Assembly and operation of the autopatcher for automated intracellular neural recording in vivo

Journal name:
Nature Protocols
Volume:
11,
Pages:
634–654
Year published:
DOI:
doi:10.1038/nprot.2016.007
Published online

Abstract

Whole-cell patch clamping in vivo is an important neuroscience technique that uniquely provides access to both suprathreshold spiking and subthreshold synaptic events of single neurons in the brain. This article describes how to set up and use the autopatcher, which is a robot for automatically obtaining high-yield and high-quality whole-cell patch clamp recordings in vivo. By following this protocol, a functional experimental rig for automated whole-cell patch clamping can be set up in 1 week. High-quality surgical preparation of mice takes ~1 h, and each autopatching experiment can be carried out over periods lasting several hours. Autopatching should enable in vivo intracellular investigations to be accessible by a substantial number of neuroscience laboratories, and it enables labs that are already doing in vivo patch clamping to scale up their efforts by reducing training time for new lab members and increasing experimental durations by handling mentally intensive tasks automatically.

At a glance

Figures

  1. The autopatcher[mdash]a robot for automated whole-cell patch clamp recordings in vivo: overview of the algorithm and schematic.
    Figure 1: The autopatcher—a robot for automated whole-cell patch clamp recordings in vivo: overview of the algorithm and schematic.

    (a) The algorithm for autopatching (adapted from Kodandaramaiah et al., 2012): the six stages of obtaining whole-cell patch clamp recordings in vivo include the following: (i) all the manual steps (Steps 12–31: installing a pipette, software initialization and so on) that need to be performed after which the autopatcher will programmatically perform the remaining steps (ii) an initial assessment of the resistance of the patch pipette to eliminate unsuitable pipettes (Box 2, autopatching step 1); (iii) lowering of the patch pipette to the region of interest followed by a second assessment of pipette tip fidelity (Box 2, autopatching step 2); (iv) the neuron-hunting stage, during which the autopatcher scans for neurons (Box 2, autopatching step 3); (v) attempting gigasealing by modulating the pressure inside the pipette and pipette voltage after contact with a cell has been established (Box 2, autopatching steps 4 and 5); and (vi) the break-in stage, during which pulses of high negative pressure are applied to achieve the whole-cell patch clamp state (Step 36). Some fraction of the autopatcher trials results in end points other than acquisition of whole-cell patched or cell-attached recordings. Such instances are highlighted with the red arrows and explained in the corresponding steps in the protocol. (b) Schematic of the autopatcher system capable of performing the autopatching algorithm (adapted from Kodandaramaiah et al., 2012): the system consists of a conventional in vivo patch setup (i.e., pipette, headstage, three-axis linear actuator, patch amplifier and computer), equipped with a few additional modules: a programmable linear motor and a custom control box for data acquisition to enable closed-loop control of the motor based upon a series of pipette resistance measurements. The control box also performs closed-loop pneumatic pressure control of the patch pipette. Adapted with permission from ref. 24.

  2. The autopatcher: equipment photographs.
    Figure 2: The autopatcher: equipment photographs.

    (a) Photograph of the autopatcher showing the general layout of major equipment. (b) Photograph focusing on the autopatcher control box and its interface with the patch amplifier and external digitizer. (c) Photograph focusing on the pipette actuator assembly. (d) Schematic of the autopatcher control box. A central digitizer board equipped with analog inputs, as well as analog and digital outputs, in the autopatcher control box sends command voltage signals to the patch amplifier and reads the patch measurements from the amplifier output. Digital outputs on the same board are sent to a bank of pneumatic valves (described in Margrie et al.7 and in the assembly manual 'autopatcher control box assembly manual.pdf' in Supplementary Data 4) to switch between different pressure states during autopatcher operation. The four pressures are generated by downregulating a compressed air source of ~2,580 mBar using manual and electronic pressure regulators whose outputs can be controlled using knobs on the front panel of the control box (potentiometers in the lower left corner). Vacuum pressures are generated using Venturi tube vacuum generators also installed inside the autopatcher control box.

