Aria Physiol Scand 1992, 144, 463468
Voltage-sensitive calcium channels in a human small-cell lung cancer cell line J. J. P A N C R A Z I O " , H. K. O I E t and Y. I. K I M " f
Departments of * Biomedical Engineering and Neurology, University of Virginia Health Sciences Center, USA, and the National Cancer Institute, Navy Medical Oncology Branch, Naval Hospital Bethesda, USA
t
PANCRAZIO, J.J., OIE,H.K. & KIM, Y.I. 1992. Voltage-sensitive calcium channels in a human small-cell lung cancer cell line. Acta Physiol Scand 144, 463468. Received 1 October 1991, accepted 24 October 1991. ISSN 0001-6772. Departments of Biomedical Engineering and Neurology, University of Virginia Health Sciences Center, and the National Cancer Institute, Navy Medical Oncology Branch, Naval Hospital Bethesda, USA. Utilizing the whole-cell patch-clamp method we assessed the Ca2+ current (Ica) in well-established cell lines from human small-cell carcinoma (SCC) of the lung, NCIH209 and NCI-H187. The Ca2+ current was readily observed in H209 tumour cells (90% of the cells tested), whereas H187 tumour cells only occasionally expressed Caz+ channels (26% of the cells tested). H209 Ca2+current was evoked by potentials greater than -30 mV and exhibited partial inactivation over the duration of a 40 ms command potential. This inward current was unchanged by alteration of the holding potential from -80 to -40 mV and the activation phase of the Ca2+current was best fitted by Hodgkin-Huxley m(t)* kinetics. H209 Ca2+current was reduced over 80% by verapamil (100 p ~ )whereas , w-conotoxin (5 p ~ appeared ) to be without effect. In contrast, H209 Ca2+ current was rapidly abolished by nifedipine (10 PM), strongly suggesting the presence of L-type Ca2+channels. Voltage-gated Ca2+channels may be important to the secretion of ectopic hormones and the etiology and pathogenesis of Lambert-Eaton syndrome, an autoimmune disorder of the motor nerve terminal in which autoantibodies directed against voltage-gated Ca2+channels are produced. Key words : CaZ+channels, Lambert-Eaton syndrome, patch-clamp, small-cell lung cancer.
Small-cell carcinoma of the lung (SCCL) is a tumour thought to be derived from secretory cells of the bronchial epithelia and, therefore, has been considered a model for the study of pulmonary endocrine electrophysiology (McCann et al. 1985). SCCL releases a diverse group of ectopic hormones including adrenoCorrespondence: Dr Y. I. Kim, Department of Biomedical Engineering, Box 377, University of Virginia Health Sciences Center, Charlottesville, VA 22908, USA. This work was supported by Javits Neuroscience Investigator Award to Y.I. K. (NIH Grant NS18607) and a research grant from the Muscular Dystrophy Association.
corticotropic hormone, gastrin-releasing peptide (GRP), and antidiuretic hormone (Russell et al. 1990). Past studies indicate that hormone secretion in small-cell carcinoma (SCC) cells occurs in a Ca2+-dependent manner (Sorenson et al. 1983, Moody et al. 1983), perhaps via exocytosis (Kim et al. 1989). Although several mechanisms may exist for the elevation of the cytoplasmic calcium concentration necessary for secretion, GRP release can be stimulated by high external K+, suggesting the participation of voltagesensitive Ca2+ channels in Ca2+ entry (Moody et al. 1983). Several past investigations have presented evidence for the existence of voltage-gated Ca2+ channels in SCC cells. McCann et al., using an
463
anodal-break stimulation method, reported that SCX; cells exhibit Ca' -spike electrogcnesis (McCann et al. 1983, 1 l c C a n n et u / . 1981). Depolarization-induced Ca2* flux, an assa!for intracellular Ca2+([Ca"Ii), in SCC cells has been shown to he reduced by a dih! drop!-ridine antagonist (Robtrts ei R / . 198.5) and w-conotosin (De.4izpurua et ml. 198X), suggesting the presence of 1.- and s-type Ca" channels. How-ever, the measurement of [Ca'*], transients may not be a complete indicator of Ca'- channel activitJ- (Fox 1'1 t i / . 1990) a n d flux studies cannot exclude the presence of low-threshold (T-tl-pe) Ca" channels. Still, with the use of h-hole-cell patchclamp techniques, direct measurements of Ca'+ channel currents in SCC cells have been elusive. O n e investigation reported an absence ofmeasurable Ca'. currents in SCC cells (Johannson ct t i / . 19889), whereas another study rei-ealed the existence o f a high-volta~e-activated (IIVA4)Ca" currcnt (Pancrazio et 01. 1989). T h e objective of the present study was to identifJ- further the type(s) of voltage-gated Ca" channels in the established SCC cell lines ?iCI-H209 and XCI-
