| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biophys J, December 2002, p. 3296-3303, Vol. 83, No. 6
*Division of Medicine and National Heart and Lung
Institute, Imperial College of Science, Technology and Medicine, MRC
Clinical Sciences Centre, London W12 0NN, United Kingdom;
Department of Physiological Sciences, University of
Newcastle, Newcastle upon Tyne, United Kingdom;
§Department of Pharmacology, University College London,
London, United Kingdom; ¶Department of Development
Genetics and Molecular Physiology, Institute of Biotechnology, National
Autonomous University of Mexico, Cuernavaca, Morelos, Mexico;
Office of Naval Research, Arlington, Virginia USA; and
**Department of Chemistry, Cambridge University, Cambridge, United
Kingdom
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
SUMMARY
REFERENCES
|
|---|
We have developed a scanning patch-clamp technique that facilitates single-channel recording from small cells and submicron cellular structures that are inaccessible by conventional methods. The scanning patch-clamp technique combines scanning ion conductance microscopy and patch-clamp recording through a single glass nanopipette probe. In this method the nanopipette is first scanned over a cell surface, using current feedback, to obtain a high-resolution topographic image. This same pipette is then used to make the patch-clamp recording. Because image information is obtained via the patch electrode it can be used to position the pipette onto a cell with nanometer precision. The utility of this technique is demonstrated by obtaining ion channel recordings from the top of epithelial microvilli and openings of cardiomyocyte T-tubules. Furthermore, for the first time we have demonstrated that it is possible to record ion channels from very small cells, such as sperm cells, under physiological conditions as well as record from cellular microstructures such as submicron neuronal processes.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
SUMMARY
REFERENCES
|
|---|
The patch-clamp technique, first described by
Neher and Sakmann (Neher and Sakmann, 1976
; Neher et al., 1978) and
initially refined by Hamill (Hamill et al., 1981) is the main
experimental approach used for obtaining information about the
characteristics and distribution of ion channels in living cells. Some
variations of this technique have since been developed for specific
experimental situations (Levitan and Kramer, 1990; Jonas et al., 1997),
and together these methods have allowed recording in an enormous
variety of situations. Data from this type of experiment, along with a host of other work, have shown that ion channels frequently associate with specific subcellular structures and are not uniformly distributed on the cell surface; this is important for their function (Joe and
Angelides, 1992; Angelides, 1986; Banke et al., 1997; Alkondon, 1996;
Tousson et al., 1989; Frosch and Dichter, 1992; Kinnamon et al., 1988;
Karpen et al., 1992; Cohen et al., 1991; Gu et al., 2002; Korchev et
al., 2000a) However, the available methods remain limited when it comes
to studying this subcellular distribution of channels. For example, it
is still difficult to record from fine structures such as microvilli
and the fine dendritic branches of neurons or to patch opaque samples
or obscured structures such as transverse tubules from muscle. A major
reason for this is the difficulty in controlling patch pipettes in
their approach to such samples. This is because an electrode is
typically positioned using manual adjustments while focusing between
the pipette and sample under a light microscope. Some samples, such as
transverse tubules, are not optically visible so these present inherent
difficulty. Ultra-fine structures also provide a considerable challenge
because high-magnification objectives with a short depth of field are used so that for much of the pipette approach, sample and electrode are
not in the same focal plane. Under these circumstances it is easy to
damage the pipette tip.
In this study we demonstrate the utility of a new patch-clamp
method that solves these problems and enables both the identification of small cellular or subcellular membrane structures as well as subsequent patch recording. Our method combines the capabilities of
conventional patch clamping with the advanced features of scanning ion
conductance microscopy (SICM) (Hansma et al., 1989). SICM allows both
ultra-fine positioning of the probe over the sample and high-resolution
imaging of living cell membranes (Korchev et al., 1997, 2000b; Shevchuk
et al., 2001). Because the SICM probe is also used as the patch
pipette, it provides its own image of the cell surface and ensures
precise positioning of the electrode relative to the cell topography.
