Methanopyrus kandleri Glutamyl-tRNA Reductase*

The initial reaction of tetrapyrrole formation in archaea is catalyzed by a NADPH-dependent glutamyl-tRNA reductase (GluTR). The hemA gene encoding GluTR was cloned from the extremely thermophilic archaeon Methanopyrus kandleri and overexpressed in Escherichia coli. Purified recombinant GluTR is a tetrameric enzyme with a nativeM r = 190,000 ± 10,000. Using a newly established enzyme assay, a specific activity of 0.75 nmol h−1 mg−1 at 56 °C with E. coliglutamyl-tRNA as substrate was measured. A temperature optimum of 90 °C and a pH optimum of 8.1 were determined. Neither heme cofactor, nor flavin, nor metal ions were required for GluTR catalysis. Heavy metal compounds, Zn2+, and heme inhibited the enzyme. GluTR inhibition by the newly synthesized inhibitor glutamycin, whose structure is similar to the 3′ end of the glutamyl-tRNA substrate, revealed the importance of an intact chemical bond between glutamate and tRNAGlu for substrate recognition. The absolute requirement for NADPH in the reaction of GluTR was demonstrated using four NADPH analogues. Chemical modification and site-directed mutagenesis studies indicated that a single cysteinyl residue and a single histidinyl residue were important for catalysis. It was concluded that during GluTR catalysis the highly reactive sulfhydryl group of Cys-48 acts as a nucleophile attacking the α-carbonyl group of tRNA-bound glutamate with the formation of an enzyme-localized thioester intermediate and the concomitant release of tRNAGlu. In the presence of NADPH, direct hydride transfer to enzyme-bound glutamate, possibly facilitated by His-84, leads to glutamate-1-semialdehyde formation. In the absence of NADPH, a newly discovered esterase activity of GluTR hydrolyzes the highly reactive thioester of tRNAGlu to release glutamate.

Plants, archaea, and the majority of bacteria form ALA in a two-step reaction from the C 5 -skeleton of glutamate bound to tRNA Glu (glutamyl-tRNA) (2). The initial enzyme of the pathway GluTR, encoded by hemA, reduces the activated ␣-carboxyl group of glutamate using NADPH to form GSA. GSA is then converted to ALA by GSA-AT (2). Investigation of GluTR catalysis has been hampered by its low cellular concentration, difficulties with the overexpression of the respective gene from various species in Escherichia coli, and the tendency of the enzyme to aggregate and precipitate at high protein concentration (3,4). Earlier characterizations with purified GluTR from barley, from the green alga Chlamydomonas reinhardtii, and from the bacteria Bacillus subtilis, Synechocystis sp. PCC 6803, and E. coli resulted in the description of highly variable molecular masses, specific activities, and catalytic properties (4 -8).
Recently, characterization of barley GluTR expressed as a fusion protein with glutathione S-transferase indicated the participation of a heme cofactor in the catalysis of the pentameric enzyme (9,10). Nevertheless, central questions concerning the exact enzymatic mechanism of GluTR and its structural basis remain to be answered. Here, we describe the utilization of recombinant GluTR from the extreme thermophilic archaeon Methanopyrus kandleri for the determination of the native molecular mass and the elucidation of essential structural features of the substrate and the cofactor NADPH for enzyme recognition. To achieve this, a new test system was established and a new GluTR-specific inhibitor was synthesized. Protein modification, site-directed mutagenesis, and enzymatic activity analysis were used to gain first insights into the molecular basis of GluTR catalysis. Evidence for an enzyme-localized thioester intermediate was obtained, allowing us to propose a mechanism for GluTR catalysis.

Cloning of the M. kandleri hemA Gene
The ALA-auxotrophic E. coli hemA strain GE1387 was transformed with a M. kandleri genomic library prepared in the vector pBluescript SKϩ, and complementing clones were identified as described before for the cloning of Methanobacterium thermoautotrophicum hemA (11).

Construction of the M. kandleri hemA Expression
Vector pMkhemA A 1228-base pair fragment encoding the 404 amino acid residues of M. kandleri GluTR was amplified by polymerase chain reaction using the primer IM1 (5Ј-GGAGGCATATGGAGGACCTGGTGTGC-3Ј) and the primer IM2 (5Ј-CCGAATTCCTAGCCGTTAAGCTCACTCGCC-3Ј). The resulting polymerase chain reaction fragment was digested with NdeI and EcoRI (underlined in the primer sequences) and ligated into the appropriately digested vector pET5a (Stratagene, Heidelberg, Germany) to generate pMkhemA.

