Artificial neural networks trained to detect viral and phage structural proteins.

Phages play critical roles in the survival and pathogenicity of their hosts, via lysogenic conversion factors, and in nutrient redistribution, via cell lysis. Analyses of phage- and viral-encoded genes in environmental samples provide insights into the physiological impact of viruses on microbial communities and human health. However, phage ORFs are extremely diverse of which over 70% of them are dissimilar to any genes with annotated functions in GenBank. Better identification of viruses would also aid in better detection and diagnosis of disease, in vaccine development, and generally in better understanding the physiological potential of any environment. In contrast to enzymes, viral structural protein function can be much more challenging to detect from sequence data because of low sequence conservation, few known conserved catalytic sites or sequence domains, and relatively limited experimental data. We have designed a method of predicting phage structural protein sequences that uses Artificial Neural Networks (ANNs). First, we trained ANNs to classify viral structural proteins using amino acid frequency; these correctly classify a large fraction of test cases with a high degree of specificity and sensitivity. Subsequently, we added estimates of protein isoelectric points as a feature to ANNs that classify specialized families of proteins, namely major capsid and tail proteins. As expected, these more specialized ANNs are more accurate than the structural ANNs. To experimentally validate the ANN predictions, several ORFs with no significant similarities to known sequences that are ANN-predicted structural proteins were examined by transmission electron microscopy. Some of these self-assembled into structures strongly resembling virion structures. Thus, our ANNs are new tools for identifying phage and potential prophage structural proteins that are difficult or impossible to detect by other bioinformatic analysis. The networks will be valuable when sequence is available but in vitro propagation of the phage may not be practical or possible.

Gene Composer in a structural genomics environment.

The structural genomics effort at the Seattle Structural Genomics Center for Infectious Disease (SSGCID) requires the manipulation of large numbers of amino-acid sequences and the underlying DNA sequences which are to be cloned into expression vectors. To improve efficiency in high-throughput protein structure determination, a database software package, Gene Composer, has been developed which facilitates the information-rich design of protein constructs and their underlying gene sequences. With its modular workflow design and numerous graphical user interfaces, Gene Composer enables researchers to perform all common bioinformatics steps used in modern structure-guided protein engineering and synthetic gene engineering. An example of the structure determination of H1N1 RNA-dependent RNA polymerase PB2 subunit is given.

Gene design, cloning and protein-expression methods for high-value targets at the Seattle Structural Genomics Center for Infectious Disease.

Any structural genomics endeavor, particularly ambitious ones such as the NIAID-funded Seattle Structural Genomics Center for Infectious Disease (SSGCID) and Center for Structural Genomics of Infectious Disease (CSGID), face technical challenges at all points of the production pipeline. One salvage strategy employed by SSGCID is combined gene engineering and structure-guided construct design to overcome challenges at the levels of protein expression and protein crystallization. Multiple constructs of each target are cloned in parallel using Polymerase Incomplete Primer Extension cloning and small-scale expressions of these are rapidly analyzed by capillary electrophoresis. Using the methods reported here, which have proven particularly useful for high-value targets, otherwise intractable targets can be resolved.

Gene composer: database software for protein construct design, codon engineering, and gene synthesis.

BACKGROUND: To improve efficiency in high throughput protein structure determination, we have developed a database software package, Gene Composer, which facilitates the information-rich design of protein constructs and their codon engineered synthetic gene sequences. With its modular workflow design and numerous graphical user interfaces, Gene Composer enables researchers to perform all common bio-informatics steps used in modern structure guided protein engineering and synthetic gene engineering. RESULTS: An interactive Alignment Viewer allows the researcher to simultaneously visualize sequence conservation in the context of known protein secondary structure, ligand contacts, water contacts, crystal contacts, B-factors, solvent accessible area, residue property type and several other useful property views. The Construct Design Module enables the facile design of novel protein constructs with altered N- and C-termini, internal insertions or deletions, point mutations, and desired affinity tags. The modifications can be combined and permuted into multiple protein constructs, and then virtually cloned in silico into defined expression vectors. The Gene Design Module uses a protein-to-gene algorithm that automates the back-translation of a protein amino acid sequence into a codon engineered nucleic acid gene sequence according to a selected codon usage table with minimal codon usage threshold, defined G:C% content, and desired sequence features achieved through synonymous codon selection that is optimized for the intended expression system. The gene-to-oligo algorithm of the Gene Design Module plans out all of the required overlapping oligonucleotides and mutagenic primers needed to synthesize the desired gene constructs by PCR, and for physically cloning them into selected vectors by the most popular subcloning strategies. CONCLUSION: We present a complete description of Gene Composer functionality, and an efficient PCR-based synthetic gene assembly procedure with mis-match specific endonuclease error correction in combination with PIPE cloning. In a sister manuscript we present data on how Gene Composer designed genes and protein constructs can result in improved protein production for structural studies.

