GLIA 41285-292 (1991)

Potassium Channels in Crustacean Glial Cells CHRISTIAN ERXLEBEN

Department of Biology, University of Konstanz, 7750 Konstanz, Federal Republic of Germany

KEY WORDS

Ion channels, Phosphorylation, Modulation

ABSTRACT Unitary currents through single ion channels in the glial cells, which ensheath the abdominal stretch receptor neurons of the crayfish, were characterized with respect to their basic kinetic properties. In cell-attached and excised patches two types of Ca++-independentK+ channels were observedwith slope conductancesof 57 pS and 96 pS in symmetrical K+ solution. The 57 pS K+ channel was weakly voltage-dependent with a slope of the Po vs. membrane potential relationship of +95 mV for an e-fold change in Po.In addition to the main conductance level, the channel displayed conductance levels of 80 and 109 pS. In excised patches, channel activity of this “subconductance”K+ channel showed “rundown” that could be prevented with 2 mM ATP-Mg on the cytoplasmic side of the membrane. The 96 pS K+ channel was strongly voltage-dependent with a slope of + 12 mV for an e-fold change in Po.Averaged single-channelcurrents elicited by voltage jumps proved the channel to be of the delayed rectifying type. Channel activity persisted in excised patches with minimal salt solution and in virtually Ca”-free saline. Because of its dependence on intracellular ATP-Mg,the subconductanceK+ channel is discussed as a target of modulation by transmitters or peptides via phosphorylation of the channel.

Moreton, 1988)and molluscan glial cells (Geletyuk and Kazachenko, 1989). It is becoming increasingly clear that glial cells are of Ion channels in vertebrate glia (Barres et al., 1990) great importance for the function and regulation of have mostly been studied in primary cell cultures and neural activity. Because of the high potassium conduc- may well be physiologically different from those in the tance of glial cells and their morphological properties, intact tissue. This is especiallytrue for glial cells, which they have been suggested to function as a spatial buffer (in situ) have extensive interaction with other glial and system (Ballanyi et al., 1987; Coles and Tsacopoulos, nerve cells. Indeed, electrophysiological properties of 1979; Orkand et al., 1966). Application of the patch- astrocytes studied in brain slices differ from those of clamp technique, initially on oligodendrocytes (Ketten- their counterparts in culture (Walz and MacVicar, mann et al., 1982, 19841, permitted studies on regional 1988), where the expression of ion channels is greatly distribution of K+channels, which led to a special model affected by culture conditions (Barres et al., 1989). of potassium homeostasis termed potassium siphoning In this study, properties of ion channels from the glial (Newman,1984;Newman et al., 1984;Nilius and Reich- cells associated with the abdominal stretch receptor enbach, 1988). Even though the electrophysiological neuron of the crayfish Orconectes limosus were investiproperties of glial cells were initially characterized in gated to provide more insight into the function of in situ the nervous system of the medicinal leech (KuMler and Potter, 1964), single-channel investigations were restricted mainly to vertebrate preparations. Exceptions Received July 26,1990; accepted September 11,1990. are recent reports on K+ channels from glial cells of the Address re rint requests to Dr. Christian Endeben, Fakultat fur Biologie der cockroach nervous system (Leech, 1986; Leech and Universitat, Jostfach 5560, 7750 Konstanz, FRG.

INTRODUCTION

0 1 9 9 1 Wiley-Liss, Inc.

286

ERXLEBEN

peripheral invertebrate glia. The glial sheath of the receptor neurons consists of flat sheets of cells often only a few microns thick (Nadol and de Lorenzo, 1968, 1973;Tao-Cheng et al., 1981).Little is known about its function or the electrophysiological properties of these cells, since conventional intracellular recordings with microelectrodes were hampered by their morphology. Preliminary accounts of some of this work have been presented in abstract form (Erxleben, 1989a). MATERIAL AND METHODS Preparation

