l-2-Haloacid dehalogenase (DehL) from Rhizobium sp. RC1

l-2-Haloacid dehalogenase (DehL) from Rhizobium sp. RC1 is a stereospecific enzyme that acts exclusively on l-isomers of 2-chloropropionate and dichloroacetate. The amino acid sequence of this enzyme is substantially different from those of other l-specific dehalogenases produced by other organisms. DehL has not been crystallised, and hence its three-dimensional structure is unavailable. Herein, we review what is known concerning DehL and tentatively identify the amino acid residues important for catalysis based on a comparative structural and sequence analysis with well-characterised l-specific dehalogenases.

produce more than one type of dehalogenase (Table 1). Attention to these bacterial dehalogenases has continually increased owing to their potential application in bioremediation of halogenated organic compounds polluted environment as well as their industrial applications, such as site-directed synthesis of isomers of halogenated organic compounds.

Classification of haloacid dehalogenases
Generally, haloacid dehalogenases are classified according to their substrate specificities and the configuration of their products. Given these criteria, Slater and colleagues classified haloacid dehalogenases as Class 1L that acts specifically on l-2-haloalkanoic acids to produce the corresponding d-2-hydroxyalkanoic acids; Class 1D that acts specifically on d-2-haloalkanoic acids to produce l-2-hydroxyalkanoic acids; Class 2I (inversion-type dehalogenase) that dehalogenates d-and l-2-haloalkanoic acids to produce the corresponding 2-hydroxyalkanoic acids with inverted configurations; and Class 2R (retentiontype dehalogenase) that dehalogenates both isomers of 2-haloalkanoic acids to produce the corresponding 2-hydroxyalkanoic acids that have the same configurations as their substrates (Slater et al. 1997).
Notably, investigating the evolutionary relationships between dehalogenases using substrate specificities as the only criterion can be misleading. For example, Rhizobium sp. RC1 DehL acts only on l-2-chloropropionate, yet its gene sequence (Cairns et al. 1996) differs substantially from the sequences of other bacterial dehalogenases with the same substrate specificity. Therefore, Rhizobium sp. RC1 DehL was tentatively suggested to be the first member of a new group (Hill et al. 1999). In addition, based on an alignment of translated amino acid sequences, Moraxella sp. dehH1 encoding fluoroacetate dehalogenase H-I (Kawasaki et al. 1981a, b), was proposed to be related to the haloalkane dehalogenase genes dhlA from Xanthobacter autotrophicus (Keuning et al. 1985) and dhaA from Rhodococcus rhodochrous (Curragh et al. 1994); and the α-hexachlorocyclohexadiene dehalogenase gene linB from Pseudomonas paucimobilis (Nagata et al. 1993), suggesting the existence of an addition group of haloalkanoic acid dehalogenases (Hill et al. 1999).

Rhizobium sp. RC1 haloacid dehalogenases
In 1979, Berry and colleagues isolated a fast-growing soil bacterium capable of using 2,2-dichloropionate as its sole carbon and energy sources, which they tentatively identified as Rhizobium sp. RC1 (Berry et al. 1979). The bacterium was reported to express three dehalogenases induced by different haloalkanoic acids (Allison et al. 1983). These dehalogenases were genetically characterized using a series of mutant strains (Leigh et al. 1986). The Rhizobium sp. RC1 mutant strain produced by chemical mutagenesis cannot use 2,2-dichloropropionate or d,l-2-chloropropionate as its sole carbon and energy sources. Three secondary mutants were isolated after culturing the original mutant strain on 2,2-dichloropropionate and/or d,l-2-chloropropionate-containing agar. In the presence of 2,2-dichloropropionate two secondary mutants, types 1 and 2 were recovered. The type 1 reverted to the wild-type phenotype (revertant), for which all three dehalogenase activities could be induced. The type 2 mutant constitutively produced DehE, but DehL and DehD were not expressed under any of the tested conditions. The selective pressure induced by the presence of d,l-2-chloropropionate resulted in the type 3 mutant that constitutively produced DehL and DehD but could not produce DehE. The mutation sites in the original mutant strain have not been identified, however they were proposed to be within the regulator gene (Leigh et al. 1986), which would affect production of the three dehalogenases provided that their genes are all controlled by this regulator. Obtaining the type 1 revertant (wild-type phenotype) requires a reversion of the original mutation in the regulator gene, or a repressor mutation in the regulator gene. Similarly to produce the type 2 and 3 secondary mutants, separate mutations in the promoter regions controlling expression of DehE and DehD/DehL are required respectively (Fig. 1).
The stereospecificities of the three Rhizobium sp. RC1 dehalogenases were characterised further by Huyop and Cooper (2003) and Huyop et al. (2004). DehL degrades l-2-chloropropionate and dichloroacetate; DehD is active against d-2-chloropropionate and monochloroacetate; and DehE dehalogenates 2,2-dichloropropionate, d,l-2-chloropropionate, monochloroacetate, dichloroacetate, and trichloropropionate. The lactates produced from d-or l-2-chloropropionate by the three dehalogenases have inverted configurations (Leigh et al. 1988). All these three dehalogenases can also act on 2,3-dichloropropionate with 2-hydroxy-3-chloropropionate being the assumed product. DehE can act on brominated substrates and does so more rapidly than it does to chlorinated substrates (Huyop et al. 2004). Huyop and colleagues assessed the specificity of DehE against mono-, di-, and tri-chloroacetates and the corresponding bromoacetates. All tested bromoacetates had greater associated specificity constants (a determinant of catalytic efficiency) than did their corresponding chloroacetates, suggesting that the brominated compounds would be the preferred DehE substrates (Huyop et al. 2004). DehL and DehD also use dibromoacetate and monobromoacetate, respectively, as substrates, although they are more active against the corresponding chlorinated substrates (Huyop and Cooper 2003). Rhizobium sp. RC1 DehD is the most kinetically active d-specific dehalogenase found (Huyop and Cooper 2003), suggesting that it would be the best d-specific dehalogenase for industrial production of l-specific products. For example, in the industrial production of herbicides and pharmaceuticals, DehD can be used instead of the d-2-haloacid dehalogenase from Pseudomonas in conjunction with a chiral feedback chemical to produce the l-2-chloropropionate intermediate (Taylor 1990).

