Enzymatic preparation of silybin phase II metabolites: sulfation using aryl sulfotransferase from rat liver.

Aryl sulfotransferase IV (AstIV) from rat liver was overexpressed in Escherichia coli and purified to homogeneity. Using the produced mammalian liver enzyme, sulfation-the Phase II conjugation reaction-of optically pure silybin diastereoisomers (silybin A and B) was tested. As a result, silybin B was sulfated yielding 20-O-silybin B sulfate, whereas silybin A was completely resistant to the sulfation reaction. Milligram-scale sulfation of silybin B was optimized employing resting E. coli cells producing AstIV, thus avoiding the use of expensive 3'-phosphoadenosine-5'-phosphate cofactor and laborious enzyme purification. Using this approach, we were able to reach 48 % conversion of silybin B into its 20-sulfate within 24 h. The sulfated product was isolated by solid phase extraction and its structure was characterized by HRMS and NMR. Sulfation reaction of silybin appeared strictly stereoselective; only silybin B was sulfated by AstIV.

A new arylsulfate sulfotransferase involved in liponucleoside antibiotic biosynthesis in streptomycetes.

Sulfotransferases are involved in a variety of physiological processes and typically use 3'-phosphoadenosine 5'-phosphosulfate (PAPS) as the sulfate donor substrate. In contrast, microbial arylsulfate sulfotransferases (ASSTs) are PAPS-independent and utilize arylsulfates as sulfate donors. Yet, their genuine acceptor substrates are unknown. In this study we demonstrate that Cpz4 from Streptomyces sp. MK730-62F2 is an ASST-type sulfotransferase responsible for the formation of sulfated liponucleoside antibiotics. Gene deletion mutants showed that cpz4 is required for the production of sulfated caprazamycin derivatives. Cloning, overproduction, and purification of Cpz4 resulted in a 58-kDa soluble protein. The enzyme catalyzed the transfer of a sulfate group from p-nitrophenol sulfate (K(m) 48.1 microM, k(cat) 0.14 s(-1)) and methyl umbelliferone sulfate (K(m) 34.5 microM, k(cat) 0.15 s(-1)) onto phenol (K(m) 25.9 and 29.7 mM, respectively). The Cpz4 reaction proceeds by a ping pong bi-bi mechanism. Several structural analogs of intermediates of the caprazamycin biosynthetic pathway were synthesized and tested as substrates of Cpz4. Des-N-methyl-acyl-caprazol was converted with highest efficiency 100 times faster than phenol. The fatty acyl side chain and the uridyl moiety seem to be important for substrate recognition by Cpz4. Liponucleosides, partially purified from various mutant strains, were readily sulfated by Cpz4 using p-nitrophenol sulfate. No product formation could be observed with PAPS as the donor substrate. Sequence homology of Cpz4 to the previously examined ASSTs is low. However, numerous orthologs are encoded in microbial genomes and represent interesting subjects for future investigations.

Identification and characterization of two novel cytosolic sulfotransferases, SULT1 ST7 and SULT1 ST8, from zebrafish.

Cytosolic sulfotransferases (SULTs) constitute a family of Phase II detoxification enzymes that are involved in the protection against potentially harmful xenobiotics as well as the regulation and homeostasis of endogenous compounds. Compared with humans and rodents, the zebrafish serves as an excellent model for studying the role of SULTs in the detoxification of environmental pollutants including environmental estrogens. By searching the expressed sequence tag database, two zebrafish cDNAs encoding putative SULTs were identified. Sequence analysis indicated that these two putative zebrafish SULTs belong to the SULT1 gene family. The recombinant form of these two novel zebrafish SULTs, designated SULT1 ST7 and SULT1 ST8, were expressed using the pGEX-2TK glutathione S-transferase (GST) gene fusion system and purified from transformed BL21 (DE3) cells. Purified GST-fusion protein form of SULT1 ST7 and SULT1 ST8 exhibited strong sulfating activities toward environmental estrogens, particularly hydroxylated polychlorinated biphenyls (PCBs), among various endogenous and xenobiotic compounds tested as substrates. pH-dependence experiments showed that SULT1 ST7 and SULT1 ST8 displayed pH optima at 6.5 and 8.0, respectively. Kinetic parameters of the two enzymes in catalyzing the sulfation of catechin and chlorogenic acid as well as 3-chloro-4-biphenylol were determined. Developmental expression experiments revealed distinct patterns of expression of SULT1 ST7 and SULT1 ST8 during embryonic development and throughout the larval stage onto maturity.

