On July 16, in Science, a Stanford team's study found that in older mice, the signal through EP2 weakened the work of tissue macrophages. When the authors disabled this receptor only in macrophages, they again cleared aging neutrophils, and the animals' heart, muscle, memory, and other organ functions were preserved. Neutrophils are the most numerous and shortest-lived white blood cells. They are the first to attack an infection and then must quickly disappear: tissue macrophages, the permanent immune cells of organs, engulf and break them down inside themselves. In one day, the human body produces more than 100 billion neutrophils. Macrophages must continuously free organs from this mass of cells. With age, this process failed in mice. In the liver, spleen, bone marrow, and other organs, neutrophils with signs of cellular aging accumulated. They released damaging enzymes and triggered NETosis - the release of networks of DNA and proteins that neutrophils use against microbes. In tissues, these networks and released substances gave inflammatory stress to neighboring cells. The authors found a brake in the macrophages themselves. Prostaglandin E2 - a signaling fatty molecule, the amount of which grows during inflammation - turns on the EP2 receptor on their surface. EP2 disables the first step of cleanup: weakens the work of integrins, proteins that grab. The macrophage worse holds the neutrophil and does not engulf it in a phagolysosome, an internal bubble for breaking down prey. A review of aging macrophages in vascular plaque had already described the loss of cleaning function in one organ; the Tan team named the molecular brake of such a failure in several tissues at once. The team genetically disabled EP2 only in tissue macrophages. In old mice, the clearance of neutrophils returned, and the indicators of fragility, loss of muscle mass, obesity, heart function, memory, and inflammation approached those of young animals. An experimental EP2 inhibitor in 22-month-old mice also reduced the number of aging neutrophils and restored their absorption by macrophages over two months. One path to therapy for aging is to directly remove aging cells. This work suggests restoring the immune system's cleanup of one particularly massive class of cells. In human liver and heart tissues, the authors saw the same age-related pattern: more EP2 in macrophages, more aging neutrophils, fewer contacts between them. The transition to humans will require a selective and safe EP2 inhibitor: the current data show a target and mechanism, but not a ready-made drug.

🔗 Read original →

On July 17 in Nature Aging, a study by Marco Demaria's group from Groningen was published. The authors linked the work of CDK4/6, retinoic acid receptor RARα, and the inflammatory secretions of senescent cells; two interventions in this chain improved physical tests in old mice. A senescent cell stops dividing after damage, but remains in the tissue. Many such cells release signaling substances that support inflammation around them. This set is called SASP, and its NF-κB-controlled part is often associated with age-related inflammation: NF-κB is a protein switch that turns on immune response genes. In June, another Nature Aging study linked SASP to the release of nuclear RNA-DNA hybrids into the cytoplasm and the activation of cGAS-STING. Demaria's group approached the same inflammatory secretions from a different side: through CDK4 and CDK6 - enzymes that normally help the cell prepare for division. The authors took already senescent cells and gave them a short course of abemaciclib. The inflammatory genes of SASP began to work weaker; genetic knockout of CDK4 and CDK6 gave a similar result. Abemaciclib changed the secretions of already aging cells, rather than removing them from the tissue. Such agents are called senomorphics: they try to suppress the harmful behavior of the cell, while keeping it in place. Further, the team traced the chain: CDK4 and CDK6 interacted with NF-κB, and CDK4 also interacted with RARα, a retinoic acid receptor, a derivative of vitamin A. Abemaciclib disrupted these interactions. A separate RARα antagonist, substance agn194310, also weakened the inflammatory secretions. The study suggests the CDK4/6-RARα-NF-κB axis as a new target for the search for senomorphics. In 22-month-old mice, both substances reduced systemic signs of this inflammatory secretions and improved physical performance. The authors administered the drugs for two months; each of the four groups had four animals. The small experience requires independent repetition, especially since abemaciclib is used against some types of cancer and has toxicity. The boundary of the result is visible in another task. In a 2022 study by the same laboratory, CDK4/6 inhibitors introduced previously dividing normal cells into a special senescent state. A fresh preprint by another team saw the later inclusion of part of the NF-κB-dependent SASP in such cells. The current article examines already existing inflammatory senescent cells. These two states require different safety checks.