Rhizobium sp. RC1 dehalogenase genes and their regulation
The Rhizobium sp. RC1 genes encoding the three dehalogenases have been sequenced, and the location of dehD is 177 non-coding base pairs upstream of dehL (Fig. 1). Conversely, the location of dehE relative to that of the other two is not known (Cairns et al. 1996). The deduced DehL and DehD amino acid sequences are only 18 % identical, indicating that these dehalogenases probably do not have many common features (Cairns et al. 1996). This degree of sequence identity is similar to that found for P. putida AJ1 HadD and HadL Jones et al. 1992).

Fig. 1
Proposed genetic organisation and regulation for the Rhizobium sp. RCI dehalogenase genes. R represents regulator gene that controls all three dehalogenases. P1 and P2 represent promoter regions of the structural genes, dehE, and dehD/dehL respectively. The arrows indicate sites of mutations. Original mutant lack the ability to express any of the three dehalogenase structural genes. Type 1 revertant regained the wild type ability to express all the three dehalogenase genes. Type 2 and 3 are constitutive for DehE and DehD/ DehL respectively The deduced amino acid sequence of DehE is not significantly similar to those of DehD and DehL, suggesting no obvious evolutionary linkage between dehE and dehD or dehL (Stringfellow et al. 1997). By characterising the expression of mutant strains of Rhizobium sp. carrying one or more mutations in dehE, dehD/dehL or dehR genes, it was found out that the dehR encoding for a regulatory protein (DehR) probably controls the three dehalogenase structural genes. The proposed regulatory model involves DehR binding to and activating the promoter of the dehalogenase structural genes thereby allowing for their transcription. However, this binding only occurs in the presence of d,l-2-chloropropionate and/or 2,2-dichloropropionate as the inducers (Leigh et al. 1986). dehR has been located upstream of dehE and its product, DehR was proved to control dehE in an engineered E. coli expression system (Huyop and Cooper 2014).
Notably, expression of cloned dehD and dehL is dependent on the presence of a cotransformed lac promoter upstream of these genes in a dehD-and dehL-containing plasmid (Cairns et al. 1996), indicating that the regulatory sequence was not cloned or that it was not functional in the E. coli host. Therefore, a single promoter possibly regulates dehD and dehL expression and physically differs from that regulating dehE. However, the regulatory mechanism(s) for these genes is not fully understood. Additional studies are needed to provide a clearer picture of how these genes are regulated.

Relationships between Rhizobium sp. RC1 dehalogenase sequence and activities
The amino acid sequence of Rhizobium sp. RC1 DehE is similar to that of P. putida PP3 DehI, suggesting that the two enzymes have similar structures, functions, and the same catalytic residues (Hamid et al. 2011). A structural model of DehE was built using DehI as the template (Hamid et al. 2013). The involvement of various amino acid residues at the presumed DehE catalytic active site was assessed by site-directed mutagenesis, which identified TYR34, PHE37, SER188, and ASP189 as catalytically important (Hamid et al. 2015b). DehE is inactive against β-haloalkanoic acids, e.g., 3-chloropropionate. However, when SER188 was mutated to VAL it gained activity against 3-chloropropionate (Hamid et al. 2015a).
In DehD, ARG134, ARG16, and TYR135 are proposed to be necessary for catalysis, with ARG134 playing the key role in stereospecific substrate binding (Sudi et al. 2014a, b). Currently, 3D structure information concerning DehL is unavailable. New studies to determine the catalytic and substrate-interacting residues of DehL, and a three-dimensional structure for it are needed to gain insight into its reaction mechanism and to maximise its industrial and environmental benefits.

