Proc. Nati. Acad. Sci. USA Vol. 90, pp. 765-769, January 1993 Plant Biology Solubilized proteins from carrot (Daucus carota L.) membranes bind calcium channel blockers and form calcium-permeable ion channels (voltage-dependent Ca2+ channel/signal transduction/cytosofic caldum/plant ion channel/liposome) PATRICE THULEAU*t, ANNICK GRAZIANA*, RAOUL RANJEVAf, AND JULIAN I. SCHROEDER* *Department of Biology and Center for Molecular Genetics, University of California, San Diego, La Jolla, CA 92093-0116 and *Centre de Biologie et Physiologie Vdgdtales, Universitd Paul Sabatier, Centre National de la Recherche Scientifique, 118 Route de Narbonne, F-31062 Toulouse Cddex, France Communicated by Bernard 0. Phinney, September 28, 1992 (receivedfor review January 1, 1992)

ABSTRACT Calcium channels have been suggested to play A plasma membrane protein of75-kDa molecular mass that a major role in the initiation of a large number of signal binds with high affinity to a phenylalkylamine azido deriva- transduction processes in higher plant cells. However, molec- tive, (-)-5-(3-azidophenylethyl[3H]methylamino)-2-(3,4,5- ular components of higher plant Ca2+ channels remain uni- trimethoxyphenyl)-2-isopropylvaleronitrile ([3H]LU 49888), dentified to date. Calcium channel blockers of the phenylal- has been identified in carrot cells (13). This protein retains its kylamine family and bepridil specifically inhibit Ca2+ influx ability to bind Ca2+ channel antagonists upon solubilization into carrot (Daucus carota L.) cells. By using a phenylalky- by a nondenaturating detergent, 3-[(3-cholamidopropyl)di- lamine azido derivative, a 75-kDa carrot membrane protein methylammonio]-l-propanesulfonate (CHAPS) (13). In the had been previously identified. Here we have partially purified present study, we investigated, using reconstitution experi- this Ca2+ channel blocker-binding protein by lectin-affinity ments and the patch-clamp technique, whether membrane and ion-exchange chromatographies. The protein fraction con- protein fractions enriched with the 75-kDa binding protein taining the 75-kDa binding protein was incorporated into gat from carrot cells contribute to Ca2+-permeable ion channel liposomes. Single-channel patch-clamp studies on these prote- function and sensitivity to Ca2+ channel blockers. This oliposomes showed the presence of Ca2+-permeable channel approach allows biochemical insight into components con- currents. These Ca2+-permeable channels were not stable. tributing to the structure of Ca2+ channels of higher plants. Recordings after durations of2-10 min showed the appearance of nonselective ion channels with a permeability to calcium and MATERIALS AND METHODS chloride ions. These nonselective Ca2+-permeable ion chan- nels, in contrast, were stable and were recorded for extended Plant Material, Membrane Preparation, and Storage. Sus- durations. The addition of the Ca2+ channel-blocker bepridil pension cultures of carrot (Daucus carota L.) cells were (10 FM) led to the inhibition of these nonselective Ca2+- prepared as described previously (11). Total membranes permeable channels by reducing the probability of channel were prepared according to a published procedure (11). opening. These results suggest that the 75-kDa Ca2+ channel Membranes were used directly after preparation or after blocker-binding protein from carrot cells plays a role in storage at -80'C. Protein concentrations were measured by channel sensitivity to Ca2+ channel inhibitors and may consti- the bicinchoninic acid assay (15). tute one of the components of Ca2+ channels in higher plants. Photolabeling. The presence of the calcium channel blocker-binding protein was verified by photoaffinity labeling A signal-induced increase in the cytosolic Ca2+ concentration with [3H]LU 49888 as described previously (13). The non- is considered to be among the most important intracellular specific component of the binding was determined in the messengers for initiation of metabolic and developmental presence of the unlabeled calcium channel blocker (-)- events in higher plants. Calcium controls various processes, bepridil at 50 ,.M. The photolabeled microsomes were used including cellular organization, ion channel gating and en- as starting material to purify