An ensemble of theta class glutathione transferases with novel catalytic properties generated by stochastic recombination of fragments of two mammalian enzymes

The correlation between sequence diversity and enzymatic function was studied in a library of Theta class glutathione transferases (GSTs) obtained by stochastic recombination of fragments of cDNA encoding human GST T1-1 and rat GST T2-2. In all, 94 randomly picked clones were characterized with respect to sequence, expression level, and catalytic activity in the conjugation reactions between glutathione and six alternative electrophilic substrates. Out of these six different compounds, dichloromethane is a selective substrate for human GST T1-1, whereas 1-menaphthyl sulfate and 1-chloro-2,4-dinitrobenzene are substrates for rat GST T2-2. The other three substances serve as substrates for both enzymes. Through this broad characterization, we have identified enzyme variants that have acquired novel activity profiles that differ substantially from those of the original GSTs. In addition, the expression levels of many clones were improved in comparison to the parental enzyme. A library of mutants can thus display a distribution of properties from which highly divergent evolutionary pathways may emerge, resembling natural evolutionary processes. From the GST library, a clone was identified that, by the point mutation N49D in the rat GST T2-2 sequence, has a 1700% increased activity with 1-menaphthyl sulfate and a 60% decreased activity with 4-nitrophenethyl bromide. Through the N49D mutation, the ratio of these activities has thus been altered 40-fold. An extensive characterization of a population of stochastically mutated enzymes can accordingly be used to find variants with novel substrate-activity profiles and altered catalytic properties. Recursive recombination of selected sequences displaying optimized properties is a strategy for the engineering of proteins for medical and biochemical applications. Such sequential design is combinatorial protein chemistry based on remodeling of existing structural scaffolds and has similarities to evolutionary processes in nature.     

Knowledge-based modeling of a bacterial dichloromethane dehalogenase

A three-dimensional structural model of the dichloromethane dehalogenase (DCMD) from Methylophilus sp. DM11 is constructed based on sequence similarities to the glutathione S-transferases (GSTs). To maximize sequence identity and minimize gaps in the alignment, a hybrid approach is used that takes advantage of the increased homology found between DM11 and domain I of the sheep blowfly θ class GST (residues 1–79) and domain II of the human α class GST (residues 81–222). The resulting structure has Cα root mean square deviations of 1.16 Å in domain I and 1.83 Å in domain II from the template GSTs, which compare well to those seen in other GST interclass comparisons. The model is further applied to explore the structural basis for substrate binding and catalysis. A conserved network of hydrogen bonds is described that binds glutathione to the G site, placing the thiol group in a suitable location for nucleophilic attack of dichloromethane. A mechanism is proposed that involves activation through a hydrogen bond interaction between Ser12 and glutathione, similar to that found in the θ-GSTs. The model also demonstrates how aromatic residues in the hydrophobic site (H site) could play a role in promoting catalysis: His116 and Trp117 are ideally situated to accept a growing negative charge on a chlorine of dichloromethane, stabilizing displacement. This scheme is consistent with experimental results of single-point mutations and comparisons with other GST structures and mechanisms. Proteins 28:217–226, 1997. © 1997 Wiley-Liss Inc.     

Structural basis of the suppressed catalytic activity of wild-type human glutathione transferase T1-1 compared to its W234R mutant

The crystal structures of wild-type human theta class glutathione-S-transferase (GST) T1-1 and its W234R mutant, where Trp234 was replaced by Arg, were solved both in the presence and absence of S-hexyl-glutathione. The W234R mutant was of interest due to its previously observed enhanced catalytic activity compared to the wild-type enzyme. GST T1-1 from rat and mouse naturally contain Arg in position 234, with correspondingly high catalytic efficiency. The overall structure of GST T1-1 is similar to that of GST T2-2, as expected from their 53% sequence identity at the protein level. Wild-type GST T1-1 has the side-chain of Trp234 occupying a significant portion of the active site. This bulky residue prevents efficient binding of both glutathione and hydrophobic substrates through steric hindrance. The wild-type GST T1-1 crystal structure, obtained from co-crystallization experiments with glutathione and its derivatives, showed no electron density for the glutathione ligand. However, the structure of GST T1-1 mutant W234R showed clear electron density for S-hexyl-glutathione after co-crystallization. In contrast to Trp234 in the wild-type structure, the side-chain of Arg234 in the mutant does not occupy any part of the substrate-binding site. Instead, Arg234 is pointing in a different direction and, in addition, interacts with the carboxylate group of glutathione. These findings explain our earlier observation that the W234R mutant has a markedly improved catalytic activity with most substrates tested to date compared to the wild-type enzyme. GST T1-1 catalyzes detoxication reactions as well as reactions that result in toxic products, and our findings therefore suggest that humans have gained an evolutionary advantage by a partially disabled active site.     

