(NZ)CH...O Contacts assist crystallization of a ParB-like nuclease
© Shaw et al; licensee BioMed Central Ltd. 2007
Received: 10 February 2007
Accepted: 07 July 2007
Published: 07 July 2007
The major bottleneck for determination of 3 D structures of proteins using X-rays is the production of diffraction quality crystals. Often proteins are subjected to chemical modification to improve the chances of crystallization
Here, we report the successful crystallization of a nuclease employing a reductive methylation protocol. The key to crystallization was the successful introduction of 44 new cohesive (NZ) CH...O contacts (3.2 – 3.7 Å) by the addition of 2 methyl groups to the side chain amine nitrogen (NZ) of 9 lysine residues of the nuclease. The new contacts dramatically altered the crystallization properties of the protein, resulting in crystals that diffracted to 1.2 Å resolution. Analytical ultracentrifugation analysis and thermodynamics results revealed a more compact protein structure with better solvent exclusion of buried Trp residues in the folded state of the methylated protein, assisting crystallization.
In this study, introduction of novel cohesive (NZ)CH...O contacts by reductive methylation resulted in the crystallization of a protein that had previously resisted crystallization in spite of extensive purification and crystallization space screening. Introduction of (NZ)CH...O contacts could provide a solution to crystallization problems for a broad range of protein targets.
The resolution and accuracy of the structural information provided by X-rays is unsurpassed when compared to other techniques employed to resolve the 3 D structure of proteins . However, not all proteins can be made available for X-ray diffraction studies because of the inherent difficulty in obtaining single crystals of adequate size and quality . A subset of such proteins that resist crystallization can be salvaged by employing a reductive methylation protocol [3–11]. A 26 kD nuclease from Pyrococcus furiosus used in the current study had resisted crystallization inspite of extensive purification and crystallization space screening. A number of techniques have been developed to improve a given target's potential for crystallization. Truncation of disordered regions [12, 13], mutagenesis of surface residues [14–16] and chemical modification of proteins  have been proven effective in this regard and a number of protein structures have been solved successfully. We decided to modify the surface lysines of the nuclease by reductive methylation in an attempt to crystallize the protein. The rationale behind targeting the amine nitrogen (NZ) of the nuclease arose from a number of considerations. Lysines harbouring free NZ atoms almost always reside on the surface of protein molecules . The thermodynamic cost for ordering the highly flexible solvent exposed side chains of lysines is exorbitant . Since crystallization is a surface phenomenon, a disordered lysine side chain has a profound negative impact on the formation of stable, uniform, inter molecular contacts essential for packaging of protein molecules in a crystal lattice. Interestingly, we found that methylation of surface lysines resulted in a decrease in the free energy of folding of the protein. This is consistent with the entropic cost of reduced flexibility in the native state due to formation of new intra molecular interactions, which in turn will lower the barrier to crystallization. We further probed the effect of methylation at the molecular level and show that cohesive (NZ)CH...O bonds assisted crystallization of the nuclease.
Generation of intra molecular contacts
Generation of inter molecular contacts
The methyl carbons of all the dimethylated lysine residues were involved in the formation of multiple new symmetry-generated intermolecular contacts. The synthesis of the intermolecular contacts was initiated by gradual evaporation of the solvent in presence of different chemicals. Commercially available sparse matrix screens under oil were used for screening the best chemical environment for formation of intermolecular bonds [19, 20]. The structure reveals 96 new symmetry-generated inter molecular contacts in the range 3.2 – 5.0 Å involving the (NZ)CH group and 28 of these inter molecular contacts are of the (NZ)CH...O type and in the range 3.2 – 4.0 Å (see Additional file 2). A significant number of these interactions were within the optimal range of 3.2 – 3.7 Å for CH...O hydrogen bonds . The intermolecular contacts involving the methyl groups of MLY112 and MLY201 are shown in Figures 2B and 2C respectively. The B factors for the NZ atoms of MLY112 and MLY201 were 25 and 15 respectively. The new intra and inter molecular contacts formed by the covalently linked methyl groups with the surrounding residues seem to have lowered the B factor values of the NZ atom indicating a localized side chain. This helps the packaging of the molecules in the crystal and the formation of a compact crystalline lattice.
In addition, a significant number of water molecules were observed to form (NZ)CH...O contacts (Figure 3C).
