Structural Insight into Dihydrodipicolinate Reductase from Corybebacterium glutamicum for Lysine Biosynthesis

Abstract

Dihydrodipicolinate reductase is an enzyme that converts dihydrodipicolinate to tetrahydrodipicolinate using an NAD(P)H cofactor in L-lysine biosynthesis. To increase the understanding of the molecular mechanisms of lysine biosynthesis, we determined the crystal structure of dihydrodipicolinate reductase from Corynebacterium glutamicum (CgDapB). CgDapB functions as a tetramer, and each protomer is composed of two domains, an Nterminal domain and a C-terminal domain. The N-terminal domain mainly contributes to nucleotide binding, whereas the C-terminal domain is involved in substrate binding. We elucidated the mode of cofactor binding to CgDapB by determining the crystal structure of the enzyme in complex with NADP⁺ and found that CgDapB utilizes both NADH and NADPH as cofactors. Moreover, we determined the substrate binding mode of the enzyme based on the coordination mode of two sulfate ions in our structure. Compared with Mycobacterium tuberculosis DapB in complex with its cofactor and inhibitor, we propose that the domain movement for active site constitution occurs when both cofactor and substrate bind to the enzyme.

Introduction

Corynebacterium glutamicum is a gram-positive and nonpathogenic bacterium that is widely used for the industrial production of amino acids, such as L-glutamate and L-lysine [5,7]. The complete genome sequence of C. glutamicum provides a large amount of information on its metabolic pathways and leads to the increased productivity [4,6]. L-Lysine is synthesized from L-aspartate in multiple steps. Aspartokinase catalyzes the phosphorylation of L-aspartate into L-aspartyl-phosphate, with the subsequent conversion of L-aspartyl-phosphate to L-aspartate-semialdehyde (ASA) by aspartate-semialdehyde dehydrogenase. These steps are common to the biosynthesis of lysine, threonine, isoleucine, and methionine. The condensation between ASA and pyruvate, catalyzed by dihydrodipicolinate synthase, produces dihydrodipicolinate (DHDP), which is then reduced to tetrahydrodipicolinate (THDP) with the cofactor NAD(P)H by DHDP reductase. At this point, four different pathways are available [15]. Most bacterial species convert THDP to meso-diaminopimelate (meso-DAP) through the succinylase pathway, which includes four enzymes. Some Bacillus species utilize the acetylase pathway and dehydrogenase pathway to directly form meso-DAP, the precursor of lysine and an essential component of peptidoglycan in bacterial cell walls [14]. The aminotransferase pathway was recently identified [11]. In a few organisms, such as C. glutamicum and Escherichia coli, two lysine biosynthetic pathways are present. Finally, meso-DAP decarboxylase catalyzes the conversion of meso-DAP to L-lysine [17,20].

Dihydrodipicolinate reductase (DapB; E.C. 1.17.1.8) catalyzes the reduction of DHDP to THDP and uses NADH or NADPH as a cofactor (Fig. 1A). The cofactor binds to Rossmann folds comprising a series of alternating β-strands and α-helices. Sequence comparison of various NAD(P)H-dependent enzymes revealed that the nucleotide binding motif is very similar and contains the highly conserved sequence G-x-x-G-x-x-G [16]. Typically, the pyridine nucleotide-dependent enzymes prefer one of the two cofactors and show a dual-cofactor specificity, but utilize NADPH over NADH [9]. The preference for NADH or NADPH is correlated to the residues involved in stabilization of the adenosyl ribose ring. For NADH, glutamate and aspartate residues stabilize the adenosyl ribose hydroxyl group and NADPH-dependent enzymes have positively charged residues, such as arginine and lysine, which interact with the 2’-phosphate monoester [8,18]. The enzymes exhibiting dual specificity have both acidic and basic residues in the adenosyl ribose-binding pocket. DapB from Mycobacterium tuberculosis utilizes dual cofactors and shows a greater preference for NADH [2]. Compared with M. tuberculosis DapB, we hypothesized the cofactor specificity of CgDapB. Here, we present the crystal structure of DapB from C. glutamicum (CgDapB) in the apoform and in complex with its cofactor NADP+. We also report the cofactor specificity of CgDapB and domain movement upon cofactor and substrate binding.

