Further-reaching optogenetics

Systemic viral delivery of a gene encoding for a light-sensitive red-shifted ion channel enables non-invasive neuronal stimulation deep in the brain of rodents.

Optogenetics — that is, the combination of the targeted expression of light-sensitive ion channels on the neuronal membrane and the temporally precise delivery of light — has spearheaded an explosion of methods for neural stimulation. Yet optogenetics requires the expression of exogenous ion channels, often virally delivered, that are typically most sensitive to short-wavelength visible light (from yellow to blue), which has poor penetration into tissue. Hence, the application of optogenetics to neural stimulation commonly requires the insertion of optical fibres into the brain. Reporting in Nature Biotechnology, Karl Deisseroth and colleagues now describe the combination of ChRmine — an ultrasensitive red-light-activated channelrhodopsin1 — and its systemic viral delivery for non-invasive optogenetic neuronal stimulation deep in the brain of rodents2. The approach is cell-type-specific, and requires light at intensities and wavelengths that are safe and effective.

Deisseroth and co-authors targeted the ventral tegmental nucleus (VTA), a region of the mouse brain which lies approximately 4.5 mm below the skull and is critical for reward learning. The authors first injected a viral construct into the brains of mice to express a ChRmine engineered to respond to tissue-penetrating red light (635 nm in wavelength). A relatively low light intensity (200 mW mm–2) applied through the skull was sufficient to generate action potentials in VTA neurons, and such transcranial VTA stimulation served as an artificially induced reward that altered the animal’s behaviour. Through single-unit recordings in rats after viral injections of ChRmine DNA at different brain depths, the authors then show that 400 mW mm–2 of light at 635 nm applied at the surface of the animals’ skull can reliably induce neuronal spiking in ChRmine-expressing neurons at depths of up to 7 mm.

When expressed in GABAergic neurons, ChRmine can functionally inhibit ongoing excitation in a targeted neuronal circuit. In mice with excitotoxin-induced epilepsy, Deisseroth and co-authors targeted ChRmine expression to inhibitory neurons in the hippocampus, which is often hyperexcitable in temporal lobe epilepsy, and then triggered light-mediated stimulation of the neurons in real time when seizures were detected. Such closed-loop induction of excitation inhibition significantly decreased the duration of ongoing seizures.

To avoid cranial surgery for the injection of ChRmine-encoding viruses, Deisseroth and colleagues used adeno-associated viruses (AAVs) engineered to cross the blood–brain barrier3,4, as these can be injected intraorbitally. In wild-type mice, they show that the animal’s brainstem (which has been challenging to stimulate thus far, owing to the difficulty of accurately and safely inserting an optic fibre in it) can be stimulated with 800 mW mm–2 of red light after the intraorbital injection of AAVs targeting serotonergic neurons (by inserting, in the viral construct, promoters specific to this type of neuron), and that such stimulation increased the animals’ social activity.

Minimally invasive neuronal control can be achieved via stimulation sources that can penetrate tissue — in particular, magnetic fields5,6, ultrasound7,8 and red-shifted light9,10,11 — paired with the viral-mediated gene delivery and expression of ion channels sensitive to heat, mechanical forces or light (Fig. 1). Nanoparticles that upconvert near-infrared light12 or X-rays13 into visible light have also been used for the activation of light-sensitive ion channels, and ultrasensitive light-sensitive ion channels can allow for long-term neuronal activation via brief light pulses14. And there are strategies that harness the physical properties of propagating waves, such as temporally interfering electric fields15 and transcranial focused ultrasound16, to generate focal regions of stimulation; these do not require the expression of exogenous ion channels. Regardless of the approach, non-invasive deep brain stimulation has to balance specificity, efficacy, the level of invasiveness (in terms of tissue damage and of genetic alteration), and safety (especially the effects of off-target activation). Deisseroth and co-authors’ strategy strikes a reasonable balance. Because light applied to tissue can increase its temperature, causing off-target stimulation effects at low levels or even tissue damage and neuronal death at higher levels, the authors show that the sensitivity of the ChRmine channel allows for safe combinations of wavelength, power and frequency. Also, in addition to the penetration depth afforded by the red-shifted and highly sensitive ChRmine — which had been only used to activate superficial cortical structures thus far9 — the systemic viral delivery, via intraorbital injection, of AAVs that cross the blood–brain barrier leads to sufficient levels of ChRmine expression for transcranial neuronal activation. Moreover, promoter sequences compatible with the compact AAV genome can be used to target viral delivery to specific cell types. With further development, the strategy should be applicable to a wider variety of brain regions and animal models, which would facilitate basic neuroscience research.

