The putative polyketide cyclase MSMEG_0129 from Mycobacterium smegmatis: purification, crystallization and X-ray crystallographic analysis

The cloning, expression, purification, crystallization and preliminary X-ray diffraction analysis of the protein MSMEG_0129 from M. smegmatis are described.

Abstract

Mycobacterium tuberculosis Rv0164 has previously been identified as a human T-cell antigen that induces significant production of IFN-γ in human peripheral blood mononuclear cells. M. smegmatis MSMEG_0129 shares 59% sequence identity with Rv0164. Based on sequence alignment, both proteins are predicted to be members of the cyclase/dehydrase family, which is part of a large group of enzymes referred to as type II polyketide synthases (PKSs). In biosynthetic pathways mediated by type II PKSs, cyclases catalyze the conversion of linear poly-β-ketones to cyclized intermediates. To date, no mycobacterial type II PKSs have been reported. Here, the goal is to determine whether these proteins adopt similar folds to reported cyclase structures, and to this end MSMEG_0129 was cloned, expressed, purified and crystallized. An X-ray diffraction data set was collected to 1.95 Å resolution from a crystal belonging to space group P6₂, with unit-cell parameters a = 109.76, b = 109.76, c = 56.5 Å, α = 90, β = 90, γ = 120°. Further crystallographic analysis should establish a basis for investigating the structure and function of this putative mycobacterial type II PKS enzyme.

1. Introduction  

Tuberculosis (TB), caused by infection with Mycobacterium tuberculosis (MTB), remains a major threat to public health. According to recent data released by the World Health Organization, there were an estimated 1.8 million deaths from TB and 10.4 million new cases of TB worldwide in 2015 (World Health Organization, 2016 ▸). The acute nature of this threat to public health is further exacerbated by the emergence of drug-resistant MTB isolates (Muller et al., 2013 ▸) and the deadly synergy that exists between TB and HIV (World Health Organization, 2016 ▸). The only licensed anti-TB vaccine to date, the Bacillus Calmette–Guerin (BCG) vaccine, is an attenuated strain of M. bovis and can effectively prevent the development of TB in early childhood, but exhibits highly variable protective efficacy against adult pulmonary TB (Moliva et al., 2015 ▸). The development of more effective vaccines and better drugs is thus urgently required to prevent the spread of TB.

The protein Rv0164 is conserved among mycobacteria (Marmiesse, 2004 ▸) and has been identified as an essential gene product for growth of M. tuberculosis (Griffin et al., 2011 ▸). As a secreted protein (Malen et al., 2007 ▸), it also serves as a human T-cell antigen (Ireton et al., 2010 ▸) and possesses the ability to induce significant production of IFN-γ in human peripheral blood mononuclear cells (Lim et al., 2004 ▸; Sable et al., 2005 ▸). Furthermore, genetic immunization of inbred mice with plasmid DNA and re-stimulation with overlapping synthetic peptides covering this protein led to the discovery of murine T-cell epitopes on Rv0164 (Eweda et al., 2010 ▸). These experimental data suggest that Rv0164 is a candidate immunogen that has potential for vaccine development and serodiagnosis.

The MSMEG_0129 protein from M. smegmatis (144 amino acids) is a close homologue of Rv0164 (161 amino acids) and shares 59% sequence identity. Pfam (Finn et al., 2016 ▸) predicts that these proteins belong to the HMM PF03364 protein family. This family contains polyketide cyclases/dehydrases that play essential roles in polyketide biosynthesis and proteins involved in lipid binding/transport. Polyketides are a large class of natural products synthesized in bacteria, fungi and plants, and have multiple biological functions (Hertweck, 2009 ▸). Their synthesis is catalyzed by polyketide synthases (PKSs), which are mainly classified into three types (Hertweck, 2009 ▸). Type I PKSs are large multifunctional polypeptides containing multiple catalytic domains. In contrast, type II PKSs are multienzyme complexes of discrete enzymes, each of which usually carries a distinct monofunctional catalytic domain. Type III PKSs are homodimeric enzymes which can mediate both polyketide-chain elongation and modification. Type II PKS aromatase/cyclases (ARO/CYCs) are responsible for the cyclization of linear poly-β-ketone intermediates and produce aromatic polyketides displaying versatile pharmacological properties, such as tetracycline and doxorubicin (Das & Khosla, 2009 ▸; Hertweck et al., 2007 ▸). Two common cyclization patterns, C7–C12 and C9–C14, have been detected in the biosynthesis of bacterial polyketides, and correspondingly there are two types of ARO/CYCs comprised of either a monodomain, as in Streptomyces sp. ZhuI, TcmN and WhiE (Ames et al., 2008 ▸, 2011 ▸; Lee et al., 2012 ▸), or double domains, as in S. steffisburgensis StfQ and Amycolatopsis orientalis BexL (Caldara-Festin et al., 2015 ▸).