Thermodynamic stability of nuclease before and after modification
Comparison of thermodynamic parameters for ParB before and after modification§
3.01 ± 0.01
19 ± 1
57.4 ± 3.0
2.74 ± 0.01
14.4 ± 0.5
39.4 ± 1.4
Structure determination holds the key to unravelling the mechanisms by which proteins drive the machinery of all living organisms for survival and propagation. In spite of several dramatic technological advances in automation and information science, the key step for successful crystallographic structure determination – production of high quality crystals – continues to remain a resource intensive bottleneck . A first step towards addressing this bottleneck would be to identify the variables involved in crystallization. The nature of the protein is the single most important factor that influences crystallization. From a crystallization point of view – size, charge, hydrophobicity, hydrophilicity, flexibility, oxidation state and post-translational modifications like phosphorylation, glycosylation and myristylation – define the nature of a protein. Since the nature of the protein often varies with function, there is no universal crystallization strategy that would work for all proteins. To maximize the chances of crystallization it is crucial to identify and target an attribute of the protein that would have a profound effect on the crystallization. Surface lysines offer one such target. Lysines almost always reside on the surface of proteins. The solvent exposed side chains of lysines are highly mobile and prevent the formation of inter molecular contacts essential for the assembly of a crystalline lattice . Defects observed in protein crystals like low resolution and twinning is also a manifestation of the flexibility of the side chains found on the surface of the protein.
Locking the side chain amine nitrogen of lysines with the electron negative carboxyl oxygens of glutamic and aspartic acid side chains via cohesive ionic interactions will result in the immobilization of these side chains (Figure 2). Although the presence of CH...O bonds has been demonstrated in proteins, nucleic acids and carbohydrates [21–27], the (NZ)CH...O bonds introduced in the current study have never been described before. Methyl groups are very effective in mediating and bridging the physical distances between the free amine nitrogen and the side chain carboxyl oxygens for the formation of cohesive ionic interactions. Methyl groups can be covalently linked to the free amine nitrogen (NZ) with very high specificity using formaldehyde as the methyl group donor in the presence of dimethylamine borane complex . The amine nitrogen sitting adjacent to the methyl carbon, (NZ)CH, has an inductive effect on the methyl carbon resulting in a highly polarized carbon, which can act as a proton donor. The negative charge required for the electro neutrality is compensated by the lone pair of electrons of carboxyl oxygens found on the side chains of neighbouring glutamic and aspartic acid residues. The net effect is a cohesive ionic (NZ)CH...O interaction between the side chains leading to a compact, rigid protein molecule with localized side chains and loops. The positive charge on the methyl carbon can also be dispersed indirectly via the participation of water molecules (Figure 3C).
The methylated protein lost its ability to cleave DNA (results not shown). The dimethylation of K221 sitting at the edge of the active site may sterically hinder access of the active site to the incoming DNA. It is also possible that an overall conformational change in the protein induced by the chemical modification affects the catalytic site and compromises the function of the protein. However, such loss-of-function due to methylation has not been reported previously. Further studies are warranted in order to determine the exact cause for the inactivation of the methylated protein.
When a protein is set up for crystallization, the molecules are moving randomly in search of compatible bonding partners. Some of the inter molecular contacts generated during the course of random collisions are sustained. As more and more solvent evaporates, a number of these interactions become permanent. Eventually it leads to one of two possible outcomes. If the inter molecular bonding is heterogeneous, as in case of protein molecules that assume more than one conformation due to the presence of unstructured domains, flexible loops and side chains, or presence of partially unfolded regions, this will result in a disordered protein aggregate commonly referred to as precipitate. Poorly diffracting or defective crystals are also a consequence of the conformational flexibility of protein molecules. A homogenous inter molecular bonding pattern between protein molecules, as in the case of structurally rigid molecules, results in optimal packing of the molecules in a crystal lattice. A direct manifestation of the chemical modification of the protein is the reduction in number of degrees of freedom available to the protein for assuming different conformations. Introduction of (NZ)CH...O bonds curbs the movement of side chains and fixes their orientation in space. This produces uniform bonding partners and decreases the steric clashes between molecules during the packing of the lattice. The observed reduction in the free energy of folding for the modified protein is consistent with reduced flexibility leading to lower entropy in the native state; ordering of water molecules around the surface-exposed methyl groups may also contribute to this. Crystallization will be favoured for the modified protein by the reduction in entropy of the solvated structure and the involvement of the methyl groups in inter molecular (NZ)CH...O interactions upon crystallization
In conclusion, introduction of (NZ)CH...O bonds by reductive methylation of surface lysines as a means to salvage targets is simple, fast, economical and non laborious. It could be the first method of choice for rescuing a target before attempting a more extensive approach involving mutagenesis.