Fig. 1: Ion channels and types of stimuli for non-invasive deep brain stimulation.

a, Red light or near-infrared light can directly stimulate light-sensitive ion channels (red), or indirectly stimulate ion channels sensitive to visible light (dark blue) or to heat (orange) via, respectively, upconversion nanoparticles or plasmonic gold nanoparticles (which generate heat via the photothermal effect). b, Ion channels activated via magnetic-field heating of magnetic nanoparticles. c, Light-sensitive ion channel activated by nanoparticles converting X-rays into visible light. d, Mechanosensitive ion channels (green) directly activated by ultrasound or indirectly enhanced via microbubbles (grey). e, Temporally interfering electric fields can directly stimulate endogenous ion channels. Representative ion channels are listed.

Naturally, neuronal activation at a depth of 7 mm with 635-nm light is insufficient for the stimulation of deep brain regions in larger animals and in humans. New or improved stimulation sources and channel sensitivities, as well as safe and effective delivery techniques and precise cell-targeting methods will be required for non-invasive optogenetics to reach deeper into bigger brains.


  1. 1.

    Marshel, J. H. et al. Science 365, eaaw5202 (2019).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  2. 2.

    Chen, R. et al. Nat. Biotechnol. (2020).

  3. 3.

    Chan, K. Y. et al. Nat. Neurosci. 20, 1172–1179 (2017).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  4. 4.

    Bedbrook, C. N. et al. Nat. Methods 16, 1176–1184 (2019).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  5. 5.

    Chen, R., Romero, G., Christiansen, M. G., Mohr, A. & Anikeeva, P. Science 347, 1477–1480 (2015).

    CAS  Article  PubMed  Google Scholar 

  6. 6.

    Stanley, S. A. et al. Nature 531, 647–650 (2016).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. 7.

    Ibsen, S., Tong, A., Schutt, C., Esener, S. & Chalasani, S. H. Nat. Commun. 6, 8264 (2015).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  8. 8.

    Qiu, Z. et al. Cell Rep. 32, 108033 (2020).

    CAS  Article  PubMed  Google Scholar 

  9. 9.

    Yizhar, O. et al. Nature 477, 171–178 (2011).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. 10.

    Lin, J. Y., Knutsen, P. M., Muller, A., Kleinfeld, D. & Tsien, R. Y. Nat. Neurosci. 16, 1499–1508 (2013).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  11. 11.

    Chuong, A. S. et al. Nat. Neurosci. 17, 1123–1129 (2014).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  12. 12.

    Chen, S. et al. Science 359, 679–684 (2018).

    CAS  Article  PubMed  Google Scholar 

  13. 13.

    Matsubara, T. et al. Preprint at bioRxiv (2020).

  14. 14.

    Gong, X. et al. Neuron 107, 38–51 (2020).

    CAS  Article  PubMed  Google Scholar 

  15. 15.

    Grossman, N. et al. Cell 169, 1029–1041 (2017).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  16. 16.

    Legon, W. et al. Nat. Neurosci. 17, 322–329 (2014).

    CAS  Article  PubMed  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Thomas J. McHugh.

Ethics declarations

Competing interests

The authors declare no competing interests.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Chen, S., McHugh, T.J. Further-reaching optogenetics. Nat Biomed Eng 4, 1028–1029 (2020).

Download citation

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing