Pflfigers Arch (1992) 421:473-485

Eu""l]ihJoumal of Physiology 9 Springer-Verlag1992

Nuclear ion channels in cardiac myocytes J. Omar Bustamante University of Ottawa Heart Institute and Department of Physiology, 1053 Carling Avenue, Ottawa, Ontario, Canada K1Y 4E9 Received March 9, 1992/Received after revision April 22, 1992/Accepted May 5, 1992

Abstract. The paradigm that nucleocytoplasmic transport of ions occurs without a diffusional barrier has been challenged by the recent demonstration with patch-clamp techniques of the existence of ion channels in the nuclear envelope of murine zygotes and hepatocytes. This report demonstrates the existence of nuclear ion channels (NIC) in murine ventricular cardiac myocytes. N I C conductance (7), calculated from current histogram peaks, was 1 0 6 - 532 pS at 2 2 - 36 ~C. In nucleus-attached patches, replacement of cytoplasmic K + with Na + reduced N I C activity within 30 s, suggesting that intranuclear-delimited mechanisms mediate this phenomenon. In excised, inside-out patches K + was as permeable as Na + through NIC. N I C activity was observed in 0 - 4 m M Mg 2+ and/or ATP 2-, with or without 0 - 1 m M Ca 2 +, indicating a minor direct role of these ions. However, in non-responsive excised inside-out patches, NIC activity appeared when the catalytic subunit of the cAMP-dependent protein kinase was applied to the nucleoplasmic side of the patch, in the presence of Mg 2+ and ATP 2-, indicating an important role for phosphorylation-dependent process(es) in NIC function - an observation supported by the depressing effects of protein kinase inhibitor on responsive NIC. The concept that nucleopore complexes are solely responsible for nucleocytoplamic transport leads to the speculation that these structures are the physical substrate for NIC.

of 10 nm, nucleopores are thought to offer a negligible diffusional barrier to the passage of ions in and out of the nucleus. Therefore, it is without surprise that not until recently little attention was paid to the transport of ions across the nuclear envelope [25, 26, 35]. Clearly, the availability of patch-clamp techniques [20] has made it possible to test these concepts. Since the hypothesis of the existence of nuclear ion channels (NIC) in cardiac myocytes is relevant to the understanding of normal cardiac structure and function as well as to the genesis of cardiac pathologies, it was deemed important and timely to test this possibility. The investigations described below provide the first demonstration of the existence of spontaneously active and voltage-gated NIC in cardiac myocytes. With properties resembling those of the gap-junction ion channels (e. g. [39, 40]), which also constitute pores through double membranes, N I C should modulate ion transport across the nuclear envelope of cardiac myocytes. Consequently, NIC should control, indirectly, intranuclear processes (e.g. transcription) responsible for the structure and function of these cells and, therefore, may play a role in the mechanisms underlying such important phenomena as hypertension and hypertrophy (e. g. [22]).

Key words: Nuclear ion channels - Cardiac myocytes Channel conductance - Patch clamp - Nuclear pore complexes - Nuclear envelope - Nuclear membrane

Nuclei isolation. A simple and rapid technique for nuclei isolation was developed following principles described elsewhere [12, 24]. A total of 29 male Swiss-Webstermice (20- 22 g, Charles River Labs./ Bausch & Lomb Inc., Montr6al, Qu6bec) were sacrificed according to the guidelines of the Canada Council on Animal Care. The hearts were extracted, placed rapidly in a solution (raM: 150 NaC1, 2 MgC12, 5 HEPES and 2.5 KOH; room temperature: 22-24~ and cleaned as reported previously [9]. After 5 min of retrograde perfusion, the ventricles were separated and minced with fine scissors [9]. The pieces of tissue were then placed in ice-cold solution (mM : I35 KC1, 2 MgC12, 5 EGTA, 5 HEPES and 17.5 KOH) kept in a 7-ml Dounce manual tissue grinder (Wheaton-33 low extractable borosilicate glass, Wheaton, Millville, N. J.). After 5 rain stabilization in this high-K, EGTA solution, tissue grinding proceeded with the loose-fitting pestle in four to eight strokes. Nuclei isolation was monitored by a high-resolution microscope [8]. Tissue homogenates

