Kinetic Characterization of the C-terminal Domain of Malonyl-CoA Reductase

Abstract

Malonyl-CoA reductase utilizes two equivalents of NADPH to catalyze the reduction of malonyl-CoA to 3-hydroxypropionic acid (3HP). This reaction is part of the carbon fixation pathway in the phototrophic bacterium Chloroflexus aurantiacus. The enzyme is composed of two domains. The C-terminal domain catalyzes the reduction of malonyl-CoA to malonic semialdehyde, while the N-terminal domain catalyzes the reduction of the aldehyde to 3HP. The two domains can be produced independently and retain their enzymatic activity. This report focuses on the kinetic characterization of the C-terminal domain. Initial velocity patterns and inhibition studies showed the kinetic mechanism is ordered with NADPH binding first followed by malonyl-CoA. Malonic semialdehyde is released first, while CoA and NADP⁺ are released randomly. Analogs of malonyl-CoA showed that the thioester carbon is reduced, while the carboxyl group is needed for proper positioning. The enzyme transfers the pro-S hydrogen of NADPH to malonyl-CoA and pH rate profiles revealed that a residue with a pK_a value of about 8.8 must be protonated for activity. Kinetic isotope effects indicated that NADPH is not sticky (that is, NADPH dissociates from the enzyme faster than the rate of product formation) and product release is partially rate-limiting. Moreover, the mechanism is stepwise with the pH dependent step occurring before or after hydride transfer. The findings from this study will aid in the development of an eco-friendly biosynthesis of 3HP which is an industrial chemical used in the production of plastics and adhesives.

1. Introduction

The decreasing reserves of petroleum and the urgent global challenge of climate change have reignited strong interest in the pursuit of sustainable technologies to replace petroleum as the primary source of fuel and chemicals [1]. This resurgence has led to an increasing emphasis on harnessing microbial processes for chemical production using renewable biomass materials [2]. However, the limited industrial-scale production of bioproducts persists due to high production costs, encompassing raw materials, fermentation, and recovery/purification processes. Although significant progress has been made in developing microorganisms capable of competing on an industrial scale, the overall low yield of microbial strains must improve to match their synthesis from fossil fuels [3].

The compound 3-hydroxypropionic acid (3HP)¹ is a building block for the synthesis of bulk chemicals such as 1,3-propanediol, acrylic acid, acrylamide [4], and polyhydroxyalkanoate poly(3-hydroxypropionic acid) [5]. These bulk chemicals are utilized in the production of coatings, absorbents, sealants, adhesives [2, 6, 7], and bio-degradable plastic [5]. Currently, the source of 3HP and acrylic acid is chemical synthesis replete with all its harmful effects on the environment [8]. A potentially more environmentally friendly method for producing 3HP involves genetically modifying bacterial metabolic pathways. Currently, the commercial potential of 3HP is significant, with estimated annual market revenue exceeding $10 billion and a production volume of around 3.6 million tons per year [9].

The compound 3HP is also an intermediate in the pathway for CO₂ fixation in the photosynthetic bacterium Chloroflexus aurantiacus [10–13]. The enzymes in the pathway that generate 3HP are shown in Figure 1.

Figure 1.

Malonyl-CoA production and function of malonyl-CoA reductase (MCR) for 3-hydroxypropionic acid production. The catalytic function of MCR is indicated in the box.

The first enzyme, acetyl-CoA carboxylase (ACC), utilizes bicarbonate to carboxylate acetyl-CoA to form the three-carbon metabolite malonyl-CoA (Figure 1). Malonyl-CoA is a substrate for malonyl-CoA reductase (MCR) which catalyzes two separate reactions. The first reaction is catalyzed by the C-terminal domain (MCR-C) and utilizes NADPH to reduce malonyl-CoA to malonic semialdehyde (MSA). In the second reaction, which takes place in the N-terminal domain (MCR-N), NADPH reduces the aldehyde to form 3HP. Previous studies established both MCR-C and MCR-N domains are in the short-chain dehydrogenases/reductase (SDR) family of enzymes [14, 15]. Enzymes in the SDR family are oxidoreductases that can catalyze reduction of C=C and C=N double bonds, as well as reduction of carbonyl/hydroxyl groups [16]. Members of the SDR family of enzymes have a canonical triad in the active site composed of residues Tyr-Lys-Ser [16].

A method for producing 3HP with a lower carbon footprint is overexpression of the gene for MCR in Escherichia coli or another microorganism. This biological approach allows for the use of renewable energy sources derived from biomass such as glycerol, glucose, or fatty acids. While a bioengineering approach to 3HP synthesis offers environmental benefits, that potential has not been realized because of several technical barriers such as the low catalytic efficiency of the MCR enzyme [17]. In fact, the rate of 3HP production increased when MCR was separated into the two domains [18]. The activity of the C-terminal part of malonyl-CoA reductase (malonyl-CoA to malonic semialdehyde) is 4–5 times lower than the activity of the N-terminal part of malonyl-CoA reductase (malonic semialdehyde to 3HP) and this prevents higher accumulation of 3HP [18]. The expression rates of both parts were optimized to increase the production of 3HP to the highest yield reported of 40.6 g/L with the productivity 0.56 g/L/h in E. coli [19]. The MCR-C was optimized by random mutagenesis producing a more active triple mutant without a significant change in the K_m for malonyl-CoA, while the MCR-N enzyme gene was incorporated into the chromosome to decrease its expression. One approach to make the biological production of 3HP commercially viable is to increase the yield of 3HP by improving the properties of the catalyst MCR. Enhancing the catalytic properties of MCR begins with a thorough understanding of the kinetics of the enzyme. A detailed kinetic analysis of the N-terminal domain of MCR has been recently reported [20]. Thus, to complete the kinetic characterization of MCR this report describes a rigorous kinetic study of the catalytic mechanism of the C-terminal domain of MCR. A determination of the detailed kinetic parameters will afford the ability to generate hypotheses for improvement of the catalytic rates. For example, if product inhibition is significant, it becomes a place to focus. This should guide new research focused on rational mutations within the MCR-C domain to enhance the catalytic efficiency of MCR.

2. Materials and methods

2.1. Materials

Yeast extract and tryptone were from either Gibco^™ or Fisher Scientific. Kanamycin, IPTG, and HEPES were obtained from Gold Biotechnology. Tris was from VWR. Ni-NTA resin was obtained from Qiagen. All other chemicals were from Sigma Aldrich unless otherwise stated.

