Catalytic mechanism of the haloalkane dehalogenase LinB from Sphingomonas paucimobilis UT26

dehalogenase UT26 (LinB), we studied the steady-state and pre-steady- state kinetics of the conversion of the substrates 1-chlorohexane, chlorocyclohexane and bromocyclohexane. The lead to a proposal of a kinetic consisting of three main steps: (i) substrate binding, (ii) cleavage of the carbon-halogen with simultaneous formation of alkyl-enzyme Release

Haloalkane dehalogenases (EC 3.8.1.5) make up an important class of enzymes that are able to cleave carbon-halogen bonds in halogenated aliphatic compounds. There is a growing interest in these enzymes due to their potential use in bioremediation, as industrial biocatalysts, or as biosensors. Structurally, haloalkane dehalogenases belong to the a/b-hydrolase fold superfamily (1). Without exception, haloalkane dehalogenases contain a nucleophile elbow (2) which is the most conserved structural feature within the a/b-hydrolase fold. The other highly conserved region in haloalkane dehalogenases is the central b-sheet. Its strands, flanked on both sides by ahelices, form the hydrophobic core of the main domain that carries the catalytic triad Asp-His-Asp/Glu. The second domain, consisting solely of a-helices, lies like a cap on top of the main domain. Residues at the interface of the two domains form the active site. Whereas there is significant similarity in the catalytic core, the sequence and structure of the cap domain diverge considerably among different dehalogenase. The cap domain is proposed to play a prominent role in determining substrate specificity (3,4). A reaction mechanism for haloalkane dehalogenase has been proposed on the basis of X-ray crystallographic (5), site-directed mutagenesis (6,7) and kinetic (8,9)  presence of three specificity classes within this family of enzymes (10,11). Three haloalkane dehalogenases representing these different classes have been isolated and characterised in detail so far: the haloalkane dehalogenase DhlA (12), the haloalkane dehalogenase LinB from Sphingomonas paucimobilis UT26 (11) and the haloalkane dehalogenase DhaA from Rhodococcus rhodochrous NCIMB 13064 (13). The kinetic mechanism has been solved for DhlA (9) and DhaA (14). Comparison of kinetic mechanism of DhlA and DhaA shows that the overall scheme is similar. The main difference was found at the rate-limiting steps. For their best substrates, it was found that as the rate of halide release represents the slowest step in the catalytic cycle of DhlA, whereas liberation of the alcohol is the rate limiting in the catalytic action of DhaA.
This study presents a detailed insight into the kinetic mechanism of the haloalkanedehalogenase LinB that is obtained using steady-state as well as pre-steady-state kinetic techniques. The similarities and differences between the kinetic mechanisms of the three well studied haloalkane dehalogenases are also discussed in this paper. Menten kinetic constants were calculated from the initial steady-state rates of product formation by nonlinear regression analysis using ORIGIN 6.1 (OriginLab Corporation, USA).

Protein Expression and Purification-To
Inhibition Kinetics-Substrate to product conversion by the action of LinB was monitored using an isothermal titration calorimeter type CSC 4200 (Calorimetry Sciences Corporation, USA). Derivation of Rates-The fluorescence traces (F) of multiple turnover experiments were fitted to a single exponential of the form F = A ´ exp(-k obs *t) + E using the program MICROMATH SCIENTIST 2.0 (ChemSW, USA). The parameter A is the amplitude, k obs is the observed rate constant and E is the floating endpoint. Equation 1 or 2 was fitted to the k obs data to obtain the respective kinetic constants. Single turnover stopped flow fluorescence and rapid quench flow data were fit to an appropriate kinetic scheme using GEPASI 3.2 (17). The single turnover data by guest on March 20, 2020 http://www.jbc.org/ Downloaded from were fitted by numerical simulation using the kinetic constants from multiple-turnover and steady-state experiments as starting values. The kinetic constants were refined by comparing fits of the single-turnover data and the k obs data. To simulate the experimental fluorescence traces with the obtained kinetic constants and reaction schemes, the traces were described as a sum of the contributions of each enzyme species to the total fluorescence.
