Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The characterization of a large conductance potassium channel in cultured rat melanotrophs Wong, Kathy 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-0432.pdf [ 3.54MB ]
Metadata
JSON: 831-1.0086935.json
JSON-LD: 831-1.0086935-ld.json
RDF/XML (Pretty): 831-1.0086935-rdf.xml
RDF/JSON: 831-1.0086935-rdf.json
Turtle: 831-1.0086935-turtle.txt
N-Triples: 831-1.0086935-rdf-ntriples.txt
Original Record: 831-1.0086935-source.json
Full Text
831-1.0086935-fulltext.txt
Citation
831-1.0086935.ris

Full Text

THE CHARACTERIZATION OF A LARGE CONDUCTANCE POTASSIUM CHANNEL IN CULTURED RAT MELANOTROPHS by KATHY WONG B.Sc, University of British Columbia, 1993 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Physiology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July 1995 © Kathy Wong, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of jjk^<m\^ tj The University of British Columbia Vancouver, Canada Date LaM* 7.7. h<r DE-6 (2/88) A B S T R A C T u The characteristics of a large conductance potassium (BK) channel found in cultured melanotrophs of the rat were studied. Single channel currents were recorded in cell-attached, inside-out and outside-out configurations using conventional patch clamp techniques. The current/voltage relationship was linear and revealed a single channel conductance of 261 ±4 pS (mean±S.E.M.) with a reversal potential of -3.2±0.9 mV (n=7) in symmetrical potassium solutions (150 mM). The BK channel was highly selective for potassium. In a physiological potassium solution, (1^=3.5, Nao=140, K;=150) the permeability ratio for sodium to potassium was <0.03. The permeability of the B K channel to ammonium ions and to the alkali metal ions, lithium, rubidium and cesium was determined using bi-ionic solutions with complete substitution of intracellular potassium by the test ions. The permeability sequence of these monovalent ions was K + (l)>Rb + (0.87)>NH 4 + (0.17)>>Na + (0.03)~Li + «Cs + . Internal cesium also caused an intermediate open channel block. B K channel gating was sensitive both to the intracellular calcium concentration and membrane potential. When open probability and membrane potential were fitted to the Boltzmann equation, the half-activation potential, representing the membrane potential where the open probability is half-maximal, was -72±7 mV, +39±19 mV and +104±14 mV in the 1 jiM (n=7), 0.5 yxM (n=5) and 0.1 uM (n=6) intracellular calcium solutions, respectively. This reveals high sensitivity to intracellular calcium between the concentration range of 0.1 to 1 JUM. The concentration of calcium required for a half-maximal open probability of the BK channel, at a membrane potential of 0 mV, was estimated to be 0.6 uM. Also from the Boltzmann fit, the voltage dependence was assessed from the values of the slope factor. The slope factor, which represents the change in membrane potential required for an e-fold change in open probability, was 12.3±1.4 mV, 13.6±2.6 mV, and 10.0±0.7 mV in the 1 juM, 0.5 uM and 0.1 JJM intracellular calcium solutions, respectively. Slope factors were not Ill significantly affected by changes in intracellular calcium concentration. Investigation of the effect of the potassium channel blocker tetraethylammonium (TEA + ) on outside-out patches revealed that exposure to external T E A + caused an intermediate open channel block. When fitted to the Hill equation the dissociation constant (K D ) , which represents the concentration of T E A + required for half-maximal block, was determined to be 0.25 mM, with 95% confidence limits (CL.) of 0.22-0.29 mM, at a membrane potential of 0 mV. The Hil l coefficient, h, which represents the number of molecules required to block the channel, was determined to be 0.81 (95% C L . 0.68-0.93), suggesting that a single molecule of T E A + was able to block the channel. When h was constrained to be one, the K D was 0.24 mM (95% C L . 0.21-0.27 mM). Internal T E A + caused a fast open channel block. The K D was 50.3 mM (95% C L . 44.8-56.0 mM) at a membrane potential of 0 mV. The Hill coefficient was 0.92 (95% C L . 0.8-1.0), again suggesting that a single molecule of T E A + is able to block the channel. When h was constrained to be one, the K D was 47.7 mM (95% C L . 44.6-50.6 mM). The effect of external 40 nM charybdotoxin was also examined on the B K channel in the outside-out configuration. The toxin caused a slow open channel block. In the rat melanotroph, the BK channel is likely to play a role in the repolarization of membrane potential following an action potential. By assisting in the repolarization of the membrane potential, the B K channel might decrease the calcium influx (required for secretion) through voltage-dependent calcium channels and thereby indirectly assist in the cessation of hormone secretion. iv T A B L E O F C O N T E N T S PAGE ABSTRACT ii TABLE OF CONTENTS iv LIST OF FIGURES..... viii LIST OF TABLES ix ACKNOWLEDGEMENTS x INTRODUCTION 1 1. ANATOMY OF THE PITUITARY 1 2. SECRETION FROM THE PARS INTERMEDIA 1 2.1 Melanotrophs as a Model for Studying Pituitary Physiology 1 2.2 Secretion in the Melanotrophs 2 Stimulus/Secretion Coupling Unstimulated "Basal" Secretion 3. ELECTROPHYSIOLOGY OF THE MELANOTROPH 4 3.1 Voltage-gated Calcium Channels .4 3.2 Voltage-gated Sodium Channels 5 3.3 Voltage-Gated Potassium Channels ; , 5 4. CALCIUM-SENSITIVE POTASSIUM CHANNELS 6 4.1 SK Channels 6 4.2 BK Channels .7 General Properties of the BK Channel Pore Structure of the BK Channel Function of the BK Channel 5. EXPERIMENTAL RATIONALE 9 METHODS AND MATERIALS 10 1. PREPARATION OF CULTURED CELLS 10 2. ELECTROPHYSIOLOGICAL EXPERIMENTS 11 2.1 Preparation of Electrodes .11 2.2 Formation of Different Types of Patch Configuration 12 2.3 Experimental Solutions 13 2.4 Data Acquisition and Analysis 16 Liquid Junction Potential Single Channel Amplitudes Open Probability 3. STAINING FOR a-MSH IN THE CELLS OF THE PARS INTERMEDIA 20 3.1 Fixation 20 3.2 Staining 20 3.3 Photomicrographs 21 RESULTS 22 1. IDENTIFICATION OF THE MELANOTROPH AND THE BK CHANNEL 22 2. PERMEATION PROPERTIES OF THE BK CHANNEL 22 2.1 Currents with Potassium as the Predominant Charge Carrier 22 I/V Relationship in Cell-attached Membrane Patches 1/V Relation with Excised Membrane Patches 1/V Relation with a Physiological Concentration Gradient of Potassium 2.2 Permeability Measurements in Bi-ionic Solutions 32 Rubidium Ammonium Lithium Cesium 3. GATING OF THE BK CHANNEL 3.1 Open Probability is Increased by Membrane Depolarization 3.2 Open Probability is Increased by Increases of Intracellular Calcium 4. PHARMACOLOGY OF THE BK CHANNEL 4.1 Open Channel Block 4.2 Tetraethylammonium Internal TEA + Caused a Fast, Open Channel Block External TEA + Caused an Intermediate, Open Channel Block 4.3 Charybdotoxin DISCUSSION 1. ION PERMEATION 1.1 Selectivity Filter 1.2 Permeation Pathway 1.3 Permeation Sequence of the Melanotroph BK channel Potassium is the Predominent Ion Traversing the BK Channel BK Channel Current Carriers: Potassium, Rubidium and Ammonium Cesium Ions are too Large to Permeate the BK Channel Lithium and Sodium Ions are too Small to Permeate the BK Channel 2. CHANNEL GATING. 2.1 Voltage Dependence .71 2.2 Calcium Dependence 2.3 Other Modulators of BK Channel Gating .74 3. BK CHANNEL BLOCKERS 3.1 Tetraethylammonium.. 3.2 Charybdotoxin .76 3.3 Significance of Blocking Agents 4. SPECULATIONS ON THE PHYSIOLOGICAL FUNCTION OF THE BK CHANNEL 4.1 BK Open Probability at Resting Conditions 4.2 BK Open Probability During an Action Potential BIBLIOGRAPHY viii LIST OF FIGURES FIGURE PAGE 1 Analysis of single channel current amplitude 18 2 Photomicrographs of cultured rat melanotrophs stained for the hormone a-MSH 23 3 The current/voltage relationship of the BK channel in the cell-attached configuration .27 4 The current/voltage relationship of the BK channel in symmetrical (150 mM) potassium solutions 29 5 The current/voltage relationship of the BK channel in a physiological concentration gradient of potassium 33 6 The BK channel is permeable to the monovalent cation rubidium 37 7 The BK channel is slightly permeable of ammonium 39 8 The BK channel has a low lithium permeability .40 9 The BK channel has a low cesium permeability 41 10 Internal cesium (150 mM) produces a flickering block of inward potassium currents .43 11 The gating of the BK channel is voltage sensitive 46 12 Channel gating is sensitive to intracellular calcium concentration 48 13 The BK channel is highly sensitive to calcium concentrations between 0.1-1 fiM 50 14 A composite graph of channel gating behaviour reveals the variability of the fitted parameters 52 15 Internal TEA + causes a fast block of the BK channel 57 16 External T E A + causes an intermediate block of the BK channel 59 17 External CTX causes a slow channel block of the BK channel 62 LIST OF TABLES TABLE 1 Composition of bath and pipette solutions 14 2 The Permeability of Monovalent Cations Through Potassium Channels .64 3 Comparison of BK Channels from Different Preparations 72 X ACKNOWLEDGEMENTS I wish to thank Dr. Steven J. Kehl for allowing me this opportunity to work in his laboratory. His patience, support, and advice during the writing of my thesis and throughout the course of the research project are greatly appreciated. I would also like to thank Dr. J. Ledsome for chairing my committee, and the rest of my thesis committee members, Dr. D. A. Mathers, Dr. P. Vaughan, Dr. K.G. Baimbridge and Dr. A. Pearson, for their advice and guidance. Special thanks to Dr. A. Pearson for helping me prepare for my seminars. Thanks to Dr. A. Buchan and Sue Shinn for their advice and assistance in the staining of the cells. A special thanks to Monika Grunert for her support and extreme patience during the endless hours of tissue culturing. Also thanks to all the members of the physiology staff, John Sanker, Joe Tay and Dave Phelan, for their help in caring for the laboratory animals and preparing slides and posters for all occasions. 1 I N T R O D U C T I O N 1. ANATOMY OF THE PITUITARY The pituitary gland lies at the base of the brain in the cavity of the sella tursica and is linked to the brain by the pituitary stalk. The gland consists of both a glandular portion (adenohypophysis) and a neural portion (neurohypophysis), each portion having a different embryonic origin. In most species, including the rat, the adenohypophysis is separated from the neurohypophysis by a thin multi-cellular partition. The cells of this partition are glandular and are collectively known as the pars intermedia (Howe, 1973). Nerve fibres descending from the hypothalamus course through the pituitary stalk and terminate in the neural and glandular portions of the gland. Fibres innervating the glandular tissue do not extend to the anterior lobe of the adenohypophysis but do terminate as "synapse-like" contacts near melanotrophs of the pars intermedia. 2. SECRETION FROM THE PARS INTERMEDIA 2.1 Melanotrophs as a Model for Studying Pituitary Physiology The melanotrophs are the hormone secreting cells of the pars intermedia and are a good model for studying pituitary physiology for several reasons. First, melanotrophs are the only secretory cells found in the pars intermedia and thus provide a homogenous population of a pituitary cell type. In culture, most of these cells remain spherical and provide an ideal shape for patch clamping in electrophysiological studies. These cells lack both axonic and dendritic processes, so that afferent signal processing and secretion occur in the same compartment. Since most patch clamping experiments are restricted to cell bodies due to technical limitations, information obtained from the melanotroph may reveal 2 information not available in other preparations where the recordings of electrical responses are remote from the site of secretion (i.e., the axon terminal). 2.2 Secretion in the Melanotrophs Stimulus/Secretion Coupling The stimulus-secretion coupling model pertains to many secretory cell types, such as chromaffin cells and various anterior pituitary cells (Douglas, 1968;1975; Schlegel et al., 1987; Law et al., 1989). In short, the model states that cell depolarization leads to activation of voltage-gated calcium channels, which ultimately leads to an increase in intracellular calcium concentration and hormone release. A similar mechanism for secretion likely applies to melanotrophs, where the influx of extracellular calcium through voltage-gated channels also plays a pivotal role in secretion (Thomas et al., 1990). Evidence for this has been provided by experiments showing a decrease in hormone release from preparations bathed in calcium-free external solutions (Tomiko et al., 1982) and an increase in hormone release from cells treated with the calcium channel agonist, BAY K 8644 (Taraskevich and Douglas, 1986;1989). Under physiological conditions, an increase in intracellular calcium is likely to accompany the firing of an action potential, when membrane potentials are able to activate the high-threshold voltage-activated (L-type) calcium channels. Furthermore, increases in hormone release have been noted to occur when melanotrophs are depolarized by extracellular potassium (Nemeth et al., 1989). Unstimulated "Basal" Secretion A unique secretory property of the melanotroph is its high basal secretory rate in vitro. Since the basal secretory rate is not effected by agents used to block action potentials 3 (Douglas and Taraskevich, 1982), it has been proposed that the high threshold calcium channels, which are typically activated only during an action potential, may not play an important role in basal secretion. Instead, this secretion is believed to be linked to the small voltage fluctuations observed around the resting membrane potential. The voltage fluctuations arise from calcium currents (Stack and Surprenant, 1991) carried by low threshold voltage-activated calcium channels (T-type), and this is known as the "spontaneous calcium entry theory" (Douglas and Shibuya, 1993). Isolated melanotrophs and isolated neurointermediate lobes have higher basal secretion than in the intact tissue because the intact gland is typically inhibited by signals from the hypothalamus. Several different neurotransmitters are involved with this inhibition, including GABA, Neuropeptide Y, and dopamine (Leenders et al., 1993; Valentijn et al., 1993; 1994). Each of these agents cause changes to cellular calcium concentration by affecting ion channels. For example, dopamine modulates multiple effectors which ultimately decreases the internal calcium concentration. The dopamine receptor (D2R), responds to the binding of the ligand by increasing the open probability of a voltage-insensitive potassium channel and thereby hyperpolarizing the cell by driving the membrane potential toward the potassium equilibrium potential, E^. The activation of this receptor also augments current passing through voltage-activated potassium channels by shifting the voltage dependence of the open probability of the channel and thus decreases the duration of the action potential. Recent data have also suggested a direct inhibitory effect on both the voltage-activated sodium and high threshold calcium channels. For a more detailed review on the multiple effects of dopamine on regulating intracellular calcium, see Valentijn et al. (1993). 