X-ray crystallography is a critical tool in the study of biological systems. It is able to provide information that has been a prerequisite to understanding the fundamentals of life. It is also a method that is central to the development of new therapeutics for human disease. Significant time and effort are required to determine and optimize many macromolecular structures because of the need for manual interpretation of complex numerical data, often using many different software packages, and the repeated use of interactive three-dimensional graphics. The Phenix software package has been developed to provide a comprehensive system for macromolecular crystallographic structure solution with an emphasis on automation. This has required the development of new algorithms that minimize or eliminate subjective input in favour of built-in expert-systems knowledge, the automation of procedures that are traditionally performed by hand, and the development of a computational framework that allows a tight integration between the algorithms. The application of automated methods is particularly appropriate in the field of structural proteomics, where high throughput is desired. Features in Phenix for the automation of experimental phasing with subsequent model building, molecular replacement, structure refinement and validation are described and examples given of running Phenix from both the command line and graphical user interface.

A density-based procedure is described for improving a homology model that is locally accurate but differs globally. The model is deformed to match the map and refined, yielding an improved starting point for density modification and further model-building.

An approach is presented for addressing the challenge of model rebuilding after molecular replacement in cases where the placed template is very different from the structure to be determined. The approach takes advantage of the observation that a template and target structure may have local structures that can be superimposed much more closely than can their complete structures. A density-guided procedure for deformation of a properly placed template is introduced. A shift in the coordinates of each residue in the structure is calculated based on optimizing the match of model density within a 6 Å radius of the center of that residue with a prime-and-switch electron-density map. The shifts are smoothed and applied to the atoms in each residue, leading to local deformation of the template that improves the match of map and model. The model is then refined to improve the geometry and the fit of model to the structure-factor data. A new map is then calculated and the process is repeated until convergence. The procedure can extend the routine applicability of automated molecular replacement, model building and refinement to search models with over 2 Å r.m.s.d. representing 65–100% of the structure.

Approximately one-third of mankind has been exposed to Mycobacterium tuberculosis, the etiological agent responsible for tuberculosis (TB). As part of an effort to develop a new generation of anti-TB agents, the chemical shifts for the 261-residue, virulence-associated protein Rv0577 from M. tuberculosis has been extensively assigned.

With over 60,000 protein structures available in the Protein Data Bank, it is frequently possible use one of them to obtain starting phase information and to solve new crystal structures. Molecular replacement1–4 procedures, which search for placements of a starting model within the crystallographic unit cell that best account for the measured diffraction amplitudes, followed by automatic chain tracing methods5–8, have allowed the rapid solution of large numbers of protein structures. Despite extensive work9–14, molecular replacement or the subsequent rebuilding usually fail with more divergent starting models based on remote homologues with less than 30% sequence identity. Here we show that this limitation can be substantially reduced by combining algorithms for protein structure modeling with those developed for crystallographic structure determination. An approach integrating Rosetta structure modeling with Autobuild chain tracing yielded high-resolution structures for 8 of 13 X-ray diffraction datasets that could not be solved in the laboratories of expert crystallographers and that remained unsolved after application of an extensive array of alternative approaches. We estimate the new method should allow rapid structure determination without experimental phase information for over half the cases where current methods fail, given diffraction datasets of better than 3.2Å resolution, four or fewer copies in the asymmetric unit, and the availability of structures of homologous proteins with >20% sequence identity.

Background

In scientific computing, Fortran was the dominant implementation language throughout most of the second part of the 20th century. The many tools accumulated during this time have been difficult to integrate with modern software, which is now dominated by object-oriented languages.

Results

Driven by the requirements of a large-scale scientific software project, we have developed a Fortran to C++ source-to-source conversion tool named FABLE. This enables the continued development of new methods even while switching languages. We report the application of FABLE in three major projects and present detailed comparisons of Fortran and C++ runtime performances.

Conclusions

Our experience suggests that most Fortran 77 codes can be converted with an effort that is minor (measured in days) compared to the original development time (often measured in years). With FABLE it is possible to reuse and evolve legacy work in modern object-oriented environments, in a portable and maintainable way. FABLE is available under a nonrestrictive open source license. In FABLE the analysis of the Fortran sources is separated from the generation of the C++ sources. Therefore parts of FABLE could be reused for other target languages.

The foundations and current features of a widely used graphical user interface for macromolecular crystallography are described.

