Directory UMM :Data Elmu:jurnal:B:Brain Research:Vol885.Issue1.2000:

www.elsevier.com / locate / bres

Research report

Enhancement of persistent sodium current by internal fluorescence in

isolated hippocampal neurons

*

George G. Somjen

Department of Cell Biology, Box 3709, Duke University Medical Center, Durham, NC 27710, USA Accepted 12 September 2000

Abstract

Following up on an earlier chance observation, voltage-dependent whole-cell currents were recorded from isolated hippocampal neurons filled with the fluorescent dyes Fluo-3 and Fura-red, that were intermittently excited by 488 nm laser light. In the absence of any ion channel blocking drugs, in most cells depolarizing voltage steps initially evoked only the ‘Hodgkin–Huxley’ type early, fast inward surge followed by sustained outward current. Over 5–20 min of intermittent electrical stimulation and laser-excited fluorescence pulses, a 1 voltage-dependent, slowly inactivating inward current also appeared and grew, while sustained outward current diminished. When K currents were blocked, a small persistent inward current was usually detectable immediately, and then it increased in amplitude. This current was blocked by tetrodotoxin (TTX) and it had current–voltage (I –V ) characteristics of a persistent sodium current, INa,P. In cells not filled with dye but illuminated by laser, and in cells with dye but not illuminated, I_Na,Premained small. There was a more than 12-fold difference in the maximal amplitude of INa,P of fluorescent compared to non-fluorescent cells. Once induced, INa,P decreased very slowly.

1

Fluorescence increased the duration but not the amplitude of the transient Na current, I_Na,T. With membrane potential clamped to a constant voltage, the laser-induced fluorescence did not evoke a membrane current. It is not certain whether fluorescence-induced INa,P

potentiation is related to photodynamic action.  2000 Elsevier Science B.V. All rights reserved.

Theme: Excitable membranes and synaptic transmission

Topic: Sodium channels

Keywords: Persistent sodium current; Sodium channel; Photodynamic action; Fluorescence; Dissociated neuron; Whole-cell current

1 1. Introduction persistent inward current grew, the presumably K

-me-diated outward currents became reduced in amplitude. It

In a previous study [19] of the effects of low sodium seemed as though the slowly inactivating opposing

cur-chloride concentration and low osmolarity, whole-cell rents interfered one with the other because, as the one

sodium and potassium currents were recorded in patch- waned, the other became more prominent [19].

clamped hippocampal neurons filled with the calcium- The slow inward current appeared to be similar to the

sensitive indicator dyes, Fluo-3 and Fura-red. The cells persistent voltage-dependent sodium current, INa,P,

re-were stimulated by series of depolarizing voltage pulses, corded in many central neurons [6,9,23], except for its

and they were intermittently excited by laser light to record growth over time and its final, unusually large amplitude. confocal images and to measure calcium-dependent fluo- Because of the possible role of INa,Pin seizures, spreading

rescence. In many trials in which no ion channel blocking depression and hypoxia [7,10,13] it seemed important to

drugs were used, a slowly inactivating, inward current define the conditions under which this current can attain

appeared. This current was not usually detectable at first, such unusual intensity.

but it increased in amplitude over 5–20 min. As the In this article I report that internal fluorescence itself

causes the growth of INa,P which, once it is induced,

reverses only very slowly. An abstract of some of the

*Tel.:11-919-681-8404; fax:11-919-684-5481.

E-mail address: g.somjen@cellbio.duke.edu (G.G. Somjen). findings has been published [20]. A companion paper [21]

0006-8993 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. P I I : S 0 0 0 6 - 8 9 9 3 ( 0 0 ) 0 2 9 4 7 - 4

1

reports the effect of elevated external K concentration on The current records were read with Clampfit (Axon

INa,P. Instruments) software. After subtraction of linear leak and

holding currents, the data were further processed with the Excel (Microsoft) program. Junction potentials were

calcu-2. Materials and methods lated with the JPCalc program [3].

2.1. Isolation of neurons 2.3. Fluorescence imaging

Hippocampal CA1 pyramidal cells were isolated accord- The non-permeant forms of the fluorescent calcium

ing to the method of Kay and Wong [14]. Briefly: rats of indicator dyes Fluo-3 (10 mM) and Fura-red (30 mM)

60–120 g body weight were decapitated under deep ether (Molecular Probes) were added to the pipette solution [11].

anesthesia. Brains were removed and 500-mm-thick slices The 488 nm excitation light was used and emission was

were cut from hippocampus. The CA1 region was cut into recorded at 520 nm (Fluo-3) as well as 640 nm (Fura-red)

smaller pieces and these tissue fragments were digested for with COMOS (Biorad) software. Fluorescence intensities

21

75 min. The digestion medium contained (in mM l ): were recorded from two intersecting elongated rectangular

NaCl 125, KCl 5, CaCl₂ 1, MgCl₂ 2, D-glucose 25, areas of interest at 10 or 20 s intervals; whole images were

[2-hydoxyethyl]piperazine-[2-ethanesulfonic acid] recorded at 60 s intervals. Background-corrected

fluores-(HEPES) 10, pH 7.0, with trypsin 0.75 mg / ml, at room cence ratios (Fluo-3 / Fura-red) were computed

subsequent-temperature (20–228C). After digestion the tissue pieces ly using Excel (Microsoft) software. Fluo-3 fluorescence

were washed and then incubated in trypsin-free oxygenated increases while Fura-red fluorescence decreases with rising 21

medium at room temperature. Tissue fragments were [Ca ] .i

dispersed by trituration with a graded series of Pasteur

pipettes. 2.4. Statistics

2.2. Recording of voltage-dependent currents Except when otherwise noted, numerical data are given

as the mean6S.E.M. Significance of differences was

Cell suspensions were placed in a chamber of about 0.7 calculated by paired two-tailed t-test. ml capacity on the stage of a Zeiss Axioskop. When Na

and K currents were recorded, the cells were maintained in

flowing HEPES-buffered medium of the following com- 3. Results

21

position: (in mM l ): NaCl 130, KCl 3.5, CaCl2 1.2,

MgCl2 1.0, glucose 25, HEPES 10, pH 7.3 or 7.35, at Whole-cell currents evoked by voltage steps were