2.2. Methods

2.2.1. The construction of the plasmid expressing the C-terminal part of MCR.

The plasmids used in this study can be found in Table 1. The construction of the plasmid pEMP11 has been previously described [20]. The plasmid for Thermoanaerobium brockii alcohol dehydrogenase was obtained directly from VectorBuilder Inc.

Table 1.

Plasmids used in this study. All plasmids code for protein bearing a N-terminal-6xHis-tag. Chloroflexus aurantiacus MCR genes are all inserted in the pET28b vector which has kanamycin resistance. TbADH is in a pET vector with ampicillin resistance.

Plasmid	Description	Source
pEMP11	malonyl-CoA reductase (Uniprot A9WIU3_CHLAA)	[20]
pMTC3	C-terminal part of malonyl-CoA reductase (residues 550–1219)	This study
pTbADH	Thermoanaerobium brockii alcohol dehydrogenase	VectorBuilder Inc.

The gene for the C-terminal part of MCR (MCR-C) was cloned into pMTC3 using the primers in Table 2. Polymerase chain reactions for the MCR-C gene (using the primers MCR_C_F and MCR_C_R), and the pET28b plasmid (using the primers pET28b_V_F and pET28b_V_R) were performed with an annealing temperature of 63 °C. The template plasmid pEMP11 for the MCR-C PCR reaction was linearized using NdeI (New England Biolabs) before the MCR-C PCR reaction. The PCR products were confirmed on a 1% agarose gel, incubated overnight at 37°C in the presence of DpnI (New England Biolabs) for template removal, and the PCR products were purified.

Table 2.

Primers used in the construction of genes for the C-terminal part of malonyl-CoA reductase. Lower case letters correspond to the pET28b plasmid sequence, while capital letters correspond to the MCR gene sequence. Start and stop codons are underlined in forward and reverse primers, respectively.

Primer	Primer sequence	Tm/°C
MCR_C_F	5′ gtgccgcgcggcagccatATGATTAGCGCGACCACGGGCGCA 3′	81.3
MCR_C_R	5′ CTCGAGTGCGGCCGCAAGCTTTTAgacggtaatcgcacgaccgcgg 3′	79.9
pET28b_V_F	5′ aagcttgcggccgcactcgag 3′	68.4
pET28b_V_R	5′ catatggctgccgcgcggcac3 ′	70.4

The HiFi DNA assembly (New England Biolabs) reaction for pMTC3 was done at 50°C for 1 h with a plasmid to gene ratio of 1:2 and subsequently transformed into NEB5a chemically competent cells (New England Biolabs). The resulting colonies were tested for the presence of pMTC3 by colony PCR. An initial hold at 95°C for 5 minutes was used to lyse the cells in the presence of water and primers, with subsequent addition of OneTaq Hot Start Quick-Load MasterMix to initiate the PCR protocol. Plasmids with PCR products of appropriate size were purified, and the sequences were confirmed by DNA sequencing. Confirmed plasmids were used to transform E. coli BL21(DE3), in-house prepared chemically competent cells.

2.2.2. Protein expression and purification of MCR-C and TbADH.

Cultures of E. coli BL21(DE3) harboring a plasmid for expression of either MCR-C or TbADH protein, were grown in LB broth containing appropriate antibiotics (MCR-C:kan^R, TbADH:amp^R) overnight. The overnight cultures were used to inoculate Lennox LB in the presence of 50 µg/mL kanamycin (MCR-C) or 50 µg/mL ampicillin (TbADH) until reaching mid-log phase at 37°C with shaking at 110 rpm. Protein expression was induced with 0.1 mM IPTG and the culture was incubated at room temperature overnight with shaking. The cells were harvested by centrifugation at 5000×g for 10 min at 4°C. The pellet was resuspended in ice-cold buffer A (10 mM HEPES-KOH, 500 mM KCl, 10 mM imidazole, pH 8.0) and sonicated in the presence of DNase I. The cell lysate was clarified at 17,000×g at 4°C for 20 min. The lysate was loaded onto Ni-NTA column equilibrated with 60 column volumes (CVs) of ice-cold buffer A. The column was extensively washed with 30 CV of ice-cold buffer A, then 30 CVs of ice-cold buffer containing 10 mM HEPES-KOH, 150 mM KCl, 40 mM imidazole, pH 8.0, and the protein was eluted by ice-cold buffer 10 mM HEPES-KOH, 150 mM KCl, 250 mM imidazole, pH 8.0. The protein was first dialyzed using Thermo Scientific^™ SnakeSkin^™ Dialysis Tubing with 3.5 kDa cutoff and cellulose membrane against the following buffer 10 mM HEPES-KOH, 150 mM KCl, 1 mM EDTA, pH 8.0 and then against the buffer, 10 mM HEPES-KOH, 150 mM KCl, pH 8.0, before concentrating with MiliporeSigma Amicon with a 30 kDa cutoff. After the purity of the protein was assessed to be above 95% by discontinuous SDS-PAGE electrophoresis, the protein was aliquoted and stored at either −80°C, −20°C or used immediately. For kinetic analyses, aliquots from −80°C or −20°C were only used once. Protein concentration was determined using the Bradford assay.

2.2.3. Kinetic assays and pH studies for MCR-C activities.

The steady-state kinetic assay for the physiological (forward) reaction of MCR-C was measured by monitoring NADPH absorbance at 340 nm (ε_340nm= 6.22 mM⁻¹ cm⁻¹). All assays were carried out in 500 μL for 5 min using a Cary 60 (Agilent) spectrophotometer at room temperature except for the alternative substrate assays and the determination of pH profiles. The alternative substrates assays were carried out in 1000 μL using a UV-1650PC (Shimadzu) connected to circulating water bath at 55°C. Assays to determine pH profiles were conducted in 100 μL using a Spectra M2 microplate reader at room temperature by observing the absorbance change for 10 min. The reactions were started by the addition of the enzyme. Specific activity was determined as mM NADPH/min/mg of MCR-C. The MCR-C activity was measured in the buffers containing 5 mM MgCl₂, and either 100 mM HEPES-KOH (8.0) or 100 mM Tris-HCl (8.0), with no significant difference in the apparent kinetic parameters between the two buffers.

The apparent kinetic parameters of MCR-C (1 μM) for malonyl-CoA were determined at saturating concentrations of NADPH (0.3 mM), and varying concentrations of malonyl-CoA (20 – 1000 μM). The apparent kinetic parameters of MCR-C for NADPH were determined at saturating concentrations of malonyl-CoA (1 mM), and varying concentrations of NADPH (15 – 300 mM). The same saturating and varying concentrations for both substrates were used in the subsequent experiments as appropriate.

