Free-Energy Landscape of a Hyperstable RNA Tetraloop
Total Page:16
File Type:pdf, Size:1020Kb
Free-energy landscape of a hyperstable RNA tetraloop Jacob C. Minera,b, Alan A. Chenc, and Angel E. Garcíaa,d,1 aCenter for Nonlinear Studies, Los Alamos National Laboratory, Los Alamos, NM 87545; bDepartment of Biological Sciences, Rensselaer Polytechnic Institute, Troy, NY 12180; cDepartment of Chemistry, University at Albany, State University of New York, Albany, NY 12222; and dDepartment of Physics, Applied Physics, and Astronomy, Rensselaer Polytechnic Institute, Troy, NY 12180 Edited by Michael L. Klein, Temple University, Philadelphia, PA, and approved April 21, 2016 (received for review February 24, 2016) We report the characterization of the energy landscape and the through multiple configurational states (19, 21, 22, 28). This folding/unfolding thermodynamics of a hyperstable RNA tetraloop flexibility makes the GNRA tetraloop a useful target for ex- obtained through high-performance molecular dynamics simula- ploring both the accuracy of RNA force fields in recapitulating tions at microsecond timescales. Sampling of the configurational folded configurations and the heterogeneity of the folded free- landscape is conducted using temperature replica exchange molec- energy landscape. ular dynamics over three isochores at high, ambient, and negative In this report, we characterize the thermodynamics and folding pressures to determine the thermodynamic stability and the free- free-energy landscape for an oligonucleotide containing a GNRA energy landscape of the tetraloop. The simulations reveal reversible tetraloop sequence (r[gcGCAAgc]) through high-performance folding/unfolding transitions of the tetraloop into the canonical simulations and replica exchange molecular dynamics (REMD) A-RNA conformation and the presence of two alternative config- (29). The conformational ensembles are described in terms of urations, including a left-handed Z-RNA conformation and a compact global and local order parameters and reveal the formation of purine Triplet. Increasing hydrostatic pressure shows a stabilizing multiple native and nonnative tetraloop configurations. The in- effect on the A-RNA conformation and a destabilization of the left- teractions of the RNA tetraloop with the solvent environment handed Z-RNA. Our results provide a comprehensive description of are probed through sampling at different hydrostatic pressures, the folded free-energy landscape of a hyperstable RNA tetraloop and its effects on the free-energy landscape are determined by and highlight the significant advances of all-atom molecular dy- studying the structure and stability of the tetraloop on three namics in describing the unbiased folding of a simple RNA second- isochores representative of high, ambient, and negative pres- ary structure motif. sures. These isochores reveal similar folded configurations, each with variations in the folded populations that lend insight into RNA | free-energy landscape | molecular dynamics | tetraloop | pressure the stabilizing effects of pressure and temperature. Changes in hydrostatic pressure modulate the interactions of the RNA he stability of a folded RNA molecule depends on a complex molecule with the solvent and ions. Our simulations describe a Tset of interactions between the RNA with itself and the free-energy landscape primarily dominated by right-handed surrounding environment. Secondary RNA structures fold by a A-RNA stems and a diverse set of loop configurations. The sim- combination of complimentary base pairing, nucleotide stacking, ulations also describe several thermodynamically stable configu- and screening of the anionic phosphate backbone (1), but these rations with noncanonical backbone configurations. The diverse interactions depend heavily on the local solution environment, ensemble of configurations adopted by a simple GNRA tetraloop and the RNA configuration can be either stabilized or destabi- hairpin serves as an example of the ability of RNA molecules to lized by a number of extrinsic factors, including temperature (2), adopt diverse conformations in response to the solution environ- pressure (3), denaturants, and salt concentration (4). Even in ment and in the context of the sequence in larger RNA molecules. simple RNA duplexes, changes in pressure and ion concentra- These conformational changes are crucial to the biological func- tions have been shown to drive the formation of alternative tion of RNA in cells. configurations (5, 6) and change the RNA folded ensemble. Characterizing the effects that these different factors have on the Significance folded free-energy landscape of RNA is crucial for determining the thermodynamic stability of different RNA motifs and the We report a multistate folding free-energy landscape