Brain Research, 568 (1991) 116-122 © 1991 Elsevier Science Pubhshers B.V. All rights reserved. 0006-8993/91/$03.50
116 BRES 17283
Effects of doxapram on ionic currents recorded in isolated type I cells of the neonatal rat carotid body C. Peers Department of Pharmacology, Leeds University, Leeds (U.K.) (Accepted 6 August 1991)
Key words: Doxapram; Carotid body; Chemoreceptor; Type I cell; Potassium current; Calcium current
Whole-cell patch-clamp recordings were used to investigate the effects of the respiratory stimulant doxapram on K ÷ and Ca2+ currents m isolated type I ceils of the neonatal rat carotid body. Doxapram (1-100/zM) caused rapid, reversible and dose-dependent inhibitions of K + currents recorded in type I cells (ICso approximately 13/zM). Inhibition was voltage-dependent, m that the effects of doxapram were maximal at test potentaals where a shoulder in the current-voltage relationship was maxamal. These K + currents were composed of both Ca2+activated and Ca2 + -mdependent components. Using high [Mg2 + ], low [Ca2 + ] solutions to inhibit Ca2+-activated K + currents, doxapram was also seen to directly inhibit Ca2+-independent K + currents. This effect was voltage-independent and was less potent (ICs0 approximately 20 /tM) than under control conditions, suggesting that doxapram was a more potent inhibitor of the Ca2÷-activated K + currents recorded under control conditions. Doxapram (10/zM) was without effect on L-type Ca2+ channel currents recorded under conditions where K + channel activity was minimized and was also without significant effect on K + currents recorded in the neuronal cell hne NG-108 15, suggesting a selective effect on carotid body type I cells. The effects of doxapram on type I cells show similarities to those of the physiological stimuh of the carotid body, suggesting that doxapram may share a similar mechanism of action in stimulating the intact organ. INTRODUCTION The carotid b o d y is the m a j o r peripheral arterial chemoreceptor. It responds to changes in levels of arterial 0 2, C O 2 and p H by modulating the frequency of discharge of afferent chemosensory fibres in the carotid sinus nerve (CSN), thereby exerting influences on the pattern of breathing 6. The type I (or giomus) cells are generally believed to play an i m p o r t a n t role in this organ's functioning; they lie in synaptic contact with afferent chemosensory fibres 14 and contain catecholaminefilled vesicles. The close correlation in b o t h time course and magnitude between catecholamine release from type I cells and elevation of CSN discharge 7'26 has contributed to the idea that type I cells are responsible for detecting and transducing chemostimuli. The patch-clamp technique has recently been applied to the study of isolated type I cells, in o r d e r to determine the possible roles ion channels may play in chemotransduction 4'9'12'25'2s. In most cases, physiological stimuli (e.g. hypoxia, acidity) have b e e n demonstrated to selectively inhibit K + channels, whilst other channel types (Ca 2÷ channels, and also N a ÷ channels in
adult type I cells) remain unaffected 3'9'12'13'19'20'23'28. In type I cells isolated from the neonatal rat at least, the effects of hypoxia and acidity are selective for charybdotoxin-sensitive (high conductance) Ca2+-activated K ÷ channels 19,20,23. D o x a p r a m hydrochloride (1-ethyl-4-(2-morpholinoethyl)-3,3-diphenyl-2-pyrrolidinone hydrochloride hydrate, hereafter referred to as d o x a p r a m ) is a drug used clinically to stimulate breathing in a variety of conditions. Its stimulatory action has b e e n shown in anaesthetized animals to be due largely to a direct action on the carotid b o d y 1°'16, where it causes an increase in the discharge of afferent chemosensory fibres in the CSN, although at higher doses it can also stimulate central respiratory neurones. The mechanism of d o x a p r a m ' s action on the carotid b o d y is at present unknown, but appears to be i n d e p e n d e n t of the b a c k g r o u n d levels of O2 x6. The present study was carried out to examine whether doxa p r a m alters the activity of ion channels in isolated type I cells of the neonatal rat, in o r d e r to c o m p a r e its effects with those of physiological stimuli (see above). A preliminary account of some of these findings has been presented in abstract form 22.
Correspondence. Dr Chns Peers, Department of Pharmacology, Worsley Mechcal and Dental Bmldmg, Leeds University, Leeds, LS2 9JT, U.K.
117 MATERIALS AND METHODS The procedures for isolation and maintenance of type I cells in short-term culture were as previously described 19"25. In brief, carotid bodies were removed from 8- to ll-day-oid rats (which were anaesthetized by breathing 4% enflurane, 96% oxygen through a face mask) and placed in ice-cold phosphate-buffered saline (PBS). They were then cut into pieces and placed in Ca 2+ and Mg3+-free PBS containing collagenase (0.03-0.05%) and trypsin (0.020.025%) at 37"C for 20-30 rain. After gentle trituration with a Pasteur pipette, the dispersed cells were resuspended in Ham's F-12 culture medium containing 10% fetal calf serum, penicillin (100 IU/rni), streptomycin (100/~g/ml) and insulin (84 U/l). They were plated onto polylysine-eoated coverslips and kept at 37"C in a humidified incubator until required for study. Electrophysiological recordings were made from cells kept in short-term culture for between 4 and 48 h. In some experiments, the effects of doxapram were investigated on ionic currents in the undifferentiated mouse neuroblastoma x rat giioma hybrid cell line NG-108 15. These cells were maintained in culture as previously described24, and recordings were made as for the type I carotid body cells. Fragments of coverslips to which cells had adhered were transferred to a recording chamber (volume approximately 80/zl) which was continually perfused at 0.5 ml/min. Whole-cell patch-clamp recordingss were made from the most numerous cell type, the phasebright, spherical type I cells, which were of approximately 10/~M diameter~. Patch electrodes had resistances of between 2 and 7 MQ. To record K + currents, they were filled with a solution of (in raM): KCI, 107; CaC12, 1; MgSO4, 2; NaCI, 10; K-EGTA, 11; HEPES, 11; ATP, 2 (pH 7.20). Although this 'intraceUular' solution overestimated the CI- concentration found in these cellsTM, it was kept high to allow direct comparisons of the effects of doxapram with previously published effects of physiological stimuli on currents in type I cells obtained from neonatal rats 19'2°'z3'~. The perfusate was of composition (in raM): NaC1, 135; KCI, 5; MgSO4, 1.2; CaCI2, 2.5; HEPES, 5 (21-24°C, pH 7.40). To record Ca 2+independent K + currents in isolation21, the extracellular [Ca2+] was reduced to 0.1 mM and the [Mg2+] was raised to 6 raM. Current flowing through Ca 2+ channels were recorded using a patch electrode filling solution of (in raM): CsCI, 130; NaC1, 10; EGTA, 1.1; MgC12, 2; CaCI2, 0.1; HEPES 10; ATP, 2 (pH 7.2); and the perfusate composition was (in raM); NaC1, i10; CsCI, 3, BaCI2, 10; MgCI2, 0.6; HEPES, 5; tetraethylammonium- (TEA-) CI, 20 (pH 7.4). Under these conditions, Ba 2+ acted as charge cartier through Ca 2+ channels and so the currents are referred to as 'Ca2+ channel currents'. It was necessary to use Ba 2+ as charge carrier, because currents recorded using physiological levels of Ca 2+ were too small to be accurately measured. Outward currents were minimized by replacement of K + with Cs + and addition of TEA [see ref. 15]. Type I cells were voltage-clamped at -70 mV holding potential, and whole-cell currents were recorded in response to 50 or 100 ms step cl~anges in the membrane potential to various test potentials as indicated in the results, applied at a frequency of 0.1 or 0.2 I-lz. Data were recorded on magnetic tape and current amplitudes measured off-line by computer (VCAN software; J. Dempster, Strathclyde University) by averaging the near-steady current level over the last 5-15 ms of the test pulse (under all conditions studied K + and Ca 2+ channel currents showed vtrtually no time-dependent decay). Current-voltage (l-V) relationships were obtained for each cell before, during and after bath application of doxapram (1-100 /tM; Dopram, A.H. Robins Co., Crawley, U.K.), and are plotted following leak subtraction, which was performed by the appropriate scaling and subtraction of the average current amplitude evoked by small depolarizing and hyperpolarizing steps applied at the beginning of each experiment in wMch a I - V relationship was obtained. In some experiments where K + currents were being studled, the membrane potential was repeatedly (0.1 or 0.2 Hz) stepped from -70 to +20 mV to examine the effects of various changes in the bathing medium with time. At the test potential of +20 mV
05
-;o -
[~..~A
0.4
~Oms
0.3
-;o
I(nA) C
I°
io
o
Fig. 1. Current-voltage (l-V) relationships obtained in the same type I cell under control conditions (O), during bath applications of 3/~M ([3) and then 30/~M (11) doxapram, and after returning to control solution (o). The inset shows example traces of the currents from this type I cell, all recorded at the same test potential of +20 mV, under the 4 conditions plotted in the I - V relationships.