20–228C. Cells were approached under the microscope recorded from patch-clamped freshly isolated hippocampal

objective with patch pipettes; tight seal was established, CA1 neurons. The recording pipette was filled either with a

and the whole-cell recording condition created by suction. KF or with a CsF based solution, that also contained the

To record Na and K currents the pipettes were filled with a fluorescent dyes Fura-red and Fluo-3. At 1-min intervals

21

solution containing (in mM l ): KF 129, NaCl 4, EGTA the cells were stimulated by a series of voltage steps to

10, CaCl 0.5, MgCl 2, HEPES 10, Na ATP 4, pH 7.1 or2 2 2 construct current–voltage (I –V ) curves, and they were also

1

7.3, tip resistance 2.5–4.5 MV. To block K currents, KF excited at 10 or 20 s intervals by 1 s pulses of laser light to 21

was substituted by 109 mM CsF and 20 mM tetraethylam- determine changes in internal Ca activity. When KF was

monium–Cl (TEA–Cl). the main electrolyte, all the voltage-gated channels were

An Axopatch 1D amplifier in voltage-clamp configura- available for activation and the records were the sums of

tion and the pClamp-6 (Axon Instruments) suite of pro- several individual currents.

grams was used to record whole-cell currents. Pipette and Fig. 1A illustrates current–voltage (I –V ) curves of the

cell capacitances were compensated in the customary slowly inactivating currents recorded with a KF-based

manner. Series resistance was compensated to 70%. The pipette from one cell over an 18-min period. The graph

holding potential was 270 mV pipette voltage. Current– shows the average currents measured during the last 15 ms

voltage (I –V ) curves were recorded usually at 1-, some- of each depolarizing step after subtracting linear leak and

times at 2-min intervals. Two different protocols were holding current and correcting pipette potentials for

junc-used: Either eight sweeps, each beginning with a pre-pulse tion potential. Fig. 1B shows sample currents evoked by a

of 100 ms to290 mV to remove inactivation, followed by depolarizing pulse to 225 mV pipette potential (V )P

200 ms depolarizing steps at 2 s intervals of 15 mV (equaling232.6 mV membrane potential (V ), after correc-m

increments, taking the pipette voltage from 270 to 135 tion for junction potential). In most cases, as in Fig. 1A,

mV; or 12 sweeps of a 200 ms hyperpolarizing pre-pulse during the first few minutes of observation the slowly

followed by 400 ms depolarizing steps in 10 mV incre- inactivating component flowed outward at all potentials

Fig. 1. Intermittent fluorescent excitation causes the growth of a persistent inward current. (A) Current–voltage (I –V ) curves of slowly inactivating current of an isolated CA1 neuron recorded at different times after establishing whole-cell recording condition, using a patch pipette filled with KF solution. After a hyperpolarizing pre-pulse, eight sweeps of 200 ms depolarizing steps at 15 mV increments were applied. The average current measured during the final 15 ms of the depolarizing pulses is plotted against the membrane potential (V ) corrected for junction potential. (B) Sample currents recorded 1 min and 18m

min after establishing whole-cells’ condition, evoked by depolarization to225 mV pipette voltage, corresponding to V ofm 232.6 mV after correction of junction potential.

1

rectifier K current, IK,Dr. After 5–10 min, the persistent pulses, a sizeable inward current appeared. Adding TTX to inward current appeared in most fluorescent laser illumi- the bathing solution completely blocked both transient and

nated cells, and then it grew in amplitude. As the persistent persistent inward currents within 3–5 min, but after

inward current increased, outward currents became smaller washing for 16 min with normal solution both components

and the reversal potential (zero-current potential) of the of the inward currents reappeared. By subtracting the

compound current shifted to more positive levels. current amplitudes recorded under the influence of TTX

The persistent inward current was blocked by 0.5 mM from those recorded at the end of the 24 min of

observa-tetrodotoxin (TTX). In the cell illustrated in Fig. 2A, a tion, the TTX-sensitive component of the persistent current

trace of a persistent inward current was recorded immedi- is revealed (Fig. 2B). Fig. 2C shows conductances

com-ately after establishing whole-cell condition. After 24 min puted by dividing the currents of Fig. 2B by the driving

Fig. 2. The persistent inward current is tetrodotoxin (TTX) sensitive. (A) I –V curves of the slowly inactivating current recorded similarly to those of Fig. 1, at 1 min and 24 min in control solution, followed by 4 min of TTX administration, and 16 min of washing with normal solution. (B) The I –V relation of the TTX-sensitive persistent inward current, obtained by subtracting the I –V curve recorded under the influence from TTX (Fig. 2A, triangles) from the I –V curve recorded at 24 min in normal solution (Fig. 2A, diamonds). (C) The voltage-dependent TTX-sensitive conductance in nanosiemens, plotted against membrane voltage in mV. Conductance was calculated by dividing the TTX-sensitive current (Fig. 2B) by the driving potential, defined as the difference between the membrane potential and the reversal potential.

1

the membrane potential and the extrapolated reversal expected, when most K currents were blocked, INa,P

potential of the TTX-sensitive current. The TTX-sensitive could readily be evoked right away in almost all cells by

persistent component activated at a level more negative prolonged depolarizing pulses. INa,P did, however, grow

than260 mV, it was maximal at 220 mV and it reversed with time and with repeated fluorescent excitation, even

1

around 130 mV. TTX sensitivity, activation and reversal when the K currents were eliminated, demonstrating that

1

potentials, and the shape of its I –V curve identify it as a rundown of K current cannot account for its growth. As

persistent sodium current, INa,P [6,9]. the persistent inward current increased in amplitude, its

The fact that outward currents became smaller while reversal shifted in the positive direction, regardless of

1

INa,P grew, suggested the possibility that the ‘rundown’ of whether K currents were blocked or not (Fig. 1A, 5A, B).