A new Python-based graphical user interface for the PHENIX suite of crystallography software is described. This interface unifies the command-line programs and their graphical displays, simplifying the development of new interfaces and avoiding duplication of function. With careful design, graphical interfaces can be displayed automatically, instead of being manually constructed. The resulting package is easily maintained and extended as new programs are added or modified.

phenix.refine is a program within the PHENIX package that supports crystallographic structure refinement against experimental data with a wide range of upper resolution limits using a large repertoire of model parameterizations. This paper presents an overview of the major phenix.refine features, with extensive literature references for readers interested in more detailed discussions of the methods.

phenix.refine is a program within the PHENIX package that supports crystallographic structure refinement against experimental data with a wide range of upper resolution limits using a large repertoire of model parameterizations. It has several automation features and is also highly flexible. Several hundred parameters enable extensive customizations for complex use cases. Multiple user-defined refinement strategies can be applied to specific parts of the model in a single refinement run. An intuitive graphical user interface is available to guide novice users and to assist advanced users in managing refinement projects. X-ray or neutron diffraction data can be used separately or jointly in refinement. phenix.refine is tightly integrated into the PHENIX suite, where it serves as a critical component in automated model building, final structure refinement, structure validation and deposition to the wwPDB. This paper presents an overview of the major phenix.refine features, with extensive literature references for readers interested in more detailed discussions of the methods.

DEN refinement and automated model building with AutoBuild were used to determine the structure of a putative succinyl-diaminopimelate desuccinylase from C. glutamicum. This difficult case of molecular-replacement phasing shows that the synergism between DEN refinement and AutoBuild outperforms standard refinement protocols.

Phasing by molecular replacement remains difficult for targets that are far from the search model or in situations where the crystal diffracts only weakly or to low resolution. Here, the process of determining and refining the structure of Cgl1109, a putative succinyl-diaminopimelate desuccinylase from Corynebacterium glutamicum, at ∼3 Å resolution is described using a combination of homology modeling with MODELLER, molecular-replacement phasing with Phaser, deformable elastic network (DEN) refinement and automated model building using AutoBuild in a semi-automated fashion, followed by final refinement cycles with phenix.refine and Coot. This difficult molecular-replacement case illustrates the power of including DEN restraints derived from a starting model to guide the movements of the model during refinement. The resulting improved model phases provide better starting points for automated model building and produce more significant difference peaks in anomalous difference Fourier maps to locate anomalous scatterers than does standard refinement. This example also illustrates a current limitation of automated procedures that require manual adjustment of local sequence misalignments between the homology model and the target sequence.

The combination of algorithms from the structure-modeling field with those of crystallographic structure determination can broaden the range of templates that are useful for structure determination by the method of molecular replacement. Automated tools in phenix.mr_rosetta simplify the application of these combined approaches by integrating Phenix crystallographic algorithms and Rosetta structure-modeling algorithms and by systematically generating and evaluating models with a combination of these methods. The phenix.mr_rosetta algorithms can be used to automatically determine challenging structures. The approaches used in phenix.mr_rosetta are described along with examples that show roles that structure-modeling can play in molecular replacement.

The first structure for a member of the DUF3349 (PF11829) family of proteins, Rv0543c from Mycobacterium tuberculosis, has been determined using NMR-based methods and some of its biophysical properties characterized. Rv0543c is a 100 residue, 11.3 kDa protein that both size exclusion chromatography and NMR spectroscopy show to be a monomer in solution. The structure of the protein consists of a bundle of five α-helices α1 (M1 - Y16), α2 (P21 - C33), α3 (S37 - G52), α4 (G58 - H65) and α5 (S72 - G87) held together by a largely conserved group of hydrophobic amino acid side chains. Heteronuclear steady-state {1H}-15N NOE, T1, and T2 values are similar through-out the sequence indicating that the backbones of the five helices are in a single motional regime. The thermal stability of Rv0543c, characterized by circular dichroism spectroscopy, indicates that Rv0543c irreversibly unfolds upon heating with an estimated melting temperature of 62.5°C. While the biological function of Rv0543c is still unknown, the presence of DUF3349 proteins predominately in Mycobacterium and Rhodococcus bacterial species suggests that Rv0543 may have a biological function unique to these bacteria, and consequently, may prove to be an attractive drug target to combat tuberculosis.

