Partial Purification and Characterization of the Recombinant Benzaldehyde Dehydrogenase from Rhodococcus ruber UKMP-5M

Background Benzaldehyde dehydrogenase (BZDH) is encoded by the xylC that catalyzes the conversion of benzaldehyde into benzoate in many pathways such as toluene degradation. Objectives In this study, the xylC gene from Rhodococcus ruber UKMP-5M was expressed in Escherichia coli, purified, and characterized. Materials and Methods The xylC was amplified and cloned in E. coli. The recombinant plasmid pGEMT-xylC was digested by NdeI and HindIII to construct plasmid pET28b-xylC and transformed in E. coli BL21 (DE3). Expression of the recombinant protein was induced by 1 mM isopropyl β-D-thiogalactoside (IPTG) at 37°C. The BZDH was purified by ion exchange chromatography, in which the product was an NAD-dependent enzyme using benzaldehyde as a substrate for enzyme characterization. The end metabolite was identified via gas chromatography mass spectrometry (GC-MS). Results The recombinant BZDH is 27 kDa, purified by ion exchange chromatography. The activity of BZDH was 9.4 U.μL-1 The optimum pH and temperature were 8.5 and 25ºC, respectively. The Michaelis constant (Km) and maximum velocity (Vmax) were 4.2 mM and 19.7 U.mL-1, respectively. The metabolite of BZDH was benzene carboxylic acid as determined by GC-MS analysis. Conclusions BZDH has the ability to degrade benzaldehyde to less toxic compounds. The BZDH is a critical enzyme for the degradation of aromatic hydrocarbons in Rhodococcus sp. The BZDH from R. ruber UKMP-5M is showed similar function with other aldehyde dehydrogenases.


Background
Rhodococcus ruber UKMP-5M is a hydrocarbon degrading bacteria through catabolic pathway crude oil and toluene (1). The toluene degradation pathway consists of two steps. The fi rst is an upper pathway induced by toluene, which catalyzes the conversion of aromatic hydrocarbons to their carboxylic acid derivatives. The second is the lower pathway induced by benzoic acid (the alternative pathway); the product that is supplied by the upper pathway (2). Benzaldehyde dehydrogenase (BZDH) is an important enzyme involved in the upper pathway of toluene and xylene degradation. This enzyme is a member of aldehyde dehydrogenases, which detoxifi es benzaldehyde to carboxylic acid compounds via irreversible oxidation reaction (3). Two types of BZDH are determined: type I induced by benzoylformate and involved in the mandelate pathway (4) and the type II induced by benzaldehyde involved in toluene and xylene degradation pathway (5). Type II of BZDH has been reported by many bacteria such as Pseudomonas putida (2), Acientobacter calcoaceticus (4), Rhodococcus rhodochrous OFS (6) and Pseudoxanthomonas spadix (7). The BZDH in P. putida mt-2 is encoded by the TOL plasmid (pWW0) to catalyze various mono aromatic alcohols and aldehydes (8). More catalytic effi ciency