Combined protein construct and synthetic gene engineering for heterologous protein expression and crystallization using Gene Composer.

BACKGROUND: With the goal of improving yield and success rates of heterologous protein production for structural studies we have developed the database and algorithm software package Gene Composer. This freely available electronic tool facilitates the information-rich design of protein constructs and their engineered synthetic gene sequences, as detailed in the accompanying manuscript. RESULTS: In this report, we compare heterologous protein expression levels from native sequences to that of codon engineered synthetic gene constructs designed by Gene Composer. A test set of proteins including a human kinase (P38alpha), viral polymerase (HCV NS5B), and bacterial structural protein (FtsZ) were expressed in both E. coli and a cell-free wheat germ translation system. We also compare the protein expression levels in E. coli for a set of 11 different proteins with greatly varied G:C content and codon bias. CONCLUSION: The results consistently demonstrate that protein yields from codon engineered Gene Composer designs are as good as or better than those achieved from the synonymous native genes. Moreover, structure guided N- and C-terminal deletion constructs designed with the aid of Gene Composer can lead to greater success in gene to structure work as exemplified by the X-ray crystallographic structure determination of FtsZ from Bacillus subtilis. These results validate the Gene Composer algorithms, and suggest that using a combination of synthetic gene and protein construct engineering tools can improve the economics of gene to structure research.

Structures of aspartate aminotransferases from Trypanosoma brucei, Leishmania major and Giardia lamblia.

The structures of three aspartate aminotransferases (AATs) from eukaryotic pathogens were solved within the Seattle Structural Genomics Center for Infectious Disease (SSGCID). Both the open and closed conformations of AAT were observed. Pyridoxal phosphate was bound to the active site via a Schiff base to a conserved lysine. An active-site mutant showed that Trypanosoma brucei AAT still binds pyridoxal phosphate even in the absence of the tethering lysine. The structures highlight the challenges for the structure-based design of inhibitors targeting the active site, while showing options for inhibitor design targeting the N-terminal arm.

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.

Structure of the Reston ebolavirus VP30 C-terminal domain.

The ebolaviruses can cause severe hemorrhagic fever. Essential to the ebolavirus life cycle is the protein VP30, which serves as a transcriptional cofactor. Here, the crystal structure of the C-terminal, NP-binding domain of VP30 from Reston ebolavirus is presented. Reston VP30 and Ebola VP30 both form homodimers, but the dimeric interfaces are rotated relative to each other, suggesting subtle inherent differences or flexibility in the dimeric interface.

Tyrosine aminotransferase from Leishmania infantum: A new drug target candidate.

Leishmania infantum is the etiological agent of zoonotic visceral leishmaniasis in the Mediterranean basin. The disease is fatal without treatment, which has been based on antimonial pentavalents for more than 60 years. Due to resistances, relapses and toxicity to current treatment, the development of new drugs is required. The structure of the L. infantum tyrosine aminotransferase (LiTAT) has been recently solved showing important differences with the mammalian orthologue. The characterization of LiTAT is reported herein. This enzyme is cytoplasmic and is over-expressed in the more infective stages and nitric oxide resistant parasites. Unlike the mammalian TAT, LiTAT is able to use ketomethiobutyrate as co-substrate. The pharmacophore model of LiTAT with this specific co-substrate is described herein. This may allow the identification of new inhibitors present in the databases. All the data obtained support that LiTAT is a good target candidate for the development of new anti-leishmanial drugs.

Structural analysis of H1N1 and H7N9 influenza A virus PA in the absence of PB1.