🔗 Read original →

Microgravity weakens cell contact with tissue and slows down protein assembly in mitochondria. On June 30, a study of cells grown on the ISS was published in Nature Communications: under microgravity, mitochondrial ribosomes assembled proteins more slowly. The authors traced the path from cell attachment to the surrounding tissue to the chemical tuning of this protein factory. Unloading and fixation of the hind limbs of mice gave the same signature in the soleus muscle. Mitochondria make some proteins themselves, with their own genome coding 13 components of the respiratory chain - a machine that converts food energy into ATP. Therefore, mitochondrial translation is important for muscle cells: the speed at which their internal ribosomes read RNA and assemble these proteins. The team of Taishi Wakigawa and Yusuke Kimura grew human cells on the ISS for 24 and 48 hours. A centrifuge on the station, which created normal Earth gravity (1g), served as a control. After 24 hours in microgravity, fewer ribosomes worked on mitochondrial RNA, while the amount of RNA itself changed little: protein assembly slowed down. In worms that lived on the station for four days, the same group measure of translation efficiency decreased. The authors reproduced the effect on Earth in a three-dimensional clinostat - a device that constantly changes the position of cells and mimics the absence of a constant gravity vector. The signal decreased within an hour, returned to its initial level under normal gravity, and increased under tenfold overload. The solution was found outside the cell. Laminin is a protein of the supporting tissue; integrins serve as molecular anchors for it. The better the cell attached to laminin, the faster the mitochondria assembled proteins. An integrin blocker gave the opposite effect and reduced oxygen consumption by mitochondria. Contact with laminin involves a chain of proteins: FAK - an enzyme at the cell attachment site, then RAC1 and PAK1. It changes the work of the BAD protein at the outer mitochondrial membrane and triggers fatty acid synthesis inside it. This process consumes malonyl-CoA, leaving fewer malonyl marks on the translation apparatus, which allows ribosomes to start working faster and elongate the protein chain. In eight-week-old mice, the authors unloaded and fixed the hind limbs for 14 days. In the soleus muscle, both mass and mitochondrial translation decreased simultaneously. The authors checked the chain in cells, worms, and young mice, but have not yet studied old muscle and strength recovery.

🔗 Read original →

Scientists appear to have found evidence that aging is a loss of cellular information, not simply the accumulation of damage.

A new preprint from the New York Genome Center has not yet been peer-reviewed. The work was performed on mouse models. Its translation to humans remains to be verified.

The researchers developed SeqTag technology and measured gene expression, chromatin accessibility, and histone marks in the same cells.

The authors found that in young cells, the three regulatory layers operate asynchronously but synchronize quickly. With age, this coordination breaks down. This is called molecular asynchrony.

This provides direct quantitative confirmation of the Information Theory of Aging (ITOA), which is being developed by David Sinclair, whose startup recently received FDA approval. His theory states that DNA barely changes during aging; the system that reads it changes.

Sinclair has long suspected that this process is reversible; experiments with partial reprogramming demonstrate tissue rejuvenation through the restoration of epigenetic patterns.

What practical changes does this make?

The damage accumulation model sets a therapeutic goal: slowing or repairing damage.

ITOA, however, has a different goal: restoring regulatory synchronization. These are different targets and different pharmacological strategies.

Altos Labs, Rejuvenate Bio, and several other companies are currently working on the latter.


https://www.biorxiv.org/content/10.64898/2026.06.02.729594v1

A review offers an explanation for why one aging target is more effective in worms than in mice. On July 12, in Mechanisms of Ageing and Development, a review was published on why anti-aging interventions often have a significant effect on simple animals and a much smaller effect on mammals. The authors collected data on worms, flies, and rodents. They suggest looking for the reason in the connections between tissues, hormones, and feedback that change the response of the entire organism. Gerontology has long seen this difference. A mutation in the signaling pathway through which insulin and IGF-1 regulate growth and metabolism can double the life of the worm C. elegans. In mice, rapamycin - one of the most studied anti-aging drugs - gave a much smaller result: when it was started at 600 days of life, the age at 90% mortality increased by 14% in females and by 9% in males. In worms, a small set of pathways simultaneously regulates nutrition, stress response, reproduction, and cell repair. The body of mammals distributes these tasks among many tissues. The liver, immune system, fat tissue, muscles, gut bacteria, and hormonal signals constantly change each other's state. The authors call the complexity of the organism the number of such connections and reserves, rather than the size of the body or the length of the genome. The authors call the strength of a single target the leverage of the pathway. The more an organism can change its life by affecting one pathway, the longer this lever. Against it works systemic buffering: backup genes, parallel signaling chains, and feedback between organs return the organism to a working state after intervention. According to this scheme, a drug can precisely alter mTOR - a cellular regulator of growth and resource expenditure - or another known pathway, but its effect will encounter several responses at once. Another tissue compensates for the shift; a hormone changes metabolism; gut bacteria change the availability of a molecule; the organism delivers and processes the drug in its own way. The more such connections, the harder it is for one target to shift the lifespan of the entire body. In mice, there is already an example of such a search: a combination of rapamycin and trametinib increased median life by approximately 30%, stronger than each drug separately. The new review raises an additional question: what combinations need to be selected, taking into account the responses of different tissues to each target. The article does not measure "buffering" with one device and does not establish a limit to human life. This is a qualitative review, not an experiment with a new therapy. It can be tested: comparing which compensatory responses are included after intervention in different tissues, and determining whether combinations of interventions enhance each other. The next generation of research should map: which tissue cancels the beneficial effect, which signal it does this with, and in what sequence to intervene, so that compensation from other organs does not cancel the result.

