Introduction

Phosphorous is one of cellular component important for many biological and biochemical processes in living organisms, such as the formations of nucleic acids (DNA and RNA) and membrane phospholipids, post-translational modifications for signal transduction, etc. [1]. The usefulness of phosphorous has been applied to extensive field such as agriculture, medicine and pharmaceuticals. In particular, phosphorous is required to secure a high level of productivity in agriculture [2, 3]. In bacteria, the major form of phosphorous is orthophosphate known as inorganic phosphates (Pi) [4]. Despite the varying importance of Pi in cellular function, it is usually found at very low concentration in natural environment [5]. Therefore, bacteria and other organisms must have relevant systems that include physiological and biochemical responses to overcome the deficiency of this nutrient [6].

A unique mechanism associated with the maintenance of Pi in bacteria as a regulatory circuit is the phosphate (Pho) regulon that is regulated by a two-component system (TCS) [5, 6]. The Pho regulon is one of the most rational and effective regulatory mechanisms. It is well-studied in model cells such as Escherichia coli [7] and Bacillus subtilis [8]. Later, it has been characterized in many other bacterial species [9]. TCSs are signal transduction pathways commonly used by bacteria to recognize and adapt to stimuli caused by environmental changes. TCS consists of histidine kinase (HK) as an inner-membrane sensor kinase that recognizes several specific environmental signals and the transcriptional response regulator (RR) protein that mediates cellular responses by regulating expression of specific genes or modulating protein functions in the cytoplasm [10]. Although these proteins are known by different names in some bacteria [11, 12], upon Pi deficiency, the RR is phosphorylated by the HK. Thereby, the phosphorylated RR can bind to specific DNA sequences and then activate or suppress the transcription of their corresponding genes [13, 14].

A number of new members of Pho regulon have been identified from several bacteria in past years, but there still remain numerous undiscovered questions such as the detailed function of the entire system and the mechanisms connecting the Pho regulon to pathogenesis [5]. Among them, enterococci are normal flora in human intestine of healthy adults, but also they are one of the major causes of hospital infections that leads a variety of diseases, including bacteremia, urinary tract and central nervous system infections [15]. Most clinical isolates of enterococci are Enterococcus faecalis along with Enterococcus faecium [16]. However, the TCS related to the Pho regulon in this strain has not been well studied except for the VanRS system that regulates the resistance of enterococci to vancomycin [17].

Most RRs have two distinct domains involving the receiver domain of N-terminus and the effector domain of C-terminus. On the basis of the structure and function of the effector domains, they can be classified into subfamilies [14]. PhoP belongs to the OmpR/PhoB subfamily of RRs, including the OmpR and PhoB as the representative members [18]. To date, only a few structures of PhoP have been reported because it is difficult to crystallize the full length of RRs in this subfamily [19]. Thus, more detailed structure information is required to understand their functional mechanisms such as the conformational changes accompanying with phosphorylation of PhoP and to compare with known PhoP structures. To determine its structure, a truncated form of PhoP from E. faecalis containing the receiver domain (EfPhoP-RD) was constructed as the first step. Here, we report the crystallization conditions and preliminary X-ray crystallographic analysis of EfPhoP-RD. Complete diffraction data sets was collected from apo-crystals at resolutions up to 3.5 Å.

Materials and methods

Overexpression and purification of EfPhoP-RD protein

The EfPhoP-RD gene was amplified from E. faecalis ATCC 29212 genomic DNA by polymerase chain reaction (PCR) using the forward and reverse primers. The primers contained respective modifications to treat suitable restriction endonucleases for insertion into the vector, where NdeI restriction site in the forward primer and the XhoI restriction site in the reverse primer are underlined in Table 1. The PCR-amplified DNA fragment was digested with NdeI and XhoI and was then inserted into the pET-28a expression vector (Novagen, USA). To enhance the solubilization of protein, the plasmid pEfPhoP-RD was generated to be expressed as EfPhoP-RD with hexa histidine-tag at the N- and C-termini. Transformed E. coli BL21 (DE3) cells (Novagen, USA) harboring pEfPhoP-RD were grown in Luria–Bertani medium with 50 μg/ml kanamycin at 25 °C to an optical density at 600 nm of 0.5–0.6. Overexpression of recombinant EfPhoP-RD protein was induced by 0.5 mM isopropyl-β-d-1-thiogalactopyrano-side and incubation at 18 °C for a further 12 h. The cells were harvested by centrifugation at 5000g for 20 min at 4 °C.

