Evidence for Coordinated Activity of Small Elementary

Published June 1, 1996

Endogenous Chloride Channels of Insect Sf9 Cells

Evidence for Coordinated Activity of Small Elementary Channel Units E. HVIID LARSEN, S.E. GABRIEL, M.J. STUTTS,J. FULLTON, E.M. PRICE, and R.C. BOUCHER From the Cystic Fibrosis Center, Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-7020 and the August Krogh Institute, University of Copenhagen, DK-2100 Copenhagen O, Denmark

INTRODUCTION Expression o f wild-type a n d e n g i n e e r e d ion c h a n n e l f o r m i n g proteins in i m m o r t a l i z e d cell lines constitutes an i m p o r t a n t tool f o r associating the c h e m i c a l structure o f the p o r e - f o r m i n g m o l e c u l e with biophysical properties o f the c o n d u c t i n g process such as gating, seE.M. Price's current address is Dalton Research Center, Research Park, University of Missouri, Colombia, MO 65211. Address correspondence to Efik Hviid Larsen, August Krogh Institute, Universitetsparken 13, DK-2100Copenhagen ~, Denmark. Fax: 45 3532 1567; E-mail:[email protected] 695

lectivity, c o n d u c t a n c e , rectification, a n d m o l e c u l a r p h a r m a c o l o g y . T h e application o f the c u l t u r e d insect Sf9 cell line for h e t e r o l o g o u s expression o f ion-channel p r o t e i n s was i n t r o d u c e d in Miller's laboratory in a study o f Shaker potassium c h a n n e l s (Klaiber et al., 1990). It was d e m o n s t r a t e d that by splicing the c l o n e d foreign c D N A into the baculovirus vector, Autographa californica, a n d using the p r o m o t e r o f the viral polyhed r o n gene, high levels o f the r e c o m b i n a n t m e m b r a n e p r o t e i n c o u l d be expressed. This expression system has b e e n used n o w also for the study o f a n i o n selective channels, e.g., the cystic fi-

j. GEN.PnYsIOL9 The RockefellerUniversityPress ~ 0022-1295/96/06/695/20 $2.00 Volume 107 June 1996 695-714

Downloaded from jgp.rupress.org on September 22, 2015

ABSTRACT The e n d o g e n o u s C1- conductance of Spodopterafrugiperda(sfg) cells was studied 20-35 h after plating out of either uninfected cells or cells infected by a baculovirus vector carrying the cloned [3-galactosidase gene ([3-Gal cells). With the cation Tris + in the pipette and Na § in the bath, the reversal potential o f whole-cell currents was governed by the prevailing Cl- equilibrium potential and could be fitted by the Goldman-Hodgkin-Katz equation with similar permeabilities for uninfected and [3-Gal cells. In the frequency range 0.12 < f < 300 Hz, the power density spectrum o f whole-cell C1- currents could be fitted by three Lorentzians. I n d e p e n d e n t o f m e m b r a n e potential, > 5 0 % of the total variance o f whole-cell current fluctuations was accounted for by the low frequency Lorentzian (f~ = 0.40 ___0.03 Hz, n = 6). Single-C1- channels showed complex gating kinetics with long lasting (seconds) openings interrupted by similar long closures. In the o p e n state, channels exhibited fast burst-like closures. Since the patches normally contained more than a single channel, it was not possible to measure o p e n and closed dwell-time distributions for comparing single-C1- channel activity with the kinetic features of whole-cell currents. However, the power density spectrum of C1- currents o f cell-attached and excised outside-out patches contained both high and low frequency Lorentzian components, with the corner frequency of the slow c o m p o n e n t ~ = 0.40 -+ 0.02 Hz, n = 4) similar to that of whole-cell current fluctuations. Chloride channels exhibited multiple conductance states with similar Goldman-Hodgkin-Katz-type rectification. Single-channel permeabilifies covered the range from ~0.6 9 10 -I4 cm3/s to ~ 6 9 10 -I4 cm3/s, corresponding to a limiting conductance (~/150/150) of N3.5 pS and ~ 3 5 pS, respectively. All states reversed near the same m e m b r a n e potential, and they exhibited similar halide ion selectivity,/91 > Pcz ~ PBr- Accordingly, C1- current amplitudes larger than current flow through the smallest channel unit resolved seem to result from simultaneous o p e n / s h u t events of two or more channel units. Key words: Sf9 cells * chloride channels * patch clamp * noise analysis 9 coordinated ion channel activity


MATERIALS

AND M E T H O D S

Preparation

retained their original spherical shape during incubation as well as during the subsequent patch-clamp studies.

Whole-Cell and Single-Channel Current Recordings Patch pipettes were constructed from borosilicate glass tubes with an inner filament (GC150F; Clark Instruments, Reading, UK) on a Narishige PP-83 or a DMZ (Zeitz Instruments, Augsburg, Germany) patch electrode puller using standard two-pull and three-pull techniques, respectively. The tip was heat polished to give an electrode resistance of 2-3 MI~ when measured with the respective standard solutions in electrode and culture dish, respectively (see "Solutions"). The culture dish was positioned on the table of an inverted microscope (IM35; Carl Zeiss, Inc., Thornwood, NY), and the cells were viewed with DIC Nmnarski optics at a magnification of 500. With a small pressure applied to the pipette solution, the tip of the patch pipette was directed with a three-dimensional hydraulic micromanipulator (MO-103; Narishige, Tokyo, Japan) to the surface of a selected cell. All four variants of the patch-clamp methods were used (Hamill et al., 1981). Whole-cell recording conditions were obtained by suction ( ~ - 0 . 2 atm) applied to cell-attached patches with a membranepipette seal-resistance >30 GI~. Excised outside-out patches were formed from the whole-cell configuration by slow withdrawal of the pipette. After giga-seal formation, patch-electrode voltage clamps and current recordings were performed with an RK-300 amplifier (Biologic, Claix, France) furnished with a headstage feedback resister of either 100 Ml) (whole-cell recordings) or 10 GI~ (singlechannel recordings). The CED patch- and voltage-clamp software and the CED 1401 interface were used for generating clamping programs and acquiring, digitizing, and analyzing resulting currents (Cambridge Electronic Design, Cambridge, UK). Singlechannel currents were recorded to videotape (modified PCM501ES, Sony Tokyo; AG-6200, Panasonic, Osaka, Japan) and subsequently low-pass filtered (f~[ - 3 dB] = 200-1,000 Hz; eight-pole Bessel, or eight-pole Butterworth when indicated [see Fig. 14]; Frequency Devices, Inc., Haverhill, MA) and digitized at 1 or 2 kHz. In some but not all experiments, capacitance neutralizations were performed. Because of the small membrane currents of Sf9 cells, the pipette access resistance was not always corrected for.

