dl-2-Haloacid Dehalogenase fromPseudomonas sp. 113 Is a New Class of Dehalogenase Catalyzing Hydrolytic Dehalogenation Not Involving Enzyme-Substrate Ester Intermediate*

dl-2-Haloacid dehalogenase fromPseudomonas sp. 113 (dl-DEX 113) catalyzes the hydrolytic dehalogenation of d- andl-2-haloalkanoic acids, producing the correspondingl- and d-2-hydroxyalkanoic acids, respectively. Every halidohydrolase studied so far (l-2-haloacid dehalogenase, haloalkane dehalogenase, and 4-chlorobenzoyl-CoA dehalogenase) has an active site carboxylate group that attacks the substrate carbon atom bound to the halogen atom, leading to the formation of an ester intermediate. This is subsequently hydrolyzed, resulting in the incorporation of an oxygen atom of the solvent water molecule into the carboxylate group of the enzyme. In the present study, we analyzed the reaction mechanism of dl-DEX 113. When a single turnover reaction of dl-DEX 113 was carried out with a large excess of the enzyme in H2 18O with a 10 times smaller amount of the substrate, either d- or l-2-chloropropionate, the major product was found to be18O-labeled lactate by ionspray mass spectrometry. After a multiple turnover reaction in H2 18O, the enzyme was digested with trypsin or lysyl endopeptidase, and the molecular masses of the peptide fragments were measured with an ionspray mass spectrometer. No peptide fragments contained 18O. These results indicate that the H2 18O of the solvent directly attacks the α-carbon of 2-haloalkanoic acid to displace the halogen atom. This is the first example of an enzymatic hydrolytic dehalogenation that proceeds without producing an ester intermediate.

DL-DEXs have been purified from Pseudomonas sp. 113 (DL-DEX 113) (14), Pseudomonas putida PP3 (15), and Rhizobium sp. (16). However, none of the reaction mechanisms of these DL-DEXs have been studied, and it is unknown whether the reaction mechanism of DL-DEX is similar to that of other halidohydrolases (dehalogenases that catalyze the hydrolytic dehalogenation). We previously determined the primary structure of DL-DEX 113 (Fig. 2), and found that it is similar to that of D-DEX from Pseudomonas putida AJ1 (17). We also showed that DL-DEX 113 has a single and common catalytic site for both D-and L-enantiomers based on a site-directed mutagenesis experiment and kinetic analysis (17). In the present study, we analyzed the reaction mechanism of DL-DEX 113 by means of 18 O incorporation experiments, and found that the reaction does not involve the formation of an enzyme-substrate ester intermediate. A water molecule is probably activated by a catalytic base of the enzyme, directly attacking the ␣-carbon of D-and L-2-haloalkanoic acids to displace the halogen atom (Fig.  1B). This is the first example of an enzymatic dehalogenation that proceeds through the mechanism shown in Fig. 1B. 2 18 O (95-98%) was obtained from Cambridge Isotope Laboratories (Andover, MA) and Nippon Sanso (Tokyo, Japan). L-and D-2-chloropropionate were purchased from Sigma. Lysyl endopeptidase of Achromobacter lyticus M497-1 and trypsin (TPCK treated) were from Wako Industry Co., Ltd. (Osaka, Japan) and Sigma, respectively. All other chemicals were of analytical grade.

Materials-H
Purification of DL-DEX 113-Recombinant Escherichia coli JM109 cells harboring p4b (1) encoding DL-DEX 113 (17) were cultivated at 37°C for 14 -18 h in a Luria-Bertani medium (1% polypeptone, 0.5% yeast extract, and 1% NaCl, pH 7.0) containing 150 g/ml ampicillin and 0.2 mM isopropyl-1-thio-␤-D-galactoside. The cells were collected by centrifugation, suspended in a 50 mM potassium phosphate buffer (pH 7.0), and disrupted by ultrasonic oscillation at 4°C for 20 min with a Seiko Instruments ultrasonic disintegrator model 7500. The cell debris was removed by centrifugation. The supernatant solution was brought to 40% saturation with ammonium sulfate, and the precipitate was removed by centrifugation. The supernatant was applied to a Butyl Toyopearl 650M column, and elution was carried out with a linear gradient of 0 -30% saturated ammonium sulfate in a 50 mM potassium phosphate buffer (pH 7.0). The active fractions were pooled and concentrated by ultrafiltration.
