- Email: [email protected]
Giant excised patch recordings of recombinant ion channel currents expressed in mammalian cells
Neuroscience Letters 329 (2002) 17–20 www.elsevier.com/locate/neulet
Giant excised patch recordings of recombinant ion channel currents expressed in mammalian cells Jonathan J. Couey a, Devon P. Ryan a, John T. Glover a, John C. Dreixler a,b, Joseph B. Young a, Khaled M. Houamed a,* a
Department of Anesthesia and Critical Care, University of Chicago, 5841 S. Maryland Avenue, Box 4028, Chicago, IL 60637, USA b Department of Neurobiology, Pharmacology and Physiology, University of Chicago, Chicago, IL 60637, USA Received 2 April 2002; received in revised form 14 May 2002; accepted 15 May 2002
Abstract The giant excised patch variant of patch clamp recording combines microsecond time resolution of macroscopic currents with rapid exchange of the experimental solutions at the intracellular membrane surface. This technique has been applied to a limited number of cell types, including Xenopus oocytes, muscle cells, and photoreceptors. We have applied this technique to recording recombinant ion channel currents expressed in membrane patches excised from HEK293 cell lines. Giant inside-out patch recordings of Na 1 channels and SKCa type calcium-activated potassium channels show high temporal resolution and excellent signal to noise characteristics. This technique will facilitate the study of recombinant ion channels expressed in mammalian cells. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: SK channel; KCNN channel; Calcium-activated potassium channel; Sodium channel; Magnesium; Channel block; Rectification
The ‘giant patch’ [2–4] has three advantages over the conventional inside-out excised patch recording technique: (1), it allows macroscopic current measurements with high signal to noise ratios; (2), because of the low access resistance, microsecond time resolution can be achieved; and (3), experimental solutions at the intracellular patch surface can be exchanged rapidly. These advantages enhance the utility of the giant patch technique for studies of gating, modulation, permeation, and blocking of ion channels and transporters [1,4,7,8]. However, this technique has been limited to certain native channels and recombinant channels expressed in Xenopus oocytes. We have applied this technique to recording recombinant ion channel currents in giant membrane patches excised from mammalian cell lines. Giant patches excised from HEK293 cells expressing recombinant K 1 and Na 1 channels show good signal to noise characteristics, high bandwidth, withstand large voltage excursions, and are mechanically stable. The manufacture of giant patch pipettes is illustrated in Fig. 1. Patch pipettes were pulled from type 8250 borosilicate glass capillaries (1.5 mm outer diameter; 0.9 mm inner * Corresponding author. Tel.: 11-773-702-7552; fax: 11-773702-4791. E-mail address: [email protected] (K.M. Houamed).
diameter; Garner Glass, Claremont CA) on a P87 puller (Sutter Instruments, Novato, CA), using a four-step program to achieve ,1 mm tip diameter (Fig. 1A). To minimize capacitance artifacts associated with bath-fluid creep, pipette tips and shanks were dipped in molten dental wax while positive pressure was applied to the back of the pipettes (Fig. 1A). A pipette is then mounted into an MF83 microforge (Narishige, Tokyo, Japan), equipped with a reticule eyepiece, and a platinum filament coated with an 8250-glass bead. The filament temperature is set so that the pipette tip melts only when it is in direct contact with the glass bead on the filament. The pipette is then inserted into the molten glass bead and filament heat is immediately turned off. The retraction of the filament upon cooling cuts the pipette cleanly at the point where the pipette tip and the glass bead fuse. If the cut is not clean, or the pipette tip is smaller than desired, the cutting process (i.e. heat, insert, fuse, cool, break) can be repeated in a stepwise manner (Fig. 1B–E). Once the desired geometry is achieved, the pipette tip is then drawn away from the filament and fire polished at a higher temperature than that used for tip cutting. The cutting and polishing process evaporates the wax coating from the lowermost ,50 mm of the giant patch pipette tip. Tip diameters of our finished giant patch pipettes typically range from 6 to 12 mm (Figs. 1E and 3).
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 2) 00 55 6- 6
18
J.J. Couey et al. / Neuroscience Letters 329 (2002) 17–20
Fig. 1. Giant patch pipette manufacture. (A) Patch pipettes before (top) and after (bottom) being dipped in molten dental wax. Scale bar, 1 mm. (B–E) Cutting and polishing a giant patch pipette. (B) A patch pipette approaches the glass-coated microforge filament. (C,D) Contact is made and the tip is progressively melted to a larger diameter. (E) Tip is cut and polished. The 10 mm scale bar in (E) applies to panels (B–E).
