Mois : janvier 2025

A new study suggests that low-intensity pulsed ultrasound (LIPUS) can be beneficial in eliminating senescent cells through the recruitment and activation of immune cells [1].

The double-edged sword of the SASP

One of the characteristics of an aging organism is the accumulation of senescent cells. Various approaches are being developed to remove or neutralize those cells.

Senescent cells produce the senescence-associated secretory phenotype (SASP), a cocktail of chemokines, pro-inflammatory cytokines, growth factors, and proteases [2]. While the SASP might have many deleterious effects, it can be a double-edged sword. Some of these many molecules have positive effects, such as attracting immune cells, which can eliminate senescent cells [3]. Precise modulation of the SASP can be a potent strategy for senescent cell elimination.

The authors of this study turned to ultrasound as a possible non-invasive therapeutic tool to eliminate senescent cells. Previous studies observed that LIPUS has positive effects on many types of tissues, including promoting wound healing or bone repair [4] and regulating the secretion of inflammation-associated cytokines [5].

Therefore, the authors of this study hypothesized that “LIPUS can modulate the secretion of SASP in senescent cells and thereby manipulate these cells” or aid in attracting immune cells.

Eating up the old cells

The authors cultured human male fibroblast cells as their research model. They made these cells replicatively senescent by allowing them to grow and divide multiple times. They divided the cells into two groups: ‘late cells,’ which had replicated many times but had not yet reached senescence, and ‘early cells,’ which had not replicated many times.

Following 20 minutes of stimulation with LIPUS, the researchers observed a marker of senescent cells, SA-β-gal, to be selectively increased in the ‘late cells’ but not in the ’early cells.’

When the researchers tested the impact of LIPUS on the expression of multiple SASP molecules, they learned that “LIPUS stimulation specifically increased the expression of immune cell attraction markers in the ‘late cells.’”

This increased expression led to an increased migration of immune cells, specifically monocytes and specific families of macrophages, towards these stimulated cells. Ultimately, it resulted in the ingestion and elimination of the ‘late cells’ by macrophages in a process called phagocytosis.

The molecules behind the scenes

In the next steps of their research, the authors investigated the molecular mechanism behind LIPUS’s selective stimulation of the SASP.

After excluding other possibilities, they tested the involvement of reactive oxygen species (ROS) since previous reports suggested increased ROS production following LIPUS stimulation [6]. These researchers confirmed that LIPUS stimulation increased intracellular ROS generation in the ‘late cells’ and observed that ROS production was required to increase SA-β-gal activity in LIPUS-stimulated ‘late cells.’

Further, they investigated the molecules that regulate the expression of SASP factors, focusing on two in particular: NF-κB and p38. NF-κB is a transcription factor family member that regulates gene expression, and p38 increases NF-κB activity.

Their experiments suggested that in ‘late cells,’ LIPUS stimulation leads to ROS-dependent activation of the p38-NF-κB pathway, activating immune cell-attracting SASP factors and leading to immune cell migration.

ROS plays a significant role in this process, but how did LIPUS stimulation cause the ROS generation? The researchers generated a few hypotheses. First, they learned that the LIPUS stimulation-generated production of extracellular ROS was not significant. Instead, LIPUS generated intracellular ROS via an enzyme called NOX4. NOX is a family of enzymes located in lipid rafts, special compartments on the plasma membranes that surround cells.

Further experiments showed that LIPUS stimulation led to perturbations in the structure and organization of the cellular membrane, creating transient pores and resulting in increased permeability, affecting the formation and localization of lipid rafts. This leads to NOX activation and ROS generation. This occurred only in ‘late cells’, whose membrane composition differs from that of ‘early cells.’

Reversing skin aging

At the end of their study, the researchers used an in vivo model of mouse skin aging to test whether LIPUS could be an efficient tool to remove senescent cells by regulating the SASP in a living organism.

After UVA-induced skin aging, LIPUS was applied for five days, and skin tissue was analyzed 10 days later. Neither UVA nor LIPUS treatment were found to impact body weight nor major organs of these mice. However, as expected, UVA resulted in extensive and deep wrinkles on the applied area and increased the levels of senescence markers.

LIPUS treatment increased SASP markers for immune cell attraction in a UVA-induced skin aging model. This increase translated into more attraction of immune cells than with UVA irradiation alone. At the same time, LIPUS treatment significantly reduced the number of cells with senescence markers, suggesting a decrease in these cells.

The researchers summarized that “these data suggest that senescent cells could be eliminated by macrophage infiltration via LIPUS stimulation.”

Optimizing for clinical use

LIPUS is a technology that can be easily applied in the clinic. Those researchers propose that it can be used to help remove senescent cells, possibly combined with senolytic treatment.

