Journal of Experimental Botany, Vol. 52, No. 362, pp. 1933-1939,
September 1, 2001
© 2001 Oxford University Press
Discontinuous single electrode voltage-clamp measurements: assessment of clamp accuracy in Vicia faba guard cells
Julius-von-Sachs-Institut für Biowisssenschaften, Lehrstuhl für Molekulare Pflanzenphysiologie und Biophysik, Universität Würzburg, Julius-von-Sachs-Platz 2, D-97082, Germany
Received 9 February 2001; Accepted 7 June 2001
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The use of a discontinuous single electrode voltage-clamp (dSEVC) offers an attractive alternative to the patch-clamp technique, since whole-cell measurements can be performed with a single sharp electrode. Comparison of current–voltage relations, however, revealed a weaker voltage dependence of channels measured with the dSEVC compared to patch clamp. The accuracy of the dSEVC was tested on Vicia faba guard cells impaled with double-barrelled electrodes. The actual clamp potential was measured independently of the dSEVC, at the second barrel. The weaker voltage dependence of ion channels appeared to be due to an overestimation of the clamp potential by the dSEVC. The deviation between the intended and actual clamp potential showed a linear relationship with the injected current; on average a 126 mV deviation was found for a clamp current of 1 nA. The deviation was probably caused by a slow settling capacity at the electrode, not compensated by the dSEVC amplifier. It is concluded that the dSEVC method in its current state is only suited for the study of small ion conductances in plant cells.
Key words: Discontinuous single electrode voltage clamp, plant cell, guard cell.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The discontinuous single electrode voltage-clamp (dSEVC) technique allows the membrane potential of a cell to be clamped with a sharp single-barrelled electrode. In electrophysiological studies on animal cells, this method is typically used to study cell types that would have been difficult to reach with patch-clamp electrodes (Richter et al., 1996
). The method is of a more general use for plant cells, since the cell wall obstructs a direct application of patch-clamp electrodes. The patch-clamp technique can only be used after enzymatic digestion of the cell wall (Elzenga et al., 1991; Hedrich, 1995) or its removal with a laser (Taylor and Brownlee, 1992; Henriksen et al., 1996). The use of the dSEVC method is an attractive alternative to the whole-cell configuration of the patch-clamp technique, since whole-cell voltage-clamp measurements can be conducted on cells with an intact cell wall.
Voltage-clamp measurements with sharp glass microelectrodes are ideally carried out with two electrodes that separate the current injection from the potential measurement (Eisenberg and Engel, 1970; Finkel and Gage, 1985). The dSEVC separates the current injection from potential measurement in time, by rapid switching between a current injection mode and potential measuring mode (Brenneke and Lindemann, 1974; Finkel and Redman, 1985; Halliwell et al., 1987). The method takes advantage of a slower time constant of the plasma membrane (
m=RmxCm) compared with the electrode. Provided the switching frequency between the current injection- and voltage measuring-mode is high enough, the plasma membrane can be clamped to a steady membrane potential.
The method was first developed for large cells with a low conductance, since these cells have a high capacity and discharge slowly (Brenneke and Lindemann, 1974; Wilson and Goldner, 1975). Physical analysis showed that the method is best used on large cells with a small conductance (Sala and Sala, 1994) and requires microelectrodes with fast settling times (Finkel and Redman, 1985). Recent developments, however, have provided amplifiers that run at high switching frequencies (up to 60 kHz) and enable voltage clamp of much smaller cells (Juusola, 1994).
The single-electrode current clamp technique was first applied on plant cells by Anderson et al., who used the method to study the membrane conductance of root cortical cells of Pisum sativum (Anderson et al., 1974). The method was criticized by Etherton et al. who compared the performance of a single-electrode current clamp with that of a conventional two-electrode current clamp in a single cell (Etherton et al., 1977). The assumption that the time constant of the electrode was small compared to that of the membrane was taken into question. Furthermore, the authors recognized that the capacitance of a microelectrode can never be fully compensated by a microelectrode amplifier. A later report (Schefczik et al., 1983) used the same approach, but a modified analysis procedure was used. The latter authors concluded that the method can be used accurately, provided pulse protocols are used that are adapted to the electrode properties.