1

K currents had unmasked the previously hidden, slow To test whether repeated stimulation by depolarizing

inward current [19]. To test this, cells were tested in which voltage steps causes the gradual increase of the INa,P

1

most of the K currents were suppressed by using pipettes (use-dependent facilitation), recordings were made from

filled with a solution containing CsF instead of KF, with cells without fluorescent dyes and without laser

illumina-1

20 mM tetraethylammonium (TEA ) added. Even after tion (three cells with KF and 14 cells with CsF in the

1 1 1

replacing K with Cs and TEA , a small non-inactivat- pipettes). In these cells the INa,P was small and it did not

ing outward current often remained visible at strongly grow during repeated stimulation. The question then was,

1

depolarized voltages (Fig. 5B). This may be due to Cs whether the dyes themselves potentiate INa,P, and,

con-1

flowing through K channels, or it may represent the versely, whether laser light in the absence of internal

containing dye (three with KF, three with CsF), and filling pA / pF recorded from cells under similar conditions but in

cells with dye but not using the laser (n511, all CsF) had the absence of fluorescence. Without exception, in all

the same negative result: in the absence of fluorescence, fluorescent cells the INa,P increased in amplitude with time

the INa,P was, and remained, small. but the magnitude of the effect was variable. In the

Fig. 3 shows a statistical summary of the maximal fluorescent, CsF group there were 18 cells from which the

persistent inward current amplitudes normalized to cell current was recorded for 5–10 min after establishing

capacitance, recorded at different times from groups of whole-cell condition; the largest current in a cell in this

cells with and without fluorescence, in the presence and in group was21700 pA (185 pA / pF) and the smallest2116

1

the absence of K mediated outward current. The differ- pA (12.2 pA / pF). The mean amplitude of all the cells in

ence between the fluorescent and the non-fluorescent cells the 5–10 min, fluorescent, CsF group was 68.6 pA / pF

was striking. In the fluorescent cells that were observed for with a standard deviation of642.1 and standard error of

more than 10 min with CsF-filled pipettes, the mean 69.9. In the non-fluorescent group, and initially also

maximal INa,P amplitude was 280.1619.7 pA / pF (n56), among the fluorescent cells, the difference between the

1

which is 12.9 times larger than the 26.261.2 (n57) INa,P amplitude recorded with and without blocking K

currents was also quite marked. With time, however, the currents recorded from fluorescent cells with KF pipettes approached those recorded with CsF. In the greater-than 10 min groups the data are too few and the variability is too large for reliable statistical comparison; for the few cells tested for more than 10 min the difference between KF and CsF recording is not significant (Fig. 3B).

There was no detectable relationship between the growth 21

of the INa,P and the cytoplasmic calcium activity, [Ca ] ,i as indicated by the ratio of the fluorescences of Fluo-3 and Fura-red. Initially the fluorescence ratio increased sharply but transiently whenever the cell was stimulated by depolarizing currents (see also [19,21]). These transient

21

responses of [Ca ] were presumably caused by voltage-i dependent calcium currents. In the course of repeated stimulation they usually subsided, probably because the pipettes did not contain the ingredients required to

coun-21

teract ‘rundown’ of Ca currents [19]. The ‘resting’ or

21

baseline [Ca ] showed upward or downward drift ini

different experiments and was not correlated with the growth of INa,P.

2

In the great majority of the trials F was the main anion in the recording pipette. K-gluconate or K-acetate replaced KF in several experiments. The seal between pipette and cell membrane often loosened within 5–7 min so that no

INa,P was seen, but in two trials with gluconate and in one with acetate the seal was maintained and INa,P appeared and grew as usual.

At this stage of the study three questions arose. First, do the laser induced pulses of fluorescence induce a mem-brane current? Second, can fluorescence by itself potentiate

Fig. 3. Statistical summary of the potentiation of the persistent inward

current by repeated pulses of fluorescence. The columns show the INa,P, or does it have to be combined with

depolariza-mean6S.E.M. of the maximal slowly inactivating inward current, normal- _{tion? And third, is the potentiation of the I reversible,}

Na,P

ized to cell capacitance. The numbers above the horizontal axis show the _{or is it an irreversibly altered state of the membrane} time in minutes elapsed from establishing whole-cell conditions until the

channel? To answer these questions, I –V curves were first

recording. ‘0–5 min’ show the largest persistent inward current measured

recorded for 6–14 min from cells that were filled with dye,

toward the end of the first 5 min of recording, ‘5–10 min’ show

measurements usually 8–10 min after the start of recording. The numbers but without turning on the laser. Then electrical

stimula-of cells in each category are shown below the columns. Pipette solutions _{tion was suspended and intermittent laser excitation was} 1

based on CsF also contained 20 mM tetraethylammonium (TEA ) to _{turned on for 7–10 min. In this period the holding current} 1

block most K currents. (A) Non-fluorescent cells include those filled

was continuously recorded. The voltage to which the

with dye but not illuminated, not filled with dye but illuminated, and free

membrane was clamped was adjusted from time to time to

of dye and not illuminated. (B) Fluorescent cells were filled with Fluo-3

dependent. Subsequently, I –V curves were again recorded there was a striking increase in amplitude of the INa,Pin all

without laser illumination for additional 28–32 min. Figs. 6 trials. The mean maximal amplitude immediately before

4 A and B show the maximal INa,P recorded with CsF laser treatment was243.7613.2 pA and immediately after

pipettes from four cells before and after laser illumination the series of illuminations29136172 pA (mean6S.E.M.).

under this protocol. Figs. 4C and D show the holding One of the six cells was lost early in the post-laser period,

current of one of these cells recorded during intermittent the persistent current of the five other cells at the end of laser illumination, reproduced on two different time-scales. 28–32 min following laser treatment was still28076162 Fig. 5 illustrates recordings obtained from individual cells pA. The decay was slow but consistent and statistically

before and after laser excitation. significant (P,0.009, paired two-tailed t-test). This slow

As illustrated by the examples of Fig. 4C and D, laser decrease could perhaps be caused by ‘rundown’ of INa,P,

1

pulse excitation of fluorescence did not evoke a detectable but the transient Na current, INa,T did not show a similar

change in the holding current, regardless of holding decay. INa,T amplitude was on average214,60761591 pA

potential. The same was true in four cells from which in these six cells before laser treatment, 212,14661406

recordings were made with CsF and two cells with KF pA immediately after laser and212,37962425 pA half an

pipettes. Recordings have been obtained at 290, 270, hour later.