Immediately before filling with recording solution, a mineral oil mixture is siphoned into, and then expelled through the giant patch pipette tip; patches formed on coated tips are mechanically more stable and can withstand large voltage excursions [4]. This mixture comprises equal parts of light and heavy mineral oil (Sigma, St. Louis, MO), into which 1% weight/volume of Parafilm (American National Can, Chicago, IL) is melted. After coating with the oil mixture, the giant patch pipette is filled with either normal mammalian Ringer’s solution of the following composition: 144 mM NaCl; 3 mM KCl; 2 mM CaCl2; 1 mM MgCl2; 5 mM glucose; 2 mM NaHCO3; 10 mM N-[2-hydroxyethyl]piperazine-N 0 -[2-ethanesulfonic acid] (HEPES); pH 7.3; or an isotonic K 1 Ringer’s of the same composition except 160 mM potassium aspartate (KAsp) replaces KCl and NaCl. The pipette is mounted into a standard patch holder with a side port connected to a three-way tap. Using the three-way tap, the patch pipette can either be connected to an air-filled micro-syringe (Captrol III; Drummond Scientific, Broomall, PA), or to the atmosphere. The micro-syringe allows fine control of positive and negative pressure applied to the patch pipette tip; this fine control greatly facilitates giant patch formation. The recording chamber (Fig. 2) is made from a 100 mm £ 75 mm £ 5 mm Delrin sheet, machined to contain a main compartment connected to a perfusion inlet and an outlet. A 75 mm £ 50 mm microscope slide forms the bottom of the chamber; the two parts are sealed together using silicone polymer (RTV 732; Corning, Midland, MI). A side
branch of the main chamber houses a fixed horizontal pipette. The horizontal pipette is pulled, cut and polished in the same manner as for the giant patch pipette, except it is not waxcoated or treated with the oil mixture. It is filled with extracellular solution (normal or isotonic K 1 Ringer’s), and connected to a 10-ml air-filled syringe via fine polyethylene tubing; Parafilm is used to seal the polyethylene tubing onto the back of the pipette. A small bead of dental wax fixes the pipette in place in the recording chamber, so that it opposes and is parallel to the long axis of the recording pipette. The fixed pipette need not be replaced after every recording; we routinely use it for more than one day. HEK293 cell lines constitutively expressing Na 1 channels or SKCa channels were grown on 35 mm dishes to ,30% confluency in Dulbecco’s modified Eagle’s medium (Invitrogen Corporation, Carlsbad, CA) supplemented with 10% fetal bovine serum and antibiotics. Prior to recording, the cells were washed with divalent ion-free phosphatebuffered saline (Invitrogen), and detached with 1 ml of a 1:1 mixture of 0.05% trypsin and versene (Invitrogen). Thirty to sixty seconds after addition of trypsin/versene, the cells lift off the dish. They are then lightly triturated, added to a polystyrene centrifuge tube containing 10 ml of the culture medium and allowed to recover for approximately 20 min. This cell suspension could be used for several hours. The recording chamber is rinsed with Ringer’s solution, and a few drops of the cell suspension are added. In normal extracellular Ca 21, the cells adhere to the bottom of the chamber in a few minutes. Giant patches are most readily formed from hemispherical, lightly adhering cells (Fig. 3). Fig. 3 depicts the process of giant patch formation. To form a giant patch, a giant patch pipette, under slight positive pressure, is brought as close to the cell as possible without contact. As the cell is approached, the positive pressure is decreased to avoid cell detachment. After contact, minimal negative pressure is applied to achieve seal formation, and the cell is lifted ,0.5 mm off the chamber bottom, so that it is in the same Z plane as the tip of the fixed horizontal
Fig. 2. Recording chamber. Schematic, drawn to scale, to illustrate the spatial arrangement of the fixed and giant patch pipettes. Symbols for the chamber parts: I, slow perfusion inlet; S, suction; A, agar bridge; F, fixed pipette; G, giant patch pipette attached to a holder and headstage; L; local perfusion outlet. Scale bar, 20 mm.
J.J. Couey et al. / Neuroscience Letters 329 (2002) 17–20
Fig. 3. Giant patch formation. A sequence of micrographs illustrating the process of giant patch formation. (A) Giant patch pipette approaches an HEK293 cell. (B) Cell-attached giant patch approaches the fixed pipette. (C) Immediately after patch excision. The 10 mm scale bar in (A) applies to panels (A–C). (D) Patch pipette is positioned within the local microperfusion device; scale bar, 50 mm. The bump approximately 50 mm from the tip of the patch pipette in (D) is the lower visible edge of the wax coating.