the peak of a shoulder is often seen in the K + I - V relationship due to the activation of Ca2+-dependent K + channels19-21 (see Fig. 1). Statistical comparisons of current amplitudes recorded under control and test conditions at any given test potential were made using paired two-tailed Student's t-test, unless otherwise indicated. RESULTS
Effects o f doxapram on K + currents in type I cells U n d e r c o n t r o l c o n d i t i o n s , d e p o l a r i z i n g steps a p p l i e d to isolated type I cells, using a h o l d i n g p o t e n t i a l of - 7 0 mV, e v o k e d o u t w a r d c u r r e n t s positive to - 3 0 m V o r - 2 0 m V (e.g. Fig. 1, c o n t r o l trace). T h e s e c u r r e n t s arise d u e to the activation of two classes o f K + c h a n n e l ; a 4-amin o p y r i d i n e - s e n s i t i v e , d e l a y e d rectifier-type c h a n n e l (IKv) 21 a n d a c h a r y b d o t o x i n - s e n s i t i v e , h i g h - c o n d u c t a n c e Ca2+-activated K + c h a n n e l (/Kca) 19. T h e s h o u l d e r s e e n in the I - V r e l a t i o n s h i p at low, positive test p o t e n t i a l values (Fig. 1) arises d u e to t h e m a x i m a l activation of / K c a following v o l t a g e - d e p e n d e n t influx of C a 2÷ into the cell t h r o u g h its o w n L - t y p e C a 2+ c h a n n e l (/Ca) 19'23. B a t h applications of d o x a p r a m ( 1 - 1 0 0 /~M) caused d o s e - d e p e n d e n t , reversible r e d u c t i o n s in K ÷ c u r r e n t amplitudes (Figs. 1 a n d 4). T h e s e effects s h o w e d a voltage d e p e n d e n c e , in that they w e r e always m a x i m a l at the test p o t e n t i a l s w h e r e the s h o u l d e r in the I - V r e l a t i o n s h i p o c c u r r e d (e.g. + 3 0 m V in Fig. 1, always b e t w e e n + 2 0 a n d + 4 0 m V ) . F o r e x a m p l e , at a dose of 10/~M, doxap r a m significantly r e d u c e d K + c u r r e n t a m p l i t u d e s o v e r the test p o t e n t i a l r a n g e 0 to + 6 0 m V ( P < 0.02 to P < 0.001, n = 10 cells), b u t its effects were m a x i m a l at + 4 0 mV, w h e r e it r e d u c e d c u r r e n t a m p l i t u d e s b y 38.2 - 2.2% ( m e a n --- S . E . M . , n = 10). This v o l t a g e - d e p e n d e n t inh i b i t i o n of K + c u r r e n t s suggested that d o x a p r a m selectively i n h i b i t e d / K c , in type I cells. H o w e v e r , at h i g h e r
118 07 I(nA)
C FR
A
0.5 ,~o,~,
04 03 O2 ,
-60
-40
J
-20
0
p
i
i
20
40
60
Vto,~ (mV)
Fig. 2. Current-voltage (I-V) relationships obtained m the same type I cell using high [Mg2+] (6 raM), low [Ca2+] (0.1 mM) solutions before application of doxapram (O), then during exposure to 10/zM (11) and then 100 #M (D) doxapram, and after returning to doxapram-free solution (o). The inset shows example traces of the currents from this type I cell, all recorded at the same test potential of +20 mV, under the 4 conditions plotted in the I - V relatmnships.
doses, K + currents were further reduced (Figs. 1 and 4), and indeed were almost completely abolished with 100 /zM doxapram. A s / K c a only accounts for up to 50% of the total K + current at a given test potential 19, this indicated that the drug was also inhibiting/Kv in type I cells. To examine any effects of doxapram on the Ca2+-in sensitive/K~ without contamination f r o m / K c a (or/ca), recordings were made from cells bathed in high [Mg2+], low [Ca 2÷] solutions (see Materials and Methods for composition). Under these conditions currents increased with increasing test potential in a near-linear fashion (Fig. 2) and no shoulder was apparent at the lower positive test potentials. In the presence of doxapram (e.g. 10 or 100/~M, Fig. 2), K ÷ current amplitudes were reversibly reduced at all activating test potential values examined, so that I - V relationships obtained in the presence of the drug were also near-linear but with a reduced slope (i.e. doxapram apparently reduced the underlying channel conductance rather than its kinetics). Unlike under control conditions (see above), the effects of doxapram o n / K ~ were voltage independent: similar degrees of current reduction were seen at every test potential value studied, for any given dose of doxapram (e.g. Fig.
3). Fig. 4 shows dose-response curves for the effects of doxapram on K ÷ currents recorded in type I cells bathed in control solutions (filled circles) and high [Mg2+], low [Ca 2+] solutions (open circles). In control conditions, doxapram caused significant reductions at all concentrations tested (see legend to Fig. 4 for details), and the
ICso was approximately 13 /zM. Doxapram also produced a dose-dependent inhibition o f / K v recorded in isolation, (IC5o approx. 20/zM), but its effects were not significant at 1/zM. Indeed, at all but the highest dose (100/zM), the effects of doxapram were significantly less on /K v than on K + currents recorded in control solutions, when K + currents are composed of b o t h / K v and /Kca (P < 0.04 to P < 0.002, unpaired t-test's). This suggested that doxapram was a more potent inhibitor of /Kca than o f / K v. The dashed line in Fig. 4 is a plot of the mean effect of doxapram on /Kc~ alone, and was calculated using the known effects of doxapram o n / K ~ (open circles, Fig. 4) along with the mean proportion of total K + current that is attributable to / K v and /Kca, taken from previous studies 19'2° (see legend for further details). Typical examples of the time course of inhibition of type I cell K + currents by doxapram are shown in Fig. 5. Under control conditions (Fig. 5A), or using high [Mg2+], low [Ca 2+] solutions (Fig. 5B), bath application always caused a rapid inhibition of K + current amplitudes, which was also quickly reversed on washout of the drug (the bath exchange time was approximately 15 s). These findings indicate that inhibition o f / K c a a n d / K v by doxapram is similarly rapid, and suggest that the effects of doxapram on both channel types may arise due to a direct blocking effect. Findings presented in Figs. 1-5 indicate that doxapram reversibly inhibits both /Kca and /Kv in type I cells, and that the drug is a more potent blocker of/Kca.
1 O0
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
80
6O
%
40
J IO0~M _
20
i
i
i
i
-20
0
20
40
60
V~.~t(mV)
Fig. 3. Plot of the percentage reduction m /K v (recorded in high [Mg2+], low [Ca2+] solutions) caused by 10 #M (o) and 100/zM (0) doxapram, over the range of activating test potentaals studied. Each point plotted is the mean (with vertical S.E.M. bars) taken from 5 cells.