250,240 and230 mV holding potentials (not all levels in Fig. 5 illustrates examples of I –V curves and current

each cells). Not surprisingly, shifting of the holding traces obtained from two of the six cells before and after

potential did evoke increases in holding current that laser-induced fluorescence. Fig. 5A, C and D were

re-inactivated slowly (Fig. 4C, D). corded with KF filled pipette from one cell, and Fig. 5B

Following 5–8 min of intermittent laser illumination, from another cell, using CsF. With KF, INa,P was not

Fig. 4. Intermittent pulses of fluorescence do not evoke a membrane current, but they cause long-lasting potentiation of the persistent inward current, INa,P. (A) Maximal persistent inward currents recorded from four dye-filled cells before illumination by laser. The period of pre-laser observation varied from 6 to

1

14 min. The recording pipettes were filled with CsF based solution containing 20 mM TEA . (B) Maximal persistent inward currents recorded from the same four cells after intermittent laser illumination. Note different ordinate scales for A and B. (C) Continuous recording of the holding current during intermittent laser illumination of one of the cells represented in A and B. The lower trace shows the switching of the laser. The broken vertical lines indicate changes in the holding potential. No voltage pulses were applied during intermittent laser illumination. (D) Part of the recording of C, on an expanded time-scale.

Fig. 5. Sample I –V curves and current recordings before and after fluorescent excitation. Parts A, C and D are from the same cell. (A) I –V curves just before and 2 min after intermittent laser illumination, recorded with KF-based pipette solution with fluorescent dyes. (B) Similar to A, from another cell,

1 1

but recorded with K currents mostly blocked, with CsF and TEA in the recording pipette. The two I –V curves recorded 2 min and 30 min after the intermittent laser treatment show the slow decrease of INa,P. (C) Transient sodium current, INa,Tevoked by depolarizing pulse of230 mV pipette voltage (corresponding to237.9 mV membrane potential) recorded before and 3 min after intermittent laser illumination. The current was sampled at 2000 Hz. (D) Currents evoked by220 mV and130 mV pipette voltage steps (227.9 mV and122.1 mV membrane potential) before and 3 min after intermittent laser treatment. The currents were sampled at 2000 Hz, but were filtered at 100 Hz off-line.

detectable before laser treatment, but it was marked slowly after laser excitation has ceased. The first of these

thereafter (Fig. 5A, D). With CsF the originally small INa,P two findings could be expected, the second was a surprise.

increased manifold following laser pulses (Fig. 5B). As Since laser illumination and intracellular fluorescence

shown in Fig. 5C, while the amplitude of INa,T did not are not usual hazards for hippocampus in situ, the effect is change, its duration increased. Such a widening of the INa,T not directly relevant to human pathophysiology.

Nonethe-trace was consistently seen in conjunction with the growth less it is of interest for the understanding of sodium

of the INa,P. The amplitude of INa,T usually showed slow currents in mammalian central neurons. The fact that the

drift up or down, unrelated to laser treatment and not persistent current can grow more than 10-fold under the

linked to the changes in other ion currents. relatively mild influence of intracellular fluorescence

sug-gests that the maximal current carrying capacity of slowly

1

inactivating Na channels is much greater than is normally manifested. It is not known whether INa,Parises because of

4. Discussion incapacitated inactivation of the channel that generates the

1

common, transient Na current, INa,T, or whether it

Two conclusions emerge. First, that depression of represents current flowing through a distinct channel type.

1

outward K currents reveals the voltage-gated persistent There is evidence favoring the former interpretations

1

Na current, INa,P, which otherwise can be concealed by [2,4,5]. Our data do not distinguish between these two

I . Second, that laser-induced internal fluorescence power-K alternatives, but the fact that INa,T duration increased fully potentiates INa,P. This potentiation accumulates with whenever INa,P amplitude increased does suggest a link

[4] W.A. Catterall, Structure and function of voltage sensitive ion

In the data there is also a message of caution.

Fluores-channels, Science 242 (1988) 50–61.

cent dyes have many uses in cell biology. One must be

[5] W.A. Catterall, Modulation of sodium and calcium channels by

mindful of the possibility that the fluorescence might alter _{protein phosphorylation and G proteins. [Review] [136 refs], Adv.}

the very function one is about to study. Second Messeng. Phosphoprot. Res. 31 (1997) 159–181.

The very slow reversibility of the fluorescence-induced [6] W.E. Crill, Persistent sodium current in mammalian central neurons, Ann. Rev. Physiol. 58 (1996) 349–362.

potentiation of INa,P raises the question whether it is an

[7] W.E. Crill, P.C. Schwindt, Role of persistent inward and outward

example of phototoxicity or of photodynamic action.

membrane currents in epileptiform bursting in mammalian neurons,

Photodynamic cell injury has been attributed to the forma- _{in: A.V. Delgado-Escueta, A.A. Ward, D.M. Woodbury, R.J. Porter} tion of reactive oxygen species [22], and as such it has (Eds.), Basic Mechanisms of the Epilepsies, Raven Press, New York,

been successfully prevented by replacing oxygen by argon 1986, pp. 225–233.

[8] Z.J. Cui, T. Kanno, Photodynamic triggering of calcium oscillation

in the solution bathing Aplysia neurons [17]. Unlike the

in the isolated rat pancreatic acini, J. Physiol. 504 (1997) 47–55.

specific effect of fluorescence in hippocampal neurons

[9] C.R. French, P. Sah, K.J. Buckett, P.W. Gage, A voltage-dependent

described here, photodynamic cell damage is typically _{persistent sodium current in mammalian hippocampal neurons, J.}

associated with non-specific increase of membrane ion Gen. Physiol. 95 (1990) 1139–1157.

