Purification and characterization of glutathione S-transferases P, S and N. Isolation from rat liver of Yb1 Yn protein, the existence of which was predicted by subunit hybridization in vitro.

The glutathione S-transferases are dimeric proteins and comprise subunits of Mr 25 500 (Ya), 26 500 (Yn), 27 000 (Yb1 and Yb2) and 28 500 (Yc). Enzymes containing Ya and/or Yc subunits have been isolated as have forms containing binary combinations of Yn, Yb1 and Yb2 subunits. To date only one enzyme, transferase S, has been described that is a YbYn heterodimer [Hayes & Chalmers (1983) Biochem. J. 215, 581-588]; the identity of the Yb monomer found in transferase S has not been reported previously. The identification and isolation of a YnYn dimer (transferase N) from rat testis is now described. This has enabled structural and functional comparisons to be made between Yb1, Yb2 and Yn monomers. Reversible dissociation experiments between the YnYn and Yb1Yb1 homodimers and between the YnYn and Yb2Yb2 homodimers demonstrated that Yn monomers can hybridize with both Yb1 and Yb2 monomers. Reversible dissociation of transferases N and C (Yb1Yb2) showed that both Yb1 and Yb2 monomers can hybridize with Yn monomers under competitive conditions. The hydridization data suggest that transferase S represents the Yb2Yn subunit combination. A knowledge of the elution position from chromatofocusing columns of the Yb1Yn hybrid that was formed in vitro enabled a purification scheme to be devised for an enzyme from rat liver (transferase P) believed to consist of Yb1Yn subunits. A comparison of the chromatographic behaviour of the YnYn, Yb1Yb1 and Yb2Yb2 dimers on chromatofocusing and hydroxyapatite columns with the behaviour of transferases P and S on the same matrices suggests these two enzymes may be identified as the Yb1Yn and Yb2Yn dimers respectively. The catalytic activities and the inhibitory effects of non-substrate ligands on transferases P and S are significantly different and again suggest they comprise Yb1 and Yn subunits and Yb2 and Yn subunits respectively; transferase P exhibits a 6-fold higher specific activity for 1,2-dichloro-4-nitrobenzene than does transferase S, whereas, conversely, transferase S possesses a 9-fold higher specific activity for trans-4-phenylbut-3-en-2-one than does transferase P. The quaternary structure of transferases P and S was verified by using peptide mapping and 'Western blotting' techniques.     

Purification and characterization of five forms of glutathione transferase from human uterus

Five glutathione transferase (GST) forms were purified from human uterus by glutathione-affinity chromatography followed by chromatofocusing, and their structural, kinetic and immunological properties were investigated. Upon SDS/polyacrylamide slab gel electrophoresis all forms resulted composed of two subunits of identical molecular size. GST V (pI 4.5) is a dimer of 23-kDa subunits. GST I (pI 6.8) and GST IV (pI 4.9) are dimers of 24-kDa subunits whereas GST II (pI 6.1) and GST III (pI 5.5) are dimers of 26.5-kDa subunits. GST V accounts for about 85-90% of the activity whereas the other isoenzymes are present in trace quantities. On the basis of the molecular mass of the subunits, amino acid composition, substrate specificities, sensitivities to inhibitors, CD spectra and immunological studies, GST V appeared very similar to transferase pi. Structural and immunological studies provide evidence that GST IV is closely related to the less 'basic' transferase (GST pI 8.5) of human skin. Extensive similarities have been found between GST II and GST III. The comparison includes amino acid compositions, subunits molecular size and immunological properties. The two enzymes, however, are kinetically distinguishable. The data presented also indicate that GST II and GST III are related to transferase mu and to transferase psi of human liver. Even though GST I has a subunit molecular mass identical to GST IV, several lines of evidence, including catalytic and immunological properties, indicate that they are different from each other. GST I seems not to be related to any of known human transferases, suggesting that it may be specific for the uterus.     

Differential tissue expression of the glutathione transferase multigene family.

The content of GSH transferase mRNAs in poly(A)-containing RNA isolated from eight rat tissues was examined by immunoprecipitation of cell-free translation products and by Northern blotting. Considerable tissue-specific distribution and heterogeneity of immunoprecipitable GSH transferase subunits 1 and 2 synthesized in vitro was observed. These results were confirmed by Northern blotting using a 32P-labelled subunit 1 cDNA probe. The same probe, used in a Southern blot analysis of genomic DNA, provided confirmation that GSH transferase subunits 1 and 2 comprise a multigenic family in the rat. The results show that the selection of cDNA clones coding for chosen subunits can be made easier by making use of qualitative and quantitative tissue differences in GSH transferase mRNAs.     

