Pflugers Arch - Eur J Physiol (2003) 446:553–558 DOI 10.1007/s00424-003-1097-9
CELL AND MOLECULAR PHYSIOLOGY
Stephan Speier · Marjan Rupnik
A novel approach to in situ characterization of pancreatic b-cells
Received: 21 January 2003 / Accepted: 14 April 2003 / Published online: 28 May 2003 Springer-Verlag 2003
Abstract The tissue-slice technique has enabled major insights into neural and neuroendocrine physiology. Our aim was to adapt this technique to study the function of the endocrine pancreas. The preparation combines an in situ approach, as in gland perfusion, with a resolution characteristic of electrophysiological studies on single cells. The membrane potential in b-cells in the slices recorded using the whole-cell patch-clamp was close to the calculated reversal potential for K+. With sufficient ATP in the recording pipette the b-cells depolarized rapidly on exposure to an increased glucose concentration or stimulation with tolbutamide. The cells preserved bursting and spiking capacity for tens of minutes despite the whole-cell dialysis. In addition, the voltage clamp was used to monitor the changes in the membrane capacitance and to allow correlation of the electrical activity and the cytosolic calcium changes. The pancreatic tissue slice preparation is a novel method for studying the function of the b- and other pancreatic endocrine and exocrine cells under near-physiological conditions. Keywords Ca2+ photometry · Insulin · Membrane capacitance · Membrane potential · Pancreas · Tissue slice · Tolbutamide
Introduction The tissue-slice technique is used routinely to investigate the electrical properties of neurons in the brain, since in slices cell-to-cell contacts, intercellular communication and tissue architecture are preserved [1]. Although similar in situ neuroendocrine preparations from the pituitary and the adrenal gland have also been used in patch-clamp experiments [2, 3], use of a pancreatic slice has not been reported to date. Standard approaches have included S. Speier · M. Rupnik ()) European Neuroscience Institute Gttingen, Waldweg 33, 37073 Gttingen, Germany e-mail:
[email protected] Fax: +49-551-392694
intracellular recording using microelectrodes and patchclamp on dispersed b-cells and isolated islets of Langerhans. In classical microelectrode experiments on mechanically isolated islets, addition of glucose has been shown to depolarize the cell membrane from the resting potential (RMP) to a threshold potential eliciting electrical activity [4, 5]. Upon reaching this threshold Ca2+-dependent action potentials develop, either as intermittent bursting or continuous spiking that can be maintained for tens of minutes [4]. In contrast, patch-clamp recordings from dispersed, cultured b-cells show remarkable rundown of electrical activity after only a few minutes of whole-cell dialysis [6]. The perforated patch-clamp version of the whole-cell technique [7], introduced to avoid the rundown of the excitability of the b-cell cells, has achieved limited success, suggesting that rundown might be a side-effect of cell culture. More advanced patch-clamp studies on enzymatically isolated pancreatic islets [8] have improved our knowledge of intact endocrine cells markedly. In particular, the development of diagnostic tools for separating the different islet cell types has made a major contribution to the understanding of basic a- and d-cell physiology [9, 10]. Although these traditional approaches have yielded valuable insights into the biophysics of the excitability and stimulus-secretion coupling in the pancreatic cells [6, 7, 11], further insight could be gained by enhancing the in situ character, by achieving access to the cytosol for biochemical modifications and by improving the data yield, which is limited in perforated patch-clamp studies. We therefore developed the preparation of pancreatic tissue slices to extend the repertoire of techniques for studying the physiology of the endocrine pancreas. The slice preparation does not disrupt the gross morphology of the pancreas—it preserves the spatial relationship between the different anatomical structures. In addition, the use of relatively large slices enabled us to co-isolate the local ganglia associated with the islets. Embedding the pancreatic tissue in agarose provided sufficient mechanical support for efficient electrophysiological work on the
554
deeper cell layers of the islets of Langerhans and enabled prolonged patch-clamp recordings. In the present study we used standard whole-cell patch-clamp techniques to record the electrical activity and stimulus-secretion coupling in single b-cells in tissue slices. The cells were stimulated using the classic secretagogues glucose and tolbutamide. At rest, b-cells in the tissue slices had a relatively high input conductance that held the membrane potential close to the equilibrium potential for K+ (EK). Single b-cells in the tissue slice maintained their ability to generate glucose-dependent electrical activity even after prolonged whole-cell dialysis, provided sufficient ATP was present in the pipette solution. The effect of perfusion with tolbutamide was similar to that reported using intracellular microelectrodes [12]. In addition, we directly correlated measured changes in membrane capacitance with intracellular [Ca2+] in the b-cells in response to a stimulation voltage-pulse protocol mimicking the spiking activity.
