crystallization communications Acta Crystallographica Section F

Structural Biology and Crystallization Communications

Crystallization and preliminary X-ray diffraction analysis of D53H mutant Escherichia coli cAMP receptor protein

ISSN 1744-3091

Jing Huang, Tong Wu, Zheng Guo, Tiantian Lou, Shaoning Yu, Weimin Gong and Chaoneng Ji* State Key Laboratory of Genetic Engineering, Institute of Genetics, School of Life Sciences, Fudan University, Shanghai 200433, People’s Republic of China

Correspondence e-mail: [email protected]

Received 29 September 2013 Accepted 18 November 2013

The Escherichia coli cyclic AMP receptor protein (CRP) is a prokaryotic global transcription activator protein that controls the expression of many different genes. Wild-type CRP can bind to special DNA sequences in the presence of cAMP. The substitution of Asp53 by His results in the CRP* phenotype, which does not require exogenous cAMP. In the present study, the D53H CRP mutant was overexpressed, purified and crystallized. cAMP-free D53H CRP crystals ˚ . Based on the systematic were obtained and diffracted to a resolution of 2.9 A absences of the crystals, the space group is likely to be P212121, with unit-cell ˚ . The asymmetric unit was parameters a = 76.66, b = 152.14, c = 176.11 A confirmed to contain four protein dimers, with a Matthews coefficient of ˚ 3 Da1 and a solvent content of 54.68%. 2.71 A 1. Introduction The Escherichia coli cyclic AMP receptor protein (CRP), also known as the catabolite gene-activator protein (CAP), is a classical protein model for structural and mechanistic studies of transcriptional activation. CRP participates in the regulation of the expression of nearly 200 genes responsive to cyclic AMP (cAMP; Busby & Ebright, 1999). CRP is a dimer composed of two identical subunits, and each monomer consists of 209 amino-acid residues (Aiba et al., 1982; Cossart & Gicquel-Sanzey, 1982). CRP is a basic protein with an isoelectric point of 9.12 (Harman, 2001). The CRP subunit is composed of two domains. The larger N-terminal domain, which includes amino-acid residues 1–133 (Fic et al., 2009), consists of eight -strands that provide a hydrophobic pocket for cAMP binding (Tutar, 2008). The smaller C-terminal domain is composed of amino acids 139–209 and is responsible for DNA binding through a characteristic helix–turn–helix (HTH) motif (Brennan & Matthews, 1989). The N-terminal and C-terminal domains are connected by a hinge region (Weber & Steitz, 1987). Our current understanding of the allosteric process of E. coli CRP activation can be summarized by a rigid-body movement that involves subunit realignment and domain rearrangement (Won et al., 2009; Harman, 2001). The main consequence of this overall transition is the protrusion and adjustment of F-helices that recognize specific DNA sites (Won et al., 2000). During past decades, biochemical and biophysical studies have provided considerable details of this process. In particular, our comprehensive understanding of the protein structure through X-ray crystallography has made a critical contribution to the current insights into CRP allostery. To date, more than 20 three-dimensional structural coordinates of CRP from E. coli have been deposited in the Protein Data Bank (PDB), and these include the structure of unligated CRP (Sharma et al., 2009), structures of the CRP–cAMP complex (McKay et al., 1982; Weber & Steitz, 1987; Passner et al., 2000), structures of the CRP–cAMP–DNA complex (Schultz et al., 1991; Parkinson et al., 1996; Passner & Steitz, 1997), the structure of the CRP–cAMP–DNA–RNAP complex (Benoff et al., 2002) and some mutant structures (CRP*) that are constitutively active even in the absence of cAMP (Weber et al., 1987). The isolation of CRP* phenotype mutants has considerably contributed to the deduction of the mechanism underlying the observed conformational change (Passner et al., 2000). CRP* designates a cAMP-independent mutant of CRP that can activate transcription even in the absence of