Introduction Current concepts, based on cell- and molecular-biology techniques, depict nuclear pore complexes as being solely responsible for traffic across the nuclear envelope [4, 6, 16, 29, 30, 31]. With an effective diameter of the order

Materials and methods

474

i(pA)

A

Vm=

+20

I00

mV, 36aC

Counts C

100

I0000

s-Histogram

5000

50

0 i

t

0

40

0

: 4.0 I00 s,

40

i 80

i(pA)

i (pA)

B

i 40

t

80

D

P (n) Binomial

N = 235

0.22

Fit

N= 14 = 0.5 20

t 0.0

t

L

p

0.5

1.0

0

t

0

(S)

L

t

10

80

f:/00

T~

Fig. 1 A - D . Spontaneous nuclear ion channel (NIC) activity. A Compressed, 80-s recording from a patch attached to the nucleus of a mouse ventricular cardiac myocyte. Sampling interval of 2 ms. The pipette and bath contained the high-K solution at 36~ The pipette potential was set to - 2 0 inV. As the resting potential was approximately 0 mV, the membrane potential, Vm, was approximated to the negative of the pipette potential or + 20 mV. B Expanded, l-s recordings obtained from the samples of the current record shown in A demonstrate the current jumps, Ai, corresponding to NIC openings. A total of 235 Ai values, measured over a period

of 100 s, gave an average outward current jump, (Ai}, of 4.0 pA. C Raw current histogram accumulated over a period of 100 s. D Normalized histogram and binomial fit. By making the assumption that one (Ai} corresponds to one channel state transition and then dividing i in C by (Ai}, one transforms the i axis in C into an n axis in D, where n is the number of channels open. By forcing the area of the histogram to 1.0, one gets the probability that n channels are open, P(n). The envelope corresponds to the binomial fit with a total number of channels, N, of 14 and an open probability per channel, p, of 0.5

containing the isolated nuclei were kept in the high-K, EGTA solution at 2 - 4 ~ for up to 4 h. The nuclei readily attached to the glass bottom of the 1-ml experimental chamber [9] and could be cleaned from debris by chamber perfusion. Each patch experiment lasted between 30 min and 300 min.

Stratford, Conn.) and with excitation and emission filters of 520 nm and 590 nm respectively (Omega Optical Inc., Brattleboro, Vt.). Ethidium-bromide-treated nuclei were not utilized in patch-clamp experiments.

Electrophysiological procedures. Experiments were conducted at Solutions. Solutions were prepared with AnalaR-grade reagents (BDH Chemicals Inc., Toronto, Ontario) and with type-I-reagentgrade water ( > 18.2 MQ cm) obtained "on demand" from a water purification system (Milli-Q-PLUS, Millipore Corp., Bedford, Mass.) that had a 0.22-gin filter unit (Millistak-GS, Millipore). All solutions had a pH of 7 . 2 - 7 . 4 and a PO2 above 150mm Hg (20 kPa), and were freshly prepared in Teflon ware (Nalgene, Nalge Co., Rochester, N. Y.) to minimize contamination. Likewise, pipettes were filled via Teflon syringes (Chromatographic Specialties Inc., Brockville, Ontario; see, for example, [28]) and filters (Nalgene). The control solution was a high-K solution consisting of (raM): 150 KC1, 2 MgC12, 5 HEPES, 2.5 KOH. A high-Na solution (raM: 150 NaC1, 2 MgC12, 5 HEPES, 2.5 NaOH) was also utilized to test the effects of Na + on channel activity. Solutions containing ATP and/or protein kinase A (PKA; catalytic subunit, Sigma) or crude PKA inhibitor (rabbit muscle, Sigma) were used within up to 30 min of preparation in control, high-K solution (i.e. not from stored stock solutions). Temperature-sensitive chemicals were transported in solid CO2 and stored in a constant-temperature, nondefrosting refrigerator/freezer (Isotemp, Fisher Scientific, Ottawa, Ontario). Solution exchange in the experimental chamber was estimated visually with dyes and electrically via measurements of electrode polarization. Complete exchange was achieved within 5 s. This time was taken into account when assessing the effects of different manoeuvres on NIC activity.

Fluorescence microscopy. Isolated nuclei were identified by their fluorescence at 520 nm excitation wavelength when treated with 1 Ixg/ml DNA marker ethidium bromide (Sigma Chemical Co., St. Louis, Mo.). These determinations were carried out in a preliminary series of experiments with a custom microscope [8] equipped with a 200-W stabilized arc lamp system (Photomax, Oriel Corp.,

2 0 - 3 8 ~ to determine the effects of temperature on NIC activity. The bath temperature was controlled with a water bath (PE-2, Haake, Mess-Technik GmbH & Co., Karlsruhe, FRG) customized to eliminate electromagnetic interference. Temperature was monitored with a thermistor (YS144004, Yellow Springs Instrument Co., Yellow Springs, Ohio) the resistance of which was measured and translated by a computer-based data-acquisition system [7]. Standard patch-clamp techniques [20] were utilized throughout the investigations. Recording pipettes were made from hard borosilicate glass (7052, Corning Glass Co., Corning, N. Y.). The 5-mm-shank pipettes were fabricated from glass capillaries (0.5 mm wall thickness, prepared by Garner Glass Co., Claremont, Calif.) in a twostage puller (PB-7 Narishige Co., Tokyo, Japan). The pipettes were heat-polished to an external diameter of 1 - 1.6 ~tm (0.2 ~tm resolution, see [8]) and had a resistance of 4 - 6 M~2. Pipettes were electrically conntected to the input headstage of a patch-clamp instrument (EPC-7, List Medical, Darmstadt/Eberstadt, FRG). However, to maximize the mechnical stability of the preparation, the pipette holder (EPC-7 accessory) was separated from the input connector of the headstage and firmly attached to the micromanipulator system. The micromanipulator system consisted of a piezo-electric block (P-280.3-Physik Instrumente GmbH & Co., Waldbronn, FRG) mounted on a standard micromanipulator (Ernst Leitz Wetzlar GmbH, Wetzlar, FRG). Further mechanical stability was achieved by mounting the chamber and micromanipulators onto a large microscope stage [8]. The experimental arrangement rested on a vibration-isolation table (Vibraplane 1201, Kinetic Systems Inc., Roslindale, Mass.). For experiments involving more than just resting potential measurements, electrode polarization was compensated for and accepted as stable if the drift in polarization was less than i mV in 15 rain (aided with overnight equilibration of the system with control solution and salt bridges made of Ag/AgC1 half-

475 cells: RC-2, World Precision Instruments Inc., New Haven, Conn., see [9]). Analog signals were continuously displayed on the screen of a conventional oscilloscope (model 2225, Tektronix, Beaverton, Ore.). The signals were filtered at 1 kHz and amplified (CyberAmp Programmable Signal Conditioner equipped with an eight-pole Bessel filter, Axon Instruments, Foster City, Calif.) prior to digitization and storage by a fast data-acquisition system [7]. Experiments were also stored on magnetic video tape with the aid of a converter (VR-10, Instrutech Corp., Elmont, N.Y.) for off-line analysis. Membrane resting potentials were determined under the zero-current mode of the EPC-7 (equivalent to the voltage-follower configuration of a standard electrometer). This procedure involved formation of a gigaseal patch with a pipette filled with high-K solution. In experiments aimed at measuring only resting potentials, stability of electrode polarization was accepted if it was less than 0.5 mV in 5 min. After gigaseal formation, the patch was perforated (to achieve contact with the nucleoplasm) with a l-s suction pulse of - 70 mm Hg ( - 9 kDa). The sudden change in membrane potential detected within the first 0.1 s was taken as the resting potential. This value was utilized to calculate the actual transmembrane potential imposed on the membrane patch during current recording in nucleus-attached patches (i. e. resting potential minus pipette potential). Single-channel activity was analysed with the aid of custom-built [7] and commercial software (pClamp, Axon). In some experiments, during offline analysis, when slow transition rates and long dwell times were observed, the signals were further filtered by software (50-100 Hz, Gauss filter, pClamp). This extra filtering considerably aided NIC analysis. The illustrations given here were obtained with 100-Hz Gauss filter.

