🧬 biology

Sub-100 ns transition paths in unmodified single protein measurements

This paper introduces temperature-jump nanoaperture optical tweezers (TJ-NOTs) to observe unmodified single proteins across a vast temporal range from nanoseconds to hours, revealing ultrafast sub-100 ns transitions in bovine serum albumin and bridging the gap between molecular dynamics simulations and biologically relevant timescales.

Original authors: Reuven Gordon, Tianyu Zhao, Xi Ren, Edona Karakaci, Ethan Bintrim, Mert Gur, Cuifeng Ying, Michael Mayer, Robert Smith

Published 2026-08-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Reuven Gordon, Tianyu Zhao, Xi Ren, Edona Karakaci, Ethan Bintrim, Mert Gur, Cuifeng Ying, Michael Mayer, Robert Smith

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Proteins are the workhorses of life, tiny molecular machines that fold themselves into precise shapes to perform tasks ranging from carrying oxygen in the blood to copying genetic instructions. For these machines to work, they must maintain a specific structure, but they are not static statues; they are constantly shifting, breathing, and occasionally changing shape entirely. Scientists have long known that these changes can happen over a vast range of time. Some shifts are slow, taking seconds or even minutes, while others are so fast they were thought to be nearly impossible to catch with current technology. The challenge has been building a tool that can watch a single protein molecule without touching it, without sticking a fluorescent tag on it that might change how it moves, and without missing the split-second moments when it flips from one shape to another. If researchers could see these rapid transitions, they could finally connect the ultrafast movements predicted by computer models with the slower, biologically relevant changes observed in the real world.

A team of researchers has now built a method capable of watching these events in real time, capturing protein movements that occur in less than one hundred nanoseconds. Using a technique called temperature-jump nanoaperture optical tweezers, they trapped single, unmodified protein molecules in a tiny hole in a metal film and heated them with a laser to trigger changes. This setup allowed them to observe the entire lifespan of a protein's movement, from the slow pauses between changes that last for minutes, down to the fleeting transitions that happen in a fraction of a microsecond. The study focused on two specific proteins: bovine serum albumin, which is found in cow blood, and conalbumin, a protein found in egg whites. By heating these molecules step-by-step, the team watched them unfold, refold, and undergo dramatic structural shifts, revealing a hidden world of speed that was previously invisible to direct observation.

When the researchers applied this method to bovine serum albumin, they witnessed a startlingly fast event. As the temperature rose to a specific range, the protein underwent a sudden, irreversible transformation. The signal from the trapped molecule showed a sharp step change that happened in just 44 nanoseconds, and on average, these fast transitions occurred in less than 100 nanoseconds. To put this speed in perspective, it is an order of magnitude faster than the previously known "speed limit" for how quickly a protein can fold. The team confirmed that this rapid change was a shift from a coiled, spiral structure known as an alpha-helix into a flat, sheet-like structure called a beta-sheet. This specific type of change is significant because it is linked to serious biological issues, such as the formation of clumps in neurodegenerative diseases, but it is also essential for normal processes like blood clotting. The researchers verified that this was a true structural shift and not just a random fluctuation by adding a chemical called urea, which is known to unfold proteins. When urea was present, the fast, sharp transition disappeared, replaced by a much slower, messy unfolding process, proving that the original fast event was a specific, organized change in the protein's shape.

In contrast to the one-way street observed with bovine serum albumin, the team watched conalbumin perform a reversible dance of unfolding and refolding. When heated, this protein unfolded in about 1.8 milliseconds, a speed that is still incredibly fast but measurable with standard equipment. Once unfolded, the molecule did not immediately snap back; it lingered in its open, disordered state for a long time, with a median wait of 68 seconds before it decided to fold again. When it finally did refold, the process took 17 milliseconds. This ability to watch a single molecule pause for over a minute and then snap back into shape provided a complete picture of the protein's energy landscape, showing both the deep valleys where it rests and the steep hills it must cross to change form. The researchers also tested ribonuclease A, another protein known for its ability to refold, and observed a slower transition taking three seconds, consistent with what is seen in larger groups of molecules.

The power of this new approach lies in its ability to bridge the gap between two different worlds of science. For years, computer simulations have predicted that proteins can make these ultrafast transitions in less than 100 nanoseconds, but experimental tools were too slow to see them, often filtering out the fastest signals as noise. At the same time, other experiments could only see the slow, average behavior of millions of molecules at once, missing the unique story of any single one. This new method captures the full range of motion, from the nanosecond flicker of a structural flip to the hour-long wait of a protein resting in a stable state, all without altering the protein with labels or tethers. By proving that these sub-100 nanosecond transitions are real and measurable, the work provides a solid foundation for refining computer models and understanding how proteins navigate their complex energy landscapes. The findings suggest that the rapid, organized shifts between different shapes are not just theoretical possibilities but are fundamental parts of how proteins function and sometimes malfunction in the body.

