BRENNEN

Industries / Frontier science / Longevity

Biology of aging

This work studies why bodies age, testing proposed causes in cells and in animals such as yeast, worms, flies and mice, and supplies the targets that aging drugs aim at.

7

Milestones achieved, from 1935 to 2022.

3

To watch for: announced, not yet confirmed outside.

2

Open deliverables. Nobody has shipped them.

The road

AchievedWatch forOpen

Achieved

  1. 1935

    Fewer calories, longer rat lives

    C. M. McCay, M. F. Crowell and L. A. Maynard fed rats a diet cut in calories but complete in every other nutrient, and both the average and the maximum lifespan rose compared with rats that ate freely. Earlier studies, such as one by Osborne, Mendel and Ferry in 1917, had lengthened rat life only by combining less food with missing nutrients; a 2010 review in The Journal of Nutrition calls the 1935 paper the first to show the effect of calories alone. The same review says calorie restriction was then the only non-genetic method known to extend lifespan in every species studied, yeast, worms, flies and rodents among them, and that almost 50 years passed before the field took it up as a model for studying aging.

    Lessonchanging one thing at a time, here calories alone, is what let later work pin the effect on energy and not on a missing nutrient.

    pmc.ncbi.nlm.nih.gov

    Credit

    C. M. McCay, M. F. Crowell and L. A. Maynard, in The Journal of Nutrition, 1935

  2. 1961

    Cells have a division limit

    At the Wistar Institute in Philadelphia, L. Hayflick and P. S. Moorhead grew 25 separate samples of human fetal fibroblasts, the cells that make connective tissue, and found that the cultures died out after about 50 subcultivations, each a split of the culture into fresh dishes, a ceiling now called the Hayflick limit. Until then, scientists believed that cells grown outside the body do not age. In 1990 C. B. Harley, A. B. Futcher and C. W. Greider showed that telomeres, the protective caps on the ends of chromosomes, get shorter as such cells keep dividing, and in 2009 E. H. Blackburn, C. W. Greider and J. W. Szostak won the Nobel Prize in Physiology or Medicine for finding how telomeres and the enzyme telomerase, which builds them, protect chromosomes.

    Lessonthe belief rested on cancer cells, many of which divide without end, and did not hold for normal cells; the Nobel committee adds that aging turned out to depend on several factors, of which telomeres are one.

    pmc.ncbi.nlm.nih.govebi.ac.uknobelprize.org

    Credit

    L. Hayflick and P. S. Moorhead; C. B. Harley, A. B. Futcher and C. W. Greider; E. H. Blackburn, C. W. Greider and J. W. Szostak

  3. 1993

    One gene doubles worm life

    In 1988 D. B. Friedman and T. E. Johnson reported that C. elegans, a small roundworm, lived on average 40% longer at 20 °C with a mutation in a gene they named age-1, found among long-lived strains M. R. Klass had collected at the University of Colorado in 1983, but had 75% fewer offspring. In 1993 C. Kenyon's lab at the University of California, San Francisco, reported that worms with mutations in a gene called daf-2 lived more than twice as long as normal while staying active and fertile, and in 1997 G. Ruvkun's lab showed that daf-2 is the worm's version of the human insulin and IGF-1 receptors, which pass on growth and nutrient signals. Other labs then lengthened life by turning down the same signal in fruit flies (median lifespan up to 48% longer, L. Partridge's group at University College London, and up to 85% longer adult life in dwarf female flies, M. Tatar's group at Brown University, both 2001) and in mice: females with one working copy of the IGF-1 receptor gene lived 33% longer, while the 16% gain in males was not statistically significant (M. Holzenberger's group at INSERM, Paris, 2003).

    LessonKenyon recalls that aging was then "generally thought to be a hopelessly intractable, even futile, problem to study"; one gene showed it is regulated, and the same signal works in worms, flies and mice.

    pmc.ncbi.nlm.nih.govebi.ac.ukebi.ac.ukebi.ac.ukebi.ac.ukebi.ac.uk

    Credit

    D. B. Friedman and T. E. Johnson; C. Kenyon and colleagues, UCSF; with the Ruvkun, Partridge, Tatar and Holzenberger groups

  4. 2009

    Antioxidant genes fail the test

    In 1956 D. Harman proposed that aging comes from damage done by free radicals, reactive molecules made during normal metabolism, and by 2009 this was the most popular explanation of aging. Over eight years, A. Richardson's group at the Barshop Institute, University of Texas Health Science Center at San Antonio, measured the lifespans of 18 kinds of mice with raised or lowered levels of antioxidant enzymes; only one, mice lacking the enzyme Sod1, had a changed lifespan (shorter), and some mice with more oxidative damage to their DNA lived normal lives. The same year J. M. Van Raamsdonk and S. Hekimi at McGill University reported that worms lacking any of their five sod genes lived at least as long as normal worms, and that those lacking sod-2 lived longer despite more damaged proteins.

