Three-dimensional RNA structure refinement by hydroxyl radical probing

Molecular modeling guided by experimentally derived structural information is an attractive approach for three-dimensional structure determination of complex RNAs that are not amenable to study by high-resolution methods. Hydroxyl radical probing (HRP), which is performed routinely in many laboratories, provides a measure of solvent accessibility at individual nucleotides. HRP measurements have, to date, only been used to evaluate RNA models qualitatively. Here we report the development of a quantitative structure refinement approach using HRP measurements to drive discrete molecular dynamics simulations for RNAs ranging in size from 80 to 230 nucleotides. We first used HRP reactivities to identify RNAs that form extensive helical packing interactions. For these RNAs, we achieved highly significant structure predictions given the inputs of RNA sequence and base pairing. This HRP-directed tertiary structure refinement approach generates robust structural hypotheses that are useful for guiding explorations of structure-function inter-relationships in RNA.     

Applying an empirical hydropathic forcefield in refinement may improve low-resolution protein X-ray crystal structures

Background

The quality of X-ray crystallographic models for biomacromolecules refined from data obtained at high-resolution is assured by the data itself. However, at low-resolution, >3.0 Å, additional information is supplied by a forcefield coupled with an associated refinement protocol. These resulting structures are often of lower quality and thus unsuitable for downstream activities like structure-based drug discovery.

Methodology

An X-ray crystallography refinement protocol that enhances standard methodology by incorporating energy terms from the HINT (Hydropathic INTeractions) empirical forcefield is described. This protocol was tested by refining synthetic low-resolution structural data derived from 25 diverse high-resolution structures, and referencing the resulting models to these structures. The models were also evaluated with global structural quality metrics, e.g., Ramachandran score and MolProbity clashscore. Three additional structures, for which only low-resolution data are available, were also re-refined with this methodology.

Results

The enhanced refinement protocol is most beneficial for reflection data at resolutions of 3.0 Å or worse. At the low-resolution limit, ≥4.0 Å, the new protocol generated models with Cα positions that have RMSDs that are 0.18 Å more similar to the reference high-resolution structure, Ramachandran scores improved by 13%, and clashscores improved by 51%, all in comparison to models generated with the standard refinement protocol. The hydropathic forcefield terms are at least as effective as Coulombic electrostatic terms in maintaining polar interaction networks, and significantly more effective in maintaining hydrophobic networks, as synthetic resolution is decremented. Even at resolutions ≥4.0 Å, these latter networks are generally native-like, as measured with a hydropathic interactions scoring tool.     

Comparative photocross-linking analysis of the tertiary structures of Escherichia coli and Bacillus subtilis RNase P RNAs.

Bacterial ribonuclease P contains a catalytic RNA subunit that cleaves precursor sequences from the 5' ends of pre-tRNAs. The RNase P RNAs from Bacillus subtilis and Escherichia coli each contain several unique secondary structural elements not present in the other. To understand better how these phylogenetically variable elements affect the global architecture of the ribozyme, photoaffinity cross-linking studies were carried out. Photolysis of photoagents attached at homologous sites in the two RNAs results in nearly identical cross-linking patterns, consistent with the homology of the RNAs and indicating that these RNAs contain a common, core tertiary structure. Distance constraints were used to derive tertiary structure models using a molecular mechanics-based modeling protocol. The resulting superimposition of large sets of equivalent models provides a low resolution (5-10 A) structure for each RNA. Comparison of these structure models shows that the conserved core helices occupy similar positions in space. Variably present helical elements that may play a role in global structural stability are found at the periphery of the core structure. The P5.1 and P15.1 helical elements, unique to the B.subtilis RNase P RNA, and the P6/16/17 helices, unique to the E.coli RNA, occupy similar positions in the structure models and, therefore, may have analogous structural function.     

Molecular modeling studies on the ribosome

In the absence of a high resolution crystal structure for the ribosome, numerous research groups are carrying out low resolution structural studies using neutron diffraction, electron microscopy, fluorescence energy transfer, chemical crosslinking, chemical footprinting studies, and other methods. We have developed a computer-based refinement method for incorporating these data into low resolution three-dimensional models. The method is based on a molecular mechanics approach, with proteins represented by spherical particles of suitable diameter and the ribosomal RNA represented by a string of spherical pseudoatoms, one for each nucleotide. Experimental data are used to derive constraints that are introduced through a special force field (potential function). Models are refined by simulated annealing. Since every term in the force field is quadratic, any model that satisfies all of the input data has an energy of zero; higher energies indicate residual unsatisfied constraints. The residual energy provides a quantitative statement of model quality and can be used to identify conflicts in the experimental data. The method has been applied to the refinement of a low resolution model for the 30S subunit (the small subunit) of theE. coli ribosome. Since this is a very underdetermined system, the range of acceptable models has also been explored. This provides an estimate of the resolution of the structure, which is about 15 Å overall, with the uncertainty in position of individual nucleotides ranging from about 5 Å to 50 Å.     

