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.     

Is myelin basic protein crystallizable?

Myelin basic protein (MBP) is the predominant extrinsic protein in both central and peripheral nervous system myelins. It is thought to be involved in the stabilizing interactions between myelin membranes, and it may play an important role in demyelinating diseases such as multiple sclerosis. In spite of the fact that this abundant protein has been known for almost three decades, its three-dimensional crystal structure has not yet been determined. In this study we report on our extensive attempts to crystallize the major 18.5 kDa isoform of MBP. We used MBP having different degrees of purity, ranging from crude MBP (that was acid or salt extracted from isolated myelin), to highest purity single isoform. We used conventional strategies in our search for a suitable composition or a crystallization medium. We applied both full and incomplete factorial searches for crystallization conditions. We analyzed the available data on proteins which have previously resisted crystallization, and applied this information to our own experiments. Nevertheless, despite our efforts which included 4600 different conditions, we were unable to induce crystallization of MBP. Previous work on MBP indicates that when it is removed from its native environment in the myelin membrane and put in crystallization media, the protein adopts a random coil conformation and persists as a population of structurally non-identical molecules. This thermodynamically preferred state presumably hinders crystallization, because the most fundamental factor of protein crystallization-homogeneity of tertiary structure-is lacking. We conclude that as long as its random coil flexibility is not suppressed, 18.5 kDa MBP and possibly also its isoforms will remain preeminent examples of proteins that cannot be crystallized.     

Thermofluor-based high-throughput stability optimization of proteins for structural studies

Production of proteins well suited for structural studies is inherently difficult and time-consuming. Protein sample homogeneity, stability, and solubility are strongly correlated with the proteins' probability of yielding crystals, and optimization of these properties will improve success rates of crystallization. In the current study, we applied the thermofluor method as a high-throughput approach for identifying optimal protein formulation for crystallization. The method also allowed optimal stabilizing buffer compositions to be rapidly identified for each protein. Furthermore, the method allowed the identification of potential ligands, physiological or non-physiological, that can be used in subsequent crystallization trials. For this study, the thermally induced melting points were determined in different buffers as well as with additives for a total of 25 Escherichia coli proteins. Crystallization trials were set up together with stabilizing and destabilizing additives identified using thermofluor screening. A twofold increase in the number of crystallization leads was observed when the proteins were cocrystallized with stabilizing additives as compared with experiments without these additives. This suggests that thermofluor constitutes an efficient generic high-throughput method for identification of protein properties predictive of crystallizability.     

Stabilizing Additives Added during Cell Lysis Aid in the Solubilization of Recombinant Proteins

Insoluble recombinant proteins are a major issue for both structural genomics and enzymology research. Greater than 30% of recombinant proteins expressed in Escherichia coli (E. coli) appear to be insoluble. The prevailing view is that insolubly expressed proteins cannot be easily solubilized, and are usually sequestered into inclusion bodies. However, we hypothesize that small molecules added during the cell lysis stage can yield soluble protein from insoluble protein previously screened without additives or ligands. We present a novel screening method that utilized 144 additive conditions to increase the solubility of recombinant proteins expressed in E. coli. These selected additives are natural ligands, detergents, salts, buffers, and chemicals that have been shown to increase the stability of proteins in vivo. We present the methods used for this additive solubility screen and detailed results for 41 potential drug target recombinant proteins from infectious organisms. Increased solubility was observed for 80% of the recombinant proteins during the primary and secondary screening of lysis with the additives; that is 33 of 41 target proteins had increased solubility compared with no additive controls. Eleven additives (trehalose, glycine betaine, mannitol, L-Arginine, potassium citrate, CuCl₂, proline, xylitol, NDSB 201, CTAB and K₂PO₄) solubilized more than one of the 41 proteins; these additives can be easily screened to increase protein solubility. Large-scale purifications were attempted for 15 of the proteins using the additives identified and eight (40%) were prepared for crystallization trials during the first purification attempt. Thus, this protocol allowed us to recover about a third of seemingly insoluble proteins for crystallography and structure determination. If recombinant proteins are required in smaller quantities or less purity, the final success rate may be even higher.     

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.     

