determinants of voltage-dependent gating and open-state … · embo [eur. mol. biol. organ.] j....

215

J. Gen. Physiol.

© The Rockefeller University Press

•

0022-1295/99/08/215/28 $5.00Volume 114 August 1999 215–242http://www.jgp.org

Determinants of Voltage-dependent Gating and Open-State Stability in

the S5 Segment of

Shaker

Potassium Channels

Max Kanevsky

and

Richard W. Aldrich

From the Howard Hughes Medical Institute and Department of Molecular and Cellular Physiology, Stanford University School of Med-icine, Stanford, California 94305

abstract

The best-known

Shaker

allele of

Drosophila

with a novel gating phenotype,

Sh

5

, differs from the wild-type potassium channel by a point mutation in the fifth membrane-spanning segment (S5) (Gautam, M., and M.A.Tanouye. 1990.

Neuron.

5:67–73; Lichtinghagen, R., M. Stocker, R. Wittka, G. Boheim, W. Stühmer, A. Ferrus, andO. Pongs. 1990.

EMBO [Eur. Mol. Biol. Organ.] J.

9:4399–4407) and causes a decrease in the apparent voltage de-pendence of opening. A kinetic study of

Sh

5

revealed that changes in the deactivation rate could account for the al-tered gating behavior (Zagotta, W.N., and R.W. Aldrich. 1990.

J. Neurosci.

10:1799–1810), but the presence of intactfast inactivation precluded observation of the closing kinetics and steady state activation. We studied the

Sh

5

muta-tion (F401I) in ShB channels in which fast N-type inactivation was removed, directly confirming this conclusion.Replacement of other phenylalanines in S5 did not result in substantial alterations in voltage-dependent gating. Atposition 401, valine and alanine substitutions, like F401I, produce currents with decreased apparent voltage de-pendence of the open probability and of the deactivation rates, as well as accelerated kinetics of opening and clos-ing. A leucine residue is the exception among aliphatic mutants, with the F401L channels having a steep voltagedependence of opening and slow closing kinetics. The analysis of sigmoidal delay in channel opening, and of gat-ing current kinetics, indicates that wild-type and F401L mutant channels possess a form of cooperativity in the gat-ing mechanism that the F401A channels lack. The wild-type and F401L channels’ entering the open state gives riseto slow decay of the OFF gating current. In F401A, rapid gating charge return persists after channels open, con-firming that this mutation disrupts stabilization of the open state. We present a kinetic model that can account forthese properties by postulating that the four subunits independently undergo two sequential voltage-sensitive tran-sitions each, followed by a final concerted opening step. These channels differ primarily in the final concertedtransition, which is biased in favor of the open state in F401L and the wild type, and in the opposite direction inF401A. These results are consistent with an activation scheme whereby bulky aromatic or aliphatic side chains at

position 401 in S5 cooperatively stabilize the open state, possibly by interacting with residues in other helices.

key words:

gating current • ion channel • site-directed mutagenesis • activation • cooperativity

i n t r o d u c t i o n

Potassium channels exert a stabilizing influence overthe membrane potential of excitable cells, shaping thepatterns of their electrical activity and serving as targetsfor modulators and drugs. To perform this role, manypotassium channels have evolved exquisite sensitivity to

transmembrane voltage. The

Shaker

channel, a member

of the family of voltage-gated (Kv)

1

potassium channels(Kamb

et al., 1987; Papazian

et al., 1987; Pongs

et al.,

1988) has been studied extensively as a model system

for mechanistic studies of K

1

channel function.Voltage-dependent gating refers to the conforma-

tional transitions that the channel protein can undergoin which intrinsic charged or dipolar groups (gatingcharges) move in response to changes in the mem-brane voltage. Structurally, voltage-gated potassiumchannels exist as tetramers of like alpha subunits(MacKinnon, 1991; Kavanaugh

et al., 1992; Liman

etal., 1992, Doyle et al., 1998), often in association with

accessory subunits (Xu

et al., 1998). Extensive site-directed mutagenesis of basic amino acids in the fourthtransmembrane segment (S4) has confirmed their im-portance in the voltage-dependent gating of

Shaker

andother potassium channels (Liman

et al., 1991; Papazianet al., 1991; Logothetis

et al., 1992) and has shown thatseveral S4 basic residues comprise a portion of the gat-ing charge (Perozo

et al., 1994; Aggarwal and MacKin-non, 1996; Seoh

et al., 1996). In addition, in both

Shaker

and muscle sodium channels, the S4 has beenshown to be translocated across the membrane in a

Portions of this work have been previously published in abstract form

(Kanevsky, M., and R.W. Aldrich. 1994.

Biophys. J.

66:A283; Kanevsky,M., and R.W. Aldrich. 1995.

Biophys. J.

68:A136).Address correspondence to Dr. R.W. Aldrich, Howard Hughes

Medical Institute and Department of Molecular and Cellular Physiol-ogy, Beckman Center B-171, Stanford University School of Medicine,

Stanford, CA 94305-5426. Fax: 650-725-4463; E-mail: [email protected]

1

Abbreviations used in this paper:

G(V), steady state voltage depen-dence of the open probability; Kv, voltage gated; Q(V), charge dis-placement versus voltage; S4, fourth transmembrane segment; S5,fifth membrane-spanning segment; wt, wild type.

Dow

nloaded from http://rupress.org/jgp/article-pdf/114/2/215/1192593/gp-7910.pdf by guest on 12 June 2021

216

S5 and Voltage Gating in Shaker

voltage-dependent fashion correlating with the activa-tion process (Yang and Horn, 1995; Larsson

et al.,1996; Mannuzzu

et al., 1996; Yang

et al., 1996; Yusaf

etal., 1996). Mutations of uncharged S4 and S4–-S5 seg-ment amino acids also exert strong effects on activationgating (Lopez

et al., 1991; McCormack

et al., 1991;Schoppa

et al., 1992; Logothetis

et al., 1993; Schoppaand Sigworth, 1998b), in particular influencing late co-operative transitions in the activation pathway (Smith-Maxwell

et al., 1998b). There is evidence that an acidicresidue in the S2 segment may be part of the gatingcharge and act as one of the countercharges for the S4gating charges (Papazian

et al., 1995; Seoh

et al., 1996;Keynes and Elinder, 1999), but the nature of interac-tions between S4 residues with other regions of thechannel remains unclear.

We have used a previously studied mutant

Shaker

al-lele as a starting point to help understand the role ofthe fifth membrane-spanning segment (S5) in activa-tion gating and as a potential interacting partner forS4.

Sh

5

is the best known mutant

Shaker

allele that af-fects voltage-dependent gating in

Drosophila

. It differsfrom the wild-type sequence by a phenylalanine-to-iso-leucine substitution located in the S5 transmembranesegment: F401I (Gautam and Tanouye, 1990; Lichting-hagen

et al., 1990). Whereas most mutant

Shaker

alleles(e.g.,

Sh

KS133

,

Sh

102

) have a loss-of-function phenotype(Salkoff, 1983), eliminating the transient “A

1

-type” po-tassium current and broadening the action potential(Tanouye

et al., 1981; Salkoff, 1983; Tanouye and Fer-rus, 1985; Wu and Haugland, 1985),

Sh

5

fly nerves firerapid spikes that, because of failure to repolarize com-pletely, occur in bursts (Tanouye

et al., 1981; Tanouyeand Ferrus, 1985). In

Sh

5

muscle fibers, A-type currentspossess novel gating properties that have been reportedas either shifting voltage dependence of activation andinactivation to a more positive range (Wu and Haug-land, 1985) or speeding up the kinetics of inactivationand recovery (Salkoff, 1983). Close examination of thevoltage dependence of steady state inactivation re-vealed that the slope is somewhat shallower in

Sh

5

thanin the wild type (Wu and Haugland, 1985; Zagotta andAldrich, 1990b), suggesting that this mutation may re-duce the apparent activation gating valence. Kineticmodeling of

Sh

5

channels showed that changes in therate and voltage dependence of deactivation could ac-count for the altered gating behavior (Zagotta and Ald-rich, 1990b). However, in the previous studies closingkinetics and steady state activation could not be directlymeasured in the native channels due to the presence ofN-type inactivation.

Using the background of an NH

2

terminus–trun-cated version of the wild-type

Shaker

channel free of fastN-type inactivation (Hoshi

et al., 1990), we introducedaliphatic point substitutions of the phenylalanine at po-

sition 401, the site of the

Sh

5

mutation. We askedwhether changes in the size of the side chain at posi-tion 401 would lead to predictable consequences forvoltage-dependent gating. The loss of apparent gatingvalence in

Sh

5

and other F401 substitutes is associatedwith a decrease in the voltage dependence of the back-ward transitions leading away from the open state. Weextended the kinetic analysis of the wt and F401 mu-tants to determine which transitions between confor-mational states are disrupted by the substitutions. Thephysical picture of the channel as a tetramer of iden-tical subunits (MacKinnon, 1991; Hurst

et al., 1992;Liman

et al., 1992; Doyle et al., 1998) restricts the inter-pretation of our results to schemes with fourfold sym-metrical alterations in the gating parameters or withchanges to concerted transitions between quaternaryconformations of the channel.

