Production and purification of a soluble hydrogenase from Ralstonia eutropha H16 for potential hydrogen fuel cell applications

Graphical abstract


G R A P H I C A L A B S T R A C T A B S T R A C T
The soluble hydrogenase (SH) from Ralstonia eutropha H16 is a promising candidate enzyme for H 2 -based biofuel application as it favours H 2 oxidation and is relatively oxygen-tolerant. In this report, bioprocess development studies undertaken to produce and purify an active SH are described, based on the methods previously reported [1][2][3][4]. Our modifications are: Upstream method optimizations were undertaken on heterotrophic growth media and cell lysis involving ultrasonication. Two anion exchangers (Q Sepharose and RESOURCE Q) and size exclusion chromatographic (Superdex 200) matrices were successfully employed for purification of a hexameric SH from R. eutropha. The H 2 oxidizing activity of the SH was demonstrated spectrophotometrically in solution and also immobilized on an EPG electrode using cyclic voltammetry. 1. Preparation of the bioreactor inoculum was initiated by picking a single colony of R. eutropha H16 from a FGN agar plate and using this to inoculate 5 mL FGN media in a 15 mL centrifuge (Falcon) tube. Cells were incubated at 30 C on a shaker overnight. The overnight inoculum was transferred to a 1 L sterile baffled shake flask containing 100 mL FGN media. The 100 mL pre-culture was grown overnight to serve as the inoculum for the bioreactor fermentation. The batch fermentation was undertaken in a laboratory scale glass bioreactor with a working volume of 5 L. Growth was monitored by measuring the OD 600nm of the fermentation broth. 2. The bioreactor was operated at 30 C with an agitation speed of 300-350 rpm and an air flow rate of 1-2 L/min. The initial pH of the culture was 7.0-7.1 and pH was not allowed to drop below 6.4 via automatic addition of 1 M NaOH when the pH reached this setpoint. For obtaining the highest SH activity, FeCl 3 and NiCl 2 were added at 15 h to a final concentration of 10 mM and 1 mM, respectively.
3. After 48 h, cells were harvested by centrifugation (10,000 g at 4 C for 10 min) followed by washing in an appropriate volume of 50 mM KPi buffer (pH 7.0). The cell pellets were stored at À80 C. 4. The cell pellet was thawed and resuspended in 50 mM KPi buffer containing cOmplete, EDTA-free Protease inhibitor and DNase I (final concentration 20 mg/ml). A ratio of cell wet weight and the resuspension buffer of 1:5 was used. The cell suspension was sonicated using a Branson Digital Sonifier equipped with 1/8 Tapered Microtip by applying 50% amplitude. Various duty cycles for finding the optimal cell disruption condition were investigated. Cell debris was removed by centrifugation at 20,000 g for 50 min at 4 C.
Step 2: soluble hydrogenase purification and identification The purification process described here for SH was designed using modern chromatography matrices not available when the original methods were described. The performance of each of the previously employed matrices (DEAE Sepharose, Phenyl Sepharose and Superdex 200) [3,6,7] and matrices not previously described for this process (Q Sepharose and RESOURCE Q) were optimized with respect to resolution and recovery. For example, the performance of Phenyl HP was investigated applying conditions as previously published [6]: a two-step gradient of 200-50 mM KPi and 10-0 mM KPi was applied for four and three column volumes, respectively, and the enzyme was expected to be eluted during the latter gradient. However, the SH failed to bind to the column (data not shown). The newly developed downstream process involved a traditional ammonium sulphate precipitation followed by two consecutive ion exchange steps (employing Q Sepharose and RESOURCE Q) and a final size exclusion chromatography step (Superdex 200). The recruitment of the second anion exchanger RESOURCE Q column following the first anion exchanger Q Sepharose resulted in significantly improved enzyme purity. The newly developed bioprocess steps designed in this study resulted in an 18.7% yield and a 13.1 fold purification to ultimately obtain the pure active SH preparation. All purification steps were performed at 4 C under aerobic conditions.