Physiological and biochemical analysis of the transformants of aerobic methylobacteria expressing the dcm A gene of dichloromethane dehydrogenase

The transformants of Methylobacterium dichloromethanicum DM4 (DM4-2cr-/pME8220 and DM4-2cr-/pME8221) and of Methylobacterium extorquens AM1 (AM1/pME8220 and AM1/pME8221) that express the dcm A gene of dichloromethane dehalogenase undergo lysis when incubated in the presence of dichloromethane and are sensitive to acidic shock. The lysis of the transformants was found to be related neither to the accumulation of Cl- ions, CH2O, and HCOOH, nor to the impairment of glutathione synthesis or to the maintenance of intracellular pH. The (exo-) Klenow fragment-mediated incorporation of [alpha-32P]dATP into the DNA of the transformants DM4-2cr-/pME8220 and AM1/pME8220 was considerably greater when the transformed cells were incubated with CH2Cl2 than when they were incubated with CH3OH, indicating the occurrence of a significant increase in the total length of gaps. At the same time, the strain AM1 (which lacks dichloromethane dehalogenase) and the dichloromethane-degrading strain DM4 incubated with CH2Cl2 showed an insignificant increase in the total length of the gaps. The transformed cells are likely to lyse due to the relatively inefficient repair of DNA lesions that are induced in response to the alkylating action of S-chloromethylglutathione, an intermediate product of CH2Cl2 degradation. The data obtained suggest that the bacterial mineralization of dichloromethane requires an efficient DNA repair system.     

Single-nucleotide polymorphic variants of human glutathione transferase T1-1 differ in stability and functional properties

We have previously expressed hexa-histidine-tagged human glutathione transferase GST T1-1 at very high levels in an Escherichia colilacZ mutagenicity assay strain. Ethylene dibromide (EDB), which is activated by GST T1-1, produces a potent response in the mutation assay. We have now constructed and expressed two SNP variants of wild-type GST T1-1:D141N and E173K. The EDB activation activities of both variant enzymes, as measured by the lacZ mutagenicity assay, are greatly reduced The D141N variant behaved similarly to the wild-type enzyme, in terms of expression level and specific activities for conjugation of glutathione with 1,2-epoxy-3-(p-nitrophenoxy)propane (EPNP), ethylene diiodide (EDI), and 4-nitrobenzyl chloride (NBCl), and for peroxidative detoxication of cumene hydroperoxide (CuOOH). In contrast, variant E173K is poorly expressed, has no detectable activity with EPNP, NBCl, or CuOOH, and has EDI activity much lower than that of the wild-type enzyme. The circular dichroism (CD) thermal denaturation profiles of the wild-type protein and variant D141N show a sharp two-state transition between native and denatured states. Variant E173K showed a very different profile, consistent with improper or incomplete protein folding. Our results show that SNP variants can give rise to GSTT1-1 proteins with significantly altered properties.     

Residue 234 is a master switch of the alternative-substrate activity profile of human and rodent theta class glutathione transferase T1-1

Background

The Theta class glutathione transferase GST T1-1 is a ubiquitously occurring detoxication enzyme. The rat and mouse enzymes have high catalytic activities with numerous electrophilic compounds, but the homologous human GST T1-1 has comparatively low activity with the same substrates. A major structural determinant of substrate recognition is the H-site, which binds the electrophile in proximity to the nucleophilic sulfur of the second substrate glutathione. The H-site is formed by several segments of amino acid residues located in separate regions of the primary structure. The C-terminal helix of the protein serves as a lid over the active site, and contributes several residues to the H-site.

Methods

Site-directed mutagenesis of the H-site in GST T1-1 was used to create the mouse Arg234Trp for comparison with the human Trp234Arg mutant and the wild-type rat, mouse, and human enzymes. The kinetic properties were investigated with an array of alternative electrophilic substrates to establish substrate selectivity profiles for the different GST T1-1 variants.

Results

The characteristic activity profile of the rat and mouse enzymes is dependent on Arg in position 234, whereas the human enzyme features Trp. Reciprocal mutations of residue 234 between the rodent and human enzymes transform the substrate-selectivity profiles from one to the other.

Conclusions

H-site residue 234 has a key role in governing the activity and substrate selectivity profile of GST T1-1.

General significance

The functional divergence between human and rodent Theta class GST demonstrates that a single point mutation can enable or suppress enzyme activities with different substrates.     