Fig. 1.Catalytic reaction and purification of CgDapB. (A) Reaction catalyzed by CgDapB. (B) SDS–PAGE of purification of recombinant CgDapB protein. Lane 1 contains molecular weight markers (labeled in kDa). Lanes 2–8 show the purification procedure of CgDapB using Ni–NTA chromatography. Lane 2, whole-cell extract; lanes 3 and 4, pellet fraction and supernatant after centrifugation of whole-cell extract, respectively; lane 5, flow-through from Ni–NTA column; lanes 6 and 7, wash with 5 and 20 mM imidazole, respectively; lane 8, elution with 300 imidazole. The purified CgDapB is indicated on the right side of the figure with an arrow.

 

Materials and Methods

Preparation of CgDapB

The forward and reverse primers had the following sequences: 5’-GCGCGCATATGGGAATCAAGGTTGGCGTTC-3’ and 5’-GCGCGCTCGAGTTACAGGCCTAGGTAATGC-3’ to introduce NdeI and XhoI restriction sites, respectively. The CgDapB coding gene (Met1-Leu248, MW 29 kDa) was amplified by polymerase chain reaction (PCR) using C. glutamicum chromosomal DNA as a template. The PCR product was then subcloned into pET30a (Life Science Research) with 6-histag at the C-terminus. The expression construct was transformed into an Escherichia coli B834 strain, which was grown in 1 L of LB medium containing kanamycin (50 mg/ml) at 37℃. After induction via the addition of 1.0 mM IPTG, the culture medium was further maintained for 20 h at 18℃. The culture was harvested by centrifugation at 5,000 ×g at 4℃. The cell pellet was resuspended in buffer A (40 mM Tris–HCl at pH 8.0 and 5 mM BME) and then disrupted by ultrasonication. The cell debris was removed by centrifugation at 11,000 ×g for 40 min, and lysate was bound to Ni-NTA agarose (Qiagen). After washing with buffer A containing 30 mM imidazole, the bound proteins were eluted with 300 mM imidazole in buffer A. A trace amount of contamination was removed by applying HiLoad 26/60 Superdex 200 prep grade (GE Healthcare) size-exclusion chromatography. The purified protein showed ~97% purity on SDS-PAGE, and was concentrated to 60 mg/ml in 40 mM Tris-HCl, pH 8.0, 1 mM dithiothreitol.

Crystallization and X-Ray Crystallographic Analysis of CgDapB

Crystallization of the purified CgDapB protein was initially performed with commercially available sparse-matrix screens from RIGAKU and Molecular Dimensions using the hanging-drop vapor-diffusion method at 20℃. Each experiment consisted of mixing 1.0 μl of protein solution (90 mg/ml in 40 mM Tris-HCl, pH 8.0) with 1.0 μl of reservoir solution and then equilibrating against 50 μl of reservoir solution. CgDapB crystals were observed from several crystallization screening conditions. After several steps that improved the crystallization process using the hanging-drop vapor-diffusion method, crystals of the best quality appeared in 2.0 M ammonium sulfate, 5% 2-propanol. The crystals were transferred to cryoprotectant solution containing 2.0 M ammonium sulfate, 5% 2-propanol, and 30% (v/v) glycerol, finished out with a loop larger than the crystals and flash-frozen by immersion in liquid nitrogen. The data were collected to a resolution of 2.5 Å at 7A beamline of the Pohang Accelerator Laboratory (PAL, Pohang, Korea), using a Quantum 270 CCD detector (ADSC, USA). All data were indexed, integrated, and scaled together using the HKL2000 software package [13]. The crystals of CgDapB apo-form belonged to the space group I4122 with unit cell parameters a = b = 107.39 Å, c = 175.67 Å. Assuming two molecules of CgDapB in an asymmetric unit, the crystal volume per unit of protein mass was 2.18 Å3 Da -1, which means the solvent content was approximately 43.69% [10]. CgDapB crystals in complex with NADP+ were crystallized with the crystallization condition of 1.85 M ammonium sulfate and 0.1 M sodium acetate trihydrate, pH 4.75, supplemented with 10 mM of NADP+. Crystals in complex with NADP+ belonged to the same space as CgDapB apo-crystals, with similar unit cell parameters. Assuming two molecules of CgDapB per asymmetric unit, the crystal volume per unit of protein mass was 2.16 Å3 Da -1, which corresponds to a solvent content of approximately 43.12%.