Materials
Saturated ammonium sulphate solution ((NH 4 ) 2 SO 4 ) for ammonium sulphate fractionation KPi 50 mM buffer with or without 1 mM EDTA (pH 7.0) Slide-A-Lyzer Dialysis cassettes 10 K MWCO (Thermo Scientific, USA) or SnakeSkin Dialysis Tubing, 10 K MWCO (Thermo Scientific, USA) for protein dialysis Unless otherwise stated, all chromatographic media and systems were obtained from GE Healthcare. All liquid chromatographic steps were performed at 4 C on a calibrated ÄKTAexplorer TM system, controlled by UNICORN TM software. All chromatography buffers were filtered through 0.22 mm filter membranes (Millipore, USA) and degassed under vacuum. The protein samples were routinely filtered through 0.22 mm filter units (Millipore, USA) prior to loading sample on to the columns. The columns used were a pre-packed HiTrap Q Sepharose FF 1 mL column or a manually packed Q Sepharose FF 7.8 mL column, RESOURCE Q 1 mL column and Superdex 200 10/300 GL (analytical grade) column. Vivaspin 1 20 10 K MWCO centrifugal concentrator (GE Healthcare, Sweden) Protein concentrations were routinely determined according to the BCA (bicinchoninic acid) method using the Pierce 1 BCA Protein Assay Kit (Thermo Fisher Scientific, USA) [8].
SDS-PAGE was performed using NuPAGE 1 Novex 1 4-12% Bis-Tris (Life Technologies, USA) pre-cast polyacrylamide gels. SeeBlue 1 Plus2 Pre-Stained Standard (Life Technologies, USA) was used as protein standard. Protein gels were stained with either the GelCode Blue Stain Reagent (based on Coomassie dye G-250, (Thermo Fisher Scientific, USA)) or the Silver Stain Plus kit (Bio-Rad, Australia) according to the manufacturer's instructions. Access to mass spectrometric analysis for protein identification Procedure 1. The ammonium sulphate fractionation was performed as described previously [6]. Briefly, the cellfree extract was fractionated by addition of saturated ammonium sulphate solution to 35% saturation. After incubation on ice and centrifugation at 10,000 g, the resulting supernatant was brought to 60% ammonium sulphate saturation.  (Fig. 1A). After a complete capture of all SH containing fractions in the first peak after elution start, two other peaks containing contaminants were eluted, and the activity assays for the fractions collected from each peak were undertaken. The active fractions were pooled, and dialyzed against 2 L 50 mM KPi buffer (pH 7.0) containing 1 mM EDTA at 4 C. 4. Second anion exchange chromatography: For further purification of the protein, the dialyzed protein solution exhibiting SH activity eluted from Q Sepharose FF was loaded on to a RESOURCE Q 1 mL column pre-packed with SOURCE TM 15Q medium. This resin is comprised of smaller monodispersed beads ($15 mm) compared to those in Q Sepharose FF (45-165 mm), which generally results in improved resolution. The eluent buffers, the flow rate and fraction sizes were equivalent to those used in the Q Sepharose FF anion exchange columns. Based on the results from the previous linear gradient run, the elution gradient was narrowed to 20%B (0-200 mM KCl). During a linear gradient of 0-20%B, the second peak spike of the first peak contains the highest SH activity (Fig. 1B). 5. Size exclusion chromatography: The combined active fractions generated after the second anion exchange chromatography column step, RESOURCE Q, were concentrated by membrane ultrafiltration in a Vivaspin 1 20 10 K MWCO centrifugal concentrator to a final volume of approximately 0.5 mL. The protein solution was then loaded on to a Superdex 200 10/300 GL (analytical grade) column using a 500 mL sample loop. Having considered that the recommended ionic strength of the buffer is !20 mS/cm according to the manufacturer's instruction, the performance of the gel filtration with a higher ionic strength buffer, 50 mM KPi containing 150 mM NaCl (with 20 mS/cm conductivity), compared to previously used 50 mM KPi only buffer (with 7 mS/cm conductivity) [3,6] was evaluated to avoid non-specific binding. Fractions were manually collected at a flow rate of 0.5 mL/min. The SH active fractions were combined and re-concentrated (Fig. 1C). Prior to storing at À80 C, the SH solution was headspace flushed with nitrogen. 6. Mass spectrometric analysis was carried out at the Bioanalytical Mass Spectrometry Facility, University of New South Wales, Australia. The five protein bands that represent the expected six subunits of the hexameric SH after Superdex 200 were excised from the Commassie Blue stained gel shown in Fig. 2    Step 3: soluble hydrogenase activity in solution and immobilized form An uptake hydrogenase activity (H 2 oxidation) assay was developed based on previously described methods [2,7,10]. Further electrochemical studies were performed using cyclic voltammetry (CV) with the enzyme immobilized on modified EPG (edge plane pyrolytic graphite) electrodes. Differently modified electrodes with the SH protein film applied were employed to study H 2 oxidizing activity. The preliminary results demonstrated that the enzyme exhibits H 2 oxidation activity at the surface of electrode and therefore this enzyme is a promising candidate for biochemical oxidation in hydrogen biofuel cell research.