1436

doi:10.1107/S174430911303145X

Acta Cryst. (2013). F69, 1436–1439

crystallization communications cAMP. CRP* mutation positions include 53, 62, 127/128 (double mutation), 138, 140, 141, 142, 144, 148 and 195 (Harman, 2001; Chu et al., 2001; Youn et al., 2006). To date, four types of constitutively active CRP* structures, namely D53H (PDB entry 1i6x; M. A. White, D. Liu, S.-H. Lin, R. O. Fox & J. C. Lee, unpublished work), S62F (PDB entry 3qop; J. Knapp, M. A. White & J. C. Lee, unpublished work), A144T (Weber et al., 1987) and T127L/S128A mutant CRP (Chu et al., 2001), have been resolved by X-ray crystallography and their conformations are highly comparable with that of CRP–cAMP. However, these mutant structures were crystallized in their cAMPbound states, and the cAMP-free forms of CRP* mutants have not been solved to date. The structures of CRP family members from four species have previously been solved. These structures can be categorized into three groups. The first group, designated the ‘off’ state, represents the ligand-free inactive conformation, as illustrated by apo CRP, CooA from Rhodospirillum rubrum (RrCooA) and CprK from Desulfitobacterium dehalogenans (DdCprK). The second group depicts a fully active ‘on’ state and includes cAMP-bound CRP and the constitutively active mutants T127L/S128A CRP (CRP*LA), G145S PrfA from Listeria monocytogenes (PrfA*145) and N127L/S128L CooA from Carboxydothermus hydrogenoformans (chCooA*LL). The other structures are considered to be in an ‘intermediate’ state, as demonstrated by ligand-bound inactive chCooA (Im–chCooA), ligand-bound inactive CprK from D. hafniense (CHPA–DhCprKox) and ligand-free active PrfA from L. monocytogenes (Won et al., 2009). Recently, the structures of several homologues from Mycobacterium tuberculosis have been solved. These include cAMP-free MtCRP (Kumar et al., 2010; Gallagher et al., 2009) and MtCRP– cAMP (Reddy et al., 2009), which belonged to the ‘off’ state and the ‘on’ state, respectively. We hypothesize that cAMP-free CRP* may represent an intermediate state of CRP and may have the same activity as CRP–cAMP but a different conformation from that of CRP–cAMP. Thus, we choose to analyze cAMP-free CRP* in this study. Although Asp53 is located neither in the cAMP-binding pocket nor in the DNA-binding domain, D53H CRP exhibits CRP*

phenotypic behaviour (Kim et al., 1992), which makes D53H CRP a very attractive target for studying the activation mechanism of CRP. An in vitro DNA-binding study indicated that both wild-type and D53H CRP contain two sets of two DNA-binding sites (Lin & Lee, 2002a). This result confirmed the positive cooperativity between the two high-affinity binding sites, although the D53H mutation evidently enhances the interaction between the sites. In DNA recognition, both wild-type and D53H CRP bind to specific DNA sequences with high affinity when the high-affinity sites are occupied; in contrast, occupancy of the weak cAMP sites leads to substantially weaker affinity. Furthermore, the order of affinity for various DNA sequences remains the same. Interestingly, the magnitude of

G, which is the difference in the energy of binding to different DNA sequences, is quite similar. In fact, Lin and Lee showed that the D53H mutation merely shifts all of the DNA-binding constants to higher values without altering the mechanism of recognition or the basic mechanism of CRP activation (Lin & Lee, 2002b). How does mutation at this site affect the properties and function of the protein? One possibility is a perturbation in the distribution of the ensemble of CRP structures (Lin & Lee, 2002b). The three-dimensional structure of cAMP-free D53H CRP may provide some insight into the mechanism. In this study, cAMP-free D53H CRP mutant was overexpressed, purified and crystallized. X-ray diffraction data for cAMP-free D53H CRP were subsequently obtained.

