It’s possible that I shall make an ass of myself. But in that case one can always get out of it with a little dialectic. I have, of course, so worded my proposition as to be right either way (K.Marx, Letter to F.Engels on the Indian Mutiny)
Monday, March 02, 2026
The physics of a squeak
High-speed imaging shows how rubber sneaker squeaks arise from supersonic detachment pulses
Harvard John A. Paulson School of Engineering and Applied Sciences
Squeaks from soft-on-rigid interfaces, like shoes on a smooth floor, are driven by opening slip pulses that rapidly detach and reattach the interface at near-supersonic speeds. The audible pitch is determined by how frequently these pulses repeat.
Tread patterns act as waveguides, trapping these pulses into a regular, periodic cycle. This geometric confinement produces a clear, well-defined tonal frequency.
The rupture dynamics of these pulses share key features with fracture fronts observed in tectonic faults, offering a surprising new model for studying earthquake mechanics.
Basketball shoes on a gym floor, bicycle brakes in need of a tune-up, or the squeal of tires are everyday examples of squeaking sounds. Ever wonder why that sound occurs?
Such sounds have long been attributed to stick-slip friction, or a cycle of intermittent sticking and sliding between surfaces. While this framework explains many rigid-on-rigid systems such as door hinges, it does not fully capture the physics of soft-on-rigid interfaces, like shoes on a floor.
To shed light on this little-understood physical process, researchers at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS), in collaboration with the University of Nottinghamand the French National Center for Scientific Research, used high-speed imaging to investigate the dynamics of soft solids sliding rapidly on rigid substrates. In a study published in Nature, the team led by first author Adel Djellouli, a postdoctoral fellow in the lab of Katia Bertoldi, the William and Ami Kuan Danoff Professor of Applied Mechanics at SEAS, reports that squeaking emerges from a previously unseen mechanism.
“This project started with a simple question: why do basketball shoes squeak?" said Djellouli. "We combined total internal reflection imaging with cameras capturing up to one million frames per second to visualize the evolving contact between rubber and glass. To drive sliding, we adapted a configuration conceptually similar to Leonardo da Vinci’s friction experiments from the 15th century."
The team’s discoveries could surface new ways to engineer and control advanced materials. "Tuning frictional behavior on the fly has been a long-standing engineering dream," Bertoldi said. "This new insight into how surface geometry governs slip pulses paves the way for tunable frictional metamaterials that can transition from low-friction to high-grip states on demand.”
Using high-speed optical imaging and synchronized audio measurements, the researchers directly visualized the contact interface between soft rubber and rigid glass. They discovered that sliding does not proceed uniformly. Instead, motion localizes into what they observed as supersonic opening slip pulses: rapid, wrinkle-like detachment fronts that propagate along the interface at high speeds.
They discovered that the audible squeak is not produced by random stick-slip events, as conventional wisdom might suggest. Rather, the squeaking sound frequency is set by the repetition rate of these propagating pulses. Beyond the sound, the study found that these opening slip pulses significantly impact the overall frictional resistance.
In another surprising twist, the high-speed images revealed an unexpected phenomenon that accompanied the squeak: lightning. Lab experiments showed that in some instances, the slip pulses are triggered by triboelectric discharges — miniature lightning bolts caused by the friction of the rubber.
Geometry also plays a decisive role in sound generation, the researchers found. During lab experiments, when rubber blocks with flat surfaces were slid along glass, the pulses were complex and irregular, resulting in broadband noise that resembled a rushing or swooshing sound. But thin ridges dramatically altered the dynamics: the pulses became confined and periodic, producing more focused pitches.
This geometric confinement forces the pulse repetition rate to lock into a characteristic frequency determined by the system dimensions. The researchers observed a scaling relationship, in which the squeak frequency depends primarily on the block height — a relationship so precise, that the researchers were able to design rubber blocks of varying heights to play the Star Wars theme song by hand.
" We were surprised that tiny surface features could so strongly reorganize frictional motion," said co-author Gabriele Albertini, of University of Nottingham. "These results challenge the assumption that friction can be fully captured by simplified one-dimensional models and highlight the critical role of interface dimensionality."
