Purification, crystallization, crystallographic analysis and phasing of the CRISPR-associated protein Csm2 from Thermotoga maritima

Csm2 is a major component of type IIIA CRISPR ribonucleoprotein complexes that confer prokaryotes with immunity against phages and plasmids via an RNA-guided interference mechanism. The Csm2 protein from T. maritima was recombinantly expressed, purified and crystallized, and its structure was solved via cadmium single-wavelength anomalous diffraction phasing.

Abstract

The clusters of regularly interspaced short palindromic repeats (CRISPR)–CRISPR-associated proteins (Cas) system consists of an intriguing machinery of proteins that confer bacteria and archaea with immunity against phages and plasmids via an RNA-guided interference mechanism. Here, the cloning, recombinant expression in Escherichia coli BL21 (DE3), purification, crystallization and preliminary X-ray diffraction analysis of Csm2 from Thermotoga maritima are reported. Csm2 is thought to be a component of an important protein complex of the type IIIA CRISPR–Cas system, which is involved in the CRISPR–Cas RNA-guided interference pathway. The structure of Csm2 was solved via cadmium single-wavelength anomalous diffraction (Cd-SAD) phasing. Owing to its involvement in the CRISPR–Cas system, the crystal structure of this protein could be of importance in elucidating the mechanism of type IIIA CRISPR–Cas systems in bacteria and archaea.

1. Introduction  

Recent reports have characterized the bacterial clusters of regularly interspaced short palindromic repeats (CRISPR)–CRISPR-associated proteins (Cas) system as a prokaryotic adaptive immune system based on RNA-guided interference (Mojica et al., 2005 ▸; Bolotin et al., 2005 ▸; Pourcel et al., 2005 ▸; Makarova et al., 2006 ▸; Barrangou et al., 2007 ▸). The CRISPR–Cas system is effective in protecting cells against nucleotide uptake from bacteriophages and plasmids (Barrangou et al., 2007 ▸; Marraffini & Sontheimer, 2008 ▸). In summary, foreign DNA is integrated into the CRISPR elements of the prokaryotic host genome (Pourcel et al., 2005 ▸; Bolotin et al., 2005 ▸; Barrangou et al., 2007 ▸) and transcripts of this DNA in the form of processed crRNA (CRISPR RNA) are involved in an RNA-guided interference mechanism to degrade target DNA or RNA (Brouns et al., 2008 ▸; Wiedenheft et al., 2009 ▸; Hale et al., 2009 ▸; Marraffini & Sontheimer, 2010 ▸).

Depending on the occurrence of three signature proteins, Cas3, Cas9 and Cas10, CRISPR–Cas systems are classified as type I, type II and type III systems, respectively (Makarova, Aravind et al., 2011 ▸; Makarova, Haft et al., 2011 ▸). Type III CRISPR–Cas systems are further subdivided into two different subtypes called IIIA and IIIB. Whereas type I and type II CRISPR–Cas systems target DNA (Garneau et al., 2010 ▸; Sinkunas et al., 2011 ▸; Gasiunas et al., 2012 ▸), type III systems are thought to target both DNA and RNA (Marraffini & Sontheimer, 2008 ▸; Hale et al., 2009 ▸; Staals et al., 2013 ▸, 2014 ▸; Taylor et al., 2015 ▸; Samai et al., 2015 ▸). The type IIIA system includes Csm Cas proteins (Makarova, Haft et al., 2011 ▸), which were identified as being expressed in Mycobacterium tuberculosis (Haft et al., 2005 ▸). The type IIIB system includes Cmr proteins (Hale et al., 2009 ▸). Processing and target recognition by crRNA is carried out by several Cas proteins, which usually form a complex of ribonucleoproteins (RNPs) like the so-called Cascade complex in type I systems (Brouns et al., 2008 ▸) and the Cmr CRISPR RNP (crRNP) complex in type IIIB systems (Hale et al., 2012 ▸; Staals et al., 2013 ▸). Electron-microscopy data and protein crystallographic studies have revealed the composition and function of the Cascade (Jore et al., 2011 ▸; Wiedenheft et al., 2011 ▸; Jackson et al., 2014 ▸) and the Cmr crRNP complexes (Staals et al., 2013 ▸; Osawa et al., 2015 ▸) in great detail. Furthermore, electron-microscopy and mass-spectrometric data have defined the Csm crRNP complex from Sulfolobus solfataricus, which is formed by up to seven different proteins (Rouillon et al., 2013 ▸). Similar experiments have defined a homologous complex in Thermus thermophilus that consists of five different proteins: Csm1–Csm5 (Staals et al., 2014 ▸). Cmr, Csm and Cascade complexes show overall structural similarity, possibly owing to the fact that they are composed of structurally related subunits (Rouillon et al., 2013 ▸). To obtain further insight into the Csm crRNP complex, we have cloned, expressed, purified, crystallized and solved the structure of the Csm2 protein from Thermotoga maritima MSB8.