Alpha repeat proteins (αRep) as expression and crystallization helpers

We have previously described a highly diverse library of artificial repeat proteins based on thermostable HEAT-like repeats, named αRep. αReps binding specifically to proteins difficult to crystallize have been selected and in several examples, they made possible the crystallization of these proteins. To further simplify the production and crystallization experiments we have explored the production of chimeric proteins corresponding to covalent association between the targets and their specific binders strengthened by a linker. Although chimeric proteins with expression partners are classically used to enhance expression, these fusions cannot usually be used for crystallization. With specific expression partners like a cognate αRep this is no longer true, and chimeric proteins can be expressed purified and crystallized. αRep selection by phage display suppose that at least a small amount of the target protein should be produced to be used as a bait for selection and this might, in some cases, be difficult. We have therefore transferred the αRep library in a new construction adapted to selection by protein complementation assay (PCA). This new procedure allows to select specific binders by direct interaction with the target in the cytoplasm of the bacteria and consequently does not require preliminary purification of target protein. αRep binders selected by PCA or by phage display can be used to enhance expression, stability, solubility and crystallogenesis of proteins that are otherwise difficult to express, purify and/or crystallize.     

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.     

A comparison of salts for the crystallization of macromolecules

Thirty-one proteins and viruses that we knew from our own experience could be crystallized, or had been reported to have been crystallized by others, were investigated. In this experiment, each protein or virus was subjected to a crystallization screen of 12 different salts, each titrated to pH 7.2 beforehand, at concentrations ranging from 20% saturation to 90% saturation. Eight macromolecules failed to crystallize at all from any salt and were omitted from consideration. From the remaining 23 proteins, each salt was scored according to how many proteins and viruses it successfully crystallized. Among several results, one was particularly striking. Sodium malonate clearly was much more successful than any other salt, resulting in the crystallization of 19 of the 23 macromolecules, almost twice as effective as the next most successful salt, which was a draw between sodium acetate, sodium tartrate, sodium formate, and ammonium sulfate (11 of 22). The high success rate of sodium malonate in producing crystals was even more impressive when an overall unique success rate with individual macromolecules was considered.     

Purification and crystallization of creatine kinase from rabbit skeletal muscle

Crystallization is the primary rate-limiting step in protein structure determination. It has been our experience over approximately 10 years that crystals are obtained in about 20% of the proteins attempted and that only about 10% of these crystals are sufficiently well ordered to permit atomic resolution structure analysis. In attempts to overcome this limitation, we have investigated the effect on crystallization of microheterogeneity in a protein regarded as pure by conventional criteria. Creatine kinase was purified from rabbit skeletal muscle and crystallized from methylpentanediol. The protein appeared to be nearly pure judging by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and high specific activity. The crystals that were obtained were of poor quality, and an extensive survey of precipitants, crystallization conditions, and additives failed to discover conditions from which usable crystals could be obtained. The enzyme was then subjected to a series of further purification steps. After each purification step, the quality of the crystals obtained under almost identical conditions improved. The final purification step was flat-bed isoelectric focusing. Crystals grown from focused creatine kinase are well ordered and diffract to approximately 3-A resolution.     

Evidence for defects in membrane traffic in Paramecium secretory mutants unable to produce functional storage granules

The ciliated protozoan Paramecium has a regulated secretory system amenable to genetic analysis. The secretory storage granules, known as trichocysts, enclose a crystalline matrix with a genetically determined shape whose biogenesis involves proteolytic maturation of a family of precursor molecules into a heterogeneous set of small acidic polypeptides that crystallize within the maturing vesicles. We have developed an original pulse-chase protocol for monoxenic Paramecium cultures using radiolabeled bacteria to study the processing of trichocyst matrix proteins in wild-type and mutant cells. In wild-type cells, proteolytic processing is blocked in the presence of monensin and otherwise rapidly completed after approximately 20 min of chase, suggesting that the conversion occurs in the trans-Golgi and/or in small vesicles soon after sorting to the regulated pathway, probably before crystallization begins. In trichless mutant cells, which contain no visible trichocysts, secretory proteins are synthesized but not processed and we report constitutive secretion of the uncleaved precursor molecules. The mutation thus appears to affect sorting to the regulated pathway and should prove useful for analysis of the sorting machinery and of the relationship between sorting and proteolytic processing of secretory proteins. In mutants bearing misshapen trichocysts with poorly crystallized contents (tam33, tam38, stubbyA), the proteolytic processing of the trichocyst matrix proteins appears to be normal, while both pulse-chase and morphological data indicate that intracellular transport is perturbed, probably between ER and Golgi. Precursor molecules are present in the mutant trichocysts but not in wild-type trichocysts and may account for the defective crystallization. Our analysis of these mutants suggests that the temporal coordination of intracellular traffic plays a regulatory role in granule maturation.     

