View source for Patch clamp

{{Short description|Technique for recording the current through single ion channels with a glass pipette sealed to a membrane}}
{{About|the electrophysiology technique|the amplifier on the Molecular Reality live stream|EDR4}}
{{Infobox technique
| name = Patch clamp
| classification = Electrophysiology
| analytes = Ion channels, in patches of membrane or in whole cells
| inventor = [[Erwin Neher and Bert Sakmann|Erwin Neher, Bert Sakmann]]
| year = 1976 (single-channel currents); 1981 (giga-seal methods)
| related = Voltage clamp; planar lipid bilayer recording; [[nanopore]] recording
}}
The '''patch clamp''' is a technique for recording the electric current through [[ion channel]]s, the protein pores in cell membranes. A glass micropipette is pressed against a cell, gentle suction seals the glass to the membrane, and an amplifier holds the voltage across the sealed patch at a chosen value while it measures the current that flows. When the seal has a resistance of a gigaohm or more, the recording is quiet enough to show the current through a single channel, about a picoampere, as the channel flips between open and shut.<ref name="hamill">{{cite journal |last1=Hamill |first1=O. P. |last2=Marty |first2=A. |last3=Neher |first3=E. |last4=Sakmann |first4=B. |last5=Sigworth |first5=F. J. |title=Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches |journal=Pflügers Archiv |volume=391 |issue=2 |pages=85-100 |year=1981 |doi=10.1007/BF00656997 |pmid=6270629}}</ref><ref name="nobel">{{cite web |url=https://www.nobelprize.org/prizes/medicine/1991/press-release/ |title=The Nobel Prize in Physiology or Medicine 1991: Press release |publisher=Nobel Assembly at the Karolinska Institute |access-date=2026-10-25}}</ref>

[[Erwin Neher and Bert Sakmann]] reported the first such recordings in 1976, from frog muscle fibers, and shared the 1991 Nobel Prize in Physiology or Medicine "for their discoveries concerning the function of single ion channels in cells".<ref name="neher">{{cite journal |last1=Neher |first1=E. |last2=Sakmann |first2=B. |title=Single-channel currents recorded from membrane of denervated frog muscle fibres |journal=Nature |volume=260 |issue=5554 |pages=799-802 |year=1976 |doi=10.1038/260799a0 |pmid=1083489}}</ref><ref name="nobel" /> The amplifiers built for the method measure picoampere currents at bandwidths of kilohertz, and the same hardware records the current through a nanopore. The experiment of 1996 that put single strands of DNA into a protein pore, and so began [[nanopore sequencing]], was recorded on an amplifier sold for physiology.<ref name="kas1996">{{cite journal |last1=Kasianowicz |first1=J. J. |last2=Brandin |first2=E. |last3=Branton |first3=D. |last4=Deamer |first4=D. W. |title=Characterization of individual polynucleotide molecules using a membrane channel |journal=Proceedings of the National Academy of Sciences |volume=93 |issue=24 |pages=13770-13773 |year=1996 |doi=10.1073/pnas.93.24.13770 |pmid=8943010}}</ref> [[Molecular Reality Corporation]] streams the current of its laboratory bench from a four-channel patch-clamp amplifier, the [[EDR4]].

== Background ==
Until the 1970s ion channels were an inference. Alan Hodgkin and Andrew Huxley held the membrane of the squid giant axon at set voltages with a feedback amplifier, a voltage clamp, and in 1952 described the sodium and potassium currents it passed in a set of equations, the [[Hodgkin-Huxley model]].<ref name="hh">{{cite journal |last1=Hodgkin |first1=A. L. |last2=Huxley |first2=A. F. |title=A quantitative description of membrane current and its application to conduction and excitation in nerve |journal=The Journal of Physiology |volume=117 |issue=4 |pages=500-544 |year=1952 |doi=10.1113/jphysiol.1952.sp004764 |pmid=12991237}}</ref> The equations described conductances and said nothing about what carried them. Estimates of the current through one channel came from indirect experiments. In 1972 Katz and Miledi analyzed the noise of the response to acetylcholine at the frog neuromuscular junction and inferred an elementary current pulse lasting about 1 ms at 20 °C and carrying about 50,000 ions.<ref name="katz">{{cite journal |last1=Katz |first1=B. |last2=Miledi |first2=R. |title=The statistical nature of the acetylcholine potential and its molecular components |journal=The Journal of Physiology |volume=224 |issue=3 |pages=665-699 |year=1972 |doi=10.1113/jphysiol.1972.sp009918 |pmid=5071933}}</ref> Antibiotics and proteins inserted into artificial lipid membranes gave single-channel steps of a few picoamperes. The background noise of a living cell, however, was thought to be at least ten times larger than that.<ref name="cat">{{cite journal |last1=Catacuzzeno |first1=L. |last2=Franciolini |first2=F. |title=Celebrating 50 years of single-channel recording with the patch clamp |journal=The Journal of Membrane Biology |volume=258 |issue=6 |pages=429-445 |year=2025 |doi=10.1007/s00232-025-00362-3 |pmid=40970955}}</ref>