Conjoined Use of EM and NMR in RNA Structure Refinement

More than 40% of the RNA structures have been determined using nuclear magnetic resonance (NMR) technique. NMR mainly provides local structural information of protons and works most effectively on relatively small biomacromolecules. Hence structural characterization of large RNAs can be difficult for NMR alone. Electron microscopy (EM) provides global shape information of macromolecules at nanometer resolution, which should be complementary to NMR for RNA structure determination. Here we developed a new energy term in Xplor-NIH against the density map obtained by EM. We conjointly used NMR and map restraints for the structure refinement of three RNA systems — U2/U6 small-nuclear RNA, genome-packing motif (Ψ^CD)₂ from Moloney murine leukemia virus, and ribosome-binding element from turnip crinkle virus. In all three systems, we showed that the incorporation of a map restraint, either experimental or generated from known PDB structure, greatly improves structural precision and accuracy. Importantly, our method does not rely on an initial model assembled from RNA duplexes, and allows full torsional freedom for each nucleotide in the torsion angle simulated annealing refinement. As increasing number of macromolecules can be characterized by both NMR and EM, the marriage between the two techniques would enable better characterization of RNA three-dimensional structures.     

Relative orientation of RNA helices in a group 1 ribozyme determined by helix extension electron microscopy.

The relative orientation of helical elements in a folded RNA molecule provides key information about its three-dimensional architecture. We have developed a method that involves extending peripheral helices of an RNA, mounting for electron microscopy in the absence of protein and measuring interhelical angles. As a control, extended anticodon and acceptor stems of tRNA(Phe) were found to form a 92 +/- 20 degrees angle, consistent with the X-ray structure. Single, double and triple extensions (50-80 bp) of helical elements P2.1, P6b and P8 of the Tetrahymena group I ribozyme did not alter its catalytic activity. The measured angle between P6b and P8 is consistent with the Michel-Westhof structural model, while the P2.1-P6b and P2.1-P8 angles allow P2.1 to be positioned in the model. The angle distributions of the ribozyme are broader than those of the tRNA, which may reflect the dynamics of the RNA. Helix extension allows low-resolution electron microscopy to provide much higher resolution information about the disposition of helical elements in RNA. It should be applicable to diverse RNAs and ribonucleoprotein complexes.     

Computer modeling 16 S ribosomal RNA

A three-dimensional structure for 16 S RNA has been produced with a computer protocol that is not dependent on human intervention. This protocol improves upon traditional modeling techniques by using distance geometry to fold the molecule in an objective and reproducible fashion. The method is based on the secondary structure of RNA and treats the molecule as a set of double-stranded helices that are linked by flexible single-strands of variable length. Data derived from chemical cross-linking studies of 16 S RNA and tertiary phylogenetic relationships provide the constraints used to fold the molecule into a compact three-dimensional form. Possibly subjective evaluation of the input data are transformed into verifiable quantitative parameters. Relationships based on general locations within the 30 S subunit or on protein-RNA interactions have been specifically excluded. The resolution of the model exceeds that of electron micrographs and approaches that obtained in preliminary X-ray crystal structures. The model size of 245 x 190 x 140 A is compatible with that of the 30 S subunit as determined by electron microscopy. The volume of the model is 1.87 x 10(6) A which is similar to that of the small subunit in a preliminary X-ray crystal structure. The radius of gyration of the model structure of 76 A is intermediate to that seen for partially denatured and fully folded 16 S RNA. Computer graphics are used to display the results in a manner that maximizes the opportunities for human visual interpretation of the models. A format for displaying the structures has been developed that will make it possible for researchers who have not devoted themselves to ribosomal modeling to comprehend and make use of the information that the models embody. On this basis the computer-generated models are compared with models developed by other researchers and with structural data not included in the folding parameter data set.     