were those of Hamill et al. (1981) with the patch electrodes (diameter about 2 p.m) coated with polystyrene (Q-Dope,GC Electronics)or Sylgard 184(Dow Corning). Data were either stored on video tape with a modified PCM recorder (Bio Logic) for later analysis or directly digitized by a 12 bit A/D converter (1401 labinterface, CED Cambridge)in connection with a PC (Tandon PCA 40). Data were filtered by a 6 pole bessel filter at frequencies indicated in the figures. The sampling rate was 5 times the filter frequency. Data acquisition and analysis were performed by a commercially available patch-clamp program (CED). Values are given as mean ? standard deviation, where appropriate. Single-channel current amplitudes and dwell times were determined using the 50% amplitude level for detection of channel openings. Steady-state open probabilities (Po),i.e., the percentage of time that the channels spent in the open state, were calculated by adding the channels’measured open times during 5 to 60 s of activity (dependingon the opening frequency of the channel) divided by the sampled time. Unless otherwise noted, downward deflections in the current traces correspond to inward currents. In preparations with high activity of voltage-independent K+ channels, the cells’ resting membrane potential was determined by switching to current clamp. This is equivalent to measurements with an intracellular microelectrode with the electrode resistance determinedby the open channel conductance. With the typical seal resistance obtained in this preparation (50 to 200 GLR), membrane potential measurements “through a 100pS K+ channel should be accurate to within 1%.

Crayfish (0.Zimosus)were obtained from a commercial supplier (Fischerei Liptow, Berlin) and kept at 13” to 17°C in running fresh water. Whole stretch-receptor organs with the surrounding glial cells of the 2nd to 5th abdominal segments of the animals were used. The preparation was pinned at the approximate in situ length in a Sylgard 184 (Dow Corning)lined experimental chamber and submerged in saline. Experiments were performed at room temperature (20-24°C). To obtain Giga-Ohm seals, the preparation was treated locally with protease (10 mg/ml, type XIV; Sigma).A polished pipette of about 25 pm opening was filled with saline containing protease and was positioned onto the receptor cell. An exposure of 5 min was sufficient to make the surface accessible for patchclamp recordings. Patches were made in the regions of the cell body, the initial segment, and the primary dendrites of the receptor neurons. In order to determine if the protease treatment affected the channels under investigation, some preparations were alternatively incubated for 30 min with collagenase(0.5mg/ml, type IA; RESULTS Sigma). Collagenase treatment with 2 mg/ml for up to 60 min has been shown not to affect the kinetics of single Patches on the glial cells were distinguished from glutamate-gated synaptic channels at the crayfish neu- those on the stretch receptor neuron (Erxleben,1989b) romuscular junction (Franke and Dudel, 1987). on the basis of their electrical properties in the wholecell current clamp mode. In contrast to the neurons, the glial cells were not electrically excitable in response to Solutions depolarizing current pulses. Furthermore, voltage reFor cell-attached and outside-out recordings, van sponses showed a fast rising phase (steeper than expoHarreveld (1936) saline with (mM) 195 NaC1, 5.4 KCl, nential or error function)and gave large apparent input 13.5CaCl,, 2.6 MgCl,, and 10 TRIS was used. Solutions resistances. The latter is typical for single electrode in the patch pipettes (outside) were either van Harre- measurements of sheet-like cellular structures (Woodveld diluted t o 90% or high K+ saline with (mM) 160 bury and Crill, 1961;Jacket al., 1975)and is consistent KCl, 13.5 CaCl,, 2.6 MgCl,, and 10 HEPES. High K+ with the morphology of the glial cells in this preparasolution (cytoplasmic side) in the pipette or bath was tion. In the glial cells, two types of channels were identi(mM) 160 KCI, 1.5 EGTA, 1MgCl,, 1CaCl, (0.1 pM free fied. They were observed with either van Harreveld or Ca++)or 5 EGTA and 0 CaC1, in some experiments,and 10 HEPES. The pH of all solutions was adjusted to 7.4. high K+ solution in the pipette. Their basic properties are demonstrated in recordings of single-channel activity during voltage ramps (Fig. 1).Both channels, later shown to be K+-selective,have similar single-channel Electrophysiological Techniques and conductances and fast gating, but the dependency of the Data Analysis open probability on the membrane potential differs Single-channelcurrents were recorded with an EPC-7 greatly. One channel (Fig. 1A) shows distinct multiple (List Electronic) patch-clamp amplifier in the cell-at- conductance levels, particularly evident from the tached or excised configuration. Recording techniques “banding pattern” of the records around 0 mV holding

