Tyr115, gln165 and trp209 contribute to the 1, 2-epoxy-3-(p-nitrophenoxy)propane-conjugating activity of glutathione S-transferase cGSTM1-1.

We investigated the epoxidase activity of a class mu glutathione S-transferase (cGSTM1-1), using 1,2-epoxy-3-(p-nitrophenoxy)propane (EPNP) as substrate. Trp209 on the C-terminal tail, Arg107 on the alpha4 helix, Asp161 and Gln165 on the alpha6 helix of cGSTM1-1 were selected for mutagenesis and kinetic studies. A hydrophobic side-chain at residue 209 is needed for the epoxidase activity of cGSTM1-1. Replacing Trp209 with histidine, isoleucine or proline resulted in a fivefold to 28-fold decrease in the k(cat)(app) of the enzyme, while a modest 25 % decrease in the k(cat)(app) was observed for the W209F mutant. The rGSTM1-1 enzyme has serine at the correponding position. The k(cat)(app) of the S209W mutant is 2. 5-fold higher than that of the wild-type rGSTM1-1. A charged residue is needed at position 107 of cGSTM1-1. The K(m)(app)(GSH) of the R107L mutant is 38-fold lower than that of the wild-type enzyme. On the contrary, the R107E mutant has a K(m)(app)(GSH) and a k(cat)(app) that are 11-fold and 35 % lower than those of the wild-type cGSTM1-1. The substitutions of Gln165 with Glu or Leu have minimal effect on the affinity of the mutants towards GSH or EPNP. However, a discernible reduction in k(cat)(app) was observed. Asp161 is involved in maintaining the structural integrity of the enzyme. The K(m)(app)(GSH) of the D161L mutant is 616-fold higher than that of the wild-type enzyme. In the hydrogen/deuterium exchange experiments, this mutant has the highest level of deuteration among all the proteins tested. We also elucidated the structure of cGSTM1-1 co-crystallized with the glutathionyl-conjugated 1, 2-epoxy-3-(p-nitrophenoxy)propane (EPNP) at 2.8 A resolution. The product found in the active site was 1-hydroxy-2-(S-glutathionyl)-3-(p-nitrophenoxy)propane, instead of the conventional 2-hydroxy isomer. The EPNP moiety orients towards Arg107 and Gln165 in dimer AB, and protrudes into a hydrophobic region formed by the loop connecting beta1 and alpha1 and part of the C-terminal tail in dimer CD. The phenoxyl ring forms strong ring stacking with the Trp209 side-chain in dimer CD. We hypothesize that these two conformations represent the EPNP moiety close to the initial and final stages of the reaction mechanism, respectively.

Co-expression of glutathione S-transferase with methionine aminopeptidase: a system of producing enriched N-terminal processed proteins in Escherichia coli.

We describe here an Escherichia coli expression system that produces recombinant proteins enriched in the N-terminal processed form, by using glutathione S-transferase cGSTM1-1 and rGSTT1-1 as models, where c and r refer to chick and rat respectively. Approximately 90% of the cGSTM1-1 or rGSTT1-1 overexpressed in E. coli under the control of a phoA promoter retained the initiator methionine residue that was absent from the mature isoenzymes isolated from tissues. The amount of initiator methionine was decreased to 40% of the expressed cGSTM1-1 when the isoenzyme was co-expressed with an exogenous methionine aminopeptidase gene under the control of a separate phoA promoter. The recombinant proteins expressed were mainly methionine aminopeptidase. The yield of cGSTM1-1 was decreased to 10% of that expressed in the absence of the exogenous methionine aminopeptidase gene. By replacing the phoA with its natural promoter, the expression of methionine aminopeptidase decreased drastically. The yield of the co-expressed cGSTM1-1 was approx. 60% of that in the absence of the exogenous methionine aminopeptidase gene; approx. 65% of the initiator methionine residues were removed from the enzyme. Under similar conditions, N-terminal processing was observed in approx. 70% of the recombinant rGSTT1-1 expressed. By increasing the concentration of phosphate in the growth medium, the amount of initiator methionine on cGSTM1-1 was decreased to 14% of the overexpressed isoenzymes, whereas no further improvement could be observed for rGSTT1-1. The initiator methionine residue does not affect the enzymic activities of either cGSTM1-1 or rGSTT1-1. However, the epoxidase activity and the 4-nitrobenzyl chloride-conjugating activity of the purified recombinant rGSTT1-1 are markedly higher that those reported recently for the same isoenzyme isolated from rat livers.

The three-dimensional structure of an avian class-mu glutathione S-transferase, cGSTM1-1 at 1.94 A resolution.

