Activation energies of selenoxide elimination from Se-substituted selenocysteine

Transition states for selenoxide elimination have been determined for a series of Se-substituted selenocysteine (RSeCys) derivatives that have potential use in the prevention and treatment of cancer, either directly or in conjunction with cisplatin (to reduce its nephrotoxic effects). Reduced activation barriers vs R=Me and R=Ph are found when the alkyl chain length is increased or when activating groups are para to the selenide. Ortho substitution of Lewis bases stabilizes the transition state by directly donating electron density to the selenoxide. The results suggest that RSeCys derivatives incorporating the properties of glutathione peroxidase mimics will, upon oxidation, rapidly eliminate selenenic acid, a precursor to chemopreventative selenols. Scheme Mechanism of selenoxide elimination from Se-substituted SeCys.     

Biosynthesis of dimethyl selenide from sodium selenite in rat liver and kidney cell-free systems.

A pathway for the synthesis of dimethyl selenide from sodium selenite was studied in rat liver and kidney fractions under anaerobic conditions in the presence of GSH, a NADPH-generating system, and S-adenosylmethionine. Chromatography of liver or kidney soluble fraction on Sephadex G-75 yielded a Fraction C (30,000 molecular weight) which synthesized dimethyl selenide, but at a low rate. Addition of proteins eluting at the void volume (Fraction A) to Fraction C restored full activity. Fractionation of Fraction A on DEAE-cellulose revealed that its ability to stimulate Fraction C was associated with two fractions, one containing glutathione reductase and the other a NADPH-dependent disulfide reductase. It was concluded that Fraction C contains a methyltransferase acting on small amounts of hydrogen selenide produced non-enzymically by the reaction of selenite with GSH, and that stimulation by Fraction A results partly from the NADPH-linked formation of hydrogen selenide catalyzed by glutathione reductase present in Fraction A. Washed liver microsomal fraction incubated with selenite plus 20 mM GSH also synthesized dimethyl selenide, but addition of soluble fraction stimulated activity. A synergistic effect was obtained when liver soluble fraction was added to microsomal fraction in the presence of a physiological level of GSH (2 mM), whereas at 20 mM GSH the effect was merely additive. The microsomal component of the liver system was labile, had maximal activity around pH 7.5, and was exceedingly sensitive to NaAsO2 (93% inhibition by 10(-6) M arsenite in the presence of a 20,000-fold excess of GSH). The microsomal activity apparently results from a Se-methyltransferase, possibly a dithiol protein, that methylates hydrogen selenide produced enzymically by the soluble fraction or non-enzymically when a sufficiently high concentration of GSH is used.     

Biosynthesis of dimethyl selenide from sodium selenite in rat liver and kidney cell-free systems

A pathway for the synthesis of dimethyl seledine from sodium selenite was studied in rat liver and kidney fractions under anaerobic conditions in the presence of GSH, a NADPH-generating system, and S-adenosylmethionine. Chromatography of liver or kidney soluble fraction on Sephadex G-75 yielded a Fraction C (30 000 molecular weight) which synthesized dimethyl selenide, but at a low rate. Addition of proteins eluting at the void volume (Fraction A) to Fraction C restored full activity. Fractionation of Fraction A on DEAE-cellulose revealed that its ability to stimulate Fraction C was associated with two fractions, one containing glutathione reductase and the other a NADPH-dependent disulfide reductase. It was concluded that Fraction C contains a methyltransferase acting on small amounts of hydrogen selenide produced non-enzymically by the reaction of selenite with GSH, and that stimulation by Fraction A results partly from the NADPH-linked formation of hydrogen selenide catalyzed by glutathione reductase present in Fraction A. Washed liver microsomal fraction incubated with selenite plus 20 mM GSH also synthesized dimethyl selenide, but addition of soluble fraction stimulated activity. A synergistic effect was obtained when liver soluble fraction was added to microsomal fraction in the presence of a physiological level of GSH (2 mM), whereas at 20 mM GSH the effect was merely additive. The microsomal component of the liver system was labile, had maximal activity around pH 7.5, and was exceedingly sensitive to NaAsO₂ (93% inhibition by 10^?6 M arsenite in the presence of a 20 000-fold excess of GSH). The microsomal activity apparently results from a Se-methyltransferase, possibly a dithiol protein, that methylates hydrogen selenide produced enzymically by the soluble fraction or non-enzymically when a sufficiently high concentration of GSH is used.     