Dichloromethane dehalogenase of Hyphomicrobium sp. strain DM2.

Dichloromethane dehalogenase, a highly inducible glutathione-dependent enzyme catalyzing the conversion of dichloromethane into formaldehyde and inorganic chloride, was purified fivefold with 60% yield from Hyphomicrobium sp. strain DM2. The electrophoretically homogeneous purified enzyme exhibited a specific activity of 17.3 mkat/kg of protein. Its pH optimum was 8.5. The enzyme was stable at -20 degrees C for at least 6 months. A subunit molecular weight of 33,000 was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Gel filtration of native dichloromethane dehalogenase yielded a molecular weight of 195,000. Subunit cross-linking with dimethyl suberimidate confirmed the hexameric tertiary structure of the enzyme. Dichloromethane dehalogenase was highly specific for dihalomethanes. Its apparent Km values were 30 microM for CH2Cl2, 15 microM for CH2BrCl, 13 microM for CH2Br2, 5 microM for CH2I2, and 320 microM for glutathione. Several chlorinated aliphatic compounds inhibited the dichloromethane dehalogenase activity of the pure enzyme. The Ki values of the competitive inhibitors 1,2-dichloroethane and 1-chloropropane were 3 and 56 microM, respectively.     

Dichloromethane dehalogenase with improved catalytic activity isolated from a fast-growing dichloromethane-utilizing bacterium.

A methylotrophic bacterium, denoted strain DM11, was isolated from groundwater and shown to utilize dichloromethane or dibromomethane as the sole carbon and energy source. The new isolate grew at the high rate of 0.22 h-1 compared with 11 previously characterized dichloromethane-utilizing bacteria (micromax, 0.08 h-1). The dichloromethane dehalogenase from strain DM11 (group B enzyme) was purified by anion-exchange chromatography. It was shown to be substantially different from the set of dichloromethane dehalogenases from the 11 slow-growing strains (group A enzymes) that had previously been demonstrated to be identical. The Vmax for the group B enzyme was 97 mkat/kg of protein, some 5.6-fold higher than that of the group A enzymes. The group A dehalogenases showed hyperbolic saturation with the cosubstrate glutathione, whereas the group B enzyme showed positive cooperativity in glutathione binding. Only 1 of 15 amino acids occupied common positions at the N termini, and amino acid contents were substantially different in group A and group B dehalogenases. Immunological assays demonstrated weak cross-reactivity between the two enzymes. Despite the observed structural and kinetic differences, there is potentially evolutionary relatedness between group A and group B enzymes, as indicated by (i) hybridization of DM11 DNA with a gene probe of the group A enzyme, (ii) a common requirement for glutathione in catalysis, and (iii) similar subunit molecular weights of about 34,000.     

Dehalogenation of dichloromethane by dichloromethane dehalogenase/glutathione S-transferase leads to formation of DNA adducts

Formation of DNA adducts following conversion of dichloromethane by bacterial dichloromethane dehalogenase/glutathione S-transferase was demonstrated. Adducts included dichloromethane carbon and glutathione sulfur atoms. A reaction with DNA occurred preferentially at guanine bases. Increased DNA degradation in a polA mutant of Methylobacterium dichloromethanicum DM4 grown with dichloromethane confirmed the genotoxicity associated with dichloromethane degradation, suggesting an important role of DNA repair in the metabolism of halogenated, DNA-alkylating compounds by bacteria.     

Effects of Bacterial Host and Dichloromethane Dehalogenase on the Competitiveness of Methylotrophic Bacteria Growing with Dichloromethane

Methylobacterium sp. strain DM4 and Methylophilus sp. strain DM11 can grow with dichloromethane (DCM) as the sole source of carbon and energy by virtue of homologous glutathione-dependent DCM dehalogenases with markedly different kinetic properties (the k_cat values of the enzymes of these strains are 0.6 and 3.3 s⁻¹, respectively, and the K_m values are 9 and 59 μM, respectively). These strains, as well as transconjugant bacteria expressing the DCM dehalogenase gene (dcmA) from DM11 or DM4 on a broad-host-range plasmid in the background of dcmA mutant DM4-2cr, were investigated by growing them under growth-limiting conditions and in the presence of an excess of DCM. The maximal growth rates and maximal levels of dehalogenase for chemostat-adapted bacteria were higher than the maximal growth rates and maximal levels of dehalogenase for batch-grown bacteria. The substrate saturation constant of strain DM4 was much lower than the K_m of its associated dehalogenase, suggesting that this strain is adapted to scavenge low concentrations of DCM. Strains and transconjugants expressing the DCM dehalogenase from strain DM11, on the other hand, had higher growth rates than bacteria expressing the homologous dehalogenase from strain DM4. Competition experiments performed with pairs of DCM-degrading strains revealed that a strain expressing the dehalogenase from DM4 had a selective advantage in continuous culture under substrate-limiting conditions, while strains expressing the DM11 dehalogenase were superior in batch culture when there was an excess of substrate. Only DCM-degrading bacteria with a dcmA gene similar to that from strain DM4, however, were obtained in batch enrichment cultures prepared with activated sludge from sewage treatment plants.     

