Alternative probe chemistries for single-molecule analysis of long non-coding RNA

This study demonstrates that optimizing probe backbone chemistries, particularly using locked nucleic acid (LNA) residues, significantly enhances the precision and accuracy of SiM-KARTS in distinguishing and classifying the structural dynamics of complex long non-coding RNAs without relying on disruptive ionic adjustments.

The Gist

Imagine you are trying to understand the shape of a giant, tangled ball of yarn (which represents a Long Non-Coding RNA, or lncRNA). These yarn balls are huge, complex, and constantly shifting. Scientists want to know exactly how the yarn is folded because the shape determines what the RNA does in our bodies.

The problem? The yarn is too big and messy to study with standard tools. It's like trying to figure out the layout of a whole city just by looking at a single street corner.

The Detective's Tool: SiM-KARTS

To solve this, scientists use a clever trick called SiM-KARTS. Think of this as sending out a tiny, glowing magnetic key (a probe) to find a specific lock on the yarn ball.

• If the lock is open (the yarn is loose there), the key snaps on and off quickly.
• If the lock is hidden (the yarn is wrapped tight around it), the key has a hard time finding it, or it can't stick at all.

By watching how often the key sticks and how long it stays, scientists can map out the shape of the yarn.

The Problem: The Keys Were Too Weak or Too Sticky

In the past, scientists used standard DNA keys. But these had a flaw: they were either too weak to stick long enough to be seen, or they were so sticky that they got stuck permanently, freezing the yarn in place and ruining the experiment.

It's like trying to test a door with a magnet that is either too weak to hold the door open, or so strong it rips the door off its hinges.

The Solution: Upgrading the Keys

This paper is about testing new types of keys to see which ones work best for these giant, complex yarn balls. The researchers tried two special upgrades:

1. LNA Keys (Locked Nucleic Acid): Imagine these keys are made of a special, rigid plastic that fits the lock perfectly. They are "locked" in place, making them stickier and more precise than regular DNA.
2. Morpholino Keys: These are like keys made of a completely different material (a neutral plastic) that doesn't react much to the environment (like salt or water).

What They Discovered

The researchers tested these keys on a model yarn ball (based on a real RNA called Braveheart) to see how well they could tell the difference between a "loose" part of the yarn and a "tight" part.

• The Regular DNA Key: It was okay, but it couldn't tell the difference between a loose spot and a tight spot very well. It was like a detective who can't tell if a door is slightly ajar or wide open.
• The Morpholino Key: It was very sticky and stayed on for a long time, but it didn't care if the door was open or closed. It stuck to everything equally. It was too "dumb" to give useful information about the shape.
• The LNA Key (The Winner!): This was the superhero. It was sensitive enough to know exactly how "open" the lock was.
  • When the lock was open, the LNA key stuck for a nice, measurable amount of time.
  • When the lock was hidden, the key barely stuck at all.
  • The Magic: Because the LNA key was so precise, the scientists could look at a single strand of data and say, "Ah, this specific piece of yarn is in a 'tight' shape," versus "This one is 'loose'." They could sort the yarn strands into different shape categories just by watching the key's behavior.

The Salt Factor

The researchers also found that changing the "water" around the yarn (adding salt) changed how the keys behaved, but not always in a straight line. Sometimes adding salt helped the key stick; other times, it actually changed the shape of the yarn itself, making the lock harder to find. This taught them that changing the key's material (chemistry) is a better way to tune the experiment than changing the water (salt).

The Big Picture

This paper is a guidebook for future scientists. It says: "If you want to study these giant, complex RNA molecules, don't just use standard DNA keys. Use LNA keys."

By using these upgraded keys, scientists can finally start mapping the shapes of these mysterious RNA molecules without breaking them or getting confused. This is a huge step forward because understanding these shapes could help us develop new medicines to treat diseases where these RNAs go wrong.

In short: They found the perfect "magnetic key" that is just sticky enough to work, but smart enough to tell the difference between a locked door and an open one, allowing us to finally see the hidden shapes of life's complex molecules.

Technical Summary

1. Problem Statement

Long non-coding RNAs (lncRNAs) are critical regulators of gene expression and are implicated in various human diseases. However, characterizing their structure and dynamics is challenging due to their large size (>200 nucleotides) and complex folding.

• Limitations of Current Methods: Standard single-molecule Förster resonance energy transfer (smFRET) requires site-specific labeling, which is costly, technically demanding, and can perturb the RNA structure.
• Limitations of Existing SiM-KARTS: Single-molecule kinetic analysis of RNA transient structure (SiM-KARTS) offers a labeling-free alternative by using short, fluorescently labeled oligonucleotide probes to monitor the accessibility of specific RNA regions. However, adapting SiM-KARTS to complex lncRNAs is difficult. Optimizing probe binding usually involves adjusting probe length or ionic conditions (salt concentration).
  • Increasing probe length reduces spatial resolution and increases the risk of disrupting RNA structure.
  • Adjusting ionic conditions (e.g., Mg²⁺) to tune binding stability can inadvertently alter the tertiary structure of the target RNA itself, leading to artifacts.
• The Gap: There is a need for a method to fine-tune probe binding stability and structural sensitivity without altering the target RNA's native environment or structure.