2. Materials and methods  

2.1. Cloning and expression  

Full-length Csm2 (GenBank entry AKE29563.1) was amplified via PCR with primer overhangs from total genomic extracts of T. maritima MSB8 (Huber et al., 1986 ▸) obtained from DSMZ (German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany). DNA oligonucleotides (Exxtend, Paulínia, Brazil) are described in Table 1 ▸. PCR fragments were cloned into the pQtev His-tag expression vector (Protein Structure Factory, Berlin, Germany) using BamHI and HindIII cloning sites. The construct was verified by sequencing (Proteobras, Paulínia, Brazil). Csm2 was expressed in LB medium for 6 h at 37°C in Escherichia coli strain BL21 (DE3) after induction with 1 mM β-d-1-thio­galactopyranoside (IPTG).

Table 1. Macromolecule-production information.

Source organism	T. maritima
DNA source	T. maritima
Forward primer (BamHI restriction site)	5-GGCCGGGGATCCGCAGTTTCTCAGGGTGTTTC
Reverse primer (HindIII restriction site)	5-GGCCGGAAGCTTATCTTCTGGCTGTCCTTCTGTTG
Cloning vector	pQtev
Expression vector	pQtev
Expression host	E. coli BL21 (DE3)
Complete amino-acid sequence of the construct produced	GSAVSQGVSLKEDLKDLVRKAEEIGRELSGKLKTNQLRKFHGHLTKIWSNYIYKKKDYRDNPEKFNEEILNELHFMKIFLAYQVGRDIEGISELKEILEPLIDEIKTPDEFEKFKKFYDAILAYHKFHSESEKSNRRTARR

2.2. Protein purification  

The cells were harvested by centrifugation and resuspended in lysis buffer consisting of 50 mM Tris–HCl pH 7.5, 400 mM NaCl, 10 mM imidazole, 1 mM PMSF, 5 mM β-mercapto­ethanol. The cells were lysed using an M-110L high-pressure homogenizer (Microfluidics, Westwood, USA). After separation of the soluble fraction by centrifugation, the protein was purified by affinity chromatography using a 5 ml HisTrap Sepharose column (GE Healthcare) on an ÄKTAprime Plus liquid-chromatography system (GE Healthcare). After elution with imidazole and dialysis in 50 mM Tris–HCl pH 7.5, 400 mM NaCl, 3 mM DTT, 10% glycerol, the His tag was cleaved using TEV (Tobacco etch virus) protease (Blommel & Fox, 2007 ▸). After a subsequent affinity-chromatography step to remove the protease and the His tag, Csm2 was concentrated using Amicon Ultra-15 centrifugal filters (Millipore) and submitted to gel-filtration chromatography on a HiLoad 26/600 Superdex 75 prep-grade column with a nominal volume of 320 ml (GE Healthcare) in buffer consisting of 25 mM Tris–HCl pH 7.5, 100 mM NaCl, 3 mM DTT, 10% glycerol. Finally, the protein was concentrated to 15 mg ml⁻¹. The molecular mass of native Csm2 (650 ng) was confirmed by electrospray-ionization (ESI) mass spectrometry using a hybrid quadrupole (Q)-IM-ToF MS instrument (Synapt G2 HDMS mass spectrometer, Waters). Deconvolution of the MS protein spectrum yielded an isotope-averaged molecular mass of 16 732.78 Da, which is similar to the value obtained by SDS–PAGE using SeeBlue Plus2 Pre-stained protein standard as a marker (Invitrogen, Life Technologies).