Neher and Sakmann, at the Max Planck Institute for Biophysical Chemistry in Göttingen, isolated a small patch of membrane with a pipette. Their pipettes had an opening of 3 to 5 µm and a resistance of 2 to 5 MΩ, and they pressed them against muscle fibers that had been cleaned with enzymes. The seal between glass and membrane reached only 50 to 100 MΩ, yet the patch current stepped between two levels as acetylcholine receptor channels opened and closed.<ref name="cat" /> Neher also spent parts of 1975 and 1976 in Charles Stevens's laboratory at Yale, where by the same history he obtained similar results.<ref name="cat" />

== The giga-seal ==
A seal is a resistor, and it generates thermal current noise of <math>\sqrt{4k_BTB/R}</math> over a bandwidth ''B''. A seal of 100 MΩ gives about 0.40 pA rms in 1 kHz, a large fraction of the current being sought.<ref name="cat" />{{efn|1=The value follows from the formula with 4''k''<sub>B</sub>''T'' = 1.6 × 10<sup>−20</sup> J at room temperature, ''B'' = 1,000 Hz and ''R'' = 10<sup>8</sup> Ω. A seal of 10 GΩ gives one tenth of that, 0.04 pA.}} The group tried different pipette glass, cleaner surfaces and enzyme treatments, with little effect. By one account they were close to abandoning the attempt when they noticed that gentle suction on the pipette produced seals of 10 to 100 GΩ. The new seals were called giga-seals, to tell them from the earlier megaohm seals.<ref name="cat" />

The 1981 paper by Hamill, Marty, Neher, Sakmann and Sigworth set down the method. It described how to make the pipettes, how to obtain giga-seals with resistances of 10<sup>9</sup> to 10<sup>11</sup> ohms, the recording circuit, and the ways of isolating patches from cells. Detached patches allowed single-channel recording with defined solutions on both sides of the membrane.<ref name="hamill" />

== Configurations ==
[[File:Patch_clamp_configurations.svg|thumb|right|upright=1.3|alt=Four drawings of a glass pipette on a cell showing cell-attached, whole-cell, inside-out and outside-out recording|The four classic configurations. In the whole-cell mode the pipette solution is continuous with the cytoplasm.]]
{| class="wikitable"
|+ Recording configurations
! Configuration !! How it is formed !! What is recorded
|-
| Cell-attached || Seal on an intact cell || Channels in the patch, with the cell interior undisturbed
|-
| Inside-out || A cell-attached patch is pulled away from the cell || Channels in the patch, with the inner face of the membrane in the bath
|-
| Whole-cell || The patch is ruptured by suction or a voltage pulse || The summed current of every channel in the cell, whose interior is dialyzed by the pipette solution
|-
| Outside-out || The pipette is withdrawn from a whole-cell recording, and the membrane reseals || Channels in the patch, with the outer face of the membrane in the bath
|-
| Perforated patch || The patch is made permeable by an antibiotic such as nystatin in the pipette || Whole-cell current without washing out the contents of the cell
|}
The 1981 paper described the cell-attached, inside-out, outside-out and whole-cell recordings, and noted that single-channel resolution is possible in the whole-cell mode when the cell is smaller than 20 µm across.<ref name="hamill" /> The perforated patch came in 1988, when Horn and Marty used nystatin in the pipette to give electrical access to a cell without the loss of the cytoplasmic compounds that its response to acetylcholine needed.<ref name="horn">{{cite journal |last1=Horn |first1=R. |last2=Marty |first2=A. |title=Muscarinic activation of ionic currents measured by a new whole-cell recording method |journal=The Journal of General Physiology |volume=92 |issue=2 |pages=145-159 |year=1988 |doi=10.1085/jgp.92.2.145 |pmid=2459299}}</ref>

