Crystal structure determination of the halogenase CtcP from Streptomyces aureofaciens

The crystal structure of the selenomethionine-derivatized halogenase CtcP from Streptomyces aureofaciens, which functions in chlortetracycline biosynthesis, is reported at 2.7 Å resolution. The structure reveals a conserved monooxygenase domain and a unique C-terminal domain. Although FAD was not observed in the structure, the monooxygenase domain has a conserved FAD-binding pocket and active center.

Abstract

Chlortetracycline (CTC), a derivative of tetracycline (TC), is a broadly used antibiotic that inhibits the synthesis of bacterial proteins by competing with the A-site tRNA on ribosomes. A recent study showed that during the biosynthesis of CTC in Streptomyces aureofaciens, the halogenase CtcP catalyzes the final chlorination reaction and transforms TC into CTC. However, the structure of this fundamental enzyme is still lacking. Here, selenomethionine-derivatized CtcP from S. aureofaciens was overexpressed and purified and its structure was determined at 2.7 Å resolution. The structure of CtcP reveals the conserved monooxygenase domain shared by all flavin-dependent halogenases and a unique C-terminal domain. Although FAD was not observed in the structure, the monooxygenase domain has a conserved FAD-binding pocket and active center. The C-terminal domain displays an α-helical bundle fold, which could contribute to substrate specificity. This work provides a molecular basis for enzyme engineering to improve the industrial production of CTC.

1. Introduction

The tetracycline (TC) family contains some of the most commonly used broad-spectrum antibiotics against numerous pathogens, including Gram-positive and Gram-negative bacteria, as well as atypical organisms (Bahrami et al., 2012 ▸; Nelson & Levy, 2011 ▸; Grossman, 2016 ▸). TC inhibits the translation of bacterial proteins by binding to 70S ribosomes and competing with A-site tRNA (Wilson, 2009 ▸; Brodersen et al., 2000 ▸; Pioletti et al., 2001 ▸; Jenner et al., 2013 ▸). To date, multiple derivatives of TC have been discovered or semi-synthesized, including chlortetracycline (CTC). As a derivative, CTC has a chlorine group at the C7 position of its tetracycline backbone. It was the first member of the TC family to be isolated from Streptomyces aureofaciens more than 70 years ago (Duggar, 1948 ▸).

Intriguingly, S. aureofaciens produces both CTC and TC (Zhu et al., 2013 ▸). Under normal culture conditions this bacterium mainly synthesizes CTC, while TC is only a side product. However, CTC production can be inhibited by adding aminopterin to switch to demeclocycline production, or by adding mercaptothiazole or potassium bromide to switch to TC production. Extensive studies have already uncovered the biosynthetic pathway of TC and CTC in this bacterium (Dairi et al., 1995 ▸; Nakano et al., 2000 ▸, 2004 ▸). Moreover, a recent study found that the rate-limiting chlorination step in CTC biosynthesis is catalyzed by two enzymes: the flavin adenine dinucleotide (FAD)-dependent halogenase CtcP and the flavin reductase CtcQ (Zhu et al., 2013 ▸).

In S. aureofaciens, CtcP depletion can completely block the production of CTC without potassium bromide supplementation. Moreover, overexpression of CtcP in this bacterium remarkably increases the yield of CTC. In addition, CtcP preferentially catalyzes the chlorination reaction on authentic CTC (4S) rather than 4-epi-CTC (4R) in vitro (Zhu et al., 2013 ▸). Thus, CtcP is an active halogenase responsible for the final chlorination reaction during CTC biosynthesis. CtcP, formerly known as Cts4, contains 555 amino-acid residues and shares 72% sequence identity with DacE (Wang et al., 2012 ▸), 47% with PrnA (Dong et al., 2005 ▸) and 41% with PltM (Zhang & Parry, 2007 ▸; Mori et al., 2019 ▸). It harbors the classic monooxygenase domain shared by all flavin-dependent halogenases, which contains all of the FAD/FMN-binding motifs that are necessary to coordinate FADH₂ during the reaction (Dym & Eisenberg, 2001 ▸; Phintha et al., 2020 ▸; Blasiak & Drennan, 2009 ▸). Similar to other halogenases, CtcP harbors a C-terminal domain that may contribute to its substrate recognition (Phintha et al., 2020 ▸; Blasiak & Drennan, 2009 ▸). However, no structural information on CtcP is yet available.