for a probability of transitions between configurations. hyperstable RNA tetraloop obtained from high-performance Small RNA hairpins have been used as a model system for molecular dynamic simulations under multiple temperature and characterizing RNA folding and stability for decades because of pressure conditions. These studies advance previous successes of their small size, their ubiquitous presence in many natural sys- de novo folding of RNA tetraloops and our understanding of tems, and the complexity of their internal interactions (7–11). the folded landscape of basic RNA motifs. In addition to the Hairpins result when a single oligonucleotide strand bends 180° canonical A-RNA conformation, our analysis describes de novo to form an antiparallel duplex through complementary base folding of a left-handed Z-RNA conformation and the formation pairing. The bent, single-strandedloopregioniscapableof of a purine Triplet, each showing differential stabilization as a forming complex hydrogen bond networks (12) or long-range function of temperature and pressure. These results provide interactions with other RNAs (13). Some of the most thermo- important insights into the ability of all-atom RNA force fields to dynamically stable RNA hairpins form loop regions of 4 nt, describe the conformational heterogeneity of the most basic called tetraloops (14), which are highly compact, structurally RNA secondary structures and represent a significant advance- conserved, and typically stabilized by the formation of a network ment in the utility of RNA molecular dynamics for describing BIOPHYSICS AND of hydrogen bonds (15) that decrease the loop conformational RNA thermodynamics. COMPUTATIONAL BIOLOGY entropy. These tetraloop structures have been explored through mul- Author contributions: J.C.M. and A.E.G. designed research; J.C.M. and A.A.C. performed tiple experimental (16–23) and computational (24–27) analyses research; J.C.M. and A.E.G. analyzed data; and J.C.M. and A.E.G. wrote the paper. that have revealed important clues about RNA thermodynamic The authors declare no conflict of interest. stability and relate to the processes that govern the formation This article is a PNAS Direct Submission. and stabilization of basic RNA molecules. Many tetraloops 1To whom correspondence should be addressed. Email: [email protected]. = containing the conserved nucleotide sequence GNRA (N A,C, This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. G,U; R = A,G) possess an inherent global flexibility that manifests 1073/pnas.1603154113/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1603154113 PNAS | June 14, 2016 | vol. 113 | no. 24 | 6665–6670 Downloaded by guest on September 25, 2021 0 Results 3.0 Thermodynamics. Characterization of ensembles sampled by the 3 -1 hairpin-forming tetraloop sequence at various thermodynamic 2.0 states (different temperatures and pressures) is first made by cal- culating the rmsd of the stem-forming residues relative to the 1.0 -2 equivalent residues in the NMR experimental structure (Protein 45 Data Bank ID code 1ZIH). These analyses were conducted across 0 -3 multiple REMD temperature ensembles (Fig. 1A)andrevealthat V2 the GCAA tetraloop stem adopts three general configurations -1.0 -4 in the low-temperature to intermediate temperature ensem- i 1 bles. These configurations are represented by ( ) a sharp peak at -2.0 6 -5 1.5-Å rmsd, (ii) a significantly populated peak centered at 5.1-Å rmsd, and (iii) a third, less-populated peak centered at 4.2-Å rmsd. 2 -3.0 -6 Using these criteria, we refer to the ensemble of configurations that fall within the first peak as the folded state and those within the -3.0 -2.0 -1.0 0 1.0 2.0 3.0 4.0 5.0 second and third peaks as alternative states. V1 The thermodynamic stabilities of these three conformations relative to the unfolded ensemble are determined by calculating Fig. 2. Free-energy landscape of GCAA tetraloop (290 K). Dihedral angle principal components determined from the lowest temperature range (290– their relative populations over the sampled temperature range B C 311.5 K) using frames every 4 ps. The six general configuration states each (Fig. 1 and ). At the lowest temperatures, the folded state show at least one peak with free-energy values of 4 kT or greater. Numbers (which we will show adopts an A-RNA hairpin) shows a non- displayed in the plot correspond to centroid identity (described in the text). cooperative unfolding transition up into the intermediate tem- Peaks containing less than four configurations were omitted for clarity. peratures, whereas conversely, the alternative (nonnative) states Histogram scale is in units of kT. show increasing stability up into the intermediate temperatures. Above 380 K, all three states show unfolding transitions, and beyond 460 K, these