|Home | About | Journals | Submit | Contact Us | Français|
Rv2780, an alanine dehydrogenase from Mycobacterium tuberculosis (MtAlaDH), catalyzes the NAD-dependent interconversion of alanine and pyruvate. Alanine dehydrogenase is released into the culture medium in substantial amounts by virulent strains of mycobacteria and is not found in the vaccine strain of tuberculosis. Crystals of recombinant MtAlaDH were grown from 2 M ammonium sulfate solution at ~12 mg ml−1 protein concentration in two crystal forms which occur in the presence and absence of NAD/pyruvate, respectively. Diffraction data extending to 2.6 Å were collected at room temperature from both apo and ternary complex crystals. Crystals of the apoenzyme have unit-cell parameters a = 173.89, b = 127.07, c = 135.95 Å. They are rod-like in shape and belong to space group C2. They contain a hexamer in the asymmetric unit. Crystals of the ternary complex belong to space group P43212 and have unit-cell parameters a = b = 88.99, c = 373.85 Å. There are three subunits in the asymmetric unit of the holoenzyme crystals.
Mycobacterium tuberculosis exhibits significant changes in gene expression during the latent/persistent stage of infection. Proteomic analysis of nutrient-starved/latent-phase bacteria suggested decreased expression of proteins involved in energy metabolism, lipid biosynthesis and cell division in addition to induction of stringent response and several other genes that may play a role in long-term survival within the host (Betts et al., 2002 ; Stewart et al., 2003 ). l-Alanine dehydrogenase, which is encoded by ald, catalyzes the oxidative deamination of l-alanine or, in the reverse reaction, the reductive amination of pyruvate. This enzyme is known to be required for normal sporulation in Bacillius subtilis (Siranosian et al., 1993 ) and for normal development of Myxococcus xanthus (Ward et al., 2000 ). In mycobacteria, l-alanine dehydrogenase is a secretory antigen that is associated with bacterial persistence during infection. Expression of ald is upregulated in M. tuberculosis upon nutrient starvation and in M. marinum during persistence within the granulomas of infected frogs (Chan et al., 2002 ). l-Alanine dehydrogenase production and activity was also increased when M. tuberculosis and M. smegmatis were shifted from aerobic to anaerobic growth conditions (Rosenkrands et al., 2002 ; Usha et al., 2002 ). The enzyme was rated among the top three targets against persistence in an analysis by Hasan et al. (2006 ). Western blotting has failed to detect l-alanine dehydrogenase in the BCG (Bacille Calmette-Guérin) strain, even though the ald gene has been identified by DNA hybridization; this absence of functional protein arises from a single-nucleotide deletion within the ald gene that causes a frame-shift mutation and disrupts the full-length protein.
The overall reaction catalysed by Ald is
l-Alanine dehydrogenases purified from sources such as M. tuberculosis (Andersen et al., 1992 ), Bacillus spp. (Ohshima et al., 1990 ), Phormidium lapideum (Sawa et al., 1994 ) and Streptomyces phaeochromogenes (Itoh & Morikawa, 1983 ) exist as homohexamers, with a subunit molecular weight of ~40 000 Da, while octameric and tetrameric oligomeric associations have been reported for the enzymes from S. aureofaciens (Vancurova et al., 1989 ) and S. fradiae (Vancura et al., 1989 ), respectively. The crystal structures of alanine dehydrogenase from P. lapideum (cyanobacteria; Baker et al., 1998 ) and Archaeoglobus fulgidis (hyperthermophilic archea; Gallagher et al., 2004 ) have been reported. The level of sequence similarity between Ald from P. lapideum and those from other eubacteria is high. These enzymes are comprised of a C-terminal NAD-binding Rossmann-fold domain and a catalytic domain. The domains are separated by a cleft. However, bacterial ALDs exhibit lower homology in sequence and in structure compared with other amino-acid dehydrogenases including phenylalanine dehydrogenase (Vanhooke et al., 1999 ), leucine dehydrogenase (Baker et al., 1995 ) and glutamate dehydrogenase (Stillman et al., 1993 ).