Ligands interacting with Mycobacterium tuberculosis recombinant proteins were identified through use of the ability of Cibacron Blue F3GA dye to interact with nucleoside/nucleotide binding proteins, and the effects of these ligands on crystallization were examined. Co-crystallization with ligands enhanced crystallization and enabled X-ray diffraction data to be collected to a resolution of at least 2.7 Å for 5 of 10 proteins tested. Additionally, clues about individual proteins’ functions were obtained from their interactions with each of a panel of ligands.

Electronic supplementary material

The online version of this article (doi:10.1007/s10969-012-9124-8) contains supplementary material, which is available to authorized users.

Exploring the function and 3D space of large multidomain protein targets often requires sophisticated experimentation to obtain the targets in a form suitable for structure determination. Screening methods capable of selecting well-expressed, soluble fragments from DNA libraries exist, but require the use of automation to maximize chances of picking a few good candidates. Here, we describe the use of an insertion dihydrofolate reductase (DHFR) vector to select in-frame fragments and a split-GFP assay technology to filter-out constructs that express insoluble protein fragments. With the incorporation of an IPCR step to create high density, focused sublibraries of fragments, this cost-effective method can be performed manually with no a priori knowledge of domain boundaries while permitting single amino acid resolution boundary mapping. We used it on the well-characterized p85α subunit of the phosphoinositide-3-kinase to demonstrate the robustness and efficiency of our methodology. We then successfully tested it onto the polyketide synthase PpsC from Mycobacterium tuberculosis, a potential drug target involved in the biosynthesis of complex lipids in the cell envelope. X-ray quality crystals from the acyl-transferase (AT), dehydratase (DH) and enoyl-reductase (ER) domains have been obtained.

Crystal and solution structures of Rv1848 protein and their implications in the biological assembly of Mtb urease is presented.

The crystal structure of the urease γ subunit (UreA) from Mycobacterium tuberculosis, Rv1848, has been determined at 1.8 Å resolution. The asymmetric unit contains three copies of Rv1848 arranged into a homotrimer that is similar to the UreA trimer in the structure of urease from Klebsiella aerogenes. Small-angle X-ray scattering experiments indicate that the Rv1848 protein also forms trimers in solution. The observed homotrimer and the organization of urease genes within the M. tuberculosis genome suggest that M. tuberculosis urease has the (αβγ)3 composition observed for other bacterial ureases. The γ subunit may be of primary importance for the formation of the urease quaternary structure.

Summary

The Mycobacterium tuberculosis protein Rv2377c (71 residues, MW = 8.4 kDa) has been characterized using nuclear magnetic resonance (NMR) and circular dichroism (CD) spectroscopy. Rv2377c was the first identified member of the MbtH-like family of proteins. MbtH-like proteins have been implicated in siderophore biosynthesis, however, their precise biochemical function remain unknown. Size exclusion chromatography and NMR spectroscopy show that Rv2377c is a monomer in solution. Circular dichroism spectroscopy indicates that Rv2377c unfolds upon heating and will reversibly fold into its native conformation upon cooling. Using NMR-based methods the solution structure of Rv2377c was determined and some of the dynamic properties of the protein studied. The protein contains a three-strand, anti-parallel β-sheet (β3:β1:β2) nestled against one C-terminal α-helix (S44-N55). Weak or absent amide cross peaks in the 1H-15N HSQC spectrum for many of the β1 and β2 residues suggest intermediate motion on the ms to μs timescale at the β1:β2 interface. Amide cross peaks in the 1H-15N HSQC spectrum are absent for all but one residue at the C-terminus (W56 - D71), a region that includes a highly conserved sequence WXDXR, suggesting this region is intrinsically disordered. The latter observation differs with the crystal structure of another MbtH-like protein, PA2412 from Pseudomonas aeruginosa, where a second ordered α-helix was observed at the extreme C-terminus.

The International Conference on Structural Genomics (ICSG 2011, http://www.icsg2011.org), held in Toronto Canada May 10–14, 2011 was a rich and exciting demonstration of how far structural genomics has come. Structural genomics has now matured into a field that includes both structure and the biology that structure enables. This has allowed targeting based on systematic approaches and on known biological importance and allows biochemical studies to be closely tied to structure determination. The wealth of purified proteins, clones, and chemical probes produced by structural genomics groups will enable a vast amount of follow-on research. The technologies, the structures, and the biology that were described at the meeting were at the cutting edge of science. Structural genomics has become a success.