The development of this method thus allows the investigation of ion
channels that have a unique spatial distribution in otherwise difficult
samples and may also lead to new methods for automated patch recording.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
SUMMARY
REFERENCES
|
|---|
Operation of the scanning patch-clamp is based on the idea that both SICM and patch-clamp recording use a glass micropipette as their probe. By combining these techniques the same micropipette can be used first in SICM protocols to image the cell surface and identify membrane structures of interest and then as a patch pipette for electrophysiological recording as shown schematically in Fig. 1. In this method the scanning micropipette is arranged vertically and manipulated by SICM computer control (Fig. 2 A). A feedback control system is in operation while the pipette approaches the cell surface. As soon as the pipette reaches a distance d from the surface, the SICM feedback control maintains a constant tip-sample separation. This procedure makes the approach straightforward and safe, because the patch pipette is prevented from touching the cell membrane until it is desired to do so to form a seal. It is important to note that a visual image of the sample is not necessary for this approach. Once the SICM protocol has obtained a topographic image of the cell surface it can be used to position the patch pipette over an exact place of interest for patch recording (Fig. 2 B). Finally, feedback control is switched off, the pipette is lowered, and suction is applied, resulting in the formation of a gigaohm seal (Fig. 2 C). Ion channel recording is then performed by conventional methods with all configurations available, e.g., cell-attached, inside-out, whole-cell, and/or outside-out mode.
|
Instrument
The basic topographic imaging equipment is as previously
described for the SICM method (Korchev et al., 1997). Briefly, the pipette is filled with electrolyte and mounted on a piezo stage that is
moved over the cell while maintaining it at a fixed distance from the
surface (Figs. 1 and 2). This is achieved
by a distance-modulated feedback control (Shevchuk et al., 2001)
designed to keep a constant ion current flowing through the pipette.
The present instrument is based on an inverted optical microscope
(Diaphot 200, Nikon Corp., Tokyo, Japan) with a mechanism for
coordinating optical and scanned images, through a video camera with
frame-grabbing hardware and software. There is a computer-controlled
three-axis translation stage with measurement and feedback systems
(East Coast Scientific, Cambridge, UK), which scans the micropipette tip over the specimen. For living cells, displacements in excess of
30-50 µm in the vertical direction are required. We have, therefore used a three-axis piezo translation stage (Tritor, Piezosystem Jena,
Germany) with a 100-µm travel distance in the x,
y, and z directions.
|
This setup was adapted for high-resolution patch clamping by replacing
the current amplifier used previously with a commercial patch-clamp
amplifier (Axopatch 200B, Axon Instruments, Foster City, CA).
Nanopipettes were made from 1.00-mm outer diameter and 0.58-mm inner
diameter borosilicate glass capillaries (Intracel, Herts, UK) using a
laser-based puller (P-2000, Sutter Instrument Co., San Rafael, CA).
Pipettes were used without any further treatment such as fire polishing
or Sylgard coating. The pipette tip radius, determined by scanning
electron microscopy, was ~100 nm and when filled with pipette
solution had an average tip resistance of 150 M
. For patch-clamp
recording, currents were sampled at 10 kHz and filtered at 2 kHz (
3
dB, 4-pole Bessel) using an Axopatch 200B amplifier and pClamp 8.0 software (Axon Instruments), which was also used to generate pulse protocols.
Cell preparations and solutions for patch-clamp recording
Sperm
Strongylocentrotus purpuratus or Lytechinus pictus sea urchins were obtained from Pacific Bio-Marine Laboratories (Long Beach, CA). Spermatozoa were obtained by intracoelomic injection of 0.5 M KCl into sea urchins and collected with a Pasteur pipette directly from the gonopores (Lee and Garbers, 1986). The cell suspension was kept on ice in physiological bath
solution composed of (mM) 486 NaCl, 10 KCl, 25 MgCl2, 10 CaCl2, 10 HEPES,
pH 8. Cells were pretreated with 0.01 mg/ml trypsin (Sigma, Poole, UK)
solution during 10 min in ice and plated on
poly-L-lysine-treated dishes in physiological
bath solution. Pipette filling solution contained (mM) 300 BaCl2, 10 HEPES, pH 8.0.
Neurons
Superior cervical ganglion cells were cultured from 17-day-old Sprague-Dawley rats using previously described methods (Dunn, 1994).
Briefly, ganglia were dissociated with collagenase followed by trypsin
and were cultured in L-15 medium supplemented with 10% fetal calf
serum and 50 ng/ml nerve growth factor. To investigate the calcium
channels, the bath solution was composed of (mM) 120 KCl, 3 MgCl2, 5 EGTA, 11 glucose, 10 HEPES, adjusted to
pH 7.4 with KOH, and the pipette solution contained 90 BaCl2, 10 HEPES, 10 tetraethylammonium(TEA)-Cl, 3 4-aminopyridine (4-AP), adjusted to pH 7.4 with TEA-OH, as described
previously (Delmas et al., 2000). To investigate
K+channels, Leibovitz's L-15 medium was used. In
several experiments we loaded the cells with DiI as previously
described (Hasbani et al., 2001).