2. Materials and methods 2.1. Site-directed mutagenesis

Overlap extension PCR was used to construct the D53H mutant with only one residue difference from the wild-type gene; the mutant gene was then cloned into a pET-30a (Novagen, USA) plasmid. The universal primers of pET-30a were chosen for both primer ends, and the other pair of primers was designed with Gene Runner: forward, 50 -CCTCTTCGTGTTTGATCAGCACAGCCACA-30 ; reverse, 50 ATCAAACACGAAGAGGGTAAAGAAATGA-30 [the mutant site is displayed in bold: GAC (Asp) to CAC (His)]. After confirmation of the mutation through sequencing, pET-30a-crpD53H was transformed into E. coli strain BL21 (DE3).

2.2. Protein expression and purification

Figure 1 The Coomassie Blue-stained 15% reducing SDS–PAGE gel shows the purity of the CRP used for crystallization. Lane 1, molecular-weight marker (labelled in kDa); lanes 2–6, CRP sample purified using a phenyl Sepharose column. Lanes 2–6 were chosen to obtain a concentration higher than 20 mg ml1.

Acta Cryst. (2013). F69, 1436–1439

The D53H CRP mutant was expressed and purified using the same protocol as used for the wild type in a previous study (Huang et al., 2013). The cells were resuspended in 250 ml buffer A (50 mM Tris– HCl, 100 mM KCl, 1 mM EDTA, 1 mM DTT, 0.2 mM PMSF, 5% glycerol pH 7.8) and lysed by pressure (JN-3000 Plus, JNBIO, People’s Republic of China) at 277 K. The cell lysate was centrifuged at 15 000 rev min1 and 277 K for 30 min. The proteins were isolated and purified through sequential chromatography on pre-equilibrated Bio-Rex 70 (Bio-Rad), hydroxyapatite (Fluka) and phenyl Sepharose (GE Healthcare) columns. We collected the effluent liquid based on the results of SDS–PAGE (Fig. 1) and a Bradford assay and then dialyzed the effluent overnight in TEK100 buffer (50 mM Tris–HCl, 100 mM KCl, 1 mM EDTA, 1 mM DTT pH 7.8). The D53H CRP solution was then injected into a centrifuge filter (Millipore) and centrifuged at 5000g and 277 K for 5 min. The centrifugation step was repeated until the protein concentration was greater than 20 mg ml1, and the concentrated protein was maintained at 203 K until use in crystallization. Huang et al.



cAMP receptor protein

1437

crystallization communications 2.3. Crystallization and X-ray data collection

The preliminary crystallization conditions were screened using Crystal Screen Lite and Crystal Screen 2 (Hampton Research, USA) by hanging-drop vapour diffusion at 277 K. The crystals were obtained by mixing 1.0 ml 20 mg ml1 protein solution with 1.0 ml reservoir solution and equilibrating the drop over 500 ml reservoir solution consisting of 0.1 M sodium citrate tribasic dihydrate pH 6.0, 0.5 M ammonium sulfate, 1.0 M lithium sulfate monohydrate. Prior to data collection, a single crystal was soaked in cryoprotectant solution (precipitant solution augmented with 30% glycerol) and flash-cooled directly in a liquid-nitrogen stream at 100 K. The X-ray diffraction data were collected using an X-ray wavelength of ˚ on beamline BL17U-MX at the Shanghai Synchrotron 0.9793 A Radiation Facility with a MAR DTB system. The complete data set included 90 images of 1 oscillation. The data from the 90 images were processed using the HKL-2000 package (Otwinowski & Minor, 1997).

