Here we present a protocol for anaerobic protein purification, anaerobic protein concentration, and subsequent kinetic characterization using an oxygen electrode system. The method is illustrated using the enzyme DesB, a dioxygenase enzyme which is more stable and active when purified and stored in an anaerobic environment.
Oxygen-sensitive proteins, including those enzymes which utilize oxygen as a substrate, can have reduced stability when purified using traditional aerobic purification methods. This manuscript illustrates the technical details involved in the anaerobic purification process, including the preparation of buffers and reagents, the methods for column chromatography in a glove box, and the desalting of the protein prior to kinetics. Also described are the methods for preparing and using an oxygen electrode to perform kinetic characterization of an oxygen-utilizing enzyme. These methods are illustrated using the dioxygenase enzyme DesB, a gallate dioxygenase from the bacterium Sphingobium sp. strain SYK-6.
Enzymes that utilize iron or other metals to activate oxygen are often susceptible to inactivation during the purification process because of their removal from the reducing environment of a cell. Therefore, these proteins must be used as cell lysates, be subjected to external reducing agents, or be purified anaerobically to ensure that they have optimal enzymatic activity1,2,3,4. For those enzymes that are oxygen-sensitive (specifically iron-containing enzymes), performing all the purification and characterization steps while maintaining anaerobic conditions is necessary to fully characterize them. This has led researchers to develop entire laboratory set-ups within the confines of anaerobic chambers for studies ranging from protein expression through crystallography5,6,7,8.
Herein, we report methods for the anaerobic purification and kinetic characterization of the enzyme DesB using an oxygen electrode system. DesB is a gallate dioxygenase from the bacterium Sphingobium sp. strain SYK-6 that is related to LigAB, a protocatecuate dioxygenase from the same organism. Both enzymes belong to the type II protocatechuate dioxygenase (PCAD) superfamily which has not been extensively studied to date9, likely in part due to enzymes of this superfamily being susceptible to inactivation when purified using standard aerobic protein purification methods. Since some of the PCAD enzymes display substrate promiscuity while others are substrate-specific2,10, further characterization of this superfamily is necessary to identify specificity determinants. As has been observed in several enzyme superfamilies11,12,13,14,15, small molecules can alter activity via direct competitive inhibition or the binding of molecules to separate allosteric pockets which causes an increase or decrease in enzymatic activity16. While kinetics alone cannot differentiate the binding location of a modulator, determining the magnitude of an activity change is important for understanding the effects. As such, methods for kinetic characterization of native DesB activity and its activity in the presence of 4-nitrocatechol (4NC), a compound commonly used to characterize and inhibit dioxygenase enzymes2,17,18, are shown.
DesB is able to break down gallate, a lignin-derived aromatic compound, via an extradiol dioxygenase (EDO) reaction in which ring opening is catalyzed using oxygen as one of the substrates10,19. This enzymatic reaction occurs within the context of the breakdown of lignin, an aromatic heteropolymer found in the cell wall of plants. Lignin can be depolymerized, yielding a variety of aromatic compounds that can be further broken down into central metabolites3,20,21,22,23,24,25,26,27,28,29,30,31,32,33. Extradiol dioxygenases (EDO) catalyze a ring opening reaction on dihydroxylated aromatic compounds, where cleavage occurs adjacent to a metal-coordinated diol; in contrast, intradiol dioxygenases cleave analogous aromatic compounds between the two hydroxyl groups (Figure 1). EDOs, like many other metalloenzymes, have a divalent metal center for coordinating Fe(II) composed of a two-histidine, one-carboxylate triad9,34,35. These metalloenzymes become oxidized, through either autoxidation or mechanism-based inactivation, whereas the enzyme is rendered inactive2,36,37,38.
In the experimental procedures described in this manuscript, we utilize DesB, a member of the PCAD superfamily from the bacterium Sphingobium sp. SYK-6, to catalyze the addition of oxygen across the C4-C5 bond of gallate (Figure 2A). The regiochemistry of this cleavage is analogous to LigAB, which is a protocatechuate-4,5-dioxygenase (Figure 2B). Thus far, investigations of this gallate dioxygenase include no reports of compounds that inhibit DesB10,19,39. With the use of aerobic purification methods, DesB exhibited variable activity, while with the use of anaerobic methods we were able to consistently obtain protein with reproducible activity. The kinetic studies described here show the methods for anaerobic purification of DesB, kinetic characterization of the reaction of DesB with gallate, and the inhibition of DesB by 4-nitrocatechol (4NC).