Influenza A viruses cause the respiratory illness influenza, which can be mild to fatal depending on the strain and host immune response. The flu polymerase acidic (PA), polymerase basic 1 (PB1), and polymerase basic 2 (PB2) proteins comprise the RNA-dependent RNA polymerase complex responsible for viral genome replication. The first crystal structures of the C-terminal domain of PA (PA-CTD) in the absence of PB1-derived peptides show a number of structural changes relative to the previously reported PB1-peptide bound structures. The human A/WSN/1933 (H1N1) and avian A/Anhui1/2013 (H7N9) strain PA-CTD proteins exhibit the same global topology as other strains in the absence of PB1, but differ extensively in the PB1 binding pocket including a widening of the binding groove and the unfolding of a ß-turn. Both PA-CTD proteins exhibited a significant increase in thermal stability in the presence of either a PB1-derived peptide or a previously reported inhibitor in differential scanning fluorimetry assays. These structural changes demonstrate plasticity in the PA-PB1 binding interface which may be exploited in the development of novel therapeutics.

Structure and function of a cyanophage-encoded peptide deformylase.

Bacteriophages encode auxiliary metabolic genes that support more efficient phage replication. For example, cyanophages carry several genes to maintain host photosynthesis throughout infection, shuttling the energy and reducing power generated away from carbon fixation and into anabolic pathways. Photodamage to the D1/D2 proteins at the core of photosystem II necessitates their continual replacement. Synthesis of functional proteins in bacteria requires co-translational removal of the N-terminal formyl group by a peptide deformylase (PDF). Analysis of marine metagenomes to identify phage-encoded homologs of known metabolic genes found that marine phages carry PDF genes, suggesting that their expression during infection might benefit phage replication. We identified a PDF homolog in the genome of Synechococcus cyanophage S-SSM7. Sequence analysis confirmed that it possesses the three absolutely conserved motifs that form the active site in PDF metalloproteases. Phylogenetic analysis placed it within the Type 1B subclass, most closely related to the Arabidopsis chloroplast PDF, but lacking the C-terminal α-helix characteristic of that group. PDF proteins from this phage and from Synechococcus elongatus were expressed and characterized. The phage PDF is the more active enzyme and deformylates the N-terminal tetrapeptides from D1 proteins more efficiently than those from ribosomal proteins. Solution of the X-ray/crystal structures of those two PDFs to 1.95 Å resolution revealed active sites identical to that of the Type 1B Arabidopsis chloroplast PDF. Taken together, these findings show that many cyanophages encode a PDF with a D1 substrate preference that adds to the repertoire of genes used by phages to maintain photosynthetic activities.

High-throughput limited proteolysis/mass spectrometry for protein domain elucidation.

High-resolution structural information is important for improving our understanding of protein function in vitro and in vivo and providing information to enable drug discovery. The process leading to X-ray structure determination is often time consuming and labor intensive. It requires informed decisions in expression construct design, expression host selection, and strategies for protein purification, crystallization and structure determination. Previously published studies have demonstrated that compact globular domains defined by limited proteolysis represent good candidates for production of diffraction quality crystals [1-7]. Integration of mass spectrometry and proteolysis experiments can provide accurate definition of domain boundaries at unprecedented rates. We have conducted a critical evaluation of this approach with 400 target proteins produced by SGX (Structural GenomiX, Inc.) for the New York Structural GenomiX Research Consortium (NYSGXRC; http://www.nysgxrc.org) under the auspices of the National Institute of General Medical Sciences Protein Structure Initiative (http://www.nigms.nih.gov/psi). The objectives of this study were to develop parallel/automated protocols for proteolytic digestion and data acquisition for multiple proteins, and to carry out a systematic study to correlate domain definition via proteolysis with outcomes of crystallization and structure determination attempts. Initial results from this work demonstrate that proteins yielding diffraction quality crystals are typically resistant to proteolysis. Large-scale sub cloning and subsequent testing of expression, solubility, and crystallizability of proteolytically defined truncations is currently underway.

Impact of the deltaF508 mutation in first nucleotide-binding domain of human cystic fibrosis transmembrane conductance regulator on domain folding and structure.