pipette (Fig. 3A–B). GigaV (GV) seals can form instantaneously but may take several minutes after initial contact. The suction is turned off once GV seal resistance is achieved, and the chamber is perfused with a nominally Ca 21-free intracellular solution of the following composition: 160 mM KAsp; 1 mM K2EGTA; 10 mM NaCl; 10 mM KCl; 5 mM HEPES; pH 7.3. The patch is excised in three steps. First, the chamber is moved using the XY translator, bringing the fixed pipette tip to within approximately one cell diameter, in the X and Y planes, from the cell-attached giant patch pipette (Fig. 3B). Second, negative pressure is applied to the fixed pipette; because of the proximity of the two pipette tips, this negative pressure can visibly deform the cell. Third, the fixed pipette (and chamber) is quickly moved past the patch pipette perpendicular to its long axis, partially sealing to and then bisecting the cell in a single motion, thus forming a giant inside-out patch (Fig. 3C). The patch is then positioned within the outlet of a local microperfusion device consisting of a 500-mm pipette connected, via a 10:1 manifold, to reservoirs containing test solutions (Fig. 3D). Using manually switched valves, the time constant for solution change of this system is ,100 ms. Excised patches formed in this manner are mechanically robust, typically lasting .20 min; they also withstand large excursions in test voltages (e.g. 2100 to 1200 mV). With practice, it is possible to achieve GV seal and patch excision rates similar to conven-
19
Fig. 4. Recombinant sodium channel currents recorded in an excised giant patch. (A) Activation and temporal stability. Top panel: linear capacity and leak currents evoked by a 40 mV hyperpolarizing test pulse from 2120 mV holding potential; the calculated resistance of this patch is 2 GV. Second and third panels: a family of membrane currents activated by 5 ms depolarizing voltage steps, in 10 mV increments, from 2100 mV holding potential, recorded 1 and 11 min after patch excision, respectively. Bottom panel: peak current versus test potential; the continuous curves are cube spline interpolations of the data points; filled squares and clear circles correspond to data recorded 1 and 11 min after patch excision, respectively. (B) Tail currents: activated by repolarizing to the different potentials indicated, following a 300 ms depolarizing pulse to 210 mV. (C) Steady-state inactivation (h1) curve. Peak current, activated by a test voltage step to 220 mV, normalized to the peak current in the absence of a prepulse, is plotted as a function of the voltage of a 1000 ms conditioning voltage step. The continuous curve is a least square fit to a Boltzmann equation, it yields a half activation potential of 288.2 mV, and a slope factor of 5.4 mV. In all cases, the data were filtered at 40 kHz and digitized at 100 kHz. Except for the top panel in (A), linear capacity and leak components were digitally subtracted using a P/4 technique. The patch pipette contained normal mammalian Ringers’ solution; the intracellular aspect of the patch was exposed to nominally Ca 21-free intracellular solution. The data are uncorrected for voltage offsets due to liquid junction potentials. Scale bars: (A), 1 nA, 1 ms; and (B), 100 ms.
tional excised patches. Thus, using 174 pipettes, we achieved 156 GV seals (89.7%) and 100 giant excised patches (57.5%). Moreover, the giant patch technique does not suffer from a common pitfall of conventional excised patches: membrane vesicle formation. We did not observe a single incidence of vesicle formation in over 300 recordings. We tested the performance of the giant patch on two different recombinant channels. To test the speed and fidelity of the clamp, we recorded macroscopic Na 1 currents (Fig. 4). The Na 1 current records show typical voltage dependence of activation (Fig. 4A), and inactivation (Fig. 4C). In five patches, we determined the voltage dependence of inactivation particle (h1): the half maximal voltage was 290.2 mV (^SD ¼ 4:4 mV; this value is uncorrected for liquid junction potential offset), and the slope factor was 5.2 mV. In addition, the recordings of sodium channel tail currents (Fig. 4B) demonstrate the microsecond time resolution of this patch configuration.