2-Acetylaminofluorene-mediated alteration in the level of liver arylsulfotransferase IV during rat hepatocarcinogenesis.

Rat liver cytosolic sulfotransferase activity forms the highly reactive sulfuric acid ester of N-hydroxy-2-acetylaminofluorene (N-OH-2AAF), an ultimate carcinogen in 2-acetylaminofluorene (2AAF) hepatocarcinogenesis. A previous report demonstrated that 2AAF-induced liver hyperplastic nodules displayed a persistent loss of cytosolic N-OH-2AAF sulfotransferase activity following a hepatocarcinogenesis-producing regimen of 2AAF administration. As an initial step in examining the mechanism responsible for lowering N-OH-2AAF sulfotransferase activity, a monospecific polyclonal antibody to aryl sulfotransferase IV (AST IV) was produced and used in the assessment of AST IV as a candidate enzyme for liver cytosolic N-OH-2AAF sulfotransferase activity. Studies comparing the levels of N-OH-2AAF sulfotransferase activity of highly purified AST IV and rat liver cytosols with corresponding immunochemical analysis of AST IV contents demonstrated that there was sufficient AST IV activity in liver cytosols to indicate that it was the primary enzyme catalyzing cytosolic N-OH-2AAF sulfation. A subsequent immunochemical survey of nine extrahepatic tissues showed no detectable AST IV content and indicated that AST IV expression may be tissue specific. An immunochemical comparison of AST IV levels in control liver cytosols (high in sulfotransferase activity) with cytosols from 2AAF-derived hyperplastic nodules (low in sulfotransferase activity) or liver tumors (no sulfotransferase activity) showed low or no detectable levels, respectively, of AST IV. In addition, an immunochemical analysis of four rat hepatoma cell lines showed they contained no detectable levels of AST IV. These results suggested a strong correlation existed between a decrease in AST IV expression and tumor development. When the liver cytosols of rats taken from early, intermediate, and late stages of 2AAF carcinogenesis were analyzed for the development of a persistent loss of N-OH-2AAF sulfotransferase activity, a parallel loss of cytosolic N-OH-2AAF sulfotransferase activity and AST IV content was observed in rats which had proceeded from a stage of low risk to high risk for liver cancer. These findings indicated that (a) AST IV, a liver-specific enzyme, was the principle enzyme comprising cytosolic N-OH-2AAF sulfotransferase activity and (b) the decrease in sulfotransferase activity in nodules and tumors resulted from a decrease in the level of AST IV expression. Furthermore, it is suggested that a persistent decrease in AST IV expression may reflect a role for AST IV as part of a resistance phenotype in which transforming liver cells are able to escape the cytotoxic effects of highly reactive 2AAF metabolites and progress to cancer.

Aryl sulfotransferase-IV-catalyzed sulfation of aryl oximes: steric and substituent effects.

The aryl sulfotransferases (EC 2.8.2.1) catalyze the sulfation of a wide variety of hydroxyl-containing molecules. The enzyme reaction requires 3'-phosphoadenosine-5'-phosphosulfate as the sulfate donor and several isozymes with broad, overlapping substrate specificities have been identified. One of the isozymes in rat hepatic cytosol, isozyme IV, is a major contributor to enzymatic sulfation. It exhibits the broadest substrate specificity of the three isozymes which have been characterized to date. Its substrates include a wide variety of phenols, certain aromatic hydroxylamines and benzylic alcohols. The latter two substrate types have implicated this isozyme in the bioactivation of several toxic compounds. Relatively little information is available, however, on substrate molecular features which account for the ability of isozyme IV to sulfate compounds not utilized by isozymes I and II. A recent investigation of isozymes I and II with a series of model aryl-oxime substrates suggested that catalysis is influenced primarily by steric factors and in particular substrate planarity and hydroxyl group orientation (Mangold et al. (1989) Biochim. Biophys. Acta 991, 453-458). In the present study, isozyme IV was investigated to characterize its substrate requirements with a more extensive series of aryl oxime substrates. The results indicated that isozyme IV has a much less stringent requirement for planarity and hydroxyl-group orientation than isozymes I or II. Isozyme IV accepted a greater variety of aryl-oxime substrates, including several classes which were not substrates for isozymes I and II. A comparison of kinetic constants and catalytic efficiencies suggested that substituent effects play a role in the sulfation of aryl oximes by isozyme IV.

