Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of divalent cations on IK(f) : a transient outward potassium current expressed in melanotrophs… Davidson, Jana-Lea 1992

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

Item Metadata

Download

Media
831-ubc_1992_fall_davidson_jana-lea.pdf [ 3.47MB ]
Metadata
JSON: 831-1.0086542.json
JSON-LD: 831-1.0086542-ld.json
RDF/XML (Pretty): 831-1.0086542-rdf.xml
RDF/JSON: 831-1.0086542-rdf.json
Turtle: 831-1.0086542-turtle.txt
N-Triples: 831-1.0086542-rdf-ntriples.txt
Original Record: 831-1.0086542-source.json
Full Text
831-1.0086542-fulltext.txt
Citation
831-1.0086542.ris

Full Text

THE EFFECT OF DIVALENT CATIONSON IK(f), A TRANSIENT OUTWARD POTASSIUMCURRENT EXPRESSED IN MELANOTROPHSOF THE RAT PITUITARY GLANDbyJANA-LEA DAVIDSONB.Sc., University of British Columbia, 1985A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Physiology)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIASeptember 1992© Jana-Lea DavidsonIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of  PhysiologyThe University of British ColumbiaVancouver, CanadaDate  DE-6 (2/88)iiABSTRACTDivalent cations are known to exert a charge screening effect on voltage-gated ionchannels either through non-specific interactions with fixed negative charges on the cellmembrane or via binding to negatively charged sites on or electrically close to the channelforming protein. In some instances, divalent cations bind directly to the gating apparatusof voltage-sensitive sodium and potassium channels thereby stabilizing the channels in aclosed conformation.Most of the investigations into the effects of divalent cations on voltage-gated ionchannels have concentrated on the Na + and delayed rectifier type K + channels. Although,there has been a recent explosion of information regarding the molecular structure of thetransient outward potassium channel, few investigators have examined the actions of divalentcations on the behaviour of the transient outward potassium current (TOC).A transient outward potassium current, (IK(f)), has been characterized inmelanotrophs, the major cell type found in the pars intermedia of the pituitary gland in rats(Kehl, 1989). IK(f) activates and inactivates rapidly. Cd2+ (5 mM) reduced the peakamplitude of I K(f) and increased the 50% rise time of this current (Kehl, 1989). The presentstudy elaborated on these observations and examined the effects of varying the extracellularconcentrations of Cd 2+ , Zn2+ , Ca2+ and Mg2+ on the behaviour of IK(f).Acutely dissociated melanotrophs were obtained from male Wistar rats and whole-cell currents were recorded, using conventional patch clamp techniques, from cellsmaintained in culture for 1-14 hrs.Divalent cations shift the activation and inactivation curves and the gating kineticsof IK(f) right-ward along the voltage axis. The cations tested varied in their ability to shiftthe potential-sensitive parameters of IK(f) and ranked in the following order: Zn 2 + > Cd2+> > Ca2 + > Mg2+ , in good agreement with previous observations of their effect on sodiumchannels.iiiThe mean control half-activation potential was -13.6 mV with a slope-factor of + 12.8mV (n = 55) and the mean control half-inactivation potential was -54.7 mV with a slope-factor of -4.4 mV (n =50). The relationships between the shift of the half-activationpotential and the divalent cation concentration indicated that the K m 's for the half-maximalshift of the activation curve were 221 µM (Cd 2 +), 92 µM (Zn2 +) and 3.4 mM (Ca2 +) and themaximal shifts of the activation curve were, respectively, +28 mV, +34 mV and + 15.6 mV.Mg2+ was far less potent than any of the other divalent cations examined. Shifts of theinactivation curve were equal to the shifts of the activation curve at each divalent cationconcentration tested. The slope-factors of the activation and inactivation curves were notaltered by the application of divalent cations. Removal of Ca 2+ from the external mediasignificantly increased the slope-factor for the activation curve. That is, zero Ca2+ resultedin a decrease in the equivalent charge transferred during the activation gating process.The prediction central to non-specific charge screening is that all divalent cations willbe equally effective. The results reported here show that this is not the case for I K(f). It isproposed that a specific binding site for divalent cations exists on or electrically close to thechannel protein. Divalent cations also slowed the rise time of I K(f) suggesting that theymight stabilize the channel conducting this current in the closed conformation.ivTABLE OF CONTENTSPAGEABSTRACT 	 iiTABLE OF CONTENTS	 ivLIST OF FIGURES 	 viiLIST OF TABLES 	 ixACKNOWLEDGEMENTS	INTRODUCTION 	 1Voltage-Gated Ion Channels And Electrical Signals In Excitable Cells	 1Negative Surface Charges Exist On Cell Membranes	 2Membrane Surface Potential Can Be Described Mathematically 	 4Different Species Of Divalent Cations Are Not Equally Effective InShifting The Voltage-Sensitive Parameters Of Voltage-Gated Ion Channels 	4Some Species Of Divalent Cations Block Ion Channels In AVoltage-Dependent Manner	 6Molecular Characterization Of Voltage-Gated Ion Channels	 7Is The Behaviour Of The Transient Outward Potassium ChannelInfluenced By Divalent Cations?	 8Melanotrophs Possess A Transient Outward Potassium Current	 9Experimental Rationale	 10METHODS	 111. Preparation Of Acutely Dissociated Melanotrophs	112. Electrophysiology	 123. Data Acquisition and Analysis 	 134. Recording Solutions 	 14VRESULTS	 18SECTION I: GENERAL PROPERTIES OF IK(f) 	 181. Voltage- and Time-Dependence Of Activation	 182. Voltage- and Time-Dependence Of Inactivation	 193. Voltage- and Time-Dependence Of The Residual Steady-State Current 	 224. Reversal Potential 	 22SECTION II: THE EFFECTS OF DIVALENT CATIONS ON I K(f) 	 271. The Effects Of Transition Metal Ions On The Gating Of I K(f) 	 271.1 Cadmium 	 271.11 Cd2 + Reduces The Peak Amplitude Of IK(f) 	 271.12 Cd2+ Causes A Right-Ward Shift Of The Activation Curve	 331.13 Concentration Dependence For Cd 2+ -Induced Shifts Of V' 	 341.14 Cd2+ Increases The 50% Rise Time Ow act) Of IK(f) 	 341.15 Cd2+ Causes A Right-Ward Shift Of The Steady-StateInactivation (h.,) Curve 	 371.16 Cd2+ Increases The Time To Half-Inactivation (t y, inact ) 	 381.2 Zinc	 391.21 Zn2+ Reduces The Peak Amplitude Of IK(f) 	 391.22 Zn2+ Causes A Right-Ward Shift Of The Activation Curve	441.23 Concentration Dependence For Zn 2+ -Induced Shifts Of V' 	 471.24 Zn2 + Increases The 50% Rise Time Of I K(f) 	 471.25 Zn2+ Causes A Right-Ward Shift Of TheSteady-State Inactivation Curve 	 471.26 Zn2+ Increases The Time To Half-Inactivation 	 50vi2. The Effects Of Alkaline Earth Metal Ions On I K(f) 	 572.1 Calcium 	 572.11 Changes in [Ca- 10 Reduce I K(f) 	 572.12 Changes in [Ca'- 1. Shift V' For The Activation Curve	 582.13 Concentration Dependence For Ca2+ -Induced Shifts Of V'	612.14 Raising [Ca 2 1. Increases ty, act For IK(f) 	 612.15 Changes In [Ca 2 +],, Shift V' For The Inactivation Curve	 612.16 Influence Of Changing [Ca'-1 ]. On ty, inact For IK(f) 	 662.2 Magnesium	 662.21 Voltage- and Time-Dependence Of Activation For IK(f)In The Presence Of Mg2+ 	 712.22 Mg2+ Shifts V' For Steady-State Inactivation 	 71DISCUSSION 	 81Divalent Cations Exert A Charge Screening Effect On TheChannel Conducting I K(f) 	 81Divalent Cations Stabilize The Closed Conformation OfThe Channel Conducting IK(f) 	 84Speculation On The Possible Characteristics Of The Binding Site 	 84High [Zn2 1. And [Cd2 1. Reduce G.	86Is Ca2+ A Necessary Co-factor For I K(f)? 	 87The Activation And Inactivation Of I K(f) Appear To Be Coupled 	 88Physico-Chemical Properties Of The Divalent Cations 	 89Summary and Future Directions 	 90Concluding Remarks 	 91BIBLIOGRAPHY 	 92LIST OF FIGURESFIGURE 	 PAGE1 	 The current-voltage relation and steady-state inactivation of I K(f) undercontrol conditions 	 202 	 The time constants for the inactivation of I K(f) 	 233 	 The reversal potential for I K(f) as determined from tail currents	 254 	 The concentration-dependent effects of Cd2+ on IK(f) 	 285 	 The effects of 500 AM and 5 mM Cd 2+ on the voltage-dependence ofactivation and inactivation for IK(f) 	 316 	 Concentration-dependent effects of Cd 2+ on the shift of the half-activationpotential 	 357 	 The time-dependence of activation and inactivation in the presence of 500and 5 mM Cd 2+ 	 408 	 Concentration-dependent effects of Zn 2+ on the current-voltage relation ofIK(f) 	 459 	 Control and treated activation and inactivation curves in Zn2+ concentrationsof 62.5 AN1, 125 AM, 250 AM, and 500 AM	 4810 	 The concentration dependence of the Zn2+ -induced shift of the half-activationpotential for IK(f) 	 5111 	 The time-dependence of activation and inactivation in the presence of 500 AMZn2 + 	 5312 	 Current-voltage relations for normalized I K(f) in response to reduced orincreased concentrations of external Ca 2+ 	 5913 	 The effect of changes of [Ca 210 on the activation and steady-state inactivationcurves for I K(f) 	 6214 	 Concentration-response for the effects of Ca 2+ on the shift of the half-activation potential (V') 	 6415 	 The effect of increased [Ca2+ ]0 on the activation and inactivation kinetics ofIK(f) 	 67viiviii16	 The effects of 10 mM (A) and 40 mM (B) Mg 2+ on the activation andinactivation curves for I K(f) 	 7217	 The time-dependence of activation in the presence of 40 mM Mg 2+ 	 75LIST OF TABLESTABLE 	 PAGE1 	 Composition of external solutions	172 	 The values of V' and k calculated for the activation and inactivation of I K(f)in the presence of Cd2+ 	 423 	 The values of V' and k calculated for the activation and inactivation of I K(f)in the presence of Zn 2 + 	 554 	 The values of V' and k calculated for the activation and inactivation of I K(f)in the presence and absence of Ca 2+ 	 695 	 The values of V' and k calculated for the activation and inactivation of I K(f)in the presence of Mg2+ 	 776 	 Shifts of the activation and inactivation curves of IK(f) caused by altering theexternal divalent cation concentration	 79ixxACKNOWLEDGEMENTSI would like to take this opportunity to thank Dr. Steven Kehl for allowing me towork with him, for his invaluable advice and input during the course of this project and forproviding me with a student fellowship. In particular, I would like to thank Dr. Kehl forsharing with me his dedication to his craft and for teaching me by his fine example how Ishould approach not only science and research but learning.My sincere thanks to Dr. Peter Vath-than for always making time for my questions,for the hours spent at the chalk board and for his encouragement and input throughout mythesis project. I would also like to thank Dr. Vaughan for excouraging me to undertake thisMaster's degree.Thank-you as well to Dr. David Mathers for Chairing my thesis committee, forkeeping all the necessary paper work flowing, for his thorough review of my thesis and forhis helpful comments and observations thereafter.A big thank-you to Dr. Raymond Pederson for his endless efforts as the GraduateStudent Advisor which made the trek through graduate studies in Physiology such awonderful experience. Memories of the Mavne Island retreats will always make me smile!A special thank-you to CA, Christine. and Jim for your friendship and encouragementand for agreeing to adjourn the Saturday morning meetings at Pauls! A very special thank-you to Jim Potts for rescuing me from a sea of numbers and helping with the statisticalanalysis.Thanks to the rest of the graduate students and staff in Physiology for making my stayseem too short, to Khaled for our wonderful discussions and to Eric for all his musicaltalents and his bad jokes!I would like to thank Clara for her technical assistance and Monica for her assistancein the preparation of the figures for this thesis.To my family, thank-you hardly seems enough, your love and support has alwaysallowed me to take on new challenges.Scoobie, I couldn't have done it without you...thanks isn't enough.1INTRODUCTIONVoltage-Gated Ion Channels And Electrical Signals In Excitable CellsVoltage-sensitive ion channels amplify electrical signals and allow them to propagatealong the membrane from one region of an excitable cell to another. These channels, whichare activated by local changes in membrane potential, are responsible for the dramaticincrease in the permeability of excitable membranes to Na+ and K + during the firing of anaction potential.Hodgkin and Huxley (1952), who first described these permeability changes in squidaxon, proposed the existence of channels selective for Na + and K+ and hypothesized thatindependent gating parameters responding to changes in membrane potential controlled theactivation and inactivation of the Na + channel and the activation of the K+ channel. Ofgreat importance to the field of electrophysiology was the testable mathematical model theydeveloped which predicted the behaviour of these gating parameters and mimicked the Na+and K + currents they observed (Hodgkin & Huxley, 1952).Hodgkin and Huxley (1952) also described the driving forces which would push ionsthrough these open channels. At rest the membranes of excitable cells are far morepermeable to IC ions than to Na+ ions. This reduces the amount of Na+ crossing themembrane and reduces the metabolic energy required to keep the intracellularconcentration of Na + low. On the other hand, the intracellular concentration of K + is kepthigh and the resting membrane potential usually lies close to EK, the equilibrium potentialfor K + . Thus, the potential of the intracellular compartment is negative with respect to theextracellular compartment (Hodgkin & Huxley, 1952). When a channel opens the drivingforce provided by this electrochemical gradient allows the movement of ions across the cellmembrane (reviewed in Hille, 1984).Since Hodgkin and Huxley's definitive work studies of excitable cells from many otherspecies and tissues have led to the discovery of voltage-gated Ca2+ and cr channels as well2as other Na + and K + channels. The use of more sophisticated techniques for analyzing thebehaviour of these channels has increased the knowledge of how they behave which isreflected in commensurately more complicated mathematical models (reviewed in Hille,1984).Excitable cells are usually studied in environments created by the investigator andconclusions regarding the manner in which these cells function in situ are often drawn fromthese experiments. Voltage-gated ion channels play an important role in signal propagationin excitable cells. Therefore, understanding how factors, such as the ionic composition ofthe recording solutions, influence the conductance through these channels or the voltage-sensitive gating of these channels becomes very important. Conversely, the manner in whichthese channels respond to changes in their environment provides some insight intostructure/function relationships.Negative Surface Charges Exist On Cell MembranesIntracellular or extracellular changes in divalent cation concentration(Frankenhaeuser & Hodgkin, 1957; McLaughlin et al., 1971; Hille et al., 1975), ionicstrength (Chandler et al., 1965; Mozhayeva and Naumov, 1972), or pH (Hille, 1968;Mozhayeva and Naumov, 1972; Woodhull, 1973) shift the potential-sensitive parameters ofvoltage-dependent ion channels along the voltage axis. Several models have been proposedto account for the voltage shifts induced by these manipulations including, screening of fixednegative surface charges on the membrane, specific binding of cations to negative surfacecharges and combinations of both (Frankenhaeuser and Hodgkin, 1957; Gilbert andEhrenstein, 1969; McLaughlin et al., 1971; D'Arrigo, 1978; Hille, 1968; Hille et al., 1975).These experimental conditions appear to alter the membrane potential surrounding thevoltage sensor of the channel without altering the bulk potential between the intracellularand extracellular compartments which is measured by an intracellular electrode.3Observations regarding the influence of divalent cations on the behaviour of excitabletissue preceded Hodgkin and Huxley's description of the Na + and K + channel. Adrian andGelfan (1933) noted that lowering external Ca 2+ caused hyperexcitability in muscle andBrink (1954) suggested, from his studies in nerves, that this increased activity resulted froma decrease in membrane resistance upon exposure to low external Ca 2+ . Highconcentrations of external Ca', on the other hand, were observed to increase membraneresistance and to stabilize excitable tissues (Weidmann, 1955).Shortly after Hodgkin and Huxley (1952) described the changes in Na + and K +permeability in squid axon during an action potential Frankenhaeuser and Hodgkin (1957)examined the effects of raising and lowering the concentration of extracellular Ca 2+ on thesepermeability changes. They observed that a five-fold reduction of the extracellular Ca 2+concentration shifted the conductance-voltage relations for both Na + and K + along thevoltage axis in a hyperpolarizing direction and was equivalent to depolarizing the membraneby 10 to 15 mV (Frankenhaeuser & Hodgkin, 1957). No measurable differences wereobserved in the potential between the bulk intracellular and extracellular solutions. Toexplain these results Huxley proposed that Ca2+ might adsorb to negative charges on theouter surface of the cell thereby altering the electric field within the membrane in a mannerwhich effectively hyperpolarized the cell (Frankenhaeuser & Hodgkin, 