2. Methodology

The authors employed a model system based on a 59-nucleotide segment of the lncRNA Braveheart (Bvht), specifically Helix 9 (Bvht H9). They compared four distinct probe chemistries:

1. DNA: Standard DNA backbone.
2. LNA2: DNA backbone with one Locked Nucleic Acid (LNA) residue at position 2.
3. LNA23: DNA backbone with LNA residues at positions 2 and 3.
4. MO: Morpholino backbone (neutral charge).

Experimental Setup:

• Target Structures: They created two target states:
  • Bvht H9: The native structure where the probe binding site is partially occluded (structured).
  • Bvht H9:cs: A hybridized state where a complementary strand opens the binding site, making it fully single-stranded and accessible.
• Imaging: Single-molecule Total Internal Reflection Fluorescence (TIRF) microscopy was used to track the binding and dissociation of fluorescent probes (Cy5 or TYE665) to Cy3-labeled target RNAs.
• Data Analysis:
  • Kinetic Traces: Hidden Markov Modeling (HMM) was used to idealize fluorescence traces into "bound" and "unbound" states.
  • Metrics: Bound and unbound dwell times were analyzed using cumulative distribution functions fitted with single- or double-exponential functions.
  • Clustering: A "kinetic fingerprint" approach was developed, plotting the average bound dwell time ($\langle t_{bound} \rangle$) vs. average unbound dwell time ($\langle t_{unbound} \rangle$) for individual traces to classify structural states.
• Orthogonal Validation:
  • Thermal Denaturation: Melting temperature ($T_M$) measurements were performed to assess hybrid stability.
  • Circular Dichroism (CD): Used to monitor global structural changes in the target RNA under varying ionic conditions.

3. Key Contributions

• Probe Chemistry Optimization: Demonstrated that modifying the probe backbone (specifically incorporating LNA) is a superior strategy for tuning binding stability compared to changing salt concentrations or probe length.
• Kinetic Fingerprinting: Introduced a robust clustering method using both bound and unbound dwell times to distinguish between different RNA structural states on a trace-by-trace basis, rather than relying solely on ensemble averages.
• Ionic Sensitivity Analysis: Revealed that divalent cations (Mg²⁺) have non-monotonic effects on probe binding kinetics, likely due to competing effects on probe-RNA hybrid stability versus target RNA structural rearrangement.

4. Key Results

• Binding Stability and Sensitivity:
  • LNA Probes: Showed the highest structural sensitivity. LNA probes exhibited significantly longer bound dwell times for the accessible target (Bvht H9:cs) compared to the occluded target (Bvht H9). They also displayed distinct unbound dwell times, allowing for clear discrimination between the two structures.
  • DNA Probes: Showed poor discrimination; bound dwell times were too short to distinguish structures effectively, and unbound dwell times were inconsistent.
  • Morpholino (MO) Probes: Exhibited high binding stability (long dwell times) but low structural sensitivity. They bound similarly to both accessible and occluded targets, failing to distinguish between the two states.
• Trace-by-Trace Classification:
  • Using the kinetic fingerprint ($\langle t_{bound} \rangle, \langle t_{unbound} \rangle$), LNA probes allowed for the accurate classification of individual traces into structural categories.
  • In a heterogeneous mixture, 85% of LNA probe traces were correctly assigned to their specific structural origin (Bvht H9 vs. Bvht H9:cs).
  • In contrast, DNA and MO probes showed significant overlap in their distributions, making trace-by-trace classification impossible (only ~29% accuracy for DNA).
• Impact of Ionic Conditions:
  • Mg²⁺ Effects: Increasing Mg²⁺ concentration (up to 1 mM) generally stabilized the probe-RNA hybrid. However, at high concentrations (20 mM), bound dwell times for LNA probes on the accessible target decreased.
  • CD Spectroscopy: Confirmed that high Mg²⁺ induces a more ordered, compact structure in the target RNA (Bvht H9:cs), which likely occludes the binding site despite the higher ionic strength stabilizing the hybrid. This explains the non-monotonic binding behavior.
  • MO Probes: Showed minimal sensitivity to ionic changes, confirming their neutral backbone limits ion-dependent effects, but this also contributed to their lack of structural discrimination.

5. Significance

This study provides a critical design framework for applying SiM-KARTS to complex lncRNAs and other large RNA systems.

• Design Principle: Researchers should prioritize probe backbone chemistry (specifically LNA incorporation) over environmental tuning (salt concentration) to optimize probe performance. This avoids the risk of perturbing the native RNA structure.
• Enhanced Resolution: The use of LNA probes enables the resolution of heterogeneous RNA populations at the single-molecule level, allowing researchers to distinguish between transiently accessible and occluded regions within a single molecule's trajectory.
• Future Applications: The established "kinetic fingerprint" analysis and LNA probe design principles will facilitate the structural and dynamic characterization of lncRNAs, which are currently under-studied due to methodological limitations. This paves the way for understanding lncRNA mechanisms in disease and developing RNA-targeted therapeutics.