20
J.J. Couey et al. / Neuroscience Letters 329 (2002) 17–20
We also recorded macroscopic SKCa currents from SK3 channels (also known as SKCa3 and KCNN3). These channels are small conductance K 1 channels activated by submicromolar intracellular Ca 21 concentrations [5]. Intracellular magnesium ions block SKCa channels in a voltage-dependent manner, causing them to inwardly rectify [6]. To activate SKCa channel currents we used an intracellular solution containing 5 mM free Ca 21; it was composed of: 160 mM KAsp; 10 mM K2HEDTA; 6.53 mM CaCl2; 10 mM HEPES; pH 7.3. To study voltage-dependent block by intracellular Mg 21, we used a solution containing 5 mM free Ca 21 and 1 mM free Mg 21; it was composed of: 160 mM KAsp; 10 mM K2EDTA; 9.7 mM MgCl2; 0.867 mM CaCl2; 5 mM HEPES; pH 7.3. Using the giant patch technique, we show a timedependent component of the onset block, proceeding exponentially with a time constant of ,190 ms (Fig. 5). Although the giant patches were mechanically stable, routinely maintaining GV seal resistances for more than 20 min, the stability of the ionic currents depended on the channel recorded. Thus, SK3 and SK2 channel currents were stable, whereas SK1 currents rapidly washed out. Sodium channels showed an intermediate rate of washout (Fig. 4A). We have described a method for giant excised patch recording from cloned channels expressed in mammalian cell lines. The utility of the giant patch method has previously been elegantly demonstrated in Xenopus oocytes [2–4]. This report extends the utility of this technique to recombinant ion channels expressed in mammalian cells. Moreover, the technique is simple and does not require specialized equipment beyond that already present in a patch clamp laboratory. The only necessary modification is the simple recording chamber, which can be made using readily available hand tools. We found it easier to cut patch pipettes to the desired size if they were pulled with longer shanks. Such pipettes are readily pulled on horizontal or vertical pullers. Fine control of pipette pressure during patch formation is critical to the success of this technique, especially with larger pipette tips. Therefore, it may be helpful to initially practice patch formation and excision with narrower giant patch pipettes. The perfusion speed of our patches was ,100 ms; however, this is most likely limited by the manual switching of the apparatus. It is likely that this speed could be considerably improved by using a faster switching apparatus. This study was funded by grants from the Brain Research Foundation, the Diabetes Research and Training Center, Digestive Diseases Research Core Center (NIH DK42086), The Whitehall Foundation, and the Midwest affiliate of the American Heart Association. J.C.D. was supported by USPHS grant T32-DA-07255. The authors would like to thank Drs J.P. Adelman and J.W. Kyle for supplying KCNN3 cDNA and Na channel cell line, respectively, Drs D.W. Hilgemann and J.D. Marks for advice, and Ms M. Garcia for technical support.
Fig. 5. Voltage-dependent block of recombinant SKCa channel currents underlies their inward rectification. (A) Membrane currents in a giant patch excised from a human SK3-channel expressing HEK293 cell line. The membrane currents were evoked by exposure of the intracellular aspect of the patch to a nominally Mg 21-free intracellular solution containing 5 mM free Ca 21. (B) The solution bathing the intracellular aspect of the patch was switched to one containing 1 mM Mg 21. This causes voltagedependent block of the SKCa current. A time-dependent component of the block appears as a rapidly decaying relaxation (t , ,190 ms) at the beginning of depolarizing pulses. In (A) and (B), the holding potential was 0 mV. One-hundred-millisecond voltage steps ranged from 2100 to 100 mV, in 10 mV increments. Data were filtered at 40 kHz, and sampled at 100 kHz. In both panels, the linear leak and capacity membrane current components were digitally subtracted, using membrane currents recorded in the nominally Ca 21-free intracellular solution. Scale bars: 5 nA; 10 ms. (C) Steady-state current–voltage relation for the control SK3 channel currents and their block in the presence of 1 mM intracellular Mg 21, determined from (A) and (B), respectively. [1] Fakler, B., Brandle, U., Glowatzki, E., Weidemann, S., Zenner, H.P. and Ruppersberg, J.P., Strong voltage-dependent inward rectification of inward rectifier K 1 channels is caused by intracellular spermine, Cell, 80 (1995) 149–154. [2] Hilgemann, D.W., Giant excised cardiac sarcolemmal membrane patches: sodium and sodium–calcium exchange currents, Pflug. Arch., 415 (1989) 247–249. [3] Hilgemann, D.W., The giant membrane patch, In B. Sakmann and E. Neher (Eds.), Single-channel Recording, Plenum Press, New York, 1995, pp. 307–327. [4] Hilgemann, D.W. and Lu, C.C., Giant membrane patches: improvements and applications, Methods Enzymol., 293 (1998) 267–280. [5] Kohler, M., Hirschberg, B., Bond, C.T., Kinzie, J.M., Marrion, N.V., Maylie, J. and Adelman, J.P., Small-conductance, calcium-activated potassium channels from mammalian brain, Science, 273 (1996) 1709–1714. [6] Soh, H. and Park, C.S., Inwardly rectifying current–voltage relationship of small-conductance Ca 21-activated K 1 channels rendered by intracellular divalent cation blockade, Biophys. J., 80 (2001) 2207–2215. [7] Womack, K.B., Gordon, S.E., He, F., Wensel, T.G., Lu, C.C. and Hilgemann, D.W., Do phosphatidylinositides modulate vertebrate phototransduction?, J. Neurosci., 20 (2000) 2792–2799. [8] Xia, X.M., Fakler, B., Rivard, A., Wayman, G., Johnson-Pais, T., Keen, J.E., Ishii, T., Hirschberg, B., Bond, C.T., Lutsenko, S., Maylie, J. and Adelman, J.P., Mechanism of calcium gating in small-conductance calcium-activated potassium channels, Nature, 395 (1998) 503–507.