1957).Subsequent evidence suggested that fixed negative surface charges exist on thecytoplasmic face of the membrane as well. Chandler et al. (1965) observed that loweringinternal ionic strength in giant axons by reducing intracellular IC . from 300 mM to 24 mMwith a non-electrolyte shifted both the activation and inactivation curve for Na + and K +currents along the voltage axis in a depolarizing direction. They attributed this effect to thepresence of a fixed layer of negative charges on the inside of the membrane as these shiftsdid not occur when K + was replaced with an ionic species such as Na + chloride or cholinechloride (Chandler et al., 1965).4Membrane Surface Potential Can Be Described MathematicallyThe Gouy-Chapman theory of fixed surface charge, developed during the early partof this century, forms the basis for theoretical discussions of the effect of charge screeningon the function of voltage-dependent ion channels. The major assumptions of the Gouy-Chapman theory are: 1) the membrane surface potential results from a uniformly smeareddensity of fixed charge per unit area; 2) the dielectric constant in the aqueous phase isassumed to be a constant and equal to its bulk value; and, 3) the ions are assumed to bepoint charges, therefore, only the charge an ion carries and not the ionic species isimportant in an ion's ability to effectively screen membrane surface charges (Gilbert &Ehrenstein, 1969; McLaughlin et al., 1971; Hille, 1984). The Gouy-Chapman theory isuseful for quantitating in mathematical terms the influence of surface charge on thebehaviour of voltage-dependent ion channels. However, the Gouy-Chapman-Stern theory,a modification of the Gouy-Chapman theory, is more biologically relevant because itaddresses not only specific binding of cations to fixed negative surface charges but also theestablishment of a diffuse double layer of ions in the transition area between the bulksolution and the surface of the membrane through non-specific interactions between cationsand negative surface charges on the membrane (Hille, 1968).Both the Gouy-Chapman theory and the Gouy-Chapman-Stern theory have been usedto determine the type and density of charges surrounding voltage sensitive ion channels andto determine the effect of varying the ionic composition of solutions on the membranesurface potential (Gilbert & Ehrenstein, 1969; Hahin & Campbell, 1983; Hille, 1968; Hilleet al., 1975; McLaughlin et al., 1971; Mozhayeva & Naumov, 1972 a,b&c).Different Species Of Divalent Cations Are Not Equally Effective In Shifting The Voltage-Sensitive Parameters of Voltage-Gated Ion ChannelsEvidence from several studies suggests that the ability of divalent cations to shift thevoltage sensitive parameters both of Na + and I(+ currents depends on the ionic species5(Arhem, 1980; Blaustein & Goldman, 1968; Cukierman & Krueger, 1990; Hille et al., 1975;Mozhayeva & Naumov, 1972c). In most instances transition metal ions are much moreeffective than alkaline earth metal ions in shifting the voltage sensitive gating parametersof these currents along the voltage axis. This suggests there is some specificity in theinteraction between these divalent cations and the ion channels whose behaviour theyinfluence.Ca2+ and the transition metal ions appear to modulate the behaviour of voltage-activated K + currents by binding to sites either near or on the channel protein (Armstrongand Cota, 1991; Begenisich, 1988; Begenisich and Lynch, 1974; Blaustein and Goldman,1968; Cukierman and Krueger, 1990; Gilly and Armstrong, 1972, Hille et al., 1975). Bindingof these cations can in theory alter both channel conductance and gating kinetics.Zn2+ and Cd2+ slow the kinetics of the K + current, reduce the amplitude of the Na +current and shift the conductance-voltage relation when applied internally in squid giantaxon (Begenisich and Lynch, 1974) or externally to myelinated nerve from Xenopus laevis (Arhem, 1980). Gilly and Armstrong (1982 a&b) observed that changing the externalconcentration of Zn2+ slowed the opening kinetics of both the Na + and K + channels withoutaltering the closing kinetics. Rather than simply interacting with fixed negative surfacecharges on the external membrane, Gilly and Armstrong (1982 a&b) suggested that Zn 2+interacts directly with a negatively charged element of the gating apparatus and prevents thenegative charge from moving inward when the membrane depolarizes thereby stabilizing thechannel in the closed position.Begenisich and Lynch (1974) observed similar effects on the kinetics of Na l- and K +channels of squid giant axon when they increased the internal concentration of Zn 2+ , Co2+ ,Cd2+ or Ni t+. Internal Ca2+ concentrations of up to 10 mM had no effect on either current(Begenisich and Lynch, 1974). The latter study also concluded that transition metal ionsmust interact with membrane constituents involved in the control of gating and found thatthe K + current was more susceptible than the Na + current.6Some Species Of Divalent Cations Block Ion Channels In A Voltage-Dependent MannerIn some cases divalent cations reduce the conductance of ion channels by causing avoltage-dependent block of the current. Divalent cations from the alkaline earth metal andthe transition metal series were found to cause a voltage-dependent block of TTX- sensitiveand -insensitive Na+ channels (Schild et al., 1991; Ravindran et al., 1991). Woodhull (1973)observed that Na + channels in frog myelinated axon were blocked by protons and Ca 2+ ina voltage-dependent manner.Based on the decrease in Na+ conductance under conditions of low external pH,Woodhull (1973) and subsequent investigators have proposed that the proton binding siteis a titratable acid moiety, most likely a carboxylic acid, lying far enough across the electricfield of the membrane to be affected by the potential difference across the membrane(Campbell, 1982; Campbell & Hille, 1976; Mozhayeva & Naumov, 1981 & 1983; Sigworth,1980; Yatani et al., 1984). Others have suggested that the binding site responsible for thedecrease in Na + channel conductance lies near the outer surface of the channel and thatprotonation of this site reduces Na + channel conductance due to electrostatic interactionswhich reduce the concentration of Na + near the entrance of the channel (Drouin &Neumcke, 1974).Voltage-dependent block by divalent cations or protons does not precludecontributions arising from surface charge interactions. Ravindran et al. (1991) found theirresults were best fitted when a single divalent binding site was combined with the Gouy-Chapman theory of surface charge. Distinguishing between a voltage-dependent block andscreening of fixed surface charges is difficult at the macroscopic current level and bothmechanisms may be involved.7Molecular Characterization Of Voltage-Gated Ion ChannelsCharacterization, at the molecular level, of the proteins which form voltage-gated ionchannels has identified specific segments and in some cases individual amino acids withinthe proteins which play critical roles in determining the gating, conductance, ionic selectivityand pharmacological profile of these channels (reviewed in Betz, 1990 and Jan & Jan, 1989).Molecular sequencing and subsequent studies utilizing site directed mutagenesis haveidentified the fourth putative transmembrane segment (the S4 region), which is ubiquitousamong all voltage-gated ion channel proteins sequenced, as the most likely candidate for thevoltage sensor (Betz, 1990; Catterall, 1988; Jan and Jan, 1989; Kamb et al., 1987; Papazianet al., 1987 & 1988; Stiihmer, 1991). The S4 region has a positively charged lysine orarginine at every third amino acid position interspersed between mostly hydrophobic aminoacids (reviewed in Betz, 1990; Jan & Jan, 1989 & 1990; and, Unwin, 1989). Based on thepresence of the positively charged amino acids in the S4 region several groups have pointedto the involvement of this region in the transfer of charge across the membrane during thevoltage-dependent transitions which occur prior to channel activation (Benndorf, 1989;Catterall, 1988; Jan & Jan, 1989). Common to each of these models is the idea that duringor prior to channel opening conformational changes of the S4 region result in the nettransfer of positive charges across the electric field of the membrane from the intracellularto the extracellular side.As discussed previously, electrophysiological studies suggest that changes in divalentcation concentration can alter the electric field sensed by this S4 region and thus shift thevoltage-dependence of activation. Zn2+ and other transition metal ions also appear tointerfere specifically with the movement of the S4 region and thus the transfer of charge8across the electric field of the membrane (Gilly and Armstrong, 1982 a&b; Begenisich &Lynch, 1974).Is The Behaviour Of The Transient Outward Potassium Channel Influenced By DivalentCations?Most of the work regarding the effects of changing divalent cation concentration onthe behaviour of voltage-dependent channels has focused on either Na 4- channels or thedelayed rectifier type K + channel. Few investigators have addressed how changing divalentcation concentration will affect the behaviour of the transient outward potassium current(TOC) despite the fact that the transient outward potassium channel is the bestcharacterized K + channel at the molecular level.Connor and Stevens (1971) first identified the TOC, which they termed the A-current(IA), in the cell body of a gastropod neurone. Pharmacologically, the TOC is oftencharacterized by its sensitivity to block by 4-aminopyridine (Rudy, 1988). The gatingbehaviour of the TOC resembles that of the voltage-dependent Na + current. That is to say,the TOC activates quickly, (0.5-20 ms) and inactivates with a time constant averaging 50 ms(Rudy, 1988). The threshold for activation is around -60 mV for most cells and steady-stateinactivation is usually complete at about -40 mV (Rudy, 1988). The TOC often activatesonly when a cell is depolarized after a period of hyperpolarization (Hille, 1984). Theproposed role for the TOC is to prolong the interval between action potentials by opposingthe depolarizing effect of inward pacemaker currents (Hille, 1984; Rudy, 1988).Melanotrophs Possess A Transient Outward Potassium CurrentMelanotrophs comprise the majority of cells found in the pars intermedia of thepituitary and form part of the pro-opiomelanotropinergic endocrine system (reviewed inO'Donohue & Dorsa, 1982). In the rat, the pars intermedia is well defined and lies in thecleft between the posterior and anterior pituitary. Pro-opiomelanocortin (POMC) is the9common pre-cursor for all peptides synthesized and released by this system (O'Donohue &Dorsa, 1982).Melanotrophs secrete several peptide hormones, however, a-me lanocyte stimulatinghormone (a-MSH), which they stain for (Back & Rechardt, 1985) is the best known product.a-MSH stimulates melanin production and dispersion in melanocytes and also has severalextrapigmentary actions (O'Donohue Dorsa, 1982). B-adrenergic agonists stimulate anddopamine inhibits the release of a-MSH from melanotrophs (Douglas & Taraskevich, 1978;O'Donohue & Dorsa, 1982). Several other neural inputs are also postulated to regulatesecretions from the pars intermedia (de Rijk et al., 1990; Kehl et al., 1987; O'Donohue &Dorsa, 1982).Melanotrophs generate spontaneous action potentials which are predominantly Na +-dependent but which have a Ca 2 + component (Douglas & Taraskevich, 1982; Tomiko et al.,1981) and they possess voltage-gated Ca currents (Cota, 1986; Kehl, 1987) which appear tobe involved in stimulus-secretion coupling (Tomiko et al., 1981).Kehl (1989) characterized two voltage sensitive outward K + currents in culturedmelanotrophs from the pars intermedia of adult rats. One is a delayed rectifier type K +current, IK(s), which activates slowly and inactivates slowly (Kehl, 1989). The other, ofinterest in the context of the present study, is a TOC, I K(f), which is sensitive to block by4-aminopyridine (Kehl, 1990). The activation threshold for I K(f), determined in highexternal Ca2+ (10 mM), was between -20 and -10 mV with steady-state inactivation completeat -10 mV (Kehl, 1989). The time constant for inactivation was between 20-35 ms and waswell fitted by a single exponential (Kehl, 1989).Experimental RationaleKehl (1989) observed that Cd 2+ (5 mM) or Co2+ (10 mM) reduced the peakamplitude and increased the 50% rise time of I K(f). Mayer and Sugiyama (1988) reportedsimilar effects of divalent cations on IA, a TOC in cultured dorsal root ganglion cells from10the rat. They observed that divalent cations evoked a right-ward shift of the activation andinactivation curves along the voltage axis, which accounted for the reduction in current, andsuggested that the divalents exert this effect by binding to the channel protein (Mayer &Sugiyama, 1988).One of the major goals of the present study was to elaborate on the findings of Kehl(1989) and to examine in greater detail the effect of varying the external concentration ofCd2+ , Zn2+ , Ca2+ and Mg2+ on the behaviour of IK(f). The aim was to determine if thesecations interacted directly with the channel protein as appears to be the case in rat sensoryneurones (Mayer & Sugiyama, 1988) or if their effects arose indirectly through non-specificscreening of fixed negative surface charges.11METHODS1. Preparation Of Acutely Dissociated MelanotrophsMelanotrophs were obtained from adult male Wistar rats (200-300g). Prior todecapitation the rats were exposed to CO 2 until unconscious. The excised pituitary glandwas placed in a 1:1 mixture of Ham's F-12 and Dulbecco's Modified Eagle's Medium(DMEM) at room temperature and the neurointermediate lobe which consists of the parsintermedia and the pars nervosa was separated from the pars anterior. Theneurointermediate lobe was then incubated in a 35° C water bath for 25-50 min in 0.5 mlCa2+ , Mg2+ -free phosphate-buffered saline (CMF-PBS) containing collagenase (type V; 1mg/ml :=-435 active units (U)/ml) and hyaluronidase (type II; 1 mg/ml 410 U/ml) andfor 10-25 min in 0.5 ml CMF-PBS containing protease (type VIII; 0.5-1 mg/mlP.', 5-15U/ml). Following enzymatic treatment the tissue was transferred to 0.5 ml of CMF-PBSand mechanically dispersed by trituration through progressively smaller syringe needles (18,21, 23 & 26 gauge). The cell suspension was then spun at approximately 50 g for 10 minon a column of CMF-PBS containing 10% (w/v) bovine serum albumin (BSA; fraction V).The supernatant was removed and the cells were resuspended in culture media comprisedof Ham's F-12:DMEM (1:1) and kept in an atmosphere- controlled (95% air : 5% CO 2 ),humidified incubator at 37° C until use (1-15 hours).For each experiment an aliquot of the cell suspension was transferred to a Perspexchamber mounted on the stage of an Olympus inverted phase contrast microscope. Thecells were viewed at 600X magnification. Melanotrophs were identified as those cells whichexhibited phase-bright membranes, contained dark inclusions (assumed to be the nucleus)and were 10-15 Am in diameter. Occasionally there were pleomorphic cells present whichwere assumed to be pituicytes and small enucleate vesicles which were assumed to be eitherdetached nerve terminals from the pars nervosa or vesicles formed spontaneously from thelipid of ruptured membranes.122. ElectrophysiologyMacroscopic ionic currents were recorded via conventional patch-clamp techniquesutilizing the whole-cell voltage clamp configuration (Hamill et al., 1981). A LIST EPC-7patch-clamp amplifier was used to measure the whole-cell currents which were then encoded(Instrutech VR-10, Elmont, NY) and stored on video cassette tape (-3 dB at 10 kHz). Thevoltage commands were derived from a custom-built digital potentiometer controlled by acustom-built digital timer. All experiments were conducted at room temperature (20-25 °C). Control, test and recovery responses were determined for each cell examined so thateach cell functioned as its own control.Patch electrodes made of borosilicate glass (Corning No. 7052, A-M Systems, WA,USA) were pulled on a Narishige PP-83 two stage puller. The electrode tip outer diameterwas 1.5-2 Am prior to fire polishing. The signals were referenced to an agar salt bridge (4%agar by weight in 150 mM NaCI) and the zero-current voltage was set on the EPC-7 oncethe electrode tip was immersed in the bath solution. The resistance of the fire-polishedelectrode tip, measured in the external control solutions, varied from 3-7 Mn. The tip wascoated with Sylgard (Dow Corning) to reduce the capacitive transients.The formation of a tight (gigohm) seal and the establishment of the whole-cellconfiguration was usually accomplished in the external control solutions described below.In some cases a modified external solution containing 10 mM Ca 2+ was used to obtain thewhole-cell configuration as high Ca 2+ appeared to promote tight seal formation. As soonas the whole-cell recording was established the external solution was changed to the externalcontrol solution.Capacitive transients arising from the cell and the pipette were compensated usingcircuitry incorporated in the EPC-7. The series resistance (R s) for the whole-cell recordingwas typically 16-25 NIn and was compensated only if the macroscopic currents were large.Rs compensation was usually 30-60%, which, with a 1 nA current (upper limit of those13examined), would result in at most a 17 mV error in the voltage command. There was nocompensation for any such