119
Effects o f doxapram on Ca2+ channel currents in type I cells O n e possible explanation for the effects of d o x a p r a m o n / K ~ (see above) is that the drug was inhibiting the Ca 2+ entry into t y p e I cells which is r e q u i r e d for activation o f this c o m p o n e n t of the K + current u n d e r the recording conditions described above. To investigate this possibility, the effects of d o x a p r a m on current flowing through type I cell Ca 2+ channels were r e c o r d e d directly, using 10 m M Ba 2+ as charge c a r d e r , and solutions designed to minimize contaminating currents flowing through o t h e r channel types (see Materials and Methods). U n d e r these conditions, inward Ca 2+ channel currents are sustained throughout 50 ms depolarizing step potentials, with no a p p a r e n t transient c o m p o n e n t (Fig. 6). Previous studies have shown t h e m to be enhanced by the d i h y d r o p y r i d i n e agonist B A Y K 864423, indicating that the channels underlying these currents are largely, if not exclusively, of the L-type 29. Bath application of 10/~M d o x a p r a m (a dose which causes significant reductions in ~ a n d / K v, see Fig. 4) was without effect on these L-type currents (Fig. 6). A lack of effect was consistently seen in 14 cells tested, indicating that the effects of d o x a p r a m o n / K c a (see above) did not result as a consequence of the drug inhibiting the Ca 2+ influx required for activation of this c o m p o n e n t of the K + current.
Effects o f doxapram on ionic currents in NG-108 15 cells Type I carotid b o d y cells are of neural crest origin it. To d e t e r m i n e w h e t h e r the inhibition by d o x a p r a m of K +
A
doxapram
100
11n.s.
80 D, . ~ so _."g 40
" ' " " "o.,
......... :
0 i
,
1
, ,
, i
,
10 [cloxapram](~M)
,
,
,
,
,
,
,
100
Fig. 4. Log dose-response curves for the effects of doxapram on K + currents obtained under control conditions (0; when both/Kc, and/K v are activated) and using high [Mg2+], low [Ca2+] solutions, (o; when/K v alone is activated). Each plotted point is the mean with vertical S.E.M. bars, taken from the number of cells tested as indicated. The test potential values used for calculating the % reduction caused by doxapram were always between +20 and +40 mV in control solutions (depending on where the shoulder in the I-V curve was maximal), and was fixed at +20 mV for high [Mg2+], low [Ca2+] solutions, as no shoulder was observed and the effects of doxapram on/Kv were voltage independent (Fig. 3). Significant reductions in K + current amplitudes caused by doxapram under control conditions are indicated by the filled stars: 3 stars, P < 0.02; two stars, P < 0.002; one star; P < 0.0001. Significant reductions found under high [Mg2+], low [Ca2+] conditions indicated by open stars: 2 stars, P < 0.0005; one star P < 0.0001 (n.s., not significant). Dashed hne connecting open square symbols represents the estimated dose-response curve for/Kc= alone. This curve was calculated from the two experimentally obtained dose-response curves, with the knowledge that ~ accounts for approx. 30% of total K + current between test potential values of +20 and +40 mVZ9.2o. B
07 ~ i ~
I~
03 I
06 I (nA)
doxapram
I (nA) "
05. 0204 O3
L
0.2
I
~_
/
01
01 0 0
' 50
, I00 time (s)
, ISO
0
, I O0
50 tlme
, 150
(s)
Fig. 5. Tune series plots of type I cell K + current amplitudes obtained by repeated step depolarizations to +20 mV (50 ms duration, 0.2 Hz), under control conditions (A) and in high [Mg2+], low [Ca2+] solutions (B), illustrating the time course of application and wash-out of 100 /~M doxapram under these two conditions. Period of doxapram appfication is indicated in each case by the horizontal bar.
120 reversible inhibitions of K ÷ current amplitudes of between 3 and 11%. These findings suggest that doxapram shows at least some selectivity for K + channels in type I carotid cells above those found in other neuronal tissue, since the dose tested on NG-108 15 cells (10 gM), caused significant reductions in both/Kc~ and/K~ in type I cells (Fig. 4). 20pA
DISCUSSION lOms
Fig. 6. Example inward Ca 2+ channel currents obtained from a type I cell, using 10 m M B a 2+ as charge carrier, under control conditions (O) and in the presence of 10/~M doxapram (0). Both currents were obtained by a step depolarization to 0 mV.
currents in type I cells (Figs. 1-5) were specific to this tissue, any effects of the drug on ionic currents recorded in other neuronal tissue were investigated, using the mouse neuroblastoma x rat glioma hybrid cell line NG108 15. These cells are known to possess an array of ion channel types, including Ca2+-activated and Ca2+-inde pendent K + currents 2. Using control solutions, 100 ms depolarizing step changes in membrane potential (0.2 Hz, holding potential -80 mV) evoked large, outward K ÷ currents (Fig. 7), which were preceded by transient inward (Na ÷ and Ca 2+) currents. In 4 out of 8 cells tested, 10/xM doxapram was without discernible effect on these K ÷ currents (e.g. Fig. 7). However, in the remaining 4 cells the same dose of doxapram caused small,
12
I(nA)
10
06°'8 J -6O
-40
-2O -02
I
20
i
I
40 60 V tesl (mY)
i
80
Fig. 7. I - V curve obtained from a cultured NG-108 15 cell under control conditions (O) and in the presence of 10/xM doxapram (o). Inset shows example traces from this same cell, under the two conditions indicated (test potential +20 mV).
Results presented here show that doxapram is a rapid and reversible inhibitor of both Ca2+-activated and Ca 2÷independent K + channels (/Kca and/Kv) found in isolated type I cells of the neonatal rat (Figs. 1-4). The drug appears more potent in inhibiting/Kca (IC5o approximately 5 gM) t h a n / K v (IC50 approximately 20 #M; Fig. 4). Its effects appear to be selective for K ÷ channels in the type I cell, as it was without significant effect on L-type Ca z+ channel currents at a dose (10 #M) which significantly reduced K + current amplitudes. Furthermore, the actions of doxapram on K + currents appears to show at least a degree of tissue selectivity: 10 #M doxapram did not significantly affect K + currents recorded in the neuronal cell line NG-108 15 (Fig. 7). This observation is in agreement with the fact that there are few observable side-effects of the drug in patients who required elevated ventilation. Although the mechanisms of the effects of doxapram reported here are at present unknown, time course studies of its blockade of b o t h / K c a a n d / K v (Fig. 5) suggest that its effects may arise either due to a direct action on these two channel types, or by interaction with a closely coupled receptor: a receptor-mediated action involving generation of second messengers would be expected to give rise to a slower inhibition of these channels. In order to explain how the effects of doxapram reported here could lead to increased CSN discharge, and hence elevated ventilation 16, it is important to compare the drug's action on type I cell ion channels with those of the physiological stimuli of the carotid body. Both hypoxia and acidity suppress K ÷ channels in type I cells3' 9,12,13,19,20,23.28. In the neonatal rat these effects have been shown by this laboratory to be selective for/Kc~ t9, 20,23, whereas in a similar preparation used by other workers 28 and also in long-term cultured type I cells from fetal rabbits 9 the K ÷ current subtype suppressed by hypoxia has not yet been fully characterized. In adult rabbit tissue the effects of these stimuli have been demonstrated on a K + current remaining after run-down of the Ca 2÷ current, suggesting that the underlying K ÷ channel is not Ca2+-activated ~3. Single channel recordings have shown, in long-term cultured type I cells obtained from fetal rabbit tissue, that hypoxia inhibits a
121 large-conductance K + channel 3, and it should be noted that the hypoxia- and a c i d - s e n s i t i v e / K ~ in neonatal rat cells is also of high conductance, as it is inhibited by charybdotoxin 19, a peptide toxin which selectively inhibits high conductance Ca2+-activated K + channels. Apamin (another peptide toxin which only inhibits low conductance Ca2+-activated K ÷ channels) has no effect in these cells 19. In every preparation thus far studied, hypoxia or acidity have been shown to be without observable effects on the inward L-type Ca 2+ current in type I cells 9'12'23'28. Thus doxapram, by inhibiting K ÷ and not Ca 2÷ channels in type I cells, shows much similarity to the effects of physiological stimuli of the carotid body. Suppression of high-conductance K + channels has been proposed as the cause of type I cell depolarization induced by hypoxia 3, and acidity, which also blocks such channels (see above), can also depolarize these cells 5. Such an effect would presumably lead to the opening of the voltage-dependent L-type Ca 2÷ channels, and hence Ca2+-dependent transmitter release. Such a model of chemotransduction in type I cells has further support, in the case of hypoxia at least, in that release of radiolabelled catecholamines in response to hypoxia can be inhibited by the L-type Ca 2÷ channel antagonist nitrendipine ~7. The observations reported in this paper would suggest that doxapram, by inhibiting b o t h / K c a a n d / K v in type I cells, may act in a similar fashion to the phys-
iological stimuli of the carotid body, but further experimentation (such as a study of the effects of doxapram on transmitter release from type I cells and the Ca 2+dependence of any such effects) is required before such a hypothesis for the actions of doxapram can be validated. Furthermore, although the above-mentioned actions of chemostimuli on ion channels in type I cells suggest a stimulus-depolarization-secretion model for chemotransduction, there are alternative possible mechanisms for such a process occurring in response to both acidic and hypoxic stimuli. Acidity has recently been demonstrated to lead to elevated intracellular [Ca 2÷] (and hence transmitter release) by reversing the operation of the Na t Ca 2÷ exchanger in the type I cell plasma membrane 27. Other workers have provided evidence that hypoxia raises cytosolic [Ca 2÷] by causing its release from intracellular stores, rather than stimulating its influx across the plasma membrane 1. The effects of doxapram on membrane transporters and intracellular stores require study before the relative importance of its actions reported here (or indeed those of hypoxia and acidity) on ion channels in type I cells can be fully determined.