Product inhibition studies for MCR-C by NADP⁺ were performed at NADP⁺ concentrations of 0, 0.1, and 0.4 mM when NADPH was varied, and 0, 0.1, 0.4, and 0.8 mM when malonyl-CoA was varied. Product inhibition studies for MCR-C of CoA were done at CoA concentrations of 0, 0.4, and 0.6 mM when NADPH was varied, and 0, 0.2, and 0.8 mM when malonyl-CoA was varied. For dead-end inhibition studies with ADP-ribose, the inhibitor was used at concentrations of 0, 0.36, and 0.72 mM when NADPH was varied, and 0, 0.72, and 1.2 mM when malonyl-CoA was varied. For dead-end inhibition studies with glyoxylic acid, the inhibitor was used at concentrations of 0, 0.01, and 0.02 mM when NADPH was varied, and 0, 0.004, and 0.01 mM when malonyl-CoA was varied.

The analyses to determine the influence of pH on the MCR-C reaction were performed in the presence of malonyl-CoA and NADPH at appropriate concentrations in a three-component buffer system consisting of 0.1 M MES, 0.051 M N-ethylmorpholine, and 0.051 M diethanolamine. The pH was adjusted by using KOH. Over the pH range of 6.0–10.0 at which the initial velocities were measured, the ionic strength of the buffer mixture remained constant at a value of 0.1 M [21].

2.2.4. Nomenclature of isotope effects.

The nomenclature used here is according to Cleland [22] where the leading superscript denotes the isotope responsible for the effect on a given kinetic parameter. For example, the deuterium isotope effect on V/K is written ^D(V/K), and this symbol represents the V/K of the protiated species relative to the rate of the deuterium species [(V/K)_H/(V/K)_D]. Multiple isotope effects are written as a combination of superscripts and subscripts. Thus, ^D(V/K)_D2O is the deuterium isotope effect on V/K with D₂O as solvent. Substrates on which the isotope effect is measured are designated by a subscript within the brackets.

2.2.4.1. Solvent and double (solvent-deuterium) isotope effects.

Solvent isotope effects and double isotope effects were done in a D₂O:H₂O ratio of 8.5:1.5 with the same components as mentioned before in the 0.1 mM Tris-H(D)Cl buffer, pL 7.8, L being H or D (pD = pH + 0.41) [23].

The apparent kinetic parameters of MCR-C for the solvent isotope effects for malonyl-CoA were determined at saturating concentrations of NADPH (0.3 mM), and varying concentrations of malonyl-CoA (0.02 – 1 mM).

2.2.5. Synthesis of deuterated NADPH and the deuterium isotope effect.

Synthesis of deuterated pro-4S-NADPD and pro-4R-NADPD were done as previously published [24] using glucose-6-dehydrogenase (Sigma Aldrich) and our recombinantly expressed TbADH, respectively, and an additional step. After precipitation of NADPH with ethanol, the pellets were lyophilized to remove any residual ethanol. Protiated NADPH was synthesized the same way as its deuterated counterpart to account for any impurities that arise from the synthesis and subsequent workup.

The apparent kinetic parameters of MCR-C for malonyl-CoA were determined at saturating concentrations of NADPH(D) (0.3 mM), and varying concentrations of malonyl-CoA (0.02 – 1 mM).

2.2.6. Analysis of malonyl-CoA analogues.

The malonyl-oxa(dethia)CoA (M-O-CoA) and malonyl-aza(dethia)CoA (M-NH-CoA) analogues were previously synthesized [25]. The MCR-C (1 µM) reactions were run in 100 µL of 100 mM Tris pH 7.5, 2 mM MgCl₂ with 200 µM of malonyl-CoA or analogue and 100 µM NADPH for malonyl-CoA or 200 µM NADPH for the analogues. The reactions were stopped by adding formic acid at a final concentration of 25% (v/v) and protein was pelleted at 16,000×g for 5 min. The reactions (50 µl) were characterized by HPLC on an Agilent 1100 with diode array UV/Vis detection at 254 nm over a Luna 5 μm C18(2) 100 Å 250 × 4.6 mm (Phenomenex) with gradient elution at 1 ml/min from 0.5% TFA in water to 20% acetonitrile over 20 minutes. Peak areas corresponding to the malonyl-CoA or malonyl-CoA analog substrate and respective CoA products were integrated to determine % conversion.

For the analysis of MCR-C activity with malonyl-CoA analogues, the MCR-C (1 µM) reactions were run for 1h in a final volume of 1 mL at 55°C as described under 2.2.3 Kinetic assays and pH studies for MCR-C activities, in 10 mM HEPES-KOH, pH 8.0, but without added magnesium and in the presence of either 2 mM of the malonyl-CoA analog, or 1 mM malonyl-CoA.

2.2.7. Data analysis.

Kinetics data were analyzed by nonlinear regression analysis using the computer programs of Cleland [26]. Where one substrate concentration was varied at a fixed concentration of the other, the velocity data were fit to equation 1, where v is experimentally determined velocity, V is the maximal velocity, A and B are the substrate concentrations, K_a and K_b are the Michaelis constants for the varied substrate and constant substrate respectively, while K_ia is the dissociation constant of the varied substrate A.

$$v=\frac{V\text{AB}}{K_{\text{ia}}{K}_{\text{b}}+K_{\text{b}}\text{A}+K_{\text{a}}\text{B}+\text{AB}}$$ (1)

The inhibition data corresponding to competitive, noncompetitive, or uncompetitive inhibition pattern were fit to equation 2, 3, or 4, respectively, where A, K, v, and V are defined the same as for equation 1 while I is the inhibitor concentration, and K_is and K_ii are the slope and intercept inhibition constants, respectively.

$$v=\frac{V\text{A}}{K\left(1+\frac{\text{I}}{{\text{K}}_{\text{is}}}\right)+\text{A}}$$ (2)

$$v=\frac{V\text{A}}{K\left(1+\frac{\text{I}}{K_{\text{is}}}\right)+\text{A}\left(1+\frac{\text{I}}{K_{\text{ii}}}\right)}$$ (3)

$$v=\frac{V\text{A}}{K+\text{A}\left(1+\frac{\text{I}}{K_{\text{ii}}}\right)}$$ (4)