Identification of a novel thyroid hormone-sulfating cytosolic sulfotransferase, SULT1 ST5, from zebrafish.

By employing RT-PCR in conjunction with 3'-RACE, a full-length cDNA encoding a novel zebrafish cytosolic sulfotransferase (SULT) was cloned and sequenced. Sequence analysis revealed that this zebrafish SULT (designated SULT1 ST5) is, at the amino acid sequence level, close to 50% identical to human and dog SULT1B1 (thyroid hormone SULT). A recombinant form of zebrafish SULT1 ST5 was expressed using the pGEX-2TK bacterial expression system and purified from transformed BL21 (DE3) cells. Purified zebrafish SULT1 ST5 migrated as a 34 kDa protein and displayed substrate specificity for thyroid hormones and their metabolites among various endogenous compounds tested. The enzyme also exhibited sulfating activities toward some xenobiotic phenolic compounds. Its pH optima were 6.0 and 9.0 with 3,3',5-triiodo-l-thyronine (l-T3) as substrate and 6.0 with beta-naphthol as substrate. Kinetic constants of the enzyme with thyroid hormones and their metabolites as substrates were determined. Quantitative evaluation of the regulatory effects of divalent metal cations on the l-T3-sulfating activity of SULT1 ST5 revealed that Fe2+, Hg2+, Co2+, Zn2+, Cu2+, Cd2+ and Pb2+ exhibited dramatic inhibitory effects, whereas Mn2+ showed a significant stimulation. Developmental stage-dependent expression experiments revealed a significant level of expression of this novel zebrafish thyroid hormone-sulfating SULT at the beginning of the hatching period during embryogenesis, which gradually increased to a high level of expression throughout the larval stage into maturity.

Human phenol sulfotransferases SULT1A2 and SULT1A1: genetic polymorphisms, allozyme properties, and human liver genotype-phenotype correlations.

Phenol sulfotransferases (PSTs or phenol SULTs) catalyze the sulfate conjugation of phenolic drugs, xenobiotics, and monoamines. Two human PST isoforms have been defined biochemically, a thermostable (TS), or phenol-preferring, and a thermolabile (TL), or monoamine-preferring form. Pharmacogenetic studies showed that levels of both TS PST activity and TS PST thermal stability (an indirect measure of variation in amino acid sequence) in the platelet were regulated by genetic polymorphisms. Subsequent molecular genetic experiments revealed the existence of three human PST genes, two of which, SULT1A1 and SULT1A2, encode proteins with "TS PST-like" activity. We recently reported common nucleotide polymorphisms for SULT1A1 that are associated with variations in platelet TS PST activity and thermal stability. In the present experiments, we set out to determine whether functionally significant DNA polymorphisms also might exist for SULT1A2, to compare the biochemical properties of all common allozymes encoded by SULT1A2 and SULT1A1, and to study phenol SULT genotype-phenotype correlations in the human liver. We phenotyped 61 human liver biopsy samples for TS PST thermal stability and activity. The open reading frames of SULT1A2 and SULT1A1 then were amplified with the polymerase chain reaction and sequenced for each of these hepatic tissue samples. We observed 13 SULT1A2 alleles that encoded 6 allozymes. These alleles were in linkage disequilibrium with alleles for SULT1A1. Biochemical characterization of common allozymes encoded by both genes suggested that SULT1A1 was primarily responsible for "TS PST phenotype" in the human liver. In summary, both SULT1A2 and SULT1A1 have a series of common alleles encoding enzymes that differ functionally and are associated with individual differences in phenol SULT properties in the liver.

Purification and immunochemical characterization of a rat liver sulphotransferase conjugating paracetamol.