Materials and methods Preparation of pancreatic tissue slices Electrophysiological experiments were performed on single b-cells within the intact islets in pancreatic slices from adult NMRI mice of either sex. Low-gelling agarose (Seaplaque GTG agarose, BMA, Walkersville, Md., USA; 0.475 g in 25 ml extracellular solution), was melted and kept at 37 C. The animals were killed by cervical dislocation, the abdominal cavity opened and agarose injected into the distally clamped bile duct. After injection, the pancreas was cooled with an ice-cold extracellular solution. The injected and hardened pancreas was then extracted, placed in ice-cold extracellular solution and, if necessary, supported with subcapsular injections of agarose. The tissue was transferred to a small dish filled with agarose and immediately cooled on ice. A small cube was cut from the agarose-embedded pancreatic tissue and glued (Super Glue, ND Industries, Troy, Mich., USA) onto the sample plate of the vibrotome (VT 1000 S, Leica, Nussloch, Germany). The tissue was cut at 0.05 mm/sec at 70 Hz into 130-m-thick slices. During slicing the tissue was kept in an ice-cold extracellular solution bubbled continuously with carbogen. After slicing the tissue slices were either used immediately for electrophysiological experiments or imaging or were incubated at 32 C in carbogenbubbled extracellular solution for up to 8 h. To measure insulin release from the superperfused slices we pooled ten slices containing 25–30 islets. Insulin release was measured using the ultrasensitive mouse insulin enzyme-linked immunosorbent (ELISA) assay (Mercodia, Uppsala, Sweden). Imaging The gross morphology of the acute tissue slices was studied using wide-field microscopy (Axioskop 2 and AxioCam; Zeiss, Oberkochen, Germany) and stereomicroscopy (SZX9; Olympus, Tokyo, Japan; Coolpix 995; Nikon, Tokyo, Japan). For immunocytochemistry, freshly prepared tissue slices were fixed and permeabilized with 4% paraformaldehyde and 0.3% Triton X-100 in PBS for 1 h at room temperature. The slices were incubated with the primary antibodies (mouse anti-insulin and rabbit anti-glucagon; Dako, Carpinteria, Calif., USA) for 2 h at 37 C. After washing with PBS, incubation with the secondary antibodies (Alexa 488 goat antimouse and Alexa 647 goat anti-rabbit; Molecular Probes, Eugene, Ore., USA) followed for either for 2 h at 37 C or overnight at 4 C. Bleaching was reduced with the SlowFade Light Antifade Kit
(Molecular Probes). Blood vessels were visualized by incubating the fresh slices for 2 h at 37 C with the panendothelial antibody (Pharmingen, San Diego, Calif., USA). After washing with PBS the secondary antibody (Alexa 488 goat anti-mouse) was added for 45 min at 37 C. Cell viability in the tissue slices was assessed with the Live/Dead kit (Molecular Probes). The immunocytochemical preparations were examined using confocal microscopy (TCS SP2, Leica) using 488 nm (Ar), 543 nm (He-Ne) and 633 nm (He-Ne) laser for excitation. Emission was detected at 505–530 nm (green channel), and >656 nm (red channel). Excitation cross-talk was minimized by sequential scanning. Images were processed using the manufacturer’s confocal software (Leica). Electrophysiology Cells from the second or the third layers in the islets were used for electrophysiological recording to increase the probability of finding b-cells. b-cells were identified by their Na+ current inactivation pattern [8] and the change in membrane potential in the presence of elevated glucose in the extracellular solution [4]. Patch pipettes were pulled (P-97; Sutter Instruments, Novato, Calif., USA) from borosilicate glass capillaries (GC150F-15; WPI, Sarasota, Fla., USA) to a resistance of 2–4 MW in KCl-based solution. The slices were transferred from the incubation beaker to the perfusion chamber and held on the bottom by a nylon-fibre net in a U-shaped platinum-wire frame. The perfusion chamber was mounted on an upright microscope (60w, NA 0.9, Eclipse E600FN; Nikon). During patch-clamp experiments the slices were superperfused continuously with carbogen-bubbled extracellular solution (32 C, 1.5 ml/ min). The standard extracellular medium consisted of (mM): 125 NaCl, 2.5 KCl, 26 NaHCO3, 1.25 NaH2PO4, 2 Na-pyruvate, 0.5 ascorbic acid, 3 myo-inositol, 6 lactic acid, 1 MgCl2, and 2 CaCl2. The glucose concentration during incubations and during the experiments was 3 mM unless otherwise indicated. Tolbutamide was added at 100 M. The pipette filling solution contained (in mM) 150 KCl, 10 HEPES (pH 7.2 with KOH), 2 MgCl2, 0.05 EGTA, and 2 or 5 ATP. All chemicals were from Sigma (St. Louis, Mo., USA) unless otherwise indicated. Recordings were performed in the standard whole-cell patchclamp mode. The voltage-clamp was used to measure membrane currents and membrane capacitance changes, the latter reflecting cell surface area variation due to exocytosis and endocytosis [13]. For this purpose a patch-clamp lock-in amplifier (SWAM II, Celica, Ljubljana, Slovenia) operating at 1.6 kHz lock-in frequency was used. Upon establishment of the whole-cell configuration, the membrane capacitance (Cm) and the access conductance (Ga) were compensated by Cm and Ga compensation controls. A sine voltage of 11 mV rms was applied. The phase angle setting was determined by applying a 1-pF pulse and monitoring the projection of the pulse from the C (signal proportional to Cm) to G outputs of the lock-in amplifier. Cm, Ga, membrane current and membrane potential were recorded after filtering (300 Hz, 4-pole Bessel). The unfiltered membrane current, C, G, membrane potential and photometry output signal were stored simultaneously (digitizer: DRA-400; Bio Logic, Claix, France; CD: PDR-W739, Pioneer, Tokyo, Japan) for off-line analysis. The current-clamp was performed to record the membrane potential changes at different glucose concentrations and during application of tolbutamide. Data was transferred to a PC via an A/D converter (PCI-6035E, National Instruments, Austin, Tex., USA). WinWCP software (John Dempster, University of Strathclyde, UK) was used to apply depolarizing pulses and to acquire and analyse data. We employed a pulse stimulation protocol that differed from the stimulation protocols employed in many previous b-cell studies, and mimicked the physiological electrical activity (see Fig. 3).
555 Ca2+ measurements Fura-6F (Molecular Probes, 0.5 mM in the pipette solution) was used to measure intracellular [Ca2+] ([Ca2+]i) changes simultaneously with the patch-clamp recordings. Fura-6F was excited at 380 nm with a monochromator (Polychrome IV; TILL Photonics, Grfelfing, Germany). A dichroic mirror centred at 400 nm reflected the monochromatic light to the perfusion chamber and transmitted the emitted fluorescence which was further filtered through a 420-nm barrier filter. The fluorescence intensity was measured by a photodiode (TILL Photonics). The filtered signal was recorded (300 Hz, 4-pole Bessel) and stored simultaneously with the unfiltered signal and voltage-clamp signals. [Ca2+]i was calculated as described previously [14].
