Genome pool strategy for structural coverage of protein families

Insights into the Structure of Rubisco from Dinoflagellates-In Silico Studies

Ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) is one of the best studied enzymes. It is crucial for photosynthesis, and thus for all of biosphere’s productivity. There are four isoforms of this enzyme, differing by amino acid sequence composition and quaternary structure. However, there is still a group of organisms, dinoflagellates, single-cell eukaryotes, that are confirmed to possess Rubisco, but no successful purification of the enzyme of such origin, and hence a generation of a crystal structure was reported to date. Here, we are using in silico tools to generate the possible structure of Rubisco from a dinoflagellate representative, Symbiodinium sp. We selected two templates: Rubisco from Rhodospirillum rubrum and Rhodopseudomonas palustris. Both enzymes are the so-called form II Rubiscos, but the first is exclusively a homodimer, while the second one forms homo-hexamers. Obtained models show no differences in amino acids crucial for Rubisco activity. The variation was found at two closely located inserts in the C-terminal domain, of which one extends a helix and the other forms a loop. These inserts most probably do not play a direct role in the enzyme’s activity, but may be responsible for interaction with an unknown protein partner, possibly a regulator or a chaperone. Analysis of the possible oligomerization interface indicated that Symbiodinium sp. Rubisco most likely forms a trimer of homodimers, not just a homodimer. This hypothesis was empowered by calculation of binding energies. Additionally, we found that the protein of study is significantly richer in cysteine residues, which may be the cause for its activity loss shortly after cell lysis. Furthermore, we evaluated the influence of the loop insert, identified exclusively in the Symbiodinium sp. protein, on the functionality of the recombinantly expressed R. rubrum Rubisco. All these findings shed new light onto dinoflagellate Rubisco and may help in future obtainment of a native, active enzyme.

The "Sticky Patch" Model of Crystallization and Modification of Proteins for Enhanced Crystallizability

Crystallization of macromolecules has long been perceived as a stochastic process, which cannot be predicted or controlled. This is consistent with another popular notion that the interactions of molecules within the crystal, i.e., crystal contacts, are essentially random and devoid of specific physicochemical features. In contrast, functionally relevant surfaces, such as oligomerization interfaces and specific protein-protein interaction sites, are under evolutionary pressures so their amino acid composition, structure, and topology are distinct. However, current theoretical and experimental studies are significantly changing our understanding of the nature of crystallization. The increasingly popular "sticky patch" model, derived from soft matter physics, describes crystallization as a process driven by interactions between select, specific surface patches, with properties thermodynamically favorable for cohesive interactions. Independent support for this model comes from various sources including structural studies and bioinformatics. Proteins that are recalcitrant to crystallization can be modified for enhanced crystallizability through chemical or mutational modification of their surface to effectively engineer "sticky patches" which would drive crystallization. Here, we discuss the current state of knowledge of the relationship between the microscopic properties of the target macromolecule and its crystallizability, focusing on the "sticky patch" model. We discuss state-of-the-art in silico methods that evaluate the propensity of a given target protein to form crystals based on these relationships, with the objective to design variants with modified molecular surface properties and enhanced crystallization propensity. We illustrate this discussion with specific cases where these approaches allowed to generate crystals suitable for structural analysis.

Is unphosphorylated Rex, as multifunctional protein of HTLV-1, a fully intrinsically disordered protein? An in silico study

Intracellularlocation of a viral unspliced mRNA in host cell is a crucial factor for normal life of the virus. Rex is a neucleo-cytoplasmic shuffling protein of Human T-cell Leukemia Virus-1(HTLV-1)which has important role in active transport of cargo-containing RNA from nucleus to cytoplasm. Therefore, it plays a crucial role in the disease development by the virus. In spite of its importance, the 3d-structurephosphorylated and unphosphorylated of this protein has not been determined. In this study, first we predicted whether Rex protein is an ordered or disordered protein. In second step protein 3Dstructure of Rex was obtained. The content of disorder-promoting amino acids, flexibility, hydrophobicity, short linear motifs (SLiMs) and protein binding regions and probability of Rex crystallization were calculated by various In Silico methods. The3D models of Rex protein were obtained by various In Silico methods, such as homology modeling, threading and ab initio, including; I-TASSER, LOMETS, SPARSKS, ROBBETA and QUARK servers. By comparing and analyzing Qmean, z-scores and energy levels of selected models, the best structures with highest favored region in Ramachandran plot (higher than 90%) was refined with MODREFINER software. In silico analysis of Rex physicochemical properties and also predicted SLiMs and binding regions sites confirms that unphosphorylated Rex protein in HTLV-1 as Rev protin in HIV is wholly disordered protein belongs to the class of intrinsically disordered proteins with extended disorder (native coils, native pre-molten globules).