287

INVERTEBRATE GLIAL POTASSIUM CHANNELS

A

Subconductance K+ Channel

-100

-50

0

+ 50

+I 00

[mVI

B

By far the most frequently observed channel in the crustacean glial cells (found in 68% of all patches) was the subconductance channel (Fig. lA), with usually two or three active channels in a patch. The channel displayed several current amplitude levels. Activity occurred in bursts with intervals of up to tens of seconds. During the bursts, rapid transitions between three preferential states were observed. The current amplitudes of these levels were not multiples of any obvious subunit. Transitions to smaller amplitude levels were much less frequent. Occasionally, bursts with openings exclusively to one conductance level were seen (Fig. 2). Probabilities for the channels’ conductance levels were determined from patches that contained single active channels (Fig. 2). Gaussian distributions were fitted to the peaks of amplitude histograms (Fig. 3B) and the probabilities derived from the relative areas. The example shows three clearly visible open levels. Additional “bumps”in the histogram between the closed level and the most frequent conductance level indicate the existence of other poorly resolved and less frequent conductance levels. Records selected to show openings to all conductance levels are shown in Fig. 3A. In all but 1of 11patches, the channel spent most of its open time at the first conductance level (Ll, 74 2 lo%, n = ll), followedby the second level (L2,25 lo%, n = 11).In 4 of these 11patches, enough openings to a third conductance level occurred for quantitative analysis (L3, 3.3 i 1.5%,n = 4). The channel’s probability of being open was only weakly voltage-dependent. For an e-fold change in Po, a +95 -+ 7 mV (n = 3) shift in membrane potential was required. The conductance of the most frequent levels (L1 to L3), measured in eight patches for outward K+ currents, was found to vary widely: 57 pS, range 37-76 pS for the most frequent level L1; 80 pS, range 59-118 pS for L2; 109 pS, range 62-195 pS for L3. The large range of conductances is not due to experimental error (they are specific for this channel-compare with the values obtained for the outward-rectifying K’ channel), but rather to the difficulty of attributing a measured conductance value to any particular conductance state. In cell-attachedpatches with van Harreveld saline in the pipette, single-channel currents reversed near the cell’s resting potential (Fig. 4A). The resting potential (-65 2 4.7 mV, n = 15) was determined either as the whole cell zero current potential after the seal was ruptured, or “through the open subconductance channel (with high K+ saline in the pipette) in the current clamp mode (see Materials and Methods). Assuming internal activities of K+ and C1- in the glial cells similar to those determined for the stretch receptor neuron (Brownet al., 1978; Moser, 1985),this suggests that the channel is either K+- or Cl--selective. With high K+ solution in the pipette, currents reversed some 70 mV positive to the cell’s resting potential, demonstrating that the channel is selective for K+ but not C1- (Fig.4B). This was confirmed in outside-out patches (see below)

*

Fig. 1. Quasi steady-state I-V relationships for sin le channel currents through the subconductanceK+ channel (A) anf outward-rectifying K+ channel (B). A Outside-out patch with van Harreveld in the bath solution and high K’ saline t 2 mMATP-M in the ipette; single channel in the patch; four sweeps superimposed.%:Ins&-out patch in symmetrical hi h K’ saline. Note increase in channel activity with depolarization. channels in the patch; three sweeps superimposed. The dotted lines mark the zero current levels with the slope indicating the seal resistances, here 100 Gn; voltage ramps from -100 mV to +lo0 mV with 1 s duration; 3 KHz bandwidth.

Lo

potential. Hence, this channel will subsequently be referred to as the subconductance channel. The second channel (Fig. 1B) shows very little activity at negative membrane potentials but the open probability is close to unity above +lo0 mV. This channel will be referred to as the outward-rectifying channel. In 8%of all patches, no channels were seen in either cell-attached configuration or after excision (inside-out) into saline with physiological Cat concentrations. In these patches as well as most others, a large number of Ca++-activatednonselective cation channels (to be described elsewhere) appeared with cytoplasmic Ca” exceeding some 10 kM. No appreciable difference of the channels’ characteristics,in particular their voltage dependence and singlechannel conductance, was seen between preparations treated locally with protease and those incubated in collagenase, the more widely used method of enzymatic treatment. Only for the subconductance K+ channel was a somewhat lower probability for conductance states other than the main state noticed. +