More recently, the dSEVC has been applied to electrically isolated plant cells. The method enabled the analysis of ionic currents of (i) laticifer protoplasts of Hevea brasiliensis, a natural rubber-producing plant (Bouteau et al., 1996), (ii) root hair cells of Medicago sativa (Kurkdjian et al., 2000) and (iii) guard cells of several species (Forestier et al., 1998). The plasma membrane conductance of the latter cell type has been studied in detail with the patch-clamp technique (Dietrich et al., 1998) and a voltage-clamp technique that uses double-barrelled electrodes (Blatt, 1992). The data obtained with these two techniques, however, differ from those measured with the dSEVC, with respect to the activation times and voltage dependence of inward- and outward K+-channels. Therefore, an experimental approach was undertaken to measure the accuracy of the dSEVC in guard cells of Vicia faba. The difference in voltage dependence of ion-channels measured with single- or double-barrelled electrodes was quantified.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Tissue preparation
3–6-week-old Vicia faba L. cv. Grünkernige Hangdown plants, were used, which were grown in the greenhouse. The abaxial epidermis was peeled from the leaves and attached to a microscope slide using Medical Adhesive (VM 355, Ulrich AG, St Gallen, Switzerland). The microscope slide was mounted in an experimental chamber filled with bath solution, 50 mM KCl and 1 mM Ca(OH)2 buffered to pH 6.0 with Mes. Guard cells were impaled on an upright microscope (Axioskop 2FS, Carl Zeiss, Göttingen, Germany), at an angle of 30°. The guard cell's long axis had a length of 38 µm (SD=2 µm), while their diameter was 13 µm (SD=1 µm). Assuming a geometry of a twisted cylinder with hemispherical caps on its ends the surface area was 2070 µm2 and the plasma membrane capacitance was
21 pF.
Electrodes and electrical system
All electrodes were pulled from borosilicate glass capillaries with a wall thickness of 0.21 mm (GC100F-10, Clarck Electromedical Instruments, Pangbourne Reading, UK) and filled with 300 mM KCl. Single-barrelled electrodes were pulled on a horizontal laser puller (P2000, Sutter Instrument Co., Novato, CA, USA). Double-barrelled electrodes made from two capillaries that were aligned, heated and twisted 360° on a horizontal puller (PD-5, Narashige, Tokyo, Japan), the electrode-tip was pulled in a two-step procedure. The tip resistance of the electrodes was measured before impalement and ranged from 40 to 120 M
, the capacitance of the electrodes was 16 pF (SD=2, n=4).
Single-barrelled electrodes were connected to a dSEVC amplifier (SEC 05 L/H, NPI Electronic, Tamm, Germany). In case double-barrelled electrodes were used, one barrel was connected to the dSEVC amplifier, while the other was connected to a regular microelectrode amplifier (VF-102, Bio-Logic, Claix, France). Voltage-clamp protocols were fed into the dSEVC amplifier using Pulse software (Heka, Lambrecht, Germany) and an ITC-16 interface (Instrutech, Corp., Elmont, NY, USA). The data were filtered with 8-pole Bessel filters present at the dSEVC amplifier or externally (type 902, Frequency Devices, Haverhill, MA, USA). The low pass filters were run at 0.3 or 1.3 kHz, while data were sampled at 1 or 10 kHz. The dSEVC amplifier was run at a switching frequency of 20 kHz, while current injection and voltage measuring time intervals were equally long (duty cycle=0.5). The capacity compensation of the dSEVC amplifier was adjusted just before impalement, with the following procedure. The amplifier was set to its switching current-clamp mode. Current pulses of 1 nA were fed through the amplifier, the capacity compensation was adjusted until deflections in the voltage signal had disappeared. At this point, the current applied during the current-injection time-interval has no effect on the potential recorded in the voltage-measuring time-interval. The capacity compensation of the regular microelectrode amplifier was set to a sub-critical level.
Numerical analysis
The voltage dependence of the inward rectifying K+-channel was determined by fitting a Boltzmann equation to the conductance–voltage relationship. The conductance was calculated as Gm=Im/(Vm-Erev), where the reversal potential Erev was assumed to be at -30 mV. The following equation was used;
 | (1) |
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
In a first series of experiments, guard cells were impaled with single-barrelled electrodes. The plasma membrane conductance was tested, by clamping the membrane potential from a holding potential of -100 mV to more negative and positive potentials (data not shown, but as in Fig. 1A
). The currents measured were similar to those of Fig. 1C, at potentials negative of -100 mV the activation of inward rectifying channels was found, while outward rectifying channels activated at more positive potentials. Current–voltage plots of these cells confirmed a smaller voltage dependence of both channels measured with the dSEVC, compared to patch- and double-barrelled voltage-clamp techniques (data not shown). The difference was quantified for the inward rectifying channel by comparison of the gating charge of the channel. Using the dSEVC method a value for zg of 0.52 (SE=0.04, n=9) was found, while zg was 1.5 in patch clamp (Dietrich et al., 1998) and 1.4 in double-barrelled voltage-clamp (Blatt, 1992). The difference in gating charge may originate from an overestimation of the clamp voltage by the dSEVC amplifier, a possibility tested in a second series of experiments.