Similar structures in gamma-carboxymuconolactone decarboxylase and beta-ketoadipate succinyl coenzyme A transferase.

gamma-Carboxymuconolactone decarboxylase (EC 4.1.1.44) and beta-ketoadipate succinyl coenzyme A transferase (EC 2.8.3.6) mediate different steps in the beta-ketoadipate pathway. Antisera prepared against the Pseudomonas putida transferase cross-reacted immunologically with the decarboxylase from the same organism. The transferase is formed by association of two nonidentical protein subunits. The NH2-terminal amino acid sequences of the two nonidentical transferase subunits resembled each other and also were similar to the NH2-terminal amino acid sequence of the decarboxylase.     

Differences in the occurrence of glutathione transferase isoenzymes in rat lung and liver

Cytosolic GSH transferases have been purified from rat lung by affinity chromatography followed by chromatofocusing. On the criteria of order of elution, substrate specificity, apparent subunit Mr, sensitivity to inhibitors, and reaction with antibodies, transferase subunits equivalent to subunits 2, 3, and 4, in the binary combinations occurring in liver, were identified. However, subunit 1 (and therefore transferases 1-1 and 1-2) was not detected. The most conspicuous difference is the presence in lung of a new form, eluting at pH 8.7, which is not detected in rat liver. This isoenzyme (transferase "pH 8.7") is characterized by its low apparent subunit Mr and high efficiency in the conjugation of glutathione with anti-benzo(a)pyrene-7,8-dihydrodiol-9,10-epoxide, considered the ultimate carcinogen of benzo(a)-pyrene.     

SPARK--a novel method to monitor ribosomal peptidyl transferase activity

The key enzymatic activity of the ribosome is catalysis of peptide bond formation. This reaction is a target for many clinically important antibiotics. However, the molecular mechanisms of the peptidyl transfer reaction, the catalytic contribution of the ribosome, and the mechanisms of antibiotic action are still poorly understood. Here we describe a novel, simple, convenient, and sensitive method for monitoring peptidyl transferase activity (SPARK). In this method, the ribosomal peptidyl transferase forms a peptide bond between two ligands, one of which is tritiated whereas the other is biotin-tagged. Transpeptidation results in covalent attachment of the biotin moiety to a tritiated compound. The amount of the reaction product is then directly quantified using the scintillation proximity assay technology: binding of the tritiated radioligand to the commercially available streptavidin-coated beads causes excitation of the bead-embedded scintillant, resulting in detection of radioactivity. The reaction is readily inhibited by known antibiotics, inhibitors of peptide bond formation. The method we developed is amenable to simple automation which makes it useful for screening for new antibiotics. The method is useful for different types of ribosomal research. Using this method, we investigated the effect of mutations at a universally conserved nucleotide of the active site of 23S rRNA, A2602 (Escherichia coli numbering), on the peptidyl transferase activity of the ribosome. The activities of the in vitro reconstituted mutant subunits, though somewhat reduced, were comparable with those of the subunits assembled with the wild-type 23S rRNA, indicating that A2602 mutations do not abolish the ability of the ribosome to catalyze peptide bond formation. Similar results were obtained with double mutants carrying mutations at A2602 and another universally conserved nucleotide in the peptidyl transferase center, A2451. The obtained results agree with our previous conclusion that the ribosome accelerates peptide bond formation primarily through entropic rather than chemical catalysis.     

Analysis of glutathione S-transferase from human liver by isoelectric focusing in a urea minigel system.

A method for the rapid analysis of isozyme subunits of glutathione transferase (GST) from human liver is described. Following purification of enzyme protein to electrophoretic homogeneity on columns of GSH-agarose, pooled transferase fractions were concentrated by ultrafiltration and subjected to further fractionation and analysis by urea-isoelectric focusing in minigels using a Hoefer Mighty Small II electrophoresis system. These methods combined with immunoblotting techniques permitted the resolution, detection, and eventual analysis of up to six different subunits of the alpha isozyme of human GST and at least three to four different forms of the pi isozyme of the transferase rapidity, accuracy, and sensitivity of the methodology may prove useful to the analysis and quantification of GST subunits in biopsies of malignant human tissue and to the development of effective chemotherapeutic regimens.     