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.     

Studies on Crystalline Growth of O2-susceptible Proteins in Space

In order to meet the requirement for crystalline growth of O2-susceptible proteins in space, crystallization conditions on the earth was optimized for the proteins using a simple and suitable device for anaerobic addition of the protein samples. Nitrogenase is susceptible to O2. ΔnifZ MoFe protein from a nifZ deleted strain and MnFe protein from mutant strain UW3 grown on a medium containing Mn were crystallized at the first time in the world using an anaerobic device equipped with plastic bags or using a small simplified box, as a replacement for the cumbersome dry box. And the proteins could be also crystallized far from laboratory by sitting-drop method using a much lighter device. It was equipped with a smaller plastic food bag and a first-aid bag filled with Ar, as a substitute for the cumbersome dry box and the Ar cylinder, respectively. The results showed that the device could meet the requirement for studies on crystal growth of the above anaerobic proteins in space.     

Synthetic symmetrization in the crystallization and structure determination of CelA from Thermotoga maritima

Protein crystallization continues to be a major bottleneck in X‐ray crystallography. Previous studies suggest that symmetric proteins, such as homodimers, might crystallize more readily than monomeric proteins or asymmetric complexes. Proteins that are naturally monomeric can be made homodimeric artificially. Our approach is to create homodimeric proteins by introducing single cysteines into the protein of interest, which are then oxidized to form a disulfide bond between the two monomers. By introducing the single cysteine at different sequence positions, one can produce a variety of synthetically dimerized versions of a protein, with each construct expected to exhibit its own crystallization behavior. In earlier work, we demonstrated the potential utility of the approach using T4 lysozyme as a model system. Here we report the successful application of the method to Thermotoga maritima CelA, a thermophilic endoglucanase enzyme with low sequence identity to proteins with structures previously reported in the Protein Data Bank. This protein had resisted crystallization in its natural monomeric form, despite a broad survey of crystallization conditions. The synthetic dimerization of the CelA mutant D188C yielded well‐diffracting crystals with molecules in a packing arrangement that would not have occurred with native, monomeric CelA. A 2.4 Å crystal structure was determined by single anomalous dispersion using a seleno‐methionine derivatized protein. The results support the notion that synthetic symmetrization can be a useful approach for enlarging the search space for crystallizing monomeric proteins or asymmetric complexes.     

Heterologous expression, purification, crystallization, X-ray analysis and phasing of the acetyl xylan esterase from Bacillus pumilus

Bacillus pumilus PS213 acetyl xylan esterase (AXE) acts as an accessory enzyme in the plant cell wall hemicellulose biodegradation pathway. It belongs to the carbohydrate esterase family 7 and hydrolyses the ester linkages of the acetyl groups in position 2 and/or 3 of the xylose moieties of the acetylated xylan fragments from hardwood. The enzyme displays activity towards a broad range of acetylated compounds including the antibiotic cephalosporin-C. In this study we report the heterologous expression, purification, physicochemical characterization and crystallization of the recombinant B. pumilus AXE. Remarkable improvement of the crystal quality was achieved by setting up crystallization conditions, at first established using the hanging drop vapor diffusion method, in a micro-batch experiment. Rod-like diffraction quality crystals were obtained using 10% PEG 6000, 0.1 M MES pH 6.0 and a wide range of LiCl concentrations (0.2-1.0 M) as precipitant agent. Two different crystal forms, both belonging to space group P2(1), were characterized, diffracting X-rays to 2.5 and 1.9 angstrom resolution. Successful molecular replacement showed 12 molecules in the asymmetric unit of either crystal forms that are arranged as two doughnut-like hexamers, each one encompassing a local 32 symmetry. A catalytic inactive mutant Ser181Ala of B. pumilus AXE was also engineered, expressed, purified and crystallized for functional and structural studies.     

