High-pH NMR to Identify Macromolecular Hydrogen-Bonds and Foldons

This study demonstrates that high-pH NMR spectroscopy is a sensitive, rapid, and accurate method for identifying macromolecular hydrogen bonds and mapping protein folding hierarchies ("foldons") in marginally stable or partially folded structures, outperforming traditional hydrogen/deuterium exchange techniques.

The Gist

The Big Idea: Finding the "Skeleton" of a Protein in a Storm

Imagine a protein is like a complex origami crane made of paper. To understand how it works, scientists need to know exactly how the paper is folded. The most important parts of the fold are held together by tiny "glue spots" called hydrogen bonds.

For decades, scientists tried to find these glue spots by putting the protein in a special liquid (heavy water) and waiting to see which parts of the paper stayed dry. If a part stayed dry, it meant it was glued shut. If it got wet, it was loose.

The Problem: This method is slow and finicky. If the origami is a bit flimsy or the paper is thin, the glue spots might dissolve before the scientist can even look at them. It's like trying to find the structural beams of a house by waiting for a slow leak; if the house is weak, it might collapse before you find the beams.

The New Solution: This paper introduces a "storm test." Instead of waiting for a slow leak, the scientists put the protein in a very strong, alkaline (soapy) solution. This creates a massive, fast storm that washes away everything that isn't glued down immediately.

How It Works: The "High-PH" Storm

Think of the protein as a group of people holding hands in a circle.

• Normal Conditions (Neutral pH): The wind is light. People who are holding hands loosely might let go eventually, but it takes a while.
• The High-pH Storm (pH 10–11): The scientists turn on a hurricane-force wind.
  • Anyone holding hands loosely (unstructured parts) gets blown apart instantly.
  • Anyone holding hands tightly (strong hydrogen bonds) stays together because their grip is too strong for the wind to break.

By looking at who is still holding hands in the middle of the hurricane, the scientists can map out exactly where the strong glue spots are.

The Results: A Better Map

The researchers tested this "storm method" on 10 different proteins (some are like balls, some are like long ropes). Here is what they found:

1. It's More Accurate: The old method (waiting for a slow leak) was about 80% accurate. The new storm method is about 91% accurate. It catches the "glue spots" that the old method missed, especially in proteins that are a bit wobbly or unstable.
2. It's Fast and Cheap: You don't need expensive equipment or special radioactive markers. You just need a standard NMR machine (a giant magnet that takes pictures of atoms) and some baking soda to raise the pH. You can get the results in minutes rather than days.
3. It Reveals the "Core": As they made the storm stronger (increasing the pH), they watched the protein fall apart layer by layer.
   • First, the loose ends blew away.
   • Then, the outer layers fell off.
   • Finally, only the very center remained.

This center is called a "foldon." It's the most stable, unbreakable part of the protein.

The "Foldon" Analogy: The Core of the Onion

Imagine peeling an onion.

• The Outer Layers: These are the parts of the protein that fall apart at pH 10. They are important, but they aren't the most stable.
• The Inner Core: As you keep peeling (increasing the pH to 11 or 12), you eventually reach the very center. This is the foldon.

In this study, they found that for some proteins, the foldon is a specific bundle of strands (like the core of a woven basket). For others, it's a specific section of a spiral (like the tightest part of a spring).

Why This Matters

This method is like having a super-powerful X-ray that only shows you the strongest parts of a building.

• For Medicine: Many diseases (like Alzheimer's) happen when proteins fold incorrectly. This method helps us see which parts of the protein are the "weak links" that might break and cause trouble.
• For Drug Design: If you want to make a drug that sticks to a protein, you need to know where the strong glue spots are. This method maps them out clearly.
• For Understanding Life: It helps us understand how proteins fold in the first place. It turns out that the parts that are hardest to break (the foldons) are often the very first parts that come together when the protein is born.

Summary

The scientists discovered that by throwing proteins into a chemical "hurricane" (high pH), they can instantly strip away the weak, floppy parts and leave only the strong, glued-together core. This is a faster, cheaper, and more accurate way to see the hidden skeleton of proteins than the old, slow methods. It's like finding the steel beams of a building by blowing the drywall away with a firehose.

Technical Summary

1. Problem Statement

Determining hydrogen bonds (H-bonds) is critical for accurate NMR structure determination, yet traditional methods face significant limitations, particularly for marginally stable proteins or folding intermediates:

• Limitations of H/D Exchange in D₂O: The standard method relies on protecting amide protons from solvent exchange in D₂O (EX2 limit). This requires the protein to be stable enough to prevent exchange over minutes to hours. Weak or transient H-bonds in unstable structures often fail to provide detectable protection, leading to "false negatives."
• Limitations of Direct H-bond Detection: Techniques like long-range HNCO experiments (detecting $^h3J_{NC'}$ couplings) are direct but require perdeuterated samples and long acquisition times, making them impractical for many systems.
• The Gap: There is a need for a sensitive, fast, and broadly applicable method to identify H-bonds in proteins that are too unstable for traditional D₂O exchange studies or to map the hierarchy of structural stability (foldons) prior to complete unfolding.