🔗 Read original →

AI can predict proteins, but therapy remains in the lab, factory, and long money - Dorothy Chow's argument. On July 16, former head of Google DeepMind's public engagement lab, Dorothy Chow, explained why computational breakthroughs alone do not accelerate biology. She suggests funding the entire path from prediction to testing and production. In a conversation with the Foresight Institute, Chow begins with AlphaFold. This system predicts a protein's three-dimensional shape from its sequence; knowing the shape helps understand where a drug molecule or antibody can bind to it. The model has made available the structures of over 200 million proteins. However, a long chain of work remains between the protein shape on the screen and a drug for humans. The success of AlphaFold relied on open data and a common way to verify results. The Protein Data Bank is a global archive of verified protein structures; CASP is a blind test where teams predict forms that have been found experimentally but not yet published. In the AlphaFold2 paper, researchers showed near-experimental accuracy for most proteins in CASP14. Such verification allowed comparing models by accuracy, not by the persuasiveness of presentations. After such a prediction, the expensive part begins: testing the candidate in a wet lab, where cells and tissues are physically worked with, and setting up production. Physical verification and the factory require time, infrastructure, and capital; Chow calls this gap the main obstacle after computation. According to Chow, a venture fund usually waits for an exit from investment within a 5-10 year horizon, and government funding often divides biology into narrow disciplines and annual budgets. She proposes mixed funding: grants and philanthropy pay for the long, risky part, and private capital scales up what has passed verification. An example is the pre-ordered vaccine purchases during COVID-19: the state reduced the risk of production, and companies could act faster. Chow suggests gathering data, independent verification, labs, production, and suitable money in advance for the next big idea. Then, AI can take successful predictions to testing and application, rather than stopping at the screen.

🔗 Read original →

PeptAI stated that its agent, after laboratory testing, independently modified the search for a peptide against VEGFR2 - a target associated with tumor vascular growth. According to the company, the program generated thousands of variants, selected several, paid for binding measurement, and received the results back. PeptAI wrote that these data forced the system to change its peptide generation approach and exclude problematic target areas in the new round. In April, PeptAI showed nine computational checks that the program runs peptides through before transferring them to an external laboratory. In May, its agent proposed 306 peptide variants of MOTS-C and an experimental plan; the candidate synthesis had not been performed yet. On July 17, PeptAI wrote in a new report: "The agent designs a candidate, pays the laboratory for verification, and receives the binding result back, and then starts the next round itself." The company claims that the agent generated thousands of peptides for VEGFR2, selected several, and paid for laboratory verification. VEGFR2 is a receptor on vascular cells: it sends a signal that helps grow new vessels, including those that feed tumors. PeptAI wrote that the positive control and a specially modified variant bound to the target, while the negative control signal did not. The agent took this result as feedback: the next round will use BindCraft - a program for designing proteins capable of binding to a given target - and will avoid known problematic areas on the VEGFR2 surface. The company remains the sole source of these data: the post lacks sequences, protocols, numerical measurements, or independent confirmation. Binding verification answers a narrow question: does the peptide bind to the target in a specific analysis? A therapeutic program then needs data on action in cells and organisms, selectivity, safety, and reproducibility. The model can rank thousands of sequences, but a physical experiment shows where its prediction met the molecule. For the agent, pairs of "sequence - result" become memories of which ideas withstood the test. If the result really changes the next search, each success and each failure narrow the space of the next experiment.

🔗 Read original →

Carl Pfleger has urged Brian Johnson to invest in companies and research on aging. Investor and founder of the AgingBiotech.info catalog, Carl Pfleger, has called on Brian Johnson to direct a significant portion of his money and public influence towards aging research. His thesis is as follows: personal measures help one survive until future therapies, and capital for laboratories, biotech, and testing regulations can bring those therapies closer. On July 17, Carl Pfleger wrote that Johnson is misallocating his priorities. Pfleger assumes that a 49-year-old person has decades of life ahead, and therefore, interventions that gerontology will create during this time will have a greater impact on his fate. "What the field of aging will be able to create in the next three to four decades will be more important for his life than everything available today," Pfleger writes. A personal protocol answers the question of how to reduce the risk of death now. Sleep, nutrition, training, disease treatment, and available medicine give a person time. Johnson has already turned this approach into a service: in June, he launched Immortals Medicine, where doctors remotely prescribe prescription medications from his protocol. Pfleger asks a different question: what does a wealthy supporter do after they have already received this time reserve? He suggests financing non-profit research, investing in companies that translate biological ideas into medicines, and advocating for regulations that allow for faster testing of therapies against age-related diseases. Gerontology studies the mechanisms of aging; without money for experiments, testing, and production, its discoveries will remain as articles and mouse models. This position has a mechanism. Personal expenses change one person's health through already available measures. Donating to a laboratory, investing in a biotech company, or a political campaign can pay for experiments, teams, and clinical trials. A successful therapy can then help many people, including the one who helped it appear. Pfleger believes that it is here that a wealthy immortalist has the greatest return on the next dollar: supporting future ways to repair the causes of aging. He also writes that he himself has invested in more than 30 rejuvenation companies; this is his own statement. Pfleger does not provide a public estimate of Johnson's expenses and does not calculate how much each of his options will accelerate the emergence of therapy. Therefore, the dispute remains a dispute over priorities. He suggests that a wealthy immortalist evaluate the next dollar based on how many research projects, companies, and trials that dollar can launch.