Both MSMEG_0129 and Rv0164 share some sequence identity (∼19%) with monodomain ARO/CYCs such as Streptomyces sp. ZhuI and TcmN, suggesting that the Mycobacterium proteins may be candidate ARO/CYCs. Crystal structures of ZhuI and TcmN have been reported (Ames et al., 2008 ▸, 2011 ▸) and both have a similar helix–grip fold that consists of a seven-stranded antiparallel β-sheet, a long C-terminal α-helix and two small helices between β1 and β2. A deep interior pocket between the interior of the β-sheet and the long α-helix is speculated to be a binding site for putative linear poly-β-ketone intermediates (Ames et al., 2011 ▸; Lee et al., 2012 ▸).

Most mycobacterial PKSs are believed to be involved in the biosynthesis of cell-envelope components, including lipids and glycolipid conjugates that are critical for virulence (Quadri, 2014 ▸). M. tuberculosis is known to have about 21 type I PKS proteins and three type III PKSs, but no type II PKSs have been reported to date. The type I PKS protein PKS13, for example, is a critical enzyme that catalyzes the last step in the biosynthesis of mycolic acids (Gavalda et al., 2009 ▸), and the type III PKSs PKS11 and PKS18 are likely to be involved in α-pyrone metabolism (Saxena et al., 2003 ▸). To determine whether Rv0164 and MSMEG_0129 are mycobacterial polyketide cyclases with an activity similar to that of the type II PKS cyclases ZhuI and TcmN, we have undertaken a study to determine whether these proteins adopt similar folds to the reported ARO/CYC structures (Ames et al., 2008 ▸, 2011 ▸). Here, we report the cloning, expression, crystallization and preliminary X-ray diffraction analysis of MSMEG_0129. The X-ray data collected to 1.95 Å resolution have established a solid basis for the structure determination of this putative mycobacterial type II PKS.

2. Materials and methods  

2.1. Macromolecule production  

The gene encoding MSMEG_0129 was amplified from M. smegmatis mc² 155 genomic DNA using PCR. Plasmid pET-28a (Novagen) was modified by replacing the thrombin cleavage site with a Tobacco etch virus (TEV) protease cleavage site. The PCR product was inserted into the modified pET-28a vector between the NdeI and NotI restriction sites. The resulting plasmid was verified by DNA sequencing before transformation into Escherichia coli strain BL21(DE3).

Bacteria were grown in LB medium containing 30 µg ml⁻¹ kanamycin at 310 K until the OD₆₀₀ reached 0.6–0.8. Expression of the recombinant His-tagged protein was then induced with 0.4 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) for 16 h at 289 K. The cells were harvested by centrifugation at 4000g for 15 min at 277 K, resuspended in lysis buffer consisting of 20 mM Tris–HCl pH 7.4, 300 mM NaCl, 10 mM imidazole, 5%(v/v) glycerol and subsequently lysed using a low-temperature ultrahigh-pressure continuous-flow cell crusher (JN-02C, JNBIO, People’s Republic of China) at 120 MPa and 277 K. After centrifugation at 16 000g for 45 min at 277 K, the supernatant of the cell lysate was subsequently loaded onto a Poly-Prep gravity-flow chromatography column (Bio-Rad) manually packed with Ni–NTA resin (Chelating Sepharose Fast Flow, GE Healthcare) pre-equilibrated with lysis buffer at 277 K. The target protein was eluted with an elution buffer consisting of 20 mM Tris–HCl pH 7.4, 300 mM NaCl, 300 mM imidazole, 5%(v/v) glycerol. After the addition of TEV protease to a mass ratio of 1:5 (TEV protease:MSMEG_0129), the column eluate was dialyzed against a buffer consisting of 20 mM Tris–HCl pH 7.4, 150 mM NaCl, 5%(v/v) glycerol overnight at 289 K, allowing parallel proteolysis. The cleaved His tag was then removed by loading the dialyzed sample onto a second Ni²⁺-affinity column. The flowthrough fractions were further purified by size-exclusion chromatography using a Superdex 75 10/300 GL column (GE Healthcare). The purified protein was concentrated to 15 mg ml⁻¹ (the protein concentration was determined by measuring A ₂₈₀ using a NanoDrop 2000c spectrophotometer) and stored at 193 K for subsequent crystallization experiments. Macromolecule-production information is summarized in Table 1 ▸.

