For a few years now, I’ve taught a course at MIT called “Principles of Neuroengineering.” The idea of the class is to get students thinking about how to create neurotechnology innovations—new inventions that can solve outstanding scientific questions or address unmet clinical needs. Designing neurotechnologies is difficult because of the complex properties of the brain: its inaccessibility, heterogeneity, fragility, anatomical richness, and high speed of operation. To illustrate the process, I decided to write a case study about the birth and development of an innovation with which I have been intimately involved: optogenetics—a toolset of genetically encoded molecules that, when targeted to specific neurons in the brain, allow the activity of those neurons to be driven or silenced by light.
A strategy: controlling the brain with lightAs an undergraduate at MIT, I studied physics and electrical engineering and got a good deal of firsthand experience in designing methods to control complex systems. By the time I graduated, I had become quite interested in developing strategies for understanding and engineering the brain. After graduating in 1999, I traveled to Stanford to begin a PhD in neuroscience, setting up a home base in Richard Tsien’s lab. In my first year at Stanford I was fortunate enough to meet many nearby biologists willing to do collaborative experiments, ranging from attempting the assembly of complex neural circuits in vitro to behavioral experiments with rhesus macaques. For my thesis work, I joined the labs of Richard Tsien and of Jennifer Raymond in spring 2000, to study how neural circuits adapt in order to control movements of the body as the circumstances in the surrounding world change.
In parallel, I started thinking about new technologies for controlling the electrical activity of specific neuron types embedded within intact brain circuits. That spring, I discussed this problem—during brainstorming sessions that often ran late into the night—with Karl Deisseroth, then a Stanford MD-PhD student also doing research in Tsien’s lab. We started to think about delivering stretch-sensitive ion channels to specific neurons, and then tethering magnetic beads selectively to the channels, so that applying an appropriate magnetic field would result in the bead’s moving and opening the ion channel, thus activating the targeted neurons.
By late spring 2000, however, I had become fascinated by a simpler and potentially easier-to-implement approach: using naturally occurring microbial opsins, which would pump ions into or out of neurons in response to light. Opsins had been studied since the 1970s because of their fascinating biophysical properties, and for the evolutionary insights they offer into how life forms use light as an energy source or sensory cue.[1. D. Oesterhelt, W. Stoeckenius, “Rhodopsin-like protein from the purple membrane of Halobacterium halobium,” Nat New Biol, 233:149-52, 1971.] These membrane-spanning microbial molecules—proteins with seven helical domains—react to light by transporting ions across the lipid membranes of cells in which they are genetically expressed. (See the illustration above.) For this strategy to work, an opsin would have to be expressed in the neuron’s lipid membrane and, once in place, efficiently perform this ion-transport function. One reason for optimism was that bacteriorhodopsin had successfully been expressed in eukaryotic cell membranes—including those of yeast cells and frog oocytes—and had pumped ions in response to light in these heterologous expression systems. And in 1999, researchers had shown that, although many halorhodopsins might work best in the high salinity environments in which their host archaea naturally live (i.e., in very high chloride concentrations), a halorhodopsin from Natronomonas pharaonis (Halo/NpHR) functioned best at chloride levels comparable to those in the mammalian brain.[2. D. Okuno et al., "Chloride concentration dependency of the electrogenic activity of halorhodopsin," Biochemistry, 38:5422-29, 1999.]
I was intrigued by this, and in May 2000 I e-mailed the opsin pioneer Janos Lanyi, asking for a clone of the N. pharaonis halorhodopsin, for the purpose of actively controlling neurons with light. Janos kindly asked his collaborator Richard Needleman to send it to me. But the reality of graduate school was setting in: unfortunately, I had already left Stanford for the summer to take a neuroscience class at the Marine Biology Laboratory in Woods Hole. I asked Richard to send the clone to Karl. When I returned to Stanford in the fall, I was so busy learning all the skills I would need for my thesis work on motor control that the opsin project took a backseat for a while.
The channelrhodopsin collaborationIn 2002 a pioneering paper from the lab of Gero Miesenböck showed that genetic expression of a three-gene Drosophila phototransduction cascade in neurons allowed the neurons to be excited by light, and suggested that the ability to activate specific neurons with light could serve as a tool for analyzing neural circuits.[3. B.V. Zemelman et al., "Selective photostimulation of genetically chARGed neurons," Neuron, 33:15-22, 2002.] But the light-driven currents mediated by this system were slow, and this technical issue may have been a factor that limited adoption of the tool.
This paper was fresh in my mind when, in fall 2003, Karl e-mailed me to express interest in revisiting the magnetic-bead stimulation idea as a potential project that we could pursue together later—when he had his own lab, and I had finished my PhD and could join his lab as a postdoc. Karl was then a postdoctoral researcher in Robert Malenka’s lab (also at Stanford), and I was about halfway through my PhD. We explored the magnetic-bead idea between October 2003 and February 2004. Around that time I read a just-published paper by Georg Nagel, Ernst Bamberg, Peter Hegemann, and colleagues, announcing the discovery of channelrhodopsin-2 (ChR2), a light-gated cation channel and noting that the protein could be used as a tool to depolarize cultured mammalian cells in response to light.[4. G. Nagel et al., "Channelrhodopsin-2, a directly light-gated cation-selective membrane channel," PNAS, 100:13940-45, 2003.]