1. General Materials and Methods
2. Preparation of the Amylose Column for Protein Purification
3. Protein Expression and Anaerobic Purification of DesB
NOTE: The DesB gene was commercially synthesized, having been placed into pET-15b, pET-32a, and pMAL- c5x vectors using the NdeI and BamHI restriction cloning sites.
4. Anaerobic Protein Concentration/Buffer Exchange
5. Desalting DesB
6. Preparing the Oxygen Electrode
7. Kinetic Assays Using the Oxygen Electrode
Shown is the SDS-PAGE gel analysis of individual fractions from purification of the DesB-maltose binding protein (MBP) fusion construct (Figure 3). The gel reveals that the protein is pure (MW = 91.22 kDa), except for the presence of DesB (MW = 49.22 kDa) and MBP protein domain (42 kDa) cleaved from each other. Fractions E2 and E3 were selected for concentration (step 4.2).
Reproducible results from DesB kinetic assays depend on correct assembly, calibration, and experimental technique. It is necessary to input the correct ambient temperature and pressure, as these variables determine the percentage of O2 expected to be dissolved in solutions during the assay (Figure 4A). The initial signal should be between 1800 and 2000 nmol/mL and should produce a stable rate close to zero, indicating that the oxygen saturation of the solution is minimally changing (Figure 4B). Once the sodium dithionate is added and the chamber is sealed, the rate of O2 consumption should rapidly and steadily increase until there is minimal O2 (<60 nmol/mL) left in the solution. A steady negative slope indicates that 1) the assembled electrode has no air leaks and 2) the electrode is adequately responding to changes in O2 concentration (Figure 4C and D). If the O2 remaining in solution is not sufficiently low, the calibration offset value will be incorrect. The calibration will have to be repeated with the electrode assembled correctly, and fresh sodium dithionite must be used.
When the electrode is correctly calibrated, kinetic assays can be performed. The first measurement establishes the activity of the freshly desalted enzyme using 100 µM gallate in 25 mM Tris buffer and 1 uL of enzyme. Before addition of the enzyme, the Tris buffer and gallate are stirred in the chamber with the plunger in position. This should produce an approximate initial signal between 250 and 350 nmol/mL, and the signal should stabilize (±5 nmol/mL/min) before the enzyme is added. A stable signal indicates that there are no variables (e.g., torn membrane, leftover sodium dithionite, dirty electrode) that may cause inconsistent measurements. When enzyme is added, the signal should rapidly and steadily decrease, indicating that the reaction of DesB is proceeding, since O2 is consumed in the reaction with gallate. The slope of the initial rate before enzyme addition and the catalytic rate were determined by using the rate tools and adjusting the window to 30 seconds (Figure 5A). After approximately 1-1.5 hours of use, the enzyme activity decreases by about 30-50%, at which point it should no longer be used (Figure 5B).
Once all kinetic measurements have been taken, the data can be plotted using a modeling program (Figure 6). The activity of DesB with gallate is obtained by measuring the rate of O2 consumption in varying gallate concentrations. The background rate before enzyme addition is subtracted from the catalytic rate to obtain the corrected rate. The background corrected rate is divided by the enzyme concentration and fitted to equation 7.3A (See step 7.3.3). A sufficient fit is dependent on the initial parameters for KM and Ksi, (initial parameters:KM = 320 μM, Ksi = 1600 μM). The result shows a hyperbolic curve that reaches a maximum plateau, then slopes downward slightly as [gallate] increases. This is a typical curve for an enzyme that exhibits substrate inhibition at high substrate concentrations.
The rate of inhibition of DesB with 4-nitrocatechol was determined by observing the reduction in oxygen consumption rate of DesB with 1 mM gallate in the presence of varying concentrations of 4-nitrocatechol (Figure 7). The 4-NC concentration was plotted against the normalized activity (the rate in the presence of inhibitor was divided by the uninhibited rate) and fit to equation 7.3B (see step 7.3.3). The results indicate that 4-NC inhibits the consumption of oxygen, thus inhibiting the DesB reaction.
