🔬 condensed matter

Nanoscale Protein Diffusion in Supercooled Cryoprotectant Solutions

This study combines X-ray Photon Correlation Spectroscopy and small-angle X-ray scattering to demonstrate that ferritin proteins in supercooled glycerol-water solutions exhibit unexpectedly high mobility below 230 K compared to larger silica tracers, a phenomenon explained by a minimal fluctuating-friction model that accounts for local variations in effective friction.

Original authors: Maddalena Bin, Anita Girelli, Mariia Filianina, Mario Reiser, Sharon Berkowicz, Milla Åhlfeldt, Michelle Dargasz, Sonja Timmermann, Jaqueline Savelkouls, Takeshi Kawasaki, Shinji Saito, Federico Zonto

Published 2026-07-14

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Maddalena Bin, Anita Girelli, Mariia Filianina, Mario Reiser, Sharon Berkowicz, Milla Åhlfeldt, Michelle Dargasz, Sonja Timmermann, Jaqueline Savelkouls, Takeshi Kawasaki, Shinji Saito, Federico Zontone, Yuriy Chushkin, Fajun Zhang, Frank Schreiber, Michael Paulus, Christian Gutt, Fivos Perakis

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 a crowded dance floor where the music suddenly slows down to a crawl. Usually, when the room gets cold and sticky, everyone moves slower. But what if the tiny dancers on the floor started moving faster than the big, clunky dancers, even though the whole room is freezing? That's exactly the weird, counter-intuitive mystery scientists solved in this study.

The researchers were looking at ferritin, a tiny, hollow protein cage that acts like a microscopic storage tank for iron. They mixed these cages into a solution of water and glycerol (a thick, syrupy liquid often used to stop things from freezing). As they cooled this mixture down from room temperature to a chilly 210 K (which is about -63°C), they watched how fast the ferritin cages bounced around.

Here is the twist: In a normal liquid, if you know how thick the liquid is (its viscosity) and how big a particle is, you can predict exactly how fast it will move. This is a rule called the Stokes-Einstein relation. Think of it like a traffic law: if the road is icy and your car is heavy, you know you'll drive slowly.

The scientists tested this law using two types of "cars":

Ferritin: Tiny protein cages with a radius of 7.3 nm.
Silica nanoparticles: Larger, solid glass-like beads with a radius of 50 nm.

They used a super-powerful X-ray camera (called XPCS) to take snapshots of the ferritin moving, and a standard light-scattering method to watch the bigger silica beads.

The Big Surprise
When the temperature dropped below 230 K, the rules seemed to break. The big silica beads slowed down exactly as the traffic law predicted. But the tiny ferritin cages? They kept zooming along much faster than they were supposed to. By the time the temperature hit 210 K, the ferritin was moving about 2.7 times faster than the math predicted it should.

What It Wasn't
Before celebrating, the scientists had to make sure they weren't seeing things. They ruled out a few obvious culprits:

Did the protein shrink? No. The X-ray pictures showed the ferritin cages stayed the same size and shape; they didn't crumple or change structure.
Did the protein break apart? No. The data showed no signs of the proteins falling apart or clumping together.
Was it just the concentration? They checked, and the amount of protein in the solution wasn't high enough to cause the weird speed-up on its own.

The Best Guess: A Bumpy Road
So, why did the tiny cages move so fast? The authors suggest a "fluctuating-friction" model. Imagine the glycerol-water mixture isn't a smooth, uniform road. Instead, as it gets super cold, it becomes a bumpy path made of patches of "thick mud" and patches of "slippery ice."

The big silica beads are so large that they average out the whole road; they feel the average stickiness and move slowly. But the tiny ferritin cages are small enough to slip through the "slippery ice" patches before they get stuck in the "thick mud." They are essentially surfing the low-friction spots in the liquid.

The math they used to describe this suggests that at 210 K, the local friction the ferritin feels can vary by nearly 80% from the average. It's not that the liquid is less sticky overall; it's that the stickiness is uneven, and the tiny particles are lucky enough to find the fast lanes.

The Bottom Line
This study doesn't claim to have solved the entire mystery of how liquids behave near freezing. Instead, it provides a solid, measured benchmark for other scientists. It proves that at the nanoscale, in super-cold liquids, things don't always follow the big-picture rules. The tiny ferritin cages are the underdogs that found a way to dance faster than the music, and now we know exactly how much faster they were going.