In February 2004, I proposed to Karl that we contact Georg to see if they had constructs they were willing to distribute. Karl got in touch with Georg in March, obtained the construct, and inserted the gene into a neural expression vector. Georg had made several further advances by then: he had created fusion proteins of ChR2 and yellow fluorescent protein, in order to monitor ChR2 expression, and had also found a ChR2 mutant with improved kinetics. Furthermore, Georg commented that in cell culture, ChR2 appeared to require little or no chemical supplementation in order to operate (in microbial opsins, the chemical chromophore all-trans-retinal must be attached to the protein to serve as the light absorber; it appeared to exist at sufficient levels in cell culture).
Finally, we were getting the ball rolling on targetable control of specific neural types. Karl optimized the gene expression conditions, and found that neurons could indeed tolerate ChR2 expression. Throughout July, working in off-hours, I debugged the optics of the Tsien-lab rig that I had often used in the past. Late at night, around 1 a.m. on August 4, 2004, I went into the lab, put a dish of cultured neurons expressing ChR2 into the microscope, patch-clamped a glowing neuron, and triggered the program that I had written to pulse blue light at the neurons. To my amazement, the very first neuron I patched fired precise action potentials in response to blue light. That night I collected data that demonstrated all the core principles we would publish a year later in Nature Neuroscience, announcing that ChR2 could be used to depolarize neurons.[5. E.S. Boyden et al., "Millisecond-timescale, genetically targeted optical control of neural activity," Nat Neurosci, 8:1263-68, 2005.] During that long, exciting first night of experimentation in 2004, I determined that ChR2 was safely expressed and physiologically functional in neurons. The neurons tolerated expression levels of the protein that were high enough to mediate strong neural depolarizations. Even with brief pulses of blue light, lasting just a few milliseconds, the magnitude of expressed-ChR2 photocurrents was large enough to mediate single action potentials in neurons, thus enabling temporally precise driving of spike trains. Serendipity had struck—the molecule was good enough in its wild-type form to be used in neurons right away. I e-mailed Karl, “Tired, but excited.” He shot back, “This is great!!!!!”
Transitions and optical neural silencers
In January 2005, Karl finished his postdoc and became an assistant professor of bioengineering and psychiatry at Stanford. Feng Zhang, then a first-year graduate student in chemistry (and now an assistant professor at MIT and at the Broad Institute), joined Karl’s new lab, where he cloned ChR2 into a lentiviral vector, and produced lentivirus that greatly increased the reliability of ChR2 expression in neurons. I was still working on my PhD, and continued to perform ChR2 experiments in the Tsien lab. Indeed, about half the ChR2 experiments in our first optogenetics paper were done in Richard Tsien’s lab, and I owe him a debt of gratitude for providing an environment in which new ideas could be pursued. I regret that, in our first optogenetics paper, we did not acknowledge that many of the key experiments had been done there. When I started working in Karl’s lab in late March 2005, we carried out experiments to flesh out all the figures for our paper, which appeared in Nature Neuroscience in August 2005, a year after that exhilarating first discovery that the technique worked.
Around that same time, Guoping Feng, then leading a lab at Duke University (and now a professor at MIT), began to make the first transgenic mice expressing ChR2 in neurons.[6. B.R. Arenkiel et al., "In vivo light-induced activation of neural circuitry in transgenic mice expressing channelrhodopsin-2," Neuron, 54:205-18, 2007.] Several other groups, including the Yawo, Herlitze, Landmesser, Nagel, Gottschalk, and Pan labs, rapidly published papers demonstrating the use of ChR2 in neurons in the months following.[7. T. Ishizuka et al., "Kinetic evaluation of photosensitivity in genetically engineered neurons expressing green algae light-gated channels," Neurosci Res, 54:85-94, 2006.],[8. X. Li et al., "Fast noninvasive activation and inhibition of neural and network activity by vertebrate rhodopsin and green algae channelrhodopsin," PNAS, 102:17816-21, 2005.],[9. G. Nagel et al., "Light activation of channelrhodopsin-2 in excitable cells of Caenorhabditis elegans triggers rapid behavioral responses," Curr Biol, 15:2279-84, 2005.],[10. A. Bi et al., "Ectopic expression of a microbial-type rhodopsin restores visual responses in mice with photoreceptor degeneration," Neuron, 50:23-33, 2006.] Clearly, the idea had been in the air, with many groups chasing the use of channelrhodopsin in neurons. These papers showed, among many other groundbreaking results, that no chemicals were needed to supplement ChR2 function in the living mammalian brain.