|
Guard cells were impaled with double-barrelled electrodes, one barrel being used to clamp the membrane potential with the dSEVC amplifier. The second barrel was used to measure the membrane potential independently. During voltage clamp, the voltage recorded by the dSEVC did not deviate from the intended clamp potential (Vint) (Fig. 1A
). The potential measured at the second barrel (Vmeas) was close to Vint at the start of the voltage clamp (Fig. 1B), but began to deviate from Vint, with time. Considering Vmeas as a representative of the true clamp potential, the cells apparently were not clamped to a steady membrane potential, but to a slowly changing potential instead. The clamp potential reached a stable value only after period of 0.5 s (Fig. 1B).
The change in Vmeas strongly correlated with the change in clamp current (Fig. 1C). The relationship between the clamp current and the deviation in voltage was determined for values measured at the end of the 2 s test pulses. A linear relation was found for currents up to 1 nA (Fig. 2A), but in some cells the relationship differed for inward and outward currents. On average, the deviation of Vmeas from Vint was 126 mV nA-1 (SD=98, n=6 cells).
|
) and against the voltage recorded independently by the regular microelectrode amplifier (Vmeas, •). Note, the increased voltage dependence of outward and inward rectifying channels when the current is plotted against Vmeas.The steady-state current–voltage relation of the cell depicted in Fig. 1
can be drawn either using Vint or Vmeas; in Fig. 2B both options are displayed. At potentials with a low conductivity of the plasma membrane, both current–voltage relations overlap. However, the curves deviate at potentials where ion channels activate, as a result of the difference between Vint and Vmeas. The voltage dependence of inward and outward rectifying channels is higher when currents are plotted against Vmeas compared with Vint. For the inward rectifier a value for zg of 0.72 (SE=0.05, n=6) was found with Vint, while zg was 1.73 (SE=0.08, n=6) based on Vmeas. Note that the value of zg based on Vmeas is closer to the patch-clamp and double-barrelled voltage-clamp values, than that based on Vint. This indicates that the dSEVC overestimates the membrane potential, during voltage clamp.
A dSEVC amplifier switches fast between current injection and voltage measurement mode. Before the voltage is sampled, the electrode needs to be discharged. Any charge remaining on the tip of the electrode will be added to the membrane potential and thus result in an overestimation of the clamp potential. An accurate functioning of the system therefore depends on a correct setting of the capacity compensation. For the experiments presented the capacity compensation was set with the electrode close to the cell. However, the resistance of the microelectrode very likely changes during impalement and the capacity compensation may need readjustment. The capacity compensation was therefore stepwise increased, while monitoring the deviation between Vint and Vmeas. Increasing the capacity compensation did indeed decrease the deviation (Fig. 3A, B), but now the voltage-clamp became unstable (ringing). This indicates that the capacity of the electrode was now overcompensated. Returning to the original capacity compensation setting, returned the current and voltage traces to the same values as in Fig. 3A (not shown). Apparently, the amplifier is not capable of compensating for the electrode capacity to its full extent.
|
The presence of a component in the electrode capacitance not properly compensated was already seen during tuning of the capacity compensation before impalement. The latter was carried out while feeding 1 nA pulses through the electrode, with the amplifier in its discontinuous current-clamp mode (Fig. 4A
). The capacity compensation was set to a value at which the recorded potential remained unchanged (Fig. 4B). At this setting the capacity is optimally compensated, since the injected current does not affect the voltage recorded in the voltage-measuring time interval. However, jumps in the potential remained after each current change, indicating that the capacity of the electrode was not entirely compensated. The potential jumps probably result from the fact that electrodes do not behave like ideal capacitors (Etherton et al., 1977; Purves, 1981). Redistribution of ions within the electrode, causes a relatively slow change in the capacitance during voltage clamp (Finkel and Redman, 1985). The electrode therefore does not resemble a resistor with a capacitor in parallel. A more realistic presentation of electrode properties is presented in Fig. 4D, a capacitor (Ce2) is added that acts in series with the resistors Rsol1 and Rsol2. The latter resistors stand for the redistribution of ions within the electrode and bath solution.