l-Stereospecific dehalogenases Diversity of l-stereospecific dehalogenases
Many organisms produce l-stereospecific dehalogenases probably because most naturally occurring halogen-containing organic compounds exist in the l-form (Martínez-Rodríguez et al. 2010). Some of the genes encoding these enzymes have been sequenced (Table 2). Although, most known dehalogenases are proteobacterial in origin, Grampositive bacteria, e.g., Rhodococcus, also degrade haloacid compounds (Jing and Huyop 2007a, b). The Gram-positive bacterium, Staphylococcus sp. produces a haloalkanoic dehalogenase (Camboim et al. 2012a) when this microbe is present in the soil of fluoroacetate-producing plants, a selective-pressure condition. Also, the thermophilic bacterium Sulfolobus tokodaii strain 7, isolated from the Beppu spring in Kyushu Japan in 1983, contains the l-stereospecific dehalogenase L-HAD. This acidophilic bacterium grows optimally at 80 °C, and its genome has been fully sequenced (Kawarabayasi et al. 2001), which is how L-HAD was initially identified. Characterisation of this dehalogenase suggested that it is maximally active at 60 °C (Rye et al. 2009). L-HAD tolerates pH environments between 4 and 10, and remains fully active after incubation at 70 °C for 4 h (Bachas-Daunert et al. 2009).

The residues catalytically important in l-specific dehalogenases
Mutation of certain residues in l-2-dehalogenases significantly affects their catalytic abilities. The catalytically important residues are highly conserved in l-2-dehalogenases. These residues are well characterised in Pseudomonas sp. YL, l-DEX . Its crystal structure (Hisano et al. 1996a, b) and those of X. autotrophicus GJ10 DhlB (Ridder et al. 1995(Ridder et al. , 1997, Burkholderia cepacia MBA4 DehIVa (Schmidberger et al. 2007) and S. tokodaii 7 L-HAD (Rye et al. 2007) have been solved. Kurihara and colleagues identified the catalytically important residues in Pseudomonas sp. YL l-DEX by site-directed mutagenesis ) that involved replacing its highly conserved charged and polar residues (except for the N-terminal Met) with other residues. The genes encoding the mutated proteins were expressed in large amounts under appropriate conditions, purified, and tested for activity towards l-2-chloropropionic acid. The replacement of ASP10, ASP180, ARG41, LYS151, SER175, SER118, THR14, TYR157, and ASN177 caused significant activity decreases. Because replacement of these residues did not cause conformational changes detectable by spectrophotometry and gel filtration, these residues are probably essential for catalysis. ASP10 was suggested to be the catalytic nucleophile ; however, its replacement with ASN did not completely inactivate the enzyme, whereas replacement with ALA, GLY, or GLU did completely inactivate the enzyme . Possibly ASN10 was deamidated, resulting in the wild-type Asp, or it served as a weaker, but still active nucleophile (Ichiyama et al. 2000;Kurihara and Esaki 2008).
The residues found to be essential in l-DEX, are conserved in DhlB and DehIVa from X. autotrophicus GJ10 and B. cepacia MBA4, respectively. The crystal structure analyses of reaction intermediates of DhlB (Ridder et al. 1997) and DehIVa (Schmidberger et al. 2007) suggest functional conservation among the conserved catalytically important residues. However, no site-directed mutagenesis studies have been performed to confirm this supposition.
At the atomic level, the release of a halide by an l-2-haloacid dehalogenase probably proceeds by an S N 2 reaction, during which the halide is replaced by a hydroxyl by one of two possible mechanisms (Fig. 3), which is based on Figure 2 in Kurihara et al. (1995). One possible reaction involves an initial nucleophilic attack on the C2 of the substrate on the (1) RCHXCOOH + OH − → RCHOHOOH + X − R = H or alkyl group, X = halogen atom side opposite that of the halide by a side-chain carboxyl of an acidic dehalogenase residue. All moieties attached to the C2 atom except the halogen are planer in the transition state, such that the nucleophilic carboxyl oxygen interacts with the C2 atom perpendicular to the plane of the transition state and inversion of the C2 stereochemistry occurs with release of the halide. Subsequently, an activated catalytic water molecule cleaves the intermediate, with retention of the C2 stereochemistry, releasing the 2-hydroxyl acid product and the intact enzyme Li et al. 1998). As noted above, ASP10 was suggested to be the nucleophile in the dehalogenases from Pseudomonas sp. YL  X. autotrophicus GJ10 (Ridder et al. 1997) and B. cepacia MBA4 DehIVa (Schmidberger et al. 2007).
A second possible mechanism involves a water molecule, activated by a basic residue, attacking the substrate to release the halide thereby producing the 2-hydroxyl acid product in a single step .