Media Type | Components |
Super Optimal broth with Catabolite repression (SOC media) | 2% (w/v) tryptone |
0.5% (w/v) yeast extract | |
10 mM NaCl | |
2.5 mM KCl | |
10 mM MgCl2 (added after sterilization) | |
20 mM glucose (added after sterilization) | |
Miller’s Lysogeny Broth with Ampicillin (LB-Amp media) | 0.5% (w/v) yeast extract |
1% (w/v) tryptone | |
1% (w/v) NaCl | |
0.1 mg/mL Ampicillin (added after sterilization) |
Table 1. Ingredients for preparation of super optimal broth with catabolite repression (SOC media) and Miller's lysogeny broth with ampicillin (LB-Amp media).
Working Solution | Component final concentrations |
Laemelli Buffer, pH 6.8 | 100 mM tris(hydroxymethyl)aminomethane (Tris) |
200 mM dithiothreitol (DTT), | |
4% sodium dodecyl sulfate (SDS) | |
0.05% bromophenol blue | |
20% glycerol | |
Running Buffer pH 8.3 | 25 mM Tris |
192 mM glycine | |
0.1% SDS |
Table 2. Ingredients for preparation of Laemelli and running buffers for SDS-PAGE analysis.
Working Solution | Component final concentrations |
Amylose Column Buffer, pH 7.4 | 20 mM tris(hydroxymethyl)aminomethane (Tris) |
1 mM ethylenediaminetetraacetic acid (EDTA) | |
200 mM NaCl | |
Elution Buffer pH 7.4 | 20 mM Tris |
1 mM EDTA | |
200 mM NaCl | |
10 mM maltose | |
Exchange Buffer, pH 7.5 | 50 mM Tris |
10% glycerol | |
Desalt Buffer, pH 7.5 | 50 mM Tris |
10% t-butanol |
Table 3. Ingredients for preparation of buffers for protein purification.
Figure 1: Comparison of the reactions catalyzed by ring cleaving dioxygenases. The wavy lines show the corresponding carbon-carbon bond that is cleaved during the dioxygenase reaction, where intradiol cleavage (red) breaks the bond between the two hydroxyl groups and extradiol cleavage (blue) breaks a carbon-carbon bond adjacent to the hydroxyl groups. Please click here to view a larger version of this figure.
Figure 2: The reactions catalyzed by DesB and LigAB, two related extradiol dioxygenases which belong to the type II protocatechuate dioxygenase (PCAD) superfamily. Please click here to view a larger version of this figure.
Figure 3: SDS-PAGE gel of DesB showing purification and concentration steps results. Lanes are labeled as follows: S = protein standard, FT1 and FT2 = the collected flow-through fractions, W1 and W2 = the collected wash fractions, E1-E4 = the collected elute fractions. Please click here to view a larger version of this figure.
Figure 4: Generation of calibration curves for the oxygen electrode. (A) Prior to calibration, ambient temperature and pressure are entered into the program to determine oxygen concentrations for air-equilibrated solutions. (B) The oxygen saturated solution reaches equilibrium and generates a prompt. (C) Sodium dithionate solution is added to utilize all the oxygen, as indicated by the decrease in oxygen signal. (D) Calibration is obtained when the signal stabilizes at zero oxygen concentration. Please click here to view a larger version of this figure.
Figure 5: Representative kinetic curves illustrating a typical loss of activity over time. (A) The reaction of 100 µm gallate using fresh DesB, with an O2 utilization rate of -56.61 µM/min. (B) The reaction of 100 µm gallate using DesB after 2 h of data collection, with an O2 utilization rate of -29.56 µM/min. Please click here to view a larger version of this figure.
Figure 6: Plot of DesB activity with gallate, fit to the Michaelis-Menton equation (black), and fit to the Haldane equation for substrate inhibition (red; Equation 7.3B). Kinetic parameters derived from the two fits are as follows: Michaelis-Menten – kcat = 316 +/- 17 sec-1, Km = 123 +/-26 μM; Haldane – kcat = 570 +/- 110 sec-1, Km = 320 +/-100 μM, Ksi = 1600 +/-600 μM. Please click here to view a larger version of this figure.
