Q&A Optogenetics: 10 years after ChR2 in neurons—views from the community Antoine Adamantidis, Silvia Arber, Jaideep S Bains, Ernst Bamberg, Antonello Bonci, György Buzsáki, Jessica A Cardin, Rui M Costa, Yang Dan, Yukiko Goda, Ann M Graybiel, Michael Häusser, Peter Hegemann, John R Huguenard, Thomas R Insel, Patricia H Janak, Daniel Johnston, Sheena A Josselyn, Christof Koch, Anatol C Kreitzer, Christian Lüscher, Robert C Malenka, Gero Miesenböck, Georg Nagel, Botond Roska, Mark J Schnitzer, Krishna V Shenoy, Ivan Soltesz, Scott M Sternson, Richard W Tsien, Roger Y Tsien, Gina G Turrigiano, Kay M Tye & Rachel I Wilson On the anniversary of the Boyden et al. (2005) paper that introduced the use of channelrhodopsin in neurons, Nature Neuroscience asks selected members of the community to comment on the utility, impact and future of this important technique. euroscientists have long dreamed of and applied to a vast array of questions both in technique has had on neuroscience, we were Nthe ability to control neuronal activ- neuroscience and beyond. curious to know how researchers in the field ity with exquisite spatiotemporal precision. In the intervening years, improvements to feel the advances in optogenetic approaches In this issue, we celebrate the tenth anniver- early techniques have provided the community have influenced their work, what they think sary of a paper published in the September with an optogenetics tool box that has opened the future holds in terms of the application 2005 issue of Nature Neuroscience by a team the door to experiments we could have once of these techniques and what they see as the Nature America, Inc. All rights reserved. America, Inc. Nature led by Karl Deisseroth (Nat. Neurosci. 8, only dreamed of. Controlling neuronal activity obstacles we need to overcome to get there. 5 1263–1268 (2005)). In this study, the authors in real time, we now have the ability to deter- Toward this end, we’ve asked a number of expressed a light-sensitive microbial protein, mine causality between activity patterns in spe- scientists to share their thoughts with us in © 201 Channelrhodopsin-2 (ChR2), in neurons and cific neuronal circuits and brain function and this Q&A. Although we weren’t able to ask showed that exposing these neurons to pulses behavior, enabling researchers to definitively test more than a small fraction of the field, their of light could activate them in a temporally pre- long-held views and advance our understanding answers give an exciting view of the power and npg cise and reliable manner. In the decade since of brain function in both health and disease. potential of optogenetic approaches for under- this paper, ‘optogenetic’ approaches have been Anniversaries are often a time to reflect standing, and even potentially repairing, the widely and enthusiastically adopted by the field and, in light of the seminal influence this nervous system. How do you define optogenetics? Ernst Bamberg: Optogenetics is the use of of genetically encoded molecules to excite and genetically encoded light-activated proteins inhibit neurons. I would prefer it to mean any John Huguenard: Sensitizing neurons to light, for manipulation of cells in an almost non- genetically encoded tool to record or perturb then manipulating neural activity in precise invasive way by light. The most prominent electrical activity of neurons and other excit- spatiotemporal patterns to answer questions tool is ChR2, which allows in a cell-specific able cells. But using a broad name to identify regarding neural circuits and behavior. way the activation of electrical excitable cells a fairly specific subset of tasks was a brilliant via the light-dependent depolarization. The stroke, nonetheless. Michael Häusser: There’s a broad definition combination of ChR2 with hyperpolarizing and a narrow definition. The broad defini- light-driven ion pumps such as the Cl- pump Dan Johnston: I suppose that the accurate tion is rooted in etymology: any approach that halorhodopsin (NphR) allows, with high tem- definition would be genetically encoded opti- combines optical interrogation with genetic poral and spatial precision, the activation or cal sensors, but it is most commonly thought of targeting qualifies as ‘optogenetic’, and that inactivation of neural cells in culture, tissue as genetically encoded light-activated channels. includes the use of genetically encoded activ- and living animals. It’s worth noting, however, and I wasn’t aware of ity sensors. However, most people generally it at the time I reviewed the Boyden paper, but use the term optogenetics to mean the use of Richard Tsien: This 10-year celebration, well- there were two papers that predated this one probes to manipulate activity, and (as is usual deserved by the authors and journal alike, that reported a genetically encoded light-acti- in English) usage normally wins. implicitly points to a narrower definition: use vated channel: Zemelman, B.V. et al. Neuron 33, 1202 VOLUME 18 | NUMBER 9 | SEPTEMBER 2015 NATURE NEUROSCIENCE Q&A 15–22 (2002) and Zemelman, B.V. et al. Proc. would really work in the manners that were clear that the technique was very robust and rep- Natl. Acad. Sci. USA 100, 1352–1357 (2003). promised. I thought it might have some finite licable. Shortly thereafter I decided to do a post- uses, but did not imagine it would be as “revo- doc focusing on this approach, and that’s how I Mark Schnitzer: lutionary” as it turned out to be. ended up at Stanford with Karl Deisseroth. The term ‘optoge- netics’ first appeared Krishna Shenoy: I Antoine Adamantidis: At that time, I was in 2006 in a short had the great pleasure finishing my PhD at the University of Liege, review article pub- of being right here at Belgium, and we were investigating the role lished in Journal Stanford and know- of a unique popula- of Neuroscience to ing Ed and Karl for tion of neurons in the Mark Schnitzer accompany a Society years, so yes, when lateral hypothalamus for Neuroscience their results came in (that is, melanin- mini-symposium that Karl Deisseroth and I and the paper came Krishna Shenoy concentrating hor- had organized. (Deisseroth, K. et al. J. Neurosci. out it was clear it mone) in controlling 26, 10380–10386 (2006)). We considered differ- would be an enormous advance. Could I have ‘dream sleep’. In mam- ent options, such as ‘photogenetics’, but even- predicted how revolutionary it would be? mals, this deep sleep Antoine Adamantidis tually settled on optogenetics as the best term No, there I’m afraid I would have underesti- stage lasts classically to describe techniques that combined optical mated the full extent to which it has been a few minutes, which made it difficult to study and genetic facets. Notably, our original intent neuroscience-wide seismic shift! with conventional ‘low temporal’ approaches was to cover both genetically targeted optical (pharmacology, KO, KI, etc.) without altering control and imaging under a single umbrella Sheena Josselyn: I thought the data were other sleep stages. Thus, when the Boyden/ term. Nevertheless, I have subsequently always interesting, but likely not replicable and defi- Deisseroth publication came out, I thought, preferred a narrower interpretation of optoge- nitely not generalizable. I thought optogenet- “This is it! That’s what we need!” Since I was netics that covers only the control approaches ics would not work reliably and, even if it did, joining the laboratory of Professor Luis de and the wonderful field that grew out of Karl’s the technique would be so complicated as to Lecea at Stanford University a few months later, seminal 2005 paper in Nature Neuroscience; be out of reach for most neuroscience labs. My I emailed Karl about this idea, who replied, “OK, the broader interpretation of optogenetics that initial impression was that optogenetics would let’s meet when you get here!” We did meet, includes imaging is so general that, in some be highly parameter-sensitive and would take and, together with Dr. Feng Zhang, brought it respects, it can be vague. My impression is that lots of fiddling to get any kind of effect. I was to brain slices and freely moving mice to publish a substantial majority of the usages of the term definitely in the camp that didn’t think it would the first in vivo optogenetic paper establishing optogenetics in the neuroscience literature fol- have an impact on my kind of neuroscience. a causal role between hypocretin/orexin cells lows the narrower interpretation. and arousal (Adamantidis, A. et al., Nature 450, Scott Sternson: I thought that if it worked as 420–424 (2007)). Thus, yes, I believe this was a Rachel Wilson: I think of optogenetic tools as Nature America, Inc. All rights reserved. America, Inc. Nature advertised, then it would be exactly the tool transformative technology since early on! 5 a set of wrenches in a larger toolkit of geneti- that I’d been looking for since I started in neu- cally encoded effec- roscience. Thomas Insel: While everyone assumes opto- © 201 tors. This includes genetics is a great technology, reviewers on NIH effectors activated Gina Turrigiano: Intense excitement. I study sections in 2005 by heat, as well as thought the work leading up to this study was did not embrace this npg effectors activated by a beautiful example of basic curiosity-driven idea. Fortunately, a designer drugs, etc. research (trying to understand the basis of very smart program Optogenetic tools are bacterial phototaxis) leading to an unantici- officer at NIMH rec- Rachel Wilson often the most useful pated transformative outcome. ognized the promise because light can be of this proposal. And modulated so rapidly and precisely. However, Kay Tye: My first Thomas Insel soon after this first we should just reach for the tool that suits the reaction was one of NIMH K award, the job.