For the pH rate profiles, experimentally determined velocities (v) were obtained at each pH by varying the concentration of a substrate (A). The resulting data were fit to equation 5 to calculate the maximal velocity (V), and the Michaelis constant (K). The pH dependence of the values for the V/K and V pH rate profiles were calculated using equation 6. In equation 6, y represents V/K or V at a particular pH and C is the pH-independent value of the parameter. K is an acid dissociation constant associated with the ionizing group on the alkaline side of the pH profile.

$$v=\frac{V\text{A}}{K+\text{A}}$$ (5)

$$\log y=\log \frac{\text{C}}{1+\frac{K}{\left[{\text{H}}^{+}\right]}}$$ (6)

For determining the deuterium and solvent isotope effects, the data were fit to equation 7. V, A, and v were as previously defined. F_i is the fraction of the solvent or deuterium label in the substrate, and E_V/K and E_V are the isotope effects minus one for the V/K and V values, respectively. In this study, F_i is 1 for the deuterium label, and 0.85 for the solvent isotope effects.

$$v=\frac{V\text{A}}{K\left(1+{\text{F}}_{\text{i}}{\text{E}}_{\text{V}/\text{K}}\right)+\text{A}\left(1+{\text{F}}_{\text{i}}{\text{E}}_{\text{V}}\right)}$$ (7)

3. Results

3.1. Initial velocity patterns reveal sequential addition of substrates

Initial velocity patterns were obtained when either NADPH or malonyl-CoA concentrations were varied at different fixed levels of the other substrate. Data were fitted to equation 1 to yield the kinetic parameters summarized in Table 3. Double reciprocal plots for both substrates are shown in Figure 2 (A and B, respectively). When either substrate is varied, the lines intersect to the left of the vertical axis, suggesting a sequential kinetic mechanism, which means that both malonyl-CoA and NADPH must be present before any product release. Interestingly, the affinity of NADPH to the active site is 2.5 times higher than the affinity of MCoA to the active site. Furthermore, the turnover rate of MCR-C is 1.4 ± 0.5 per second at room temperature.

Table 3.

Kinetic parameters of the C-terminal part of malonyl-CoA reductase obtained by initial velocity patterns. MCoA stands for malonyl-CoA.

Enzyme	NADPH		MCoA		V / μmol min−1 mg−1	k cat /s −1	kcat/KNADPH (s−1/mM)	kcat/KMCoA (s−1/mM)
MCR-C	Ka/μM	Kia/μM	Kb/μM	Kib/μM
	30 ± 5	97 ± 16	75 ± 8	243 ± 55	1.21 ± 0.04	1.4 ± 0.5	46.7	18.7

Figure 2.

Initial velocity patterns of the C-terminal domain of MCR (MCR-C). (A) NADPH (0.03 – 0.3 mM) was varied at several fixed concentrations of malonyl-CoA (0.02 – 0.4 mM). (B) malonyl-CoA was varied (0.02 – 0.4 mM) at several fixed concentrations of NADP⁺ (0.03 – 0.3 mM). Measurements were done at 25°C. The data from both data sets were fit to equation 1 to obtain parameters summarized in Table 3. The units for v and [substrate] are µmol min⁻¹ mg⁻¹ and mM, respectively.

3.2. Product and dead-end inhibition

Product inhibition studies were used to determine the order of product released. NADP⁺ and CoA were assessed, whereas MSA is not commercially available. The data for the product and dead-end inhibition studies is summarized in Table 4. Measurements at two concentrations of each inhibitor were used in calculating the parameters in Table 4 by the best fit of data to equation 2, 3 or 4. Product inhibition by NADP⁺ exhibited competitive inhibition versus NADPH and was noncompetitive versus malonyl-CoA. Likewise, product inhibition by CoA exhibited competitive inhibition versus NADPH and was noncompetitive versus malonyl-CoA.

Table 4.

Inhibition patterns and constants for the C-terminal domain of malonyl-CoA reductase (MCR-C). All measurements were done at 25°C. n.a. not applicable

Inhibitor	Substrate varied	Pattern	Kii/μM	Kis/μM
NADP+	NADPH	competitive	n.a.	104 ± 15
NADP+	malonyl-CoA	noncompetitive	1019 ± 150	382 ± 57
CoA	NADPH	competitive	n.a.	238 ± 14
CoA	malonyl-CoA	noncompetitive	1879 ± 417	797 ± 218
ADP-ribose	NADPH	competitive	n.a.	1098 ± 140
ADP-ribose	malonyl-CoA	noncompetitive	2268 ± 180	2157 ± 420
glyoxylic acid	NADPH	uncompetitive	8.7 ± 0.3	n.a.
glyoxylic acid	malonyl-CoA	competitive	n.a.	6.0 ± 0.8

To determine the order of substrate binding, dead-end inhibition analyses of MCR-C were undertaken. ADP-ribose was chosen as a dead-end inhibitor for NADPH, due to its structural similarity. The inhibition with the ADP-ribose is competitive versus NADPH and noncompetitive versus malonyl-CoA (Table 4). For malonyl-CoA, glyoxylic acid was used as a dead-end inhibitor because it is a slow alternate substrate for MCR-C (see later in text). A slow alternate substrate can be used as a dead-end inhibitor if the activity of the reaction is less than 2% compared to the natural substrate [27]. The activity of MCR-C with glyoxylic acid is 1% the activity of malonyl-CoA. Inhibition with glyoxylic acid was competitive versus malonyl-CoA and uncompetitive inhibition versus NADPH (Table 4).

3.3. pH rate profiles

The effect of pH on the kinetic parameters of MCR-C was assessed to gain insight into the chemical mechanism of catalysis. The pH dependence of the kinetic parameters V and V/K was determined by varying either NADPH over the pH range of 6.1–10.0 or malonyl-CoA over the pH range of 6.1–9.8. When NADPH was the varied substrate and malonyl-CoA was held constant at 1 mM, the activity for the V/K_NADPH parameter decreased with increasing pH (Figure 3 A). The data were fitted to equation 6 to yield a pK value of 8.6 ± 0.1. When malonyl-CoA was varied, and NADPH was held constant at 0.3 mM, the activity for the V/K_MCoA parameter also decreased with increasing pH (Figure 3 B). The data were fitted to equation 6 to yield a pK value of 8.9 ± 0.1. Lastly, the pH dependence of the maximal velocity, V, decreased with increasing pH. The data were fitted to equation 6 to yield a pK of 8.8 ± 0.1 the V profile. The ionizable group observed in each of these profiles is probably acting as a general acid since it must be protonated for maximal MCR-C activity. No significant decrease in activity was detected at pH levels as low as 6.0.