Several gating mechanisms for potassium channelsthat incorporate independent and cooperative steps inthe activation process have been proposed (Koren

etal., 1990; Zagotta and Aldrich, 1990a; Tytgat and Hess,1992; Bezanilla

et al., 1994; McCormack

et al., 1994;Zagotta

et al., 1994b; Schoppa and Sigworth, 1998c).We use the aliphatic substitutions at F401 to test theability of the type of models exemplified by Zagotta

etal. (1994b) and Schoppa and Sigworth (1998c) to pre-dict the mutants’ divergent gating properties on the ba-sis of physically interpretable alterations of just a fewtransitions. We conclude that F401 is involved in the co-operative stabilization of the open state.

m a t e r i a l s a n d m e t h o d s

Terminology

All mutant channel constructs were made in the ShB

D

6-46 back-ground (Hoshi

et al., 1990), a deletion mutant in which fast N-typeinactivation has been disrupted. These parent channels will bereferred to as wild type (wt), and channels containing furthersingle amino acid substitutions will be designated AxxxB, wherexxx is the position of the amino acid in the deduced sequence ofShB (Tempel

et al., 1987). For gating current measurements, weused a version of the wild-type channel (free of N-type inactiva-tion) containing the mutation W434F in the pore region thatnearly completely abolishes ion conduction but not charge move-ment (Perozo

et al., 1993; Yang

et al., 1997). This construct willbe referred to as wf, and point substitutions in its backgroundwill be termed wfAxxxB.

Site-directed Mutagenesis and Oocyte Expression

All conducting versions of constructs containing point substitu-tions in the S5 region were generated by synthetic oligonucle-otide-directed cassette mutagenesis using the polymerase chainreaction. To record gating charge movement, a high-expressionvector containing the W434F mutation (Perozo

et al., 1993; Yanget al., 1997) was obtained from Ligia Toro (UCLA School ofMedicine, Los Angeles, CA). We subcloned inserts containingthe alanine and leucine substitutions for F401 into the W434Fconstruct (wf). Fidelity of DNA synthesis was verified by dideoxy

Dow


217

Kanevsky and Aldrich

termination sequencing (Sanger

et al., 1977) of the region span-ning the cassette insert. cRNAs were transcribed in vitro fromplasmid templates, linearized with SacI or NdeI (HindIII for wf-based constructs), using the mMessage mMachine kit with T7RNA polymerase (Ambion Inc.) and injected into

Xenopus laevis

oocytes 2–14 d before recording.

Electrophysiology

Patch-clamp recordings from oocytes were carried out using theAxopatch 200A amplifier (Axon Instruments) with borosilicateglass pipettes (initial tip resistances between 0.4 and 2 M

V

). Mac-roscopic ionic currents recorded in the inside-out and outside-out excised configurations (Hamill

et al., 1981) were low-passfiltered at 5–10 kHz with an eight-pole Bessel filter (FrequencyDevices, Inc.) and acquired online with sampling frequencies be-tween 10 and 100 kHz using an ITC-16 interface board (In-strutek) and a Macintosh computer running Pulse software(HEKA Electronik). In all experiments, care was taken to allowtail current kinetics to settle to a steady level for 3–5 min afterpatch excision to the inside-out configuration before acquiringdata. Patches that showed significant drift in the tail current timeconstant over the course of the experiment were excluded fromanalysis. No series resistance compensation was used.

To improve the signal-to-noise ratio for gating current experi-ments, we used a high-performance cut-open oocyte clamp (CA-1;Dagan Inc.) (Taglialatela

et al., 1992; Stefani

et al., 1994). Goodvoltage control and dynamic response were obtained by permeabi-lizing the lower dome with 0.3% saponin solution and using agarbridges filled with 1 M NaMES containing fine platinum-iridiumwire. Low-resistance (

,

1 M

V

) glass microelectrodes were filledwith 3 M KCl. Online series-resistance compensation was used.Linear leak and capacitative currents were subtracted using a P/

2

5to P/

2

8 protocol from a holding voltage of

2

120 mV. Resultingtraces were periodically compared with those obtained with a P/4subtraction protocol from the holding voltage of

1

50 mV, andno consistent differences were noted. Records were low-pass fil-tered at 5–10 kHz.

A holding voltage of

2

100 mV was used except as noted in thetext. All experiments were carried out at 20.0

6

0.2

8

C, unless oth-erwise indicated, using a feedback temperature controller device.

Solutions

For patch-clamp recordings, we used chloride-containing solu-tions. The external solution contained (mM): 140 NaCl, 5 MgCl

2

,2 KCl, 10 HEPES (NaOH), pH 7.1. The internal solution con-tained (mM): 140 KCl, 2 MgCl

2

, 11 EGTA, 1 CaCl

2

, 10 HEPES (

N

-methylglucamine), pH 7.2. To reduce the slowly activating nativeoocyte chloride conductances when using the cut-open clamp,we perfused nominally chloride-free solutions containing (topand guard chambers, mM): 110 NaOH, 2 KOH, 2 Mg(OH)

2

, 5HEPES (MES), pH 7.1; (bottom chamber, mM): 110 KOH, 2Mg(OH)

2

, 1 Ca(OH)

2

, 10 EGTA, 5 HEPES (MES), pH 7.1.

Off-Line Analysis

Linear components of leak and capacitative currents were digi-tally subtracted. Macroscopic ionic and gating current recordswere analyzed further using Igor Pro (WaveMetrics) and custom-written software. Comparisons of the relative open probabilityversus voltage relationship among the wt and mutant channelswere based on the isochronal (between 0.5 and 1 ms post-pulse)amplitude of their tail currents after variable test pulses becausethis approach does not rely on assumptions about the linearity ofthe open-channel i(V) or the reversal potential. This type of mea-surement is termed a steady state voltage dependence of the

open probability [G(V)] relation in this paper. We fit the G(V)data with Boltzmann functions raised to the fourth power (seeZagotta

et al., 1994a), according to the equation,

where

z

is the apparent gating valence per channel subunit, V

1/2

is the apparent mid-point of the voltage-dependent transition ineach subunit, and

R

,

T

, and F have their usual thermodynamicmeanings. The time course of current activation was fit with theexponential

beginning with the time of the half-maximal current amplitude.As a measure of delay in current turn-on, tdelay (the time-axis in-tercept of the fitted exponential function) was found to be widelyvariable between patches for the same channel species. There-fore, the independently measured time-to-half-maximum wasused as an alternative indicator of the activation delay. Decayingexponential fits to the kinetics of tail currents were obtained us-ing the equation

Fits to experimental data and model simulations were performedusing a Levenberg-Marquardt nonlinear least squares optimiza-tion algorithm.

Model simulations were done using BigChannel software,courtesy of Toshinori Hoshi and Dorothy Perkins (HowardHughes Medical Institute, Stanford University, Stanford, CA). Inbrief, simulated macroscopic ionic and gating currents were cal-culated numerically using an Euler integration method, digitallyfiltered to match the corner frequency of an eight-pole Bessel fil-ter used in obtaining the corresponding data, and subsequentlyanalyzed in the manner identical to experimental traces. Modelparameters were allowed to vary slightly in fitting individual fami-lies of records. The goodness of fit was ultimately assessed by eye.Model parameters used in each simulation are given in the figurelegends.

r e s u l t s

The Sh5 Replica Mutation Alters Activation Gating of Shaker

A point mutation converting the first phenylalanine ofthe fifth transmembrane segment to isoleucine servedas a replica of the Sh5 mutation (Fig. 1 C). Familiesfrom patches expressing either wild-type or mutantF401I currents activate over a similar range of voltagesand deactivate completely at the relatively depolarizedtail potential of 265 mV (Fig. 1 A). Compared with thewt G(V) curve, F401I activation has a noticeably shal-lower voltage dependence (Fig. 1 B). The fourth powerBoltzmann fits to the G(V) curves yield values of z ofapproximately four elementary charges (e0) per wild-type subunit, which is similar to the estimated totalcharge displacement per channel of 12.5–14 e0, ob-tained from direct gating current measurements(Schoppa et al., 1992; Aggarwal and MacKinnon, 1996;Noceti et al., 1996; Seoh et al., 1996). By contrast, the

P0rel V( ) 1 1 e

V V1 2⁄–( )zF RT⁄–+( )⁄[ ]4,=

I t( ) Imax 1 et– tdelay+( ) τ⁄–( ) ,=

I t( ) Imax et– tdelay+( ) τ⁄( ) ,=

Dow


218 S5 and Voltage Gating in Shaker

value of z for F401I is decreased to 2.4 e0, which impliesthat the mutation either reduces the amount of chargedisplacement in the channel, or alters the coupling be-tween the charge-moving transitions and the channelopening.

If we consider a voltage-sensitive transition with asso-ciated charge displacement z in terms of transition-statetheory, the voltage dependence of the forward andbackward rates is determined by the charge movementbefore and after the transition state, respectively, andneed not be equal. We asked if the diminished voltagedependence of the F401I mutant is associated primarilywith forward or reverse transitions. A method to assessthe forward rates in relative isolation from the back-ward transitions is illustrated in Fig. 2. The currentsfrom wt and F401I channels activate with a sigmoidaldelay, reflecting a multistep opening process. As thetest potential is stepped to more positive values, chan-nel opening kinetics accelerate for both the wt andF401I families. With sufficiently depolarizing voltagesteps (i.e., more positive than 210 mV where the prob-ability of opening for both channels nears saturation),the reverse rates can be considered negligible and thekinetics of activation are almost entirely determined bythe forward rates. In this voltage range, the time courseof activation has a complex multiexponential behaviorbut, for a class of models commonly used to describe

Shaker gating, the slowest exponential component has atime constant that is the inverse of the slowest forwardrate (Zagotta et al., 1994a; Schoppa and Sigworth,1998c). We find that a good single-exponential fit canbe obtained to the latter phase of the trace beginningwith the time at which currents reach their half-maxi-mal amplitude (Fig. 2 A). The F401I mutant activatesmore rapidly and with less sigmoid delay than the wildtype. Time constants are voltage dependent, but the de-duced amount of charge moved for these forward tran-sitions is small and essentially unchanged between thewt and F401I channels (Fig. 2 B): 0.36 (see also Zagottaet al., 1994a) and 0.31 e0, respectively. F401I also pro-duces a consistent decrease in the time-to-half-maxi-mum current over the depolarized voltage ranges (Fig.2, C and D).