  3. Optimum pipettes used for autopatching.
    Figure 3: Optimum pipettes used for autopatching.

    (a,b) Photomicrographs of an ideal patch pipette pulled using a Flaming-Brown pipette puller focusing on the pipette tip with a 0.9-μm tip diameter (6.2 M resistance) visualized with a 40× magnification objective (left) and a 100× water-immersion objective (right) (a) in comparison with a patch pipette with 1.5-μm tip diameter (3.3 M resistance) visualized with a 40× magnification objective (left) and 100× water-immersion objective (right) (b). (c) Comparison of a convex tapered pipette, which is ideal for autopatching (left), versus concave tapered pipettes (right). (d) Illustration of an ideal patch pipette exhibiting broad cone angle. Larger tip angles, as measured in the image at the very tip of the pipette, are ideal for rapid gigasealing, stable recordings and easier break-in attempts.

  4. Surgical procedure for headplate implantation.
    Figure 4: Surgical procedure for headplate implantation.

    (a) top (left) and side (right) views of the Delrin headplate to be affixed to the skull for head stabilization during autopatching. Scale bar, 5 mm. (bf) Preparation and surgery of the anesthetized mouse after administering approved anesthetic, followed by shaving and sterilizing the scalp. First, perform longitudinal incision of the scalp (b,c) to expose the skull (d). Clear the skull further around the desired recording region (d; red circle). Use a burr drill bit to drill three anchor holes (~500 μm in diameter, three small red circles; e) and implant skull screws. (f) Finally, implant the headplate by applying freshly mixed dental acrylic cement around the skull screws and around the periphery of the headplate window. (g) Photographs of the mouse skull after implantation of the skull screws (top) and after implantation of the headplate (bottom). Scale bars, ~2 mm. (hk) Illustration of the craniotomy procedure. Identify the desired craniotomy location (h), thin down a 1- to 2-mm-wide pit at the desired craniotomy location until the remaining bone is ~100 μm thick (i), carefully dislodge the flaky bone tissue with a tip of a needle (j) and lift off the bone tissue and clear any remaining bone fragments (k). All animal use should comply with institutional and governmental regulations. (e.g., Massachusetts Institute of Technology (MIT) Department of Comparative Medicine Committee on Animal Care).

  5. Autopatcher software GUI.
    Figure 5: Autopatcher software GUI.

    Red dotted lines outline different panels of the GUI. (Panel 1) Pipette status indicators that display the instantaneous pipette position (in μm) from the surface of the brain, pressure applied to the pipette (in kPa) and the holding voltage (in mV) applied to the pipette during autopatcher operation. (Panel 2) Text indicators that display the current status of the autopatcher trial, and a log of all autopatching trials attempted during an experiment, as well as an entry box where the experimenter can log comments. (Panel 3) Includes control elements that allow the experimenter to control the programmable motor. This displays the absolute position of the pipette in motor coordinates. (Panel 4) The interactive elements displayed in this panel depend on the stage of autopatching, and they include methods for setting the beginning and ending depth ranges within which the autopatcher will scan for neurons, among other things (see Figs. 6,7 and Supplementary Video 1 to see how this box changes throughout the protocol).

  6. Autopatcher software graphical user interface: neuron hunting.
    Figure 6: Autopatcher software graphical user interface: neuron hunting.

    (a) The autopatcher software GUI displayed during the neuron-hunting stage (Box 2, autopatching step 3) of autopatching: red dotted lines highlight the indicators and user controls in panel 4 of Figure 5 that can be accessed during this stage. The measured resistances after each step taken during neuron hunting are plotted on the 'NEURON HUNTING RESISTANCE MONITOR' graph. The 'SKIP TO GIGASEALING' button allows the experimenter to override the neuron detection algorithm of the autopatcher and proceed to the gigasealing stage. The resistance threshold that the algorithm uses to switch from neuron hunting to gigasealing mode can be changed as needed in the 'Neuron detection threshold (M-Ohms)' numerical entry box (default, 0.25 M (ref. 24)). (b) Representative screenshot of the 'NEURON HUNTING RESISTANCE MONITOR' graph displaying the resistance measurements logged during a successful trial resulting in a whole-cell patch recording. The autopatcher initially lowered the pipette to a depth of 650 μm and scanned to a depth of 720 μm before encountering a neuron, as indicated by the monotonic increase in pipette resistance in the last data points.