Current Processing for Fluctuation Analyses

We used cultured insect Spodopterafrugiperda (Sf9, InVitrogen Corp., San Diego, CA) cells, uninfected or infected with recombinant virus carrying the [3-galactosidase gene. Infected cells were included to investigate whether expression of viral proteins influences the magnitude of endogenous whole-cell chloride conductance. The cells were maintained at 27~ in Petri dishes (Nunc, Roskilde, Denmark; 4 cm OD, 2 ml, 3 9 106 cells/ml) in Grace's medium supplemented with yeastolate and lactalbumin hydrolysate, to which we added 10% fetal bovine serum and antibiotics (50 ~zg/ml gentamycin). Electrophysiological recordings were made at room temperature between 20 and 35 h after plating and replacing of the tissue-culture medium with either of the bath solutions indicated below ("Solutions"). Practically all cells

1Abbreviations used in thispaper. CFTR, cystic fibrosis transmembrane conductance regulator; GHK, Goldman-Hodgkin-Katz; PDS, power density spectrum. 696

Current fluctuations were analyzed according to standard methods (Anderson and Stevens, 1973; Lindemann and Van Driessche, 1978) using the SPAN program of Dempster's patch-and voltageclamp software package (Strathclyde Electrophysiology, Glasgow, UK). For obtaining power density spectra (PDS) of sufficient resolution and at the same time covering the considerable range of 0.12-1,000 Hz, it was necessary to process, with different bandwidth and time record length, the same stationary current segment twice: Stationary whole-cell currents of 150-180 s duration stored on videotape were low-pass filtered ( - 3 dB, eight-pole Bessel) at 990 Hz or 245 Hz and digitized at an amplification of 10 at a sampling rate of 2,000 Hz or 500 Hz, respectively. The high and low frequency spectra were then calculated by a 4,096-point fast Fourier transform over square-pulse time windows covering 2.048-s or 8.192-s current records, respectively. Each spectrum was obtained by averaging such 16-180 individually transformed records and merged to a single PDS before curve fitting. Some

EndogenousC1- Channels of Sf9 Cells


brosis t r a n s m e m b r a n e c o n d u c t a n c e r e g u l a t o r (CFTR1; K a r t n e r et al., 1991; Sarkadi et al., 1992; G a b r i e l et al., 1992), a n d o f r e c o m b i n a n t s u b u n i t s o f the h u m a n ~ - a m i n o b u t y r i c acid r e c e p t o r ( B i r n i r et al., 1992; N i e l s e n et al., 1995). It has b e e n a d v a n t a g e o u s for these studies that the h e t e r o l o g o u s l y e x p r e s s e d p r o t e i n s c o u l d be s t u d i e d s u p e r i m p o s e d u p o n small e n d o g e n o u s i o n p e r m e a b i l i t i e s . I n o u r studies of Ca~'TRe x p r e s s i n g Sf9 cells, however, it t u r n e d o u t that the b a s e l i n e C1- p e r m e a b i l i t y o f CFTR-infected a n d u n i n fected cells was a b o u t the same (Larsen et al., 1992). F u r t h e r m o r e , s i n g l e - c h a n n e l studies i n d i c a t e d the prese n c e o f a small l i n e a r C1- c h a n n e l (Gabriel et al., 1992) o f a c o n d u c t a n c e similar to that r e p o r t e d for C F F R ( K a r t n e r et al., 1991; Bear et al., 1992). T h e g a t i n g kinetics o f e n d o g e n o u s c h a n n e l s was c o m p l e x , a n d single c h a n n e l s e x i b i t e d very l o n g dwell-times i n o p e n state, which also characterizes C F T R - m e d i a t e d c h a n n e l activity (Larsen et al., 1993; Fisher a n d M a c h e n , 1994; Gadsby et al., 1995). We t h e r e f o r e d e t e r m i n e d that t h e r e was a n e e d for a m o r e t h o r o u g h c h a r a c t e r i z a t i o n of e n d o g e n o u s c h l o r i d e c h a n n e l s o f Sf9 cells, b o t h at w h o l e cell a n d at i n d i v i d u a l p r o t e i n c h a n n e l level. I n the p r e s e n t study, we analyzed a n d c o m p a r e d p r o p e r ties of e n d o g e n o u s c h l o r i d e c h a n n e l s by f l u c t u a t i o n analyses o f m a c r o s c o p i c m e m b r a n e c u r r e n t s a n d by sing l e - c h a n n e l r e c o r d i n g s . O u r results i n d i c a t e that small e l e m e n t a r y C1- c h a n n e l u n i t s are f o u n d i n close proximity to o n e a n o t h e r a n d that several units, transiently, e x h i b i t c o o r d i n a t e d gating. T h e r e s u l t i n g c o n d u c t a n c e t r a n s i t i o n s r e s e m b l e the activity of larger c h a n n e l s . Furt h e r m o r e , o u r results will aid f u r t h e r studies o f the ani o n p e r m e a b i l i t y o f Sf9 cells after e x p r e s s i o n of a n i o n t r a n s p o r t e r s like, for e x a m p l e , CFTR a n d ~/-aminobutyric acid r e c e p t o r s u b u n i t s .


spectra were also calculated by multiplying the fast Fourier transforms with a 10% cosinus bell window time function, but the results obtained were not significantly affected by this type of windowing (not shown). Generally, therefore, fast Fourier transformations were performed without multiplying with any time window function. All spectra are shown and were analyzed without notch-filtering. Two to three Lorentzian curves (see Eq. 7a) were fitted to the calculated PDS using the Levenberg-Marquardt nonlinear least-square routine. The stationary mean-current component used for calculating the variance of fluctuations (see Eq. 5b) was obtained by additional sampling and digitizing through a low-gain (X 1) DC-coupled channel.

