type P. patens, the two APR isoforms were not affected by treatments that strongly regulate this enzyme in flowering plants. The data thus suggest that in P. patens APS reduction is not the major control step of sulfate assimilation.     

Analysis of cytosolic and plastidic serine acetyltransferase mutants and subcellular metabolite distributions suggests interplay of the cellular compartments for cysteine biosynthesis in Arabidopsis

In plants, the enzymes for cysteine synthesis serine acetyltransferase (SAT) and O-acetylserine-(thiol)-lyase (OASTL) are present in the cytosol, plastids and mitochondria. However, it is still not clearly resolved to what extent the different compartments are involved in cysteine biosynthesis and how compartmentation influences the regulation of this biosynthetic pathway. To address these questions, we analysed Arabidopsis thaliana T-DNA insertion mutants for cytosolic and plastidic SAT isoforms. In addition, the subcellular distribution of enzyme activities and metabolite concentrations implicated in cysteine and glutathione biosynthesis were revealed by non-aqueous fractionation (NAF). We demonstrate that cytosolic SERAT1.1 and plastidic SERAT2.1 do not contribute to cysteine biosynthesis to a major extent, but may function to overcome transport limitations of O-acetylserine (OAS) from mitochondria. Substantiated by predominantly cytosolic cysteine pools, considerable amounts of sulphide and presence of OAS in the cytosol, our results suggest that the cytosol is the principal site for cysteine biosynthesis. Subcellular metabolite analysis further indicated efficient transport of cysteine, γ -glutamylcysteine and glutathione between the compartments. With respect to regulation of cysteine biosynthesis, estimation of subcellular OAS and sulphide concentrations established that OAS is limiting for cysteine biosynthesis and that SAT is mainly present bound in the cysteine–synthase complex.     

Manipulation of thiol contents in plants

As sulfur constitutes one of the macronutrients necessary for the plant life cycle, sulfur uptake and assimilation in higher plants is one of the crucial factors determining plant growth and vigour, crop yield and even resistance to pests and stresses. Inorganic sulfate is mostly taken up as sulfate from the soil through the root system or to a lesser extent as volatile sulfur compounds from the air. In a cascade of enzymatic steps inorganic sulfur is converted to the nutritionally important sulfur-containing amino acids cysteine and methionine (Hell, 1997; Hell and Rennenberg, 1998; Saito, 1999). Sulfate uptake and allocation between plant organs or within the cell is mediated by specific transporters localised in plant membranes. Several functionally different sulfate transporters have to be postulated and have been already cloned from a number of plant species (Clarkson et al., 1993; Hawkesford and Smith, 1997; Takahashi et al., 1997; Yamaguchi, 1997). Following import into the plant and transport to the final site of reduction, the plastid, the chemically relatively inert sulfate molecule is activated through binding to ATP forming adenosine-5'-phosphosulfate (APS). This enzymatic step is controlled through the enzyme ATP-sulfurylase (ATP-S). APS can be further phosphorylated to form 3'-phosphoadenosine-5'-phosphosulfate (PAPS) which serves as sulfate donor for the formation of sulfate esters such as the biosynthesis of sulfolipids (Schmidt and J?ger, 1992). However, most of the APS is reduced to sulfide through the enzymes APS-reductase (APR) and sulfite reductase (SIR). The carbon backbone of cysteine is provided through serine, thus directly coupling photosynthetic processes and nitrogen metabolism to sulfur assimilation. L-serine is activated by serine acetyltransferase (SAT) through the transfer to an acetyl-group from acetyl coenzyme A to form O-acetyl-L-serine (OAS) which is then sulhydrylated using sulfide through the enzyme O-acetyl-L-serine thiol lyase (OAS-TL) forming cysteine. Cysteine is the central precursor of all organic molecules containing reduced sulfur ranging from the amino acid methionine to peptides as glutathione or phytochelatines, proteines, vitamines, cofactors as SAM and hormones. Cysteine and derived metabolites display essential roles within plant metabolism such as protein stabilisation through disulfide bridges, stress tolerance to active oxygen species and metals, cofactors for enzymatic reactions as e.g. SAM as major methylgroup donor and plant development and signalling through the volatile hormone ethylene. Cysteine and other metabolites carrying free sulfhydryl groups are commonly termed thioles (confer Fig. 1). The physiological control of the sulfate reduction pathway in higher plants is still not completely understood in all details. The objective of this paper is to summarise the available data on the molecular analysis and control of cysteine biosynthesis in plants, and to discuss potentials for manipulating the pathway using transgenic approaches.     

Adenosine 5'-phosphosulfate reductase (APR2) mutation in Arabidopsis implicates glutathione deficiency in selenate toxicity

APR2 is the dominant APR (adenosine 5'-phosphosulfate reductase) in the model plant Arabidopsis thaliana, and converts activated sulfate to sulfite, a key reaction in the sulfate reduction pathway. To determine whether APR2 has a role in selenium tolerance and metabolism, a mutant Arabidopsis line (apr2-1) was studied. apr2-1 plants had decreased selenate tolerance and photosynthetic efficiency. Sulfur metabolism was perturbed in apr2-1 plants grown on selenate, as observed by an increase in total sulfur and sulfate, and a 2-fold decrease in glutathione concentration. The altered sulfur metabolism in apr2-1 grown on selenate did not reflect typical sulfate starvation, as cysteine and methionine levels were increased. Knockout of APR2 also increased the accumulation of total selenium and selenate. However, the accumulation of selenite and selenium incorporation in protein was lower in apr2-1 mutants. Decreased incorporation of selenium in protein is typically associated with increased selenium tolerance in plants. However, because the apr2-1 mutant exhibited decreased tolerance to selenate, we propose that selenium toxicity can also be caused by selenate's disruption of glutathione biosynthesis leading to enhanced levels of damaging ROS (reactive oxygen species).