🔗 Read original →

A vaccine against six breaks in the KRAS gene was tested in people before pancreatic cancer appeared. On July 16, in an article in Cancer Discovery, a team from Johns Hopkins University described the first small clinical trial, phase I, with 20 people who have a hereditary risk of pancreatic cancer and suspicious changes on scans, often a small cyst. In 18 participants, T-cells - cells of the immune system that recognize the altered KRAS - appeared; such cells were found in the blood for up to two years. Over a median observation period of 16.5 months, cancer did not develop in any of the participants. In people with a hereditary risk of pancreatic cancer, doctors usually try to catch it with regular scans. If a cyst or another lesion starts to look dangerous, doctors may remove part of the pancreas. Early pre-cancerous cells can be too small to see on a scan. The Johns Hopkins University team chose KRAS - a gene whose breaks trigger the growth of more than 90% of such tumors - as a target. The same breaks often appear in pre-tumor cells as well. The mKRAS-VAX vaccine contains fragments of the six most common KRAS variants. After vaccination, the immune system learns to recognize these fragments and seek out cells that display them on their surface. People have already been given a vaccine against a common tumor mutation after diagnosis: in the IDH1-vac trial, 33 patients with brain tumors received it along with surgery, radiation, and chemotherapy. The participants in the new study had not yet developed cancer. All had a hereditary predisposition and a change in the pancreas that doctors usually monitor. Participants received three initial doses under the skin and one booster. In 18 out of 20, two types of T-cells appeared: some ready to immediately attack the target, others storing immune memory. Part of the cells induced by the vaccine persisted in the blood for up to two years. Undesirable phenomena did not exceed the first or second degree. Researchers also looked at cysts. In the vaccination group, they decreased or disappeared in 37.5% of participants; in a comparable unvaccinated group - in 6.8%. People were not randomly assigned to groups, so this number does not separate the action of the vaccine from differences between groups. The first phase tested safety and the ability to induce an immune response. The authors reported on a recruiting study where they will check if T-cells from the blood reach pre-tumor lesions in the pancreas itself. A larger comparison should measure whether the number of cancer cases decreases.

🔗 Read original →

Exosomes delivered RNA through the brain's defense and turned off inflammatory neuronal death in mice. On July 11, a study on siRNA - a short RNA that makes a cell stop producing a selected protein - was published in Advanced Science. The authors packaged siRNA against RIPK3 into exosomes, administered them to mice intravenously, and obtained a signal in the brain, suppression of the target in neurons, and better results in memory tests in a mouse model of Alzheimer's disease. It can be easier to come up with a molecule against a brain disease than to deliver it to a neuron. The blood-brain barrier consists of tightly connected vascular cells and barely allows large RNAs from the blood to pass through. Free siRNA is quickly destroyed. Without delivery, even precise RNA remains outside the target cell. A team from Harbin Medical University took exosomes, natural bubbles that cells use to exchange substances. They placed the RVG peptide on their surface: it binds to receptors on the wall of brain vessels and neurons. They loaded siRNA against RIPK3 - a protein involved in triggering necroptosis, an inflammatory form of cell death - inside. During necroptosis, a neuron bursts and releases alarm signals; they raise inflammation around it. The authors checked the entire chain. In a barrier model, a fluorescent label passed into the "brain" chamber. After injection into mice, the exosome signal reached the brain in about six hours. In neurons, siRNA exited from lysosomes - intracellular bubbles where foreign cargo is usually digested - into the cytoplasm and reduced the activity of RIPK3 and its partner MLKL. Further, they checked the disease. Triple APP/PS1/TAU mice carry mutations that cause them to accumulate amyloid and pathological tau protein early, lose synapses, and lose memory. After intravenous administration of siRNA-exosomes, the animals better recognized a new object and found the previous platform location in a water labyrinth. In the brain, the proportion of cells with active RIPK3 decreased from 28.48% to 12.19%, and with active MLKL - from 24.42% to 12.45%. In human cortical organoids, the system also reduced the necroptosis signal and restored the synaptic protein GRIA1. The study addresses two separate challenges: blocking the death pathway and delivering a gene switch to a neuron. Necroptosis has long been suspected in neuronal loss in Alzheimer's disease: in a 2023 experiment, human neurons in a mouse brain were protected by suppressing RIPK1, RIPK3, or MLKL. The new study checks transport: can a gene switch be brought from a vein directly to a vulnerable neuron. Organoids do not have vessels and a full-fledged immune system, and exosomes for experiments were obtained from mouse cell lines. The fluorescent signal in the brain does not yet measure the exact proportion of the administered dose in human tissues. For human trials, data from primates are needed: long-term safety, organ distribution, and production of identical batches of exosomes.