Acid-volatile selenium formation catalyzed by glutathione reductase.

The production of acid-volatile selenide (apparently H2Se) was catalyzed by glutathione reductase in an anaerobic system containing 20 mM glutathione, 0.05 mM sodium selenite, a TPNH-generating system, and microgram quantities of highly purified yeast glutathione reductase. H2Se production in this system was proportional to glutathione reductase concentration and was maximal at pH 7. Significant nonenzymic H2Se production occurred in the system lacking glutathione reductase and TNPH. A concentration of arsenite (0.1 mM) which does not inhibit glutathione reductase inhibited selenide volatilization, as did bovine serum albumin (1.67 mg/ml). Both appear to inhibit Se volatilization by reacting with the selenide product(s). The selenotrisulfide derivative of glutathione (GSSeSG) was readily converted to H2Se by glutathione reductase and TPNH without the addition of glutathione. These results suggest that GSSeSG formed nonenzymically from glutathione and selenic undergoes stepwise reduction by glutathione reductase (or excess GSH) to GSSeH and finally to H2Se. The same pathway operates when glutathione is used as the reducing agent but to a lesser extent.     

Catalytic action of L-methionine gamma-lyase on selenomethionine and selenols.

We examined the catalytic action of L-methionine gamma-lyase (EC 4.4.1.11) on selenomethionine (2-amino-4-(methylseleno)butyric acid), methaneselenol, l-hexaneselenol, and benzeneselenol. The enzyme catalyzes alpha, gamma-elimination of selenomethionine to yield alpha-letobutyrate, ammonia, and methaneselenol, and also its gamma-replacement reaction with various thiols to produce S-substituted homocysteines. Selenomethionine is an even better substrate than methionine in alpha, gamma-elimination but is less effective in gamma-replacement. In addition, L-methionine gamma-lyase catalyzes gamma-replacement reaction of methionine and its derivatives with selenols to form the corresponding Se-substituted selenohomocysteines, although selenols are less efficient substituent donors than thiols. This is the first proven mechanism for the incorporation of selenium atom into amino acids.     

77Se NMR characterization of 77Se-labeled ovine erythrocyte glutathione peroxidase

Lambs, maintained on a selenium-deficient diet supplemented with 94 atom % Na2 27SeO3, have been used as a source of 77Se-enriched erythrocyte glutathione peroxidase. After 5 months on this diet, the percentage of selenium in the enzyme derived from the supplement had reached 88%. From each monthly bleeding of two sheep, approximately 20 mg of 77Se-enriched glutathione peroxidase could be isolated in pure form. Although attempts to observe 77Se NMR signals from the native enzyme labeled with 6,6'-[77Se]diselenobis-(3-nitrobenzoic acid) failed, due to the low solubility of the enzyme, two 77Se resonances were observed after unfolding the enzyme with 8 M urea and reaction with iodoacetamide. These resonances, at 195 and 377 ppm, were from the selenoether alkylamide derivative and from protein cross-linked selenide sulfide species, respectively. Relaxation time measurements on the selenoether at 4.7 and 9.4 teslas enabled an estimate of the chemical shift anisotropy to be made. A value of less than or equal to 262 ppm was determined. Reduction of the denatured selenide sulfide species with dithiothreitol gave an observable 77Se resonance from the Se- moiety at pH 8 and from SeH at pH 4.2. The chemical form of the selenocysteine residue in the resting state enzyme most consistent with formation of the acetamide derivative and the selenide sulfide is Se- or SeH. From the magnitudes of the estimated chemical shift anisotropies, it is predicted that direct observation of selenium in the native enzyme will be feasible if the enzyme concentration can be increased to 0.25 mM tetrameric glutathione peroxidase.     

Reduction of Selenate to Selenide by Sulfate-Respiring Bacteria: Experiments with Cell Suspensions and Estuarine Sediments

Washed cell suspensions of Desulfovibrio desulfuricans subsp. aestuarii were capable of reducing nanomolar levels of selenate to selenide as well as sulfate to sulfide. Reduction of these species was inhibited by 1 mM selenate or tungstate. The addition of 1 mM sulfate decreased the reduction of selenate and enhanced the reduction of sulfate. Increasing concentrations of sulfate inhibited rates of selenate reduction but enhanced sulfate reduction rates. Cell suspensions kept in 1 mM selenate were incapable of reducing either selenate or sulfate when the selenate/sulfate ratio was ≥0.02, indicating that irreversible inhibition occurs at high selenate concentrations. Anoxic estuarine sediments having an active flora of sulfate-respiring bacteria were capable of a small amount of selenate reduction when ambient sulfate concentrations were low (<4 mM). These results indicate that sulfate is an inhibitor of the reduction of trace quantities of selenate. Therefore, direct reduction of traces of selenate to selenide by sulfate-respiring bacteria in natural environments is constrained by the ambient concentration of sulfate ions. The significance of this observation with regard to the role sediments play in sequestering selenium is discussed.     