4 3. ELECTROPHYSIOLOGY OF THE MELANOTROPH Electrophysiological studies performed on the melanotroph have been undertaken in part to gain insights on the coupling between electrical events and hormone secretion. Melanotrophs demonstrate spontaneous electrical activity, consisting both of subthreshold fluctuations of the resting membrane potential and the firing of action potentials (Nemeth et al. 1989; Stack and Surprenant, 1991). The action potentials have a width that is typically 4 ms at 0 mV and reach peak potentials of approximately 35 mV. These action potentials are then followed by a marked after-hyperpolarization. Underlying these electrical activities is a repertoire of voltage-gated ion channels selective for calcium, sodium or potassium ions (McBurney and Kehl, 1988). 3.1 Voltage-gated Calcium channels Melanotrophs express low and high threshold voltage-sensitive calcium channels (Cota, 1986; McBurney and Kehl, 1989; Williams et al. 1991). The high threshold channels are activated at membrane potentials above -10 mV (McBurney and Kehl, 1989). Under physiological conditions, these channels would most likely open only during the depolarizing phase of an action potential. These channels may therefore play a role in stimulus-secretion coupling of the cells since, as discussed above, the action potential triggers hormone secretion. The low threshold calcium channels are activated at membrane potentials of -50 mV (Kehl and McBurney, 1989; Williams et al, 1991). Since the resting membrane potentials spontaneously fluctuate between, -50 to -35 mV (McBurney and Kehl, 1989), calcium influx occurs through low threshold channels at resting membrane potentials. As mentioned above, these channels play an important role in the basal secretion of the melanotrophs. 5 Since calcium plays a central role in hormone secretion, studies of melanotroph electrophysiology have focussed in particular on calcium channels (Williams et al., 1991;1993; Nussinovich and Kleinhaus, 1992; Nemeth et al. 1989; Taraskevich and Douglas, 1986;1989; Stack and Surprenant, 1991). 3.2 Voltage-gated Sodium Channels The membrane also contains voltage-gated sodium channels. The sodium current has an activation threshold of around -35 mV and underlies the depolarizing phase of the action potential in the melanotroph. Its behaviour is qualitatively similar to that of the sodium current of nerve cells (Kehl, 1994). 3.3 Voltage-gated Potassium Channels Potassium channels contribute both to the resting membrane potential and the repolarizing phase of the action potential. From whole-cell studies in cultured melanotrophs (Kehl, 1989) or thin pituitary slices (Schneggenburger and Lopez-Barneo, 1992) , two potassium currents have been isolated; a slowly-inactivating outward current IK(s), and a fast-activating, fast-inactivating outward current, IK(f)-The IK(s) current is similar to the delayed rectifier current in other cells (McBurney and Kehl, 1989). It has an activation threshold between -30 to -10 mV N(Kehl, 1989) and therefore is not active at resting membrane potentials. Inactivation of this current is measurable only under experimental situations where the cell is depolarized for hundreds of milliseconds and would not occur, to any significant extent, under physiological conditions where action potentials typically last for 5 ms. This current plays an important part in the repolarizing phase of the action potential. Thus, when external T E A + is used to block this 6 current, the action potential duration is greatly increased (Kehl, 1990). The IK(f) has properties similar to those of A currents in other excitable cells. It has an activation threshold between -10 and -20 mV, and therefore would not be activated at resting potentials. The steady-state half-inactivation potential is between -20 to -30 mV, and therefore the majority of these channels are in the resting (versus inactivated) state at typical resting potentials. The biophysical properties of this current suggest that it acts to decrease the amplitude and duration of the action potential (Kehl, 1989; Schneggenburger and Lopez-Barneo, 1992). 4. CALCIUM-SENSITIVE POTASSIUM CHANNELS Calcium-sensitive potassium channels have been observed in a wide range of cell types (Latorre et al., 1989; Kolb, 1990). These channels have been categorized into at least two types based on their single channel conductances: a small conductance type (SK) and a large conductance type (BK). 4.1 SK channels Members of the small conductance type typically have single channel conductances of 4-80 pS (Latorre et al., 1989). Aside from the smaller conductance, this class of channels also has biophysical and pharmacological properties different from those of the large conductance type. SK channels are more sensitive to intracellular calcium than the BK channels (Blatz and Magleby, 1987) but have little or no voltage sensitivity in anterior pituitary cells (Lang and Ritchie, 1988), olfactory neurons (Maue and Dionne, 1987), rat skeletal muscle (Blatz and Magleby, 1986), and human red blood cells (Grygorzyk et al, 1984). While most of the SK channels studied have shown insensitivity to external 7 tetraethylammonium (Kolb, 1990; for exceptions see Latorre et al, 1989), the channels are typically blocked by the external application of apamin, a peptide extracted from bee venom (Blatz and Magleby, 1986; Lang and Ritchie, 1988; Kolb, 1990)! Due to the differences in both physiology and pharmacology of the S K and B K channels, it has been possible to study the two channels in isolation in most preparations (Latorre et al, 1989; Leinders et al, 1992) The S K channels of rat skeletal muscle has been studied in detail by Blatz and Magleby (1986). In that preparation, and several others (Kolb, 1990), the S K channel is responsible for the slow afterhyperpolarization following an action potential. 4.2 BK Channels General Properties of the BK channel Large conductance calcium-dependent potassium channels were first identified by Marty (1981) in bovine chromaffin cells and have subsequently been observed in diverse cell types (Blatz and Magleby, 1984; 1987; Kolb, 1990; Latorre et al. 1989). B K channels have been found, for example, in skeletal and smooth muscle cells (Blatz and Magleby, 1986; Miller et al, 1987; Wang and Mathers, 1993), neurons (Maue and Dionne, 1987; Reinhart et al., 1989), endocrine and exocrine cells (Atwater et al., 1979; Fenwick et al. 1982; Trautman and Marty, 1984; Squire and Peterson, 1987), epithelial cells (Guggino, 1986), and hair cells (Sugihara, 1994). These channels have three common characteristics. First, the single channel conductances are large, ranging from 120 pS to 300 pS in symmetrical potassium solutions (120 to 160 mM). Second, the gating of the channel is sensitive both to membrane potential and the intracellular calcium concentration. Third, B K channels are typically blocked by the pharmacological agents tetraethylammonium ( T E A + ) and charybdotoxin (CTX) . 8 Pore Structure of the BK channel The BK channel has the unique and seemingly paradoxical property of being able to pass large currents, greater than 108 ions/s, and exhibiting high ion selectivity (PNa/PK<.05). This has led investigators (e.g., Latorre et al., 1989) to propose that the pore is constructed in such a way that there are large, funnel-shaped vestibules on either side of a short and narrow tunnel. In this view, the channel resembles an hourglass. The large vestibules are believed to be lined with negative charges. The mouths of the channel therefore attract cations into the pore region and cause a local concentration of cations above that occurring in the bulk solution (Latorre et al., 1989). Evidence for the large vestibules lies in the observation that the open channel can be blocked by large organic ions such as TEA + . The vestibules are large enough to accommodate most hydrated cations and therefore do not discriminate between the different cations. The short and narrow tunnel connecting the vestibules contains the selectivity filter of the channel and is the rate-limiting step for conduction. Functions of the BK channel BK channels from different tissue preparations have shown wide variability both of their biophysical and pharmacological properties (Blatz and Magleby, 1987; Latorre et al, 1989). Perhaps reflecting these diverse properties, BK channels have been proposed to have different functions in the different types of tissue. In many cell types BK channels are believed to be involved with the repolarizing phase of the action potential. In endocrine cells such as those of the adrenal medulla and anterior pituitary, BK channels regulate secretion by affecting the duration of the action potential (Peterson and Maruyama, 1984; Douglas and Shibuya, 1993; Latorre et al., 1989). BK channels are activated in response to 9 influxes of calcium through voltage-gated calcium channels; the efflux of potassium then hyperpolarizes the cell and consequently prevents further calcium influx. Thus BK channels act indirectly to attenuate calcium-dependent hormone release. In other cell types such as the non-excitable epithelial cells of the kidney tubule (Gitter, 1987), the BK channel contributes to the coupling between sodium reabsorption and potassium secretion. And, in the epithelial cells of the choroid plexus (Christensen, 1987), BK channels are believed to assist in the secretion of potassium into cerebrospinal fluid. 5. EXPERIMENTAL RATIONALE To date, no studies of a large conductance potassium-selective channel in the rat melanotroph have been undertaken. The primary goal of the present study was therefore to determine the biophysical properties of the channel, including the sensitivity of the channel to membrane potential and to intracellular calcium concentration. Additionally, the selectivity of the pore of the channel for potassium ions over sodium ions and the monovalent cations, Li +,Cs +, Rb+ and NH 4 + was examined. To determine the pharmacological profile of the channel, its sensitivity to channel blockers, T E A + and CTX was also investigated. By characterizing the channel, its activity under physiological conditions, i.e., in the resting state and during an action potential, can be better predicted as can its role in the control of hormone secretion. 10 M E T H O D S A N D M A T E R I A L S I. PREPARATION OF CULTURED CELLS Experiments were performed on cultured melanotrophs isolated from the pars intermedia of the rat pituitary gland. Adult male Wistar rats (200-300 g) were rendered unconscious with C 0 2 gas prior to decapitation. Excision of the pituitary gland was performed by opening the skull, removing the brain, and exposing the pituitary gland. The entire gland was removed and placed into chilled C a 2 + - and Mg 2 +-free phosphate-buffered saline (CMF-PBS) . Subsequent manipulations were performed under sterile conditions. The pituitary tissue, usually from two rats, was first washed in C M F - P B S and then the neurointermediate lobe, consisting of the pars nervosa and the pars intermedia, was separated from the pars anterior. The neurointermediate lobe was placed in a 2 ml vial containing approximately 1 ml of C M F - P B S with the enzymes hyaluronidase (Type II, 1 mg/ml, Sigma, M O , USA) and collagenase (Type 1A, 1 mg/ml, Sigma, M O , USA) ; and incubated in a 37°C water bath for 50 minutes. A second incubation was then performed for 15 minutes in C M F - P B S solution containing the enzyme protease (Type VIII, 1 mg/ml, Sigma, M O , U S A ) . By the end of this second digestion, the tissue was ready to be mechanically dissociated to single cells. Mechanical separation was achieved by trituration, a process in which the tissue was drawn up and down needles of decreasing diameter (18, 21, 23, 25, and 26 gauge). The cell suspension was then layered onto a C M F - P B S solution containing 4% (w/v) bovine serum albumin (Sigma, M O , U S A ) and centrifuged at =*50 g for 10 minutes to form a loose pellet at the bottom of the microcentrifuge tube. Cells were resuspended in 1.2 ml of the culture medium consisting of a 1:1 mixture of Ham's F-12 and Dulbecco's Modified Eagle Medium 11 (DMEM) solution supplemented with 10% fetal bovine serum (Gibco, NY, USA), 2 mM glutamine (Sigma, MO, USA), insulin-transferrin-selenium (5 /Ltg/ml, 5 jLig/ml, 5 ng/ml, Sigma, MO, USA) and penicillin-streptomycin (100 units/ml, 100 units/ml, Gibco, NY, USA). A hundred microlitres of this suspension were then plated onto each of twelve collagen-coated Aclar (Proplastics, NJ, USA) coverslips along with 0.5 ml of the culture medium. Cultures were then kept in a humidified incubator containing 6% C0 2 at 37°C until use. The culture medium was first changed 5-7 days after plating, and thereafter every 2-3 days. Penicillin and streptomycin were present in the culture medium until the first change. Cytosine 6-D-Arabinofuranoside (Ara-C) was added to the culture solution 8-10 days after plating to arrest the mitotic division of pituicytes. 