Characterization of porcine alpha-class glutathione transferase A1-1

An Alpha-class glutathione transferase (GST) has been cloned from pig gonads. In addition to two conservative point mutations our nucleotide sequence presents a frame shift resulting from a missing A as compared to a previously published porcine GST A1-1 sequence. The deduced C-terminal amino-acid segment of the protein differs between the two variants. Repeated sequencing of cDNA isolated from different tissues and animals ruled out the possibility of a cloning artifact, and the deduced amino acid sequence of our clone showed higher similarity to related mammalian GST sequences. Hereafter, we refer to our cloned enzyme as GST A1-1 and to the previously published enzyme as GST A1-1^∗. The study of the tissue distribution of the GSTA1 mRNA revealed high expression levels in many organs, in particular adipose tissue, liver, and pituitary gland. Porcine GST A1-1 was expressed in Escherichia coli and its kinetic properties were determined using alternative substrates. The catalytic activity in steroid isomerization reactions was at least 10-fold lower than the corresponding values for porcine GST A2-2, whereas the activity with 1-chloro-2,4-dinitrobenzene was approximately 8-fold higher. Differences in the H-site residues of mammalian Alpha-class GSTs may explain the catalytic divergence.     

The evolution of catalytic efficiency and substrate promiscuity in human theta class 1-1 glutathione transferase

Theta class glutathione transferases (GST) from various species exhibit markedly different catalytic activities in conjugating the tripeptide glutathione (GSH) to a variety of electrophilic substrates. For example, the human theta 1-1 enzyme (hGSTT1-1) is 440-fold less efficient than the rat theta 2-2 enzyme (rGSTT2-2) with the fluorogenic substrate 7-amino-4-chloromethyl coumarin (CMAC). Large libraries of hGSTT1-1 constructed by error-prone PCR, DNA shuffling, or saturation mutagenesis were screened for improved catalytic activity towards CMAC in a quantitative fashion using flow cytometry. An iterative directed evolution approach employing random mutagenesis in conjunction with homologous recombination gave rise to enzymes exhibiting up to a 20,000-fold increase in k(cat)/K(M) compared to hGSTT1-1. All highly active clones encoded one or more mutations at residues 32, 176, or 234. Combinatorial saturation mutagenesis was used to evaluate the full complement of natural amino acids at these positions, and resulted in the isolation of enzymes with catalytic rates comparable to those exhibited by the fastest mutants obtained via directed evolution. The substrate selectivities of enzymes resulting from random mutagenesis, DNA shuffling, and combinatorial saturation mutagenesis were evaluated using a series of distinct electrophiles. The results revealed that promiscuous substrate activities arose in a stochastic manner, as they did not correlate with catalytic efficiency towards the CMAC selection substrate. In contrast, chimeric enzymes previously constructed by homology-independent recombination of hGSTT-1 and rGSTT2-2 exhibited very different substrate promiscuity profiles, and showed a more defined relationship between evolved and promiscuous activities.     

Cloning and characterization of dichloromethane dehalogenase from Methylobacterium rhodesianum for dichloromethane degradation

The gene encoding dichloromethane dehalogenase from Methylobacterium rhodesianum was cloned. Bioinformatic analysis showed that dichloromethane dehalogenase gene sequence from M. rhodesianum is almost identical to the one from Methylobacterium extorquens, with only one base difference. Dichloromethane dehalogenase was subsequently expressed in Escherichia coli BL21 (DE3) and purified. It was found that enzyme activity in recombinant cells was 3 times higher than that in the wild-type M. rhodesianum. Further investigation showed that recombinant dichloromethane dehalogenase was most active at 40°C at pH 7–8, and its K_M was 10.96 mM when treated with dichloromethane as substrate. The fitted curve of dichloromethane degradation gave a V_max of 0.43 mM/h of in 0.01 M phosphate buffer. Degradation efficiency of dichloromethane reached 86.11% within 20 h. In addition, it was found that degradation efficiency of dichloromethane was highly associated with glutathione concentration, supporting the reports that glutathione functions as coenzyme of dichloromethane dehalogenase for dichloromethane degradation.     