Paracetamol sulphotransferase (ST) was purified 250-fold from male rat liver, and the pure enzyme used to elicit antibodies in rabbit. The enzyme was active towards paracetamol at pH 9.0, as well as towards several commonly used drugs, and formed sulphates at both O- and N-atoms. Comparison of the substrate specificity of paracetamol ST with that of aryl sulphotransferases isolated by other workers suggested that we have purified a previously unknown isoenzyme of rat liver ST, although the difficulties of characterization of STs based on their substrate specificities is noted. The antibody preparation recognized only one polypeptide (Mr = 35,000) on immunoblot analysis of rabbit liver cytosol, corresponding to purified paracetamol ST. Analysis of the tissue distribution of this protein demonstrated that its expression was restricted to the liver, as was the enzyme activity. The observed sex difference in paracetamol ST (males greater than females) was determined by immunoblot analysis to be the result of reduced enzyme protein levels in females. In human liver cytosol, the antibody recognized two polypeptides, probably corresponding to M- and P-phenol STs, suggesting significant sequence similarity between rat and human phenol sulphotransferases.

Sulfation in male reproductive organs. Bull and boar testis phenol sulfotransferases.

Phenol sulfotransferases (PST) from bull and boar testis were partially purified and characterized. A single form of PST adsorbed on DEAE-cellulose was found in the bull testis, whereas from boar testis two different peaks of PST activity were separated. The bull testis PST and both boar testis enzymes were active with p-nitrophenol and adrenalin. They all showed higher affinity to pNP than to adrenalin and were inhibited by these substrates at higher concentrations. Their optimal pH was at 8.5. Bull testis PST and boar PST II which were adsorbed on DEAE-cellulose were thermostable, whereas boar PST I was thermolabile. Those three PST forms differed in sensitivity to 2,6-dichloro-4-nitrophenol (DCNP), N-ethyl maleimide (NEM), iodoacetamide (IAA) and phenylglyoxal (PG). Bull and boar PST II were more rapidly inactivated in the presence of DCNP than boar PST I. In the presence of NEM, the--SH groups reagent, the bull phenol sulfotransferase and boar PST I lost their activity, whereas the activity of boar PST increased. Also iodoacetamide, another--SH group modificator, raised boar PST II activity and decreased boar PST I activity. DTT, which protects thiol groups, had an opposite effect on the enzymes studied than NEM. Phenylglyoxal, a reagent specific for arginine residues inhibited bull testis PST and both boar phenol sulfotransferases. Substrate protection experiments were also performed to determine the localization of reactive groups in bull and boar testis phenol sulfotransferases.

Purification and reaction mechanism of arylsulfate sulfotransferase from Haemophilus K-12, a mouse intestinal bacterium.

A novel type of sulfotransferase, arylsulfate sulfotransferase [EC 2.8.2.22], was purified to homogeneity from Haemophilus K-12, a mouse intestinal bacterium. The purified enzyme (M(r) 290,000) is composed of four subunits (M(r) 70,000). The best donor substrate was 4-methylumbelliferyl sulfate, followed by beta-naphthyl sulfate, p-nitrophenyl sulfate (PNS), and alpha-naphthyl sulfate. The best acceptor substrate was alpha-naphthol, followed by phenol and resorcinol. The apparent Km for PNS using phenol as an acceptor and that for phenol and resorcinol. The apparent Km for PNS using phenol as an acceptor and that for phenol using PNS as a donor substrate were determined to be 0.095 and 0.71 mM, respectively. One of the reaction products, p-nitrophenol inhibited the enzyme noncompetitively with respect to PNS, but competitively with respect to alpha-naphthol. The Ki values of PNP for PNS and alpha-naphthol were 0.89 and 0.12 mM, respectively. The other reaction product, alpha-naphthyl sulfate, inhibited the enzyme competitively with respect to PNS, but non-competitively with respect to alpha-naphthol. The Ki values of alpha-naphthyl sulfate for PNS and for alpha-naphthol were 2.72 and 1.7 mM. These results suggest that the sulfate transfer reaction proceeds according to a ping pong bi bi mechanism.

Improved purification of arylsulfate sulfotransferase from human intestinal bacterium by using polyclonal antibody.