== The amplifier ==
A patch-clamp amplifier is a [[transimpedance amplifier]] in a small headstage next to the pipette, followed by gain, filtering and a digitizer. The element in its feedback loop sets the noise. A resistor adds thermal noise, which falls as the resistance rises, and the very large resistors that picoampere work calls for are imperfect parts. A capacitor adds almost none.<ref name="axo">{{cite web |url=https://www.moleculardevices.com/sites/default/files/en/asset/dd/data-sheets/axon-axopatch-200b-datasheet-rev-b.pdf |title=Axon Axopatch 200B Microelectrode Amplifier (data sheet) |publisher=Molecular Devices |access-date=2026-10-25}}</ref>

The [[Axopatch 200B]], made by Axon Instruments and now sold by [[Molecular Devices]], is built on that argument. In patch mode its feedback element is a 1 pF capacitor, so the headstage measures current as the rate at which voltage rises across it. The voltage cannot rise forever, so the circuit resets when it reaches the supply limit, in 50 µs, and cooling the critical components to about −15 °C lowers their thermal noise. Whole-cell mode adds a resistor of 500 MΩ, or 50 MΩ for large currents, in parallel with the capacitor, because the cell and its surroundings make more noise than the amplifier does.<ref name="axo" /> The data sheet gives the maximum noise of the headstage in patch mode, with the input open and an 8-pole Bessel filter:

{| class="wikitable"
|+ Axopatch 200B noise, patch mode, headstage without pipette holder
! Bandwidth !! Noise, rms
|-
| 0.1 to 1 kHz || 0.015 pA
|-
| 0.1 to 5 kHz || 0.060 pA
|-
| 0.1 to 10 kHz || 0.130 pA
|}
With a pipette holder attached the figure at 10 kHz is 0.145 pA.<ref name="axo" /> A 1 GΩ resistor alone would add 0.41 pA in 10 kHz, three times as much, and a 1 MΩ resistor 12.9 pA.{{efn|1=Calculated with the formula in the section on the giga-seal at 300 K.}} The measured noise rises almost in proportion to the bandwidth, ten times the bandwidth giving 8.7 times the noise. That is the behavior of dielectric and capacitive noise, and thermal noise alone would rise only as the square root of the bandwidth.<ref name="fragasso">{{cite journal |last1=Fragasso |first1=A. |last2=Schmid |first2=S. |last3=Dekker |first3=C. |title=Comparing current noise in biological and solid-state nanopores |journal=ACS Nano |volume=14 |issue=2 |pages=1338-1349 |year=2020 |doi=10.1021/acsnano.9b09353 |pmid=32049492}}</ref> The sources are set out in [[Nanopore noise]].

The amplifier also corrects for the voltage drop across the pipette (series-resistance compensation), cancels the capacitance of the pipette and of the cell, with a whole-cell range of 100 pF or 1,000 pF, and filters with a 4-pole Bessel filter set between 1 and 100 kHz, the limit of its recording bandwidth. It can apply a pulse of +1.3 V to rupture a patch on the way to the whole-cell mode.<ref name="axo" />

== Uses ==
The Nobel Assembly's announcement of 1991 said that the new knowledge had helped explain the cellular mechanisms behind several diseases, among them diabetes and cystic fibrosis, and that drugs commonly used to stimulate insulin secretion act directly on ATP-controlled ion channels.<ref name="nobel" />

The method is used to measure what a drug does to an ion channel. The safety guideline ICH S7B (2005) recommends an in vitro IKr assay, which evaluates a compound's effect on the ionic current through a native or expressed channel such as the one encoded by hERG, because inhibition of that current is the most common way a drug delays ventricular repolarization.<ref name="s7b">{{cite web |url=https://ema.europa.eu/en/documents/scientific-guideline/ich-s-7-b-nonclinical-evaluation-potential-delayed-ventricular-repolarization-qt-interval_en.pdf |title=ICH S7B: Nonclinical evaluation of the potential for delayed ventricular repolarization (QT interval prolongation) by human pharmaceuticals |publisher=European Medicines Agency |date=2005 |access-date=2026-10-25}}</ref> Such currents are recorded by patch clamp.