tI 1 87. \ l A T E K I A L , S AVI) M E T H O D S Crll cxlturr. The human S(X:I, cell lines NCI-I I209 and NCI-IlI8i were generousl!. provided b! A. Gazdar of the National Cancer Institute, Bethesda. hlaryland, USA. Cells were maintained in KPIII 1h4O Lvith 5(),, heat-inacti\atcd foetal hot ine serum supplemented with 2 n1.I i.-glutamine. Flasks containing the cells were incubated at 3 i " C in 5",, Co, and air. Cultures were subcultured once ewr! 5-7 days hy mechanical dissociation. In order to immobilize the tumour cells for each esperiment, the cells were affixed tn glass coverslips coated with pol!-i.-lysine. .Yolicrions. The external bathing solution contained (in mht): 130 NaCI, 5. KCI, 10 CaCI,, and 10 .\-2h!-drosyeth~-lpiperazine-~'-2-ethancsulfonic acid (1lEPI.3). Tetrodotoxin (5 /m) was routinel! added to the bath solution to suppress voltage-gated S a current. The internal pipette solution contained (in 1. 20 tetraeth!-lammonium-CI, 10 :IL, and 11 ethl-leneglycol his($aminoeth! I ether)-.V-,\-,\-'-.Y-tetraacetic acid. :\I1 solutions Mere adjusted to pH 7 . 2 with SaOII and were filtered through a 0.2pm Millipore filter before use. l,harmltr-c/loRirr/ q t ' n t s . Nifedipine was dissoli-ed in lOO",, ethanol as a stock solution of1.18 m u . Control experiments \!ere conducted with the same amount of
ethanol present in the external solution as the treatment experiments. The s y t h e t i c form of the marine snail poison m-conotoxin GVIA (aCgTX) obtained from Pcninsula Cornpan!- was dissolved in distilled water a t a concentration of -5OO/mi and diluted to a final concentration in the recording chamber. Both nifedipine and mCgTX were stored at 4 "C in the absence of light. fF'holect11 putci~-r~/urnp rtcording und nnu,lysis. Whole-cell recordings were made at room temperature (22-23 "C;) with standard patch-clamp techniques (IIamill 1'1 u I . 1081) using a 1,ist EPC-i patch-clamp amplifier (I>ist Electronic, Darmstadt, Germany). Patch electrodes, fashioned from Kimas capillarltubes (outer diameter 1.5-1.8 mm), were coated w-ith S! Igard-184 resin to reduce capacitance. Patch electrode tips were heat-polished using the List I./MCPZ-101 pipette forging system (1.ist Electronic, Darmstadt, Germany). A digital pulse stimulator based on an AIXI-0.5 microcomputer was used to produce command voltage pulses at a rate of 0.5 flz. Kecordings were initiated 1 min after establishing the u hole-cell recording configuration. Unless otherwise noted, currents were evoked by depolarizing step potentials of 40 or 200 msec in duration from a holding potential of -80 mV. Current records were loxs-pass filtered ( - 3 dB at 3 kIIz) with an X-pole Ikssel filter and digitized for analysis by a PDP 11/73 computer. Currents elicited by 10 ms step potentials were sampled at 10 kHz, w-hereas 200 ms depolarizations ncre digitized at 2 kHz. Series resistance errors, usuall!- small, were subtracted from the command potentials to determine the actual clamp 1-oltages. 1,inear leak currents and capacitive artifacts \yere determined b!- calculating the ensemble average of 16 consecutive current traces triggered by a 20 niV depolarizing step. Esponential curve fitting was performed on an IBIf 386 compatible microcomputer equipped with a floating point coprocessor using the s w r AI LOSLIK software module. W-here appropriate, data are presentcd as meankstandard error of the mean (XU) and the number ofcells tested (n). Unless otherw-ise noted, measurements were made from separate control and treatment groups in order to avoid Ca" current (Z(.:,) rundown (Fenwick et ul. 19x2). For statistical comparisons, whole-cell currents were normalized with respect to cell membrane capaci tance (p.l/pF).