error.The stimulus frequency was 0.25 Hz and the holding potential (V H) was -80 mVunless otherwise noted.3. Data Acquisition and AnalysisFor analysis, current signals were passed through a 4-pole low-pass Bessel filter witha -3 dB cut-off frequency of 2 kHz and digitized at a sampling frequency at least twice thefilter cut-off frequency using a 12-bit analog/digital converter (Scientific SolutionsLabmaster, OH) interfaced with an AT clone. Current and potential recordings wereanalyzed off-line with an Intel-based computer using BASIC-Fastlab routines (INDECSystems, Sunnyvale, Calif.).IK(f) was separated from the leak and residual I K(s) currents using a two-pulseprotocol. When a two pulse protocol was used the first pulse is referred to as theconditioning pulse and the second pulse in referred to either as the conditioned or non-conditioned test pulse. No less than three current traces were averaged for eachconditioned and non-conditioned test pulse. A non-conditioned 350 ms test pulse from theholding potential to -40 mV or more elicited an outward current composed of IK(f), residualIK(s) and leak current. Since the time dependence of inactivation determined at -40 mV waswell fitted by a single exponential with a time constant (r) of 54.7 ms a conditioning pulselength of 600 ms (approximately 11 time constants) was judged sufficient to virtuallyeliminate IK(f). Consequently it was possible to isolate I K(f) by subtracting the averagedconditioned pulse from the averaged non-conditioned pulse. For measuring purposes IK(f)was defined as the difference between the peak outward current and the steady-statecurrent.Curve fitting was done by using non-linear regression routines in BASIC-Fastlab.Values are expressed as the mean ± the standard error of the mean (s.e.m.) and "n"14represents the number of cells studied. Statistical analysis of the results was completedusing SAS (Cary, NC). Where there were more than two concentrations of a given divalentcation being examined, analysis of variance (ANOVA) was employed with p =0.01. If onlyone concentration of a given divalent cation was tested, significance was determined by usinga paired t-test, again with p =0.01.4. Recording SolutionsThe compositions of the external solutions are summarized in Table 1. The pH ofeach of these solutions was adjusted to 7.4 with 1 M NaOH. Tetraethylammonium chloride(TEA) was included to block the slowly-activating, slowly-inactivating K+ current (I K(s);Kehl, 1989), and tetrodotoxin (TTX; 1-2 AM) was included to block inward Na + currents(McBurney & Kehl, 1988). Cd 2+ was not included in the standard external solution to blockCa2+ currents because even at the low concentrations (300 AM) necessary, Cd 2 + affected thekinetics and conductance-voltage relation of I K(f). Inward Ca 2+ currents are normally muchsmaller than the outward K + currents and in almost all cells examined inward Ca currentswere not detected.In some cases sucrose was added to the external solutions to make themhyperosmotic relative to the pipette solutions as this appeared to aid in tight seal formationand in prolonging the membrane integrity of the cell once whole-cell recording wasestablished.For some of the experiments a three-buffer system was utilized in which 10 mMHEPES was replaced by 5 mM HEPES, 5 mM CHES (2-[N-cyclohexylamino] ethanesulfonicacid) and 5 mM propionic acid-Na salt. In these cases a three-buffer system was alsoemployed in the pipette-filling solution (see below). The three-buffer solutions wereprepared in order to conduct pH experiments which are not discussed here. I K(f) recordedin the three-buffered solutions at pH 7.4 exhibited the same behaviour as in HEPES15buffered solutions at pH 7.4. For subsequent divalent cation experiments the moreconventional HEPES buffered solutions were prepared.The standard patch pipette solution contained (in mM): 140 KCI, 5 MgCI,-6H 20, 1CaCl2, 10 HEPES, 11 EGTA (ethyleneglycol-bis-(B-aminoethyl ether)-N,N,N'N'-tetraaceticacid), 1 Na,ATP; pH was adjusted to 7.4 with 1M KOH, pCa > 8. For the experiments inwhich a three-buffer system was used, 10 mM HEPES in the pipette solution was replacedby 5 mM HEPES, 5 mM CHES and 5 mM propionic acid-Na + salt. The standard HEPESbuffered pipette solution is designated "1" and the three-buffer pipette solution is designated"3" in the figure legends. Pipette solutions were passed through a 0.2 Am filter prior tofilling the electrode. The term "pipette solution" is used interchangeably with "internalsolution" because in whole-cell recording the pipette solution diffuses into the cell tobecome the internal solution.The following convention will be used to describe the solutions used for eachexperiment in the figure legends: external///internal. External solutions are referred to bythe letter assigned to them in Table 1 and the concentration (in mM or AM) of the divalentcation of interest will be given. The pipette solutions will be referred to by their numbercode as described above. If the external solution was changed during the experiment thefollowing convention will be used in the figure legends: control//test [divalentcation]///internal.The Na salt of Phenol Red (200 AM; Sigma, tissue culture grade) was added to thetest solutions so that solution exchanges in the recording chamber could be monitoredvisually. Solution exchange was accomplished using a stop-flow system in which the testsolution flowed into one end of the recording chamber while solution was drawn off at theother end of the chamber at a rate sufficient to maintain a constant fluid level in the bath.Solution exchange continued until the solution in the bath was completely replaced and anychanges in the behaviour of the current, evoked as a result of changes in the solution16composition, had reached a steady-state - usually within 1-2 min - after which the flow wasstopped and the recording continued.Chemicals were obtained either from Sigma or Aldrich. Solutions were stored at 4°C and allowed to warm to room temperature prior to use.17Zcn1/40tocnoNNIHIll 01-1 N o.4.I-1CIN1-IInr)0NNIHIn 0HN o0NT--1te)el0NNIHU.) 0H N HI-11/40NHIn.CI0NNI1-iIn In tn in I 0Li-)H O(-1HtrINHLO.r0NNiHIn Ln LC) In 0 H OcnH InZoNHtoCIoNNIHIn 0H NHcn 	 •0 00I0\01--1.00NUI 0NNIIn 01-1 NH 0\0HHtoCI H44•NHt--.1u",CIoNNIHin 0H N •-•1 inInPI1•NHtor)oNNIH10 0HN H LU,-1H0VDNHIn.el0NNiHin In In to o 1--1 N OM00NHIn•el0NNHIn Ul inIIn N H H No 1OMCOONto•oNNIin in in In N H OcnH CI H0 In 0 N In 0 N r--I4 N • N I HH CI HW 0o 	 Z 	 -H(1)U) 	la0 ZE•I 	 v 	 0 	 (1) 	 •H I"Ca H 	 0 14 u) 	 ai ro HPa 0 H 4 x o a 4) o -H UO 0 	 (,) 	 LL1 	 E-I 	 ,-1 	 ril 	 Z 	 0 	 faCO Z 	 El PI 0 x 0 (:14 4 0CM 	 CV 	 CMH H HO 0 	 0b' 	 rliZ o N0)u)O0oU)18RESULTSSECTION I: GENERAL PROPERTIES OF IK(.01. Voltage- and Time-Dependence of ActivationThe activation threshold for I K(f) under control conditions was between -40 and -30mV and the peak amplitude of the current increased monotonically with increasingmembrane depolarization (Fig. la,b) as previously reported (Kehl, 1990). From -30 mV to+60 mV the current-voltage relation was non-linear but approached linearity at membranepotentials above +60 mV as the maximum conductance (G.) was reached (Fig. lb).The data points (filled circles) in Fig. ld indicate the normalized chord conductance(g) derived from the I K(f) at each test potential. The reversal potential of -65 mVdetermined in standard control solution (discussed in detail in Section I part 4) was usedto calculate the chord conductance. The curve fitted to the data points by a least squaresregression routine to the Boltzman equation,g/G. = [1 + exp((V'-V)/k)] -1 ,indicated that half-activation (V') occurred at -13.6 mV with a slope-factor (k) of + 12.8 mV(n =55).The slope of the conductance-voltage relation, which represents the voltage-sensitivityof activation gating, can be converted to an equivalent gating charge by solving for Z, theeffective valency of the gating particle, in the relationship,k = KT/Ze,where k is the slope-factor, K is Boltzman's constant (1.38 x 10 -23 J/K), T is the absolutetemperature and e is the size of the electronic charge (1.6 x 10 -'9 C). At 20°C, KT/e isapproximately 25 mV. Solving for Z indicates that an equivalent of 2 gating charges mustmove completely across the membrane field during activation.19The rate of rise of I K(f) was strongly influenced by depolarization (Fig. la). Undercontrol conditions, the 50% rise time (t y, a„) decreased from 2.8 ± 0.1 ms (n = 20) at -20 mVto 0.8 ± 0.1 ms (n =30) at + 60 mV in agreement with previous results (Kehl, 1989).2. Voltage- and Time-Dependence of InactivationSteady-state inactivation was analyzed by using a two-pulse protocol in which the cellwas held at -80 mV and stepped to 0 mV for 350 ms immediately after a 600 msconditioning pulse of varying intensity. The conditioned test currents for one cell are shownin Fig. lc. In Fig. id the peak current measured during the conditioned test pulse is plottedagainst the pre-pulse potential. The curve fitted to the data points (open circles) by a leastsquares regression routine represents the solution for the Boltzman equation,I/I. = [1+ exp((V-V)/k)] -1 ,and indicates that half-inactivation (V') occurred at -54.7 mV. As discussed previously, theslope-factor (k) of -4.4 mV (n=50) reflects the voltage sensitivity of inactivation gating.Determination of the equivalent gating charge for inactivation indicates that the equivalentof 6 gating charges moves across the membrane electric field during inactivation suggestingthat inactivation is more steeply voltage-dependent than activation.At -20 mV, the decay of IK(f) was well fitted by least squares regression analysis toa single exponential with a time constant of decay (r) of 25.8 ± 1.1 ms (n = 10) as shown forone cell in Fig. 2a. At potentials above -20 mV, the decay of I K(f) was often better fittedby two exponentials as illustrated for one cell in Fig. 2b. The slow component at + 60 mVwas 66.2 ± 1.8 ms (n= 10) and accounted for 66 ± 1.6% of the decay, whereas the fastcomponent at + 60 mV was 14.2 ± 2.3 ms (n=10) and accounted for 34 ± 1.6% of thedecay.20Figure 1. The current-voltage relation and steady-state inactivation of I K(f) under controlconditions. A. Whole-cell currents recorded during 350 ms steps to -20, 0, +20, +40 and+ 60 mV from a holding potential of -80 mV. B. The current-voltage relationship for I K(f).Data points represent the peak amplitude of IK(f), as defined in the methods, at eachpotential. C. Superimposed currents during test pulses to 0 mV following 600 msconditioning pulses to -100, -80, -60, -55, -50, -45 and -30 mV; the current traces followingpulses to -100 and -80 mV are superimposed. The current traces were not leak subtracted.D. The control activation (filled circles) and steady-state inactivation (open circles) curvesfor IK(f). For the activation curve the normalized conductance was plotted against themembrane potential during the test pulse. The curve represents the solution to theBoltzman equation where the potential for half-activation was -15.5 mV and the slope-factorwas 10.3 mV. The inactivation curve is the normalized test current plotted against themembrane potential during the conditioning pulse. The half-inactivation potential was -55.6mV and the slope-factor was -3.7 mV. The straight line in A and C represents the zerocurrent level. Cell 160492B. Recording solutions E///1 (see Table 1).5W223. Voltage-Dependence of the Residual Steady-State CurrentThe magnitude of the time-dependent residual steady-state current (e.g., Fig. 1d)increased with membrane depolarization. This current probably arises from I K(s) channelswhich are not blocked in 20 mM TEA. Some evidence suggests that the block of delayedrectifier type K+ channels by TEA is voltage-dependent (Hille, 1984; Clay, 1985) and thismight account for the observation that the residual steady-state current increases withdepolarization.4. Reversal Potential For IK(f)The reversal potential for I K(f) was determined by using a two pulse protocol. Aconditioned pulse to 0 mV for 10 ms maximally activated I K(f) but produced very littleinactivation of the current. This pulse was immediately followed by a test pulse to thepotentials indicated beside the current traces in Fig. 3. At maximal activation a majorityof the channels are open and the test pulse will drive current through the open channel ina direction which is dictated by the difference between the membrane potential and theequilibrium potential for the current. The potential at which no time-dependent current isevident is taken to be the reversal potential (E R). This was measured either directly or byinterpolation. It should be noted that a time-independent current (the leak current) alsocontributes to the tail current. ER varied from -60 to -70 mV among the cells examined.For the purposes of calculating the chord conductance an ER of -65 mV was chosen as thisvalue fell between the extremes. EK, calculated from the Nernst Equation to be -93.2 mV,was substantially more hyperpolarized than ER. This suggests that the channel conductingIK(f) is not completely selective for K.23Figure 2. The time constants for the inactivation of IK(f). A. The decay of the currentevoked at -20 mV was well fitted by a single exponential with T = 25.07 ms. B. The decayof the current at +60 mV was better fitted by two exponentials. The slow component (r si,,,,,)was 61.4 ms and accounted for 63% of the current decay and the fast component (r fast ) was14.1 ms and accounted for 37% of the current decay. The curve representing the sum ofthe fast and slow components of decay overlays the current trace. Separate curvesrepresenting the fast and slow components are illustrated by the dashed lines. Cell140492A. Recording solutions E///1.2425Figure 3. The reversal potential for IK(f) as determined from tail currents following a 10ins pulse to 0 mV. The potential at which each current was recorded is indicated beside thetrace. The trace which exhibited no time-dependent current illustrates the ER. Theestimated ER is indicated by an arrow. Vh was -80 mV. Cell 140591A. Solutions: B///3.-50-60-65--■ -70-75-80-1002627SECTION II: THE EFFECTS OF DIVALENT CATIONS ON I KLODivalent cations have been shown to shift the activation and inactivation curves ofvoltage sensitive Na 4. and K + channels. This effect of divalent cations is attributed to eitherthe non-specific screening of fixed negative surface charges or the specific interaction ofthese cations with sites close to or on the channel protein. The purpose of the present studywas to elaborate on the observations of Kehl (1989) and to determine whether divalentcations affect I K(f) in a manner analogous to that reported for I A, a TOC in rat sensoryneurones (Mayer & Sugiyama, 1988).1. THE EFFECTS OF TRANSITION METAL IONS ON THE GATING OF I KkO1.1 CadmiumUsing a single concentration of Cd 2+ (5 mM) Kehl (1989) observed that there wasa reduction of the peak amplitude of IK(f) and an increase of the 50% rise time of thecurrent. This effect was examined further over a range of Cd 2+ concentrations.1.11 Cd2+ Reduces The Peak Amplitude Of I K(f)The concentration-dependent reduction of the peak amplitude of I K(f) by Cd2+ (100AM to 1600 MM), is illustrated in the current-voltage relations summarized in Fig. 4a wherethe data points represent the means from 5 cells. The reduction of peak I K(f) by Cd2+ wasparticularly noticeable at 0 mV and traces illustrating the dose-dependent decline of I K(f)at this potential are presented in Fig. 4b. Overall, the peak amplitude of I K(f) at 0 mV wasreduced 25 ± 2.6%, 37 ± 2.2%, 50 ± 2.4%, 63 ± 2.0% and 73 ± 1.5% by a Cd 2+concentration of 100, 200, 400, 800 and 1600 MM, respectively, (n =5; e.g., Fig. 4b).The effect of Cd2+ on the peak amplitude of I K(f) was less pronounced at moredepolarized potentials. The relationship between the reduction of I K(f) at + 60 mV and theexternal concentration of Cd 2+ is illustrated in Fig. 4c. The line fitted to the data pointsrepresents the best fit to the Michaelis-Menton equation, R = R./(1 + (Ka[Cd21)n),28Figure 4. The concentration-dependent effects of Cd 2 '. A. Normalized current-voltagerelations for 1K(f) in control (open circles) and in the presence of 100 AM (crosses), 200 AM(diamonds), 400 AM (squares), 800 AM (triangles) and 1600 AM Cd 2+ (filled circles). Datapoints represent the mean for five cells. 1 K(f) was normalized against the maximum currentobtained in control. B. Superimposed whole-cell currents evoked by a 350 ms test pulseto 0 mV from the holding potential illustrating the dose-dependent reduction of 1 K(f) byCd2+ . Also apparent from these traces is the slowing of the activation rate as theconcentration of Cd 2+ is increased. The concentration of Cd 2+ is indicated beside theindividual traces. The straight line below the current traces identifies the zero current level.Cell 040991C. C. A concentration-response curve comparing the reduction of normalizedIK(f) (1-Ino,„) to Cd 2 + concentration. Data points represent the mean ± s.e.m. for five cells.The curve fitted to the data points is a solution to the Michaelis-Menton equation assuminga single binding site where the maximum current reduction is 38.4% and the K m for theeffect of Cd2+ is 256 MM. D. The dose-dependent shift of the activation curve by Cd 2+ inwhich normalized conductance is plotted against the test potential for each concentrationof Cd2+ (symbols represent the same Cd 2+ concentrations as for A above). The chordconductance was determined at each potential in each concentration of Cd 2 +. Curves werenormalized to the maximum conductance at each concentration of Cd 2+ to correct forpossible channel block (Perozo & Bezanilla, 1990). Curves fitted to the data pointsrepresent solutions to the Boltzman equation where the potentials for half-activation, inascending order from control to 1600 AM Cd 2 +, were -11.9 mV, -5.0 mV, -0.9 mV, 4.1 mV,10.1 mV and 12.3 mV, respectively. The slope-factor for these activation curves did notappear to be altered by the presence of Cd 2+ and was approximately 12.4 mV. Cell040991C. Recording solutions B//C: [Cd 2+ ] indicated below///3.29C; 0(..) 	 