Cardiac myocytes
Left ventricular cells from the rat myocardium were isolated as previously described (Harding et al., 1988). Cells were allowed to
attach to polystyrene cell culture dishes filled with bath solutions
composed of (in mM) 120 K-glutamate, 25 KCl, 2 MgCl2, 1 CaCl2, 2 EGTA, 10 glucose, 10 HEPES, adjusted to pH 7.4 with NaOH for the
Ca2+ experiments. Pipette filling solutions were
composed of (in mM) 70 BaCl2, 10 HEPES, 110 sucrose, adjusted to pH 7.4 with TEA-OH for the
Ca2+ experiments. Ionic compositions were similar
to those previously used (Fan et al., 2000). Experiments were started
after cells ceased to contract due to depolarization by high
[K+]bath.
Epithelial kidney cells
A single A6 cell line was kindly provided by Dr. DeSmet. All experiments were carried out on cells between 127 and 134 passages. Cells were cultured as described previously (Sariban-Sohraby et al., 1984) on high-pore-density membrane filters (Falcon, Bedford, MA).
Cells were grown and kept in a 1:1 mixture of modified Ham's F-12
medium and Leibovitz's L-15 medium, modified to contain 105 mM NaCl
and 25 mM NaHCO3. The mixture was supplemented
with 10% fetal calf serum, 200 µg/ml streptomycin, and 200 U/ml
penicillin. Cells were maintained at 28°C in an atmosphere of
humidified air plus 1% CO2. Cells were passaged
and used between days 4 and 5 when they were 90-95% confluent.
Single-channel recordings were performed using a bath solution and
pipette backfill solution both composed of (in mM) 140 NaCl, 5 KCl, 0.8 MgCl2, 1.2 CaCl2, 10 HEPES,
pH 7.4.
Aorta
Samples of aorta obtained from male Sprague-Dawley rats, 200-300 g, were kindly provided by Susan Nourshargh (Imperial College, London). The aorta was mounted in a petri dish and imaged in Leibovitz's L-15 medium.| |
RESULTS AND DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
SUMMARY
REFERENCES
|
|---|
The key feature of our method is a patch electrode that first gathers topographic information and can then be used to seal onto the cell at a precise location. For this reason we call the method smart patch. To explore the potential of this smart-patch technique we chose various experimental conditions and artificially divided them into three groups: first, small samples, visible optically but patched under feedback control; second, samples that are too small to be properly resolved or structures that are undetectable by light microscopy; and finally, opaque samples, where optical imaging is impossible. We show that we can make channel recordings in all of these situations.
Samples visible optically but patched with feedback control
Direct electrophysiological characterization of the ion channels
in sperm by conventional patch-clamp is difficult because of their
small size. In examining sea urchin sperm, previous studies have tried
to get round this issue by using osmotic swelling to facilitate
high-resistance seal formation. However, the cell-attached seals
obtained in this fashion lasted only for a few minutes, making the
characterization of ion channels difficult (Sanchez et al., 2001). As
another alternative, spermatogenic cells, the progenitors of sperm that
are larger, have been used to study sperm ion channel electrophysiology
(Munoz-Garay et al., 2001). The disadvantage of this approach is that
during the last stages of spermatogenesis the type and localization of
ion channels may significantly change (Serrano et al., 1999). As a
third alternative, ion channels from sperm have been reconstituted into
artificial lipid bilayers. Unfortunately this type of reconstitution
makes the definition of channel orientation problematic (Lievano et al., 1990).
Fig. 3 A shows the first
direct recordings of a channel on a sea urchin's sperm using our
smart-patch setup. This recording shows a voltage-dependent multi-state
Ca2+ channel with a high main-state conductance
that is similar to the multi-state conductances seen in planar bilayer
recordings (Lievano et al., 1990). This recording allows us to assign
the polarity of this sperm channel indicating that it opens at positive potentials and then relaxes to lower-conductance states. Such a channel
may thus participate in the uptake of Ca2+
required for the acrosome reaction (Lievano et al., 1990). The probability of obtaining a seal on the sperm cell body using our smart-patch method is 0.45 (in a total of 40 patches), which is 15 times higher than the probability reported in previous studies using
conventional patch-clamp setup (Guerrero et al., 1987). The duration of
cell-attached seals was ~20 min. Also, for the first time, we managed
to obtain an ion channel recording from sea urchin sperm cells in
inside-out configuration.
|
By using infrared video microscopy and differential interference
contrast optics, it is possible to obtain patch recordings from the
dendrites of some neurons in slice preparations (Johnston et al., 1996;
Stuart et al., 1993). However, by the methods currently available it is
difficult to patch the finest dendritic processes or structures, so for
technical reasons, most recordings are usually made on larger features,
e.g., dendrites >1 µm in diameter (Stuart et al., 1993).