Giant excised patch recordings of recombinant ion channel currents expressed in mammalian cells Jonathan J. Couey a, Devon P. Ryan a, John T. Glover a, John C. Dreixler a,b, Joseph B. Young a, Khaled M. Houamed a,* a
Department of Anesthesia and Critical Care, University of Chicago, 5841 S. Maryland Avenue, Box 4028, Chicago, IL 60637, USA b Department of Neurobiology, Pharmacology and Physiology, University of Chicago, Chicago, IL 60637, USA Received 2 April 2002; received in revised form 14 May 2002; accepted 15 May 2002
Abstract The giant excised patch variant of patch clamp recording combines microsecond time resolution of macroscopic currents with rapid exchange of the experimental solutions at the intracellular membrane surface. This technique has been applied to a limited number of cell types, including Xenopus oocytes, muscle cells, and photoreceptors. We have applied this technique to recording recombinant ion channel currents expressed in membrane patches excised from HEK293 cell lines. Giant inside-out patch recordings of Na 1 channels and SKCa type calcium-activated potassium channels show high temporal resolution and excellent signal to noise characteristics. This technique will facilitate the study of recombinant ion channels expressed in mammalian cells. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: SK channel; KCNN channel; Calcium-activated potassium channel; Sodium channel; Magnesium; Channel block; Rectification
The ‘giant patch’ [2–4] has three advantages over the conventional inside-out excised patch recording technique: (1), it allows macroscopic current measurements with high signal to noise ratios; (2), because of the low access resistance, microsecond time resolution can be achieved; and (3), experimental solutions at the intracellular patch surface can be exchanged rapidly. These advantages enhance the utility of the giant patch technique for studies of gating, modulation, permeation, and blocking of ion channels and transporters [1,4,7,8]. However, this technique has been limited to certain native channels and recombinant channels expressed in Xenopus oocytes. We have applied this technique to recording recombinant ion channel currents in giant membrane patches excised from mammalian cell lines. Giant patches excised from HEK293 cells expressing recombinant K 1 and Na 1 channels show good signal to noise characteristics, high bandwidth, withstand large voltage excursions, and are mechanically stable. The manufacture of giant patch pipettes is illustrated in Fig. 1. Patch pipettes were pulled from type 8250 borosilicate glass capillaries (1.5 mm outer diameter; 0.9 mm inner * Corresponding author. Tel.: 11-773-702-7552; fax: 11-773702-4791. E-mail address: [email protected] (K.M. Houamed).
diameter; Garner Glass, Claremont CA) on a P87 puller (Sutter Instruments, Novato, CA), using a four-step program to achieve ,1 mm tip diameter (Fig. 1A). To minimize capacitance artifacts associated with bath-fluid creep, pipette tips and shanks were dipped in molten dental wax while positive pressure was applied to the back of the pipettes (Fig. 1A). A pipette is then mounted into an MF83 microforge (Narishige, Tokyo, Japan), equipped with a reticule eyepiece, and a platinum filament coated with an 8250-glass bead. The filament temperature is set so that the pipette tip melts only when it is in direct contact with the glass bead on the filament. The pipette is then inserted into the molten glass bead and filament heat is immediately turned off. The retraction of the filament upon cooling cuts the pipette cleanly at the point where the pipette tip and the glass bead fuse. If the cut is not clean, or the pipette tip is smaller than desired, the cutting process (i.e. heat, insert, fuse, cool, break) can be repeated in a stepwise manner (Fig. 1B–E). Once the desired geometry is achieved, the pipette tip is then drawn away from the filament and fire polished at a higher temperature than that used for tip cutting. The cutting and polishing process evaporates the wax coating from the lowermost ,50 mm of the giant patch pipette tip. Tip diameters of our finished giant patch pipettes typically range from 6 to 12 mm (Figs. 1E and 3).
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 2) 00 55 6- 6
18
J.J. Couey et al. / Neuroscience Letters 329 (2002) 17–20
Fig. 1. Giant patch pipette manufacture. (A) Patch pipettes before (top) and after (bottom) being dipped in molten dental wax. Scale bar, 1 mm. (B–E) Cutting and polishing a giant patch pipette. (B) A patch pipette approaches the glass-coated microforge filament. (C,D) Contact is made and the tip is progressively melted to a larger diameter. (E) Tip is cut and polished. The 10 mm scale bar in (E) applies to panels (B–E).