IntraceUular Potentials of Sf9 Cells

Solutions Standard bath solution contained the following(mM): 127 Na +, 4 Ca z+, 5 Mg 2+, 150 CI-, 10 glucose, 50 mM sucrose, and 10 N-tris (hydroxymethyl) methyl-2-aminoethane sulfonic acid (TES), pH -= 6.8. In sodium-free bathing solution, Na + was replaced by Tris +. Standard pipette solution contained the following (mM): 130 Tris +, 0.8 Mg 2+, 0.38 Ca z+ + 1 mM EGTA (100 nM free Ca2+), 40 CI-, 90 gluconate, 3 ATP, 0.1 ADP, 0.05 GTP, 1 pyruvate, 1 acetate, 1 L-aspartate, 1 D-glucose, 5 TES, pH = 7.2 adjusted with NaOH resulting in a final [Na +] of 12 mM. In experiments with pipette [CI-] raised to 140 mM, Tris + was raised to 140 mM and added as Tris-HC1. Channel selectivity studies were performed with Br- or I- partially replacing Cl- in the bath: 100 mM of either Br- or I-, 50 mM CI-, otherwise identical to the standard bath solution indicated above. 10 mM of either NaBr or NaI was added to standard pipette solution (above), maintaining [C1-]p = 40 mM. Only reagent-grade chemicals were used.

Correctionfor Liquid Junction Potentials With the pipette in the bath, the liquid junction potential (VLj) between the bath and pipette solutions as well as other external

1

Bath solution*

Pipette solution*

VLj+,

150 mM CI50 mM C1-/100 mM Br50 mM CI-/100 mM I-

40 mM C140 mM C1-/10 mM Br40 mM CI /10 mM I-

+8.0 mV +6.3 mV +5.8 mV

The values are means of three to five determinations which, with the floating KCI bridges and large solution reservoirs, showed practically no variation among subsequent measurements. *Only halide ion concentrations are listed. Other components are listed in Materials and Methods. ZThe sign convention of VLjis such that bath potential is given relative to pipette potential.

electromotive forces was nulled before the pipette tip was pressed against a selected cell. Thus, in cell-attached recordings with different solutions in bath and pipette ( VU r 0), and with pipette potential, Vp, the patch-membrane potential (VM) becomes (Neher, 1991) VM =

VC -

Vp-~- VLj.

(la)

Here, Vc is the intracellular potential (cf. above), and VL3 indicates bath potential relative to pipette potential. With similar conventions, in the whole-cell and excised outside-out configurations, VMis given by: VM = V p - VLj.

(lb)

VLj for bath/pipette solution pairs used in the present study was measured with floating 2 M KC1 bridges and large solution reservoirs. The values obtained are listed in Table I.

Description of Ion Flow through Chloride Channels It turned out that the C1 current-voltage relationship of a homogenous constant field membrane could be fitted to currents obtained from both whole-cell and single-channel voltage-clamp recordings. The current, ~l, carried by such a channel obeys the Goldman-Hodgkin-Katz (GHK) current equation (Goldman, 1944; Hodgkin and Katz, 1949): Pc~"/?" V. { [ C l - ] c - [C1-] o. e x p ( - F " V/(R. 7) ) } icl

R . T. {1 - e x p ( - F . VI(R" 7 ) ) }

(2) where Pcl is the single-channel permeability (cm3/s per channel), V is the membrane potential with reference to external bath, [C1 ] denotes concentration on the external (o) and internal (c) side of the membrane, and F, R, and T have their usual meanings. The integral (or chord) conductance (~o) of the channel is given by the following (e.g., Finkelstein and Mauro, 1963; StenKnudsen, 1978):

Pcl'F 3" V" { [C1-]~ - [CI-]~ 9 e x p ( - F . V/(R. 7 ) ) } R z . T 2- {exp(-F.V/(R. T))-I } -In{[Cl-]c/([Cl 697

L A R S E N ET AL.

lc'exp(-F" VI(R. T))) } ( V C E ) o

(3a)


During the first day of incubation, Sf9 cells settled on the bottom of the Petri dish, and they could be impaled with a conventional single-barreled microelectrode (2 M KC1 filling solution, 80-120 MII tip resistance) connected to a high impedance electrometer (N-707A; World Precision Instruments, Inc., Sarasota, FL) and using a piezoelectric transducer (PM20H; Elke Nagel-Biomedizinische Instrumente, Mfinchen, Germany) for advancing the tip of the electrode through the membrane of a visually selected cell. With standard bath outside (150 mM CI-, see below), intracellular potentials ranged from - 1 0 to - 2 5 mV with a mean value, Vc = - 1 4 . 2 + 4.9 mV (+SEM, n = 10 cells). With the likely assumption that CI- is in equilibrium (see below), implying that at steady state the membrane potential is set by the distribution of the small diffusible cations and their respective membrane permeabilities, the above value of Vc would indicate an intracellular CI- concentration of ~80 raM.

TABLE

LiquidJunction Potentialsof Bath~PipetteSolution Pairs


Pc~.F

s" V .

[ C 1 - ] o " [C1-]~

~ci =

(V=Ecl)

(3b)

R 2" T ~" { [C1-]o - [C1-] r with /c] = "/c] " ( V - E e l ) , and the e q u i l i b r i u m potential, Ecl = - ( R 9 T / b ) 9 l n ( [ C 1 - ] o / [ C I - ] ~ ) . In symmetrical 150-mM C1- solutions, a condition often prevailing in studies of excised insideout patches, the c h a n n e l conductance would be given by:

Y150/150 - -

Pc," F2. [ C l - ] R' T

([Cl-] = 150mM).

(3c)

In the text, "/150/L~0denotes the limiting conductance. Halide ion selectivity of single channels studied in the outside-out patch configuration was estimated from the c h a n n e l ' s reversal potential, Vr, in experiments with mixtures of C1- a n d a n o t h e r halide ion, A = Br- or I-, in b o t h bath a n d pipette solution. The relative permeability, PA/Pc], was calculated by the following equation derived from the GHK potential equation:

[CI-]~.exp(-V~'F/(R" T)) [A] . e x p ( - V , . F / ( R 9 T) )-[A] o

[C1- ]o-

(4)

where Vr is the reversal potential corrected for the liquid j u n c t i o n potential (cf. above). Whole-cell currents, Ic~, were also fitted by Eq. 2, providing the chloride permeability of the cell m e m b r a n e (cm3/s per cell), which entered into Eqs. 3a a n d 3b for calculating the associated macroscopic C1- conductances (Gc]).

Passive C i r c u i t

10 z

22 pF

(1) S,9-Ce,,

tl

I

R,

Interpretation of Stationary Current Fluctuations Noise data were analyzed in terms of their variance, (s and power spectral densities, S0r), where (s was obtained by integrating the theoretical (fitted) PDS (Christensen a n d Bindslev, 1982; Lindemann, 1984): (T2 =

I~i'S(j) . df

1~176 ~ 10 -1

10"4

1) over the interval, 1 300 Hz, the unavoidable noise contributed by the recording network is also evident. Biological noise sources. For a single population of channels exhibiting stochastic behavior, the PDS of macroscopic current fluctuations, caused by random o p e n / s h u t events of individual channels, is given by the following (Stevens, 1972; Neher and Stevens, 1977; Colquhoun and Hawkes, 1977):

L = l/(2w"r).