288

ERXLEBEN

11.6 2 3.2 mV (n = 8) depolarization. Upon step changes in membrane potential, time-dependent relaxation of the ensemble currents was observed (Fig. 71, indicating that the channel is a delayed rectifier. Current through the delayed rectifying K+ channel reversed close to the cell’s resting potential with van Harreveld saline in the pipette, at +69 2 6 mV (n = 6) with high K+ solution, or at 0 mV in the excised patch and symmetrical K+ saline (Fig. 8). The channel was thus concluded to be K+-selective.For symmetrical K+ solutions, the average slope conductance was 96 12 pS (n = 6). No obvious change in the channel’s activity was noticed in virtually Ca++-free(5 EGTA, 0 CaC1, added) cytoplasmic solution.

*

Fig. 2. Continuous traces (1s) with inward K+ current through the subconductance channel. Note traces 6 and 7 from the to with openingsonly to the large conductance level. Cell-attached patckat the cell’s resting potential (-68 mV) with high K+ saline in the pipette; inward current shows as upward deflection; 3 KHz bandwidth.

where a change from external van Harreveld solution to high K+ saline led to a shift in the current reversal consistent with that expected from the Nernst equation. One rather surprising feature of the subconductance channel was noticed upon excision of patches. Channel activity decreased gradually and, within 5 min, only brief openings (< 0.1 ms) remained, much like the “rundown” or “washout”that has been described for a number of other channels, in particular Ca++ channels (Byerly and Hagiwara, 1988). This process is often B 25000 attributed to dephosphorylation of the channel (or closely associated proteins) and can be prevented or at least slowed down by including ATP-Mg in the solution 20000 on the cytoplasmic side of the membrane. Therefore, channel activity was compared in excised outside-out patches with or without ATP-Mg in the pipette solution. In patches with no ATP-Mg (n = 6),channel activity ran 3a 15000 down within 5 min (Fig. 5). In the patches with 1-2 mM w?J ATP-Mgpresent (n = 8),however, no rundown occurred +I. 10000 in five patches, with the activity persisting as long as the patches (up to 60 min). In two cases, ATP-Mg initially preserved channel activity, but then the chan5000 nels abruptly stopped opening. This phenomenon could also be seen in cell-attached patches and may be due to degradation or loss of the channel. 0 -2 0 2 amplitude 4 [pAI 6 8 In contrast to this dependence of channel activity on internal ATP-Mg, no dependence on the cytoplasmic Ca++concentration was seen. Fig. 3. Records (A) and amplitude histogram (B)of single-channel

n

I\

Outward-Rectifying K+ Channel Some 24% of all patches contained channels that responded to depolarization with an increase in open probability(Figs. lB, 6). The percentage of time spent in the open state was found to increase e-fold for a

activity of the subconductance channel in an outside-out patch. A: Traces selected to show the range of conductance levels as indicated by the dotted lines; traces filtered at 3 KHz. B: Histogram from 60 s of channel activity. The large peak at zero current corresponds to = 0.50). The dotted line is a fit of gaussian the closed channel (PClased distributions to the data: open levels L1, L2, and L3 at 4.4,6.3,and 7.75 r A with o en robabilities of 0.24,0.13,and 0.02, respectively.Smaller evels w !be El (bumps) make up for the remaining 11%. Excised outside-out atch at +60 mV in symmetrical high K’ saline (+Z mh4 ATP-Mg in t\e pipette).

289

INVERTEBRATE GLIAL POTASSIUM CHANNELS

d-50

50

0

I

150

100

HP (mV)

0.9-

0.8-

0.70.6-

2 0.5-

Po=l /(I +exp(-s*V-k))

0.4-

0.3-

0.24

6

0.1

0’ -50

---

0

100

50

0

HP (mV) 0

50

150

100 HP (mV)

Fi . 6 Steady-state open probability (Po) of the outward-rectifym

200

K’ caannel as a function of the membrane potential. A Single-channe!

Fig. 4. I-V relationships for the subconductance channel in cellattached patches with van Harreveld (A) and hi h K+ saline (B) in the pipette. A: Outward K’ currents. B: Inward ancfoutward K+ currents; reversal potentials +76, +74, and 1-62 mV for current levels L1, L2, and L3, respectively. Potentials are plotted relative to the cell’s membrane resting potential of,about -65 mV. Currents fitted by linear regression with the slopes indicated in the figure.