Selenium in chemistry and biochemistry in comparison to sulfur

What makes selenoenzymes--seen from a chemist's view--so special that they cannot be substituted by just more analogous or adapted sulfur proteins? This review compiles and compares physicochemical properties of selenium and sulfur, synthetic routes to selenocysteine (Sec) and its peptides, and comparative studies of relevant thiols and selenols and their (mixed) dichalcogens, required to understand the special role of selenium in selenoproteins on the atomic molecular level. The biochemically most relevant differences are the higher polarizability of Se- and the lower pKa of SeH. The latter has a strikingly different pH-dependence than thiols, with selenols being active at much lower pH. Finally, selected typical enzymatic mechanisms which involve selenocysteine are critically discussed, also in view of the authors' own results.     

Evidence for substrate channeling in the early steps of cholesterogenesis

We have directly tested the ability of acetoacetate, upon activation to the CoA thioester, to channel into the cholesterogenic pathway prior to scrambling of its carbon skeleton with the acetate pool. The approach relies upon trapping [3-13C]acetoacetate-derived hydroxymethylglutaryl-CoA, hydrolyzing this metabolite, and esterifying the resulting hydroxymethylglutaric acid to allow gas chromatography/mass spectrometry analysis of the dimethyl esters for the 13C enrichment and labeling pattern. 99% enriched [3-13C] and [1,3,5-13C]hydroxymethylglutaric acid samples were synthesized, providing standards against which physiological samples could be compared. Cytosolic extracts from brain and liver of cholestyramine-fed rats were incubated with [3-13C]acetoacetate (2 mM) or with [1-13C]acetate (5 mM). In contrast to [13C]acetate-derived hydroxymethylglutarate, which shows the expected triple labeling pattern, [13C]acetoacetate-derived hydroxymethylglutarate from both liver and brain extracts is predominantly monolabeled. These data suggest that, after acetoacetate is activated to the CoA thioester, cytosolic hydroxymethylglutaryl-CoA synthase effectively commits much of this acetoacetyl-CoA to cholesterogenesis before thiolase can scramble the carbon skeleton of the acetoacetyl moiety into the acetate pool. This chemical approach represents an alternative method for testing the channeling of metabolites through sequential steps in a metabolic pathway. Such a method may be useful when physical or kinetic techniques prove to be unsuitable.     

Novel method for in vivo hydroxyl radical measurement by microdialysis in fetal sheep brain in utero.

Hydroxyl radical (.OH) is a reactive oxygen species produced during severe hypoxia, asphyxia, or ischemia that can cause cell death resulting in brain damage. Generation of .OH may occur in the fetal brain during asphyxia in utero. The very short half-life of .OH requires use of trapping agents such as salicylic acid or phenylalanine for detection, but their hydroxylated derivatives are either unstable, produced endogenously, or difficult to measure in the small volume of microdialysis samples. In the present study, we used terephthalic acid (TA), hydroxylation of which yields a stable and highly fluorometric isomer (excitation, 326 nm; emission, 432 nm). In vitro studies using .OH generated by the Fenton reaction showed that hydroxylated TA formed quickly (<10 s), was resistant to bleaching (<5% change in fluorescence), and permitted detection of <0.5 pmol .OH. In vivo studies were performed in fetal sheep using microdialysis probes implanted into the parasagittal cortex. The probe was perfused at 2 mul/min with artificial cerebrospinal fluid containing 5 mM TA, and samples were collected every 30 min. Fluorescence measured in 10 mul of dialysate was significantly greater than in the efflux from probes perfused without TA. High-performance liquid chromotography analysis showed that the fluorescence in dialysis samples was entirely due to hydroxylation of TA. Thus this study shows that it is possible to use TA as a trapping agent for detecting low concentrations of .OH both in vitro and in vivo and that low concentrations of .OH are present in fetal brain tissue and fluctuate with time.     