Acknowledgements. This work was supported by the Wellcome Trust. I would like to thank J. Dempster (Strathclyde University) for his analysis software.
REFERENCES 1 Biscoe, T.J. and Duchen, M.R., Responses of type I cells dissociated from rabbit carotid body to hypoxia, J. Physiol., 428 (1990) 39-59. 2 Brown, D.A. and Higashida, H., Voltage- and calcium-activated potassium currents in mouse neuroblastoma × rat glioma hybrid cells, J Physml., 397 (1988) 149-166. 3 Delpiano, M.A. and Hescheler, J., Evidence for a PO2-senslrive K + channel in the type-I cell of the rabbit carotid body, FEBS Lett., 249 (1989) 195-198. 4 Duchen, M R., Caddy, K.W.T., Kirby, G.C., Patterson, D.L., Ponte, J. and Biscoe, T.J., Biophysical studies of the cellular elements of the rabbit carotid body, Neuroscience, 26 (1989) 291-311. 5 Eyzagmrre, C., An overview of mechamsms associated with the onset of sensory &scharges in the carotid nerve. In C. Beimonte, D.J. Pallot, H. Acker and S. Fidone (Eds.), Arterial Chemoreceptors, Leicester University Press, U.K., 1981, pp. 20-44 6 Fldone, S.J. and Gonzalez, C., Initiation and control of chemoreceptor activity m the carotid body. In A.P. Fishman, N.S. Cherniack, J.G. Widdicome and S.R. Geiger (Eds.), Handbook of Physiology. III: The Respiratory System, Vol 2, Part 1, American Physiological Society, Bethesda, MD, 1986, pp. 247-312. 7 Fidone, S., Gonzalez, C. and Yoshizaki, K., Effects of low oxygen on the release of dopamine from the rabbit carotid body m wvo, J PhysioL, 333 (1982) 93-110. 8 Hamill, O.P., Marry, A., Neher, E., Sakmann, B. and Slgworth, EJ., Improved patch clamp techmques for hlgh-resolu-
9
10 11 12
13
14
15 16 17
tion current recording from cells and cell-free membrane patches, Pfl~gers Arch., 391 (1981) 85-100. Hescheler, J., Delpiano, M.A., Acker, H. and Pietruschka, F., Iomc currents on type-I cells of the rabbit carotid body measured by voltage-clamp experiments and the effect of hypoxia, Brain Research, 486 (1989) 79-88. Hirsch, K. and Wang, S.C., Selective respiratory stimulating action of doxapram compared to pentylenetetrazol, J. Pharmacol. Exp. Ther., 189 (1974) 1-11. Le Douarin, N., Le L~evre, C. and Fontmne, J., Recherches expenmentales sur l'origme embryologique du corps carotidien chez les oiseaux, C.R. Acad. Sci. Ser. D., 275 (1972) 583-586. Lopez-Barneo, J., Lopez-Lopez, J.R., Urena, J. and Gonzalez, C., Chemotransductlon m the carotid body: K + current modulated by PO2 in type I chemoreceptor cells, Science, 241 (1988) 580-582. Lopez-Lopez, J., Gonzalez, C., Urena, J. and Lopez-Barneo, J., Low PO2 selectively inhibits K channel activity in chemoreceptor cells of the mammalian carotid body, J. Gen. Physiol., 93 (1989) 1001-1015. McDonald, D.M. and Mitchell, R.A., A quantitative analysis of synaptic connections in the rat carotid body. In M.J. Purves (Ed.), The Pertpheral Arterial Chemoreceptors, Cambridge University Press, London, 1975, pp. 101-131. Narahashi, T., Tsunoo, A. and Yoshli, M., Characterization of two types of calcium channels in mouse neuroblastoma cells, J. Physiol., 383 (1987) 231-249. Nishml, T., Mokashi, A. and Lahiri, S., Stimulation of carotid chemoreceptors and ventilation by doxapram in the cat, J. Appl. Physiol., 52 (1982) 1261-1265. Obeso, A., Fldone, S. and Gonzalez, C,, Pathways for calcium
122
18
19
20
21
22
23
entry into type I cells: slgniticance for the secretory response. In J.A. Ribeiro and D.J. Pallot (Eds.), Chemoreceptors m Respiratory Control, Croom Helm, London, 1987, pp. 91-97. Oyama, Y., Walker, J.L. and Eyzaguirre, C., The intracellular chloride activity of giomus cells in the isolated rabbit carotad body, Brain Research, 368 (1986) 167-169 Peers, C., Selective effect of lowered extracellular pH on Ca 2+dependent K ÷ currents in type I cells isolated from the neonatal rat carotid body, J. Physiol., 422 (1990) 381-395. Peers, C., Hypoxic suppression of K + currents in type I carotid body cells: selective effect on the Ca2+-activated K + current, Neurosct. Lett, 119 (1990) 253-256. Peers, C., Effects of D600 on hyporac suppression of K + currents in isolated type I carotid body cells of the neonatal rat, FEBS Lett., 271 (1990) 37-40. Peers, C., Effects of doxapram on ionic currents in isolated type I cells of the neonatal rat carotad body, J. Physiol., 438 (1991) 230P. Peers, C. and Green, EK., Intracellular acidosis inhlbtts Ca 2+dependent K + currents in isolated type I cells of the neonatal rat caroUd body, J. Phystol., 437 (1991) 589-602.
24 Peers, C., Lang, B., Newsom-Davls, J. and W.-Wray, D., Selectwe action of myasthenic syndrome ant~bodles on calcium channels in a rodent neuroblastoma x ghoma cell line, J. Phystol, 421 (1990) 293-308. 25 Peers, C. and O'Donnell, J., Potassmm currents recorded m type I carotid body cells from the neonatal rat and their modulation by chemoexcitatory agents, Bram Research, 522 (1990) 259-266. 26 Rigual, R., Gonzalez, E., Fldone, S. and Gonzalez, C., Effects of low pH on synthesis and release of catecholammes in the cat carotid body m vitro, Brain Research, 309 (1984) 178-181. 27 Rocher, A , Obeso, A., Gonzalez, C and Herreros, B., Ionic mechamsms for the transductlon of acl&c stimuli in rabbit carotid body glomus cells, J. Physiol., 433 (1991) 533-548. 28 Stea, A. and Nurse, C.A., Whole-cell and perforated-patch recordings from O2-sensitive rat carotid body cells grown in short-, and long-term culture, Pfl~igers Arch., 418 (1991) 93-101. 29 Ts~en, R.W., Llpscombe, D., Ma&son, D.V., Bley, K.R. and Fox, A.P., Multiple types of neuronal calcium channels and their selective modulation, Trends Neuroscz., I1 (1988) 431-438.