Table 1. Macromolecule-production information.

Source organism	M. smegmatis mc2 155
DNA source	M. smegmatis genomic DNA
Forward primer†	GAGTCCATATGGTGAGCAAGACTGTCGAGGTCG
Reverse primer‡	TACTAGCGGCCGCTCAGCTCTGGGTGAGCTGCT
Cloning vector	pET-28a(+) modified by replacing the thrombin cleavage site with a TEV site
Expression vector	pET-28a(+) modified by replacing the thrombin cleavage site with a TEV site
Expression host	E. coli strain BL21(DE3)
Complete amino-acid sequence of the construct produced§	MGSSHHHHHHSSGENLYFQG HMVSKTVEVAASAETITSIVSDFEAYPQWNPEIKGCWILARYNDGRPSQLRLDVEIQGQSGVFITAVYYPAENQIFTMLQQGDHFTKQEQRFSIVPLGPDSTLLQVDLDVEVKLPVPGPMVKKLAGETLEHLAKALEGRVEQLTQS

^†The NdeI site is underlined.

^‡The NotI site is underlined.

^§The His-tag sequence and the cleavage site of TEV protease are underlined. The amino-acid sequence in italics is the crystallized protein amino-acid sequence.

2.2. Crystallization  

The protein concentration was adjusted to 5 mg ml⁻¹ before crystallization. Initial crystallization trials were carried out at 289 K using the sitting-drop vapour-diffusion method. Crystallization conditions were screened using five commercial kits from Hampton Research (Index, Crystal Screen, PEG/Ion, PEGRx and SaltRx) with a Mosquito crystallization robot (TTP Labtech). Crystals were observed under two conditions: (i) 3.5 M sodium formate, 0.1 M bis-tris propane pH 7.0 and (ii) 1.3 M ammonium tartrate, 0.1 M bis-tris propane pH 7.0. Optimization was subsequently performed using the hanging-drop vapour-diffusion method in 24-well plates at 289 K by mixing 1 µl protein solution with 1 µl reservoir solution and equilibrating against 500 µl reservoir solution. Crystal optimization was performed by altering the precipitant concentration and the pH and by the use of various additives from the Silver Bullets, Additive Screen and Detergent Screen HT kits (Hampton Research). Additive optimization was performed by mixing 1 µl protein solution with 0.2 µl additive and 0.8 µl reservoir solution. The best diffracting crystals were grown with 2.0 M sodium formate, 0.1 M MES pH 6.5. Crystallization information is summarized in Table 2 ▸.

Table 2. Crystallization.

Method	Hanging-drop vapour diffusion
Plate type	24-well hanging-drop plate
Temperature (K)	289
Protein concentration (mg ml−1)	5
Buffer composition of protein solution	20 mM Tris–HCl pH 7.4, 150 mM NaCl, 5%(v/v) glycerol
Composition of reservoir solution	2.0 M sodium formate, 0.1 M MES pH 6.5
Volume and ratio of drop	2 µl (1:1)
Volume of reservoir (µl)	500

2.3. Data collection and processing  

Crystals were soaked in cryoprotectant solution consisting of 2.0 M sodium formate, 0.1 M MES pH 6.5, 5%(v/v) glycerol for several seconds before mounting them and flash-cooling them in a liquid-nitrogen stream at 100 K. Diffraction data were collected from a single crystal on beamline BL19U1 at Shanghai Synchrotron Radiation Facility (SSRF), People’s Republic of China. A total of 360 images were recorded with a 1° oscillation step and 0.6 s exposure time per frame. The diffraction data were indexed, integrated and scaled using XDS (Kabsch, 2010 ▸). Data-collection and data-processing statistics are summarized in Table 3 ▸.

Table 3. Data collection and processing.

Values in parentheses are for the outer shell.