Redox reactions of hydrogen selenide ion

Hydrogen selenide ion (HSe^-) is an important product in the metabolism of the essential trace element selenium. Although its role in selenium metabolism is recognized, aspects of the basic chemistry of selenide have been ignored, particularly the tendency of selenide to undergo rapid redox reactions with biological oxidants. Using polarography, we have found that selenide reacts in vitro with a variety of compounds including dehydroascorbic acid, quinones like vitamin K₁ and FAD (flavin adenine dinucleotide), and disulfides such as oxidized glutathione and lipoic acid. The fact selenide reacts readily in vitro suggests similar reactions may also occur in vivo with important biological consequences. Contrary to expectations, selenide was found not to reduce the disulfide bond of oxidized dithiothreitol (trans-4,5-dihydroxyl-1,2-dithiane), indicating the commonly published value for the standard electrode potential of the selenium/hydrogen selenide ion couple is in error. The electrode potential is an important parameter to aid in anticipating possible redox reactions of selenide in vivo.     

Methylcitrate synthase from Aspergillus nidulans: implications for propionate as an antifungal agent

Aspergillus nidulans was used as a model organism to investigate the fungal propionate metabolism and the mechanism of growth inhibition by propionate. The fungus is able to grow slowly on propionate as sole carbon and energy source. Propionate is oxidized to pyruvate via the methylcitrate cycle. The key enzyme methylcitrate synthase was purified and the corresponding gene mcsA, which contains two introns, was cloned, sequenced and overexpressed in A. nidulans. The derived amino acid sequence of the enzyme shows more than 50% identity to those of most eukaryotic citrate synthases, but only 14% identity to the sequence of the recently detected bacterial methylcitrate synthase from Escherichia coli. A mcsA deletion strain was unable to grow on propionate. The inhibitory growth effect of propionate on glucose medium was enhanced in this strain, which led to the assumption that trapping of the available CoA as propionyl-CoA and/or the accumulating propionyl-CoA itself interferes with other biosynthetic pathways such as fatty acid and polyketide syntheses. In the wild-type strain, however, the predominant inhibitor may be methylcitrate. Propionate (100 mM) not only impaired hyphal growth of A. nidulans but also synthesis of the green polyketide-derived pigment of the conidia, whereas in the mutant pigmentation was abolished with 20 mM propionate.     

Specificity of the surface aminopeptidase in Moniliformis moniliformis (Acanthocephala)

The hydrolysis of various oligopeptides in solution by intact Moniliformis moniliformis was examined using paper chromatographic analysis of the incubation medium. In the presence of transport inhibitors, the respective peptide sub-units and/or amino acid residues accumulated in the bathing medium. Only peptides with serine, methionine, leucine or alanine at the NH2-terminal end of the peptide were hydrolysed. There was no hydrolysis when these amino acids were located internally or at the COOH-terminus indicating genuine aminopeptidase activity of the class, alpha-aminoacylpeptide hydrolase. Hydrolysis was negligible when the NH2-terminus was arginine, aspartic acid, glutamic acid, glycine, histidine, lysine, phenylalanine, proline, tryptophan, tyrosine, or valine. In separate experiments, mediated uptake of 0.1 mM 3H-leucine by the worms in 2 min was inhibited 100% by 5 mM unlabelled leucine or tri-serine, but only partially inhibited by 5 mM Ser-Gly (66%), 10 mM Ser-Gly (74%), 5 mM Leu-Leu (69%), 10 mM Leu-Leu (70%), 5 mM Leu-Gly (58%) or 5 mM Met-Met (69%). Because the inhibitions produced by 5 mM Leu-Leu plus 5 mM Met-Met (79%) or 5 mM Leu-Leu plus 5 mM Ser-Gly (76%) were not additive, a single enzyme is indicated. The name serine aminopeptidase is proposed because of its preference for serine.     

Exploring a synthetic organoselenium compound for antioxidant pharmacotherapy—toxicity and effects on ROS-production

The organoselenium antioxidant 1 was previously found to act as a catalytic antioxidant in a two-phase lipid peroxidation system. In aqueous environment, selenide 1 quenched ABTS-radicals more efficiently than Trolox and ascorbic acid. The selenide dose-dependently scavenged reactive oxygen and nitrogen species more efficiently than Trolox for neutrophils and PMA-stimulated macrophages, with 50% inhibitory concentrations in the low micromolar range. In addition no sign of toxicity or effect on cell viability was seen when culturing five human cell lines in concentrations up to 200 μM of selenide 1 for up to seven days. We therefore feel that the compound would be a good candidate for future drug development for prevention or treatment of disorders caused by or involving free radical-mediated or oxidative tissue damage.     