Cystic fibrosis is caused by defects in the cystic fibrosis transmembrane conductance regulator (CFTR), commonly the deletion of residue Phe-508 (DeltaF508) in the first nucleotide-binding domain (NBD1), which results in a severe reduction in the population of functional channels at the epithelial cell surface. Previous studies employing incomplete NBD1 domains have attributed this to aberrant folding of DeltaF508 NBD1. We report structural and biophysical studies on complete human NBD1 domains, which fail to demonstrate significant changes of in vitro stability or folding kinetics in the presence or absence of the DeltaF508 mutation. Crystal structures show minimal changes in protein conformation but substantial changes in local surface topography at the site of the mutation, which is located in the region of NBD1 believed to interact with the first membrane spanning domain of CFTR. These results raise the possibility that the primary effect of DeltaF508 is a disruption of proper interdomain interactions at this site in CFTR rather than interference with the folding of NBD1. Interestingly, increases in the stability of NBD1 constructs are observed upon introduction of second-site mutations that suppress the trafficking defect caused by the DeltaF508 mutation, suggesting that these suppressors might function indirectly by improving the folding efficiency of NBD1 in the context of the full-length protein. The human NBD1 structures also solidify the understanding of CFTR regulation by showing that its two protein segments that can be phosphorylated both adopt multiple conformations that modulate access to the ATPase active site and functional interdomain interfaces.

New York-Structural GenomiX Research Consortium (NYSGXRC): a large scale center for the protein structure initiative.

Structural GenomiX, Inc. (SGX), four New York area institutions, and two University of California schools have formed the New York Structural GenomiX Research Consortium (NYSGXRC), an industrial/academic Research Consortium that exploits individual core competencies to support all aspects of the NIH-NIGMS funded Protein Structure Initiative (PSI), including protein family classification and target selection, generation of protein for biophysical analyses, sample preparation for structural studies, structure determination and analyses, and dissemination of results. At the end of the PSI Pilot Study Phase (PSI-1), the NYSGXRC will be capable of producing 100-200 experimentally determined protein structures annually. All Consortium activities can be scaled to increase production capacity significantly during the Production Phase of the PSI (PSI-2). The Consortium utilizes both centralized and de-centralized production teams with clearly defined deliverables and hand-off procedures that are supported by a web-based target/sample tracking system (SGX Laboratory Information Data Management System, LIMS, and NYSGXRC Internal Consortium Experimental Database, ICE-DB). Consortium management is provided by an Executive Committee, which is composed of the PI and all Co-PIs. Progress to date is tracked on a publicly available Consortium web site (http://www.nysgxrc.org) and all DNA/protein reagents and experimental protocols are distributed freely from the New York City Area institutions. In addition to meeting the requirements of the Pilot Study Phase and preparing for the Production Phase of the PSI, the NYSGXRC aims to develop modular technologies that are transferable to structural biology laboratories in both academe and industry. The NYSGXRC PI and Co-PIs intend the PSI to have a transforming effect on the disciplines of X-ray crystallography and NMR spectroscopy of biological macromolecules. Working with other PSI-funded Centers, the NYSGXRC seeks to create the structural biology laboratory of the future. Herein, we present an overview of the organization of the NYSGXRC and describe progress toward development of a high-throughput Gene-->Structure platform. An analysis of current and projected consortium metrics reflects progress to date and delineates opportunities for further technology development.

Structure of nucleotide-binding domain 1 of the cystic fibrosis transmembrane conductance regulator.

Cystic fibrosis transmembrane conductance regulator (CFTR) is an ATP-binding cassette (ABC) transporter that functions as a chloride channel. Nucleotide-binding domain 1 (NBD1), one of two ABC domains in CFTR, also contains sites for the predominant CF-causing mutation and, potentially, for regulatory phosphorylation. We have determined crystal structures for mouse NBD1 in unliganded, ADP- and ATP-bound states, with and without phosphorylation. This NBD1 differs from typical ABC domains in having added regulatory segments, a foreshortened subdomain interconnection, and an unusual nucleotide conformation. Moreover, isolated NBD1 has undetectable ATPase activity and its structure is essentially the same independent of ligand state. Phe508, which is commonly deleted in CF, is exposed at a putative NBD1-transmembrane interface. Our results are consistent with a CFTR mechanism, whereby channel gating occurs through ATP binding in an NBD1-NBD2 nucleotide sandwich that forms upon displacement of NBD1 regulatory segments.