Whereas the voltage dependence of the forward ratesand, therefore, the amount of charge movement be-fore the transition state, appears unaffected by theF401I mutation, the voltage dependence of the closing(deactivation) transitions, reflecting the charge move-ment “after” the transition state, is very sensitive to thischange. The kinetics of deactivation were studied fromcurrents recorded during channel closing (tail cur-rents) at hyperpolarized potentials (negative to 260mV) after maximally activating prepulses (Fig. 3, A andB). Deactivation follows a nearly single-exponential

Figure 1. Steady state voltagedependence of the wild-typeShBD6-46 K channel and theF401I mutant. (A) Activationfamilies were elicited from theholding potential of 2100 mVwith steps in 10-mV incrementsbetween 280 to 150 mV, and tailcurrents were recorded at 265mV, as indicated schematicallyabove the traces. Patch-clamprecords were obtained in theinside-out excised configuration(Hamill et al., 1981), digitized at20 kHz, and low-pass filtered at 8(wt) or 3 (F401I) kHz. (B) Rela-tive conductance is plotted ver-sus voltage. Conductances werenormalized to the maximal valuein each family and the resultsfrom different patches were aver-aged to obtain the means andstandard errors shown (wt: n 5 5;F401I: n 5 6). Averaged G(V)curves were fit with the fourthpower of a Boltzmann function(see methods). The apparent

gating valence per subunit from these fits is reduced from 4.18 e0 for the wild type to 2.40 e0 for F401I. V1/2 parameters from the fits are256.3 mV for wt and 257.6 mV for F401I. (C) The amino acid sequence difference between wild-type Shaker and the neomorphic Sh5 al-lele is localized to a single substitution in the S5 region: phenylalanine is mutated to isoleucine at position 401 (Gautam and Tanouye,1990; Lichtinghagen et al., 1990). The sequence is shown using the standard single-letter amino acid code; dashes in the Sh5 sequence in-dicate amino acid identity with wt. Asterisks mark the residues that were mutated in this study.

Dow


219 Kanevsky and Aldrich

time course in both channels, with the time constantsfrom the fits displayed in Fig. 3 C against tail voltage. wttail currents are not simply slower compared withF401I; the difference is greatest at 260 mV, but dimin-ishes at very negative voltages and largely disappearsbelow 2160 mV (Fig. 3 C, inset). Kinetics of the tailcurrents in the wt are steeply potential dependent, withthe apparent charge associated with the backward tran-sitions, zr , of 1.2 e0, consistent with a previous report(Zagotta et al., 1994a). This number may be an overes-timate of the actual charge associated with the rate ofany one individual backward transition because of thetendency of channels to reopen in a voltage-sensitivefashion at all but the most negative tail voltages(Schoppa and Sigworth, 1998a). In contrast to the wt,the apparent valence derived from voltage dependenceof tail time constants in F401I is only 0.68 e0. Thus,

while the F401I mutant appears to move roughly thesame amount of charge during the forward transitionslate in the activation process, the mutation nearlyhalves the apparent charge movement associated withthe early backward transitions. Therefore, the domi-nant effect on the kinetics of the F401I channels’ re-turn to the closed state is the speeding of the tail cur-rents over all but the most negative voltages at whichthe determination of the tail time constant can becomelimited by the clamp response time. Our ability to ob-serve deactivation in the absence of superimposed fastN-type inactivation allows us to study reverse transitionsin relative isolation from other kinetic processes in thechannel. Our results lend direct support to the earlierproposal by Zagotta and Aldrich (1990b) that the Sh5

mutation affects the magnitude and voltage depen-dence of the reverse rate.

Figure 2. Kinetics and voltagedependence of forward transi-tions in the wild-type Shaker chan-nel and the F401I mutant. (A)The time course of activation wascompared in current traces ob-tained as in Fig. 1, with pulse volt-ages indicated on the right. Cur-rents from wild-type and mutantchannels were scaled to matchmaximal amplitudes at each ofthree voltage levels and fit with asingle exponential function be-ginning at the time point corre-sponding to half-maximal cur-rent amplitude (see methods).Fits are superimposed on thetraces as dotted curves. (B) Volt-age dependence of the activationtime constant. Values of the timeconstant, t, derived from single-exponential fits from a numberof patches (wt: n 5 7; F401I: n 55) were averaged and plottedagainst pulse voltage on semilog-arithmic axes. Error bars repre-sent the standard error of themean. The time constant versusvoltage relation was fit with adecaying exponential functionabove 210 mV, shown as solid(F401I) and dashed (wt) lines.The apparent charge associatedwith forward transitions late inactivation, zf, was calculated fromthe slope of the fitted line, andfound to be 0.36 e0 for the wt and0.31 e0 for F401I. (C) Differencein activation delay between wt

and mutant currents. Sets of representative traces recorded from two patches different from those in A are shown, with pulse voltage to theright of the traces. Current amplitudes for the 120-mV traces in each family were set to unity. Arrows indicate the time points at which cur-rent amplitudes are half-maximal. (D) As a way to quantify the absolute activation delay, the mean time to reach half-maximum is dis-played against pulse voltage for wt (n 5 8) and F401I (n 5 5) currents. The standard error of the mean is shown as error bars when it ex-ceeds the size of a symbol.

Dow



Comparison of S5 Phenylalanine Substitutions

F401 is one of five phenylalanines in the Shaker S5 se-quence (at positions 401, 402, 404, 410, and 416; seeFig. 1 C). To investigate whether other amino acid sub-stitutions in S5 have similar effects on activation gating,we conducted alanine mutagenesis of the four phenyla-lanines downstream (towards the carboxyl terminus) ofF401 as well as other S5 hydrophobic residues (leucinesat positions 396, 398, 399, 403, and 409, and serines at

positions 411 and 412), noting that it was an alaninesubstitution at F401 that resulted in the greatest effects(see below). Only F404A, F416A, L403A, S411A, andS412A gave rise to reliable ionic current expression.The results are shown in Fig. 4. The mutants’ steadystate activation voltage dependence shows few differ-ences from the wt other than a small 1–10-mV depolar-izing shift in most of the G(V) curves. The apparent va-lence of activation was not altered in any of the mu-

Figure 3. Alterations in the ki-netics and voltage dependenceof the closing transitions associ-ated with the F401I mutation. (Aand B) Deactivation familiesfrom wt (A) and F401I (B)patches were obtained with brief(8–10 ms) depolarizations to150 mV, followed by steps tovoltages between 260 and 2160mV in 10-mV increments, as indi-cated schematically above thetraces. (C) Relaxations of the tailcurrents were fitted with a singleexponential function (see meth-ods). Means and standard errorsof deactivation time constants, t,from the fits are plotted againsttail potential for patches contain-ing wt (n 5 8) and F401I (n 5 6)channels. Average t vs. voltagecurves were computed and fittedwith an exponential over the volt-age range below 260 mV. Fits areshown as solid (F401I) and bro-ken (wt) lines. The apparentcharge associated with reversetransitions, zr, was calculatedfrom the steepness of the expo-nential voltage dependence of t.zr for the wt is 1.30 e0 whereas zrfor F401I is 0.68 e0. (Inset)Shown are superimposed wt andmutant tail currents recorded at260 (left) and 2160 (right) mV.The currents were scaled tomatch their peak amplitudesduring the 150-mV prepulse.Note that the time scales are dif-ferent so as to enable compari-son of kinetic detail for the twospecies at both voltage extremes.

Dow



tants. For L403A, these findings confirm earlierobservations on channels with intact inactivation (Lo-pez et al., 1991; McCormack et al., 1991) that this resi-due, the fifth leucine in a putative heptad motif span-ning the S4–S5 regions, plays at most a minor role involtage-dependent gating. Results from the two serinesubstitutions imply that removal of the hydroxyl groupsfrom the respective side chains does not alter the acti-vation process.

Because the F404 residue is the least well conservedof S5 phenylalanines among the family of potassiumchannels, with alanine occurring at the equivalent sitein, for example, Kv2.1, fShal and fShab, we were notsurprised that the F404A substitution did not signifi-cantly alter activation or deactivation kinetics. In con-trast, the position equivalent to F416 in Shaker channelsonly contains aromatic amino acids among voltage-gated potassium channels. We noted small but consis-

Figure 4. Alanine replace-ments in the S5 region and thesteady state activation in Shaker.On the left are representativefamilies recorded from inside-out patches containing the mu-tant channels indicated. Teststeps were given in the voltageranges shown under the traces in10-mV increments. Tail voltagewas 265 mV for all families. Onthe right, corresponding G(V)relations are shown for each mu-tant; curves fitted to the wt G(V)are included for comparison(shown as broken lines). For theF416A mutant, the mean of 12experiments with its standard er-ror is displayed. For the othermutants, combined data fromthree (F404A), two (L403A), two(S411A), and three (S412A)patches are shown. Thin linesthrough the data represent fits ofthe Boltzmann function raised tothe fourth power (see meth-ods), yielding the following esti-mates of apparent gating valenceand midpoint of voltage-depen-dent transition: z 5 3.81 e0 andV1/2 5 255.5 mV (F404A); z 53.97 e0 and V1/2 5 251.1 mV(F416A); z 5 3.86 e0 and V1/2 5249.1 mV (L403A); z 5 4.16 e0and V1/2 5 250.4 mV (S411A);z 5 4.10 e0 and V1/2 5 249.6 mV(S412A).

