Characterization of Alkyl-Nickel Adducts Generated by Reaction of Methyl-Coenzyme M Reductase with Brominated Acids

Abstract

Methyl-coenzyme M reductase (MCR) from methanogenic archaea catalyzes the final step in the biological synthesis of methane. Using coenzyme B (CoBSH) as the two-electron donor, MCR reduces methyl-coenzyme M (methyl-SCoM) to methane and the mixed disulfide, CoB-S-S-CoM. MCR contains Coenzyme F₄₃₀, an essential redox-active nickel tetrahydrocorphin at its active site. The active form of MCR (MCR_red1) contains Ni(I)-F₄₃₀. When 3-bromopropane sulfonate (BPS) is incubated with MCR_red1, an alkyl-Ni(III) species is formed that elicits the MCR_PS EPR signal. Here we used EPR and UV-visible spectroscopy and transient kinetics to study the reaction between MCR from Methanothermobacter marburgensis and a series of brominated carboxylic acids, with chain lengths from 4 to 16 carbons long. All of these compounds give rise to an alkyl-Ni intermediate with an EPR signal similar to that of the MCR_PS species. Reaction of the alkyl-Ni(III) adduct, formed from brominated acids with 8 or less total carbons, with HSCoM as nucleophile at pH 10.0 results in the formation of a thioether coupled to regeneration of the active MCR_red1 state. When reacted with 4-bromobutyrate, MCR_red1 forms the alkyl-Ni(III) MCR_XA state and then, surprisingly, undergoes “self-reactivation” to regenerate the Ni(I) MCR_red1 state and a bromocarboxy ester. The results demonstrate an unexpected reactivity and flexibility of the MCR active site to accommodate a broad range of substrates, which act as molecular rulers for the substrate channel in MCR.

Methanogenic archaea are anaerobic microbes that form methane as an end product of their metabolism. Methanogens produce approximately 1 billion tons of methane every year, providing a valuable source of renewal energy. On the other hand, methane is a greenhouse gas that is 20 times more potent than CO₂ and therefore poses potential environmental problems (1). Methyl coenzyme M reductase (MCR), found in all methanogens, is a nickel containing enzyme that catalyzes the terminal step in biological methane formation, which is the reduction of methyl-coenzyme M (2-(methylthiol)ethane sulfonate, methyl-SCoM) with the two-electron donor coenzyme B, (N-7-mercaptoheptanolyl-threonine phosphate, CoBSH), to form methane and the heterodisulfide CoBS-SCoM (1-3). There is strong evidence that MCR or a MCR-like enzyme also catalyzes the first step in the anaerobic oxidation of methane (4). The active site of MCR contains a redox-active Ni-tetrapyrrolic cofactor called coenzyme F₄₃₀ (or simply F₄₃₀) that is thought to play an essential role in catalysis (5-7). The crystal structure(s) of MCR reveals that F₄₃₀ is non-covalently bound to MCR and sits at the bottom of a 30 Å hydrophobic channel (8) that is sufficiently deep to accommodate both substrates.

The Ni atom of F₄₃₀ in MCR can exist in 3 different oxidation states, each exhibiting several coordination states. Active MCR_red1 contains Ni(I), which is the form that can react with methyl-SCoM to catalyze methane formation (9, 10). In addition to MCR_red1, two other states of MCR that could be relevant for the catalytic mechanism are MCR_ox1 and MCR_PS.

Based on spectroscopic and computational studies, MCR_ox1 is best described as a high spin Ni(II)/thiyl-radical that is in resonance with a Ni(III)/thiolate species (11, 12). MCR_ox1 resembles a proposed catalytic intermediate and is called the “ready state” (9) because it can be converted to MCR_red1 in vitro by incubation with the low potential reductant, titanium(III) citrate (Ti(III) citrate) (10). MCR_ox1 is generated in vivo by treating the actively growing cells with sodium sulfide (13) or alternatively, by switching the gas from 80%H₂/20%CO₂ to 80%N₂/20%CO₂ (14). MCR_ox1 can be generated in vitro by treating MCR_red2 (MCR_red1 in the presence of HSCoM and CoBSH) with polysulfide (15).

The second state of MCR that may be relevant for catalysis is MCR_PS (formerly MCR_BPS) (16). MCR_PS has been described as a high-spin Ni(II)/alkylsulfonate radical species resonating with a Ni(III)/alkyl species (17, 18), and, thus, resembles the proposed methyl-Ni catalytic intermediate. MCR_PS is generated in vitro by incubating MCR_red1 with 3-bromopropane sulfonate (BPS) (16), which is a potent inhibitor (apparent K_i = 50 nM) (19, 20) of MCR that also has been classed as an irreversible inhibitor (21). However, when it was recognized that the MCR_PS state can be converted back to the active MCR_red1 state by reacting with thiolates, reversing inhibition, BPS was classified as a reversible redox inactivator (22). Thus, both MCR_PS and MCR_ox1 are “ready” states that can generate MCR_red1 in vitro; furthermore, they exhibit similar UV-visible and EPR spectra.

The two mechanisms for methane formation can be distinguished by the first step of catalysis (Figure 1). In Mechanism I, which is based on the crystal structure and mechanistic work with F₄₃₀ model complexes (23-25), nucleophilic attack of Ni(I)-MCR_red1 on the methyl group of methyl-SCoM generates a methyl-Ni intermediate (26). Although a true methyl-Ni intermediate has not been described upon reaction of MCR_red1 with the native substrate methyl-SCoM, the reaction of BPS and other alkyl halides with MCR_red1 generates an alkyl-Ni(III) species that is subsequently attacked by free organic thiolates to generate the thioether product and MCR_red1 (22). This reaction is analogous to that of the reaction catalyzed by methionine synthase, where homocysteine acts as a nucleophile to reduce methyl-Co(III) to form Co(I) and methionine (27).