Glutathione S-transferase cGSTM1-1, an avian class-mu enzyme with high sequence identity with rGSTM3-3, was expressed heterologously in Escherichia coli. The three-dimensional structure of this protein that co-crystallized with an inhibitor, S-hexylglutathione, was determined by the molecular replacement method and refined to 1.94 A resolution. The three-dimensional structure and the folding topology of the dimeric cGSTM1-1 closely resembles those of other class-mu GSTs. The bound inhibitor, S-hexylglutathione, orients in disparate directions in the two subunits. The combined space occupied by the hexyl moiety of the inhibitors overlaps with that reported for rGSTM1-1 co-crystallized with (9 S,10 S)-9-(S-glutathionyl)-10-hydroxy-9,10-dihydrophenanthrene. Conformational differences at a flexible loop (residue 35 to 40) were also observed between the crystal structures of cGSTM1-1 and rGSTM1-1.cGSTM1-1 has the highest epoxidase activity among all the class-mu enzymes reported. Tyr115, has been identified as a residue that participates in the epoxidase activity of class-mu glutathione S-transferase and is conserved in cGSTM1-1. The epoxidase and trans-4-phenyl-3-buten-2-one conjugating activity of cGSTM1-1 are decreased drastically but not abolished by replacing Tyr115 with phenylalanine. The specificity constant of the cGSTM1-1(Y115F) mutant, with 1-chloro-2,4-dinitrobenzene as substrate, is 15-fold higher than that of the wild-type enzyme.

NMR investigation of isotopically labeled cyanide derivatives of lignin peroxidase and manganese peroxidase.

The 1H NMR spectroscopy was used to study lignin peroxidase (LiP) and manganese peroxidase (MnP) containing deuterated histidines. LiP and MnP were obtained from a histidine auxotroph of the fungus Phanerochaete chrysosporium grown in the presence of deuterated histidines. The derivatives with deuterated histidines have allowed a firm assignment of the protons of the distal and proximal histidines. We have also found that the LiP from this strain exhibits different orientations of the 2-vinyl group compared to the LiP from the strain previously studied. Mobility of the group has also been detected, thus explaining the apparent inconsistency between X-ray solid-state and NMR solution data. The 15N shift values of 15N-enriched CN- in the cyanide derivatives of LiP and MnP have also been measured. The shift patterns, both for 15N and for the proximal histidine protons of several peroxidases, are consistent with predominant contact shift contributions which reflect the bond strength of the metal-axial ligand. Finally, our results confirm a correlation between shift values of 15N and those of proximal histidine protons and the Fe3+/Fe2+ redox potentials.

Kinetic analysis of manganese peroxidase. The reaction with manganese complexes.

Manganese peroxidase from the lignin-degrading fungus Phanerochaete chrysosporium catalyzes the H2O2-dependent oxidation of Mn2+ to Mn3+. Presteady-state methods were employed to characterize the reactions of free and chelated Mn2+ with the 2-electron and 1-electron oxidized forms of the enzyme, compounds I and II, respectively. At pH 4.5, the optimum pH for steady-state turnover, the reaction of compound I with Mn2+, either free or complexed, is too rapid to measure by stopped flow methods. The reactions of compound I with Mn2+ can only be monitored under non-optimal conditions of pH 2.5. The reaction of compound II with Mn2+ is much slower than compound I. Chelators such as oxalate, lactate, and malonate facilitated the reaction of Mn2+ with compound II. In contrast, succinate, which does not readily form a complex with Mn2+, and polyglutamate, which is polymeric, were ineffective in stimulating the reaction of Mn2+ with compound II. The 1:1 chelator-Mn2+ complex is the preferred substrate for compound II; this conclusion is based on known formation constants for the various Mn2+ complexes. Steady-state kinetics studies were performed by directly measuring the initial rate of Mn3+ formation. The kcat values for the formation of Mn(3+)-oxalate, Mn(3+)-lactate, and Mn(3+)-malonate are 308, 211, and 220 s-1, respectively. The Km values for Mn(2+)-oxalate, Mn(2+)-lactate, and Mn(2+)-malonate are 13, 41, and 18 microM, respectively. These results collectively indicate that manganese peroxidase does not readily oxidize free (hexa-aquo) Mn2+ as previously proposed (Wariishi, H., Valli, K., and Gold, M. H. (1992) J. Biol. Chem. 267, 23688-23695), but the Mn2+ has to be chelated to support steady-state turnover.

Glyoxylate-supported reactions catalyzed by Mn peroxidase of Phanerochaete chrysosporium: activity in the absence of added hydrogen peroxide.

