This article is mentioned in:

Introduction

Tuberculosis (TB) is caused by Mycobacterium tuberculosis infection and remains one of the major infectious diseases affecting people worldwide, particularly in developing countries. The morbidity and mortality of TB has been exacerbated by the human immunodeficiency virus, the emergence of multidrug-resistant TB, and extensively drug-resistant strains of M. tuberculosis (1). Therefore, it is important to understand the disease and to investigate novel anti-TB drugs.

Unlike the synthesis of cysteine from methionine through homocysteine in fungi and animals (2), serine acetyltransferase (CysE) catalyzes the conversion of acetyl-coenzyme A (CoA) and L-serine to CoA and O-acetyl-L-serine, which is the first product and a key amino acid in the biosynthesis of L-cysteine in microorganisms and plants (3,4). Cysteine is required for the biosynthesis of essential compounds, including L-methionine, thiamine, biotin, coenzyme A and other sulfur-containing compounds. Notably, CysE has no known biological function in humans, as different pathways are used in the anabolism of cysteine in microorganisms and humans. The M. tuberculosis CysE is therefore considered as a potential anti-TB drug target (5,6). Our previous study demonstrated that the Rv2335 protein was the M. tuberculosis homolog of CysE, encoded by the cysE gene, and determined and described the kinetic parameters and optimal catalytic conditions of the enzyme (7). However, to the best of our knowledge, there is currently no information regarding the structure and active sites of M. tuberculosis CysE. Determination of the protein structure is crucial for investigating the relationship between structure and function of any protein. Homology modeled structures may characterize conserved domains and active amino acids, which are essential to the function of the protein.

In the present study, the CysE amino acid sequence from five species, M. tuberculosis, Escherichia coli, Salmonella typhimurium, Haemophilus influenza and Entamoeba histolytica, was compared. This comparison allowed for the prediction of the secondary structure of M. tuberculosis CysE and the construction of a 3D model. Three residues, D67, H82 and H117, were identified as significant sites for activity. Such identifications and characterizations may facilitate screening for enzyme inhibitors to catalytic processes in M. tuberculosis, and thus therapy for patients with TB.

Materials and methods

Materials

Reagents and enzymes involved in gene cloning and site-directed mutagenesis were purchased from Takara Bio, Inc. (Otsu, Japan). The chemicals for all enzyme assays were purchased from Sigma-Aldrich (Merck Millipore, Darmstadt, Germany). The Ni-NTA agarose column used for protein purification was obtained from Qiagen GmbH (Hilden, Germany).

Sequence analysis and alignments

The amino acid sequences of M. tuberculosis CysE (NP_216851) were retrieved from the National Center for Biotechnology Information (NCBI) GenBank database (Bethesda, MD, USA) (8). The ProtParam tool in the ExPASy server (http://web.expasy.org/protparam) was used to predict the physiochemical properties of M. tuberculosis CysE (9). The parameters included: Molecular weight, theoretical isoelectric point (pI), amino acid composition, atomic composition, extinction coefficient, estimated half-life, instability index, aliphatic index and grand average of hydropathy (GRAVY). CysE protein sequences of M. tuberculosis CysE, E. coli CysE (AJF44936), S. typhimurium CysE (KPF36926.1), H. influenzae CysE (KMZ31837.1) and E. histolytica CysE (BAA82868.1) were aligned using the MultAlin sequence alignment tool (http://multalin.toulouse.inra.fr/multalin) (10).

Secondary structure prediction and functional annotation of the M. tuberculosis CysE protein

Based on sequence similarity searches with orthologous family members, several bioinformatics tools were used to analyze the conserved domains of CysE: The European Bioinformatics Institutes' (EBI) InterProScan tool, which combines different protein signature recognition methods from InterPro consortium member databases into one resource (11); the University College London's PSIPRED protein structure prediction server (http://bioinf.cs.ucl.ac.uk/psipred), which is based on a neural network algorithm for predicting structural information about a protein from its amino acid sequence alone (12); and the NCBI Conserved Domains Database (NCBI-CDD; https://www.ncbi.nlm.nih.gov/cdd), which is a protein annotation resource that consists of a collection of well-annotated multiple sequence alignment models for ancient domains and full-length proteins (13).

