Back

Epigenetic inheritance of complex learning abilities in the mammalian brain

Zidan, S.; Andreyanov, M.; Saleh, A.; Barnea, N.; Kundu, S.; Kumar, A.; Bakman, N.; Rosenberg, T.; Hovav, M.; Johenning, F.; Schmitz, D.; Nabavi, S.; Kushner, S.; Marco, A.; Berlin, S.; Schiller, J.; Barkai, E.

2025-08-31 neuroscience
10.1101/2025.08.31.673327 bioRxiv
Show abstract

For several decades, the question of whether cognitive and learning capacities can be inherited through non-genetic mechanisms has been the subject of ongoing debate. Here, we provide the first evidence of transgenerational inheritance of enhanced ability to learn complex tasks in the mammalian rodent brain. These inherited learning enhancements are not limited to specific stimuli, sensory modalities, or learning paradigms. Using behavioral, cellular biophysical, methylomics, genetics and molecular methods, we find that the inherited epigenetic modifications reflect an enhanced neuronal learning state, driven by increased intrinsic neuronal excitability in most pyramidal neurons in the relevant neuronal networks. This enhancement is mediated by persistent downregulation of the muscarinic M-current and is associated with widespread changes in DNA methylation, notably within coding genes associated with the M-current, the Kv7 pathway, in the hippocampi of trained F0 rats, as well as in non-coding RNAs in their sperm samples. Remarkably, a significant portion of these DNA methylation changes were also observed in the hippocampi of their untrained F1 offspring. These findings suggest that complex learning abilities can be inherited in the mammalian brain, as the offspring of trained rodents are born with the biophysical modifications that enable them to become super-learners, the exact change that occurs in their parents brains only after the rule learning.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.