Mn peroxidases are H2O2-utilizing hemeproteins secreted by the lignin-degrading fungus Phanerochaete chrysosporium. We show here that glyoxylate is capable of supporting Mn peroxidase activity without added H2O2. This glyoxylate-supported activity is dependent upon Mn2+ and dioxygen. The involvement of superoxide is demonstrated by inhibition by superoxide scavenging agents, superoxide dismutase, or tetranitromethane. The addition of catalase resulted in dioxygen evolution, indicating that H2O2 is an intermediate in the reaction. Formate is one of the oxidation products of glyoxylate as detected by the formate dehydrogenase assay. The generation of H2O2 in the presence of Mn2+ and Mn peroxidase results in Mn3+ formation. Consequently, we show that the direct reaction between glyoxylate and Mn3+ also results in formate formation. The stoichiometry of this reaction approaches 1:1. Electron spin resonance, spin-trapping studies show formation of the formate radical CO2-. in the reaction of Mn3+ and glyoxylate.

Stimulation of Mn peroxidase activity: a possible role for oxalate in lignin biodegradation.

Oxalate is produced by numerous wood-degrading fungi. Our studies here show that the white-rot fungus Phanerochaete chrysosporium produces extracellular oxalate under conditions that induce synthesis of the ligninolytic system. Little or no oxalate was detected in cultures grown under high nutrient nitrogen or carbon. This extracellular oxalate was identified and quantitated by HPLC. Its identity was further substantiated by its decomposition by the enzyme oxalate oxidase. The oxalate content of the extracellular fluid (peaking at 60 microM) paralleled the extracellular activity of the lignin-degrading enzyme, Mn peroxidase. Significantly, we demonstrated that oxalate, at physiological concentrations, substantially stimulated Mn peroxidase-catalyzed phenol red oxidation, presumably by its ability to chelate Mn. Stopped flow studies also indicate that oxalate accelerates the turnover of Mn peroxidase. Furthermore, we discovered that oxalate can support Mn peroxidase-catalyzed oxidations in the absence of exogenous H2O2 and in the presence of dioxygen. These results allow us to propose an important role for oxalate, a ubiquitous compound produced by wood-destroying fungi, in lignin biodegradation.

Phosphorylation of lignin peroxidases from Phanerochaete chrysosporium. Identification of mannose 6-phosphate.

Many of the extracellular lignin-degrading peroxidases from the wood-degrading fungus Phanerochaete chrysosporium are phosphorylated. Immunoprecipitation of the extracellular fluid of cultures grown with H2K32PO4 with a polyclonal antibody raised against one of the lignin peroxidase isozymes, H8 (pI 3.5), revealed the incorporation of H2K32PO4 into lignin peroxidases. Analyses of the purified isozymes from labeled cultures by isoelectric focusing showed that, in addition to isozyme H8, lignin peroxidase isozymes H2 (pI 4.4), H6 (pI 3.7), and H10 (pI 3.3) are also phosphorylated. These analyses also showed that lignin peroxidase isozyme H1 (pI 4.7) and manganese-dependent peroxidase isozymes H3 (pI 4.9) and H4 (pI 4.5) are not phosphorylated. Phosphate quantitation indicated the presence of one molecule of phosphate/molecule of enzyme for all of the phosphorylated isozymes. To locate the site of phosphorylation, one-dimensional phosphoamino acid analysis was performed with hydrolyzed 32P-protein. However, phosphotyrosine, phosphoserine, and phosphothreonine could not be identified. Coupled enzyme assays of acid hydrolysate indicated the presence of mannose 6-phosphate as the phosphorylated component on the lignin peroxidase isozymes. Digestion of the isozymes with N-glycanase released the phosphate component, indicating that the mannose 6-phosphate is contained on an asparagine-linked oligosaccharide.

Characterization of two lignin peroxidase clones from Phanerochaete chrysosporium.

Two cDNA clones encoding lignin peroxidase isozymes from Phanerochaete chrysosporium have been isolated and characterized. One of the clones, lambda ML-4, encodes isozyme H8 as does the previously reported clone lambda ML-1 [Tien, M. and Tu, C.-P.D. Nature 326 (1987) 520-523; 328, 742]. Our data are consistent with lambda ML-1 and lambda ML-4 being allelic variants. The other clone, lambda ML-5, encodes a homologous isozyme. We have also isolated the genomic clone corresponding to lambda ML-4 cDNA. Conserved residues thought to be essential for peroxidase function were identified in the predicted amino acid sequences of both cDNA clones. Northern blot analyses indicate that these isozymes are expressed during secondary metabolism, appearing on day 4 of growth and increasing on days 5 and 6.