3. Results and discussion In this study, D53H CRP protein was successfully overexpressed in E. coli. The purified mutant protein was subjected to extensive crystallization screening. Wedge-shaped crystals of D53H CRP could only be obtained using a solution consisting of 0.1 M sodium citrate tribasic dihydrate pH 6.0, 0.5 M ammonium sulfate, 1.0 M lithium sulfate monohydrate. The dimensions of the crystals were approximately 0.8  0.3  0.1 mm (Fig. 2). The preliminary crystallographic data statistics for the D53H CRP crystal are given in Table 1 and are based on global indicators of the quality of the diffraction data (Weiss, 2001). The likely space group of the crystal is P212121, with ˚ . The Matthews unit-cell parameters a = 76.66, b = 152.14, c = 176.11 A ˚ 3 Da1 and corresponds to a solvent coefficient, which was 2.71 A content of 54.68%, indicated that there are eight CRP molecules in the asymmetric unit. In subsequent studies, we will attempt to determine the structure of the D53H CRP mutant using the molecular-replacement method. The crystal structure may provide more precise information on the mechanism underlying the cAMP-induced allostery that is required for DNA binding and transcription. We predict that there may be some distinct structural differences between apo CRP, cAMP–CRP and D53H CRP. The three-dimensional structure of D53H CRP may

Figure 2 Crystal of D53H CRP. The crystal was grown in a reservoir solution composed of 0.1 M sodium citrate tribasic dihydrate pH 6.0, 0.5 M ammonium sulfate, 1.0 M lithium sulfate monohydrate. The dimensions of the wedge-shaped crystal are approximately 0.8  0.3  0.1 mm.

1438

Huang et al.



cAMP receptor protein

Table 1 Statistics of the collected data. Values in parentheses are for the outermost resolution shell. Source ˚) Wavelength (A ˚) Resolution range (A Likely space group ˚) Unit-cell parameters (A a b c Observed reflections Total Unique Completeness (%) Molecules per asymmetric unit ˚ 3 Da1) VM (A Rmerge† (%) Average I/(I) Rr.i.m.‡ (%) Rp.i.m.§ (%) Mosaicity ( ) Solvent content (%)

Beamline BL17U, SSRF 0.9793 50–2.90 (2.95–2.90) P212121 76.66 152.14 176.11 158361 45116 99.0 (92.5) 8 2.71 6.70 (36.8) 20.76 (3.81) 6.9 (43.4) 3.6 (23.1) 0.410 54.68

P P P P † Rmerge = hkl P i jIi ðhklÞ  hIðhklÞij= hkl i Ii ðhklÞ, P where Ii(hkl) is the intensity of reflection hkl, hkl is the sum over all reflections, i is the sum over i measurements of reflection hkl and hI(hkl)i is the weighted average intensity ofPall observations i of P 1=2 reflection hkl. ‡ Rr.i.m. = hkl fNðhklÞ=½NðhklÞ  1g P i jIi ðhklÞ  hIðhklÞij= P P 1=2 I ðhklÞ, where N(hkl) is the redundancy. § R = p.i.m. hkl i i hkl f1=½NðhklÞ  1g P P P  i jIi ðhklÞ  hIðhklÞij= hkl i Ii ðhklÞ.

provide insight into the process of the CRP conformational change and enhance our understanding of the mechanism underlying the allostery of CRP activation induced by cAMP binding. We thank the staff of beamline BL17U at SSRF for their help during data collection. This work was supported by the National Basic Research Program of China (2009CB825505), the National Natural Science Foundation of China (30770427), the New Century Excellent Talents in University (NCET-06-0356), the Shanghai Leading Academic Discipline Project (B111) and the National Talent Training Fund in Basic Research of China (J0630643).

