Whole-Cell Recording of Neuronal Membrane Potential during Behavior
2017; Cell Press; Volume: 95; Issue: 6 Linguagem: Inglês
10.1016/j.neuron.2017.06.049
ISSN1097-4199
Autores Tópico(s)Neuroscience and Neural Engineering
ResumoNeuronal membrane potential is of fundamental importance for the mechanistic understanding of brain function. This review discusses progress in whole-cell patch-clamp recordings for low-noise measurement of neuronal membrane potential in awake behaving animals. Whole-cell recordings can be combined with two-photon microscopy to target fluorescently labeled neurons, revealing cell-type-specific membrane potential dynamics of retrogradely or genetically labeled neurons. Dual whole-cell recordings reveal behavioral modulation of membrane potential synchrony and properties of synaptic transmission in vivo. Optogenetic manipulations are also readily integrated with whole-cell recordings, providing detailed information about the effect of specific perturbations on the membrane potential of diverse types of neurons. Exciting developments for future behavioral experiments include dendritic whole-cell recordings and imaging, and use of the whole-cell recording pipette for single-cell delivery of drugs and DNA, as well as RNA expression profiling. Whole-cell recordings therefore offer unique opportunities for investigating the neuronal circuits and synaptic mechanisms driving membrane potential dynamics during behavior. Neuronal membrane potential is of fundamental importance for the mechanistic understanding of brain function. This review discusses progress in whole-cell patch-clamp recordings for low-noise measurement of neuronal membrane potential in awake behaving animals. Whole-cell recordings can be combined with two-photon microscopy to target fluorescently labeled neurons, revealing cell-type-specific membrane potential dynamics of retrogradely or genetically labeled neurons. Dual whole-cell recordings reveal behavioral modulation of membrane potential synchrony and properties of synaptic transmission in vivo. Optogenetic manipulations are also readily integrated with whole-cell recordings, providing detailed information about the effect of specific perturbations on the membrane potential of diverse types of neurons. Exciting developments for future behavioral experiments include dendritic whole-cell recordings and imaging, and use of the whole-cell recording pipette for single-cell delivery of drugs and DNA, as well as RNA expression profiling. Whole-cell recordings therefore offer unique opportunities for investigating the neuronal circuits and synaptic mechanisms driving membrane potential dynamics during behavior. The temporal pattern of action potential firing in diverse neurons distributed across the brain forms the basis of the neuronal code (Figure 1A). Action potential firing in sensory brain areas typically correlates with specific features of sensory input, whereas spiking in motor areas precedes and correlates with movement. Methods have been developed to simultaneously measure the action potential firing of many neurons using silicon probes with a high density and large number of recording sites (Buzsáki, 2004Buzsáki G. Large-scale recording of neuronal ensembles.Nat. Neurosci. 2004; 7: 446-451Crossref PubMed Scopus (886) Google Scholar, Rossant et al., 2016Rossant C. Kadir S.N. Goodman D.F. Schulman J. Hunter M.L. Saleem A.B. Grosmark A. Belluscio M. Denfield G.H. Ecker A.S. et al.Spike sorting for large, dense electrode arrays.Nat. Neurosci. 2016; 19: 634-641Crossref PubMed Scopus (329) Google Scholar). Action potential firing is invariably accompanied by calcium influx through voltage-gated calcium channels, and imaging of calcium-sensitive fluorophores can therefore also yield important, although indirect, measures of spiking activity in neuronal networks with single-cell resolution (Yuste and Katz, 1991Yuste R. Katz L.C. Control of postsynaptic Ca2+ influx in developing neocortex by excitatory and inhibitory neurotransmitters.Neuron. 