2.3. Crystallization  

Csm2 was crystallized by the hanging-drop method in 24-well plates after screening trials using Crystal Screen 2 (Hampton Research, Aliso Viejo, USA) at 20°C. Initial screening conditions were optimized, leading to optimal crystallization conditions consisting of 100 mM cadmium chloride, 21% PEG 400, 100 mM sodium acetate buffer pH 4.6, as described in Table 2 ▸.

Table 2. Crystallization.

Method	Hanging drop
Plate type	24-well VDX plate
Temperature (K)	293.15
Protein concentration (mgml1)	15
Buffer composition of protein solution	25mM TrisHCl pH 7.5, 100mM NaCl, 3mM DTT, 10% glycerol
Composition of reservoir solution	100mM sodium acetate pH 4.6, 100mM CdCl2, 21% PEG 400
Volume and ratio of drop	4l, 1:1 ratio
Volume of reservoir (ml)	1

2.4. Data collection and processing  

Diffraction data were collected on the MX-1 beamline at the Brazilian Synchrotron Light Laboratory (LNLS, Campinas, Brazil; Polikarpov et al., 1998 ▸) after flash-cooling in liquid nitrogen using crystallization buffer supplemented with 12% glycerol. A highly redundant data set was collected at a wavelength of 1.458 Å. Crystallographic data were processed with XDS (Kabsch, 2010 ▸) and Adxv (Arvai, 2015 ▸). The structure of Csm2 was solved by single-wavelength anomalous diffraction (SAD) using Phaser (McCoy et al., 2007 ▸) within the AutoSol module of PHENIX (Adams et al., 2010 ▸), using the fact that the crystallization conditions contained cadmium ions. Electron-density maps were inspected using Coot (Emsley et al., 2010 ▸) and figures were rendered using PyMOL (DeLano, 2002 ▸).

3. Results and discussion  

In recent years, several outstanding crystal structures of RNA-processing enzymes and protein complexes have been obtained using isoforms from thermophilic bacteria (Wang et al., 2008 ▸; Mulepati & Bailey, 2011 ▸; Lintner et al., 2011 ▸; Cocozaki et al., 2012 ▸; Osawa et al., 2013 ▸). To obtain further data for the Csm CRISPR–Cas type IIIA crRNP complex, we have cloned, expressed, purified and crystalized a major protein from this complex, Csm2.

T. maritima MS8 Csm2 was expressed in soluble form in E. coli strain BL21 (DE3) in relatively high quantities of up to 30 mg per litre of culture. However, the protein could initially not be concentrated owing to precipitation. This problem was circumvented by using glycerol (10%) and high concentrations of NaCl (up to 400 mM) during all purification steps. To improve the chance of crystallization, after purifying the protein the concentration of NaCl in the final gel-filtration step was reduced to 100 mM (Fig. 1 ▸). The protein could then be concentrated to 15 mg ml⁻¹. Small monocrystals obtained in initial screens could be improved in size through refinement of the crystallization conditions, as shown in Fig. 2 ▸.

Figure 1.