The implications extend beyond squeaky shoes. The physics governing these slip pulses, specifically how two surfaces move relative to each other, mirror earthquake dynamics, where ruptures and slip pulses propagate along tectonic faults at extremely high speeds, approaching and sometimes exceeding the speed of sound.
"These results bridge two fields that are traditionally disconnected: the tribology of soft materials and the dynamics of earthquakes," said co-author Shmuel Rubinstein, professor of physics at Hebrew University and visiting professor at SEAS. “Soft friction is usually considered slow, yet we show that the squeak of a sneaker can propagate as fast as, or even faster than, the rupture of a geological fault, and that their physics is strikingly similar.”
The research was carried out through an international collaboration among Harvard University (USA), CNRS/Université du Mans (France), the Hebrew University of Jerusalem (Israel), and the University of Nottingham (UK), with support from the U.S. National Science Foundation (Harvard MRSEC; and an NSF Graduate Research Fellowship), the Simons Collaboration on Extreme Wave Phenomena Based on Symmetries, BASF, and the Swiss National Science Foundation.
In this illustration, all 13 classical vitamins grow from a tree rooted in DNA, representing a nutritional genomics framework that connects genetic information to precision vitamin therapies.
By the 1890s, just as the medical world was beginning to understand that invisible microbes caused infectious diseases, the idea that a missing nutrient could kill you was even harder to fathom.
Then the evidence started piling up. Dutch physician Christiaan Eijkman noticed chickens fed polished rice developed beriberi-like paralysis, cured by adding back the husks. It was the first clue that what we now call vitamin B1 was essential. Naval ships were losing a third of their sailors to scurvy on long voyages. Give them citrus fruit, rich in vitamin C, and deaths dropped to zero.
Over the following decades, researchers began searching for other essential nutrients that the body can't produce on its own. They called them vitamins, and the 'vitamin hunters' who identified them saved millions of lives, earning over a dozen Nobel Prizes along the way. By 1948, all 13 classical vitamins were known.
With readily available vitamins, much of our food started getting supplemented. While this had the obvious benefit of reducing dietary deficiencies, one downside was that people stopped paying attention. Vitamins became something you could buy at the grocery store without a clear rationale of how much, why, or the underlying mechanism for their benefits. There are case reports of physicians giving patients vitamin cocktails for various diseases, and sometimes it works and sometimes it doesn't. The literature is messy, but there are enough glimmers of hope that there's a real signal in there.
Our lab wants to tackle vitamins more systematically. In our new paper in Cell, we flip the traditional approach to finding cures. Instead of starting with one disease and spending a decade searching for a treatment, we start with a treatment that's already safe and accessible and ask: what are all the genetic diseases it can treat?
A genome-wide screen for vitamin-treatable diseases
We started with vitamins B2 and B3 because they're common components of cell culture media that can easily be omitted or added. Taking advantage of this, we performed genome-wide CRISPR screens in K562 cancer cells in media with or without one of these B vitamins. We then consulted the OMIM database of known monogenic diseases to nominate candidates from the genes that grew well with the vitamin but poorly without it. This yielded a list of dozens of disease genes that could be rescued by high vitamin B2 or B3 levels.
For B2, the top hit was SLC52A2, a riboflavin transporter whose mutations cause Brown-Vialetto-Van Laere syndrome, a disorder already known to be treatable with high-dose riboflavin. Another top hit, FLAD1, also corresponds to a riboflavin-responsive condition. Two known positive controls sitting right at the top of our list told us the framework works.
But the screen also picked up GPX4, an enzyme that prevents a type of cell death called ferroptosis, which was unexpected. We validated the connection in cell culture and then in a mouse model, where GPX4-deficient mice on a B2-deficient diet showed accelerated motor decline. This represented a brand new vitamin-disease interaction.
NAXD and vitamin B3
From the vitamin B3 screen, the top hit was the NAXD gene. Mutations in it cause a very rare disease, not a lot of labs study NAXD, and it was completely new to us at the time. What we learned is pretty fascinating, and it led us on a years-long, often challenging, investigation.