  7. Autopatcher software graphical user interface: gigasealing and break-in.
    Figure 7: Autopatcher software graphical user interface: gigasealing and break-in.

    (a) Panel 4 in Figure 5, now showing the autopatcher software GUI displayed during the gigasealing and break-in stages of autopatching (Box 2, autopatching steps 4 and 5; Steps 34 and 35 of the main PROCEDURE). The 'GIGASEALING RESISTANCE MONITOR' graph displays the recorded seal resistances during a gigasealing attempt. The 'RETURN TO NEURON HUNT' button allows the experimenter to over-ride the autopatcher operation and return to the neuron-hunting stage of autopatching. The 'MANUAL APPLICATION OF SUCTION' button allows the experimenter to over-ride the autopatcher algorithm's negative pressure application to manually apply additional negative pressure at any time during gigasealing, and it allows exploration of alternate strategies for gigasealing used in in vivo and in vitro slice patching25, 34. Once a successful gigaseal is formed, break-in can be attempted by using the 'ATTEMPT BREAK IN' button, which causes the autopatcher to apply pulses of negative pressure of selected time duration. Alternatively, break-in can be attempted by applying voltage pulses using the 'ZAP!' button. The 'WHOLE CELL CURRENTS MONITOR' graph displays the currents in response to injected voltage square waves after a break-in attempt so that the user can assess whether it has achieved the whole-cell configuration. (b) Screen capture of seal resistance measurements displayed in the 'GIGASEALING RESISTANCES MONITOR' graph in the autopatcher software GUI during a successful gigasealing attempt. A G seal was obtained at time point 'i'. The 'ATTEMPT BREAK IN' button was used by the experimenter at t = 105 s to break into the cell. Whole-cell configuration was obtained at time point 'ii'. (c) Illustration of currents measured and displayed in the 'WHOLE CELL CURRENTS MONITOR' graph in the autopatcher software GUI after a successful break-in attempt resulting in a whole-cell patched neuron.

  8. Autopatcher software graphical user interface: recording.
    Figure 8: Autopatcher software graphical user interface: recording.

    (a) Panel 4 in Figure 5, now showing the autopatcher software GUI displayed during the recording stage after autopatching (Steps 37–39). A toggle switch allows the user to use either the built-in data acquisition feature under 'Autopatcher Control' or to use an external data acquisition software under 'External Control'. If the 'Autopatcher Control' option is used, a second toggle switch allows the user to switch between voltage clamp and current clamp mode. A numerical entry box allows the user to set the holding voltage (in mV, if recording in voltage clamp mode) or holding current (in pA, if recording in current clamp mode). When the 'RECORD' button is pressed, the measured currents (if recording in voltage clamp mode) or measured voltage (if recording in current clamp mode) are displayed in the graph indicator. The 'REPEAT' button can be used to acquire data continuously, and the 'SAVE TO FILE' button saves the file to disk. Pressing the 'START OVER' button ends the recording, and the autopatcher will retract the patch pipette back to the surface in one quick step (lasting ~500 ms for pipette retrieval and replacement) to start a new trial. To recover the morphology of the recorded cell, biocytin filling is attempted by pressing the 'SLOWLY RETRACT PIPETTE' button, which will result in the autopatcher withdrawing the patch pipette back to the surface of the brain at a rate of 3 μm/s.