Counts

+82 pA

500

A

Vm= +40 mV +1o2 pA

250

O

+40 pA ~

pA

+62

-0 i

i

o

75 -2,5

150

pA

400

o

-71 pA Vm= - 4 0 mV o -47 pA

200

i

i

-75

-15o

i(pa) A ~ (pA)

Results

30

B

General characteristics of the preparation L i g h t - m i c r o s c o p y observations revealed that the isolated nuclei fluoresced readily w h e n treated with ethidium bromide ( n o t shown). A t the resolution o f the light microscope used ( ~ 0.2 tam), the isolated nuclei seldom showed remains f r o m the endoplasmic reticulum. Further, these nuclei always had s h a r p - c o n t r a s t and s m o o t h contours. The c o n t o u r s were rugged, however, w h e n the tight-fitting pestle o f the tissue-grinder was used for isolation in preliminary tests (results f r o m these experiments are n o t included here). The nuclei h a d ellipsoidal shape with m a j o r and m i n o r axes o f (mean _+ SD): 13.6 + 3.0 ~tm and 5.2 _+ 1.1 ~tm respectively ( N = 15). The experimental results reported here were o b t a i n e d f r o m nuclei displaying m a j o r k n o w n features (e. g. nucleoli) and devoid o f endoplasmic reticulum remains. N u c l e a r patches with resistances a b o v e 1 Gf2 (i. e. gigaseals) f o r m e d readily on c o n t a c t o f the p a t c h pipette with the nuclear envelope ( ~ 80% success rate versus 100% in nuclei isolated f r o m hepatocytes - n o t shown). Resting potentials yielded a negligible value: 0.8 • 1.3 m V (mean _+ SD, N = 15).

Spontaneous activity Figure I A - D illustrates the s p o n t a n e o u s activity o f a nucleus-attached patch when the h i g h - K resolution was used in b o t h the pipette and bath. Since, u n d e r these conditions, the m e a s u r e d resting potential was negligible it was considered reasonable to assume that the nuclear

~

4

~/~

pS N = 236 R = 0.996

-30 -60

0

60

vm (mv) Fig. 2A, B. Current/amplitude histograms and current/voltage relation for a patch containing spontaneously active NIC. Control conditions as in Fig. 1. Temperature was 33.2~C. A Amplitude histograms, counts versus i, constructed from selected segments of current records collectetd at two different values of membrane potentials, Vm (top: + 40 mV, bottom: - 4 0 mV). B Single-channel current, Ai, versus Vmplot for the whole range of Vmtested. O, Mean values; vertical error bars, the standard deviation of the data for each Vm. The total number of values pooled, iV, was 236 (excluding those for Vm = 0 mV). The high value of the regression coefficient, R, demonstrates the linearity of the conductance, 7

m e m b r a n e potential (Vm) was approximately the negative o f the pipette potential. Figure 1 A, B shows, respectively, compressed (80 s) and e x p a n d e d (1 s) current (/) recordings. Figure 1 C shows the current histogram (counts vs /) a c c u m u l a t e d during 100 s. The h i s t o g r a m displays nine conspicuous peaks spread u n i f o r m l y along the horizontal current axis, i. Figure 1 D shows the normalized histo-

476 A

Counts

~(pA)

i

I

I

0.0

1.5

3.0

L_140

B

I

-- 50

,tJl

f

,,,,.,,.,lib .....i, .......ItI o I

140

170

200

f

200

]

I

[

I

]

