Cloning and characterization of aspartate-β-semialdehyde dehydrogenase from Mycobacterium tuberculosis H37 Rv

Authors


Rupinder Tewari, Microbial Biotechnology Laboratory, Department of Biotechnology, Panjab University, Chandigarh, Punjab 160014, India (e-mail: rupinder@pu.ac.in).

Abstract

Aims:  To clone and characterize the aspartate-β-semialdehyde dehydrogenase of Mycobacterium tuberculosis H37Rv.

Methods and Results:  The asd gene of M. tuberculosis H37Rv was cloned in pGEM-T Easy vector, subcloned in expression vector pQE30 having a T5 promoter, and overexpressed in Escherichia coli. The ASD enzyme was expressed to levels of 40% but was found to be inactive. Functional ASD was obtained by altering induction and growth conditions and the enzyme was purified to near homogeneity using nickel–nitrilotriacetic acid (Ni-NTA) affinity chromatography. The Km and Vmax values for the three substrates l-ASA, NADP and Pi, the turnover number and specific activity of the enzyme were determined.

Conclusions:  Functional ASD enzyme of M. tuberculosis was obtained by gene cloning and protein purification using affinity chromatography. The Kcat and specific activity of the enzyme were 8·49 s−1 and 13·4 μmol min−1 μg−1 respectively.

Significance and Impact of the Study:  The ASD enzyme is a validated drug target. We characterized this enzyme from M. tuberculosis and future work would focus on deducing the three-dimensional structure of the enzyme and design of inhibitors, which could be used as drugs against TB.

Introduction

Mycobacterium tuberculosis is one of the leading causes of infectious disease in the world and kills more than 50 000 people every week (Chopra et al. 2003). The high proportion of tuberculosis (TB)-related mortality has been attributed to a variety of reasons: BCG inefficacy, HIV co-infection and emergence of multiple drug-resistant strains of M. tuberculosis. The situation is so grim that the WHO has constituted a programme called ‘Global Alliance for TB Drug Development’ (http://www.tballiance.org). The main goal of this programme is to identify novel drugs against this pathogen. One modern avenue for developing such compounds is structure-based drug design, in which three-dimensional structure of the target is determined, and then using this structural information for the design of new compounds. In the case of antibacterial compounds, selection of targets usually involves identifying enzymes of metabolic pathways essential for the micro-organism but not present or less important in human metabolism (Zhang and Amzel 2002). The availability of genome sequence of many bacteria has given an impetus to the search for novel antibacterial drugs (Jackson et al. 2000; McKinney et al. 2000).

The aspartate amino acid family pathway is responsible for the biosynthesis of lysine, threonine, isoleucine and methionine. The pathway is present in plants and bacteria and is essential for their viability, but absent in humans (Cohen 1983). Therefore developing inhibitors of key enzymes of this pathway have been suggested to control the growth or kill various bacterial pathogens (Pavelka et al. 1997).

Diaminopimelic acid (DAP) is one of the intermediates for synthesis of lysine and also an essential constituent of many bacterial cell walls including M. tuberculosis. Absence of DAP makes the cell wall so fragile that DAP bacteria lyse immediately. The absolute requirement of DAP by microbes along with nonrequirement as well as nonproduction of this compound by humans makes DAP biosynthetic enzymes an excellent target for designing specific enzyme inhibitors that could have potential as new anti-mycobacterial agents (Cirillo et al. 1994). Aspartate-β-semialdehyde dehydrogenase (ASD, EC 1.2.1.11) is one of the key enzymes of DAP biosynthetic pathway and we report for the first time the heterologous expression, purification and characterization of ASD enzyme (Rv3708c) of M. tuberculosis H37Rv.

Materials and methods

Bacterial strains and plasmids

Mycobacterium tuberculosis H37Rv cells (kindly provided by Dr G.K. Khuller, Department of Biochemistry, P.G.I.M.E.R, Chandigarh, Punjab India) were grown on Lowenstein–Jensen (LJ) medium. Escherichia coli JM109 (Promega, Madison, WI, USA) and E. coli M15 (pREP4) (Qiagen, Valencia, CA, USA) cells were routinely cultured in Luria–Bertani (LB) medium (Difco Laboratories, Detroit, MI, USA) containing, if needed, the appropriate concentrations of antibiotics: 100 μg ml−1 ampicillin and 25 μg ml−1 kanamycin (Sigma, St Louis, USA). The plasmids pGEM-T Easy (Promega) and pQE30 (Qiagen) were used for cloning and subcloning respectively.