1991; 6: 333-344Abstract Full Text PDF PubMed Scopus (421) Google Scholar, Stosiek et al., 2003Stosiek C. Garaschuk O. Holthoff K. Konnerth A. In vivo two-photon calcium imaging of neuronal networks.Proc. Natl. Acad. Sci. USA. 2003; 100: 7319-7324Crossref PubMed Scopus (816) Google Scholar). Electrophysiological and optical methods are therefore being developed to record from very large numbers of neurons simultaneously, providing invaluable information about the cellular structure of the neuronal code of action potentials. Action potentials are fired when the membrane potential (Vm) is depolarized beyond the threshold. The biophysical conductances underlying the action potential were quantitatively investigated at the giant axon of the squid by Hodgkin and Huxley, 1952Hodgkin A.L. Huxley A.F. A quantitative description of membrane current and its application to conduction and excitation in nerve.J. Physiol. 1952; 117: 500-544Crossref PubMed Google Scholar, who found that the rapid recruitment of voltage-gated sodium channels at the action potential threshold outpaced the delayed activation of voltage-gated potassium channels, giving rise to a positive feedback all-or-none event. The Hodgkin and Huxley formalism of the action potential has turned out to be generally applicable across species and, with small variations, accurately describes the action potential waveform in many types of neurons. Although the biophysical mechanisms underlying the action potential waveform itself are well-understood, what physiologically drives a neuron to the action potential threshold requires measurement of Vm in behaving animals. Such measurements reveal a wealth of subthreshold Vm fluctuations, some of which cross threshold-evoking action potentials (Figure 1B). The action potential might therefore be considered just the “tip of the iceberg” in terms of neuronal activity. Subthreshold changes in Vm are largely driven by the spatiotemporal integration of synaptic inputs arriving across the somatodendritic arborization. A unitary synaptic input can be considered the result of a single action potential in a single presynaptic neuron. Typically, unitary excitatory postsynaptic potentials (uEPSPs) and unitary inhibitory postsynaptic potentials (uIPSPs) are small in amplitude (∼1 mV) and have a short duration (∼10 ms) (Figure 1C). The large (∼20 mV) Vm fluctuations observed in vivo (Figure 1B) on both short (millisecond) and longer (second) timescales must therefore result from the integration of many synaptic inputs distributed across the complex morphology of single neurons. In addition, neuromodulatory input serves to change various membrane conductances, typically on longer timescales, affecting input resistance, excitability, and synaptic transmission. The fundamental computation a single neuron carries out is thus to integrate the incoming uEPSPs and uIPSPs to decide whether to fire an action potential. This synaptic computation can be investigated with whole-cell Vm recordings, and the application of this technique to behaving animals is the focus of this review. The patch-clamp recording technique was developed by Erwin Neher and Bert Sakmann to measure single-channel currents (Neher and Sakmann, 1976Neher E. Sakmann B. Single-channel currents recorded from membrane of denervated frog muscle fibres.Nature. 1976; 260: 799-802Crossref PubMed Google Scholar, Sakmann and Neher, 1995Sakmann B. Neher E. Single-channel recording. Plenum Publishing Corporation, New York, USA1995Google Scholar). They found that glass pipettes with a tip diameter of ∼1 μm filled with isotonic ionic solutions could make a tight electrical seal with lipid bilayer membranes. After establishing a seal with a membrane, the leak resistance between the inside of the pipette and the reference electrode could be greater than 1 gigaohm (GΩ), termed a “giga-seal.” The giga-seal resulted in low-noise recordings showing the opening and closing of single ion channels present in the patch of membrane sealed onto the end of the electrode (Hamill et al., 1981Hamill O.P. Marty A. Neher E. Sakmann B. Sigworth F.J. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches.Pflugers Arch. 1981; 391: 85-100Crossref PubMed Scopus (14435) Google Scholar). Not only has the giga-seal been found to be applicable to many different membranes, it has also turned out to be surprisingly stable mechanically. The remarkable mechanical stability of the giga-seal together with the low electrical noise of the recordings has made the patch-clamp method useful in many different experiments. Today, one of the most widely used configurations of the patch-clamp method is the whole-cell recording technique (Hamill et al., 1981Hamill O.P. Marty A. Neher E. Sakmann B. Sigworth F.J. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches.Pflugers Arch. 1981; 391: 85-100Crossref PubMed Scopus (14435) Google Scholar). The whole-cell recording configuration is established in a three-step process (Figure 1D). First, the glass pipette is filled with an intracellular solution into which a silver/silver chloride (Ag/AgCl) electrode is inserted, forming the electrochemical interface with the headstage of a patch-clamp amplifier. The patch pipette is further connected to an air-filled tube, allowing the pressure inside the pipette to be controlled. Using a micromanipulator, the glass patch-clamp electrode is inserted into the extracellular solution, which fills the recording chamber. Positive pressure is applied to the inside of the pipette, making a small flow of solution out of the electrode, which helps keep the tip of the glass electrode clean. This is essential because any dirt on the pipette tip will prevent giga-seal formation. The electrical circuit is completed by an Ag/AgCl electrode placed in the recording chamber, which serves as the reference electrode for the patch-clamp amplifier set to 0 mV as the extracellular potential. The electrical tip resistance is constantly monitored by applying voltage steps to the pipette in the voltage-clamp mode of the patch-clamp amplifier while measuring the current flow. In this “search” mode, the tip of the patch electrode has free access to the extracellular solution and typically has a resistance of ∼5 MΩ. Remarkably, the outflow of pipette solution can also move tissue out of the way of the advancing patch electrode, as found in recordings from brain slices in vitro (Blanton et al., 1989Blanton M.G. Lo Turco J.J. Kriegstein A.R. Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex.J. Neurosci. Methods. 1989; 30: 203-210Crossref PubMed Scopus (783) Google Scholar, Stuart et al., 1993Stuart G.J. Dodt H.U. Sakmann B. Patch-clamp recordings from the soma and dendrites of neurons in brain slices using infrared video microscopy.Pflugers Arch. 1993; 423: 511-518Crossref PubMed Google Scholar). As the patch-electrode tip touches a cell membrane, the experimenter removes the positive pressure from the inside of the pipette and gently sucks. The negative pressure in the pipette now pulls a small patch of membrane into the pipette, and the resistance between the inside of the electrode and the extracellular solution increases. As the glass and the lipid bilayer begin to form a tight seal, the pipette resistance increases rapidly, typically reaching many gigaohms. A giga-seal is thus established with the plasma membrane in the so-called “cell-attached” configuration of the patch-clamp method. Using the voltage-clamp amplifier, the potential of the patch electrode is typically set to a negative value close to the expected resting potential of the cell. Suction pulses are then applied to the inside of the patch pipette to rupture the membrane without affecting the tight electrical giga-seal. The solution inside the patch electrode now has direct intracellular access, and the currents flowing across the cell membrane can thus be measured with very little leak current because of the giga-seal. In this “whole-cell” patch-clamp recording configuration, voltage steps give rise to large transient currents because of the charging of the membrane capacitance. The steady-state current flow is also larger than in the cell-attached configuration, reflecting the larger conductance across the membrane of the whole cell compared with the small patch of membrane in the pipette tip. The patch-clamp amplifier can be used to measure whole-cell currents at different membrane potentials in the voltage-clamp mode, or, when the amplifier is switched to current-clamp mode, the experimenter can now measure Vm. Current injections can be used to measure the excitability, firing patterns, input resistance, and intrinsic conductances of the neuron, all of which can be modulated during behavior on a wide range of timescales. The whole-cell recording technique offers reliable, low-noise recording of membrane current and Vm, accompanied by high mechanical stability, allowing long-lasting recordings. In vivo whole-cell recordings from neurons in the mammalian brain were first obtained from the visual cortex of anesthetized cats (Pei et al., 1991Pei X. Volgushev M. Vidyasagar T.R. Creutzfeldt O.D. Whole cell recording and conductance measurements in cat visual cortex in-vivo.Neuroreport. 