¨

permeability and consequently with depolarization [16,22]. [10] A.K.M. Hammarstrom, P.W. Gage, Inhibition of oxidative metabo-lism increases persistent sodium current in rat CA1 hippocampal

Other, more subtle photodynamic effects, which do not

neurons, J. Physiol. 510 (1998) 735–741.

necessarily destroy cells have, however, also been

de-[11] R.P. Haugland, Handbook of Fluorescent Probes and Research

scribed. Among them are inactivation of the mitochondrial _{Chemicals, Molecular Probes, Eugene, 1996.}

permeability transition pore [18]; the triggering of calcium [12] Y.-K. Ju, D.A. Saint, P.W. Gage, Hypoxia increases persistent

oscillations in pancreatic cells [8]; and light-induced sodium current in rat ventricular myocytes, J. Physiol. 497 (1999) 337–347.

potentiation of NMDA currents in cultured neurons [15].

[13] H. Kager, W.J. Wadman, G.G. Somjen, G.G., Simulated seizure

Each of these effects differs from the one described here,

discharges and spreading depression-like depolarization in a neuron

yet each exemplifies the novel fact, that light energy can _{model incorporating interstitial space and ion concentration changes,}

change channel function. J. Neurophysiol. (2000) (in press).

Gage and associates [10,12] reported enhancement of [14] A.R. Kay, R.K.S. Wong, Isolation of neurons suitable for patch clamping from adult mammalian central nervous system, J.

Neuro-INa,P by hypoxia and cyanide poisoning in neurons and

sci. Meth. 16 (1986) 227–238.

heart muscle cells. We have found a similar effect in

[15] D.N. Leskiewicz, W.K. Potthoff, K.K. Kandler, E. Aizenman,

1

hippocampal pyramidal cells by elevated [K ]o [20,21]. _{Selective enhancement of NMDA receptor mediated currents by}

1

The effects of hypoxia and high [K ] were weaker ando light, Soc. Neurosci. Abstr. 25 (1999) 1978.

more readily reversible than those of fluorescence. This [16] F. Mendez, R. Penner, Near-visible ultraviolet light induces a novel ubiquitous calcium-permeable cation current in mammalian cell

quantitative difference does not, however, exclude a ‘final

lines, J. Physiol. 507 (1999) 365–377.

common path’ of the enhancement of INa,P by these three _{[17] T.D. Parsons, D. Kleinfeld, F. Raccuia-Behling, B.M. Salzberg,}

agents. _{Optical recording of the electrical activity of synaptically interacting}

Aplysia neurons in culture using potentiometric probes, Biophys. J.

56 (1989) 213–221.

[18] C. Salet, G. Moreno, F. Ricchelli, P. Bernardi, Singlet oxygen

Acknowledgements

produced by photodynamic action causes inactivation of the mito-chondrial permeability transition pore, J. Biol. Chem. 272 (1997)

Supported by NIH grant NS 18670. _{21938–21943.}

[19] G.G. Somjen, Low external NaCl concentration and low osmolarity enhance voltage gated Ca currents but depress K currents in freshly isolated rat hippocampal neurons, Brain Res. 851 (1999) 189–197.

References ₁

[20] G.G. Somjen, Intracellular fluorescence and elevated [K ] poten-o

tiate persistent sodium current in isolated hippocampal neurons, [1] C. Alzheimer, A novel voltage-dependent cation current in rat _{FASEB J. 14 (2000) A107.}

neocortical neurones, J. Physiol. 479 (1994) 199–205. _{[21] G.G. Somjen, M. Muller, Potassium-induced enhancement of persis-¨} 1

[2] C. Alzheimer, P.C. Schwindt, W.E. Crill, Modal gating of Na _{tent inward current in hippocampal neurons in isolation and in tissue} 1

channels as a mechanism of persistent Na current in pyramidal _{slices, Brain Res. 885 (2000) 102–110.}

neurons from rat and cat sensorimotor cortex, J. Neurosci. 13 (1993) _{[22] K.G. Specht, M.A.J. Rodgers, Plasma membrane depolarization and} 660–673. _{calcium influx during cell injury by photodynamic action, Biochim.} [3] P.H. Barry, JPCalc, a software package for calculating liquid _{Biophys. Acta 1070 (1991) 60–68.}

junction potential corrections in patch-clamp, intracellular, epithelial 1

[23] C.P. Taylor, Na currents that fail to inactivate, Trends Neurosci. 16 and bilayer measurements and for correcting junction potential _{(1993) 455–460.}

Fig. 1. Intermittent fluorescent excitation causes the growth of a persistent inward current. (A) Current–voltage (I –V ) curves of slowly inactivating current of an isolated CA1 neuron recorded at different times after establishing whole-cell recording condition, using a patch pipette filled with KF solution. After a hyperpolarizing pre-pulse, eight sweeps of 200 ms depolarizing steps at 15 mV increments were applied. The average current measured during the final 15 ms of the depolarizing pulses is plotted against the membrane potential (V ) corrected for junction potential. (B) Sample currents recorded 1 min and 18m

min after establishing whole-cells’ condition, evoked by depolarization to225 mV pipette voltage, corresponding to V ofm 232.6 mV after correction of junction potential.

1

rectifier K current, IK,Dr. After 5–10 min, the persistent pulses, a sizeable inward current appeared. Adding TTX to inward current appeared in most fluorescent laser illumi- the bathing solution completely blocked both transient and nated cells, and then it grew in amplitude. As the persistent persistent inward currents within 3–5 min, but after inward current increased, outward currents became smaller washing for 16 min with normal solution both components and the reversal potential (zero-current potential) of the of the inward currents reappeared. By subtracting the compound current shifted to more positive levels. current amplitudes recorded under the influence of TTX The persistent inward current was blocked by 0.5 mM from those recorded at the end of the 24 min of observa-tetrodotoxin (TTX). In the cell illustrated in Fig. 2A, a tion, the TTX-sensitive component of the persistent current trace of a persistent inward current was recorded immedi- is revealed (Fig. 2B). Fig. 2C shows conductances com-ately after establishing whole-cell condition. After 24 min puted by dividing the currents of Fig. 2B by the driving of stimulation by voltage pulses alternated with laser voltage, the latter being defined as the difference between

Fig. 2. The persistent inward current is tetrodotoxin (TTX) sensitive. (A) I –V curves of the slowly inactivating current recorded similarly to those of Fig. 1, at 1 min and 24 min in control solution, followed by 4 min of TTX administration, and 16 min of washing with normal solution. (B) The I –V relation of the TTX-sensitive persistent inward current, obtained by subtracting the I –V curve recorded under the influence from TTX (Fig. 2A, triangles) from the I –V curve recorded at 24 min in normal solution (Fig. 2A, diamonds). (C) The voltage-dependent TTX-sensitive conductance in nanosiemens, plotted against membrane voltage in mV. Conductance was calculated by dividing the TTX-sensitive current (Fig. 2B) by the driving potential, defined as the difference between the membrane potential and the reversal potential.

