RNA structure and folding

Drosophila germ granules enrich mRNAs critical for fly development. Within germ granules, mRNAs form multi-transcript clusters marked by increased mRNA concentration, creating an elevated potential for intermolecular base pairing. However, the type and abundance of intermolecular base pairing in mRNA clusters is poorly characterized. Using single-molecule super-resolution microscopy, chemical probing for base accessibility, phase separation assays, and simulations, we demonstrated that mRNAs remain well-folded upon localization to germ granules. While most base pairing is intramolecular, mRNAs still display the ability for intermolecular base pairing, facilitating clustering without high sequence complementarity or significant melting of secondary structure. This base pairing among mRNAs is driven by scattered and discontinuous stretches of bases appearing on the surface of folded RNAs, providing multivalency to clustering but exhibits low probability for sustained interactions. Notably, engineered germ granule mRNAs with exposed GC-rich complementary sequences (CSs) presented within stable stem loops induce sustained base pairing in vitro and enhanced intermolecular interactions in vivo. However, the presence of these stem loops alone disrupts fly development, and the addition of GC-rich CSs exacerbates this phenotype. Although germ granule mRNAs contain numerous GC-rich CSs capable of stable intermolecular base pairing, they are primarily embedded by RNA folding. This study emphasizes the role of RNA folding in controlling the type and abundance of intermolecular base pairing, thereby preserving the functional integrity of mRNAs within the germ granules.

Interplay between divalent cations (Mg2+ and Ca2+) and single-stranded DNA (ssDNA) and double-stranded DNA (dsDNA), as well as stacking interactions, is important in nucleosome stability and phase separation in nucleic acids. Quantitative techniques accounting for ion–DNA interactions are needed to obtain insights into these and related problems. Toward this end, we created a sequence-dependent computational TIS-ION model that explicitly accounts for monovalent and divalent ions. Simulations of the rigid 24 base-pair (bp) dsDNA and flexible ssDNA sequences, dT30 and dA30, with varying amounts of the divalent cations show that the calculated excess number of ions around the dsDNA and ssDNA agree quantitatively with ion-counting experiments. Using an ensemble of all-atom structures generated from coarse-grained simulations, we calculated the small-angle X-ray scattering profiles, which are in excellent agreement with experiments. Although ion-counting experiments mask the differences between Mg2+ and Ca2+, we find that Mg2+ binds to the minor grooves and phosphate groups, whereas Ca2+ binds specifically to the minor groove. Both Mg2+ and Ca2+ exhibit a tendency to bind to the minor groove of DNA as opposed to the major groove. The dA30 conformations are dominated by stacking interactions, resulting in structures with considerable helical order. The near cancellation of the favorable stacking and unfavorable electrostatic interactions leads to dT30 populating an ensemble of heterogeneous conformations. The successful applications of the TIS-ION model are poised to confront many problems in DNA biophysics.

Repeat RNA sequences self-associate to form condensates. Simulations of a coarse-grained single-interaction site model for (CAG)n (n = 30 and 31) show that the salt-dependent free energy gap, ΔGS, between the ground (perfect hairpin) and the excited state (slipped hairpin (SH) with one CAG overhang) of the monomer for (n even) is the primary factor that determines the rates and yield of self-assembly. For odd n, the free energy (GS) of the ground state, which is an SH, is used to predict the self-association kinetics. As the monovalent salt concentration, CS, increases, ΔGS and GS increase, which decreases the rates of dimer formation. In contrast, ΔGS for shuffled sequences, with the same length and sequence composition as (CAG)31, is larger, which suppresses their propensities to aggregate. Although demonstrated explicitly for (CAG) polymers, the finding of inverse correlation between the free energy gap and RNA aggregation is general.

Low-complexity nucleotide repeat sequences, which are implicated in several neurological disorders, undergo liquid–liquid phase separation (LLPS) provided the number of repeat units, n, exceeds a critical value. Here, we establish a link between the folding landscapes of the monomers of trinucleotide repeats and their propensity to self-associate. Simulations using a coarse-grained Self-Organized Polymer (SOP) model for (CAG)n repeats in monovalent salt solutions reproduce experimentally measured melting temperatures, which are available only for small n. By extending the simulations to large n, we show that the free-energy gap, ΔGS, between the ground state (GS) and slipped hairpin (SH) states is a predictor of aggregation propensity. The GS for even n is a perfect hairpin (PH), whereas it is a SH when n is odd. The value of ΔGS (zero for odd n) is larger for even n than for odd n. As a result, the rate of dimer formation is slower in (CAG)30 relative to (CAG)31, thus linking ΔGS to RNA–RNA association. The yield of the dimer decreases dramatically, compared to the wild type, in mutant sequences in which the population of the SH decreases substantially. Association between RNA chains is preceded by a transition to the SH even if the GS is a PH. The finding that the excitation spectrum—which depends on the exact sequence, n, and ionic conditions—is a predictor of self-association should also hold for other RNAs (mRNA for example) that undergo LLPS.

Divalent cations are often required to fold RNA, which is a highly charged polyanion. Condensation of ions, such as Mg2+ or Ca2+, in the vicinity of RNA renormalizes the effective charges on the phosphate groups, thus minimizing the intra RNA electrostatic repulsion. The prevailing view is that divalent ions bind diffusively in a nonspecific manner. In sharp contrast, we arrive at the exact opposite conclusion using a theory for the interaction of ions with the phosphate groups using RISM theory in conjunction with simulations based on an accurate three-interaction-site RNA model. The divalent ions bind in a nucleotide-specific manner using either the inner (partially dehydrated) or outer (fully hydrated) shell coordination. The high charge density Mg2+ ion has a preference to bind to the outer shell, whereas the opposite is the case for Ca2+. Surprisingly, we find that bridging interactions, involving ions that are coordinated to two or more phosphate groups, play a crucial role in maintaining the integrity of the folded state. Their importance could become increasingly prominent as the size of the RNA increases. Because the modes of interaction of divalent ions with DNA are likely to be similar, we propose that specific inner and outer shell coordination could play a role in DNA condensation, and perhaps genome organization as well.

RNA molecules cannot fold in the absence of counterions. Experiments are typically performed in the presence of monovalent and divalent cations. How to treat the impact of a solution containing a mixture of both ion types on RNA folding has remained a challenging problem for decades. By exploiting the large concentration difference between divalent and monovalent ions used in experiments, we develop a theory based on the reference interaction site model (RISM), which allows us to treat divalent cations explicitly while keeping the implicit screening effect due to monovalent ions. Our theory captures both the inner shell and outer shell coordination of divalent cations to phosphate groups, which we demonstrate is crucial for an accurate calculation of RNA folding thermodynamics. The RISM theory for ion–phosphate interactions when combined with simulations based on a transferable coarse-grained model allows us to predict accurately the folding of several RNA molecules in a mixture containing monovalent and divalent ions. The calculated folding free energies and ion-preferential coefficients for RNA molecules (pseudoknots, a fragment of the rRNA, and the aptamer domain of the adenine riboswitch) are in excellent agreement with experiments over a wide range of monovalent and divalent ion concentrations. Because the theory is general, it can be readily used to investigate ion and sequence effects on DNA properties.