RESULTS
It kolr-rell Cn" cliirnnrl tharirrttrictiLs i f i t h outlsdrd K * and inward Na+ currents suppressed, fell of the M187 cells (7 of 27 cells tested) and most H209 turnour cells (17 of 19 cellr terted) exhibited a partial11 inactivating
465
CaZi channels in lung cancer cells
-20
+ l o 4 \
O +20
0
1 I 25 ms
10 rns
-20
4
+ 1O/
+20
+40
25 ms
Fig. 1. Voltage-gated Ca2+ currents (ZrJ measured from I1209 tumour cells. Depolarizing pulses 40 ms in duration were applied every 2 s from a holding potential of -80 mV. (a) Currents elicited from a typical cell; the absolute voltages are given for each trace; C, = 5.7 pF. (b) Current-voltage relationship for peak (") inward currents for the cell shown in 1a.
Ca2+current for potentials greater than - 30 mV and 40 ms in duration (Fig. 1). Peak H187 I,,, which was evoked by command pulses of 10 to 20 mV, reached only 4.2 _+ 0.7 p A / p F with cell membrane capacitance (C,) of 9.4 f 1.4 p F (n = 7). Subsequent studies focused on the H209 cell line which was relatively more prominent. Peak H209 I,,, which exhibited a similar voltage sensitivity as H187, was substantially larger reaching 13.4-t 1.5 p A / p F (a = 13) with C, = 5.7 f0.7 p F (n = 13). This high-voltage-activated (HVA) Ca2+ current observed in H209 cells was insensitive to depolarized prepulse potentials of -40 mV (4 of 4 cells tested). Test pulses 200 ms in length were used to resolve more clearly the time course of H209 I,.&. T h e decaying phase of the inward current was well-fitted to a single exponential with T ranging from 100 to 400 ms with little or no dependence on the magnitude of the applied test pulse. T h e rates of inactivation varied among the cells listed. For example, the H209 cell in Figure 2a
+
+
Fig. 2. Variable degree of inactivation of voltagedependent Ca2+currents from H209 cells. Depolarizing voltages 200 ms in duration were applied every 2 s from a holding potential of -80 mV. Zca was elicited by four test pulses whose absolute voltage is given in each record. (a) Current from a cell with an apparent steady-state component: C, = 7.1 pF. (b) Current from a cell with little or no steady-state component: C,, = 4.5 pF. exhibits incomplete inactivation, whereas the cell depicted in Figure 2 b shows nearly complete inactivation. T h e activation time course ofl,., was measured for test voltages ranging from 0 to 40 mV from a holding potential of -80 mV. T h e timedependent Ca2' conductance was fitted to the Hodgkin-Huxley (1952) activation parameters m(t) raised to the integer power, p, using a nonlinear least squares routine. I n more than 80% of the traces, the activation phase of H209 Icawas best fitted with m(t)z kinetics. T h e tirnc constant of activation was very brief; 7 , was 0.7kO.l ms ( n = 6) and 0.4fO.l ms for 0 and 30 mV, respectively.
+
+
Pharmacological sensitivity of H209 Ira I t is now clear that the time course of a Ca2+ current is not conclusive evidence for a class of Ca2+ channels (Hess 1990). Ca'+ channel characterization requires the determination of pharmacological sensitivity to such agents as dihydropyridines, phenylalkylamines, and Dconotoxin GVIA (mCgTX). Therefore, H209
t!pe C a 2 +current. Clearly, not all S C X cell lines express Ca" channels consistentlj-. Two cell lines in which we have readily observed (;a" currents are NCI-H146 (Pancrazio et al. 1089) and NCI-H209 as described in the present Istudy. 50 ms I3209 I( i( eshibited rii(i)' actii-ation kinetics similar to rat pituitary gonadotrophs (Stutzsin r / ti/. 1989), although the rates of inactivation seemed to \-ar!-. Interestingly, €3209 and H187 I , ,, bear superficial resemblaiice to thc Ck" current previously reported tbr H69 and H128 SCCT, tumour cells (Pancrazio et ill. 1989), which also shons partial inactit ation o t er a 40 Ins test potential. 1,ike E-1187,neither 1169 nor M 128 consistentll- express voltage-gated (h2' channels. Such inactivation of an H V A ciirreiit 5 ms . 3. Actions of nifedipine and iLg'TZ on SCXl obser\ ed in H200 cells initially suggested the presence of N-t!-pe L a 2 +channels. Howc\er, the )O (;J" current. (a) Peak I, evoked h! a cominand ; t i i l \ i to T I ( J ni\ from a holding potential of - 50 m \ I1209 transient current appeared to be insensitive hr