Structure Determination of CgDapB

The structure of apo-form of CgDapB was determined by molecular replacement with the CCP4 version of MOLREP [19], using the structure of DapB from Mycobacterium tuberculosis (PDB code 1YL5) as a search model. Further model building was performed manually using the program WinCoot [3], and refinement was performed with CCP4 refmac5 [12] and CNS [1]. The structure of CgDapB in complex with NADP+ was solved by molecular replacement using the crystal structure of the apo-form of CgDapB. The data statistics are summarized in Table 1. The refined model of the apo-form of CgDapB and that in complex with NADP+ will be deposited in the Protein Data Bank with PDB codes of 5EER and 5EES, respectively.

Table 1.The numbers in parentheses are statistics from the highest resolution shell. Rsym = Σ|Iobs - Iavg| / Iobs, where Iobs is the observed intensity of individual reflection and Iavg is the average over symmetry equivalents. Rwork = Σ||Fo| - |Fc|| / Σ|Fo|, where |Fo| and |Fc| are the observed and calculated structure factor amplitudes, respectively. Rfree was calculated with 5% of the data.

 

Results and Discussion

Overall Structure of CgDapB

To determine the molecular mechanism of dihydrodipicolinate reductase from C. glutamicum (CgDapB), we purified, crystallized, and determined the crystal structure of the enzyme at 2.5 Å resolution (Fig. 1B). The crystal belonged to the I4122 space group, and the asymmetric unit of the crystal contained two CgDapB molecules. The overall structure of CgDapB was homologous to that of dihydrodipicolinate reductase from Mycobacterium tuberculosis (MtDHPR), and the amino acid sequence identity between CgDapB and MtDHPR was 61% (Fig. 2A). The CgDapB monomer consists of two distinct domains, the N-terminal domain (NTD) and the C-terminal domain (CTD). The NTD (Met1–Phe107 and Asn219–Leu248) comprises seven parallel β-strands (β1–β6 and β12) and six α-helices (α1–α5 and α8) and is mainly involved in the binding of the NAD(P)H cofactor. The center of the NTD contains a 7-stranded β-sheet, with six α-helices covering both sides of the sheet. The CTD (Ala108–Arg218) comprises five β-strands (β7–β11) and two α-helices (α6 and α7). The five β-strands form an anti-parallel β-sheet and the two α-helices are located on the NTD side covering one side of the β-sheet (Fig. 2B). The domain constitutes the substrate binding site, which will be described later.

Fig. 2.Overall shape of CgDapB. (A) Alignment of amino acid sequences of DapB proteins. Secondary structure elements are shown and labeled based on the structure of CgDapB. Identical and highly conserved residues are presented in red and blue colored characters, respectively. The G-x-G-x-x-G motif is indicated by rectangles of a red color, and the N- and C-terminal domains are identified and labeled. Residues involved in the NAD+ and substrate binding are marked with purple and red colored triangles, respectively. (B) Monomer structure of CgDapB. NTD and CTD are distinguished with cyan and orange colors, respectively. NADP+ and two sulfate molecules bound in the enzyme are shown as stick models and labeled. (C) Tetramer structure of CgDapB. The tetramer structure of CgDapB is presented as a cartoon diagram. Mol I is presented with colors of cyan and orange for NTD and CTD, respectively. Other three molecules are shown with colors of yellow, green, and light-blue. NADP+ bound in the enzyme is shown as a stick model with magenta colors and labeled. The right side figure is a 90 degree rotation from the left side figure in horizontal direction.

