Data mining crystallization databases: knowledge-based approaches to optimize protein crystal screens

Protein crystallization is a major bottleneck in protein X-ray crystallography, the workhorse of most structural proteomics projects. Because the principles that govern protein crystallization are too poorly understood to allow them to be used in a strongly predictive sense, the most common crystallization strategy entails screening a wide variety of solution conditions to identify the small subset that will support crystal nucleation and growth. We tested the hypothesis that more efficient crystallization strategies could be formulated by extracting useful patterns and correlations from the large data sets of crystallization trials created in structural proteomics projects. A database of crystallization conditions was constructed for 755 different proteins purified and crystallized under uniform conditions. Forty-five percent of the proteins formed crystals. Data mining identified the conditions that crystallize the most proteins, revealed that many conditions are highly correlated in their behavior, and showed that the crystallization success rate is markedly dependent on the organism from which proteins derive. Of the proteins that crystallized in a 48-condition experiment, 60% could be crystallized in as few as 6 conditions and 94% in 24 conditions. Consideration of the full range of information coming from crystal screening trials allows one to design screens that are maximally productive while consuming minimal resources, and also suggests further useful conditions for extending existing screens.     

应用动态光散射对一些蛋白质结晶性能的研究

动态光散射是研究生物大分子溶液物化行为和结晶性能的重要工具。以溶菌酶为模型蛋白 ,对其溶液的DLS研究表明 ,一定范围的蛋白质浓度不会影响蛋白质聚合性质 ,而起沉淀剂作用的盐对多色散性的影响较大。考虑这些研究结果 ,应用DLS研究对空间微重力下晶体生长样品的质量做了考察 ,这对于提高空间样品准备水平 ,进而提高空间实验成功率是一重要途径。     

Crystallization of biological macromolecules from flash frozen samples on the Russian Space Station Mir

One hundred eighty-three flash frozen, liquid-liquid diffusion and batch method protein and virus crystallization samples were launched aboard the Space Shuttle Discovery on June 27 (STS-71) and transferred to the Russian Space Station Mir on July 1, 1995. They were returned to earth November 20, 1995 (STS-74). Subsequent examination showed that of the 19 types of proteins and viruses investigated, 17 were crystallized during the period on Mir. The experiment demonstrates the utility of this very simple and inexpensive approach for the crystallization of biological macromolecules in space over extended time periods. The distribution of crystals among the three types of containers used indicated small samples yielded results equal or better than larger samples and that long diffusion path lengths were clearly better. Distribution of crystals within the container tubes showed a striking gradient of quality and size that indicated long, narrow tubes yield superior crystals, as predicted from other work based on crystallization in capillaries.     

Use of detergents in two-dimensional crystallization of membrane proteins

Structure determination at high resolution is actually a difficult challenge for membrane proteins and the number of membrane proteins that have been crystallized is still small and far behind that of soluble proteins. Because of their amphiphilic character, membrane proteins need to be isolated, purified and crystallized in detergent solutions. This makes it difficult to grow the well-ordered three-dimensional crystals that are required for high resolution structure analysis by X-ray crystallography. In this difficult context, growing crystals confined to two dimensions (2D crystals) and their structural analysis by electron crystallography has opened a new way to solve the structure of membrane proteins. However, 2D crystallization is one of the major bottlenecks in the structural studies of membrane proteins. Advances in our understanding of the interaction between proteins, lipids and detergents as well as development and improvement of new strategies will facilitate the success rate of 2D crystallization. This review deals with the various available strategies for obtaining 2D crystals from detergent-solubilized intrinsic membrane proteins. It gives an overview of the methods that have been applied and gives details and suggestions of the physical processes leading to the formation of the ordered arrays which may be of help for getting more proteins crystallized in a form suitable for high resolution structural analysis by electron crystallography.     