116 BRES 17283
Effects of doxapram on ionic currents recorded in isolated type I cells of the neonatal rat carotid body C. Peers Department of Pharmacology, Leeds University, Leeds (U.K.) (Accepted 6 August 1991)
Key words: Doxapram; Carotid body; Chemoreceptor; Type I cell; Potassium current; Calcium current
Whole-cell patch-clamp recordings were used to investigate the effects of the respiratory stimulant doxapram on K ÷ and Ca2+ currents m isolated type I ceils of the neonatal rat carotid body. Doxapram (1-100/zM) caused rapid, reversible and dose-dependent inhibitions of K + currents recorded in type I cells (ICso approximately 13/zM). Inhibition was voltage-dependent, m that the effects of doxapram were maximal at test potentaals where a shoulder in the current-voltage relationship was maxamal. These K + currents were composed of both Ca2+activated and Ca2 + -mdependent components. Using high [Mg2 + ], low [Ca2 + ] solutions to inhibit Ca2+-activated K + currents, doxapram was also seen to directly inhibit Ca2+-independent K + currents. This effect was voltage-independent and was less potent (ICs0 approximately 20 /tM) than under control conditions, suggesting that doxapram was a more potent inhibitor of the Ca2÷-activated K + currents recorded under control conditions. Doxapram (10/zM) was without effect on L-type Ca2+ channel currents recorded under conditions where K + channel activity was minimized and was also without significant effect on K + currents recorded in the neuronal cell hne NG-108 15, suggesting a selective effect on carotid body type I cells. The effects of doxapram on type I cells show similarities to those of the physiological stimuh of the carotid body, suggesting that doxapram may share a similar mechanism of action in stimulating the intact organ. INTRODUCTION The carotid b o d y is the m a j o r peripheral arterial chemoreceptor. It responds to changes in levels of arterial 0 2, C O 2 and p H by modulating the frequency of discharge of afferent chemosensory fibres in the carotid sinus nerve (CSN), thereby exerting influences on the pattern of breathing 6. The type I (or giomus) cells are generally believed to play an i m p o r t a n t role in this organ's functioning; they lie in synaptic contact with afferent chemosensory fibres 14 and contain catecholaminefilled vesicles. The close correlation in b o t h time course and magnitude between catecholamine release from type I cells and elevation of CSN discharge 7'26 has contributed to the idea that type I cells are responsible for detecting and transducing chemostimuli. The patch-clamp technique has recently been applied to the study of isolated type I cells, in o r d e r to determine the possible roles ion channels may play in chemotransduction 4'9'12'25'2s. In most cases, physiological stimuli (e.g. hypoxia, acidity) have b e e n demonstrated to selectively inhibit K + channels, whilst other channel types (Ca 2÷ channels, and also N a ÷ channels in
adult type I cells) remain unaffected 3'9'12'13'19'20'23'28. In type I cells isolated from the neonatal rat at least, the effects of hypoxia and acidity are selective for charybdotoxin-sensitive (high conductance) Ca2+-activated K ÷ channels 19,20,23. D o x a p r a m hydrochloride (1-ethyl-4-(2-morpholinoethyl)-3,3-diphenyl-2-pyrrolidinone hydrochloride hydrate, hereafter referred to as d o x a p r a m ) is a drug used clinically to stimulate breathing in a variety of conditions. Its stimulatory action has b e e n shown in anaesthetized animals to be due largely to a direct action on the carotid b o d y 1°'16, where it causes an increase in the discharge of afferent chemosensory fibres in the CSN, although at higher doses it can also stimulate central respiratory neurones. The mechanism of d o x a p r a m ' s action on the carotid b o d y is at present unknown, but appears to be i n d e p e n d e n t of the b a c k g r o u n d levels of O2 x6. The present study was carried out to examine whether doxa p r a m alters the activity of ion channels in isolated type I cells of the neonatal rat, in o r d e r to c o m p a r e its effects with those of physiological stimuli (see above). A preliminary account of some of these findings has been presented in abstract form 22.
Correspondence. Dr Chns Peers, Department of Pharmacology, Worsley Mechcal and Dental Bmldmg, Leeds University, Leeds, LS2 9JT, U.K.
117 MATERIALS AND METHODS The procedures for isolation and maintenance of type I cells in short-term culture were as previously described 19"25. In brief, carotid bodies were removed from 8- to ll-day-oid rats (which were anaesthetized by breathing 4% enflurane, 96% oxygen through a face mask) and placed in ice-cold phosphate-buffered saline (PBS). They were then cut into pieces and placed in Ca 2+ and Mg3+-free PBS containing collagenase (0.03-0.05%) and trypsin (0.020.025%) at 37"C for 20-30 rain. After gentle trituration with a Pasteur pipette, the dispersed cells were resuspended in Ham's F-12 culture medium containing 10% fetal calf serum, penicillin (100 IU/rni), streptomycin (100/~g/ml) and insulin (84 U/l). They were plated onto polylysine-eoated coverslips and kept at 37"C in a humidified incubator until required for study. Electrophysiological recordings were made from cells kept in short-term culture for between 4 and 48 h. In some experiments, the effects of doxapram were investigated on ionic currents in the undifferentiated mouse neuroblastoma x rat giioma hybrid cell line NG-108 15. These cells were maintained in culture as previously described24, and recordings were made as for the type I carotid body cells. Fragments of coverslips to which cells had adhered were transferred to a recording chamber (volume approximately 80/zl) which was continually perfused at 0.5 ml/min. Whole-cell patch-clamp recordingss were made from the most numerous cell type, the phasebright, spherical type I cells, which were of approximately 10/~M diameter~. Patch electrodes had resistances of between 2 and 7 MQ. To record K + currents, they were filled with a solution of (in raM): KCI, 107; CaC12, 1; MgSO4, 2; NaCI, 10; K-EGTA, 11; HEPES, 11; ATP, 2 (pH 7.20). Although this 'intraceUular' solution overestimated the CI- concentration found in these cellsTM, it was kept high to allow direct comparisons of the effects of doxapram with previously published effects of physiological stimuli on currents in type I cells obtained from neonatal rats 19'2°'z3'~. The perfusate was of composition (in raM): NaC1, 135; KCI, 5; MgSO4, 1.2; CaCI2, 2.5; HEPES, 5 (21-24°C, pH 7.40). To record Ca 2+independent K + currents in isolation21, the extracellular [Ca2+] was reduced to 0.1 mM and the [Mg2+] was raised to 6 raM. Current flowing through Ca 2+ channels were recorded using a patch electrode filling solution of (in raM): CsCI, 130; NaC1, 10; EGTA, 1.1; MgC12, 2; CaCI2, 0.1; HEPES 10; ATP, 2 (pH 7.2); and the perfusate composition was (in raM); NaC1, i10; CsCI, 3, BaCI2, 10; MgCI2, 0.6; HEPES, 5; tetraethylammonium- (TEA-) CI, 20 (pH 7.4). Under these conditions, Ba 2+ acted as charge cartier through Ca 2+ channels and so the currents are referred to as 'Ca2+ channel currents'. It was necessary to use Ba 2+ as charge carrier, because currents recorded using physiological levels of Ca 2+ were too small to be accurately measured. Outward currents were minimized by replacement of K + with Cs + and addition of TEA [see ref. 15]. Type I cells were voltage-clamped at -70 mV holding potential, and whole-cell currents were recorded in response to 50 or 100 ms step cl~anges in the membrane potential to various test potentials as indicated in the results, applied at a frequency of 0.1 or 0.2 I-lz. Data were recorded on magnetic tape and current amplitudes measured off-line by computer (VCAN software; J. Dempster, Strathclyde University) by averaging the near-steady current level over the last 5-15 ms of the test pulse (under all conditions studied K + and Ca 2+ channel currents showed vtrtually no time-dependent decay). Current-voltage (l-V) relationships were obtained for each cell before, during and after bath application of doxapram (1-100 /tM; Dopram, A.H. Robins Co., Crawley, U.K.), and are plotted following leak subtraction, which was performed by the appropriate scaling and subtraction of the average current amplitude evoked by small depolarizing and hyperpolarizing steps applied at the beginning of each experiment in wMch a I - V relationship was obtained. In some experiments where K + currents were being studled, the membrane potential was repeatedly (0.1 or 0.2 Hz) stepped from -70 to +20 mV to examine the effects of various changes in the bathing medium with time. At the test potential of +20 mV
05
-;o -
[~..~A
0.4
~Oms
0.3
-;o
I(nA) C
I°
io
o
Fig. 1. Current-voltage (l-V) relationships obtained in the same type I cell under control conditions (O), during bath applications of 3/~M ([3) and then 30/~M (11) doxapram, and after returning to control solution (o). The inset shows example traces of the currents from this type I cell, all recorded at the same test potential of +20 mV, under the 4 conditions plotted in the I - V relationships.