Site-directed Mutagenesis of M. kandleri hemA
To exchange amino acid residues of M. kandleri GluTR, the Quickchange TM kit (Stratagene, Heidelberg, Germany) was used according to the manufacturer's instructions. The following oligonucleotides, with newly introduced codons underlined, were employed to generate the mutants indicated: for C6S GGAGGACCTGGTGAGCGTCGGTATCAC-CCACAAGG, for C42S CCTTCGGACTCTCCGGCAGCGTCCTCCT-TCAGACATGC, for C48S CCTCCTTCAGACAAGCAACCGCGTCGAG-GTGTACGCC, for C90S CCTATTCCGTGTCGCTAGCGGGCTGGAGT-CGATGATGG, for C393S CCGTAGAGCGGCGAGCAGGGCCCTAAGA  CGG, for H84A GGAAGCTGTGAGGGCCCTATTCCGTGTCGCTTGT-GGGC, and for H84N GGAAGCTGTGAGGAACCTATTCCGTGTCGC-TTGTGGGC. Mutant M. kandleri hemA genes were expressed, and recombinant proteins were purified to apparent homogeneity as described below.
Overexpression of M. kandleri hemA and Purification of GluTR BL21(DE3) carrying pMkhemA was cultivated in LB medium containing 100 g/ml ampicillin at 37°C to an A 578 of 0.7. After the addition of IPTG to a final concentration of 1 mM, the cultures were further incubated until an A 578 of 2 was attained and subsequently the cells were harvested by centrifugation. The bacterial cell pellet was re-suspended in 10 ml of 50 mM Na-HEPES, pH 8.1, including 1 mM ␤-mercaptoethanol and disrupted by sonication. Cell debris was removed by centrifugation for 60 min at 23,000 ϫ g at 4°C. The protein solution (10 mg/ml) was loaded with a flow rate of 1.5 ml/min onto a 50-ml Red Sepharose CL-6B column (diameter 3 cm), which had previously been equilibrated with 20 mM Na-HEPES, pH 8.1, including 1 mM ␤-mercaptoethanol (buffer A). The column was washed with 100 ml of the same buffer to remove unbound proteins, and bound proteins were eluted using a 500-ml linear gradient of 0 -1 M NaCl in buffer A. Fractions containing GluTR were pooled, dialyzed against buffer A, and loaded at a concentration of 5 mg of protein per ml of column volume onto a MonoQ HR 10/10 column previously equilibrated with buffer A. Bound proteins were eluted using a linear 100-ml gradient of 0 -1 M NaCl in buffer A. Fractions containing GluTR were pooled and concentrated by ultrafiltration (Centriprep 30; Amicon, Witten, Germany). The concentrated solution (4 ml; 7 mg/ml) was applied to a HiLoad Superdex 200 HR 26/60 prep grade gel permeation chromatography column pre-equilibrated with 150 mM NaCl in buffer A at a flow rate of 1.5 ml/min. Fractions containing GluTR were pooled, dialyzed against buffer A, and concentrated to 7 mg/ml by ultrafiltration as described above.

Determination of the Native Molecular Mass Using Gel Permeation Chromatography and Glycerol Gradient Centrifugation
The native molecular mass of recombinant GluTR in the absence of tRNA substrate was determined as described previously (12).
The GluTR assay mix (200 l) contained 1 M purified recombinant M. kandleri GluTR, 4 M E. coli [ 14 C]Glu-tRNA Glu in 20 mM Na-HEPES, pH 8.1, 10 M dithiothreitol, and 2 mM NADPH unless stated otherwise. Reactions were usually incubated at 56°C for the indicated time periods and stopped by pipetting aliquots of 30 l from the assay mixture onto Whatman no. 3MM filters. The tRNA was precipitated, washed, and quantified as described previously (15). Reactions without GluTR for each employed condition served as background controls for spontaneous substrate hydrolysis. The logarithm of measured substrate depletion was plotted against time, and the GluTR specific activity was deduced from this first-order equation by a method described previously (16). This procedure was necessary since the limiting amounts of pure substrate did not allow the excess of substrate required for classical Michaelis-Menten kinetics.