Correlating SHAPE signatures with three-dimensional RNA structures

Selective 2'-hydroxyl acylation analyzed by primer extension (SHAPE) is a facile technique for quantitative analysis of RNA secondary structure. In general, low SHAPE signal values indicate Watson-Crick base-pairing, and high values indicate positions that are single-stranded within the RNA structure. However, the relationship of SHAPE signals to structural properties such as non-Watson-Crick base-pairing or stacking has thus far not been thoroughly investigated. Here, we present results of SHAPE experiments performed on several RNAs with published three-dimensional structures. This strategy allows us to analyze the results in terms of correlations between chemical reactivities and structural properties of the respective nucleotide, such as different types of base-pairing, stacking, and phosphate-backbone interactions. We find that the RNA SHAPE signal is strongly correlated with cis-Watson-Crick/Watson-Crick base-pairing and is to a remarkable degree not dependent on other structural properties with the exception of stacking. We subsequently generated probabilistic models that estimate the likelihood that a residue with a given SHAPE score participates in base-pairing. We show that several models that take SHAPE scores of adjacent residues into account perform better in predicting base-pairing compared with individual SHAPE scores. This underscores the context sensitivity of SHAPE and provides a framework for an improved interpretation of the response of RNA to chemical modification.     

Modeling the Structure of RNA Molecules with Small-Angle X-Ray Scattering Data

We propose a novel fragment assembly method for low-resolution modeling of RNA and show how it may be used along with small-angle X-ray solution scattering (SAXS) data to model low-resolution structures of particles having as many as 12 independent secondary structure elements. We assessed this model-building procedure by using both artificial data on a previously proposed benchmark and publicly available data. With the artificial data, SAXS-guided models show better similarity to native structures than ROSETTA decoys. The publicly available data showed that SAXS-guided models can be used to reinterpret RNA structures previously deposited in the Protein Data Bank. Our approach allows for fast and efficient building of de novo models of RNA using approximate secondary structures that can be readily obtained from existing bioinformatic approaches. We also offer a rigorous assessment of the resolving power of SAXS in the case of small RNA structures, along with a small multimetric benchmark of the proposed method.     

Analysis of the tertiary structure of the ribonuclease P ribozyme-substrate complex by site-specific photoaffinity crosslinking.

Bacterial ribonuclease P (RNase P), an endonuclease involved in tRNA maturation, is a ribonucleoprotein containing a catalytic RNA. The secondary structure of this ribozyme is well-established, and a low-resolution model of the three-dimensional structure of the ribozyme-substrate complex has been proposed based on site-specific crosslinking and phylogenetic comparative data [Harris ME et al., 1994 EMBO J 13:3953-3963]. However, several substructures of that model were poorly constrained by the available data. In the present analysis, additional constraints between elements within the Escherichia coli RNase P RNA-pre-tRNA complex were determined by intra- and intermolecular crosslinking experiments. Circularly permuted RNase P RNAs were used to position an azidophenacyl photoactive crosslinking agent specifically at strategic sites within the ribozyme-substrate complex. Crosslink sites were mapped by primer extension and confirmed by analysis of the mobility of the crosslinked RNA lariats on denaturing acrylamide gels relative to circular and linear RNA standards. Crosslinked species generally retained significant catalytic activity, indicating that the results reflect the native ribozyme structure. The crosslinking results support the general configuration of the structure model and predicate new positions and orientations for helices that were previously poorly constrained by the data set. The expanded library of crosslinking constraints was used, together with secondary and tertiary structure identified by phylogenetic sequence comparisons, to refine significantly the model of RNase P RNA with bound substrate pre-tRNA. The crosslinking results and data from chemical-modification and mutational studies are discussed in the context of the current structural perspective on this ribozyme.     

Protein structure determination using a database of interatomic distance probabilities

The accelerated pace of genomic sequencing has increased the demand for structural models of gene products. Improved quantitative methods are needed to study the many systems (e.g., macromolecular assemblies) for which data are scarce. Here, we describe a new molecular dynamics method for protein structure determination and molecular modeling. An energy function, or database potential, is derived from distributions of interatomic distances obtained from a database of known structures. X-ray crystal structures are refined by molecular dynamics with the new energy function replacing the Van der Waals potential. Compared to standard methods, this method improved the atomic positions, interatomic distances, and side-chain dihedral angles of structures randomized to mimic the early stages of refinement. The greatest enhancement in side-chain placement was observed for groups that are characteristically buried. More accurate calculated model phases will follow from improved interatomic distances. Details usually seen only in high-resolution refinements were improved, as is shown by an R-factor analysis. The improvements were greatest when refinements were carried out using X-ray data truncated at 3.5 A. The database potential should therefore be a valuable tool for determining X-ray structures, especially when only low-resolution data are available.     