the [3H]LU 49888-binding zyme activities, during physiological responses to external protein complex. Bepridil was from Laboratoire Cerm stimuli (1-5). The influx of calcium into the cytosol of plant (Riom, France), and [3H]LU 49888 (3.15 TBq/mmol) was cells is mediated by Ca2+-permeable signal-regulated ion from Knoll (Ludwigshafen, F.R.G.). channels which traverse the plasma membrane and vacuolar Solubilization and Purification of the Calcium Channel membrane (refs. 6 and 7; for reviews, see refs. 8 and 9). Blocker-Binding Protein. Membrane proteins were solubi- In spite of the demonstration that Ca2+ influx systems are lized as described (13). The solubilized proteins were diluted involved in physiological processes in plants, the molecular 10-fold with 20 mM Tris HCl, pH 7.5/5% (vol/vol) glycerol structure of Ca2+ channel components remains unidentified and loaded onto a concanavalin A (Con A)-Sepharose column in higher plant cells. Pharmacological agents, referred to as (Pharmacia LKB) equilibrated in 20 mM Tris HCl, pH Ca2+ channel blockers or antagonists, are known to interfere 7.5/5% glycerol/0.1% CHAPS (buffer A). The column was with a variety of plant functions (1, 4, 9). High-affinity washed with buffer A and the calcium channel blocker- membrane for these have been char- binding protein was eluted with 0.25 M methyl a-D- receptors compounds mannopyranoside in buffer A and loaded onto a DEAE- acterized in different plants (10-13). Site occupancy of these trisacryl column (IBF) equilibrated in buffer A. Unbound receptors results in the inhibition of Ca2+ uptake into carrot materials were washed with 100 mM NaCl in buffer A, and protoplasts (11, 14). The most potent antagonists of Ca2+ the retained the ion were eluted a influx found to date for higher plants are phenylalkylamines proteins by exchanger by and bepridil (11, 14). Abbreviation: CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]- 1-propanesulfonate. The publication costs of this article were defrayed in part by page charge tPresent address: Centre de Biologie et Physiologie Vdgdtales, payment. This article must therefore be hereby marked "advertisement" Universitt Paul Sabatier, Centre National de la Recherche Scien- in accordance with 18 U.S.C. §1734 solely to indicate this fact. tifique, F-31062 Toulouse CUdex, France. 765 Downloaded by guest on September 28, 2021 766 Plant Biology: Thuleau et al. Proc. Natl. Acad Sci. USA 90 (1993)

gradient 100-300 mM NaCl in buffer A. The fractions con- interval of 100 ,usec, and analyzed by using the program taining the binding protein (eluted at 200 mM NaCl) were PCLAMP (Axon Instruments). For all experiments corrections pooled, diluted 2-fold with buffer A, and loaded onto a second for ionic activities (22) and liquidjunction potentials (23) were Con A-Sepharose column, which was equilibrated with 100 determined and incorporated into the analysis. Amplitude mM NaCl in buffer A. The column was washed with 100 mM histograms were measured by determining the amplitude of NaCl in buffer A and the binding protein was eluted with 0.25 all samples digitized by the computer and therefore reveal M methyl a-D-mannoside/100 mM NaCi in buffer A. The single-channel conductances and changes in open probabil- final extract (partially purified extract) was desalted by using ities (20). a PD-10 Sephadex G-25M column (Pharmacia), equilibrated in 20 mM Tris HCl, pH 7.5/5% glycerol/1% CHAPS and was RESULTS concentrated by using a Centricon 10 microconcentrator (Amicon). All procedures were performed at 4TC. Purification and Photolabeling. Previous data demon- Reconstitution and Formation of Large Liposomes. For strated that carrot cell membranes could be specifically reconstitution experiments, solubilization and purification of photolabeled by using the phenylalkylamine azido derivative the calcium channel blocker-binding protein was performed [3H]LU 49888 (13). Therefore a strategy, developed to purify according to the procedure described above without previous the calcium channel from skeletal muscle transverse tubules, treatment with [3H]LU 49888. was used for this study (24). After photolabeling, the [3H]LU Total solubilized proteins or partially purified extracts 49888-binding protein complex was