Single Turnover Reaction of DL-DEX 113 in H 2 18 O-For a single turnover experiment, 200 nmol of DL-DEX 113 in 50 l of a 400 mM Tris-H 2 SO 4 buffer (pH 9.5) was lyophilized. The reaction was initiated by dissolving the dried enzyme in 50 l of H 2 18 O containing 20 nmol of D-or L-2-chloropropionate (neutralized with NaOH), and the mixture was incubated at 30°C for 24 h. The reaction mixtures were ultrafiltered, diluted 10-fold with 50% acetonitrile/H 2 O (1:1), and then introduced into the mass spectrometer using a Harvard Apparatus syringe infusion pump operating at 2 l/min. The molecular mass of the produced lactate was measured with a PE-Sciex API III triple quadrupole mass spectrometer equipped with an ionspray ion source in the negative ion mode (Sciex, Thornhill, Ontario, Canada).
Digestion of DL-DEX 113 with Trypsin-Lyophilized 10 nmol of DL-DEX 113, 1.2 mol of D-or L-2-chloropropionate (neutralized with NaOH), and 1.25 mol of Tris-H 2 SO 4 (pH 9.5) were mixed in 50 l of H 2 18 O and incubated at 30°C for 24 h. The enzyme was inactivated by incubating the reaction mixture for 10 min at 80°C and then denatured by the addition of 100 l of 5 M urea in 100 mM Tris-H 2 SO 4 (pH 7.5). Subsequently, the volume was adjusted to 500 l with 120 mM Tris-H 2 SO 4 (pH 7.5) in order to reduce the urea concentration to 1 M and the pH to approximately 8.0. DL-DEX 113 in this solution was digested with 5 g of TPCK-treated trypsin at 37°C for 12 h.
To digest DL-DEX 113 in H 2 18 O, the protein was denatured by the addition of 100 l of 3 M urea in 100 mM Tris-H 2 SO 4 (pH 7.0) prepared with H 2 18 O. Thereafter, 150 l of 100 mM Tris-H 2 SO 4 (pH 7.5) containing 5 g of trypsin prepared with H 2 18 O was added to this solution and incubated at 37°C for 12 h.
Digestion of DL-DEX 113 with Lysyl Endopeptidase-Lyophilized 10 nmol of DL-DEX 113, 1.2 mol of D-or L-2-chloropropionate (neutralized with NaOH), and 2.5 mol of Tris-H 2 SO 4 (pH 9.5) were mixed in 50 l of H 2 18 O and incubated at 30°C for 24 h. The enzyme was inactivated by incubating the reaction mixture at 80°C for 10 min, denatured with 8 M urea, and subsequently digested with 825 pmol of lysyl endopeptidase at 37°C for 12 h.
To digest the enzyme in H 2 18 O, 10 nmol of DL-DEX 113, 1.0 mol of D-or L-2-chloropropionate (neutralized with NaOH), and 20 mol of Tris-H 2 SO 4 (pH 9.5) were mixed in 50 l of H 2 18 O and incubated at 37°C for 24 h. A 20-l aliquot was lyophilized and dissolved with 20 l of H 2 18 O containing 8 M urea. After incubation at 37°C for 1 h, 30 l of 1 M Tris-H 2 SO 4 (pH 9.0) in H 2 18 O and 10 l of 33 M lysyl endopeptidase in H 2 18 O were added to this solution, and incubation was carried out at 37°C for 12 h.
Liquid Chromatography/Mass Spectrometry Analysis of the Proteolytic Digests-The proteolytic digests of the enzyme were loaded onto a YMC-PackC4-AP column (100 ϫ 1.0-mm inner diameter) (YMC Co., Kyoto, Japan) connected to the mass spectrometer and then eluted with a linear gradient of 0 -80% acetonitrile in 0.05% trifluoroacetic acid over 80 min at a flow rate of 10 l/min. A total ion current chromatogram was recorded in the single-quadrupole mode with a PE-Sciex API III mass spectrometer equipped with an ionspray ion source. The quadrupole was scanned from 300 to 2000 atomic mass units with a step size of 0.25 atomic mass units and a 0.5-ms dwell time per step. Ionspray voltage was set at 5 kV, and the orifice potential was 80 V. The molecular mass of each peptide was calculated with MacSpec software supplied by Sciex.
Site-directed Mutagenesis of DL-DEX 113-Plasmid p4b (1)  113, L-DEX YL (19), and L-DEX from Pseudomonas putida no. 109 (20) (final 0.45 mg/ml) were mixed with 1 M hydroxylamine in 1 M Tris-H 2 SO 4 (pH 9.0) in the presence or absence of 100 mM L-2-chloropropionate and incubated at 30°C for 60 min. After dialysis against a 50 mM potassium phosphate buffer (pH 7.5), the remaining activities of the enzyme were measured by the standard assay method as described below.
