Chemical Probing

Chemical probes selectively modify RNA nucleotides based on their local structural environment. Flexible, unpaired nucleotides react more readily than those constrained by base pairing or tertiary interactions. The resulting reactivity profile provides a per-nucleotide readout of structural dynamics.

• SHAPE reagents (e.g., 1M7, NMIA) acylate the 2'-hydroxyl group, with reactivity correlating to local nucleotide flexibility
• DMS (dimethyl sulfate) methylates the N1 position of adenine and N3 position of cytosine, requiring these atoms to be solvent-accessible
• CMCT (1-cyclohexyl-3-(2-morpholinoethyl)carbodiimide metho-p-toluenesulfonate) modifies N3 of uridine and N1 of guanine at accessible positions

Reactivity data is converted into pseudo-energy terms and incorporated into thermodynamic folding algorithms. This soft constraint penalizes structures inconsistent with experimental data, dramatically improving prediction accuracy beyond sequence-only methods.

• SHAPE reactivity is transformed using the Deigan equation: ΔG_SHAPE = m × ln(SHAPE reactivity + 1) + b
• The pseudo-energy term is added to the standard Turner nearest-neighbor energy model
• Partition function calculations incorporate reactivity to refine base pairing probabilities across the thermodynamic ensemble

Modified RNA is analyzed through primer extension with reverse transcriptase, which stalls at adduct sites. The resulting cDNA fragments are separated by capillary electrophoresis to produce an electropherogram where peak intensities correspond to modification frequency.

• Fluorescently labeled primers enable multiplexed detection across multiple channels
• ddNTP sequencing ladders provide single-nucleotide resolution alignment
• Normalization against no-reagent controls corrects for background reverse transcriptase drop-off

Chemical probing can be performed under native cellular conditions or in purified systems, revealing how the intracellular environment—including molecular crowding, protein binding, and metabolite interactions—modulates RNA structure.

• DMS-MaPseq (mutational profiling) uses DMS modification followed by thermostable group II intron reverse transcriptase (TGIRT) to introduce mutations during cDNA synthesis, readable by next-generation sequencing
• icSHAPE (in vivo click SHAPE) uses cell-permeable reagents with biotinylation and streptavidin enrichment for transcriptome-wide structural analysis
• In vivo probing captures ribonucleoprotein (RNP) footprints where protein binding occludes chemical modification

Coupling chemical probing with high-throughput sequencing enables simultaneous structural interrogation of thousands of RNAs. These methods generate genome-scale reactivity profiles that inform global RNA structure-function relationships.

• SHAPE-Seq and SHAPE-MaP adapt SHAPE chemistry for next-generation sequencing readout
• PARS (parallel analysis of RNA structure) uses structure-specific nucleases (RNase V1 for helices, S1 nuclease for loops) with deep sequencing
• DMS-Seq applies DMS probing transcriptome-wide, detecting structural signatures of riboswitches, IRES elements, and regulatory motifs

Chemical probing serves as the primary experimental benchmark for validating RNA structure predictions. Agreement between predicted base pairing and observed reactivity patterns provides confidence in model accuracy without requiring crystallography or cryo-EM.

• Area under the ROC curve (AUC) quantifies how well predicted base pairs discriminate between reactive and unreactive nucleotides
• RNA-Puzzles and CASP-RNA assessments use chemical probing data as evaluation criteria
• Discrepancies between prediction and probing highlight structural dynamics and alternative conformations not captured by static models

Selective 2'-Hydroxyl Acylation analyzed by Primer Extension measures nucleotide flexibility by acylating the 2'-OH group of unconstrained nucleotides. The resulting adducts block reverse transcription, creating a detectable stop at each modified position.

• Reactivity correlates inversely with base-pairing stability
• Data integrated as pseudo-energy restraints in folding algorithms
• SHAPE-MaP variant uses mutational profiling for higher throughput

Dimethyl sulfate methylates the N1 position of unpaired adenines and N3 of unpaired cytosines, providing complementary structural information to SHAPE. DMS cannot penetrate Watson-Crick base pairs, making it a direct probe of solvent accessibility.

• Detects secondary structure and tertiary interactions
• DMS-MaPseq enables genome-wide RNA structure probing
• Combined with SHAPE for multi-reagent reactivity profiles

Chemical probing reactivity values are converted into pseudo-free energy penalties that bias folding algorithms toward experimentally consistent structures. High SHAPE reactivity at a nucleotide adds a penalty if that position is predicted as base-paired.

• Implemented in RNAstructure and ViennaRNA packages
• Restraint strength calibrated against known structures
• Enables accurate prediction of long-range tertiary contacts

MaP techniques replace reverse transcriptase stops with read-through mutations, converting chemical adducts into sequence mutations detected by high-throughput sequencing. This transforms a binary stop signal into a quantitative mutation rate per nucleotide.

• SHAPE-MaP and DMS-MaPseq are key implementations
• Enables multiplexed probing of thousands of transcripts
• Provides per-nucleotide reactivity with single-molecule resolution

Exploits the intrinsic pH-dependent self-cleavage of RNA, where flexible, unconstrained nucleotides undergo spontaneous phosphodiester bond scission at higher rates than structured regions. No chemical reagent is required.

• Particularly useful for riboswitch conformational analysis
• Measures global structural dynamics under varying ligand conditions
• Cleavage products quantified by capillary electrophoresis or sequencing

Chemical probing data represents a population-weighted average of all conformations in solution. Computational methods like DREEM and MOHCA-seq deconvolve these signals to reconstruct multiple coexisting RNA structures.

• Reveals transient folding intermediates
• Critical for understanding co-transcriptional folding pathways
• Combines clustering algorithms with reactivity pattern analysis