g0-11-1to 	030where n is assumed to be 1. From the fitted curve, the reduction (R) of the normalizedcurrent at + 60 mV saturated (Rmax) at 38.4% and the concentration of Cd2 + at which half-maximal reduction occurred (KM ) was 255.8 AM (n =5). For the five cells examined, thereduction of IK(f) by Cd2+ at + 60 mV ranged from 13 ± 2.3% in 100 AM Cd2+ to 35 ± 2.3%in 1600 AM Cd2+ (Fig. 4c).Following the analysis of the dose-response to Cd2+ , the effects of 500 AM, 2 mMand 5 mM Cd2+ on IK(f) were more thoroughly examined. At a concentration of 500 M,Cd2+ caused a rightward shift of the current-voltage relation for I K(f) along the voltage axis.The reduction of peak IK(f) and the shift of the threshold for activation from -40 mV to -20mV caused by 500 µM Cd2+ are illustrated for a representative cell in Fig. 5a. Overall, 500AM Cd2+ reduced the peak I K(f) by 50 ± 3.8% at 0 mV and 9 ± 2.2% at +60 mV (n =5) andcaused a slight, 4 ± 2.7% (n=5), reduction of G.. For the cell shown in Fig. 5a G max wasreduced from 13.6 nS to 12.3 nS. The reduction of peak I K(f) and G. in the presence of500 AM Cd2+ is most likely due to the right-ward shift of the activation curve. The effectsof 500 AM Cd 2+ were completely reversed within two minutes of returning to Cd 2+ -freecontrol solution (e.g., filled circles; Fig. 5a).As expected from the apparent saturation of the Cd 2+ effect at 1.6 mM in the dose-response experiments (see Fig. 4c), there were no measurable differences in the effectsexerted by 2 mM and 5 mM Cd 2+ on IK(f). For this reason, only the results for 5 mM Cd 2+are discussed in detail here. Fig. 5b illustrates the effect of 5 mM Cd 2+ on a representativecell. At a concentration of 5 mM, Cd 2+ shifted the threshold for activation of I K(f) from -40mV to 0 mV. In the 8 cells tested the peak amplitude of the current was decreased 90 ±2.0% at 0 mV and 30 ± 4.2% at + 60 mV and G. was reduced by 23 ± 4.4% (n =8) in thepresence of 5 mM Cd2+ . For the cell illustrated in Fig. 5b, G. decreased from 9.8 nS to7.2 nS.31Figure 5. The effects of 500 AM and 5 mM Cd 2+ on the voltage-dependence of activationand inactivation for IK(f). A. Effect of 500 AM Cd 2 + on the normalized current-voltagerelation for IK(f) in control (open circles), 500 AM Cd 2+ (diamonds) and recovery (filledcircles). Cell 140492C. Solutions: E//E&F combined to [Cd 2 1=500 AM///1. B. As forA but with 5 mM Cd2+ . Solutions: E//F///1. C. The activation and steady-stateinactivation curves for IK(f) in control (filled and open circles, respectively) and in thepresence of 500 AM Cd 2+ (filled and open diamonds, respectively). Same cell as in A. V'was -11.5 mV in control and shifted to +9.6 mV in the presence of 500 AM Cd2+ . Theslope-factor increased from a control value of 11.5 mV to 12.2 mV. From the inactivationcurves, V' was -46.5 mV in control and shifted to -27.6 mV in control. In the presence of500 AM Cd2+ there was little change in the slope-factor which was -3.9 mV in control and-3.4 mV in 500 AM Cd2+ . D. As for C but with 5 mM Cd 2+ . Same cell as in B. The half-activation potential, was -10.9 mV in control and shifted to +24.5 mV in 5 mM Cd 2+ . Theslope-factor remained relatively constant, with a value of 11.4 mV in control and 11.2 mVin the test conditions. For steady-state inactivation, the half-inactivation potential shiftedfrom -50.1 mV in control to -15.3 mV in 5 mM Cd 2+ . The slope-factor was -3.4 mV underboth conditions.CO'.;E2WA32EEon'a. 64o .42A<EWC.)ELU331.12 Cd2+ Causes A Right-Ward Shift Of The Activation CurveThe voltage-dependent reduction of I K(f) observed in the presence of Cd 2+ isexplained in part by the Cd 2+ -induced right-ward shift of the activation curve along thevoltage axis. The values for V' and k, (estimated as described in Section I), weredetermined in each concentration of Cd 2+ tested and are summarized in Table 2 (pp 42-43).The magnitudes of the shifts of the half-activation potential (V') with respect to Cd 2+concentration are summarized in Table 6 (pp 79-80).The progressive right-ward shift of the conductance-voltage relation as theconcentration of Cd 2+ was increased from 100 AM to 1600 AM is illustrated for one cell inFig. 4d. In the concentration-response experiments (n =4), V' ranged from -12.9 ± 0.6 mVin control (e.g., circles, Fig. 4d) to -4.4 ± 0.5 mV in 100 AM Cd 2+ (e.g., crosses, Fig. 4d) and+ 11.4 ± 1.3 mV in 1600 AM Cd2+ (e.g., filled circles, Fig. 4d; values for other concentrationsare summarized in Table 2). The slope-factor of the conductance-voltage relation increasedfrom the control value of 15.7 ± 0.3 mV to 11.4 ± 0.2 mV in 100 AM Cd2+ , the lowestconcentration of Cd2+ tested; however, although there was some variability in k values atother Cd2+ concentrations, increases of the Cd 2+ concentration beyond 100 AM did notfurther increase the slope of the conductance-voltage relation (see Table 2).In 500 AM Cd2+ , V' for the conductance-voltage relation shifted + 24.6 ± 2.3 mV(p < 0.01), from -13.7 ± 1.6 mV in control responses (filled circles) to 10.9 ± 1.7 mV (n=5;filled squares) (Fig. 5c). The slope-factor (k) did not change in the presence of 500 AMCd2+ (see Table 2).At a concentration of 5 mM, Cd2 + shifted the activation curve +29.2 ± 3.7 mV (n=7;p <0.01; see Table 6), as V' shifted from a control value of -7.4 ± 2.6 mV (filled circles, Fig.5d) to + 21.8 ± 1.1 mV (filled diamonds, Fig. 5d). The slope-factor estimated from theactivation curve decreased from 12.3 ± 0.5 to 11.1 ± 0.4 (Table 2) in 5 mM Cd2+.341.13 Concentration Dependence For Cd2+ -Induced Shifts Of V'The relationship between the shift of V' (estimated as described in Section I, Fig. 1)and the external Cd2+ concentration is illustrated in Fig. 6. The line fitted to the data pointsrepresents the best fit to the Michaelis-Menton equation,S = S./(1 + (Km/[Cd21)"),in which n is assumed to be one. From the fitted curve, the shift (S) of V' for theconductance-voltage relation, caused by Cd 2+ , saturated (S.) at +28 mV and theconcentration of Cd 2+ at which the half-maximal shift occurred (K M) was 221 AM. Thisvalue for KM is in close agreement with the KM value (256 AM) estimated from thereduction of peak 1K(f) by Cd2+ .1.14 Cd2+ Increases The 50% Rise Time (t) Of I K(f)The time to half-activation was compared since the precise timing of the current peakwas ambiguous in the presence of Cd 2+ (especially at potentials close to the activationthreshold) and this made an accurate determination of time to - peak IK(f) difficult. The 50%rise times of IK(f) in the presence of 500 AM (diamonds) and 5 mM Cd 2+ (squares) andtheir respective controls (open and filled circles) were compared over a 60-80 mV range andare summarized in Fig. 7a (pp 40-41). As the membrane potential was increased to moredepolarized levels there was a decrease in the rise time in control and treated responses.In control conditions, there was approximately a 4-fold decrease of ty2 act at +60 mV (0.7 ±0.1 ms & 0.6 ± 0.1 ms) compared to t y, act at -20 mV (2.6 ± 0.1 ms). Cd2 + appeared toaugment these differences. In 500 AM Cd 2+ , ty2 act decreased approximately 8-fold from 10.1± 1.4 ms at -20 mV to 1.2 ± 0.7 ms (n=5) at +60 mV and in 5 mM Cd 2+ , ty, act decreasedapproximately 11-fold from 16.5 ± 2.9 ms at 0 mV to 1.6 ± 0.1 ms (n=5) at +60 mV.35Figure 6. Concentration-dependent effects of Cd 2+ on the shift of the half-activationpotential. Voltage shifts of the half-activation potential as summarized in Table 6 are plottedas a function of the Cd2+ concentration. Data points represent the mean ± s.e.m. for thenumber of cells tested at each concentration of Cd 2+ , "n" is indicated in brackets on thefigure. A concentration-response curve fitted to the Michaelis-Menton equation indicatesa maximum Cd 2+ -induced shift of +28 mV and a K M of 220 AM.IIIIIIIIIIT361.0	 2.0	 3.0	 4.0Cce+ concentration (mM)37The Cd2+ -induced decreases in the activation rate of IK(f) are readily apparent fromthe current traces of Fig. 7c,d which show superimposed the control current traces and thecurrent traces for 500 AM Cd2+ and 5 mM Cd 2+ , respectively. The current trace recordedin 500 AM Cd2+ at 0 mV was scaled up by 2.25 in order to allow a comparison of the risingphase of the current. In 5 mM Cd2+ current traces at + 20 mV were compared because theactivation threshold for I K(f) was shifted to 0 mV. The current trace recorded in 5 mMCd2+ at +20 mV was scaled up by 2.71.At membrane potentials close to the activation threshold, Cd 2+ substantially slowedthe 50% rise time of I K(f). At -20 mV, 500 AM Cd 2 + increased Liz act 3.9-fold, from 2.6 ± 0.1ms in control to 10.1 ± 1.4 ms (n=5) and at 0 mV, 5 mM Cd 2 + increased t act 9.1-fold from1.8 ± 0.3 ms in control to 16.5 ± 2.9 ms (n=5). At +60 mV, 500 AM Cd 2+ increased ty, act1.7-fold and 5 mM Cd 2+ increased t act 2.6-fold compared to control values (n=5).1.15 Cd2+ Causes A Right-Ward Shift Of The Steady-State Inactivation (h.) CurveThe effects of Cd2+ on steady-state inactivation (h.) were examined at 500 AM, 2mM and 5 mM Cd2+ and the values for V' and k, (estimated as described in Section I) aresummarized in Table 2. The magnitudes of the shifts of V' for the h o, curve with respectto Cd2+ concentration are summarized in Table 6. As previously mentioned, the effects of2 mM Cd2+ on IK(f) were the same as those reported for 5 mM Cd 2+ .Steady-state inactivation was examined by using a two pulse protocol as discussed inSection I. The data points, for the responses in control (open circles) and 500 AM Cd 2+(open diamonds) are summarized for one cell in Fig. 5c. Overall, curves fitted to the datapoints (as described in Section I) indicated the h, curve shifted +21.2 ± 3.9 mV (n=5;p<0.01) from a V' of -52.4 ± 1.9 mV in control to -31.3 ± 2.0 mV in 500 AM Cd 2+ . Theslope-factor of the inactivation curve decreased from -4.1 ± 0.2 to -3.4 ± 0.1 (see Table 2)which reflects an increase in the voltage-sensitivity of inactivation.38The data points, for the responses in control (open circles) and 5 mM Cd 2+ (opendiamonds), comparing the conditioning pulse potential to the peak current elicited duringa subsequent test pulse to +30 mV are summarized for one cell in Fig. 5d. The 11,,, curve,fitted as described in Section I, shifted +32.8 mV (n=4; p<0.01) with V' increasing froma control value of -48.4 ± 3.6 mV to -15.6 ± 0.3 mV in 5 mM Cd 2+ . The slope of theinactivation curve did not change with 5 mM Cd 2 + present (see Table 2).1.16 Cd2+ Increases The Time To Half-Inactivation (4 inact)The times from the peak of the current to half-inactivation were compared ratherthan the time constants for decay because at some potentials the decay was well fitted bytwo exponentials and at other potentials by a single exponential as illustrated in Section I.In Fig. 7d the t1/2 inact times are plotted against the membrane voltage for 500 AM Cd 2+(diamonds) and its control (open circles) and from 0 mV to +60 mV for 5 mM Cd2+(squares) and its control (filled circles).In general, the Cd2+ -induced increase in t y, inact was not as great as the increase oft1/2 act. Moreover, the increase of t1/2 inact by Cd2+ did not appear to be concentration-dependent in the range of concentrations used, as shown in Fig. 7b. In the presence of 500AM Cd2+ , ty, inact at 0 mV increased 1.6-fold from 20.9 ± 0.8 ms in control to 29.2 ± 0.9 ms(n=5) and in 5 mM Cd 2+, inact increased 1.9-fold from 18.6 ± 1.4 ms in control to 36.0 ±6.3 ms (n =5). At +60 mV, 500 AM Cd2+ increased t-1/2 inact 1.2-fold from a control value of30.5 ± 1.4 ms to 36.3 ± 2.6 ms and 5 mM Cd 2 + increasedt Y2 inact 1.3-fold from 29.2 ± 1.0 msin control to 39.0 ± 2.5 ms (n=5).1.2 ZincZn2+ has been reported to exert effects on voltage-gated channels in a mannerqualitatively similar to that of Cd2+ (Arhem, 1980; Begenisich & Lynch, 1975; Gilly &39Armstrong, 1982 a&b). Therefore, the following experiments were undertaken to examinethe effect of Zn2+ on IK(f).1.21 Zn2+ Reduces The Peak Amplitude Of I K(f)The normalized current-voltage relations from three representative cells in Fig. 8illustrate the concentration-dependent reduction of I K(f) by Zn2+ . The reduction of I K(f)was less pronounced at more depolarized potentials as was the case for Cd 2+ . An externalZn2+ concentration ([Zn210) of 31 AM (e.g., squares, Fig. 8a) reduced I K(f) by 27 ± 1.8%at 0 mV and by 16 ± 1.5% at +60 mV (n =3). (In one cell examined at 31 AM Zn 2+ therewas a 9% decrease at 0 mV and a 7% increase in IK(f) at +60 mV; this cell was notincluded in the calculation of the mean). Raising the [Zn 2+ ]. to 62.5 AM (e.g., triangles, Fig.8a) reduced IK(f) an additional 7.0 ± 2.0% at 0 mV but had no additional effect at +60 mV.In two cells examined, there was in 62.5 ,uM Zn2 +, respectively, a 18% and 71% reductionof IK(f) at 0 mV and a 17% increase and 61% decrease at +60 mV; these cells were notincluded in the calculation of the mean. Increasing [Zn 2+ ] 0 to 125 AM (e.g., diamonds, Fig.8a) reduced IK(f) by 55 ± 1.2% at 0 mV and by 23 ± 0.5% at + 60 mV (n=3) and in one cellexamined, 250 AM Zn2+ reversibly reduced I K(f) by 58% at 0 mV and 2% at +60 mV (Fig.8b). At a concentration of 500 µM, Zn2+ reduced peak IK(f) by 71 ± 5.4% at 0 mV and by14 ± 3.4% at + 60 mV (n=4) as illustrated for one cell in Fig. 8c. The peak amplitude ofIK(f) was unaffected by 3 AM Zn 2+ (not shown).G. was reduced by 13% with 31 AM Zn 2+ in the external medium (n =3; in one cellit increased 17%) and by approximately 20% when the concentration of Zn2+ was increasedto 62.5 AM (n=5), 125 AM (n=2) or 500 AM (n=4).40Figure 7. The time-dependence of activation and inactivation in 500 AM and 5 mM Cd 2 +.A. The time to half-activation lt 12 act )1 is plotted against the test pulse potential for 500 AM1Cd2+ (diamonds) and its control (open circles) and for 5 mM Cd 2+ (squares) and its control(filled circles). Data points represent the mean ± s.e.m. of five cells at each concentrationof Cd2+ . The standard error bars were not included if they were smaller than the symbols.B. The time to half-inactivation lt inact) is plotted against the test pulse potential. Symbolsfor each concentration of Cd 2+ and the controls are the same as in A. Data pointsrepresent the mean ± s.e.m. of five cells at each concentration of Cd 2+ . C. Superimposedwhole-cell currents evoked at 0 mV under control conditions and in the presence of 500 AMCd2+ . The Cd2+ trace is scaled x2.25 to allow comparison of the rise times of the current.At 0 mV, th act increased from 1.8 ms in control to 4.8 ms in the presence of 500 AM Cd 2 +.Calibration bars apply to the control trace. The zero current level is indicated by thestraight line. Cell 140492C. D. Superimposed whole-cell currents evoked by a test pulseto +20 mV from the holding potential under control conditions and in the presence of 5mM Cd2+ . The current trace in 5 mM Cd 2+ was scaled x2.71 in order to compare the risetimes of the current. At +20 mV, ty, act increased from 1.2 ms in control to 5.1 ms in thepresence of 5 mM Cd2+ . Cell 1404921. Solutions: E//E&F [Cd 2 1 =500 AM///1 and E//F[Cd2+ ] =5 mM///1.4....1< U00c ;'I I 	 I 	 1 	 IC.) 	 C.) 	 C.)lil 	 en 	 r	 4Cf)E......te,„.co4142Table 2. Values listed are the MEAN (±s.e.m.) of the cells examined at eachconcentration of Cd 2+ . The number of cells tested "n" is indicated in parentheses in the leftcolumn. The data summarized is from four sets of experiments. The control values for eachset of experiments are in the row above the treated data. The shift of the activation andsteady-state inactivation curves, given as the difference between the mean values of V'obtained in control and those obtained in the Cd 2+ containing test solutions are summarizedin Table 6. N.T. means not tested. Levels of significance for the shifts of the activation andinactivation curves are included in Table 6 (pp 79-80).I V' represents the membrane potential for half-maximal activation of the current and krepresents the slope-factor of the curve fitted to the conductance-voltage relation asdescribed in Section I.2 V' represents the membrane potential at which steady-state half-inactivation of the currentis attained and k represents the slope-factor of the curve fitted to the normalized current-voltage relation as described in Section I.3 n = 7 for activation and n =4 for inactivation43TABLE 2.ANDSOLUTION[Cd2+ ]THE VALUES OF V' AND k CALCULATED FOR THE ACTIVATIONINACTIVATION OF I K (f) IN THE PRESENCE OF Cd2+ACTIVATION 	 INACTIVATION2V'(±S.E.M.)k(±S.E.M.)V'(±S.E.M.) (±S.E.M.)Control -13.7 mV 11.5 -52.4 mV -4.1(n=5) (±0.6) (±0.9) (±1.9) (±0.2)500 gM +10.9 mV 13.2 -31.3 mV -3.4(±1.7) (±0.4) (±2.0) (±0.1)Control -12.7 mV 15.1 -52.9 mV -4.1(n=5) (±1.1) (±0.9) (±0.6) (±0.2)2 mM +17.5 mV 9.8 -19.0 mV -4.3(±0.6) (±0.4) (±0.5) (±0.1)Control -7.4 mV 12.3 -48.4 mV -4.0(n=7/4) 3 (±2.6) (±0.5) (±3.6) (±0.3)5 mM +21.8 mV 11.1 -15.6 mV -3.8(±1.1) (±0.4) (±0.3) (±0.4)Control -12.9 mV 15.7 N.T.