Arylsulfate sulfotransferase (ASST) from a human intestinal bacterium stoichiometrically catalyzed the transfer of the sulfate group of phenylsulfate esters to phenolic compounds. Polyclonal antibodies against ASST were obtained from rabbit sera. These antisera did not inhibit ASST activity. ASST was recognized by the IgG fraction of the antisera, but rat liver phenol sulfotransferase did not show cross-reactivity to ASST on Western blot (immunoblot) analysis. The ASST was purified by an anti-ASST immobilized affinity column chromatography to homogeneity on SDS-PAGE. The NH2-terminal amino acid and partial sequence of the purified enzyme were serine and SVKYSFEDHIINRQYEAEQAMLAKF, respectively. We corrected the previous result that the NH2-terminal of ASST was arginine.

Human liver triiodothyronine sulfotransferase: copurification with phenol sulfotransferases.

To ascertain whether triiodothyronine (T3) sulfotransferase coeluted with the known phenol sulfotransferases (PSTs) during purification, human liver thermostable PST, thermolabile PST, and T3 sulfotransferase were assayed with p-nitrophenol, dopamine, and T3, respectively. Thermostable PST eluted from an ion-exchange column in two sequential peaks of activity (Peaks I and II), followed by a peak of thermolabile PST activity. There were three peaks of T3 sulfotransferase with thermostable PST: two within thermostable PST Peak I, and one peak of T3 sulfotransferase activity within thermostable PST Peak II. There was a minor peak of T3 sulfotransferase with thermolabile PST. Further purification of thermostable Peak I showed coelution of T3 sulfotransferase with thermostable PST during gel filtration and affinity chromatography. SDS-PAGE revealed a major protein band at 31 kDa. Dehydroepiandrosterone sulfotransferase comprised only 4% of the final activity. This report demonstrates coelution of T3 sulfotransferase with thermostable PST, shows a potential additional isozyme of T3 sulfotransferase, and points out the apparent minimal role of dehydroepiandrosterone sulfotransferase in T3 sulfation. The findings support the hypothesis that thermostable PST is the predominant human liver T3 sulfotransferase activity.

Central monoamine oxidase and phenolsulfotransferase activities in spontaneously hypertensive rats.

The activities of monoamine oxidase and phenolsulfotransferase in the hypothalamus and anterior pituitary gland of spontaneously hypertensive rats and the normotensive control (Wistar Kyoto rat) rats were investigated. The monoamine oxidase activity (determined using dopamine as substrate) in both these tissues was not significantly different between the normo- and hypertensive animals. Hypothalamic phenolsulfotransferase does not sulfate-conjugate dopamine at pH of 6.5 and pituitary phenolsulfotransferase does not sulfate-conjugate dopamine or 3,4-dihydroxyphenylacetic acid at the same pH. Hypothalamic phenolsulfotransferase activity determined using 3,4-dihydroxyphenylacetic acid as substrate was significantly higher in the spontaneously hypertensive than the Wistar Kyoto rats, while pituitary enzyme (determined using phenol as substrate) was the same in both strains of animals. We proposed that in the spontaneously hypertensive rats the higher level of hypothalamic phenolsulfotransferase could (by removing 3,4-dihydroxyphenylacetic acid as sulfated acid) increase the deamination of dopamine by monoamine oxidase. This could in turn result in the presence of high amount of sulfated 3,4-dihydroxyphenylacetic acid in the anterior pituitary gland reported in our earlier study, and be partly responsible for the reduced central dopaminergic activity found in the hypertensive rats.

Substrate specificity and some properties of phenol sulfotransferase from human intestinal Caco-2 cells.

The phase II metabolic reactions, sulfation and glucuronidation, were studied in a human colon carcinoma cell line (Caco-2), which has been developed as a model of intestinal enterocytes. Phenol sulfotransferase (PST, EC 2.4.2.1) was isolated from Caco-2 cells cultured for 7, 14 and 21 days. The enzyme catalyzed the sulfation of both p-nitrophenol and catecholamines (e.g., dopamine) as well as most catecholamine metabolites. The affinity (Km) of PST for dopamine was much higher than for p-nitrophenol, and the specific activity of PST with both substrates increased with the age of the cells. The thermal stability of Caco-2 PST increased with cell age and was not dependent on the acceptor substrate used. The thermolabile PST from 7-day old cells was more sensitive to NEM than was the thermostable enzyme from 21-day old cells. No UDP-glucuronyltransferase (EC 2.4.1.17) activity was detected in 7-, 14- and 21-day old Caco-2 cells with any of the methods used.