Accordingly, with N channels having open probability, po, and carrying current, i, in its open state of mean duration, "r~ en, we have (e.g., Neher and Stevens, 1977):

2

(Sb)

,opt, = ,~/(1 - Po),

(8c)

i = (r / [ I m 9 (1 - p o ) ]

where I~ is the mean whole-cell current. Performing the integration in Eq. 5a, with S(f) defined by Eq. 7a, it follows that the bandwidth-unlimited variance entering into Eq. 8b can be calculated from:

(7a)

1 + (f/f)~

(8a)

I,n = i . p o . N

So S(f)

(7b)

where So is the low frequency asymptote, and the corner frequency, f~, is related to the time constant ('r) of the fluctuations by:

2

= -rr. S0 "f~/2.

(9)

[m -

P/B: 4 0 / 1 5 0 mM CI-

(pAl "

P/B: t 4 0 / 1 5 0

Vh = 0 mV

mM CI-

Vh = 0 mV

250

-250

50 ms

50 ms

-500

B

~

Sf9-Cell- UninfectedI

[Ctle/[Cl]p o 15o/4o o 1=50/14] ~

S f 9 Cells -

TIm (pA)

[Ct-]e/[Cl-lp 9 150/40 150/140

i

~ 200 Vc (mY)

o

100 -100 '

-100 ~

f

-200 Q

t

!8-Galactosidase

I

Irn(pA) 2O0 ,.~,4j

.~'/ ...l:~,

510 Yp [rnVi~0

I -lOO -2oo

FIGURE 2. Whole-cell currents of SJ9 cells. (A) Uninfected cells. Shown are current responses to 200-ms square voltage pulses generated from 0 mV holding potential and with pulse amplitude varying in steps of 10 mV from - 8 0 to +60 mV, Currents were filtered at 500 Hz and digitized at l-ms intervals. (Left) With pipette/bath [C1-] = 40/150 mM, die membrane rectified in the outward direction; outward currents were more noisy. (Right) With pipette/bath [C1-] = 140/150 mM, current amplitudes were near-symmetrical for voltage steps of similar amplitude and opposite signs, and the noisy appearance no longer depended on direction of current flow. (B) //Vrelationships constructed from currents generated in the way shown in A by sampling from 150 to 190 ms after onset of voltage pulse. Raising [C1-]p from 40 to 140 mM resulted in rightward displacement of reversal potential towards zero, as expected for currents carried predominantly by chloride. The full lines show the fit ofEq. 2 to experimental data. 699

LARSEN ET AL.


A


TABLE

II

A

Membrane Currents of Sf 9 Cells, Uninfectedor ExpressingRecombinant cDNAfrom the Baculovirus Vector,Autographa californica Cells

V~

V~~ mV

Pci

l lT s2cm3/(s.cell)

Uninfected -34.1 _+2.0 -42.1 [3-Galactosidase -31.7 -+ 2.1 -39.7

2.4 -+ 0.3 2.4 -+ 0.5

G,

300

Vc-Ecl

DIGITIZED MEMBRANE CURRENTS

+50 mV

Im=223•

pA (n=146)

-

I~0 z l~J

I0-2

10-4

rr IJJ

B 10 G O h m

0

Q_

10 -0

0.01

0.1

1 FREQUENCY

10

100

1000

1U

(Hz)

FIGURE 14. PDS o f CI c u r r e n t flow in m e m b r a n e p a t c h e s o f SJ9 cells. Blocksize 4096, b a n d w i d t h 200 Hz ( - 3 dB, eight-pole Butterworth). (A) Cell-attached p a t c h with identical ( s t a n d a r d ) s o l u t i o n s in pipette a n d b a t h (Materials a n d M e t h o d s ) , a n d pipette p o t e n tial c l a m p e d at t h e t h r e e d i f f e r e n t values indicated. T h e t h r e e s p e c t r a s h o w n are f r o m t h e s a m e patch. D a s h e d lines are calculated as t h e s u m o f t h r e e L o r e n t z i a n c o m p o n e n t s (Eq. 6b) with t h e following p a r a m e t e r s . - Vp = + 4 0 mV: f~l = 0.42 Hz, SOl = 0.30 p A ~" s;f~2 = 9.25 Hz, Sos = 1 . 3 . 1 0 -3 p A 2. s;f~3 = 85.6 Hz, Sos = 2 9 10-4 p A 2 . s - Vp = - 40 mV: f~l = 0.43 Hz, SOl = 0.13 p A 2. s; fez = 11.3 Hz, SO2 = 6.2 - 10-4 p A z 9 s ; f ~ = 5 6 H z , SO~ = 8 9 10 -5 p A 2 9 s. H e r e , a n d in B, t h e h i g h f r e q u e n c y powers were o m i t t e d for ohm i n i n g t h e L o r e n t z i a n fits to t h e calculated s p e c t r a . T h e PDS obt a i n e d at Vp = 0 m V ( Vm = Eel) indicates t h e noise floor o f t h e rec o r d i n g s e t u p w h i c h is n o t "white" a n d s o m e w h a t larger t h a n t h e theoretical m i n i m u m level given by t h e J o h n s o n - N y q u i s t noise o f t h e h e a d s t a g e ' s f e e d b a c k resister (10 GO; Eq. 6a). T h e line freq u e n c y h e r e is 50 Hz (University o f C o p e n h a g e n ) c o m p a r e d with 60 Hz in Fig. 3 (University o f N o r t h C a r o l i n a at C h a p e l Hill). (B) Excised o u t s i d e - o u t p a t c h with s t a n d a r d b a t h ([C1-] = 150 m M ) a n d pipette ([C1 ] = 40 raM) solutions (Materials a n d M e t h o d s ) . T h e two s p e c t r a were o b t a i n e d f r o m t h e s a m e patch. T h e Lorentzian p a r a m e t e r s u s e d for c o n s t r u c t i o n o f t h e d a s h e d relationships are Vp = + 2 0 mV:f~l = 0.33 Hz, SOl = 0.61 p A 2" s ; f 2 = 40 Hz, SO2 = 709

2

LARSEN ET A L

~ -100

Vp (mY), 50 100

-1 FIGURE 15. Analysis o f B r - / C I selectivity o f c o n d u c t a n c e levels resolved in excised o u t s i d e - o u t p a t c h o f insect SJ9 cell. [Br-]B = 100 m M ; [C1-]B = 50 m M ; [ B r - ] r = 10 m M ; [C1-]p = 40 m M . (A) C u r r e n t s o b t a i n e d at pipette potentials o f - 8 0 m V (two upper traces) a n d + 8 0 m V (two/ower traces). C o n d u c t a n c e levels a r e indicated by c u r r e n t j u m p s f r o m closed state resolved by Guassian fits to h i s t o g r a m s o f c u r r e n t distributions. (B) C u r r r e n t - v o l t a g e relationships o f t h r e e resolved c o n d u c t a n c e levels indicate a c o m m o n reversal potential.

