Crystallization and preliminary X-ray analysis of the C-type lectin domain of the spicule matrix protein SM50 from Strongylocentrotus purpuratus

The C-type lectin domain of the spicule matrix protein SM50 from S. purpuratus was purified and crystallized as a fusion protein with SUMO.

Abstract

Sea urchin spicules have a calcitic mesocrystalline architecture that is closely associated with a matrix of proteins and amorphous minerals. The mechanism underlying spicule formation involves complex processes encompassing spatio-temporally regulated organic–inorganic interactions. C-type lectin domains are present in several spicule matrix proteins in Strongylocentrotus purpuratus, implying their role in spiculogenesis. In this study, the C-type lectin domain of SM50 was overexpressed, purified and crystallized using a vapour-diffusion method. The crystal diffracted to a resolution of 2.85 Å and belonged to space group P2₁2₁2₁, with unit-cell parameters a = 100.6, b = 115.4, c = 130.6 Å, α = β = γ = 90°. Assuming 50% solvent content, six chains are expected to be present in the asymmetric unit.

1. Introduction  

The phenomenon of biomineralization encompasses diverse and widespread processes that involve organic–inorganic interactions by which organisms form composite hierarchical materials (Lowenstam & Weiner, 1989 ▶; Mann, 2001 ▶; Fratzl & Weinkamer, 2007 ▶). The sea urchin spicule is a model system for investigating the mechanisms of calcium carbonate biomineralization (Wilt, 2002 ▶). The spicule has a mesocrystalline architecture composed of crystallographically aligned calcite particles (50–200 nm) organized in a matrix of proteins and amorphous calcium carbonate (Seto et al., 2012 ▶). On account of this structure, the spicule has unique properties such as single-crystal-like diffraction and a conchoidal fracture surface typical of amorphous materials. Hence, the mechanisms underlying spicule formation have attracted interest from multiple disciplines (Knapp et al., 2012 ▶; Rao et al., 2013 ▶; Tester et al., 2013 ▶).

Among the proteins regulating spicule formation, SM50 is a 48.5 kDa nonglycosylated, secreted protein with an alkaline pI (Benson et al., 1987 ▶; Killian & Wilt, 1996 ▶). The N-terminal region of SM50 harbours a C-type lectin (CTL) domain (13.6 kDa), which can affect calcium carbonate mineralization (Killian & Wilt, 2008 ▶; Rao et al., 2013 ▶). Although their functions are as yet unknown, CTL domains are also present in other proteins associated with the sea urchin spicule, such as the SM30 family (Livingston et al., 2006 ▶). Thus, CTL domains appear to play an important role in calcium carbonate biomineralization. In nature, proteins with the CTL fold, such as the type II antifreeze proteins, phospholipase receptors and coagulation-factor-binding proteins, have diverse functions (Drickamer, 1999 ▶). These proteins can also bind to carbohydrate ligands, Ca²⁺ ions and form oligomers (Drickamer, 1993 ▶). To elucidate their role in bio­mineralization, structural and functional understanding of the CTL domain with respect to calcium carbonate mineralization is important. The highest sequence identity of the CTL domain of the SM50 protein from Strongylocentrotus purpuratus to a protein of known structure is 32% to snake-type CTL (PDB entry 3ubu; Gao et al., 2012 ▶). The snake CTL binds specifically to blood platelet glycoproteins and inhibits adhesion and aggregation (Gao et al., 2012 ▶). Here, we report the purification, crystallization and preliminary X-­ray analysis of the CTL domain of SM50 spicule matrix protein in fusion with SUMO protein.

2. Experimental procedures  

2.1. Overexpression and purification  

Protein expression and purification were carried out as described previously (Rao et al., 2013 ▶). Briefly, the CTL domain (13.6 kDa) of the larval spicule matrix protein SM50 from S. purpuratus N-­terminally fused with a cleavable (Ulp1 protease) His₆-SUMO tag was cloned in pET-24a vector (Fig. 1 ▶). Escherichia coli BL21 CodonPlus-RIL cells were transformed with the pET-24a vector and cultured in LB medium at 30°C. On reaching an OD₆₀₀ of 0.6, expression was induced with 0.5 mM IPTG overnight at 20°C. Cells were harvested and suspended in lysis buffer consisting of 20 mM HEPES, 50 mM NaCl, 10 mM β-mercaptoethanol (β-ME) pH 7.0, 10 pg ml⁻¹ pepstatin, 0.5 mM PMSF, 0.5 mg ml⁻¹ lysozyme (Roth, Germany) and were incubated for 30 min on ice. The resulting cell lysate was centrifuged at 16 000 rev min⁻¹ for 30 min and the resulting supernatant was passed over an Ni–NTA column (Qiagen). After initial washing with five bed volumes of 20 mM HEPES, 50 mM NaCl, 0.5 mM imidazole, 10 mM β-ME pH 7.0, the protein was eluted with 20 mM HEPES, 50 mM NaCl, 200 mM imidazole, 10 mM β-ME pH 7.0. The eluted protein was concentrated with Vivaspin concentrators (5 kDa cutoff) and loaded onto a Superdex 75 column (GE Healthcare) for final purification in 20 mM HEPES, 50 mM NaCl, 10 mM β-­ME pH 7.0. A single peak was observed and the peak fractions were analyzed by SDS–PAGE (Fig. 2 ▶). When the SUMO tag was cleaved, the CTL domain was unstable and aggregated; therefore, crystallization was performed using the SUMO-CTL fusion protein.