I

0,0

0.5

1.0

140

170

200

200

Fig. 3A, B. NIC subconductance

f

i,o

, ,h,,,,,J. ,..a..~

........

i

I

I

I

I

1.0

1.5

2.0

140

170

200

170 2.0

2,5 L

t(s)

3.0 [

30

o

1

I

30

15

140

,illJl'[l~[} . . . . . . . . . . . . . . . , I 140

gram and the closest binomial fit (e. g. [13, 21]), the solidline envelope, drawn according to the formula: N~ P(n) - n !(N - n) !p" (1 _p)N - , where p is the probability that a unit opens, n is the number of units open, P(n) is the probability of finding n units simultaneously open and N is the total number of functional units. This fit suggests a total number of functional units or channels, N, of 14. Note that the binomial assumption provided a different value for the number of functional channels than that estimated by counting the m a x i m u m number of current j u m p s (Ai). The latter yielded a value of N = 12, suggesting that the record was not long enough to capture the event at which all the channels were simultaneously open [because of the low probability of this event, i.e. when N = 14, P(14) = 0.00006 and P(12) = 0.0055]. Also note that: (a) a correction of 1 unit m a y be needed should the first current level correspond to the closed state, and (b) the fact that P(n) ~ 0 for n < 3 indicates that there are two channels permanently open. The symmetry of the histogram

[11]'iii ll'l~lll

I

I

170

200

0

states. A Current records from an experiment with an attached patch at 37.4~C and with high-K solution in both the pipette and bath. The membrane potential, Vm,was about § 40 mV. The top diagram shows the full record, obtained from 0 to 3 s. The lower diagrams show the same record with an expanded time scale for 0 - 1, 1 - 2, and 2 - 3 s. B Amplitude histograms obtained from the records in A. Each histogram corresponds to the record on the left. The vertical arrowheads indicate the putative subconductance states. Current level was negligible at Vm=0

i(pA)

speaks in favour of a probability of opening per single channel, p, of a b o u t 0.5. In the experiment of Fig. 1 (at 35.9~ the average outward current jump, ( A i ) , was 3.95 pA (SD = 0.34 pA, N - - 235) when Vm = + 20 mV. The conductance, 7, was determined by linear regression of the A i corresponding to the histogram peaks, and was 198 pS (regression coefficient, R = 0.998). A similar analysis was conducted for each experiment. Figure 2A, B gives an example of the analysis followed to determine the current peaks and the current jumps in order calculate ? for a patch containing spontaneously active N I C . Figure 2A provides the raw histograms (open circles) and their corresponding Gaussian fit (solid lines) for two different m e m b r a n e voltages, Vm, of + 40 mV and - 40 mV. Figure 2 B gives the plot of A i versus Vm for the complete experiment. The plot in Fig. 2B demonstrates the linearity of N I C conductance under control conditions. Under these experimental conditions, measurements from another nucleus-attached patch yielded a quite different value of ? equal to 445 pS (R = 0.997, N = 26). At 22-24~ 7 = 132 + 32 pS (mean_+ SD, range: 9 9 198 pS) in 15 out of the 22 spontaneously active patches where current levels were not contaminated with open-