  9. Example data acquired by the autopatcher.
    Figure 9: Example data acquired by the autopatcher.

    (a) Current clamp recording from autopatched cortical neuron during current injection (2-s-long pulses of −30, 0, +30, +60, +90, +120, +150 and +180 pA current injection). Access resistance, 44 M; input resistance, 66 M; depth of cell 442 μm below the surface of the brain. (bd) Voltage clamp recordings from an autopatched cortical neuron clamped at −80 mV showing spontaneous excitatory postsynaptic potentials (EPSCs) (b), zooming in on a single synaptic event indicated by black arrow in b (c), and zooming in on a synaptic barrage event indicated by a black bar in b (d). Access resistance, 23 M; input resistance, 124 M; depth of cell, 544 μm from the surface of the brain. (e) Biocytin fill of the autopatched cortical pyramidal neuron recorded in (a). (f) Autopatching in awake head-fixed mice. Current clamp recording from a layer-4 cortical neuron in barrel cortex of an awake head-fixed mouse showing persistent depolarization of membrane potential during active whisking (periods of whisker movements indicated by black bar). Access resistance, 37 M; input resistance, 88 M; depth of cell, 468 μm below the surface of the brain. (g,h) Simultaneous whole-cell recording and optogenetic stimulation in vivo. (g) Jaws-expressing neuron in the cortex showing hyperpolarization at onset of red light delivery. Reproduced with permission from ref. 26. (h) Channelrhodopsin-2–expressing cortical neuron in a Thy1-ChR2 mouse, showing evoked spiking in response to 20-ms blue light pulses. All animal use complied with institutionaland governmental regulations (MIT Department of Comparative Medicine Committee on Animal Care). (g) Adapted from Chuong et al.26.

Videos

  1. Computer screen broadcast of the autopatcher software GUI during a representative autopatching trial.
    Video 1: Computer screen broadcast of the autopatcher software GUI during a representative autopatching trial.