Figure 3.

The pH dependence of V/K_NADPH (A), V/K_MCoA (B) and V (C) of MCR-C activity. The apparent kinetic parameters of MCR-C at each pH for NADPH were determined at saturating levels of 1 mM malonyl-CoA, and varying concentrations of NADPH (0.015– 0.3 mM). The apparent kinetic parameters of MCR-C at each pH for malonyl-CoA were determined at saturating levels of 0.3 mM NADPH, and varying concentrations of malonyl-CoA (0.02 – 1 mM). Solid lines represent a fit to equation 6 and the calculated pK is written near each curve. The units for V and V/K are mM min⁻¹ mg⁻¹ and min⁻¹ mg⁻¹, respectively.

3.4. Stereospecificity of hydrogen transfer from NADPH

NAD(P)H has two hydrogen atoms at position C4 of the nicotinamide ring. These two hydrogen atoms are in the pro-R and pro-S positions. Every NAD(P)H-dependent family of enzymes uses one or the other. Since MCR-C belongs to the SDR family, it probably prefers the pro-S hydrogen [15]. To test the stereospecificity of MCR-C, pro-S and pro-R NADPD at position C4 were made, using glucose-6-dehydrogenase from yeast and alcohol dehydrogenase from Thermoanaerobium brockii (TbADH), respectively. As a control, protiated NADPH was synthesized in the same way as the deuterated nucleotide. When malonyl-CoA was varied at 0.3 mM of pro-4R-NADPD, the values for ^DV/K and ^DV were 1.06 ± 0.06 and 1.13 ± 0.03, respectively. By contrast, when malonyl-CoA was varied at 0.3 mM of pro-4S-NADPD, the values obtained for ^DV/K and ^DV were 1.92 ± 0.02 and 1.63 ± 0.07, respectively. Since no significant deuterium effect was observed with pro-4R-NADPD and relatively significant deuterium effects were observed with pro-4S-NADPD, MCR-C likely catalyzes the transfer of a hydride in the pro-S orientation, in accordance with other members of the SDR family of enzymes [16]. The pro-S stereoisomer of NADPD was used for all subsequent isotope experiments.

3.5. Kinetic isotope effects

Isotope effects, both deuterium and solvent isotope effects were determined to gain insight into the rate determining step of the reaction catalyzed by MCR-C. To determine both the deuterium and solvent isotope effects, the method of direct comparison of initial velocities was used.

For the deuterium isotope effects, the substrate NADPD was held constant at a saturating concentration and malonyl-CoA was varied at pH 7.8. The ^D(V/K) value was 1.92 ± 0.02, while ^DV was 1.63 ± 0.07 (Table 5).

Table 5.

Approximate lower and upper limits for C_f and C_VR in H₂O.

Value for Dk5	Cf	CVR
2.5	0.67	1.5
2.75	0.94	1.92

The deuterium isotope effects were also determined at pH 9.0, which is on the descending limb of the pH rate profiles (Figure 4). At pH 9.0, the isotope effect on ^D(V/K) decreased to 1.0, while the isotope effect on ^DV decreased to 1.24 ± 0.02.

Figure 4.

HPLC chromatograms of the MCR-C reaction with A) malonyl-CoA (as a control) after 10 min, and of the MCR-C reactions with B) malonyl-O-CoA and C) malonyl-NH-CoA after 12 h.

The solvent isotope effect was carried out using a 0.85 mol fraction of D₂O. The substrate NADPH was held constant at a saturating concentration while malonyl-CoA was varied. The solvent isotope effect on ^D2O(V/K) was 1.4 ± 0.2, while ^D2OV was 1.27 ± 0.05.

A multiple solvent isotope effect was carried out using a constant concentration of NADPD and varying concentrations of malonyl-CoA. Initial velocities were measured in H₂O and in D₂O. The data were fitted to equation 7. The double solvent isotope effect was calculated by dividing the kinetic parameter, V or V/K, measured in water by the corresponding value measured in deuterium oxide. The double solvent isotope effect on ^D₂^OV_D was 1.24 ± 0.04, while ^D₂^O(V/K)_D was 1.

The multiple deuterium isotope effect was determined in D₂O. The initial velocity was measured when NADPH was held constant at the varying concentrations of malonyl-CoA and when NADPD was held constant at the varying concentrations of malonyl-CoA. The data were fit to equation 7. The double deuterium isotope effect was calculated by dividing the kinetic parameter measured with NADPH by the corresponding value measured with NADPD. The double deuterium isotope effect on ^D(V/K)_D2O was 1.7 ± 0.2, while ^DV_D2O was 1.43 ± 0.04.

In summary, both solvent and deuterium isotope effects decreased in the presence of the other heavy atom which suggest they do not occur on the same step [22].

3.6. Specificity of MCR-C reaction using malonyl-CoA analogs

In order to determine the reductive capacity, we tested MCR-C with analogs where the thioester is replaced by an ester or amide [25]. The M-O-CoA analog has an oxygen in place of sulfur to form an oxygen ester. In the M-NH-CoA analog, the sulfur is replaced with nitrogen to form an amide. Either of the two analogs or malonyl-CoA were incubated with MCR-C and the reactions were analyzed by HPLC. The M-O-CoA analog showed a peak corresponding to oxa(dethia)CoA, while the M-NH-CoA analog did not show an aza(dethia)CoA peak on the HPLC chromatogram (Figure 4). Moreover, the area of the M-NH-CoA peak does not change significantly as a function of time. A small peak at 6.25 min is observed in the chromatograms for both analogs, which corresponds to NADPH.

In addition to the HPLC analysis of the reaction, the reduction of the malonyl-CoA analogs by NADPH was examined by monitoring the absorbance at 340 nm. This analysis showed 86 ± 6% residual activity with M-O-CoA compared with malonyl-CoA, while the M-NH-CoA analog showed 2.5 ± 0.2% residual activity which is consistent with the background level of NADPH oxidation. Thus, both HPLC and spectroscopic analyses revealed that M-O-CoA is turned over by MCR-C, but at a slower rate than malonyl-CoA, while with M-NH-CoA no turnover is observed. The relative reactivity can be explained by the inferior ability of the sulfur to stabilize against protonation and reduction of the thioester ketone, compared to an ester or amide which is highly resonance stabilized. These experiments will help elucidate which analogs might be useful in follow-up structural studies.