Results Features of mice pancreatic slices Agarose around the pancreatic tissue and inside the ductal system stabilized the tissue in the slice and made it mechanically suitable for slicing, transferring and electrophysiological experiments. Autolysis due to the digestive enzymes from the exocrine part did not occur and no enzyme inhibitors were needed. Cell viability studies using the Live/Dead kit showed that the endocrine cells survived in the tissue slices for at least 24 h (data not shown). The tissue slices were 130 m thick and about 40–100 mm2 in area. The three-dimensional architecture of the mouse pancreas was well preserved and the acinar branches could be distinguished easily (Fig. 1A). The major part of these branches consisted of large, highly polarized cells representing the cells of the exocrine part of the pancreas. The capillaries reaching the islets could be visualized with the panendothelial antibody: the islets of Langerhans were vascularized more densely than the surrounding exocrine tissue (Fig. 1D, E). Adipose tissue (Fig. 1A) and local ganglia were also detected in the slices (not shown). Incident-light stereomicroscopy revealed average sized islets of Langerhans that appeared as white structures in the slices (Fig. 1A). With transmitted light the islets appeared as brownish cell clusters in the surrounding dark green exocrine tissue (Fig. 1B, D). The size of the islets was from 50 to >500 m (longest axis). The number of islets of Langerhans in the average slice varied from zero to ten, irrespective of the part of the pancreas from which the slices were obtained. A combination of several slices sufficed for correlation between the ELISA based perfusion assay and single-cell electrical activity (Fig.2D, inset). The typical microanatomy of the islets was consistent with previous description, with the non-b-cells being peripheral to the b-cells (Fig. 1C). The islets varied strongly in size, shape and structure. About 50% of the islets were round, like islets isolated using the standard collagenase isolation procedure. Islets of more complex shapes were, however, found frequently. The islets located close to the pancreatic ducts or the blood vessels were often reflected around these structures. The islet cell
Fig. 1A–E Features of fresh pancreatic tissue slices. A Stereomicrograph of the slice. The bulk of the tissue consists of exocrine cells. The islets of Langerhans (round structures) and the adipose tissue (right edge of the slice) are seen as bright white structures. A large blood vessel is indicated by an arrow. The longer axis of the slice is 10 mm. B Transmitted light micrograph of an islet of Langerhans surrounded by the exocrine tissue. Next to the islet a cut blood vessel and exocrine ducts can be seen. Capillaries connect the blood vessel and the islet (arrows). C Confocal-based 3-D reconstruction of the immunostained islet of Langerhans. The cells are stained with insulin antibodies (Ab) (green) and glucagon Ab (red). D Transmission image of the slice showing the exocrine tissue and the islet of Langerhans. E Fluorescence micrograph of D stained with the panendothelial antibody. Scale bar 50 m
density also varied. Most islets were densely packed with cells, but almost hollow islets containing fewer cells were also found in adult pancreatic preparations. These islets were not included in the electrophysiological characterization. Electrical activity of the b-cells in the pancreatic tissue slices A hallmark of the electrical activity of the b-cells is the hyperpolarized membrane potential at rest, followed by depolarization and spiking activity on exposure to elevated glucose concentrations [4]. The resting membrane potential (RMP) depended primarily on EK and values obtained using intracellular microelectrodes are close to the theoretical value of EK [15]. Similarly, under our experimental conditions with the extracellular solu-
556
Fig. 2A–D Current-clamp recordings of the electrical activity of bcells in fresh pancreatic tissue slices. The slices were perfused initially with a solution containing 3 mM glucose. Arrows indicate the times at which the secretagogues were added. A, B 2 mM ATP; C, D 5 mM ATP was included in the pipette solution. A Slight depolarization of the membrane potential with 16.7 mM glucose. B The effect of diazoxide on membrane potential (Vm) after the cell had been depolarized by tolbutamide and high glucose. C Electrical activity stimulated with two different glucose concentrations. The