🔗 Read original →

On July 16, Science journal editor-in-chief Holden Thorp described a bottleneck in scientific AI. The generation of hypotheses, analyses, and manuscripts is accelerating, while the verification of their origin and validity remains the work of humans. In a July 16 editorial in Science, Thorp suggests measuring scientific AI by two speeds. The first shows how many hypotheses, calculations, and articles a machine produces per day. The second shows how many of them researchers can verify to rely on the conclusions. AI agents are accelerating the first speed; the second is limited by the number of checks that people can perform. The reason is simple. A finished article hides a long chain of decisions: which data the system chose, which it excluded, which metric it assigned as primary, how many options it tried before achieving a successful result. If this chain was assembled by an agent, the editor needs more than just to read the smooth text. They need the original data, code, and action log to reproduce the work process. In December 2025, the authors of a study on two open autonomous scientific systems found four classes of hidden errors: inappropriate testing, data leakage, incorrect success metrics, and result selection after the experiment. The final manuscript may hide such decisions. A complete action log and code allow them to be seen. In May, the authors of a preprint on fictional bibliographic references, i.e., works before peer review, checked 111 million references in 2.5 million works. They estimated the number of non-existent references in 2025 to be at least 146,932. Each such reference forces the editor, reviewer, or reader to manually search for the real basis of the statement. A scientific agent must transmit a verifiable trail along with the result: the original question, data, intermediate decisions, and final conclusion. Such a trail turns verification into part of the research, not just a review of beautiful text after its appearance. Dorothy Chow described the path from protein prediction to a drug: after calculation, laboratory testing, production, and long funding remain. Thorp adds editorial verification to this chain. In biology, an error in target, model, or biomarker selection leads to months of laboratory work and money needed for the next intervention. The speed of scientific AI is measured by the speed of verification of its conclusions. Journals, laboratories, and foundations need a reproducible chain that leads to each conclusion.

🔗 Read original →

Nerve endings in the skin send a signal to collagen-producing cells that helps to restrain their aging. On July 17, a study on the connection between nerve endings and skin condition was published in Cell. The study, which involved 148 people, found an age-related decline in Nefh and collagen; the authors then tested the causal chain in skin models. It involves glutamate, a chemical signal that nerve cells use to transmit messages to neighboring cells. Fibroblasts produce collagen, which keeps tissue elastic. As people age, they enter senescence - they stop dividing normally and change their environment. The new work checks how the nerve fiber signal affects fibroblasts in a skin model. The authors add a nerve ending to this picture. Fibroblasts often came into contact with nerve fibers carrying Nefh. This protein was predominantly found in neurons that release glutamate. Nefh is a neurofilament heavy chain protein that supports the neuron's framework. The loss of Nefh changed the state of fibroblasts through a neuronal signal. When researchers removed Nefh from neurons that release glutamate in skin models, fibroblasts more often entered a senescent state, and the skin lost collagen. Denervation of the skin had a similar effect: signs of aging intensified along with the loss of collagen. Further, the authors dissected the pathway. Cdk5 interacts with both Nefh and Vglut2 - a transporter that loads glutamate into vesicles for release from the neuron. In fibroblasts, the action of glutamate depended on Slc1a3, a carrier protein. In models, the addition of glutamate weakened signs of senescence and maintained collagen production. In skin models, disruption of the signal from a neighboring nerve changed the state of fibroblasts and collagen content. Studies of skin aging can now test how to preserve the connection between the nerve ending and the tissue it serves. In human samples, the work shows an age-related correlation, and the removal of Nefh and addition of glutamate were investigated in skin models. Separate data on signal delivery, safety, and duration of effect are needed for therapy. In models, the loss of Nefh in neurons that release glutamate reduced this signal, enhanced fibroblast senescence, and decreased collagen content.

🔗 Read original →

SonoThera claims to have delivered a full-size dystrophin gene to monkeys' muscles without a virus. Gene therapy is hindered by delivery: a large gene needs to be introduced into the necessary cells, and a common viral vector often cannot be administered repeatedly. SonoThera demonstrates an alternative approach in animals: DNA is introduced with microbubbles, and then focused ultrasound temporarily opens a pathway into cells. The company has no clinical data or peer-reviewed article. Duchenne muscular dystrophy occurs due to defects in the dystrophin gene. Without this protein, muscle fibers gradually deteriorate. The full version of the gene is too large for a conventional viral vector, so viral therapies usually carry a shortened micro-dystrophin. On July 2, Elixirgen described a different approach to the same goal: localized mRNA with full-size dystrophin should remain at the injection site. SonoThera is attempting to deliver DNA with ultrasound directly into multiple muscle tissues. According to biologist Avi Ro, the company introduces DNA with microscopic gas bubbles and directs ultrasound to the target tissue. The bubbles help make the cell membrane permeable for a short time, after which the DNA enters. No virus is needed; theoretically, such a procedure can be repeated. On a poster, the company showed expression of full-size human dystrophin in the muscles, heart, and diaphragm of mice with a Duchenne model. In the skeletal muscle of monkeys, SonoThera claims up to 50% of the normal protein level. This figure has a specific limitation: the used antibody detects both the monkey's natural protein and the delivered human protein. Therefore, the authors separately searched for human RNA and a specific label on the delivered protein in the treated muscle fibers. On July 17, STAT quoted two Duchenne researchers who found the results in animals too unusual to accept without further verification. SonoThera has no human trials yet; the company plans its first trial for 2027, and the disclosed data exists in the form of a conference poster. Ultrasound needs to prove two properties in human muscles: to transport large DNA into a sufficient number of cells and to withstand repeated dosing without hazardous consequences. Then, part of gene therapy may be able to move beyond the size of a viral shell and transport more complete repair programs into tissues.