0

40

80

120

160

200

240

time (see) Fig. 5. Activity of the subconductance channel in excised patches persists in the presence of 2 mMATP-M (s uares) and shows washout in minimal high K’ saline without iT8-Mg (circles). Each point represents channel open probabilities (P ) determined from 10-30 s of channel activity; both patches mthvan garreveld saline on the outside at 0 mV holding potential.

o enin s otentials shown with each trace; dotted lines indicate the clanne!’j&sed level; 3 KHz bandmdth. B: P determined from 30-60 s sequences as shown in A. The line through tbe points is a Boltzmann curve fitted to the data, withV being the membrane potential and s the voltage sensitivity (+15.3 mV for an e-fold chan e in membrane otential; k = -5.5). Potentials are relative to the cil’s resting mem!ran, potential; cell-attached patch with high K+ saline in the pipette; current reversal +66 mV.

d

I

Fi 7. Dela ed rectification of the outward-rectifyingK’ channel. Singfe-channercurrents (middle trace) were elicited by 50 ms voltage umps from + l o 0 to +150 mV (lower trace), averaged ( ~ 2 0 1 and , eak-subtracted to give the average current (top trace). Cell-attached patch with high K+ saline in the pipette; otential relative to the cell’s membrane resting potential; in the mid& trace, outward current is downward; current calibration is 10 PA for the top and 5 PA for the middle trace; same patch as Fig. 6.

I

290

ERXLEBEN

--

0.2 ms at the cell's resting potential to 4 ms when depolarizedby 130 mV. For the same voltage range, two exponentials ranging from 250 to 0.2 ms are required to fit the closed times. The dependence of open and closed times on the membrane potential is shown in Fig. 9C. The data suggest that the outward-rectifying K+ channel has at least one open and two closed states.

I

DISCUSSION -12

-50

50

0

150

100 HP (mV)

200

Fig. 8. I-V relationshi s for the outward-rectifying K' channel in cell-attached patches wit{ van Harreveld (open squares) and high K+ saline (closed squares) in the pi ette. Current reversal with uasi symmetricalK+ saline is indicate: by the arrow. Potentials are,plotted relative to the cell's membrane resting potential. Currents fitted by linear regression with the slopes indicated in the figure.

Records of the outward-rectifying K+ channel (Fig. 6A) show that channel openings occur in bursts. This behavior is reflected in the dwell time distributions (Fig. 9A,B). The open time distribution of the channel can be described by a single exponential ranging from

0.14

0

The present investigation demonstrates the presence of two prominent Cat +-independentK+ channels in the glial cells of the peripheral nervous system of the crayfish. Two previous reports on invertebrate glial cells have also described Ca++-independentK+ channels: a fast flickering 100 pS channel with little voltage dependence in glial cells of the central nervous system of the cockroach (Leech, 1986;Leech and Moreton, 1988)and a 100 pS channel in the central nervous system of the mollusc Lymneu stagnulis (Geletyuk and Kazachenko, 1989). Its voltage dependence (about +60 mV for an e-fold change in Po)is comparable to that of the subconductance K+ channel presented here. Interestingly, the molluscan K+ channel also displays multiple conductance states.

0.014 2

4

6

8

10

12

14

16

closed time (msec)

-50

0

50 HP (mV)

100

I

150

Fig. 9. Dwell time distributions for the outward-rectif 'ng K+ channel and their dependence on membrane otential CloseGA) and open (B) time distributions at +90 mV witg T~~ = 0134 ms, T~ - 4.3 ms, and 'T,~~,, = 1.9 ms; the continuous curves were calcufaced by a non-linear regression of the data to double exponentials (closed times) and a single exponential (open times). C: Dwell times as a function of membrane potential. Re ession lines are least square fits to the data. Cell-attached patch wighigh K+ solution in the pipette; potentials relative to the cell's resting potential.