PCR amplification of asd gene

Genomic DNA of M. tuberculosis H37Rv was extracted from harvested log phase culture by the method of van Soolingen and Hermans (1995). PCR amplification of the coding region of the asd structural gene (excluding start codon ATG) was carried out using two sets of forward (F) and reverse (R) primers (Microsynth, Balgach, Switzerland). Bold letters signify recognition sequences for restriction enzymes.

Primer F1 (30 mer): 5′-GGC CTG TCA ATA GGG ATC GTG GGG GCC ACC-3′.

Primer R1 (30 mer): 5′-TCA CAA GTC GGC GGT CAG CAG CTC GGC GAT-3′.

Primer F2 (36 mer): 5′-CGG GGT ACC GGC CTG TCA ATA GGG ATC GTG GGG GCC-3′.

Primer R2 (48 mer): 5′-GGG AAG CTT CAA GTC GGC GGT CAG CAG CTC GGC GAT CTG GAT GGT GTT-3′.

These primers were designed based on the sequence of asd gene as present in the genome sequence of M. tuberculosis H37Rv (Cole et al. 1998) accessed from NCBI's Entrez Genome database. PCR kit was purchased from Roche (Mannheim, Germany). The thermocycler (Techne Progene, Cambridge, UK) was programmed for a hot start of 94°C for 5 min followed by 30 cycles at 94, 63 and 72°C each for 1 min and a final cycle of 10 min at 72°C. A 10% (v/v) dimethylsulphoxide (DMSO) was used in the PCR reaction. Amplified DNA fragments were purified from agarose gel using Qiagen Gel Extraction Kit.

Cloning of asd gene

The asd amplicons and inserts obtained from restriction digestion were ligated into pGEM-T Easy vector and pQE30 vector respectively (4°C, overnight, ligation water bath). Ligation products were used to transform competent cells (E. coli JM109 or E. coli M15) prepared by CaCl2 method (Sambrook et al. 2000). Transformed cells were plated on LB agar plates supplemented with isopropyl-β-d-thiogalactopyranoside (IPTG) (4 μl from a stock of 200 mg ml−1), X-gal (40 μl from a stock of 20 mg ml−1) and ampicillin for E. coli JM109 and ampicillin + kanamycin for E. coli M15 (pREP4).

Expression of recombinant ASD

Overnight log phase culture (1%) of E. coli M15 carrying lacI plasmid pREP4 and asd+ recombinant plasmid pSST1 was inoculated in LB broth + amp + km and allowed to grow to O.D600 0·7–0·8 (37°C, 4 h, 200 rev min−1). The expression of (His)6-ASD was induced by the addition of IPTG (0·15–1·0 mm final concentration) to the culture medium (23–37°C, 3–6 h, 200  rev min−1).

SDS-PAGE and Western blotting

A 1·5-ml of the induced culture was pelleted, resuspended in sample buffer and heated for 5 min and analysed by SDS-PAGE (4% stacking, pH 6·8, 10% resolving; pH 8·8). The gels were stained for 4 h by standard Coomassie brilliant blue R250 and destained with 25% methanol/7% acetic acid solution. Images were saved on a gel documentation system. Western blot analysis of the N-terminal His-tagged ASD was performed using mouse anti-His antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The protein was transferred to nitrocellulose membrane (Amersham Biosciences, Piscataway, NJ, USA), which was then developed by the alkaline phosphatase reaction using 5-bromo, 4-chloro, 3-indolylphosphate/nitroblue tetrazolium (BCIP/NBT), following incubation with primary and secondary antibodies.

ASD purification

For purification of the enzyme, IPTG-induced cells were pelleted and redissolved in 20% (w/v) lysis buffer (50 mm potassium phosphate buffer (KPB); pH 8·0, 1 mm phenyl methyl sulfonyl fluoride (PMSF), 1 mm dithiothreitol (DTT), 1 mm lysozyme) and kept in ice for half an hour before being sonicated for 5 min (30-s pulses followed by 30-s intervals). The cellular debris was removed by centrifugation (15 000 g, 15 min, 4°C). The purification of ASD was carried out using chromatography on a Bio-Rad Econosystem (Bio-Rad Laboratories, Richmond, CA, USA) using an imidazole gradient (0–250 mm) prepared in 50 mm KPB; pH 8·0. The crude cell extract was loaded onto 2 ml (50% w/v) slurry of nickel-chelating agarose beads (Novagen, Madison, USA), which was also equilibrated with the same buffer. The flow through was collected and after washes with the KPB, the gradient former was switched on allowing the buffer with increasing imidazole gradient to pass through the column. One-millilitre fractions were collected and checked for purified protein on SDS-PAGE. The fractions showing purified protein were then pooled and passed through a Sephadex G-25 column and eluted using 50 mm KPB (pH 8·0), to remove imidazole. The desalted samples were again checked by SDS-PAGE and protein concentrations were determined by the Bradford's (1976) method.