A homogeneous resonance energy transfer assay for phosphopantetheinyl transferase

Phosphopantetheinyl transferase plays an essential role in activating fatty acid, polyketide, and nonribosomal peptide biosynthetic pathways, catalyzing covalent attachment of a 4′-phosphopantetheinyl group to a conserved residue within carrier protein domains. This enzyme has been validated as an essential gene to primary metabolism and presents a target for the identification of antibiotics with a new mode of action. Here we report the development of a homogeneous resonance energy transfer assay using fluorescent coenzyme A derivatives and a surrogate peptide substrate that can serve to identify inhibitors of this enzyme class. This assay lays a blueprint for translation of these techniques to other transferase enzymes that accept fluorescent substrate analogues.     

Studies of the relationship between the catalytic activity and binding of non-substrate ligands by the glutathione S-transferases.

The dimeric enzyme glutathione S-transferase B is composed of two dissimilar subunits, referred to as Ya and Yc. Transferase B (YaYc) and two other transferases that are homodimers of the individual Ya and Yc subunits were purified from rat liver. Inhibition of these three enzymes by Indocyanine Green, biliverdin and several bile acids was investigated at different values of pH (range 6.0-8.0). Indocyanine Green, biliverdin and chenodeoxycholate were found to be effective inhibitors of transferases YaYc and YcYc at low (pH 6.0) but not high (pH 8.0) values of pH. Between these extremes of pH intermediate degrees of inhibition were observed. Cholate and taurochenodeoxycholate, however, were ineffective inhibitors of transferase YcYc at all values of pH. The observed differences in bile acids appeared to be due, in part, to differences in their state of ionization. In contrast with the above results, transferase YaYa was inhibited by at least 80% by the non-substrate ligands at all values of pH. These effects of pH on the three transferases could not be accounted for by pH-induced changes in the enzyme's affinity for the inhibitor. Thus those glutathione S-transferases that contain the Yc subunit are able to act simultaneously as both enzymes and binding proteins. In addition to enzyme structure, the state of ionization of the non-substrate ligands may also influence whether the transferases can perform both functions simultaneously.     

The major isozyme of rat cardiac glutathione transferases. Its correspondence to hepatic transferase X

A major isozyme of rat heart glutathione transferase was purified to homogeneity by Sephadex G-200 gel filtration, ammonium sulfate precipitation, CM-cellulose chromatography and affinity chromatography on S-hexylglutathione-linked Sepharose 6B. The purified isozyme was a dimer with an apparent relative molecular mass of 50 000 composed of two Yb-size subunits (Mr = 26 500). The isozyme is immunologically related to rat liver glutathione transferase X and 3-3, especially closely to transferase X, and no immunological cross-reactivity with subunits 1 and 2 of hepatic glutathione transferases was observed. The isoelectric point (pI = 6.9) of the isozyme was identical with and the substrate specificity was very similar to transferase X. Thus, the cardiac near-neutral isozyme is considered to be identical to glutathione transferase X recognized in rat liver. The amount of this near-neutral isozyme estimated to be present in heart tissue is 70 micrograms/g. The isozyme has relatively high activities towards alpha, beta-unsaturated carbonyl compounds such as trans-4-phenyl-3-buten-2-one and trans-4-hydroxynon-2-enal. The latter is a cytotoxic product resulting from lipid peroxidation of polyunsaturated fatty acids, and the cardiac isozyme may play a physiologically significant role with glutathione conjugation of this compound. In addition to the near-neutral isozyme, acidic forms with isoelectric points of 4.9, 5.2 and 5.5 were partially purified; some of them are considered to consist of subunits immunologically related to transferase X.     

Nonequivalence of the two subunits of horse erythrocyte glutathione transferase in their reaction with sulfhydryl reagents

Glutathione transferase (EC 2.5.1.18) from horse erythrocytes has been purified and some molecular and kinetic properties have been investigated. It appears to be a dimeric protein composed of subunits of about 23 kDa, indistinguishable either in sodium dodecyl sulfate or in urea electrophoresis. Amino acid composition, substrate specificities, sensitivity to inhibitors, CD spectra, and immunological studies provide evidence that the horse enzyme is related to the pi class transferases. This enzyme has only two reactive thiol groups/dimer whose integrity appears to be essential for the activity. A peculiar feature of these protein thiol groups is that they react nonidentically with a number of thiol blocking reagents, i.e. iodacetamide, bromopyruvate, N-ethylmaleimide, and 1-chloro-2,4-dinitrobenzene. Also many disulfides react with one thiol group 5- to 10-fold more rapidly than with the other. The two mixed disulfides so formed also have different rates of reactivation by dithiothreitol. All the structural and kinetic data reported in this paper indicate a nonsymmetrical association of two identical subunits, or alternatively heterodimeric structure with subunits of very similar charge and size.     