Searching for silver bullets: an alternative strategy for crystallizing macromolecules

Based on a hypothesis that various small molecules might establish stabilizing, intermolecular, non covalent crosslinks in protein crystals and thereby promote lattice formation, we carried out three separate experiments. We assessed the impact of 200 chemicals on the propensity of 81 different proteins and viruses to crystallize. The experiments were comprised of 18240 vapor diffusion trials. A salient feature of the experiments was that, aside from the inclusion of the reagent mixes, only two fundamental crystallization conditions were used, 30% PEG 3350, and 50% Tacsimate at pH 7. Overall, 65 proteins (85%) were crystallized. Most significant was that 35 of the 65 (54%) crystallized only in the presence of one or more reagent mixes, but not in control samples lacking any additives. Among the most promising types of reagent mixes were those composed of polyvalent, charged groups, such as di and tri carboxylic acids, diamino compounds, molecules bearing one or more sulfonyl or phosphate groups, and a broad range of common biochemicals, coenzymes, biological effectors, and ligands. We propose that an alternate approach to crystallizing proteins might be developed, which employs a limited set of fundamental crystallization conditions combined with a broad screen of potentially useful small molecule additives.     

Statistical Analysis of Crystallization Database Links Protein Physico-Chemical Features with Crystallization Mechanisms

X-ray crystallography is the predominant method for obtaining atomic-scale information about biological macromolecules. Despite the success of the technique, obtaining well diffracting crystals still critically limits going from protein to structure. In practice, the crystallization process proceeds through knowledge-informed empiricism. Better physico-chemical understanding remains elusive because of the large number of variables involved, hence little guidance is available to systematically identify solution conditions that promote crystallization. To help determine relationships between macromolecular properties and their crystallization propensity, we have trained statistical models on samples for 182 proteins supplied by the Northeast Structural Genomics consortium. Gaussian processes, which capture trends beyond the reach of linear statistical models, distinguish between two main physico-chemical mechanisms driving crystallization. One is characterized by low levels of side chain entropy and has been extensively reported in the literature. The other identifies specific electrostatic interactions not previously described in the crystallization context. Because evidence for two distinct mechanisms can be gleaned both from crystal contacts and from solution conditions leading to successful crystallization, the model offers future avenues for optimizing crystallization screens based on partial structural information. The availability of crystallization data coupled with structural outcomes analyzed through state-of-the-art statistical models may thus guide macromolecular crystallization toward a more rational basis.     

Cloud-point temperature and liquid-liquid phase separation of supersaturated lysozyme solution

The detailed understanding of the structure of biological macromolecules reveals their functions, and is thus important in the design of new medicines and for engineering molecules with improved properties for industrial applications. Although techniques used for protein crystallization have been progressing greatly, protein crystallization may still be considered an art rather than a science, and successful crystallization remains largely empirical and operator-dependent. In this work, a microcalorimetric technique has been utilized to investigate liquid-liquid phase separation through measuring cloud-point temperature T(cloud) for supersaturated lysozyme solution. The effects of ionic strength and glycerol on the cloud-point temperature are studied in detail. Over the entire range of salt concentrations studied, the cloud-point temperature increases monotonically with the concentration of sodium chloride. When glycerol is added as additive, the solubility of lysozyme is increased, whereas the cloud-point temperature is decreased.     

Introduction to protein crystallization

Biological macromolecules can be crystallized by a variety of techniques, and using a wide range of reagents which produce supersaturated mother liquors. These may, in turn, be applied under different physical conditions such as temperature. The fundamental approaches to devising successful crystallization conditions and the factors that influence them are summarized here. For the Novice, it is hoped that this brief review might serve as a useful introduction and a stepping-stone to a successful X-ray structure determination. In addition, it may provide a framework in which to place the articles that follow.     

Randomness in crystallization of proteins from Staphylococcus aureus

Of many factors affecting protein crystallization, randomness in proteins has been given less attention although highly structured proteins would be at low entropy state. The factors, which impact on protein crystallization, are almost exclusively related to non-random amino acid properties such as physiochemical properties of amino acids. In this study, we used logistic regression and neural network to model the success rate of crystallization of 420 proteins from Staphylococcus aureus with each of non-random and random amino acid properties in order to determine whether randomness in a protein plays a role in the crystallization process. The results show that randomness is indeed involved in the crystallization process, and this rationale would enrich our knowledge on crystallization process and enhance our ability to crystallize more important proteins.     

