Crystallization and X-ray diffraction analysis of the CH domain of the cotton kinesin GhKCH2

The cloning, expression, purification and crystallization of the CH domain of the plant-specific kinesin GhKCH2 is reported.


Introduction
Microfilaments and microtubules, two vital cytoskeleton systems in cells, together take part in a variety of cellular activities, such as cell division and proliferation, transportation of organelles and vesicles, and the organization and formation of plant preprophase bands, phragmoplasts and cell plates (Wasteneys & Galway, 2003;Petrá šek & Schwarzerová , 2009). The kinesins are a superfamily of microtubule-based motor proteins (Howard, 1996), some of which (KCHs) contain a single N-terminal calponin homology (CH) domain that is able to bind to both microfilaments and microtubules. In 2004, the first KCH (GhKCH1) was identified and demonstrated to be involved in the elongation of cotton fibres (Preuss et al., 2004). Subsequently, our laboratory and others identified further KCHs (GhKCH2, O12/OsKCH1, AtKinG and NtKCH1), all of which were able to bind to both microfilaments and microtubules in vitro or in vivo (Frey et al., 2009;Xu et al., 2009;Buschmann et al., 2011;Umezu et al., 2011;Klotz & Nick, 2012). Recently, Ram Dixit deduced that KCH might be involved in the transportation of actin fragments (Dixit, 2012).
The actin-binding CH domain, named after its first identification in calponin, consists of about 100 amino-acid residues (Takahashi & Nadal-Ginard, 1991 one structure available (fimbrin from Arabidopsis thaliana; PDB entry 1pxy; Klein et al., 2004). GhKCH2 (GenBank accession No. EF432568), a KCH previously cloned from cotton (Gossypium hirsutum) fibres in our laboratory, has a motor domain with microtubulestimulated ATPase activity and a CH domain that strongly interacts with microfilaments, suggesting it to be a candidate for a linker between microfilaments and microtubules (Xu et al., 2007(Xu et al., , 2009. The CH domain of GhKCH2 shares 31% amino-acid sequence identity with human calponin 1 and 28% amino-acid sequence identity with Arabidopsis fimbrin. Our previous studies revealed that GhKCH2-N (amino acids 1-306), containing the CH domain, had a higher affinity for F-actin (K d = 0.42 AE 0.02 mM) than most other CH-domaincontaining proteins (K d = $5-50 mM) (Gimona et al., 2002;Xu et al., 2009). To elucidate the mechanisms of this unique biochemical feature, further exploration of the structure of GhKCH2-CH was performed. Here, the expression, purification, crystallization and preliminary X-ray diffraction studies of the CH domain of GhKCH2 are described.

Protein expression and purification
The CH domain of GhKCH2 was cloned using the forward primer 5 0 -GAG AGT CCA TAT GGA TTT GGA ATC TAG AAA AGC TG-3 0 (NdeI site in bold) and the reverse primer 5 0 -CAA CTC GAG TTA CGA GAG CCT CCA CTC GTT ATA G-3 0 (XhoI site in bold). The PCR product was digested and inserted into a modified pGEX-4T-2 vector (kindly provided by Professor Zhongzhou Chen, China Agricultural University) at the NdeI and XhoI restriction sites with a TEV protease cleavage site between GST and the target gene (Table 1). The correct certified constructs were transformed into Escherichia coli strain BL21(DE3). Cells were grown at 37 C in LB medium containing 100 mg ml À1 ampicillin to an A 600 of 0.6-0.8 and were induced with 0.1 mM isopropyl -d-1thiogalactopyranoside (IPTG) at 22 C overnight. The cells were harvested by centrifugation and lysed by gentle sonication in lysis buffer (0.1 M Tris-HCl pH 7.5, 150 mM NaCl, 1 mM DTT, 1 mM PMSF). After high-speed centrifugation, the supernatant was loaded onto a home-made Glutathione Sepharose 4B column (GE Healthcare) and incubated for 1 h. After washing with wash buffer ( Table 1 Macromolecule-production information.  150 mM NaCl), TEV protease was added to cleave the fusion protein overnight. The collected flowthrough was purified by Mono Q chromatography and further polished by gel filtration on a HiLoad 16/60 Superdex 75 pg column (GE Healthcare). The purified proteins were pooled, concentrated to 10-15 mg ml À1 , flash-cooled in liquid nitrogen and stored at 80 C. Protein purity and identity were assessed by SDS-PAGE with Coomassie Bright Blue staining (Fig. 1). All of the purification procedures described above were conducted at 4 C.