A

/(pA) 80

channel noise (which would, in turn, cause underestimation of current and, therefore, 7). Patch excision was difficult in most instances as the nuclei adhered firmly to the recording glass pipette. However, in one of such experiments, where excision was possible, the process did not affect 7, suggesting little intranuclear control of this variable under these conditions. In another experiment, a Q1o of 3.3 was estimated for 7, supporting the idea that the apparent temperature independence of 7 (i. e. 198 pS at 22~ and 36~ in two separate patches) results from a change in the probability of finding n channels open, P(n), which, in turn, depends on the probability, p, of a NIC being open. Close inspection of the current records revealed (in 9 patches at 22-37~ less frequent and smaller Ai values, as the current records and histograms of Fig. 3 A, B illustrate (putative substates indicated with downward arrowheads). Similar small Ai values were reported previously as substates [25]. Figure 4 A - E illustrates the effects of exchanging the bath solution from high K to high Na on NIC activity from a nucleusattached patch. As is evident from Fig. 4 A - E , not only are 7 and p depressed by the increased cytoplasmic (extranuclear) Na § concentration but so also is the number of functional channels, N (see Fig. 4 D). In three independent experiments designed to determine the effects of Na § on resting potentials (i.e. to assess properly the effects of Na § on 7), switching from high-K to high-Na solution caused a resting potential shift of approximately + 3 mV within the time span of the test illustrated in Fig. 4 A - D . These Na+-induced changes in resting potential parallel the small changes observed in reversal potentials (inside-out excised patches), as Fig. 4E shows. Note, however, that the permeability ratio (i. e. PK/PNa) calculated from reversal potentials under conditions such as the present one (e. g. [21]) does not match the reduction by almost one-half in the patch conductance, suggesting that more complex mechanisms are involved or that the slope conductance takes into account a non-NIC component such as a leak pathway. Also note that NIC activity did not disappear in the inside-out, excised patches, an interesting contrast with the results obtained in nucleus-attached patches. Since these tests were conducted in excised patches, one may conclude that the discordant results may be due to the complex nature of NIC gating and selectivity filter (e.g. [21]). Whatever mechanisms are involved, it is clear that the resting potential change does not account for the reduction in 7 (from 150 pS to 54 pS), p (from 0.85 to 0.65), and N (from 11 to 2) associated to the Na depression of NIC activity observed in the nucleus-attached patch illustrated in Fig. 4 A - D (and 2 other attached patches). The relatively long time taken for this effect to reach a steadystate points towards a mechanism involving intranuclear pathways. Further investigations must be conducted to clarify this point. Similar ion substitution in 4 excised, inside-out patches produced a change in patch "resting" or zero-current potential of about 6 mV within 5 min, leading to an estimate of the "passive" (i. e. when NIC were closed) permeability ratio for K + over Na § PK/ PN,, of around 1.3. Note that NIC (i. e. their opening) per se did not contribute to this change in resting potential;

A t t a c h e d P a t c h , V m = + 4 0 mV, T= 22~ P i p e t t e S o l u t i o n : 150 KC1 + 2 MgC12

c

0

115o KCll

lao NaCl

[

I

0

60

r I

120 i ( p A )

R

150 KC1, < A 4 > =

6.0

F

pA

[

] 15

0

C

~50 Nacl, < A~ >= 2.0 pA

r 105

2

t (s)

Dcounts 800

40 F 40

L

0

lao KCI-120 s N=ll, p=0.85

NaC1-30 s N=2, p=0.65

80

Counts F 9000

L1450~

150 t 0

L0

i

i

35

70

i (pA)

E

i(pA) 3O

J l -40

k 0

--,o l 40

Vm(mV) Fig. 4 A - E . Modification of spontaneous NIC activity upon replacement of cytoplasmic or nucleoplasmic K + with Na +. A Compressed, 120-s recording from a nucleus-attached patch showing a reduction in NIC activity conductance upon replacement of K + with Na +. Sampling interval of 2 ms. High-K solution in the pipette. Experiment conducted at 22 ~C. B Expanded recording constructed with the samples used in the first 15 s of the record shown in A. The record illustrates the activity during bath perfusion with high-K solution. C Expanded recording corresponding to the samples for the last 15 s of the record shown in A and demonstrating the depression of NIC activity upon ion replacement by bath perfusion with high-Na solution. Note that was reduced from 6.0 pA to 2.0 pA (a decrease in 7 from i50 pS to 54 pS). D Histograms accumulated during 120-s perfusion with high-K solution (right) and during 30 s perfusion with high-Na solution (left). Histogram envelopes were drawn according to the best binomial fit (scaled to match the raw histograms) and suggest a reduction in the number, N, of functional NIC and the probability, p, that a NIC opens. E Experiment conducted to assess the direct effects of intranuclear Na + on NIC activity. High-K solution in the pipette. Current signals elicited with 2-s voltage ramps and obtained from an excised, insideout patch under high-K (control) and after 3 min of high-Na + perfusion of the nucleoptasmic side of the patch are indicated with the ion species labels (i. e. K + and Na +). Reversal potential for highK/high-K (pipette/bath) condition was 0.7 mV, and for high-K/ high-Na was 3.2 mV