Enzyme and Protein Assay-DL-DEX 113 was assayed with 25 mM Dor L-2-chloropropionate as a substrate. The chloride ions released were measured spectrophotometrically (21). One unit of enzyme was defined as the amount of the enzyme that catalyzes the dehalogenation of 1 mol of the substrate/min. The protein assay was done with a Bio-Rad protein assay kit.

Single Turnover Reaction of DL-DEX 113 in H 2
18 O-We conducted the single turnover reaction of DL-DEX 113 in H 2 18 O with D-or L-2-chloropropionate as a substrate, using an excess amount of the enzyme. We found that a majority of the lactate produced was labeled with 18 O (Fig. 3, A and B). This makes a clear contrast with our results on the L-DEX YL reaction, which proceeds through the mechanism involving an ester intermediate ( Fig. 1A) (5). Only 10% of the D-lactate produced from L-2-chloropropionate by L-DEX YL was labeled with 18 O (Fig.  3C). This suggests that in the DL-DEX 113 reaction an oxygen atom of the solvent water was directly incorporated into the product. While supporting the mechanism shown in Fig. 1B, this is not compatible with the mechanism in Fig. 1A, in which an oxygen atom of the solvent water is first incorporated into the enzyme.
Liquid Chromatography/Mass Spectrometry Analysis of the Peptides Proteolytically Formed from DL-DEX 113-A multiple turnover reaction of DL-DEX 113 was carried out in H 2 18 O with D-or L-2-chloropropionate as a substrate. After completion of the reaction, the enzyme was digested with TPCK-treated trypsin, and the resulting peptide fragments were separated on a reversed phase column interfaced with an ionspray mass spectrometer as a detector. If the reaction proceeds through the mechanism in Fig. 1B, 18 O should not be detected in the proteolytic fragments. The spectrometer was used in the single quadrupole mode, and the total ion current chromatogram was obtained as shown in Fig. 4 Fig. 2). The molecular masses of all peptides were virtually indistinguishable from the predicted ones whether the reaction was conducted with D-or L-2-chloropropionate. Since peptides containing amino acid residues 1, 106 -107, 135-142, 181-183, 229 -238, 250 -254, 284 -285, and 299 -300 were not found in the trypsin-digested sample, we also analyzed lysyl endopeptidase-digested enzyme by the same method. Peptides 120 -142, 232-285, and 299 -306 were identified, and their molecular masses were virtually indistinguishable from the predicted ones (Table II, Fig. 2). The molecular masses of the peptides containing amino acid residues 1, 106 -107, 181-183, and 229 -231 could not be measured.
Liquid Chromatography/Mass Spectrometry Analysis of the Proteolytic Fragments Formed in H 2 18 O-An oxygen atom of Step size was 0.1 atomic mass unit, and dwell time was 10 ms/step. Ionspray voltage was set at Ϫ3.5 kV, and the orifice potential was Ϫ50 V.  the catalytic carboxylate group of haloalkane dehalogenase from X. autotrophicus GJ10 is rapidly replaced by an oxygen atom of the solvent water even in the absence of the substrate (9). If this is the case for DL-DEX 113, 18 O once incorporated into the acidic amino acid residue during the dehalogenation should have been replaced by the 16 O of the solvent water during the treatment with trypsin or lysyl endopeptidase, which raised the possibility that the increase in the molecular mass of the peptides might not be detectable in the above experiments.
To examine this possibility, we carried out the denaturation and trypsin digestion of the enzyme in H 2 18 O. If there is an oxygen atom ( 16 O) in the enzyme which is readily exchangeable for an oxygen ( 18 O) of the solvent water, an increase in the molecular mass of the proteolytic peptides is expected. The molecular masses of the proteolytic peptides, 2-8, 9 -21, 22-34, 35-43, 44 -66, 67-76, 77-105, 108 -119, 120 -134, 144 -151,  152-162, 184 -196, 197-210, 211-228, 239 -249, 255-272, and 273-283 were approximately 4 Da higher than the predicted molecular masses in the system with either D-or L-2-chloropropionate as a substrate (Table III, Fig. 2). A peptide containing amino acid residues 163-180 appeared as a dimeric form, and its molecular mass was approximately 8 Da higher than the predicted one. These increases are attributed to the incorporation of two 18 O atoms into the ␣-carboxylate group of the C-terminal amino acid residue of each tryptic fragment. The molecular mass of peptide 301-307 derived from the C-terminal region of the enzyme was almost identical to the predicted value. Peptide fragments prepared with lysyl endopeptidase were also analyzed, and the molecular masses of peptides 120 -142, 232-285, 286 -298, and 299 -306 were about 4 Da higher than the predicted ones (Table IV, Fig. 2). These show that the 18 O of the solvent water was incorporated only into the ␣-carboxylate group of each peptide but not into the side chain carboxylate group.