the peak of a shoulder is often seen in the K + I - V relationship due to the activation of Ca2+-dependent K + channels19-21 (see Fig. 1). Statistical comparisons of current amplitudes recorded under control and test conditions at any given test potential were made using paired two-tailed Student's t-test, unless otherwise indicated. RESULTS
Effects o f doxapram on K + currents in type I cells U n d e r c o n t r o l c o n d i t i o n s , d e p o l a r i z i n g steps a p p l i e d to isolated type I cells, using a h o l d i n g p o t e n t i a l of - 7 0 mV, e v o k e d o u t w a r d c u r r e n t s positive to - 3 0 m V o r - 2 0 m V (e.g. Fig. 1, c o n t r o l trace). T h e s e c u r r e n t s arise d u e to the activation of two classes o f K + c h a n n e l ; a 4-amin o p y r i d i n e - s e n s i t i v e , d e l a y e d rectifier-type c h a n n e l (IKv) 21 a n d a c h a r y b d o t o x i n - s e n s i t i v e , h i g h - c o n d u c t a n c e Ca2+-activated K + c h a n n e l (/Kca) 19. T h e s h o u l d e r s e e n in the I - V r e l a t i o n s h i p at low, positive test p o t e n t i a l values (Fig. 1) arises d u e to t h e m a x i m a l activation of / K c a following v o l t a g e - d e p e n d e n t influx of C a 2÷ into the cell t h r o u g h its o w n L - t y p e C a 2+ c h a n n e l (/Ca) 19'23. B a t h applications of d o x a p r a m ( 1 - 1 0 0 /~M) caused d o s e - d e p e n d e n t , reversible r e d u c t i o n s in K ÷ c u r r e n t amplitudes (Figs. 1 a n d 4). T h e s e effects s h o w e d a voltage d e p e n d e n c e , in that they w e r e always m a x i m a l at the test p o t e n t i a l s w h e r e the s h o u l d e r in the I - V r e l a t i o n s h i p o c c u r r e d (e.g. + 3 0 m V in Fig. 1, always b e t w e e n + 2 0 a n d + 4 0 m V ) . F o r e x a m p l e , at a dose of 10/~M, doxap r a m significantly r e d u c e d K + c u r r e n t a m p l i t u d e s o v e r the test p o t e n t i a l r a n g e 0 to + 6 0 m V ( P < 0.02 to P < 0.001, n = 10 cells), b u t its effects were m a x i m a l at + 4 0 mV, w h e r e it r e d u c e d c u r r e n t a m p l i t u d e s b y 38.2 - 2.2% ( m e a n --- S . E . M . , n = 10). This v o l t a g e - d e p e n d e n t inh i b i t i o n of K + c u r r e n t s suggested that d o x a p r a m selectively i n h i b i t e d / K c , in type I cells. H o w e v e r , at h i g h e r
118 07 I(nA)
C FR
A
0.5 ,~o,~,
04 03 O2 ,
-60
-40
J
-20
0
p
i
i
20
40
60
Vto,~ (mV)
Fig. 2. Current-voltage (I-V) relationships obtained m the same type I cell using high [Mg2+] (6 raM), low [Ca2+] (0.1 mM) solutions before application of doxapram (O), then during exposure to 10/zM (11) and then 100 #M (D) doxapram, and after returning to doxapram-free solution (o). The inset shows example traces of the currents from this type I cell, all recorded at the same test potential of +20 mV, under the 4 conditions plotted in the I - V relatmnships.
doses, K + currents were further reduced (Figs. 1 and 4), and indeed were almost completely abolished with 100 /zM doxapram. A s / K c a only accounts for up to 50% of the total K + current at a given test potential 19, this indicated that the drug was also inhibiting/Kv in type I cells. To examine any effects of doxapram on the Ca2+-in sensitive/K~ without contamination f r o m / K c a (or/ca), recordings were made from cells bathed in high [Mg2+], low [Ca 2÷] solutions (see Materials and Methods for composition). Under these conditions currents increased with increasing test potential in a near-linear fashion (Fig. 2) and no shoulder was apparent at the lower positive test potentials. In the presence of doxapram (e.g. 10 or 100/~M, Fig. 2), K ÷ current amplitudes were reversibly reduced at all activating test potential values examined, so that I - V relationships obtained in the presence of the drug were also near-linear but with a reduced slope (i.e. doxapram apparently reduced the underlying channel conductance rather than its kinetics). Unlike under control conditions (see above), the effects of doxapram o n / K ~ were voltage independent: similar degrees of current reduction were seen at every test potential value studied, for any given dose of doxapram (e.g. Fig.
3). Fig. 4 shows dose-response curves for the effects of doxapram on K ÷ currents recorded in type I cells bathed in control solutions (filled circles) and high [Mg2+], low [Ca 2+] solutions (open circles). In control conditions, doxapram caused significant reductions at all concentrations tested (see legend to Fig. 4 for details), and the
ICso was approximately 13 /zM. Doxapram also produced a dose-dependent inhibition o f / K v recorded in isolation, (IC5o approx. 20/zM), but its effects were not significant at 1/zM. Indeed, at all but the highest dose (100/zM), the effects of doxapram were significantly less on /K v than on K + currents recorded in control solutions, when K + currents are composed of b o t h / K v and /Kca (P < 0.04 to P < 0.002, unpaired t-test's). This suggested that doxapram was a more potent inhibitor of /Kca than o f / K v. The dashed line in Fig. 4 is a plot of the mean effect of doxapram on /Kc~ alone, and was calculated using the known effects of doxapram o n / K ~ (open circles, Fig. 4) along with the mean proportion of total K + current that is attributable to / K v and /Kca, taken from previous studies 19'2° (see legend for further details). Typical examples of the time course of inhibition of type I cell K + currents by doxapram are shown in Fig. 5. Under control conditions (Fig. 5A), or using high [Mg2+], low [Ca 2+] solutions (Fig. 5B), bath application always caused a rapid inhibition of K + current amplitudes, which was also quickly reversed on washout of the drug (the bath exchange time was approximately 15 s). These findings indicate that inhibition o f / K c a a n d / K v by doxapram is similarly rapid, and suggest that the effects of doxapram on both channel types may arise due to a direct blocking effect. Findings presented in Figs. 1-5 indicate that doxapram reversibly inhibits both /Kca and /Kv in type I cells, and that the drug is a more potent blocker of/Kca.
1 O0
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
80
6O
%
40
J IO0~M _
20
i
i
i
i
-20
0
20
40
60
V~.~t(mV)
Fig. 3. Plot of the percentage reduction m /K v (recorded in high [Mg2+], low [Ca2+] solutions) caused by 10 #M (o) and 100/zM (0) doxapram, over the range of activating test potentaals studied. Each point plotted is the mean (with vertical S.E.M. bars) taken from 5 cells.