3D modeling of CysE protein

A 3D structure of the M. tuberculosis CysE protein was modeled by comparative protein modeling methods using the SWISS-MODEL server (http://swissmodel.expasy.org) (14). A suitable template for homology modeling was selected using NCBI Position Specific Iterated-Basic Local Alignment Search Tool (PSI-BLAST) (15) against the Research Collaboratory for Structural Bioinformatics Protein Data Bank (PDB) (http://rcsb.org). The ExPASy Swiss-PdbViewer 3.5 (http://spdbv.vital-it.ch) was used to produce a structure-based alignment and SWISS-MODEL was used in the optimized mode to minimize energy (14,16). For structural evaluation and stereochemical analyses, the 3D model was submitted to the EBI PDBsum pictorial database (17). The M. tuberculosis CysE protein model was further evaluated using the University of California, Los Angeles-Department of Energy Institute programs PROCHECK (18), Verify3D and ERRAT (http://services.mbi.ucla.edu).

Site-directed mutagenesis of M. tuberculosis cysE

Three pairs of mutated primers were designed according to the DNA sequence of M. tuberculosis cysE (Table I). Every pair mutated primer was characterized with their 5′ends adjacent and 3′ends in opposite directions according to the instructions of the Takara MutanBEST kit (Takara Bio, Inc.). The recombinant plasmid pMD18-Mtb-cysE, which contains NdeI and XhoI restriction sites, was constructed previously (7) and was used as the template. The linear mutated plasmids were amplified using Pfu DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA). The amplification reaction was performed in a final volume of 50 µl mixture including 0.5 µl Pfu DNA polymerase (5 U/µl), 5 µl Pfu Buffer (10X), 5 µl dNTP mixture (2.5 mM each), 1 µl template DNA (100 ng), 3 µl mutated primers (20 µM) and ddH2O. The reaction conditions consisted of pre-denaturation (94°C, 5 min), 30 cycles of amplification (denaturation at 94°C for 30 sec, annealing at 65/61°C for 30 sec and primer extension at 72°C for 5 min) and final extension at 72°C for 10 min. Plasmids were recircularized by blunting kination ligation and self-ligation using Takara MutanBEST kit (Takara Bio, Inc.). pMD18-Mtb-cysE plasmids containing site-directed mutations were transformed and grown in E. coli Novablue competent cells (Merck Millipore). The plasmid was transformed into competent cells by heat shock at 42°C. The E. coli Novablue cells harboring plasmid were cultured in lysogeny medium supplemented with 100 ug/ml ampicillin (Merck Millipore) at 37°C. Positive clones were identified by restriction endonuclease digestion (Takara Bio, Inc.) and DNA sequencing as those harboring the required mutations. The sequencing work was collaborated with Takara Bio Inc.

Table I. Sequences of oligonucleotide primers used for constructing site-direct mutations of M. tuberculosis cysE gene.

Table I.

Sequences of oligonucleotide primers used for constructing site-direct mutations of M. tuberculosis cysE gene.

Name	Primer orientation	Primer sequencea (5′-3′)
D67A	F	ATCCTGACCGGTGTAGCCATCC
		ACCCCGGTGCC
	R	TACACCGGTCAGGATGCG
H82A	F	CGCGTGTTCATCGACGCCGCG
		ACCGGCGTGGTG
	R	GTCGATGAACACGCGAGCACCG
H117A	F	GTTGGCGGGAAACGCGCCCCC
		ACCGTCGGTGAC
	R	GCGTTTCCCGCCAACCATGCC

a Mutated residues are underlined. F, forward; R, reverse.