purified by a combination were mixed with L-a-lecithin lipids from soybean (type II-S; of Con A-Sepharose and ion-exchange chromatographies, Sigma) dissolved in 20 mM Tris-HCl, pH 7.5/5% glycerol/1% resulting in a final purification factor of 24.6-fold and 30% CHAPS (protein-to-lipid ratio: 1 mg/100 mg). The reconsti- recovery (based on the starting soluble radioactive material) tution was performed at 40C by removing the detergent by (Table 1). This purification procedure allowed solubilization dialysis for 72 hr against three 1-liter batches of 20 mM and purification of the calcium channel blocker-binding pro- Tris HCl, pH 7.5. Subsequently, the proteoliposomes were tein from membranes which had not been photolabeled with transformed into large liposomes, suitable for patch-clamp [3H]LU 49888. The partially purified fraction contained sev- studies, by using the dehydration-rehydration technique (16, eral polypeptides (Fig. 1A, lane 1) among which only the 17). The rehydration was carried out with 20 mM Hepes-Tris, 75-kDa peptide was specifically photolabeled (Fig. 1B). pH 6.5/90 mM CaCl2. NaDodSO4/PAGE and silver nitrate staining of the par- Patch-Clamp Recording and Data Analysis. All experiments tially purified fraction showed the presence of four major were performed at room temperature (21 ± 20C). For patch- polypeptides (Fig. 1A, lane 1). However, in control experi- clamp measurements, 2.5 .l oflarge liposomes was diluted in ments, without loading carrot protein extracts onto the Con 1.5 ml of 20 mM Hepes-Tris, pH 6.5/90 mM CaC12 (bath A-Sepharose column, three ofthese polypeptides (66, 32, and solution), resembling the solution used for identification of 16 kDa) were eluted from the column with the equilibration the molecular components contributing to skeletal muscle buffer and were therefore contaminants from the Con Ca2+ channel function (18, 19). Single-channel currents were A-Sepharose column (Fig. 1A, lane 2). These polypeptides measured by patch-clamp techniques (20). Kimax-51 glass could be the result ofthe nonspecific elution ofthe Con A not pipettes (Kimble Glass, Vineland, NJ) were filled with 20 mM covalently bound to the support column. The presence of Hepes-Tris, pH 6.5/9 mM CaCl2 containing 200 mM D-man- these contaminating polypeptides in the purified sample nitol to adjust the osmolality of the pipette solution to the indicates that our estimated final purification factor can be same osmolality as the bath solution (240 mmol kg-1). After viewed as a minimal estimate. The protein preparation par- the pipette was sealed, withdrawal of the pipette from the tially purified from carrot membranes was enriched with the membrane surface and quickly passing the tip through the 75-kDa peptide. air/buffer interface resulted in an excised membrane patch Electrophysiological Studies: Control Experiments. To ex- ("inside-out configuration"; see ref. 20). Single-channel cur- clude any nonspecific electrical activities, patch-clamp ex- rents were measured and recorded with an Axopatch 200 periments were performed on different control liposomes. No amplifier (Axon Instruments, Burlingame, CA). Membrane channel activity was observed at pipette holding potentials potentials are referred to as the potential on the interior ofthe ranging from -150 to + 150 mV on giant liposomes made as pipette with respect to the bath electrode, which was held to described in Materials and Methods in the absence of carrot virtual ground. Ion channel recordings were stored and protein extracts (n = 10) and on giant liposomes reconstituted analyzed after pulse code modulation and subsequently dig- with thermally denaturated crude (n = 10) or thermally itized by using an Axolab TL-1 DMA interface (Axon In- denatured partially purified solubilized proteins (n = 5). In struments) as described previously (21). Recorded single- addition, patch-clamp analysis of liposomes reconstituted channel data were filtered at 0.5 kHz, digitized at a sampling with the contaminating proteins eluted from the Con A-Seph- Table 1. Purification of the complex of calcium channel blocker-binding protein with [3H]LU 49888 from membranes of carrot cells Binding,* Recovery, Protein, Specific activity, Purification, Step dpm x 10- % mg dpm x 