Properties of the selenium-containing moiety of nicotinic acid hydroxylase from Clostridium barkeri

The nicotinic acid hydroxylase from Clostridium barkeri is a selenoenzyme, as evidenced by the copurification of selenium with enzyme activity. This conclusion is supported by data showing a 23-fold increase in nicotinic acid hydroxylase activity when C. barkeri was cultured in media supplemented with selenium. A labile, selenium-containing compound was released from the native protein by treatment with either chaotropic agents and heat or by heating alone. A stable selenium compound was formed when the enzyme was alkylated prior to denaturation. This compound had the same chromatographic properties as dialykyl selenide in a number of systems. The formation of dialkyl selenide upon alkylation is not consistent with the selenium moiety being selenocysteine. Thus, nicotinic acid hydroxylase represents a new type of selenoenzyme.     

Regioselective hydroxylation of quinolinic acid,lutidinic acid and isocinchomeronic acid by resting cells of pyridine dicarboxylic acid-degrading microorganisms

Microorganisms aerobically degrading quinolinic acid, lutidinic acid or isocinchomeronic acid were isolated and the microbial regioselective hydroxylation of these pyridine dicarboxylic acids was studied. Alcaligenes sp. UK21 cells converted quinolinic acid into 6-hydroxypicolinic acid, suggesting the involvement of two enzyme reactions catalyzing hydroxylation at position C6 and decarboxylation at position C3 of quinolinic acid. Resting cells of Alcaligenes sp. UK21 accumulated 94.9 mM 6-hydroxypicolinic acid (13.2 g l(-1)), with a 96% molar conversion yield by 48 h incubation. Rhizobium sp. LA17 and Hydrogenophaga sp. IMA01 catalyzed the regioselective hydroxylation of lutidinic acid and isocinchomeronic acid into 6-hydroxylutidinic acid and 6-hydroxyisocinchomeronic acid, respectively. 6-Hydroxylutidinic acid accumulated up to 95.4 mM (17.5 g l(-1)) by 24 h incubation in the resting cells reaction, using Rhizobium sp. LA17, with a 99% molar conversion yield. Resting cells of Hydrogenophaga sp. IMA01 produced 88.7 mM 6-hydroxyisocinchomeronic acid (16.2 g l(-1)) by 24 h incubation, with a 81% molar conversion yield.     

Growth and shape transformations of giant phospholipid vesicles upon interaction with an aqueous oleic acid suspension

The interaction of two types of vesicle systems was investigated: micrometer-sized, giant unilamellar vesicles (GUVs) formed from 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and submicrometer-sized, large unilamellar vesicles (LUVs) formed from oleic acid and oleate, both in a buffered aqueous solution (pH 8.8). Individual POPC GUVs were transferred with a micropipette into a suspension of oleic acid/oleate LUVs, and the shape changes of the GUVs were monitored using optical microscopy. The behavior of POPC GUVs upon transfer into a 0.8 mM suspension of oleic acid, in which oleic acid/oleate forms vesicular bilayer structures, was qualitatively different from the behavior upon transfer into a 0.3 mM suspension of oleic acid/oleate, in which oleic acid/oleate is predominantly present in the form of monomers and possibly non-vesicular aggregates. In both cases, changes in vesicle morphology were observed within tens of seconds after the transfer. After an initial increase of the vesicle cross-section, the vesicle started to evaginate, spawning dozens of satellite vesicles connected to the mother vesicle with narrow necks or tethers. In 60% of the cases of transfer into a 0.8 mM oleic acid suspension, the evagination process reversed and proceeded to the point where the membrane formed invaginations. In some of these cases, several consecutive transitions between invaginated and evaginated shapes were observed. In the remaining 40% of the cases of transfer into the 0.8 mM oleic acid suspension and in all cases of vesicle transfer into the 0.3 mM oleic acid suspension, no invaginations nor subsequent evaginations were observed. An interpretation of the observed vesicle shape transformation on the basis of the bilayer-couple model is proposed, which takes into account uptake of oleic acid/oleate molecules by the POPC vesicles, oleic acid flip-flop processes and transient pore formation.     

Polarized infrared attenuated total reflectance spectroscopy of the Ca(2+)-ATPase of sarcoplasmic reticulum.