Technical Summary: Nanoscale Protein Diffusion in Supercooled Cryoprotectant Solutions

Problem Statement
Vitrification during cryopreservation relies on a quantitative understanding of protein transport within deeply supercooled cryoprotectant solutions. While glycerol-water mixtures are widely used to suppress ice formation and extend the supercooled regime, the microscopic mechanisms governing molecular transport in these environments remain poorly understood. Specifically, the relationship between molecular transport and structural relaxation as the system approaches the glass transition is complex. In simple liquids, the Stokes–Einstein (SE) equation relates the translational diffusion coefficient $D(T)$ to solvent viscosity $\eta$ . However, this relation is known to break down in supercooled and glass-forming liquids, including supercooled water and protein solutions. The specific behavior of nanoscale protein dynamics relative to larger tracer particles in deeply supercooled cryoprotectants has remained an open challenge due to the scarcity of direct measurements at molecular length scales.

Methodology
To address these gaps, the authors employed a combination of X-ray Photon Correlation Spectroscopy (XPCS) and Small-Angle X-ray Scattering (SAXS).

Sample Preparation: Ferritin proteins (equine spleen) were prepared in glycerol–water mixtures (23 mol% glycerol) at a volume fraction of $\phi \approx 0.05$ . For comparative purposes, silica nanoparticle suspensions (radius 50 nm) were prepared in the same solvent at $\phi \approx 0.002$ .
Experimental Setup:
- SAXS: Performed at the DELTA synchrotron (12 keV) to monitor structural stability and protein conformation across temperatures from 293 K down to 193 K.
- XPCS: Conducted at the ESRF ID10 beamline (9 keV) to probe nanoscale dynamics. This technique measures temporal fluctuations in scattered X-ray intensity to determine collective diffusion on molecular length scales.
Temperature Control: Samples were mounted in a nitrogen gas cryostat, allowing measurements down to 210 K for XPCS (limited by freezing onset) and 193 K for SAXS.
Analysis: Diffusion coefficients were extracted by fitting the intensity autocorrelation functions $g_2(q,t)$ to a single-exponential decay model, yielding the relaxation rate $\Gamma(q) = Dq^2$ .

Key Results

Structural Stability: SAXS data revealed nearly invariant scattering profiles from 293 K to 193 K. The absence of significant changes in the scattering profile indicates that ferritin maintains its structural integrity, with no evidence of cold denaturation, aggregation, or nanocrystallization. The samples remained in a supercooled liquid state throughout the investigated range.
Diffusion Behavior: XPCS measurements of ferritin diffusion coefficients ( $D$ ) showed a continuous slowing down upon cooling, well-described by the Vogel–Fulcher–Tammann (VFT) relation.
Breakdown of Stokes–Einstein Scaling: When comparing ferritin diffusion to that of larger silica nanoparticles under identical solvent conditions, the data overlapped down to approximately 230 K. Below this temperature, ferritin exhibited a "relative enhancement" in mobility compared to the prediction based on hydrodynamic scaling (SE relation) derived from the silica tracers.
- At 210 K, the ratio of the measured ferritin diffusion to the SE-predicted reference ( $D/D_0$ ) increased to approximately 2.7.
- The VFT fitting parameter $B$ (related to effective activation energy) was significantly higher for ferritin ( $B \approx 26$ ) than for silica nanoparticles ( $B \approx 11$ ), indicating a stronger apparent temperature dependence for the nanoscale protein.
Modeling the Deviation: The authors applied a minimal fluctuating-friction model, where the local friction coefficient fluctuates due to temporal and spatial heterogeneities in the solvent. In the slow-fluctuation limit, the effective diffusion is enhanced relative to the mean friction. This model successfully reproduced the observed $D/D_0$ enhancement, requiring a local friction fluctuation amplitude ( $\delta$ ) of approximately 0.57 at 220 K and 0.79 at 210 K.

Significance and Claims
The paper claims to provide the first direct experimental benchmarks for molecular transport in deeply supercooled cryoprotectant solutions at nanoscale length scales. By demonstrating that XPCS can resolve protein diffusion at cryogenic temperatures, the study closes a critical gap in understanding protein dynamics in vitrified environments.

The primary finding is that nanoscale objects (ferritin) retain higher mobility upon cooling than larger tracers (silica) would predict via standard hydrodynamic scaling, with deviations emerging below ~230 K. The authors posit that this behavior arises from local variations in effective friction within the heterogeneous solvent environment. While the specific microscopic mechanism (e.g., dynamical heterogeneity vs. activated hopping) is not uniquely identified, the results offer direct constraints for future theoretical and simulation studies. The work establishes XPCS as a powerful tool for investigating molecular transport under cryogenic conditions, relevant to optimizing cryopreservation strategies for biological and pharmaceutical systems.

More like this