High Performance Liquid Chromatography (HPLC) Analysis of GluTR Product Formation
Standard assay mixtures were performed as outlined above with an incubation at 56°C for 7 min. In most cases the reaction product GSA was further identified by its specific conversion into ALA using purified E. coli GSA-AT (17). Reaction products were analyzed on a Waters Bondapack TM C 18 reversed phase column (3.9 ϫ 150 mm, 125 Å pore size, 10 m particle diameter) as described before (18). The radioactive GluTR substrate [ 14 C]Glu-tRNA Glu still contained residual amounts of free [ 14 C]Glu. Under the employed separation conditions, the commercially available [ 14 C]Glu (ICN Pharmaceuticals, Irvine, CA) appeared as a double peak at 3 and 4.5 min retention time, with a small third peak at 8 min retention time (Fig. 3).

RESULTS AND DISCUSSION
Isolation of the M. kandleri hemA Gene by Complementation of an E. coli ALA Auxotrophic hemA Mutant-Functional complementation of the ALA-auxotrophic E. coli hemA strain GE1387 with an M. kandleri genomic library to ALA prototrophy yielded one slow-growing colony. Restriction enzyme analysis of the complementing plasmid revealed the presence of an insert of approximately 3500 base pairs, which was subjected to complete DNA sequence analysis. A 1215-base pair open reading frame encoding a 404-amino acid protein with a calculated molecular mass of 45,448 Da was detected. The deduced protein showed a high degree of amino acid sequence identity to archaeal, bacterial, and plant GluTRs (data not shown).
Expression of M. kandleri hemA and Purification of Recombinant GluTR-The M. kandleri hemA gene was expressed using a T7 RNA polymerase-driven expression system. Recombinant GluTR was purified to apparent homogeneity as described under "Experimental Procedures". Fig. 1 shows an SDS-polyacrylamide gel electrophoresis analysis of the various stages of the expression and purification procedure. In the final concentrated Superdex 200 fraction (Fig. 1, lane 5), a single protein after Coomassie Blue staining was visible. The apparent homogeneity of M. kandleri GluTR was confirmed by electrospray ionization mass spectrometry and isoelectric focusing. One liter of bacterial culture yielded approximately 5 mg of apparently homogeneous GluTR. The calculated molecular mass for one GluTR monomer deduced from the gene sequence (45,448 Da) was experimentally confirmed by electrospray ionization mass spectrometry to be 45,436 Ϯ 30 Da (data not shown). The N-terminal sequence of the purified enzyme determined by Edman degradation (MEDLVXVGITHKEAE-VEELEKARFESDEAVXDIVESFGLSG) was found to be identical to the amino acid sequence deduced from the cloned hemA gene sequence.
A Novel Substrate Depletion Assay for Testing GluTR Activity-Purified recombinant GluTR was used to establish a new GluTR assay based on substrate utilization measurements. The assay originates from the well established and frequently used aminoacyl-tRNA synthetase assay, in which the product aminoacyl-tRNA is recovered after the reaction by acid precipitation (15). However, due to the principle of substrate detection, real product formation was always verified using HPLC analysis. This was necessary when significant changes in supplied cofactors, inhibitors, or reaction conditions were introduced to the assay. Under the standard assay conditions, outlined under "Experimental Procedures", M. kandleri GluTR showed a specific activity of 0.75 nmol h Ϫ1 mg Ϫ1 of protein. This specific activity was usually set to 100% activity, and other values obtained were related to that. A reaction temperature of 56°C instead of the measured temperature optimum of 90°C (see below) was used to stabilize the employed substrate glutamyl-tRNA. The heterologous substrate E. coli glutamyl-tRNA was routinely used, since the supply of the extremely thermophilic archaeon M. kandleri for tRNA preparation was limited. The specific activity measured with homologous substrate M. kandleri glutamyl-tRNA at 90°C was only approximately 3-4 times higher compared with the routinely used conditions with E. coli glutamyl-tRNA at 56°C (data not shown).
As shown in Fig. 2, recombinant M. kandleri GluTR efficiently utilized the E. coli glutamyl-tRNA substrate. Under the assay conditions at 56°C, only slow substrate hydrolysis (approximately 10% after 5 min) was observed. The reduction of glutamyl-tRNA to the product GSA was visualized using HPLC analysis (Fig. 2, panel B). Addition of purified E. coli GSA-AT to the reaction mixture resulted in the almost complete conversion of GSA into ALA (Fig. 2, panel B).