5 9 10 -4 p A 2 " s;fcs = 75 Hz, SO3 = 1 " 10 -4 p A 2 " s. Vp = 0 m V : f d = 0.40 Hz, S01 = 0.10 p A 2" S;fc2 = 2 7 H z , Sos = 1.3 9 10-4 p A 2 9 s;f~3 = 79 Hz, SOl = 2 9 10 -5 p A 2" s.


E-oo


the variance o f the whole-cell m e m b r a n e c u r r e n t fluctuations is g o v e r n e d by slowly g a t e d c h l o r i d e channels.

Halide Ion Selectivity of Individual Conductance States

~

C

..........................................................

C

.............................. - ~, -..... - . . . . . . - ~ C 710

EndogenousC1- Channels of Sf9 Cells

............

C

FIGURE 16. Analysis of I - / C1 selectivity of conductance levels resolved in excised outside-out patch of insect Sjq9cell. [I-] B = 100 mM; [C1-] B = 50 mM; [I-]p = 10 mM; [Cl-]p = 40 raM. At a constant Vp = 50 mV, current jumps to several different levels are seen. ~e~ = 91 Gf/.


It was shown above that it was possible, in the same patch, to retrieve i n f o r m a t i o n o n a set o f c h a n n e l events at different m e m b r a n e potentials for the construction o f families o f i/V relationships (Figs. 6 a n d 13). However, after a c h a n g e in external halide ion c o m p o s i t i o n , it was n o l o n g e r possible to associate the individual c o n d u c t a n c e levels with those identified before the c h a n g e o f solution. T h e r e f o r e , with such a protocol we c o u l d n o t analyze shifts in reversal potentials to obtain i n f o r m a t i o n o n the selectivity o f the c o n d u c tance levels. It t u r n e d out, however, that families o f i/V curves c o u l d be o b t a i n e d in e x p e r i m e n t s where chloride ions were partly r e p l a c e d by either b r o m i d e o r iodide. With this protocol, we n e e d e d j u s t a single set o f i/Vrelationships for calculating c h a n n e l selectivities. As can be seen by the r e c o r d i n g s shown in Fig. 15 A, with b o t h B r - a n d C1- in bath a n d pipette solution, several different c h a n n e l amplitudes were observed in the same patch. T h r e e c h a n n e l s were resolved, a n d they were all f o u n d to reverse n e a r - 2 0 m V (Fig. 15 B). After c o r r e c t i o n for a liquid j u n c t i o n potential o f 6.3 m V (Table I; Eq. l b ) the reversal potential e n t e r i n g Eq. 4 is

V r e v ~-- - 2 6 . 3 mY, f r o m which we calculate a selectivity ratio o f PsJPo -~ 0.90 (PB~/Pcl = 0.93 -- 0.08, n = 3; the ratio is n o t statistically different f r o m unity). Also, in patches studied with a C 1 - / I - ion pair, several different event amplitudes were resolved. Fig. 16 shows c u r r e n t segments o b t a i n e d with pipette potential held at + 5 0 mV, i.e., with a large inwardly directed driving force i m p o s e d o n the halide ions. Despite c l a m p i n g the driving force to a c o n s t a n t value, the curr e n t j u m p e d to several levels, a p p a r e n t l y in a r a n d o m fashion. T h e largest event was ~ 1 . 8 pA, b u t several o t h e r o p e n states o f l o n g dwell-times are also seen. Similar sets o f observations were o b t a i n e d at o t h e r m e m b r a n e potentials. F r o m inspection o f the c u r r e n t r e c o r d s shown in Fig. 17, it can be seen that c u r r e n t s all reversed at a pipette potential n e a r - 3 0 inV. T h e family o f i/Vrelationships c o n s t r u c t e d f r o m these c u r r e n t records is shown in Fig. 18 A. Individual current-voltage curves were near-linear, with slopes r a n g i n g f r o m 3.7 to 22.1 pS. Fig. 18 B shows that individual c o n d u c tance levels were a multiple o f the smallest c o n d u c tance resolved (r 2 = 0.993). O n the average, with this ion pair a reversal potential o f 30.0 + 2.7 mV (n = 8) was obtained. With VLj = + 5 . 8 mV (Table I), the estim a t e o f the reversal potential o f all c o n d u c t a n c e levels is V~ ~ - 3 5 . 8 inV. F r o m this value we calculate a selectivity ratio o f P1/Pcl = 2.0.


DISCUSSION

This study has shown that fluctuations of whole-cell chloride currents in insect Sf9 cells cannot simply be the result of r a n d o m opening and closing o f just a single population of two-state ionic channels. It has further been demonstrated that m e m b r a n e patches of these cells contain channels of multiple time constants and multiple conductance levels. Here, we shall discuss causal relationships between the gating features revealed in whole.cell and in single-channel recordings, respectively. The o t h e r question to be addressed is whether different levels of single-channel conductance result from simultaneous opening and closing of a varying n u m b e r of smaller but similar channels.

Whole.cell recordings showed that the major membrane current was carried by channels specific for C1(Fig. 2, Table II). In this situation and with a well-defined high frequency background noise, it was fairly easy to verify that fluctuations of m e m b r a n e currents r e c o r d e d in the band, Af = 2-1,000 Hz, were caused by chloride channel activity (Fig. 3, A and B). However, the membrane currents also displayed low frequency fluctuations, and they contributed relatively significantly to the total variance at all potentials (Table III). Therefore, although some spectra became fairly flat below 1 Hz, it is important to consider to what extent extraneous noise contributed to the low frequency power. To deal with this question in detail, we analyzed the relative contributions of different background noise sources to the variance of current fluctuations in the entire frequency range of interest. At higher frequencies, the spectra generated by nonbiological sources were well predicted by theoretical considerations. For f < 2 Hz, the background spectrum deviated from theory by rising rather steeply with decreasing frequency, however, with unpredictable "roll-off" (cf. Figs. 1 and 3). With 970-pA DC current flowing to ground, the PDS of the nonbiological noise source was several orders of magnitude smaller than that of whole-cell currents of similar magnitude (Fig. 1). Thus, it is indicated that, also at the lowest frequencies, the PDS of Sf9 membrane currents is p r o d u c e d by electrical activity of ion channels. Single-channel recordings supported this contention. Chloride channels studied both in cell-attached (Fig. 5), in excised inside-out (Fig. 9), and in excised outside-out patches (Figs. 11, 15 A, 16, and 17) entered into o p e n states of 0.1-5-s durations interrupted by similar long closed states. The channels also exhibited faster gating modes, seen as brief closures within the long periods of openings. Furthermore, the resulting 711

LARSEN ET AL.