Possible catalytically important residues in Rhizobium sp. RC1 DehL
The crystal structure of Rhizobium sp. RC1 DehL is not available. Nor has any study directly identified the residues involved in DehL catalysis. Even though no obvious sequence identity between DehL and other l-2-haloacid dehalogenases exists, 3D structure comparison of DehL and l-DEX; and multiple alignment of the DehL sequence with those of l-DEX, DhlB, and DehIVa allow us to infer the possibly catalytically important DehL residues. 3D structure of DehL (Fig. 5a) was predicted by Modeller 9.15 using l-DEX (PDB IB: 1JUD) as template. Structural superposition of DehL with l-DEX (Fig. 5b) and multiple sequence alignment of DehL, l-DEX, DhlB, and DehIVa ( Fig. 4; Table 3) show ASP10, THR14, ARG41 and SER175, which have been shown to affect catalytic activity in l-DEX are conserved in DehL. However, THR14, ARG41 and SER175 were observed to be only structurally but not sequentially conserved in DehL. This is probably due to variation in size of the aligned sequences. The ARG51 of DehL is not in directly similar structural position as the ARG41 of l-DEX, although the positions of the two ARG residues are in relative position and pointing at the same direction in the active site. The variation in positions of the ARG residues in the two dehalogenases might be due to the difference in size of the active site, which is dependent on the range of substrate specificities. For example l-DEX activity is not limited to short-carbon-chain 2-haloacids a b Fig. 3 Proposed catalytic reaction mechanisms for l-2-haloacids dehalogenases. a Attack on the C2 of the substrate by a dehalogenase side-chain carboxyl to produce an ester intermediate with subsequent attack by a water molecule on the intermediate to produce the corresponding hydroxyacid with the opposite stereoconfiguration. b Water is activated by a basic residue and attacks the substrate to produce the hydroxyacid with simultaneous release of halide ion. Adapted from Kurihara et al. (1995) such as monochloroacetate but it also acts on long-carbon-chain of 2-haloacids such as 2-bromohexadecanoate in n-heptane ; whereas l-2-chloropropionate is the longest carbon-chain DehL ever reported to acts on.
Conservation in amino acid often confers functional conservation. Therefore, it can be hypothesise that the catalytically important residues of l-DEX that are conserved in DehL may also be catalytically important and probably have similar functions. This was reported to be the case among the nine conserved catalytically important residues in l-DEX, DhlB and DehIVa. ASP10 in l-DEX that corresponds to ASP13 in DehL plays a nucleophilic role by attacking C2 of L-2-chloropropionate during dehalogenation catalysis . The corresponding residues in DhlB (ASP8) (Ridder et al. 1997) and DehIVa (ASP11) (Schmidberger et al. 2007) were reported to have similar function. SER 175 in l-DEX (SER183 in DehL) and its corresponding residue, SER171 in DehIVa both involve in a hydrogen bond with ASP10 to probably maintain the orientation of its carboxyl group in a way suitable to attack the C2 of the substrate (Hisano et al. 1996a, b;Schmidberger et al. 2007). As a positively charged polar residue, ARG41 in l-DEX (ARG 51in DehL) accepts the released chloride ion by electrostatic interaction (Kondo et al. 2014). Furthermore, the corresponding residue in DehIVa (ARG42) was proposed to play key role in substrate "lock down" mechanism; and also acts a member of the halidebinding cradle together with ASN120 and TRP180 (Schmidberger et al. 2007). The role of THR14 in l-DEX (THR17 in DehL) is not yet determined, however its corresponding residue in DhlB (THR12) together with SER171 and ASN173 were reported to firmly hold the ASP8 in a position that favours the nucleophilic attack (Ridder et al. 1997). On the other hand, the rest of the catalytically important residues of l-DEX (SER118, LYS 151, TYR157, ASN177 and ASP180) not conserved in DehL may be probably the same as those in l-DEX but in different positions in the active site or substituted by similar residues. To fully elucidate the mechanism of DehL dehalogenation and the contributions of the specific residues, additional work is needed.

Conclusions
l-2-Haloacid dehalogenases have been found in many different bacteria; many of these enzymes have been sequenced, and for some, their substrate specificities and kinetics have been well characterized. In addition, four have been crystallised and their threedimensional structures solved, which is informative concerning their possible catalytic mechanism(s). Although, DehL from Rhizobium sp. RC1 dehalogenates the same substrates as l-2-haloacid dehalogenases from other organisms do, its amino acid sequence is quite different from those of the other enzymes. Results of our pairwise DehL amino acid sequence comparison with those of the crystallised proteins; and the structural