The CgDapB tetramer was generated by crystallographic symmetry operation, and the results of size-exclusion chromatography confirmed that CgDapB functions as a tetramer (data not shown). The tetramer is exclusively constituted by interactions between the CTDs. First, to form a dimer, five β-strands from Mol I interact with those from Mol II, forming a 10-stranded β-sheet, and four α-helices (two from each monomer) covering one side of the β-sheet. The tetramer is then formed by interaction between the two dimers, where the two 10-stranded β-sheets of two dimers form a flattened 20-stranded β-barrel by face-to-face pairing (Fig. 2C).

NADP+ Binding Mode of CgDapB

To identify the NADP+ binding mode, we determined the crystal structure of CgDapB in complex with the NADP+ cofactor at 2.12 Å resolution. When we superimposed the NADP+-bound form of CgDapB with the apo-form of the enzyme, regional structural changes were observed, although overall the two structures were nearly identical. In particular, three connecting loops (β1-α1, β2-α2, and β3-α3) underwent structural transformations to stabilize the NADP+ molecule with a maximum distance of 3 Å (Fig. 3A). The NADP+ binding site was located at the G-x-x-G-x-x-G nucleotide binding motif, comprising residues Gly9-Ala10-Lys11-Gly12-Arg13-Val14-Gly15, and the hydroxyl groups of a pyrophosphate moiety were hydrogen-bonded with the main-chain nitrogen atoms of Arg13 and Val14. The pyrophosphate moiety was also stabilized by the side chain of Arg13. The nicotinamide and the ribose rings of NADP+ were stabilized through hydrogen-bond interactions mediated by the side chain of Thr79 and main chains of Thr79 and Ala104. The adenine moiety of NADP+ was positioned at the hydrophobic pocket formed by hydrophobic residues such as Leu8, Val36, Leu40, and Ala59. One exception was Asn63, which assists the binding of the adenine moiety of NADP+ through a water-mediated hydrogen bond (Fig. 3B).

Fig. 3.Cofactor binding mode of CgDapB. (A) Structural change upon the binding of NADP+. The CgDapB structures in the apo-form and in complex with NADP+ were superimposed, and are shown as cartoon models with gray and cyan colors, respectively. The bound NADP+ is shown as a stick model with magenta color and labeled. (B) Cofactor binding mode of CgDapB. The CgDapB structure in complex with NADP+ is shown as a cartoon model with cyan color. Residues involved in the NADP+ stabilization are shown as stick models. The G-x-x-G-x-x-G motif is distinguished with green color. Hydrogen bonds formed between NADP+ and neighboring residues are shown with red-colored dotted lines. (C) Electron density map of NADP+. The electron density map of the bound NADP+ is shown as a gray-colored mesh and contoured at 1.0 σ. The phosphate group that showed no visible electron density map is distinguished with a redcolored dotted circle.

Interestingly, the phosphate group attached to the adenosine nucleotide did not interact with the neighboring residues. Moreover, we did not observe an electron density map of the phosphate group region, indicating that the phosphate group might be highly flexible in the bound form to the enzyme (Fig. 3C). Since MtDHPR utilizes both NADH and NADPH as cofactors, we speculate that CgDapB utilizes both molecules as cofactors as well.