Dow



tent differences between the F416A mutant and thewild type. F416A currents have a more pronounced sig-moid delay in activation and more rapid deactivationkinetics. In summary, neither of the two downstreamS5 phenylalanine-to-alanine mutations that producedfunctional channel expression, and none of the leucineand serine substitutions, influenced voltage-dependentgating to the degree evident for the F401 mutations.

Correlating Effects of F401 Mutations with Side Chain Properties

Because of the striking effects of the F401I substitution,we substituted other amino acids for the phenylalanineat 401 to investigate the role of side chain structure on

gating. We introduced individually three progressivelysmaller aliphatic amino acids leucine, valine, and ala-nine at that site. Fig. 5 A shows representative currentfamilies from these channels on different time scales tobring out the distinctive kinetic features of each chan-nel type. In Fig. 5 B, the range of change induced bythese mutations in the steady state voltage dependenceof the relative open probability is shown. For compari-son, previously described fits of a fourth power of theBoltzmann function to the wt and F401I data are alsoincluded. The F401V G(V) relationship is shallowerthan that of the wt, and similar in slope (zapp 5 2.5) tothe F401I mutant. However, the V1/2 in F401V is posi-tively shifted by z5 mV compared with F401I. Steadystate activation of the F401A mutant is the shallowest

Figure 5. Aliphatic side-chain substitutions at position 401 affect the steady state voltage dependence of activation. (A) Families of macro-scopic currents from patches expressing (left to right) F401L, F401V, and F401A channels were recorded as in Fig. 1. The voltage ranges forthe three families were as follows (step increment in parentheses): 285 to 125 mV (10 mV) for F401L, 280 to 160 mV (10 mV) for F401V,and 280 to 1160 mV (20 mV) for F401A. Note the faster time scale for the F401A family used to resolve its kinetic features. (B) Relativeconductance versus voltage relationships for the three mutants are shown. For comparison, fits to G(V) curves for the wt and F401I chan-nels are reproduced from Fig. 1. F401L and F401V data are plotted as means 6 SEM of eight and nine families, respectively. Averaged G(V)data were fitted with the fourth power of a Boltzmann function, as previously described, giving apparent z of 4.25 and 2.52 e0, and V1/2 of269.7 and 253.1 mV for F401L and F401V, respectively. For the F401A experiment shown (representative of 11 patches), G(V) was mea-sured either as the chord conductance assuming the reversal potential of 280 mV (denoted “pulse”), or as the isochronal tail current am-plitude (denoted “tail”). The characteristic failure of the steady state conductance to reach a maximum within the attainable voltage rangeprevented meaningful normalization of the F401A G(V).

Dow



(zapp , 0.5 e0); in fact, the G(V) relationship fails toreach saturation at voltages in excess of 1150 mV infive patches, and is therefore displayed on a dimension-less y axis. Unexpectedly, introduction at position 401of a leucine, an amino acid chemically most similar tothe isoleucine, carried nearly opposite consequencescompared with the Sh5 replica mutation F401I. TheF401L mutant has a G(V) relation as steep as that ofthe wt (zapp 5 4.25) but with the midpoint of the activat-ing transition shifted negatively (V1/2 5 269.7 mV), theonly mutant in this study to do so.

A look at the activation time course on the expandedtime scale in Fig. 6 A underscores that all channelsbearing aliphatic substitutions for phenylalanine at po-sition 401 activate more rapidly than the wt for a givenvoltage. Whereas F401L channels are least differentfrom the wt, 401 isoleucine and valine channels aresimilar to each other and have faster kinetics than leu-cine channels; alanine channels are the fastest by farover all voltages. Quantitatively, Figs. 2 B and 6 B showthat the voltage dependence of activation time con-stants measured late in the activation process is simi-larly weak no matter which of the five residues is at po-sition 401, with the apparent valence associated withthe forward transitions, zf, ranging from 0.32 to 0.42 e0.The absolute values of the time constants, t, are compa-rable except for F401A, in which they are significantlydiminished. Regardless of whether F401 mutations di-minish steady state voltage dependence of the currents,the voltage dependence of the forward rate, zf, remainsin the wt range.

We expected that, as in the case of Sh5 (F401I), theother aliphatic substitutions would preferentially per-turb deactivating transitions. In Fig. 7 A, time constantsfrom fits to tail current relaxations are plotted for theleucine, valine, and alanine mutations. For compari-son, fits to the voltage dependence of the deactivationtime constant, t, from wt and F401I are also included.The tail time constants of F401L currents are slowerthan those of the wt but have similarly steep voltage de-pendence. Deactivation kinetics of F401V (zr 5 0.74 e0)are nearly the same as those of F401I, and F401A deac-tivation appears to be nearly voltage independent tothe best of our ability to analyze its very rapid kinetics.This finding provides a ready explanation for the veryshallow G(V) of F401A. In wild-type Shaker, a greaterproportion of the total gating charge movement occursafter the transition state (Zagotta et al., 1994a), and itsloss will be reflected in the diminished voltage depen-dence of the steady state gating parameters. On theother hand, the notable decrease in the backward ratesand modest increase in the forward rates seen in F401Limply that some of the gating equilibria for this chan-nel are biased toward the open state compared with thewt, which is consistent with the finding of a negatively

shifted G(V) relation. Fig. 7 B depicts families of cur-rent traces from the two mutants that differ the most intheir tail kinetics. Currents from the F401L and F401Afamilies are shown on the same time scale to illustratethat there is more than an order of magnitude differ-ence in the tendency of these channels, once activated,to remain in (or near to) the open state long after theend of a depolarizing voltage pulse. Note that even at

Figure 6. Effects of replacing phenylalanine 401 with aliphaticresidues on the activation kinetics and voltage dependence. (A)Representative current traces from wt, F401I, F401V, F401A, andF401L patches were superimposed and scaled to match at theirpeaks. Pulse voltages are indicated on the left. (B) A semilogarith-mic plot of the activation time constants (mean 6 SEM), deter-mined from fitting an exponential function to the activation timecourse (see methods), is shown for the F401L (n 5 12), F401V(n 5 12), and F401A (n 5 11) mutants. From the steepness of theexponential fit to the t versus voltage relation, apparent valence as-sociated with the forward transitions, zf, was found to be 0.41 e0 forF401L, 0.42 e0 for F401V, and 0.32 e0 for F401A. Fits are shown asthin solid and broken lines; for comparison, a fit to the wt activa-tion t is reproduced from Fig. 2 as a thick broken line.

Dow



fairly depolarized tail potentials F401A channels relaxto a new steady state level on a very rapid time scale.

Because several F401 mutants accelerate deactivationof macroscopic ionic currents, we hypothesized thefaster rates for leaving the open state by deactivationshould decrease the mean time spent in the open state.F401I has a unitary conductance similar to the wt butbriefer open times (mean 2 vs. 4 ms in the wt), consis-tent with its faster deactivation kinetics. Single F401Achannels show extremely brief, incompletely resolvedopenings that are seen promptly at the start of the testpulse (data not shown). Bandwidth limitations of therecording equipment did not allow us to pursue quanti-tative analysis of these channels, but qualitatively theirbehavior supports the hypothesis that isoleucine and

especially alanine mutants accelerate transitions fromthe open state that reverse the activation sequence.

Gating Charge Movement in the F401 Mutants

One possible explanation for the reduction in the ap-parent valence of channel opening in Sh5 and relatedF401 mutants is an alteration in the coupling amongcharge-moving transitions. This could take the form ofa transition (or transitions) that the channel mustundergo during opening that has a voltage midpointshifted far in the positive direction relative to the wt,such as has been proposed for several S4 and S4–S5linker mutations (Schoppa et al., 1992; Perozo et al.,1994; Schoppa and Sigworth, 1998b; Smith-Maxwell et

Figure 7. Substitutions at position 401 affect deactivation. (A) Deactivation time constants for the F401L, F401V, and F401A mutantchannels were obtained from single exponential fits to currents during channel closing. The time constants, t, are plotted as a function oftail potential and fitted with an exponential. For F401L and F401V, results are shown as the mean 6 SEM of 8 and 10 experiments, respec-tively. Due to the bandwidth limitations on our ability to fully resolve closing kinetics in the F401A mutant, results from the two bestpatches are shown at selected voltages. For comparison, fits to wild-type and F401I deactivation curves from Fig. 3 are also included. Thevoltage dependence of deactivation t yielded apparent zr values of 1.15 e0 for F401L, 0.74 e0 for F401V, and ,0.5 e0 for F401A. (B) Differ-ences of deactivation kinetics are illustrated with the tail families of the F401A and F401L mutants, shown on the same time scale. TheF401A family (left) was recorded under standard ionic conditions (see methods) from an inside-out patch, sampled at 100 kHz and low-pass filtered at 9.5 kHz. Tail voltage ranged between 220 and 2170 mV in 10-mV increments. The F401L currents (right) were recordedfrom an outside-out patch with symmetrical 140 mM K1 as the permeant cation. Increasing external K1 from 2 to 140 mM does not signif-icantly affect tail kinetics of this channel (data not shown, but see Zagotta et al. (1994a). Tail voltage was stepped to between 280 and2180 mV in 10-mV increments. For both families, a 10-ms pulse to 150 mV preceded the steps to the tail potentials.

Dow



al., 1998a,b). Gating charge measurements from such amutant will reveal a separation of charge componentsalong the voltage axis, giving rise to an inflection or afrank shoulder in the total steady state charge displace-ment versus voltage [Q(V)] relationship (Ledwell andAldrich, 1999). Therefore, we studied Q(V) relation-ships for the wt and two mutants with very different ap-parent gating valences.