0.14

0

2

4

6

8

10

open time (msec)

12

14

16

INVERTEBRATE GLIAL POTASSIUM CHANNELS

In the crustacean glial cells as well as other invertebrate glia (Geletyuk and Kazachenko, 1989; Leech, 1986; Leech and Moreton, 1988) the predominant K+ channels seem to be only weakly voltage-dependent. For vertebrate glia, K+ channels with comparable voltage sensitivity have been reported in oligodendrocytes by some groups (Kettenmann et al., 1982,1984;McLarnon and Kim, 1989),whereas others find inwardly rectifying K+ channels to be predominant (Barres et al., 1988). Outward-Rectifying K+ Channel A specificfunction of the outward-rectifyingK+ channel is not obvious, since it is not known if the glial cells ever depolarize to such an extent that the channel will contribute significantly to the membrane conductance. They are, however not unique to the crustacean glial cells and similar conductances/channels have been described in various vertebrate glial cells (Barres et al., 1990). Subconductance K’ Channel The subconductance K+ channel is more likely to be of significance for glial cell function. First, because it is the predominant channel in this preparation and the only channel with appreciable open probability at the cell’s resting membrane potential, it probably determines the latter. Second,the ability of internal ATP-Mgto prevent rundown of channel activity suggests that the subconductance K+ channel depends on phosphorylation. A stimulation of channel activity by ATP binding-unprecedented, but similar to the inhibition of activity by ATP that can be observed in a class of ATP-sensitiveK+ channels from a variety of cells (Ashcroft, 1988tseems less likely. In this context, it is worth pointing out that the subconductance K+ channel is not sensitive to changes in cytoplasmic Ca++ or Mg++,both of which decrease the frequency of channel openings of ATPsensitive K+ channels (Ashcroft and Kakei, 1989;Kakei and Noma, 1984).Further experiments, which are beyond the scope of this paper, are required to prove that phosphorylation of the channel is actually involved. In neuronal Ca++ and K+ channels, modulation by neurotransmitters and peptides is frequently mediated by phosphorylation of the channels (Levitan,1988).For invertebrate glial preparations, there is also considerable evidence for up- and down-regulation of K+ channels. In leech neuropile glia, serotonin increases the K+ permeability (Walz and Schlue, 1982). In glial cells forming the blood-brain barrier of the cockroach, octopamine reduces the K+ permeability, possibly through stimulation of the adenylate cyclase (Schofield and Treherne, 1986).Similarly, long-lasting hyperpolarizations of Schwann cells of the squid giant axon are mediated by a K+ conductance activated by acetylcholine Nillegas. 1975). octopamine (Reale et al., 19861, and variocs neuropeptides (Villegas et al., 1988).All of

291

these responses appear to be mediated by increasing intracellular levels of cyclic AMP (Evans et al., 1985). In all preparations mentioned above, the channels that are subject to modulation were not identified in single-channel recordings, but they appear to be those responsible for the membrane resting potential, commonly classified as “background K+ channels. The subconductance K+ channel presented here is such a background K+ channel, since it is only weakly voltagedependent and open at the normal resting potential. While any connection between the subconductance channel in the crayfish glial cells and its modulation by transmitters or peptides is speculative at this time, the present data show that the last part of the cellular machinery for such a pathway-an ion channel that can be regulated by phosphorylation-does exist. Such regulation of the subconductance channel’s activity might play a role in the K+-clearanceproperties of the glial cells by coupling high K+ permeability to nervous activity, as discussed by Walz (1988) for the serotonin response of leech neuropile glial cells. Multi-Barreled K+ Channel or Conductance Substates? The properties of the subconductance channel presented in this paper argue for “true” conductance substates rather than a cluster of coupled, conducting, identical units like the two-barreled C1- channel from Torpedo electroplax(Miller,1982),the large anion channel from alveolar epithelial cells with seven conductance levels (Krouse et al., 19861,or the multi-barrelled K+ channel in renal tubules (Hunter and Giebisch, 1987). The current amplitudes of the subconductance K+ channel are not integer multiples of any obvious subunit nor the result of superposition of a “large”and a “small” channel, since the small channel cannot be expected to open almost exclusively on top of the large one (Figs. 1 3 ) . Furthermore, transitions between conducting states seem to be governed by different rate constants, as seen in the current records. Transitions between the large conductance levels are very fast and often only visible as channel flicker (Figs. lA, 3A). ACKNOWLEDGMENTS I thank Dr. E. Florey for his continuing interest and support, Dr. A. Wenning for comments on the manuscript, and M.A. Cahill for editorial assistance. This work was supported by the Deutsche Forschungsgemeinschaft (SFB 156A31). REFERENCES Ashcroft, F.M. (1988)Adenosine 5‘-trisphosphate-~ensitive potassium channels. Annu. Reu. Neurosci., 11:97-118. Ashcroft, F.M. and Kakei, M. (1989)ATP-sensitiveK’ channels inrat pancreatic p-cells: Modulation by ATP and Mg2’ ions. J. Physzol., 416:349-367.