1991; 2: 485-488Crossref PubMed Google Scholar, Ferster and Jagadeesh, 1992Ferster D. Jagadeesh B. EPSP-IPSP interactions in cat visual cortex studied with in vivo whole-cell patch recording.J. Neurosci. 1992; 12: 1262-1274Crossref PubMed Google Scholar, Jagadeesh et al., 1993Jagadeesh B. Wheat H.S. Ferster D. Linearity of summation of synaptic potentials underlying direction selectivity in simple cells of the cat visual cortex.Science. 1993; 262: 1901-1904Crossref PubMed Google Scholar, Nelson et al., 1994Nelson S. Toth L. Sheth B. Sur M. Orientation selectivity of cortical neurons during intracellular blockade of inhibition.Science. 1994; 265: 774-777Crossref PubMed Scopus (215) Google Scholar) and later from anesthetized rats (Moore and Nelson, 1998Moore C.I. Nelson S.B. Spatio-temporal subthreshold receptive fields in the vibrissa representation of rat primary somatosensory cortex.J. Neurophysiol. 1998; 80: 2882-2892Crossref PubMed Google Scholar, Zhu and Connors, 1999Zhu J.J. Connors B.W. Intrinsic firing patterns and whisker-evoked synaptic responses of neurons in the rat barrel cortex.J. Neurophysiol. 1999; 81: 1171-1183Crossref PubMed Google Scholar, Margrie et al., 2002Margrie T.W. Brecht M. Sakmann B. In vivo, low-resistance, whole-cell recordings from neurons in the anaesthetized and awake mammalian brain.Pflugers Arch. 2002; 444: 491-498Crossref PubMed Scopus (352) Google Scholar). These pioneering studies used the so-called “blind” recording technique (Blanton et al., 1989Blanton M.G. Lo Turco J.J. Kriegstein A.R. Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex.J. Neurosci. Methods. 1989; 30: 203-210Crossref PubMed Scopus (783) Google Scholar), in which the patch pipette is advanced into the brain in small steps (typically ∼2 μm) while monitoring the tip resistance but without any visual control of the pipette approaching individual cells. A rapid increase in tip resistance after the pipette is moved forward indicates possible contact of the electrode tip with a cell membrane. Suction can then be applied and a whole-cell recording established. The same methods used for recording under anesthesia can also be applied to obtain whole-cell recordings from awake animals (Figure 1E). This is important because anesthesia changes the brain state, alters many aspects of neuronal activity, and prevents movements. Recordings from awake animals allow the activity of neurons to be directly correlated with behavior, which is an essential step for understanding brain function. The key to obtaining whole-cell recordings in behaving animals is to minimize the relative motion of the brain and the patch-clamp electrode. The simplest method is to implant a metal head fixation post directly to the skull under anesthesia. For mice, such a post can simply be cemented to the cleaned skull using cyanoacrylate glue reinforced with dental cement. When the headpost is attached to a stable holder mounted on an air table, the movement of the skull relative to the table is typically on the order of a few micrometers, even when the mouse is active. Mice readily adapt to head restraint within a few daily training sessions. Additionally, placing the mouse on a floating ball may help to further minimize skull movement (Dombeck et al., 2007Dombeck D.A. Khabbaz A.N. Collman F. Adelman T.L. Tank D.W. Imaging large-scale neural activity with cellular resolution in awake, mobile mice.Neuron. 