Diffraction source	BL19U1, SSRF
Wavelength (Å)	0.97853
Temperature (K)	100
Detector	PILATUS3 6M
Crystal-to-detector distance (mm)	300
Rotation range per image (°)	1
Total rotation range (°)	360
Exposure time per image (s)	0.6
Space group	P62
a, b, c (Å)	109.76, 109.76, 56.5
α, β, γ (°)	90, 90, 120
Mosaicity (°)	0.206
Resolution range (Å)	50–1.95 (2.00–1.95)
Total no. of reflections	416719 (31987)
No. of unique reflections	55131 (4031)
Completeness (%)	99.3 (97.5)
Multiplicity	7.6 (7.9)
〈I/σ(I)〉	19.12 (3.60)
R meas (%)	7.7 (62.2)
Overall B factor from Wilson plot (Å2)	31.23

3. Results and discussion  

Recombinant MSMEG_0129 protein was obtained in soluble form, with a yield of ∼15 mg protein being obtained from 1 l bacterial culture. Initial crystallizations were carried out using the His-tagged protein. Although the crystals obtained diffracted to 2.8 Å resolution, they were strongly anisotropic even after thorough optimization. To overcome this problem, the His tag was removed by TEV proteolysis before another round of crystallization trials was performed. TEV protease cleavage efficiencies of approximately 50 and 30% were obtained after overnight digestion at 289 K (Fig. 1 ▸ a) and 277 K, respectively. The elution volume of MSMEG_0129 during size-exclusion chromatography using a HiLoad Superdex 75 10/300 column (GE Healthcare) was 11.2 ml, indicating that the protein exists as a dimer in solution (Fig. 1 ▸ b). The final purity of MSMEG_0129 was greater than 90%, as determined by 15% SDS–PAGE (Fig. 1 ▸ c).

Figure 1.

(a) 15% SDS–PAGE of purified MSMEG_0129. Lane M, protein markers (labelled in kDa); lane 1, elution fraction from an Ni–NTA chelating column; lane 2, TEV protease-cleaved protein (cleaved overnight at 289 K); lane 3, flowthrough fraction from the second round of affinity chromatography. (b) Elution profile of MSMEG_0129 from a Superdex 75 10/300 GL column. (c) Content analysis of MSMEG_0129 crystals on a 15% SDS–PAGE gel. Lane M, protein markers (labelled in kDa); lane 1, purified MSMEG_0129; lane 2, dissolved MSMEG_0129 crystals obtained using ammonium tartrate as a precipitant; lane 3, dissolved MSMEG_0129 crystals obtained using sodium formate as a precipitant.

Two crystallization conditions were obtained from five commercial screening kits. The first condition was 1.3 M ammonium tartrate, 0.1 M bis-tris propane pH 7.0, under which branched rod-shaped crystals were obtained. After initial optimization by altering the precipitant concentration, the buffer composition and the pH, we obtained larger crystals with visible branches in a condition consisting of 1.3 M ammonium tartrate, 0.1 M HEPES pH 7.0, but the crystal monomorphism was not improved (Fig. 2 ▸ a). After one week of crystal growth, we manually removed the branches before soaking the main body of the crystals for several seconds in cryoprotectant solution composed of the reservoir solution supplemented with 5%(v/v) glycerol. One crystal of 300 × 40 × 40 µm in size diffracted to 3 Å resolution (Fig. 3 ▸ a) when exposed to X-rays on beamline BL19U1 at SSRF. Indexing of the X-ray data showed that the crystal belonged to the hexagonal point group P6, with unit-cell parameters a = 112.91, b = 112.91, c = 56.48 Å, α = 90, β = 90, γ = 120°. Two protomers are predicted to be present in the asymmetric unit based on a V _M value (Matthews, 1968 ▸) of 3.18 ų Da⁻¹ with a solvent content of 61.38%. This is in agreement with the observation of a noncrystallographic twofold axis in the self-rotation function (data not shown) and with the observation of a dimeric form in solution based on the elution volume from size-exclusion chromatography.

Figure 2.

Crystallization of MSMEG_0129 using ammonium tartrate (a, b, c) or sodium formate (d, e) as a precipitant. (a) Branched rod-shaped crystals obtained using 1.3 M ammonium tartrate, 0.1 M HEPES pH 7.0. (b, c) Single polyhedron crystals grown using 1.2 M ammonium tartrate, 0.1 M bis-tris propane pH 7.3 with CTAB (b) or MnCl₂ (c). (d) Needle clusters grown using 3.5 M sodium formate, 0.1 M bis-tris propane pH 7.0. (e) A single crystal obtained using 2.0 M sodium formate, 0.1 M MES monohydrate pH 6.5.

Figure 3.

X-ray diffraction patterns of MSMEG_0129 recorded on beamline BL19U1 at Shanghai Synchotron Radiation Facility, People’s Republic of China from a crystal grown under the condition 1.3 M ammonium tartrate, 0.1 M HEPES pH 7.0 with branches removed (a) and a single crystal grown with 2.0 M sodium formate, 0.1 M MES monohydrate pH 6.5 (b).