292

ERXLEBEN

Ballanyi, K., Grafe, P., and Ten Bruggencate, G. (1987) Ion activities Leech, C.A. and Moreton, R.B. (1988) Fast flickering of a potassium channel in lial cells from the cockroach central nervous system. and potassium uptake mechanisms of glial cells in guinea-pig olfacComp. Bwcfem. Ph siol., 9OA297-301. tory cortex slices. J. Ph siol ,382:159-174. Barres, B.A., Chun, L.L.$, and Corey, D.P. (1988)Ion channel expres- Levitan, I.B. (1988) dodulation of ion channels in neurons and other cells.Annu. Rev. Neurosci., 11:119-136. sion by white matter glia: I. Type 2 astrocytes and oligodendrocytes. McLarnon, J.G. and Kim, S.U. (1989) Single channel potassium curGlia, 1:10-30. rents in cultured adult bovine oligodendrocytes.Glia, 2:298-307. Barres, B.A., Chun, L.L.Y., and Corey, D.P. (1989)Acalciumcurrent in cortical astrocytes: Induction by CAMPand neurotransmitters, and Miller, C. (1982) Open-state substructure of single chloride channels from Torpedo electroplax. Philos. Trans. R. SOC. Lond. B, permissive effect of serum factors. J. Neuroscz., 9:3169-3175. 299:401411. Barres, B.A., Chun, L.L.Y., and Corey, D.P. (1990) Ion channels in Moser, H. (1985)Intracellular pH re lation in the sensory neurone of vertebrate glia. Annu. Rev. Neurosci., 13:441474. the stretch receptor of the crayfis8Astacus fluuzatzlis). J. Physiol., Brown, H.M., Ottoson, D., and R dqvist, B. (1978) Crayfish stretch 362:23-38. receptor: An investigation wit{ voltage-clamp and ion-sensitive Nadol, J.B. and de Lorenzo A.J.D. (1968)Observations on the abdomelectrodes. J. Physiol., 284:155-180. inal stretch receptor and the fine structure of associated axo-denByerl , L. and Hagiwara, S. (1988) Calcium channel diversity. In: dritic synapses and neuromuscular junction in homarus. J . Comp. Cazium and Ion Channel Modulation. A.D. Grinnel, D. Armstrong, Neurol., 132:419-444. and M.B. Jackson, eds. Plenum Press New York, pp. 3-18. Coles, J.A. and Tsacopoulos, M. (1979)Potassium activit in photore- Nadol, J.B. and de Lorenzo,A.J.D. (1973)Observations on the organization of the dendritic processes and receptor terminations in the ceptors, $ial cells and extracellular space in the &one retina: abdominal muscle receptor organ of homarus. J . Comp. Neurol., Changes uring photostimulation. J . Physiol., 290525-549. 137:19-58. Erxleben, C. (1989a)ATP-dependentK channel with multiple conducNewman, E.A. (1984)Regional specializationof retinal glial cell memtance substates in crustacean glial cells. Biophys. J., 55:542a. brane. Nature, 309:155-157. Erxleben, C. (1989b) Stretch-activated current through sin le ion channels in the abdominal stretch receptor organ of the crayfish. J. Newman, E.A., Frambach, D.A., and Odette, L.L. (1984) Control of extracellular potassium levels by retinal glial cell K' siphoning. Gen. Ph siol, 94:1071-1083. Science, 225:1174-1175. Evans, P . 6 , Reale, V., and Villegas, J. (1985) The role of cyclic nucleotidesin modulationof the membrane otential of the Schwann Nilius, B. and Reichenbach, A. (1988)Efficient K' buffering by mammalian retinal glial cells is due to cooperation of specialized ion cell of the squid giant nerve fibre. J. Physiof, 363:151-167. channels. Pfliigers Arch., 411:654-660. Franke, C. and Dudel, J. (1987) Single glutamate- ated synaptic channels at the crayfish neuromuscular junction. f'flugers Arch., Orkand, R.K., Nieholls, J.G., and Kuf€ler S.W. (1966) Effect of nerve 