Table 1 Macromolecule-production information
Full size table

The harvested cell pellets were resuspended in pre-equilibrium buffer A (0.02 M Tris–HCl, pH 7.5, 0.5 M NaCl, 10% glycerol) adding 1 mM phenylmethylsulfonyl fluoride and ruptured by ultrasonication at 4 °C. The crude lysate was centrifuged at 25,000g for 20 min at 4 °C. The supernatant was loaded onto a nickel (Ni2+) charged HisTrap HP column (GE Healthcare, USA) equilibrated in buffer A. The bound EfPhoP-RD on the column was eluted with a linear gradient of elution buffer containing 0.02 M Tris–HCl, pH 7.5, 0.5 M imidazole, 0.5 M NaCl, 10% glycerol. The collected each fraction was confirmed by 15% SDS-PAGE, and subsequently purified by size exclusion chromatography on a HiLoad Superdex 200 column (GE Healthcare, USA) pre-equilibrated with buffer containing 0.02 M Tris–HCl pH 7.5, 0.15 M NaCl, 10% glycerol. The collected fractions containing EfPhoP-RD were pooled and concentrated to 7.4 mg/ml using an Amicon Ultra-15 centrifugal filter device (Millipore, USA).

Protein crystallization

Preliminary screening for the crystallization of EfPhoP-RD was performed by the hanging-drop vapour-diffusion method in 96-well microplates at 21 °C using various commercial screening kits such as Crystal Screen 1 and 2, PEGRx 1 and 2 (Hampton Research, USA), and Wizard Classic 1, 2, 3 and 4 (Rigaku Reagents Inc., USA). Initial crystals were obtained from two solutions as follows: the condition No. 30 of Crystal Screen 2 [0.1 M HEPES, pH 7.5, 5% (v/v) (+/−)-2-methyl-2,4-pentanediol, 10% (w/v) polyethylene glycol (PEG) 6000] and the condition No. 19 of PEGRx 2 [0.1 M Bis–Tris–propane, pH 9.0, 0.1 M NaCl, 25% (w/v) PEG 1500]. Optimization of EfPhoP-RD crystal was performed with the hanging-drop vapour-diffusion method in 24-well VDX plates (Hampton Research, USA) under conditions containing various PEGs and pH ranges; Each hanging drop was made by adding 1 μl protein solution to 1 μl reservoir solution to be a total volume of 2 μl and was then equilibrated against 500 μl reservoir solution.

Collection and analysis of X-ray diffraction data

For the collection of X-ray diffraction data in cryogenic condition, all EfPhoP-RD crystals were transferred to a cryoprotection solution with 25% (v/v) ethylene glycol added to each reservoir solution. The cryoprotected crystal was then rapidly cooled at − 180 °C in a steam of liquid nitrogen. The sets of X-ray diffraction data were collected at beamline 7A in the Pohang Light Source (Pohang, South Korea) using an ADSC Quantum 270r CCD detector. A total range of 180° was covered using 1° oscillation and 5 s exposure/frame for thick plate-shape crystal, whereas a total range of 360° was covered using 1° oscillation and 2 s exposure/frame for thin rod-shape crystal. The wavelength of synchrotron radiation was 1,0000 Å. The crystal-to-detector distances of thick plate-shape and thin rod-shape crystals were 450 and 350 mm, respectively. All diffraction data sets were indexed to identify the unit cell and space group of the crystal, and then scaled after integration of the indexed data using the HKL2000 software package [20]. Detailed information on data collection is shown in Table 2.