|
Testing the dSEVC amplifier on an electrical circuit that resembled that of Fig. 4D
, indeed revealed voltage jumps (Fig. 4C), similar to those seen with microelectrodes. In the absence of Ce2, such voltage jumps did not occur (data not shown). Using this circuit, the voltage at Vm was clamped with the dSEVC amplifier and measured independently with a second microelectrode amplifier. In the presence of Ce2, Vmeas deviated from Vint, while Vmeas and Vint were identical in the absence of Ce2.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Based on the present results it was concluded that dSEVC amplifiers in their current state are not suited to study large conductance ion-channels in plant cells. For smaller conductances, however, the method can be used. Under the conditions applied in this report, a plasma membrane conductance of up to 0.8 nS can be measured with an error in the clamp potential of less than 10%.
There are two major problems in studying large conductance K+-currents in guard cells with the dSEVC method. First, the voltage measured by the dSEVC amplifier at the end of the 2 s test pulses, deviates from the actual clamp potential. This gives rise to current–voltage plots that display an underestimated voltage dependence of ion-channels. Second, the clamp potential slowly changes in time as voltage-dependent channels activate and the clamp current increases (Fig. 1B, C). The measured activation of voltage-dependent channels is, in fact, a complex mix of changes in membrane conductance and a concurrent change of the clamp potential. Activation and deactivation kinetics obtained with this technique must therefore be interpreted with caution.
The reason that the dSEVC does not function properly at large clamp currents, was most likely due to its inability to compensate fully for the electrode capacitance. Fundamental problems of the method described by Etherton et al. are to some extent still relevant (Etherton et al., 1977). The first problem concerns the difference in time constant of the electrode and plasma membrane, which should differ sufficiently to separate settling of the electrode and membrane capacitance in time. These time constants should differ at least two orders of magnitude (according to Etherton et al., 1977). This requirement can be met for guard cells when the membrane potential is clamped to a value at which inward- and outward rectifying channels remain deactivated. At this potential, the resistance of the plasma membrane is high and therefore the membrane will only slowly discharge. Now, in principle, the capacity compensation can be set, based on the voltage response to current pulses.
More problematic is the slow settling capacitance of the electrodes. Electrodes do not behave like simple resistors with a capacitance in parallel, but a change in voltage will cause a redistribution of ions in the electrode, resulting in a slow change of the capacitance (Finkel and Redman, 1985). This feature of microelectrodes has received little attention, but is probably the cause for deviation between the potential measured by the dSEVC amplifier and the actual membrane potential.
The performance of the dSEVC can be improved by the use of microelectrodes with a lower resistance or capacitance. The capacitance may be reduced by coating of the electrode, but for the present experiments the electrode capacitance was already minimized by using thick-walled glass. A lower resistance can be achieved by pulling electrodes with a more blunt tip or by using a higher concentration of electrolyte. Electrodes with a more blunt tip, however, are difficult to impale through the cell wall, while the use of high electrolyte concentrations will load the cells with salt, causing a depolarization of the membrane potential (Blatt, 1987).
Alternatively, dSEVC amplifiers may be further developed. A circuit may be incorporated that corrects for the slow settling capacitance of microelectrodes. Such a circuit will have to be tuned to the electrode properties. The modified dSEVC should be designed in such way that it can correct for the change in electrode resistance, occurring upon impalement.
|
Acknowledgments |
|---|
We thank U Schliwa (University of Würzburg) for technical support, P Dietrich (University of Würzburg) for her useful suggestions concerning the analysis and HBA Prins (University of Groningen) for his helpful comments on the manuscript. The work was supported with grants from the Deutsche Forschungs Gemeinschaft to RH.
|
Notes |
|---|
1 To whom correspondence should be addressed. Fax: +49 931 8886157. E-mail: hedrich{at}botanik.uni\|[hyphen]\|wuerzburg.de
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Anderson WP, Hendrix DL, Higinbotam N. 1974. Higher plant cell membrane resistance by a single intracellular electrode method. Plant Physiology53, 122–124.