    Lessonfinding that long-lived animals resist damage did not show that damage sets the pace of aging; the direct tests in mice and worms did not support it.

    pmc.ncbi.nlm.nih.govebi.ac.ukebi.ac.uk

    Credit

    V. I. Pérez, A. Richardson and colleagues; J. M. Van Raamsdonk and S. Hekimi

  5. 2013

    Nine, then twelve, hallmarks

    C. López-Otín, M. A. Blasco, L. Partridge, M. Serrano and G. Kroemer proposed in Cell nine hallmarks of aging, processes common to aging across organisms, among them genomic instability, telomere attrition, loss of proteostasis (the upkeep of a cell's proteins), cellular senescence and stem cell exhaustion. Their 2023 update listed twelve, adding disabled macroautophagy (the cell's recycling of its own parts), chronic inflammation and dysbiosis (an upset gut microbe population), and states three tests each must meet: it appears with age, making it worse speeds aging, and treating it slows, stops or reverses aging; by Crossref's count on 20 November 2026 the two papers had been cited 14,917 and 7,125 times. D. Gems and J. P. de Magalhães argued in 2021 that the list is a useful checklist but does not explain what causes aging, and in 2023 M. Keshavarz, D. Ehninger and colleagues at the German Center for Neurodegenerative Diseases (DZNE) in Bonn found links to cancer prevention for more than 80% of the interventions the 2013 paper cites as lengthening mouse life.

    Lessonin mice, where, by the studies Keshavarz's team cites, cancer causes about 70 to 90% of natural deaths in old age, a longer life may mean less cancer rather than slower aging.

    ebi.ac.ukebi.ac.ukebi.ac.ukpmc.ncbi.nlm.nih.gov

    Credit

    C. López-Otín, M. A. Blasco, L. Partridge, M. Serrano and G. Kroemer

  6. 2015

    A youth factor fails retest

    In 2005 I. M. Conboy, T. A. Rando and colleagues at Stanford joined the blood circulation of old and young mice, a method called heterochronic parabiosis, and found that the old animals' muscle stem cells regained their power to repair muscle and their liver cells divided more, as in young animals; in 2014 S. A. Villeda (UCSF) and T. Wyss-Coray (Stanford) reported that young blood also improved learning and memory in old mice. In 2013 and 2014 teams at the Harvard Stem Cell Institute and Brigham and Women's Hospital, led by A. J. Wagers and R. T. Lee, named one blood protein, GDF11, as a factor that falls with age and reverses heart thickening and muscle decline in old mice. In 2015 a Novartis team led by D. J. Glass reported that the GDF11 claims could not be reproduced and that the tests used to detect GDF11 were not specific to it; with a test that was, GDF11 tended to rise with age in rats and people, and in mice it hindered muscle repair. A 2024 review in NPJ Aging notes that a later study also failed to replicate GDF11's effect on the heart and found minimal effects of GDF11 injections.

    Lessonbefore trusting a measured change, check that the test detects only the molecule it is meant to.

    ebi.ac.ukebi.ac.ukebi.ac.ukebi.ac.ukebi.ac.ukpmc.ncbi.nlm.nih.gov

    Credit

    I. M. Conboy, T. A. Rando and colleagues (2005); M. A. Egerman, D. J. Glass and colleagues, Novartis Institutes for Biomedical Research (2015)

  7. 2022

    Big money for aging biology

    On 18 September 2013 Google announced Calico, a company to work on "the challenge of aging and associated diseases", led by A. D. Levinson, then chairman of Genentech and of Apple. On 19 January 2022 Altos Labs launched with US$3 billion committed by investors it did not name, to study what it calls cellular rejuvenation programming, with institutes in the San Francisco Bay Area, San Diego and Cambridge, UK, and S. Yamanaka as senior scientific advisor. In June 2022 MIT Technology Review reported that the Hevolution Foundation, a not-for-profit set up by Saudi royal order in December 2018, was authorized to spend up to US$1 billion a year on aging biology and healthy lifespan, against about US$325 million a year spent by the division of the US National Institute on Aging that funds basic aging biology.