Figure 7: Plot of 4-nitrocatechol inhibition of DesB dioxygenation of gallate (1 mM). Please click here to view a larger version of this figure.
Figure 8: Coordination of gallate by DesB (pdb ID = 3wr3). As seen in many dioxygenases, the ligand coordinates to an active site iron(II) atom in a bidentate fashion. Based on the structural similarities between gallate and 4-nitrocatechol (4NC), the inhibition of DesB by 4NC was predicted. Please click here to view a larger version of this figure.
The critical steps in obtaining active, purified DesB protein involve the forming and maintaining of the reduced Fe(II) active site in the enzyme. As such, correct performance of the induction, purification, concentration, and desalting steps are essential to successfully obtaining active enzyme. Inducing protein expression in the presence of 1 mM ferrous ammonium sulfate ensures that Fe(II) is correctly incorporated into the active site of DesB. This method is inspired by studies like those with amidohydrolase metalloenzymes, which often require the addition of metal to growth media to allow proper folding and full occupancy of the metal binding site6,41,42,43.
The anaerobic purification and concentration in the glove box are perhaps the most critical and technically difficult steps in this protocol. It is essential to maintain an oxygen-free atmosphere in the glove box. This requires maintaining the deoxygenation catalyst in the glovebox as prescribed by the manual, periodically changing the nitrogen tanks that are attached to the box when they are low, ensuring that there are no holes in the gloves attached to the glovebox, and keeping a tray of fresh desiccant in the glovebox. Oxygen-sensitive strips that change color when exposed to ambient O2 may be placed in the box to test for an O2-free atmosphere. Furthermore, it is necessary to fully degas all buffers and solutions that will be used during the purification and concentration process. To ensure minimal O2 in the buffers, take great care to limit exposure of the degassed buffers to air when switching between nitrogen bubbling and vacuuming cycles.
When running the column in the glove box, a good test to determine whether there is oxygen present in the buffers is to take a small portion of one of the wash fractions (approximately 0.1-0.5 mL) and place it in a microcentrifuge tube with a small portion of ferrous ammonium sulfate and DTT (less than the amount necessary for the concentration step). It is then important to mix the contents of the tube well and observe the color change. If the solution turns black, dark yellow, or orange, there is oxygen present in the protein fractions, and there will likely be reduced catalytic activity after the complete purification and concentration process. If the solution is a light lavender or very light-yellow color, there may be minimal O2 present, but it is likely that the enzyme will still have high activity. The dark yellow to orange color change when oxygen is present is likely caused by oxidation of the iron in the ferrous ammonium sulfate to Fe(III), forming rust44. To fix this issue, it is recommended to degas the buffers and allow nitrogen to bubble through crude the protein solution for 1-2 extra minutes before placing them in the glove box. Also, it is important to ensure that the atmosphere in the glove box is truly O2-free by using O2-sensitive strips.
The last critical step, desalting the enzyme prior to kinetic characterization, requires an oxygen-free atmosphere and must be done on ice, as the purified protein is prone to denaturation at room temperature. Determining when the desalted protein comes off the column is essential to successful kinetic assays, as the most concentrated fractions give the most reproducible results. The fraction where the desalted protein is eluted must be determined every time a new batch of protein is purified and when the column is repacked with new resin. If activity is low during kinetic assays and the purification and concentration steps have been technically mastered, the issue may lie in the desalting step. When troubleshooting this step, it is crucial to check that the 50 mM Tris/10% t-butanol buffer is thoroughly degassed, repack the column with fresh Sephadex resin, and ensure that there are no holes in the glove bag.