3.7. Structure-activity relationship of MCR-C

The previous results indicated that the free carboxyl group of malonyl-CoA is not involved in the chemistry of the reaction, therefore, the question arises as to what role, if any, does it play in the MCR-C mechanism of action? To investigate the role of the free carboxyl group, different structural analogs of the malonyl moiety of malonyl-CoA were tested as alternate substrates. According to the data in Figure 5, both the size and the presence of the carboxyl group are essential for the MCR-C reaction.

Figure 5.

Activity of the C-terminal part of malonyl-CoA reductase (MCR-C) with structural analogues of malonyl-CoA, and their structures for comparison. The concentrations of malonyl-CoA and the analogues were 2 mM, while the concentration of NADPH was 0.3 mM. The data shows the average of three replicate measurements at 55°C.

Formic acid, the smallest acid of the compounds tested, does not get reduced even though it possesses a carboxyl group. Glyoxylic acid, which is one methylene group shorter than the malonic acid moiety of malonyl-CoA, shows 1% residual activity of malonyl-CoA. In contrast, acetyl-CoA has hydrogen in place of carboxyl group and does not show activity with MCR-C. Propionyl-CoA has the same number of carbons as malonyl-CoA, minus the carboxyl group, and does not show activity with MCR-C. This indicates that the carboxyl group is important for catalysis and plays a role in the proper positioning of the substrate thioester ketone for hydride transfer.

4. Discussion

4.1. Initial velocity patterns and inhibition studies

The initial velocity patterns indicated that the MCR-C enzyme has a sequential kinetic mechanism, which means that both substrates must bind to the enzyme before any chemical reaction can occur. This is in contrast to the archaeal counterpart of MCR-C where the acyl moiety of malonyl-CoA forms a covalent intermediate with an active site cysteine, which may manifest as a ping-pong initial velocity pattern [28]. The sequential kinetic mechanism for Chloroflexus aurantiacus MCR-C would be consistent with either ordered or random addition of substrates to the enzyme.

A dead-end inhibitor for malonyl-CoA was needed to establish whether substrate binding was ordered or random. Glyoxylic acid is a product of the metabolic pathway in which MCR plays a role, and it was found to be a slow alternate substrate of the MCR-C enzyme. Since glyoxylic acid reacted at 1% the activity of malonyl-CoA, it was used as a dead-end inhibitor. Glyoxylic acid exhibits competitive inhibition versus malonyl-CoA and uncompetitive inhibition versus NADPH. This combination of inhibition patterns suggests that MCR-C has an ordered addition of substrates with NADPH binding first (Figure 7). ADP-ribose was used as a dead-end inhibitor of NADPH. As expected, ADP-ribose exhibited competitive inhibition versus NADPH and noncompetitive inhibition versus malonyl-CoA which is consistent with the ordered binding of substrates shown on Figure 7. The noncompetitive inhibition of ADP-ribose versus malonyl-CoA is also consistent with the sequential initial velocity pattern. If the kinetic mechanism was ping-pong (i.e., involved a covalent intermediate), the inhibition pattern of ADP-ribose would have been uncompetitive.

Figure 7.

Electron density and distances for the product ligands of Roseiflexus castenholzii MCR-C (PDB ID 8hi5). NADP+ is in white and malonate semialdehyde is in cyan. The black dashes indicate favorable interactions with the product aldehyde. The red dash indicates a close contact. The distance between C4 hydrogen of NADP⁺ to the thioester carbon (4.2 Å) and to the carboxyl carbon (6.0 Å) of malonic semialdehyde (MSA) in the C-terminal part of malonyl-CoA reductase (MCR-C) from Roseiflexus castenholzii (PDB ID 8hi5) [29]. NADP⁺ molecule is on the left and up, and the malonic semialdehyde is on the bottom side of the figure.

The competitive inhibition patterns observed for NADP⁺ and CoA versus NADPH suggested both products compete for the same enzyme form. Since NADPH binds first, this indicates that both NADP⁺ and CoA bind to the free enzyme. The possible explanation for these results is that NADP⁺ and CoA are released from the enzyme in a random fashion.

The noncompetitive inhibition patterns observed for NADP⁺ and CoA versus malonyl-CoA indicates those products bind to a different enzyme form than malonyl-CoA. Combining this observation with the fact that NADP⁺ and CoA bind to the free enzyme suggests that MSA is released first from the enzyme and is followed by the random release of NADP⁺ and CoA. Therefore, the kinetic mechanism of MCR-C can be written as shown in Figure 6.

Figure 6.

Order of substrate binding and product release for the C-terminal part of malonyl-CoA reductase (MCR-C). E represents MCR-C, MCoA is malonyl-CoA, and MSA is malonic semialdehyde.

The random release of the substrates CoA and NADP⁺ is reinforced by the structural analyses of the MCR-C enzyme. The MCR-C enzyme structures were determined in complex with either CoA or NADP⁺, but not the MSA substrate [14]. This observation is consistent with the proposed random release of the CoA and NADP⁺ substrates because it shows they both can bind to free MCR-C enzyme. On the other hand, the structure in complex with MSA is only solved along with the NADP⁺ substrate present in the active site. This reinforces the notion that MSA is released first, followed by the random release of NADP⁺ and CoA [14, 29].

4.2. Structure-activity relationship of MCR-C reaction.

The steady-state kinetic studies indicate the catalytic mechanism of the MCR-C enzyme does not involve a covalent intermediate but instead occurs via direct hydride transfer from NADPH to malonyl-CoA. The question is which carbonyl carbon of the malonyl moiety of malonyl-CoA is reduced, the carboxyl carbon or the thioester carbon? The reduction of the malonyl-CoA analog M-O-CoA suggested that the thioester carbon is being reduced. If the carboxyl carbon were to be reduced, we would expect to observe turnover in the MCR-C reaction with M-NH-CoA, as the presence of the amide group would not terminate the reduction of this analog on the carboxyl carbon. Precedent for the thioester carbon being reduced without formation of a covalent intermediate is the reaction catalyzed by 3-hydroxy-3-methylglutaryl-CoA reductase (HMG-CoA reductase) [30–32]. Further support for thioester carbonyl reduction is a structure of MCR-C from Roseiflexus castenholzii [29] that was modeled with malonic semialdehyde and NADP⁺ bound, Figure 7. Based on the modest density for malonic semialdehyde product (which was not included in the crystallization buffer) the authors suggest that the canonical active site Ser and Tyr interacts with the thioester oxygen. Furthermore, their model has a distance between the C4 hydrogen of the nicotinamide ring of NADP⁺ to the thioester carbon is shorter (4.2 Å) than to the carboxyl carbon (6.0 Å) (Figure 7).