119
Effects o f doxapram on Ca2+ channel currents in type I cells O n e possible explanation for the effects of d o x a p r a m o n / K ~ (see above) is that the drug was inhibiting the Ca 2+ entry into t y p e I cells which is r e q u i r e d for activation o f this c o m p o n e n t of the K + current u n d e r the recording conditions described above. To investigate this possibility, the effects of d o x a p r a m on current flowing through type I cell Ca 2+ channels were r e c o r d e d directly, using 10 m M Ba 2+ as charge c a r d e r , and solutions designed to minimize contaminating currents flowing through o t h e r channel types (see Materials and Methods). U n d e r these conditions, inward Ca 2+ channel currents are sustained throughout 50 ms depolarizing step potentials, with no a p p a r e n t transient c o m p o n e n t (Fig. 6). Previous studies have shown t h e m to be enhanced by the d i h y d r o p y r i d i n e agonist B A Y K 864423, indicating that the channels underlying these currents are largely, if not exclusively, of the L-type 29. Bath application of 10/~M d o x a p r a m (a dose which causes significant reductions in ~ a n d / K v, see Fig. 4) was without effect on these L-type currents (Fig. 6). A lack of effect was consistently seen in 14 cells tested, indicating that the effects of d o x a p r a m o n / K c a (see above) did not result as a consequence of the drug inhibiting the Ca 2+ influx required for activation of this c o m p o n e n t of the K + current.
Effects o f doxapram on ionic currents in NG-108 15 cells Type I carotid b o d y cells are of neural crest origin it. To d e t e r m i n e w h e t h e r the inhibition by d o x a p r a m of K +
A
doxapram
100
11n.s.
80 D, . ~ so _."g 40
" ' " " "o.,
......... :
0 i
,
1
, ,
, i
,
10 [cloxapram](~M)
,
,
,
,
,
,
,
100
Fig. 4. Log dose-response curves for the effects of doxapram on K + currents obtained under control conditions (0; when both/Kc, and/K v are activated) and using high [Mg2+], low [Ca2+] solutions, (o; when/K v alone is activated). Each plotted point is the mean with vertical S.E.M. bars, taken from the number of cells tested as indicated. The test potential values used for calculating the % reduction caused by doxapram were always between +20 and +40 mV in control solutions (depending on where the shoulder in the I-V curve was maximal), and was fixed at +20 mV for high [Mg2+], low [Ca2+] solutions, as no shoulder was observed and the effects of doxapram on/Kv were voltage independent (Fig. 3). Significant reductions in K + current amplitudes caused by doxapram under control conditions are indicated by the filled stars: 3 stars, P < 0.02; two stars, P < 0.002; one star; P < 0.0001. Significant reductions found under high [Mg2+], low [Ca2+] conditions indicated by open stars: 2 stars, P < 0.0005; one star P < 0.0001 (n.s., not significant). Dashed hne connecting open square symbols represents the estimated dose-response curve for/Kc= alone. This curve was calculated from the two experimentally obtained dose-response curves, with the knowledge that ~ accounts for approx. 30% of total K + current between test potential values of +20 and +40 mVZ9.2o. B
07 ~ i ~
I~
03 I
06 I (nA)
doxapram
I (nA) "
05. 0204 O3
L
0.2
I
~_
/
01
01 0 0
' 50
, I00 time (s)
, ISO
0
, I O0
50 tlme
, 150
(s)
Fig. 5. Tune series plots of type I cell K + current amplitudes obtained by repeated step depolarizations to +20 mV (50 ms duration, 0.2 Hz), under control conditions (A) and in high [Mg2+], low [Ca2+] solutions (B), illustrating the time course of application and wash-out of 100 /~M doxapram under these two conditions. Period of doxapram appfication is indicated in each case by the horizontal bar.
120 reversible inhibitions of K ÷ current amplitudes of between 3 and 11%. These findings suggest that doxapram shows at least some selectivity for K + channels in type I carotid cells above those found in other neuronal tissue, since the dose tested on NG-108 15 cells (10 gM), caused significant reductions in both/Kc~ and/K~ in type I cells (Fig. 4). 20pA
DISCUSSION lOms
Fig. 6. Example inward Ca 2+ channel currents obtained from a type I cell, using 10 m M B a 2+ as charge carrier, under control conditions (O) and in the presence of 10/~M doxapram (0). Both currents were obtained by a step depolarization to 0 mV.
currents in type I cells (Figs. 1-5) were specific to this tissue, any effects of the drug on ionic currents recorded in other neuronal tissue were investigated, using the mouse neuroblastoma x rat glioma hybrid cell line NG108 15. These cells are known to possess an array of ion channel types, including Ca2+-activated and Ca2+-inde pendent K + currents 2. Using control solutions, 100 ms depolarizing step changes in membrane potential (0.2 Hz, holding potential -80 mV) evoked large, outward K ÷ currents (Fig. 7), which were preceded by transient inward (Na ÷ and Ca 2+) currents. In 4 out of 8 cells tested, 10/xM doxapram was without discernible effect on these K ÷ currents (e.g. Fig. 7). However, in the remaining 4 cells the same dose of doxapram caused small,
12
I(nA)
10
06°'8 J -6O
-40
-2O -02
I
20
i
I
40 60 V tesl (mY)
i
80
Fig. 7. I - V curve obtained from a cultured NG-108 15 cell under control conditions (O) and in the presence of 10/xM doxapram (o). Inset shows example traces from this same cell, under the two conditions indicated (test potential +20 mV).
Results presented here show that doxapram is a rapid and reversible inhibitor of both Ca2+-activated and Ca 2÷independent K + channels (/Kca and/Kv) found in isolated type I cells of the neonatal rat (Figs. 1-4). The drug appears more potent in inhibiting/Kca (IC5o approximately 5 gM) t h a n / K v (IC50 approximately 20 #M; Fig. 4). Its effects appear to be selective for K ÷ channels in the type I cell, as it was without significant effect on L-type Ca z+ channel currents at a dose (10 #M) which significantly reduced K + current amplitudes. Furthermore, the actions of doxapram on K + currents appears to show at least a degree of tissue selectivity: 10 #M doxapram did not significantly affect K + currents recorded in the neuronal cell line NG-108 15 (Fig. 7). This observation is in agreement with the fact that there are few observable side-effects of the drug in patients who required elevated ventilation. Although the mechanisms of the effects of doxapram reported here are at present unknown, time course studies of its blockade of b o t h / K c a a n d / K v (Fig. 5) suggest that its effects may arise either due to a direct action on these two channel types, or by interaction with a closely coupled receptor: a receptor-mediated action involving generation of second messengers would be expected to give rise to a slower inhibition of these channels. In order to explain how the effects of doxapram reported here could lead to increased CSN discharge, and hence elevated ventilation 16, it is important to compare the drug's action on type I cell ion channels with those of the physiological stimuli of the carotid body. Both hypoxia and acidity suppress K ÷ channels in type I cells3' 9,12,13,19,20,23.28. In the neonatal rat these effects have been shown by this laboratory to be selective for/Kc~ t9, 20,23, whereas in a similar preparation used by other workers 28 and also in long-term cultured type I cells from fetal rabbits 9 the K ÷ current subtype suppressed by hypoxia has not yet been fully characterized. In adult rabbit tissue the effects of these stimuli have been demonstrated on a K + current remaining after run-down of the Ca 2÷ current, suggesting that the underlying K ÷ channel is not Ca2+-activated ~3. Single channel recordings have shown, in long-term cultured type I cells obtained from fetal rabbit tissue, that hypoxia inhibits a
121 large-conductance K + channel 3, and it should be noted that the hypoxia- and a c i d - s e n s i t i v e / K ~ in neonatal rat cells is also of high conductance, as it is inhibited by charybdotoxin 19, a peptide toxin which selectively inhibits high conductance Ca2+-activated K + channels. Apamin (another peptide toxin which only inhibits low conductance Ca2+-activated K ÷ channels) has no effect in these cells 19. In every preparation thus far studied, hypoxia or acidity have been shown to be without observable effects on the inward L-type Ca 2+ current in type I cells 9'12'23'28. Thus doxapram, by inhibiting K ÷ and not Ca 2÷ channels in type I cells, shows much similarity to the effects of physiological stimuli of the carotid body. Suppression of high-conductance K + channels has been proposed as the cause of type I cell depolarization induced by hypoxia 3, and acidity, which also blocks such channels (see above), can also depolarize these cells 5. Such an effect would presumably lead to the opening of the voltage-dependent L-type Ca 2÷ channels, and hence Ca2+-dependent transmitter release. Such a model of chemotransduction in type I cells has further support, in the case of hypoxia at least, in that release of radiolabelled catecholamines in response to hypoxia can be inhibited by the L-type Ca 2÷ channel antagonist nitrendipine ~7. The observations reported in this paper would suggest that doxapram, by inhibiting b o t h / K c a a n d / K v in type I cells, may act in a similar fashion to the phys-
iological stimuli of the carotid body, but further experimentation (such as a study of the effects of doxapram on transmitter release from type I cells and the Ca 2+dependence of any such effects) is required before such a hypothesis for the actions of doxapram can be validated. Furthermore, although the above-mentioned actions of chemostimuli on ion channels in type I cells suggest a stimulus-depolarization-secretion model for chemotransduction, there are alternative possible mechanisms for such a process occurring in response to both acidic and hypoxic stimuli. Acidity has recently been demonstrated to lead to elevated intracellular [Ca 2÷] (and hence transmitter release) by reversing the operation of the Na t Ca 2÷ exchanger in the type I cell plasma membrane 27. Other workers have provided evidence that hypoxia raises cytosolic [Ca 2÷] by causing its release from intracellular stores, rather than stimulating its influx across the plasma membrane 1. The effects of doxapram on membrane transporters and intracellular stores require study before the relative importance of its actions reported here (or indeed those of hypoxia and acidity) on ion channels in type I cells can be fully determined.