Figure 1.

Relationship between MCR_ox1, MCR_PS and the two proposed mechanisms for MCR.

Mechanism II, which is based on density function theory computations (28, 29), avoids the methyl-Ni(III) species because cleavage of the strong methyl-S bond of methyl-SCoM to form a relatively weak methyl-Ni(III) species was calculated to be extremely endothermic (45 kcal/mol). Therefore, Mechanism II proposes attack of Ni(I) on the sulfur atom adjacent to the methyl group of methyl-SCoM resulting in homolytic cleavage of the methyl-sulfur bond to generate a methyl radical and a Ni(III)-thiolate/Ni(II)-thiol-radical complex (MCR_ox1-like species) (Figure 1).

It was recently shown (22) that the alkyl-Ni(III) species (MCR_PS) can be converted to the active MCR_red1 state when it is reacted with various thiols at pH 10.0. The MCR_PS intermediate is chemically similar to the proposed alkyl-Ni intermediate in the first step of Mechanism I; however, the position of attack on BPS is stereochemically comparable to the first step in Mechanism II. The position of attack by Ni(I) is probably influenced by the placement of BPS (or other analogs) in the active site. Based on the MCR_ox1-silent crystal structure, the sulfonate group of HSCoM is firmly anchored by three contacts at the “back” of the active site and the thiolate is ligated to Ni. Similarly, the reactivity of CoBSH would be influenced by its location in the substrate channel relative to the position of the Ni-bound ligand. CoBSH has phosphate and carboxylate groups that anchor it to the upper lip of the channel and an alkyl chain that leads into the active site and terminates in a thiol group 8.7 Å from the nickel atom of F₄₃₀ (8).

In order to better understand the selectivity of the MCR reaction toward nucleophilic attack by Ni(I), one could compare the reactivity of a series of brominated sulfonates of varying chain length. However, such a series of compounds is not commercially available. On the other hand, substitution of the sulfonate of BPS with a carboxylate still allows formation of a MCR_PS-like signal (22), which we call MCR_XA. Furthermore, methylmercaptopropionate is a substrate for MCR (k_cat/K_M = 26 M^-1 s^-1), albeit ~ 110-fold less reactive than the natural substrate, methyl-SCoM (k_cat/K_M = 2.8 × 10³ M^-1s^-1) (30), and bromopropionate (like BPS) is an inhibitor (18), indicating that brominated carboxylic acids mimic the interactions of BPS and methyl-SCoM at the MCR active site. Therefore, in order to better understand the selectivity of the MCR reaction, we have in this paper compared the reactivity of a series of brominated compounds of varying chain lengths in formation of the alkyl-Ni(III) complexes and in the subsequent reaction with organic thiol to regenerate the active enzyme.

Here we show that various brominated acids ranging from the relatively small bromobutyric acid (Br4A) to the relatively large bromohexadecanoic acid (Br16A) can all react with MCR_red1 to form an EPR-active species, MCR_XA, that is nearly identical to MCR_PS. Based on the present studies a model has been proposed for the mode of binding of various brominated acids of different chain lengths that can be classified as: (a) methyl-SCoM-like and (b) CoBSH-like. The results provide information on the selectivity of MCR for its substrates (methyl-SCoM, CoBSH and BPS) and may aid in the development of other substrate analogs or inhibitors of MCR.

Materials and Methods

Materials and Organism

Methanothermobacter marburgensis (f. Methanobacterium thermoauotrophicum strain Marburg) was obtained from the Oregon Collection of Methanogens (Portland, OR) catalogue as OCM82. All buffers, media ingredients, and other reagents were acquired from Sigma-Aldrich (St. Louis, MO) unless otherwise stated and were of the highest purity available. Solutions were prepared using Nanopure deionized water. N₂ (99.98%), H₂S (99.0 %), H₂/CO₂ (80%/20%), and Ultra High Purity (UHP) H₂ (99.999%) were obtained from Linweld (Lincoln, NE). Ti(III) citrate solutions were prepared from a stock solution of 200 mM Ti(III) citrate, which was synthesized by adding an aqueous solution of sodium citrate (5.8 g sodium citrate added to 20 ml H₂O) to Ti(III) trichloride (30 weight % solution in 2 M hydrochloric acid) (Acros Organics, Morris Plains, NJ) under anaerobic conditions and adjusting the pH to 7.0 with a saturated solution of sodium bicarbonate (31). The concentration of Ti(III) citrate was determined routinely by titrating a methyl viologen solution.

M. marburgensis growth, harvest and MCR_red1 purification conditions

MCR_red1 was isolated from M. marburgensis cultured on H₂/CO₂/H₂S (80%/20%/0.1%) at 65 °C in a 14-L fermentor (New Brunswick Scientific Co., Inc. New Brunswick, NJ) (9). Culture media were prepared as previously described (32). MCR_red1 was generated in vivo and purified as described earlier (22). The purification procedure routinely generates 60-70 % MCR_red1 as determined by UV-visible and EPR spectroscopy described earlier (22)

Spectroscopy of MCR

UV-visible spectra of MCR were recorded in the anaerobic chamber using a diode array spectrophotometer (Model DT 1000A, Analytical Instrument Systems, Inc., Flemingron, NJ). EPR spectra were recorded on a Bruker EMX spectrometer (Bruker Biospin Corp., Billerica, MA), equipped with an Oxford ITC4 temperature controller, a Hewlett-Packard Model 5340 automatic frequency counter and Bruker gaussmeter. Unless otherwise noted, the EPR spectral parameters included: temperature, 70 K; microwave power, 10 mW; microwave frequency, 9.43 GHz; receiver gain, 2 × 10⁴; modulation amplitude, 10.0 G; modulation frequency, 100 kHz. Double integrations of the EPR spectra were performed and referenced to a 1 mM copper perchlorate standard.