2007; 56: 43-57Abstract Full Text Full Text PDF PubMed Scopus (497) Google Scholar). However, even after complete stabilization of the skull, the brain can still move a small amount inside the skull (for example, when a thirsty mouse is vigorously licking a reward spout), but heartbeat pulsation and breathing-related brain movements can also be observed, depending upon the preparation. A craniotomy, followed by opening of the dura, is essential to introduce the patch-clamp recording electrode, and making craniotomies as small as possible is helpful in reducing brain movement. Despite these small brain movements on the order of a few microns, high-quality whole-cell recordings can be obtained from awake rats (Margrie et al., 2002Margrie T.W. Brecht M. Sakmann B. In vivo, low-resistance, whole-cell recordings from neurons in the anaesthetized and awake mammalian brain.Pflugers Arch. 2002; 444: 491-498Crossref PubMed Scopus (352) Google Scholar) and mice (Petersen et al., 2003Petersen C.C.H. Hahn T.T.G. Mehta M. Grinvald A. Sakmann B. Interaction of sensory responses with spontaneous depolarization in layer 2/3 barrel cortex.Proc. Natl. Acad. Sci. USA. 2003; 100: 13638-13643Crossref PubMed Scopus (449) Google Scholar), presumably because of the remarkable mechanical stability of the giga-seal between glass and lipid membrane. Whole-cell recordings of Vm in awake mammals have now been obtained using this blind patch-clamp technique under a wide variety of interesting experimental conditions. Early recordings in awake head-restrained mice highlighted the surprising prevalence of slow, large-amplitude Vm fluctuations during quiet wakefulness (Petersen et al., 2003Petersen C.C.H. Hahn T.T.G. Mehta M. Grinvald A. Sakmann B. Interaction of sensory responses with spontaneous depolarization in layer 2/3 barrel cortex.Proc. Natl. Acad. Sci. USA. 2003; 100: 13638-13643Crossref PubMed Scopus (449) Google Scholar). In a more complex virtual reality setup, whole-cell recordings were obtained from the hippocampus of head-restrained mice running on a floating trackball during virtual navigation (Harvey et al., 2009Harvey C.D. Collman F. Dombeck D.A. Tank D.W. Intracellular dynamics of hippocampal place cells during virtual navigation.Nature. 2009; 461: 941-946Crossref PubMed Scopus (403) Google Scholar). Blind whole-cell recordings of the Vm correlates of tactile sensory perception have also been investigated in mice trained to lick a reward spout in response to detected whisker stimuli (Sachidhanandam et al., 2013Sachidhanandam S. Sreenivasan V. Kyriakatos A. Kremer Y. Petersen C.C.H. Membrane potential correlates of sensory perception in mouse barrel cortex.Nat. Neurosci. 2013; 16: 1671-1677Crossref PubMed Scopus (116) Google Scholar, Zagha et al., 2015Zagha E. Ge X. McCormick D.A. Competing neural ensembles in motor cortex gate goal-directed motor output.Neuron. 2015; 88: 565-577Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar, Yang et al., 2016Yang H. Kwon S.E. Severson K.S. O’Connor D.H. Origins of choice-related activity in mouse somatosensory cortex.Nat. Neurosci. 2016; 19: 127-134Crossref PubMed Scopus (29) Google Scholar). The blind whole-cell recording technique is not limited to rodents, and, indeed, such Vm recordings have also been obtained from the primary visual cortex of awake head-restrained macaque monkeys trained to fixate while viewing drifting gratings (Tan et al., 2014Tan A.Y. Chen Y. Scholl B. Seidemann E. Priebe N.J. Sensory stimulation shifts visual cortex from synchronous to asynchronous states.Nature. 2014; 509: 226-229Crossref PubMed Scopus (85) Google Scholar). In general, the blind whole-cell recording technique is simple and robust, requiring relatively little equipment, but involves extensive training and technical skill of the experimenter. Whole-cell recording is labor-intense, and one drawback of the whole-cell patch-clamp technique is the need to use a new pipette for each attempt to obtain a recording. Thus, the electrode must be withdrawn from the brain, changed for a new pipette, and then reinserted into the target location. Some of the steps involved in obtaining a whole-cell recording in vivo have now been automated through the development of patch-clamp robots (Kodandaramaiah et al., 2012Kodandaramaiah S.B. Franzesi G.T. Chow B.Y. Boyden E.S. Forest C.R. Automated whole-cell patch-clamp electrophysiology of neurons in vivo.Nat. Methods. 