After DesB has been successfully purified and desalted, kinetic assays using the oxygen electrode must be performed carefully to obtain data on catalytic activity of the enzyme. Kinetic measurements are typically reproducible when a catalytically active and concentrated protein is used. If data points are not reproducible when repeating a run for a substrate or inhibitor concentration data point, the issue may be that the protein has denatured or oxidized after extended use. Freshly desalted protein can be generated to allow continuation of data collection. It is important to retain all vials of the enzyme, so the exact protein concentration can be determined after completion of the data collection using a Bradford assay. This step is performed after the kinetics measurements because the enzyme loses activity over time, so performing it first may lead to lower activity and inaccurate kinetics measurements. The protein concentration is then used to convert observed catalytic rates into the reaction rates that are needed to determine the turnover number. In addition to concerns about the loss of enzyme activity leading to irreproducible kinetics results, degradation of the membrane covering the electrode after extended use may also cause challenges when reproducing data. Membrane degradation is typically indicated by an increase in the initial signal (from 250-350 nmol/mL to >350 nmol/mL) or an inability to attain a stable background rate before enzyme addition (>±5 nmol/mL/min). If either is observed, it is recommended to disassemble the electrode, clean any oxides off the electrode using the supplied cleaning powder, reassemble the electrode, and recalibrate.
The method of anaerobic purification is very important for enzymes with a metal center that can be oxidized, especially for those that have tightly bound metals which cannot be exchanged after the protein folds. Although enzymes have evolved to protect themselves by coordinating oxygen only after substrate coordination, they have done so in a cellular environment – the saturating amounts of oxygen in an in vitro environment can lead to rapid oxidation of metals and the conversion of Fe(II) into the inactive Fe(III) form31. This oxidation/inactivation can lead to skewed results in which the enzyme is not in its catalytically active state. The kinetic assays using an oxygen-sensitive electrode can be applied to enzymes that rely on oxygen as a substrate. Rather than obtaining kinetics parameters using the change in intensity of substrate or product absorption, this method allows for the visualization of oxygen consumption saturated in solution. This method has been used previously with LigAB, another extradiol dioxygenase in the protocatechuate dioxygenase superfamily that similarly relies on a Fe(II) in its active site to coordinate and cleave its substrates.
This manuscript also provides additional information about the enzyme DesB from Sphingobium sp. strain SYK-6. Following the work that defined the enzymatic function and structure of DesB10,19,39, it was determined herein that DesB is a competent catalyst of the dioxygenation of gallate, with kcat of 17.8 ± 1.0 s-1 and Km of 45 ± 13 µM, resulting in kcat/Km of 3.98 x 105 M/s. These rates are comparable to those determined for other dioxygenase enzymes, including LigAB (kcat of 51 s-1 and kcat/KM of 4.26 x 106 M-1s-1) and other dioxygenases which have kcat/Km values ranging from 105-108 M-1s-1 36,45,46,47,48,49,50.
The DesB active site was previously shown to be at the dimer interface, with residues from both monomers contributing to the coordination of substrate [residues originating from the monomer that binds the Fe(II) are indicated by their residue number, while residues from the other monomer are indicated by their residue number and a prime (i.e., Glu377ʹ)]6. In the absence of structural information showing the binding interactions of 4NC with DesB, the structural similarities and differences between gallate and 4NC can provide insight into how DesB might be inhibited by 4NC. Gallate has three hydroxyl substituents at C3, C4, and C5, with its C3 and C4 hydroxyls being coordinated to the Fe(II) center, and the C5 hydroxyl being coordinated by Glu377ʹ (Figure 8). The carboxylic acid group at C1 is coordinated by Tyr-391ʹ, Tyr-412ʹ, Thr-13, and Thr-267 in the DesB active site. 4NC, which proved to be a modest inhibitor of DesB, has two hydroxyls available to coordinate the Fe(II) center and one C1 nitro group that is isosteric to a carboxylate (while also having two oxygens for coordination by residues 391, 412, 13, and 267), but is much more electron-withdrawing than the carboxylic acid on gallate. Since 4NC displayed only 36.6% inhibition of the DesB dioxygenation of gallate when the inhibitor was present in 5-fold excess over substrate, it is unsurprising that it was not a very potent inhibitor (with a Ki of 2.3 ± 0.3 mM). This suggests that the C5 hydroxyl and C1 substituent play a significant role in promoting the enzyme-ligand complex. Since residues Glu377ʹ, Tyr-391ʹ, and Tyr-412ʹ are all implicated in these interactions, this suggests that DesB active site contacts with adjacent monomers are important for the placement of a compound and structuring the active site.