Multiple Conductance States Indicate the Presence of Elementary Channel Units Identification of channel currents. In cell-attached patches, e n d o g e n o u s K + channels of Sf9 cells reverse at pipette potentials between 85 and 100 mV and carry outward currents when Vp is held at 0 mV (Claus Nielsen and E. Hviid Larsen, unpublished observations). Thus, K + currents are easily distinguished from C1- currents which, with [C1-]p = [C1-]B, reversed near Vp = 0 mV (Fig. 5). In a g r e e m e n t with this notion, when [C1-]p was lowered significantly, i.e., from 150 to 40 mM, currents carried by cell-attached channels reversed at -Vp > 0 mV (Fig. 7). In excised outside-out patches with Tris + as major cation in pipette solution, cation (Na +) channel currents would reverse at ~ + 60 mV. However, all channel currents resolved in this configuration reversed near the major anion's negative reversal potential (e.g., Fig. 13 B). Thus, it was not difficult to identify the conducting ion in electrically active channels. In the present study, we never observed m e m b r a n e patches with cation-selective channels. With this information and results of the whole.cell current analysis (Fig. 2 and Table II), it seems that Sf9 cells grown u n d e r the present conditions express a relatively very low cation permeability. Since they also generate a fairly small m e m b r a n e potential, Vc = - 1 5 m V , 2 cultured Sf9 cells share macroscopic m e m b r a n e electrophysiology with mammalian red cells. They exhibit a small m e m b r a n e potential, a relatively very low cation permeability, and

2Similar to all whole-cell and single-channel current recordings, the m e m b r a n e potential was measured in cells bathed in a solution made slightly hypertonic by 50 mM sucrose; see Materials and Methods.


Fluctuations of Whole-Cell Currents Reveal Multiple Gating Modes of Chloride Channels

PDS, whether calculated from current flow in cellattached (Fig. 14 A) or excised outside-out (Fig. 14 B) patches, shows sufficient similarity with that of the whole-cell current fluctuations to allow the conclusion that whole-cell currents and their fluctuations about their stationary mean value--over the entire frequency range covered in this study--are generated by the electrical activity of chloride channels. This said, it is equally important to note that the very long o p e n states exemplified in Fig. 13 A might well contribute to additional power below the frequency range we are able to study. T h e lower frequency limit for spectral analysis of whole-cell current fluctuations is defined by the slowly developing undesired current drift (e.g., Conti et al., 1980; H e l m a n and Kizer, 1990). An investigation of the possible contribution of the very slow gating to the macroscopic PDS would require sampiing of currents for 20 min or more. We could not acquire stationary whole.cell currents for such a long period of time.


+80 mV +20 mV -30 mV

c ,~'~-~-.":~'~:~. "~'~,'--'%~';:';Yr 7:: ~.',,:.,~--.:'~_'.,:,~9

+60 mY +10 mV

- 4 0 mV

mV

+50

0

-50

mV

-60

mV

-70

mV

mV

-10 mV

+30 mV

-20 mV

Fmul~ 17. Analysis of I /CI- selectivityof conductance levels resolved in the excised outside-out patch of Fig. 16. Current recordings at 15 pipette potentials covering the range from +80 to - 7 0 mV with Vvindicated above each trace (the trace shown for Vp = +50 mV is the same as the one in Fig. 16, upper-left. [I ]B = 100 mM; [C1-]B = 50 mM; [I-]p = 10 raM; [C1-]p = 40 raM. tion, r e c t i f i c a t i o n s w i t c h e d f r o m o u t w a r d (Figs. 5 a n d 13) to i n w a r d w h e n e x t e r n a l [C1-] was m a d e signific a n t l y l o w e r t h a n t h e e s t i m a t e d i n t e r n a l [C1-] (Fig. 7). Like t h e m a c r o s c o p i c m e m b r a n e c u r r e n t s (Fig. 2), t h e i/V r e l a t i o n s h i p b e c a m e linear, as p r e d i c t e d , for symm e t r i c a l C1- c o n c e n t r a t i o n s (Fig. 9). A useful c o n s e -

an intracellular c h l o r i d e c o n c e n t r a t i o n o f ~ 7 5 m M (Benn e k o u a n d S t a m p e , 1988; P a r k e r a n d D u n h a m , 1989). Rectification and permeability. I n d e p e n d e n t o f m e m brane patch configuration, open-channel currents were well d e s c r i b e d by t h e G H K e q u a t i o n b o t h for inw a r d a n d o u t w a r d c u r r e n t s . I n a g r e e m e n t vdth this no-

A

B 30

pS 22.1

[]

. i(pA)

173

o

~

-

0= 20

13.4 10.6

=

O

7.1

=o

*

3.7

~= lO o

0 I

,

-100

,

,

J

.

.

9

.

I-1

tJ

Vp (mY)

~

=

0 1

0.993

,

,

,

,

,

2

3

4

5

6

Conductance

712

,

Lever

Endogenous C1 Channels of Sfl9 Cells

,

FIGURE 18. Analysis of I-/C1- selectivity of conductance levels resolved in the excised outsideout patch of Figs. 16 and 17. (A) The patch exhibited at least six different conductance levels. The associated current-voltage relationships were near-linear with a common reversal potential far more left than that obtained with Br-/C1- as halide ion pair (Fig. 15 B), indicating a strong preference for iodide over chloride and bromide. (Inset) Slope conductances obtained by linear regression analyses. (B) The slopes of individual current-voltage relationships depicted a straight line as a function of conductance level with unitary conductance " ~ s l o p e = 3.6 pS.