References

  1. Bruno, R.M. & Sakmann, B. Cortex is driven by weak but synchronously active thalamocortical synapses. Science 312, 16221627 (2006).
  2. Arenz, A., Silver, R.A., Schaefer, A.T. & Margrie, T.W. The contribution of single synapses to sensory representation in vivo. Science 321, 977980 (2008).
  3. Brecht, M., Schneider, M., Sakmann, B. & Margrie, T.W. Whisker movements evoked by stimulation of single pyramidal cells in rat motor cortex. Nature 427, 704710 (2004).
  4. Chadderton, P., Margrie, T.W. & Hausser, M. Integration of quanta in cerebellar granule cells during sensory processing. Nature 428, 856860 (2004).
  5. Eberwine, J. et al. Analysis of gene expression in single live neurons. Proc. Natl. Acad. Sci. USA 89, 30103014 (1992).
  6. Rancz, E.A. et al. Transfection via whole-cell recording in vivo: bridging single-cell physiology, genetics and connectomics. Nat. Neurosci. 14, 527532 (2011).
  7. Margrie, T.W., Brecht, M. & Sakmann, B. In vivo, low-resistance, whole-cell recordings from neurons in the anaesthetized and awake mammalian brain. Pflugers Arch. 444, 491498 (2002).
  8. Chadderton, P., Agapiou, J.P., McAlpine, D. & Margrie, T.W. The synaptic representation of sound source location in auditory cortex. J. Neurosci. 29, 1412714135 (2009).
  9. Chadderton, P., Margrie, T.W., Hausser, M Integration of quanta in cerebellar granule cells during sensory processing. Nature 428, 856860 (2004).
  10. Crochet, S. & Petersen, C.C.H. Correlating whisker behavior with membrane potential in barrel cortex of awake mice. Nat. Neurosci. 9, 608610 (2006).
  11. Crochet, S., Poulet, J.F., Kremer, Y. & Petersen, C.C. Synaptic mechanisms underlying sparse coding of active touch. Neuron 69, 11601175 (2011).
  12. Gentet, L.J., Avermann, M., Matyas, F., Staiger, J.F. & Petersen, C.C. Membrane potential dynamics of GABAergic neurons in the barrel cortex of behaving mice. Neuron 65, 422435 (2010).
  13. Gentet, L.J. et al. Unique functional properties of somatostatin-expressing GABAergic neurons in mouse barrel cortex. Nat. Neurosci. 15, 607612 (2012).
  14. Harvey, C.D., Collman, F., Dombeck, D.A. & Tank, D.W. Intracellular dynamics of hippocampal place cells during virtual navigation. Nature 461, 941946 (2009).
  15. Hromadka, T., DeWeese, M.R. & Zador, A.M. Sparse representation of sounds in the unanesthetized auditory cortex. PLoS Biol. 6, 124137 (2008).
  16. Schaefer, A.T. & Margrie, T.W. Spatiotemporal representations in the olfactory system. Trends Neurosci. 30, 92100 (2007).
  17. Kitamura, K., Judkewitz, B., Kano, M., Denk, W. & Hausser, M. Targeted patch-clamp recordings and single-cell electroporation of unlabeled neurons in vivo. Nat. Methods 5, 6167 (2008).
  18. Komai, S., Denk, W., Osten, P., Brecht, M. & Margrie, T.W. Two-photon targeted patching (TPTP) in vivo. Nat. Protoc. 1, 647652 (2006).
  19. Margrie, T.W. et al. Targeted whole-cell recordings in the mammalian brain in vivo. Neuron 39, 911918 (2003).
  20. Lee, A.K., Epsztein, J. & Brecht, M. Head-anchored whole-cell recordings in freely moving rats. Nat. Protoc. 4, 385392 (2009).
  21. Lee, A.K., Manns, I.D., Sakmann, B. & Brecht, M. Whole-cell recordings in freely moving rats. Neuron 51, 399407 (2006).
  22. Lee, D., Lin, B.-J. & Lee, A.K. Hippocampal place fields emerge upon single-cell manipulation of excitability during behavior. Science 337, 849853 (2012).
  23. Long, M.A., Jin, D.Z. & Fee, M.S. Support for a synaptic chain model of neuronal sequence generation. Nature 468, 394399 (2010).
  24. Kodandaramaiah, 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 9, 585587 (2012).
  25. DeWeese, M.R. Whole-cell recording in vivo. Curr. Protoc. Neurosci. 38, 6.22.16.22.15 (2007).
  26. Chuong, A.S. et al. Noninvasive optical inhibition with a red-shifted microbial rhodopsin. Nat. Neurosci. 17, 11231129 (2014).
  27. Schramm, A.E., Marinazzo, D., Gener, T. & Graham, L.J. The touch and zap method for in vivo whole-cell patch recording of intrinsic and visual responses of cortical neurons and glial cells. PLoS ONE 9, e97310 (2014).
  28. Pak, N. et al. Closed-loop, ultraprecise, automated craniotomies. J. Neurophysiol. 113, 39433953 (2015).
  29. Kodandaramaiah, S.B., Boyden, E.S. & Forest, C.R. In vivo robotics: the automation of neuroscience and other intact-system biological fields. Ann. NY Acad. Sci. 1305, 6371 (2013).
  30. Arenkiel, B.R. et al. In vivo light-induced activation of neural circuitry in transgenic mice expressing channelrhodopsin-2. Neuron 54, 205218 (2007).
  31. Harrison, R.R. et al. Microchip amplifier for in vitro, in vivo, and automated whole-cell patch-clamp recording. J. Neurophysiol. doi:10.1152/jn.00629.2014 (2014).
  32. Poulet, J.F.A., Fernandez, L.M.J., Crochet, S. & Petersen, C.C.H. Thalamic control of cortical states. Nat. Neurosci. 15, 370372 (2012).
  33. Polack, P.O., Friedman, J. & Golshani, P. Cellular mechanisms of brain state-dependent gain modulation in visual cortex. Nat. Neurosci. 16, 13311339 (2013).
  34. Plant, T.D., Eilers, J. & Konnerth, A. in Patch-Clamp Applications and Protocols. Vol. 26 (eds. Boulton, A., Baker, G. & Walz, W.) 233258 (Humana Press, 1995).

Download references

Author information

Affiliations

  1. Media Lab, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA.

    • Suhasa B Kodandaramaiah,
    • Ian R Wickersham,
    • Annabelle C Singer,
    • Giovanni Talei Franzesi &
    • Edward S Boyden
  2. McGovern Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA.

    • Suhasa B Kodandaramaiah,
    • Ian R Wickersham,
    • Annabelle C Singer &
    • Edward S Boyden
  3. Departments of Biological Engineering and Brain and Cognitive Sciences, MIT, Cambridge, Massachusetts, USA.

    • Suhasa B Kodandaramaiah,
    • Ian R Wickersham,
    • Annabelle C Singer &
    • Edward S Boyden
  4. George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA.