(n=4) (±0.6) (±0.3)100 AM -4.4 mV 11.4 N.T.(±0.5) (±0.2)200 AM +0.2 mV 12.5 N.T.(±0.8) (±0.4)400 AM +3.7 mV 11.8 N.T.(±0.8) (±0.6)800 AM +8.5 mV 12.3 N.T.(±1.2) (±0.2)1.6 mM +11.4 mV 11.3 N.T.(±1.3) (±0.4)441.22 Zn2+ Causes A Right-Ward Shift Of The Activation CurveThe apparent voltage-dependent reduction of I K(f) in the presence of Zn2+ , as withCd2+ , is also explained by a Zn2+ -induced right-ward shift of the activation curve along thevoltage axis. The values for V' and k, (estimated as described in Section I), weredetermined in each concentration of Zn2 + tested and are summarized in Table 3 (pp 55-56).The magnitudes of the shifts of V' with respect to Zn2 + concentration are summarized inTable 6 (pp 79-80).The filled symbols in Figs. 9a, 9b & 9d represent the mean (±s.e.m.) normalizedchord conductance at each potential in the presence of 62.5 MM, 125 AM and 500 AM Zn 2 +,respectively. The data points (filled circles and filled diamonds) in Fig. 9c represent thenormalized chord conductance calculated at each potential in the presence of 250 AM Zn 2 +for one cell only.At a concentration of 31 MM, Zn 2 + did not significantly (p > 0.01) shift the activationcurve nor did 3 AM Zn 2 + shift the activation curve (see Tables 3 and 6). A [Zn 2+ ]. of 62.5AM shifted V' by + 12.8 ± 2.8 mV (p <0.01) from -12.7 ± 1.9 mV in control (filled circles)to +0.1 ± 0.9 mV (filled diamonds, Fig. 9a; n=4). Increasing the [Zn 2 1. to 125 AM shiftedV' an average of +20.9 mV from -14.7 ± 0.9 mV in control (filled circles) to +6.2 ± 0.5 mV(filled diamonds, Fig. 9b; n=2). For the one cell examined at 250 AM Zn2+ , V' was shiftedby +27.5 mV, from -17.7 mV (filled circles) in control to +9.8 mV (filled diamonds, Fig.9c). The Zn2+ effect on the activation curve appeared to saturate near 250 AM sinceincreasing [Zn2+ ]. to 500 AM shifted V' by +26.1 ± 4.3 mV (p <0.01), from a control valueof -11.0 ± 2.0 mV (filled circles) to + 15.1 ± 2.3 mV (n=4; filled diamonds; Fig. 9d).Zn2+ did not alter the slope-factor of the conductance-voltage relation at any of theconcentrations tested (see Table 3).45Figure 8. Concentration-dependent effects of Zn 2 + on the current-voltage relation of I K(f).A. Normalized current-voltage relations for I K(f) under control conditions (open circles)and in the presence of 31 AM (squares), 62.5 AM (triangles) and 125 AM Zn 2  (diamonds).Cell 271191B. The effects of Zn 2+ on IK(f) were completely reversible (filled circles). B.The normalized current-voltage relation for IK(f) in one cell (271191A) in 250 AM Zn 2+(diamonds). The current was normalized against maximum IK(f) in control. The normalizedcurrent was plotted against the test pulse potential. A Zn2+ concentration of 250 AM shiftedthe threshold for activation from the control (open circles) value of -40 mV to -20 mV.Consistent with this shift the current evoked at each test potential up to +60 mV wasreduced in the presence of 250 AM Zn 2 . At test potentials greater than + 60 mV thecurrent-voltage relations overlap. The effects of 250 AM Zn2+ on IK(f) were reversible(filled circles). C. The normalized current-voltage relation elicited in the same manner asin A and B but in the presence of 500 AM Zn 2 + (diamonds). The threshold for activationshifted from the control (open circles) value of -40 mV to -10 mV. Commensurate with thisshift, the current evoked at each test potential in the presence of 500 AM Zn2+ was reduced.The effects of 500 AM Zn2+ were partially reversible (filled circles). Solutions: G//H[Zn2+ ]=31 to 500 AM///1.A46-40 	 0 	 40	80EM (mV)B-40 	 0	 40 	 80EM (mV)C-40	40 	 80EM (mV)471.23 Concentration Dependence For Zn 2+ -Induced Shifts Of V'The concentration dependence of the Zn2+ -induced shift of V' is illustrated in Fig.10. A least squares fit of the Michaelis-Menton equation to the data showed that the Zn 2+-induced shift of V' saturated at +34 mV with a KM of 92 AM.124 Zn2+ Increases The 50% Rise Time Of I K(f)At all concentrations examined, with the exception of 3 AM, Zn2+ increased the risetime of the current especially at membrane potentials close to the threshold for activation.Half-activation times were examined at 500 AM Zn2+ to allow qualitative comparisons tothe results observed with Cd2+ .The 50% rise times, determined at four potentials, are summarized in Fig. 11a. At0 mV, there was a 4.7-fold increase (n = 4) in ty, act from 2.1 ± 0.2 ms in control (circles; Fig.11a) to 9.7 ± 0.9 ms in 500 AM Zn2 + (diamonds; Fig. 11a). This effect of Zn 2+ on theactivation kinetics is evident in the current traces in Fig. 11c where the current tracerecorded in 500 AM Zn2+ at 0 mV was scaled up 5.35-fold, in order to allow a comparisonof the rising phase of the current. As the potential was stepped to more depolarized levelsthe slowing of the activation kinetics by Zn 2 + was less pronounced as indicated by the graphof Fig. lla and by the current traces of Fig. 1 ld at +60 mV. At +60 mV, there was a 2-fold increase in ty, act (n=4) from 0.9 ± 0.1 ms in control (circles; Fig. 11a) to 2.0 ± 0.1 msin 500 AM Zn2+ (diamonds; Fig. 11a).1.25 Zn2+ Causes A Right-Ward Shift Of The Steady-State Inactivation CurveThe effect of Zn2+ on steady-state inactivation was examined at 3-500 AM [Zn2+ ] andthe values for V' and k, (estimated as described in Section I) are summarized in Table 3.The magnitudes of Zn2+ -induced shifts of V' for the 11,,, curve are summarized in Table48Figure 9. Control and treated activation and inactivation curves in Zn2+ concentrations of62.5 AM (A), 125 AM (B), 250 AM (C), and 500 AM (D). A. The data points represent themean ± s.e.m. of four cells. For the activation curve, control V' was -12.7 ± 1.9 mV and theslope-factor (k) was 13.2 ± 0.6 mV (filled circles) versus V' = -0.1 ± 0.9 mV and k =12.8 ±0.6 mV (filled diamonds) in 62.5 AM Zn2 +. For the inactivation curves, V' = -54.0 ± 0.3 mVand k = -4.7 ± 0.1 mV in control (open circles) versus V' = -39.2 ± 0.9 mV and k = -5.0 ± 0.3mV (open diamonds). B. The data points represent the mean ± s.e.m. of two cells. V' foractivation shifted from -14.7 ± 0.9 mV in control (filled circles) to +6.2 ± 0.5 mV in mediumcontaining 125 AM Zn2+ (filled diamonds). The respective slope-factors were 12.1 ± 1.8 mVand 13.9 ± 0.1 mV. The inactivation curves were best fit with V' = -56.5 ± 2.5 mV andk=-4.7 ± 0.2 mV in control (open circles) versus V' = -37.0 ± 3.1 mV and k = -5.0 ± 0.0 mVin 125 AM Zn2+ (open diamonds). C. The activation and inactivation curves in 250 AMZn2+ were determined for one cell (271191A). For the activation curve, control V' and kwere -17.7 mV and 11.0 mV, respectively (filled circles) versus V' = + 9.8 mV and k =12.3mV (filled diamonds). For the inactivation curve, in control (open circles), V' = -54.9 mVand k=-4.2 my versus V'=-28.8 mV and k = -5.1 mV (open diamonds). D. The data pointsrepresent the mean ± s.e.m. of four cells in 500 AM Zn 2+ . For control activation (filledcircles), V' =41.0 ± 2.0 mV and k =12.5 mV versus V' = + 15.1 ± 2.3 my and k =12.4 mV(filled diamonds). Steady-state half-inactivation shifted from control (open circles) V' = -53.1± 1.2 mV to V' = -27.2 ± 1.4 mV (open diamonds). The respective slope-factors were k = -4.5± 0.1 mV and k = -5.1 ± 0.3 mV.EWj 8 iEW49EWC.)EW506. The data points (open symbols) in Fig. 9a,b,d represent the mean normalized current ±s.e.m. at each potential in the presence of 62.5 AM, 125 AM and 500 AM Zn 2+ , respectively.The data points (open circles and open diamonds) in Fig. 9c represent the normalizedcurrent calculated at each potential in the presence of 250 AM Zn2+ for one cell only.Bath application of 3 AM Zn2+ had no effect on the h a, curve. However, 31 AM Zn2+which did not significantly (p > 0.01) shift the activation curve did shift V' of the h., curveby +8.5 ± 3.3 mV (see Table 3). Shifts of + 14.9 ± 1.2 mV (n=4; p < 0.01) and + 19.4 mV(n=2) for V' of the h., curve were obtained with 62.5 AM Zn 2+ and 125 AM Zn2+ ,respectively (open symbols; Figs. 9a & 9b, respectively). The Zn 2+ -induced shift of V' forinactivation also appeared to saturate near 250 AM Zn 2 +. For the one cell examined, 250AM Zn2+ shifted V' by +26 mV from -54.9 mV in control (open circles; Fig. 9c) to -28.8 mV(open diamonds; Fig. 9c). When the [Zn 2-1, was increased to 500 AM there was no furtherincrease in the shift of V'. At a concentration of 500 AM, Zn 2+ also shifted V' by + 26.0 ±2.5 mV (n=4; p <0.01) from -53.1 ± 1.2 mV in control (open circles; Fig. 9d) to -27.2 ± 1.4mV (open diamonds; Fig. 9d).1.26 Zn2+ Increases The Time To Half-InactivationTime to half-inactivation increased approximately 1.4-fold at all potentials in thepresence of 500 AM Zn 2+ . At 0 mV, ty, inact increased from 27.9 ± 0.7 ms in control (circles,Fig. lib) to 42.7 ± 5.3 ms in 500 AM Zn 2+ (diamonds, Fig. 11b) and at +60 mV t 1/2 inactincreased from 40.2 ± 0.2 ms in control to 54.4 ± 1.6 ms in 500 AM Zn2+.51Figure 10. The concentration dependence of the Zn 2 + -induced shift of the half-activationpotential for I K(f). Data points represent the mean shift ± s.e.m. unless there wasinsufficient data to calculate s.e.m. in which case no error bar is included; the bracketednumber indicates the value for "n". Shifts of the activation curve by Zn 2+ (as summarizedin Table 6) are plotted as a function of Zn 2+ concentration. The line represents a solutionto the Michaelis-Menton equation in which S max was +34 mV and the K M was 92 MM.35 —(1)025 —(4)     (2)15—(5)52-5 — (2)(4)I 	 i 	 I 	 I 	 I 	 I 	 1 	 1 	 1 	 1100	 200 	 300	400	 500Zn2 + concentration (AM)53Figure 11. The time-dependence of activation and inactivation in the presence of 500 AMZn2+ . (A) The time to half-activation (ty, act ) (mean ± s.e.m., n=4) is plotted against thetest pulse potential for currents evoked in 500 AM Zn 2+ (diamonds) or in control medium(open circles). B. The time for half-inactivation (t,,, inact) is plotted against the membranepotential during the test pulse. Symbols are the same as in A. Data points represent themean ± s.e.m. (n=4). C. Superimposed whole-cell currents elicited by a test pulse to 0 mVfrom the holding potential under control conditions and in 500 AM Zn 2+ . The Zn2+ traceis scaled 5.35-fold to allow comparison of the rise times of the current. At this membranepotential, ty, act increased from 2.0 ms in control to 9.8 ms in the presence of 500 AM Zn 2+ .Calibration bars apply to the control trace. Cell 201191D. D. Superimposed whole-cellcurrents evoked by a test pulse to +60 mV from the holding potential for the same cell asin C, in control medium and modified control containing 500 AM Zn 2+ . The current tracein 500 AM Cd2+ was scaled 1.25-fold in order to compare the rise times of the current underthe control and test conditions. The effect of Zn 2 + on the rise time of I K(f) is lesspronounced at more depolarized potentials. At +60 mV, act increased from 0.87 ms incontrol to 1.87 ms in the presence of 500 AM Zn 2 +. Cell 201191D. Solutions: G//H[Zn2+ ] =500 AM///1.C.)54EW- 0cc0F 	 100	 v.) 	 0.--.1d55Table 3. 	 Values summarized are the MEAN (±s.e.m.). The number of cells "n"examined is indicated in parentheses below each solution. The control values for each setof experiments precede the values for the test conditions. The shift of the activation andsteady-state inactivation curves, given as the difference between the mean values of V' inthe control and Zn2 + containing solutions, respectively, are summarized in Table 6. Levelsof significance for the shifts of the activation and inactivation curves are included in Table6 (pp 79-80).1 n =4 for activation and n = 5 for inactivation56TABLE 3.ANDSOLUTION[Ee]THE VALUES OF V' AND k CALCULATED FOR THE ACTIVATIONINACTIVATION OF I K (f) IN THE PRESENCE OF Zn2+ACTIVATION 	 INACTIVATIONV'(±S.E.M.)k(±S.E.M.)V'(±S.E.M.) (±S.E.M.)Control -18.1 mV 12.2 -60.5 mV -4.3(n=2) (±4.7) (±4.0) (±4.5) (±0.1)3 AM -20.2 mV 14.7 -62.4 mV -4.1(±1.3) (±1.1) (±6.3) (±0.3)Control -14.5 mV 11.8 -53.7 mV -5.1(n=4) (±4.2) (±1.7) (±1.5) (±0.6)31 AM -7.4 mV 11.9 -45.2 mV -4.8(±5.3) (±1.8) (±1.7) (±0.3)Control -12.7 mV 13.2 -54.0 mV -4.7(n=4/5) 1 (±1.9) (±0.6) (±0.3) (±0.1)62.5 AM +0.1 mV 12.8 -39.2 mV -5.0(±0.9) (±0.6) (±0.9) (±0.3)Control -14.7 mV 12.1 -56.5 mV -4.7(n=2) (±0.9) (±1.8) (±2.5) (±0.2)125 AM +6.2 mV 13.9 -37.0 mV -5.0(±0.5) (±0.1) (±3.1) (±0.0)Control(n=1)-17.7 mV 11.0 -54.9 mV -4.2250 AM +9.8 mV 12.3 -28.8 mV -5.1Control -11.0 mV 12.5 -53.1 mV -4.5(n=4) (±2.0) (±0.7) (±1.2) (±0.1)500 +15.1 mV 12.4 -27.2 mV -5.1(±2.3) (±0.3) (±1.4) (±0.3)572. The Effects Of Alkaline Earth Metal Ions On I K(f)Divalent cations from Group IIB, though less potent than the transition metal ions,have been reported to exert qualitatively similar effects on the behaviour of voltage-gatedion channels (McLaughlin et al., 1971; Hahin & Campbell, 1983; Cukierman & Krueger,1990). Of the alkaline earth metal ions, Ca 2+ is reported to have the greatest and Mg2+ thesmallest effect. The following results indicate that the same pattern persists for the I K(f) ofmelanotrophs.2.1 CalciumCa2+ is postulated to play a role in the regulation of Na + channel gating (Armstrong& Cota, 1991), the activation of some TOC-like currents (Zbicz & Weight, 1985), and themaintenance of K + channel integrity (Armstrong & LOpez-Barneo, 1987). External Ca 2+also causes a voltage-dependent block of Na+ channels (Woodhull, 1973) in addition toproducing a non-specific screening of surface charges (Frankenhaeuser & Hodgkin, 1957).Thus, it was of interest to determine whether Ca t} might exert similar effects on thebehaviour of IK(f).2.11 Changes in [Ca21. Reduce IK(f)The reduction of IK(f) in external medium containing zero Ca- is illustrated in thecurrent-voltage relation for one cell in Fig. 12a (note that the solution is nominally Ca 2+ -freeas a Ca2+ buffer was not included - the contaminating concentration of Ca 2+ could be ashigh as 2.5 AM). In the four cells examined, zero Ca 2+ reduced IK(f) by 18 ± 2.1% at 0 mVand 21 ± 3.0% at + 60 mV and decreased Gmax by 21 ± 3.0%. This decline of I K(f) probablyreflects a left-ward shift of the 11,,, curve. During the protocol to determine the current-voltage relation the holding potential was -80 mV; due to the left-ward shift of the h,,, curvein zero Ca2+ approximately 20% of the current is inactivated at this potential which isconsistent with the reduction in peak I K(f) and Gmax observed.58Increasing [Ca 2 10 to 5 mM, 10 mM and 20 mM also reduced the peak amplitude ofIK(f) as shown for one cell in Fig. 12b. The peak amplitude of I K(f) and the chordconductance at +60 mV for this cell were reduced by 5%, 7% and 21% in 5, 10 and 20 mMCa2+ , respectively. For purposes of comparison, the chord conductance was determined at+ 60 mV in all concentrations of Ca2+ tested. In control medium, Gmax is reached at + 60mV, however, increasing [Ca2 10 shifts the conductance-voltage relation right-ward along thevoltage axis and it is unlikely that G. was attained at + 60 mV in the test conditions.Therefore, an unequivocal conclusion regarding the effect of increased [Ca 2 1„ on Gmaxcannot be made.2.12 Changes in [Ca2+]. Shift V' For The Activation CurveThe values for V' and k, (estimated as described in Section I), determined in 0-20mM Ca2+ are summarized in Table 4 (pp 69-70). The magnitudes of the shifts of V' withrespect to the external Ca 2+ concentration are summarized in Table 6 (pp 79-80).The conductance-voltage relation shifted left-ward along the voltage axis in zero Ca 2+as illustrated for a representative cell in Fig. 13a (filled symbols). For the four cellsexamined, removing Ca2+ from the external medium shifted half-activation by -11.7 ± 2.7mV from -15.2 ± 1.6 mV in control to -26.9 ± 1.1 mV. In zero Ca 2+ the slope-factor (k) ofthe conductance-voltage relation increased from 13.6 ± 0.9 mV to 22.0 ± 1.3 mV (see Table4) which reflects a decrease in the voltage-sensitivity of activation.Increasing [Ca210, from the control concentration of 2 mM, caused a concentration-dependent right-ward shift of the activation curve (Fig. 13b). A [Ca 2+ ]. of 5 mM shifted V'+2.5 ± 1.8 mV from -10.6 ± 1.7 in control to -8.1 ± 0.0 (n=2). Raising the Ca2+concentration to 10 mM shifted V' by +8.4 ± 2.3 mV from the control value above to -2.2± 0.6 mV (n=2). A Ca2+ concentration of 20 mM shifted V' + 12.4 ± 2.2 mV from thecontrol value of -10.6 ± 1.7 to + 1.8 ± 0.5 mV (n=3).59Figure 12. Current-voltage relations for normalized IK(f) in response to reduced orincreased concentrations of external Ca'-+ . A. The peak amplitude of 1 K(f) was reduced inzero Ca2+ (diamonds) relative to the control responses (circles) when 2 mM Ca2+ waspresent. At 0 mV, IK(f) was reduced 15.7% in zero Ca 2+ relative to the control and at + 60mV IK(f) was reduced by 24.1% relative to control. Solutions: B//I///3. Cell 260891D.B. As the external concentration of Ca 2 + was increased from the control concentration of2 mM (circles) to 5 mM (diamonds), 10 mM (squares) and 20 mM Ca 2+ (triangles) therewas a 5%, 7%, and 21% decrease, respectively, in the normalized peak amplitude of I K(f).Solutions: B//J [Ca2+ ] =5-20 mM///3. Cell 280891H.40 60-20	20EM (mV)40 60-20 	 0 	 20EM (mV)A6061Increasing [Ca2+ ]o decreased the slope-factor (k) of the conductance-voltage relation from14.7 ± 0.4 mV in control to 11.8 ± 0.3 mV in 5 mM Ca - +, 10.7 ± 0.3 mV in 10 mM Ca2+and 9.5 ± 0.5 mV in 20 mM Ca 2+ .2.13 Concentration Dependence For Ca2+-Induced Shifts of V'The curve fitted to the data points of Fig. 14 indicates that the maximal Ca 2+ -inducedshift of V' (from the control concentration of 2 mM) was + 15.6 mV and that the half-maximal shift occurred at a [Ca21c, of 3.4 mM. The Km for the shift of V' is approximatelyone order of magnitude greater than that calculated for either Cd 2 + or Zn2+ and themaximum depolarized shift is approximately half that possible with the two transition metalions.2.14 Increasing [Ca21. Increases ty, ad For IK(f)Increasing [Ca210 increased the time to half-activation at potentials close to thethreshold for activation (Fig. 15a). The increase in the rise time of I K(f) was mostnoticeable in 20 mM Ca2+ where th act increased approximately 2-fold at 0 mV. At +60mV, there was no measurable difference in the rise time of the current at any concentrationof Ca2+ examined.2.15 Changes in [Ca2+ ]o Shift V' For The Inactivation CurveThe effect of Ca2 + on steady-state inactivation was examined and the values for V'and k (estimated as described in Section I) are summarized in Table 4.In zero Ca2+ there was a left-ward shift of the inactivation curve (Fig. 13a, opensymbols). V' was shifted -19.5 ± 5.2 mV (p < 0.01) from a control value of -54.1 ± 1.3 mV(open circles) to -73.5 ± 3.9 mV (open diamonds; n=4).62Figure 13. The effect of changes of [Ca"],„ on the activation and steady-state inactivationcurves for IK(f). A. The normalized chord conductance is plotted against the test pulsepotential and the activation curve is a solution to the Boltzman equation where V' = -12.4mV and the slope-factor = 15.0 mV for control (filled circles) versus V' = -25.6 mV anda slope-factor = 23.4 mV in zero Ca2 + (filled diamonds). For steady-state inactivation, thecurve is a solution to the Boltzman equation where V' = -51.1 mV and the slope-factor(k)=-4.0 mV for control (open circles) whereas in zero Ca 2+ V' = -67.8 mV and k = -5.2 mVin zero Ca" (open diamonds). Solutions: B//I///3. Cell 260891D. B. The details ofthese activation and steady-state inactivation curves are given in the text. Activation: control(filled circles); 5 mM Ca" (filled diamonds); 10 mM Ca' (filled squares); 20 mM Ca"(filled triangles). Inactivation: control (open circles); 5 mM Ca" (open diamonds); 10 mMCa" (open squares); 20 mM Ca" (open triangles). Solutions: B//J [Ca21. = 5-20mM///3. Cell 280891H.63-80 	 -40 	 0	40	80EM (mV)  -80 	 -40 	 0	 40 	 80EM (mV) 64Figure 14. Concentration-response for the effects of Ca 2+ on the shift of the half-activationpotential (V'). Shifts of V' as summarized in Table 6 are plotted against external Ca 2+concentration. Data points represent the mean ± s.e.m.; "n" is indicated in brackets. Thefitted curve is a solution to the Michaelis-Menton equation where the maximal shift is + 15.6mV and the KM for the half-maximal shift is 3.4 mM.(3)10 	 15 	 20Ca2+ concentration M)6566When [Ca-+ ]0 was raised there was a concentration-dependent right-ward shift of theinactivation curve along the voltage axis (Fig. 13b). Increasing [Ca2 +]. to 5 mM shifted V'by approximately +6.3 mV from -52.7 ± 1.5 mV in control (open circles) to -46.5 ± 0.4 mV(open diamonds; n=2). After changing to 10 mM Ca2+ , V' shifted approximately + 11.6 mVfrom the same control value to -41.1 ± 0.4 mV (open squares; n =2). Finally, a 20 mMconcentration of Ca2 + shifted V' by approximately + 18.8 mV (n=5; p < 0.01) from thecontrol value of -52.7 ± 1.5 mV (open circles) to -33.9 ± 1.7 mV (open triangles). The shiftsobserved with Ca2+ in the millimolar range were not as great as those observed with athousand-fold lower concentration of either Cd 2+ or Zn2+ .When the concentration of Ca2+ was raised or lowered the steady-state inactivationcurves shifted in a parallel fashion and there was no change in the slope (k) of the curves(see Table 4).2.16 Influence Of Changing [Ca2+ ]0 On 4, thati For IK(f)The time to half-inactivation did not change from control values with 5 mM or 10mM external Ca2+ . A 20 mM concentration of Ca2+ increased t- 'h inact approximately 1.2-foldat 0 mV and 1.0-fold at +60 mV in the two cells examined (Fig. 15b).2.2 MagnesiumAmong the divalent cations, Mg2 + is one of the least potent in causing right-wardshifts of the activation and inactivation curves of the Na + and delayed rectifier K + channel(Blaustein & Goldman, 1968; Hille et al., 1975). The effect of Mg 2+ was examined todetermine if this finding was consistent with respect to IK(f).67Figure 15. The effect of increased [Ca210 on the activation and inactivation kinetics ofIK(f). A. The half-activation times (t y, act) in control (circles) and medium containing 5 mM(diamonds), 10 mM (squares) or 20 mM (triangles) Ca2+ . Each point represents the meanfor two cells. At -20 mV, actt increased approximately 1.3-fold, 1.9-fold, and 3.2-fold,respectively, in the presence of 5, 10 and 20 mM Ca 2+ relative to a ty, act of 3.32 ms incontrol at -20 mV. At +60 mV, t y, act was increased maximally (approximately 1.5-fold) inthe presence of 20 mM Ca2 +. B. The half-inactivation time for IK(f) is plotted as in A.There was no difference in ty, inact between the control values and the values with 5 and 10mM Ca2+ present. The shift in tY2 inact with 20 mM Ca2+ in the bath is described in the text.Solutions: B//J [Ca2+ ] =5-20 mM///3.40 60-20	0	20EN! (mV)10 — tio act (ms)45 — tv2 inact (ms)15—68AB40 60I 	 I-20 20EM (mV)69Table 4. 	 Values summarized are the MEAN (±s.e.m.). The number of cells "n"examined is indicated in parentheses below each solution. The shift of the activation andsteady-state inactivation curves, given as the difference between the mean values of V' inthe control and test solutions, respectively, are summarized in Table 6. Levels ofsignificance for the shifts of the activation and inactivation curves are included in Table 6(pp 79-80).1 As described in the methods, the solution identified as 0 mM Ca2+ can only be considerednominally Ca2+ free as a Ca2+ buffer was not included in the solution.2 The conductance-voltage relation was completed for n =2 cells and steady-state inactivationwas completed for n =4 cells.3 The conductance-voltage relation was examined for 3 cells and steady-state inactivationwas determined for 4 cells.t The slope of the activation curve in zero Ca 2+ is significantly greater than in control(p < 0.01). Significance was determined using a paired "t"-test with p = 0.01.t The slope of the inactivation curve in zero Ca 2+ is not significantly greater than in controlwith (p < 0.03). The significance level was set at p = 0.01 to account for the small samplesize. However, as the change would be significant with p = 0.05 the increase in the slope ofthe inactivation curve warrants closer examination.70TABLE 4. THE VALUES OF V' AND k FOR THE ACTIVATION ANDINACTIVATION OF I K (f) IN THE PRESENCE AND ABSENCE OF Ca 2+ACTIVATION 	 INACTIVATIONSOLUTION(Ca2+ ]V'(±S.E.M.)k(±S.E.M.)V'(±S.E.M.)k(±S.E.M.)Control(n=4)-15.2 mV(±1.6)13.6(±0.9)-54.1 mV(±1.3)-3.8(±0.3)0 mM -26.9 mV 22.0f -73.5 mV -5.4*(n=4) (±1.1) (±1.3) (±3.9) (±0.6)Control -10.6 mV 14.7 -52.7 mV -4.3(2 mM)(n=3)(±1.7) (±0.4) (±1.5) (±0.1)5 mM -8.1 mV 11.8 -45.7 mV -4.1(n=2\4) 2 (±0.0) (±0.3) (±1.8) (±0.0)10 mM -2.2 mV 10.7 -40.5 mV -3.8(n=2\4) 2 (±0.6) (±0.3) (±1.8) (±0.1)20 mM 1.8 mV 9.5 -34.5 mV -3.5(n=3\4) 3 (±0.5) (±0.5) (±2.1) (±0.1)712.21 Voltage- and Time-Dependence of Activation For I K(f)In The Presence of Mg2+Mg2 + at a concentration of 10 mM had very little effect on the current-voltagerelation for IK(f). Consistent with these observations, V' and k (estimated as in Section I)for the conductance-voltage relation were unchanged from the control values (Fig. 16a).In 40 mM Mg 2+ the threshold for activation shifted from -40 mV to -30 mV and atpotentials below + 60 mV the peak amplitude of I K(f) was reduced. At potentials above+ 60 mV, 40 mM Mg2 + had no effect on the peak amplitude of IK(f). These results areconsistent with the right-ward shift of the activation curve caused by 40 mM Mg 2+ : the half-activation potential shifted by 2.7 ± 3.9 mV (p < 0.01) from -13.8 ± 1.9 mV in control (Fig.16b, filled circles) to -11.2 ± 2.0 mV in 40 mM Mg 2 + (Fig. 16b, filled diamonds; n=4). A40 mM concentration of Mg2 + shifted the activation and inactivation curves along thevoltage axis to approximately the same degree as 20 mM Ca 2+ , 200 AM Cd2+ or 62.5 AMZn2+ (see Table 6). The slope-factor (k) of the conductance-voltage relation was not alteredin the presence of 40 mM Mg2 + (see Table 5 - pp 77-78).As illustrated in Fig. 17a, 40 mM Mg2+ increased ty, act in a manner similar to thatobserved with the other divalent cations. The action of 40 mM Mg2+ to slow the activationof IK(f) is illustrated in the current traces of Fig. 17b in which the treated current has beenscaled and superimposed on the control current. At 0 mV, 40 mM Mg2 + increased ty, a„ 1.8-fold from 1.4 ± 0.1 ms to 2.5 ± 0.2 ms (n=4). As the membrane was stepped to moredepolarized levels there was less of an effect on the rise time by Mg 2+ consistent with theresults obtained with Cd2+ , Zn2+ and Ca2+ .2.22 Mg2+ Shifts V' For Steady-State InactivationThe effects of Mg2+ on steady-state inactivation were comparable to its effects onactivation. Thus, a concentration of 10 mM Mg2+ had very little effect on steady-state72Figure 16. The effects of 10 mM (A) and 40 mM (B) Mg2+ on the activation andinactivation curves for IK(f). A. In 10 mM Mg2+ , V' shifted from -12.4 mV in control (filledcircles) to -13.2 mV in 10 mM Mg 2+ (filled diamonds), and the slope-factor decreased froma control value of 13.3 mV to 10.9 mV. The inactivation curve shifted from a half-inactivation potential of -55.0 mV with a slope-factor of -3.9 mV in control (open circles)to a half-inactivation of -50.8 mV with a slope-factor of -4.0 mV (open diamonds). Circlesinset with crosses represent the recovery responses. Cell 241091B. Solutions: K//L&M[Mg21 ].= 10 mM///1. B. A 40 mM concentration of Mg 2+ shifted the activation andsteady-state inactivation curves right-ward along the voltage axis. The activation curve forthis cell shifted from V'= -16.5 mV with a slope-factor of 11.6 mV in control (filled circles)to V' = -1.0 mV with a slope-factor of 12.3 mV in 40 mM Mg2 + (filled diamonds). Theinactivation curve shifted from V' =-59.8 mV with a slope-factor of -4.4 mV in control (opencircles) to V' =-45.9 mV with a slope-factor of -5.1 mV in 40 mM Mg 2+ (open diamonds).Cell 171091D. Solutions: K//M [Mg 2+ ] =40 mM///1.60-20 	 20EM (mV)A0.5 -B-20 	 20EM (mV)7374inactivation as illustrated for one cell in Fig. 16a (open circles and diamonds). The shift ofthe ho, curve caused by 40 mM Mg 2 + was equivalent to the shift of the conductance-voltagerelation and is illustrated for one cell in Fig. 16b. V' shifted by + 17.2 ± 3.8 mV from -60.4± 2.3 mV in control to -43.2 ± 1.5 mV in 40 mM Mg 2+ (n =4).The half-inactivation time was unaffected by either 10 mM or 40 mM Mg 2+ (notshown).75Figure 17. The time-dependence of activation in the presence of 40 mM Mg 2+ . A. Thetime to half-activation is plotted against the test pulse potential. Data points represent themean ± s.e.m. for 3 cells recorded in solutions: K//M [Mg - +J=40 mM///1. A 40 mMconcentration of Mg2+ (diamonds) increased the time to half-activation compared to thecontrol values (circles) at all potentials. See text for details. B. Superimposed whole-cellcurrents evoked at 0 mV. The current trace obtained in 40 mM Mg2 + was scaled x1.53 inorder that the rise times of the currents could be compared. Mg -+ , even at a concentrationof 40 mM, was less potent than any of the other divalents tested in slowing the rise of I K(f).At 0 mV, ty, act increased from 1.4 ± 0.1 ms in control to 2.5 ± 0.2 ms in 40 mM Mg2+.A 6.0--20I 	 I20EM (mV)Bt1/2 act (ms)	 —4.0 -2.0-I40I 11 	 I607677Table 5. 	 Values summarized are the MEAN (±s.e.m.). The number of cells examined"n" is indicated in parentheses below each solution. Experiments using 10 mM Mg2+ wererun separately from those with 40 mM Mg 2+ . The control values for each experiment aresummarized in the row above the treated values. The shifts of the activation and steady-stateinactivation curves caused by Mg 2 +, given as the difference between the mean values of V'in the control and Mg2 + containing solutions, respectively, are summarized in Table 6.Levels of significance for the shifts of the activation and inactivation curves are included inTable 6 (pp 79-80).78TABLE 5.ANDSOLUTION[mg2+ ]THE VALUES OF V' AND k CALCULATED FOR THE ACTIVATIONINACTIVATION OF I K (f) IN THE PRESENCE OF Mg2+ACTIVATION 	 INACTIVATIONV'(±S.E.M.)k(±S.E.M.)V'(±S.E.M.) (±S.E.M.)Control -15.2 mV 11.3 -57.2 mV -4.1(1 Me)(n=6)(±2.1) (±1.0) (±0.6) (±0.2)10 mM -15.9 mV 11.8 -54.2 mV -4.0(n=6) (±2.5) (±0.9) (±1.2) (±0.1)Control -13.8 mV 12.8 -60.4 mV -6.0(1 Me)(n=4)(±1.9) (±0.9) (±2.3) (±1.0)40 mM -1.2 mV 12.1 -43.2 mV -4.8(n=4) (±2.0) (±1.7) (±1.5) (±0.2)7 9Table 6. A summary of the results presented in Tables 2, 3, 4 and 5. The voltage shiftswere determined by the difference between the mean half-activation or half-inactivationpotentials in the control medium, and the medium containing the divalent cation indicated.The number of cells from which the mean (± s.e.m.) was determined is indicated to the rightin parenthesis. N.T. means not tested.1 Significance of the shifts was not examined for this series of experiments since the steady-state inactivation was not tested and complete data was obtained at three otherconcentrations of Cd2+ .2 Zero Ca2+ solutions were actually nominally Ca 2+ -free. EGTA was not included in themodified external medium to buffer a possible contaminating concentration of Ca 2+ . In twocells, zero Ca2+ caused a positive shift of the activation and inactivation curves. The valuesfor V' for these cells were not included in the calculation of the shift.t These shifts, relative to the control values, are significant p < 0.0001. Significance wasdetermined using analysis of variance.t These shifts were not signficant.1 This shift was significant (p < 0.0001). Significance was determined using a paired "t"-test.80TABLE 6. SHIFTS OF THE ACTIVATION AND INACTIVATION CURVES OFI k (f) CAUSED BY ALTERING THE EXTERNAL DIVALENT CATIONCONCENTRATIONDIVALENT CATIONPRESENT SHIFT OF THE 	 SHIFT OF THEACTIVATION CURVE 	 INACTIVATION CURVE(mV) 	 (mV)      100.0 AM Cd2+ +8.6 ± 	 1.2 (4) 1 N.T.200.0 AM Cd2+ +13.1 ± 1.5 (4) N.T.400.0 AM Cd2+ +16.6 ± 	 1.4 (4) N.T.800.0 AM 0:12+ +21.4 + 1.9 (4) N.T.1.6 mM Cd2+ +24.3 + 1.9 (4) N.T.500.0 AM Cd2+ +24.6 ± 2.3 (5)t +21.2 ± 3.9 (5)t2.0 mM 0:12+ +30.1 ± 	 1.6 (5) t +33.9 ± 1.0 (5)t5.0 mM Cd2+ +29.2 ± 	 3.7 (7) t +32.8 ± 3.9 (4)t3 AM Zn2+ -2.0 (2) -1.9 (2)31 AM Zn2+ +7.2 + 9.5 (4) * +8.5 + 	 3.3 (4)t62.5 AM Zn2+ +12.8 + 2.8 (5) t +14.9 + 1.2 (4)t125.0 AM Zn2+ +20.9 (2) +19.4 (2)250.0 AM Zn24. +27.5 (1) +26.1 (1)500.0 AM Zn2+ +26.1 ± 4.3 (4)t +26.0 + 2.5 (4)t0 mM Ca2+ (2) -11.7 ± 	 2.7 (4) 1 - 19.5 + 5.2 (4) t5 mM Ca2+ +2.5 (2) +7.0 + 	 3.3 (4)t10 mM Ca2+ +8.4 (2) +12.3 + 3.3 (4)t20 mM Ca 2+ +12.4 ± 2.2 (3) +18.2 + 3.6 (4)t10 mM Mg2+ -0.8 + 4.7 (6) * +3.0 + 1.8 (6)*40 mM Mg2+ +12.7 + 3.9 (4)t +17.2 + 3.8 (4)t81DISCUSSIONDivalent Cations Exert A Charge Screening Effect On The Channel Conducting IK(f)Changes in the intracellular or extracellular concentration of divalent cations causethe potential-sensitive parameters of voltage-dependent ion channels, such as the activationand inactivation curves, to shift along the voltage axis. Increased divalent cationconcentration has also been shown to slow the activation kinetics of voltage-gated ionchannels. There are two hypotheses to account for the effects of divalent cations on thegating of voltage-dependent channels. The first suggests that divalent cations neutralizedfixed negative charges on the membrane of the cell in a non-specific manner and thus alterthe electric field surrounding the voltage sensor (Frankenhaeuser & Hodgkin, 1957; Gilbert& Ehrenstein, 1969; McLaughlin et al., 1971; D'Arrigo, 1978; Hahin & Campbell, 1983).The second hypothesis invokes the specific binding of divalent cations, either to a negativelycharged component of the channel protein or to a negatively charged site on or electricallyclose to the channel (Hille, 1968; Begenisich & Lynch, 1974; Hille et al., 1975; Gilly &Armstrong, 1982 a&b; Mayer & Sugiyama, 1988). Specific binding of divalent cations doesnot preclude the contribution of non-specific effects as well and surface chargeneutralization as a result of both has been proposed to account for the action of divalentcations on sodium channels (Cukierman & Krueger, 1990; Schild et al., 1991). Binding andscreening are different ways of neutralizing surface charge but will exert the same effect onthe electric field across the membrane (Gilbert & Ehrenstein, 1984).The charge screening actions of the divalent cations examined here were observedas a reduction in the peak amplitude of IK(f) resulting from the right-ward shift of both theactivation and inactivation curves. In the present study, divalent cation substitution wasemployed to distinguish between these two mechanisms. The prediction central to non-specific screening of fixed negative surface charges is that cations with the same valenceshould be equally effective (Hille, 1984). The effects of divalent cations on I K(f) violate this82prediction. The results reported here suggest that divalent cations likely exert their effectsthrough specific binding either to a site on the channel protein or to a site electrically closeto the channel and its voltage sensor.The divalent cations tested varied widely in their ability to shift the