Cloning of the xylC Gene for BZDH
The total DNA of bacteria was extracted using Wizard genomic DNA-purifi cation kit (Promega, Madison, USA). The xylC gene was amplifi ed in an automated thermal cycler (Bio-Rad, California USA) using specifi c primers designed based on genome sequences from R. ruber UKMP-5M. Restriction enzyme recognition sites were underlined in the sequences. The forward oligonucleotide containing an NdeI site (CAꞌTATG) and reverse oligonucleotide with a HindIII site (AꞌAGCTT).
The purifi ed DNA (~0.8 kb) was ligated into pGEM ® -T Easy vector (Promega, Madison, USA) and transformed into competent cells of E. coli DH5α using a heat shock method at 42ºC for 50 s. The transformed E. coli was cultured on LB agar containing ampicillin (Sigma, Saint Louis USA) (50 μg.mL -1 ), 50 mg.mL -1 5-bromo-4-chloro-3-indoyl--D-galactopyranoside (X-Gal) (Promega, Madison, USA) and 100 mM isopropyl -D-1-thiogalacto pyranoside (IPTG) (Sigma -Aldrich, Taufkirchen, Germany) for 16 h. The positive transformants were screened from white colonies by PCR. The plasmid pGEMT-xylC was extracted via QIAprep Miniprep kit (Qiagen, Hilden, Germany) according to manufacturer's instructions and analyzed using 1% agarose gel. Subsequently, the size and accuracy of plasmid pGEMT-xylC was determined by a supercoiled ladder (Promega, Madison, USA) as standard and PCR. The xylC fragments from pGEMT-xylC and pET 28b (Novagen, Madison, USA) were excised using the NdeI and HindIII restriction enzymes and recovered from agarose gel. The purifi ed xylC was inserted into linearized pET 28b using T4 DNA ligase for 16 h at 16°C. The cloned was transformed into E. coli DH5α using heat shock method and the transformed cells were cultured onto LB agar containing kanamycin (50 g.mL -1 ). The recombinant plasmid pET28b-xylC was extracted from positive transformants and screened by PCR. The nucleotide sequences of the plasmids pGEMT-xylC and pET28b-xylC were determined by DNA sequencing using xylC, M13 and T7 primers (universal primers). The sequencing data were analyzed by VecScreen, BLASTP, and BLASTN (9).

Expression of BZDH
The plasmid pET28b-xylC was transformed into E. coli BL21 (DE3). A pre-culture from the transformant was prepared in LB broth and incubated at 37ºC to reach optical density OD 550 ~ 0.5. The standard inoculums (10%) were diluted to minimal salt medium (MSM) (10) induced by 0.5-2 mM benzaldehyde. The culture was incubated at 30°C, 150 rpm for 3 days and OD 550 was measured. The control was run in parallel condition with E. coli BL21 (DE3) without recombinant plasmid. The pre-culture was prepared by inoculating a single colony of E. coli BL21 (DE3) into 10 mL LB broth containing 50 g.mL -1 kanamycin and shaken at 37°C, 250 rpm for 16 h. The cells were centrifuged at 4ºC, 4000 rpm for 15 min and the supernatant was discarded. The resuspended pellet was diluted 5-fold (50 mL) and incubated to adjust an OD 550 0.6. Culture (1 mL) was collected as an uninduced sample (control) and the culture was induced by adding IPTG (0.01-1 mM) at 37°C after 1, 2, 4, 6 and 16 h of incubation. The harvested cells were dissolved in lysis buff er (50 mM NaH 2 PO 4 with 300 mM NaCl) (pH 8.0) and 1 μg.mL -1 lysozyme and 0.1 mM phenylmethyl sulfonyl fl uoride (PMSF) (Sigma-Aldrich, Taufkirchen, Germany) added to the mixture and incubated in ice for 30 min. The cells were disrupted by sonicator (Sonics-vibra cell, Ontario, Canada) at 20 s pulses with 5 min rest for 30 min. The crude lysate was centrifuged at 12000 ×g for 60 min at 4°C. The supernatant and the pellet were separated and 15 L of each sample loaded into 12% sodium dodecyl sulphate-polyacrylamide gel and run at 150 V. Expression was confi rmed through western blot when the protein was transferred from the gel onto a nitrocellulose membrane at a constant voltage of 15 V for 45 min using Trans-Blot SD semi-dry electrophoretic transfer cell (Bio-rad, California USA). The target protein was determined after reaction with a monoclonal antibody.