478

A

(pA)

i (rA) S e t 2, V m = §

-200

S e t I0, V m = - 6 0 mV

mV [ 2 0 0

= i3O

pA -I00

L0

0 -200

S e t 8, V m = - 4 0 mV

S e t l , V m = + 4 0 mV [ 2o0 = +23 pi

= -2t pa

k

-I00

I

L- 0 Set

Vm= +30

3,

= +11

0

~00

mV

-2:00

S e t 6, V m = - 2 0 mV < A ~ > = - 1 1 pA

pA

-100

1oo

0

0 S e t 4, V m = =

309

+tO m V + 5 pA

-200

S e t 5, V m = - 1 0 mV = 5 pA

-I00 1000

0

t

E

500

i000

i

1500 t (ms)

i

i

i

i

2000

2500

0

500

B

i

~

i000 1500 t (ms)

J

i

2000

2500

(pA) S e t 3, V m = T =

+60 mV 545 m s

< i > (pa) -200

200 S e t 10, V ~f~--= 1 6 7 1TIS mV

-100

100

0

0 S e t 1, V m = 7" =

+4,0 mV 693 m s

200

S e t 8, V m = 7- =

-200

-40 mV 401 m s

-lOO

100

7

I

o 200

Set 3, V m =

+20 mV 7" = 2 4 2 6 m s

Set 6. Vm= -20 mV

-lOO ......

0 Vm= T=

4,

o -zoo

"7" = 1063 m s 100

Set

0

200

+lO mV ?

...........

S e t 5, V m = T=

L

o -zoo

-10 mV ?

- lOO

100 " "l 0 i

,

~

i

i

i

0

500

1000

1500

2000

2500

t (ms)

i

0

i

500

i

i

1000

1500

i

2000

i

0

Fig. 5A, B. Voltage-gated NIC activity. A Current transients, i(t), selected from ensembles of raw records elicited in a nucleus-attached patch with voltage pulses applied from a membrane potential, Vm, of 0 to the Vmlevel shown in each trace. A linear conductance, 7, of 532 pS was calculated for this patch (Fig. 6 A - C). Experiment with high-K solution in both pipette and bath. The temperature was set to 33~ B Ensemble averages often current transients, (i(t)), corresponding to A. Single-exponential fits were obtained for each average (superimposed). The voltage dependence of the time constants of these monoexponential fits, z, suggests a voltage-dependent inactivation process. The activation process was too fast to allow assessment. At low voltages (Vm = _ 10 mV in the panel) the z values were too long to permit proper estimation

E5o0

t (ms)

from which it is c o n c l u d e d that the N I C p e r m e a b i l i t y ratio for K § a n d N a § is a p p r o x i m a t e l y 1. Finally, spont a n e o u s activity was observed in the presence of 0 - 4 m M M g 2-- a n d / o r A T P 2 - , regardless o f whether or n o t 0 1 m M C a 2 § was present, i n d i c a t i n g a m i n o r direct role o f these ions in N I C activity.

Voltage-gated activity W h e n n o s p o n t a n e o u s activity was observed, N I C could be i n d u c e d to o p e n by a p p l y i n g voltage pulses (11 patches: 7 n u c l e i - a t t a c h e d a n d 5 excised, inside-out, one of which was originally attached). F i g u r e 5 A shows cur-

479 Counts

C o u n t s

A

600

~

Set 2, V m = +60 m V

S e t 10, V m = - 6 0 mV [ 9 0 ~

300

,

i

I00

200

-100

-200

2:50

Set 1, V m = +40 mV

S e t 8, V m = - 4 0 mV i 2 5 ~ 125

i

i

i

I

i

I00

200

-100

-200

140

Set 3, V r n = +20 m V

2