DNA sequencing

The cloned asd gene in the recombinant plasmid pSST1 was first checked by restriction digestion using KpnI and HindIII restricrion enzymes and then subjected to forward and reverse sequencing using sequencing primers for pQE vectors by the fluorescent dideoxyterminator method using an ABI 3100 capillary sequencer (ACTG, Inc., Northbrook, IL, USA) and the sequence submitted to GenBank.

ASD enzyme assay

The ASD enzyme assay method was adopted from Moore et al. (2002). Enzyme assays were performed on a Shimadzu UV 160A spectrophotometer (Shimadzu, Tokyo, Japan). The assay reaction monitors the oxidative phosphorylation of ASA in the nonphysiological direction in 200 mm N-cyclohexyl-2-aminoethanesulfonic acid (CHES); pH 9·0, 50 mm KPi, 1 mm NADP, 1 mm EDTA, 1 mm DTT at 30°C. l-ASA (obtained as a generous gift from Dr A.T. Hadfield, University of Bristol, UK) was finally added to initiate the reaction, which was monitored at 340 nm (ɛ = 6·22 mm−1 cm−1) to measure the increase in NADPH absorbance. The kinetic parameters Km and Vmax were determined by the Lineweaver–Burk representation of the Michaelis–Menten model. The results were recorded after fitting the data to an Enzyme Kinetics software package adapted from Cleland (1967). The Kcat values were calculated based on a molecular weight of 37 kDa for the M. tuberculosis ASD.

Results

Cloning of asd gene

The asd gene of M. tuberculosis H37Rv, lacking its original start codon was PCR amplified using chromosomal DNA as template and a set of primers F1 and R1. F1 primer sequence matched the 5′-end of asd structural gene lacking only the initiation codon, ATG. R1 primer sequence matched the last 30 nucleotides of asd gene. High annealing temperature (63°C) and use of DMSO in the PCR reaction are attributed to the high GC content of the M. tuberculosis genome. The PCR product corresponded with the expected size (1038 bp) of M. tuberculosis H37Rv asd. The amplified fragments were gel purified and ligated into dT vector pGEM-T Easy, transformed into competent E. coli JM109 cells and plated on LB agar containing ampicillin. The recombinant plasmid asd+ pGEM-T Easy was extracted from white E. coli colonies and used as template for PCR amplification using a new set of modified primers F2 and R2. F2 primer carried KpnI restriction site in F1 primer (excluding the last three nucleotides), whereas R2 primer carried HindIII restriction site in R1 primer (excluding the first three nucleotides). The amplified DNA fragments were again gel purified and ligated into pGEM-T Easy vector and transformed into E. coli JM109 and recombinant colonies were selected on LB agar + amp plates. A few white colonies were picked and the asd+ recombinant plasmids were extracted from these clones and digested with KpnI and HindIII restriction enzymes (New England Biolabs, Beverly, MA, USA). The DNA insert was ligated into the expression vector pQE30, which had been digested with same enzymes. The ligation mix was transformed into E. coli M15 carrying lacI plasmid pREP4. This plasmid carries Km marker and lacI repressor gene. Transformants were screened for the presence of asd insert by PCR as well as restriction digestion with KpnI and HindIII. One clone showing a high level of expression of ASD was subjected to forward and reverse DNA sequencing. The results showed 100% homology with asd nucleotide sequence present in the complete genome sequence of M. tuberculosis H37Rv (Cole et al. 1998). This asd+ plasmid construct was named pSST1 (Fig. 1).

Figure 1.

Plasmid map of the construct pSST1 (pQE30/Mycobacterium tuberculosisasd)

Sequence submission

The asd gene nucleotide sequence was submitted to GenBank and appears with accession number AY372113.