Owing to the implied importance of Ald in the virulence of M. tuberculosis, we have cloned, expressed, purified and crystallized the M. tuberculosis protein (MtAlaDH). The structure of the apoenzyme has been solved using molecular replacement and reveals a hexamer in the asymmetric unit of the crystal. Similar calculations show that there are three subunits in the asymmetric unit of the crystals of the holoenzyme. The structures are expected to be a valuable tool in identifying novel inhibitors using structure-based strategies.
The ald gene (Rv2780) was amplified from M. tuberculosis H37Rv genomic DNA using the following primers: the sense primer 5′-ATGCGCTAGCATGCGCGTCGGTATTTCCG-3′ and the antisense primer 5′-TATAAAGCTTGGCCAGCACGCTGG-3′ containing NheI and HindIII restriction sites (nucleotides shown in bold). The amplified PCR product was digested with NheI and HindIII and ligated into pET23a (Novagen) vector digested with the same enzyme. This attaches a tag comprising ACGRTRAHHHHHH at the C-terminal end and allows purification using affinity chromatography. Clones were screened by restriction digestion and the resulting construct was transformed into Escherichia coli C41 (DE3). The integrity of the insert was verified by sequencing.
A single colony was inoculated into 50 ml YT medium containing 50 µg ml−1 ampicillin and grown overnight at 310 K. These cells were used to inoculate 1 l YT medium containing 50 µg ml−1 ampicillin and expression was induced with 0.2 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) at an optical density of 0.6 at 600 nm. Cells were grown at 303 K for 6 h after induction with IPTG. The IPTG-induced cells were harvested, resuspended in ice-cold 50 mM Tris–HCl pH 7.2, 300 mM NaCl, 10 mM imidazole (buffer A) and sonicated. The crude lysate was centrifuged at 27 000g for 30 min. The supernatant was applied onto a Ni–IDA column pre-equilibrated with buffer A. The protein was eluted using the same buffer supplemented with 500 mM imidazole. Fractions containing the protein were pooled and precipitated by ammonium sulfate (60%) saturation. The pellet was resuspended in 50 mM Tris–HCl pH 7.2, 50 mM NaCl, 5 mM EDTA, 2 mM β-mercaptoethanol (buffer B) and further purified on a Superdex S-200 (Amersham Biosciences) gel-filtration column equilibrated with buffer B. The protein was pooled and concentrated to 12 mg ml−1 using a 30 kDa cutoff Centricon (Amicon). The purified protein with a hexahistidine tag at the C-terminus was assayed for activity and protein concentrations were determined with the Bradford reagent (Bradford, 1976 ) using bovine serum albumin as a standard. MtAlaDH remained stable at 277 K without degradation for up to one week. The purity of the protein was confirmed using 12% SDS–PAGE (Fig. 1 ). Photometric determination of enzyme activity was accomplished by measuring the rate of the production of NADH at 340 nm that accompanies the conversion of alanine into pyruvate in the oxidative deamination.
MtAlaDH was concentrated to ~12 mg ml−1 in a buffer comprising 50 mM Tris pH 7.2, 50 mM NaCl, 5 mM EDTA and 2 mM β-mercaptoethanol. All crystallization experiments were carried out by the hanging-drop vapour-diffusion method in 24-well Linbro tissue-culture plates (ICN Inc.). The experiments were set up exploiting two different sparse-matrix screening strategies (Mazeed et al., 2003 ; Jancarik & Kim, 1991 ; Cudney et al., 1994 ). Each hanging drop was prepared using 2 µl protein sample and 1 µl reservoir solution and was equilibrated against 500 µl reservoir solution. The apo and ternary complex crystals were obtained under similar crystallization conditions, except that 5 mM NAD and 10 mM pyruvate were added to the crystallization solution in the case of the ternary complex crystals. Good diffraction-quality crystals of MtAlaDH were obtained from a reservoir solution consisting of 2 M ammonium sulfate. Single rod-shaped crystals corresponding to the apo form (Fig. 2 ) grew to typical dimensions of 1.4 × 0.7 × 0.4 mm. In the case of the ternary complex, large diamond-shaped crystals started appearing within 3–4 d and grew to typical maximum dimensions of 1.7 × 1 × 0.7 mm.
