Binding Site Identification and Flexible Docking of Single Stranded RNA to Proteins Using a Fragment-Based Approach

Protein-RNA docking is hampered by the high flexibility of RNA, and particularly single-stranded RNA (ssRNA). Yet, ssRNA regions typically carry the specificity of protein recognition. The lack of methodology for modeling such regions limits the accuracy of current protein-RNA docking methods. We developed a fragment-based approach to model protein-bound ssRNA, based on the structure of the protein and the sequence of the RNA, without any prior knowledge of the RNA binding site or the RNA structure. The conformational diversity of each fragment is sampled by an exhaustive RNA fragment library that was created from all the existing experimental structures of protein-ssRNA complexes. A systematic and detailed analysis of fragment-based ssRNA docking was performed which constitutes a proof-of-principle for the fragment-based approach. The method was tested on two 8-homo-nucleotide ssRNA-protein complexes and was able to identify the binding site on the protein within 10 Å. Moreover, a structure of each bound ssRNA could be generated in close agreement with the crystal structure with a mean deviation of ~1.5 Å except for a terminal nucleotide. This is the first time a bound ssRNA could be modeled from sequence with high precision.     

Pushing the limits of what is achievable in protein–DNA docking: benchmarking HADDOCK’s performance

The intrinsic flexibility of DNA and the difficulty of identifying its interaction surface have long been challenges that prevented the development of efficient protein–DNA docking methods. We have demonstrated the ability our flexible data-driven docking method HADDOCK to deal with these before, by using custom-built DNA structural models. Here we put our method to the test on a set of 47 complexes from the protein–DNA docking benchmark. We show that HADDOCK is able to predict many of the specific DNA conformational changes required to assemble the interface(s). Our DNA analysis and modelling procedure captures the bend and twist motions occurring upon complex formation and uses these to generate custom-built DNA structural models, more closely resembling the bound form, for use in a second docking round. We achieve throughout the benchmark an overall success rate of 94% of one-star solutions or higher (interface root mean square deviation ≤4 Å and fraction of native contacts >10%) according to CAPRI criteria. Our improved protocol successfully predicts even the challenging protein–DNA complexes in the benchmark. Finally, our method is the first to readily dock multiple molecules (N > 2) simultaneously, pushing the limits of what is currently achievable in the field of protein–DNA docking.     

Accuracy of Protein-Protein Binding Sites in High-Throughput Template-Based Modeling

The accuracy of protein structures, particularly their binding sites, is essential for the success of modeling protein complexes. Computationally inexpensive methodology is required for genome-wide modeling of such structures. For systematic evaluation of potential accuracy in high-throughput modeling of binding sites, a statistical analysis of target-template sequence alignments was performed for a representative set of protein complexes. For most of the complexes, alignments containing all residues of the interface were found. The full interface alignments were obtained even in the case of poor alignments where a relatively small part of the target sequence (as low as 40%) aligned to the template sequence, with a low overall alignment identity (<30%). Although such poor overall alignments might be considered inadequate for modeling of whole proteins, the alignment of the interfaces was strong enough for docking. In the set of homology models built on these alignments, one third of those ranked 1 by a simple sequence identity criteria had RMSD<5 Å, the accuracy suitable for low-resolution template free docking. Such models corresponded to multi-domain target proteins, whereas for single-domain proteins the best models had 5 Å<RMSD<10 Å, the accuracy suitable for less sensitive structure-alignment methods. Overall, ∼50% of complexes with the interfaces modeled by high-throughput techniques had accuracy suitable for meaningful docking experiments. This percentage will grow with the increasing availability of co-crystallized protein-protein complexes.     

Fabrication and characterization of RNA aptamer microarrays for the study of protein-aptamer interactions with SPR imaging

RNA microarrays were created on chemically modified gold surfaces using a novel surface ligation methodology and employed in a series of surface plasmon resonance imaging (SPRI) measurements of DNA–RNA hybridization and RNA aptamer–protein binding. Various unmodified single-stranded RNA (ssRNA) oligonucleotides were ligated onto identical 5′-phosphate-terminated ssDNA microarray elements with a T4 RNA ligase surface reaction. A combination of ex situ polarization modulation FTIR measurements of the RNA monolayer and in situ SPRI measurements of DNA hybridization adsorption onto the surface were used to determine an ssRNA surface density of 4.0 × 10¹² molecules/cm² and a surface ligation efficiency of 85 ± 10%. The surface ligation methodology was then used to create a five-component RNA microarray of potential aptamers for the protein factor IXa (fIXa). The relative surface coverages of the different aptamers were determined through a novel enzymatic method that employed SPRI measurements of a surface RNase H hydrolysis reaction. SPRI measurements were then used to correctly identify the best aptamer to fIXa, which was previously determined from SELEX measurements. A Langmuir adsorption coefficient of 1.6 × 10⁷ M⁻¹ was determined for fIXa adsorption to this aptamer. Single-base variations from this sequence were shown to completely destroy the aptamer–fIXa binding interaction.     