Families of gating currents from the wf and thewfF401L and wfF401A mutants are shown in Fig. 8. Thegating currents are shown superimposed and staggeredto facilitate comparison of the development of kineticfeatures with changes in voltage. The wf ON gating cur-

rents (IgON) have a rising phase, appear at negative volt-ages, and show a slow decaying component in the volt-age range where channels open. This latter componentaccelerates with further depolarizations. The overalltime course of the IgON decay becomes faster in the or-der wf, wfF401L, and wfF401A, consistent with thefaster time course of ionic current activation observedin the corresponding conducting species, although theF401A gating currents are accelerated to a lesser extentthan the corresponding ionic currents. A prominentrising phase and slow decay appear in the wf OFF cur-rents (IgOFF) at the voltages where there is a slow phaseof the IgON decay, consistent with published observa-

Figure 8. Gating charge move-ment in the wild type and chan-nels containing F401L andF401A mutations. Gating cur-rents were recorded from non-conducting channels containingthe indicated mutations using acut-open oocyte voltage clamp,digitized at 20 kHz (42 kHz forwfF401A) and low-pass filtered at10 kHz. Traces elicited by steps tothe voltages indicated on the left,beginning from and returning tothe holding potential of 2100mV, are shown staggered to facil-itate the kinetic comparison.Chloride-free standard solutionswere used (see methods).

Dow



tions from the cut-open oocyte clamp. (Perozo et al.,1992; Bezanilla et al., 1994; Stefani et al., 1994). Themost notable change introduced by the wfF401L muta-tion is the profound slowing of the OFF gating chargereturn at voltages where the channel opens. In con-trast, wfF401A all but abolishes the slowing of IgOFF. Thecorrelation between faster OFF gating charge returnand faster deactivation of ionic currents is more consis-tent with the slowing of OFF charge due to a stabiliza-tion of the open state rather than to channels enteringa C-type inactivated state (see Chen et al., 1997).

Steady state Q(V) relationships for the three chan-nels were computed from the integral of the IgOFF aftera test pulse. The integral of the IgON, while not shown,agreed closely. Fig. 9 A shows that the wf Q(V) curvehas a characteristically shallow base and steeper upperportion (see also Stefani et al., 1994). The wfF401LQ(V) relation is negatively shifted relative to the wf,similar to the relationship between the G(V) of F401Land the wt. For this mutant, channel opening seems tofollow closely the displacement of the voltage-sensingcharges. wfF401A has detectable charge movement atmore negative voltages than the wf, but its Q(V) slopeis somewhat shallower. However, we did not detect anyobvious inflections reflecting the movement of an addi-tional component of the gating charge in the wfF401AQ(V) curve at voltages above 0 mV and extending evenbeyond 1100 mV. Therefore, we conclude that the re-duction in the apparent voltage dependence of activa-tion is not the result of altered coupling of activationpathway transitions carrying significant amounts ofcharge movement. It is likely that this mutation affectsan activating transition that moves only a small amountof charge and thus would not perturb the overall shapeof the Q(V) curve.

We can compare the kinetics of the ON gating cur-rents by fitting their decay phase with an exponentialtime constant. IgON is not well described by a single ex-ponential at all voltages but, above 220 mV, these fitsare useful as a way of assessing the overall kinetics offorward transitions in the channel. When these timeconstants are plotted against voltage for the wf and thewfF401L and wfF401A mutants (Fig. 9 B), the rates offorward transitions are the fastest for wfF401A, fol-lowed by wfF401L and the wf channel. This mirrors therelationship among the activation time constants ofionic currents from the corresponding conductingchannels. The voltage dependence for the movementof the charge “before” the transition states (zf) is con-served for the three channels, ranging between 0.57and 0.63 e0. These values are similar to those estimatedfrom ionic current measurements (Fig. 6) and placeimportant constraints on kinetic modeling of the earlysteps in channel activation.

The changes in both the ionic and gating currents re-

veal alterations in the voltage dependence and magni-tude of the reverse rates out of the open state with theF401 residue replacements. The time course and volt-age dependence of the forward activation rates aremuch less affected. These results suggest that the F401mutations alter the energetic stability of the openstate relative to closed states. Alterations in gating by

Figure 9. F401 substitutions affect the steady state charge trans-fer and kinetics of the forward transitions. (A) Relative charge ver-sus voltage relations were computed for the wf, wfF401L, andwfF401A channels by integrating the gating current during the2100-mV post-pulse (IgOFF). Normalized Q(V) relations from anumber of experiments were averaged and plotted as a function oftest voltage as mean 6 SEM (wf: n 5 17; wfF401L: n 5 10;wfF401A: n 5 12). The fitted lines through the data are the Boltz-mann function predictions for channels with independent andidentical subunit transitions with the following values of voltagemidpoint (V1/2) and associated charge displacement (z): 243 mVand 3.73 e0, 261.6 mV and 4.14 e0, and 252.3 mV and 2.76 e0, forwf, wfF401L, and wfF401A, respectively. (B) The decay phase ofthe IgON was fitted with a single exponential and the time constantsare plotted against pulse voltage for the wf (n 5 11), wfF401L (n 58), and wfF401A (n 5 8) channels. From the exponential fits ofthe voltage dependence, beginning at 210, 225, and 220 mV, theestimate zf of the charge associated with the forward transitions is0.57, 0.61, and 0.63 e0, for wf, wfF401L, and wfF401A, respectively.

Dow



changes of noncharged residues in the S4 segmenthave been interpreted in terms of a change in the ener-getics of a final cooperative opening step (Smith-Max-well et al., 1998b; Ledwell and Aldrich, 1999). An alter-ation in the cooperative stabilization of the open statecould likewise lead to the observed behavior of theF401 mutations. In the following section, we test thishypothesis more directly using previously developedexperimental protocols to elucidate the cooperative in-teractions between channel subunits (Zagotta et al.,1994a; Smith-Maxwell et al., 1998b; Ledwell and Ald-rich, 1999).

Behavior of Ionic Currents Indicates Nonindependence of Subunits

Shaker ionic currents activate upon depolarization witha delay, giving rise to a sigmoidal time course. This sig-moidicity arises from the multi-step nature of the acti-vation process. Voltage dependence in the sigmoidicityof ionic currents is a diagnostic feature of deviation

from subunit independence in activation (Zagotta etal., 1994a; Smith-Maxwell et al., 1998a,b). Sigmoidicityis preserved in the mutant channels F401L and F401A(Fig. 10 A), although their overall kinetics and the ab-solute amount of delay vary. For a noncooperativemodel of channel activation that postulates n indepen-dent first-order voltage-dependent processes (Hodgkinand Huxley, 1952; Cole and Moore, 1960), it can beshown that sigmoidicity will be the same for a given n atall test voltages, and the time- and amplitude-scaledtraces will superimpose. Over the voltage ranges shown,wt and F401L channels clearly display deviations fromthe independent scheme. Sigmoidicity is greater athigher than at lower voltages, but appears to reach asaturating value when the voltage is in the range ofmaximal steady state activation (reached near or below0 mV for both of these channels). Zagotta et al. (1994a)used this observation in wt Shaker to argue for a form ofcooperativity that acts to slow the first closing transitionfrom the open state causing the current waveform atlower voltages to be close to a monoexponential func-

Figure 10. Mutations of F401 and cooperative gating: sigmoidicity. (A) Activation families of wt currents taken at voltages between 265and 25 mV, F401L (280 to 130 mV), and F401A (-70 mV to 190 mV) in 10-mV increments. (B) Data transformed to normalize the cur-rent amplitude and overall speed of activation for the voltages shown. Sigmoidicity is defined as the amount of initial delay relative to theoverall rate of activation. Its voltage dependence can be visualized by following transformation (Zagotta et al., 1994a). The currents fromdifferent voltage steps are first scaled to match at their peaks. Then the time derivative of the current waveform is determined at the timewhen current amplitude is half maximal. In general, it is at that point that the slope is the steepest. All the records in the family are thenexpanded or compressed along the time axis so that their derivatives at half-maximum match. This transformation results in parallelcurves whose relative displacement along the x axis is the measure of sigmoidicity. The noisier traces correspond to lower voltages. Datawere filtered at between 8 and 10 kHz and sampled every 20–50 ms.

Dow



tion because it is limited by the slow final step (see alsoSmith-Maxwell et al., 1998a,b; Ledwell and Aldrich,1999). From the F401L records, it is apparent that thesmaller amount of sigmoidicity at the lower voltages isat least as pronounced as in the wt (note that the steadystate activation in F401L is shifted negatively relative tothe wt by 15–20 mV). On the other hand, over the volt-age range between 240 and 190 mV, channels with analanine substitution at F401 do not display the clear in-crease in the amount of sigmoidicity with higher depo-larizations seen with the wt and F401L. The F401A mu-tant acts as though there is no rate-limiting transitionpresent at the lower voltages, implying that the leavingrate from the open state is not slowed relative to thepredictions of a mechanism with independently gatingsubunits.