Both optically and by the smart-patch technique we observed the
formation of focal swellings (varicosities) on the dendrite-like structures of superior cervical ganglion (SCG) neurons in culture. These varicosities (0.5-1 µm) were visible in a light microscope and
may be related to similar structures seen in organotypic slice and
cultured cell preparations of central nervous system neurons where it
has been demonstrated that sustained transmitter stimulation can
mediate their formation (Park et al., 1996). We have patched this type
of varicosity on the processes of cultured SCG neurons and found that a
22-pS cation channel is present Fig. 3 B (right panel). The current-voltage curve (data not shown) is
characteristic of a calcium channel. It is perhaps surprising that with
such a fine pipette we are still able to obtain channel recordings. However, conventional patch recording of channels in dendrites has
shown that the calcium channels tend to cluster (Lipscombe et al.,
1988), so if the right location is selected channels can be found.
Structures not clearly resolved by light microscopy
A second type of experimental sample that presents difficulties for conventional patch-clamp recording consists of those that are not clearly resolved by light microscopy. We illustrate this with a neuronal structure whose vertical dimension is too small to be properly resolved and with transverse tubules that cannot be suitably detected using light microscopy. Under these conditions the scanning patch clamp was implemented fully as described in Fig. 1; the SICM topographical image of the cell surface was acquired and then the pipette was positioned to the selected region of interest according to known coordinates.
In our first example, a flat membranous protrusion (~100 nm in height) formed on a SCG dendrite-like structure was scanned and patched. As with the neurite varicosities we were able to find channels on the surface of this structure; Fig. 4 A shows the recording of ion channel current. This channel was identified as a potassium channel on the basis of its reversal potential. The probability of obtaining a seal on SCG neurites was 0.62 (in a total of 29 patches).
|
The transverse tubular system of cardiac muscle is a structure that
allows rapid propagation of excitation into the cell interior (Cheng et
al., 1994). It has been shown by immunofluorescence and immunoelectron
microscopy that L-type Ca2+ channels are
distributed along the T-tubule membrane (Carl et al., 1995; Sun et al.,
1995). However, no direct electrophysiological recordings of ion
channels using conventional methods in transverse tubule opening
regions have been reported to date. This is mainly because of the
difficulty in visualizing transverse tubule openings using light
microscopy. Furthermore, it is not possible to position the patch-clamp
pipette into such small openings to obtain ion channel recordings with
conventional methods. Fig. 4 B presents an example of ion
channel patch-clamp recording in a T-tubule opening of a cardiac
myocyte. This channel has the current-voltage and temporal
characteristics of an L-type Ca2+ channel, which
is to be expected in the T-tubule opening regions. Recently, we have
been able to use this technique to show that the L-type
Ca2+ and Cl
channels are
distributed and co-localized in the region of T-tubule openings, but
not in other regions of the myocyte. The probability of obtaining a
seal was 0.70 at the regions of the Z-groove, 0.59 at the T-tubule
opening, and 0.96 at the scallop crest (Gu et al., 2002).
Nontransparent samples
Using conventional optical microscopy there is no satisfactory way
to patch nontransparent samples such as intact tissues and cells grown
on nontransparent material. Although it is possible to use the
blind-patch approach this provides no information about the position or
type of cellular structure from which a recording is made. In this
study we have applied the smart-patch approach to two nontransparent
samples. The first is an intact tissue preparation of rat aorta, and
the second is a layer of A6 cells cultured on a nontransparent filter.
Fig. 5 shows topographical images
obtained by using the smart patch to scan the rat aortic endothelial
cells (Fig. 5 A) and epithelial cells grown on a
nontransparent filter, with high pore density (Fig. 5 B).
With these images the patch pipette can be easily positioned at a
region of interest on the cell surface and then used to obtain a
gigaohm seal and thus to make ion channel recordings. Fig. 5
C shows a higher-resolution image of the one presented in
Fig. 5 B where a successful patch clamp current recording
was made of a channel located at the tip of a microvillus
(right panel). This channel was identified as a
K+ channel on the basis of the reversal potential
of the current-voltage curve under the experimental conditions used.