However, before this therapy can enter the clinic, LIPUS parameters need to be optimized and tested for side effects, as it is known that different LIPUS parameters can elicit different effects in different types of cells. Additionally, its limitations, such as penetration efficiency, and patients’ aged immune systems, which might not be as effective in clearing senescent cells, need to be considered.

Literature

[1] Gwak, H., Hong, S., Lee, S. H., Kim, I. W., Kim, Y., Kim, H., Pahk, K. J., & Kim, S. Y. (2025). Low-Intensity Pulsed Ultrasound Treatment Selectively Stimulates Senescent Cells to Promote SASP Factors for Immune Cell Recruitment. Aging cell, e14486. Advance online publication.

[2] Watanabe, S., Kawamoto, S., Ohtani, N., & Hara, E. (2017). Impact of senescence-associated secretory phenotype and its potential as a therapeutic target for senescence-associated diseases. Cancer science, 108(4), 563–569.

[3] Burton, D. G. A., & Stolzing, A. (2018). Cellular senescence: Immunosurveillance and future immunotherapy. Ageing research reviews, 43, 17–25.

[4] Schortinghuis, J., Bronckers, A. L., Stegenga, B., Raghoebar, G. M., & de Bont, L. G. (2005). Ultrasound to stimulate early bone formation in a distraction gap: a double blind randomised clinical pilot trial in the edentulous mandible. Archives of oral biology, 50(4), 411–420.

[5] Li, J. K., Chang, W. H., Lin, J. C., Ruaan, R. C., Liu, H. C., & Sun, J. S. (2003). Cytokine release from osteoblasts in response to ultrasound stimulation. Biomaterials, 24(13), 2379–2385.

[6] Duco, W., Grosso, V., Zaccari, D., & Soltermann, A. T. (2016). Generation of ROS mediated by mechanical waves (ultrasound) and its possible applications. Methods (San Diego, Calif.), 109, 141–148.

Scientists have presented GroceryDB, an open-access online database that measures the degree of processing of tens of thousands of products sold in three major US grocery chains [1].

What is ultra-processed food?

While there is no universally accepted definition, the NOVA food classification system is widely used, and it defines ultra-processed food as “industrially manufactured food products made up of several ingredients including sugar, oils, fats and salt (generally in combination and in higher amounts than in processed foods) and food substances of no or rare culinary use (such as high-fructose corn syrup, hydrogenated oils, modified starches and protein isolates).”

In other words, ultra-processed food involves breaking down “real food” and creating something mostly new, such as instant soup or candy, with a nutrient makeup unlike anything encountered in nature. This food equivalent of Frankenstein’s monster fools the gratification circuits developed by millions of years of evolution, tricking people into ingesting too much unhealthy and too little healthy food while generally overeating.

Even though ultra-processed food encompasses a wide range of different products, from beverages to sausages, as a category, it has been consistently linked to adverse health outcomes, such as cancer [2], cardiovascular disease [3], and obesity [4]. Alarmingly, people in developed nations consume up to 60% of their calories from ultra-processed food.

Getting to know your food

It is not always straightforward to know the processing of foods in a local grocery store. It is sometimes possible to make an educated guess, such as with sugary beverages, but other products are not as obvious. Unfortunately, minimalistic food labels don’t offer much help.

The group led by Prof. Albert-László Barabási, whichi included researchers from the Harvard Medical School and Northeastern University, has been studying ultra-processed food for several years. In their new paper published in Nature Food, the researchers describe GroceryDB, an open online database that contains information on more than 50,000 products offered by three major US chains: Walmart, Whole Foods Market, and Target.

Last year, the group published FPro, a food processing score that they had developed using machine learning techniques, that translates the nutritional content of a food item into its degree of processing. FPro, which powers GroceryDB, is mostly based on NOVA (it was trained to predict a NOVA category of the product from its ingredients) but can accommodate other food processing classification systems. The reliance on the list of nutrients is due to several reasons, such as that in unprocessed food, their quantities are constrained by biochemistry-determined physiological ranges.

Browsing GroceryDB, available at Truefood.tech, and comparing favorite foods to less or more processed alternatives is a captivating pastime. One of the main takeaways is that the degree of food processing can vary a lot even within a single category. The paper provides several examples, starting with bread.

The tale of the two cheesecakes

A multi-grain bread from Manna Organics, sold by Whole Foods, which mostly contains whole wheat, barley, and brown rice without any additives, salt, oil, and yeast, has an FPro of 0.314 (the index ranges from 0 to 1). The two less health-oriented chains, Walmart and Target, both carry Aunt Millie’s and Pepperidge Farmhouse breads (FPros of 0.732 and 0.997, respectively) with ingredients including soluble corn fiber, sugar, resistant corn starch, wheat gluten, and monocalcium phosphate.