Stopped-Flow Studies

Stopped-flow experiments were carried on an Applied Photophysics spectrophotometer (SX.MV18, Leatherhead, UK) equipped with a photodiode array detector. Rigorous measures were taken to remove oxygen from the stopped-flow instrument. Buffered solutions of enzymes and inhibitors were made in the anaerobic chamber in 0.5 M Tris-HCl, pH 7.6 containing 0.2 mM Ti(III) citrate at 25 °C. All buffered solutions contained 0.2 mM Ti(III) citrate as an oxygen scavenger, since exclusion of Ti(III) citrate resulted in oxidation of MCR_red1. The solutions were then loaded into tonometers, which had been incubated in the anaerobic chamber for at least 4 days and served as reservoirs for the drive syringes of the stopped-flow instrument. The drive syringes and mixing chamber were made anaerobic by flushing the syringe chamber with a dithionite:resazurin (1 mM: 0.02 mM) solution in 0.1M NaOH. MCR_red1 and varied concentrations of brominated acids were rapidly mixed at 25 °C in a 1:1 ratio. The reaction was monitored in the single wavelength mode by following the decrease of MCR_red1 at 385 nm and MCR_XA formation was followed at 420 nm, with a 1 cm pathlength. Data were fit to a single exponential decay and exponential rise to maximum functions, respectively, using software provided by Applied Photophysics (version SX MV.18) or using Sigma Plot 2001 (Point Richmond, CA). Reported rate constants are the average of at least three different rapid-mixing experiments.

Mass spectromety

All mass spectra (LCMS) were collected in negative ion modes by direct infusion to a [4000 Qtrap (ABS)] mass spectrometer using a kD Scientific micro flow syringe pump. The data was acquired and processed using Analyst 1.4.1 software. Data was acquired in Q1 (quadrupole one) and Product ion scan in negative ion mode. The mass range of 50-300 amu (atomic mass units) was scanned in 1s. The ionspray voltage was set to -4500 V, the temperature was set to 22 °C and a declustering potential (DP) value was -50 V. MCR_red1 (10 μM) was incubated for 10 min with 2.0 mM 4-bromobutyric acid (Br4A was prepared as a stock solution in 0.1 M formic acid; thus, the final concentration of formic acid in the reaction mixture is 0.4 mM) in 20 mM ammonium carbonate, pH 10.0, and the reaction was followed by UV-visible spectroscopy. For, the regeneration of MCR_red1 from MCR_XA with HSCoM, typically 20 μM MCR_red1 was reacted with 1 mM brominated acid in 50 mM ammonium carbonate, pH 10 to form MCR_XA, which was then reacted with 20 mM HSCoM. This reaction was monitored by UV-visible spectroscopy. After MCR_red1 was fully reactivated, assessed by reaching a stable maximum absorbance of the 385 nm band, the reaction was stopped by freezing in liquid nitrogen. For analysis, the frozen liquid was thawed and, subsequently, enzyme and ligands were separated using a 0.5 mL Microcon centrifugal filter device with a 30 KDa cut-off filter (Millipore, Billerica, MA). The filtrate containing the product was collected and stored at 4 °C until it was assayed by mass spectrometry. All MS samples were prepared by mixing the ligand solution with an equal volume of acetonitrile.

Reactions between MCR_red1 and brominated acids

MCR_red1 (typically 0.01-0.02 mM) was incubated in 50 mM Tris-HCl (pH 7.6) at 25 °C for 0.5-2 min, with various brominated acids at concentrations between 0.1 mM and 2 mM: 0.2 mM 4-bromobutyric acid (Br4A) in 0.1 mM final formic acid concentration, 0.2 mM 5-bromovaleric acid (Br5A), 0.2 mM 6-bromohexanoic acid (Br6A), 1.0 mM 7-bromoheptanoic acid (Br7A) (Karl Industries Inc., Aurora, OH and Matrix Scientific, Columbia, SC), 0.1 mM 8-bromooctanoic acid (Br8A), 0.2 mM 9-bromononanoic acid (Br9A) (Matrix Scientific), 0.2 mM 10-bromodecanoic acid (Br10A) (Matrix Scientific), 1.0 mM 11-bromoundecanoic acid (Br11A) (Fluka, St.Louis, MO), 1.0 mM 12-bromododecanoic acid (Br12A) (Fluka), 1.0 mM 15-bromopentadecanoic acid (Br15A) (Fluka), and 0.5 mM 16-bromohexadecanoic acid (Br16A). All brominated acid solutions, with the exception of Br4A, were made in 100 % ethanol. The rate and extent of formation of new MCR_XA complexes was measured by EPR spectroscopy (Table 1).

Table 1.

Kinetic parameters for the conversion of MCR_red1 to MCR_XA.

Chemical Name	Chemical Structure	MCRred1 → MCRXA (% Conversion)c	kmax (s-1)	KM (mM)	kmax/KM (M-1 s-1)a
Br4A		85 ± 15	15 ± 0.6	88 ± 5	170 ± 12
Br5A	n = 2	56 ± 10	2.0 ± 0.5	65 ± 27	31 ± 15
Br6A	n = 3	53 ± 10	0.27 ± 0.05	5.2 ± 3.6	53 ± 38
Br7A	n = 4	93 ± 7	2.2 ± 0.3	21 ± 7	140 ± 57
Br8A	n = 5	93 ± 1	2.1 ± 0.5	92 ± 34	23 ± 10
Br9A	n = 6	60 ± 2	0.074 ± 0.003	1.3 ± 0.2	56 ± 7
Br10A	n = 7	12 ± 8	0.050 ± 0.002	2.2 ± 0.5	22 ± 2
Br11A	n = 8	<5 (13 ± 5)	380 ± 170	200 ± 110	1900 ± 1300 1500 ± 80b
Br12A	n = 9	<5 (10 ± 5)	0.28 ± 0.02	0.93 ± 0.26	500 ± 92
Br15A	n = 12	6 (20 ± 5)	0.26 ± 0.04	0.16 ± 0.11	1600 ± 1000 470 ± 64b
Br16A	n = 13	8 (41 ± 2)	ND	ND	57 ± 18b

^aSecond-order rate constants were determined by a hyperbolic fit.