2. ELECTROPHYSIOLOGICAL EXPERIMENTS All experiments were performed at room temperature (=22 °C). The age of the culture when experiments were performed ranged from the day of the dissociation to 1 month after culture, most frequently between 1 to 3 weeks. The properties of the BK channel were not obviously affected by the age of the cultures but this was not systematically studied. 2.1 Preparation of Electrodes Glass tubing with outer diameter 1.2 mm and internal diameter 0.68 mm and containing a filament (Corning No. 7052, A-M Systems, WA, USA) were pulled on a two-step electrode puller (Narishige PP-83, Tokyo, Japan) to form electrodes with an outer tip diameter of 1-3 /xm. The shoulder of the electrode was then coated with Sylgard (Dow Corning, MI, USA), to within a few hundred micrometers of the tip, to decrease the capacitative coupling between the electrode and the bath solution. Just prior to use, the 12 electrodes were fire-polished to produce a smooth tip and optimize seal formation. Patch electrode resistance, measured in the high calcium "sealing" solution (see below), ranged between 5 to 20 megohms (MH) and was typically about 10 Mfl. Signals were referenced to an agar bridge containing 150 m M NaCl. 2.2 Formation of Different types of Patch Configuration Three different recording configurations were utilized during the course of the experiments: cell-attached, inside-out and outside-out patches. Tight seal formation was attempted in a sealing solution containing 128 m M N a + , 10 m M C a 2 + , 3.5 m M K + , 1 m M M g + , 10 m M N-[2-Hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] (HEPES) , 10 m M glucose and at a p H of 7.4. Aclar coverslips with adherent cells were transferred from the incubator to the recording chamber which was mounted on the stage of a phase contrast inverted microscope (Olympus LH50A, Japan). Using a micromanipulator system, an electrode was brought up to the cell and light suction was then applied to the electrode through a side port on the electrode holder. The formation of a tight seal was detected by a large decrease in current elicited by a 1 mV test command. The resistance under these conditions ranged from 1 to 10 gigohms. In this cell-attached, tight-seal configuration, single channel activity can be seen and has the advantage that the internal integrity of the cell is conserved. A disadvantage of this configuration is that neither the concentration of ions and other intracellular constituents nor the potential across the patch is known. For experiments where the inside-out configuration was required the pipette was pulled away from the cell taking with it a patch of membrane. This manoeuvre resulted in either the formation of an inside-out patch or, more often, the formation of a vesicle. Vesicle formation was detected as distortions of the single channel currents. When a vesicle was 13 present, brief exposure of the patch to air was sufficient to rupture the outer membrane and form an inside-out patch. For experiments where the outside-out configuration was required, strong suction was applied to rupture the patch of membrane in the electrode tip to form the whole-cell configuration. At this point the electrode was withdrawn to form an outside-out patch. 2.3 Experimental Solutions Tables IA and IB summarize the compositions of the bathing and pipette solutions, respectively. The pipette solution represents the external solution for inside-out and cell-attached experiments. Conversely, the pipette solution represents the cytoplasmic or internal solution in outside-out patches. In the figure legends, the solutions used in the experiments are presented as: internal solution /// external solution. Solutions are referred to by the code names assigned in Tables IA and IB. Ionized or free calcium concentrations were determined from an equilibrium between total calcium and ethyleneglycol-bis(6-aminoethyl ether)N,N,N',N'-tetraacetic acid (EGTA) as determined by the program, Max Chelator v.4.1T (Stanford University, CA, USA). A problem with this method of calculation was that small errors in measurement of either total calcium or EGTA can produce significant errors in the free calcium concentration. For a discussion on the imprecision inherent in the determination of the free calcium concentration see Pallota et al, (1992). Additional efforts were not made in our study to directly determine the free calcium concentration. When the expression "internal calcium concentration" is used, it refers to the free calcium level. Potassium hydroxide was used in most solutions to adjust the pH to 7.4, this potassium was accounted for as part of total potassium in the solution. In potassium-free solutions, 91 16 N-methyl-D-glutamate was used to adjust the pH. Chemicals were obtained from Sigma Chemical Co. (MO, USA) except for charybdotoxin which was purchased from Peninsula Laboratories (CA, USA). Bath solutions were changed using a gravity-fed perfusion system in which the different control and test solutions were contained in reservoirs above the level of the microscope stage. Silicon tubing carried the solutions from the reservoirs through a multiway valve to the bath inlet. After formation of a patch, the electrode was moved to the mouth of the inlet tube. The aim here was to facilitate rapid exchange of solutions. To maintain a more-or-less constant level in the bath, fluid was removed manually by applying light suction at the outlet at the opposite end of the bath. 2.4 Data Acquisition and Analysis Single channel currents were recorded on a List EPC-7 amplifier and filtered at a cut-off frequency (-3dB, 4-pole Bessel) of 2 kHz. Single channel currents were monitored on-line using a Hitachi Digital Oscilloscope and were digitized (VR-10 Instrutech, NY, USA) and stored on video cassette tape for off-line data analysis. By convention, inward membrane currents are shown as downward deflections from the 0 current level, while outward membrane currents were shown as upward deflections (regardless of patch configuration). Data acquisition and analysis were done by using pCLAMP (v.5.5 and v.6.0, Axon Instruments, Inc., CA, USA) controlling a Labmaster 12-bit anolog-to-digital converter (Scientific Solutions, OH, USA). Curve fitting and graphing was performed using Axum (Trimetrix, WA, USA). The goodness of the fit is represented by the co-efficient of determination (r2). 17 Liquid Junction potential Liquid junction potentials arise from differences in mobility of the ions in the patch solution and the bathing solution. The junction potential causes a small error, typically <5 mV, between the command potential and the actual pipette potential (Neher, 1992). Efforts were not made to correct for the junction potential. Single Channel Current Amplitudes All-points amplitude histograms were generated in pCLAMP to show the current amplitude distribution of the acquired data. Figure IA shows a distribution with three peaks representing the current level with none, one or two channels open. An events list was generated with a threshold for event detection set at 50% of the mean single open channel current amplitude. Single channel current events from a 5 second recording period (at 10 kHz) were then fitted to a Gaussian distribution. The modal values were taken from the fit and used as the single channel current amplitudes. Figure IB shows the current amplitude distribution of a patch clamped at -50 mV and fitted to a single Gaussian. The modal current for this patch is -14.4 pA. Open Probability Open probabilities were determined using pCLAMP. In the analysis, the user sets the criterion for the closed and open states; the time spent in each state is then tracked by the computer. The mean open probability (P0) of a channel in a patch was determined using the equation, 18 Figure 1. Analysis of single channel current amplitude. A) All-points amplitude histogram generated in pCLAMP. The three peaks represent the current amplitudes of the closed and two open states of a patch containing two BK channels voltage-clamped at +40 mV. Note that there was a leak current in the patch of approximately 4.5 pA which is detected as the shift in zero current level of the closed state. B) The current amplitude distribution of a patch, voltage-clamped at -50 mV. The distribution was fitted to a single Gaussian term with a modal value of -14.4 pA. 19 A l l - P o i n t s A m p l i t u d e H i s t o g r a m O o c D 0 o 1 o Q_ 6000 5000 4000 H 3000 2000 1000 1 I 10 15 20 Amp 1I t ude ( p A ) 25 30 B 1 20 100 c 3 CT 0> O cr iii a> o D -25 0 -20 0 15 0 -10 0 -5 0 A m p l i t u d e ( p A ) 20 p (Ti+T2+... + TN) where N is the number of channels in the patch, Tv T 2, T N are the times for which at least 1,2...N channels are open and T t o t is the sampling time. N was determined for a patch by using recording conditions that maximized the channel open probability. This was achieved by depolarizing the patch and exposing it to a high internal calcium solution. 3. STAINING FOR a-MSH IN THE CELLS OF THE PARS INTERMEDIA 3.1 Fixation Melanotrophs, on Aclar coverslips, were removed from the incubator and fixed in 2% paraformaldehyde in 0.1 M phosphate buffer solution (pH adjusted with NaOH to 7.4). The coverslips were placed in the fixative for ten minutes and then washed three times in phosphate-buffered saline (PBS) solution (five minutes per wash). 3.2 Staining Cells were stained with antibodies to a-MSH (A660/R2y, UCB Bioproducts, Ont, Canada) obtained from rabbits. Lyophilized pure serum was resuspended with distilled water; pure serum was then diluted 1:20 with a 0.05 M phosphate buffer solution (pH of 7.4) containing 1% BSA, 0.01% sodium azide and 0.3 mg/ml EDTA and then further diluted with PBS containing 0.1% Triton X to give final dilutions of 1:500, 1:1000 and 1:2000. Coverslips were incubated with the antibody overnight at 4° C. After the incubation, the coverslips were washed three times in PBS (five minutes per wash). A secondary stain was performed to localize the primary antibody. Peroxidase-21 conjugated AffiniPure Goat anti-Rabbit IgG (Bio/Can Scientific, Ont., Canada) was used at a dilution of 1:500 in PBS with 0.1% Triton X. Coverslips were incubated with the secondary antibody for one hour at room temperature and then washed three times in PBS solution (five minutes per wash). A diaminobenzidene (DAB, 0.25 mg/ml) / H 2 0 2 (0.001%) mixture was then added to the peroxidase-stained cells. Here, peroxidase activates the H 20 2 , which in turn activates the chromogen, diaminobenzidene. Coverslips were left in the mixture for approximately 10 minutes. Haematoxylin was used to stain the nuclei both of a-MSH containing and other cells. The coverslips were placed in haematoxylin for five minutes, washed in water for a minute, dipped in acid alcohol, washed in water for another minute and then placed in lithium carbonate for an additional minute. Coverslips were finally removed and placed face-down onto a glass slide with a drop of PBS/glycerine (1:9). 33 Photomicrograph Photomicrographs were taken on a phase contrast, inverted microscope (Zeiss, Germany). Cells containing a-MSH are dark brown; haematoxylin-stained nuclei are blue. 22 R E S U L T S 1. IDENTIFICATION OF THE MELANOTROPH AND THE BK CHANNEL All electrophysiological recordings were made on cultured cells from the pars intermedia of the rat, selected on the basis of their appearance. Cells chosen were spherical (10-20 /xm in diameter), had no processes, occurred singly or in clusters of up to 20 cells and were phase bright. Immunohistochemical staining demonstrated that cells with these characteristics contained the hormone a-MSH (Figure 2). Inside-out patches excised from these cells displayed at least three types of single channel currents in symmetrical (150 mM) potassium solutions. The focus of the work described here was on the channel population having the largest single channel conductance. This large or Big conductance potassium channel was easily identified by its much larger single channel conductance (=260 pS in symmetrical 150 mM K +), when compared to the other single channel conductances that were 1/10 to 1/2 as large, and is subsequently referred to as the BK channel. 2. PERMEATION PROPERTIES OF THE BK CHANNEL 2.1 Currents with Potassium as the Predominant Charge Carrier Current/Voltage (I/V) Relationship in Cell-attached Membrane Patches The I/V relationship was determined for the BK channel in the cell-attached configuration. In this configuration, single channel currents and channel gating are observed without (knowingly) disturbing the internal constituents. The single channel current (I) is determined by the equation, 23 Figure 2. Photomicrographs of cultured rat melanotrophs stained for the hormone a-MSH. Cells containing a-MSH appear dark brown. The nuclei of all cells were stained with haematoxylin and appear blue. Under low magnification (top), cells containing a-MSH are seen singly or in clusters of up to 20 cells. With higher magnification (bottom), the hormone containing cells are seen to have small nuclei. Cells not stained with a-MSH typically have much larger nuclei and are presumed to be the pituicytes. The cells shown here were one-week-old at the time of staining. 24 400 fim * . A 100 /u-rn 25 where g s c is the single channel conductance and E K is the reversal potential determined from the Nernst equation; V, the voltage across the patch of membrane, is equal to V R -V P where V R is the resting membrane potential and V P is the pipette voltage. When the open channel current is zero in the above equation, VR-Vp-E^is equal to zero (since g^O). In other words, in a graph of I vs -VP, the x-intercept is determined by the values for V R and E K . If potassium concentration is the same on both sides of the patch of membrane then EK=0 and at the zero current level VR=V P. To give an example, if V R is -40 mV then the zero current level would occur when VP=-40 mV or -VP=40 mV. For the graphs of Figure 3 the single channel current is plotted against -VP. In the experiments with cell-attached patches, the potassium concentration at the external face of the channel was 150 mM (solution PI in Table IB) and the intracellular potassium was unknown but assumed to be between 100-150 mM. E K was assumed to be near zero. Currents through cell-attached patches were measured under either of two conditions. In one condition (Figure 3A) the bathing solution contained 140 Naand 3.5 K to mimic physiological conditions (NaCl in Table 1A). As noted above, the potential across a patch under these conditions would be equal to V R -V P (EK=*0). The resting membrane potential was thus estimated to be -41 ±2 mV (n=6) from the interpolated or extrapolated zero current levels. In the other recording condition (Figure 3B), the bathing solution contained 150 mM potassium (150K in Table 1A) to depolarize the cell to a potential that depends on the value for internal potassium, but which is usually taken to be near 0 mV. Since the resting membrane potential is =»0 mV, the potential across the patch is approximately equal to the pipette potential but of opposite sign (V=-VP). 