Here, selenomethionine-derivatized CtcP was overexpressed, purified and crystallized. Using single-wavelength anomalous diffraction (SAD), its structure was solved at 2.7 Å resolution in an apo form. Structure analysis revealed that the N-terminus of CtcP contains the classic flavin-dependent halogenase fold and the C-terminus forms an α-helical bundle, which together with the N-terminal domain forms the substrate-binding pocket. In summary, the structure provides a molecular basis for enzyme engineering to improve the industrial production of CTC.

2. Materials and methods

2.1. CtcP protein purification

The CtcP gene was amplified from S. aureofaciens genomic DNA using the following primers: forward primer, GATCGCGGATCCATGACCGACACAACCGCGGATCAG; reverse primer, GCACCGGAATTCCTACGGCTTGACCCGCATCGCCG. The DNA fragment corresponding to the correct CtcP gene was purified and then digested using BamHI and EcoRI restriction enzymes overnight at 37°C. Digested fragments were inserted into a modified pET-28a vector. The final CtcP construct (pET-28a-CtcP) contains a 6×His tag followed by a SUMO protein tag at its N-terminus. Macromolecule-production information is summarized in Table 1 ▸.

Table 1. Macromolecule-production information.

Source organism	Streptomyces aureofaciens
DNA source	Streptomyces aureofaciens genomic DNA
Forward primer	GATCGCGGATCCATGACCGACACAACCGCGGATCAG
Reverse primer	GCACCGGAATTCCTACGGCTTGACCCGCATCGCCG
Expression vector	pET-28a
Expression host	E. coli BL21 (DE3)
Complete amino-acid sequence of construct produced	SMTDTTADQTRHGDRPYDVVIIGSGLSGTMLGSILAKHGFRIMLLDGAHHPRFAVGESTIGQTLVVLRLISDRYGVPEIANLASFQDVLANVSSSHGQKSNFGFMFHRDGEEPDPNETSQFRIPSIVGNAAHFFRQDTDSYMFHAAVRYGCDARQYYRVENIEFDDGGVTVSGADGSTVRARYLVDASGFRSPLARQLGLREEPSRLKHHARSIFTHMVGVDAIDDHVDTPAELRPPVPWNDGTMHHIFERGWMWIIPFNNHPGATNPLCSVGIQLDERRYPARPDLTPEEEFWSHVDRFPAVQRQLKGARSVREWVRTDRMQYSSSRTVGERWCLMSHAAGFIDPLFSRGLSNTCEIINALSWRLMAALREDDFAVERFAYVEELEQGLLDWNDKLVNNSFISFSHYPLWNSVFRIWASASVIGGKRILNALTRTKETGDDSHCQALDDNPYPGLWCPLDFYKEAFDELTELCEAVDAGRTTAEEAARVLEQRVRESDWMLPALGFNDPDTHHINPTADKMIRIAEWATGHHRPEIRELLAASAEEVRAAMRVKP

The plasmids were transformed into Escherichia coli BL21 (DE3) cells by heat-shock transformation and the cells were cultured in 10 ml LB medium overnight. They were then transferred to 2 l 2×YT medium supplemented with kanamycin (50 µg ml⁻¹) and culture continued at 37°C. When the OD₆₀₀ of the culture reached 1.0, expression of the CtcP protein was induced by adding 0.1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) at 15°C overnight. The cells were harvested by centrifugation, resuspended in lysis buffer (20 mM Tris–HCl pH 8.0, 300 mM NaCl) and lysed using a French press. The cell lysate was then centrifuged at 15 000 rev min⁻¹ for 15 min and the supernatant was loaded onto Ni–NTA resin (GE Healthcare) pre-equilibrated with lysis buffer. The bound protein was washed three times with wash buffer (20 mM Tris–HCl pH 8.0, 300 mM NaCl, 20 mM imidazole) and digested on-column by adding Ulp1 protease overnight to elute the CtcP protein. The cleaved protein was further purified by ion-exchange chromatography and size-exclusion chromatography, concentrated using an Amicon Ultra concentrator and stored in buffer consisting of 10 mM Tris–HCl pH 8.0, 150 mM NaCl, 5 mM DTT.

To purify selenomethionine-derivatized protein, after the overnight preculture the transformants were transferred into M9 medium and culture continued at 37°C. When the OD₆₀₀ of the culture reached 0.6–0.8, the cells were switched into a shaker at 15°C. The target CtcP protein was induced by adding 0.1 mM IPTG and 50 mg l⁻¹ selenomethionine to the medium overnight.