To improve the crystal quality, we also performed several optimization measures, including seeding, additive screening and post-crystallization treatments. Microseeding using a Seed Bead kit (Hampton Research) failed to produce single crystals. Additive screening using the Silver Bullets, Additive and Detergent Screen kits (Hampton Research) did change the crystal shape significantly; the addition of the detergent CTAB and MnCl₂, for example, gave single polyhedron crystals within 3–5 d (Figs. 2 ▸ b and 2 ▸ c). Unfortunately, however, these seemingly better-shaped crystals showed even poorer diffraction to 4–5 Å resolution on beamlines BL17U1 and BL19U1 at SSRF. Even after post-crystallization treatments such as annealing, dehydration (Heras & Martin, 2005 ▸) and cryoprotectant screening, the crystal diffraction quality did not improve. These additives may change the protein–protein interactions and lead to looser packing and higher solvent content in the crystals.

The second condition from the initial screening was 3.5 M sodium formate, 0.1 M bis-tris propane pH 7.0. Clusters of needle-shaped crystals grew under this condition. Subsequent optimization was performed with grid screening of the sodium formate concentration from 2.0 to 4.5 M and of the pH from 5.0 to 8.6 using sodium acetate, MES, HEPES and Tris as buffers. However, no crystals were observed two weeks after drop setup. We therefore discontinued coarse grid screening and refined the sodium formate concentration in grids from 3.1 to 3.9 M and the pH from 6.3 to 7.7 using bis-tris propane. The protein did not readily nucleate under these conditions; only a few crystals grew on the surface of some impurities that were accidently mixed into the crystallization drops, and these crystals still formed as clusters of needles (Fig. 2 ▸ d). After nearly one month, we unexpectedly found a single fusiform crystal in a drop equilibrated against a reservoir solution consisting of 2.0 M sodium formate, 0.1 M MES monohydrate pH 6.5 (Fig. 2 ▸ e).

The crystal grew rather slowly, reaching maximum dimensions of 120 × 20 × 20 µm approximately 20 d later. After soaking in cryoprotectant solution consisting of 2.0 M sodium formate, 0.1 M MES monohydrate pH 6.5, 5%(v/v) glycerol, this crystal was exposed to X-rays on beamline BL19U1 at SSRF. Diffraction data were collected over 360° of rotation in 1° steps. The crystal showed good diffraction but had strong ice rings (Fig. 3 ▸ b). We tried to reduce the impact of the ice rings by annealing, but the crystal soon lost diffraction. Unexpectedly, however, the data-evaluation statistics were satisfactory and the data set processed to 1.95 Å resolution was not greatly affected by the ice rings, as indicated by the reasonable completeness, I/σ(I) and R factors (Table 3 ▸). Similar to the crystal grown under the ammonium tartrate condition, this crystal belongs to point group P6, with unit-cell parameters a = 109.76, b = 109.76, c = 56.5 Å, α = 90, β = 90, γ = 120°. Two protein protomers are likely to reside in the asymmetric unit, as estimated from the V _M value (Matthews, 1968 ▸) of 3.01 ų Da⁻¹ and solvent content of 59.15%. All data-evaluation statistics are given in Table 3 ▸.

The structure was determined by molecular replacement using phenix.rosetta_mr. The structure of ZhuI from Strepto­mcyes sp. R1128 (PDB entry 3tfz; Ames et al., 2011 ▸) was used as a search model. A definite solution was obtained (LLG = 957.43 and R _free = 0.29 after autobuilding), confirming the presence of two protein protomers in the asymmetric unit and indicating the space group to be P6₂.

In this study, we fortuitously obtained a single crystal of MSMEG_0129 and subsequently collected a data set from it to atomic resolution, providing a good opportunity to determine the structure of a putative type II PKS from a mycobacterium. Structural information on MSMEG_0129 will provide mechanistic insights into the biosynthesis of aromatic polyketides in mycobacteria, and will likely facilitate our understanding of the biological roles played by Rv0164 in the metabolic pathways involved in cell-envelope formation in M. tuberculosis (Quadri, 2014 ▸).

Our experience with the crystallization trials implies that the crystal quality of MSMEG_0129 is closely linked to the crystal-growth rate. Branched rod-shaped crystals grew within several days under an ammonium tartrate condition; however, when sodium formate was used as the major precipitant growth was much slower (requiring months), and a considerable shrinkage in the unit-cell size (from 112.91 to 109.76 Å for the a and b axes) and a decrease in the solvent content (from 61.38 to 59.15%) occurred. These changes indicate more closely packed protein molecules in the crystal grown in the latter condition, which in turn significantly improved its diffraction quality (Frey, 1994 ▸).