26 In either condition, the single channel slope conductance was less than that of BK channels in excised patches (see below). From linear regression routines, the single channel conductances were 194 ±7 pS (n=6) and 186 ±6 pS (n=7) in the NaCl and 150K solutions, respectively (Figure 3). For the majority of the patches obtained in the cell-attached configuration, open probability was quite low even when the patches were strongly depolarized (up to 120 mV). Only in rare cases were there detectible channel openings prior to depolarization. I/V Relation with Excised Membrane Patches The I/V relationship for the BK channel derived from recordings from inside-out patches (n=7) is shown in Figure 4. In symmetrical potassium solutions (150 mM), the I/V relationship was linear over the voltage range of -100 mV to +80mV. Using a linear regression routine, the mean slope conductance was determined to be 261 ± 4 pS (mean±S.E.M.) and the reversal potential was -3.2 ± 0.9 mV (Figure 4B). Figure 4A shows current traces from an inside-out patch containing two BK channels. The data were fitted to the Goldman-Hodgkin-Katz (GHK) equation, 7_ VF2(PXX!+PKK/-(PXX0+PKR'0) exp(-VF/RT)) RT (l-exp(- VFIRT) where I is the single channel current, V is the membrane voltage, F is Faraday's constant (96,480 C/mole), T is the temperature (295K for our calculations), R is the gas constant (8.314 VC/mole/K), PK is the permeability of potassium, P x is the permeability of the test ion (e.g., Na, Rb, NH4), K; and K^ are the activities (the product of the concentration and the activity coefficient) of potassium on the inside and outside, respectively, and X; and X^, 27 Figure 3. The current/voltage relationship of the BK channel in the cell-attached configuration. Potassium concentration was approximately symmetrical, with 150 K^ , and an assumed concentration of 100-150 mM potassium within the cell. A. The bathing solution contained 140 Na and 3.5 Kto mimic physiological conditions. For this figure, current along the y-axis refers to the single channel current amplitude recorded at the indicated negative pipette potential (-VP) along the x-axis. Note that the voltage across these patches was the sum of the -V P and the V R . Using a linear regression routine, the mean slope conductance was determined to be 194 ±7 pS (n=6) and the resting membrane potential was estimated from the zero current level to be -41 ±2 mV. B. The concentration of potassium in the bathing solution was increased to 150 mM by substituting KC1 for NaCl. Under these conditions the resting potential of the cell was =0 mV and the indicated negative pipette potential is the same as the voltage across the patch (V=-VP). The mean slope conductance was 186 ±6 pS (n=7). Symbols represent data from individual patches. 29 Figure 4. The current/voltage relationship of the BK channel in symmetrical (150 mM) potassium solutions. A. Current records from an inside-out patch containing two BK channels. The value for the holding potential is indicated to the left of each trace. Arrows mark the closed state. In this and subsequent figures inward (negative) currents are shown as downward deflections; outward (positive) currents appear as upward deflections. B. For this and subsequent figures, current, along the y-axis, refers to the single channel current amplitudes recorded at the indicated membrane potentials, along the x-axis. The current/voltage relationship was linear over the voltage range of -100 mV to +80 mV. Using a least squares linear regression routine, the mean slope conductance was determined to be 261±4 pS (n=7) and the reversal potential was -3.2±0.9 mV. Recording solutions from Table 1: 150K///P1 30 -80 mV I FT 31 are the activities of the test ion on the inside and outside, respectively. Activity coefficients were obtained from Robinson and Stokes (1965). In a symmetrical potassium solution (K i =K Q =[K] and with X i = X o = 0 ) , the G H K equation simplifies to, 7 VF*PK [K] RT In a graph of I versus V , the single channel conductance is equal to the slope ( F^P^K] / RT) and the zero current voltage (reversal potential) is 0 mV, regardless of the potassium activity ([K]) and permeability (P K). The permeability of the channel to potassium, P K , was determined from the G H K fit of the data of Figure 4 to be 6.2 x 10 1 3 cm3/s. I/V Relation with a Physiological Concentration Gradient of Potassium The I/V relationship was also studied in inside-out patches using solutions approximating physiological conditions, e.g., 1^=3.5 m M , Na o=140 m M and 1^=150 m M . Unlike the situation with symmetrical potassium, the I/V relationship was non-linear. Outward currents were carried by potassium; inward currents were not seen (Figure 5A). When the data (n=6 patches) were fitted to the G H K equation (Figure 5B, open circles), an extrapolated reversal potential of -82 mV was obtained. The permeabilities of the channel to potassium (PK) and sodium (P N a ) were determined to be 6.3 x 10"13 and 9.8 x 10"15 cm3/s, respectively. The permeability ratio of sodium to potassium is therefore < 0.03, indicating that the channel is highly selective for potassium ions. The permeability of the sodium ion was also examined under experimental conditions opposite of the one described above (1^=150, K;=3.5 and Na ;=140). In this situation 32 potassium is the predominant carrier of inward current and again sodium is not seen to carry (outward) current. These data (n=5 patches) were also fitted to the GHK equation (Figure 5B, filled circles) and the permeabilities of the channel to potassium and sodium were 4.9 x 10"13 and 9.8 x 1015, respectively. The permeability ratio of sodium to potassium is again <0.03. 2.2 Permeability Measurements in Bi-ionic Solutions To gain more insight into the nature of the channel pore, experiments were performed to quantitate the ability of other alkali metal ions and ammonium to permeate the BK channel. Single channel currents were recorded under bi-ionic solutions in which the intracellular potassium was replaced with an equimolar concentration of the test ion (150 mM) and the extracellular (patch pipette) solution contained 150 mM potassium. Rubidium The I/V relationship recorded when rubidium completely replaced internal potassium is shown in Figure 6. The I/V data were poorly fitted by the GHK equation. This is perhaps not surprising since there are assumptions, concerning the permeation pathway, underlying the derivation of the GHK equation that may not be valid for the BK channel. One such assumption is that ions can traverse a channel independently of other ions of the same or different species. This assumption has been shown to be invalid for some types of BK channel (see Discussion). The GHK equation was nevertheless used to fit the permeation data in absence of a more suitable alternative and since, with a few exceptions, the data fitted well to the GHK equation. And, as noted by Hille (1975), the GHK theory has the important advantage that it summarizes measurements in terms of a single 33 Figure 5. The current/voltage relationship of the B K channel in a physiological concentration gradient of potassium (Kj=150 m M , K„=3.5 m M , Na o=140 m M ) and in a high internal sodium solution (1^=150 m M , Kj=3.5, Na—140 mM) . A . Single channel currents from an inside-out patch containing 2 B K channels, recorded in the physiological saline solution. Outward currents (upward deflections) were carried by potassium and no inward currents were seen. Arrows mark the closed state. B. In contrast to the situation with symmetrical potassium, the current/voltage relationships in the physiological saline solution ( o ) and the high internal sodium solution (•) were non-linear. The stippled line represents the current/voltage relationship in symmetrical 150 m M potassium solution. The data were fitted to the G H K equation (n=7, r2=0.98 and n=6, r 2=.99, respectively). The extrapolated reversal potentials were -82 and 77 mV, respectively. The absolute permeabilities for potassium and sodium were 6.3 x 10"13 and 9.8 x 10 1 5 cm3/s, for the physiological saline solution and 4.9 x 10 1 3 and 9.8 x 10"15 cm3/s, for the high internal sodium solution, respectively. From both fits, the permeability ratio of sodium to potassium was determined to be <0.03. This indicates that the channel is highly selective for potassium over sodium. Solutions: for (o) 150K///P3 and for (•) NaCl / / /P l 35 parameter, absolute permeability or permeability ratios. Barrier models based on Eyring rate theory can provide a more complete description of ion permeation through ion channels (e.g. Yellen, 1984a) but are much more complex and involve a large number of free parameters (e.g., French et al., 1994, where one model had 24 parameters). For rubidium, when both outward and inward current values were used, the reversal potential obtained from the GHK fit (=20 mV) was clearly an overestimation (Figure 6A). The GHK equation was then used to fit only the inward currents. This second fit was still unable to accurately represent the entire I/V data, but gave a more reasonable estimate of the reversal potential (Figure 6B). The reversal potential obtained from the second fit was 4 mV and the absolute permeabilities of the channel to potassium (P K ) and rubidium (PRE) were 5.5 x 10~13 and 4.8 x 1013 cm3/s, respectively. This gave a permeability ratio of rubidium to potassium of 0.87. It was also observed in these experiments, that although the channel was highly permeable to rubidium, rubidium did not carry large outward currents. Using a linear regression routine, the slope conductance for outward rubidium current was determined to be «40pS. This gave a conductance ratio of rubidium to potassium (gRb/gK) of 0.16 which was much less then the permeability ratio. Ammonium Internal potassium was replaced with the monovalent cation ammonium. Inward currents were carried by potassium, while outward currents were carried by ammonium. The current/voltage relationship was fitted to the GHK equation (n=6) (Figure 7). The reversal potential was determined to be 45 mV and the absolute permeabilities of potassium and ammonium ( P N H 4 ) were 5.1 x 10"13 and 8.5 x 10"14 cm3/s, respectively. P N H 4/PR W A S 0-17-The channel is therefore permeable to ammonium, but to a lesser extent than to potassium 36 or rubidium. Using a linear regression routine, the slope conductance for the outward ammonium current was approximately 80 pS. This gave a conductance ratio of NH 4 /K of 0.4. Although the channel is more permeable to rubidium than it is to ammonium, the ammonium conductance is greater than that of rubidium. Lithium When lithium was the major internal cation, only inward potassium currents were seen. Lithium did not carry outward currents. The I/V relationship (n=6) was recorded and fitted to the GHK equation (Figure 8). The reversal potential derived from the fitted curve was 60 mV and the absolute permeabilities of the channel to potassium and lithium were 5.1 x 1013 and 4.6 x 10 1 4 cm3/s, respectively, with a PLi/PK of 0.09. Although the reversal potential was estimated to be 60 mV, outward currents were not seen at positive membrane potentials up to +100 mV. The fitted value for the reversal potential might be an underestimate. Since outward currents were not seen at a membrane potentials up to +100 mV, the PLi/PK and/or the gu/gK r a u o f ° r the BK channel is very low. Cesium When cesium was used as the major cation in the internal solution, outward currents were again not seen. The I/V relationship (n=6) was recorded and fitted to the GHK equation (Figure 9). A reversal potential was extrapolated from the fitted curve to be 50 mV and the absolute permeabilities for potassium and cesium were 4.77 x 10 1 3 and 7 x 1014 cm3/s, respectively. The Cs/K permeability ratio was 0.15. Although this implies that the cesium permeability is similar to that of ammonium, it is unlikely to be the case. First, unlike ammonium (Figure 7), outward currents were not seen with cesium at potentials as 37 Figure 6. The BK channel is permeable to the monovalent cation rubidium. In a bi-ionic solution (1^ =150 mM, Rb;=150 mM), the inward currents are carried by potassium, while the outward currents are carried by rubidium. Inside-out patches (n=5) were used to record the single channel currents at the potentials indicated. A. The current/voltage data did not fit well to the GHK equation (see text). B. The reversal potential was therefore estimated by fitting only the inward currents (1^ =0.98), and was determined to be 4 mV. The second fit was still not able to accurately represent the entire range of the current/voltage data, but did give a more reasonable estimation of the reversal potential. The absolute permeabilities of Rb + and K + were determined from the latter fit to be 4.8 x IO"13 and 5.5 x IO13 cm3/s, respectively. The Rb +/K + permeability ratio is 0.87. Notice that although the channel is highly permeable to rubidium, the outward rubidium currents are much smaller than the inward potassium currents with an equivalent driving force; the estimated Rb/K conductance ratio (see text) is only 0.16. Solutions: RbCl///Pl 39 Figure 7. The BK channel is slightly permeable to ammonium. Inside-out patches (n=8) were placed in bi-ionic solutions with 150 mM KC1 on the external face and 150 mM NH4C1 on the cytoplasmic face of the membrane. Under these conditions, inward currents were carried by potassium and outward currents were