Figure 1.

Sequence of the SUMO-CTL fusion protein used for crystallization. The SUMO domain with an N-terminal His₆ tag is shown in green and the CLT domain of the SM50 protein (NCBI Reference Sequence NP_999775.1) is shown in magenta.

Figure 2.

SDS–PAGE showing purified SUMO-CTL after size-exclusion chromatography purification (Superdex 75). The individual molecular weights of SUMO and the CTL domain are 12.1 and 13.6 kDa, respectively. SUMO-CTL runs at an apparent molecular weight near 25 kDa. The right lane contains molecular-weight marker (labelled in kDa).

2.2. Crystallization  

Prior to crystallization, the protein was concentrated to 5 mg ml⁻¹ (Bradford assay) and filtered through a 0.1 µm filter (Millipore). Crystallization screening was performed with the sitting-drop method in a 96-well format using a Gryphon Robot (Art Robbins Instruments). Crystals were obtained in a condition with reservoir solution consisting of 0.1 mM HEPES pH 7.5, 70% MPD using a protein:reservoir ratio of 1:1(v:v) (0.2 µl protein solution and 0.2 µl reservoir solution) at 18°C. Further optimization was performed in both hanging drops and sitting drops.

2.3. Data collection and processing  

The harvested crystals were directly flash-cooled in liquid nitrogen. Data were collected on the X06DA beamline at the Swiss Light Source (SLS), Villigen, Switzerland. The best crystal diffracted isotropically to a resolution of 2.85 Å. The diffraction data set was processed with the X-ray Detector Software (XDS program package; Kabsch, 2010a ▶,b ▶). The space-group assignment was performed with POINTLESS (Evans, 2006 ▶) and further analysis of data quality was carried out with phenix.xtriage (Zwart et al., 2005 ▶).

3. Results and discussion  

The SUMO-CTL crystals appeared after three weeks. The crystals were very fragile. When the wells were opened, the crystals tended to lose their crystalline shape and changed to spherical droplets resembling the droplets observed during phase separation of organic solvents such as PEG and MPD. Addition of glycerol and ethylene glycol did not improve the crystal quality. Crystals were optimized in hanging-drop and sitting-drop conditions, and the best diffracting crystals were obtained using the sitting-drop method (Fig. 3 ▶). Crystals were quickly harvested using a CryoLoop (Hampton Research) directly for flash-cooling in liquid nitrogen. The best crystal diffracted to a resolution of 2.85 Å and belonged to the orthorhombic space group P2₁2₁2₁ (Table 1 ▶). The SUMO-CTL fusion protein with a total molecular weight of 25.9 kDa is monomeric in solution as shown by analytical ultracentrifugation (Rao et al., 2013 ▶). Further analysis with phenix.xtriage based on sequence composition gave a solvent content of 50.2% and a Matthews coefficient of 2.47 ų Da⁻¹ (Matthews, 1968 ▶), with six chains in the asymmetric unit. Attempts to solve the structure using the structures of SUMO protein (PDB entries 3qht, 3pge and 3tix; Gilbreth et al., 2011 ▶; Freudenthal et al., 2011 ▶; Schalch et al., 2011 ▶), snake CTL (PDB entry 3ubu; Gao et al., 2012 ▶) and a combination of both as molecular-replacement models were unsuccessful. Therefore, we are presently working on the crystallization of selenomethionine-labelled SUMO-CTL protein.

Figure 3.

Crystals of SUMO-CTL fusion protein grown by the sitting-drop vapour-diffusion method with a protein:reservoir ratio of 1:1 in 0.1 mM HEPES pH 7.5, 70% MPD. The crystals grew to a length of 150 µm.

Table 1. Diffraction data statistics.

Values in parentheses are for the outer shell.

Diffraction source	X06DA, SLS
Wavelength (Å)	1.000
Rotation range per image (°)	0.1
Total rotation range (°)	180
Exposure time per image (s)	0.1
Space group	P212121
Unit-cell parameters (Å)	a = 100.6, b = 115.4, c = 130.6
Mosaicity (°)	0.228
Resolution (Å)	2.85 (3.02–2.85)
Total No. of reflections	242290 (35688)
No. of unique reflections	36606 (5688)
Completeness (%)	99.5 (97.5)
Multiplicity	6.6 (6.2)
Mean I/σ(I)	6.38 (1.07)
CC1/2 † (%)	98.4 (42.7)
R merge (%)	29.0 (166.1)

^†Karplus & Diederichs (2012 ▶).