If the carboxyl carbon is not being reduced, the question remains what role, if any, does it have in the reaction catalyzed by MCR-C? The structure-activity relationship studies presented here suggest that a free carboxyl is necessary for catalysis. Molecules without a carboxyl group showed no activity with MCR-C. The only molecule with a carboxyl group, glyoxylic acid, did exhibit activity, albeit very slowly. Thus, a carboxyl group appears to be required for catalysis. The structure of MCR-C from R. castenholzii with MSA bound shows an interaction between an arginine and the carboxyl group of MSA (Figure 7) [29]. The arginine is in a domain that undergoes a conformational change upon binding of malonic semialdehyde suggesting the role of the carboxyl group of malonic semialdehyde and, by inference, malonyl-CoA is to properly position the substrate in the active site for catalysis.

4.3. pH profiles

The pH profiles for V/K_NADPH, V/K_MCoA and V show that a single ionizable group must be protonated for catalysis with a pK value between 8.6 – 8.9. The similar shapes and pK values for all three profiles suggest the same ionizable group is being observed. Since V/K pH profiles determine ionizable groups involved in binding and catalysis, while V pH profiles detect ionizable groups involved in catalysis only, the group observed in all three profiles likely plays a catalytic role and is not involved in binding. Moreover, neither NADPH nor malonyl-CoA have any ionizable groups with a pK of 8.6–8.9 [33, 34]. Enzymes in the SDR family contain a canonical triad of residues, tyrosine, lysine and serine in the active site. Mutagenesis of any of the three residues to alanine abolishes enzymatic activity indicating their significance in the catalytic mechanism [29]. Two of the residues, tyrosine, and lysine, have ionizable groups with pK values near the pK values observed in the pH profiles. The tyrosine residue is postulated to stabilize the oxyanion on the carbonyl group that is being reduced and is the most conserved residue in the SDR superfamily of the enzymes [15, 16], while lysine plays a role in stabilizing the incipient negative charge on the tyrosyl oxygen via an electrostatic interaction. Since both the tyrosine and lysine must be in their protonated state for catalysis, either one could correspond to the ionizable group observed in the pH rate profiles.

4.4. Interpretation of kinetic isotope effects

Based on the initial velocity pattern and inhibition studies, the kinetic mechanism for the MCR-C enzyme can be written as shown in Figure 8. The isotope sensitive step is k₅, while rates for the release of all three products are simplified by combining them into a single macroscopic rate constant k₆.

Figure 8.

Kinetic mechanism for the C-terminal part of malonyl-CoA reductase (MCR-C).

For the deuterium isotope effects, the concentration of NADPH was held constant, and the concentration of malonyl-CoA was varied. In that case, ^D(V/K) can be described by equation (8)

$$D_{\left(V/K\right)}=\frac{{\text{D}}_{k_5+\frac{k_5}{k_4}}}{1+\frac{k_5}{k_4}}$$ (8)

where the ratio k₅/k₄ is the commitment to catalysis (C_f) which is the ratio of the forward rate of catalysis (k₅) to the rate constant (k₄) for dissociation of malonyl-CoA.

For the kinetic mechanism in the Figure 9, the ^DV isotope effect is described by equation (9)

$${\text{D}}_V=\frac{{\text{D}}_{k_5+\frac{k_5}{k_6}}}{1+\frac{k_5}{k_6}}$$ (9)

where the ratio k₅/k₆ is the ratio to catalysis (C_VR) which is the ratio of the forward rate of catalysis (k₅) to the rate of product release (k₆).

Figure 9.

The proposed kinetic mechanism for the C-terminal part of malonyl-CoA reductase from Chloroflexus aurantiacus.

The calculation of C_f and C_VR requires a value for the intrinsic isotope effect, ^Dk₅. The intrinsic isotope effect is the maximum value for the isotope effect and would be observed only when the isotope sensitive step is completely rate-limiting. The intrinsic isotope effect for the reaction catalyzed by MCR-C is not known. Moreover, the intrinsic isotope effect for pyridine nucleotide catalyzed reactions varies widely between 2.5 and 5.7 [35].

In the absence of a value for the intrinsic isotope effect, the relative values for the C_f and C_VR can be determined. The K_b and K_D for a substrate are related to ^DV and ^D(V/K_b) by equation 11 [34].

$$\frac{{\text{D}}_{V-1}}{\text{D}\left(\frac{V}{{\text{K}}_b}\right)-1}=\frac{{\text{K}}_b}{{\text{K}}_D}$$ (11)

In equation 11, K_b and K_D are the Michaelis constant and dissociation constant for malonyl-CoA, respectively. Using the values for ^DV, ^D(V/K_b) and K_b from the initial velocity pattern a value of 112.5 μM for the K_D was calculated.

If the K_D of a substrate is greater than the K_b, it means the substrate is not “sticky” and it dissociates from the enzyme faster than it reacts to form products [36]. Moreover, because ^D(V/K_b) is greater than ^DV, and K_b is less than K_D, the step following the isotope sensitive step is slow, and C_f is smaller than C_VR. If malonyl-CoA is not sticky, then C_f is less than 1, which limits the range of values for the intrinsic isotope effect (Table 5).

The values for C_f show that the release of malonyl-CoA from MCR-C is 1.06 to 1.5 times faster than hydride transfer from NADPH to malonyl-CoA. The off-rate constant of the first substrate NADPH can be calculated using equation 12 where V/E_t is 1.43 s⁻¹ and K_ia and K_a are the dissociation constant and K_m for NADPH, respectively and were obtained from the initial velocity patterns. The off-rate constant k₂ of NADPH was calculated to be 4.62 s⁻¹ which is 3.2 times greater than the catalytic rate k₅. Thus, the data suggest that neither of the two substrates NADPH or malonyl-CoA are sticky. That is, the release of the substrates is greater than the forward catalytic rate k₅. Substrates that are not “sticky” are usually observed in enzymes with a random addition of substrates. However, when one substrate, in this case NADPH, binds tighter when malonyl-CoA is bound (the K_ia is greater than the K_a in Table 4) the binding of substrates is synergistic and manifest as an ordered addition of substrates in other kinetic analyses such as inhibition studies.

$$k_2=\frac{V{K}_{\text{ia}}}{E_{\text{t}}{K}_{\text{a}}}$$ (12)

Using the upper and lower intrinsic isotope effects calculated above for ^D(V/K), the limits of the ratio to catalysis (C_VR), can also be calculated. The values are shown in Table 5. These values show that the hydride transfer from NADPH to MCoA is 1.5–1.92 times faster than the release of the products CoA, NADP⁺ and malonic semialdehyde, indicating that product release is at least partially rate-limiting. Since CoA and NADP⁺ are released randomly, the release of those two products is not likely to be rate-limiting. Instead, the release of MSA is likely to be partially rate-limiting which is consistent with the fact that the C_VR is larger than C_f. Rate-limiting product release is commonly observed in dehydrogenases [36].