Protein production and purification
The 26 kD ParB nuclease was expressed with a N-terminal hexa Histidine tag. The gene was PCR amplified from the genomic DNA of Pyrococcus furiosus and cloned into pET-28a vector (Invitrogen). E coli BL21 cells containing the plasmid were grown in LB cultures. The protein was purified using affinity chromatography followed by size exclusion chromatography. Nucleic acids were removed by hydroxyapatite chromatography (GE Healthcare). Further purification was achieved by an ion exchange step. The protein was exchanged into crystallization buffer (20 mM Tris, pH8.0, 200 mM NaCl) using a size exclusion column.
Methylation of the protein was done as described before  using formaldehyde and dimethylamine-borane complex (DMAB). In brief, 10 mg/ml of protein in a 1.5 ml eppendorf tube covered with aluminium foil was mixed with 40 μl of 1 M solution of formaldehyde (Sigma) and 20 μl of 1 M solution of DMAB (Sigma) in the dark at 4°C. The reaction mixture was incubated under shaking conditions for 2 h, after which the chemical additions were repeated. Finally, 10 μl of DMAB was added and the reaction mixture was incubated overnight. Excess chemicals were removed by size exclusion chromatography.
Methylated and non-methylated protein was set up for crystallization under oil as described before [19, 20]. 1-μl crystallization drops contained 0.5 μl protein mixed with 0.5 μl of crystallization solution. Commercially available sparse matrix screens (Hampton Research, Molecular Dimensions) were used for crystallization screening. Crystals for structure determination were produced using a precipitant solution consisting of 600 mM sodium di-hydrogen phosphate, 2.4 M di-potassium hydrogen phosphate, 200 mM sodium chloride, 100 mM HEPES, p H 7.3.
Data collection and structure determination
Data collection and structure determination will be described elsewhere.
Analytical sedimentation velocity experiments were carried out using a ProteomeLab™ XL-I protein characterization system (Beckman Coulter). An-60Ti rotor was used to centrifuge a 10 mg/ml protein sample suspended in 50 mM phosphate buffer, 150 mM NaCl, pH 7.2, at 60,000 rpm. Absorbance was read at 280 nm. A set of 93 scans were collected at 1 min intervals. Data was analyzed using Sedfit software
Equilibrium denaturation experiments
All fluorescence denaturation experiments were performed in 20 mM phosphate buffer pH 7.4 at 25°C, with a final protein concentration of 1.2 μM. Samples of modified and unmodified protein were mixed with different concentrations of GdmCl and allowed to equilibrate overnight before measurements were taken. Refolding experiments were also performed, by denaturing the protein for 8 h in 6 M GdmCl, then diluting the protein to give different final concentrations of GdmCl as for unfolding experiments. The intrinsic fluorescence spectra were recorded between 300 and 400 nm after excitation at 280 nm in a Hitachi F-4500 spectrofluorimeter. The fluorescence data were plotted as the centre of spectral mass as described previously . GdmCl denaturation was found to be reversible and the data were fitted to a 2-state model .
Far-UV CD experiments were performed on a Pi-star 180 instrument (Applied Photophysics, UK) using a cell of 1 mm optical path length and the same buffer as for fluorescence experiments. The protein concentration was 25 μM. The temperature was changed at a rate of 1°C per 10 min, with a step size of 0.5°C, for both heating and cooling. Thermal denaturation was found to be reversible and the data were analyzed as described previously 
Coordinates for the structure reported in the current study have been deposited in the Protein Data Bank under accession code 1VK1
List of abbreviations used
Side chain amine nitrogen
This work was funded by the 863 (Grant 2006AA02A316) and 973 (Grants 2006CB910901, 2006CB910903, 2006CB500703) Projects of the Ministry of Science and Technology of China, the National Natural Science Foundation of China (Grants 30470363, 30670427), the National Institutes of Health (Grant 1P50 GM62407), the University of Georgia Research Foundation, the Georgia Research Alliance. Crystallographic data was collected at Southeast Regional Collaborative Access Team (SER-CAT) 22-ID beamline at the Advanced Photon Source, Argonne National Laboratory. Use of the Advanced Photon Source was supported by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract number W-31-109-Eng-38.