2. Methodology

The authors propose High-pH NMR as an alternative strategy to identify H-bonds and structural stability hierarchies.

• Principle: Hydrogen exchange (HX) is base-catalyzed. As pH increases (specifically pH 10–11), the intrinsic exchange rate ($k_{int}$) increases by a factor of 10 per pH unit.
• Mechanism Shift: At high pH, the exchange mechanism shifts from the EX2 limit (stability-dependent) to the EX1 limit (opening-rate dependent). In this regime, the observed exchange rate ($k_{ex}$) is governed by the rate of H-bond breaking ($k_{op}$) rather than the equilibrium stability.
• Experimental Approach:
  • Proteins are dissolved in H₂O (not D₂O) and titrated to pH 10–11.
  • Standard 2D $^1$H-$^{15}$N HSQC spectra (for labeled proteins) or TOCSY (for unlabeled proteins) are acquired.
  • Logic: Unstructured regions exchange rapidly and their signals disappear due to fast exchange broadening or transfer to the solvent. Only amide protons involved in stable H-bonds (which slow down the opening rate) survive to be detected in the spectrum.
• Dataset: The study analyzed ~750 amide sites across 10 proteins with known X-ray, cryo-EM, or NMR structures (including Ubiquitin, GCN4p, CspA, LysN, SN, P22iD, CUS3iD, Kinesin neck, etc.).
• Validation: Results were compared against:
  1. Known structural H-bonds (from PDB).
  2. Traditional D₂O exchange data.
  3. Contact density calculations (Ca-Ca distances < 5 Å).
  4. Sequence propensity for secondary structure.

3. Key Contributions

• Novel Identification Method: Established that amide proton survival at pH 10–11 is a robust indicator of H-bonding, offering a "tunable" approach where increasing pH progressively strips away less stable structures.
• Foldon Mapping: Demonstrated that high-pH titration reveals a hierarchy of structural stability ("foldons"), identifying the most persistent sub-structures (resistant to alkaline unfolding) before total denaturation.
• Methodological Advantages: The approach is inexpensive, fast (minutes per spectrum), requires no specialized labeling (works with unlabeled samples via TOCSY), and is applicable to marginally stable proteins where D₂O exchange fails.

4. Key Results

• Accuracy of H-bond Identification:
  • High-pH NMR: Achieved ~91% accuracy in identifying H-bonds.
    • False Positives (FP): ~6% (surviving but not H-bonded).
    • False Negatives (FN): ~3% (H-bonded but not detected).
  • Traditional D₂O Exchange: Achieved only ~80% accuracy for the same dataset.
    • Higher FN rate (12%) due to insufficient protection in marginally stable structures.
    • Higher FP rate (8%) due to arbitrary time thresholds.
• Unfolding Hierarchies (Foldons):
  • Coiled Coils (GCN4p, Kinesin neck): Showed a gradient of stability from N- to C-terminus. The most persistent regions corresponded to "trigger sites" with the highest $\alpha$-helical propensity, consistent with nucleation sites for folding.
  • $\beta$-Sheet Proteins (P22iD, CUS3iD): The most persistent structures were the core six-stranded antiparallel $\beta$-barrels, despite being non-contiguous in sequence. This matched foldons identified in neutral pH NSHX studies.
  • Ubiquitin: Persistent amides mapped to a ladder of secondary structure near the N-terminus ($\beta$1-$\beta$2 hairpin and $\alpha$-helix), while the C-terminal region (rich in basic residues) unfolded earlier, likely due to electrostatic repulsion at high pH.
  • Global Unfolding: Some proteins (e.g., SN) showed cooperative global unfolding where all signals disappeared within a narrow pH range.
• Determinants of Stability:
  • For $\beta$-sheet proteins, the pH at which amides disappear correlates strongly with inter-residue contact density (Ca-Ca contacts < 5 Å).
  • For coiled coils, stability is determined by sequence propensity for $\alpha$-helix formation rather than contact density.

5. Significance

• Superiority for Unstable Systems: High-pH NMR outperforms traditional D₂O exchange for identifying H-bonds in marginally stable proteins and folding intermediates, where weak H-bonds are often missed by standard methods.
• Insight into Folding Pathways: The method provides a rapid way to map the "stability hierarchy" of a protein, revealing which structural elements (foldons) are the most robust. This aligns with folding intermediates observed under neutral conditions, suggesting that alkaline unfolding follows native-like pathways rather than random disruption.
• Broad Applicability: The technique is accessible to most NMR labs as it uses standard pulse sequences and does not require perdeuteration or complex sample preparation.
• Future Directions: Offers a new window to explore protein dynamics, misfolding mechanisms (relevant to amyloid diseases), and the stability of pharmaceutical proteins under basic conditions.

In conclusion, the paper validates high-pH NMR as a powerful, high-accuracy tool for mapping hydrogen-bonded networks and structural stability hierarchies, effectively overcoming the sensitivity limitations of traditional hydrogen-deuterium exchange experiments.