Protein structure prediction using sparse NOE and RDC restraints with Rosetta in CASP13

Computational methods that produce accurate protein structure models from limited experimental data, for example, from nuclear magnetic resonance (NMR) spectroscopy, hold great potential for biomedical research. The NMR-assisted modeling challenge in CASP13 provided a blind test to explore the capabilities and limitations of current modeling techniques in leveraging NMR data which had high sparsity, ambiguity, and error rate for protein structure prediction. We describe our approach to predict the structure of these proteins leveraging the Rosetta software suite. Protein structure models were predicted de novo using a two-stage protocol. First, low-resolution models were generated with the Rosetta de novo method guided by nonambiguous nuclear Overhauser effect (NOE) contacts and residual dipolar coupling (RDC) restraints. Second, iterative model hybridization and fragment insertion with the Rosetta comparative modeling method was used to refine and regularize models guided by all ambiguous and nonambiguous NOE contacts and RDCs. Nine out of 16 of the Rosetta de novo models had the correct fold (global distance test total score > 45) and in three cases high-resolution models were achieved (root-mean-square deviation < 3.5 å). We also show that a meta-approach applying iterative Rosetta + NMR refinement on server-predicted models which employed non-NMR-contacts and structural templates leads to substantial improvement in model quality. Integrating these data-assisted refinement strategies with innovative non-data-assisted approaches which became possible in CASP13 such as high precision contact prediction will in the near future enable structure determination for large proteins that are outside of the realm of conventional NMR.     

Automated real-space refinement of protein structures using a realistic backbone move set

Crystals of many important biological macromolecules diffract to limited resolution, rendering accurate model building and refinement difficult and time-consuming. We present a torsional optimization protocol that is applicable to many such situations and combines Protein Data Bank-based torsional optimization with real-space refinement against the electron density derived from crystallography or cryo-electron microscopy. Our method converts moderate- to low-resolution structures at initial (e.g., backbone trace only) or late stages of refinement to structures with increased numbers of hydrogen bonds, improved crystallographic R-factors, and superior backbone geometry. This automated method is applicable to DNA-binding and membrane proteins of any size and will aid studies of structural biology by improving model quality and saving considerable effort. The method can be extended to improve NMR and other structures. Our backbone score and its sequence profile provide an additional standard tool for evaluating structural quality.     

Use of structure comparison methods for the refinement of protein structure predictions. I. Identifying the structural family of a protein from low-resolution models.

Predicting the three-dimensional structure of proteins is still one of the most challenging problems in molecular biology. Despite its difficulty, several investigators have started to produce consistently low-resolution predictions for small proteins. However, in most of these cases, the prediction accuracy is still too low to make them useful. In the present article, we address the problem of obtaining better-quality predictions, starting from low-resolution models. To this end, we have devised a new procedure that uses these models, together with structure comparison methods, to identify the structural family of the target protein. This would allow, in a second step not described in the present work, to refine the predictions using conserved features of the identified family. In our approach, the structure database is investigated using predictions, at different accuracy levels, for a given protein. As query structures, we used both low-resolution versions of the native structures, as well as different sets of low accuracy predictions. In general, we found that for predictions with a resolution of > or =5-7 A, structure comparison methods were able to identify the fold of a protein in the top positions.     

Elastic network models capture the motions apparent within ensembles of RNA structures

The role of structure and dynamics in mechanisms for RNA becomes increasingly important. Computational approaches using simple dynamics models have been successful at predicting the motions of proteins and are often applied to ribonucleo-protein complexes but have not been thoroughly tested for well-packed nucleic acid structures. In order to characterize a true set of motions, we investigate the apparent motions from 16 ensembles of experimentally determined RNA structures. These indicate a relatively limited set of motions that are captured by a small set of principal components (PCs). These limited motions closely resemble the motions computed from low frequency normal modes from elastic network models (ENMs), either at atomic or coarse-grained resolution. Various ENM model types, parameters, and structure representations are tested here against the experimental RNA structural ensembles, exposing differences between models for proteins and for folded RNAs. Differences in performance are seen, depending on the structure alignment algorithm used to generate PCs, modulating the apparent utility of ENMs but not significantly impacting their ability to generate functional motions. The loss of dynamical information upon coarse-graining is somewhat larger for RNAs than for globular proteins, indicating, perhaps, the lower cooperativity of the less densely packed RNA. However, the RNA structures show less sensitivity to the elastic network model parameters than do proteins. These findings further demonstrate the utility of ENMs and the appropriateness of their application to well-packed RNA-only structures, justifying their use for studying the dynamics of ribonucleo-proteins, such as the ribosome and regulatory RNAs.     