🔗 Read original →

According to mathematician Kerger, GPT-5.6 Sol proposed a proof of the limits of optimization; Lean checked a new lower bound. On July 14, Berkeley instructor Philip Kerger posted a preprint on the problem of convex optimization, where the algorithm receives only numerical answers to a function. According to Kerger, GPT-5.6 Sol found a proof construction in a session lasting around 2.5 hours; the open Lean project gathers a new lower bound with precisely specified accuracy. Let's present the problem: one needs to find the lowest point of an unknown "bowl" in a space with d parameters. The algorithm chooses a point and receives one number - the height of the bowl at that point. The next move must be chosen based on the history of such numbers. How many measurements are inevitably required before one can guarantee approaching the minimum? At an accuracy of the order of 1/√d, Protasov's 1996 method gave an algorithm with approximately d² log²d measurements. The strictly proven ban was much weaker: around d measurements. Between these estimates, there remained the possibility of an algorithm that would be radically more economical. In Kerger's preprint, a new lower bound is stated: any deterministic algorithm will require approximately d²/log(d+1) queries in the worst case. It almost closes the previous gap to within logarithms. Here, Lean appears - a language and program for formal proof verification. Each line of the proof receives an exact logical type, and the program's core re-checks whether the conclusion follows from the previous steps. The open code conducts this procedure for precisely the new lower bound: for each algorithm with a limited number of queries, an admissible convex function is constructed on which that algorithm misses the specified accuracy. The Lean project checks the lower bound. The upper estimate from Protasov's work, the two-sided formula for complexity, and the transfer to mixed-integer problems remain in the manuscript; they will be evaluated by mathematical reviewers. On July 16, the chief editor of the Science journals, Holden Thorp, wrote that AI can write more science than people can check. For arguments that can be recorded in strict logic, Lean sets a sequence: the model proposes a proof construction, the core checks a specific statement, and mathematicians evaluate the setup and conclusions.

🔗 Read original →

When scientists turned off nine old genes, dietary restriction extended the life of flies more strongly. On July 16, a study of eight species of Drosophila was published in EMBO Reports. The authors found a common response to dietary restriction, selected 15 evolutionarily old genes, and temporarily turned them off in adult flies. In nine interventions, lifespan on a restricted diet increased. In fly experiments, dietary restriction looks like a series of diets with different proportions of yeast - the main source of protein. A team from the University of Sheffield gave eight species of Drosophila five such diets. In five species, life was longer on less rich food, with each species having its own optimal proportion of yeast. In flies with a diet for maximum lifespan, egg laying also decreased. Then the researchers looked at which genes changed activity under such nutrition. In six species, they compared 6,926 common genes. The response turned out to be similar in direction: the same groups of genes in different flies more often turned on or off together. Most of the most responsive genes, however, arose relatively recently in evolution and are found mainly in arthropods. The common RNA signature does not yet indicate a common mechanism of longevity. This signature mixes ancient processes and local adaptations of each species to food deficiency. The authors identified 15 genes that changed activity consistently and existed even in distant animal ancestors. Each of them was temporarily turned off by RNA interference: this method forces the cell to destroy the RNA of the chosen gene and reduces the production of the corresponding protein. In 12 interventions, lifespan changed. Nine enhanced the effect of dietary restriction. Five of these nine genes usually became more active under the most restrictive diet; their shutdown gave flies an even greater gain. The authors suggest that part of the response to food deficiency protects current physiology: the cell tries to conserve and redistribute rare substances, and this compensation partially offsets the life extension. Among the selected genes, sulfur-containing amino acid metabolism - cysteine, methionine, and serine - was particularly common. These molecules are involved in protein assembly, antioxidant protection, and one-carbon group metabolism. The authors received a direction for the next check: to determine which of such compensatory reactions limit the effect of diet and whether it is possible to influence them more precisely. A review of how tissues and feedbacks weaken single interventions against aging discussed this problem in complex organisms. The new work shows a possible special case even in flies: the organism itself includes responses to food deficiency, and some of them reduce the gain in longevity. The work covers adult female flies and diets with different proportions of yeast. The path to humans will require testing these targets in mammals and a separate assessment of safety. Currently, the most valuable result is the ability to search for a common mechanism among a huge list of changed genes: comparing species, selecting ancient candidates, and immediately testing whether they change the life of the animal.