inset shows, on an expanded time scale, two typical bursts after 10 min of whole-cell dialysis. D Response of Vm to addition of tolbutamide or tolbutamide plus high glucose. Note the bursting pattern at 3 mM glucose and continuous spiking at 13 mM glucose. Inset left: two bursts during tolbutamide application, expanded time scale. Inset right: insulin release from the superfused pancreatic slice stimulated with high glucose. Insulin was measured by enzyme-linked immunosorbent assay (ELISA)
tion containing 2.5 mM K+ zero current potential was between 90 and 100 mV. RMP depended critically on the leak conductance, which we measured at 110 mV (EK, no current through K+ channels). Cells with an RMP more positive than 80 mV (about 25%) were excluded from current-clamp analysis. The standard pipette solution contained 2 mM ATP, a concentration not sufficient to lower the input conductance of the cells by closure of the KATP channels and, hence, to depolarize the membrane (Fig. 2A). In fact, after a few minutes of whole-cell dialysis with 2 mM ATP, the membrane potential hyperpolarized by 5 to 10 mV. This indicates that the resting ATP concentration in the murine b-cell probably exceeds 2 mM, as has been suggested previously [16]. In experiments with 5 mM ATP in the pipette, the KATP channel conductance decreased, resulting usually in slight depolarization (Fig. 2). Superfusion of the slice with an elevated glucose concentration generally did not trigger electrical activity in cells dialysed with 2 mM ATP (Fig. 2A). Indeed, only 10% of patch-clamped cells responded to glucose with
suprathreshold depolarization and electrical activity. In contrast, with 5 mM ATP in the pipette, b-cells were more readily depolarized by exposure to glucose. This depolarization was associated with electrical activity, both sustained and intermittent (Fig. 2C, D). The pattern of slow depolarization and action potentials was comparable to that observed with intracellular microelectrodes (see insets in Fig. 2C and [4]). The membrane potential repolarized rapidly after removal of the glucose. Insulin release during the glucose challenge consisted of a profound first phase with no obvious second phase (inset Fig. 2D). The ability of cells to show the electrical activity was preserved even after several tens of minutes of whole-cell dialysis. The classical secretagogue tolbutamide depolarized the membrane potential rapidly in all tested cells and induced a bursting pattern at substimulatory glucose levels, and constant firing activity at elevated glucose levels (Fig. 2D). Membrane repolarization was also observed in the presence of diazoxide, the classical KATP channel opener (Fig. 2B).
557
Fig. 3 Typical secretory activity of b-cells in response to two successive trains of voltage pulses (lower trace). Upper trace: typical capacitance changes during voltage stimulation. Middle trace: change in cytosolic Ca2+ measured with Fura-6F
Secretory activity of b-cells in pancreatic tissue slices To study the secretory activity of the b-cells in our preparation we employed a voltage pulse protocol mimicking the spiking activity recorded on exposure to an elevated glucose concentration (Fig. 2), i.e. a train of 50 pulses from the RMP to +10 mV (pulse duration 100 ms) at a frequency of 3.3 Hz (total duration 15 s). During these pulses cell capacitance and changes in [Ca2+]i were measured simultaneously (Fig. 3). The resting capacitance of the cells was 5.6€0.2 pF (n=190), consistent with earlier results [6]. [Ca2+]i increased by 680€20 nM (n=4) during the voltage protocol. This change in [Ca2+]i did not run down when applying a second depolarization train (Fig. 3). One-third of the cells studied did not react to the voltage protocol with a detectable capacitance change. The remaining two-thirds (n=22) responded to the voltage protocol with a mean capacitance change of 165€39 fF. This corresponds to a release of about 40 vesicles in 15 s. With the second identical train of pulses after a 10-s interval the capacitance change was clearly reduced (111€28 fF), indicating lack of recycling rather than run-down, a feature reported previously [11].