1

the membrane potential and the extrapolated reversal expected, when most K currents were blocked, INa,P

potential of the TTX-sensitive current. The TTX-sensitive could readily be evoked right away in almost all cells by persistent component activated at a level more negative prolonged depolarizing pulses. INa,P did, however, grow than260 mV, it was maximal at 220 mV and it reversed with time and with repeated fluorescent excitation, even

1

around 130 mV. TTX sensitivity, activation and reversal when the K currents were eliminated, demonstrating that 1

potentials, and the shape of its I –V curve identify it as a rundown of K current cannot account for its growth. As persistent sodium current, INa,P [6,9]. the persistent inward current increased in amplitude, its The fact that outward currents became smaller while reversal shifted in the positive direction, regardless of

1

INa,P grew, suggested the possibility that the ‘rundown’ of whether K currents were blocked or not (Fig. 1A, 5A, B). 1

K currents had unmasked the previously hidden, slow To test whether repeated stimulation by depolarizing inward current [19]. To test this, cells were tested in which voltage steps causes the gradual increase of the INa,P

1

most of the K currents were suppressed by using pipettes (use-dependent facilitation), recordings were made from filled with a solution containing CsF instead of KF, with cells without fluorescent dyes and without laser

illumina-1

20 mM tetraethylammonium (TEA ) added. Even after tion (three cells with KF and 14 cells with CsF in the

1 1 1

replacing K with Cs and TEA , a small non-inactivat- pipettes). In these cells the INa,P was small and it did not ing outward current often remained visible at strongly grow during repeated stimulation. The question then was,

1

depolarized voltages (Fig. 5B). This may be due to Cs whether the dyes themselves potentiate INa,P, and, con-1

flowing through K channels, or it may represent the versely, whether laser light in the absence of internal non-specific cation current described by Alzheimer [1]. As fluorescence has this effect. Laser illumination of cells not

cells with dye but not using the laser (n511, all CsF) had the absence of fluorescence. Without exception, in all the same negative result: in the absence of fluorescence, fluorescent cells the INa,P increased in amplitude with time the INa,P was, and remained, small. but the magnitude of the effect was variable. In the Fig. 3 shows a statistical summary of the maximal fluorescent, CsF group there were 18 cells from which the persistent inward current amplitudes normalized to cell current was recorded for 5–10 min after establishing capacitance, recorded at different times from groups of whole-cell condition; the largest current in a cell in this cells with and without fluorescence, in the presence and in group was21700 pA (185 pA / pF) and the smallest2116

1

the absence of K mediated outward current. The differ- pA (12.2 pA / pF). The mean amplitude of all the cells in ence between the fluorescent and the non-fluorescent cells the 5–10 min, fluorescent, CsF group was 68.6 pA / pF was striking. In the fluorescent cells that were observed for with a standard deviation of642.1 and standard error of more than 10 min with CsF-filled pipettes, the mean 69.9. In the non-fluorescent group, and initially also maximal INa,P amplitude was 280.1619.7 pA / pF (n56), among the fluorescent cells, the difference between the 1 which is 12.9 times larger than the 26.261.2 (n57) INa,P amplitude recorded with and without blocking K

currents was also quite marked. With time, however, the currents recorded from fluorescent cells with KF pipettes approached those recorded with CsF. In the greater-than 10 min groups the data are too few and the variability is too large for reliable statistical comparison; for the few cells tested for more than 10 min the difference between KF and CsF recording is not significant (Fig. 3B).

There was no detectable relationship between the growth

21 of the INa,P and the cytoplasmic calcium activity, [Ca ] ,i

as indicated by the ratio of the fluorescences of Fluo-3 and Fura-red. Initially the fluorescence ratio increased sharply but transiently whenever the cell was stimulated by depolarizing currents (see also [19,21]). These transient

21

responses of [Ca ] were presumably caused by voltage-i

dependent calcium currents. In the course of repeated stimulation they usually subsided, probably because the pipettes did not contain the ingredients required to

coun-21

teract ‘rundown’ of Ca currents [19]. The ‘resting’ or

21

baseline [Ca ] showed upward or downward drift ini

different experiments and was not correlated with the growth of INa,P.

2

In the great majority of the trials F was the main anion in the recording pipette. K-gluconate or K-acetate replaced KF in several experiments. The seal between pipette and cell membrane often loosened within 5–7 min so that no INa,P was seen, but in two trials with gluconate and in one with acetate the seal was maintained and INa,P appeared and grew as usual.