Almost immediately after I finished my PhD in October 2005, two months after our ChR2 paper came out, I began the faculty job search process. At the same time, I started a position as a postdoctoral researcher with Karl and with Mark Schnitzer at Stanford. The job-search process ended up consuming much of my time, and being on the road, I began doing bioengineering invention consulting in order to learn about other new technology areas that could be brought to bear on neuroscience. I accepted a faculty job offer from the MIT Media Lab in September 2006, and began the process of setting up a neuroengineering research group there.
Around that time, I began a collaboration with Xue Han, my then girlfriend (and a postdoctoral researcher in the lab of Richard Tsien), to revisit the original idea of using the N. pharaonis halorhodopsin to mediate optical neural silencing. Back in 2000, Karl and I had planned to pursue this jointly; there was now the potential for competition, since we were working separately. Xue and I ordered the gene to be synthesized in codon-optimized form by a DNA synthesis company, and, using the same Tsien-lab rig that had supported the channelrhodopsin paper, Xue acquired data showing that this halorhodopsin could indeed silence neural activity. Our paper[11. X. Han, E.S. Boyden, "Multiple-color optical activation, silencing, and desynchronization of neural activity, with single-spike temporal resolution," PLoS ONE, 2:e299, 2007.] appeared in the March 2007 issue of PLoS ONE; Karl’s group, working in parallel, published a paper in Nature a few weeks later, independently showing that this halorhodopsin could support light-driven silencing of neurons, and also including an impressive demonstration that it could be used to manipulate behavior in Caenorhabditis elegans.[12. F. Zhang et al., "Multimodal fast optical interrogation of neural circuitry," Nature, 446:633-39, 2007.] Later, both our groups teamed up to file a joint patent on the use of this halorhodopsin to silence neural activity. As a testament to the unanticipated side effects of following innovation where it leads you, Xue and I got married in 2009 (and she is now an assistant professor at Boston University).
I continued to survey a wide variety of microorganisms for better silencing opsins: the inexpensiveness of gene synthesis meant that it was possible to rapidly obtain genes codon-optimized for mammalian expression, and to screen them for new and interesting light-drivable neural functions. Brian Chow (now an assistant professor at the University of Pennsylvania) joined my lab at MIT as a postdoctoral researcher, and began collaborating with Xue. In 2008 they identified a new class of neural silencer, the archaerhodopsins, which were not only capable of high-amplitude neural silencing—the first such opsin that could support 100 percent shutdown of neurons in the awake, behaving animal—but also were capable of rapid recovery after having been illuminated for extended durations, unlike halorhodopsins, which took minutes to recover after long-duration illumination.[13. B.Y. Chow et al., "High-performance genetically targetable optical neural silencing by light-driven proton pumps," Nature, 463:98-102, 2010.] Interestingly, the archaerhodopsins are light-driven outward pumps, similar to bacteriorhodopsin—they hyperpolarize neurons by pumping protons out of the cells. However, the resultant pH changes are as small as those produced by channelrhodopsins (which have proton conductances a million times greater than their sodium conductances), and well within the safe range of neuronal operation. Intriguingly, we discovered that the H. salinarum bacteriorhodopsin, the very first opsin characterized in the early 1970s, was able to mediate decent optical neural silencing, suggesting that perhaps opsins could have been applied to neuroscience decades ago.
Beyond luck: systematic discovery and engineering of optogenetic toolsAn essential aspect of furthering this work is the free and open distribution of these optogenetic tools, even prior to publication. To facilitate teaching people how to use these tools, our lab regularly posts white papers on our website* with details on reagents and optical hardware (a complete optogenetics setup costs as little as a few thousand dollars for all required hardware and consumables), and we have also partnered with nonprofit organizations such as Addgene and the University of North Carolina Gene Therapy Center Vector Core to distribute DNA and viruses, respectively. We regularly host visitors to observe experiments being done in our lab, seeking to encourage the community building that has been central to the development of optogenetics from the beginning.
As a case study, the birth of optogenetics offers a number of interesting insights into the blend of factors that can lead to the creation of a neurotechnological innovation. The original optogenetic tools were identified partly through serendipity, guided by a multidisciplinary convergence and a neuroscience-driven knowledge of what might make a good tool. Clearly, the original serendipity that fostered the formation of this concept, and that accompanied the initial quick try to see if it would work in nerve cells, has now given way to the systematized luck of bioengineering, with its machines and algorithms designed to optimize the chances of finding something new. Many labs, driven by genomic mining and mutagenesis, are reporting the discovery of new opsins with improved light and color sensitivities and new ionic properties. It is to be hoped, of course, that as this systematized luck accelerates, we will stumble upon more innovations that can aid in dissecting the enormous complexity of the brain—beginning the cycle of invention again.Edward Boyden leads the Synthetic Neurobiology Group at MIT, where he is the Benesse Career Development Professor and associate professor of biological engineering and brain and cognitive science at the MIT Media Lab and the MIT McGovern Institute.
This article is adapted from a review in F1000 Biology Reports, DOI:10.3410/B3-11 (open access at http://f1000.com/reports/b/3/11). For citation purposes, please refer to that version.