🔬 optics

Label-free Imaging of Single-Biomolecule Structure and Interaction by Stimulated Raman Photothermal Encoded Scattering

This paper introduces Stimulated Raman Photothermal Encoded Scattering (SRPSCAT) microscopy, a label-free technique that combines interferometric scattering with stimulated Raman processes to achieve quantitative, chemically specific imaging of single biomolecule structures, masses, interactions, and conformational dynamics in native environments.

Original authors: Pin-Tian Lyu, Yifan Zhu, Qing Xia, Guangrui Ding, Arvind Pillai, Xinru Wang, Jianpeng Ao, Haonan Lin, David Baker, Ji-Xin Cheng

Published 2026-03-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Pin-Tian Lyu, Yifan Zhu, Qing Xia, Guangrui Ding, Arvind Pillai, Xinru Wang, Jianpeng Ao, Haonan Lin, David Baker, Ji-Xin Cheng

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to identify a single, tiny grain of sand on a beach, but you can't touch it, and you can't paint it a bright color to make it stand out. You just have a very powerful camera.

Most current ways of looking at single molecules are like that painting analogy. Scientists usually have to stick a fluorescent "tag" or "glow-in-the-dark sticker" onto a protein to see it. But these stickers have problems: they can fade (photobleach), they might change how the protein behaves, and they don't tell you much about what the protein actually is or what it's made of.

Other methods can see the protein without a sticker, but they are like looking at a silhouette in the fog. You can see that something is there and measure its size, but you can't tell if it's a rock, a shell, or a piece of driftwood. You lack the "chemical ID."

Enter SRPSCAT: The "Thermal Fingerprint" Scanner

This paper introduces a new superpower called SRPSCAT (Stimulated Raman Photothermal Encoded Scattering). Think of it as a high-tech, non-invasive scanner that can identify a single molecule by its unique "thermal fingerprint" without ever touching it or painting it.

Here is how it works, broken down into simple analogies:

1. The "Warm-Up" Trick (The Photothermal Effect)

Imagine you have a specific song that only a specific type of guitar can play. If you play that song near a guitar, the guitar starts to vibrate and get slightly warm.

In the lab: Scientists shine two laser beams at a molecule. These beams are tuned to a specific "frequency" that matches the vibration of a specific chemical bond inside the molecule (like the C-H bond in proteins).
The Result: The molecule absorbs this energy, vibrates, and gets a tiny bit warmer. This is the "Photothermal" part. It's like the molecule is saying, "I'm a protein, and I'm heating up because I recognize this specific frequency!"

2. The "Invisible Ripple" (Interferometric Scattering)

Now, how do we see that tiny bit of heat? Heat makes things expand and changes how light bends through them.

The Analogy: Imagine dropping a pebble in a still pond. You see ripples. Now, imagine the molecule is the pebble. When it gets warm, it creates a tiny, invisible "ripple" in the way light bounces off it.
The Magic: The SRPSCAT microscope is incredibly sensitive. It acts like a super-sonar that detects these tiny ripples in light. It compares the light bouncing off the molecule with a reference beam. When the molecule heats up and changes slightly, the interference pattern shifts. This shift is the signal.

3. Why This is a Game-Changer

The paper shows that this method can do three amazing things that previous methods couldn't do all at once:

It Weighs Tiny Things: Just by looking at how strong the signal is, the microscope can tell you the exact mass of a single protein. It's like weighing a single grain of sand on a scale that's sensitive enough to detect a feather landing on it. They tested this on proteins ranging from small to huge (like the massive IgM antibody), and it worked perfectly.
It Reads the "Shape" of the Molecule: Proteins fold into different shapes (like origami). Some fold into spirals (alpha-helices), and others into sheets (beta-sheets). These shapes vibrate differently. SRPSCAT can "listen" to these vibrations and tell the difference between a spiral protein and a sheet protein, even if they weigh the same. It's like being able to tell a violin from a flute just by the sound of the air moving through them, even if they are the same size.
It Watches Movies of Life: Because it's so fast and doesn't need stickers, the scientists could watch proteins interact in real-time.
- Example: They watched a lock (an antibody) and a key (a protein) find each other, click together, and then let go. They could measure exactly how long the key stayed in the lock.
- Example: They watched a protein change its shape (like a folding chair snapping open) in milliseconds.