    • Gregory L Holst &
    • Craig R Forest
  5. Department of Physiology, School of Medicine, Emory University, Atlanta, Georgia, USA.

    • Michael L McKinnon

Contributions

S.B.K., I.R.W., G.L.H., C.R.F. and E.S.B. designed, built and tested the autopatcher system. A.C.S. and G.T.F. assisted with experiments. M.L.M. developed the software included with the manuscript. S.B.K., I.R.W., G.L.H., A.C.S., G.T.F., C.R.F. and E.S.B. wrote the manuscript.

Competing financial interests

A.C.S., G.T.F., M.L.M., C.R.F. and E.S.B. declare no competing interests. I.R.W., S.B.K. and G.L.H. received financial remuneration from Neuromatic Devices for technical consulting services provided in 2012, 2013 and 2012–2015, respectively.

Corresponding authors

Correspondence to:

Author details

Supplementary information

Video

  1. Video 1: Computer screen broadcast of the autopatcher software GUI during a representative autopatching trial. (19.37 MB, Download)

PDF files

  1. Supplementary Text and Figures (180 KB)

    Supplementary Methods

Zip files

  1. Supplementary Data 1: File archive consisting of software required for running the Autopatcher. (2,115 KB)

    Includes two Labview library files – ‘Autopatcher 2000.llb’ and ‘Hardware.llb’ that can be opened using Labview installed in Step 8 of the protocol. Also included is a corresponding ‘Autopatcher software configuration manual.pdf’ that provides detailed instructions on installation of software and configuring the software settings to control the autopatcher control box.

  2. Supplementary Data 2: File archive consisting of mechanical drawings and computer aided design (CAD) files for making the custom head fixation base and headplate. (365 KB)

    ‘Headfixation fixation base CAD.pdf’ is a mechanical drawing of the headfixation base, while ‘Head fixation base CAD.SLDPRT’ is the 3D drawing that can be opened in Solidworks software. ‘Head Plate CAD.pdf’ is a mechanical drawing of headplate implant, and ‘Head Plate CAD.SLDPRT’ is the 3D drawing that can be opened in Solidworks software.

  3. Supplementary Data 3: File archive consisting of mechanical drawings and computer aided design (CAD) files and instructions for assembling the autopatcher pipette actuator assembly. (3,022 KB)

    Assembly instructions are provided in ‘Autopatcher Robotic Arm Assembly Manual.pdf’. ‘adapter plate 1.PDF’ and ‘adapter plate 1.SLDDRW’ are mechanical drawings of the adaptor plate used for mounting programmable linear stage onto Sutter manipulator. ‘adapter plate 1.SLDPRT’ is the corresponding 3D CAD file that can be opened in Solidworks. ‘adapter plate 2.PDF’ and ‘adapter plate 2.SLDDRW’ are mechanical drawings of the adaptor plate used to mount the amplifier headstage onto the programmable linear stage. ‘adapter plate 2.SLDPRT’ is the corresponding 3D CAD file that can be opened in Solidworks.

  4. Supplementary Data 4: File archive consisting of mechanical drawings, computer aided design (CAD) files and instructions for assembling the autopatcher control box. (10,129 KB)

    Assembly instructions are provided in the ‘Autopatcher control box assembly manual.pdf’ while ‘Autopatcher control box parts list.xlsx’ provides complete list of parts required for assembling the control box. Details of each part include description, name of vendor, catalog number, price/unit (as on Aug 2015), and quantity of each part. The sub-folder ‘Laser cutter files’ contains the ‘Autopatcher panels front & back.ai’ and ‘Autopatcher structural base, platform, & manometer clamp.ai’ files which can be used to cut two structural elements used for control box assembly (See the ‘Autopatcher control box assembly manual.pdf’). The sub-folder ‘Circuit board files’ contains: ‘Autopatcher PCB parts list.xlsx’ – a full parts list of all components on the pressure control printed circuit board (PCB) and valve relay PCB. ‘pressure_board.brd’ and ‘pressure_board.sch’ are the pressure control PCB CAD files, while ‘valve-relay_board.brd’ and ‘valve-relay_board.sch’ are the valve relay PCB CAD files.

Additional data