Discussion Pancreatic tissue slices Freshly prepared pancreatic slices are a novel in situ preparation for electrophysiological and imaging studies on b- and other cell types in the pancreas. The b-cells were not disturbed enzymatically or physically and the islet capsule remained intact, providing near-physiological conditions. The slice technique made it possible to record physiological parameters from a single cell whilst preserving the gross anatomy of the pancreas. Slices can be prepared irrrespective of the size, age or genetic background of the animals. As whole glands are used,
significantly fewer animals are needed for experiments, enabling developmental studies as well as experiments on rare genetic phenotypes with perinatal mortality. The mechanical stability of the tissue slice preparation allowed us to employ various techniques for studying endocrine function down to the molecular level. The fresh tissue slice preparation does not restrict investigations to the distinctive size and type of islets yielded by the standard islet isolation. Indeed, this approach enabled us to study b-cell function in knockout mice that lack obvious islet morphology and are therefore not suitable for use with the standard procedures (e.g. Pax6 knock-out mice, data not shown). It is also possible to monitor physiological or pathophysiological states that result in modified islet morphology and physiology. Because of its mechanical stability, the fresh preparation is suitable for staining and imaging. The major advantage of pancreatic slice is realized in electrophysiological experiments. The patch-clamp technique in fresh slices makes it possible to study b-cells in the deeper cell layers of the islets. In contrast, in isolated islets only the surface cells can be patch-clamped [8]. Whole-cell dialysis and electrical activity of b-cells Traditionally, the perforated patch-clamp technique [7] has been used to study excitability in the pancreatic bcell, while it is generally accepted that cell metabolism contributes significantly to stimulus-secretion coupling. Wash-out of key intermediates may therefore interfere with this coupling. Wash-out in whole cells depends on the pipette resistance, series resistance and, particularly, the size of soluble components [17, 18]. The pipette opening through which we gain access to the cell cytosol is, at most, 1% of the total cell membrane area and thus constitutes the major diffusion barrier for dialysis [18]. In addition, the eukaryotic cell is not an empty container, thus limiting diffusion further. The wash-out/wash-in time for a molecule of about 500 Da in b-cells with standard patch-pipette is more than 30 s [17]. Whole-cell currents and/or capacitance responses measured during the 1st min of whole-cell recording should therefore not differ significantly from those measured in the perforated patch-clamp configuration. In fact, they present a valuable internal control for the subsequent test measurements. The major advantage of the standard whole-cell configuration compared with the perforated patch-clamp configuration is the access to the cell cytosol and the possibility of replacing the cytosol with test solutions. Another argument for the use of the whole-cell configuration to study b-cells in the slice is that, despite the expected wash-out phenomena, the cells maintain the capability to generate electrical activity reversibly and fire action potentials when exposed to glucose and tolbutamide, provided sufficient ATP is supplied (Fig. 2). During dialysis the concentration of important metabolites could still change, although the fluctuations
558
might be slightly damped. It would, of course, be desirable to perform perforated patch-clamp experiments as on the isolated b-cell preparation [7]. Working on the deeper cell layers of the islet to gain the information from the intact cells requires the use of substantial positive pressure on the patch-pipette. It is therefore very difficult to control the precise time at which the perforating agent reaches the tip of the pipette. In conclusion, whole-cell dialysis is a valuable approach for studying the excitability and the secretory activity of single b-cells. It provides an initial internal control period and the possibility of changing the composition of the cytosol. The electrical activity of b-cells in pancreatic slices followed the patterns observed in earlier intracellular microelectrode studies [4, 12, 16]. The glucose-dependent electrical activity depended critically on the ATP concentration in the patch pipette, 2 mM being insufficient to support electrical activity (Fig. 2A). At 5 mM ATP the standard secretagogues glucose and tolbutamide induced standard electrical activity (Fig. 2C, D), with both burst activity (see insets in Fig. 2C and D) and continuous firing (Fig. 2D). The KATP channel opener diazoxide rapidly repolarized the b-cell membrane to levels close to EK (Fig. 2B). b-Cell secretory activity in slices Mimicking the spiking activity recorded during exposure to elevated glucose concentrations failed to change membrane capacitance in one-third of the cells studied. This could reflect a population of b-cells not able to develop the electrical activity. The other two-thirds of the cells studied reacted with a capacitance change corresponding to the secretion of approximately 40 vesicles during the stimulus protocol, i.e. exocytosis of about three vesicles/s, which is close to the calculated fusion frequency measured in gland perfusions [19]. The discrepancy could be related to the inactive b-cells inside the islets that do not react to elevated glucose levels and falsify the calculation of the in vivo insulin secretion. The reason for that may be differential innervation or inhomogeneity between the b-cells. In summary, our fresh, murine pancreas tissue slice preparation is fast and robust and facilitates the use of several different methods for studying function and development of b-cells in the islets of Langerhans, as well as the secretory cells in the exocrine pancreas. Acknowledgements We thank Marion Niebeling and Heiko Rhse for excellent technical assistance, and S. Gpel and P. Rorsman of the University of Lund for valuable discussions. This work was supported by the European Commission (Grant QLG1-CT-200102233 to M.R.) and the Max Planck Society.