The authors have nothing to disclose.
We would like to thank Dr. Camille Keller of Wesleyan University for technical support. Special thanks to Professor Lindsay D. Eltis and Jenna K. Capyk from the University of British Columbia, as well as Christian Whitman from the University of Texas at Austin, for their advice regarding anaerobic protein purification methods and the use of an O2-sensitive electrode.
Isopropyl β-D-1-thiogalactopyranodise | Gold Bio Technologies | I2481C50 | |
Coomassie Brilliant Blue R-250 | Bio-Rad | 161-0400 | |
Ammonium persulfate | Bio-Rad | 161-0700 | |
30% Acrylamide | Bio-Rad | 161-0158 | |
N,N'tetramethyl-ethylenediamine | Bio-Rad | 161-0801 | |
Amylose Resin High Flow | New England Biolabs | E8022S | |
BL21 (DE3) competent Escherichia coli cells | New England Biolabs | C2527I | |
L-cysteine | Sigma Aldrich | C7352 | |
gallic acid | Sigma Aldrich | G7384 | |
4-nitrocatechol | Sigma Aldrich | N15553 | |
Ferrous ammonium sulfate | Mallinckrodt | 5064 | |
Sodium dithionite | Alfa Aesar | 33381-22 | |
wheaton serum bottles | Fisher Scientific | 06-406G | |
25 mm Acrodisc PF Syringe Filter with Supor Membrane | Pall Corportation | 4187 | |
400 mL Amicon Stirred Cell Concentrator | EMD Millipore | UFSC40001 | |
76 mm Millipore Ultracel 10 kDa cutoff reconsituted cellulose membrane filter | EMD Millipore | PLGC07610 | |
DL-dithiothreitol | Gold Bio Technologies | DTT50 | |
Sephadex G-25 coarse desalting gal column | GE Healthcare | 17-0033-01 | |
2 mL Crimp-Top Vials | Fisher Scientific | 03-391-38 | |
Oxygraph Plus Electrode Control Unit | Hansatech Instruments | OXYG1 Plus | |
Oxygen Eletrode Chamber | Hansatech Instruments | DW1 | |
Electrode Disc | Hansatech Instruments | S1 | |
PTFE (0.0125 mmX25mm) 30m reel | Hansatech Instruments | S4 | |
Electrode cleaning Kit | Hansatech Instruments | S16 | |
Spacer paper | Zig Zag | available at any gas station | |
He-series Dri-Lab glove box | Vacuum/Atmospheres Company | ||
HE-493 Dri-Train | Vacuum/Atmospheres Company | ||
Double-Ended Micro-Tapered Stainless Steel Spatula | Fisher Scientific | 21-401-10 | |
DWK Life Sciences Kimble Kontes Flex Column Economy Column | Fisher Scientific | k420400-1530 | |
10 μL, Model 701 N SYR, Cemented NDL 26s ga, 2 in, point stlye 2 syringe | Hamilton | 80300 | |
DWK Life Sciences Kimble Kontes Flex Column Economy Column | Fisher Scientific | K420401-1505 | |
Emulsiflex-C5 high-pressure homogenizer | Avestin | ||
B-PER Complete Bacterial Protein Extraction Reagent | Thermo Fisher Scientific | 89821 | |
Lysozyme from chicken egg white | Sigma Aldrich | 12650-88-3 | |
Sodium dodecyl sulfate | Thermo Fisher Scientific | 151-21-3 | |
ampicillin | Sigma Aldrich | 7177-48-2 | |
Tryptone | Fisher Scientific | BP-1421-500 | |
Yeast extract | Fisher Scientific | BP1422-2 | |
Sodium Chloride | Fisher Scientific | S271-10 | |
Potassium Chloride | Fisher Scientific | P217-3 | |
Magnesium Chloride | Fisher Scientific | M33-500 | |
Dextrose | Fisher Scientific | D16-3 | |
Sodium Hydroxide | Fisher Scientific | S318-1 | |
Tris hydrochloride | Fisher Scientific | BP153-500 | |
Maltose | Fisher Scientific | BP684-500 | |
Glycine | Fisher Scientific | G46-500 |