We sought to confirm further that the mutations atF401 alter the deviation from independence seen in wtchannels upon entering the open state. As originally de-

scribed by Cole and Moore (1960), multistep activationgating results in greater delay in current turn-on whenthe channels are subjected to progressively more nega-tive voltages before the test pulse. The greater delay oc-curs as the equilibrium distribution of channels amongclosed states shifts in favor of the states most distantfrom the open state. For an independent gating scheme,a family of current waveforms corresponding to differ-ent prepulse voltages becomes superimposable simplyby a translation along the time axis that allows for theamount of delay lost or gained. These transformationsof wt, F401L, and F401A currents are shown in Fig. 11.In wt, over the voltage range between 2140 and 270mV, there is almost no current activation during theprepulse, and the corresponding traces at 0 mV are par-allel and superimposable by a shift along the time axis.However, at the prepulse voltages of 250 and 240 mV,wt channels open with nonnegligible probability andthe 0-mV current waveforms cannot be superimposed

Figure 11. Mutations of F401 and cooperative gating: Cole-Moore analysis. (A) Families of currents from the wt, F401L, and F401A chan-nels were obtained as follows: pulses to 0 mV were preceded by 1-s-long prepulses to voltages between 2140 and 240 mV (wt, left), 2170and 250 mV (F401L, center), or between 2130 and 220 mV (F401A, right). A few of the most depolarizing prepulses are in the activationrange of the channels, as evident from the steady current recorded during the prepulse. The most negative prepulses induce greatest delayat the onset of activation. (B) The voltage families above were scaled in amplitude to match at the peak to compensate for steady stateC-type inactivation incurred at the more depolarized prepulses. Scaled traces were then shifted along the time axis to obtain the best su-perposition. Note that, whereas in wt and F401L (left, center) traces in which the prepulse opens channels with detectable probability arenot superimposable on the rest of the family, in F401A such behavior is not seen. At the onset of current during the main pulse, gating cur-rents can be seen as bumps in the F401A family.

Dow



on the rest of the family by this procedure (Fig. 11 B,left; Zagotta et al., 1994a). This property is also observedwith F401L channels (Fig. 11, middle). In contrast, inF401A, the Cole-Moore shift is present with complete su-perimposability over the prepulse voltage range of 2130to 220 mV (Fig. 11, right). During the more depolar-ized prepulses in this family, F401A channels are signifi-cantly activated, but their entry into the open state con-fers no new kinetic features to the 0-mV traces thatwould suggest nonindependent subunit behavior.

Gating Charge Movement Reveals Stability of the Open Conformation

Kinetics of the return of gating charge (IgOFF) uponstepping from a positive to a negative voltage provideimportant information about the voltage-dependent re-verse transitions (those leading away from the openstate). In potassium channels, their dependence on theduration and amplitude of the preceding depolariza-tion has been extensively studied (Bezanilla et al., 1991;Perozo et al., 1992, 1993; Bezanilla et al., 1994; McCor-mack et al., 1994; Zagotta et al., 1994b; Hurst et al.,1997; Schoppa and Sigworth, 1998c). In Shaker, thetime course of gating charge return at the end of a volt-age pulse depends strongly on the voltage and durationof the depolarization. After depolarizing steps to nega-tive voltages that would result in a low probability ofchannel opening, gating charge return follows a rapidnearly exponential time course, reflecting redistribu-tion of channels among closed states. At more positivetest voltages, increased pulse durations give rise to IgOFF

that is characterized by a rising phase and a slowly de-caying time course. This observation fits with the ideathat channels entering the open state leave it onlyslowly, “trapping” charged domains in the activatedconformation.

Fig. 12 A contrasts the effect of test-pulse duration atthree voltage levels on IgOFF in wf, wfF401L, andwfF401A channels. With pulses to 250 mV, charge re-turn upon repolarization to 2100 mV remains rapid inwf and wfF401A for pulse durations between 1 and 57ms (wf) and 41 ms (wfF401A). F401L channels are sig-nificantly open at this voltage, however, and the OFFgating currents in wfF401L accordingly display progres-sively diminished amplitude and a prolonged decliningphase as pulse length exceeds z3 ms. A pulse ampli-tude of 230 mV (Fig. 12 A, middle left) marks a transi-tion zone for the kinetics of the wf IgOFF. Pulses of a fewmilliseconds duration do not impede subsequent rapidcharge return, those longer than z10 ms give rise toOFF currents with complex time courses in which atleast three kinetic components can be recognized, andthose .25 ms produce a rising phase and exponentialdecay. wfF401L OFF gating currents with all but theshortest 230 mV pulses are notable for the greatly

slowed charge return that is incomplete after up to 30ms at 2100 mV (Fig. 12 A, middle center). The familiesof wf and wfF401L channels show progression of thesame trends when the pulse amplitude is 0 mV. In fact,IgOFF becomes nearly “immobilized” in wfF401L, dis-playing protracted decay after pulses lasting longerthan 3–4 ms.

wfF401A gating currents are unlike those of the othertwo species. With pulses to 0 and 1100 mV, the latter ofwhich are sufficient to activate many F401A channels,the time course of the OFF currents is rapid and unaf-fected by pulse length (Fig. 12 A, right). The smallsteady state outward current seen at 1100 mV is anionic current contaminant, likely of native Xenopus oo-cyte origin because its appearance at that voltage is vari-able among different cells and does not depend on thelevel of channel expression. Its tail current also ac-counts for a very small slow phase on the OFF gatingcurrents after longer pulse durations. The results forthe wf (left), wfF401L (center), and wfF401A (right)channels are summarized in Fig. 12 B, which plots thetime constants from exponential fits to the decayingphase of IgOFF as a function of the length of pulses atthe different pulse voltages. The time course of theOFF gating currents has a complex waveform, andthese single-exponential fits are not meant to implythat there is an underlying first-order kinetic process;rather, they provide a ready means to document a tran-sition from a predominantly fast to a slow process. ForwfF401A, they additionally illustrate that as the proba-bility of channel opening is changing over a voltagerange of 150 mV, the kinetics of charge return at 2100mV are barely altered.

Kinetic Mechanisms for Nonindependent Gating

Ionic and gating current results described in the pre-ceding section imply that mutating phenylalanine toalanine at position 401 drastically diminishes the coop-erative stabilization of the open state characteristic ofthe wild type, whereas the leucine mutation augmentsit. In the following, we investigate this hypothesis quan-titatively, using kinetic principles and models previ-ously developed for Shaker gating.

Several general features of Shaker gating have beenestablished by diverse means in different laboratories(e.g., Bezanilla and Stefani, 1994; Sigg et al., 1994; Sig-worth, 1994; Zagotta et al., 1994a,b; Aggarwal andMacKinnon, 1996; Baker et al., 1998; Schoppa andSigworth, 1998a,c). These include: (a) activation is amulti-step process involving more than a single transi-tion per subunit, (b) activation requires the transloca-tion of z14 elementary charges across the electricalfield of the membrane, (c) charge movement is spreadover a number of transitions and its quantity is not thesame for all transitions, (d) for most transitions toward

Dow



Figure 12. Mutations of F401 and cooperative gating: return of the gating charge before and after channel opening. (A) Families of gat-ing currents recorded in the cut-open oocyte clamp are shown. In all families, steps were given from the holding voltage of 2100 mV. Forthe wf (left) and wfF401L (center) channels, pulse voltages were (top to bottom) 250, 230, and 0 mV. For wfF401A (right), pulse voltageswere 250, 0, and 1100 mV. Voltage electrode traces are included above each family to indicate the voltage protocol used. In each family,gating currents resulting from different durations of the pulse are overlaid such that the changes with the pulse length can be observed asthe envelope of OFF gating currents. Data were filtered at 8–10 kHz. (B) Voltage and pulse-length dependence of charge return kinetics.For the wf (left), wfF401L (center), and wfF401A (right) channels, records obtained as in Fig. 8 were fitted with an exponential function todescribe the time course of IgOFF decay. The time constants from the fits for the different levels of pulse voltage are plotted as a function ofduration of the pulse.

Dow



the open state, more charge moves after the transitionstate than before it, (e) some of the transitions late inactivation carry a greater amount of charge than theearlier transitions, (f) gating among closed states canbe approximated by independent action of the sub-units, (g) open channels can close to states that werenot necessarily traversed during the activation process,(h) transitions that involve the open state disobey thepredictions of subunit independence.

One relatively simple formalism that has been putforth to account for these features of gating is thescheme of Zagotta et al. (1994b), which will be referredas the ZHA model, which explicitly introduces a coop-erative factor u by which the first closing transition rateis divided. Otherwise, the activation pathway in thismodel is an independent process involving four gatingsubunits, each undergoing two sequential charge-mov-ing transitions. The ZHA model is shown in an abbrevi-ated form in Fig. 13 A, emphasizing the fourfold sym-metry. We used the parameters of the original ZHAmodel (Zagotta et al., 1994b) to simulate the G(V)(Fig. 13 B), ionic currents for steps to 150 mV followedby a tail voltage of 265 mV (Fig. 13 C), steady statecharge-voltage relation [Q(V)] (Fig. 13 D), and the de-pendence of the time course of the OFF gating cur-rents on the duration of 230-mV pulses (Fig. 13 E). Ineach case, the variable parameter was the factor u,which was either 1, 9.4, or 50. These were chosen togive the best overall approximations of the F401A, wt,and F401L channels’ behavior, respectively. For F401A,the cooperativity factor (u) was set to a value of 1, theequivalent of complete subunit independence. Themodel succeeds in correctly describing the order of rel-ative steepness of the G(V) curves, of the deactivationkinetics, and of the duration dependence of IgOFF. How-ever, the extremely shallow voltage dependence of F401Achannel opening (Fig. 5) could not be reproduced bythe model, even with the introduction of a modestamount of negative cooperativity (u 5 0.4; Fig. 13 B).