🔗 Read original →

A radio frequency field preserved pig cartilage cells during warming after vitrification. On July 16, a manuscript about the experience with pig joint cartilage was accepted for publication in Biofabrication. A piece of cartilage approximately 15 × 10 × 2.5 mm in size was vitrified in 10 ml of cryoprotective solution; after warming with an electric field, around 90% of cells throughout the cartilage thickness remained alive. Vitrification translates water in tissues into a glassy state and does not allow ice crystals to grow. The most dangerous section begins when returning to warmth. A water bath quickly heats the edge of the cartilage, while the center lags behind. The temperature difference creates mechanical stress and gives ice time to grow. Joint cartilage adds a second task: it has no vessels. The cryoprotective solution VS55 must penetrate from the surface to the center to protect cells during cooling and warming. A higher concentration helps avoid ice, but damages cells. After a three-hour staged loading, the authors achieved more than 85% of the target concentration in the center of the cartilage piece. A controlled gradient of cryoprotector concentration has already helped cells, oocytes, and skin pieces about 2 mm thick survive thawing. In the new experiment, the solution had to pass through avascular cartilage 2.5 mm thick. Then, the piece along with the solution was placed between electrodes. A 40 MHz field heated the cryoprotective solution throughout the system volume, including the solution inside the cartilage. The polar molecules of the solution followed the field oscillations and converted its energy into heat. The apparatus accounted for the change in the electrical properties of the solution and adjusted the power. In the dangerous temperature range, the difference between the edge and the center remained below 10 °C; a water bath gave a difference of about 76 °C. A water bath warmed the system with cartilage in 10 ml of solution at a rate of about 31 °C per minute, while the radio frequency setup warmed at about 110 °C per minute. In the center of the cartilage pieces after the water bath, cell viability dropped to around 60%; after radio frequency warming, about 90% was preserved throughout the thickness. By the third day of culture, metabolic activity exceeded 70% of the level of fresh control cartilage. In experiments with smaller cartilage disks, radio frequency warming preserved more than 90% of glycosaminoglycans - matrix components that hold water and help cartilage withstand load. The elasticity and permeability after the procedure were close to those of fresh cartilage. Earlier, the same research line heated cartilage through magnetic nanoparticles. Molecules of cryoprotective solution are approximately an order of magnitude smaller than particles and more easily penetrate the dense matrix. Now, the field heats the solution already delivered to the tissue. The authors checked pig cartilage pieces and observed them for five days in culture. Warming of a system with a volume of up to 65 ml exists as a calculation for now; the authors have not performed transplantation with this material. Several co-authors are affiliated with TDA Research, and patent applications have been filed for the system and loading protocol. Vitrification of a whole rabbit brain with solution M22 also gave preservation of fine structure after warming. The cartilage experiment adds dimensions necessary for dense tissue: did the cryoprotector reach the center and how much did the center lag in temperature.

🔗 Read original →

Jennifer Doudna's group created compact genome editors using AI that worked better than the natural TnpB protein in human cells. On July 16, a study was published in Science about SynTnpB, new variants of a small enzyme that guide RNA brings to the desired DNA site. The authors built these variants based on the three-dimensional form of the protein and evolutionary constraints, and then tested them in living cells. The genome editor works as a molecular tool with an address: the guide RNA recognizes a DNA site, and the enzyme cuts or changes it. For many tasks, such a tool needs to be delivered inside the cell. The size of the enzyme determines whether it can fit in the delivery vehicle along with the necessary RNA. TnpB belongs to compact ancestors of part of the CRISPR-Cas12 system, making it suitable for systems where both the enzyme and guide RNA need to fit. Jennifer Doudna's group took TnpB and separately designed two surfaces of the protein: one binds to DNA, the other holds the guide RNA. Evolutionary data suggested which amino acids could not be changed without losing function. A protein inverse folding model selected new sequences to fit the given three-dimensional shape of the remaining parts. Then, the researchers assembled combinations of these parts, filtered out non-working ones in bacteria, and tested the best ones in human and plant cells. The best SynTnpB in human cells retained or exceeded the activity of natural TnpB. One of the most active variants matched the original protein only 77% of the amino acids. Cryo-electron microscopy showed how the created enzyme holds RNA and DNA in two different positions. The laboratory obtained a working molecular machine with a new sequence and proven function. The study tested editors in bacteria, plants, and human cells. Delivery to the desired tissues, off-target DNA cuts, and human safety remain to be tested separately. The model created strongly differing designs, and experimental testing selected those that actually cut DNA. Thus, the search for genome editors combines the creation and testing of new proteins. Source: Science; independent analysis: Chemical & Engineering News.

🔗 Read original →

REGENXBIO reported stable or improved vision five years after surgery with Sura-vec in wet age-related macular degeneration. The data pertains to the third and fourth groups of an early study. The published press release does not include the size of these groups and detailed five-year follow-up metrics; the company is waiting for the primary results of two trials where Sura-vec is compared to standard treatments in the fourth quarter of 2026. In wet age-related macular degeneration, permeable vessels grow in the central zone of the retina. Fluid and blood from them damage vision used for reading and recognizing faces. Their growth is stimulated by the VEGF protein. Treatments blocking VEGF preserve vision but require repeated injections into the eye. Sura-vec is an investigational gene therapy, so it is administered under the retina during surgery under local anesthesia: the surgeon first removes the clear gel inside the eye, then delivers the treatment. Such surgery is usually performed on an outpatient basis. The AAV8 viral vector delivers genetic instructions to retinal cells for a fragment of an antibody blocking VEGF. Its goal is to reduce the number of subsequent injections. In 2024, The Lancet published two-year results of a study with 42 previously treated patients. All patients were administered a single dose of the treatment, then called RGX-314, under the retina. In most patients in the groups with higher doses, vision remained stable or improved, and additional injections of VEGF-blocking treatments were rarely or not at all required. In one patient from the maximum dose group, after 12 months, pigment changes appeared in the central zone of the retina, and vision sharply deteriorated; the connection to treatment was considered possible. Asymptomatic pigment changes in the lower periphery of the retina were more frequently noted at higher doses. Five-year follow-up checks if the clinical effect is preserved. In a July 18 statement, REGENXBIO writes that in the third and fourth groups, vision remained stable or improved, and the need for additional injections decreased. The company separately mentions a patient with polypoidal choroidal vasculopathy - a form of the disease with abnormal vessels under the retina; their disease did not respond to VEGF-blocking treatments. It also reports that in long-term follow-up, it found no new safety signals, and no intraocular inflammation was observed in participants who completed the study. The published press release does not include the size of the groups, changes in visual acuity, and frequency of injections, so the results cannot be quantitatively compared to standard treatment yet. Such a comparison should be provided by the ATMOSPHERE and ASCENT studies. In these studies, participants are randomly assigned to either Sura-vec or standard treatments: ranibizumab or aflibercept. According to REGENXBIO, over 1,200 people were enrolled in the two trials. The company expects the primary results in the fourth quarter of 2026. The main test is whether the surgery will preserve visual acuity with properly prescribed glasses or lenses at least as well as regular injections.