Voltage-sensitive calcium channels in a human small-cell lung cancer cell line J. J. P A N C R A Z I O " , H. K. O I E t and Y. I. K I M " f
Departments of * Biomedical Engineering and Neurology, University of Virginia Health Sciences Center, USA, and the National Cancer Institute, Navy Medical Oncology Branch, Naval Hospital Bethesda, USA
t
PANCRAZIO, J.J., OIE,H.K. & KIM, Y.I. 1992. Voltage-sensitive calcium channels in a human small-cell lung cancer cell line. Acta Physiol Scand 144, 463468. Received 1 October 1991, accepted 24 October 1991. ISSN 0001-6772. Departments of Biomedical Engineering and Neurology, University of Virginia Health Sciences Center, and the National Cancer Institute, Navy Medical Oncology Branch, Naval Hospital Bethesda, USA. Utilizing the whole-cell patch-clamp method we assessed the Ca2+ current (Ica) in well-established cell lines from human small-cell carcinoma (SCC) of the lung, NCIH209 and NCI-H187. The Ca2+ current was readily observed in H209 tumour cells (90% of the cells tested), whereas H187 tumour cells only occasionally expressed Caz+ channels (26% of the cells tested). H209 Ca2+current was evoked by potentials greater than -30 mV and exhibited partial inactivation over the duration of a 40 ms command potential. This inward current was unchanged by alteration of the holding potential from -80 to -40 mV and the activation phase of the Ca2+current was best fitted by Hodgkin-Huxley m(t)* kinetics. H209 Ca2+current was reduced over 80% by verapamil (100 p ~ )whereas , w-conotoxin (5 p ~ appeared ) to be without effect. In contrast, H209 Ca2+ current was rapidly abolished by nifedipine (10 PM), strongly suggesting the presence of L-type Ca2+channels. Voltage-gated Ca2+channels may be important to the secretion of ectopic hormones and the etiology and pathogenesis of Lambert-Eaton syndrome, an autoimmune disorder of the motor nerve terminal in which autoantibodies directed against voltage-gated Ca2+channels are produced. Key words : CaZ+channels, Lambert-Eaton syndrome, patch-clamp, small-cell lung cancer.
Small-cell carcinoma of the lung (SCCL) is a tumour thought to be derived from secretory cells of the bronchial epithelia and, therefore, has been considered a model for the study of pulmonary endocrine electrophysiology (McCann et al. 1985). SCCL releases a diverse group of ectopic hormones including adrenoCorrespondence: Dr Y. I. Kim, Department of Biomedical Engineering, Box 377, University of Virginia Health Sciences Center, Charlottesville, VA 22908, USA. This work was supported by Javits Neuroscience Investigator Award to Y.I. K. (NIH Grant NS18607) and a research grant from the Muscular Dystrophy Association.
corticotropic hormone, gastrin-releasing peptide (GRP), and antidiuretic hormone (Russell et al. 1990). Past studies indicate that hormone secretion in small-cell carcinoma (SCC) cells occurs in a Ca2+-dependent manner (Sorenson et al. 1983, Moody et al. 1983), perhaps via exocytosis (Kim et al. 1989). Although several mechanisms may exist for the elevation of the cytoplasmic calcium concentration necessary for secretion, GRP release can be stimulated by high external K+, suggesting the participation of voltagesensitive Ca2+ channels in Ca2+ entry (Moody et al. 1983). Several past investigations have presented evidence for the existence of voltage-gated Ca2+ channels in SCC cells. McCann et al., using an
463
anodal-break stimulation method, reported that SCX; cells exhibit Ca' -spike electrogcnesis (McCann et al. 1983, 1 l c C a n n et u / . 1981). Depolarization-induced Ca2* flux, an assa!for intracellular Ca2+([Ca"Ii), in SCC cells has been shown to he reduced by a dih! drop!-ridine antagonist (Robtrts ei R / . 198.5) and w-conotosin (De.4izpurua et ml. 198X), suggesting the presence of 1.- and s-type Ca" channels. How-ever, the measurement of [Ca'*], transients may not be a complete indicator of Ca'- channel activitJ- (Fox 1'1 t i / . 1990) a n d flux studies cannot exclude the presence of low-threshold (T-tl-pe) Ca" channels. Still, with the use of h-hole-cell patchclamp techniques, direct measurements of Ca'+ channel currents in SCC cells have been elusive. O n e investigation reported an absence ofmeasurable Ca'. currents in SCC cells (Johannson ct t i / . 19889), whereas another study rei-ealed the existence o f a high-volta~e-activated (IIVA4)Ca" currcnt (Pancrazio et 01. 1989). T h e objective of the present study was to identifJ- further the type(s) of voltage-gated Ca" channels in the established SCC cell lines ?iCI-H209 and XCI-