The NAXD protein is a metabolic proofreading enzyme. Hundreds of enzymes rely on NAD/NADH (carriers of electrons for oxidation-reduction reactions) for everything from basic energy production to DNA repair. But sometimes, whether through mistakes by enzymes like GAPDH, or due to lower pH and increased temperature, NADH gets erroneously hydrated into something called NADHX. If that's not corrected, it can cyclize into a dead-end product called cyclic NADHX. These rogue metabolites can competitively block the enzymes that normally use NAD and NADH.
NAXD's job is to convert NADHX back to usable NADH, correcting the errors before they cause damage. If people lack NAXD function, they develop severe neurodevelopmental disease that is often fatal in early childhood, though the timeline varies depending on the mutation.
We validated the screen hit in cell culture first, finding that NAXD knockout cells grew poorly in B3-deficient media, and this phenotype was rescued by high B3. But cancer cells in a dish aren't representative of what happens in a living body. We needed an animal model, and no mouse model of NAXD disease existed. That had been a major roadblock for the field.
We used CRISPR to generate knockout mice with frameshift mutations in exon 2 of Naxd. The mice were born indistinguishable from their littermates, but within a day or two, their health deteriorated rapidly.
Understanding and reversing NAXD disease
It was a challenge, but we optimized LC-MS methods to detect all the relevant metabolites — NAD, NADH, R-NADHX, S-NADHX, and cyclic NADHX. In the knockout mice, we saw massive accumulation of every form of NADHX across all tissues, while these metabolites were essentially undetectable in wild-type animals.
We were particularly intrigued to find that NAD was specifically depleted in the brain and skin, which is where the disease shows up in human patients. Other organs like liver, heart, and kidney had the error metabolites piling up but were maintaining their NAD levels. Why some tissues are more vulnerable than others is still something we're working to understand.
Broader metabolomics on knockout brains and MALDI-TOF imaging revealed that the most dramatically depleted compounds beyond NAD itself were serine and phosphoserine. Serine depletion was most severe in cortical regions of the brain. But single-nucleus RNA sequencing revealed that the most affected cell types weren't neurons. They were brain endothelial cells, mural cells, and astrocytes – the brain's vasculature. Endothelial cells showed a strong cell death signature that was completely reversed by B3 treatment.
That brought us back to our original question: can the vitamin actually treat this disease?
First we tested what happens when you remove it entirely. We put pregnant mothers on a B3-deficient diet. On regular chow, about a quarter of pups were knockouts, as expected. On the deficient diet, across about 70 pups, not a single knockout was born alive.
Then we tried the opposite. Getting a vitamin to actually increase in the tissues of a tiny newborn mouse pup is harder than you'd think, and we spent over a year on this problem alone. But when we got it right, with daily intraperitoneal injections of nicotinamide riboside (one form of vitamin B3) starting at birth, the rescue was dramatic. Knockout mice were indistinguishable from their littermates at day 50 and beyond. Eight of nine treated animals continued to survive, more than a 40-fold improvement in lifespan. Every aspect of disease we measured was rescued, including body weight, brain pathology, cell death.
When we delayed treatment by just two days, postnatal day 2 instead of day 0, there was no benefit. That's why we advocate including NAXD in neonatal screening panels.
A new framework for vitamin biology
There are clinical case reports where NAXD patients were given supplements including vitamin B3 and some appeared to improve. Those are anecdotal, but they're consistent with what we've found. Translating the right dosing from mice to humans will require careful work, but we're eager to get these findings to the patient and clinical communities as quickly as possible.
We also hope this work inspires others to revisit vitamins with fresh eyes. We'd love some company doing this. Our screens nominated dozens of additional diseases that may respond to B2 or B3 therapy, and we've only tested two out of more than 50 known micronutrients.
More broadly, we think the field should be thinking more creatively about therapies, not just developing drugs, but asking whether there are things we can change about how we live to treat disease. Whether it's what we breathe or what we eat.