10-5/mg fold Solubilized proteins 11.2 100 1.31 8.5 1 Con A-Sepharose 8.51 76 0.1 85.1 10 DEAE-trisacryl 3.87 34.5 0.021 184.4 21.6 Con A-Sepharose 3.36 30 0.016 210.4 24.6 The membranes (200 ,g) were photolabeled with 5 nM [3H]LU 49888 in the absence (assay) or the presence of 50 AM bepridil (control). Subsequently, from both photolabeled samples, the "LU 49888 receptor" was solubilized and purified as described in Materials and Methods. *At each step of the purification, the radioactivity associated with the different protein fractions was determined and the presence ofthe complex was calculated by the difference (specific binding) between the radioactivity measured in the assay and the one found in the control. The dpm values were automatically determined and calculated by the electronic counter, which was calibrated for 3H. Downloaded by guest on September 28, 2021 Plant Biology: Thuleau et al. Proc. Natl. Acad. Sci. USA 90 (1993) 767 (Fig. 3 B and C), between the Nernst potential for Cl- (-47 B3 mV) and the Nernst potential for Ca2+ (+24 mV). Therefore, these channels displayed a permeability to both calcium and chloride ions. The permeability ratio for Ca2+ with respect to Cl- (PCa2+/PCj-), calculated from the Goldman-Hodgkin- Katz equation for divalent ions over monovalent ions (25),

a P 7 was 0.7. The single-channel conductance of these currents was 67 ± 10 pS (±SD, n = 7) (Fig. 3). In contrast to the Ca2+-selective channels, these nonse- lective Ca2+- and Cl--permeable channels were more stable and could be recorded for durations ofup to 60 min, allowing pharmacological studies. When the calcium channel blocker AM -, I 1 I'd. * - (-)-bepridil (10 t&M) was added in the bath solution, single- channel currents through these nonselective channels were immediately blocked (n = 6) (Fig. 4). Bepridil led to a substantial reduction of the open-state probability of this channel (Fig. 4). The inhibition was reversed (n = 5) by perfusing the bath solution with the buffer devoid of bepridil (data not shown). An identical inhibition was also observed with the phenylalkylamine Ca2+ channel blocker verapamil at FIG. 1. Partial purification of the calcium channel blocker- 10 ,uM (n = 2). binding protein from membranes of carrot cells. (A) Lane 1, silver nitrate staining of NaDodSO4/PAGE (12.5% acrylamide gel) of the partially purified protein fraction; lane 2 (control), contaminating DISCUSSION proteins eluted from the Con A-Sepharose column by the equilibra- In higher plants, Ca2+ channel activation by physiological tion buffer (see text). The migration of the molecular mass standards stimuli such as plant hormones, light, and fungal elicitors has (kDa) is indicated to the left. (B) After the last purification step the been suggested to play a primary role in the initiation of protein extract was incubated with [3H]LU 49888 and UV-irradiated. Subsequently, the radiolabeled peptides were analyzed by Na- signal-transduction processes (1, 4, 6, 9). However, the DodSO4/PAGE (12.5% acrylamide gel) and fluorography. Lane 1, molecular components of higher plant Ca2+ channels remain the incubation was performed in the absence of the calcium channel unknown. In the present study, solubilization, partial puri- blocker bepridil; lane 2, the incubation was done in the presence of fication, reconstitution ofa membrane Ca2+ channel blocker- (-)-bepridil (50 ,uM). The migration of standard proteins is indicated binding protein from carrot cells, and patch clamp analysis to the right. The radioactivity present in the front (bottom) ofthe gel have allowed us to address the question of whether this may be due to nonspecific labeling of phospholipids (13). protein may contribute to Ca2+ channel function. Previous studies have shown that Ca2+ channel blockers of arose column with the equilibration buffer (Fig. 1A, lane 2) the phenylalkylamine family block Ca2+ influx into carrot showed that no current was detectable at pipette holding cells (11) and that a 75-kDa Ca2+ channel antagonist-binding potentials ranging from -150 to + 150 mV (n = 8). protein solubilized from membranes of these cells is photo- Patch-Clamp Studies on Reconstituted Liposomes with the labeled with the phenylalkylamine azido derivative [3H]LU LU 49888-Binding