2-Arylindole-3-acetamides: FPP-competitive inhibitors of farnesyl protein transferase

A series of 2-arylindole-3-acetamide farnesyl protein transferase inhibitors has been identified. The compounds inhibit the enzyme in a farnesyl pyrophosphate-competitive manner and are selective for farnesyl protein transferase over the related enzyme geranylgeranyltransferase-I. A representative member of this series of inhibitors demonstrates equal effectiveness against HDJ-2 and K-Ras farnesylation in a cell-based assay when geranylgeranylation is suppressed.     

Acidic glutathione transferase from human heart. Characterization and N-terminal sequence determination

The anionic glutathione transferase of human heart has been purified to homogeneity by using DEAE-cellulose, affinity chromatography, and FPLC. The enzyme has an isoelectric point at pH 4.75 and has an electrophoretic mobility on SDS-PAGE identical to placental transferase pi, indicating that the heart enzyme is formed by two similar subunits of 23,000 Mr. Upon isoelectric focusing on ampholine PAG plates the enzyme recovered from FPLC gave two bands of activity at pH 4.75 and 4.9 which were reduced to essentially a single band at pH 4.75 after incubation with dithiothreitol. In the immunodiffusion experiment, the heart enzyme gave a positive precipitin line with the antibodies against transferase pi but not with antibodies prepared against the "basic" transferase of human skin or against the "near-neutral" transferase of human uterus. The substrate specificities, the sensitivities to characteristic inhibitors, the amino acid composition, together with the immunological studies, strongly indicate that the anionic enzyme of human heart is closely related to the transferase pi of human placenta. The N-terminal amino acid sequence of the first 48 residues was determined and compared with the N-terminal region of other reported human glutathione transferase sequences. The heart enzyme differs from the placental enzyme in a single residue (Trp instead of Arg in the 28th position) further supporting their similarity.     

The molecular basis of coupling of translocation and N-glycosylation

Protein translocation and N-glycosylation are essential coordinated cellular processes that are mediated by the translocon and the oligosaccharyl transferase (OT), respectively. The recent identification of several specific interactions between the OT subunits and the translocon provides a molecular basis for the coupling of these two processes. Data suggest that multiple OT isoforms with different affinities for the translocon and ribosome and with heterogeneous subunit composition might exist in the endoplasmic reticulum (ER) membrane, thereby providing a means of regulating protein N-glycosylation.     

Glutathione transferases in primary rat hepatomas: the isolation of a form with GSH peroxidase activity

A previously uncharacterized glutathione (GSH) transferase which is not apparent in normal liver, accounts for at least 25% of the soluble GSH transferase content of primary hepatomas induced by feeding N,N-dimethyl-4-aminoazobenzene. This enzyme is readily isolated, has an isoelectric point of 6.8, is composed of two identical subunits of apparent M_r 26 000 and has GSH transferase activity towards a number of substrates including benzo(a)pyrene-7,8-diol-9,10-oxide. It is unusual in that it has GSH peroxidase activity towards fatty acid hydroperoxides but not towards the model substrates, cumene hydroperoxide and t-butyl hydroperoxide. It has been shown by tryptic peptide analysis to be distinct from GSH transferases composed of subunits 1, 2, 3,4 or 6 and has been designated GSH transferase 7-7.     

Influence of cholesteryl 14-methylhexadecanoate on some ribosomal functions required for peptide elongation

1. Polyribosomes and ribosomal subunits from rat liver were adsorbed on a cellulosic ion-exchange adsorbent, freeze-dried and extracted with organic solvents. The activity of extracted particles in peptide elongation was tested in the presence of purified peptideelongation factors. 2. Chloroform-methanol mixture (2:1, v/v) extracted 1.87+/-0.15 pmol of cholesteryl 14-methylhexadecanoate/pmol of the smaller ribosomal subunit and 0.92+/-0.11 pmol/pmol of the larger subunit. 3. In the presence of transferase I, extracted polyribosomes and 40S subunits bound more phenylalanyl-tRNA than did control non-extracted particles. The same binding as in control mixtures was obtained with extracted particles supplemented with cholesteryl 14-methylhexadecanoate in quantities corresponding to those extracted. 4. The polymerization of phenylalanine was greatly decreased with extracted polyribosomes and subunits and addition of the cholesteryl ester could not fully restore the original activity. 5. Extraction significantly decreased the activity of the P site of peptidyl transferase and normal activity was recovered after the addition of the ester. The A site of peptidyl transferase in extracted polyribosomes showed an increased activity when compared with non-extracted polyribosomes. 6. Cholesteryl 14-methylhexadecanoate apparently affects the function of the ribosomal A site and peptidyl transferase site and probably also that of the guanosine triphosphatase site and P site. The presence of different amounts of the ester in polyribosomes may be one of the mechanisms modulating peptide elongation at the ribosomal level.     

