Using novel technology to elucidate neocortical microcircuits with multiple simultaneous whole-cell recordings

In recent years, the paired-recording technique has been developed to yield large databases on neuronal connectivity (e.g. 1, 2, 3) and synaptic weight distributions (4), revealing e.g. the information storage capacity of pyramidal neurons (5) and enabling full-scale simulations of cortical columns (6). However, paired recordings are hampered by the sparse connectivity (1, 2, 4, 5), so multiple simultaneous whole-cell recordings are typically used to yield data fast and of high statistical order (1). Now, the number of recordings ranges from four up to twelve (1-3). Such recordings obviously place demands on the experimenter, but they are quite tractable with new technology. The project outlined here below aims to achieve six to eight simultaneous recordings to rapidly elucidate the microcircuit of the mouse visual cortex by benefiting from already existing prototype equipment and an already existing collaboration between Scientifica UK and the Sjöström lab. For example, Scientifica has already developed dedicated software, 'Follow,' that enables automatic cross mapping of the different coordinate systems of multiple manipulators and the microscope/XY stage. Some of the more technological aspects of this project are already underway, being funded by a Royal Society Industrial Fellowship to Dr. Dale Elgar. The present proposal is more biological, but will still require a close report with R&D at Scientifica UK. The aims are: Aim 1: To Build Rig with Six to Eight High-Performance Slim-Line Robotic Micromanipulators By developing a slim-line manipulator, more electrodes can readily be positioned around the sample. Since the number of possible connected pairs of neurons scales as n*(n-1), where n is the number of electrodes, the yield of actual connected cells scales very favourably with n. Presently, Scientifica has two prototype slim line manipulators that we are about to test. This coming fall, this design will be ready for experiments at UCL, but setting such a complex rig up will require time and some additional development. A technically minded CASE PhD student will be able to do this; this straightforward low-risk aim will provide Scientifica UK will important feedback. Aim 2: To Elucidate the Visual Cortical Microcircuitry While investigating long-term plasticity, Dr Sjöström previously mapped the layer-5 pyramidal network of rat visual cortex using quadruple recordings (1). Here, however, the effort will be dedicated solely at elucidating the microcircuit of mouse visual cortex (cf. 3). In recordings lasting a few minutes, connectivity, short-term plasticity, and synaptic efficacy will be assessed (not long-term plasticity). We already use transgenic mice expressing GFP in somatostatin and parvalbumin-positive interneurons; these permit targeting of e.g. Martinotti cells (2). Connections across all layers will be examined (3). Morphologies will be imaged using already an existing custom-built confocal attachment. Morphologies and spiking patterns will be used to classify cells. This project is high profile and synergistic with the Blue Brain project (6). Though whole-cell recordings are considered difficult, a PhD student will be successful with Dr Sjöström's expertise (through experiment optimization & 'tricks', software and hardware solutions, etc) and Scientifica's novel technology. It may be possible to examine two ages: developing and mature. If progress is slow, however, he student could also focus just on layers 4 and 6 in developing cortex, since we are already working on layers 2, 3, and 5 in other contexts. References 1. S. Song, P. J. Sjöström, S. Nelson, D. B. Chklovskii, PLoS Biol 3, e68 (2005). 2. G. Silberberg, H. Markram, Neuron 53, 735 (2007). 3. S. Lefort et al, Neuron 61, 301 (2009). 4. B. Barbour et al, Trends Neurosci (2007). 5. L. R. Varshney, P. J. Sjöström, D. B. Chklovskii, Neuron 52, 409 (2006). 6. H. Markram, Nat Rev Neurosci 7, 153 (2006).