DNA phosphate crowding correlates with protein cationic side chain density and helical curvature in protein/DNA crystal structures

Sequence-specific binding of proteins to their DNA targets involves a complex spectrum of processes that often induce DNA conformational variation in the bound complex. The forces imposed by protein binding that cause the helical deformations are intimately interrelated and difficult to parse or rank in importance. To investigate the role of electrostatics in helical deformation, we quantified the relationship between protein cationic residue density (Cpc) and DNA phosphate crowding (Cpp). The correlation between Cpc and Cpp was then calculated for a subset of 58 high resolution protein–DNA crystal structures. Those structures containing strong Cpc/Cpp correlation (>±0.25) were likely to contain DNA helical curvature. Further, the correlation factor sign predicted the direction of helical curvature with positive (16 structures) and negative (seven structures) correlation containing concave (DNA curved toward protein) and convex (DNA curved away from protein) curvature, respectively. Protein–DNA complexes without significant Cpc/Cpp (36 structures) correlation (-0.25<0<0.25) tended to contain DNA without significant curvature. Interestingly, concave and convex complexes also include more arginine and lysine phosphate contacts, respectively, whereas linear complexes included essentially equivalent numbers of Lys/Arg phosphate contacts. Together, these findings suggest an important role for electrostatic interactions in protein–DNA complexes involving helical curvature.     

The ends of a large RNA molecule are necessarily close

We show on general theoretical grounds that the two ends of single-stranded (ss) RNA molecules (consisting of roughly equal proportions of A, C, G and U) are necessarily close together, largely independent of their length and sequence. This is demonstrated to be a direct consequence of two generic properties of the equilibrium secondary structures, namely that the average proportion of bases in pairs is ∼60% and that the average duplex length is ∼4. Based on mfold and Vienna computations on large numbers of ssRNAs of various lengths (1000–10 000 nt) and sequences (both random and biological), we find that the 5′–3′ distance—defined as the sum of H-bond and covalent (ss) links separating the ends of the RNA chain—is small, averaging 15–20 for each set of viral sequences tested. For random sequences this distance is ∼12, consistent with the theory. We discuss the relevance of these results to evolved sequence complementarity and specific protein binding effects that are known to be important for keeping the two ends of viral and messenger RNAs in close proximity. Finally we speculate on how our conclusions imply indistinguishability in size and shape of equilibrated forms of linear and covalently circularized ssRNA molecules.     

Ligand Docking to Intermediate and Close-To-Bound Conformers Generated by an Elastic Network Model Based Algorithm for Highly Flexible Proteins

Incorporating receptor flexibility in small ligand-protein docking still poses a challenge for proteins undergoing large conformational changes. In the absence of bound structures, sampling conformers that are accessible by apo state may facilitate docking and drug design studies. For this aim, we developed an unbiased conformational search algorithm, by integrating global modes from elastic network model, clustering and energy minimization with implicit solvation. Our dataset consists of five diverse proteins with apo to complex RMSDs 4.7–15 Å. Applying this iterative algorithm on apo structures, conformers close to the bound-state (RMSD 1.4–3.8 Å), as well as the intermediate states were generated. Dockings to a sequence of conformers consisting of a closed structure and its “parents” up to the apo were performed to compare binding poses on different states of the receptor. For two periplasmic binding proteins and biotin carboxylase that exhibit hinge-type closure of two dynamics domains, the best pose was obtained for the conformer closest to the bound structure (ligand RMSDs 1.5–2 Å). In contrast, the best pose for adenylate kinase corresponded to an intermediate state with partially closed LID domain and open NMP domain, in line with recent studies (ligand RMSD 2.9 Å). The docking of a helical peptide to calmodulin was the most challenging case due to the complexity of its 15 Å transition, for which a two-stage procedure was necessary. The technique was first applied on the extended calmodulin to generate intermediate conformers; then peptide docking and a second generation stage on the complex were performed, which in turn yielded a final peptide RMSD of 2.9 Å. Our algorithm is effective in producing conformational states based on the apo state. This study underlines the importance of such intermediate states for ligand docking to proteins undergoing large transitions.     