Strategies for crystallizing membrane proteins

Crystallizing membrane proteins remains a challenging endeavor despite the increasing number of membrane protein structures solved by X-ray crystallography. The critical factors in determining the success of the crystallization experiments are the purification and preparation of membrane protein samples. Moreover, there is the added complication that the crystallization conditions must be optimized for use in the presence of detergents although the methods used to crystallize most membrane proteins are, in essence, straightforward applications of standard methodologies for soluble protein crystallization. The roles that detergents play in the stability and aggregation of membrane proteins as well as the colloidal properties of the protein-detergent complexes need to be appreciated and controlledbefore and during the crystallization trials. All X-ray quality crystals of membrane proteins were grown from preparations of detergent-solubilized protein, where the heterogeneous natural lipids from the membrane have been replaced by ahomogeneous detergent environment. It is the preparation of such monodisperse, isotropic solutions of membrane proteins that has allowed the successful application of the standard crystallization methods routinely used on soluble proteins. In this review, the issues of protein purification and sample preparation are addressed as well as the new refinements in crystallization methodologies for membrane proteins. How the physical behavior of the detergent, in the form of micelles or protein-detergent aggregates, affects crystallization and the adaptation of published protocols to new membrane protein systems are also addressed. The general conclusion is that many integral membrane proteins could be crystallized if pure and monodisperse preparations in a suitable detergent system can be prepared.In memory of Glenn D. Garavito.     

Crystallization and preliminary crystallographic analysis of San Miguel sea lion virus: an animal calicivirus

The Caliciviridae is a family of nonenveloped, icosahedral, positive-sense single-stranded RNA viruses. This family of viruses consists of both animal and human pathogens. Adapting human caliciviruses to cell culture has not been successful, whereas some animal caliciviruses, including San Miguel sea lion virus, have been successfully propagated in vitro. Here we report the crystallization of San Miguel sea lion virus serotype 4 (SMSV4) and the preliminary X-ray crystallographic analysis of the crystals. SMSV4 have been crystallized using the hanging-drop method. These crystals diffracted to approximately 3A resolution using a synchrotron radiation source. A single crystal under cryo-conditions yielded a complete set of diffraction data. Data processing of the diffraction patterns showed that SMSV crystals belong to I23 space group with cell dimensions a=b=c=457 A. The crystallographic asymmetric unit includes five icosahedral asymmetric units, each consisting of three capsid protein subunits. In the space group I23, given the icosahedral symmetry and the size of the virus particle, the location of the particle is constrained to be at the point where the crystallographic 2- and 3-fold axes intersect. The orientation of the virus particle in the unit cell was ascertained by self-rotation function calculations.     

Model for calcium binding to gamma-carboxyglutamic acid residues of proteins: crystal structure of calcium alpha-ethylmalonate

The crystal structure of a Ca2+ salt of alpha-ethylmalonic acid was determined from three-dimensional X-ray diffraction data. The dicarboxylate anion represents the functional side chain of gamma-carboxyglutamic acid (Gla) residues, which are implicated as essential calcium-binding ligands in a variety of proteins. The alpha-ethylmalonate ion chelates the Ca2+ ion in a bidentate manner that involves an O atom from each of the two malonate carboxylate groups. This type of binding arises from the constrained arrangement of carboxylate ligands in the malonate group and may be of significance to the calcium-binding properties of Gla-containing sites in proteins. The Ca2+-malonate chelation forms a six-membered ring, which is stabilized by interactions that are consistent with the preferred stereochemistries of both calcium-carboxylate and metal-malonate complexes. No other interactions are observed between Ca2+ ions and alpha-ethylmalonate ions that depend upon the malonate juxtaposition of two carboxylate groups. The potential for this type of binding distinguishes Gla residues from the monocarboxylate residues, aspartate and glutamate, and confers a novel calcium-chelation ability upon Gla-containing sites in proteins.     