408:30&306. impulses on the membrane otential of glial cells in the central Geletyuk, V.I. and Kazachenko, V.N. (1989) Single potential-depennervous system of amphibia. Neurophyszol., 29:78%806. dent K' channels and their oligomers in molluscan glial cells. Reale, V., Evans, P.D., and Ville as, J. (1986)OctopaminergicmodulaBiochim. Biophys. Acta, 981:343-350. tion of the membrane potentiat of the Schwann cell of the squid giant Hamill, O.P., Marty, A., Neher, E., Sakmann, B., and Sigworth, F.J. nerve fibre. J. Ex .Bzol., 121:421-443. (1981)Improved patch-clam techniques for high-resolution current Schofield, P.K. antTreherne, J.E. (1986) Octo amine reduces potasrecording from cells a n 8 cell-free patches. Pflugers Arch., sium permeability of the glia that form t i e insect blood-brain 391:85-100. barrier. Brain Res., 360:344-348. Harreveld, A. van (1936) A hysiological solution for fresh water Tao-Cheng, J.-H., Hirosawa, K., and Nakajima, Y. (1981)UltrastrucExp. %iol. Med., 34:428-432. crustaceans. Proc. SOC. ture of the crayfish stretch receDtor in relation to its function. J. Hunter, M. and Giebisch,G. (1987)Multi-barreled K channels in renal Comp. Neur01.,"200:1-21. tubules. Nature, 327522424. Ville as, J. (1975) Characterization of acetylcholine receptors in the Jack, J.J.B., Noble, D., and Tsien, R.W. (1975)ElectricCurrentFlow in Scfwann cell membrane of the sauid nerve fibre. J. Phvsiol.. " , Excitable Cells. Clarendon Press, Oxford, pp. 83-97. 249:679-689. Kakei, M. and Noma, A. (1984) Adenosine 5'-tri hos hate-sensitive Villegas, J., Evans, P.D., and Reale, V. (1988) Electrophysiology of ptassium channel in the atrioventricular no& ceg of the rabbit Schwann cell receptors. In: Glial Cell Receptors. H.K. Kimelberg,ed. eart. J . Ph siol ,352:265-284. Raven Press, Ltd., New York, pp. 141-157. Kettenmann, Orkand, R.K., Lux, H.D., and Schachner,.M. (1982) Walz, W. (1988) Ph siolo 'cal come uences of activating serotonin Single otassium channel currents in cultured mouse oligodendroreceptors. In: GliaTCell i%ceptors. 2.K. Kimelberg, ed. Raven Press, cytes. Ifeurosci. Lett., 32:41-46. Ltd., New York, pp. 121-130. Kettenmann, H., Orkand, R.K., and Lux, H.D. (1984)Some pro erties Walz, W. and Schlue, W.R. (1982)Ionic mechanism of h er olarizing of single potassium channels in cultured oligodendrocytes.Pftugers 5-hydroxytryptamineeffect on leech neuropile glial cef'ps.Aain Res., Arch., 400:215-221. 25O:lll-121. Krouse, M.E., Schneider, G.T., and Gage, P.W. (1986)A large anion- Walz, W. and MacVicar,B.A. (1988)Electro hysiological roperties of selective channel has seven conductancelevels. Nature, 319:58-60. $ial cells. Comparison of brain slices wit{ primary cuiures. Brain Kuf€ler, S.W. and Potter, D.D. (1964)Glia in the leech central nervous es., 443:321-324. s stem: Physiological properties and neuron-glia relationship. J. Woodbury,, J.W. and Crill, W.E. (1961) On the ,problem of impulse euro hysiol., 27:29&320. conduction in the atrium. In: Nervous Inhibition. E. Florey, ed. Leech, $A. (1986)Patch-clamp recording of potassium channels from Plenum Press, New York, pp. 124-135. glial cells in the cockroach central nervous system. J . Exp. Biol., 122:443-446.

5.

fI:,

K

Potassium channels in crustacean glial cells.

Unitary currents through single ion channels in the glial cells, which ensheath the abdominal stretch receptor neurons of the crayfish, were character...
839KB Sizes 0 Downloads 0 Views