At this stage of the study three questions arose. First, do the laser induced pulses of fluorescence induce a mem-brane current? Second, can fluorescence by itself potentiate Fig. 3. Statistical summary of the potentiation of the persistent inward

current by repeated pulses of fluorescence. The columns show the INa,P, or does it have to be combined with depolariza-mean6S.E.M. of the maximal slowly inactivating inward current, normal- _{tion? And third, is the potentiation of the I reversible,}

Na,P

ized to cell capacitance. The numbers above the horizontal axis show the _{or is it an irreversibly altered state of the membrane} time in minutes elapsed from establishing whole-cell conditions until the

channel? To answer these questions, I –V curves were first recording. ‘0–5 min’ show the largest persistent inward current measured

recorded for 6–14 min from cells that were filled with dye, toward the end of the first 5 min of recording, ‘5–10 min’ show

measurements usually 8–10 min after the start of recording. The numbers but without turning on the laser. Then electrical stimula-of cells in each category are shown below the columns. Pipette solutions _{tion was suspended and intermittent laser excitation was}

1

based on CsF also contained 20 mM tetraethylammonium (TEA ) to _{turned on for 7–10 min. In this period the holding current}

1

block most K currents. (A) Non-fluorescent cells include those filled

was continuously recorded. The voltage to which the with dye but not illuminated, not filled with dye but illuminated, and free

membrane was clamped was adjusted from time to time to of dye and not illuminated. (B) Fluorescent cells were filled with Fluo-3

dependent. Subsequently, I –V curves were again recorded there was a striking increase in amplitude of the INa,Pin all without laser illumination for additional 28–32 min. Figs. 6 trials. The mean maximal amplitude immediately before 4 A and B show the maximal INa,P recorded with CsF laser treatment was243.7613.2 pA and immediately after pipettes from four cells before and after laser illumination the series of illuminations29136172 pA (mean6S.E.M.). under this protocol. Figs. 4C and D show the holding One of the six cells was lost early in the post-laser period, current of one of these cells recorded during intermittent the persistent current of the five other cells at the end of laser illumination, reproduced on two different time-scales. 28–32 min following laser treatment was still28076162 Fig. 5 illustrates recordings obtained from individual cells pA. The decay was slow but consistent and statistically before and after laser excitation. significant (P,0.009, paired two-tailed t-test). This slow As illustrated by the examples of Fig. 4C and D, laser decrease could perhaps be caused by ‘rundown’ of INa,P,

1

pulse excitation of fluorescence did not evoke a detectable but the transient Na current, INa,T did not show a similar change in the holding current, regardless of holding decay. INa,T amplitude was on average214,60761591 pA potential. The same was true in four cells from which in these six cells before laser treatment, 212,14661406 recordings were made with CsF and two cells with KF pA immediately after laser and212,37962425 pA half an pipettes. Recordings have been obtained at 290, 270, hour later.

250,240 and230 mV holding potentials (not all levels in Fig. 5 illustrates examples of I –V curves and current each cells). Not surprisingly, shifting of the holding traces obtained from two of the six cells before and after potential did evoke increases in holding current that laser-induced fluorescence. Fig. 5A, C and D were re-inactivated slowly (Fig. 4C, D). corded with KF filled pipette from one cell, and Fig. 5B Following 5–8 min of intermittent laser illumination, from another cell, using CsF. With KF, INa,P was not

Fig. 4. Intermittent pulses of fluorescence do not evoke a membrane current, but they cause long-lasting potentiation of the persistent inward current, INa,P. (A) Maximal persistent inward currents recorded from four dye-filled cells before illumination by laser. The period of pre-laser observation varied from 6 to

1

14 min. The recording pipettes were filled with CsF based solution containing 20 mM TEA . (B) Maximal persistent inward currents recorded from the same four cells after intermittent laser illumination. Note different ordinate scales for A and B. (C) Continuous recording of the holding current during intermittent laser illumination of one of the cells represented in A and B. The lower trace shows the switching of the laser. The broken vertical lines indicate changes in the holding potential. No voltage pulses were applied during intermittent laser illumination. (D) Part of the recording of C, on an expanded time-scale.

Fig. 5. Sample I –V curves and current recordings before and after fluorescent excitation. Parts A, C and D are from the same cell. (A) I –V curves just before and 2 min after intermittent laser illumination, recorded with KF-based pipette solution with fluorescent dyes. (B) Similar to A, from another cell,

1 1

but recorded with K currents mostly blocked, with CsF and TEA in the recording pipette. The two I –V curves recorded 2 min and 30 min after the intermittent laser treatment show the slow decrease of INa,P. (C) Transient sodium current, INa,Tevoked by depolarizing pulse of230 mV pipette voltage (corresponding to237.9 mV membrane potential) recorded before and 3 min after intermittent laser illumination. The current was sampled at 2000 Hz. (D) Currents evoked by220 mV and130 mV pipette voltage steps (227.9 mV and122.1 mV membrane potential) before and 3 min after intermittent laser treatment. The currents were sampled at 2000 Hz, but were filtered at 100 Hz off-line.

detectable before laser treatment, but it was marked slowly after laser excitation has ceased. The first of these thereafter (Fig. 5A, D). With CsF the originally small INa,P two findings could be expected, the second was a surprise. increased manifold following laser pulses (Fig. 5B). As Since laser illumination and intracellular fluorescence shown in Fig. 5C, while the amplitude of INa,T did not are not usual hazards for hippocampus in situ, the effect is change, its duration increased. Such a widening of the INa,T not directly relevant to human pathophysiology. Nonethe-trace was consistently seen in conjunction with the growth less it is of interest for the understanding of sodium of the INa,P. The amplitude of INa,T usually showed slow currents in mammalian central neurons. The fact that the drift up or down, unrelated to laser treatment and not persistent current can grow more than 10-fold under the linked to the changes in other ion currents. relatively mild influence of intracellular fluorescence sug-gests that the maximal current carrying capacity of slowly

1

inactivating Na channels is much greater than is normally manifested. It is not known whether INa,Parises because of

4. Discussion incapacitated inactivation of the channel that generates the

1

common, transient Na current, INa,T, or whether it Two conclusions emerge. First, that depression of represents current flowing through a distinct channel type.

1

outward K currents reveals the voltage-gated persistent There is evidence favoring the former interpretations 1

Na current, INa,P, which otherwise can be concealed by [2,4,5]. Our data do not distinguish between these two I . Second, that laser-induced internal fluorescence power-K alternatives, but the fact that INa,T duration increased fully potentiates INa,P. This potentiation accumulates with whenever INa,P amplitude increased does suggest a link repeated laser pulse excitation and it reverses only very between these two.

[4] W.A. Catterall, Structure and function of voltage sensitive ion In the data there is also a message of caution.