Preparation of mutant M. tuberculosis CysE proteins

Site-mutated M. tuberculosis cysE genes were subcloned into pET29b (Merck Millipore) at NdeI and XhoI restriction sites, and transformed into E. coli BL21 (DE3) cells (Merck Millipore) to express the three different CysE mutated proteins using the heat shock method, each containing a polyhistidine (His)-tag. CysE mutant proteins were prepared and purified using Ni2+ affinity chromatography according to the expression protocol of wild-type M. tuberculosis CysE protein, as previously described (7). Total protein (~10 mg) was loaded onto a 1 ml Ni-NTA agarose column (Qiagen, Hilden, Germany). Then the column was washed with 20 ml wash buffer (20 mM Tris-HCl pH 8.0, 500 mM NaCl, 20% glycerol, 60 mM imidazole and 1 mM PMSF). Finally, the mutant CysE proteins with a His-tag at the C-terminus were eluted with 10 ml elution buffer (20 mM Tris-HCl pH 8.0, 500 mM NaCl, 20% glycerol, 300 mM imidazole and 1 mM PMSF). The concentration of purified mutant CysE proteins was measured using the BCA Protein Assay Kit (P0012S; Beyotime Institute of Biotechnology, Haimen, China). The purified mutant CysE proteins were identified by western blot analysis. The protocol was as followed: Purified proteins (20 µg) per lane were loaded and sepearated by 12% SDS-PAGE (P0012A; Beyotime Institute of Biotechnology). The proteins were transferred to nitrocellulose membranes (Thermo Fisher Scientific, Inc.) after electrophoresed with SDS-PAGE. Following blocking in 5% skimmed milk buffer for 2 h at 37°C, the membrane was incubated with mouse antipolyhistidine monoclonal antibody (dilution, 1:5,000; H1029; Merck Millipore) for 2 h at 37°C and followed by horse-anti-mouse IgG conjugated with alkaline phosphatase (dilution, 1:2,000; ZB-2310; ZSGB-Bio, Beijing, China) for 45 min at room temperature. After staining in the solution mixture of BCIP/NBT (ZSGB-Bio), the immunoreactive protein bands were visualized and collected.

Enzymatic activity assay for mutant CysE proteins

The enzymatic activity assay of the various CysE mutants was conducted by the Ellman's reagent [5,5′-dithiobis-(2-nitrobenzoic acid); DTNB] method, as previously described (19). Briefly, equal concentrations of the purified CysE mutant proteins and the wild-type M. tuberculosis CysE protein were used. The relative activities of the mutated CysE proteins were calculated against activity of the wild-type M. tuberculosis CysE, which was used as the control (set to 100%). All data were analyzed using GraphPad Prism 6 (GraphPad, La Jolla, CA, USA).

Results and Discussion

Sequence analysis and alignment

ProtParam was used to analyze various physiochemical properties from the amino acid sequence of CysE protein. The molecular weight and pI of the 229-amino-acid-long sequence were predicted as 23,770.3 Da and pH 7.11, respectively. An instability index of 27.56 indicated that the CysE protein was stable. The GRAVY index of 0.217 was indicative of a hydrophobic protein. Sequence alignment demonstrated that the M. tuberculosis CysE protein shared identity with E. coli CysE (45%), S. typhimurium CysE (45%), H. influenzae CysE (45%) and E. histolytica CysE (49%) (Fig. 1A).

Figure 1. Amino acid sequence alignment and secondary structure prediction of M. tuberculosis CysE protein (A) Amino acid sequence alignment of homologous CysE proteins from Mycobacterium tuberculosis, Escherichia coli, Salmonella typhimurium, Haemophilus influenza and Entamoeba histolytica. Analysis of the conserved domains of CysE protein using (B) NCBI-CDD, (C) InterProScan and (D) PSIPRED. CysE, serine acetyltransferase; NCBI-CDD, National Center for Biotechnology Information-Conserved Domains Database.

Secondary structure prediction and functional annotation of CysE protein

The bioinformatics tools InterProScan, PSIPRED and NCBI-CDD were used to search the conserved domains and potential functions of the M. tuberculosis CysE protein. The results revealed that the CysE protein possessed a left-handed-β-helix (LβH) domain and belonged to the hexapeptide acetyltransferase family (Fig. 1B-D), as previously shown for Rhizobium leguminosarm (20). This family of acetyltransferases is characterized by imperfect tandem repeats of the hexapeptide motif [LIV]-[GAED]-X2-[STAV]-X (21) with a conserved LβH domain, which is folded by three hexapeptides (22,23).

3D modeling of the CysE protein

Homology modeling is a standard method for structure prediction and contributes to the understanding of the relationship between protein structure and function. The SWISS-MODEL server was used to model proteins based on protein-structure homology (24). A PSI-BLAST search revealed 3P47A as an available template that can be used for homology modeling of the CysE protein. 3P47A is a refined X-ray diffraction model of the A-chain of E. histolytica CysE at a resolution of 1.78 Å. Homology modeling of M. tuberculosis CysE was constructed using 3P47A as a template (Fig. 2A).

Figure 2. 3D structure prediction and assessment of M. tuberculosis CysE protein (A) Predicted 3D structure of CysE protein. (B) Ramachandran plot for the predicted model of CysE. CysE, serine acetyltransferase.