tI 1 87. \ l A T E K I A L , S AVI) M E T H O D S Crll cxlturr. The human S(X:I, cell lines NCI-I I209 and NCI-IlI8i were generousl!. provided b! A. Gazdar of the National Cancer Institute, Bethesda. hlaryland, USA. Cells were maintained in KPIII 1h4O Lvith 5(),, heat-inacti\atcd foetal hot ine serum supplemented with 2 n1.I i.-glutamine. Flasks containing the cells were incubated at 3 i " C in 5",, Co, and air. Cultures were subcultured once ewr! 5-7 days hy mechanical dissociation. In order to immobilize the tumour cells for each esperiment, the cells were affixed tn glass coverslips coated with pol!-i.-lysine. .Yolicrions. The external bathing solution contained (in mht): 130 NaCI, 5. KCI, 10 CaCI,, and 10 .\-2h!-drosyeth~-lpiperazine-~'-2-ethancsulfonic acid (1lEPI.3). Tetrodotoxin (5 /m) was routinel! added to the bath solution to suppress voltage-gated S a current. The internal pipette solution contained (in 1. 20 tetraeth!-lammonium-CI, 10 :IL, and 11 ethl-leneglycol his($aminoeth! I ether)-.V-,\-,\-'-.Y-tetraacetic acid. :\I1 solutions Mere adjusted to pH 7 . 2 with SaOII and were filtered through a 0.2pm Millipore filter before use. l,harmltr-c/loRirr/ q t ' n t s . Nifedipine was dissoli-ed in lOO",, ethanol as a stock solution of1.18 m u . Control experiments \!ere conducted with the same amount of
ethanol present in the external solution as the treatment experiments. The s y t h e t i c form of the marine snail poison m-conotoxin GVIA (aCgTX) obtained from Pcninsula Cornpan!- was dissolved in distilled water a t a concentration of -5OO/mi and diluted to a final concentration in the recording chamber. Both nifedipine and mCgTX were stored at 4 "C in the absence of light. fF'holect11 putci~-r~/urnp rtcording und nnu,lysis. Whole-cell recordings were made at room temperature (22-23 "C;) with standard patch-clamp techniques (IIamill 1'1 u I . 1081) using a 1,ist EPC-i patch-clamp amplifier (I>ist Electronic, Darmstadt, Germany). Patch electrodes, fashioned from Kimas capillarltubes (outer diameter 1.5-1.8 mm), were coated w-ith S! Igard-184 resin to reduce capacitance. Patch electrode tips were heat-polished using the List I./MCPZ-101 pipette forging system (1.ist Electronic, Darmstadt, Germany). A digital pulse stimulator based on an AIXI-0.5 microcomputer was used to produce command voltage pulses at a rate of 0.5 flz. Kecordings were initiated 1 min after establishing the u hole-cell recording configuration. Unless otherwise noted, currents were evoked by depolarizing step potentials of 40 or 200 msec in duration from a holding potential of -80 mV. Current records were loxs-pass filtered ( - 3 dB at 3 kIIz) with an X-pole Ikssel filter and digitized for analysis by a PDP 11/73 computer. Currents elicited by 10 ms step potentials were sampled at 10 kHz, w-hereas 200 ms depolarizations ncre digitized at 2 kHz. Series resistance errors, usuall!- small, were subtracted from the command potentials to determine the actual clamp 1-oltages. 1,inear leak currents and capacitive artifacts \yere determined b!- calculating the ensemble average of 16 consecutive current traces triggered by a 20 niV depolarizing step. Esponential curve fitting was performed on an IBIf 386 compatible microcomputer equipped with a floating point coprocessor using the s w r AI LOSLIK software module. W-here appropriate, data are presentcd as meankstandard error of the mean (XU) and the number ofcells tested (n). Unless otherw-ise noted, measurements were made from separate control and treatment groups in order to avoid Ca" current (Z(.:,) rundown (Fenwick et ul. 19x2). For statistical comparisons, whole-cell currents were normalized with respect to cell membrane capaci tance (p.l/pF).