Physiological and Biochemical Analysis of the Transformants of Aerobic Methylobacteria Expressing the dcmA Gene of Dichloromethane Dehalogenase

Transformants of Methylobacterium dichloromethanicum DM4 (DM4-2cr^–/pME 8220 and DM4-2cr^–/pME8221) and of Methylobacterium extorquens AM1 (AM1/pME8220 and AM1/pME8221) that express the dcm A gene of dichloromethane dehalogenase undergo lysis when incubated in the presence of dichloromethane and are sensitive to acidic shock. The lysis of the transformants was found to be related neither to the accumulation of Cl^– ions, CH₂O, or HCOOH, nor to the impairment of glutathione synthesis or to the disturbance of intracellular pH homeostasis. The (exo^–) Klenow fragment–mediated incorporation of [-³²P]dATP into the DNA of the transformants DM4-2cr^–/pME8220 and AM1/pME8220 was considerably greater when the transformed cells were incubated with CH₂Cl₂ than when they were incubated with CH₃OH, indicating the occurrence of a significant increase in the total length of gaps. At the same time, the strain AM1 (which lacks dichloromethane dehalogenase) and the dichloromethane-degrading strain DM4 incubated with CH₂Cl₂ showed an insignificant increase in the total length of the gaps. The transformed cells are likely to lyse due to the relatively inefficient repair of DNA lesions that are induced in response to the alkylating action of S-chloromethylglutathione, an intermediate product of CH₂Cl₂ degradation. The data obtained suggest that the bacterial mineralization of dichloromethane requires an efficient DNA repair system.     

Characterization of a Flavobacterium glutathione S-transferase gene involved reductive dechlorination.

The gene pcpC, encoding tetrachloro-p-hydroquinone (TeCH) reductive dehalogenase, was cloned from Flavobacterium sp. strain ATCC 39723 and sequenced. The gene was identified by hybridization with a degenerate oligonucleotide designed from the N-terminal sequence of the purified protein. An open reading frame of 747 nucleotides was found, which predicts a translational product of 248 amino acids having a molecular weight of 28,263, which agrees favorably with the sodium dodecyl sulfate-polyacrylamide gel electrophoresis-determined molecular weight of 30,000 reported for the purified protein. The predicted translational product of pcpC matched the N-terminal sequence of the purified protein exactly. From the nucleotide sequence, the protein appears to have a processed formylmethionyl. An Escherichia coli pcpC overexpression clone was shown to produce dichlorohydroquinone and trichlorohydroquinone from TeCH. Protein data base searches grouped the predicted translational sequence of pcpC with two previously reported plant glutathione S-transferases but less significantly with any of the mammalian glutathione S-transferases or the glutathione-utilizing, hydrolytic dechlorinating enzyme from Methylobacterium sp. strain DM4.     

Emergence of a novel highly specific and catalytically efficient enzyme from a naturally promiscuous glutathione transferase

Redesign of glutathione transferases (GSTs) has led to enzymes with remarkably enhanced catalytic properties. Exchange of substrate-binding residues in GST A1-1 created a GST A4-4 mimic, called GIMFhelix, with >300-fold improved activity with nonenal and suppressed activity with other substrates. In the present investigation GIMFhelix was compared with the naturally-evolved GSTs A1-1 and A4-4 by determining catalytic efficiencies with nine alternative substrates. The enzymes can be represented by vectors in multidimensional substrate-activity space, and the vectors of GIMFhelix and GST A1-1, expressed in kcat/Km values for the alternative substrates, are essentially orthogonal. By contrast, the vectors of GIMFhelix and GST A4-4 have approximately similar lengths and directions. The broad substrate acceptance of GST A1-1 contrasts with the high selectivity of GST A4-4 and GIMFhelix for alkenal substrates. Multivariate analysis demonstrated that among the diverse substrates used, nonenal, cumene hydroperoxide, and androstenedione are major determinants in the portrayal of the three enzyme variants. These GST substrates represent diverse chemistries of naturally occurring substrates undergoing Michael addition, hydroperoxide reduction, and steroid double-bond isomerization, respectively. In terms of function, GIMFhelix is a novel enzyme compared to its progenitor GST A1-1 in spite of 94% amino-acid sequence identity between the enzymes. The redesign of GST A1-1 into GIMFhelix therefore serves as an illustration of divergent evolution leading to novel enzymes by minor structural modifications in the active site. Notwithstanding low sequence identity (60%), GIMFhelix is functionally an isoenzyme of GST A4-4.