^bSecond-order rate constants determined by fitting data to a linear equation.

^cThe % conversion was determined by EPR. The number in the parenthesis was obtained when the reaction was performed on ice. For details, refer to experimental section

Special precautions were required for working with 4-bromobutyric acid (Br4A). This compound is unstable in neutral and alkaline solutions; therefore, all experiments were performed by dissolving it in 0.1M formic acid. Attempts to make a solution of Br4A in neutral or high pH solutions or in ethanol resulted in degradation of the acid as we inferred by a decrease in reactivity of Br4A with MCR_red1 as a function of time.

Results

MCR_XA formation with brominated acids, studied by EPR and UV-visible spectroscopy

Kinetic and spectroscopic studies indicate that, when MCR_red1 reacts with BPS, bromide is eliminated to generate a six-coordinate Ni(III)-complex with propylsulfonate as the upper axial ligand, which undergoes protonation to form propanesulfonate (17, 18, 22). This reaction is analogous to the reaction of MCR_red1 to generate a methyl-Ni(III) intermediate in methane formation with the natural substrate, methyl-SCoM. Here we expand on a previous study to determine the range of BPS analogs that are accommodated by MCR (22). Instead of sulfonates, we used a series of commercially available brominated carboxylic acids. As described in the Introduction, we hypothesized that the reaction between MCR_red1 and brominated acids is similar to that with BPS. To test this hypothesis, EPR and UV-visible spectroscopies were used to observe the formation of the Ni(III)-alkanoic acid complex(s), and stopped-flow experiments were conducted to characterize the kinetic parameters for the reactions. Since MCR_XA (alkyl-Ni(III)) and MCR_silent (Ni(II)) exhibit similar UV-visible spectra, unambiguous quantification of the amount of MCR_red1 and of MCR_XA was performed by EPR spectroscopy by comparing the doubly integrated signal intensity with that of a Cu standard. The concentration of MCR_silent is then equal to the difference between the initial concentration of MCR_red1 and the amount of MCR_XA. Brominated acids used in this study are abbreviated as BrXA, where X refers to total length of the chain (between 4 and 16 carbons long) including the terminal carboxyl group (A).

MCR_red1 reacts with brominated acids that are 4 to 16 carbons in length according to Equation 1 to form MCR_XA (X = 4-16), which elicits an EPR signal nearly identical to that of MCR_PS (Table 1). A representative EPR spectrum of MCR_red1 and MCR_7A, the product of the reaction between MCR_red1 and Br7A, is shown in Figure 2.

$${\mathrm{MCR}}_{\text{red}1}+\mathrm{BrXA}\to {\mathrm{MCR}}_{\mathrm{XA}}+{\mathrm{Br}}^{\text{-}}$$ Equation (1)

Figure 2.

EPR spectra of MCR_7A formed from the reaction of MCR_red1 with Br7A. MCR_red1 (20.6 μM (solid line)) was incubated with 1 mM Br7A (dashed line) or 548 μM BPS (dotted line) in 50 mM Tris-HCl (pH 7.6) with 0.1 mM Ti(III) citrate at 20 °C. *Contamination from MCR_ox1.

The near identity of the EPR spectrum of the alkyl-Ni(III) MCR_PS state with that of MCR_XA strongly indicates that the reaction of MCR_red1 with the bromoacids also generates an alkyl-Ni(III) species. The different MCR_XA complexes accumulate to varying amounts, as reported in Table 1. Those generated from the shorter brominated acids, Br10A and below, form quickly and accumulate to fairly high amounts during a 2-min reaction at room temperature. The shortest acid tested is Br4A, which had earlier been reported to form a MCR_PS-like species (17, 18, 22). When reacted with MCR_red1, Br4A completely converts to the MCR_4A species; however, this reaction is unique and is discussed in more detail below. 3-Bromopropionate (Br3A) was not tested although it has been reported not to form the MCR_XA signal (18). Br3A is an analog of 2-bromoethanesulfonate (BES) and like BES ([I]_{0.5 V} = 2 μM), quenches the MCR_red1 signal by the oxidation of the enzyme from Ni(I) to an EPR-silent Ni (34) state, thus BES and hence, Br3A appears to act as irreversible inhibitors of MCR (18).

The MCR_XA species formed from the longer (X = 11, 12, 15, and 16) brominated acids accumulate to less than 10% within 1 min at 25 °C, indicating that these complexes form transient species that form and decay more rapidly than the time it takes to hand mix and freeze the samples. When these brominated acids were reacted with MCR_red1 at 4 °C and frozen within ~30 secs in an EPR tube, accumulation of the corresponding MCR_XA species markedly increased (Table 1). We suggest that the MCR_XA complexes formed with the longer brominated acids are not well anchored to the active site and, thus, escape from the radical cage and dissociate rapidly from the substrate binding channel in MCR. To better understand the reactions of MCR_red1 with the bromoacids, stopped flow experiments were performed.

Reaction of MCR_red1 with brominated acids and MCR_XA formation followed by stopped-flow

Formation of the EPR-active MCR_XA state is accompanied by a 35 nm red shift relative to MCR_red1 (Figure 3). Exhibiting absorption maxima at 420 nm, the UV-visible absorption spectra of MCR_XA resemble those of MCR_PS (alkyl-Ni(III)), MCR_ox1 (RS-Ni(III)), and MCR_silent (Ni(II)).

Figure 3.