Spectroscopic Features of M. kandleri GluTR-Purified recombinant GluTR was analyzed at high concentrations of active enzyme (up to 50 mg/ml) using absorption and fluorescence spectroscopy. No typical spectra for hemes, flavins, NADPH, NADP ϩ , iron-sulfur clusters, or pyridoxal 5Ј-phosphate/pyridoxamine 5Ј-phosphate were observed. No stimulation of GluTR activity by the addition of FAD, FMN, pyridoxal 5Јphosphate, pyridoxamine 5Ј-phosphate, or heme to the assay was observed (Table I). From these results we conclude that M. kandleri GluTR does not possess a chromophoric prosthetic group. This is a striking difference to barley GluTR, which was shown to carry a bound heme cofactor (9). This difference is not completely unexpected since several methanogenic archaea including the phylogenetic group of M. kandleri do not synthesize heme and consequently lack the enzymes usually converting uroporphyrinogen III into protoheme.
Native Molecular Mass of M. kandleri GluTR-Since purification and characterization of GluTR from various sources yielded highly diverse native molecular masses, three independent methods were used to determine the native molecular mass of M. kandleri GluTR. Gel permeation chromatography, glycerol gradient centrifugation, and dynamic light scattering revealed a relative molecular mass for native GluTR of 190,000 Ϯ 10,000 with a Stokes radius of 48 Å. Based on these results and the calculated subunit molecular mass of 45,448 Da, we conclude that M. kandleri GluTR is a tetrameric protein.
Influence of Temperature, pH, and Ionic Strength on M. kandleri GluTR Activity and Stability-A pH optimum of 8.1 and an optimal temperature of 90°C were determined for M. kandleri GluTR (Table II). No significant stimulation of M. kandleri GluTR activity by high salt concentrations was observed (Table II). However, the enzyme clearly tolerated the high ionic strength conditions employed. GluTR stability at high temperatures was found to be salt-independent (Table II). Isoelectric focusing of M. kandleri GluTR gave a single major band corresponding to a pI of 5.6.
Metal Ions Are Not Required for M. kandleri GluTR Activity-Treatment of M. kandleri GluTR with EDTA, EGTA, 1,10phenanthroline, and 2,2Ј-dipyridyl did not affect enzyme activity (Table I). Atomic absorption spectroscopic analysis of GluTR protein for Mg 2ϩ , Zn 2ϩ , Fe 2ϩ /Fe 3ϩ , and Mn 2ϩ revealed the absence of these metal ions from highly active enzyme preparations. Treatment of M. kandleri GluTR with Mg 2ϩ , Ca 2ϩ , Co 2ϩ , and Ni 2ϩ did not change enzymatic activity (Table I). From these observations, we conclude that M. kandleri GluTR does not contain detectable amounts of these metals and performs a metal-independent catalysis. Heavy metal compounds like PbCl 2 , PtCl 4 , and KPdCl 4 as well as Zn 2ϩ (250 M) inhibited the enzyme ( Table I). The functional implications of these observations are discussed below.
Structural Features of the Substrate Glutamyl-tRNA for GluTR Recognition-To determine the structural features of glutamyl-tRNA required for enzyme recognition, a variety of glutamate analogues, tRNAs, the reaction product GSA, and a newly synthesized inhibitor mimicking the intact 3Ј end of the substrate were tested in GluTR inhibition experiments. Similar to results obtained from comparable investigations with the barley enzyme (8), GSA inhibited enzymatic glutamyl-tRNA reduction to 50% at a concentration of 1 mM (Table I). However, none of the tested glutamate analogues (L-and D-glutamate, 4,5-dioxovaleric acid, 4,5-diaminovaleric acid, glutaric acid) and tRNA Glu preparations from E. coli and M. kandleri inhibited enzymatic activity when present in 200 -2000-fold molar excess over the substrate (Table I). These results suggested that an intact chemical bond between glutamate and the tRNA-Glu part of the substrate is required for efficient substrate binding.
Due to the labile nature of the ester bond between the 3Ј end of tRNA Glu and glutamate, a substrate analogue representing the last adenosine residue of the tRNA coupled via a stable amide bond to glutamate was synthesized. Although this potential inhibitor was originally suggested by Dr. Kannangara (19), no data concerning inhibitor synthesis or its quantitative analysis were available. We synthesized the inhibitor from the structurally related puromycin aminonucleoside and named it glutamycin (Fig. 3). As shown in Table I, glutamycin was successfully used for specific inhibition of M. kandleri GluTR. These findings for the first inhibitory substrate analogue of GluTR confirm the proposal that the chemical bond between the glutamate and tRNA Glu portions of the substrate is essential for substrate recognition.