The ZHA model provided a convenient starting pointfor arriving at kinetic descriptions of the wt and mutantcurrents. Manipulating only the degree of cooperativeslowing of the first closing transition with changes tothe factor u does surprisingly well in describing the ba-sic properties of the F401 mutant channels. However,the model proves inadequate for the quantitative agree-ment with the macroscopic ionic and gating currentdata. Proper fits to the kinetics of channel opening,steady state G(V) relations, ON gating currents, andthe steady state Q(V) curves all require manipulationof additional parameters of the ZHA model and in anumber of cases were unattainable without altering thefourfold symmetric structure that had made it so con-ceptually attractive. Therefore, we broadened our con-sideration of candidate models to include ones where a

separate concerted transition (or transitions) connectsthe four parallel and independent activation pathways(one per subunit) to the open state. Precedents for thismechanism can be found in earlier models for potas-sium channels (Koren et al., 1990; Zagotta and Aldrich,1990a; Schoppa and Sigworth, 1998c; Ledwell and Ald-rich, 1999). This class of models provides the concep-tual advantage of preserving the symmetric nature ofthe activation pathway while introducing only a few ad-ditional free parameters compared with the model ofZagotta et al. (1994b) and Smith-Maxwell et al. (1998b).

A detailed kinetic model of this class has been pro-posed for Shaker and a mutant channel (V2) (Schoppaand Sigworth, 1998c), which argues for the necessity toinclude a third charge-translocating step per subunit aswell as two sequential concerted transitions precedingchannel opening (a so-called 3129 scheme). We optedfor the simpler (2119) model because of the limitedexperimental means to constrain a more elaboratescheme for all three channel species in this study. Ourgoal is to provide a robust description of the main as-pects of the channels’ gating while minimizing thenumber of transitions that differ among the models forthe wt, F401L, and F401A channels. Our ability to do sosupports the hypothesis that F401 mutations do not dis-rupt the wt gating mechanism in a global sense, butonly target specific aspects of it. Our model providesreasonable fits to the wt and mutant channels, with ma-jor differences among the three species primarily lim-ited to the first closing transition, as suggested by ourdata and the predictions of the original ZHA model(Fig. 13). Fig. 14 shows the connectivity of the modeland illustrates that despite the large number (17) of ki-netic states, only 12 free parameters are needed to con-strain the mechanism up to and including the con-certed Closed ↔ Open transition, compared with 9 forthe ZHA model and 20 for the 3129 model of Schoppaand Sigworth (1998c). These are the zero-voltage rate(k0) and associated valence (zk) of the two forward ratesa and g and the two reverse rates b and d for each ofthe four subunits and of the forward rate k and the re-verse rate l for the concerted transition. The rates areassumed to be instantaneous exponential functions ofvoltage, according to:

The total charge displacement for each channel is con-strained to be z14 e0 (14.47 e0 for the wt, 14.37 e0 forF401A, and 14.07 e0 for F401L), in agreement with pre-viously published measurements in wt (Aggarwal andMacKinnon, 1996; Noceti et al., 1996; Seoh et al., 1996;Schoppa et al., 1992).

Kinetic transitions that follow channel opening at de-polarized potentials are to states that are not obligato-

k k0ezkFV RT⁄ .=

Dow



rily traversed during the activation process. These havebeen characterized using single-channel recordings ofwt Shaker (Hoshi et al., 1994; Schoppa and Sigworth,1998a). The predominant fast component of closed du-rations seen at depolarized voltage can be accountedfor by including a state (shown as Cf… in Fig. 14, top)that channels can enter after opening by a nearly volt-age-independent transition. For our modeling of wt, weused the rate parameters for this transition given inZagotta et al. (1994b). Because the open single F401Achannels tend to have very brief flickery openings thatare incompletely resolved in our records, we were un-able to conduct quantitative analysis of the open andclosed durations for this mutant and, therefore, lackthe basis for detailed comparison with the wt data. Wechose to assign the wt values for the O ↔ Cf… transi-

tion to the same values in all three channel speciesrather than let them vary among the wt, F401L, andF401A.

Testing Model Predictions in wt and Mutant Shaker Channels

Fig. 15 A shows the fits of the model shown in Fig. 14for the wf, wfF401L, and wfF401A channel’s steady statecharge vs. voltage curves. Equilibrium constants for thetwo charge-moving transitions in each subunit and forthe concerted step were optimized to obtain the de-sired steepness and position along the voltage axis. Thetotal charge displacement for a given transition is thesum of the charges that move before and after the tran-sition state or, equivalently, that are associated with theforward and backward rates of that transition. The

Figure 13. Predictions of theZHA model at different levels ofcooperativity. (A) Connectivity ofthe ZHA model is shown in thecollapsed format (for expandedversion showing all nondegener-ate states, see Fig. 7 of Zagotta etal., 1994b). Note that the rate ofthe first backward transition forleaving the open state is slowedby the cooperative factor u. Simu-lations were performed on theversions of the ZHA model thatvaried the value of the factor uas indicated. Simulated currentswere then analyzed in the man-ner identical to the experimentaldata. Shown are the simulationsfor the normalized steady stateG(V) relationship (B), ionic tailcurrents at 265 mV (C), normal-ized steady state Q(V) relation(D), and pulse-length depen-dence of the OFF gating currentat 230 mV (E).

Dow



three channels differ the most in their equilibria forthe concerted opening transition. The zero-voltageequilibrium constants for this step are 55, 125, and 0.5for the wf, wfF401L, and wfF401A, respectively. Themarked decrease in wfF401A provides part of the expla-nation for the shallowness of the slope of its Q(V)curve, even though this transition carries only z1/16of the total charge displacement in the channel. To de-scribe accurately the relatively shallow lower portion ofeach of the curves, it is necessary to make the chargedisplacement associated with the second of the two se-quential subunit transitions (zgd) greater than that ofthe first (zab) (Bezanilla et al., 1994). For all channel

species, the quantity zgd is 2.0–2.1 e0. The first transitioncarries the charge of 1.35 e0. Models for wfF401L andwfF401A channels make the zero-voltage equilibriumconstant for the first transition,

approximately twice as great, and that for the secondtransition,

nearly three times as great as those of the wf model inorder to account for the more negative voltage rangeover which the initial gating charge movement occursin the mutants.

In Fig. 15 B, model fits are shown superimposed onfamilies of gating currents. The fits are a good descrip-tion of the time course of IgON obtained at depolarizedvoltages and of the IgOFF. However, the models predicttoo rapid a rise and decay of the ON gating currents inthe activating voltage range (approximately 280 to240 mV for all three channels). The data suggest thepresence of a rising phase in the IgON records even atthese negative voltages. We were able to qualitativelyimprove on these fits by introducing a three-step persubunit activation sequence after Schoppa and Sig-worth (1998c), but this approach was not pursued fur-ther as discussed above. The time course of the decayphase of the IgON at depolarized voltages is a reflectionpredominantly of the forward rates of charge-movingtransitions among closed states. Fitting single-exponen-tial functions to IgON at 0 mV provides a way to estimatea weighted average of the rates a0 and g0. The wf timeconstant from such fits is z2.5 ms (n 5 11), which isover two times greater than the time constants forwfF401L and wfF401A. This provides the rationale forassigning the values of a0 5 560 s21 and g0 5 1,340 s21

for wt, or about one half of the corresponding zero-voltage rates for the two mutants. As described earlier,the kinetics of IgOFF are very sensitive to the amplitudeof the voltage step in the wf and especially wfF401L, butnot in wfF401A. The model is able, by the large differ-ences in the first closing rate l, to account for the timecourse of the OFF gating current in the three families.

Inspection of the time course of the gating charge re-turn as a function of pulse duration (Fig. 12) revealsimportant differences among the three channel spe-cies. Model fits to these data, shown in Fig. 16, indicatethat our simulations are adequate to describe the timecourse of IgOFF for a variety of pulse durations at 250and 0 mV. In particular, the observed slow decay of OFFgating currents observed in wfF401L at 250 mV atlonger pulse durations as the consequence of signifi-cant open probability of the channels at that voltage is

K0αβk0αk0β-------,=

K0γδk0γk0δ-------,=

Figure 14. A kinetic model for Shaker and the F401 mutants.Note that, compared with the ZHA model, independent transi-tions within each of the four subunits are completely symmetricaland that a concerted transition precedes channel opening. Aclosed state in which all four subunits are in the C2 state is implicitin the model; is it referred to as the Cn state in the text. (Bottom)The model estimates of the zero-voltage rate constants (k0, in s21)and associated quantities of charge movement (zk, in e0) for eachof the rates for the wt, F401L, and F401A channels are shown. Theinitial closing transition parameters are enclosed in a box to high-light the differences among the models.

Dow



observed in simulated traces. The complex waveformof wf IgOFF after 0-mV pulses lasting z10 ms deviatessomewhat from the predictions of our model andwould probably be better fit with the introduction ofadditional steps in the activation pathway (see above).The model predicts that the gating charge return willbe rapid and independent of pulse duration inwfF401A, which is clearly a feature of our data.

The predictions of the models for the macroscopicionic currents are shown in Fig. 17. Representative fam-ilies of activating currents recorded from patches con-taining wt, F401L, and F401A channels are qualitativelycomparable to model simulations at matching voltagesin terms of the overall sigmoidal character and the volt-age range over which activation kinetics are most no-ticeably changing. A consistent finding for all threechannels is that the model traces appear to have aslower overall time course than the patch data. This dis-crepancy is quantified in Fig. 17 B, which displaysmodel predictions for the time constant of activationderived from fitting the late phase of current timecourse (Zagotta et al., 1994a) and in Fig. 17 C in whichthe time-to-half-maximum current is plotted for thethree models. The relative order of magnitudes of ex-perimentally derived values for the three channels arepreserved in the model simulations. While a shift of ap-proximately 210 mV between the model and data re-sults in closer agreement, these kinetic measurements

obtained in excised inside- and outside-out patches arenot easily superimposable on the model simulations bya simple voltage offset. We wondered if the model pa-rameters that are mostly determined using gating cur-rent recordings obtained with cut-open oocyte clampsystematically predict slower ionic currents as the resultof differences inherent in the two recording tech-niques. Stefani et al. (1994) demonstrated that for thekinetics of gating currents, the cut-open oocyte clamptechnique produced very similar results to those ob-tained with cell-attached macropatches. In our expe-rience, activation kinetics of Shaker in cell-attachedpatches are somewhat slower (and deactivation isfaster) than the excised patches we used. We electednot to alter the models extensively to try to accommo-date both the cut-open oocyte clamp and excised patchdata sets, but instead focused on qualitative agreementbetween experimental ionic current results and modelsimulations.