+40 mV


tudes, i n d e p e n d e n t of absolute value, could be resolved into an integer n u m b e r of an elementary channel unit, it is m o r e likely that several channel molecules contributed to the C1- current flow, with the n u m b e r of electrically active channels varying f r o m patch to patch and m o m e n t to m o m e n t . T h e n u m b e r of active units could be quite large; e.g., six levels were resolved in Figs. 8 and 18, a n d 10 levels were calculated f r o m the permeability coefficient of the largest channel studied (Fig. 12). Mazzanti et al. (1991) p r o p o s e d a m o d e l accounting for Ca 2+ c u r r e n t - d e p e n d e n t block of L-type Ca 2+ channels that assumed that calcium channels occur in clusters sharing a s u b m e m b r a n e Ca z+ pool buildup by inflowing Ca 2+ (see also DeFelice, 1993). Cell surface transport studies of m u r i n e glycoproteins (Holifield and Jacobson, 1991), of the low density lipoprotein receptor (Ghosh and Webb, 1993), of e p i d e r m a l growth factor receptor and E-cadherin (Kusumi et al. 1993), and of the receptors for transferrin and az-macroglobulin (Sako and Kusumi, 1994) have indicated that these integral plasma m e m b r a n e proteins are also confined to small m e m b r a n e domains, e.g., governed by interactions with the cytoskeletal network adjacent to the m e m b r a n e (Kusumi et al., 1993), rather than having a r a n d o m horizontal distribution on the surface of the m e m b r a n e . T h e results obtained in the present study with insect cells in culture indicate that (a) the plasma m e m b r a n e chloride c o n d u c t a n c e consists of elementary chloride channel units, and that (b) single events of amplitudes larger than the elementary current flow result f r o m c o o r d i n a t e d o p e n / c l o s e transitions of two or m o r e elementary channel units. T h e finding that the n u m b e r of units exhibiting simultaneous o p e n / close activity varies f r o m patch to patch, and f r o m mom e n t to m o m e n t in the same patch, would be consistent with the o c c u r r e n c e of channel units in m e m b r a n e domains allowing for coordinated activity a m o n g two or m o r e channel units.

This work was supported by National Institutes of Health grants HL34322 a n d P30DK34987 a n d by Danish National Science Research Council grants 110017, 119270, a n d 110971.

Original version received 27June 1995 and accepted version received 2 7 February 1996.

REFERENCES Anderson, C.R., and C.F. Stevens. 1973. Voltage clamp analysis of acetylcholine-produced end-plate current fluctuations at frog neuromuscularjunction.J. Physiol. (Camb.). 236:655-691. Bear, C.E., C. Li, N. Kartner, R.J. Bridges, T.J. Jensen, M. Ramjoesingh, andJ.R. Riordan. 1992. Purification and functional reconstitution of the cystic fibrosis transmembrane conductance regulator (CFTR). Cell. 68:809-818. 713

LARSEN ET AL.

Bennekou, P., and P. Stampe. 1988. The effect ofATP, intracellular calcium and the anion exchange inhibitor DIDS on conductive anion fluxes across the h u m a n red cell membrane. Biochim. Biophys. Acta. 1030:18.%187. Birnir, B., M.L. Tierney, S.M. Howitt, G.B. Cox, and P.W. Gage. 1992. A combination of h u m a n cd and [31 subunits is required for formation of detectable GABA-activated chloride channels in


quence of this consistent ion p e r m e a t i o n m o d e is that one can calculate an intrinsic c o n d u c t a n c e property, such as permeability or limiting c o n d u c t a n c e (~h50/1~0), to c o m p a r e channels resolved in different configurations, bathing solutions, or range of potentials. Chloride channels resolved in cell-attached as well as in excised patches exhibited several conductance levels (e.g., Figs. 6, 8, 10, 11, 16, and 17), indicating a hallm a r k of the e n d o g e n o u s chloride ion p e r m e a t i o n system in Sf9 cells. Thus, permeabilities obtained f r o m single-channel i/V curves r a n g e d f r o m N0.5 9 10 -14 cm3/s to N5.5 9 10 -14 cm3/s in cell-attached patches as well as in excised outside-out patches with the associated limiting c o n d u c t a n c e levels spanning f r o m 3-4 pS to ~ 3 5 pS. Interrelationship between conductance levels. With the driving force for current flow held constant, currents exhibited transitions of different amplitudes with large as well as smaller c u r r e n t transitions occurring from, a n d to, the closed level, as well as f r o m o n e o p e n level to ano t h e r (Figs. 10, 16, and 17). Different c o n d u c t a n c e levels reversed n e a r the same potential (Figs. 6 and 12) a n d exhibited a halide ion p e r m e a t i o n sequence of PI > Pcl -~ PB~ = 2.0:1:0.9 (Figs. 15 and 18 A). In o t h e r words, the selectivity is c o m m o n for all conductance levels studied. F u r t h e r m o r e , the smallest c o n d u c t a n c e resolved was similar in cell-attached a n d excised outside-out patches, i.e., ~/1~0/150 = ~ 3 . 5 pS. These observations lead to the question of whether different conductance levels are associated with j u m p s f r o m a main conductance state to substates and superstates of a singlechannel molecule or are caused by simultaneous activity of several channel molecules. If a j u m p f r o m one c o n d u c t a n c e level to a n o t h e r is associated with a transition of a single-channel molecule, one would have to postulate six or m o r e states of the conducting p o r e (Figs. 12 a n d 18), with every state separated f r o m its two n e i g h b o r i n g states by exactly the same distance (Fig. 18 B). Thus, because individual current ampli-


St'9 cells. Proc. R. Soc. Lond. Ser. B Biol. Sci. 250:307-312. Christensen, O., and N. Bindslev. 1982. Fluctuation analysis of short-circuit current in a warm-blooded sodium-retaining epithelium: site, current, density, and interaction with triamterene. J. Memb. Biol. 65:19-30. Colquhoun, D., and A.G. Hawkes. 1977. Relaxation and fluctuations of m e m b r a n e currents that flow through drug-operated ion channels. Proc. R. Soc. Lond. Ser. B Biol. Sci. 199:231-262. Conti, F., B. Neumcke, W. Nonner, and R. St~mpfli. 1980. Conductance fluctuations from the inactivation process of sodium channels in myelinated nerve fibres.J. Physiol. (Camb.). 308:217-239. DeFelice, LJ. 1981. Introduction to Membrane Noise. Plenum Press, New York and London. 500 pp. DeFelice, LJ. 1993. Molecular and biophysical view of the Ca channel: a hypothesis regarding oligomeric structure, channel clustering, and macroscopic current.J. Membr. Biol. 133:191-202. Finkelstein, A., and A. Mauro. 1963. Equivalent circuits as related to ionic systems. Biophys.J. 3:215-237. Fisher, H., and T.E. Machen. 1994. CFTR displays voltage dependence and two gating modes during stimulation. J. Gen. Physiol.