    googlepress.blogspot.comprnewswire.comtechnologyreview.com

    Credit

    Google and A. D. Levinson (Calico); R. Klausner, H. Bishop and H. Barron (Altos Labs); M. Khan, chief executive (Hevolution Foundation)

Watch for

  1. Bowhead whale DNA repair

    Bowhead whales can live more than 200 years, by age estimates from their ovaries, baleen and eye lenses. V. Gorbunova and A. Seluanov's lab at the University of Rochester reported that bowhead fibroblasts repair breaks across both strands of DNA better and more accurately than the cells of other mammals and pick up fewer mutations, and that whale cells and tissues hold high levels of a protein called CIRBP, which was largely undetectable in other mammals except the humpback whale. Whale CIRBP improved DNA repair when added to human cells, and fruit flies made to produce extra CIRBP lived longer. The tests used fibroblasts, though most human cancers start in a different cell type, and the authors say how CIRBP helps repair is not yet known.

    Achieved whena lab outside the author group measures the high CIRBP and better DNA repair in bowhead cells, or repeats the fly lifespan gain.

    pmc.ncbi.nlm.nih.govrochester.edu

    Claimed by

    D. Firsanov, M. Zacher, A. Seluanov, V. Gorbunova and colleagues, in Nature, October 2025

  2. Telomere particles in blood

    A. Lanna's team reported that CD4+ T cells, a kind of immune cell, take up telomeres from other immune cells and release them into the blood in particles the authors call telomere Rivers. In their preprint, a single transfer of such particles, collected from treated old mice, into 20-month-old mice extended median lifespan by about 17 months, with several mice living to about 58 months; the treated groups held 8 to 10 mice each, against 26 untreated old mice. The work has not been peer reviewed and was funded by Sentcell, the company Lanna leads and holds shares in; he is named as sole inventor of methods to make artificial particles, for which Sentcell is the patent applicant.

    Achieved whenthe work passes peer review with its full survival data, and a lab with no tie to Sentcell reports a median lifespan gain in old mice from the same particles.

    biorxiv.orgapi.biorxiv.org

    Claimed by

    A. Lanna (Sentcell, Siena, and University College London), with S. Valvo and M. L. Dustin (University of Oxford) and F. Rinaldi, on bioRxiv, November 2025

  3. Greenland shark genome read

    The Greenland shark is thought to be the longest-lived vertebrate; the team puts its estimated lifespan at 392 ± 120 years. A team led from the University of Tokyo assembled its genome, 5.9 billion DNA letters long, and reported gene families that have expanded, or show signs of natural selection, in immune defence, cancer resistance and DNA repair, plus amino-acid changes in histone H1.0, a protein that packs DNA, which they predict make the packed DNA more stable. These are predictions from comparing sequences; a July 2026 review of the species says the functional meaning of such genome features has yet to be tested.

    Achieved whena laboratory test, in shark cells or by an outside group, shows that the H1.0 changes make the packed DNA more stable, or that one of the DNA repair changes works as predicted.

    ebi.ac.ukebi.ac.uk

    Claimed by

    K. Yang, S. Kinoshita and colleagues (University of Tokyo, with the Norwegian Polar Institute, RIKEN and others), in PNAS, May 2026

Open

  1. 01

    Score the twelve hallmarks

    A free, downloadable table with one row for each experiment that the 2023 hallmarks paper cites as evidence that a hallmark drives aging. Each row gives the hallmark; which of the paper's three tests it is cited for (appears with age, making it worse speeds aging, treating it slows aging); species and sex; whether lifespan was measured or only signs of health; the age at which the change began; whether young animals were tested as well as old; whether the animals had a disease-like shortened life; and whether a second, separate lab reported the same result. A summary page counts, for each hallmark, how many experiments meet each test.

    The brief
    Credit · unclaimed
  2. 02

    Young-blood factors, checked

    A public table of every molecule in blood that a peer-reviewed paper has reported to reverse, or to drive, an aging change in animals. It has one row for each first claim and one for each later test of it by a separate lab, giving species, sex, tissue, what was measured, whether the test could tell the factor apart from close relatives (the GDF11 problem), and the outcome: confirmed, contradicted or mixed. Each factor gets a one-line status, and the table refreshes every quarter.