Primer-dependent synthesis by poliovirus RNA-dependent RNA polymerase (3Dpol)

Properties of poliovirus RNA-dependent RNA polymerase (3D^pol) including optimal conditions for primer extension, processivity and the rate of dissociation from primer-template (k_off) were examined in the presence and absence of viral protein 3AB. Primer-dependent polymerization was examined on templates of 407 or 1499 nt primed such that fully extended products would be 296 or 1388 nt, respectively. Maximal primer extension was achieved with low rNTP concentrations (50–100 µM) using pH 7 and low (<1 mM) MgCl₂ and KCl (<20 mM) concentrations. However, high activity (about half maximal) was also observed with 500 µM rNTPs providing that higher MgCl₂ levels (3–5 mM) were used. The enhancement observed with the former conditions appeared to result from a large increase in the initial level or active enzyme that associated with the primer. 3AB increased the number of extended primers at all conditions with no apparent change in processivity. The k_off values for the polymerase bound to primer-template were 0.011 ± 0.005 and 0.037 ± 0.006 min^–1 (average of four or more experiments ± SD) in the presence or absence of 3AB, respectively. The decrease in the presence of 3AB suggested an enhancement of polymerase binding or stability. However, binding was tight even without 3AB, consistent with the highly processive (at least several hundred nucleotides) nature of 3D^pol. The results support a mechanism whereby 3AB enhances the ability of 3D^pol to form a productive complex with the primer-template. Once formed, this complex is very stable resulting in highly processive synthesis.     

Dead-End Elimination with a Polarizable Force Field Repacks PCNA Structures

A balance of van der Waals, electrostatic, and hydrophobic forces drive the folding and packing of protein side chains. Although such interactions between residues are often approximated as being pairwise additive, in reality, higher-order many-body contributions that depend on environment drive hydrophobic collapse and cooperative electrostatics. Beginning from dead-end elimination, we derive the first algorithm, to our knowledge, capable of deterministic global repacking of side chains compatible with many-body energy functions. The approach is applied to seven PCNA x-ray crystallographic data sets with resolutions 2.5–3.8 Å (mean 3.0 Å) using an open-source software. While PDB_REDO models average an R_free value of 29.5% and MOLPROBITY score of 2.71 Å (77th percentile), dead-end elimination with the polarizable AMOEBA force field lowered R_free by 2.8–26.7% and improved mean MOLPROBITY score to atomic resolution at 1.25 Å (100th percentile). For structural biology applications that depend on side-chain repacking, including x-ray refinement, homology modeling, and protein design, the accuracy limitations of pairwise additivity can now be eliminated via polarizable or quantum mechanical potentials.     

Kinetics of formation of hypoxanthine containing base pairs by HIV-RT: RNA template effects on the base substitution frequencies

Hypoxanthine (H), the deamination product of adenine, has been implicated in the high frequency of A to G transitions observed in retroviral and other RNA genomes. Although H·C base pairs are thermodynamically more stable than other H·N pairs, polymerase selection may be determined in part by kinetic factors. Therefore, the hypoxanthine induced substitution pattern resulting from replication by viral polymerases may be more complex than that predicted from thermodynamics. We have examined the steady-state kinetics of formation of base pairs opposite template H in RNA by HIV-RT, and for the incorporation of dITP during first- and second-strand synthesis. Hypoxanthine in an RNA template enhances the k_2app for pairing with standard dNTPs by factors of 10–1000 relative to adenine at the same sequence position. The order of base pairing preferences for H in RNA was observed to be H·C >> H·T > H·A > H·G. Steady-state kinetics of insertion for all possible mispairs formed with dITP were examined on RNA and DNA templates of identical sequence. Insertion of dITP opposite all bases occurs 2–20 times more frequently on RNA templates. This bias for higher insertion frequencies on RNA relative to DNA templates is also observed for formation of mispairs at template A. This kinetic advantage afforded by RNA templates for mismatches and pairing involving H suggests a higher induction of mutations at adenines during first-strand synthesis by HIV-RT.     