Structural similarity and diversity of sulfotransferases.

In the present study, four new forms of aryl sulfotransferase cDNAs have been isolated and their structures determined. A compilation of primary structures of 16 different sulfotransferases, including enzymes metabolizing endogenous chemicals and xenobiotics, showed a considerable extent of similarity among bacterial, plant and mammalian species, and indicates that these enzymes constitute a supergene family. Aryl sulfotransferase and estrogen sulfotransferase are shown to belong to a single gene family (ST1) which consists of at least four subfamilies, whereas, based on the sequence similarity, hydroxysteroid sulfotransferases constitute a distinct family (ST2). Little or no clear similarity was observed between the primary structures of enzymes N-sulfating aminosugars and those sulfating hydrophobic chemicals such as phenols, alcohols or amines, indicating that both types of enzymes diverged early in their evolutionary history. Two regions in the C-terminal parts are, however, conserved among all enzymes examined, which suggests a possibly essential role of these sites for the binding of a PAPS cofactor or for sulfate transfer.

Heterologous systems for expression of mammalian sulfotransferases.

This paper describes the use of both mammalian and bacterial expression systems as tools to study the structural and functional relationships of proteins encoded by cDNAs to both rat and human aryl sulfotransferases. In particular, we describe the use of the mammalian COS cell system for functional expression studies, and the use of Escherichia coli for the expression and purification of a sulfotransferase fusion protein suitable as an antigen for the generation of sulfotransferase antibodies.

Identification of a novel flavonoid glycoside sulfotransferase in Arabidopsis thaliana.

The discovery of sulfated flavonoids in plants suggests that sulfation may play a regulatory role in the physiological functions of flavonoids. Sulfation of flavonoids is mediated by cytosolic sulfotransferases (SULTs), which utilize 3'-phosphoadenosine 5'-phosphosulfate (PAPS) as the sulfate donor. A novel SULT from Arabidopsis thaliana, designated AtSULT202B7 (AGI code: At1g13420), was cloned and expressed in Escherichia coli. Using various compounds as potential substrates, we demonstrated, for the first time, that AtSULT202B7 displayed sulfating activity specific for flavonoids. Intriguingly, the recombinant enzyme preferred flavonoid glycosides (e.g. kaempferol-3-glucoside and quercetin-3-glucoside) rather than their aglycone counterparts. Among a series of hydroxyflavones tested, AtSULT202B7 showed the enzymatic activity only for 7-hydroxyflavone. pH-dependency study showed that the optimum pH was relatively low (pH 5.5) compared with those (pH 6.0-8.5) previously reported for other isoforms. Based on the comparison of high performance (pressure) liquid chromatography (HPLC) retention times between sulfated kaempferol and the deglycosylated product of sulfated kaempferol-3-glucoside, the sulfation site in sulfated kaempferol-3-glucoside appeared to be the hydroxyl group of the flavonoid skeleton. In addition, by using direct infusion mass spectrometry, it was found that the sulfated product had one sulfonate group within the molecule. These results indicated that AtSULT202B7 functions as a flavonoid glycoside 7-sulfotransferase.

Cloning, expression and purification of arylsulfate sulfotransferase from Eubacterium A-44.

A gene (astA) encoding arylsulfate sulfotransferase (ASST), which transfers a sulfate group from phenolic sulfate esters to phenolic acceptors, was cloned from a Eubacterium A-44 genomic library. The probe (1.5 kb fragment) for the astA gene was prepared from the PCR product of the primers produced using two internal amino acid sequences of ASST, which had been purified from Eubacterium A-44. The astA gene was cloned into the pKF3 vector. Its sequence revealed a 1863 bp open reading frame (ORF) encoding a protein containing 620 amino acids with a secretary signal peptide, and showed 91% homology (identity) to Eubacterium rectale IIIH previously reported. The cloned astA gene was expressed under the T7 promoter of the expression vectors, pET-39b(+) and pET-26b(+), in Escherichia coli BL21 (DE3), and the expressed ASSTs were purified using His Bind column chromatography. The specific activities of the purified ASSTs were 25.6 micromol/min/mg and 37.1 micromol/min/mg, respectively.