UV-visible spectral changes associated with reaction of MCR_red1 with Br7A and BPS. MCR_red1 (8.2 μM (solid line)) was incubated with 400 μM Br7A (dashed line) or 219 μM BPS (dotted line) in 50 mM Tris-HCl (pH 7.6) with 0.1 mM Ti(III) citrate at 20 °C.

Stopped-flow UV-visible spectroscopy was used to obtain kinetic parameters for the reaction of MCR_red1 with the brominated acids by monitoring the absorption bands at 385 nm and 420 nm to follow the decay of MCR_red1 and formation of MCR_XA, respectively. For the reaction with each of the brominated acids, the rate constant (k) could be obtained by a single exponential equation, which adequately fit the data. The rate of decay of MCR_red1 matches the rate of MCR_XA formation, which indicates that there are no intermediates in the conversion of MCR_red1 to MCR_XA, or that any intermediates are too transient to be observed. These rate constants were plotted versus the BrXA concentrations to determine the second order rate constant (k_max/K_M) for MCR_XA formation as exemplified for Br7A in Figure 4. The second order rate constants for formation of the MCR_XA complexes with the shorter and medium length acids (X = 5-10) are between 20-60 M^-1 s^-1 (Table 1, Figures S1-S5), with the exception of MCR_7A, which has a second order rate constant of 140 M^-1 s^-1. The second order rate constants for formation of MCR_11A, MCR_12A and MCR_15A (Table 1, Figures S6-S8) are significantly higher. The k_max and K_M values for MCR_16A could not be determined due to solubility issues involved with Br16A concentrations above 0.5 mM. However, with the available concentrations, a linear fit of the data established a second order rate constant of 60 M^-1 s^-1 (Table 1, Figure S9).

Figure 4.

Dependence of k_obs for MCR_7A on Br7A concentration. 10 μM MCR_red1 was converted to MCR_7A by the addition of different concentrations of Br7A in 0.5 M Tris-HCl, pH 7.6 containing 0.2 mM Ti(III) citrate. The data were fit to a 2-parameter hyperbolic equation with a k_max of 2.8 ± 0.4 s^-1 and K_M of 21 ± 8 mM and a second order rate constant of 140 ± 57 M^-1 s^-1.

The kinetics for formation of these MCR_XA complexes can be compared to those for formation of their similarly sized sulfonate cousins, MCR_PS and MCR_BS, and to the rate of methane formation from methyl-SCoM. MCR_4A is formed with an overall second order rate constant of 170 M^-1 s^-1, which is 1000 times lower than that for formation of MCR_PS (2.3 × 10³ M^-1 s^-1 at 25 °C) (17, 18, 22). However, the k_max values for these reactions are nearly identical (MCR_PS, 17s^-1, MCR_4A, 15 s^-1); therefore, the decreased catalytic efficiency for the formation of MCR_4A relative to MCR_PS comes exclusively from the difference in K_M values for these two substrates, ~90 mM for Br4A and ~0.09 mM for BPS. On the other hand, the k_max/K_M for MCR_4A formation is only 5.4-fold lower than the k_cat/K_M for methane formation from methyl-SCoM (930 M^-1 s^-1 at 20 °C, 1.9 × 10⁴ M^-1 s^-1 at 65 °C) and the k_max value is actually 3-fold higher than the k_cat for methane formation at 20 °C. The second order rate constant for MCR_5A formation (31 M^-1s^-1) is 70-fold lower than that for formation of MCR_BS (2.3 × 10³ M^-1s^-1), the analogous sulfonate complex, resulting from a 7-fold increase in K_M and a ~10-fold decrease in k_max. The reactions of MCR_red1 with Br7A and Br8A exhibit kinetic parameters similar to those of the shorter bromoacids with near or complete conversion to MCR_7A and MCR_8A. For example, MCR_7A formation occurs with a second order rate constant of 140 M^-1 s^-1, the k_max values are rather high (~2-3 s^-1) and the K_M values are similar to those of methyl-SCoM. We conclude that the short chain brominated acids are positioned in the active site much like methyl-SCoM and HSCoM and react similarly to BPS.

There is a marked drop in both k_max and K_M values for the reactions of MCR_red1 with Br9A, Br10A, Br12A, Br15A and Br16A (Table 1, Figures S4, S5 and S7-S9). The K_M values for most of these BrXA fall in the range of the K_M for CoBSH (~0.2 mM) (21) and the values of k_max/K_M for the reactions with Br11A, Br12A and Br15A are similar to the k_cat/K_M for CoBSH (2.2 × 10³ M^-1s^-1, 20 °C) in methane formation (26), suggesting that these long chain bromoacids are recognized by MCR as CoBSH analogs. Based on a simple structural analysis using ChemDraw (Figure 5), the negatively charged oxygen on the carboxyl group of Br10A would be in the same position as the carboxylate oxygen of CoBSH and the oxygen from the carboxylate of Br12A would occupy the same position as the phosphate oxygen of CoBSH. The corresponding MCR_XA complexes decay more rapidly than those formed from the shorter BPS-like bromoacids, with <10 % of the MCR_XA complex accumulating when mixed at room temperature (Table 1). The yield of EPR-active MCR_XA could be increased to ~20-40 % by performing the reaction at 4 °C.

Figure 5.

Comparison of the structures of Br11A and CoBSH. Two alternative conformations (a and b) are shown by the solid and dotted lines.

Although Br11A reacts very rapidly with MCR_red1, we did not observe saturation even at 60 mM concentrations; therefore, there is a relatively high standard error associated with both the K_M (200 mM) and k_max (380 s^-1) values shown in Table 1. The second order rate constant for MCR_11A formation determined from a linear fit (1500 ± 80 M^-1 s^-1) is more accurate and is in reasonable agreement with that determined from a hyperbolic fit (1900 M^-1 s^-1). Due to the low solubility of Br15A and Br16A, kinetic parameters were determined using brominated acid concentrations no greater than 0.5 mM. Regardless, the k_obs/K_M values are comparable to that obtained with Br11A; for example, for the reaction of Br15A with MCR_red1, the second order rate constant is 1600 M^-1s^-1 (Figure S8). Then, with Br16A, saturation could not be reached and the k_max/K_M value, determined from a linear fit is 57 M^-1s^-1.