Purifi cation of BZDH
The BZDH was overexpressed in a 3L batch in optimal condition. The inclusion bodies were formed after high expression of target protein. Thus, some treatments such as use of lysozyme and sonication were applied to reduce the viscosity of the suspension followed by centrifugation at 12000 ×g for 60 min at 4ºC. The supernatant (25 mL) contained the desired protein and fi ltered with 0.22 m membrane fi lter before start of purifi cation. Purifi cation was carried out by AKTA prime (No 1314455 Sweden; GE Healthcare, Uppsala, Sweden) using ion exchange chromatography according to the manuals' instructions. In the fi rst step, machine and column were washed stepwise with buff er A (wash buff er) containing 20 mM Tris or bis-tris (pH 6-8) with fl ow rate of 1 mL.min -1 and fi ltered sample was injected into the Hi Trap DEAE column (i.d × 0.7 × 2.5 cm) with fl ow rate of 0.8 mL.min -1 The enzyme was eluted with a linear gradient of NaCl (1 M) in buff er B (elution buff er) containing buff er A with 1 M NaCl. The machine and column were re-equilibrated with buff er A in the last step. The collected fractions from bound and unbound proteins were analyzed using SDS-PAGE and western blot. The protein was concentrated by a vivaspin or Amicon (Millipore, Hannover, Germany) protein column to maximum volume of 3 mL and the concentration of purifi ed BZDH was measured using bicinchoninic acid (BCA) method at OD 562 (11). Protein degradation during purifi cation was reduced by adding dithiothreitol (DTT) or 2-mercaptoethanol (2-ME) into buff ers at a low concentration of 0.5 mM.

Characterization of Purifi ed BZDH
In BZDH reaction, nicotineamide adenine dinucleotide (β-NAD) was converted to a reduced form of NADH. All assay measurements were performed in triplicate.

Enzyme Assay
The mixture was prepared as follows: buff er (50 mM glycin-NaOH) 20 L; H 2 O 119.15 l and substrate (benzaldehyde) at fi nal concentration of 0.85 mM; -NAD (5 mM) 40 L. The reaction was started after that 20 L of purifi ed BZDH was added and the reaction was monitored for 10 min with 1 min intervals at 25°C at 340 nm. The activity of enzyme was calculated by general equation based on unit.mL -1 .
One-unit enzyme activity is the amount of enzyme that catalyzes the conversion of 1.0 M of substrate to the expected product per min at a standard assay condition (12).

Determination of Optimal pH and Temperature
The optimum pH was prepared in the standard assay conditions described before, except the following buff er systems from 6-11 were used: 50 mM K 2 HPO 4 (pH 6-8), 50 mM Tris buff er (pH 8-10), and 50 mM NaHCO 3 (pH 10-11) at 0.5 intervals. The mixture was incubated for 3 min at 25ºC and OD 340 was measured.
At optimal pH, the temperature was adjusted to 4, 20, 25, 37, 40, 50, 60, 70 and 80°C and enzyme assays were carried out. The reaction was stopped after 3 min and the absorbance was measured at OD 340 .

The Kinetic Study of the Enzyme
The eff ect of substrate concentration on enzyme activity was evaluated by maximum velocity (V max ) and Michaelis constant (K m ) by varying concentration of benzaldehyde at the range of 0.005-4 mM to a total adjusted volume of 200 L. The mixture incubated for 3 min at 25°C and OD 340 was measured. K m and V max were calculated from Lineweaver-Burk plots.

Gas Chromatography-Mass Spectrometry (GC-MS) Analysis
The mixture was pre-incubated for 5 min and reaction was stopped by adding 300 L of 0.1 M HCl. The protein was separated by Vivspin 500 (Sartorius, Gottingen, Germany) and 300 L diethyl ether was added to residual liquid. The upper layer (1 L) of volatile phase was injected into the injection port of the GC device.