RESULTS
It kolr-rell Cn" cliirnnrl tharirrttrictiLs i f i t h outlsdrd K * and inward Na+ currents suppressed, fell of the M187 cells (7 of 27 cells tested) and most H209 turnour cells (17 of 19 cellr terted) exhibited a partial11 inactivating
465
CaZi channels in lung cancer cells
-20
+ l o 4 \
O +20
0
1 I 25 ms
10 rns
-20
4
+ 1O/
+20
+40
25 ms
Fig. 1. Voltage-gated Ca2+ currents (ZrJ measured from I1209 tumour cells. Depolarizing pulses 40 ms in duration were applied every 2 s from a holding potential of -80 mV. (a) Currents elicited from a typical cell; the absolute voltages are given for each trace; C, = 5.7 pF. (b) Current-voltage relationship for peak (") inward currents for the cell shown in 1a.
Ca2+current for potentials greater than - 30 mV and 40 ms in duration (Fig. 1). Peak H187 I,,, which was evoked by command pulses of 10 to 20 mV, reached only 4.2 _+ 0.7 p A / p F with cell membrane capacitance (C,) of 9.4 f 1.4 p F (n = 7). Subsequent studies focused on the H209 cell line which was relatively more prominent. Peak H209 I,,, which exhibited a similar voltage sensitivity as H187, was substantially larger reaching 13.4-t 1.5 p A / p F (a = 13) with C, = 5.7 f0.7 p F (n = 13). This high-voltage-activated (HVA) Ca2+ current observed in H209 cells was insensitive to depolarized prepulse potentials of -40 mV (4 of 4 cells tested). Test pulses 200 ms in length were used to resolve more clearly the time course of H209 I,.&. T h e decaying phase of the inward current was well-fitted to a single exponential with T ranging from 100 to 400 ms with little or no dependence on the magnitude of the applied test pulse. T h e rates of inactivation varied among the cells listed. For example, the H209 cell in Figure 2a
+
+
Fig. 2. Variable degree of inactivation of voltagedependent Ca2+currents from H209 cells. Depolarizing voltages 200 ms in duration were applied every 2 s from a holding potential of -80 mV. Zca was elicited by four test pulses whose absolute voltage is given in each record. (a) Current from a cell with an apparent steady-state component: C, = 7.1 pF. (b) Current from a cell with little or no steady-state component: C,, = 4.5 pF. exhibits incomplete inactivation, whereas the cell depicted in Figure 2 b shows nearly complete inactivation. T h e activation time course ofl,., was measured for test voltages ranging from 0 to 40 mV from a holding potential of -80 mV. T h e timedependent Ca2' conductance was fitted to the Hodgkin-Huxley (1952) activation parameters m(t) raised to the integer power, p, using a nonlinear least squares routine. I n more than 80% of the traces, the activation phase of H209 Icawas best fitted with m(t)z kinetics. T h e tirnc constant of activation was very brief; 7 , was 0.7kO.l ms ( n = 6) and 0.4fO.l ms for 0 and 30 mV, respectively.
+
+
Pharmacological sensitivity of H209 Ira I t is now clear that the time course of a Ca2+ current is not conclusive evidence for a class of Ca2+ channels (Hess 1990). Ca'+ channel characterization requires the determination of pharmacological sensitivity to such agents as dihydropyridines, phenylalkylamines, and Dconotoxin GVIA (mCgTX). Therefore, H209
t!pe C a 2 +current. Clearly, not all S C X cell lines express Ca" channels consistentlj-. Two cell lines in which we have readily observed (;a" currents are NCI-H146 (Pancrazio et al. 1089) and NCI-H209 as described in the present Istudy. 50 ms I3209 I( i( eshibited rii(i)' actii-ation kinetics similar to rat pituitary gonadotrophs (Stutzsin r / ti/. 1989), although the rates of inactivation seemed to \-ar!-. Interestingly, €3209 and H187 I , ,, bear superficial resemblaiice to thc Ck" current previously reported tbr H69 and H128 SCCT, tumour cells (Pancrazio et ill. 1989), which also shons partial inactit ation o t er a 40 Ins test potential. 1,ike E-1187,neither 1169 nor M 128 consistentll- express voltage-gated (h2' channels. Such inactivation of an H V A ciirreiit 5 ms . 3. Actions of nifedipine and iLg'TZ on SCXl obser\ ed in H200 cells initially suggested the presence of N-t!-pe L a 2 +channels. Howc\er, the )O (;J" current. (a) Peak I, evoked h! a cominand ; t i i l \ i to T I ( J ni\ from a holding potential of - 50 m \ I1209 transient current appeared to be insensitive hr