The researchers saw a similar picture with yogurts: Seven Stars Farm yogurt made from grade A pasteurized organic milk has an FPro of 0.355. Siggi’s yogurt, with an FPro of 0.436, uses pasteurized skim milk, which, according to the paper, requires more food processing to eliminate fat. The two pale in comparison to Chobani Cookies and Cream yogurt with its whopping FPro of 0.918, thanks to loads of cane sugar and multiple additives such as caramel color, fruit pectin, and vanilla bean powder.

Some food categories are inherently highly processed, so it is unlikely to find cookies with a low FPro. However, even in those categories, the distribution of FPro scores is quite wide, and healthier alternatives are available. Unsurprisingly, the prevalence of less processed food was much higher in Whole Foods Market than in the other two chains.

The researchers highlighted one of the reasons for the abundance of ultra-processed food: processing decreases the cost per calorie. Across all GroceryDB categories, a 10% increase in FPro leads to an 8.7% decrease in the price per calorie. In some categories, the decline is much steeper, however, with milk, the relationship is reversed, probably due to more expensive plant milks also being more processed.

To illustrate the makeup of the FPro score for every product in GroceryDB, its ingredients are presented as a tree. This allows accounting for “ingredients of ingredients,” such as in this example with two cheesecakes. While both are highly processed, the one on the left has an FPro of 0.953 and the one on the right – 0.720. The former, along with many additives, contains sour cream which, in turn, contains a number of ingredients, such as modified food starch.

The highly processed ingredients are designated by red dots. The researchers mention, however, that this does not necessarily mean they are harmful. For example, xanthan gum, guar gum, and locust bean gum are considered generally safe. The purpose of GroceryDB is to allow people to dig into the ingredients of pantry staples and make informed decisions.

Informing the public’s choices

“GroceryDB serves as a proof of concept, demonstrating the potential of accessible, algorithm-ready data to advance nutrition research,” said Dr. Giulia Menichetti, a Principal Investigator and Junior Faculty at Harvard Medical School, and a co-author of the study. “This is especially significant in a field where much of the work still depends on labor-intensive manual curation, relying on descriptive definitions that suffer from poor inter-rater reliability and lack of reproducibility.”

“While the general population is increasingly aware of the potential health impacts of ultra-processed foods, they lack the knowledge to distinguish minimally processed foods, which have no known health consequences, from ultra-processed ones,” said Prof. Barabási. “Here, we set out to offer this resource by measuring the degree of processing for the foods that constitute a significant fraction of the US food supply. Most importantly, through this online resource, consumers are empowered to replace the ultra-processed foods they consume with brands that are less processed.”

Menichetti, too, underscored the potential societal benefits of their project. “Our vision with GroceryDB is not just to build a database, but to catalyze a global effort toward open-access, internationally comparable data that advances nutrition security and ensures equitable access to healthier food options for all,” she said.

Another nutrition scientist, Barry M. Popkin, Distinguished Professor of Nutrition at the University of North Carolina, who was not involved in this study, voiced some critique regarding its design: “Rather than doing an exact study of the ingredients list to find those colors, flavors and other additives identified in NOVA as identifying ultra-processed food, they guess on a set of foods that they were ultra-processed and then the machine identified the other ultra-processed foods.”

However, according to Menichetti, the approach suggested by Popkin “is not currently practical from an algorithmic perspective, due to the poor standardization of ingredient lists worldwide and the absence of definitions tied to robust, measurable variables across food composition databases.”

“Implementing such an approach,” she noted, “would have required significant manual curation, more than double the funding and time, and the incorporation of subjective opinions into the classification process. We see these challenges firsthand with our friends at Open Food Facts, a non-profit initiative powered by thousands of volunteers globally, which grapples with these same limitations daily.”

Literature

[1] Ravandi, B., Ispirova, G., Sebek, M., Mehler, P., Barabási, A. L., & Menichetti, G. (2025). Prevalence of processed foods in major US grocery stores. Nature Food, 1-13.

[2] Fiolet, T., Srour, B., Sellem, L., Kesse-Guyot, E., Allès, B., Méjean, C., … & Touvier, M. (2018). Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort. bmj, 360.

[3] Srour, B., Fezeu, L. K., Kesse-Guyot, E., Allès, B., Méjean, C., Andrianasolo, R. M., … & Touvier, M. (2019). Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (NutriNet-Santé). bmj, 365.

[4] Hall, K. D., Ayuketah, A., Brychta, R., Cai, H., Cassimatis, T., Chen, K. Y., … & Zhou, M. (2019). Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell metabolism, 30(1), 67-77.

In Cell Metabolism, researchers have described how a microRNA (miRNA) derived from exosomes generated by human embryonic stem cells (hESCs) restores function and fights senescence in cell cultures and mice.

Looking for a better senomorphic

This study begins with a discussion of cellular senescence and its role in aging, focusing on the main approaches to dealing with it: senolytics, which kill senescent cells, and senomorphics, which transform them. Senolytics may have side effects because their number goes up with advanced age [1], and some of them are necessary for life. Even such techniques as directly affecting the SASP may have adverse impacts on the immune system [2].