Table 2 Data-collection statistics
Full size table

Results and discussion

The gene encoding EfPhoP-RD (residue 1–130) from E. faecalis ATCC 29212 was successfully cloned into bacterial expression vector pET-28a. The recombinant EfPhoP-RD protein containing 130 residues with a calculated molecular weight of 17.8 kDa, was overexpressed in E. coli BL21 (DE3). The purification of protein was carried out in two steps using a nickel-charged HisTrap HP and size exclusion columns. The purified EfPhoP-RD protein showed a single band on 15% SDS-PAGE, with estimated purity over 95% (Fig. 1). Amino acid sequence comparison of EfPhoP-RD was performed by multiple alignment analysis with receiver domains of PhoPs from E. coli (EcPhoP-RD) [21], Mycobacterium tuberculosis (MtPhoP-RD) [19] and B. subtilis (BsPhoP-RD) [22] as known structures of the OmpR/PhoB subfamily. EfPhoP-RD (residue 1–130) shared low sequence identities with EcPhoP-RD (residue 1–121), MtPhoP-RD (residue 19–140), and BsPhoP-RD (residue 3–132) (36%, 41% and 54%, respectively) (Fig. 2).

Fig. 1
figure 1

15% SDS-PAGE analysis of EfPhoP-RD. Lane 1, molecular-weight marker (labelled in kDa); lane 2, purified EfPhoP-RD protein

Full size image
Fig. 2
figure 2

Multiple alignment analysis of EfPhoP-RD with orthologous PhoP-RDs, as available structure of the OmpR/PhoB subfamily. Amino acid sequences are as follows: PhoP-RD from E. coli (EcPhoP-RD), PhoP-RD from M. tuberculosis (MtPhoP-RD) and PhoP-RD from B. subtilis (BsPhoP-RD). Hydrophobic, polar, acidic, and basic residues are shown in red, green, blue, and magenta, respectively. (*) identical residues, (:) conserved residues, and (.) semi-conserved residues

Full size image

Based on initial crystallization conditions, crystals of EfPhoP-RD suitable size for X-ray diffraction were obtained within 4 days using two optimized reservoir solutions as follows. Several thick plate-shape and thin rod-shape crystals were obtained under reservoir solutions consisting of 0.1 M Bis–Tris–propane, pH 9.0, 6–8% (w/v) PEG 6 K and 0.1 M HEPES, pH 7.5, 4–5% (w/v) PEG 10 K, respectively. Dimensions of thick plate-shape and thin rod-shape crystals were approximately 0.3 × 0.3 × 0.1 mm (Fig. 3a) and 0.1 × 0.1 × 0.8 mm (Fig. 3b), respectively. Thick plate-shape crystal of EfPhoP-RD diffracted at a low resolution of 5.0 Å (Fig. 4a). It was found to belong to the orthorhombic space group P212121, with unit-cell parameters a = 188.632, b = 187.706, c = 197.488 Å. Assuming the presence of forty monomers per asymmetric unit, the Matthews coefficient VM value [23] was calculated to be 2.45  Å3 Da−1, with estimated solvent content of 49.87%. On the other hand, thin rod-shape crystal of EfPhoP-RD diffracted to better than 3.5Å resolution (Fig. 4b). It was found to belong to the orthorhombic space group C2221, with unit-cell parameters a = 118.743, b = 189.826, c = 189.882 Å. Assuming the presence of twelve monomers per asymmetric unit, the Matthews coefficient VM value [23] was calculated to be 2.50  Å3 Da−1, with estimated solvent content of 50.85%.

Fig. 3
figure 3

Crystals of PhoP-RD from E. faecalis ATCC 29212. a Thick plate-shape crystals of EfPhoP-RD; b Thin rod-shape crystals of EfPhoP-RD. The crystal dimensions of EfPhoP-RD are approximately 0.3 × 0.3 × 0.1 mm for thick plate-shape crystals and 0.1 × 0.1 × 0.8 mm for thin rod-shape crystals

Full size image
Fig. 4
figure 4

X-ray diffraction patterns of EfPhoP-RD thick plate-shape a and thin rod-shape b crystals obtained using an ADSC Quantum 270r CCD detector. A resolution circle is shown at 5.0 Å (a) and 3.5 Å (b). The red boxes show the magnification of an area containing high-resolution spots indicated by the arrows

Full size image

Molecular replacement was attempted using MOLREP program [24] in the CCP4 [25] with the crystal structure of BsPhoP-RD (PDB ID: 1MVO) [22] indicating 54% sequence identity as a search model. However, our attempts could not provide a clear solution of structure for further refinement. This implies that the structure of EfPhoP-RD might contain a novel or different fold compared to other PhoP-RDs, although all data sets had low completeness and resolution. Therefore, the structure of EfPhoP-RD will be further determined by the MAD method [26] using selenomethionine substituted protein to solve the phase problem.