Immediately before filling with recording solution, a mineral oil mixture is siphoned into, and then expelled through the giant patch pipette tip; patches formed on coated tips are mechanically more stable and can withstand large voltage excursions [4]. This mixture comprises equal parts of light and heavy mineral oil (Sigma, St. Louis, MO), into which 1% weight/volume of Parafilm (American National Can, Chicago, IL) is melted. After coating with the oil mixture, the giant patch pipette is filled with either normal mammalian Ringer’s solution of the following composition: 144 mM NaCl; 3 mM KCl; 2 mM CaCl2; 1 mM MgCl2; 5 mM glucose; 2 mM NaHCO3; 10 mM N-[2-hydroxyethyl]piperazine-N 0 -[2-ethanesulfonic acid] (HEPES); pH 7.3; or an isotonic K 1 Ringer’s of the same composition except 160 mM potassium aspartate (KAsp) replaces KCl and NaCl. The pipette is mounted into a standard patch holder with a side port connected to a three-way tap. Using the three-way tap, the patch pipette can either be connected to an air-filled micro-syringe (Captrol III; Drummond Scientific, Broomall, PA), or to the atmosphere. The micro-syringe allows fine control of positive and negative pressure applied to the patch pipette tip; this fine control greatly facilitates giant patch formation. The recording chamber (Fig. 2) is made from a 100 mm £ 75 mm £ 5 mm Delrin sheet, machined to contain a main compartment connected to a perfusion inlet and an outlet. A 75 mm £ 50 mm microscope slide forms the bottom of the chamber; the two parts are sealed together using silicone polymer (RTV 732; Corning, Midland, MI). A side
branch of the main chamber houses a fixed horizontal pipette. The horizontal pipette is pulled, cut and polished in the same manner as for the giant patch pipette, except it is not waxcoated or treated with the oil mixture. It is filled with extracellular solution (normal or isotonic K 1 Ringer’s), and connected to a 10-ml air-filled syringe via fine polyethylene tubing; Parafilm is used to seal the polyethylene tubing onto the back of the pipette. A small bead of dental wax fixes the pipette in place in the recording chamber, so that it opposes and is parallel to the long axis of the recording pipette. The fixed pipette need not be replaced after every recording; we routinely use it for more than one day. HEK293 cell lines constitutively expressing Na 1 channels or SKCa channels were grown on 35 mm dishes to ,30% confluency in Dulbecco’s modified Eagle’s medium (Invitrogen Corporation, Carlsbad, CA) supplemented with 10% fetal bovine serum and antibiotics. Prior to recording, the cells were washed with divalent ion-free phosphatebuffered saline (Invitrogen), and detached with 1 ml of a 1:1 mixture of 0.05% trypsin and versene (Invitrogen). Thirty to sixty seconds after addition of trypsin/versene, the cells lift off the dish. They are then lightly triturated, added to a polystyrene centrifuge tube containing 10 ml of the culture medium and allowed to recover for approximately 20 min. This cell suspension could be used for several hours. The recording chamber is rinsed with Ringer’s solution, and a few drops of the cell suspension are added. In normal extracellular Ca 21, the cells adhere to the bottom of the chamber in a few minutes. Giant patches are most readily formed from hemispherical, lightly adhering cells (Fig. 3). Fig. 3 depicts the process of giant patch formation. To form a giant patch, a giant patch pipette, under slight positive pressure, is brought as close to the cell as possible without contact. As the cell is approached, the positive pressure is decreased to avoid cell detachment. After contact, minimal negative pressure is applied to achieve seal formation, and the cell is lifted ,0.5 mm off the chamber bottom, so that it is in the same Z plane as the tip of the fixed horizontal
Fig. 2. Recording chamber. Schematic, drawn to scale, to illustrate the spatial arrangement of the fixed and giant patch pipettes. Symbols for the chamber parts: I, slow perfusion inlet; S, suction; A, agar bridge; F, fixed pipette; G, giant patch pipette attached to a holder and headstage; L; local perfusion outlet. Scale bar, 20 mm.
J.J. Couey et al. / Neuroscience Letters 329 (2002) 17–20
Fig. 3. Giant patch formation. A sequence of micrographs illustrating the process of giant patch formation. (A) Giant patch pipette approaches an HEK293 cell. (B) Cell-attached giant patch approaches the fixed pipette. (C) Immediately after patch excision. The 10 mm scale bar in (A) applies to panels (A–C). (D) Patch pipette is positioned within the local microperfusion device; scale bar, 50 mm. The bump approximately 50 mm from the tip of the patch pipette in (D) is the lower visible edge of the wax coating.
