Sight restored by turning back the epigenetic clock

Neurons progressively deteriorate with age and lose resilience to injury. It emerges that treatment with three transcription factors can re-endow neurons in the mature eye with youthful characteristics and the capacity to regenerate.
Andrew D. Huberman is in the Department of Neurobiology and the Department of Ophthalmology, Stanford University School of Medicine, Stanford, California 94305, USA.

Search for this author in:

Ageing has negative consequences for all the cells and organs in our bodies. Our brains are no exception. Neurons in the developing brain form circuits that can adapt to change and regenerate in response to injury. These capacities have long been known1 to diminish over time, but the molecular shifts that underlie this deterioration have remained mysterious. Lu et al.2 show in a paper in Nature that neurons of the eye can be programmed to revert to a youthful state in which they reacquire their ability to resist injury and to regenerate. The authors’ findings shed light on mechanisms of ageing and point to a potent therapeutic target for age-related neuronal diseases.

Retinal ganglion cells (RGCs) reside in the eyes and thus outside the skull, but they are bona fide brain neurons. They initially develop as part of the forebrain. Subsequently, RGCs extend projections called axons out of the eye to make connections with neurons in the brain itself. These axons — which join together to form the optic nerve — survive and regenerate if they are damaged early in development, but not after they reach maturity3,4. Evidence indicates3,5 that this shift is intrinsic to RGCs, rather than reflecting changes in the surrounding cells.

Myriad studies have searched for factors that can prevent or promote RGC survival and regeneration. A handful of such factors have been identified that can endow mature RGCs with some degree of survival and regenerative capacity — but not enough to fully maintain or restore vision after damage to the opticnerve4.

Lu et al. asked whether it is possible to revert RGCs to a younger ‘age’, and whether doing so would allow the cells to regenerate. They infected RGCs in mice with adeno-associated viruses. These harmless viruses had been genetically engineered to induce expression of three of the ‘Yamanaka factors’ — a group of four transcription factors (Oct4, Sox2, Klf4 and c-Myc) that can trigger mature cell types to adopt an immature state6. Such an approach normally comes with hazards in vivo: Yamanaka factors can cause cells to adopt unwanted new identities and characteristics, leading to tumours or death7. Fortunately, Lu and co-workers found that they could circumvent these hazards by expressing just Oct4, Sox2 and Klf4 (together called OSK).

The authors tested the infected RGCs’ ability to regenerate if the cells’ axons were crushed. They found that the OSK-expressing viruses triggered RGC regeneration and long-distance axon extension following damage to the optic nerve (Fig. 1), with no apparent alterations to RGC identity, formation of retinal tumours or any other ill effects.

Figure 1

Figure 1 | Restoring vision in mice. Retinal ganglion cells (RGCs) transmit visual information from the eye to the brain along projections called axons. Damage to the RGC axons prevents transmission of this information, leading to sight loss. Lu et al.2 report that treatment of damaged RGCs with a transcription-factor cocktail called OSK restores the cells to a youthful state, leading to axon regeneration and restoration of sight in mice.

OSK expression had beneficial effects on RGC axon regeneration in both young and aged mice. In some cases, the regenerated axons extended all the way from the eye to the optic chiasm (the location at the base of the brain at which the optic nerves from each eye cross to the opposite brain hemisphere). It is notable that the effects of OSK are seen in older animals, because studies of RGC regeneration are often conducted in relatively young animals, which have a residual natural regenerative ability. Thus, the evidence suggests that Lu and colleagues’ approach can fully restore long-distance regenerative capacity in mature RGCs — a milestone for the field.

Almost all techniques previously used to enhance RGC survival and axon regrowth had to be performed before optic-nerve damage4 — a restriction incompatible with using a technique therapeutically. Excitingly, Lu and colleagues showed that they could induce OSK expression at different time points — even after axon injury — and still improve RGC survival and regeneration. These effects were not limited to optic-nerve injury; OSK expression also effectively reversed RGC and vision loss in a mouse model of glaucoma (the most common cause of human blindness). Expression of OSK in RGCs after axon and vision loss (but before the RGCs died) fully restored vision in these animals. The same was true for wild-type old mice: OSK allowed old mice to regain youthful eyesight.