References 1. Sakmann B, Stuart G (1995) Patch-pipette recordings from the soma, dendrites and axon of neurons in brain slices. In: Sakmann B, Neher E (eds) Single channel recording. Plenum Press, New York, 199–211 2. Schneggenburger R, Lopez-Barneo J (1992) Patch-clamp analysis of voltage-gated currents in intermediate lobe cells from rat pituitary thin slices. Pflugers Arch 420:302–312 3. Moser T, Neher E (1997) Rapid exocytosis in single chromaffin cells recorded from mouse adrenal slices. J Neurosci 17:2314– 2323 4. Dean PM, Matthews EK (1970) Glucose-induced electrical activity in pancreatic islet cells. J Physiol (Lond) 210:255–264 5. Ashcroft FM, Rorsman P (1989) Electrophysiology of the pancreatic beta-cell. Prog Biophys Mol Biol 54:87–143 6. Trube G, Rorsman P (1986) Calcium and potassium currents recorded from pancreatic beta-cells under voltage clamp control. Adv Exp Med Biol 211:167–175 7. Gillis KD, Misler S (1992) Single cell assay of exocytosis from pancreatic islet B cells. Pflugers Arch 420:121–123 8. Gpel S, Kanno T, Barg S, Galvanovskis J, Rorsman P (1999) Voltage-gated and resting membrane currents recorded from Bcells in intact mouse pancreatic islets. J Physiol (Lond) 521:717–728 9. Gpel SO, Kanno T, Barg S, Weng XG, Gromada J, Rorsman P (2000) Regulation of glucagon release in mouse-cells by KATP channels and inactivation of TTX-sensitive Na+ channels. J Physiol (Lond) 528:509–520 10. Gpel SO, Kanno T, Barg S, Rorsman P (2000) Patch-clamp characterisation of somatostatin-secreting-cells in intact mouse pancreatic islets. J Physiol (Lond) 528:497–507 11. mml C, Eliasson L, Bokvist K, Larsson O, Ashcroft FM, Rorsman P (1993) Exocytosis elicited by action potentials and voltage-clamp calcium currents in individual mouse pancreatic B-cells. J Physiol (Lond) 472:665–688 12. Henquin JC, Meissner HP (1982) Opposite effects of tolbutamide and diazoxide on 86Rb+ fluxes and membrane potential in pancreatic B cells. Biochem Pharmacol 31:1407–1415 13. Neher E, Marty A (1982) Discrete changes of cell membrane capacitance observed under conditions of enhanced secretion in bovine adrenal chromaffin cells. Proc Natl Acad Sci USA 79:6712–6716 14. Carter TD, Ogden D (1994) Acetylcholine-stimulated changes of membrane potential and intracellular Ca2+ concentration recorded in endothelial cells in situ in the isolated rat aorta. Pflugers Arch 428:476–484 15. Atwater I, Ribalet B, Rojas E (1978) Cyclic changes in potential and resistance of the beta-cell membrane induced by glucose in islets of Langerhans from mouse. J Physiol (Lond) 278:117–139 16. Ashcroft SJ, Weerasinghe LC, Randle PJ (1973) Interrelationship of islet metabolism, adenosine triphosphate content and insulin release. Biochem J 132:223–231 17. Pusch M, Neher E (1988) Rates of diffusional exchange between small cells and a measuring patch pipette. Pflugers Arch 411:204–211 18. Mathias RT, Cohen IS, Oliva C (1990) Limitations of the whole cell patch clamp technique in the control of intracellular concentrations. Biophys J 58:759–770 19. Straub SG, Sharp GW (2002) Glucose-stimulated signaling pathways in biphasic insulin secretion. Diabetes Metab Res Rev 18:451–463