- Stevens R, Wilson I: Industrializing structural biology. Science 2001, 293: 519–520. 10.1126/science.293.5529.519View ArticlePubMedGoogle Scholar
- Dale G, Oefner C, D'Arcy A: The protein as a variable in protein crystallization. J Struct Biol 2003, 142: 88–97. 10.1016/S1047-8477(03)00041-8View ArticlePubMedGoogle Scholar
- Rayment I: Reductive alkylation of lysine residues to alter crystallization properties of proteins. Methods in Enzymology 1997, 276: 171–179.View ArticlePubMedGoogle Scholar
- Geoghegan K, Cabacungan J, Dixon H, Feeney R: Alternative reducing agents for reductive methylation of amino groups in proteins. Int J Pept Protein Res 1981, 17: 345–352.View ArticlePubMedGoogle Scholar
- Gerken T, Jentoft J, Jentoft N, Dearborn D: Intramolecular interactions of amino groups in 13C reductively methylated hen egg-white lysozyme. J Biol Chem 1982, 257: 2894–2900.PubMedGoogle Scholar
- Means G, Feeney R: Reductive alkylation of amino groups in proteins. Biochemistry 1968, 7: 2192–2201. 10.1021/bi00846a023View ArticlePubMedGoogle Scholar
- Rypniewski W, Holden H, Rayment I: Structural consequences of reductive methylation of lysine residues in hen egg white lysozyme: An X-ray analysis at 1.8Å resolution. Biochemistry 1993, 32: 9851–9858. 10.1021/bi00088a041View ArticlePubMedGoogle Scholar
- Schubot F, Waugh D: A pivotal role for reductive methylation in the de novo crystallization of a ternary complex composed of Yersinia pestis virulence factors YopN, SycN and YscB. Acta Cryst 2004, D60: 1981–1986.Google Scholar
- Kobayashi M, Kubota M, Matsuura Y: Crystallization and improvement of crystal quality for X-ray diffraction of maltooligosyl trehalose synthase by reductive methylation of lysine residues. Acta Cryst 1999, 55(Pt 4):931–933.Google Scholar
- Walter T, Meier C, Assenberg R, Au K-F, Ren J, Verma A, Nettleship J, Owens R, Stuart D, Grimes J: Lysine methylation as a routine rescue strategy for protein crystallization. Structure 2006, 14: 1617–1622. 10.1016/j.str.2006.09.005View ArticlePubMedGoogle Scholar
- Liu Z-J, Tempel W, Ng J, Lin D, Shah A, Chen L, Horanyi P, Habel J, Kataeva I, Xu H, Yang H, Chang J, Huang L, Chang S-H, Zhou W, Lee D, Praissman J, Zhang H, Newton G, Rose J, Richardson J, Richardson D, Wang B-C: The high-throughput protein-to-structure pipeline at SECSG. Acta Crystallogr 2005, 61(Pt 6):679–684.Google Scholar
- Dale G, Kostrewa D, Gsell B, Stieger M, D'Arcy A: Crystal engineering: deletion mutagenesis of the 24 kDa fragment of the DNA gyrase B subunit from Staphylococcus aureus . Acta Crystallogr D Biol Crystallogr 1999, 55: 1626–1629. 10.1107/S0907444999008227View ArticlePubMedGoogle Scholar
- Kwong P, Wyatt R, Robinson J, Sweet R, Sodroski J, Hendrickson W: Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 1998, 393: 648–659. 10.1038/31405View ArticlePubMedGoogle Scholar
- Derewenda Z: Rational protein crystallization by mutational surface engineering. Structure 2004, 12: 529–535. 10.1016/j.str.2004.03.008View ArticlePubMedGoogle Scholar
- Longenecker K, Garrard S, Sheffield P, Derewenda Z: Protein crystallization by rational mutagenesis of surfaceresidues: Lys to Ala mutations promote crystallization of RhoGDI. Acta Crytallogr D Biol Crytallogr 2001, 57(Pt 5):679–688. 10.1107/S0907444901003122View ArticleGoogle Scholar
- Mateja A, Devedjiev Y, Krowarsch D, Longenecker K, Dauter Z, Otlewski J, Derewenda Z: The impact of Glu→Ala and Glu→Asp mutations on the crystallization properties of RhoGDI: the structure of RhoGDI at 1.3 Å resolution. Acta Crytallogr D Biol Crytallogr 2002, 58(Pt 12):1983–1991. 10.1107/S090744490201394XView ArticleGoogle Scholar