Why might reprogramming old RGCs to a younger state promote regeneration and restore vision? An emerging model in the field of ageing is that, over time, cells accumulate epigenetic noise — molecular changes that alter patterns of gene expression8, including transcriptional changes and shifts in the patterns of methyl groups on DNA. Collectively, these changes cause cells to lose their identity and so to lose the DNA-, RNA- and protein-expression patterns that once promoted their youthful resilience9,10. Given the growing excitement about DNA methylation as a marker of cell age, the authors asked whether OSK expression somehow counteracts the negative effects of ageing or axon injury on DNA methylation.

The RNA components of a cell’s protein-synthesizing machine, called the ribosome, are encoded by ribosomal DNA genes that steadily accrue methyl marks with age. The ribosomal ‘DNA methylation clock’ is therefore considered to be a reliable estimate of cell age11. Lu et al. found that damaging the axons of RGCs accelerated ribosomal DNA methylation in a way that mimicked accelerated cellular ageing, whereas OSK expression counteracted that acceleration, indicating that tissue injury in general might be a form of accelerated ageing.

The group also tested whether the removal of DNA methylation is required for OSK to regenerate axons or restore vision in old mice. The TET enzymes (TET1, TET2 and TET3) catalyse the removal of DNA methylation12. The authors showed that OSK induced expression of TET1 and TET2 genes, and that reducing TET1 and TET2 production blocked the effects of OSK on RGC regeneration and vision restoration in old mice. Thus, changes in DNA methylation seem essential for the effects of OSK. Indeed, Lu et al. found that OSK restored youthful DNA-methylation patterns across a broad set of genes involved in neuron survival, outgrowth and connectivity. These patterns occur at chromosomal regions that have high levels of PRC2 — a protein complex that alters methylation during development and ageing13. Going forward, it will be important to determine the exact extent to which the positive effects of OSK are mediated by resetting DNA-methylation patterns, and the downstream mechanisms that guide the cellular reset.

Are Lu and colleagues’ findings likely to be relevant to humans? The authors found that OSK expression enhanced axon regrowth and cell survival in human neurons in vitro. The effects of OSK in people remain to be tested, but the existing results suggest that OSK is likely to reprogram brain neurons across species.

Future research should also address whether OSK expression can have the same remarkable effects on neurons elsewhere in the brain and spinal cord. Given that RGCs are bona fide brain neurons, there is good reason to think they will. As such, the current findings are bound to ignite great excitement, not only in the field of vision restoration but also in those looking to understand epigenetic reprogramming of neurons and other cell types generally. For decades, it was argued that understanding normal neural developmental processes would one day lead to the tools to repair the aged or damaged brain. Lu and colleagues’ work makes it clear: that era has now arrived.

Nature 588, 34-36 (2020)


  1. 1.

    D.-B., D. Nature 125, 230–231 (1930).

  2. 2.

    Lu, Y. et al. Nature 588, 124–129 (2020).

  3. 3.

    Goldberg, J. L., Klassen, M. P., Hua, Y. & Barres, B. A. Science 296, 1860–1864 (2002).

  4. 4.

    Laha, B., Stafford, B. K. & Huberman, A. D. Science 356, 1031–1034 (2017).

  5. 5.

    Horsburgh, G. M., Lund, R. D. & Hankin, M. H. J. Comp. Neurol. 327, 323–340 (1993).

  6. 6.

    Takahashi, K. & Yamanaka, S. Cell 126, 663–676 (2006).

  7. 7.

    Abad, M. et al. Nature 502, 340–345 (2013).

  8. 8.

    Oberdoerffer, P. et al. Cell 135, 907–918 (2008)

  9. 9.

    López-Otín, C., Blasco, M. A., Partridge, L., Serrano, M. & Kroemer, G. Cell 153, 1194–1217 (2013).

  10. 10.

    Horvath, S. Genome Biol. 14, 3156 (2013).

  11. 11.

    Wang, M. & Lemos, B. Genome Res. 29, 325–333 (2019).

  12. 12.

    Rasmussen, K. D. & Helin, K. Genes Dev. 30, 733–750 (2016).

  13. 13.

    Laugesen, A., Højfeld, J. W. & Helin, K. Mol. Cell 74, 8–18 (2019).

Download references

Nature Briefing

An essential round-up of science news, opinion and analysis, delivered to your inbox every weekday.