potential-sensitiveparameters of I K(f) along the voltage axis and ranked in the following order:Zn" > Cd' > > Ca2+ > Mg'.The fact that the transition metal ions, Zn" and Cd", were far more potent than thealkaline earth metal ions, Ca" and Mg" is consistent with the rank order reported for theeffect of these ions on the sodium channels of both the lobster giant axon and the node ofRanvier of frog myelinated nerve (Blaustein & Goldman, 1968; Hille et al., 1975).The relationships between the shift of the half-activation potential and the divalentcation concentration, [X210, were well fitted by the Michaelis-Menton equation,S = Smax/(1 + Km/[X21),assuming a single binding reaction. The Km 's for the half-maximal shift of the activationcurve varied from 92 AM for Zn 2+ to 221 AM for Cd2+ and 3.4 mM for Ca". The right-ward shift (S) of the activation curve was saturable but at different concentrations of eachspecies of divalent cation in keeping with the differing KM values. The positive shifts of V'appeared to saturate at 250 AM Zn2+ , 2 mM Cd", and 20 mM Ca". The effect of Mg"was substantially less potent than the other divalents examined and at 40 mM, the highestconcentration of Mg2+ tested, it is unlikely the action of Mg2+ on IK(f) had saturated.Given the low potency of Mg2+ in shifting the activation and inactivation curvesobserved, it would be prudent to examine the effect of Mg 2+ without Ca2+ present.Removing Ca2+ from the external media resulted in approximately a 12 mV and 19 mV left-ward shift of the activation and inactivation curves, respectively. Examining the magnitudesof the Me-induced right-ward shifts of the activation and inactivation curves in zero Ca 2 +would be a more sensitive test of the actions of Mg2+.83The results reported here suggest quite strongly that the divalent cations areinteracting specifically with a saturable binding site. However, this study did not elucidatewhether each of the divalent cations tested was interacting at the same site or whetherdifferent binding sites were involved depending on the species of cation. Although 2 mMCa2+ and 1 mM Me + were present in both the control solutions and the test solutions,other than the solutions in which [Ca 24 ]0 was changed, the effect of each divalent cation onIK(f) was essentially examined in isolation. It would be of interest to determine whether ornot the effects of Zn 2+ and Cd2+ or Zn2+ and Ca2+ or Ca2+ and Mg2+ , for example, wereadditive. The effects of changing pH against varied divalent cation concentrations shouldalso shed some light on the nature of the binding site.Experiments with the frog Node of Ranvier, in which the concentrations of twodivalents were varied simultaneously (Mozhayeva & Naumov, 1972c) or the [Ca 2+ ]o and pH.were varied simultaneously (Hille et al., 1975), suggest that competition exists amongdivalent cations and between divalent cations and protons for the binding sites present onor near both the potassium channel (Mozhayeva & Naumov, 1972c) and the sodium channel(Hille et al., 1975). If the actions of divalent cations on I K(f) were additive at saturatingconcentrations one could postulate that each divalent cation was reacting at a uniquesaturable binding site, as suggested by Gilly and Armstrong (1982b) based on theirobservations of the additive effects of Hg 2+ and Zn2+ on the K + channel of the squid giantaxon. If the effects of the divalents were not additive at saturating concentrations this wouldsuggest that the cations were either binding at the same saturable site or that binding of onedivalent to its site precluded, by an allosteric effect, binding of the second divalent to its site.Divalent Cations Stabilize The Closed Conformation Of The Channel Conducting 'aThe kinetic parameters, th act and 42 inac„ measured for IK(f) were shifted along thevoltage axis to a greater degree than the activation and inactivation curves. For example,500 AM Zn2+ evoked a + 60 mV shift of both ty, act and ty, inact compared to a + 26 mV shift84of the activation and inactivation curves. The rightward shift of the activation kinetics forIK(f) in the presence of low concentrations of Cd 2+ and Zn2+ and high concentrations ofCa2+ and Mg2+ suggest that divalent cations may stabilize the closed conformation of thechannel conducting I K(f), perhaps by binding to a negatively charged component of thegating apparatus in a manner similar to that proposed by Gilly and Armstrong (1982 a &b) for the action of Zn2 + on the activation of both Na+ and IC channels in squid axon.Speculation On The Possible Characteristics Of The Binding SiteDivalent cations could be exerting their action on IK(f) by binding to a site either onthe channel protein itself or near enough to the channel to alter the membrane electric fieldsensed by the voltage sensor. The results provided in this study suggest the presence of atleast one divalent cation specific binding site. The transition metal ions, Cd2+ and Zn2+ ,exhibited a much higher affinity for the site than did the alkaline earth metal ions, Ca 2+ andmg2+ .Zn2+ demonstrates a 2-fold higher binding affinity for lipid than Ca2+ (Blaustein,1967) however, this does not seem to account for the 40-fold greater effect of Zn 2+ andCd2+ in altering the behaviour of IK(f). On the other hand, transition metal ions bind toproteins with a much higher affinity than Ca 2 + (Begenisich & Lynch, 1974). Zn 2+ and Cd2+ ,have respectively, a 234- and 400-fold higher affinity for imidazole groups, a 148- and 280-fold higher affinity for amino groups and a 6.3x10 6- and 6.3x107-fold higher affinity forsulfhydryl groups relative to Ca2+ (Begenisich & Lynch, 1974). The differences in bindingaffinity to proteins more than accounts for the different potency of the ions tested andsuggests strongly that the binding site, for the transition metal ions at least, is part of thechannel protein and not on the lipid surrounding the channel.Begenisich and Spires (1991) exposed the A-type K + channel from ShakerDrosophila, which has been molecularly characterized, to histidine- or sulfhydryl-modifyingagents but found that these agents did not alter the effects of Zn 2+ on channel behaviour.85However, the amino specific reagent trinitrobenzenesulfonic acid (TNBS), which convertsamino groups with a high pK to trinitrobenzene derivatives, completely abolished the Zn 2 +effect. It was postulated that the external divalent cation binding site contains one or moreimportant amino residues such as a lysine or a terminal amine (Begenisich & Spires, 1991).Cahalan and Pappone (1981) utilized TNBS, applied externally to frog skeletalmuscle fibres, to increase the negative surface charge on the membrane and observed thatTNBS resulted in a left-ward shift of both the activation and inactivation curves for the Nacurrent. In the Begenisich and Spires (1991) study, it is possible that rather than definingan amino component of the divalent cation binding site, TNBS simply counteracted thecharge screening effect of the Zn2+ which could have been interacting at an altogetherdifferent site.Schild and Moczydlowski (1991) exposed cardiac sodium channels to iodoacetamide(IAA), a sulfhydryl-specific alkylating agent and found it modified saxitoxin binding andcompletely abolished the block by Zn2 +. Sulfhydryl groups are often present at high-affinityZn2+ binding sites to coordinate the ion and Schild and Moczydlowski (1991) postulated thatthe binding site on the cardiac sodium channel contains at least one or more cysteine (Cys)residues. Examination of the amino acid sequence of several potassium channel clonesreveals three highly conserved Cys residues located in the first (S1), second (S2) and sixth(S6) transmembrane spanning regions (Bulter et al., 1989; Chandy et al., 1990; Kamb et al.,1988; Schwarz et al., 1988; Stiihmer et al., 1989; Swanson et al., 1990; Tempel et al., 1987;Tempel et al., 1988; Wei et al., 1990). Therefore, it is feasible that Cys residues areinvolved in the Zn2+ \Cd2+ binding site of IK(f) although the evidence provided byBegenisich & Spires (1991) argues against this possibility.Ca2+ and Mgt} were less potent than either Zn2+ or Cd2+ is shifting the voltage-sensitive parameters of I K(f). It is conceivable that more than one binding site exists on thechannel protein which display different affinities and selectivity for the divalent cations. Thedose-response curves fitted to a single binding site do not preclude this possibility as each86divalent cation was essentially examined in isolation. As discussed, the existence of at leasttwo sites for divalent cation action on the delayed rectifier IC channel in squid axons hasbeen put forward by Gilly and Armstrong (1982b).High [Zn21. And [Cd21. Reduce G.In most instances the apparent reduction of G. observed in the presence of a given[X210 was accounted for by the right-ward shift of the activation curve. If the right-wardshift of the activation curve was substantial G. may not have been attained since the cellswere not depolarized to potentials above + 80 mV. However, 500 AM Zn 2 + and 5 mM Cd2+caused respectively, a 23% and a 20% reduction in G max. In both cases, Gmax was calculatedat a membrane potential less depolarized than + 80 mV, the maximum potential to whichthe membrane was depolarized. These results are analogous to those of Mayer andSugiyama (1988) where they observed a 5-11% and 16-26% reduction in G max with 10 mMand 20 mM [Mn210, respectively, but not with [Mn2 10 < 5 mM.There is not a clear explanation for the reduction of G max in the presence of highconcentrations of Zn2+ and Cd2+ . However, several possibilities suggest themselves. At themacroscopic current level it is difficult to distinguish unequivocally between possible chargescreening effects and the voltage-dependent block of I K(f) by divalent cations. Studies ofthe effect of Zn2 + and Cd 2 + on the conductance of the cardiac sodium channel indicate thatthese ions reduce channel conductance by inducing a voltage-dependent conversion of thechannel to a subconductance state (Ravindran et al., 1991; Schild et al., 1991). It is possiblethat Zn2+ and Cd2+ are having a similar action on IK(f), however, single channel analysis ornoise analysis of macroscopic currents would be required to study this possibility.The presence of divalent cations in the double layer could decrease the concentrationof permeant ions near the mouth of the channel due to electrostatic interactions whichwould also reduce conductance (Green & Andersen, 1991). For example, Cooper andShrier (1989) reported that the conductance of A channels in cultured sensory neurones87varied with the square root of the external K + concentration. If the local concentration ofK + was reduced due to the presence of divalent cations then it is reasonable to expect thatthe conductance of the channel could be reduced. The reduction in conductance may bethe result of the specific interaction of the divalent cations with a site on or close to thechannel as well as, non-specific electrostatic effects created when the extracellularconcentration of these cations is increased.In the calculation of the chord conductance ER was assumed not to have changedwhen divalent cation concentration was altered, however, the effects of divalent cations onER remain to be tested.Is Ca2+ A Necessary Co-factor For I K(f)? Removing Cao2 + significantly increased (p < 0.0001) the slope-factor of the activationcurve for IK(f) and decreased the peak amplitude of the current. The slope-factor of theinactivation curve was also increased, however, this increase was not significant (p > 0.01).The increase in the slope-factor of the activation curve translates into a decrease in theequivalent gating charge transferred across the membrane during the activation process.The reduction of peak IK(f) caused by zero Ca' is explained by the left-ward shift of theinactivation curve. Although Ca" has been identified as an essential co-factor in the gatingof Na channels in GH 3 clonal pituitary cells (Armstrong and Cota, 1991) and IA-like currents(Begenisich, 1988) it is unclear whether Ca 2+ fulfils a similar role in the gating of I K(f). Thechange in the slope-factors of the activation curve, in zero Ca 2 +, suggests that Ca" may playsome role in the movement of the gating charge across the membrane during theactivation/inactivation processes.Lack of Ca2+ in the bath solution also resulted in an increase in the leak current inmelanotrophs. Armstrong, has provided evidence which suggests that Ca 2+ ions stabilize theclosed conformation of delayed rectifier type IC channels (Armstrong & Matteson, 1986)and are necessary factors in maintaining the functional integrity of these channels88(Armstrong and LOpez-Barneo, 1987). It is not clear from the present study whether theincrease in the leak current was due specifically to a loss in the selectivity of the channelcarrying IK(f), as suggested by Armstrong and II•pez-Barneo (1987), or whether the lack ofCa2+ in the external media simply destabilized the membrane and consequently resulted inan increased leakage conductance.The Activation and Inactivation Of I K(f) Appear To Be CoupledEqual shifts of both the activation and inactivation curves of I K(f) suggests that theseprocesses are coupled. The Debye length, which acts as a guide as to how far into asolution the electrostatic effects of a charge can be felt, is less than 10 A in standard Ringersolution (Hille, 1984). If activation and inactivation were not coupled it is quite feasiblethat the gating mechanisms governing these two processes would experience very differentlocal electric fields even if they were only 20 A apart (Hille, 1984). This difference wouldbe observed in the degree to which activation and inactivation were influenced by thepresence of divalent cations. As both the activation and inactivation of I K(f) were affectedequally it is more likely that the mechanisms governing these processes experienced thesame local electrical field. Inactivation of I K(f) would become voltage-dependent if it werestrongly coupled to activation which is a highly voltage-sensitive process (Armstrong &Bezanilla, 1977). Activation and inactivation could be coupled due to physical constraintsimposed by the conformation of the channel protein. Movement of the activation gatingparticle might be necessary in order to expose a site needed for inactivation to proceed, asproposed in the "ball and chain" model of inactivation (Armstrong and Bezanilla, 1977;Hoshi et al., 1990; Zagotta et al., 1990).The results with zero Ca2+ , on the other hand, suggest that Ca2+ ions in particularare capable of affecting charge transfer during the gating process of activation withoutinfluencing the gating charge movement which occurs during inactivation. Zero Ca 2+significantly increased the slope-factor of activation which translates into a reduction in the89equivalent gating charge transferred during the activation process, whereas, there was nosignificant effect on the slope-factor of the inactivation curve (p > 0.01). This result suggeststhat the activation process can be acted on separately from the inactivation process by Ca",however, further investigation into the actions of zero Ca" are required to determinewhether the action on the slope-factor for inactivation is truly insignificant. Armstrong andMatteson (1986) have proposed that in addition to binding to a negatively charged portionof the gating apparatus, Ca" and perhaps Mg' normally occupies the closed K + channelin squid axon. Ca" may operate in a similar fashion at the channel conducting I K(f),movement of this ion out of the channel prior to activation could account for some of thecharge movement during the activation processes and could explain the reduction in chargemovement observed in zero Ca".Physico-Chemical Properties Of The Divalent CationsDifferences in physico-chemical properties, such as, ionic radius, electronegativity,and hydration energy, among the divalent cations tested may explain in part the disparityin their relative abilities to shift the voltage-dependent parameters of I K(f). Most notably,Mg", which was the least effective, has a mean hydration shell of approximately 13 watermolecules (Arhem, 1980b) and exhibits a water substitution rate of 10 5 s-1 (HiIle, 1984) whichis the slowest among the divalent cations examined. The slow replacement of waters aroundMg" may be a factor in the ability of Mg" to approach and bind at the putative divalentcation binding site in a manner similar to that described by Hille (1984) to account for thereduced permeability of small ionic channels to Mg", Ni" and Co", each of which holdsonto oxygen ligands longer than other inorganic cations. It would be of interest to examine,in detail, the effects of Ni" and Co" on the behaviour of I K(f) to determine whetherhydration energy is a determining factor in the interaction of divalent cations with thechannel conducting IK(f). Differences in electronegativity are unlikely to account for theresults presented here as Mg" is less electronegative than Zn" or Cd" but more90electronegative than Ca2+ . However, the increased electronegativity of Zn 2+ and Cd2+versus Ca2 + and Mg2 + might account in part, for the greater effect on I K(f) caused bytransition metal ions. Differences in ionic radii are not consistent with the results reportedhere (Arhem, 1980a & 1980b).Summary and Future DirectionsDivalent cations evoke a right-ward shift of the potential-sensitive parameters of IK(f)along the voltage axis. The ions are not