    The brief
    Credit · unclaimed

Open 01 · the brief

Score the twelve hallmarks

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A free, downloadable table with one row for each experiment that the 2023 hallmarks paper cites as evidence that a hallmark drives aging. Each row gives the hallmark; which of the paper's three tests it is cited for (appears with age, making it worse speeds aging, treating it slows aging); species and sex; whether lifespan was measured or only signs of health; the age at which the change began; whether young animals were tested as well as old; whether the animals had a disease-like shortened life; and whether a second, separate lab reported the same result. A summary page counts, for each hallmark, how many experiments meet each test.

Why it is still open

M. Keshavarz, D. Ehninger and colleagues at the DZNE in Bonn did a version of this for the 2013 paper's nine hallmarks, in spreadsheets attached to their 2023 paper. They found, for example, that about 67% of the studies cited for genomic instability, 63% for cellular senescence and 62% for stem cell exhaustion were not about normal aging but about models of unclear relevance to it. The 2023 update has twelve hallmarks and, by Crossref's count, 300 references; a search of Europe PMC titles and of GitHub on 20 November 2026 found reviews, diagrams and a qualitative 2026 review of how senescence links to the other hallmarks, but no table that scores each cited experiment of the 2023 paper.

pmc.ncbi.nlm.nih.govapi.crossref.orgebi.ac.uk

Built with what exists

The 2023 paper's reference list from Crossref (294 of its 300 references carry a DOI), and the paper itself, free to read since January 2024 according to the licence in its Crossref record; the Europe PMC REST API for abstracts and open full texts; Keshavarz's spreadsheets as a template and as an answer key where the two papers cite the same studies; a large language model such as Claude to draft each row from the cited paper, with a person checking every field; a CSV file in a public GitHub repository. The Note "Findings that cite notes" covers an agent that cites the source behind each finding, the habit each row needs.

  1. First step

    Download the 300 references from Crossref, mark which ones report experiments in animals or cells, and count them by hallmark.

  2. Done looks like

    All 300 references are marked as reporting an experiment or not; every experimental one has a complete row linked to the page it came from; a second person's check of 50 random rows finds at least 95% of fields match their sources; and on studies both papers cite, the table agrees with Keshavarz's spreadsheets in at least 90% of rows.

Open 02 · the brief

Young-blood factors, checked

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A public table of every molecule in blood that a peer-reviewed paper has reported to reverse, or to drive, an aging change in animals. It has one row for each first claim and one for each later test of it by a separate lab, giving species, sex, tissue, what was measured, whether the test could tell the factor apart from close relatives (the GDF11 problem), and the outcome: confirmed, contradicted or mixed. Each factor gets a one-line status, and the table refreshes every quarter.

Why it is still open

A 2024 review in NPJ Aging by F. A. Lagunas-Rangel of Uppsala University tabulates eight named factors found by heterochronic parabiosis (CCL11, β2M, PF4, HSF1, GDF11, MANF, miR-29c-3p and TPM1) with organ, mouse age, pairing time and effect, but has no column for later tests that confirmed or contradicted them; its text notes that a later study failed to replicate GDF11's effect on the heart and calls for standard protocols to ensure reproducibility. Searches of Europe PMC titles for systematic reviews, databases or replication studies of these factors, and of GitHub, on 20 November 2026 found none that tracks replication.

pmc.ncbi.nlm.nih.gov

Built with what exists

The Europe PMC REST API, whose citations service lists every indexed paper that cites a given one (on 20 November 2026 it listed 797 for the 2013 GDF11 heart paper, 699 for the 2014 muscle paper and 452 for the 2015 Novartis paper); a large language model such as Claude to sort citing papers into "reports a new test of the factor" or "only mentions it" and to draft rows, each checked by a person; a CSV file and static page in a public GitHub repository, rebuilt by a scheduled job such as GitHub Actions. The Note "Findings that cite notes" covers an agent that cites the source behind each finding.

ebi.ac.uk

  1. First step

    Pull the Europe PMC citation lists for the three GDF11 papers and sort the first 100 citing records by hand, to set the rules the model will follow.

  2. Done looks like

    All eight factors in the 2024 table are covered; every paper Europe PMC lists as citing each first claim is sorted; every test row links to its source; a second person's check of 50 random rows agrees with at least 95%; and the table has refreshed on schedule twice.