Materials
Cary 100 UV-vis Spectrophotometer with Temperature Controller (Varian, Australia) at 30 C Septum-sealed special optical glass cuvettes (Starna, Australia) 50 mM H 2 -saturated Tris/HCl buffer (pH 8.0) NAD+ solution (1 mM final concentration) Argon and H 2 gas cylinders EPG electrodes Polymyxin for the electrode modification A platinum wire and Ag/AgCl as a control and a reference electrode, respectively MPG-2 potentiostat supplied with EC-Lab 1 software (Bio-Logic, USA) Procedure 1. In-solution enzyme activity assay: The hydrogen-oxidizing activity of a SH preparation was measured anaerobically with NAD + as an electron acceptor on a Cary 100 UV-vis Spectrophotometer with Temperature Controller at 30 C (NAD + ). The SH solution, 100 mL, was added to 2.9 mL reaction mixture containing 50 mM H 2 -saturated Tris/HCl buffer (pH 8.0) and 1 mM NAD + after 10 min Argon and 10 min H 2 sparging to a septum-sealed special optical glass cuvette. For all gas sparging (both argon and H 2 ) a pressure of 20 psi and flowrate of 4-5 mL/min was applied. The mixture was then pre-incubated at 30 C (NAD + ) for 5 min. NADH formation (e = 6.22 mM À1 cm À1 ) was monitored with continual incubation for 10 min at 0.5 min intervals at 340 nm using the Cary WinUV software. The unit of enzyme activity is defined as the reduction of 1 mmol NAD + per min. The absorbance change at 340 nm per minute (dA 340nm /min) is calculated using the data obtained from the slope of a linear absorbance versus time plot. 2. Immobilized enzyme activity assay: Electrochemical analyses of the SH preparations purified in the present study were carried out by the method of CV [11]. The H 2 oxidation activity of the enzyme preparations were investigated by forming a protein film on modified EPG electrodes. Polymyxin was used to immobilize the SH on an EPG electrode. In all electrochemical experiments, 50 mM KPi buffer at pH 7.0 was used. A platinum wire and Ag/AgCl were employed as a control and a reference electrode, respectively, and potentials were normalised vs. NHE (Normal Hydrogen Electrode). Potentials were controlled using the MPG-2 potentiostat with the scan rate of 50 mV s À1 . Fig. 3 shows cyclic voltammograms recorded in H 2 -equilibrated 50 mM KPi buffer (pH 7.0) for a film of the SH immobilized on EPG electrodes. H 2 oxidation activity was examined on the NADH activated electrodes in the presence and absence of polymyxin. 5 mL of activated hydrogenase did not show any activity in the absence of polymyxin, however, it displayed H 2 oxidizing activity in the presence of polymyxin when an external potential was applied (Fig. 3A). The dependence of H 2 oxidation activity of the SH (5 mL) on incubation time of the reaction at the polymyxin-coated electrode surface was also analyzed by applying a potential of À550 mV (vs. NHE). The hydrogenase activity was observed to increase as the incubation time was prolonged (Fig. 3B). These findings confirm that the enzyme preparation purified in this study is capable of oxidizing H 2 when immobilized on this electrode.

Acknowledgements
MethodsX thanks the reviewers (anonymous) of this article for taking the time to provide valuable feedback. The authors wish to acknowledge AusAID for providing scholarship support for BEJ. Polymyxin with SH. Both electrodes were activated by NADH. 5 mL of activated SH only (a in blue) and mixture of 3 mL of activated hydrogenase and 2 mL of 2% of Polymyxin (b in red) are shown. B) H 2 oxidation following different incubation times at an applied potential of À550 mV (vs. NHE). 5 mL of SH was applied on an EPG electrode. 5 min (a in pink), 10 min (b in green) and 30 min (c in blue). All cyclic voltametric experiments were carried out in 50 mM KPi buffer (pH 7.0) at a scan rate 50 mV s À1 . I À Current, E À potential, NHE À Normal Hydrogen Electrode. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)