DMS footprinting of structured RNAs and RNA-protein complexes

We describe a protocol in which dimethyl sulfate (DMS) modification of the base-pairing faces of unpaired adenosine and cytidine nucleotides is used for structural analysis of RNAs and RNA-protein complexes (RNPs). The protocol is optimized for RNAs of small to moderate size (< or = 500 nt). The RNA or RNP is first exposed to DMS under conditions that promote formation of the folded structure or complex, as well as 'control' conditions that do not allow folding or complex formation. The positions and extents of modification are then determined by primer extension, polyacrylamide gel electrophoresis and quantitative analysis. From changes in the extent of modification upon folding or protein binding (appearance of a 'footprint'), it is possible to detect local changes in the secondary and tertiary structure of RNA, as well as the formation of RNA-protein contacts. This protocol takes 1.5-3 d to complete, depending on the type of analysis used.     

3Drefine: Consistent protein structure refinement by optimizing hydrogen bonding network and atomic‐level energy minimization

One of the major limitations of computational protein structure prediction is the deviation of predicted models from their experimentally derived true, native structures. The limitations often hinder the possibility of applying computational protein structure prediction methods in biochemical assignment and drug design that are very sensitive to structural details. Refinement of these low‐resolution predicted models to high‐resolution structures close to the native state, however, has proven to be extremely challenging. Thus, protein structure refinement remains a largely unsolved problem. Critical assessment of techniques for protein structure prediction (CASP) specifically indicated that most predictors participating in the refinement category still did not consistently improve model quality. Here, we propose a two‐step refinement protocol, called 3Drefine, to consistently bring the initial model closer to the native structure. The first step is based on optimization of hydrogen bonding (HB) network and the second step applies atomic‐level energy minimization on the optimized model using a composite physics and knowledge‐based force fields. The approach has been evaluated on the CASP benchmark data and it exhibits consistent improvement over the initial structure in both global and local structural quality measures. 3Drefine method is also computationally inexpensive, consuming only few minutes of CPU time to refine a protein of typical length (300 residues). 3Drefine web server is freely available at http://sysbio.rnet.missouri.edu/3Drefine/ . Proteins 2013. © 2012 Wiley Periodicals, Inc.     

Computational methods for defining the allowed conformational space of 16S rRNA based on chemical footprinting data.

Structural models for 16S ribosomal RNA have been proposed based on combinations of crosslinking, chemical protection, shape, and phylogenetic evidence. These models have been based for the most part on independent data sets and different sets of modeling assumptions. In order to evaluate such models meaningfully, methods are required to explicitly model the spatial certainty with which individual structural components are positioned by specific data sets. In this report, we use a constraint satisfaction algorithm to explicitly assess the location of the secondary structural elements of the 16S RNA, as well as the certainty with which these elements can be positioned. The algorithm initially assumes that these helical elements can occupy any position and orientation and then systematically eliminates those positions and orientations that do not satisfy formally parameterized interpretations of structural constraints. Using a conservative interpretation of the hydroxyl radical footprinting data, the positions of the ribosomal proteins as defined by neutron diffraction studies, and the secondary structure of 16S rRNA, the location of the RNA secondary structural elements can be defined with an average precision of 25 A (ranging from 12.8 to 56.3 A). The uncertainty in individual helix positions is both heterogeneous and dependent upon the number of constraints imposed on the helix. The topology of the resulting model is consistent with previous models based on independent approaches. The result of our computation is a conservative upper bound on the possible positions of the RNA secondary structural elements allowed by this data set, and provides a suitable starting point for refinement with other sources of data or different sets of modeling assumptions.     

Determining RNA three-dimensional structures using low-resolution data

Knowing the 3-D structure of an RNA is fundamental to understand its biological function. Nowadays X-ray crystallography and NMR spectroscopy are systematically applied to newly discovered RNAs. However, the application of these high-resolution techniques is not always possible, and thus scientists must turn to lower resolution alternatives. Here, we introduce a pipeline to systematically generate atomic resolution 3-D structures that are consistent with low-resolution data sets. We compare and evaluate the discriminative power of a number of low-resolution experimental techniques to reproduce the structure of the Escherichia coli tRNA(VAL) and P4-P6 domain of the Tetrahymena thermophila group I intron. We test single and combinations of the most accessible low-resolution techniques, i.e. hydroxyl radical footprinting (OH), methidiumpropyl-EDTA (MPE), multiplexed hydroxyl radical cleavage (MOHCA), and small-angle X-ray scattering (SAXS). We show that OH-derived constraints are accurate to discriminate structures at the atomic level, whereas EDTA-based constraints apply to global shape determination. We provide a guide for choosing which experimental techniques or combination of thereof is best in which context. The pipeline represents an important step towards high-throughput low-resolution RNA structure determination.