Someday, maybe every newborn's genome gets sequenced at birth, and they receive custom dietary recommendations based on their genetics. We see this as picking up where the original vitamin hunters left off, bringing modern tools to the question of which vitamins can treat which diseases, and the mechanisms behind them.
Garg, A., Blume, S.Y., Huynh, H., Barrios, A.M., Karabulut, O.O., Zhao, Q., Midha, A.M., Turner, A., Resnick, B.V., Chen, X., Agrawal, A., Kim, J., Chen, L., Ran, Q., Ryan, A.M., Larson, R.C., Negahban, M., Nelson, S.C.K., Yang, A.C., Traglia, M., Thomas, R., Sun, R., Paredes, M., Corces, R., Lin, H., & Jain, I.H. (2026). Nutritional Genomics Uncovers Vitamin B3 Therapy as Curative for NAXD Disease. Cell. https://doi.org/10.1016/j.cell.2026.01.022
Authors:
Senior author Isha Jain (X: @ishahjain) is an Arc Institute Core Investigator, an Associate Investigator at Gladstone Institutes and an Associate Professor at UCSF.
First authors:
Ankur Garg was a postdoctoral fellow in the Jain Lab at Gladstone Institutes and is now a staff scientist at Arc Institute
Skyler Blume is a Research Associate in the Jain Lab at Gladstone Institutes.
A research group at the University of Lausanne (Unil) has identified a new mechanism that exposes the vulnerability of tumor cells when they are deprived of vitamin B7.
The ability of cells to adapt to fluctuations in nutrient availability is essential to life. Yet, some cells become highly dependent on glutamine, an amino acid that plays a central role in cellular metabolism. This nutrient provides the essential building blocks for protein and DNA synthesis, and cells stop proliferating without it.
This is the case for cancer cells: “glutamine addiction” is a well-known vulnerability of tumors, but many cancers manage to bypass it. In a study published in the journal Molecular Cell, the team led by Alexis Jourdain, assistant professor in the Department of Immunobiology (DIB) at Unil’s Faculty of Biology and Medicine (FBM), advances the understanding of cellular mechanisms that until now were poorly understood.
Cells, show your papers!
The work led by Dr. Miriam Lisci, a postdoctoral researcher in Prof. Jourdain’s laboratory, highlighted carbon-rich molecules —particularly pyruvate — that allow cells to continue dividing even in the absence of glutamine. The experiments reveal that this effect requires a mitochondrial enzyme called pyruvate carboxylase, which itself needs vitamin B7 (or biotin) to function. Without this vitamin, the enzyme is inactive and cells remain stalled. Biotin thus acts as a true “metabolic license,” enabling pyruvate to fuel the cells’ energy cycle and compensate for the lack of glutamine.
The key role of the FBXW7 gene
The researchers also discovered a previously unknown role for FBXW7, a gene known for its involvement in many cancers. “When FBXW7 is mutated — a situation that is frequent in certain cancers — pyruvate carboxylase partially disappears, pyruvate can no longer be used efficiently, and cells become dependent on glutamine,” explains Miriam Lisci, first author of the article. The scientists even showed that certain FBXW7 mutations found in patients directly induce this metabolic addiction. The results were obtained thanks to collaborations with the FBM’s metabolomics and proteomics platforms, as well as with the team of Prof. Owen Skinner at Northeastern University in the United States.
This publication also highlights why some therapies targeting glutamine fail: cancer cells can activate alternative metabolic pathways. “In the longer term, this research opens up new avenues for better understanding the metabolic vulnerabilities of cancers and for designing innovative therapeutic strategies that take into account the great metabolic flexibility of tumor cells, notably by targeting several metabolic pathways simultaneously,” concludes Alexis Jourdain, senior author of the study.
Functional nutrient-genetic profiling reveals biotin and FBXW7 are essential to bypass glutamine addiction
Article Publication Date
25-Feb-2026
COI Statement
AAJ is listed as an inventor on a patent related to system Xc- inhibition. OSS previously served as a consultant for Handshake AI. The other authors declare no conflict of interest.