Protein purification by bulk crystallization: the recovery of ovalbumin

Crystallization is used industrially for the recovery and purification of many inorganic and organic materials. However, very little is reported on the application of bulk crystallization for proteins. In this work, ovalbumin was selected as a model protein to investigate the feasibility of using bulk crystallization for the recovery and purification of proteins. A stirred 1-L seeded batch crystallizer was used to obtain the crystal growth kinetics of ovalbumin in ammonium sulfate solutions at 30 degrees C. The width of the metastable region, in which crystal growth can occur without any nucleation, is equivalent to a relative supersaturation of about 20. The bulk crystallizations were undertaken within this range (using initial relative supersaturations less than 10) and nucleation was not observed. The ovalbumin concentration in solution was measured by UV absorbance and checked by crystal content measurement. Crystal size distributions were measured both by using a Malvern Mastersizer and by counting crystals through a microscope. The crystal growth rate was found to have a second-order dependence upon the ovalbumin supersaturation. While there is no discernible effect of ammonium sulfate concentration at pH 4.90, there is a slight effect at higher pH values. Overall the effect of ammonium sulfate concentration is small compared to the effect of pH, for which there is a 10-fold increase in the growth rate constant, k(Gsigma) over the range pH 4.6-5.4. To demonstrate the degree of purification which can be achieved by bulk crystallization, ovalbumin was crystallized from a solution containing conalbumin (80,000 Da) and lysozyme (14, 600 Da). After one crystallization and a crystal wash, ovalbumin crystals were produced with a protein purity greater than 99%. No contamination by the other proteins was observed when using overloaded sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) stained with Coomassie blue stain and only trace amounts of lysozyme were observed using a silver stain. The presence of these other proteins in solution did not effect the crystal growth rate constant, k(Gsigma). The study demonstrates the feasibility of using bulk crystallization for the recovery and purification of ovalbumin. It should be readily applicable to other protein systems. (c) 1995 John Wiley & Sons, Inc.     

Rapid Isolation of Single-chain Antibodies for Structural Genomics

High throughput approaches to structural genomics requires expression, purification, and crystallization of proteins derived from predicted open reading frames cloned into a host organism, typically E. coli. Early results from this approach suggest that the success rate of obtaining well diffracting crystals from eukaryotic proteins is disappointingly low. A proven method of improving the odds of crystallization is formation of a complex with a conformation-stabilizing partner of known structure that is easily crystallized. Such complexes are also able to engage in different crystal contacts than the original protein by itself. Fab fragments derived from monoclonal antibodies have been successfully used for this purpose for a variety of proteins, however conventional methods for the isolation of monoclonal antibodies from hybridomas are time consuming and expensive. We are exploring the use of phage display to generate recombinant antibodies to target proteins that can be used to obtain co-complexes to facilitate crystallization and structural determination. We are using a large, human single-chain Fv (scFv) library to select for antibodies that bind to a panel of Leishmania major target proteins. Thirteen out of 16 target proteins yielded good binders after three rounds of enrichment. A total of 55 distinct scFvs were identified, with five targets each yielding at least five different scFvs. Individual clones were analyzed for binding specificity and soluble scFv can be readily produced and purified via the appended His₆ epitope tag. Using immunoaffinity chromatography, eight scFv target protein pairs were identified that exhibit stable complex formation and are suitable for co-crystallization trials.     

Diagnostic of precipitant for biomacromolecule crystallization by quasi-elastic light-scattering

The translational diffusion coefficient D25,w of hen egg-white lysozyme and concanavalin A from the jack bean is measured in various precipitating agent solutions as a function of salt and protein concentration using quasi-elastic light-scattering. With some precipitants, in undersaturated protein solutions, a protein or salt concentration dependence of the diffusion coefficient of the scatters is observed. It can be correlated with the inability of the protein to crystallize in this precipitant once the solution is supersaturated. These variations of D25,w are interpreted in terms of non-specific interactions and/or aggregation that prevent the protein from making appropriate contacts to form a crystal. With other precipitants known to lead to crystallization, no significant variation of the diffusion coefficient with increasing concentration was observed, indicating that under such conditions up to saturation the proteins remain essentially monodisperse. Application of this technique to find crystallization conditions of other proteins is discussed.     

Development of an alternative approach to protein crystallization

We are developing an alternate strategy for the crystallization of macromolecules that does not, like current methods, depend on the optimization of traditional variables such as pH and precipitant concentration, but is based on the hypothesis that many conventional small molecules might establish stabilizing, intermolecular, non covalent crosslinks in crystals, and thereby promote lattice formation. To test the hypothesis, we carried out preliminary experiments encompassing 18,240 crystallization trials using 81 different proteins, and 200 chemical compounds. Statistical analysis of the results demonstrated the validity of the idea. In addition, we conducted X-ray diffraction analyses of some of the crystals grown in the experiments. These clearly showed incorporation of conventional molecules into the protein crystal lattices, and further validated the underlying hypothesis. We are currently extending the investigations to include a broader and more diverse set of proteins, an expanded search of conventional and biologically active small molecules, and a wider range of precipitants. The strategy proposed here is essentially orthogonal to current approaches and has an objective of doubling the success rate of today.     

