Automated whole-cell patch-clamp electrophysiology of neurons in vivo

Abstract

Whole-cell patch-clamp electrophysiology of neurons is a gold-standard technique for high-fidelity analysis of the biophysical mechanisms of neural computation and pathology, but it requires great skill to perform. We have developed a robot that automatically performs patch clamping in vivo, algorithmically detecting cells by analyzing the temporal sequence of electrode impedance changes. We demonstrate good yield, throughput and quality of automated intracellular recording in mouse cortex and hippocampus.

Figure 1. The autopatcher: a robot for in vivo patch clamping

(a) The four stages of the automated in vivo patch algorithm (detailed in Supplementary Fig. 3).(b) Schematic of a simple robotic system capable of performing the autopatching algorithm, consisting of a conventional in vivo patch setup, equipped with a programmable linear motor (note that if the vertical axis of the 3 axis linear actuator is computer-controlled, this can be omitted), a controllable bank of pneumatic valves for pressure control, and a secondary computer interface board (if the patch amplifier provides direct access to these measurements, this can be omitted). (c) Current clamp traces during current injection (top; 2 s-long pulses of −60, 0, and +80 pA current injection), and at rest (bottom; note compressed timescale relative to the top trace), for an autopatched cortical neuron. Access resistance, 44 MΩ; input resistance, 41 MΩ; depth of cell 832 µm below brain surface. (d) Current clamp traces during current injection (top; 2 s-long pulses of −60, 0, and +40 pA current injection), and at rest (bottom), for an autopatched hippocampal neuron. Access resistance, 55 MΩ; input resistance, 51 MΩ; depth of cell, 1,320 µm. (e) Biocytin fill of a representative autopatched cortical pyramidal neuron. Scale bar, 50 µm.

Figure 2. Autopatcher operation and performance

(a) Representative timecourse of pipette resistance during autopatcher operation, top, with zoomed-in view of the neuron hunting phase, bottom. Roman numerals: i, the first of the series of resistance measurements that indicate neuron detection; ii, the last of the series; iii, when positive pressure is released; iv, when suction is applied; v, when holding potential starts to ramp from −30 mV to −65 mV; vi, when it hits −65 mV; vii, break-in. (b) Raw traces showing patch pipette currents, while a square voltage wave (10 Hz, 10 mV) is applied, at the events flagged by Roman numerals in Fig. 2a. (c–f) Quality of recordings obtained with the autopatcher vs. by manual whole cell patch clamping. (c) left, Plot of access resistances obtained versus pipette depth and right, bar graph summary of access resistances (mean ± s.d.), for the final autopatcher whole cell patch validation test set (closed symbols; n = 23), the test set in which the autopatcher concludes in the gigaseal state (open symbols, n = 24; data acquired after manual break-in), and the test set acquired via manual whole cell patch clamp (grayed symbols; n = 15), for cortical (circles) and hippocampal (triangles) neurons. (d) left, Resting potential versus pipette depth, and right, summary data, plotted as in c. (e) left, Holding current versus pipette depth, and right, summary, plotted as in c. (f) left, Holding times versus pipette depth, and right, summary, plotted as in c (including recordings that were deliberately terminated, as well as recordings terminated spontaneously).