4. Real-World Applications

The researchers didn't just stop at theory. They used this to:

Sort Viruses: They looked at Adeno-associated viruses (used for gene therapy). Some are "full" (carrying the cure) and some are "empty" (useless trash). SRPSCAT could instantly tell the difference by detecting the DNA inside the full ones, without opening them up.
Study New Designs: They watched "de novo" (newly designed) proteins switch shapes when a drug was added, proving the design worked.

The Bottom Line

Think of SRPSCAT as a universal translator for the microscopic world.

Old way: "I see a dot. It's heavy. I hope it's the right thing." (Requires painting the dot).
SRPSCAT way: "I see a dot. It weighs exactly 950 kDa. It is made of beta-sheets. It is currently binding to a key. And it is a protein, not dust."

It allows scientists to see the invisible world of single molecules with chemical clarity, speed, and without disturbing the delicate dance of life happening right under the lens.

1. Problem Statement

Current single-molecule detection methods face a fundamental trade-off between sensitivity and chemical specificity:

Fluorescence-based methods: Offer high specificity but require labeling, which can alter native biomolecule function, suffer from photobleaching, and provide limited structural information about the host molecule.
Label-free scattering methods (e.g., iSCAT): Provide high sensitivity and mass quantification but lack chemical specificity, making it difficult to distinguish specific biomolecules from background noise (e.g., dust, salts) or to determine secondary structures.
Raman-based methods (e.g., SERS): Provide chemical fingerprints but often require plasmonic "hot spots" (nanostructured surfaces) for signal enhancement, limiting their applicability to native biological environments and causing signal non-uniformity.
Existing Vibrational Imaging: Techniques like Stimulated Raman Excited Fluorescence (SREF) are limited to specific dye molecules with electronic pre-resonance.

The Gap: There is a lack of a label-free, far-field technique capable of achieving single-molecule sensitivity with quantitative chemical specificity (vibrational spectroscopy) to image protein mass, secondary structure, and binding kinetics in native environments.

2. Methodology: SRPSCAT Microscopy

The authors developed Stimulated Raman Photothermal Encoded Scattering (SRPSCAT) microscopy, which synergizes Stimulated Raman Scattering (SRS) and Interferometric Scattering (iSCAT).

Physical Principle:
1. SRS Excitation: Pump and Stokes beams induce stimulated Raman gain and loss, depositing vibrational energy ( $\omega_p - \omega_s$ ) into the target molecule.
2. Photothermal Effect: The deposited energy relaxes non-radiatively, causing a localized temperature rise ( $\Delta T$ ), thermal expansion ( $\Delta r$ ), and a change in the refractive index ( $\Delta n$ ).
3. Encoding: These photothermal changes modulate the scattering field of the molecule.
4. Detection (iSCAT): A probe beam interferes with the scattered light from the molecule and the reference field from the substrate interface. The photothermal modulation alters the interference pattern, encoding the vibrational spectroscopic information into the iSCAT signal.
Signal Advantage: Theoretical modeling indicates the modulation depth of SRPSCAT ( $\sim 10^{-5}$ ) is two orders of magnitude higher than direct SRS detection ( $\sim 10^{-7}$ ), enabling single-molecule sensitivity.
Instrumentation:
- Intensity-modulated pump/Stokes pulses combined with a collinear probe beam.
- 1-MHz modulation to reject low-frequency noise and achieve shot-noise limited detection.
- High-refractive-index substrates ( $n=1.7$ ) to maximize reference photons.
- Deep Learning Denoising: Utilization of a self-supervised SPEND (Self-Permutation Noise2Noise Denoiser) framework to recover signal-to-noise ratio (SNR) from complex system noise, improving SNR by 8-fold.