•

Physico-chemical properties of Rex protein were confirmed unphosphorilated Rex protein is a wholly intrinsically disordered protein.

•

The 3d-structure model of Rex protein was determined.

Increasing the structural coverage of tuberculosis drug targets

High-resolution three-dimensional structures of essential Mycobacterium tuberculosis (Mtb) proteins provide templates for TB drug design, but are available for only a small fraction of the Mtb proteome. Here we evaluate an intra-genus "homolog-rescue" strategy to increase the structural information available for TB drug discovery by using mycobacterial homologs with conserved active sites. Of 179 potential TB drug targets selected for x-ray structure determination, only 16 yielded a crystal structure. By adding 1675 homologs from nine other mycobacterial species to the pipeline, structures representing an additional 52 otherwise intractable targets were solved. To determine whether these homolog structures would be useful surrogates in TB drug design, we compared the active sites of 106 pairs of Mtb and non-TB mycobacterial (NTM) enzyme homologs with experimentally determined structures, using three metrics of active site similarity, including superposition of continuous pharmacophoric property distributions. Pair-wise structural comparisons revealed that 19/22 pairs with >55% overall sequence identity had active site Cα RMSD <1 Å, >85% side chain identity, and ≥80% PSAPF (similarity based on pharmacophoric properties) indicating highly conserved active site shape and chemistry. Applying these results to the 52 NTM structures described above, 41 shared >55% sequence identity with the Mtb target, thus increasing the effective structural coverage of the 179 Mtb targets over three-fold (from 9% to 32%). The utility of these structures in TB drug design can be tested by designing inhibitors using the homolog structure and assaying the cognate Mtb enzyme; a promising test case, Mtb cytidylate kinase, is described. The homolog-rescue strategy evaluated here for TB is also generalizable to drug targets for other diseases.

Improving the chances of successful protein structure determination with a random forest classifier

Obtaining diffraction quality crystals remains one of the major bottlenecks in structural biology. The ability to predict the chances of crystallization from the amino-acid sequence of the protein can, at least partly, address this problem by allowing a crystallographer to select homologs that are more likely to succeed and/or to modify the sequence of the target to avoid features that are detrimental to successful crystallization. In 2007, the now widely used XtalPred algorithm [Slabinski et al. (2007), Protein Sci. 16, 2472-2482] was developed. XtalPred classifies proteins into five `crystallization classes' based on a simple statistical analysis of the physicochemical features of a protein. Here, towards the same goal, advanced machine-learning methods are applied and, in addition, the predictive potential of additional protein features such as predicted surface ruggedness, hydrophobicity, side-chain entropy of surface residues and amino-acid composition of the predicted protein surface are tested. The new XtalPred-RF (random forest) achieves significant improvement of the prediction of crystallization success over the original XtalPred. To illustrate this, XtalPred-RF was tested by revisiting target selection from 271 Pfam families targeted by the Joint Center for Structural Genomics (JCSG) in PSI-2, and it was estimated that the number of targets entered into the protein-production and crystallization pipeline could have been reduced by 30% without lowering the number of families for which the first structures were solved. The prediction improvement depends on the subset of targets used as a testing set and reaches 100% (i.e. twofold) for the top class of predicted targets.

Correlation between protein sequence similarity and crystallization reagents in the biological macromolecule crystallization database

The protein structural entries grew far slower than the sequence entries. This is partly due to the bottleneck in obtaining diffraction quality protein crystals for structural determination using X-ray crystallography. The first step to achieve protein crystallization is to find out suitable chemical reagents. However, it is not an easy task. Exhausting trial and error tests of numerous combinations of different reagents mixed with the protein solution are usually necessary to screen out the pursuing crystallization conditions. Therefore, any attempts to help find suitable reagents for protein crystallization are helpful. In this paper, an analysis of the relationship between the protein sequence similarity and the crystallization reagents according to the information from the existing databases is presented. We extracted information of reagents and sequences from the Biological Macromolecule Crystallization Database (BMCD) and the Protein Data Bank (PDB) database, classified the proteins into different clusters according to the sequence similarity, and statistically analyzed the relationship between the sequence similarity and the crystallization reagents. The results showed that there is a pronounced positive correlation between them. Therefore, according to the correlation, prediction of feasible chemical reagents that are suitable to be used in crystallization screens for a specific protein is possible.