References Aiba, H., Fujimoto, S. & Ozaki, N. (1982). Nucleic Acids Res. 10, 1345–1361. Benoff, B., Yang, H., Lawson, C. L., Parkinson, G., Liu, J., Blatter, E., Ebright, Y. W., Berman, H. M. & Ebright, R. H. (2002). Science, 297, 1562–1566. Brennan, R. G. & Matthews, B. W. (1989). J. Biol. Chem. 264, 1903–1906. Busby, S. & Ebright, R. H. (1999). J. Mol. Biol. 293, 199–213. Chu, S. Y., Tordova, M., Gilliland, G. L., Gorshkova, I., Shi, Y., Wang, S. & Schwarz, F. P. (2001). J. Biol. Chem. 276, 11230–11236. Cossart, P. & Gicquel-Sanzey, B. (1982). Nucleic Acids Res. 10, 1363–1378. Fic, E., Bonarek, P., Gorecki, A., Kedracka-Krok, S., Mikolajczak, J., Polit, A., Tworzydlo, M., Dziedzicka-Wasylewska, M. & Wasylewski, Z. (2009). J. Mol. Microbiol. Biotechnol. 17, 1–11. Gallagher, D. T., Smith, N., Kim, S.-K., Robinson, H. & Reddy, P. T. (2009). J. Biol. Chem. 284, 8228–8232. Harman, J. G. (2001). Biochim. Biophys. Acta, 1547, 1–17. Huang, J., Liu, J., Tao, W., Yang, Z., Qiu, R., Yu, S. & Ji, C. (2013). Acta Cryst. F69, 562–565. Kim, J., Adhya, S. & Garges, S. (1992). Proc. Natl Acad. Sci. USA, 89, 9700– 9704. Kumar, P., Joshi, D. C., Akif, M., Akhter, Y., Hasnain, S. E. & Mande, S. C. (2010). Biophys. J. 98, 305–314. Lin, S.-H. & Lee, J. C. (2002a). Biochemistry, 41, 11857–11867. Lin, S.-H. & Lee, J. C. (2002b). Biochemistry, 41, 14935–14943. McKay, D. B., Weber, I. T. & Steitz, T. A. (1982). J. Biol. Chem. 257, 9518–9524. Otwinowski, Z. & Minor, W. (1997). Methods Enzymol. 276, 307–326. Parkinson, G., Wilson, C., Gunasekera, A., Ebright, Y. W., Ebright, R. H., Ebright, R. E. & Berman, H. M. (1996). J. Mol. Biol. 260, 395–408. Passner, J. M., Schultz, S. C. & Steitz, T. A. (2000). J. Mol. Biol. 304, 847–859. Passner, J. M. & Steitz, T. A. (1997). Proc. Natl Acad. Sci. USA, 94, 2843–2847. Acta Cryst. (2013). F69, 1436–1439

crystallization communications Reddy, M. C. M., Palaninathan, S. K., Bruning, J. B., Thurman, C., Smith, D. & Sacchettini, J. C. (2009). J. Biol. Chem. 284, 36581–36591. Schultz, S. C., Shields, G. C. & Steitz, T. A. (1991). Science, 253, 1001–1007. Sharma, H., Yu, S., Kong, J., Wang, J. & Steitz, T. A. (2009). Proc. Natl Acad. Sci. USA, 106, 16604–16609. Tutar, Y. (2008). Protein J. 27, 21–29. Weber, I. T., Gilliland, G. L., Harman, J. G. & Peterkofsky, A. (1987). J. Biol. Chem. 262, 5630–5636.

Acta Cryst. (2013). F69, 1436–1439

Weber, I. T. & Steitz, T. A. (1987). J. Mol. Biol. 198, 311–326. Weiss, M. S. (2001). J. Appl. Cryst. 34, 130–135. Won, H.-S., Lee, Y.-S., Lee, S.-H. & Lee, B.-J. (2009). Biochim. Biophys. Acta, 1794, 1299–1308. Won, H.-S., Yamazaki, T., Lee, T.-W., Yoon, M.-K., Park, S.-H., Kyogoku, Y. & Lee, B.-J. (2000). Biochemistry, 39, 13953–13962. Youn, H., Kerby, R. L., Conrad, M. & Roberts, G. P. (2006). J. Biol. Chem. 281, 1119–1127.

Huang et al.



cAMP receptor protein

1439

Copyright of Acta Crystallographica: Section F (International Union of Crystallography IUCr) is the property of International Union of Crystallography - IUCr and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.