Precipitants and additives for membrane crystallization of lysozyme

Membrane crystallization is a newly developed crystallization technique that has proven to be superior in producing good crystal forms under operating conditions that are not appropriate to perform the crystallization process by other traditional techniques. In this work, static membrane crystallization was carried out on lysozyme, with hollow-fiber microporous hydrophobic membranes. Numerous precipitant and additive types and concentrations were employed in the crystallization processes in order to select the most appropriate precipitant and additive types and to find their corresponding concentration levels that can yield the best crystal forms. The crystallization processes were analyzed in two ways: firstly, by evaluation of the transmembrane fluxes obtained by using different precipitants and additives; secondly, by utilization of the images and results obtained from the micrography and IR spectra in comparisons and evaluations of the crystals formed under all kinds of conditions. Moreover, the size distributions of the crystals yielded under several typical crystallization conditions were analyzed, and turbidity and induction time periods obtained during typical crystallization experiments were also measured. Amongst the numerous precipitants and additives tested, the most appropriate precipitant type and additive were chosen and their concentrations were optimized. Good lysozyme crystals were obtained using a certain precipitant and additive. The obtained results from this work further support the advantages of utilizing the membrane crystallization technique for macromolecule crystallizations.     

Atomic force microscopy of crystalline insulins: the influence of sequence variation on crystallization and interfacial structure.

The self-association of proteins is influenced by amino acid sequence, molecular conformation, and the presence of molecular additives. In the presence of phenolic additives, LysB28ProB29 insulin, in which the C-terminal prolyl and lysyl residues of wild-type human insulin have been inverted, can be crystallized into forms resembling those of wild-type insulins in which the protein exists as zinc-complexed hexamers organized into well-defined layers. We describe herein tapping-mode atomic force microscopy (TMAFM) studies of single crystals of rhombohedral (R3) LysB28ProB29 that reveal the influence of sequence variation on hexamer-hexamer association at the surface of actively growing crystals. Molecular scale lattice images of these crystals were acquired in situ under growth conditions, enabling simultaneous identification of the rhombohedral LysB28ProB29 crystal form, its orientation, and its dynamic growth characteristics. The ability to obtain crystallographic parameters on multiple crystal faces with TMAFM confirmed that bovine and porcine insulins grown under these conditions crystallized into the same space group as LysB28ProB29 (R3), enabling direct comparison of crystal growth behavior and the influence of sequence variation. Real-time TMAFM revealed hexamer vacancies on the (001) terraces of LysB28ProB29, and more rounded dislocation noses and larger terrace widths for actively growing screw dislocations compared to wild-type bovine and porcine insulin crystals under identical conditions. This behavior is consistent with weaker interhexamer attachment energies for LysB28ProB29 at active growth sites. Comparison of the single crystal x-ray structures of wild-type insulins and LysB28ProB29 suggests that differences in protein conformation at the hexamer-hexamer interface and accompanying changes in interhexamer bonding are responsible for this behavior. These studies demonstrate that subtle changes in molecular conformation due to a single sequence inversion in a region critical for insulin self-association can have a significant effect on the crystallization of proteins.     

Crystallization and liquid‐liquid phase separation of monoclonal antibodies and fc‐fusion proteins: Screening results

Crystallization holds the potential to be used for protein purification and low‐viscosity drug substance and drug product formulations. Twenty‐two different proteins (20 monoclonal antibodies and two Fc‐fusions) were examined to determine the breadth of applicability of crystallization to these therapeutic proteins. Vapor diffusion technique and an evaporative screening method were used to identify crystallization conditions using around a 100 initial conditions based on reagents that are generally regarded as safe (GRAS). Of 16 IgG₂s examined, at least four formed diffraction‐quality crystals and four others formed crystal‐like particles. At least three of the IgG₂s that crystallized well were also crystallized under the same set of operating conditions using inexpensive GRAS reagents. The crystals were formed to high‐yields in a few hours and were dissolved quickly without impacting product quality. Although only a fraction of the proteins examined crystallized, all exhibited liquid‐liquid phase separation (LLPS), which could be used for their concentration or possibly purification. One of the Fc‐fusions, for example, was concentrated by LLPS to a self‐buffering solution at 150 g/L. Crystallization and LLPS in the salting‐in region were shown to be feasible. © 2011 American Institute of Chemical Engineers Biotechnol. Prog., 2011     