pipette (Fig. 3A–B). GigaV (GV) seals can form instantaneously but may take several minutes after initial contact. The suction is turned off once GV seal resistance is achieved, and the chamber is perfused with a nominally Ca 21-free intracellular solution of the following composition: 160 mM KAsp; 1 mM K2EGTA; 10 mM NaCl; 10 mM KCl; 5 mM HEPES; pH 7.3. The patch is excised in three steps. First, the chamber is moved using the XY translator, bringing the fixed pipette tip to within approximately one cell diameter, in the X and Y planes, from the cell-attached giant patch pipette (Fig. 3B). Second, negative pressure is applied to the fixed pipette; because of the proximity of the two pipette tips, this negative pressure can visibly deform the cell. Third, the fixed pipette (and chamber) is quickly moved past the patch pipette perpendicular to its long axis, partially sealing to and then bisecting the cell in a single motion, thus forming a giant inside-out patch (Fig. 3C). The patch is then positioned within the outlet of a local microperfusion device consisting of a 500-mm pipette connected, via a 10:1 manifold, to reservoirs containing test solutions (Fig. 3D). Using manually switched valves, the time constant for solution change of this system is ,100 ms. Excised patches formed in this manner are mechanically robust, typically lasting .20 min; they also withstand large excursions in test voltages (e.g. 2100 to 1200 mV). With practice, it is possible to achieve GV seal and patch excision rates similar to conven-
19
Fig. 4. Recombinant sodium channel currents recorded in an excised giant patch. (A) Activation and temporal stability. Top panel: linear capacity and leak currents evoked by a 40 mV hyperpolarizing test pulse from 2120 mV holding potential; the calculated resistance of this patch is 2 GV. Second and third panels: a family of membrane currents activated by 5 ms depolarizing voltage steps, in 10 mV increments, from 2100 mV holding potential, recorded 1 and 11 min after patch excision, respectively. Bottom panel: peak current versus test potential; the continuous curves are cube spline interpolations of the data points; filled squares and clear circles correspond to data recorded 1 and 11 min after patch excision, respectively. (B) Tail currents: activated by repolarizing to the different potentials indicated, following a 300 ms depolarizing pulse to 210 mV. (C) Steady-state inactivation (h1) curve. Peak current, activated by a test voltage step to 220 mV, normalized to the peak current in the absence of a prepulse, is plotted as a function of the voltage of a 1000 ms conditioning voltage step. The continuous curve is a least square fit to a Boltzmann equation, it yields a half activation potential of 288.2 mV, and a slope factor of 5.4 mV. In all cases, the data were filtered at 40 kHz and digitized at 100 kHz. Except for the top panel in (A), linear capacity and leak components were digitally subtracted using a P/4 technique. The patch pipette contained normal mammalian Ringers’ solution; the intracellular aspect of the patch was exposed to nominally Ca 21-free intracellular solution. The data are uncorrected for voltage offsets due to liquid junction potentials. Scale bars: (A), 1 nA, 1 ms; and (B), 100 ms.
tional excised patches. Thus, using 174 pipettes, we achieved 156 GV seals (89.7%) and 100 giant excised patches (57.5%). Moreover, the giant patch technique does not suffer from a common pitfall of conventional excised patches: membrane vesicle formation. We did not observe a single incidence of vesicle formation in over 300 recordings. We tested the performance of the giant patch on two different recombinant channels. To test the speed and fidelity of the clamp, we recorded macroscopic Na 1 currents (Fig. 4). The Na 1 current records show typical voltage dependence of activation (Fig. 4A), and inactivation (Fig. 4C). In five patches, we determined the voltage dependence of inactivation particle (h1): the half maximal voltage was 290.2 mV (^SD ¼ 4:4 mV; this value is uncorrected for liquid junction potential offset), and the slope factor was 5.2 mV. In addition, the recordings of sodium channel tail currents (Fig. 4B) demonstrate the microsecond time resolution of this patch configuration.