The mean orientations of the transition dipole moments associated with vibrational modes of the proteins and phospholipids of sarcoplasmic reticulum were determined on dry and hydrated membrane multilayers deposited on germanium or zinc selenide crystals, using polarized infrared attenuated total reflectance spectroscopy (P-IR-ATR). For preservation of the enzymatic activity of the Ca(2+)-ATPase the films were prepared from solutions containing 0.05 M KCl, 5 mM imidazole (pH 7.4), 0.5 mM MgCl2, 1-10 mM trehalose and dithiothreitol. The anisotropy was highest in dry films containing congruent to 7.5 micrograms protein/cm2, and decreased with increasing membrane thickness or hydration. The dichroic ratio of the CH2 vibrations (2923 cm-1) of extracted sarcoplasmic reticulum phospholipids on Ge plate was 1.56, compared with a dichroic ratio of 1.68 obtained on dry films of whole sarcoplasmic reticulum. The dichroic ratios of the amide I band (1650 cm-1) of the Ca(2+)-ATPase in the Ca2-E1 state and in the EGTA and vanadate stabilized E2-V state were nearly identical (1.60 vs. 1.62). The dichroism of the amide I, amide II and lipid CH2 vibrations was not affected by changes in the concentration of KCl (25-100 mM) or Ca2+ (approximately equal to 10(-8)-10(-4) M) and by the addition of vanadate (1 mM) or Pi (5 mM) in a calcium-free medium containing 0.5 mM EGTA. The dichroic ratio of the C-C (1033 cm-1) or CO stretching band (1046 cm-1) of trehalose incorporated into SR films was 1.2 on Ge plate; this corresponds to a mean angle of approximately 70 degrees between the plane of the trehalose ring and the normal of the film plane, suggesting that the trehalose molecules are surprisingly well oriented in the polar headgroup region of the phospholipids. The orientation of the trehalose was not affected by the presence of Ca(2+)-ATPase.     

Enzymatic methylation of sulfide, selenide, and organic thiols by Tetrahymena thermophila

Cell extracts from the ciliate Tetrahymena thermophila catalyzed the S-adenosylmethionine-dependent methylation of sulfide. The product of the reaction, methanethiol, was detected by a radiometric assay and by a gas-chromatographic assay coupled to a sulfur-selective chemiluminescence detector. Extracts also catalyzed the methylation of selenide, and the product was shown by gas chromatography-mass spectrometry to be methaneselenol. The sulfide and selenide methyltransferase activities copurified with the aromatic thiol methyltransferase previously characterized from this organism (A.-M. Drotar and R. Fall, Pestic. Biochem. Physiol. 25:396-406, 1986), but heat inactivation experiments suggested the involvement of distinct sulfide and selenide methyltransferases. Short-term toxicity tests were carried out for sulfide, selenide, and their methylated derivatives; the monomethylated forms were somewhat more toxic than the nonmethylated or dimethylated compounds. Cell suspensions of T. thermophila exposed to sulfide, methanethiol, or their selenium analogs emitted methylated derivatives into the headspace. These results suggest that this freshwater protozoan is capable of the stepwise methylation of sulfide and selenide, leading to the release of volatile methylated sulfur or selenium gases.     

Enzymatic methylation of sulfide, selenide, and organic thiols by Tetrahymena thermophila.

Cell extracts from the ciliate Tetrahymena thermophila catalyzed the S-adenosylmethionine-dependent methylation of sulfide. The product of the reaction, methanethiol, was detected by a radiometric assay and by a gas-chromatographic assay coupled to a sulfur-selective chemiluminescence detector. Extracts also catalyzed the methylation of selenide, and the product was shown by gas chromatography-mass spectrometry to be methaneselenol. The sulfide and selenide methyltransferase activities copurified with the aromatic thiol methyltransferase previously characterized from this organism (A.-M. Drotar and R. Fall, Pestic. Biochem. Physiol. 25:396-406, 1986), but heat inactivation experiments suggested the involvement of distinct sulfide and selenide methyltransferases. Short-term toxicity tests were carried out for sulfide, selenide, and their methylated derivatives; the monomethylated forms were somewhat more toxic than the nonmethylated or dimethylated compounds. Cell suspensions of T. thermophila exposed to sulfide, methanethiol, or their selenium analogs emitted methylated derivatives into the headspace. These results suggest that this freshwater protozoan is capable of the stepwise methylation of sulfide and selenide, leading to the release of volatile methylated sulfur or selenium gases.