An approach to crystallizing proteins by metal-mediated synthetic symmetrization

Combining the concepts of synthetic symmetrization with the approach of engineering metal-binding sites, we have developed a new crystallization methodology termed metal-mediated synthetic symmetrization. In this method, pairs of histidine or cysteine mutations are introduced on the surface of target proteins, generating crystal lattice contacts or oligomeric assemblies upon coordination with metal. Metal-mediated synthetic symmetrization greatly expands the packing and oligomeric assembly possibilities of target proteins, thereby increasing the chances of growing diffraction-quality crystals. To demonstrate this method, we designed various T4 lysozyme (T4L) and maltose-binding protein (MBP) mutants and cocrystallized them with one of three metal ions: copper (Cu²⁺, nickel (Ni²⁺), or zinc (Zn²⁺). The approach resulted in 16 new crystal structures--eight for T4L and eight for MBP--displaying a variety of oligomeric assemblies and packing modes, representing in total 13 new and distinct crystal forms for these proteins. We discuss the potential utility of the method for crystallizing target proteins of unknown structure by engineering in pairs of histidine or cysteine residues. As an alternate strategy, we propose that the varied crystallization-prone forms of T4L or MBP engineered in this work could be used as crystallization chaperones, by fusing them genetically to target proteins of interest.

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.

The JCSG high-throughput structural biology pipeline

The Joint Center for Structural Genomics high-throughput structural biology pipeline has delivered more than 1000 structures to the community over the past ten years. The JCSG has made a significant contribution to the overall goal of the NIH Protein Structure Initiative (PSI) of expanding structural coverage of the protein universe, as well as making substantial inroads into structural coverage of an entire organism. Targets are processed through an extensive combination of bioinformatics and biophysical analyses to efficiently characterize and optimize each target prior to selection for structure determination. The pipeline uses parallel processing methods at almost every step in the process and can adapt to a wide range of protein targets from bacterial to human. The construction, expansion and optimization of the JCSG gene-to-structure pipeline over the years have resulted in many technological and methodological advances and developments. The vast number of targets and the enormous amounts of associated data processed through the multiple stages of the experimental pipeline required the development of variety of valuable resources that, wherever feasible, have been converted to free-access web-based tools and applications.

Prediction of protein crystallization outcome using a hybrid method

The great power of protein crystallography to reveal biological structure is often limited by the tremendous effort required to produce suitable crystals. A hybrid crystal growth predictive model is presented that combines both experimental and sequence-derived data from target proteins, including novel variables derived from physico-chemical characterization such as R(30), the ratio between a protein's DSF intensity at 30°C and at T(m). This hybrid model is shown to be more powerful than sequence-based prediction alone - and more likely to be useful for prioritizing and directing the efforts of structural genomics and individual structural biology laboratories.

Comprehensive classification of nucleotidyltransferase fold proteins: identification of novel families and their representatives in human

This article presents a comprehensive review of large and highly diverse superfamily of nucleotidyltransferase fold proteins by providing a global picture about their evolutionary history, sequence-structure diversity and fulfilled functional roles. Using top-of-the-line homology detection method combined with transitive searches and fold recognition, we revised the realm of these superfamily in numerous databases of catalogued protein families and structures, and identified 10 new families of nucleotidyltransferase fold. These families include hundreds of previously uncharacterized and various poorly annotated proteins such as Fukutin/LICD, NFAT, FAM46, Mab-21 and NRAP. Some of these proteins seem to play novel important roles, not observed before for this superfamily, such as regulation of gene expression or choline incorporation into cell membrane. Importantly, within newly detected families we identified 25 novel superfamily members in human genome. Among these newly assigned members are proteins known to be involved in congenital muscular dystrophy, neurological diseases and retinal pigmentosa what sheds some new light on the molecular background of these genetic disorders. Twelve of new human nucleotidyltransferase fold proteins belong to Mab-21 family known to be involved in organogenesis and development. The determination of specific biological functions of these newly detected proteins remains a challenging task.

Structural genomics target selection for the New York consortium on membrane protein structure

The New York Consortium on Membrane Protein Structure (NYCOMPS), a part of the Protein Structure Initiative (PSI) in the USA, has as its mission to establish a high-throughput pipeline for determination of novel integral membrane protein structures. Here we describe our current target selection protocol, which applies structural genomics approaches informed by the collective experience of our team of investigators. We first extract all annotated proteins from our reagent genomes, i.e. the 96 fully sequenced prokaryotic genomes from which we clone DNA. We filter this initial pool of sequences and obtain a list of valid targets. NYCOMPS defines valid targets as those that, among other features, have at least two predicted transmembrane helices, no predicted long disordered regions and, except for community nominated targets, no significant sequence similarity in the predicted transmembrane region to any known protein structure. Proteins that feed our experimental pipeline are selected by defining a protein seed and searching the set of all valid targets for proteins that are likely to have a transmembrane region structurally similar to that of the seed. We require sequence similarity aligning at least half of the predicted transmembrane region of seed and target. Seeds are selected according to their feasibility and/or biological interest, and they include both centrally selected targets and community nominated targets. As of December 2008, over 6,000 targets have been selected and are currently being processed by the experimental pipeline. We discuss how our target list may impact structural coverage of the membrane protein space.