We describe an in vitro colony screen to identify Escherichia coli expressing soluble proteins and stable, assembled multiprotein complexes. Proteins with an N-terminal 6His tag and C-terminal green fluorescent protein (GFP) S11 tag are fluorescently labeled in cells by complementation with a coexpressed GFP 1–10 fragment. After partial colony lysis, the fluorescent soluble proteins or complexes diffuse through a supporting filtration membrane and are captured on Talon® resin metal affinity beads immobilized in agarose. Images of the fluorescent colonies convey total expression and the level of fluorescence bound to the beads indicates how much protein is soluble. Both pieces of information can be used together when selecting clones. After the assay, colonies can be picked and propagated, eliminating the need to make replica plates. We used the method to screen a DNA fragment library of the human protein p85 and preferentially obtained clones expressing the full-length ‘breakpoint cluster region-homology' and NSH2 domains. The assay also distinguished clones expressing stable multi-protein complexes from those that are unstable due to missing subunits. Clones expressing stable, intact heterotrimeric E.coli YheNML complexes were readily identified in libraries dominated by complexes of YheML missing the N subunit.

RuvA, a protein from M. tuberculosis H37Rv involved in recombination, has been cloned, expressed, purified and analysed by X-ray crystallography.

The process of recombinational repair is crucial for maintaining genomic integrity and generating biological diversity. In association with RuvB and RuvC, RuvA plays a central role in processing and resolving Holliday junctions, which are a critical intermediate in homologous recombination. Here, the cloning, purification and structure determination of the RuvA protein from Mycobacterium tuberculosis (MtRuvA) are reported. Analysis of the structure and comparison with other known RuvA proteins reveal an octameric state with conserved subunit–subunit interaction surfaces, indicating the requirement of octamer formation for biological activity. A detailed analysis of plasticity in the RuvA molecules has led to insights into the invariant and variable regions, thus providing a framework for understanding regional flexibility in various aspects of RuvA function.

Summary

Central to crystallographic structure solution is obtaining accurate phases in order to build a molecular model, ultimately followed by refinement of that model to optimize its fit to the experimental diffraction data and prior chemical knowledge. Recent advances in phasing and model refinement and validation algorithms make it possible to arrive at better electron density maps and more accurate models.

A decade of structural genomics, the large-scale determination of protein structures, has generated a wealth of data and many important lessons for structural biology and for future large-scale projects. These lessons include a confirmation that it is possible to construct large-scale facilities that can determine the structures of a hundred or more proteins per year, that these structures can be of high quality, and that these structures can have an important impact. Technology development has played a critical role in structural genomics, the difficulties at each step of determining a structure of a particular protein can be quantified, and validation of technologies is nearly as important as the technologies themselves. Finally, rapid deposition of data in public databases has increased the impact and usefulness of the data and international cooperation has advanced the field and improved data sharing.

Overproduction of soluble and stable proteins for functional and structural studies is a major bottleneck for structural genomics programs and traditional biochemistry laboratories. Many high-payoff proteins that are important in various biological processes are “difficult to handle” as protein reagents in their native form. We have recently made several advances in enabling biochemical technologies for improving protein stability (http://www.lanl.gov/projects/gfp/), allowing stratagems for efficient protein domain trapping, solubility-improving mutations, and finding protein folding partners. In particular split-GFP protein tags are a very powerful tool for detection of stable protein domains. Soluble, stable proteins tagged with the 15 amino acid GFP fragment (amino acids 216–228) can be detected in vivo and in vitro using the engineered GFP 1–10 “detector” fragment (amino acids 1–215). If the small tag is accessible, the detector fragment spontaneously binds resulting in fluorescence. Here, we describe our current and on-going efforts to move this process from the bench (manual sample manipulation) to an automated, high-throughput, liquid-handling platform. We discuss optimization and validation of bacterial culture growth, lysis protocols, protein extraction, and assays of soluble and insoluble protein in multiple 96 well plate format. The optimized liquid-handling protocol can be used for rapid determination of the optimal, compact domains from single ORFS, collections of ORFS, or cDNA libraries.

A method for rapid chain tracing of polypeptide backbones at moderate resolution is presented.

A method for the rapid tracing of polypeptide backbones has been developed. The method creates an approximate chain tracing that is useful for visual evaluation of whether a structure has been solved and for use in scoring the quality of electron-density maps. The essence of the method is to (i) sample candidate Cα positions at spacings of approximately 0.6 Å along ridgelines of high electron density, (ii) list all possible nonapeptides that satisfy simple geometric and density criteria using these candidate Cα positions, (iii) score the nonapeptides and choose the highest scoring ones, and (iv) find the longest chains that can be made by connecting nonamers. An indexing and storage scheme that allows a single calculation of most distances and density values is used to speed up the process. The method was applied to 42 density-modified electron-density maps at resolutions from 1.5 to 3.8 Å. A total of 21 428 residues in these maps were traced in 24 CPU min with an overall r.m.s.d. of 1.61 Å for Cα atoms compared with the known refined structures. The method appears to be suitable for rapid evaluation of electron-density map quality.