Acknowledgements. This work was supported by the Wellcome Trust. I would like to thank J. Dempster (Strathclyde University) for his analysis software.
REFERENCES 1 Biscoe, T.J. and Duchen, M.R., Responses of type I cells dissociated from rabbit carotid body to hypoxia, J. Physiol., 428 (1990) 39-59. 2 Brown, D.A. and Higashida, H., Voltage- and calcium-activated potassium currents in mouse neuroblastoma × rat glioma hybrid cells, J Physml., 397 (1988) 149-166. 3 Delpiano, M.A. and Hescheler, J., Evidence for a PO2-senslrive K + channel in the type-I cell of the rabbit carotid body, FEBS Lett., 249 (1989) 195-198. 4 Duchen, M R., Caddy, K.W.T., Kirby, G.C., Patterson, D.L., Ponte, J. and Biscoe, T.J., Biophysical studies of the cellular elements of the rabbit carotid body, Neuroscience, 26 (1989) 291-311. 5 Eyzagmrre, C., An overview of mechamsms associated with the onset of sensory &scharges in the carotid nerve. In C. Beimonte, D.J. Pallot, H. Acker and S. Fidone (Eds.), Arterial Chemoreceptors, Leicester University Press, U.K., 1981, pp. 20-44 6 Fldone, S.J. and Gonzalez, C., Initiation and control of chemoreceptor activity m the carotid body. In A.P. Fishman, N.S. Cherniack, J.G. Widdicome and S.R. Geiger (Eds.), Handbook of Physiology. III: The Respiratory System, Vol 2, Part 1, American Physiological Society, Bethesda, MD, 1986, pp. 247-312. 7 Fidone, S., Gonzalez, C. and Yoshizaki, K., Effects of low oxygen on the release of dopamine from the rabbit carotid body m wvo, J PhysioL, 333 (1982) 93-110. 8 Hamill, O.P., Marry, A., Neher, E., Sakmann, B. and Slgworth, EJ., Improved patch clamp techmques for hlgh-resolu-
9
10 11 12
13
14
15 16 17
tion current recording from cells and cell-free membrane patches, Pfl~gers Arch., 391 (1981) 85-100. Hescheler, J., Delpiano, M.A., Acker, H. and Pietruschka, F., Iomc currents on type-I cells of the rabbit carotid body measured by voltage-clamp experiments and the effect of hypoxia, Brain Research, 486 (1989) 79-88. Hirsch, K. and Wang, S.C., Selective respiratory stimulating action of doxapram compared to pentylenetetrazol, J. Pharmacol. Exp. Ther., 189 (1974) 1-11. Le Douarin, N., Le L~evre, C. and Fontmne, J., Recherches expenmentales sur l'origme embryologique du corps carotidien chez les oiseaux, C.R. Acad. Sci. Ser. D., 275 (1972) 583-586. Lopez-Barneo, J., Lopez-Lopez, J.R., Urena, J. and Gonzalez, C., Chemotransductlon m the carotid body: K + current modulated by PO2 in type I chemoreceptor cells, Science, 241 (1988) 580-582. Lopez-Lopez, J., Gonzalez, C., Urena, J. and Lopez-Barneo, J., Low PO2 selectively inhibits K channel activity in chemoreceptor cells of the mammalian carotid body, J. Gen. Physiol., 93 (1989) 1001-1015. McDonald, D.M. and Mitchell, R.A., A quantitative analysis of synaptic connections in the rat carotid body. In M.J. Purves (Ed.), The Pertpheral Arterial Chemoreceptors, Cambridge University Press, London, 1975, pp. 101-131. Narahashi, T., Tsunoo, A. and Yoshli, M., Characterization of two types of calcium channels in mouse neuroblastoma cells, J. Physiol., 383 (1987) 231-249. Nishml, T., Mokashi, A. and Lahiri, S., Stimulation of carotid chemoreceptors and ventilation by doxapram in the cat, J. Appl. Physiol., 52 (1982) 1261-1265. Obeso, A., Fldone, S. and Gonzalez, C,, Pathways for calcium
122
18
19
20
21
22
23
entry into type I cells: slgniticance for the secretory response. In J.A. Ribeiro and D.J. Pallot (Eds.), Chemoreceptors m Respiratory Control, Croom Helm, London, 1987, pp. 91-97. Oyama, Y., Walker, J.L. and Eyzaguirre, C., The intracellular chloride activity of giomus cells in the isolated rabbit carotad body, Brain Research, 368 (1986) 167-169 Peers, C., Selective effect of lowered extracellular pH on Ca 2+dependent K ÷ currents in type I cells isolated from the neonatal rat carotid body, J. Physiol., 422 (1990) 381-395. Peers, C., Hypoxic suppression of K + currents in type I carotid body cells: selective effect on the Ca2+-activated K + current, Neurosct. Lett, 119 (1990) 253-256. Peers, C., Effects of D600 on hyporac suppression of K + currents in isolated type I carotid body cells of the neonatal rat, FEBS Lett., 271 (1990) 37-40. Peers, C., Effects of doxapram on ionic currents in isolated type I cells of the neonatal rat carotad body, J. Physiol., 438 (1991) 230P. Peers, C. and Green, EK., Intracellular acidosis inhlbtts Ca 2+dependent K + currents in isolated type I cells of the neonatal rat caroUd body, J. Phystol., 437 (1991) 589-602.
24 Peers, C., Lang, B., Newsom-Davls, J. and W.-Wray, D., Selectwe action of myasthenic syndrome ant~bodles on calcium channels in a rodent neuroblastoma x ghoma cell line, J. Phystol, 421 (1990) 293-308. 25 Peers, C. and O'Donnell, J., Potassmm currents recorded m type I carotid body cells from the neonatal rat and their modulation by chemoexcitatory agents, Bram Research, 522 (1990) 259-266. 26 Rigual, R., Gonzalez, E., Fldone, S. and Gonzalez, C., Effects of low pH on synthesis and release of catecholammes in the cat carotid body m vitro, Brain Research, 309 (1984) 178-181. 27 Rocher, A , Obeso, A., Gonzalez, C and Herreros, B., Ionic mechamsms for the transductlon of acl&c stimuli in rabbit carotid body glomus cells, J. Physiol., 433 (1991) 533-548. 28 Stea, A. and Nurse, C.A., Whole-cell and perforated-patch recordings from O2-sensitive rat carotid body cells grown in short-, and long-term culture, Pfl~igers Arch., 418 (1991) 93-101. 29 Ts~en, R.W., Llpscombe, D., Ma&son, D.V., Bley, K.R. and Fox, A.P., Multiple types of neuronal calcium channels and their selective modulation, Trends Neuroscz., I1 (1988) 431-438.