carried by ammonium. The data were fitted to the GHK equation (^ =.86) and the reversal potential was determined to be 45 mV. The absolute permeabilities of NH 4 + and K + are 8.5 x 1044 and 5.1 x 10"13 envy's, respectively. The NH 4 + /K + permeability ratio is 0.17. Solutions: NH4C1///P1 40 10r Figure 8. The B K channel has a low lithium permeability. Inside-out patches (n=6) recorded in bi-ionic solutions (1^=150 m M , Li ; = 150 mM) displayed inward currents carried by potassium, but no outward currents. The data were fitted to the G H K equation (r^O.95) and the reversal potential was determined to be 60 mV. The absolute permeabilities are 4.6 x 10 u and 5.1 x 10"13 cm3/s for L i + and K + , respectively and the L i + / K + permeability ratio is 0.09. Although the reversal potential from the fit to the G H K equation was 60 mV, outward currents were not seen at potentials above the reversal potential, as was the case with ammonium. Patches were clamped at voltages up to +100 mV without observing outward current. Solutions: L i C l / / / P l 41 Figure 9. The BK channel has a low cesium permeability. Recordings were made from inside-out patches (n=6) in bi-ionic solutions (1^ = 150 mM, Cs;=150 mM). Inward but not outward currents were seen. Internal cesium caused an intermediate open channel block that decreased the potassium current amplitudes; this was apparent in the all-points histogram (not shown) as a pronounced skewing of the amplitude distribution. A. Single channel currents are shown for a patch containing a single BK channel. The inward currents (downward deflections) were carried by potassium. Notice that outward currents were not seen at a membrane potentials above 50 mV. Arrows mark the closed state. B. The data were fitted to the GHK equation (^ =0.90) and a reversal potential of 50 mV was obtained. Such a reversal potential predicts outward currents comparable to the ones observed for ammonium (Figure 7). However reversal currents were not seen with cesium at potentials up to +100 mV which suggests that cesium is not able to permeate the channel. Solutions: CsCl///Pl -80 mV -50 mV -20 mV 20 mV 50 mV 80 mV 20 mS B io 10 pA -120 120 Voltage (mV) 43 L i C I 20 ms 2.5 pA Figure 10. Internal 150 mM cesium produces a flickering block of inward potassium currents. The block, which is also evident in Figure 9 (especially at 20 mV), is evident here in the traces obtained after the internal (bath) solution was switched from one containing 150 mM L i + as the predominant cation to one containing 150 mM Cs+. The external solution contained 150 mM K + and the patch was held at 0 mV. Solutions: LiCl///Pl or CsCl///Pl 44 high as +100 mV. Second, as discussed below, cesium has a blocking effect on inward potassium currents. The latter effect could lead to a reduction of potassium currents at positive membrane potentials (particularly if the cesium block has any significant voltage dependence) and an underestimate of the extrapolated reversal potential. Internal cesium caused a flickery block of the inward potassium current (Figure 9A and 10). The block was similar to the intermediate open channel block described for external T E A + (see below) and was more apparent at more positive membrane potentials, suggesting that the block might have some voltage dependence. As noted earlier, the GHK equation assumes independent ion movement and will not account for such a channel block. 3. GATING PROPERTIES OF THE BK CHANNEL 3.1 Open Probability is Increased by Membrane Depolarization Changes in membrane voltage greatly affected the gating of the BK channel. Figure 11A shows an inside-out patch containing a single BK channel, as the membrane potential was altered from -50 mV to 50 mV. The internal calcium concentration was 0.5 /xM. When fitted to the Boltzmann equation, P p _ max V - V l +e*p(-*—) where P0 is the open probability, Pmaxis the maximum open probability, V is the membrane voltage, Vy2 is the voltage at which the open probability is half-maximal, and S is the slope factor, the Pm a x was determined to be 0.98, Vy2 was -13.1 mV, and S was 13.1 mV (Figure 11B). The slope factor represents the change in membrane potential required for an e-fold change in open probability. The gating of all the BK channels tested demonstrated a similar 45 sensitivity to membrane voltage, represented by the value S, regardless of the value for V/2 (see below). 3.2 Open Probability is Increased by Increases in Intracellular Calcium Changes in intracellular calcium concentration also affected the gating of the BK channel. Figure 12 shows the open probability of a channel in an inside-out patch recorded in solutions with calcium concentrations ranging from 0.01 to 100 fiM. The channel is most sensitive to changes in intracellular calcium concentration in the range of 0.1 to 1 /i,M (Figures 12A and 13). The data relating the open probability to the calcium concentration were fitted to the equation, i i * = J y In ( - * * - ) S [Ca2+] (Wong et ai, 1982) where, V 1 / 2 is the half-activation potential, S is the slope factor, K D is the dissociation constant, [Ca2+] is the concentration of internal calcium and N represents the number of calcium binding sites on the channel (Figure 12B). Values for V 1 / 2 and S were obtained from the fits of the Boltzmann equation to the data of Figure 14. The K D , which represents the concentration of calcium required for a half-maximal open probability of the BK channel at a membrane potential of 0 mV, was 0.6 /xM with 95% confidence limits (95% CL.) of 0.25-0.93 JUM. The value for JV was determined to be 4 (95% C L . 3-6) and reflects the steepness of the calcium dependence of the gating. Although all of the BK channels demonstrated sensitivity to intracellular calcium concentration, individual patches showed differences in open probability at a given calcium 46 Figure 11. The gating of the BK channel is voltage sensitive. Recordings were made in symmetrical potassium (150 mM) solutions with 0.5 /iM internal free calcium. A. Currents from an inside-out patch containing a single BK channel. The value for the holding potential is indicated to the left of each trace. Notice that at -50 mV the channel is more often closed, whereas at 50 mV the channel is open most of the time. Arrows mark the closed state. B. The open probability is a function of the membrane voltage. Data fitted to the Boltzmann equation (Pmax=0.98, V% =-13.1 mV). The slope factor, S, which represents the change in membrane potential required for an e-fold change in the open probability, was 13.1 mV. Solutions: 150K//P1 -50 mV -20 mV 20 mV 50 mV B 1.00 0 . 7 5 p 0 . 5 0 h 20 ms 10 pA 0 . 2 5 0 . 0 0 - 8 0 - 4 0 0 4 0 8 0 Vo l tage (mV) 48 Figure 12. Channel gating is sensitive to the intracellular calcium concentration. A. A patch containing 2 BK channels was depolarized in recording solutions containing 0.01 /xM (•), 0.1/xM (•), 1/xM (T) and 100/iM (•) internal calcium. All four sets of data were fitted to Boltzmann equations (Pmax=1.04, 0.97, 0.95, 0.96; Vy,=76.1, 59.6, -90.5, -99.5; S=15.2,12.3, 9.2, 11.0; for the 0.01 yuM, 0.1 /i,M, 1 uM and 100 /zM internal calcium solutions, respectively). B. The data concerning the sensitivity of BK channel to internal calcium were fitted to the equation from Wong et al, (1982)(r2=0.56). From the equation the K D , which represents the concentration of calcium required for a half-maximal open probability of the channel at 0 mV, was 0.6 /zM and TV, which represents the number of calcium binding sites on the channel, was determined to be 4 (95% C L . 3-6). Solutions: 150K///P1 49 50 Figure 13. The BK channel is highly sensitive to calcium concentrations between 0.1 and 1 fiM. A . A patch containing a single BK channel was placed into recording solutions containing 0.1 /xM ( • ) , 0.5 ttM ( A ) and 1 /JM (•) calcium. The membrane voltage was maintained at -30 mV. The open probability is much higher in the 1 /xM solution than in the 0.1 /zM solution. Arrows mark the closed state. B. The changes in open probability are shown for the same channel as the membrane was depolarized in the three calcium solutions. Data were fitted to Boltzmann equations (Pmax=0.84, 0.98, 0.92; V%= 125.2, -13.2, -69.0; S=7.0, 13.4, 11.9; for the 0.1 ,0.5 and 1 /xM calcium solutions, respectively). Solutions: 150K///P1 51 A 0.1 /xM Voltage (mV) 52 Figure 14. A composite graph of channel gating behavior reveals the variability of the fitted parameters. Although the responses to membrane voltage and [Ca2+]; of all of the channels studied were qualitatively similar, there were, as shown here, quantitative differences of the half-activation potentials. The mean half-activation potentials in A (1 fiM Ca2+), B (0.5 fiM. Ca2+) and C (0.1 fiM Ca2+) were -72±7 mV, 39 ±19 mV and 104 ±14 mV, respectively. There was no significant difference (p > .05) between the mean slope factors in A(12.3±1.4 mV), B(13.6±2.6 mV) and C(10.0±0.7 mV). Symbols represent data from individual patches. Solutions: 150K///P1 53 1.00 [ -100 0 100 200 Voltage (mV) 54 concentration and voltage. Figure 14 demonstrates the variability between the patches under recording conditions presumed to be identical. 4. PHARMACOLOGY OF THE BK CHANNEL 4.1 Open Channel Block Open channel blockers are defined as agents that decrease membrane current but which bind to the channel only when it is in the open state. Open channel blockers decrease current flow by physically blocking and completely occluding the channel pore. Consequently, channels can be either in a conducting state or a non-conducting (blocked) state. In single channel recordings open channel block can be categorized, for convenience, as one of three types. The block can appear as a decrease in single channel current amplitude, as a decrease in single channel current amplitude associated with flickering or as a decrease in open probability. The explanation for the appearance of three types of block lies in the blocking kinetics. If the block is fast, on-rates and off-rates for the channel binding are fast, and the frequency of the blocked-unblocked transitions exceeds the frequency response of the system. In other words, the system is not able to resolve individual blocking events. The signal that is recorded represents the time-averaged amplitude of the single channel current. If the block is of an intermediate speed, the average duration of the blocked state is sufficiently long that channel blocking events can be partially detected. Thus for an intermediate block the changes in states are slow enough to be detected but are still too fast to be fully resolved, e.g. the current does not reach the zero level before unblocking occurs. For slow blockers, the dwell time of .the blocker is long enough that blocking events can be distinguished by the recording system as individual openings and closings. This block is 55 measured as a decrease in the open probability; the amplitude of the single channel current is not changed. Examples of all three blocking modes were obtained with the pharmacological agents used to characterize the B K channel of the rat melanotroph. 4.2 Tetraethylammonium Internal TEA + Caused a Fast, Open Channel Block T E A + is a well-known potassium channel blocker and has been extensively used in the characterization of delayed rectifier and B K channels (Latorre et al, 1989). The effects of the application of millimolar concentrations of TEA + to the internal face of the B K channel were examined first. Since external potassium has been reported to affect the blocking action of internal TEA + (Armstrong, 1975), these experiments were performed in the absence of external potassium (Nao=140 mM, 1^ =110 mM). Shown in Figure 15 are single channel currents recorded at 0 mV, in solutions containing 0, 10, 20 and 40 mM TEA + . Internal T E A + caused a concentration-dependent, reversible reduction of the single channel current amplitude. Unlike the situation with external T E A + (see below), the decrease in amplitude was not associated with flickering of the channel current. The data concerning the internal TEA + block were fitted to the Hill equation, / . i norm i + ( 1ZMV where I n o r m is the normalized current (IxEA/Icontroi)) K D is the dissociation constant, [TEA+] is the concentration of TEA + , and h is the Hill coefficient. The concentration of TEA + producing a half-maximal block, K D , was determined from a non-linear least squares fitting routine to be 50.3 mM (95% CL. 44.8-56 mM) (n=4). The Hill coefficient, h, which 56 represents the number of TEA + binding sites on the channel, was determined to be 0.92 (95% C L . 0.8-1.0). This suggests that a single molecule of T E A + is able to block the channel. When the data were fitted to the Hill equation under the constraint that h was equal to one, the K D was 47.7 mM (95% CL. 44.6-50.6 mM). The voltage dependence of the block by internal TEA + was not studied systematically but appeared to be weak. External TEA* Caused an Intermediate, Open Channel Block The effects of external TEA + were examined in outside-out patches recorded in a physiological potassium concentration gradient with an internal calcium concentration of 1 /xM. Figure 16A shows current traces derived from a patch containing a single BK channel recorded at 0 mV, in TEA + concentrations ranging from 50 to 800 /xM. Application of T E A + caused a concentration-dependent, reversible decrease of single channel currents. The concentration of TEA + required for a half-maximal block (KD), was determined to be 0.25 mM (95% C L . 0.22- 0.29 mM) (n=6) and the Hill coefficient (h) was 0.81 (95% C L . 0.68-0.93) (Figure 16B). As with internal TEA + , this suggests that a single molecule of TEA + is able to block the channel. When h was constrained to be one, the K D was 0.24 mM (95% C L . 0.21-.27 mM) The block by external TEA + was weakly voltage-dependent (data not shown). 