Fluores-channels, Science 242 (1988) 50–61. cent dyes have many uses in cell biology. One must be

[5] W.A. Catterall, Modulation of sodium and calcium channels by mindful of the possibility that the fluorescence might alter _{protein phosphorylation and G proteins. [Review] [136 refs], Adv.} the very function one is about to study. Second Messeng. Phosphoprot. Res. 31 (1997) 159–181.

The very slow reversibility of the fluorescence-induced [6] W.E. Crill, Persistent sodium current in mammalian central neurons, Ann. Rev. Physiol. 58 (1996) 349–362.

potentiation of INa,P raises the question whether it is an

[7] W.E. Crill, P.C. Schwindt, Role of persistent inward and outward example of phototoxicity or of photodynamic action.

membrane currents in epileptiform bursting in mammalian neurons, Photodynamic cell injury has been attributed to the forma- _{in: A.V. Delgado-Escueta, A.A. Ward, D.M. Woodbury, R.J. Porter} tion of reactive oxygen species [22], and as such it has (Eds.), Basic Mechanisms of the Epilepsies, Raven Press, New York, been successfully prevented by replacing oxygen by argon 1986, pp. 225–233.

[8] Z.J. Cui, T. Kanno, Photodynamic triggering of calcium oscillation in the solution bathing Aplysia neurons [17]. Unlike the

in the isolated rat pancreatic acini, J. Physiol. 504 (1997) 47–55. specific effect of fluorescence in hippocampal neurons

[9] C.R. French, P. Sah, K.J. Buckett, P.W. Gage, A voltage-dependent described here, photodynamic cell damage is typically _{persistent sodium current in mammalian hippocampal neurons, J.} associated with non-specific increase of membrane ion Gen. Physiol. 95 (1990) 1139–1157.

¨

permeability and consequently with depolarization [16,22]. [10] A.K.M. Hammarstrom, P.W. Gage, Inhibition of oxidative metabo-lism increases persistent sodium current in rat CA1 hippocampal Other, more subtle photodynamic effects, which do not

neurons, J. Physiol. 510 (1998) 735–741. necessarily destroy cells have, however, also been

de-[11] R.P. Haugland, Handbook of Fluorescent Probes and Research scribed. Among them are inactivation of the mitochondrial _{Chemicals, Molecular Probes, Eugene, 1996.}

permeability transition pore [18]; the triggering of calcium [12] Y.-K. Ju, D.A. Saint, P.W. Gage, Hypoxia increases persistent oscillations in pancreatic cells [8]; and light-induced sodium current in rat ventricular myocytes, J. Physiol. 497 (1999)

337–347. potentiation of NMDA currents in cultured neurons [15].

[13] H. Kager, W.J. Wadman, G.G. Somjen, G.G., Simulated seizure Each of these effects differs from the one described here,

discharges and spreading depression-like depolarization in a neuron yet each exemplifies the novel fact, that light energy can _{model incorporating interstitial space and ion concentration changes,} change channel function. J. Neurophysiol. (2000) (in press).

Gage and associates [10,12] reported enhancement of [14] A.R. Kay, R.K.S. Wong, Isolation of neurons suitable for patch clamping from adult mammalian central nervous system, J. Neuro-INa,P by hypoxia and cyanide poisoning in neurons and

sci. Meth. 16 (1986) 227–238. heart muscle cells. We have found a similar effect in

[15] D.N. Leskiewicz, W.K. Potthoff, K.K. Kandler, E. Aizenman, 1

hippocampal pyramidal cells by elevated [K ]o [20,21]. _{Selective enhancement of NMDA receptor mediated currents by} 1

The effects of hypoxia and high [K ] were weaker ando light, Soc. Neurosci. Abstr. 25 (1999) 1978.

more readily reversible than those of fluorescence. This [16] F. Mendez, R. Penner, Near-visible ultraviolet light induces a novel ubiquitous calcium-permeable cation current in mammalian cell quantitative difference does not, however, exclude a ‘final

lines, J. Physiol. 507 (1999) 365–377.

common path’ of the enhancement of INa,P by these three _{[17] T.D. Parsons, D. Kleinfeld, F. Raccuia-Behling, B.M. Salzberg,} agents. _{Optical recording of the electrical activity of synaptically interacting}

Aplysia neurons in culture using potentiometric probes, Biophys. J.

56 (1989) 213–221.

[18] C. Salet, G. Moreno, F. Ricchelli, P. Bernardi, Singlet oxygen

Acknowledgements

produced by photodynamic action causes inactivation of the mito-chondrial permeability transition pore, J. Biol. Chem. 272 (1997) Supported by NIH grant NS 18670. _{21938–21943.}

[19] G.G. Somjen, Low external NaCl concentration and low osmolarity enhance voltage gated Ca currents but depress K currents in freshly isolated rat hippocampal neurons, Brain Res. 851 (1999) 189–197.

References ₁

[20] G.G. Somjen, Intracellular fluorescence and elevated [K ] poten-o

tiate persistent sodium current in isolated hippocampal neurons, [1] C. Alzheimer, A novel voltage-dependent cation current in rat _{FASEB J. 14 (2000) A107.}

neocortical neurones, J. Physiol. 479 (1994) 199–205. _{[21] G.G. Somjen, M. Muller, Potassium-induced enhancement of persis-¨}

1

[2] C. Alzheimer, P.C. Schwindt, W.E. Crill, Modal gating of Na _{tent inward current in hippocampal neurons in isolation and in tissue}

1

channels as a mechanism of persistent Na current in pyramidal _{slices, Brain Res. 885 (2000) 102–110.}

neurons from rat and cat sensorimotor cortex, J. Neurosci. 13 (1993) _{[22] K.G. Specht, M.A.J. Rodgers, Plasma membrane depolarization and} 660–673. _{calcium influx during cell injury by photodynamic action, Biochim.} [3] P.H. Barry, JPCalc, a software package for calculating liquid _{Biophys. Acta 1070 (1991) 60–68.}

junction potential corrections in patch-clamp, intracellular, epithelial 1

[23] C.P. Taylor, Na currents that fail to inactivate, Trends Neurosci. 16 and bilayer measurements and for correcting junction potential _{(1993) 455–460.}