Protein crystallization
Initial screening for crystallization conditions was carried out in 48-well sitting-drop plates using the commercially available kits Crystal Screen, Crystal Screen 2 and Index (Hampton Research, USA) at 4 C. Crystals of GhKCH2-CH were initially grown from a mixture of 1 ml protein solution (10-15 mg ml À1 in 20 mM Tris-HCl pH 7.5, 150 mM NaCl) and 1 ml precipitatant solution equilibrated against 100 ml reservoir solution at 277 K. Subsequent optimizations were performed using 24-well sitting-drop plates, and the size of the crystals was enlarged by streak-seeding using a cat whisker (Figs. 2a and 2b). Detailed information on GhKCH2-CH crystallization is given in Table 2.

Data collection
Mounted crystals were dehydrated with a solution consisting of 0.1 M Tris-HCl pH 7.0, 20%(w/v) PEG 8000, 10%(v/v) DMSO for 5 min, transferred to a solution consisting of 0.1 M Tris-HCl pH 7.0, 20%(w/v) PEG 8000, 20%(v/v) DMSO for a further 5 min and finally flash-cooled in liquid nitrogen. The diffraction data set was collected at 100 K on BL17U1 at Shanghai Synchrotron Radiation Facility (SSRF) using an ADSC Q315 CCD. A total of 360 images with an oscillation angle of 1 each were collected using a 250 mm crystal-to-detector distance and an exposure time of 1 s per frame (Fig. 3). Detailed information on data collection is given in Table 3.

Results and discussion
Indexing with XDS (Kabsch, 2010) indicated that GhKCH2-CH crystallized in space group P2 1 . Analysis of the Patterson function with phenix.xtriage revealed a significant off-origin peak that was 78.6% of the height of the origin peak (Adams et al., 2010). Analysis of average intensities with TRUNCATE showed that the crystals had strong reflections with indices of l = 2n [even reflections, I/(I) = 23.6] and weak reflections with indices of l = 2n + 1 [odd reflections, I/(I) = 13.3] (French & Wilson, 1978). HHpred (http://toolkit.tuebingen.mpg.de/ hhpred; Hildebrand et al., 2009) was used to identify homologues of the GhKCH2 CH domain, and the ten PDB hits (PDB entries 1p2x,1ujo,1h67,1p5s,1wym,3i6x,1wyr,213g,1wyp and 1wyn) with the highest scores were used as search models. When determining the crystal structure of  Figure 3 X-ray diffraction pattern from a crystal of GhKCH2-CH. A resolution circle at 2.5 Å is shown. GhKCH2-CH using Phaser (McCoy et al., 2007), the structure of human calponin 2 (PDB entry 1wyn; RIKEN Structural Genomics/Proteomics Initiative, unpublished work) gave the best results after further refinement among these models. Four molecules were found in the asymmetric unit with a Matthews coefficient of 2.22 Å 3 Da À1 , corresponding to a solvent content of 44.8%. The entire reflection data set was split into odd (l = 2n + 1) and even components (l = 2n) using MTZUTILS from the CCP4 package . The positions of the four molecules in the asymmetric unit were optimized using rigidbody refinement with odd reflections, and restrained refinement was subsequently performed using REFMAC or phenix.refine with even reflections (Murshudov et al., 2011;Oksanen et al., 2006;Adams et al., 2010). After several cycles of such refinement, the model was refined against all data to an R free of about 35%. Owing to the low amino-acid sequence identity of the GhKCH2 CH domain to other CH domaincontaining proteins, crystallization of selenomethioninelabelled protein is in progress.