Protein. Patch-clamp analysis was per- 49888 (13). This property of the binding protein was used in formed on proteoliposomes reconstituted with either crude this study to follow the polypeptide during purification and (i.e., total solubilized proteins) or partially purified proteins subsequently to establish a purification procedure. (Fig. 1A, lane 1) from unlabeled membranes. In both cases, The combination ofmembrane protein reconstitution tech- we observed Ca2+-permeable single-channel currents (Fig. niques and patch-clamp studies has led to the identification 2), in asymmetrical solutions with a 10-fold gradient ofCaCl2 and characterization of a large number of different ion (Fig. 2A Inset). Single-channel currents measured by patch- channel proteins from mammalian systems (for review, see clamp recordings from liposomes reconstituted with crude (n ref. 26) and more specifically the main ion channel conduct- = 11) and partially purified (n = 15) proteins were similar ing subunits of Ca2+ channels (18, 19, 27-29). The reconsti- (compare Fig. 2 A and B). These channels showed reproduc- tution of membrane proteins into liposomes, as pursued in ible conductances of 49 + 7 pS (±SD, n = 5) and 38 ± 4 pS this study, is an approach which allows the association of (n = 4), respectively (Fig. 2 C and D). The direction of biological functions to defined membrane proteins at a res- single-channel currents reversed at a pipette potential of olution level of the molecular dynamics of single proteins approximately +25 mV, which is the range ofthe equilibrium (26). In the present study reconstitution of the partially potential (Nernst potential) for Ca2+ (+24 mV) under the purified protein fraction containing the 75-kDa polypeptide imposed experimental conditions (Fig. 2 C and D). These into liposomes and subsequent patch-clamp analysis showed data show that the measured single-channel currents were that the protein extract displayed Ca2+-permeable single- due to the passage of calcium ions through reconstituted channel currents (Figs. 2 and 3). Reconstitution of either Ca2+-permeable ion channels. contaminating proteins eluted from Con A-Sepharose col- However, these Ca2+-permeable channels were not stable. umns (Fig. 1A, lane 2) or thermally denaturated protein In both types of reconstituted liposomes (i.e., reconstituted extracts did not show Ca2+ channel activity. either with the solubilized extract or with the partially Reconstituted ion channels selective for Ca2+ over Cl- purified protein) Ca2+ channels disappeared after the single- (Fig. 2) were unstable after seal formation, as these single- channel currents had been recorded for durations of 2-10 channel currents vanished after recording for durations of min, thus hampering detailed cationic selectivity measure- several minutes. Due to this instability, determination of ments. In such experiments, an apparently new type of permeability ratios for Ca2+ with respect to other cations was nonselective Ca2+-permeable channel appeared after disap- hampered. The large single-channel conductance for Ca2+ of pearance of the above-described channel currents (n = 6) the partially purified extract (Fig. 2) and previous studies (Fig. 3). In several experiments (n = 7) these nonselective ion showing block of Ca2+ uptake into carrot cells by phenyla- channels were the first to appear in membrane patches. The lkylamines and bepridil (11) support the hypothesis that the reversal potential of these channel currents was -10 mV partially purified protein plays a role in Ca2+ influx. Single- Downloaded by guest on September 28, 2021 768 Plant Biology: Thuleau et al. Proc. Natl. Acad Sci. USA 90 (1993)

9 mm \>C.C12 pipette pipette A Mpotential B potential 90 mm (mV) (mV) C : : : +100 c _ ,,, +100 _^., Ad ~~+80 VII ~ ^$ I +80 +60 +80 ._. +40

~ 0 'wtsov -20 -20 "wv . y ---- -I -40 -40 ~~~-60 -60 C- 1- a- -80 Anu 0 *... {| he of@ 8jr20 m1 p ...._, ,_ _ . _,I.-- -7;00 7 10 pA O .... Ho 20 msec 0 7 10 pA C A.- rub'

4a.01 0)

%..O 40