Effect of the Replacement of Asp 181 on Enzyme Activity-The molecular masses of peptides 1 (M), 106 -107 (LK), 143 (R), 181-183 (DIR), and 229 -231 (IRK) could not be determined in the above experiments. Therefore, we could not exclude the possibility that 18 O was incorporated into Asp or Glu in these peptides. However, among these peptides, only peptide 181-183 contains an acidic residue, Asp 181 . We replaced Asp 181 with Ala, Arg, and Glu by site-directed mutagenesis to clarify whether Asp 181 is involved in the catalytic reaction shown in Fig. 1A as Asp 10 of L-DEX YL is. The activities of these mutant enzymes were similar to that of the wild-type enzyme ( Table  V), indicating that Asp 181 is not essential for the catalysis.
Effect of Hydroxylamine on Enzyme Activity-We previously found that hydroxylamine performs a nucleophilic attack on the active site aspartate residue (Asp 10 ) of L-DEX YL (6). This inactivation was observed only in the presence of the substrate, and an ester intermediate formed from Asp 10 and the substrate was thought to be a target of hydroxylamine. This was confirmed by mass spectrometric analysis of the inactivated enzyme, which showed that the modified Asp 10 residue contained both hydroxylamine-and substrate-derived moieties. The inactivation of 4-chlorobenzoyl-CoA dehalogenase by hydroxylamine was also reported (13). In contrast, no inactivation of DL-DEX 113 was observed (Table VI).

DISCUSSION
The reaction mechanism of DL-DEX 113 was studied by 18 O incorporation experiments and site-directed mutagenesis. Single turnover reactions carried out in H 2 18 O indicated that an oxygen atom of the solvent water is directly incorporated into the product (Fig. 3, A and B). We also found that an oxygen atom of the solvent water is not incorporated into the side chain carboxylate groups of the acidic amino acid residues of the enzyme in the dehalogenation reaction (except for Asp 181 , whose molecular mass could not be measured) (Tables I-IV). A site-directed mutagenesis experiment showed that Asp 181 is not essential in the catalysis (Table V). These results are consistent with the general base mechanism shown in Fig. 1B, but not with the mechanism shown in Fig. 1A. This applies to the dehalogenations of both enantiomers of 2-haloalkanoic acids because the results obtained for both enantiomers were virtually the same. We previously reported that DL-DEX 113 has a single and common catalytic site for both L-and D-enantiomers based on a site-directed mutagenesis experiment and kinetic analysis (17). This conclusion is supported by our present data showing that the enzymatic dehalogenations of both enantiomers proceed through the same mechanism as shown in Fig. 1B.
The reaction mechanisms of three kinds of halidohydrolases have been analyzed: haloalkane dehalogenase from X. autotrophicus GJ10 (8 -10), 4-chlorobenzoyl-CoA dehalogenases from Pseudomonas sp. strain CBS3 (11,12), and Arthrobacter sp. 4-CB1 (13) and L-DEX YL (5,7). The reactions of these dehalogenases proceed as shown in Fig. 1A. Although there are no similarities in their primary and tertiary structures (12,(22)(23)(24)(25)(26), they resemble one another in their catalytic reactions, in which an essential ester intermediate is produced from the active site nucleophilic carboxylate and the substrate molecule. So far, no halidohydrolases have been shown to catalyze the reaction without the formation of the ester intermediate. Thus, DL-DEX 113 is unique in that its reaction does not involve the formation of an ester intermediate. Since D-DEX from P. putida AJ1 (27) shows sequence similarity with DL-DEX 113 (17), the reaction of D-DEX probably proceeds through the mechanism shown in Fig. 1B. Although DL-DEX and L-DEX can catalyze the same reaction (hydrolysis of L-2-haloalkanoic acids), our present data clearly show that the reaction mechanisms of DL-DEX and L-DEX are completely different from each other.
Recently, we found that Glu 69 and Asp 194 are essential for the catalysis of DL-DEX 113 by site-directed mutagenesis (17). Several hydrolases such as aspartic proteases (28) and glycosyl hydrolases (29) have been shown to possess Glu or Asp as a catalytic base that activates a water molecule to attack the substrate molecule. In these enzymatic hydrolyses, ester intermediates are not produced. In this respect, DL-DEX 113 resem-bles these hydrolases, and Glu 69 and/or Asp 194 may be involved in the activation of a water molecule that attacks the ␣-carbon atom of the substrate. Further studies including crystallographic analysis of the enzyme are now being carried out to identify the active site residues and to clarify their roles in the catalysis.