Results
Rhodococcus ruber UKMP-5M was isolated from oilcontaminated soils in Malaysia. The xylC gene was amplifi ed at 64°C and the resulting product was 0.8 kb (Fig. 1A). The recombinant plasmid pGEMT-xylC was constructed, successfully transformed into E. coli DH5α and extracted from the positive transformants (Fig. 1B). The inserted fragment xylC was excised with double digestion using NdeI and HindIII (Fig. 1C). The xylC fragment was ligated into pET 28b at 15ºC to form the recombinant plasmid pET 28b-xylC, which was 6.2 kb, consisting of pET 28b (5.4 kb) and xylC (792 bp) (Fig. 1D). The highest growth of transformant E. coli BL21 containing pET 28b-xylC was determined when the cells incubated in 0.5 mM benzaldehyde for 24 h.
The transformed E. coli BL21 (DE3) was induced with 1 mM IPTG at 37ºC, which BZDH was successfully expressed for 2, 4, 6, and 16 h (Figs. 2A,  B). The expression reached its highest level at 4 h post induction. A residual amount of BZDH was shown in pellet after SDS-PAGE, because of inclusion bodies formation. Using higher concentration of lysozyme in lysis buff er, high speed and long-time centrifugation reduced the protein aggregation. A BZDH protein has an approximate molecular weight of 27 kDa. The purifi ed BZDH showed a single band on SDS-PAGE and the result was confi rmed by western blot (Figs.  2C, D). The total concentration of BZDH was 1.18  mg.mL -1 as determined by BCA method. The results of BZDH purifi cation from R. ruber UKMP-5M by anion exchange chromatography is summarized in Table 1, showing that the BZDH protein was purifi ed at 14 folds with 85% yield.
Optimum BZDH activity was at 9 min with the highest activity of 9.4 U.mL -1 . The eff ect of diff erent pHs on BZDH activity exhibited the highest level at pH 8.5 (Fig. 3A). The optimum temperature for BZDH was 25°C (Fig. 3B). Enzymatic activity decreased when the incubation temperature reached 50ºC (50% reduction in maximum activity). V max for the BZDH was 19.72 U.mL -1 and K m was 4.2 mM. The BZDH utilized benzaldehyde and the product was benzene carboxylic acid at a retention time of 12.5 min as determined by GC-MS.
The optimum temperature for BZDH activity from R. ruber UKMP-5M and P. stutzeri  was 25°C. The activity of BZDH was not stable for a long period and BZDH from R. ruber UKMP-5M, P. putida pWW0 MT53 and A. calcoaceticus (17) were losing 50% of enzyme activity at 50ºC within a period of one to 5 min.
The K m value of BZDH from R. ruber UKMP-5M (4.2 mM) was much higher to what was reported earlier. The K m value for BZDH was 460 μM for P. putida (16), 1.4 μM for P. putida CSV86 (2), 2.5 M for P. putida pWW0 (19), 0.63 M for A. calcoaceticus, 0.79 M for P. putida (17) and 7 M reported for P. stutzeri , indicate a diff erent variation in specifi city for BZDH, even within the same genus. The high K m value for BZDH from R. ruber UKMP-5M could be explained by a limited number of active sites when compared with other aldehyde dehydrogenases. As a result, the enzyme showed low affi nity for benzaldehyde as a substrate, which requires to have greater concentration of substrate to achieve V max and the enzyme activity was highly dependent on substrate. It is possible that BZDH from R. ruber UKMP-5M has a preference for other substituted of benzaldehyde than benzaldehyde, which also showed in other BZDHs (21). The BZDH from R. ruber UKMP-5M showed lower V max compared to similar BZDH in P. putida (104 U.mL -1 ), A. calcoaceticus (63.5 U.mL -1 ) (17) and 48 U.mL -1 for P. putida (19), which lead to the low rate of catalysis. The high K m value and low V max for BZDH from R. ruber UKMP-5M suggest that this enzyme may be active in high concentration of benzaldehyde, although it is slow in catalytic reaction and may be applicable for biodegradation in high contaminated area with hydrocarbons. The BZDH from R. ruber UKMP-5M is able to convert benzaldehyde to benzene derivatives. The products of BZDH in P. putida CSV86 was benzoic acid (22), benzoate and its derivatives (16,2) as determined by GC-MS.
The achievements of this paper show that benzaldehyde dehydrogenase is a NAD-dependent enzyme, important for hydrocarbon degradation through R. ruber UKMP-5M. The enzyme revealed similar characteristics to other aldehyde dehydrogenase even though it has smaller mass than others. However, it is apparent that benzaldehyde dehydrogenase has a catalytic mechanism diff ering from classical mechanisms, resulting in low affi nity and slow catalysis for benzaldehyde. In spite of this fact, it could be possible that the other enzymes of R. ruber UKMP-5M interfere in hydrocarbon biodegradation. The previous studies indicated that although aldehyde dehydrogenases are similar to each other in terms of many properties, they are diff erent with respect to features such as cofactor, substrate specifi ties, or genetic regulation. The results presented in this paper provide a starting point for a detailed molecular comparison of isolated BZDH R. ruber UKMP-5M with other BZDH.