4.3 Effects of the BK channel Blocker, Charybdotoxin (CTX), on the BK Channel We encountered considerable difficulty in trying to maintain steady (stationary) open probabilities when working with outside-out patches. In this configuration the channel would frequently enter long-lived closed states. In some patches, the channel would return from the closed states after several minutes and in other patches the channel appeared to 57 Figure 15. Internal TEA + caused a fast block of the BK channel. Current traces show outward membrane currents recorded from an inside-out patch held at 0 mV. The block appears as a decrease in the single channel amplitude. When the data from four patches were fitted to the Hill equation (r2=.99), the K D was determined to be 50.3 mM. The Hill coefficient, h, was determined to be 0.92, suggesting that a single molecule of T E A + is able to block the channel. When h was constrained to be one in the fitting routine (not shown), the K D was 47.7 mM. Solutions: TEAj///P4 58 control Ai iff 10 mM T E A 20 mM T E A + 40 mM T E A + 5 pA 10 ms a IH t-l O T3 N •rH •e t-l o 1.0 r 0.8 0.6 0.4 0.2 0.0 1 1 1 • ' 1 i 111—,—i i 1111111,1 i i i 111 , i i 11 1 111 1 11 i i i 111 10 100 1000 T E A + concentration (in mM) 59 Figure 16. External TEA + caused an intermediate block of the BK channel current. A. Currents recorded in a physiological potassium concentration gradient and at a membrane potential of 0 mV. Single channel currents of an outside-out patch in solutions containing 0, 0.2, 0.4, or 0.8 mM TEA + . Arrows mark the closed state. The actions of T E A + were correlated with a flickery block. B. From the Hill plot (r2=.99), the K D was determined to be 0.25 mM and the Hill coefficient was 0.81. This suggests that a single molecule of T E A + is able to block the channel. When the data were fitted to the Hill equation with the constraint that h was one (not shown), the K D was 0.24 mM. Solutions: P2///TEA,, Q I i i . i 11 .1—i i i i i 11 i i i 111,1 i i i i 111 • i . I . I . I I , t , i , , ,1 i i i i 111 .001 .01 .1 1 10 TEA concentration (mM) 61 enter and stay in the closed state. The basis for this instability in channel gating in outside-out patches is not known. Inside-out patches, which are also excised patches, do not display the same problem which suggests that the problem is not simply related to the wash-out of intracellular components. These difficulties did not affect the analysis of the effects of external T E A + data since the block was measured as a decrease in single channel current amplitude. The non-stationary behaviour of the open probability in the outside-out patches did however make quantitative analysis of CTX data impossible. CTX binds to an external site on the BK channel and because of its long dwell time in the channel decreases the open probability, i.e. it is a slow blocker. Therefore in the absence of a uniform control open probability, a decrease in open probability in the presence of the blocker could often not be unequivocally attributed either to spontaneous closure or to block by CTX. The analysis of the effects of CTX was necessarily qualitative and limited to a few patches that did have a relatively stable open probability. The effects of external CTX on the BK current were examined on outside-out patches. A CTX concentration of 40 nM was chosen which, based on the results of others (Miller et al, 1985), is well above the K D for the block of BK channels. Physiological potassium solutions were used in these experiments with 1^ =3.5, NaQ= 140 and 150 mM. As reported by others (Miller et al, 1985), CTX appeared to act as a slow open channel blocker. Figure 17 shows a single patch with four BK channels. The open probability decreased from 0.48 to 0.01 upon exposure to 40 nM CTX at a membrane potential of 0 mV. In this patch, near complete recovery to the control PQ was obtained after returning to CTX-free medium. Similar results were obtained from the experiments performed on two other patches. 62 200 ms 10 pA Figure 17. External CTX caused a slow channel block. Current traces show an outside-out patch containing 4 BK channels recorded in a physiological potassium concentration gradient and at a patch membrane potential of 0 mV. Arrows mark the closed state. The open probability was initially 0.46. The channel blocker CTX at a concentration of 40 nM, decreased the open probability to 0.01 but did not affect the single channel current amplitude. The block was reversible and the open probability returned to 0.36 when the patch was returned to the control solution. Solutions: P2///CTX 63 DISCUSSION 1. ION PERMEATION The BK channel is a transmembrane protein that provides an aqueous route for the selective movement of ions across the membrane. The selectivity is conferred by a narrow region of the channel pore which filters oncoming ions and hence is referred to as the selectivity filter. 1.1 Selectivity Filter The mouths of the BK channel pore are believed to be large and therefore do not confer a size constraint on the alkali metal ions and NH 4 ions entering the pore. For these regions, the selectivity sequence is the same as the mobility sequence in water. For cations the sequence is Cs>Rb>NH4=K>Na>Li with a ratio of Cs to Li of 2 (Hille, 1975). However, potassium channels, including the BK channel, have selectivity sequences that differ from the mobility sequence. It is believed that the diameter of the pore at the selectivity filter must be smaller than 3A (Bezanilla and Armstrong, 1972; Hille, 1973) to account for the observed differential constraints on the flux of hydrated ions of similar charge and sign. Furthermore, although the narrow region acts as a sieve that excludes some ions, the selection process is not strictly related to the size of the ion. Permeation also depends on the ability of the hydrated ion species to interact in a energetically-favourable way with the pore wall (Eisenman and Horn, 1983; Eisenman, 1987). If selectivity filters discriminated simply according to size, it would not be possible to account for a pore selective for potassium over sodium, since potassium is the larger ion (Table 2). Both potassium and sodium ions are however hydrated in aqueous solutions and < o cr cr cr cr cr 2? S o o o o o 2 ~ 3 3 3 3 3 | g <» o. 2. K 3 £ ° £ ^ 1 B 5 * a \ 0 3 (6 O O 3 3 O 5 * 8 a r s. 5 £ - H H * 5. o 5' 3 A o o un A o o UJ A o o un A o o Un A o o un UJ o UJ o © c c r s* H* 3 vo ON o © o ^0 o o VO 00 -4 EL £' B N> 00 un UJ H 13 o P NJ ON ON o D-5° 3 VO A o o un A o o UJ A o o un A o o un A o o un IA o © UJ c 3 A o o un A o o UJ A o o un IA o VO 00 -a c o s 50 O co co a sr co o. H S o n ? c& C 00 1° o (5 03 C 00 2.9 o O ST CO — 03 co n 133 2. s 5d * O £0 g r . 3 -§ a. h3 W r o £ ' o z s z H O o r Ni fr9 65 neither ion in its fully hydrated form is able to pass through the narrow region without interacting with the pore wall (Begenisich, 1987;1992). It has been suggested (Hille, 1975) that the pores of ion channels are lined with carboxyl groups, possibly arranged in a ring structure, the oxygen component of which is able to interact with the cations. The interaction between the ion and the carboxyl oxygens makes it energetically favorable for an ion to loose some of the water molecules of its hydration shell and reduce its effective size. This interaction with the pore wall underlies the catalytic effect of the channel on ion transport. In potassium channels, the interaction between the pore wall and the-potassium ion is much more favourable than with sodium, presumably because the sodium ions are too small to interact effectively with the ring of carboxyl groups. Thus, in a potassium channel pore, being too small, as with sodium and lithium ions, is as detrimental as being too big, as appears to be the case with cesium ions (Table 2). 1.2 Permeation Pathway Ion permeation is a concept that is not clearly understood. Many theories attempt to explain this concept (Wagoner and Oxford, 1987; Lester, 1991). The Eyring rate theory has been successful at explaining most of the data collected on permeability. Eyring (1949) viewed the permeation pathway as a series of energy barriers and wells. By definition, a permeating ion must be able to pass over all the barriers to exit the other side of the membrane. In this theory, interactions of the ion with the pore wall would be considered energy barriers. Since the potassium conductance of the BK channel is high, the durations of the interactions (binding and unbinding) between a permeant ion and the pore wall must be brief. That is to say the energy changes must be quite small throughout the channel to 66 accommodate the high flux rates. 1.3 Permeation Sequence of the Melanotroph BK channel Aside from potassium, ammonium and rubidium ions were also able to carry currents through the BK channel of the rat melanotroph. Sodium, lithium and cesium did not carry currents under the recording conditions. The permeation sequence, K>Rb>NH4>Na,Li,Cs, is similar to that of most BK channels examined (Table 2) and to most potassium channels in general (Latorre et al., 1989; for exceptions see, Kolb, 1990). For example, the permeability sequences for the BK channels in skeletal cells (Eisenman et-al., 1986) and chromaffin cells (Yellen, 1984a) are Tl>K>Rb>NH4>>Na, Li, Cs and K>Rb>>Na, Cs, respectively. The similarity in permeation sequences between the potassium channels suggests that the channel pore is a highly conserved structure. Potassium is the Predominent Ion Traversing the BK Channel All permeant ions met the size requirement for potassium pores as suggested by Hille (1975) and have ionic diameters between 2.66 to 3.0 A (Table 2). Of the six cations tested, the BK channel was most permeable to potassium ions. This is also the predominant ion traversing the channel under physiological conditions. In symmetrical potassium solutions (150 mM), the single channel conductance was 261 pS. However, the single channel conductance was smaller (186-194 pS), when it was studied in the cell-attached configuration. There are several possible reasons for the smaller single channel conductance in the cell-attached patches. First, the potassium concentration across the membrane may not be symmetrical. The internal potassium concentration may have been much lower than 150 mM; internal potassium concentration is less than 100 mM (Lee and Fozzard 1975) in 67 heart muscle cells. A similar decrease in single channel conductance has been observed for the cell-attached to 6-escin induced "open" cell-attached patches in BK channels of smooth muscle cells (Muraki et al., 1992) and was attributed to a low intracellular potassium concentration. Another factor that may account for the smaller conductance in cell-attached recordings is the possibility that there are fast channel blocking agents at the cytoplasmic face of the membrane. Internal sodium, for example, has been shown to block the BK channel in the chromaffin cells (Yellen, 1984a). Arguing against this possibility is the fact that high external potassium, as was used in our experiments, is able to relieve the block by internal sodium (Yellen, 1984b and see below). The existence of an unknown, fast blocking agent within the cytoplasm of the cell is also possible. BK Channel Current Carriers: Potassium, Rubidium and Ammonium Permeabilities for rubidium and potassium were similar (PRB/PK=0.87), however, outward currents carried by rubidium were small (gRb/gK=0-16). The similar permeabilities suggest that the energy barriers for the two ions are comparable. The dwell time of the two ions in the energy wells may however be different, with rubidium occupying the pore longer on average. Thus, rubidium is limiting its own current flow by prolonging its stay in the pore. Similar results were obtained for the BK channel in chromaffin cells (Yellen, 1984a) and in smooth muscle cells (Benham, 1986). In the smooth muscle BK channels, PRb/Pic was 0.7, while the conductance ratio was only 0.15. If rubidium does have long dwell times in the pore, it would be expected to have some effect on inward potassium currents, but this was not the case. The rubidium block did not affect potassium currents through the BK channels of any one of the three 68 preparations studied, including the melanotrophs (Yellen, 1984a; Benham et al., 1986). Yellen attributes this lack of effect on the potassium current to a strong voltage dependence of the rubidium "block". Further support for a voltage dependent block by rubidium was provided by Eisenman et al. (1986). They showed that the addition of 10 mM rubidium to the external face of the channel (in 50 mM symmetrical potassium solutions), was able to greatly decrease inward potassium currents, but only at negative membrane potentials. Ammonium ions were much less permeable than potassium ions (PNHl/TK =0.17). This is likely due to more favourable interactions between potassium and the pore lining, which is not surprising since the ammonium ion is large. Ammonium in fact-borders on the size limit of permeable ions. Hille (1975), suggested that ammonium would not be able to permeate the potassium pore if not for the fact that it is able to form hydrogen bonds with the pore. The hydrogen bonds effectively shorten the distance between the center of the ion and the pore and thus facilitates ion transport. Unlike the case for rubidium, gNnVgn (0.4) was larger than PNH^/PK (0.17). This suggests that although energy barriers were larger for ammonium then they were for potassium, the energy wells for ammonium were not as deep. Therefore, as long as an ammonium ion is able to get past the energy barriers, its exit is fast. The conductance sequence (K(l)>NH4(0.4)>Rb(0.15)) for the three ions is not the same as the permeability sequence (K(l)>Rb(0.86)>NH4(0.17)). Since the two measurements represent different properties of the permeation pathway (Hille, 1975), they need not be identical. The discrepency between conductance ratios and permeability ratios is in agreement with others, for example, Eisenman et al., (1986) noted that the conductance sequence (K(1)>NH4 (0.18)>Rb(.07)) for the BK channels of skeletal muscle was different from its permeability sequence (K(l)>Rb(0.7)>NH4(0.1)). 