Substrate Binding Mode of CgDapB

We could not determine the CgDapB structure in complex with the dihydrodipicolinate substrate; however, we can infer the substrate binding mode of the enzyme from the MtDHPR structure in complex with 2,6-pyridinedicarboxylate (2,6-PDC), a competitive inhibitor of DHPR. In MtDHPR, the substrate binding site is located in the CTD; the aromatic ring of 2,6-PDC is stacked against the nicotinamide ring of bound NADP+ cofactor, and two carboxyl groups form hydrogen bonds with neighboring residues such as His133, Lys136, and Thr143. Among these residues, Lys136 is thought to be a catalytic residue. Interestingly, in our CgDapB structure in complex with NADP+, we observed two sulfate molecules at the substrate binding site (Fig. 4A). When we superimposed the CgDapB structure in complex with NADP+ with the MtDHPR structure in complex with 2,6-PDC, the sulfate molecules were found to be located at positions similar to the two carboxyl groups of 2,6-PDC (Fig. 4B). Moreover, the sulfate molecules in CgDapB were stabilized in a manner similar to the carboxyl groups of 2,6-PDC in MtDHPR. Sulfate I was coordinated through hydrogen bonding with the side chains of Lys138 and Thr145 and the main chains of Gly144 and Thr145. Additionally, sulfate II was stabilized by residues His135 and Lys138 through hydrogen-bond interactions (Fig. 4B). Based on these observations, we propose that CgDapB shares both the catalytic mechanism and substrate binding mode with MtDHPR.

Fig. 4.Substrate binding mode of CgDapB. (A) Electron density map of two sulfate molecules. The electron density map of two bound sulfate molecules are shown as a light blue-colored mesh and contoured at 1.0 σ. (B) Substrate binding mode of CgDapB. The CgDapB structure in complex with NADP+ and the MtDHPR in complex with NADP+ and 2,6-PDC were superimposed, and are shown with cyan and gray colors, respectively. The 2,6-PDC and sulfate molecules are shown as stick models and labeled. The residues involved in the stabilization of the sulfate molecules in CgDapB are shown as a stick model, and the corresponding residues in MtDHPR are as a line model.

Domain Movement and Sequence of Cofactor and Substrate Binding

It has been reported that MtDHPR undergoes an open/closed domain movement upon binding of the cofactor and substrate. Interestingly, we did not observe domain movement upon binding of the NADP+ cofactor, but rather the position of NTD in the apo-form was nearly identical to that in the NADP+-bound form (Fig. 5). This observation indicates that domain movement to the closed conformation requires binding of both the cofactor and the substrate. Here, we propose the sequence of binding events of the cofactor and substrate to the enzyme as follows. First, the cofactor binds to the NTD of the enzyme. Binding of the cofactor to the enzyme may increase the binding affinity of the substrate, since one key contributor of 2,6-PDC stabilization in the substrate binding site is a p-p interaction between the aromatic ring of 2,6-PDC and the nicotinamide ring of NADP+. The binding of substrate to the NADP+-bound enzyme then induces domain movement to the closed conformation for enzyme catalysis.

Fig. 5.Domain movement of CgDapB. (A) Relative position of NTDs. The CgDapB structures in the apo-form and in complex with NADP+ and the MtDHPR structures in the apo-form and in complex with NADP+/2,6-PDC were superimposed based on the CTDs. The CTDs of these structures are shown with a gray color, and the NTDs of these structures are distinguished with different colors and labeled. (B-E) Surface models of CgDapB MtDHPR. The MtDHPR structures in the apo-form (B) and in complex with NADP+ and 2,6-PDC (C) and the CgDapB structures in the apo-form (D) and in complex with NADP+ (E) are presented as surface models. The NTDs and CTDs of these structures are shown in the same color scheme as shown in (A). The bound NADP+ and 2,6-PDC are shown as stick models and labeled.

Our study provides structural insight into the molecular mechanism of CgDapB, one of the key enzymes in lysine biosynthesis inC. glutamicum. Through the CgDapB structure in complex with NADP+ cofactor, we determined how the enzyme utilizes both NADH and NADPH as cofactors. We also determined the substrate binding mode of the enzyme through the coordination mode of two sulfate molecules in our structure. Finally, we propose that the cofactor binds to the enzyme prior to substrate for enzyme catalysis, and domain movement to the active conformation requires binding of both the cofactor and the substrate.