Expression of ASD

The expression of recombinant ASD in E. coli M15 (pREP4) carrying pSST1 plasmid was induced by adding 1 mm IPTG (final concentration) to LB broth +amp + km (37°C, 4 h, 200  rev min−1), when the culture had grown to O.D.600 0·7–0·8. A 1·5-ml of this culture was aliquoted, centrifuged and used for SDS-PAGE analysis. A thick band of ca 40 kDa size was clearly visible after Commassie blue staining of the gel (Fig. 2a) which corresponded with the expected size of ASD protein as deduced from nucleotide sequence of M. tuberculosis H37Rv asd. Densitometric analysis of the lane indicated the protein to be expressed to levels of as high as 40%. Western blot analysis confirmed the expression of His-tagged ASD as the developed nitrocellulose membrane clearly showed bands corresponding to the recombinant protein in different lanes.

Figure 2.

(a) Lane 1 and 2 show IPTG induced samples with a thick band corresponding to recombinant Mycobacterium tuberculosis ASD. Lane 3 is the uninduced control of the clone. (b) Lane 1 shows the purified M. tuberculosis ASD using Ni-NTA affinity chromatography

The cell paste of the remaining culture was isolated (12 000g, 10 min, 4°C). These cells were suspended in 20% KPB (pH 8·0) and disrupted by sonication. The sonicated samples were then centrifuged (15 000g, 15 min, 4°C) and the cell free supernatant was used for ASD enzyme assay. However, the supernatant did not show any significant increase in the ASD activity compared with control nonrecombinant E. coli M15 (pREP4), suggesting that the enzyme was rendered inactive by being trapped in inclusion bodies. To express the enzyme in active form, IPTG induction of the recombinant ASD was carried out at different incubation temperatures (23–37°C), IPTG concentrations (0·15–1 mm) and incubation periods (3–6 h). We found that the maximum soluble ASD was obtained from the culture grown for 5 h at 23°C after induction by 0·15 mm IPTG.

Purification of ASD

Escherichia coli M15 (pREP4) cells harbouring asd+ pSST1 were grown in 200 ml LB broth + amp + km and the expression of ASD induced under optimal conditions (23°C, 5 h, 0·15 mm IPTG). The cells were harvested, resuspended in lysis buffer, sonicated and centrifuged as mentioned above. The crude extract was then passed through the nickel–nitrilotriacetic acid (Ni-NTA) metal affinity column. The flow through and wash fractions were collected before the gradient former was switched on and fractions collected at increasing imidazole concentration (0–250 mm). The aliquots from the fractions were run on SDS-PAGE. Bands pertaining to the purified protein were obtained at ca 60–65% imidazole gradient (ca 150–160 mm imidazole). Pure protein containing fractions were pooled and passed through a G-25 Sephadex desalting column to get rid of imidazole. The collected fractions were then run on SDS-PAGE (Fig. 2b) and analysed to be 96% pure by densitometry scanning.

Kinetic properties of ASD

The reaction catalysed by ASD was studied by monitoring the increase in NADPH absorbance at 340 nm. ASD exhibited a specific activity of 13·4 μmol NADP reduced min−1 μg−1 of protein. With a molecular mass of ca 37 kDa, the turnover number (Kcat) of the enzyme was calculated to be 8·49 s−1. The apparent Km and Vmax values for the three substrates ASA, NADP and Pi are given in Table 1. Lineweaver–Burk double-reciprocal plots of the substrates are given in Fig. 3. These values were deduced by varying the concentration of the substrate of interest and keeping the concentration of the other two substrates constant at saturating levels.

Table 1.  Apparent kinetic parameters of purified recombinant ASD of Mycobacterium tuberculosis H37Rv
 KmVmaxKcat/Km
  1. Maximum standard errors ±15%. Each value represents the mean of three different measurements.

l-ASA0·955 mm51·84 μm min−1 mg−18·89 × 103 M−1 s−1
Pi11·39 mm44·04 μm min−1 mg−17·45 × 102 M−1 s1
NADP0·065 mm43·75 μm min−1 mg−11·31 × 105 M−1 s−1
Figure 3.

Double reciprocal (Lineweaver–Burk) plots of the substrates ASA (a), Pi (b) and NADP (c). Each dot represents the mean value of three different measurements

Discussion

Aspartate-β-semialdehyde dehydrogenase is a key enzyme of DAP biosynthetic pathway and reduces l-aspartyl-β-phosphate into l-aspartic-β-semialdehyde. Blocks in DAP biosynthetic pathway are lethal, as demonstrated in Salmonella typhimurium (Galan et al. 1990) and Streptococcus mutans (Cardineau and Curtiss 1987). Mycobacteria have been suggested to have an absolute requirement for this pathway even when growing in rich medium containing DAP and all the members of the aspartate family (Pavelka and Jacobs 1996). Recently, the ASD enzyme of bacterial pathogens Haemophilus influenzae, Pseudomonas aeruginosa, Vibrio cholerae has been characterized (Moore et al. 2002) in order to design ASD inhibitors which could act as selective drugs. Considering the highly validated status of ASD as a drug target and the increasing menace of TB, it was highly desirable to characterize the ASD enzyme of M. tuberculosis.