Modelling ion binding to AA platform motifs in RNA: a continuum solvent study including conformational adaptation

Binding of monovalent and divalent cations to two adenine–adenine platform structures from the Tetrahymena group I intron ribozyme has been studied using continuum solvent models based on the generalised Born and the finite-difference Poisson–Boltzmann approaches. The adenine–adenine platform RNA motif forms an experimentally characterised monovalent ion binding site important for ribozyme folding and function. Qualitative agreement between calculated and experimental ion placements and binding selectivity was obtained. The inclusion of solvation effects turned out to be important to obtain low energy structures and ion binding placements in agreement with the experiment. The calculations indicate that differences in solvation of the isolated ions contribute to the calculated ion binding preference. However, Coulomb attraction and van der Waals interactions due to ion size differences and RNA conformational adaptation also influence the calculated ion binding affinity. The calculated alkali ion binding selectivity for both platforms followed the order K⁺ > Na⁺ > Rb⁺ > Cs⁺ > Li⁺ (Eisenman series VI) in the case of allowing RNA conformational relaxation during docking. With rigid RNA an Eisenman series V was obtained (K⁺ > Rb⁺ > Na⁺ > Cs⁺ > Li⁺). Systematic energy minimisation docking simulations starting from several hundred initial placements of potassium ions on the surface of platform containing RNA fragments identified a coordination geometry in agreement with the experiment as the lowest energy binding site. The approach could be helpful to identify putative ion binding sites in nucleic acid structures determined at low resolution or with experimental methods that do not allow identification of ion binding sites.     

DOCK 6: Combining techniques to model RNA–small molecule complexes

With an increasing interest in RNA therapeutics and for targeting RNA to treat disease, there is a need for the tools used in protein-based drug design, particularly DOCKing algorithms, to be extended or adapted for nucleic acids. Here, we have compiled a test set of RNA–ligand complexes to validate the ability of the DOCK suite of programs to successfully recreate experimentally determined binding poses. With the optimized parameters and a minimal scoring function, 70% of the test set with less than seven rotatable ligand bonds and 26% of the test set with less than 13 rotatable bonds can be successfully recreated within 2 Å heavy-atom RMSD. When DOCKed conformations are rescored with the implicit solvent models AMBER generalized Born with solvent-accessible surface area (GB/SA) and Poisson–Boltzmann with solvent-accessible surface area (PB/SA) in combination with explicit water molecules and sodium counterions, the success rate increases to 80% with PB/SA for less than seven rotatable bonds and 58% with AMBER GB/SA and 47% with PB/SA for less than 13 rotatable bonds. These results indicate that DOCK can indeed be useful for structure-based drug design aimed at RNA. Our studies also suggest that RNA-directed ligands often differ from typical protein–ligand complexes in their electrostatic properties, but these differences can be accommodated through the choice of potential function. In addition, in the course of the study, we explore a variety of newly added DOCK functions, demonstrating the ease with which new functions can be added to address new scientific questions.     

Anatomy of noncovalent interactions between the nucleobases or ribose and π-containing amino acids in RNA–protein complexes

A set of >300 nonredundant high-resolution RNA–protein complexes were rigorously searched for π-contacts between an amino acid side chain (W, H, F, Y, R, E and D) and an RNA nucleobase (denoted π–π interaction) or ribose moiety (denoted sugar–π). The resulting dataset of >1500 RNA–protein π-contacts were visually inspected and classified based on the interaction type, and amino acids and RNA components involved. More than 80% of structures searched contained at least one RNA–protein π-interaction, with π–π contacts making up 59% of the identified interactions. RNA–protein π–π and sugar–π contacts exhibit a range in the RNA and protein components involved, relative monomer orientations and quantum mechanically predicted binding energies. Interestingly, π–π and sugar–π interactions occur more frequently with RNA (4.8 contacts/structure) than DNA (2.6). Moreover, the maximum stability is greater for RNA–protein contacts than DNA–protein interactions. In addition to highlighting distinct differences between RNA and DNA–protein binding, this work has generated the largest dataset of RNA–protein π-interactions to date, thereby underscoring that RNA–protein π-contacts are ubiquitous in nature, and key to the stability and function of RNA–protein complexes.     

Prevalent RNA recognition motif duplication in the human genome

The sequence-specific recognition of RNA by proteins is mediated through various RNA binding domains, with the RNA recognition motif (RRM) being the most frequent and present in >50% of RNA-binding proteins (RBPs). Many RBPs contain multiple RRMs, and it is unclear how each RRM contributes to the binding specificity of the entire protein. We found that RRMs within the same RBP (i.e., sibling RRMs) tend to have significantly higher similarity than expected by chance. Sibling RRM pairs from RBPs shared by multiple species tend to have lower similarity than those found only in a single species, suggesting that multiple RRMs within the same protein might arise from domain duplication followed by divergence through random mutations. This finding is exemplified by a recent RRM domain duplication in DAZ proteins and an ancient duplication in PABP proteins. Additionally, we found that different similarities between sibling RRMs are associated with distinct functions of an RBP and that the RBPs tend to contain repetitive sequences with low complexity. Taken together, this study suggests that the number of RBPs with multiple RRMs has expanded in mammals and that the multiple sibling RRMs may recognize similar target motifs in a cooperative manner.