Blatt MR. 1987. Electrical characteristics of stomatal guard cells: the ionic basis of the membrane potential and the consequence of potassium chloride leakage from microelectrodes. Planta170, 272–287.
Blatt MR. 1992. K+ channels of stomatal guard cells. Characteristics of the inward rectifier and its control by pH. Journal of General Physiology99, 615–644.
Bouteau F, Bousquet U, Pennarun A-M, Convert M, Dellis O, Cornel D, Rona J-P. 1996. Time-dependent K+ currents through plasmalemma of laticifer protoplasts from Hevea brasiliensis. Physiologia Plantarum98, 97–104.
Brennecke R, Lindemann B. 1974. Theory of a membrane-voltage clamp with discontinuous feedback through a pulsed current clamp. Reviews of Scientific Instruments45, 184–188.
Dietrich P, Dreyer I, Wiesner P, Hedrich R. 1998. Cation sensitivity and kinetics of guard-cell potassium channels differ among species. Planta205, 277–287.
Eisenberg RS, Engel E. 1970. The spacial variation of membrane potential near a small source of current in a spherical cell. Journal of General Physiology55, 736–757.
Etherton B, Keifer DW, Spanswick RM. 1977. Comparison of three methods for measuring electrical resistances of plant cell membranes. Plant Physiology60, 684–688.
Elzenga JTM, Keller CP, Van Volkenburgh E. 1991. Patch clamping protoplasts from vascular plants. Plant Physiology97, 1573–1575.
Finkel AS, Gage PW. 1985. Conventional voltage clamping with two intracellular microelectrodes. In: Smith TG, Lecar H, Redman SJ, Gage PW, eds. Voltage and patch clamping with microelectrodes. Baltimore, USA: Williams and Wilkins Co., 47–94.
Finkel AS, Redman SJ. 1985. Optimal voltage clamping with single microelectrode. In: Smith TG, Lecar H, Redman SJ, Gage PW, eds. Voltage and patch clamping with microelectrodes. Baltimore, USA: Williams and Wilkins Co., 95–120.
Forestier C, Bouteau F, Leonhardt N, Vavasseur A. 1998. Pharmacological properties of slow anion currents in intact guard cells of Arabidopsis. Application of the discontinuous single-electrode voltage-clamp to different species. European Journal of Physiology436, 920–927.
Halliwell JV, Plant TD, Standen NB. 1987. Microelectrode techniques and plant cells. In: Standen, NB, Gray PTA, Whitaker MJ, eds. Microelectrode techniques, 13–28.
Hedrich R. 1995. Technical approaches to studying properties of ion channels in plants. In: Sakmann B, Neher E, eds. Single channel recording, 2nd edn. New York: Plenum Press, 277–305.
Henriksen GH, Taylor AR, Brownlee C, Assmann SM. 1996. Laser microsurgery of higher plant cell walls permits patch-clamp access. Plant Physiology110, 1063–1068.[Abstract]
Juusola M. 1994. Measuring complex admittance and receptor current by single electrode voltage-clamp. Journal of Neuroscience Methods53, 1–6.[Web of Science][Medline]
Kurkdjian A, Bouteau F, Pennarum A-M, Convert M, Cornel D, Rona J-P, Bousquet U. 2000. Ion currents involved in early Nod factor response in Medicago sativa root hairs: a discontinuous single electrode voltage-clamp study. The Plant Journal22, 9–17.[Web of Science][Medline]
Purves RD. 1981. Microelectrode methods for intracellular recording and ionophoresis. London: Academic Press Inc.
Richter DW, Pierrefiche O, Lalley PM, Polder HR. 1996. Voltage-clamp analysis of neurons within deep layers of the brain. Journal of Neuroscience Methods67, 121–131.[Web of Science][Medline]
Sala F, Sala S. 1994. Sources of errors in different single-electrode voltage-clamp techniques: a computer simulation study. Journal of Neuroscience Methods53, 189–197.[Web of Science][Medline]
Schefczik K, Simonis W, Schiebe M. 1983. Continuous registration of membrane input resistance of small plant cells using a double-pulse current clamp technique for single-electrode impalements. Plant Physiology72, 368–375.
Taylor AR, Brownlee C. 1992. Localized patch clamping of plasma membrane of a polarized plant cell. Plant Physiology99, 1686–1688.
Wilson WA, Goldner MM. 1975. Voltage clamping with a single microelectrode. Journal of Neurobiology6, 411–422.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||