2012; 9: 585-587Crossref PubMed Scopus (83) Google Scholar, Kolb et al., 2016Kolb I. Stoy W.A. Rousseau E.B. Moody O.A. Jenkins A. Forest C.R. Cleaning patch-clamp pipettes for immediate reuse.Sci. Rep. 2016; 6: 35001Crossref PubMed Scopus (6) Google Scholar), and with each step of automation, the technique should become easier for the experimenter. Membrane potential can also be measured using classical sharp glass microelectrodes, which have very fine tip diameters and are filled with a high concentration of ions (typically ∼3 M KCl). Because of their very thin tip, these electrodes can directly penetrate the cell membrane, and they can be used to record Vm in vivo simply by advancing the electrode. However, microelectrode recordings typically have a large leak conductance because of the absence of a tight electrical seal between the electrode and the cell membrane. In addition, the small-diameter electrodes can bend in response to brain movements, which may create artifactual electrical signals in measurements. Typically, recording reliability, quality, and duration are improved with the use of whole-cell recordings compared with classical microelectrodes, but there are also examples of sharp microelectrode recordings in awake animals; for example, in cats (Timofeev et al., 2001Timofeev I. Grenier F. Steriade M. Disfacilitation and active inhibition in the neocortex during the natural sleep-wake cycle: an intracellular study.Proc. Natl. Acad. Sci. USA. 2001; 98: 1924-1929Crossref PubMed Google Scholar), songbirds (Long et al., 2010Long M.A. Jin D.Z. Fee M.S. Support for a synaptic chain model of neuronal sequence generation.Nature. 2010; 468: 394-399Crossref PubMed Scopus (174) Google Scholar), and also mice (Schneider et al., 2014Schneider D.M. Nelson A. Mooney R. A synaptic and circuit basis for corollary discharge in the auditory cortex.Nature. 2014; 513: 189-194Crossref PubMed Scopus (134) Google Scholar). Neurons in different parts of the brain process distinct information, and knowing the exact location of the cell body of a recorded neuron can thus be helpful. Even nearby neurons can have very different functional properties, which, in some cases, may correlate with dendritic and axonal morphology as well as gene expression. Recordings from anatomically identified neurons are therefore of critical importance for a detailed understanding of the cellular mechanisms of brain function. During whole-cell recording, the pipette solution diffuses into the recorded cell. If the pipette solution contains biocytin (a biotin-lysine complex), then the biocytin will enter the cell, filling the soma, dendrites, and axons. At the end of the recording, the patch electrode is withdrawn very slowly while applying voltage-clamp pulses to monitor tip resistance. If a patch of membrane reseals on the pipette tip, forming a so-called “outside-out” patch, then the neuronal membrane also reseals, leaving the recorded neuron intact. Next, the brain is chemically fixed by transcardial perfusion of paraformaldehyde and sectioned into thin slices. The recorded neuron can then be stained by applying avidin conjugated to enzymes or fluorescent molecules (Horikawa and Armstrong, 1988Horikawa K. Armstrong W.E. A versatile means of intracellular labeling: injection of biocytin and its detection with avidin conjugates.J. Neurosci. Methods. 1988; 25: 1-11Crossref PubMed Scopus (734) Google Scholar). These methods have been applied to neurons recorded in awake head-restrained mice; for example, in the context of investigating active touch. Primary sensory cortices have precise maps that process specific sensory information. The primary somatosensory cortex of mice contains an anatomical map of the facial whiskers, with each whisker being individually represented in layer 4 by a so-called “barrel” (Woolsey and Van der Loos, 1970Woolsey T.A. Van der Loos H. The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units.Brain Res. 1970; 17: 205-242Crossref PubMed Scopus (1672) Google Scholar). Each cortical barrel column processes sensory information predominantly from its aligned whisker; for example, neurons in the C2 barrel column primarily process information related to the C2 whisker (Petersen, 2007Petersen C.C.H. The functional organization of the barrel cortex.Neuron. 