Modelling of Protein-Ligand Complexes-Crystal structure coordinates of the LinB enzyme were obtained from Protein Data Bank (PDB ID 1CV2). The polar hydrogens were added to the structure by WHATIF 5.0 (18). The catalytic His272 was singly protonated in the Nd on accordance with its catalytic function. The script q.kollua was used for addition of charges on all enzyme atoms and script addsol for addition of solvation parameters to the carbon atoms in the protein structure (19). The complex of LinB with halide bound in the active site cavity was obtained using the superimpose function of the visualising program INSIGHTII 95 (Accelrys, USA). All rotatable bonds were specified in each of the substrates using the program AUTOTORS of AUTODOCK 3.0 (19). Enzyme(-halide)-substrate complexes were modelled by AUTODOCK 3.0 using the Lamarckian Genetic Algorithm (19). Fifty docking runs were performed for each enzyme(-halide)-substrate complex. Calculated substrate orientations were clustered with a clustering tolerance for the root mean square positional deviation of 0.5 Å , but orientations of 1-chlorohexane and 1-bromohexane in LinB-halide complexes were clustered with a clustering tolerance of 1 Å in order to reduce number of obtained orientations. Site-directed mutagenesis of LinB suggests that Asp108, His272 and Glu132 comprise the catalytic triad in this enzyme (20,21). The proposed reaction cycle (Scheme I) is initiated by binding of the substrate in the Michaelis complex (E.RX).The binding site for the halogen that is cleaved off is formed by Asn38, Trp109 and Pro208. The binding step is followed by a nucleophilic attack of Asp108 on the carbon atom to which the halogen is bound, leading to cleavage of the carbon-halogen bond and formation of alkyl-enzyme intermediate (E-R.X -). The intermediate is subsequently hydrolysed by activated water, with His272 acting as a base catalyst, which gives formation of the enzyme-product complex (E.ROH.X -). The function of Glu132 is to keep His272 in the proper orientation and to stabilise the positive charge that develops on histidine imidazole ring during the reaction. The final step is release of the products.
Scheme I Four substrates were selected as model compounds for a detailed kinetic study of LinB reaction. Chlorocyclohexane is structuraly similar to the natural substrate for LinB 1,3,4,6tetrachloro-1,4-cyclohexadiene. Bromocyclohexane was selected as a brominated analogue of chlorocyclohexane. Furthermore, 1 -chlorohexane was selected as one of the best known substrates for LinB with a k cat /K m = 1.6´10 5 M -1 .s -1 , and 1-bromohexane as its brominated analogue. However, 1-bromohexane was later excluded from the set of substrates because its high volatility and low water solubility, which prevented usage of the rapid-mixing instruments for transient kinetic measurement. [here Figure 2] Product Binding and Release-The product inhibition patterns were examined to obtain information about the mechanism of product release. The LinB reaction was assayed in the presence and absence of a fixed amount of the corresponding products using isothermal titration calorimetry. Inhibition of LinB bromocyclohexane hydrolysis by bromide and cyclohexanol ( Figure 3A and 3B) and inhibition of LinB 1-chlorohexane hydrolysis by chloride and 1-hexanol ( Figure 3C and 3D) was examined using Lineweaver-Burk plots. For all products the same intercept was observed, whereas the slopes differed suggesting that the inhibitors had no effect on the initial velocity but only increased the apparent K m . From these results, the competitive inhibition pattern was deduced for both halide and alcohol inhibition. The kinetics of 1chlorohexane conversion in the presence of chloride also showed substrate inhibition. The data were further analysed by fitting the equations for competitive, noncompetitive and mixed inhibition models. The best fit was evaluated by comparison of standard errors and variance. The halide inhibition were data fitted best with the competitive model, which yielded inhibition constants for chloride (K i = 888 ± 45 mM) and for bromide (K i = 282 ± 21 mM.) Also 1-hexanol (K i = 3.3 ± 0.1 mM) and cyclohexanol (K i = 43 ± 1.3 mM) were competitive inhibitors of LinB reaction.