Similar K+ channels in A6 cells were previously
described (Nilius et al., 1995). The probability of obtaining a seal by
scanning patch clamp was 0.64 on the top of the microvillus (in a total
of 25 patches). We were also able to patch between microvilli. We have
previously been able to show that Cl channels
are also located at the top of microvilli in aldosterone-stimulated A6
cells, and it has also been suggested that other channel types are
located here (Smith et al., 1997).
|
| |
SUMMARY |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
SUMMARY
REFERENCES
|
|---|
We have demonstrated that the smart-patch approach can be used to make single-channel recordings from a large variety of preparations and to direct the patch pipette to precise locations on cellular or subcellular structures. The implementation of the SICM feedback control by itself eases many of the technical problems associated with sealing patch pipettes on target cells, particularly when using very fine pipettes. Because the SICM protocols are well suited for use with pipettes that are much smaller than those conventionally used in patch-clamp recording, our method is a natural way to make recordings from small and precisely located areas of membrane. Furthermore, these structures need not be visible by light microscopy, and indeed opaque samples can be dealt with easily. Although very small pipettes do not have to be used for the smart-patch method, they do provide the highest-resolution images. The pipettes we use, having a diameter of ~200 nm, will sample an area of membrane that is only ~0.03 µm2. However, we have shown that in a variety of preparations, channels can be detected and patch recordings made, because ion channels do not distribute evenly over the cell surface of the cell but cluster in specific subcellular locations.
Potential applications of the smart-patch technique and the use of its feedback control are quite varied. It can be applied to small cells such as sperm cells or to subcellular structures such as microvilli. It can be used to study nontransparent cell samples such as intact tissues, brain slice preparations, epithelial cultures, and hair cells. The spatial precision of the method allows a much improved ability to examine the special localization of ion channels, and the feedback system opens up the possibility of automated patch recording with possible applications in drug screening.
As the microscope nanopipette probe can be used to deliver defined chemical, electrical, or mechanical stimuli to precisely selected areas on the cell surface, this technique could, in the future, also be adapted to investigate the distribution of ligand-gated or mechano-gated ion channels or the effects of localized drug application.
| |
ACKNOWLEDGMENTS |
|---|
We are very grateful to Dr. S. Nourshargh (National Heart and Lung Institute, Imperial College) for providing aortic tissue.
This work was supported by the Biotechnology and Biological Science Research Council, the British Heart Foundation, the Office of Naval Research and the Wellcome Trust, Mexican Council for Science and Technology (CONACyT), National Autonomous University of Mexico (DGAPA-UNAM), and the Third World Academic of Science (TWAS).
| |
FOOTNOTES |
|---|
Address reprint requests to Dr. Yuri E. Korchev, Imperial College of Science, Technology and Medicine, Hammersmith Campus, 5th Floor, MRC Clinical Sciences Centre, DuCane Road, London W12 ONN, UK. Tel.: 44-208-383-2362; Fax: 44-208-383-8306; E-mail: y.korchev{at}ic.ac.uk.
Submitted June 28, 2002, and accepted for publication September 3, 2002.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
SUMMARY
REFERENCES
|
|---|
channels.
J. Physiol.
487:367-378
Biophys J, December 2002, p. 3296-3303, Vol. 83, No. 6
© 2002 by the Biophysical Society 0006-3495/02/12/3296/08 $2.00
This article has been cited by other articles:
 |
|
 |
|
  D. Sanchez, N. Johnson, C. Li, P. Novak, J. Rheinlaender, Y. Zhang, U. Anand, P. Anand, J. Gorelik, G. I. Frolenkov, et al. Noncontact Measurement of the Local Mechanical Properties of Living Cells Using Pressure Applied via a Pipette Biophys. J., September 15, 2008; 95(6): 3017 - 3027. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Dutta, Y. E. Korchev, A. I. Shevchuk, S. Hayashi, Y. Okada, and R. Z. Sabirov Spatial Distribution of Maxi-Anion Channel on Cardiomyocytes Detected by Smart-Patch Technique Biophys. J., March 1, 2008; 94(5): 1646 - 1655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bruckbauer, P. James, D. Zhou, J. W. Yoon, D. Excell, Y. Korchev, R. Jones, and D. Klenerman Nanopipette Delivery of Individual Molecules to Cellular Compartments for Single-Molecule Fluorescence Tracking Biophys. J., November 1, 2007; 93(9): 3120 - 3131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Gorelik, L. Q. Yang, Y. Zhang, M. Lab, Y. Korchev, and S. E. Harding A novel Z-groove index characterizing myocardial surface structure Cardiovasc Res, December 1, 2006; 72(3): 422 - 429. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