A method for rapid model building of β-sheets at moderate resolution is presented.

A method for rapidly building β-sheets into electron-density maps is presented. β-Strands are identified as tubes of high density adjacent to and nearly parallel to other tubes of density. The alignment and direction of each strand are identified from the pattern of high density corresponding to carbonyl and Cβ atoms along the strand averaged over all repeats present in the strand. The β-strands obtained are then assembled into a single atomic model of the β-sheet regions. The method was tested on a set of 42 experimental electron-density maps at resolutions ranging from 1.5 to 3.8 Å. The β-­sheet regions were nearly completely built in all but two cases, the exceptions being one structure at 2.5 Å resolution in which a third of the residues in β-sheets were built and a structure at 3.8 Å in which under 10% were built. The overall average r.m.s.d. of main-chain atoms in the residues built using this method compared with refined models of the structures was 1.5 Å.

A method for rapid model building of α-helices at moderate resolution is presented.

A method for the identification of α-helices in electron-density maps at low resolution followed by interpretation at moderate to high resolution is presented. Rapid identification is achieved at low resolution, where α-helices appear as tubes of density. The positioning and direction of the α-helices is obtained at moderate to high resolution, where the positions of side chains can be seen. The method was tested on a set of 42 experimental electron-density maps at resolutions ranging from 1.5 to 3.8 Å. An average of 63% of the α-helical residues in these proteins were built and an average of 76% of the residues built matched helical residues in the refined models of the proteins. The overall average r.m.s.d. between main-chain atoms in the modeled α-helices and the nearest atom with the same name in the refined models of the proteins was 1.3 Å.

Protein production in Escherichia coli involves high-level expression in a culture, followed by harvesting of the cells and finally their disruption, or lysis, to release the expressed proteins. We compare three high-throughput chemical lysis methods to sonication, using a robotic platform and methodologies developed in our laboratory [1]. Under the same expression conditions, all lysis methods varied in the degree of released soluble proteins. With a set of 96 test proteins, we used our split GFP to quantify the soluble and insoluble protein fractions after lysis. Both the amount of soluble protein and the percentage recovered in the soluble fraction using SoluLyse® were well correlated with sonication. Two other methods, Bugbuster® and lysozyme, did not correlate well with sonication. Considering the effects of lysis methods on protein solubility is especially important when accurate protein solubility measurements are needed, for example, when testing adjuvants, growth media, temperature, or when establishing the effects of truncation or sequence variation on protein stability.

We show that Cibacron Blue F3GA dye resin chromatography can be used to identify ligands that specifically interact with proteins from Mycobacterium tuberculosis, and that the identification of these ligands can facilitate structure determination by enhancing the quality of crystals. Four native Mtb proteins of the aldehyde dehydrogenase (ALDH) family were previously shown to be specifically eluted from a Cibacron Blue F3GA dye resin with nucleosides. In this study we characterized the nucleoside-binding specificity of one of these ALDH isozymes (recombinant Mtb Rv0223c) and compared these biochemical results with co-crystallization experiments with different Rv0223c-nucleoside pairings. We found that the strongly interacting ligands (NAD and NADH) aided formation of high-quality crystals, permitting solution of the first Mtb ALDH (Rv0223c) structure. Other nucleoside ligands (AMP, FAD, adenosine, GTP and NADP) exhibited weaker binding to Rv0223c, and produced co-crystals diffracting to lower resolution. Difference electron density maps based on crystals of Rv0223c with various nucleoside ligands show most share the binding site where the natural ligand NAD binds. From the high degree of similarity of sequence and structure compared to human mitochondrial ALDH-2 (BLAST Z-score = 53.5 and RMSD = 1.5 Å), Rv0223c appears to belong to the ALDH-2 class. An altered oligomerization domain in the Rv0223c structure seems to keep this protein as monomer whereas native human ALDH-2 is a multimer.

Electronic supplementary material

The online version of this article (doi:10.1007/s10969-009-9073-z) contains supplementary material, which is available to authorized users.