2.2. Crystallization and data collection

Initial crystallization screening was carried out using commercially available kits from Hampton Research at 4 and 18°C. The final crystals of selenomethionine-derivatized CtcP protein were grown at 18°C using the hanging-drop vapor-diffusion method by mixing equal volumes of protein solution and crystallization buffer consisting of 0.1 M bis-Tris pH 6.0, 0.2 M magnesium sulfate, 20%(w/v) polyethylene glycol 3350. Crystals were slowly equilibrated with a cryoprotectant buffer containing reservoir buffer plus 5%(v/v) glycerol and were flash-cooled in liquid nitrogen. All data sets were collected on beamline BL17U1 at the Shanghai Synchrotron Radiation Facility (SSRF) using an ADSC Quantum 315r detector. The exposure time was 0.5 s per frame and the oscillation range was 1° per frame. The data were processed using HKL-2000 (Otwinowski & Minor, 1997 ▸). In order to obtain crystal structures of CtcP with FAD or TC, crystal soaking was carried out. Different concentrations of FAD and TC were incubated with the CtcP crystals for different times. However, none of the conditions that were tried preserved the diffraction of the crystals. The crystals even dissolved when incubated with higher concentrations of the molecules for longer times. Thus, in the end, no complex structures could be obtained. Crystallization information is provided in Table 2 ▸.

Table 2. Crystallization.

Method	Hanging-drop vapor diffusion
Plate type	24-well plate
Temperature (K)	291
Protein concentration (mg ml−1)	10
Buffer composition of protein solution	10 mM Tris–HCl pH 8.0, 150 mM NaCl, 5 mM DTT
Composition of reservoir solution	0.1 M bis-Tris pH 6.0, 0.2 M magnesium sulfate heptahydrate, 20%(w/v) polyethylene glycol 3350
Volume and ratio of drop	2 µl, 1:1 ratio
Volume of reservoir (µl)	200

2.3. Structure determination

The structure was determined by the single-wavelength anomalous dispersion (SAD) method with Phaser (McCoy et al., 2007 ▸) from the Phenix suite (Liebschner et al., 2019 ▸). A total of 49 selenomethionine residues could be found in one asymmetric unit. The atomic model was built manually using Coot (Emsley et al., 2010 ▸) and was refined with phenix.refine (Adams et al., 2010 ▸). The structure was validated using MolProbity (Chen et al., 2010 ▸). All figures were generated by ChimeraX (Goddard et al., 2018 ▸). The data-collection and final refinement statistics are summarized in Tables 3 ▸ and 4 ▸.

Table 3. Data collection and processing.

Values in parentheses are for the outer shell.

Diffraction source	BL17U, SSRF
Wavelength (Å)	0.97923
Temperature (K)	100
Detector	ADSC Q315r
Crystal-to-detector distance (mm)	400
Rotation range per image (°)	1
Total rotation range (°)	360
Exposure time per image (s)	0.5
Space group	P22121
a, b, c (Å)	116.393, 121.281, 179.194
α, β, γ (°)	90, 90, 90
Mosaicity (°)	0.651
Resolution range (Å)	50–2.7 (2.8–2.7)
Total No. of reflections	1038277
No. of unique reflections	70271
Completeness (%)	100
Multiplicity	14.8 (14.9)
〈I/σ(I)〉	15.5 (4.4)
R r.i.m. (%)	0.177 (0.812)
Overall B factor from Wilson plot (Å2)	37.030

Table 4. Structure refinement.

Values in parentheses are for the outer shell.

Resolution range (Å)	46.11–2.70 (2.797–2.700)
Completeness (%)	99.77 (98.15)
No. of reflections, working set	68153 (6598)
No. of reflections, test set	2010 (206)
Final R cryst (%)	0.1766 (0.2324)
Final R free (%)	0.2240 (0.3227)
No. of non-H atoms
Total	17336
Solvent	288
Protein	17048
R.m.s. deviations
Bond lengths (Å)	0.010
Angles (°)	1.16
Average B factors (Å2)
Overall	38.35
Macromolecules	38.44
Solvent	33.08
Ramachandran plot
Favored (%)	97.00
Additionally allowed (%)	2.82
Outliers (%)	0.18

3. Results

3.1. The overall structure of CtcP

To obtain the crystal structure, the full-length CtcP protein with one extra serine residue at the N-terminus left by Ulp1 protease cleavage was used. Wild-type and selenomethionine-derivatized CtcP proteins were purified to homogeneity, which showed that CtcP is a monomer in solution. Both native and selenomethionine-derivatized CtcP crystals could be obtained, but only the latter diffracted well. Using a data set collected from a selenomethionine-derivatized CtcP crystal, the structure was determined by the single-wavelength anomalous dispersion (SAD) method at 2.7 Å resolution. Four molecules were observed in one crystallo­graphic asymmetric unit packed in space group P22₁2₁. Although the full-length protein was used for crystallization, only residues ranging 7–551 could be built in the final model. Unfortunately, despite the purified protein displaying a yellowish color indicative of the presence of FAD, no reliable density for FAD molecules could be built. Soaking crystals in solutions with high concentrations of FAD or TC also failed to lead to the solution of structures of their complexes with CtcP.