In evaluating model predictions for the steady stateopen probability vs. voltage, we took notice of the ob-served differences in the voltage positions of G(V)curves obtained using wt cell-attached and excisedpatches. The former tended to be shifted positively byz6 mV (data not shown). When displaying the G(V)curves for wt, F401L, and F401A with their respectivemodels in Fig. 18, we similarly offset the simulatedcurves by between 26 and 27 mV to compare them

Figure 15. (A) Model descrip-tion of the steady state gatingcharge movement. The aver-aged Q(V) curves for the wf,wfF401L, and wfF401A channelsare replotted from Fig. 9. Fits ofthe model in Fig. 14 are shownsuperimposed as solid lines.These were obtained by integrat-ing the simulated OFF gatingcurrents at 2100 mV, low-pass fil-tered at 10 kHz, following stepsto the test potentials shown onthe x axis in 2-mV incrementsand normalizing by the chargeobtained at the most positivevoltages. This analysis follows thenormal procedure we use for theexperimental data. (B) Compari-son of gating current kineticswith the model predictions. Rep-resentative families of gating cur-rents from oocytes containing wf(top, from 280 to 140 mV),wfF401L (middle, 280 to 140mV), and wfF401A (bottom, 295to 125 mV) channels are shown.Voltage increment is 20 mV in

each case. Simulated gating currents are superimposed (thin lines). Models shown in Fig. 14 were used for the three channels, with the fol-lowing modifications to obtain better fits to these experimental families: in the wf model, l0 5 101 s21; in the F401L model, l0 5 39 s21;and in the F401A model, k0 5 3,500 s21, l0 5 8,000 s21, and zg 5 0.8 e0.

Dow



with the experimental results. With this correction,both the steepness and the midpoint of the voltage de-pendence for wt and F401L channels were well de-scribed by the model. As earlier, the shallowness of theG(V) relation in the F401A mutant precludes us fromobserving saturation of the open probability within theattainable voltage range. Therefore we cannot mean-ingfully normalize G(V) data from different patches fordirect comparison. Instead, G(V) relations from a rep-resentative F401A patch obtained by two means [iso-chronal tail G(V) and pulse G(V)] are shown in Fig. 18.Model traces for F401A were analyzed in the analogousmanner, and the resulting G(V) curves are shown scaledto match F401A curves at 1100 mV after a 26-mV shiftalong the x axis. The model, which postulates that theextremely shallow G(V) curve results from a destabiliza-tion of the open state by greatly speeding the initial clos-ing transition, is qualitatively supported by these data.

The proposed alterations in a backward transitionleading away from the open state are expected to affectthe time course of macroscopic ionic tail currents. Fig.19 (top) shows, for the wt and the two F401 mutants,the decay in the relative open probability as a functionof time when voltage is stepped from a depolarized

value of 150 mV to hyperpolarized potentials. All cur-rent traces are normalized to match their initial ampli-tudes. In Fig. 19 (bottom), these data are re-plotted ona logarithmic time axis. These two transformations al-low a closer examination of the kinetics of deactivationat the more hyperpolarized voltages at which tail cur-rents are very small and rapid. Additionally, a logarith-mic time scale would bring out the convergence of theopen probability traces to an asymptotic value at verylow voltages if a single closing transition were to be-come rate limiting. We used the models for the wt andF401L in which initial closing rates were modifiedsomewhat to reflect the slower deactivation kinetics inexcised patches. With l0 set to 50 s21 for the wt and 29s21 for F401L, the open probability decay is well de-scribed by the model predictions for both channelsover the range 260 to 2160 mV (2180 mV for F401L).The valence of 0.6 e0 assigned to this rate is the mini-mum required to produce the necessary spacing of thetraces at voltages below 2120 mV (Fig. 19, bottom).Less charge associated with the first closing step pre-dicts a rate-limiting step that is not evident in the data.The analysis of F401A deactivation (Fig. 19, right) iscomplicated by the extreme rapidity of its tail currents

Figure 16. Model predictions for the pulse duration dependence of the OFF gating charge movement. Experimental records of wf(left), wfF401L (center), and wfF401A (right) gating currents obtained at 250 and 0 mV are shown as in Fig. 12. The fits of the corre-sponding models are shown overlaid (thin lines). Modifications of the models needed for these simulations were as follows (s21): for wf, b05 153, d0 5 14, and l0 5 113 at 0 mV; for wfF401L, a0 5 1,150, d0 5 11.5, and l0 5 39 at 250 mV and b0 5 136 and d0 5 11.5 at 0 mV; andfor wfF401A, a0 5 1,200 and l0 5 8,000 at both 250 and 0 mV.

Dow



over a wide voltage range. Fitting exponential functionsto their time course yields time constants in the rangeof 100–500 ms, which are difficult to resolve well whenthe current amplitudes are small. Additionally, there isa prominent component of OFF gating current (notethe lowermost trace in the panel which was taken at290 mV) that is slower than the ionic tail in this mu-tant and significantly distorts its kinetics. SimulatedF401A traces have a multi-exponential decay, with thevery rapid component near the limit of resolution inour recordings (and perhaps more rapid than the ex-cised patch data), and a slower component that at 230and 250 mV cannot be reliably distinguished from thesteady state component seen in the experimental data.Because of the nonionic components of the current de-

cay (which the model does not take into account),more detailed comparison of the model simulations tothe tail currents in F401A at hyperpolarized voltageswas not undertaken.

Sigmoidal activation kinetics are a cardinal feature ofShaker channel gating and, as shown in Fig. 10, they re-main present in the F401 mutants. The amount of sig-moidicity, as defined earlier, and the way it varies withpulse potential is a sensitive means to assess the pres-ence of a slow first reverse transition from the openstate (Zagotta et al., 1994a). In Fig. 20, the models forthe wt, F401L, and F401A that differ mainly in the rateof that transition are used to generate activation fami-lies that are scaled in amplitude and time, as describedin the discussion of Fig. 10. The relative spacing of

Figure 17. (A) Model predictions for the activation of ionic currents. On the left, representative families of currents recorded at be-tween 290 and 150 mV (wt), 295 to 125 mV (F401L), and 280 to 1140 mV (F401A) are displayed. Voltage increments were 10 mV (20mV for F401A). Corresponding simulated ionic currents are shown on the right. Model parameters were not altered from those in Fig. 14,B and C. Comparison of the model with measured ionic current activation kinetics. Activation time constants (B) and time-to-half-maxi-mum (C), obtained as described in the text, are plotted as a function of test voltage for wt, F401L, and F401A channels. Predictions of themodels are shown as solid and dashed lines. These were obtained by the same analysis procedures applied to the simulated currents as tothe experimental records.

Dow



these traces gives a measure of sigmoidicity over thevoltage ranges of 235 to 155 mV (wt model), 255 to145 mV (F401L model), and 240 to 1120 mV (F401Amodel). As in the experimental records, F401L and wt

model simulations show the progression from lower toasymptotically higher sigmoidicity, which is a predic-tion for a forward biased cooperative step with the slowrate of leaving the open state. F401A simulations dis-

Figure 18. Model descriptionof the conductance–voltage rela-tionships. Experimental data forwt, F401L, and F401A are shownplotted as in Figs. 1 and 5. Simu-lated G(V) relations were gener-ated from current families takenevery 2 mV. The resultant curveswere then shifted negatively by 6(wt and F401L) or 7 (F401A) mVand shown superimposed aslines. The predicted G(V) curvesfrom pulse and tail (see meth-ods) for F401A were normalizedto match the maximum relativeconductance of the data.

Figure 19. Models predict the deactivation kinetics in Shaker and F401 mutants. Ionic tail currents that arise during channel deactivationat negative voltages are shown as relative open probability as the function of time by scaling to match the instantaneous amplitudes afterthe end of the 150-mV test pulse (top). The current families for the tail voltages of 260 to 2160 mV (wt, left), 260 to 2180 mV (F401L,center), and 230 to 290 mV (F401A, right) are shown. Records at 2100 mV are omitted because the current amplitudes are very low nearthe reversal potential. Note the inward current at the start of the tail at 290 mV in F401A, which is largely due to the OFF gating currentcomponent. Model predictions for the time course of the open probability decay are superimposed as thin lines. (Bottom) The data andmodel traces for the wt and F401L are replotted on a logarithmic time axis to enable closer comparison of the model to the data at the low-est voltages. For these simulations, l0 was 50 and 29 s21 for the wt and F401L models, respectively.

Dow



play nearly identical sigmoidicity over a wide voltagerange characteristic of this mutant’s currents.

d i s c u s s i o n

In this paper, we have confirmed, with direct evidenceobtained from channels without N-type inactivation,the hypothesis put forth by Zagotta and Aldrich(1990b) that Sh5 (F401A) selectively reduced the volt-age dependence of deactivation with little effects onthe forward transitions. Any effects of Sh5 on inactiva-tion can be most economically explained by the tightcoupling of intrinsically weakly voltage-dependent inac-tivation to the altered activation process (Zagotta andAldrich, 1990a,b).

We used F401, an important residue with known ef-fects but a poorly

determinants of voltage-dependent gating and open-state … · embo [eur. mol. biol. organ.] j....

Documents