Rationalization of membrane protein crystallization with polyethylene glycol using a simple depletion model

Based on the importance of crystallizing membrane proteins in a rational way, cytochrome bc(1) complex (BC1) was crystallized using polyethylene glycol (PEG) as a sole crystallization agent. Interaction between protein-detergent complexes of BC1 was estimated by dynamic light scattering, and was compared with the numerical calculation using the Derjaguin-Landau-Verwey-Overbeek potential plus a depletion potential, without considering specific surface properties of the protein-detergent complexes. The experiments and calculation were found to be consistent and we obtained a relation between PEG molecular weight M and the range of depletion zone delta as delta approximately M(0.48+/-0.02). The stability of liquid phase of BC1 solutions was controlled by a ratio of (the range of depletion zone)/(the radius of a BC1 particle), which was consistent with recent theoretical predictions. The crystallization was most successful under a condition where the stability of the liquid phase changed from stable to unstable. The PEG molecular weight that fulfilled this condition coincided with the one used empirically to crystallize BC1 in the past by a number of groups. These results are compared to the fact that membrane proteins were often successfully crystallized close to the detergent cloud point.     

Strategies for the crystallization of viruses: using phase diagrams and gels to produce 3D crystals of Grapevine fanleaf virus

The small icosahedral plant RNA nepovirus Grapevine fanleaf virus (GFLV) is specifically transmitted by a nematode and causes major damage to vineyards worldwide. To elucidate the molecular mechanisms underlying the recognition between the surface of its protein capsid and cellular components of its vector, host and viral proteins synthesized upon infection, the wild type GFLV strain F13 and a natural mutant (GFLV-TD) carrying a Gly???Asp mutation were purified, characterized and crystallized. Subsequently, the geometry and volume of their crystals was optimized by establishing phase diagrams. GFLV-TD was twice as soluble as the parent virus in the crystallization solution and its crystals diffracted X-rays to a resolution of 2.7 ?. The diffraction limit of GFLV-F13 crystals was extended from 5.5 to 3 ? by growth in agarose gel. Preliminary crystallographic analyses indicate that both types of crystals are suitable for structure determination. Keys for the successful production of GFLV crystals include the rigorous quality control of virus preparations, crystal quality improvement using phase diagrams, and crystal lattice reinforcement by growth in agarose gel. These strategies are applicable to the production of well-diffracting crystals of other viruses and macromolecular assemblies.     

The challenge of protein structure determination--lessons from structural genomics

The process of experimental determination of protein structure is marred with a high ratio of failures at many stages. With availability of large quantities of data from high-throughput structure determination in structural genomics centers, we can now learn to recognize protein features correlated with failures; thus, we can recognize proteins more likely to succeed and eventually learn how to modify those that are less likely to succeed. Here, we identify several protein features that correlate strongly with successful protein production and crystallization and combine them into a single score that assesses "crystallization feasibility." The formula derived here was tested with a jackknife procedure and validated on independent benchmark sets. The "crystallization feasibility" score described here is being applied to target selection in the Joint Center for Structural Genomics, and is now contributing to increasing the success rate, lowering the costs, and shortening the time for protein structure determination. Analyses of PDB depositions suggest that very similar features also play a role in non-high-throughput structure determination, suggesting that this crystallization feasibility score would also be of significant interest to structural biology, as well as to molecular and biochemistry laboratories.     

Osmotic second virial cross‐coefficient measurements for binary combination of lysozyme,ovalbumin, and α‐amylase in salt solutions

Interactions measurement is a valuable tool to predict equilibrium phase separation of a desired protein in the presence of unwanted macromolecules. In this study, cross‐interactions were measured as the osmotic second virial cross‐coefficients (B₂₃) for the three binary protein systems involving lysozyme, ovalbumin, and α‐amylase in salt solutions (sodium chloride and ammonium sulfate). They were correlated with solubility for the binary protein mixtures. The cross‐interaction behavior at different salt concentrations was interpreted by either electrostatic or hydrophobic interaction forces. At low salt concentrations, the protein surface charge dominates cross‐interaction behavior as a function of pH. With added ovalbumin, the lysozyme solubility decreased linearly at low salt concentration in sodium chloride and increased at high salt concentration in ammonium sulfate. The B₂₃ value was found to be proportional to the slope of the lysozyme solubility against ovalbumin concentration and the correlation was explained by preferential interaction theory. © 2013 American Institute of Chemical Engineers Biotechnol. Prog., 29:1203–1211, 2013