🔗 Read original →

Michael Darwin proposes calculating in advance when to send a biostasis team to a patient. On July 13, Darwin published a 381-page review of medical methods to assess the risk of death on the horizon of months, weeks, days, and hours. He suggests gathering these assessments on one panel to prepare the cryopreservation team based on the patient's changing condition. In the new review, Darwin describes the task that begins even before cryopreservation. Biostasis preserves a person after legal death in anticipation of future medical care. The team must quickly start cardiopulmonary support, cooling, and cryopreservation preparation to reduce the time without brain circulation. According to Darwin's estimate, one team deployment within the US costs from $60,000. A doctor can tell a family "days or weeks," but such a response does not indicate when to gather specialists and transport equipment. Some medical assessments consider the risk of death in the coming months, others track how a person loses the ability to walk and care for themselves, and third ones predict the last days in palliative care. In the ICU, doctors monitor whether organ failure is accelerating. Darwin wants to link these responses to three decisions: when to alert the team, when to put it on standby, and when to send it to the patient. In February, Cryonics Monitoring introduced a ring that sends an alarm when a hand remains still for a long time. It signals after a disturbing sign appears. Darwin describes a previous stage: how to see from the course of the disease, analyzes, breathing, and organ function that the likelihood of cardiac arrest is rapidly increasing. An early forecast gives the team time to be nearby before cardiac arrest. The panel will have to be tested on real cases. Darwin takes medical assessments from nursing homes, palliative oncology, and intensive care units; for people who have chosen biostasis, their thresholds still need to be verified with real cases. Several signs of dying may come from one common failure of the body, so the program should consider their connection, rather than turning each mark into a separate confirmation. Darwin has gathered indicators and time horizons for such a system. It will need data from real cases, a comparison of forecasts with the actual time of cardiac arrest, and rules that link the risk level to the time of deployment, team composition, and equipment preparation.

🔗 Read original →

The protein TTYH3 regulates chloride flow in lysosomes, cellular cleanup, and signs of aging. On July 17, a study on TTYH3, a protein in the membrane of lysosomes, intracellular vesicles for breaking down and reusing molecules, was published in Cell Reports. In cellular experiments, TTYH3 was linked to chloride movement, autophagy - the cell's internal cleanup - and signs of cellular aging. In their new study, Chunlei Cang's group sought to answer a narrow question: who else, besides the known channel CLN7, provides chloride flow through the lysosome membrane. In part of the lysosomes of cells without CLN7, this flow was preserved. The authors associated this residual flow with TTYH3, a lysosome membrane protein. The authors measured chloride flow but have not yet established whether TTYH3 forms the pore channel itself or regulates another protein. The lysosome stores a lot of chloride inside. When TTYH3 was overexpressed, there was less calcium in it, and blocking TRPML1 - a channel that releases calcium from lysosomes - weakened the TTYH3-induced lysosome fusion. The authors associate these results with the release of calcium through TRPML1. TTYH3 also reduced the phosphorylation of proteins in the AKT/mTOR signaling pathway, which inhibits autophagy. TFEB, a protein regulator of the cell cleanup program, was more often found in the nucleus, and other measurements indicated an increase in autophagic flux. In a 2021 study, the same research line showed that CLN7 works as a lysosomal chloride channel. Overexpression of TTYH3 partially corrected the consequences of CLN7 deficiency in cells: restored chloride flow, normalized lysosome pH, and reduced oxidative lipid damage and cell autofluorescence. The authors checked the connection to aging in human neuroblastoma cells SH-SY5Y. After TTYH3 was turned off in them, the levels of p53 and p21 - proteins that stop cell division - grew, and the proportion of cells with a marker of cellular aging increased. The active form of the signaling protein ERK removed part of the effect of TTYH3 shutdown. In the brains of mice, the amount of TTYH3 decreased with age, but only the protein level was measured. In a living organism, TTYH3 has been studied so far only in terms of protein quantity in the mouse brain. The authors suggest tissue-specifically turning off its gene in mice to test its role in neurodegeneration and regulation of lifespan. In a study on ASAP3, the shutdown of another autophagy regulator in mice was accompanied by fewer signs of liver aging and increased lifespan.

🔗 Read original →