69 Cesium Ions are too Large to Permeate the BK Channel The BK channel had a very low cesium permeability, which may be a reflection of the large size of the cesium ion (Table 2). The low permeability is in agreement with the data collected from smooth muscle cells (Benham et al., 1986) and chromaffin cells (Yellen, 1984a). Internal cesium caused a flickery block of the inward potassium currents. Cesium has been reported to block other potassium channels including the delayed rectifier channels (Chandler and Meves, 1965; Adelman and French, 1978) and other BK channels (Yellen, 1984a; Benham et al. 1986). In smooth muscle cells (Benham et al. 1986), the internal cesium block was very' similar to that reported here with melanotroph BK channels in that the open channel noise was increased. In both chromaffin cells (Yellen, 1984a) and smooth muscle cells (Benham et al. 1986), external application of cesium was also able to block the BK channel; the block was stronger when cesium was applied from external side. Both internal and external blocks were voltage dependent, which suggests that the binding site is within the transmembrane electric field. The internal cesium block in the melanotroph BK channel also appeared to be voltage sensitive, but this was not systematically studied. Lithium and Sodium Ions are too Small to Permeate the BK Channel The permeability of the BK channel to both sodium and lithium were also very low. Perhaps, as previously discussed for sodium, these small ions (Table 2) are unable to interact with the carboxyl oxygens of the pore wall. These findings are consistent with the results of others (Yellen, 1984a; Blatz and Magleby, 1984; Lang and Ritchie,1988). In the delayed rectifier channel of a snail neuron (Reuter and Stevens, 1980), lithium had a slightly higher permeability than sodium. This was unexpected since lithium is an even smaller ion than sodium, its interactions with the pore wall would theoretically be more difficult. For the 70 melanotroph BK channel, the absolute permeabilities obtained from the GHK agree with this finding. However, since no outward currents were recorded with lithium being the predominent internal cation, the permeability of lithium may be lower than that estimated by the GHK equation. Yellen (1984) observed a block of potassium currents by internal sodium (5-20 mM) in BK channels of chromaffin cells. In this present study, 150 mM internal sodium was not observed to have an effect on inward potassium currents. However, subsequent experiments (Mitton and Kehl, unpublished observations) have shown that internal 20 mM sodium produces a strongly voltage-dependent block of outward potassium currents in the absence of external potassium. That no blocking effect of sodium was evident during the study of permeation is most likely due to the ionic composition of the recording solutions. Yellen (1984b) has shown that external potassium (5-20 mM) is able to relieve the block by internal sodium and, in this context, the fact that K„ was 150 mM in the permeation experiments might be significant. No data is available yet on the ability of external K to relieve the sodium block of melanotroph BK channels. In any case, the sodium block of BK channels in chromaffin cells and melanotrophs provides unequivocal evidence against independent ion flux through the channel and reinforces the concept of the BK channel as a multi-ion pore. A multi-ion pore has several ion binding sites within the pore region, each binding site being equivalent to an energy well. In this view, a pore can simultaneously be occupied by several ions. But how can the existence of many binding sites along the pore account for the high fluxes that can be measured in a BK channel? Hille (1975) and others (Yellen, 1984b; Blatz and Magleby, 1984) have proposed that as a new ion enters at one end of the pore, the other ions in the pore are destabilized by ion-ion interactions, a process that promotes rapid single-file "hopping" between binding sites and consequently rapid 71 movements through the channel. Yellen (1984b) proposed for the sodium block, that internal sodium can enter the cytoplasmic opening of the pore and block the channel. With a high concentration of potassium cations on the outside, external potassium enters the channel and interacts with the sodium ion at its binding site. At this point, repulsion from the charges of the two ions promotes the exit of the sodium ion back to the internal solution. Besides potassium, other cations have also been seen to relieve the sodium block; these cations include rubidium and cesium. This theory requires that there be at least two ions present in the pore for potassium to be able to "knock off' the sodium ion. 2. CHANNEL GATING 2.1 Voltage Dependence The gating of melanotroph BK channels is sensitive to changes in membrane potential. The open probability of the channel has a upper limit of increasing e-fold as the membrane potential is depolarized 10-13 mV. Similar sensitivities have been observed for BK channels in other cell preparations (Table 3) including muscle cells (Barret et al., 1981; Latorre et al, 1982; Blatz and Magleby, 1984; Benham et al., 1986; Wang and Mathers, 1993), chromaffin cells (Marty, 1983), and pituitary cells (Wong et al, 1982; Wang et al., 1992). The BK channel is not as sensitive to membrane potential as some other voltage-gated potassium channels. The delayed rectifier, for example, typically has a upper limit of increasing e-fold as the membrane potential is depolarized 2-6 mV (Kolb, 1990). Molecular studies have suggested that a chain of evenly spaced positive charges in the membrane-spanning, S4 region of the channel structure constitutes the voltage-sensor of the BK channel (Salkoff et al., 1992), which detects changes in the electrical field. Such II II II K. — D. O s H I I §• I ° 8 SL S § o § Oi tt « 3 3 " T3 o i l 8 1:1 3 * § K |i (i 5" " T3 3 C/ O O D. (t "> S" 3 a. _, §" I § 3 K •n O oi -a s « « 3 sr o 3 » D O 3 •a 00 ^ II © ON t * 50 s-^ P p ° TO s £ a. a 2 NO S. B is s?1 oo OO II B + O % Z H jo P* NO 00 o o p 1 P P C o- JV i l ON re 3 S P 1 3 NO B < o U . II > U l B < o . II B o 00 B 2 o to B 2 2 p o NO 00 C L > -^N NO p ON & B T3 0O oo II D N> B a. B P P m £ o B 13 5? C« to B S8 B x z Z H 2 1 2 P a D. O 5" Z g s 3 P NO « 00 B 13 5? V) o o o B, o ^ B o FT a ^ so N P P P £ °- sr & E 3- 2 NO n 2 S i 00 + II o ^ B D 3 II B P < w o B 2 o o 50 ^ fts P . ^ ft) <6 a. °- B s r — >-» p t?1 8 C - n h-1 NO 00 8 o g 3 1 3 00 oo II B x < 5 _ o o II 3 ** < P a o B < NO B 2 jo P W I - H P IT tJ* 0 1 I I a -r- B S i oo er a s c CZI to to ON g B 1 3 " 0O II X w O O 3 B-2 Z H P p OQ P P c p p-a* ]3 p* NO fi> NO P O B "a S< oo 0O i f ^ H B O U o w B * 5^ o _ a + II S P B SS < B 2 ^ i t/3 3 o s a e Si s a I I51 B ~ (ft CTQ B » » o re 5— B 2. & S S S ZL 73 changes in the electrical field may lead to physical shifting of the S4 region and therefore result in conformational changes of the channel. Moczydlowski and Latorre (1983) suggested that this conformational change does not directly affect gating but facilitates calcium binding. The binding of the calcium in turn affects the gating and, as occupancy increases, the likelihood of the channel being open also increases. In this view, the transition between open and closed states is not directly voltage dependent, only the calcium binding is voltage-dependent. In accordance with such a scheme, it has been shown that depolarization alone cannot activate the channel in the absence of calcium. 2.2 Calcium dependence BK channels are typically sensitive to intracellular calcium concentration (Methfessel and Boheim, 1982; Latorre et al., 1989). Structurally, BK channels are similar to other voltage-gated potassium channels (Salkoff et al., 1992). The molecular structure of the BK channel consists of a core region with the six putative transmembrane segments, similar to those described for other potassium channels. The primary sequence of the BK channel is however much larger than that of most other potassium channels. This additional length is believed to be due to additions to the carboxyl terminal of the channel (Wei et al., 1994). This appendage is believed to consist of four additional transmembrane domains and the sensitivity to calcium is believed to be due to the cytoplasmic segments on this "tail domain". Sensitivity to calcium can vary in different cell types (Table 3). For example, the calcium concentration required to produce a half-maximal open probability (Po=0.5 or P05), at 0 mV, ranges from 5 x 105 M in rabbit muscle cells (Moczydlowski and Latorre, 1983; Vergara et al. 1984) to 10* M in secretory cells ( Maruyama et al., 1983; Findlay, 1984), with rabbit smooth muscle cells and anterior pituitary cells having an intermediate sensitivity, 5 74 x 107 M (Benham et al. 1986; Wong et al, 1982). These differences in sensitivity range may reflect different physiological calcium concentrations in the respective cells as well as different functions of the channels. The melanotrophs showed a calcium sensitivity (P0 5 = 6 x 10"7 M at 0 mV) similar to that of the smooth muscle and anterior pituitary cells. The BK channel in the melanotroph was most sensitive to calcium concentrations between 0.1 to 1 /xM. Concentrations of calcium above 1 /xM did not greatly increase the open probability of the channel, presumably because the binding sites were maximally occupied. The open probability is however a steep function of the intracellular calcium concentration over the range of 0.1-1 /xM, suggesting that the binding of many calcium ions is required for the activation. The results suggested that there are three to six calcium binding sites on the BK channel of the rat melanotroph. This is in agreement with BK channels from other preparations which have estimated the existence of up to six calcium binding sites (McManus, 1991). A more detailed discussion of the calcium dependence of the BK channels can be found in the reviews by McManus and Magleby (1991) and McManus (1991). 23 Other Modulators of BK Channel Gating Aside from voltage and calcium, the BK channels gating may also be modulated by other factors. From Figure 14, it can be seen that at a given calcium concentration, there is considerable variability in the gating behavior of individual patches. BK channels have been observed to be affected by either phosphorylation or dephosphorylation (Egan et al., 1993; Tamaguchi et al., 1993; and Reinhart et al, 1991). BK channels from canine artery smooth muscle cells can be phosphorylated. The half-activation potential of the phosphorylated channel is shifted 25 mV in the negative direction at a 75 given concentration of calcium (Tamihuchi et al., 1993). Since some BK channel types respond to phosphatases, at least some of the channels are in the phosphorylated state prior to experimental manipulations (Reinhart et al., 1991). The differences in gating observed in our study might be due to different phosphorylation states of the individual channels. Also, with regard to the calcium dependence of BK channels, subtypes of BK channels have been found in other cell preparations. Subtypes may have similar conductances and even voltage and calcium dependence. Subtypes of rat brain BK channels have been distinguished based on the reaction to phosphatases; one subtype was up-regulated upon phosphorylation, while the other type was down-regulated (Reinhart et al., 1991). Subtypes of BK channels may also be present in the rat melanotroph. The variability observed in our experiment may be due to one or more of these factors. 3. BK CHANNEL BLOCKERS 3.1 Tetraethylammonium T E A + is a well-known potassium channel blocker. It has been reported to block most types of voltage-gated potassium channels and has been seen to block BK channels in most, but not all, preparations from both sides of the membrane (Latorre et al., 1989). The concentrations required for a half-maximal block vary a great deal (Table 3). For example, the KQ at 0 mV for the internal block can range from 0.08 mM in a pituitary cell line (Wong and Adler, 1986) to 60 mM in rat myotubules (Blatz and Magleby, 1984); and the K D at 0 mV for an external block can range from 0.2 mM in chromaffin cells (Marty, 1983) to 52 mM in a pituitary cell line (Wong and Adler, 1986). The large difference in the K D values might reflect small changes in the primary sequence of the pore region, since single amino acid substitutions of some potassium channels can produce large changes of sensitivity to 76 T E A + (Yellen et al, 1991.) External TEA + caused an intermediate open channel block; the K D was 0.25 mM, at 0 mV. Internal TEA + acted as a fast blocker with a K,, of 50 mM, at 0 mV. Thus, the profile of the BK channel block by internal and external TEA + closely resembles that in another secretory cell, the chromaffin cell of the adrenal medulla (Yellen, 1984a). 3.2 Charybdotoxin CTX is an extract of the venom of the scorpion, Leiurus quinquestriatus, that blocks, at nanomolar concentrations, the BK channel of several different preparations (Latorre et al, 1989; Miller et al, 1985; for an exception see, Wang et al, 1992). CTX was once perceived to be a selective blocker for BK channels and has been used extensively in the classification of potassium channels. More recently, CTX has been shown also to block SK channels (Blatz and Magleby, 1986; Hermann and Erxleben, 1987). Our data are consistent with previous results showing that CTX acts as a slow open channel blocker (Miller et al., 1985). 33 Significance of Blocking Agents Why is it important to determine which blockers affect the BK channel? First pharmacological information provides a means of classifying channels. It also provides structural information about the channel. The fact that a large molecule like T E A + can cause an open channel block has been used to extract information concerning the dimensions of the pore vestibule (Armstrong, 1975). Blocking experiments have also provided evidence to support the view of the BK channel as a multi-ion pore, as noted previously for the block of the BK channel by internal sodium ions. Perhaps most 77 importantly, finding a selective blocker for the BK channel would enable a pharmacological dissection to determine the functional role of the current. 