Reaction of MCR_red1 with Br4A - self-reactivation of MCR_4A to form MCR_red

When the reaction between MCR_red1 and Br4A was monitored by UV-visible and EPR spectroscopy, MCR_red1 was shown to quantitatively convert to MCR_4A., as shown earlier (17, 18, 22); however, when lower concentrations (~0.2 mM) of Br4A were used, a rapid decrease in absorbance at 385 nm due to formation of MCR_4A was observed, followed by a gradual increase in absorbance at 385 nm after ~1 min (Figure 6), which corresponds to the reformation of MCR_red1. The regeneration of MCR_red1 occurred at either pH 10.0 or pH 7.6 in the absence of free thiols, which were shown to convert MCR_PS to MCR_red1 at pH 10.0 (17, 18, 22).

Figure 6.

Formation of MCR_4A followed by regeneration of MCR_red1. The formation or decay of MCR_4A and MCR_red1 were monitored at 385 nm (solid line) and 420 nm (open triangles) by UV-visible spectroscopy. MCR_red1 (10 μM) was reacted with 0.2 mM Br4A in 200 mM ammonium carbonate, pH ~ 10. The data were fit to three-component sequential equations with the following parameters: for the absorbance changes at 385: k₁, 0.037 ± 0.003 s^-1; k₂ = 0.0067 ± 0.0003 s^-1; for the absorption changes at 420 nm: k₁, 0.036 ± 0.001 0 s^-1, k₂ 0.0054 ± 0.0007 s^-1.

The rate of MCR_4A generation depends on the Br4A concentration, with a second order rate constant of 170 ± 12 M^-1 s^-1 (k_max, 15 s^-1; K_M, 89 mM) (Figure S10), while the rate constant for regeneration of MCR_red1 (0.0067 s^-1) is 2200-fold slower and is independent of the Br4A concentration. At low concentrations of Br4A (i.e., 0.2 mM), there is complete (100%) conversion of MCR_4A to MCR_red1 (Figure 6). However, at higher concentrations of Br4A, MCR_4A is only partially converted to MCR_red1 and, at Br4A concentrations greater than 8 mM, there is no appreciable regeneration and the absorbance at 420 nm increases to a limiting value in a single exponential fashion. The apparent conundrum is that conversion of alkyl-Ni(III) to Ni(I) and an alkane requires a two-electron input into the Ni(III) center and no redox mediators are present; thus, it was important to determine the product of this MCR_red1 regeneration reaction.

Mass spectrometric (MS) evidence for ester formation from the self-reactivation reaction of MCR_4A

Mass spectrometry was used to identify the product of the self-reactivation reaction of MCR_red1 with Br4A (Figure 7). In the negative ion mode (NIM), the major peak has an m/z value of 250.8. This corresponds to the molecular formula C₈H₁₃O₄Br with an exact mass of 252, which can be assigned to the bromo carboxy ester, 4-(4-bromobutanoyloxy)butanoic acid, Br-(CH₂)₃-COO-(CH₂)₃-COOH. In NIM, the loss of a proton from the carboxylic terminal group of the bromo-carboxy ester product, Br-(CH₂)₃-COO-(CH₂)₃-COOH, will give the expected m/z value of 251, i.e., (252 - 1)/1, consistent with the experimentally observed value (Inset, Figure 7). Confirmation that the product is the ester was obtained by MS-MS analysis of the parent ion peak (m/z 250.8).

Figure 7.

NIM MS-MS identification of the fragmented/daughter ions from the bromocarboxy ester product formed from the “self-reactivation” of MCR_4A to MCR_red1. MCR_red1 (10 μM) was reacted with 2.0 mM Br4A in 20 mM ammonium carbonate, pH ~ 10. Inset: NIM ES-MS identification of the bromo-carboxy ester product formed from the “self-reactivation” of MCR_4A to MCR_red1.

Daughter ion peaks were primarily observed at m/z 149.5, 164.8, 171.7 (Figure 7). The peak at m/z 171.7 arises from loss of bromine from the parent ion to give C₈H₁₃O₄, with an exact mass of 173.2, followed by deprotonation of the carboxylic acid to give the negative ion, [C₈H₁₃O₄]^-, with a calculated m/z of 172, i.e, (252 - 79 = [173 - 1/1]=172), the fragment labeled ‘c’ in Figure 7. Thus, the ester can cleave at different positions marked ‘a’, ‘b’, and “c” (Figure 7) to form daughter ions having molecular formulae C₄H₆O₂Br, C₄H₆OBr and C₈H₁₃O₄, and the peaks marked ‘a’ and ‘b’ agree the experimentally observed m/z values of 164.8 and 149.5, respectively.

When the negative control experiment was performed in absence of enzyme, the product was not observed in either NIM or positive ion mode (PIM); however, a peak corresponding to the bromide was observed in NIM and the remaining fragment after de-bromination, which presumably is a positive species, was observed in PIM. Thus, the reaction of MCR_red1 with Br4A generates the alkyl-Ni(III) species MCR_4A. Another molecule of Br4A reacts with MCR_4A to generate the ester, 4-(4-bromobutanoyloxy)butanoic acid. Two possible mechanisms by which this condensation could occur are considered in the Discussion.

MCR_XA is reactivated to MCR_red1 by thiols

It has been recently observed that the MCR_PS complex can be reconverted to active MCR_red1 by nucleophilic attack of various thiolates on the carbon bound to nickel. To determine if MCR_XA reacts similarly with thiols, we incubated the MCR_XA complexes with HSCoM and followed the decrease in absorbance of MCR_XA at 420 nm and the increase in MCR_red1 at 385 nm (Figure 8). Since MCR_XA and MCR_silent have similar UV-visible spectra, we also followed the reaction by EPR spectroscopy (inset, Figure 8).