3. Key Contributions

Development of SRPSCAT: A novel microscopy modality that achieves quantitative, bond-selective imaging of single biomolecules without labels.
Chemical Specificity in Scattering: Successfully encodes vibrational spectroscopic information (Raman fingerprints) into an interferometric scattering signal, overcoming the chemical blindness of traditional iSCAT.
Single-Molecule Mass Quantification: Demonstrated a linear relationship between SRPSCAT intensity and molecular mass, achieving mass accuracy within 2% (error <5 kDa) for proteins ranging from 66.5 kDa to 950 kDa.
Secondary Structure Determination: Resolved protein secondary structures ( $\alpha$ -helix vs. $\beta$ -sheet) at the single-molecule level by detecting shifts in the Raman fingerprint window (e.g., Amide I band).
Dynamic Kinetics & Conformational Switching: Enabled real-time tracking of protein-protein binding kinetics and observed instantaneous conformational changes in de novo designed allosteric proteins.

4. Key Results

A. Performance Validation

Sensitivity: Successfully imaged single 29 nm polystyrene (PS) nanoparticles. The SRPSCAT signal showed clear Raman signatures at 3055 cm⁻¹, whereas standard SRS imaging showed no contrast under identical power conditions.
Size Dependence: Signal intensity followed a $d^3$ power law (interference-dominated), confirming the mechanism.
Chemical Differentiation: Distinguished between PS and PMMA nanoparticles of identical size based on their distinct Raman signatures.

B. Quantitative Mass Imaging

IgM Analysis: Differentiated single, double, and triple IgM clusters (950 kDa) from non-protein contaminants (dust/salts) which appeared in iSCAT but not in SRPSCAT.
Mass Calibration: Established a linear intensity-mass relationship across five proteins (BSA, transferrin, IgG, IgA, IgM).
Sensitivity: Detected single BSA molecules (66.5 kDa) in the C-H region.

C. Secondary Structure Elucidation

Differentiation: Distinguished Apoferritin (rich in $\alpha$ -helix, peak at 1656 cm⁻¹) from Fibronectin (rich in $\beta$ -sheet, peak at 1671 cm⁻¹).
Spectral Unmixing: In mixed samples where iSCAT could not differentiate the two proteins (due to similar mass), SRPSCAT successfully unmixed them based on their secondary structure Raman fingerprints.

D. Binding Kinetics and Dynamics

Protein-Protein Interaction: Monitored the binding of IgA to immobilized anti-IgA.
Kinetic Parameters: Calculated association ( $k_{on}$ ), dissociation ( $k_{off}$ ), and equilibrium dissociation constants ( $K_D$ ) for single molecules. For IgA/anti-IgA, $K_D$ was determined to be 6.1 nM, consistent with literature.
Heterogeneity: Revealed diverse binding behaviors (tight binding, transient binding, rapid switching) among individual molecules.

E. Conformational Dynamics of De Novo Proteins

System: Studied sr312, a computationally designed allosteric protein that switches from a triangular trimer (X3 state) to a square tetramer (Y4 state) upon effector binding.
Observation: SRPSCAT tracked the mass increase (from ~142 kDa to ~317 kDa) and observed an instantaneous transition (<30 ms) from X3 to Y4 states in real-time, matching the characteristic timescale of natural allostery.

F. Application to Viral Vectors

AAV Differentiation: Differentiated between empty and full Adeno-Associated Virus (AAV) capsids by detecting the C-H stretching band of DNA (2959 cm⁻¹) in full capsids, offering a label-free quality control method for gene therapy vectors.

5. Significance and Impact

Label-Free Structural Biology: SRPSCAT provides a powerful tool for studying biomolecular structures and interactions in their native environments without the perturbations caused by fluorescent labels.
Overcoming Photobleaching: Unlike fluorescence, the photothermal mechanism does not suffer from photobleaching, allowing for long-term observation of single molecules.
Universal Applicability: Unlike SERS, it does not require plasmonic substrates, making it applicable to a broader range of biological samples and nanoparticles.
Multimodal Insight: It uniquely combines the mass sensitivity of iSCAT with the chemical specificity of Raman spectroscopy, enabling the simultaneous determination of mass, secondary structure, and binding kinetics.
Future Potential: The technique holds promise for high-throughput screening of drug candidates, studying membrane protein dynamics, and analyzing the structural heterogeneity of viral vectors for clinical applications. The authors suggest further improvements via cryogenic conditions or advanced AI denoising could push sensitivity to single-bond detection.