4. SPECULATIONS ON THE PHYSIOLOGICAL FUNCTION OF THE BK CHANNEL 4.1 BK Open Probability at Resting Conditions The open probability of most of the BK channels was virtually zero in cell-attached recordings (Figure 3A) when the membrane patch was not depolarized (-VP=0). Assuming a resting membrane potential of approximately -40 mV (McBurney and Kehl, 1988) and a resting intracellular calcium concentration of 0.1 to 0.3 /xM (Shibuya and Douglas, 1993), these data are consistent with the data obtained from inside-out patches (Figure 14), which predict that the majority of the BK channels would not be active under these conditions. The activity of these BK channels is however likely to increase upon cell depolarization, and therefore may play a role during an action potential. A few of the cell-attached patches did show spontaneous BK channel activity at membrane potentials more negative than -80 mV (Figure 3A). The cells from which the patches were taken appeared to be healthy and were not phenotypically different from other cells tested. Since the resting level of calcium is not identical for every cell (Douglas and Shibuya, 1993), these few occurrences might have reflected local increases in calcium concentration. Melanotroph BK channels are extremely sensitive to small changes in calcium around the resting concentrations. Consequently, small changes in calcium concentration due to influx through the low threshold calcium channels and/or release from internal stores might account for the observed changes in BK activity. In smooth muscle cells (Muraki et al., 1992), caffeine-induced release of calcium from internal stores was effective in activating BK channels. 78 4.2 BK Open Probability During an Action Potential The function of the majority of the BK channels in the rat melanotroph is most likely related to the repolarization of the membrane following an action potential and the cessation of hormone secretion. From the result of the gating experiments, BK activity greatly increases both with membrane depolarization and increases in intracellular calcium concentration. Thus, there are conditions that will increase BK activity both during and after an action potential. During the action potential, opening of sodium channels drives the membrane potential close to E^, (40-50mV). At the same time, the depolarization also leads to the opening of voltage-gated calcium channels. Calcium then enters-through these channels which then increases the intracellular calcium concentration. Calcium imaging experiments (Nemeth et al., 1989), have shown that potassium-induced depolarization of the melanotroph can produce calcium concentrations several-fold higher than the resting level. However, calcium measurements, using imaging systems, represent the mean increase of calcium over the entire cell and much of the recent literature has suggested that the increase in calcium is not uniformly distributed throughout the cell. Calcium entering through the voltage-gated channels would be expected, on theoretical grounds, to concentrate near the point of entry. Therefore, if the BK channel is in close proximity to the voltage-gated calcium channels, it might transiently be exposed to a higher level of calcium than that assessed by the calcium imaging studies. To determine if BK channels are involved in the repolarizing phase of the action potential, future experiments are needed to determine if brief depolarization and/or brief increases in calcium concentration, such as are believed to occur during an action potential, are sufficient to significantly affect the open probability of the BK channel. 82 Hille, B. (1973). Potassium channels in myelinated nerve. Selective permeability to small cations. /. Gen. Physiol. 61:669-686. Hille, B. (1975). In Membranes, A series of Advances, ed. Eisenman, G. pp. 256-323. Ionic selectivity of Na and K channels of nerve membranes. New York: Marcel Dekker. Hille, B. (1992). Ionic Channels of Excitable Membranes. Sinauer Assoc. Inc., Sunderland, Mass. Second Edition. Howe, A. (1973). The mammalian pars intermedia: a review of its structure and function. /. Endocrin. 59:385-409. Kehl, S.J. (1989). Cultured melanotrophs of the adult rat pituitary possess a voltage-activated fast transient outward current. /. Physiol. (Lond.) 411:457-468. Kehl, S.J. (1990). 4-aminopyridine causes a voltage-dependent block of the transient outward potassium current in rat melanotrophs. /. Physiol. (Lond.) 431:515-528. Kehl, S.J., and McBurney, R.N. (1989). The firing patterns of rat melanotrophs recorded using the patch clamp technique. Neurosci. 33:579-586. Kehl, S.J. (1994). Voltage-clamp analysis of the voltage-gated sodium current of the rat pituitary melanotroph. Neurosci. Lett. 165:67-70. Kolb, H. (1990). Potassium channels in excitable and non-excitable cells. Rev. Physiol.Biochem.Pharmacol. 115:51-76. Lang, D.G., and Ritchie, A.K. (1988). Pharmacological sensitivities of large and small conductance calcium-activated potassium channels. Biophys. J: 37:170a Latorre, R., Oberhauser, A., Labarca, P., and Alvarez, O. (1989). Varieties of calcium-activated potassium channels. Anna. Rev. Physiol. 51:385-399. 83 Latorre, R., Vergara, C , and Hidalgo, C. 1982. Reconstitution in planar lipid bilayers of a calcium dependent potassium channel from transverse tubule membranes isolated from rabbit skeletal muscle. Proc. Natl. Acad. Sci, USA. 77:7484-86. Law, G.J., Pachter, J.A., and Dannies, P.S. (1989). Calcium transients induced by thyrotropin-releasing hormone rapidly lose their ability to cause release of prolactin. Mol. Endocrin. 3:539-546. Lee, CO. and Fozzard, H.A. (1975). Activities of potassium and sodium ions in rabbit heart muscle. /. Gen. Physiol. 65:695-708. Leenders, H., de Koning, H.P., Ponten, S.P., Jenks, B.G., and Roubos, E.W. (1993). Diffential effects of coexisting dopamine, GABA, and NPY on a-MSH secretion from melanotrope cells oiXenopus laevis. Life Sciences. 52:1969-1975. Leinders, T., van Kleef, R.G.D.M., and Vijverberg, H.P.M. (1992). Divalent cations activate small-(SK) and large-conductance (BK) channels in mouse neuroblastoma cells: selective activation of SK channels by cadmium. Pfliigers Arch. 422:217-222. Lester, H.A. (1991). Strategies for studying permeation at voltage-gated ion channels. Annu. Rev. Physiol. 53:477-96. Marty, A. (1981). Calcium-dependent potassium channel with large unitary conductance in chromaffin cell membranes. Nature 291:497-500. Marty, A. (1983). Calcium-dependent potassium channel with large unitary conductance. Trends in Neurosci. 6:262-65. Maruyama, Y., Gallecher, D.V., and Petersen, O.H. (1983). Voltage and calcium-activated potassium channel in basolateral acinar cell membranes of mammalian salivary glands. Nature 302:827-829. Maue, R. A., and Dionne, V.E. (1987). Patch-clamp study of isolated mouse olfactory receptor neurons. /. Gen. Physiol. 90:95-125. 84 McBurney, R.N., and Kehl, S. (1989). Electrophysiology of neurosecretory cells from the pituitary intermediate lobe. /. Exp. Biol. 139:317-328. McManus, O.B. (1991). Calcium-activated potassium channels: regulation by calcium. /. Bioenerg. Biomembr. 23:537-560. McManus, O.B. and Magleby, K.L. (1991). Accounting for the calcium-dependent kinetics of single large-conductance calcium activated potassium channels in rat skeletal muscle. /. Physiol, (lond.) 443:739-777. Methfessel, C , and Boheim, G. (1982). The gating of single calcium-dependent potassium channels is described by an activation/blockade mechanism. Biophys. Struc. Mech. 9:35-60. Miller, C , Latorre, R., and Reisin, I. (1985). Charybdotoxin, a protein inhibitor of calcium-activated potassium channels from mammalian skeletal muscle. Nature 313:316-318. Miller, C , Latorre, R., and Reisin, I. (1987). Coupling of voltage-dependent gating and barium block in the high conductance calcium activated potassium channel. /. Gen. Physiol. 90:427-449. Moczydlowski, E., and Latorre, R. (1983). Gating kinetics of calcium-activated potassium channels from rat muscle incorporated into planar lipid bilayers./. Gen. Physiol. 82:511-542. Muraki, K., Imaizumi, Y., and Watanabe, M. (1992). Calcium-dependent potassium channels in smooth muscle cells permeabilized by B-escin recorded using the cell-attached patch-clamp technique. Pflugers Arch. 420:461-469. Neher, E. (1992). Correction for liquid junction potentials in patch clamp experiments. Methods Enzymol. 207:123-131. Nemeth, E.F., Taraskevich, S., and Douglas, W.W. (1989). Cytosolic calcium in melanotrophs: pharmacological insights into regulatory influences of electrical activity and ion channels. Endocrinol. 126(2):754-758. Nussinovitch, I., and Kleinhaus, A. L. (1992). Dopamine inhibits voltage-activated calcium channel currents in rat pars intermedia pituitary cells. Brain Res. 574(l-2):49-55). 85 Pallotta, B.S., Blatz. A.L., and Magleby, K.L. (1992). Recording from calcium-activated potassium channels. Methods in Enzymol. 207:194-207. Peterson, O.H., and Maruyama, Y. (1984). Calcum-activated potassium channels and their role in secretion. Nature 307:693-697. Reinhart, P.H., Chung, S., and Levitan, I. (1989). A family of calcium-dependent potassium channels from rat brain. Neuron 2:1031-1041. Reinhart, P.H., Chung, S., Martin, B., Brautigan, D.L. and Levitan, I. (1991). Modulation of calcium-activated potassium channels from rat brains by protein kinase A and phosphatase 2A. J. Neurosci. 11:1627-1635. Reuter, H., and Stevens, C.F. (1980). Ion conductance and ion selectivity of potassium channels in snail neurones. /. Memb. Biol. 57:103-118. Robinson, R.A., and Stokes, R. H. (1965). Electrolyte Solutions. London:Butterworths. Salkoff, L., Baker, K., Butler, A., Covarrubias, M., Pak, A., and Wei, A. (1992). An essential set of K channels conserved in flies, mice and humans. Trends in Neurosci. 15:161-166. Schlegel,W., Winiger, B.P., Mollard, P., Vacher, P., Wuarin, F., Zahnd, G.R., Wollheim, C.B., and Dufy, B. (1987). Oscillations of cytosolic calcium in pituitary cells due to action potentials. Nature 329:719-721. Schneggenburger, R., and Lopez-Barneo, J. (1992). Patch clamp analysis of voltage-gated currents in intermediate lobe cells from rat pituitary thin slices. Pfliigers Archiv. 420:302-312. Shibuya, I., and Douglas, W.W. (1993). Spontaneous cytostolic calcium pulsing detected in Xenopus melanotrophs: modulation by secreto-inhibitory and stimulatory ligands. Endocrin. 132:2166-2175. 86 Squire, L.G., and Petersen, O.H. (1987). Modulation of calcium and voltage-activated potassium channels by internal magnesium in salivary acinar cells. Biochimica et Biophysica Acta 899:171-175. Stack, J. and Surprenant, A. (1991). Dopamine actions of calcium currents, potassium currents and hormone release in rat melanotrophs. /. Physiol. (Lond.) 439:37-58. Sugihara, I. (1994). Calcium-activated potassium channels in goldfish hair cells. /. Physiol. (Lond.) 476:373-390. Tamaguchi, J., Furudawa, K.I. and Shigekawa, M. (1993). Maxi K channels are stimulated by cyclic guanosine monophosphate-dependent protein kinase in canine artery smooth muscle cells. Pflugers Archiv. 423(3-4):167-172. Taraskevich, P.S., and Douglas, W.W. (1986). Effects of BAY K 8644 and other dihydropyridines on basal and potassium-evoked output of MSH from mouse melanotrophs in vitro. Neuroendocrin. 44:384-389. Taraskevich, P.S., and Douglas, W.W. (1989). Effects of BAY K 8644 on calcium-channel currents and electrical activity in mouse melanotrophs. Brain Res. 491:102-108. Thomas, P., Surprenant, A., and Aimers, W. (1990). Cytosolic calcium, exocytosis and endocytosis in single melanotrophs of the rat pituitary. Neuron 5:723-733. Tomiko, S.A., Taraskevich, P.S., and Douglas, W.W. (1982). Potassium induced secretion of melanocyte-stimulating hormone from isolated pars intermedia cells signals participation of voltage-dependent calcium channels in stimulus-secretion coupling. Neurosci. 6:2259-2267. Trautman, A., and Marty, A., (1984). Activation of calcium-dependent K channels by carbamoylcholine in rat lacrimal glands. Proc. Natl. Acad. Sci. USA. 81:611-15. Valentijn, J.A., Vaudry, H., and Cazin, L. (1993). Multiple control of calcium channel gating by dopamine D 2 receptors in frog pituitary melanotrophs. Ann. N.Y. Acad. Sci. 680:211-227. 87 Valentijn, J. A., Vaudry, H., Kloas, W., and Cazin, L. (1994). Melanostatin inhibited electrical activity in frog melanotrophs through modulation of potassium, sodium and calcium currents. /. Physiol. (Lond.) 475:185-195. Vergara, CM. , Moczydlowski, E., and Latorre, R. (1984). Conduction, blockade and gating in a calcium-activated potasium channel incorporated into planar lipid bilayers. J. Gen. Physiol. 82:543-568. Wagoner, K., and Oxford, G.S. (1987). Cation permeation through the voltage-dependent potassium channel in the squid axon. /. Gen. Physiol. 90:261-290. Wang, G., Thorn, P., and Lemos, J.R. (1992). A novel large-conductance calcium-activated potassium channel and current in nerve terminals of the rat neurohypophysis. /. Physiol. (Lond.) 457:47-74. Wang, Y., and Mathers, D.A. (1993). Calcium-dependent potassium channels of high conductance in smooth muscle cells isolated from rat cerebral arteries. J. Physiol. (Lond.) 462:529-545. Wei, A., Solaro, C , Lingle, C , and Salkoff, L. (1994). Calcium sensitivity of BK-type channels determined by a separable domain. Neuron 13:671-681. Williams, P.J., Pittman, Q.J., and MacVicar, B.A. (1991). Calcium and voltage-dependent inactivation of calcium currents in rat intermediate pituitary. Brain Res. 564:12-18. William, P.J., Pittman, Q.J., and MacVicar, B.A. (1993). Blockade by funnel web toxin of a calcium current in the intermediate pituitary of the rat. Neurosci. Lett. 157:171-4. Wong, B.S., Lecar, H., and Adler, M. (1982). Single calcium dependent potassium channels in clonal anterior pituitary cells. Biophys. J. 39:313-317. Wong, B.S., and Adler, M. (1986). Tetraethlyammonium blockade of calcium-activated potassium channels in clonal anterior pituitary cells. Pfliigers Archiv. 407:279-84. Yellen, G. (1984a). Ionic permeation and blockade in Caz+-activated K + channels of bovine chromaffin cells. /. Gen. Physiol. 84:157-186. 88 Yellen, G. (1984b). Relief of sodium block of calcium-activated potassium channels by external cations. /. Gen. Physiol. 84:187-199. Yellen, G., Jurman, M.E., Abramson, T. and Mackinon, R. (1991). Mutations affecting internal TEA blockade identify the probable pore-forming region of a potassium channel. Science 251:939-942. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086935/manifest

Comment

Related Items