20
J.J. Couey et al. / Neuroscience Letters 329 (2002) 17–20
We also recorded macroscopic SKCa currents from SK3 channels (also known as SKCa3 and KCNN3). These channels are small conductance K 1 channels activated by submicromolar intracellular Ca 21 concentrations [5]. Intracellular magnesium ions block SKCa channels in a voltage-dependent manner, causing them to inwardly rectify [6]. To activate SKCa channel currents we used an intracellular solution containing 5 mM free Ca 21; it was composed of: 160 mM KAsp; 10 mM K2HEDTA; 6.53 mM CaCl2; 10 mM HEPES; pH 7.3. To study voltage-dependent block by intracellular Mg 21, we used a solution containing 5 mM free Ca 21 and 1 mM free Mg 21; it was composed of: 160 mM KAsp; 10 mM K2EDTA; 9.7 mM MgCl2; 0.867 mM CaCl2; 5 mM HEPES; pH 7.3. Using the giant patch technique, we show a timedependent component of the onset block, proceeding exponentially with a time constant of ,190 ms (Fig. 5). Although the giant patches were mechanically stable, routinely maintaining GV seal resistances for more than 20 min, the stability of the ionic currents depended on the channel recorded. Thus, SK3 and SK2 channel currents were stable, whereas SK1 currents rapidly washed out. Sodium channels showed an intermediate rate of washout (Fig. 4A). We have described a method for giant excised patch recording from cloned channels expressed in mammalian cell lines. The utility of the giant patch method has previously been elegantly demonstrated in Xenopus oocytes [2–4]. This report extends the utility of this technique to recombinant ion channels expressed in mammalian cells. Moreover, the technique is simple and does not require specialized equipment beyond that already present in a patch clamp laboratory. The only necessary modification is the simple recording chamber, which can be made using readily available hand tools. We found it easier to cut patch pipettes to the desired size if they were pulled with longer shanks. Such pipettes are readily pulled on horizontal or vertical pullers. Fine control of pipette pressure during patch formation is critical to the success of this technique, especially with larger pipette tips. Therefore, it may be helpful to initially practice patch formation and excision with narrower giant patch pipettes. The perfusion speed of our patches was ,100 ms; however, this is most likely limited by the manual switching of the apparatus. It is likely that this speed could be considerably improved by using a faster switching apparatus. This study was funded by grants from the Brain Research Foundation, the Diabetes Research and Training Center, Digestive Diseases Research Core Center (NIH DK42086), The Whitehall Foundation, and the Midwest affiliate of the American Heart Association. J.C.D. was supported by USPHS grant T32-DA-07255. The authors would like to thank Drs J.P. Adelman and J.W. Kyle for supplying KCNN3 cDNA and Na channel cell line, respectively, Drs D.W. Hilgemann and J.D. Marks for advice, and Ms M. Garcia for technical support.
Fig. 5. Voltage-dependent block of recombinant SKCa channel currents underlies their inward rectification. (A) Membrane currents in a giant patch excised from a human SK3-channel expressing HEK293 cell line. The membrane currents were evoked by exposure of the intracellular aspect of the patch to a nominally Mg 21-free intracellular solution containing 5 mM free Ca 21. (B) The solution bathing the intracellular aspect of the patch was switched to one containing 1 mM Mg 21. This causes voltagedependent block of the SKCa current. A time-dependent component of the block appears as a rapidly decaying relaxation (t , ,190 ms) at the beginning of depolarizing pulses. In (A) and (B), the holding potential was 0 mV. One-hundred-millisecond voltage steps ranged from 2100 to 100 mV, in 10 mV increments. Data were filtered at 40 kHz, and sampled at 100 kHz. In both panels, the linear leak and capacity membrane current components were digitally subtracted, using membrane currents recorded in the nominally Ca 21-free intracellular solution. Scale bars: 5 nA; 10 ms. (C) Steady-state current–voltage relation for the control SK3 channel currents and their block in the presence of 1 mM intracellular Mg 21, determined from (A) and (B), respectively. [1] Fakler, B., Brandle, U., Glowatzki, E., Weidemann, S., Zenner, H.P. and Ruppersberg, J.P., Strong voltage-dependent inward rectification of inward rectifier K 1 channels is caused by intracellular spermine, Cell, 80 (1995) 149–154. [2] Hilgemann, D.W., Giant excised cardiac sarcolemmal membrane patches: sodium and sodium–calcium exchange currents, Pflug. Arch., 415 (1989) 247–249. [3] Hilgemann, D.W., The giant membrane patch, In B. Sakmann and E. Neher (Eds.), Single-channel Recording, Plenum Press, New York, 1995, pp. 307–327. [4] Hilgemann, D.W. and Lu, C.C., Giant membrane patches: improvements and applications, Methods Enzymol., 293 (1998) 267–280. [5] Kohler, M., Hirschberg, B., Bond, C.T., Kinzie, J.M., Marrion, N.V., Maylie, J. and Adelman, J.P., Small-conductance, calcium-activated potassium channels from mammalian brain, Science, 273 (1996) 1709–1714. [6] Soh, H. and Park, C.S., Inwardly rectifying current–voltage relationship of small-conductance Ca 21-activated K 1 channels rendered by intracellular divalent cation blockade, Biophys. J., 80 (2001) 2207–2215. [7] Womack, K.B., Gordon, S.E., He, F., Wensel, T.G., Lu, C.C. and Hilgemann, D.W., Do phosphatidylinositides modulate vertebrate phototransduction?, J. Neurosci., 20 (2000) 2792–2799. [8] Xia, X.M., Fakler, B., Rivard, A., Wayman, G., Johnson-Pais, T., Keen, J.E., Ishii, T., Hirschberg, B., Bond, C.T., Lutsenko, S., Maylie, J. and Adelman, J.P., Mechanism of calcium gating in small-conductance calcium-activated potassium channels, Nature, 395 (1998) 503–507.