Figure 8.

UV-visible spectral changes associated with the regeneration of MCR_red1 from MCR_7A incubated with 20 mM HSCoM. MCR_red1 (11.52 μM (solid line)) was treated with 100 μM Br7A (dotted line) and the MCR_7A formed was incubated with 20 mM HSCoM in 1M TAPS-Na (pH ~ 10) at 20 °C and the absorbance was recorded every 6 mins for over 40 mins. Inset: Representative EPR spectra of MCR_red1 (solid line) MCR_7A (dotted line) and regeneration of MCR_7A to MCR_red1 (dashed line).

The rate constant for conversion of the different MCR_XA species to MCR_red1 is in the range 0.003-0.012 s^-1 (Table 2), which is similar to the rate constant for regeneration by HSCoM of MCR_BS (k_max of 0.006 s^-1) and MCR_PS (k_max is 0.011 s^-1). We did not observe a HSCoM concentration dependence for the rate of conversion of MCR_XA to MCR_red1; however, the reactions with MCR_BS (K_M = 17 mM, Figure S11) and MCR_PS (K_M = 2.8 mM (21)) exhibit clear hyperbolic binding curves.

Table 2.

Regeneration of MCR_red1 from MCR_XA using 20 mM HSCoM. Experiments were performed in 0.5 M TAPS (pH 10) at 25 ^oC.

Chemical Name	Reactivation of MCRXA	MCRXA → MCRred1 (% reactivation)a	kobs (s-1)
Br5A	Yes	46	0.012 + 0.001
Br6A	Yes	65	0.006 ± 0.001
Br7A	Yes	81	0.006 ± 0.001
Br8A	Yes	31	0.003 ± 0.001
Br9A	No	Negligible	NA

^aThe % reactivation represents the percentage conversion of MCR_XA to MCR_red1. For details of experimental conditions, refer to experimental section.

Among the different brominated acids shown in Table 2, the percent conversion of MCR_XA to MCR_red1 vary, with a maximum of about 80 % conversion of MCR_7A to MCR_red1 (inset, Figure 8). Any EPR-inactive Ni species that is formed most likely decayed into Ni(II)-MCR_silent. Furthermore, we did not observe conversion of MCR_XA, formed from the longer brominated acids (9, 10, 11, 12, 15 and 16-BrA), to MCR_red1 when these MCR_XA species are incubated with HSCoM. Perhaps the MCR_XA complex formed from the longer acids did not reactivate because the active site may be plugged, preventing access of thiols to the active site.

Mass spectrometry evidence that a thioether product is formed from the reaction of MCR_XA and HSCoM

We used MS in NIM to identify the product of the reaction between MCR_5A and HSCoM, as CoM-S-5A, with an m/z value 241.3, which corresponds to the molecular formula C₇H₁₄O₅S₂ with an exact mass of 241.3 after loss of a proton (inset, Figure 9). This assignment was confirmed by MS-MS analysis of the parent ion peak at m/z 241.3, which yields the fragmentation pattern shown in Figure 9.

Figure 9.

NIM-MS/MS spectrum of the product of the reaction of MCR_5A with HSCoM. 20 μM MCR_red1 was reacted with 1mM Br5A to form MCR_5A, which was reacted with 1 mM HSCoM in 500 mM ammonium carbonate pH ~ 10.0. Inset: Original NIM-MS spectrum of the reaction product.

Daughter ion peaks were observed at m/z 140.9, 132.9, 107.0, 99.0 and 81.0. The daughter ion at m/z 132.9, labeled “b” is assigned to the 5-thiopentanoic acid fragment, with a molecular formula C₅H₉O₂S in which the C-S bond is cleaved. The remaining fragment ion is desulfo-CoM with a molecular formula C₂H₄SO₃ and m/z of 107.0 (fragment labeled ‘c’). The daughter ion peak at m/z 140.9 (peak labeled ‘a’) corresponds to C₂H₆O₃S₂ with an exact mass of 141.97, which is formed when the thioether (S-C) bond is cleaved on the pentanoic acid side of the thioether group. A secondary ion that is generated from this S-C fragmentation is pentanoate, with an m/z of 99.0 (fragment and peak labeled ‘d’). The peak at m/z 81.0, labeled ‘e’, is assigned to the sulfonate group that is cleaved from various ions.

A negative control experiment performed in the absence of enzyme gave a peak at m/z 242.3, which is only one mass unit higher than the expected product (CoM-S-5A, m/z 241.3). In order to confirm if the 242.3 peak corresponds to the product, MS-MS was performed; the resulting fragmentation pattern was not consistent with that observed for CoM-S-5A product, which rules out the possibility of the thioether product in the absence of enzyme.

MS coupled with MS-MS also verified formation of the thioether products from the reactions of MCR_6A and MCR_7A with HSCoM. Thus, the MS data reveal that the reaction of HSCoM with MCR_XA generates a thioether product, like the reaction with MCR_BS and MCR_PS.

Discussion

We studied the reaction of MCR_red1 with a series of brominated carboxylic acids by EPR and UV-visible spectroscopy and by kinetics. These reactions, as with BPS, apparently involve nucleophilic attack of Ni(I)-MCR_red1 on the terminal carbon adjacent to the bromine atom to eliminate bromide and generate the EPR-active MCR_XA species. Compounds that give rise to the MCR_XA EPR signal can be abbreviated as BrXA and are generalized by the following structure: Br-(CH₂)_3-15-COO^-; where X is the alkyl linker and is ~3-15 carbons long, and A is the terminal anionic carboxylate group. Based on the near identity of the MCR_XA EPR signal to that of the MCR_PS (21), we can confidently assign MCR_XA as a Ni(III)-alkyl carboxylate. As with MCR_PS, the lack of detectable hyperfine splitting from the bromide (I = 3/2) demonstrates that it is not near the paramagnetic center, suggesting that the bromide undergoes elimination in formation of the MCR_XA state.