Heme Inhibition of GluTR-Inhibition of GluTR by the end product of the pathway heme has been described for GluTRs from various species (7,8). M. kandleri GluTR was also inhibited by heme in the micromolar range (Table I). Any biological significance attributed to this observation must be interpreted with caution since heme in the higher micromolar range nonspecifically inhibits various other enzymes including restriction endonucleases and DNA polymerases (20).
Structural Features of the NADPH Cofactor for GluTR Utilization-The utilization of NADH instead of NADPH by GluTR from Synechocystis, has been reported (7); however, it should be noted that NADH functioned with low efficiency. We tested several commercially available NADPH analogues for their reduction and inhibition capacities (Table II). NADH neither sustained nor inhibited M. kandleri GluTR-catalyzed reduction of glutamyl-tRNA. Removal of the amino group localized on the adenine ring (nicotinamide hypoxanthine  a Indicated substances were pH adjusted and pre-incubated with purified M. kandleri GluTR under standard assay conditions as outlined under "Experimental Procedures" for 20 min before reactions were started with the addition of the substrate, radioactive E. coli glutamyl-tRNA. b GluTR activity was measured using the substrate depletion assay as outlined under "Experimental Procedures." Product formation was verified using HPLC analysis. GluTR specific activity obtained without additions was set to 100%, and all other measured GluTR specific activities were related to that. c Purified M. kandleri GluTR was pre-treated for 45 min at room temperature with indicated amounts of modifying reagent or potential inhibitor. Residual reagent was removed from the assay by centrifugation through a G50 spin column under test conditions before activity determination. d ND, not detectable. dinucleotide phosphate, reduced form) converted the cofactor into a potential inhibitor with 50% inhibition at a concentration of 1 mM. Removal of the adenosine phosphate part (␤-nicotinamide mononucleotide, reduced form) completely abolished cofactor utilization by the enzyme. Interestingly, flexibility for the localization of the phosphoryl group of the adenosine ribose between the 2Ј (NADPH) and 3Ј (3Ј NADPH) position was observed. However, as mentioned above, experiments using NADH demonstrated the absolute requirement for the presence of the phosphoryl group. These experiments showed an overall requirement for all the major determinants of NADPH for efficient recognition and utilization by GluTR. Tight NADPH coordination by the enzyme was concluded.
GluTR Modification Using Iodoacetamide, TPCK, and 5,5Ј-Dithiobis(2-nitrobenzoic Acid) Inhibits Enzyme Activity-An initial proposal suggested the back reaction catalyzed by GAPDH as a mechanistic model for GluTR activity (2). GAPDH catalysis involves an active site cysteinyl residue, which forms a thioester with the substrate after NAD ϩ -dependent oxidation (21). To analyze M. kandleri GluTR catalysis for a potential participation of active site-localized nucleophilic amino acid residues, chemical protein modification experiments were performed. As shown in Table I, treatment of M. kandleri GluTR with iodoacetamide, TPCK, and 5,5Ј-dithiobis(2-nitrobenzoic acid) totally abolished enzyme activity. In agreement with the observed heavy metal sensitivity, these results suggested the b For the pH titration, the following buffer systems were used: 50 mM phosphate, 50 mM Na-BICINE, and 50 mM Na-HEPES (see footnotes e-g). c All assays were performed in 5 times the usual assay volume for periods of 30 s to 2 min. GluTR activity was measured using the substrate depletion assay as outlined under "Experimental Procedures." Product formation was verified using HPLC analysis. GluTR specific activity obtained without additions was set to 100%, and all other measured GluTR specific activities were related to that. Background controls without enzyme addition measuring substrate hydrolysis were performed for each employed condition and subtracted.
d During incubation at 85-95°C, assays were overlaid with mineral oil to prevent evaporation. e 50 mM phosphate. f 50 mM Na-BICINE. g 50 mM Na-HEPES. h ND, not detectable. presence of one or more nucleophilic cysteinyl residues involved in catalysis. Additionally, potential histidinyl residue involvement in enzyme activity was suggested.