-64 -32 0 +32 +64 -120 -60 0 60 120 pipette potential (mV) pipette potential (mV) FIG. 2. Recordings ofsingle Ca2+-permeable ion channel currents in inside-out membrane patches excised from liposomes reconstituted with either solubilized proteins (crude extract) (A) or the partially purified calcium channel blocker-binding protein from carrot membranes (B). (A and B) Single-channel currents recorded at different pipette potentials, indicated to the right. Solutions were 20 mM Hepes-Tris, pH 6.5/9 mM CaCl2/200 mM mannitol in the pipette and 20 mM Hepes-Tris, pH 6.5/90 mM CaC12 in the bath (see Inset). Closed and open states ofion channels are indicated by C and 0, respectively. (C) Current-voltage relationship of single Ca2+-permeable channel currents recorded after reconstitution of the partially purified protein. Ca2+ channels were recorded on the same patch by four successive overlapping voltage ramps between -80 mV and +80 mV (ramp duration 0.4 sec). C and 0 indicate closed and open states. (D) Single-channel current amplitudes plotted as a function ofthe applied pipette potential from liposomes reconstituted with the solubilized proteins (o) or with the partially purified protein (o). The arrows indicate the theoretical reversal potentials for Ca2+ (+24 mV) and for Cl- (-47 mV), after correction for ionic activities (22). channel currents, appearing subsequently in these membrane partial denaturation of the channel protein, thereby affecting patches, were nonselective, showing a permeability to both ion selectivity. In studies which led to the identification of calcium and chloride ions (Fig. 3). One possible explanation functional subunits of skeletal muscle Ca2+ channels, insta- for this modification of single ion channel currents may be a bility ofchannel function after insertion into lipid vesicles and

pipette 0 A potential B (mV) C c +80 rA-vnd414'TUJ~-~.A.m+60 4 pA| II , . . FiG. 3. Nonselective Ca2+-permeable ~ +40 -64 -32 0 +32 +64 single ion channel currents were recorded in pipette potential (m\/ inside-out membrane patches excised from liposomes reconstituted with the partially purified calcium channel blocker-binding ______, __ -20 8 C / protein. (A) Single-channel recordings are shown at different pipette potentials, indi- -40 4. cated to the right. (B) Current-voltage rela- !/ tionship of nonselective channel activity de- __--60 a- termined on the same patch by four succes- sive overlapping voltage ramps between -80 :3, .0 mV and +80 mV (ramp duration 0.4 sec). (C) H a _ -80 0 -4- Amplitude of single-channel currents as a function of applied potential. The arrows C .... i-100 % indicate the theoretical reversal potential for -120 -60 0 60 120 20 msec Ca2+ (+24 mV) and for Cl- (-47 mV). The pA pipette potential (mV) pipette and bath solutions were identical to 710 those in Fig. 2. Downloaded by guest on September 28, 2021 Plant Biology: Thuleau et A Proc. Natl. Acad. Sci. USA 90 (1993) 769

A B We thank Drs. N. Bush and R. Massingham (Laboratoire Cerm. ..0 ° Riom, France) for gifts of bepridil and Dr. M. Traut for [3H]LU - c c 49888. We also thank Drs. W. A. Catterall, J. M. Ward, M. Montal, and H. Hofte for comments on the manuscript. This work was ... supported by the Powell Foundation and National Science Founda- tion Grant DCB-9004977 (J.I.S.), by the European Economic Com- munity (Eclair Programme) (R.R.), by the French Ministry of For- eign Affairs (Bourse Lavoisier), and by a European Molecular Biology Organization Postdoctoral Fellowship (P.T.). *> 1 , ' f > 1. Hepler, P. K. & Wayne, R. 0. (1985) Annu. Rev. Plant Physiol. 36, 397-473.

0 2. McAinsh, M. R., Brownlee, C. & Hetherington, A. M. (1990) --c Nature (London) 343, 186-188. 20,r c. 2,-" -sec 3. Bush, D. S. & Jones, R. L. (1990) Plant Physiol. 93, 841-845. I k., n)u Ic ;-A 4. Leonard, R. T. & Hepler, P. K., eds. (1990) Calcium in Plant 0 Growth and Development (Am. Soc. Plant Physiol., Rockville, (n MD). E E c m10. 5. Gilroy, S., Fricker, M. D., Read, N. D. & Trewavas, A. J. U) 'n 0 (1991) Plant Cell 3, 333-344. -5 o 'r 6. Schroeder, J. I. & Hagiwara, S. (1990) Proc. Natl. Acad. Sci. 0 b-X 5 USA 87, 9305-9309. w 7. Johannes, E., Brosnan, J. M. & Sanders, D. (1992) Plant J. 2, E avE 97-102. _C :3 0 ---.- ..... 8. Tester, M. (1990) New Phytol. 114, 305-340. 