The BrXA's were characterized by their reactivity with MCR_red1 to form MCR_XA (Table 1), their ability to accumulate MCR_XA (Table 1), and by the rate at which the respective MCR_XA complexes react with thiolates (^-SCoM) to form MCR_red1 and the CoMS-XA thioether (Table 2). Accumulation of MCR_XA (in the absence of the thiolate) is a function of the rate of MCR_XA formation and its decay to form XA and the EPR-silent Ni(II) MCR_silent state. All of these bromoacids form EPR-active MCR_XA species and fall into two classes, based on their reactivity (Figure S12).

The shorter brominated acids, Br4A to Br8A, react rapidly with MCR_red1 to form the MCR_XA state, exhibit high values of K_M and k_max for MCR_XA formation, and accumulate nearly quantitatively in this state; furthermore, these compounds react with HSCoM to form a thioether and regenerate MCR_red1. These properties indicate that the short bromoacids mimic binding of methyl-SCoM (K_M = 5.0 mM, (31)), with their carboxylate groups interacting with Arg120.

The MCR_4A complex formed from the short bromoalkanoic acid, Br4A, undergoes conversion to MCR_red1 in the absence of thiolate. One would expect Br4A to be reactive, since like BPS (the most potent inhibitor of MCR), its bromide is adjacent to an electrophilic carbon atom that is four bonds (~ 4.8 Å) from the negatively charged oxygen of the carboxylate. Accordingly, the k_max values for formation of MCR_PS and MCR_4A are nearly identical (~15 s^-1). Therefore, we propose that MCR_red1 reacts with Br4A to form MCR_4A, an alkyl-Ni(III) species as with the other brominated substrate analogs. However, unlike BPS, with three oxygen atoms on the sulfonate that firmly anchor it to the active site, the carboxylate of Br4A is short of one oxygen, and thus would be held by weaker interactions (as indicated by its high K_M value). Regardless, it is likely that the alkanoate group is positioned in the active site with the methylene group bound to Ni and the carboxylate group bound to Arg120. Then, two pathways could be considered for formation of 4-(4-bromobutanoyloxy)butanoic acid, the product that is observed by mass spectrometry. In Pathway I, a second molecule of Br4A could enter the active site and nucleophilically attack the nickel-bound electrophilic carbon-4 to form the bromocarboxy ester and MCR_red1. The nucleophilic attack of the second Br4A is similar to the reaction of HSCoM on the MCR_XA species (where X = 5-8). In fact, the rate constants for formation of the thioethers and the bromocarboxy ester are similar. Another possibility, shown in pathway II, Figure S13 is that the electrophilic methylene group bound to Ni in MCR_4A could undergo intramolecular attack by the negatively charged oxygen of the carboxylate to form a five-membered butyrolactone and MCR_red1. Then, another molecule of Br4A would react with the butyrolactone in solution or in the enzyme active site to generate the bromocarboxy ester.

The longer bromoacids, Br9A to Br16A, apparently mimic CoBSH (K_M = 0.2 mM, 21) since they bind more tightly to MCR, exhibit second-order rate constants for formation of MCR_XA that are similar to those for CoBSH in methane synthesis, and accumulate to a much lower extent than the shorter brominated acids. These compounds likely bind with their carboxyl group interacting with the solvent and the positively charged residues at the upper lip of the channel with the bromoalkyl chain reaching toward the nickel center, where it could react rapidly and form the MCR_XA complex. The unstable alkyl-Ni(III) complex then rapidly decays to an EPR silent Ni(II) state. Perhaps this is due to homolytic cleavage of the nickel carbon bond giving Ni(II)-MCR_silent and the corresponding alkanoic acid radical, which could abstract a hydrogen atom from the environment of protein to form the alkanoic acid (21).

Conclusion

The present studies reveal that MCR_red1 can react with a wide range of brominated acids giving rise to an alkyl-Ni(III) complex, called MCR_XA, which exhibits UV-visible and EPR spectra that are nearly identical to the Ni(III)-alkylsulfonate species formed from MCR_PS. Thus, the MCR_XA complex resembles the methyl-nickel(III) species in the first step of one of the mechanisms (Mechanism I) proposed for methane formation from the natural substrates. The present studies reveal the flexibility of the active site of MCR to accommodate a broad array of substrates. The stable MCR_XA species (where X = 5-8) can undergo attack by nucleophilic thiols to form a thioether product and regenerate the active Ni(I) MCR_red1 state, while MCR_4A converts to MCR_red1 in the absence of a thiol. The unexpected reactivity and flexibility of the MCR active site to accommodate a broad range of substrates provides a molecular ruler for the substrate channel in MCR.

Abbreviations

MCR

Methyl-coenzyme M reductase

methyl-SCoM

methyl-coenzyme M or, 2-(methylthio)ethanesulfonate

HSCoM

coenzyme M or, 2-thioethanesulfonate

CoBSH

coenzyme B or N-7-mercaptoheptanoylthreonine phosphate

BPS

3-bromopropane sulfonate

BBS

4-bromobutane sulfonate

MCR_red1

MCR exhibiting the Ni (I) EPR signal

MCR_PS

MCR exhibiting alkyl-Ni(III) signal arising from reaction of MCR_red1 with BPS

MCR_BS

MCR exhibiting the alkyl-Ni(III) signal from the reaction of MCR_red1 with BBS

MCR_XA

MCR exhibiting the alkyl-Ni(III) signal from reaction of MCR_red1 with bromocarboxylic acids

NIM

negative ion mode