Identification of influenza virus inhibitors which disrupt of viral polymerase protein-protein interactions

Due to their ability to rapidly mutate, influenza viruses quickly develop resistance against many antiviral substances, leading to an urgent need for new compounds. The trimeric viral polymerase complex, a major target for the development of new inhibitors, must be assembled from the PB1, PB2, and PA subunits for successful infection. Here, we describe ELISA-based assays which allow the identification of peptides which impair polymerase complex formation. Since the protein-protein interaction domains of the viral polymerase are highly conserved, these inhibitors are also predicted to be active against a broad range of influenza strains. Using this method, identification of small molecules and lead compounds against influenza A and B viruses should be feasible.     

Inhibition of hepatic and extrahepatic glutathione S-transferases by primary and secondary bile acids.

Glutathione S-transferases are a complex family of dimeric proteins that play a dual role in cellular detoxification; they catalyse the first step in the synthesis of mercapturic acids, and they bind potentially harmful non-substrate ligands. Bile acids are quantitatively the major group of ligands encountered by the glutathione S-transferases. The enzymes from rat liver comprise Yk (Mr 25 000), Ya (Mr 25 500), Yn (Mr 26 500), Yb1, Yb2 (both Mr 27 000) and Yc (Mr 28 500) monomers. Although bile acids inhibited the catalytic activity of all transferases studied, the concentration of a particular bile acid required to produce 50% inhibition (I50) varies considerably. A comparison of the I50 values obtained with lithocholate (monohydroxylated), chenodeoxycholate (dihydroxylated) and cholate (trihydroxylated) showed that, in contrast with all other transferase monomers, the Ya subunit possesses a relatively hydrophobic bile-acid-binding site. The I50 values obtained with lithocholate and lithocholate 3-sulphate showed that only the Ya subunit is inhibited more effectively by lithocholate than by its sulphate ester. Other subunits (Yk, Yn, Yb1 and Yb2) were inhibited more by lithocholate 3-sulphate than by lithocholate, indicating the existence of a significant ionic interaction, in the bile-acid-binding domain, between (an) amino acid residue(s) and the steroid ring A. By contrast, increasing the assay pH from 6.0 to 7.5 decreased the inhibitory effect of all bile acids studied, suggesting that there is little significant ionic interaction between transferase subunits and the carboxy group of bile acids. Under alkaline conditions, low concentrations (sub-micellar) of nonsulphated bile acids activated Yb1, Yb2 and Yc subunits but not Yk, Ya and Yn subunits. The diverse effects of the various bile acids studied on transferase activity enables these ligands to be used to help establish the quaternary structure of individual enzymes. Since these inhibitors can discriminate between transferases that appear to be immunochemically identical (e.g. transferases F and L), bile acids can provide information about the subunit composition of forms that cannot otherwise be distinguished.     

Role for the highly conserved region of domain IV of 23S-like rRNA in subunit-subunit interactions at the peptidyl transferase centre.

The function of the highly conserved and accessible region of domain IV of 23S rRNA (positions 1900-1981 in Escherichia coli 23S rRNA) was investigated by subjecting it to a random mutagenesis procedure that produced single-site mutations efficiently. Nine single-site mutants were selected that were recessive lethal. High levels of mutated 23S rRNA were expressed in E. coli and extracted ribosomes were investigated for their content of mutated rRNA. The peptidyl transferase activity of the ribosomes was also estimated using a newly developed method involving selective inhibition of chromosome-encoded ribosomes by clindamycin. Two of the mutants, U1940A and U1955G, yielded 50S subunits that were defective in subunit-subunit association but active in peptidyl transferase activity and five, U1926C, U1946C, U1979C, U1982A and G1984A, produced 50S subunits that were defective in both subunit-subunit interactions and peptidyl transferase activity. We infer that the large conserved rRNA region generates a complex structure that plays an essential role in maintaining and modulating subunit-subunit interactions and argue that its involvement in the peptidyl transferase centre is secondary, possibly involving the correct alignment of protein L2.