Overexpression of recombinant proteins sometimes results in the formation of functionally inactive inclusion bodies. Lowering incubation temperature and IPTG concentrations in the nutrient medium have been shown to produce functionally active and soluble recombinant proteins in E. coli. The Km value for ASA is in close proximity with ASD of the actinomycete Amycolatopsis mediterranei U32 (Zhang et al. 2000). However, this value is several times higher than those reported for Gram-negative bacteria E. coli (Ouyang and Viola 1995), H. influenzae, P. aeruginosa and V. cholerae (Moore et al. 2002). The Km value of 11·39 mm for Pi is several times higher than the values reported for the above mentioned bacteria except for the V. cholerae II ASD which has been reported to have a Km of 22 mm. However, the Km values for NADP in these bacteria are approximately two to four times higher than the value in M. tuberculosis, which is 0·065 mm. The turnover number of this enzyme is 8·49 s−1, which is much lower than the values reported for E. coli, H. influenzae, P. aeruginosa, V. cholerae I and V. cholerae II (610, 330, 160, 120 and 58 s−1 respectively).

The E. coli ASD enzyme has been crystallized and its structure solved to 2·5 A° resolution (Hadfield et al. 1999). In this enzyme, Cys135 has been identified as active site nucleophile (Karsten and Viola 1992). Gln162 and Arg267 have been shown to assist in catalysis and participate in substrate recognition and binding respectively. His274 is likely the acid/base catalyst in the enzymic reaction (Hadfield et al. 1999). Multiple sequence alignment of ASD from M. tuberculosis and seven other bacteria including E. coli shows all these four amino acid residues are conserved (Fig. 4) suggesting these key residues may be involved in maintenance of structural stability and/or biological function in the other ASD as well. However, no structural data is available to substantiate these presumptions and our future work would focus on solving the three-dimensional structure of this enzyme from M. tuberculosis. The amino acid sequence of ASD of M. tuberculosis and closely related Amycolatopsis mediterranei shared the highest homology (69%). The ASD of E. coli and M. tuberculosis share a homology of 26%. Search in the NCBI Conserved Domain Database suggested the presence of two semialdehyde dehydrogenase domains in the M. tuberculosis ASD, the N-terminal NADP-binding domain and the dimerisation domain, as reported for E. coli ASD. The N-terminal pyridine nucleotide coenzyme-binding domain of the E. coli enzyme possesses a nucleotide-binding motif GxxGxxG (Hadfield et al. 1999). This motif is also conserved in all the ASD including M. tuberculosis.

Figure 4.

ClustalX sequence alignment between ASD of Mycobacterium tuberculosis, Amycolatopsis mediterranei, Vibrio cholerae, Streptococcus mutans, Escherichiacoli, Salmonella typhimurium, Haemophilus influenzae and Pseudomonas aeruginosa. Identical amino acids are indicated by asterisks and similar by dots

As a result of its essential role in amino acid biosynthesis and its relatively well-characterized mechanism, ASD is a potentially attractive target for the identification of compounds that can serve as leads for the development of new antibiotics. Even in herbs and plants, the central enzymes of aspartate pathway are considered targets for inhibition and enhancement, respectively, with the aim of developing new herbicides and improving the nutritional value of crops (Viola 2001). Recently the ASD enzyme of the plant Arabidopsis thaliana has been overproduced in E. coli and characterized (Paris et al. 2002). We plan to use this M. tuberculosis ASD for the design of inhibitors against this deadly pathogen.

Acknowledgements

We sincerely thank Dr A.T. Hadfield (Department of Biochemistry, University of Bristol, UK), for generously providing the enzyme substrate l-ASA. We are extremely grateful to Dr K.V.S. Rao (Immunology Group Leader, I.C.G.E.B) and Dr V.S. Chauhan (Director, I.C.G.E.B) for providing the requisite facilities for this work to be carried out. Department of Biotechnology (Government of India) is acknowledged for financial aid to Dr Pawan Sharma's laboratory.

Ancillary