Technical Summary: Sub-100 ns Transition Paths in Unmodified Single Protein Measurements

Problem Statement
Observing the full dynamical range of single protein transitions—from nanoseconds to hours—presents a significant challenge in molecular biology and biophysics. Existing methods face a trade-off between temporal resolution and the ability to study unmodified proteins. While techniques like single-molecule FRET and high-bandwidth photon counting have resolved transition paths in the microsecond to sub-microsecond range (e.g., 700 ns), they often require protein labeling or tethering, which can perturb the energy landscape. Conversely, label-free techniques such as nanopore sensing and optical tweezer force spectroscopy typically suffer from bandwidth limitations (e.g., low-pass filtering at 2 MHz or microsecond-scale resolution), preventing the observation of ultrafast events. Furthermore, the energy landscape of protein dynamics involves long dwell times between quasi-stable states (hundreds of seconds) and rapid transition paths (potentially sub-100 ns). Computational approaches like replica-exchange molecular dynamics can sample these fast transitions but rely on non-physical temperature exchanges, raising questions about whether the inferred timescales match real physical timescales. There is a critical need for a label-free method capable of resolving sub-100 ns transition paths in unmodified proteins to bridge the gap between ultrafast molecular dynamics simulations and biologically relevant timescales.

Methodology: Temperature-Jump Nanoaperture Optical Tweezers (TJ-NOT)
The authors introduce Temperature-Jump Nanoaperture Optical Tweezers (TJ-NOT) to address these limitations. The system utilizes a double-nanohole aperture in a gold film to trap single, unmodified proteins via optical scattering.

Mechanism: The trapping laser intensity is stepped up in 1-second intervals. The absorption of laser energy by the gold film around the nanoaperture induces rapid local temperature jumps (approximately 1.3°C per step).
Detection: The signal is derived from the transmission of the laser through the aperture, which fluctuates as the protein's conformation changes its optical polarizability.
Resolution: The system employs a detector with a 50 MHz bandwidth and a sampling period of 4 ns, achieving a time resolution of 20 ns (limited by detector response). This allows for the observation of dynamics across a temporal dynamic range of 10 orders of magnitude (nanoseconds to hours) without the need for fluorescent labels or chemical tethers.
Validation: The study utilizes Bovine Serum Albumin (BSA) and conalbumin. Molecular dynamics (MD) simulations of specific $\alpha$ -helical regions (H19 and H22) of BSA at 353.2 K were performed to provide a theoretical basis for the observed transition times.

Key Results

Sub-100 ns Irreversible Transitions in BSA:
- Upon heating single BSA molecules to 65–68°C, the authors observed fast, stepwise signal changes indicative of conformational transitions.
- Using high-speed sampling, they resolved transition times ranging from 12 ns to 160 ns. The average transition time across 11 events was 90 ± 60 ns.
- These transitions were found to be irreversible on the experimental timescale (no return to the initial state after cooling).
- Urea Validation: The addition of urea (a denaturant) altered the dynamics. At 1 M and 4 M urea, only slow transitions (average ~1.7 ms) were observed. At 8 M urea, no transitions were observed, consistent with total denaturation. This supports the hypothesis that the ultrafast events in pure BSA correspond to a specific structural transition (likely $\alpha$ -helix to $\beta$ -sheet) rather than general denaturation.
- MD Correlation: Simulations of H19 and H22 regions showed $\alpha$ -helix to $\beta$ -sheet transitions occurring on timescales of ~700 ns and ~1150 ns, respectively. While these simulation times are longer than the experimental average, they demonstrate comparable conformational transition mechanisms and timescales within the ultrafast regime.
Reversible Dynamics in Conalbumin:
- Unlike BSA, conalbumin exhibited reversible unfolding and refolding under repeated temperature jumps.
- Unfolding: Occurred at ~65°C with a median transition time of 1.8 ms.
- Dwell Time: Once unfolded and cooled to 38°C, the protein remained in the unfolded state for a median dwell time of 68 s before refolding.
- Refolding: The initial refolding transition occurred with a median time of 17 ms.
- Ribonuclease A: As a secondary example, Ribonuclease A showed a slower refolding transition of 3 s, consistent with ensemble measurements, though it displayed heterogeneous dynamics including quasi-periodic oscillations.

Significance and Claims
The paper claims that TJ-NOT successfully captures the full dynamical range of heterogeneous thermally-induced conformational changes in structural biology. By resolving sub-100 ns transition paths in unmodified proteins, the method provides the first experimental observation of ultrafast irreversible transitions in BSA, an order of magnitude faster than the previously estimated folding "speed limit."

The authors posit that this approach connects the ultrafast timescales of molecular dynamics simulations with biologically relevant timescales (microseconds to hours). Because the technique is label-free and does not require tethering, it avoids perturbing the protein's energy landscape. The authors suggest these findings will inform computational models of protein dynamics and potentially aid in training machine learning approaches. They note that while the current setup uses continuous heating, the method's nanosecond thermal equilibration suggests that with pulsed laser modifications, it could further explore fast-folding proteins. The work addresses the fundamental challenge of validating whether the fast timescales inferred from non-physical computational models (like replica-exchange) correspond to real physical transition paths.

Technical Summary: Sub-100 ns Transition Paths in Unmodified Single Protein Measurements

More like this