Loading today’s stories
Science4 distinct publishers3 min readUpdated
Five years in a dish, four years of cortex: the organoid clock runs at human speed
Arlotta's group at Harvard cultured cortical organoids past five years and got the gene activity of a 4-year-old. The model is longitudinal now, and priced in calendar years.
The Scientist · Science desk
What happened
- The work is published in Nature as "Human brain organoids record the passage of time over multiple years."
- Paola Arlotta, a developmental neurobiologist at Harvard University, is senior author of the study.
- Cortical organoids grown for five years are the longest-lived brain organoids that have been studied in detail.
- The team documented the organoids' progress over five years before submitting the research, then continued culturing some for two more years before ending the experiment.
- Arlotta's team has grown the same kind of organoids for seven years, but there are too few of these older organoids to reliably analyse them.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
Count from conception. A typical 4-year-old is roughly 57 months past that point, and the organoids whose gene activity matched that child had been in culture for about 60 [11][27]. Developmental time in the dish therefore ran at something close to 95 percent of wall-clock time [27]. The earlier checkpoints agree: at three to six months of culture the tissue looked like a fetal brain three to six months post-conception [10], and features that normally appear only after birth showed up at about a year [12]. Nothing in the data accelerates. Ten years of culture would buy about nine and a half years of development [28].
That is the number that prices every program built on this method. Arlotta's group says it is already using the organoids to screen for drugs that might alter the progression of conditions such as epilepsy [22], and a screen at a late-maturation phenotype means paying the years before the first dose.
It also sharpens a disagreement between the write-ups. Scientific American reports Madeline Lancaster of Cambridge, who was not involved, describing signs of maturation "and even aging like you see in an actual brain," with adult-onset disorders including schizophrenia, Alzheimer's and Parkinson's as the eventual payoff [23][19]. What was measured is a developmental trajectory in gene activity and DNA methylation that lands at age four [11][14]. A four-year-old cortex is not an aged one. Andras Lakatos, also at Cambridge and also uninvolved, puts it operationally: finding ways to speed up organoid ageing will matter [29], and his own disease examples, autism and epilepsy in later development, sit inside the window the tissue actually reached [20].
The cell-intrinsic result is the part that constrains the workaround. Dissociate an old organoid and the cells rebuild late-stage types; mix old cells with young ones and the old cells resume making neurons, but only the later-stage kinds [16]. Age travels with the cells rather than with the flask, so maturity cannot be borrowed by co-culture, and reseeding does not reset it.
Against the previous published ceiling of two years, reported in 2021, five analysed years is a 2.5-fold extension and the unanalysed seven-year cultures are 3.5-fold [7][26]. The enabling change was unglamorous: standard media did not sustain neuronal activity over long periods, and reformulating the medium to support spontaneous firing kept the populations alive, according to Noelia Anton-Bolanos, now at UMC Utrecht [8][9]. The seven-year cohort exists but is too small to analyse reliably [5]. STAT's account of the origin is telling on that point: the old organoids sat there until a junior scientist, Irene Faravelli, analysed them and then feared she had made a grave mistake [25]. Long culture was tolerated before it was designed for.
What the tissue still lacks bounds the claim in the other direction. There are no blood vessels, no large-scale neural architecture and no sensory input from a body, and Arlotta's own summary is that "at some point, you need much more" [17]. Cortical layering and vascularisation remain on the fix list [18]. On consciousness, Arlotta calls it highly unlikely that organoids are even faintly aware, while saying the question needs re-examining as the systems get more complex [24].
Loading topic controls
Loading related stories