Causes, functions, and therapeutic possibilities of RNA secondary structure ensembles and alternative states

Abstract

RNA secondary structure plays essential roles in encoding RNA regulatory fate and function. Most RNAs populate ensembles of alternatively paired states and are continually unfolded and refolded by cellular processes. Measuring these structural ensembles and their contributions to cellular function has traditionally posed major challenges, but new methods and conceptual frameworks are beginning to fill this void. In this review, we provide a mechanism- and function-centric compendium of the roles of RNA secondary structural ensembles and minority states in regulating the RNA life cycle, from transcription to degradation. We further explore how dysregulation of RNA structural ensembles contributes to human disease and discuss the potential of drugging alternative RNA states to therapeutically modulate RNA activity. The emerging paradigm of RNA structural ensembles as central to RNA function provides a foundation for a deeper understanding of RNA biology and new therapeutic possibilities.

Figure 1:

Causes of RNA secondary structure heterogeneity A. Representative RNA free energy landscape. Biologically accessible alternative states occur as local minima in the energy landscape. The relative population of each state is determined by its energy, and the rate of exchange is controlled by the height of the energy barrier. Cells can redistribute RNA ensembles by stabilizing or destabilizing certain states (dashed line). B. Cellular processes that redistribute RNA structural ensembles. During transcription of nascent RNA or after alternative splicing, the unavailability of downstream sequences effectively shifts the ensemble towards different states. Ligand binding, intermolecular pairing, helicases, and post-transcriptional modifications are examples of additional factors that can stabilize different structured or unstructured states. C. Differences in cell environments, including changes in RBP expression, differential localization to phase-separated granules, and post-transcriptional modifications drive redistribution of RNA structural ensembles across cell populations.

Figure 2:

Methods for probing RNA structural ensembles in cells A. Conventional chemical probing experiments report per-nucleotide reactivities averaged over every state in a structural ensemble. Partition function and sample-and-select approaches can be used to estimate structural ensembles. B. Alternative allele- and isoform-resolved chemical probing can reveal changes in reactivity profiles indicative of shifts in structural ensembles across different isoforms. C. Single-molecule chemical probing measures correlated modification events that directly identify distinctive structural states. The data can be deconvoluted into reactivity profiles and populations of each ensemble state. D. Proximity ligation methods directly measure RNA base-paired duplexes, including long-range intra- and inter-molecular interactions and rare states. These methods can thus reveal alternative pairing interactions but provide limited information about populations.

Figure 3:

Operating principles of RNA structural ensembles A. RNA structural ensembles modulate RNA functional output by tuning the population of functional substates. Ensembles are often sensitive to cellular factors and can be shifted toward or away from functional states for dynamic, context-specific control. B. Types of functional substates. Left, effector states feature structural motifs that recruit proteins or directly modulate the function of translocating machines. Middle, masking states sequester sequence motifs, preventing binding by factors that recognize single-stranded nucleotides. Right, proximity-inducing states bring distal RNA elements into proximity to facilitate communication between factors.

Figure 4:

Examples of functional control by RNA structural ensembles A. The FMN riboswitch in Bacillus subtilis modulates transcription based on the cellular concentration of FMN metabolite. FMN binding stabilizes an effector substate that induces transcription termination. B. In Drosophila oocytes, localization of oskar mRNA is controlled in part by an effector structure that forms across the exon 1-exon 2 junction in spliced transcripts. This effector structure likely recruits a still-to-be-identified RBP. C. Translation of the human CDKN1B 3’ UTR is regulated by a masking state equilibrium that modulates accessibility of Pumilio RBP and miRNA binding sites. Cell cycle dependent changes in Pumilio proteins (PUM1 and PUM2) shift the CDKN1B ensemble towards unmasked states, permitting coordinate translation downregulation by PUM1/2 and miR-221/222. D. The repeat A domain of the mammalian Xist lncRNA populates a heterogenous structural ensemble that scaffolds assembly of numerous proteins to mediate X chromosome inactivation. The different states collectively function as proximity-inducing states that concentrate protein binding sites and may provide flexibility for interacting with different subsets of proteins. E. In Flavivirus RNAs, replication and translation are regulated by a structural equilibrium involving long-range pairing interactions between the 5’ and 3’ UTR. The “circular” form (right) functions as a proximity-inducing state that brings the 5’ and 3’ ends into proximity for replication and also acts as a masking state that prevents translation. F. Stability of maternal mRNAs in zebrafish embryos is regulated by 3’ UTR masking states that modulate binding by the stabilizing protein Elavl1a.

Figure 5:

Dysregulation of RNA ensembles in disease A. Alternative splicing of the MAPT mRNA is regulated by a structural equilibrium that masks the 5’ splice site downstream of exon 10. In some tauopathies, intronic mutations disrupt this structural equilibrium and result in pathological overproduction of the 4R MAPT isoform. B. In repeat expansion diseases, an increased number of repeat sequence elements stabilizes inter-molecular pairing interactions, resulting in formation of neurotoxic granules that sequester RNAs and proteins. C. Translation of many mRNAs is regulated by structural equilibria in 5’ UTRs that block ribosome scanning or mask translation start sites. Dysregulation of helicases that modulate these 5’ UTR structural equilibria results in aberrant translation and has been linked to cancers and neurological diseases. D. The 7SK lncRNA codes for a structural ensemble that controls the population of an effector state that binds and inhibits the transcription factor P-TEFb. During HIV infection and potentially in cancer, aberrant expression of viral and host factors shifts the 7SK structural equilibrium towards anti-effector P-TEFb-released states, supporting upregulated transcription.

Figure 6:

Therapeutical potential of RNA structural ensembles A. ASOs and small molecules can be used to stabilize specific ensemble substates to effect positive or negative changes in RNA function. B. Nusinersen is a clinically approved ASO for treatment of spinal muscular atrophy that targets the SMN2 pre-mRNA to increase inclusion of exon 7 during splicing. Nusinersen functions by destabilizing a masking structure and blocking binding of hnRNP proteins that antagonize recognition of the exon 7 splice site. C. A number of small molecules have been developed to downregulate translation of cancer genes by stabilizing G-quadruplex structures in mRNA 5’ UTRs that block ribosome scanning. D. Small molecules and ASOs have been used to disrupt −1 frameshifting in SARS-CoV-2 as potential antiviral therapies. SARS-CoV-2 frameshifting is induced by a pseudoknot effector structure that is in competition with multiple other anti-effector structures. While the exact mechanisms of action for these compounds remain incompletely understood, it is likely that they bind to and stabilize anti-effector states.