0 5 10 0 5 10 9. Schroeder, J. I. & Thuleau, P. (1991) Plant Cell 3, 555-559. currents (pA) currents (pA) 10. Andrejauskas, E., Hertel, R. & Marmi, D. (1985) J. Biol. Chem. 260, 5411-5414. FIG. 4. Inhibition of nonselective ion channel currents by the 11. Graziana, A., Fosset, M., Ranjeva, R., Hetherington, A. M. & calcium channel blocker (-)-bepridil in liposomes reconstituted with Lazdunski, M. (1988) Biochemistry 27, 764-768. the partially purified LU 49888-binding protein. Continuous single- 12. Harvey, H. J., Venis, M. A. & Trewavas, A. J. (1989) Bio- channel current recordings were made before (A) and after (B) the chem. J. 257, 95-100. addition of 10 AM bepridil. Amplitude histograms on the bottom 13. Thuleau, P., Graziana, A., Canut, H. & Ranjeva, R. (1990) show the distributions of samples in the open state (0) and the closed Proc. Natl. Acad. Sci. USA 87, 10000-10004. state (C) before (A) and after (B) addition of(-)-bepridil. The pipette 14. Thuleau, P., Graziana, A., Rossignol, M., Kauss, H., Auriol, potential was +60 mV. The pipette and bath solutions were identical P. & Ranjeva, R. (1988) Proc. Natl. Acad. Sci. USA 85, to those in Figs. 2 and 3. 5932-5935. 15. Smith, P. K., Krohn, R. L., Hermanson, G. T., Mallia, P. K., bilayers has also been reported (18, 29). In the present study, Gartner, F. H., Provenzano, M. D., Fujimoto, E. K., Goehe, N. M., Olson, B. J. & Klenk, D. C. (1985) Anal. Biochem. 150, in several experiments the nonselective channels were the 76-85. first to appear in membrane patches. We currently cannot 16. Criado, M. & Keller, B. U. (1987) FEBS Lett. 224, 172-176. distinguish whether these two recorded ion channels repre- 17. Riquelme, G., Lopez, E., Garcia-Segura, L. M., Ferragut, sent two different physiological forms of the same protein or J. A. & Gonzalez-Ros, J. M. (1990) Biochemistry 29, 11215- two different proteins. 11222. The nonselective Ca2+-permeable channels were inhibited 18. Flockerzi, V., Oeken, H.-J., Hofmann, F., Pelzer, D., Cavalie, A. & Trautwein, W. (1986) Nature (London) 323, 66-68. by addition of the calcium channel blockers bepridil and 19. Home, W. A., Abdel-Ghany, M., Racker, E., Weiland, G. A., verapamil, which are known to inhibit 45Ca2+ influx into Oswald, R. E. & Cerione, R. A. (1988) Proc. Natl. Acad. Sci. carrot protoplasts (11). Direct recordings of inhibition of USA 85, 3718-3722. Ca2+ channels by blockers have not been obtained in higher 20. Hamill, 0. P., Marty, A., Neher, E., Sakmann, B. & Sigworth, plants to date. The only data allowing comparison to the F. (1981) Pflugers Arch. 391, 85-100. mechanism of block come from animal systems. The block 21. Schroeder, J. I., Raschke, K. & Neher, E. (1987) Proc. Natl. mechanisms of carrot and Acad. Sci. USA 84, 4108-4112. (Fig. 4) mammalian (19, 30) Ca2+ 22. Robinson, R. A. & Stokes, R. H. (1955) Electrolyte Solutions channels by calcium channel blockers are similar. Thus, it (N.Y. Acad. Sci., New York), pp. 480-499. was observed that the antagonists did not lead to changes in 23. Fenwick, E. M., Marty, A. & Neher, E. (1982) J. Physiol. the single channel current amplitudes through open Ca2+ (London) 331, 577-597. channels but rather to blocker-induced reductions in the 24. Curtis, B. M. & Caterall, W. A. (1984) Biochemistry 23, 2113- open-state probability of channels (Fig. 4; refs. 19 and 30). 2118. 25. Hodgkin, A. L. & Katz, B. (1949) J. Physiol. (London) 108, In conclusion, the present study demonstrates that the 37-77. 75-kDa binding polypeptide contributes to Ca2+ channel 26. Miller, C. (1986) Ion Channel Reconstitution (Plenum, New function after reconstitution and plays a role in the sensitivity York). to Ca2+ channel blockers. In addition, the presented findings 27. Curtis, B. M. & Caterall, W. A. (1986) Biochemistry 25, 3077- provide evidence for a membrane protein that may constitute 3083. one of the molecular components of Ca2+ channels in higher 28. Nelson, M. T., French, R. J. & Krueger, B. K. (1984) Nature plants. Further studies on the function and sequence (London) 308, 77-80. primary 29. Rosenberg, R. L., Hess, P., Reeves, J. P., Smilowitz, H. & ofthe Ca2+ channel blocker-binding protein and direct patch- Tsien, R. W. (1986) Science 231, 1564-1566. clamp recordings of carrot cells may allow new insights to 30. Hess, P., Lansman, J. B. & Tsien, R. W. (1984) Nature (Lon- plant Ca2+ channel structure and regulation. don) 311, 538-544. Downloaded by guest on September 28, 2021