2007; 56: 339-355Abstract Full Text Full Text PDF PubMed Google Scholar). It is therefore essential to know the location of a recorded neuron within the somatotopic barrel map. The cortex is also divided into different layers containing different types of cells, and it is therefore also necessary to know the laminar location of a recorded neuron. As an example, Crochet et al., 2011Crochet S. Poulet J.F.A. Kremer Y. Petersen C.C.H. Synaptic mechanisms underlying sparse coding of active touch.Neuron. 2011; 69: 1160-1175Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar obtained whole-cell Vm recordings from neurons in the primary whisker somatosensory cortex of awake head-restrained mice actively palpating objects with their C2 whiskers (Figure 2A). Biocytin was introduced into the neurons during the whole-cell recordings, and the fixed tissue was subsequently stained with avidin conjugated to peroxidase, forming a brown precipitate in a reaction with diaminobenzidene. The labeled somata, descending main axons, and dendrites were subsequently digitally reconstructed in three dimensions in the context of the barrel maps labeled with cytochrome oxidase. The Vm dynamics during active touch of different post hoc-identified neurons in the C2 barrel column differed significantly depending upon the subpial depth of the recorded neuron, with long-lasting, slow depolarizations in layer 2 and faster, transient active touch responses in layer 3 (Crochet et al., 2011Crochet S. Poulet J.F.A. Kremer Y. Petersen C.C.H. Synaptic mechanisms underlying sparse coding of active touch.Neuron. 2011; 69: 1160-1175Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar). Many further whole-cell recording studies in awake head-restrained rodents have used biocytin to anatomically identify the recorded neurons, including the characterization of stellate cells in the medial entorhinal cortex during virtual navigation (Schmidt-Hieber and Häusser, 2013Schmidt-Hieber C. Häusser M. Cellular mechanisms of spatial navigation in the medial entorhinal cortex.Nat. Neurosci. 2013; 16: 325-331Crossref PubMed Scopus (119) Google Scholar), differentiation of neurons in layer 5 of the mouse motor cortex (Schiemann et al., 2015Schiemann J. Puggioni P. Dacre J. Pelko M. Domanski A. van Rossum M.C. Duguid I. Cellular mechanisms underlying behavioral state-dependent bidirectional modulation of motor cortex output.Cell Rep. 2015; 11: 1319-1330Abstract Full Text Full Text PDF PubMed Google Scholar), differentiation of direct and indirect pathway neurons in the dorsolateral striatum during a whisker detection task (Sippy et al., 2015Sippy T. Lapray D. Crochet S. Petersen C.C.H. Cell-type-specific sensorimotor processing in striatal projection neurons during goal-directed behavior.Neuron. 2015; 88: 298-305Abstract Full Text Full Text PDF PubMed Google Scholar), localization of neurons in layer 4 of the mouse barrel cortex during a whisker-dependent object localization task (Yu et al., 2016Yu J. Gutnisky D.A. Hires S.A. Svoboda K. Layer 4 fast-spiking interneurons filter thalamocortical signals during active somatosensation.Nat. Neurosci. 2016; 19: 1647-1657Crossref PubMed Scopus (4) Google Scholar), and localization of neurons in different layers of the mouse forepaw somatosensory cortex (Zhao et al., 2016Zhao W.J. Kremkow J. Poulet J.F.A. Translaminar cortical membrane potential synchrony in behaving mice.Cell Rep. 2016; 15: 2387-2399Abstract Full Text Full Text PDF PubMed Google Scholar). The blind whole-cell recording technique allows recordings from many different brain regions, but, within a given brain area, there can be many different types of neurons. Neuron types that are very prevalent are more likely to be recorded. For example, in the neocortex, most neurons are excitatory pyramidal neurons, and the majority of neurons recorded by stepping blindly through the cortex with a patch electrode are pyramidal neurons. One approach to selectively record from specific types of neurons is to label them with fluorescence and then target whole-cell recording electrodes to these cells under visual guidance. However, brain tissue strongly scatters visible light, and therefore high-resolution optical imaging in the mammalian brain is challenging. Two-photon microscopy can resolve micron-scale fluorescently labeled st
Referência(s)