GluTR Possesses Glutamyl-tRNA Esterase Activity-The enzyme efficiently utilized glutamyl-tRNA in the absence of NADPH. However, HPLC analysis of the resulting reaction products revealed the liberation of glutamate from the tRNA (Fig. 2). The catalytic rate of GluTR-dependent glutamyl-tRNA esterase activity observed in the absence of NADPH was comparable to the catalytic rate of the GluTR reductase activity in the presence of NADPH. Esterase activity is a reaction typical of other enzymes forming a covalent acyl-enzyme intermediate involving an active site cysteinyl residue like GAPDH, thiol proteinases, and aldehyde dehydrogenases (22)(23)(24).
Mutation C48S Totally Abolishes GluTR Reductase and Esterase Activity-To identify definitely the cysteinyl residues important in GluTR catalysis, all 5 cysteinyl residues of the enzyme located at positions 6, 42, 48, 90, and 393 were individually changed to serine residues. Only GluTR mutant C48S had completely lost its reductase and esterase activities (Table  III). In agreement with this result, Cys-48 was identified as the only cysteinyl residue conserved in all known GluTR enzymes. All other mutants retained their full catalytic activity (Table  III). These observations provided evidence for a role of Cys-48 as the active site nucleophile.
The Role of His-84 in GluTR Catalysis-Inhibition of M. kandleri GluTR by TPCK indicated a potential involvement of histidinyl residues in catalysis. In the case of cysteine proteinases, the formation of a Cys Ϫ /His ϩ ion pair between the active site cysteinyl residue and a conserved histidinyl residue was found essential for the nucleophilic character of the catalytic cysteine. A conserved histidine of GAPDH acted as a base catalyst facilitating hydride transfer toward NAD ϩ (25). All known GluTR enzymes contain one completely conserved histidinyl residue, which is located at position 84 of M. kandleri GluTR. To distinguish between the potential functions of His-84, the two mutant GluTRs H84A and H84N were analyzed. GluTR-H84A had significantly reduced enzymatic activity indicating the general importance of His-84 for GluTR catalysis (Table III). The GluTR-H84N mutant still possessed 30% of wild type reductase activity and 15% of esterase activity. Analogous to the findings recently obtained for similar mutants of GAPDH these results make the participation of His-84 in Cys-48 reactivity enhancement very unlikely (25). The potential role of His-84 as a base catalyst in facilitating hydride transfer toward the activated glutamate will be subject to future investigations.
An Enzymatic Mechanism for M. kandleri GluTR: A Proposal-Based on the results obtained, we postulate that in analogy to the formal back reaction of GAPDH the nucleophilic Cys-48 of GluTR attacks the ␣-carbonyl group of glutamate activated by an ester linkage to tRNA Glu . Subsequently, a highly reactive enzyme-localized thioester is formed and free tRNA Glu is released. Direct hydride transfer from NADPH, potentially facilitated by His-84, leads to the formation and the release of GSA and NADP ϩ . In the absence of NADPH, the thioester is hydrolyzed by GluTR esterase activity and glutamate is released (Fig. 4). FIG. 4. Postulated enzymatic mechanism for GluTR. The highly reactive sulfhydryl group of Cys-48 acts as a nucleophile to attack the ␣-carbonyl group of glutamate bound to tRNA Glu . An enzymelocalized thioester intermediate is formed with the release of free tRNA Glu . In the presence of NADPH, hydride transfer mediated by the reductase activity of GluTR leads to GSA formation. In the absence of NADPH, the esterase activity hydrolyzes the ester bond with the subsequent release of glutamate. In the C48S mutant, thioester formation is abolished. The presence of the catalytically facilitating histidinyl residue His-84 is indicated. a GluTR mutants were constructed, produced, purified, and analyzed as outlined under "Experimental Procedures." b GluTR activity was measured in the presence and absence of NADPH using the substrate depletion assay as outlined under "Experimental Procedures." Product formation was verified using HPLC analysis. GluTR specific reductase and esterase activity obtained with the wild type enzyme were set to 100%, and all other measured specific activities of the mutant GluTRs were related to that. c ND, not detectable.
kandleri cells and Drs. S. Shima and R. K. Thauer (Max-Planck-Institute Marburg, Germany) for the M. kandleri genomic library and helpful discussions. We thank Dr. Gary Sawer (John Innes Center, Norwich, United Kingdom) for proofreading the manuscript.