Manual patch clamping takes a trained experimenter and a micromanipulator. In 2002 Fertig, Blick and Behrends reported whole-cell recording on a planar glass chip, which replaced the pipette with a hole in the chip; automated patch-clamp systems use chips of this kind.<ref name="fertig">{{cite journal |last1=Fertig |first1=N. |last2=Blick |first2=R. H. |last3=Behrends |first3=J. C. |title=Whole cell patch clamp recording performed on a planar glass chip |journal=Biophysical Journal |volume=82 |issue=6 |pages=3056-3062 |year=2002 |doi=10.1016/S0006-3495(02)75646-4 |pmid=12023228}}</ref> [[Nanion Technologies]] sells instruments of both the planar patch-clamp and the bilayer kind, and supplies the bilayer platform with the software that Elements developed for its amplifiers.<ref name="nanion">{{cite web |url=https://nanion.com/products/orbit-16-tc/software/ |title=Software |website=Orbit 16 TC |publisher=Nanion Technologies |access-date=2026-10-25}}</ref>

== Nanopore recording ==
A [[nanopore]] experiment is a patch-clamp experiment with a different membrane. Two chambers of salt solution are separated by a membrane with one pore, each chamber holds a silver chloride electrode ([[Ag/AgCl electrode]]), an amplifier holds the voltage and reads the current, a low-pass filter and a digitizer follow, and a metal box shields the headstage from interference.<ref name="fragasso" /> The channel of a patch recording is replaced by a protein channel in an artificial bilayer or by a hole in a chip, and the open-channel current becomes the [[open-pore current]]. Recording single channels in artificial bilayers is older than the patch clamp, and nanopore laboratories use the Montal-Mueller bilayer method.<ref name="cat" /><ref name="fragasso" />

In the paper of 1996 the current "was converted to voltage and amplified by either an Axopatch 200A (Axon Instruments, Foster City, CA) or a Dagan 3900A patch clamp amplifier", then filtered with a Bessel filter and sampled at typically 64,000 samples a second, and the bilayer apparatus sat in a μ-metal box to shield it from "electric and magnetic noise sources".<ref name="kas1996" /> The Axopatch 200B data sheet has a section for bilayers, which says the headstage is stable with an input capacitance of 1,000 pF and that its integrating design lets the bilayer voltage be stepped quickly.<ref name="axo" /> In a 2020 comparison of noise in biological and solid-state nanopores the example traces of both kinds were recorded with an Axopatch 200B, and the one solid-state recording at 500 kHz needed a dedicated high-bandwidth amplifier, the VC100 of Chimera Instruments.<ref name="fragasso" />

Where nanopore work differs is the load on the input. A membrane chip adds capacitance, over 300 pF for standard silicon chips and under 1 pF for the best glass ones, and that capacitance largely sets the noise above 10 kHz.<ref name="fragasso" />

== At Molecular Reality ==
[[File:Stuff-edr4.jpg|thumb|right|upright=0.8|alt=A silver amplifier box in a 3D-printed bracket above a white flow cell with red wing nuts|The EDR4 unit in its bracket above the flow cell on the company's bench.]]
The company's guide describes its amplifier in the terms of the field: "the instrument that applies the voltage and hears the current." It samples at 200 kHz, one reading every five microseconds.<ref name="faq">{{cite web |url=https://molecularreality.com/FAQ |title=Keltar's Field Guide to Molecular Reality |publisher=Molecular Reality Corporation |date=August 2026}}</ref> In a post of March 2026 the founder said that the stream page should take data from any patch-clamp amplifier or nanopore device, and from other kinds of molecular sensor.<ref name="marching">{{cite web |title=Marching orders |author=Kemmishtree |website=EQBio-OS |date=March 28, 2026}}</ref>

The first recordings in the stream's archive are of air, a standard first check: with no liquid there is no ionic path, and the amplifier reports its own floor of leakage and pickup.<ref name="p6">{{cite news |title=Nineteen Billion Ions a Second, Live on the Internet |work=Signal & Noise |author=Lenora Ashgrove |date=2026-07-29 |url=/press/signal-noise/2026/07/29/nineteen-billion}}</ref> In its data sheet the Axopatch states its noise with the headstage input open. On the stream the pore is a [[200-micron laser-drilled pore|laser-drilled hole 200 µm across]],<ref name="p6" /> more than 100,000 times wider than the narrowest part of a protein nanopore, which is about 1 nm.

== See also ==
* [[Ion channel]]
* [[Hodgkin-Huxley model]]
* [[Lipid bilayer]]
* [[Picoampere]]
* [[Bessel filter]]

== Notes ==
{{Notelist}}

== References ==
{{Reflist}}

== Further reading ==
* {{cite book |last1=Sakmann |first1=B. |last2=Neher |first2=E. |title=Single-Channel Recording |publisher=Plenum Press |year=1995}}

{{Nanopores}}
{{Portal bar|Nanopores|Molecular Reality}}

[[Category:Electrophysiology]]
[[Category:Neuroscience]]
[[Category:Nanopores]]