Purification of Csm2. (a) Gel-filtration chromatography profile of Csm2 using a HiLoad 26/600 Superdex 75 prep-grade column (GE Healthcare). The last peak corresponds to the crystallized Csm2. (b) Molecular-mass determination of native Csm2 by mass spectrometry. Deconvolution of the MS protein spectrum yielded an isotope-averaged molecular mass of 16 732.78 Da. (c) SDS–PAGE of Csm2 after gel filtration using SeeBlue Plus2 Pre-stained protein standard (Invitrogen, Life Technologies) as a marker (labelled in kDa).

Figure 2.

Csm2 crystals and X-ray diffraction pattern. (a) Representative native crystals of Csm2; scale bar = 0.5 mm. (b) Diffraction image of a Csm2 crystal.

A highly redundant data set was collected from one of these crystals on the MX-1 beamline at the Brazilian Synchrotron Light Laboratory (LNLS), Campinas, Brazil. Crystallographic data for this crystal are given in Table 3 ▸. The crystals tested diffracted to a limited resolution of 2.9 Å in space group P3₁21 and contained an estimated three subunits in the asymmetric unit (Matthews coefficient of 2.7 ų Da⁻¹ and 54% solvent content). As there are no known close structural homologues of Csm2 in the PDB, the structure of Csm2 could not be solved by molecular replacement. However, as our crystals grew at high concentrations (100 mM) of CdCl₂, we hypothesized that Cd²⁺ ions possibly formed strong linkages with the numerous charged amino acids in Csm2. As several structures have been solved using Cd-SAD and/or S-SAD (Yogavel et al., 2010 ▸; Robbins et al., 1991 ▸; Medlock et al., 2009 ▸), Cd-SAD was used together with the PHENIX software package (Adams et al., 2010 ▸) to try to solve the structure. Indeed, the measured data set indicated the presence of an anomalous signal, quantified by an overall anomalous correlation of 31% and an overall mean anomalous difference in units of standard deviation (SigAno in XDS) of 1.085. Consequently, phenix.hyss was able to detect six cadmium ions. Automatic refinement and phasing with Phaser and subsequent phase density modifications then led to the first experimental electron-density maps. This solution was characterized by an overall figure of merit of 0.319, a correlation of local r.m.s. density of 0.82 and a meaningful map skew of 0.10.

Table 3. Data collection and processing.

Values in parentheses are for the outer shell.

Diffraction source	MX-1, LNLS
Wavelength ()	1.458
Temperature (K)	100
Detector	MAR CCD 165
Crystal-to-detector distance (mm)	120
Rotation range per image ()	1
Data range (images)	1190
Space group	P3121
a, b, c ()	77, 77, 160
, , ()	90, 90, 120
Mosaicity ()	0.2
Resolution range ()	502.9 (3.082.90)
Total No. of reflections	143907 (22874)
No. of unique reflections	23467 (3813)
Completeness (%)	99.9 (99.6)
Multiplicity	6.13 (5.99)
I/(I)	13.77 (2.62)
R meas (%)	11.0 (81.3)
Overall B factor from Wilson plot (2)	81.2

Most importantly, the experimental electron-density maps after density modification promptly showed meaningful electron density, and several α-helical structures could be detected in these maps (Fig. 3 ▸). However, the resolution of the data set has thus far hindered the assignment of amino-acid side chains. Improvement of the diffraction resolution will thus be necessary in order to build and refine an unambiguous model.

Figure 3.

Electron-density maps. Stereoview of the AutoSol density-modified electron-density map of the Csm2 crystal structure contoured at a 1.5σ (a) and 3σ (b) level showing a cadmium ion (yellow) linked to an α-helix with a C^α trace (green).

As Csm2 is a homologue of S. solfataricus SSo1424 (18% sequence identity) and T. thermophilus Csm2 (22% sequence identity), which play important roles in the assembly of Csm crRNP complexes (Rouillon et al., 2013 ▸; Staals et al., 2014 ▸), the results described here have an impact in helping to understand the structure and function of Csm crRNP complexes of type IIIA CRISPR–Cas systems.