While senomorphics appear to be a viable strategy, they are still in their infancy. One potential strategy involves exosomes: messenger particles that are regularly sent by cells. Previous work with hESC-derived exosomes (hESC-Exos) has found that their contents are instrumental in rejuvenating multiple tissues [3]. Therefore, these researchers decided to examine them as a senomorphic, discovering which of their many components is best at restoring senescent cells.

Cells regained the power to proliferate

The first experiment was a basic test of hESC-Exos on IMR-90 cells, a line of human fibroblasts. After 30 divisions, this cell line remains youthful, but at 50, it has reached replicative senescence. However, treating those senescent cells with hESC-Exos almost entirely reversed their senescent biomarkers, downregulating SASP-related genes such as those for inflammatory interleukins, restoring cellular proliferation, and sharply decreasing the senescence biomarker SA-β-gal. Other genes related to senescence were inhibited, while those relating to proliferation were enhanced.

These findings were recapitulated at the single-cell level. First, the researchers created a population of human cells that were modified to express the fluorescence protein along with the senescence marker p21. Then, they drove these cells senescent by exposing them to doxycycline, after which they exposed some of them to hESC-Exos.

Compared to an untreated control group, the treated cells had substantially less p21-related fluorescence and less SA-β-gal, and some of them regained the ability to proliferate. Just like with the replicatively senescent cells, hESC-Exos diminished senescence-related gene expression and enhanced proliferation-related expression instead.

Mice regained youthful attributes

From the 20th month to the 30th month of life, wild-type mice on a normal diet were injected with hESC-Exos. Compared to a control group, the exosome-treated mice performed better on both fixed and accelerating rotarod tests, had higher body weight, and retained normal bladder activity. In the Morris water maze test, they both found the platform more quickly and remembered its location more accurately. They retained their hair color and skin smoothness as well.

Biomarkers of senescence, just as in the cellular culture, were substantially reduced in the treated mice. Inflammatory factors, such as cytokines and TNF-α, were also substantially reduced in their circulation, and γ-H2AX, a marker of genomic damage, was notably reduced. Just as in cellular cultures, the mice’s biomarkers of cellular proliferation were improved. Overall, the researchers found the treated mice to be substantially rejuvenated as a whole.

Finding the key molecule

The researchers hypothesized that much of this rejuvenative power can be distilled down to the exosomes’ individual components. They selected one of them, miR-302b, which is abundantly found in hESC-Exos and is documented to play a role in cellular proliferation [4]. However, it had remained untested against aging.

This particular miRNA was found to directly regulate expression of the Cdkn1a gene, which is related to cellular senescence. Exposing IMR-90 cells to miR-302b recapitulated the effects of hESC-Exos, reducing senescence and promoting proliferation.

Encouraged, the researchers then turned to mice. This time, they injected 25- to 30-month-old mice with artificially transfected exosomes containing miR-302b. They found that this approach recapitulated the results found in hESC-Exos as well, reducing inflammatory factors and SA-β-gal, substantially improving the results of rotarod and Morris water maze tests, and restoring cellular proliferation.

Llifespan itself was also improved by this approach. The median lifespan of the mice treated with miR-302b was 137 days greater than that of the control group. While the difference was not significant, the effect seemed to be stronger in males.

While hESC-Exos and miR-302b were not compared directly, they appear to be largely similar in terms of their effects. Still, this is a cell and mouse study, and further work needs to be done to determine if this approach is safe for clinical use. It is also not known which of these approaches will be the most scalable and suitable for mass production.

Lifterature

[1] Huang, W., Hickson, L. J., Eirin, A., Kirkland, J. L., & Lerman, L. O. (2022). Cellular senescence: the good, the bad and the unknown. Nature Reviews Nephrology, 18(10), 611-627.

[2] Zhang, L., Pitcher, L. E., Yousefzadeh, M. J., Niedernhofer, L. J., Robbins, P. D., & Zhu, Y. (2022). Cellular senescence: a key therapeutic target in aging and diseases. The Journal of Clinical Investigation, 132(15).

[3] Bi, Y., Qiao, X., Liu, Q., Song, S., Zhu, K., Qiu, X., … & Ji, G. (2022). Systemic proteomics and miRNA profile analysis of exosomes derived from human pluripotent stem cells. Stem Cell Research & Therapy, 13(1), 449.

[4] Subramanyam, D., Lamouille, S., Judson, R. L., Liu, J. Y., Bucay, N., Derynck, R., & Blelloch, R. (2011). Multiple targets of miR-302 and miR-372 promote reprogramming of human fibroblasts to induced pluripotent stem cells. Nature biotechnology, 29(5), 443-448.