equally potent in their actions which providescompelling evidence that divalent cations exert their effects through interaction with at leastone specific and saturable binding site. The presence of divalent cations slows the activationkinetics of IK(f) perhaps by stabilizing the I K(f) channel in the closed conformation.Single channel analysis could be used to determine whether divalent cations exerttheir effects on IK(f) solely by shifting the voltage-dependence of gating or, alternatively, theextent of the contribution by voltage-dependent block, or transformation to one or moresubconductance states.At the macroscopic current level, experiments to determine whether divalent cationsare competing for the same binding site on the channel conducting I K(f) would helpquantitate the number of the binding sites.Exposing melanotrophs to iodoacetamide or TNBS to determine whether either ofthese agents prevents Zn2+ or Cd2+ from affecting the behaviour of IK(f) might help todetermine the amino acid components of the binding site for these transition metal ions onthe channel conducting IK(f).Site directed mutagenesis could potentially be used to examine whether specificamino acid residues are important in defining a divalent cation binding site in the knownpotassium channel clones. However, given that the residues involved in forming the bindingsite might also be involved in forming the channel pore, this technique might not yielddefinitive results. Moreover, results using this technique must be analyzed keeping in mind91that single amino acid changes can have global effects on channel configuration and that thestructure of potassium channels even within a subclass of K+ channels such as thosesubserving A-like currents are quite diverse.Concluding RemarksDivalent cations, particularly Cd 2 +, have been used to block voltage-sensitive Ca 2+currents in order to isolate other currents of interest (Kehl, 1989) and as a diagnostic toolto determine whether Ca2+ influx through voltage-gated Ca2+ channels is involved in theactivation of other currents (Douglas & Taraskevich, 1982). The results of this studyemphasize that even at the concentrations typically used to block Ca 2+ currents (e.g., 100-300 AM Cd2 +) the gating of K+ channels can be substantially altered.Protein phosphorylation has been implicated as a mechanism by which the activityof channels conducting the TOC are modulated (DiFrancesco & Tortora, 1991; Braun et al.,1990). The modulatory action of phosphorylation appears in part to involve an alterationof the surface potential sensed by voltage-gated channels through interactions at thecytoplasmic face of the membrane. Phosphorylation of the delayed rectifier K + channel insquid axon has been observed to shift its voltage-dependent parameters and result inmodification of its kinetic and conductive properties suggesting that electrostatic interactionsmay play an important role in modulating the behaviour of voltage-dependent channels(Perozo et al., 1989; Perozo & Bezanilla, 1990). Rudy et al. (1988) have observed that A-currents expressed in Xenopus oocytes required the presence of a large (6-7 kb) and a small(2-4 kb) mRNA species in order to display normal kinetics and pharmacology. They havesuggested that the small mRNA species encodes a second subunit of the A-channel whichmay modulate its behaviour. It is quite feasible that the channel conducting I K(f) is alsomodulated by phosphorylation and that part of this modulation involves electrostaticinteractions similar to those observed in this study.92BIBLIOGRAPHYAdrian, E.D. and Gelfan, S. (1933) Rhythmic activity in skeletal muscle fibres. Journal ofPhysiology 78, 271-287.Arhem, P. (1980a) Effects of some heavy metal ions on the ionic currents of myelinatedfibres from Xenopus laevis. Journal of Physiology 306, 219-231.Arhem, P. (1980b) Effects of rubidium, caesium, strontium, barium and lanthanum on ioniccurrents in myelinated nerve fibres from Xenopus laevis. Acta Physiol Scand 108, 7-16.Armstrong, C. and Bezanilla, F. (1977) Inactivation of the sodium channel II. Gating currentexperiments. Journal of General Physiology 70, 567-590.Armstrong, C. and Cota, G. (1991) Calcium ion as a cofactor in Na channel gating.Proceedings of the National Academy of Sciences USA 88, 6528-6531.Armstrong, C. and LOpez-Barneo, J. (1987) External calcium ions are required for potassiumchannel gating in squid neurons. Science 236, 712-714.Armstrong, C. and Matteson, D.R. (1986) The role of calcium ions in the closing of Kchannels. Journal of General Physiology 87, 817-832.Back, N. and Rechardt, L. (1985) The effect of reserpine on the pars intermedia of the ratpituitary. Cell and Tissue Research 241, 1-8.Begenisich, T. and Lynch, C. (1974) Effects of internal divalent cations on voltage-clampedsquid axons. Journal of General Physiology 63, 675-689.Begenisich, T. and Spires, S. (1991) The divalent cation binding site on potassium channels.Society For Neuroscience Abstracts 17(2), 1474.Benndorf, K. (1989) A reinterpretation of Na channel gating and permeation in terms of aphase transition between a transmembrane S4 a-helix and a channel-helix. EuropeanBiophysical Journal 17(5), 257-271.93Betz, H. (1990) Homology and analogy in transmembrane channel design: lessons fromsynaptic membrane proteins. Biochemistry 29(15), 3591-3599.Blaustein, M.P. and Goldman, D.E. (1968) The action of certain polyvalent cations on thevoltage-clamped lobster axon. Journal of General Physiology 51, 279-291.Braun, A.P., Fedida, D., Clark, R.B. and Giles, W.R. (1990) Intracellular mechanisms foraradrenergic regulation of the transient outward current in rabbit atrial myocytes. Journalof Physiology 431, 689-712.Brink, F. (1954) The role of calcium ions in neural processes. Pharmacological Review 6,243-298.Butler, A., Wei, A., Baker, K., and Salkoff, L. (1989) A family of putative potassium channelgenes in Drosophila. Science 243, 943-947.Cahalan, M.D. and Pappone, P.A. (1981) Chemical modification of sodium channel surfacecharges in frog skeletal muscle by trinitrobenzene sulphonic acid. Journal of Physiology 321,127-139.Campbell, D.T. (1982) Do protons block Na + channels by binding to a site outside the pore?Nature 298, 165-167.Campbell, D.T. and Hille, B. (1976) Kinetic and pharmacological properties of frog skeletalmuscle. Journal of General Physiology 67, 309-323.Catterall, W. (1988) Genetic analysis of ion channels in vertebrates. Annual Review ofPhysiology 50, 395-406.Chandler, W.K., Hodgkin, A.L. and Meves, H. (1965) The effect of changing the internalsolution on sodium inactivation and related phenomena in giant axons. Journal of Physiology180, 821-836.Chandy, K.G., Williams, C.B., Spencer, R.H., Aguilar, B.A., Ghanshani, S., Tempel, B., andGutman, G.A. (1990) A family of three mouse potassium channel genes with intronlesscoding regions. Science 247, 973-975.94Clay, J.R. (1985) Comparison of the effects of internal TEA + and Cs + on potassium currentin squid giant axons. Biophysical Journal 34, 885-892.Connor, J.A. and Stevens, C.F. (1971) Voltage clamp studies of a transient outwardmembrane current in Gastropod neural somata. Journal of Physiology  213, 21-30.Cooper, E. and Shrier, A. (1989) Inactivation of A currents and A channels on rat nodoseneurones in culture. Journal of General Physiology 94, 881-910.Cota, G. (1986) Calcium currents in pars intermedia cells of the rat pituitary gland. Journalof General Physiology 88, 83-105.Cukierman, S. and Krueger, B.K. (1990) Modulation of sodium channel gating by externaldivalent cations: differential effects on opening and closing rates. Pfliigers Archivs 416, 360-367.D'Arrigo, J.S. (1978) Screening of membrane surface charges by divalent cations: an atomicrepresentation. American Journal of Physiology 235, c109-c117.de Rijk, E.P.C., Jenks, B.G., Vaudry, H., and Roubos, E.W. (1990) GABA andNeuropeptide Y co-exist in axons innervating the neurointermediate lobe of the pituitary ofXenopus Laevis - an immunoelectron microscopic study. Neuroscience 38(2), 495-502.DiFrancesco, D. and Tortora, P. (1991) Direct activation of cardiac pacemaker channels byintracellular cyclic AMP. Nature 351, 145-147.Douglas, W.W. and Taraskevich, P.S. (1978) Action potentials in gland cells of rat pituitarypars intermedia: inhibition by dopamine, an inhibitor of MSH secretion. Journal ofPhysiology 285, 171-184.Douglas, W.W. and Taraskevich, P.S. (1982) Slowing effect of dopamine and calciumchannel blockers on frequency of sodium spike in rat pars intermedia cells. Journal ofPhysiology 326, 201-211.Drouin, H. and Neumcke, B. (1974) Specific and unspecific charges at the sodium channelsof the nerve membrane. Pfliigers Archivs 351, 207-229.95Frankenhaeuser, B. and Hodgkin, A.L. (1957) The action of calcium on the electricalproperties of squid axons. Journal of Physiology  137, 218-244.Gilbert, D.L. and Ehrenstein, G. (1969) Effects of divalent cations on potassiumconductance of squid axons: determination of surface charge. Biophysical Journal 9, 447-469.Gilbert, D.L. and Ehrenstein, G. (1984) Membrane surface charge. Current Topics inMembrane Transport 22, 407-421.Gilly, W.F. and Armstrong, C.M. (1982a) Slowing of sodium channel opening kinetics insquid axon by extracellular zinc. Journal of General Physiology  79, 935-964.Gilly, W.F. and Armstrong, C.M. (1982b) Divalent cations and the activation kinetics ofpotassium channels in squid Giant axons. Journal of General Physiology  79, 965-996.Green, W.N. and Andersen, O.S. (1991) Surface charges and ion channel function. AnnualReview of Physiology 53, 341-359.Hahin, R. and Campbell, D.T. (1983) Simple shifts in the voltage dependence of sodiumchannel gating caused by divalent cations. Journal of General Physiology  82, 785-805.Hamill, 0.P., Neher, M.E., Sakmann, B., and Sigworth, F.J. (1981) Improved patch-clamptechniques for high-resolution current recording from cells and cell-free membrane patches.Pflagers Archivs 391, 85-100.Hille, B. (1968) Charges and potential at the nerve surface: divalent ions and pH. Journalof General Physiology 51, 221-236.Hille, B. (1984) Ionic Channels of Excitable Membranes, Sinauer Assc. Inc., Sunderland,Mass.Hille, B., Woodhull, A.M., and Shapiro, B.I. (1975), Negative surface charges near sodiumchannels of nerve: divalent ions, monovalent ions and pH. Philosophical Transactions of theRoyal Society London B. 270, 301-318.96Hodgkin, A.L. and Huxley, A.F. (1952) Currents carried by sodium and potassium ionsthrough the membrane of the giant axon of Loligo. Journal of Physiology 116, 449-472.Hoshi, T., Zagotta, W.N., and Aldrich, R.W. (1990) Biophysical and molecular mechanismsof Shaker potassium channel inactivation. Science 250, 533-538.Jan, L.Y., and Jan, Y.N. (1989) Voltage-sensitive ion channels. Cell 56, 13-25.Jan, L.Y. and Jan, Y.N. (1990) How might the diversity of potassium channels be generated?Trends in Neuroscience  13(10), 415-419.Kamb, A., Iverson, L.E. and Tanouye, M.A. (1987) Molecular characterization of Shaker,a Drosophila gene that encodes a potassium channel. Cell 50, 405-413.Kamb, A., Tseng-Crank, J., and Tanouye, M.A. (1988) Multiple products of the DrosophilaShaker gene may contribute to potassium channel diversity. Neuron 1, 421-430.Kehl, Si. (1987) A high-threshold non-inactivating calcium current in isolated melanotrophsin vitro. Journal of Physiology 391, 74P.Kehl, S.J. (1989) Cultured melanotrophs of the adult rat pituitary possess a voltage-activatedfast transient outward current. Journal of Physiology 411, 457-468.Kehl, S.J. (1990) 4-Aminopyridine causes a voltage-dependent block of the transient outwardK + current in rat melanotrophs. Journal of Physiology 431, 515-528.Kehl, Si., Hughes, D. and McBurney, R.N. (1987) A patch clamp study of GABA-inducedmacroscopic currents in rat melanotrophs in cell culture. British Journal of Pharmacology92, 573-585.Mayer, M.L. and Sugiyama, K. (1988) A modulatory action of divalent cations on a transientoutward current in cultured rat sensory neurones. Journal of Physiology 396, 417-433.McBurney, R.N. and Kehl, S.J. (1988) Electrophysiology of neurosecretory cells from thepituitary intermediate lobe. Journal of Experimental Biology  139, 317-328.97McLaughlin, S.G.A., Szabo, G. and Eisenman, G. (1971) Divalent ions and the surfacepotential of charged phospholipid membranes. Journal of General Physiology 58, 667-687.Mozhayeva, G.N. and Naumov, A.P. (1972a) Effect of the surface charge on the steadypotassium conductivity of the membrane of a Node of Ranvier - I. Changes in pH ofexternal solution. Biofizika 17, 412-420.Mozhayeva, G.N. and Naumov, A.P. (1972b) Effect of the surface charge on the steadypotassium conductivity of the membrane of a Node of Ranvier - II. Changes in ionicstrength of the external solution. Biofizika 17, 618-622.Mozhayeva, G.N. and Naumov, A.P. (1972c) Influence of the surface charge on the steadypotassium conductivity of the membrane of a Node of Ranvier - III. Effect of bivalentcations. Biofizika 17, 801-808.Mozhayeva, G.N. and Naumov, A.P. (1983) The permeability of sodium channels tohydrogen ions in nerve fibres. Pflugers Archivs 396, 163-173.Mozhayeva, G.N., Naumov, A.P. and Negulyaeu, Yu.A. (1981) Evidence for the existenceof two acid groups controlling the conductance of sodium channel. Biochimica et BiophysicaActa. 643, 251-255.O'Donohue, T.L. and Dorsa, D.M. (1982) The opiomelanotropinergic neuronal andendocrine system. Peptides 3, 353-395.Papazian, D.M., Schwarz, T.L., Tempel, B.L., Jan, Y.N., and Jan, L.Y. (1987) Cloning ofgenomic and complementary DNA from Shaker, a putative potassium channel gene fromDrosophila. Science 237, 749-753.Papazian, D.M., Schwarz, T.L., Tempel, B.L., Timpe, L.C. and Jan, L.Y. (1988) Ion channelsin Drosophila. Annual Review of Physiology  50, 379-394.Perozo, E., Bezanilla, F., and Dipolo, R. (1989) Modulation of K channels in dialysed squidaxons. Journal of General Physiology  93, 1195-1218.Perozo, E. and Bezanilla, F. (1990) Phosphorylation affects voltage gating of the delayedrectifier K + channel by electrostatic interactions. Neuron 5, 685-690.98Ravindran, A., Schild, L. and Moczydlowski, E. (1991) Divalent cation selectivity for externalblock of voltage-dependent Na + channels prolonged by Batrachotoxin. Journal of GeneralPhysiology 97, 89-115.Rudy, B. (1988) Diversity and ubiquity of K channels. Neuroscience 25, 729-749.Rudy, B., Hoger, J.H., Lester, H.A., and Davidson, N. (1988) At least two mRNA speciescontribute to the properties of rat brain A-type potassium channels expressed in Xenopusoocytes. Neuron 1, 649-658.Schild, L. and Moczydlowski, E. (1991) Competitive binding interaction between Zn 2+ andsaxitoxin in cardiac Na + channels evidence for a sulfhydryl group in the Zn2 Vsaxitoxinbinding site. Biophysical Journal  59, 523-537.Schild, L., Ravindran, A. and Moczydlowski, E. (1991) Zn 2 +-induced subconductance eventsin cardiac Na + channels prolonged by Batrachotoxin. Journal of General Physiology  97, 117-142.Schwarz, T.L., Tempel, B.L., Papazian, D.M., Jan, Y.N. and Jan, L.Y. (1988) Multiplepotassium-channel components are produced by alternative splicing at the Shaker locus inDrosophila. Nature 331, 137-142.Sigworth, S.J. (1980) The conductance of sodium channels under conditions of reducedcurrent at the Node of Ranvier. Journal of Physiology 307, 131-142.Stiihmer, W. (1991) Structure-function studies of voltage-gated ion channels. Annual Reviewof Biophysics and Biophysical Chemistry  20, 65-78.Stilhmer, W., Ruppersberg, J.P., SchrOter, K.H., Sakmann, B., Stocker, M., Giese, K.P.,Perschke, A., Baumann, A., and Pongs, 0. (1989) Molecular basis of functional diversity ofvoltage-gated potassium channels in mammalian brian. The EMBO Journal 8(11), 3235-3244.Swanson, R., Marshall, J., Smith, J.S., Williams, J.B., Boyle, M.B., Folander, K., Luneau,C.J., Antanavage, J., Oliva, C., Buhrow, S.A., Bennett, C., Stein, R.B., and Kaczmarek, L.K.(1990) Cloning and expression of cDNA and genomic clones encoding three delayed rectifierpotassium channels in rat brain. Neuron 4, 929-939.99Tempel, B.L., Papazian, D.M., Schwarz, T.L., Jan, Y.N. and Jan, L.Y. (1987) Sequence ofa probable potassium channel component encoded at Shaker locus of Drosophila. Science237, 770-775.Tempel, B.L., Jan, Y.N. and Jan, L.Y. (1988) Cloning of a probable potassium channel genefrom mouse brain. Nature 332, 837-839.Tomiko, S.A., Taraskevich, P.S. and Douglas, W.W. (1981) Potassium-induced secretion ofmelanocyte-stimulating hormone from isolated pars intermedia cells signals participation ofvoltage-dependent calcium channels in stimulus-secretion coupling. Neuroscience 6(11),2259-2267.Unwin, N. (1989) The structure of ion channels in membranes of excitable cells. Neuron 3,665-676.Wei, A., Covarrubias, M., Butler, A., Baker, K., Pak, M., and Salkoff, L. (1990) K + currentdiversity is produced by an extended gene family conserved in Drosophila and mouse.Science 248, 599-603.Weidmann, S. (1955) Effects of calcium ions and local anaesthetics on electrical propertiesof Purkinje fibres. Journal of Physiology  129, 568-582.Zagotta, W.N., Hoshi, T., and Aldrich, R.W. (1990) Restoration of inactivation in mutantsof Shaker potassium channels by a peptide derived from ShB. Science 250, 568-571.Zbicz, K.L. and Weight, F.F. (1985) Transient voltage and calcium-dependent outwardcurrents in hippocampal CA3 pyramidal neurons. Journal of Neurophysiology 53, 1038-1058.

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.831.1-0086542/manifest

Comment

Related Items