The observation of a KSIE on the reaction catalyzed by MCR-C suggests a proton is involved in catalysis and is partially rate-limiting. This is consistent with the pH profiles which showed a single ionizable group must be protonated for catalysis.

To investigate the relationship between the pH dependance and the isotope effects, the deuterium isotope effect was measured at pH 9.0 which was well below the pH optimum (Figure 4) and was the highest pH the isotope effect could be measured. The ^DV/K was 1, while ^DV was 1.2 ± 0.02. According to the seminal studies of Cook and Cleland, when the ^DV/K isotope effect decreases to 1 at the pH extremes, that means the pH dependence and the deuterium isotope step are independent of each other [37]. This suggests a stepwise mechanism with the pH dependent step occurring before or after the isotope sensitive step [36, 37]. At pH 9 the proton in the proton transfer step is lost, which commits the reaction to go forward rendering C_f equal to infinity. Moreover, when the pH dependent step and the isotope sensitive step are independent proton transfer has been proposed to be part of a rate-limiting conformational change [36].

The multiple isotope effects are consistent with the pH dependent step and the isotope sensitive step occurring in different transition states along the reaction coordinate. First, the value of ^D(V/K) in D₂O is lower than the value in H₂O suggesting a stepwise mechanism with the pH dependent step occurring before hydride transfer. Second, according to Pitsawong et al. if a double kinetic isotope effect does not equal the product of the deuterium and solvent isotope effects then the proton transfer and hydride transfer occur in different steps [38]. The double kinetic isotope effect for MCR-C was calculated by dividing the V from NADPH in H₂O water with the value for V from NADPD in D₂O water and equals 1.8. The product of the deuterium and solvent isotope effects is 2.2. Thus, the proton transfer and hydride transfer take place in different transition states.

4.5. Proposed catalytic mechanism of MCR-C

The kinetic studies presented here suggest a stepwise mechanism for catalysis by MCR-C. The pH dependent step could precede or follow the isotope hydride transfer step. We propose that the pH dependent step precedes hydride transfer (Figure 9). Comparison of structural analyses of holo MCR without substrates or products bound with the structure of MCR-C with the products NADP⁺ and MSA bound, shows that the canonical triad of the SDR family, Tyr-Lys-Ser, moves towards the products and, by inference, the substrates NADPH and malonyl-CoA upon substrate binding. Once the residues of the triad are in place to stabilize the oxyanion hydride transfer can occur. While stepwise mechanisms for SDR enzymes have been proposed previously mostly based on structural studies, to our knowledge, this is the first detailed kinetic evidence in support of a stepwise mechanism of this kind [29].

In contrast, the MCR-N domain employs a concerted mechanism where hydride transfer and proton transfer occur simultaneously within a single step. In this process, there is a complete transfer of a proton from the catalytic tyrosine to MSA, resulting in the formation of 3HP. However, in the MCR-C catalytic site, we propose a stepwise mechanism without direct proton transfer from the catalytic tyrosine. Nonetheless, the presence of the proton is crucial as it serves to stabilize the negative charge of the oxygen on the thioester group. The MCR-C reaction is more intricate due to the necessity for multiple products to exit the catalytic site and the occurrence of multiple reactions—namely, the reduction of the thioester carbon and the release of CoA by breaking the carbon-sulfur bond. These complexities may contribute to the observed differences in catalytic rates between the MCR-N and MCR-C domains.

4.6. Comparison of MCR-C to similar enzymes

Interestingly, malonyl-CoA/succinyl-CoA reductase is part of the 3-hydroxypropionate/4-hydroxybutyrate cycle for CO₂ fixation in Archaea. The archaeal counterpart solely catalyzes the reduction of malonyl-CoA to MSA [39], while the reduction of MSA to 3-hydroxypropionate (3HP) is performed by malonic semialdehyde reductase [40]. This reduction proceeds through an acyl-enzyme intermediate, involving a conserved cysteine residue in the active site, in a ping-pong mechanism where CoA and NADP⁺ share the same binding site [28, 39]. The K_m of the archaeal enzyme from the extremophile Sulfolobus tokodaii for malonyl-CoA is 40 μM, and the K_m for NADPH is 25 μM [39]. The K_m for NADPH is quite like that for MCR-C (30 μM), although the K_m for malonyl-CoA for MCR-C is slightly higher at 75 μM. The archaeal enzyme has a turnover rate (k_cat) of 26 s⁻¹ for succinyl-CoA and 28 s⁻¹ for malonyl-CoA at 65°C, with an optimal temperature of 90°C [40]. While MCR-C was assayed at room temperature, making direct comparisons of turnover rates difficult, the k_cat of MCR-C at its optimal temperature of 57°C has been determined as 18.6 s⁻¹ [10]. Although MCR-C has a slightly lower turnover rate, when it is fused with MCR-N, the overall reaction turnover rate increases to 50 s⁻¹ at the optimum temperature of 57°C [10]. This enhancement possibly occurs because MCR-N shifts the equilibrium towards MSA and NADP⁺ by depleting MSA, as it has a k_cat that is 3.5 times higher than MCR-C [20] which is consistent with a previous study that shows that MCR-C is a rate-limiting step in the MCR enzyme [18].

The detailed kinetic analyses of MCR-C described here, and MCR-N described previously [20] suggest the evolutionary rationale for the higher activity of the N-terminal domain compared to MCR-C is to decrease the likelihood that the highly reactive MSA intermediate escapes from the whole MCR complex. The kinetic study of MCR-C in this report showed that release of MSA is partially rate-limiting. From an evolutionary point of view this makes sense but from a biotechnology view this is a hindrance. Therefore, mutations directed at eliminating the slow product release and increasing the overall production of 3HP can now be explored.