A single crystal was mounted in a capillary for both the apo protein and the complex with NAD and pyruvate. Diffraction data were collected using a MAR-DTB image-plate detector and a Rigaku Micromax-007 X-ray generator coupled with Varimax-HR optics. Crystal-to-detector distances of 200 and 240 mm were used for the apo and ternary crystal forms, respectively, and exposure times of 60 and 10 s per frame were used in the respective data collections. An oscillation range of 1° was used for the apoenzyme crystals. One cell edge of the ternary complex crystals was large; the oscillation range was therefore reduced to 0.5° and the data could be processed successfully. The resolution of the larger cell edge was possible with the use of the Rigaku instrument and associated optics, which allow resolution of cell edges typically up to 450 Å. Although the holoenzyme crystals diffract to better than 2.6 Å resolution (Fig. 3 ), the geometry and limitations of the home source only allowed data collection to this resolution. Diffraction images were indexed, integrated and scaled using the HKL-2000 package (Otwinowski & Minor, 1997 ). The crystals of the apoenzyme were monoclinic, with unit-cell parameters a = 173.9, b = 127.1, c = 135.9 Å, and belonged to space group C2. The ternary complex crystallized in space group P43212, with unit-cell parameters a = b = 88.99, c = 373.85 Å. Both crystal forms diffracted to 2.6 Å and the statistics are summarized in Table 1 .
The crystals of the apoenzyme were phased by the molecular-replacement approach using AMoRe (Navaza, 1994 ) as implemented in the CCP4 package (Collaborative Computational Project, Number 4, 1994 ). Sequence alignment with known structures in the PDB showed that the alanine dehydrogenase from P. lapideum (Baker et al., 1998 ; PDB code 1pjb) has nearly 45% sequence similarity. A search model for molecular-replacement calculations was generated from 1pjb in which non-identical amino acids were replaced by alanine. A clear solution was obtained with a correlation coefficient of 56.2% and an R factor of 47.8% using data in the resolution range 15–3 Å. The solution exhibits good packing with no steric hindrances. The asymmetric unit contains a hexamer, which corresponds to a Matthews coefficient (Matthews, 1968 ) of 2.8 Å3 Da−1 and a solvent content of 56.3%. Our previous experiments involving size-exclusion chromatography were consistent with a hexameric association in MtAlaDH. The initial molecular-replacement calculations and maps in the case of the holoenzyme reveal clear density for the NAD and pyruvate moieties, confirming the occurrence of the ternary complex in the crystal. The presence of NAD in the dissolved crystals was also spectroscopically confirmed. There are three subunits in the asymmetric unit, corresponding to a solvent content of about 59.8%. Phase improvement and model building of the structures are presently in progress.
Note added in proof: After submission of the manuscript, we were alerted that structures of the apo and holo forms of the enzyme had been published in the meantime by another group (Agren et al., 2008 ). The crystal forms described differ from those reported here.
This work was supported by a Council of Scientific and Industrial Research, India (CSIR) network grant NWP0038 (formerly SMM003). Computational support was from CSIR grant CMM0017. We thank Dr Amit Sharma (International Centre for Genetic Engineering and Biotechnology, New Delhi, India) for use of the Rigaku Micromax-007 X-ray generator. SMT is a senior research fellow of the University Grants Commission, India. This is communication No. 7472 from CDRI.