Energy minimization, quality assessment and visualization

The reliability of the predicted 3D model of the M. tuberculosis CysE was assessed by Ramachandran plot, using PROCHECK, and a quality assessment confirmed using PDBsum. The Ramachandran plot revealed a distribution of 160 amino acid residues of M. tuberculosis CysE, with the majority of residues (98.4%) in the favored and allowed regions, and only two of the residues (1.6%) in the generously allowed region (Fig. 2B). The g-factor demonstrates how unusual a stereochemical property is: Values below −0.5 represent unusual property, whereas values below −1.0 indicate high unusualness. The g-factors for dihedral angles and main chain covalent forces were −0.10 and 0.31, respectively. The overall average g-factor for the M. tuberculosis CysE model was 0.07. Taken together, the Ramachandran plot and g-factors indicated that the backbone dihedral angles, phi and psi in the 3D model of M. tuberculosis CysE were within acceptable limits.

Following the construction of a predicted 3D structure, the final protein model was verified using Verify3D, which analyzed the compatibility of the 3D (atomic) model of CysE with its own amino acid sequence (1D). The scores of a sliding 21-residue window (between −10 and +10) were added and plotted for the individual residues (25,26). The M. tuberculosis CysE protein was consistent with the respective amino acids, with 88.82% of the residues having an averaged 3D-1D and a score >0.2. ERRAT is a protein structure verification algorithm that is well suited for evaluating the quality of crystallographic model building and refinement. The final model had a quality factor of 91.156% in ERRAT, indicating that the model was acceptable and credible.

Site-directed mutagenesis of M. tuberculosis CysE

The conserved residues D67, H82 and H117 of M. tuberculosis CysE were selected for site-directed mutation. To maintain the native spatial structure, these three residues were substituted with alanine. Three linear pMD18 plasmid vectors containing the mutated M. tuberculosis cysE gene (pMD18-Mtb-cysE) were amplified using mutated primers. Restriction endonuclease digestion and sequencing analysis confirmed that the three specific site-directed mutations were obtained successfully (data not shown).

Preparation of mutant M. tuberculosis CysE proteins

The pET29b-Mtb-cysE expression plasmids containing the subcloned site-directed mutations were constructed and transformed into E. coli. Three different soluble CysE mutant proteins with fused His-tags were expressed in E. coli BL21 (DE3) cells, and purified by Ni2+ affinity chromatography. The purified mutant proteins were identified by western blot analysis. The band of CysE mutant proteins appeared at ~30 kDa, which was similar to that of wild-type M. tuberculosis CysE protein (7) (Fig. 3A and B).

Figure 3. The identification of purified mutated CysE protein and serine acetyltransferase activity (A and B) Western blot analysis of purified mutated CysE proteins from E. coli BL21 (DE3). Lane 1, protein marker; lane 2, mutated D67A CysE protein; lane 3, mutated H82A CysE; lane 4, mutated H117A CysE protein ; lane 5, wild-type CysE protein. (C) Relative enzymatic activity of CysE with different mutated sites compared with wild-type CysE (**P<0.01). CysE, serine acetyltransferase.

Enzymatic activity assay for mutant CysE proteins

CysE enzymatic activity of the three mutant proteins was detected using a colorimetric assay with DTNB (19), which indicated that the mutant M. tuberculosis CysE proteins exhibited different enzymatic activity compared with wild-type CysE. The D67A mutant demonstrated only 9.79% activity, whereas the H82A and H117A mutants had 14.31 and 20.82% activity, respectively (Fig. 3C). These data suggested that the three residues may be involved in the functional activity of the enzyme.

In conclusion, site-directed mutagenesis of CysE and the enzymatic activity assay indicated that the conserved amino acid residues D67, H82 and H117 may be involved in the catalytic activity and, thus, the function of CysE. The prediction for M. tuberculosis CysE structure may facilitate the elucidation of the functional mechanism of this enzyme. However, further studies regarding the interaction between Cys as a substrate and M. tuberculosis CysE as an enzyme are required. Furthermore, determination of the active amino acids of M. tuberculosis CysE will immensely promote the development of structure-guided enzyme inhibitors for use as novel and effective anti-TB drugs.

Acknowledgements

The present study was supported by the National Natural Science Foundation of China (grant no. 31070066).

References