Like other FAD-dependent halogenases, the overall structure of CtcP can be clearly divided into N-terminal and C-terminal domains (Phintha et al., 2020 ▸; Dym & Eisenberg, 2001 ▸; Blasiak & Drennan, 2009 ▸). The N-terminal domain displays a Rossmann-like fold and harbors all of the FAD-binding motifs, which are supposed to bind to FADH₂ during the reaction (Phintha et al., 2020 ▸; Dym & Eisenberg, 2001 ▸; Blasiak & Drennan, 2009 ▸). The C-terminal domain displays an α-helical bundle fold that contains seven protein α-helices. It binds closely to the N-terminal domain and contributes to the formation of the substrate-binding pocket (Fig. 1 ▸).

Figure 1.

The crystal structure of CtcP. The crystal structure of CtcP from S. aureofaciens is shown in ribbon representation in two different views. The N-terminal (green) and C-terminal (blue) domains of CtcP are color-coded. α-Helix 10 from the N-terminal domain and α-helices 11–17 from the C-terminal domain are labeled. Due to the slight difference among the four molecules in the asymmetric unit, only residues 15–551 were built in this chosen copy.

3.2. Structure comparison of CtcP with other halogenases

The S. aureofaciens CtcP structure solved here was then compared with available structures of all types of halogenases from various species. The results revealed that the S. aureofaciens CtcP structure had the highest similarity to that of the phenolic halogenase PltM from Pseudomonas protegens (Mori et al., 2019 ▸). A low r.m.s.d. value (1.026 Å) was obtained by a direct structural alignment of 2465 atoms between CtcP and PltM (PDB entry 6bza), confirming their high conformational similarity (Mori et al., 2019 ▸). Although both the N-terminal and the C-terminal domains are structurally comparable in CtcP and PltM, the C-terminus of CtcP contains three extra α-helices that closely fold together with the rest of the C-terminal domain and form a larger helical bundle fold (Fig. 2 ▸).

Figure 2.

Structural comparison of CtcP with PltM. Two different views of a structural comparison between the P. protegens halogenase PltM in complex with phloroglucinol (PDB entry 6bza) and CtcP (this study). The CtcP protein is rainbow-colored, while PltM is colored gray.

The bacterial halogenase PltM is capable of using diverse halides for the halogenation of various types of phenolic compounds, including phloroglucinol (Mori et al., 2019 ▸). In a previous study, its structure was solved with a FAD ligand and a phloroglucinol substrate. Comparison with the structure of PltM in complex with FAD shows that all of the essential residues that recognize FAD are highly conserved, indicating that CtcP is likely to be capable of binding FAD. Moreover, the conformation of the residue that transfers the halogen atom to the substrate (Lys87 in PltM and Lys98 in CtcP) in the active center is also conserved (Fig. 3 ▸).

Figure 3.

Structural comparison of the FAD- and substrate-binding sites. (a) All of the essential residues which coordinate FAD in both CtcP and PltM are shown in stick representation. CtcP is colored green, while PltM is colored gray. The FAD molecule and the Cl⁻ ion are from the model of PltM. (b) Comparison of the substrate-binding site in CtcP and PltM. The enzymatic active-center residues (Lys87 in PltM and Lys98 in CtcP) and the phloroglucinol from the PltM structure are shown in stick representation. CtcP is colored green and blue, while PltM is colored gray.

Interestingly, the structure of the substrate-binding pocket of PltM with phloroglucinol differs from that of S. aureo­faciens CtcP. Due to the extra three α-helices in the C-terminal domain and a longer internal loop region (124–128) in the N-terminal domain, the substrate-binding pocket of CtcP is wider and thus allows tetracycline to be accommodated, which is larger than phloroglucinol (Fig. 3 ▸ b).

In conclusion, the crystal structure of the halogenase CtcP from S. aureofaciens, which is responsible for the final crucial step in CTC biosynthesis, was solved. In future studies, molecular-dynamics simulation or small-molecular docking can be applied to reveal its substrate specificity. Since CTC is a widely used antibiotic both in human medicine and animal feed, the structure can provide structural information to guide enzyme engineering to improve the production of CTC or TC.

Supplementary Material

PDB reference: CtcP, 7xgb