Thermodynamics of the hydrophobicity in crystallization of insulin

For insight into the solvent structure around protein molecules and its role in phase transformations, we investigate the thermodynamics of crystallization of the rhombohedral form of porcine insulin crystals. We determine the temperature dependence of the solubility at varying concentration of the co-solvent acetone, Cac=0%, 5%, 10%, 15%, and 20%, and find that, as a rule, the solubility of insulin increases as temperature increases. The enthalpy of crystallization, undergoes a stepwise shift from approximately -20 kJ mol(-1) at Cac=0%, 5%, and 10% to approximately -55 kJ mol(-1) at Cac=15% and 20%. The entropy change upon crystallization is approximately 35 J mol(-1) K(-1) for the first three acetone concentrations, and drops to approximately -110 J mol(-1) K(-1) at Cac=15% and 20%. DeltaS degrees cryst>0 indicates release of solvent, mostly water, molecules structured around the hydrophobic patches on the insulin molecules' surface in the solution. As Cac increases to 15% and above, unstructured acetone molecules apparently displace the waters and their contribution to DeltaS degrees cryst is minimal. This shifts DeltaS degrees cryst to a negative value close to the value expected for tying up of one insulin molecule from the solution. The accompanying increase in DeltaH degrees cryst suggests that the water structured around the hydrophobic surface moieties has a minimal enthalpy effect, likely due to the small size of these moieties. These findings provide values of the parameters needed to better control insulin crystallization, elucidate the role of organic additives in the crystallization of proteins, and help us to understand the thermodynamics of the hydrophobicity of protein molecules and other large molecules.     

RNA crystallization

RNA molecules may be crystallized using variations of the methods developed for protein crystallography. As the technology has become available to synthesize and purify RNA molecules in the quantities and with the quality that is required for crystallography, the field of RNA structure has exploded. The first consideration when crystallizing an RNA is the sequence, which may be varied in a rational way to enhance crystallizability or prevent formation of alternate structures. Once a sequence has been designed, the RNA may be synthesized chemically by solid-state synthesis or it may be produced enzymatically using RNA polymerase and an appropriate DNA template. Purification of milligram quantities of RNA can be accomplished by HPLC or gel electrophoresis. As with proteins, crystallization of RNA is usually accomplished by vapor diffusion techniques. There are several considerations that are either unique to RNA crystallization or more important for RNA crystallization. Techniques for design, synthesis, purification, and crystallization of RNAs will be reviewed here.     

A novel "reverse screening" to identify refolding additives for activin-A

A general approach for refolding recombinant proteins from inclusion bodies (IBs) is to screen conditions, that facilitate a conversion of unfolded to folded structure and minimize a conversion of unfolded to misfolded and aggregated structures. In this simplified model, such conditions may be those that stabilize the native protein and/or reduce aggregation. In this paper, a novel screening approach, termed reverse screening, was developed using a native activin. Activin-A, a member of transforming growth factor beta superfamily, is a homodimeric protein with nine disulfide bonds. We examined partial unfolding process of native activin-A dissolved in a buffer containing moderate concentrations of denaturant and reducing reagent (i.e., 1.5 M urea and 0.2 mM dithiothreitol). The recovery of the protein was followed by reverse-phase high performance chromatography analysis. Without additives, activin-A showed about 60% loss of the protein due to aggregation after 12-h incubation in the above condition. We then tested various additives for their effects on the recovery after partial unfolding. One of these additives, sodium taurodeoxycholate (TDCA), greatly increased recovery and suppressed aggregation of the protein. These additives were then tested for refolding activin-A from IBs. TDCA among others is proved to be a highly effective refolding additive. These results strongly suggest that reverse screening using native proteins, if available, may be another approach to discovering effective refolding additives.