An Image-Free Opto-Mechanical System for Creating Virtual Environments and Imaging Neuronal Activity in Freely Moving Caenorhabditis elegans Serge Faumont 1 , Gary Rondeau 2 , Tod R. Thiele 3 , Kristy J. Lawton 4 , Kathryn E. McCormick 1 , Matthew Sottile 5 , Oliver Griesbeck 6 , Ellie S. Heckscher 7 , William M. Roberts 1 , Chris Q. Doe 7 , Shawn R. Lockery 1 * 1 Institute of Neuroscience, University of Oregon, Eugene, Oregon, United States of America, 2 Applied Scientific Instrumentation, Eugene, Oregon, United States of America, 3 University of California San Francisco, San Francisco, California, United States of America, 4 Department of Neurobiology and Behavior, Cornell University, Ithaca, New York, United States of America, 5 Galois Inc., Portland, Oregon, United States of America, 6 Max-Planck-Institute of Neurobiology, Martinsried, Germany, 7 Howard Hughes Medical Institute, Institute of Neuroscience, Institute of Molecular Biology, University of Oregon, Eugene, Oregon, United States of America Abstract Non-invasive recording in untethered animals is arguably the ultimate step in the analysis of neuronal function, but such recordings remain elusive. To address this problem, we devised a system that tracks neuron-sized fluorescent targets in real time. The system can be used to create virtual environments by optogenetic activation of sensory neurons, or to image activity in identified neurons at high magnification. By recording activity in neurons of freely moving C. elegans, we tested the long-standing hypothesis that forward and reverse locomotion are generated by distinct neuronal circuits. Surprisingly, we found motor neurons that are active during both types of locomotion, suggesting a new model of locomotion control in C. elegans. These results emphasize the importance of recording neuronal activity in freely moving animals and significantly expand the potential of imaging techniques by providing a mean to stabilize fluorescent targets. Citation: Faumont S, Rondeau G, Thiele TR, Lawton KJ, McCormick KE, et al. (2011) An Image-Free Opto-Mechanical System for Creating Virtual Environments and Imaging Neuronal Activity in Freely Moving Caenorhabditis elegans. PLoS ONE 6(9): e24666. doi:10.1371/journal.pone.0024666 Editor: Aravinthan Samuel, Harvard University, United States of America Received July 25, 2011; Accepted August 15, 2011; Published September 28, 2011 Copyright: ß 2011 Faumont et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by funding from the National Institutes of Health (NIH) grant MH051383-13. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have read the journal’s policy and have the following conflicts: Gary Rondeau is employed by Applied Scientific Instrumentation (Eugene, OR) which builds a commercial version of the tracking system. This does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials. Matthew Sottile is employed by Galois Inc. (Portland, OR) which has no financial or intellectual-property interests in the tracking system or in the software that supports it. Dr. Sottile’s efforts on the project were and are entirely external to Galois Inc. Thus, his relationship to this company does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials. The other authors have declared that no competing interests exist. * E-mail: shawn@chinook.uoregon.edu Introduction The ability to record the activity of particular neurons non- invasively in untethered, freely moving animals would greatly accelerate studies of the neuronal basis of behavior, but such recordings remain generally elusive. Head-mounted telemetry systems have made single unit recordings possible in untethered mammals and birds [1], but this approach is invasive and the identities of recorded neurons are often uncertain. Truly non- invasive recordings have been made in freely moving zebrafish larvae by means of an activity-dependent bioluminescent probe that is transgenically expressed in target neurons and monitored by a wide-field photodetector [2]. However, this approach reports the summed activity of all neurons expressing the probe, making it impossible to assign signals to individual neurons. Thus, current technology has allowed only limited progress toward recording the activity of identified neurons in freely moving animals. To record from identified neurons in untethered animals it is usually necessary to genetically target an optical probe to known subsets of neurons. Such recordings have been achieved from groups of neurons, or spatially isolated single neurons, in freely moving Caenorhabditis elegans. This was done by periodically recentering the target in the field of view either manually [3] or by means of image processing software that controlled a motorized stage [4], an approach developed more than a decade ago for behavioral tracking experiments [5]. A fundamental limitation of this method is that under conditions of normal locomotion, the target neuron moves large distances during the time taken to process the neuronal image and recenter the stage. As a result, the neuron escapes the field of view unless a wide-field microscope objective is used. However, the magnification of such objectives (#20 6 ) is insufficient to resolve the somata of individual C. elegans neurons unless they happen to be far apart. Thus, the image processing approach is not a general method for recording from neurons of interest in freely moving C. elegans and other model organisms. To develop a general method for recording from identified neurons in freely moving C. elegans and other model organisms, we devised an image-free, opto-mechanical system that recenters fluorescent targets moving at speeds of up to 500 mm/s. Importantly, the new system is compatible with high performance microscope objectives (63 6 –100 6 , 1.3–1.4 N.A.), making it possible to resolve densely packed neurons in freely moving animals for the first time. The system is based on a high-speed feedback loop between the location of the target in the field of view and compensating movements of the microscope stage, reducing the latency of recentering movements by a factor of up to 40 PLoS ONE | www.plosone.org 1 September 2011 | Volume 6 | Issue 9 | e24666