- Baud F, Karlin S: Measures of residue density in protein structures. Proc Natl Acad Sci USA 1999, 96: 12494–12499. 10.1073/pnas.96.22.12494PubMed CentralView ArticlePubMedGoogle Scholar
- Avbelj F, Fele L: Role of main-chain electrostatics, hydrophobic effect and side-chain conformational entropy in determining the secondary structure of proteins. J Mol Biol 1998, 279: 665–684. 10.1006/jmbi.1998.1792View ArticlePubMedGoogle Scholar
- Chayen N, Shaw Stewart P, Maeder D, Blow D: An automated system for micro-batch protein crystallization and screening. J Appl Cryst 1990, 23: 297–302. 10.1107/S0021889890003260View ArticleGoogle Scholar
- D'Arcy A, Elmore C, Stihle M, Johnston J: A novel approach to crystallising proteins under oil. J Crystal Growth 1996, 168: 175–80. 10.1016/0022-0248(96)00351-XView ArticleGoogle Scholar
- Derewenda Z, Lee L, Derewenda U: The occurrence of C-HO hydrogen bonds in proteins. J Mol Biol 1995, 252: 248–262. 10.1006/jmbi.1995.0492View ArticlePubMedGoogle Scholar
- Magalhaes A, Maigret B, Hoflack J, Gomes J, Scheraga H: Contribution of unusual arginine-arginine short-range interactions to stabilization and recognition in proteins. J Protein Chem 1994, 13: 195–215. 10.1007/BF01891978View ArticlePubMedGoogle Scholar
- Kump R, Damewood J: Do nitromethane and molononitrile form C-H...O hydrogen bonds? Implications for molecular recognition by crown ethers. J Chem Soc Chem Commun 1988, 621–622. 10.1039/c39880000621Google Scholar
- Taylor R, Kennard O: Crystallographic evidence for the existence of C-H...O, C-H...N and C-H...Cl hydrogen bonds. J Amer Chem Soc 1982, 104: 5063–5070. 10.1021/ja00383a012View ArticleGoogle Scholar
- Wahl M, Sundaralingam M: CH...O hydrogen bonding in biology. Trends Biochem Sci 1997, 22: 97–102. 10.1016/S0968-0004(97)01004-9View ArticlePubMedGoogle Scholar
- Desiraju G: The C-H...O hydrogen bond in crystals: What is it? Acc Chem Res 1991, 24: 290–296. 10.1021/ar00010a002View ArticleGoogle Scholar
- Donati A, Ristori S, Bonechi C, Panza L, Martini G, Rossi C: Evidences of strong C-HO bond in an o-carboranyl beta-lactoside in solution. J Am Chem Soc 2002, 124: 8778–8779. 10.1021/ja016697lView ArticlePubMedGoogle Scholar
- McPherson A: Introduction to Protein Crystallography. Methods: A Companion to Methods in Enzymology 2004, 34: 254–265. 10.1016/j.ymeth.2004.03.019View ArticlePubMedGoogle Scholar
- Lima L, Zingali R, Foguel D, Monteiro R: New insights into conformational and functional stability of human alpha-thrombin probed by high hydrostatic pressure. Eur J Biochem 2004, 271: 3580–3587. 10.1111/j.0014-2956.2004.04295.xView ArticlePubMedGoogle Scholar
- Maxwell K, Wildes , Zarrine-Afsar A, De Los Rios M, Brown A, Friel C, Hedberg L, Horng J-C, Bona D, Miller E, Vallée-Bélisle A, Main E, Bemporad F, Qiu L, Teilum K, Vu N-D, Edwards A, Ruczinski I, Poulsen F, Kragelund B, Michnick S, Chiti F, Bai Y, Hagen S, Serrano L, Oliveberg M, Raleigh D, Wittung-Stafshede P, Radford S, Jackson S, Sosnick T, Marqusee S, Davidson A, Plaxco K: Protein folding: Defining a "standard" set of experimental conditions and a preliminary kinetic data set of two-state proteins. Protein Science 2005, 14: 602–616. 10.1110/ps.041205405PubMed CentralView ArticlePubMedGoogle Scholar
- Su X, Sanbo Q, Xian-Ming P: Thermal and conformational stability of Ssh10b protein from archaeon Sulfolobus shibattae . Biochem J 2004, 382: 433–440. 10.1042/BJ20040191View ArticleGoogle Scholar
- Potterton E, Briggs P, Turkenburg M, Dodson E: A graphical user interface to the CCP 4 program suite. Acta Cryst 2003, 59(Pt 7):1131–1137.Google Scholar