714

Endogenous C1- Channels of Sf9 Cells


104:541-566.

Fishman, H.M. 1973. Relaxation spectra of potassium channel noise from squid axon membranes. Proc. Nat. Acad. Sci. USA. 70: 876-879. Gabriel, S.E., E.M, Price, R.C. Boucher, and M.J. Stutts. 1992. Small linear chloride channels are endogenous to nonepithelial cells. Am. J. of Physiol. (Cell Physiol.). 263:C708-C713. Gadsby, D.C., G. Nagel, and T.-C. Hwang. 1995. The C ~ R chloride channel of mammalian heart. Annu. Rev. Physiol. 57:387-416. Ghosh, R.N., and W.W. Webb. 1993. Automated detection and tracking of individual and clustered cell surface low density lipoprotein receptor molecules. Biophys. J. 66:1301-1318. Goldman, D.E. 1944. Potential, impedance, and rectification in membranes.J. Gen. Physiol. 27:37-60. Hamill, O.P., A. Marty, E. Neher, B. Sakmann, and F.J. Sigworth. 1981. Improved patch-clamp techniques for high resolution current recording from cells and cell-free m e m b r a n e patches. Pfliigers Arch. Eur. J. Physiol. 391:85-100. Helman, S.I., and N.K. Kizer. 1990. Apical sodium ion channels of tight epithelia as viewed from the perspective of noise analysis. Curt. Top, Membr. 37:117-155. Hodgkin, A.L., and B. Katz. 1949. The effect of sodium ions on the electrical activity of the giant axon of the squid. J. Physiol. (Camb.). 108:37-77. Holifield, B.F., and K. Jacobson. 1991. Mapping trajectories of Pgp-1 membrane protein patches on surfaces of motile fibroblasts reveals a distinct boundary separating capping on the lamella and forward transport on retracting tail. J. Cell Sci. 98: 191-203. Karmer, N.,J.W. Hanrahan, TJ.Jensen, A.L. Naismith, S. Sun, C.A. Ackerley, E.F. Reyes, L.-C. Tsui, J.M. Rommens, C.E. Bear, and J.R. Riordan. 1991. Expression of the cystic fibrosis gene in nonepithelial invertebrate cells produces a regulated anion conductance. Cell. 64:681-691. Klaiber, K., N. Williams, T.M. Roberts, D.M. Papazian, L.Y.Jan, and C. Miller. 1990. Functional expression ofShakerK + channels in a Baculovirus-infected insect cell line. Neuron. 5:221-226.

Kusumi, A., Y. Sako, and M. Yamamoto. 1993. Confined lateral diffusion of m e m b r a n e receptors as studied by single particle tracking (nanovid microscopy). Effects of calcium-induced differentiation in cultured epithelial cells. Biophys. J. 65:2021-2040. Larsen, E.H., J. Fullton, M.J. Stutts, R.C. Boucher, and E.M. Price. 1992. Kinetics of CI- currents of Sf9 cells infected with h u m a n CFI'R. Pediatr. Pulmonol. 267(Suppl. 8):119. (Abstr.) Larsen, E.H., S.E. Gabriel, J. Fullton, R.C. Boucher, E. Price, and M.J. Stutts. 1993. Multiple single channel conductances and gating time constants of CFTR-mediated chloride currents. FASEBJ. 7:A426:2468. (Abstr.) Larsen, E.H., and B. Harvey. 1994. Chloride currents of single mitochondria-rich cells of toad skin epithelium. J. Physiol. (Camb.). 478:7-15. Lindemann, B. 1984. Analysis of additively contaminated Lorentzians by integration. Biophys. J. 46:409-411. Lindemann, B., and W. Van Driessche. 1978. The mechanism of Na-uptake through Na-selective channels in the epithelium of frog skin. In Membrane Transport Processes. Vol. 1. J.F. Hoffman, editor. Raven, NewYork. 155-178. Marty, A., and E. Neher. 1995. Tight-seal whole-cell recording. In Single Channel Recording. B. Sakmann and E. Neher, editors. Plenum Press, New York. 31-52. Mazzanti, M., L.J. DeFelice, and Y.-M. Liu. 1991. Gating of L-type Ca 2+ channels in embryonic chick ventricle cells: dependence on voltage, current and channel density.J. Physiol. (Camb.). 443:307334. Neher, E. 1991. Correction for liquid junction potentials. Methods Enzymol. 207:123-131. Neher, E., and C.F. Stevens. 1977. Conductance fluctuations and ionic pores in membranes. Annu. Rev. Biophys. Bioeng. 6:345-381. Nielsen, C., L. Thomsen, M. Nielsen, S.E. Westh-Hansen, R.M. Witt, P.B. Rasmussen, S. Hastrup, and E.H. Larsen. 1995. Multiple versus single conductance states in ~313272sand c~3132members of the GABAA receptor family. FASEBJ. 9:A390; 2258. (Abstr.) Parker, J.C., and P.B. Dunham. 1989. Passive cation transport In Red Blood Cell Membranes. P. Agre and J.C. Parker, editors. Marcel Dekker, Inc., NewYork. 507-561. Poussart, DJ.M. 1971. Membrane current noise in lobster axon under voltage clamp. Biophys. J. 11:211-234. Sako, Y., and A. Kusumi. 1994. Compartmentalized structure of the plasma m e m b r a n e for receptor movements as revealed by nanometer-level motion analysis. J. Cell Biol. 125:1251-1264. Sarkadi, B., D. Bauzon, W.R. Huckle, H.S. Earp, A. Berry, H. Sushindran, E.M. Price, J.C. Olsen, R.C. Boucher, and G.A. Scarborough. 1992. Biochemical characterization of the cystic fibrosis transmembrane conductance regulator in normal and cystic fibrosis epithelial cells.J. Biol. Chem. 267:2987-2095. Sten-Knudsen, O. 1978. Passive transport processes. In Membrane Transport in Biology. Concepts and Models. G. Giebisch, D.C. Tosteson, and H.H. Ussing, editors. Springer-Verlag, New York. 5-113. Stevens, C.F. 1972. Inferences about membrane properties from electrical noise measurements. Biophys. J. 12:1028-1047. Van Driessche, W., and W. Zeiske. 1985. Ion channels in epithelial membranes. Physiol. Rev. 65:833-903.