[here Figure 3] [here Figure 4] The amplitude of the fluorescence change that is observed upon mixing of LinB with cyclohexanol and 1-hexanol was not sufficient for a detailed fluorescence evaluation of alcohol binding. Similar to what was found with halide, only steady-state changes of fluorescence could be observed, suggesting that also alcohol binding reaches rapid equilibrium in the dead time of the stopped flow instrument. Considering that the release of both alcohol and halide reach rapid equilibrium and that the enzyme-bound complexes of both products have high dissociation constants, the reaction mechanism from Scheme I can be simplified to Scheme II.
Since no products of the reverse reaction have been detected with DhlA (9) or DhaA (14) Figure   5A), yielding an observed rate constant for each substrate concentration (k obs ). A relationship between k obs and the microscopic rate constants (Equation 1) was derived using Scheme II. The dependency of k obs on the 1-chlorohexane concentration showed saturation ( Figure 5B). Thus, k 2 and K s could be determined separately. The dependency of k obs on the bromocyclohexane and chlorocyclohexane concentrations ( Figure 5C and 5D) showed no saturation behavior even after reaching the limit of substrate solubility. This indicates that the K s is high, leading to a simplification of the function for k obs (Equation 2). In this case, only the ratio k 2 /K s could be determined and only the lower limits of k 2 and K s could be obtained for these substrates.
[here Figure 5] The cleavage of the carbon-halogen bond was found to be the fastest step in the catalytic cycle (Table 1). The rate of cleavage of carbon-bromide bond is faster than cleavage of carbon-chloride bond, as exemplified from comparison of kinetic constants determined with 1-chlorocyclohexane and 1-bromocyclohexane. This observation is in agreement with bromine being a better leaving group than chlorine in a bimolecular substitution reaction. The same result was found for the [here Table 1] by guest on March 20, 2020 http://www.jbc.org/

Downloaded from
Hydrolysis of the alkyl-enzyme intermediate was found to be the slowest step in the kinetic mechanism of LinB using chlorocyclohexane, bromocyclohexane and 1-chlorohexane as model substrates. Hydrolysis is therefore the rate-limiting step for the overall kinetics of LinB and it is highly correlated with k cat . It is interesting to note that although dehalogenation of the substrates bromocyclohexane and chlorocyclohexane results in the same alkyl-enzyme intermediate, the rates of their hydrolysis differ significantly, being 33-times slower in case of chlorocyclohexane.
This might be attributed to the presence of different halide ion in the active site after the first reaction step and its interaction with the transition state structure of the hydrolytic reaction.
Both products are released from the enzyme active site after the hydrolysis. Considering the high rate of product release and the high dissociation constants of halide and alcohol from the enzyme the release of products was not included in the kinetic mechanism as a distinct step.
Alcohol and halide do not compete for the same binding site and the release of one product is not dependent on the presence of the other product in the active site of LinB. Taken together with the finding that both an alcohol and a halide are competitive inhibitors for LinB, this implies a random sequential mechanism for the release of products.
A comparison of the kinetic mechanism of LinB with the mechanisms reported for DhlA (8) and DhaA (14) Table 2). The main and very important difference in kinetic mechanism is in the rate-limiting step. Whereas the halide release is the predominant rate- [here Table 2] The high rate of export of products out of LinB can be explained by the presence of three Another structural difference linked to the export of halide ions from the active site is found in the halide-stabilising residues (26). One of two primary halide-stabilising residues (Trp125 in DhlA, Trp107 in DhaA and Trp109 in LinB) is fully conserved in all haloalkane dehalogenases, while the second residue differs both by its nature and location within the protein structures  Table 1 and Scheme II were used for the simulation.