- Email: [email protected]
Molecular aspects of human brain sodium channel
Newoscience Vol. 38, No. 3, pp. 809-817, Printed in Great Britain
MOLECULAR
0306-4522/90 53.00 + 0.00 Pergamon Press plc 0 1990 IBRO
I990
ASPECTS OF HUMAN CHANNEL
BRAIN SODIUM
C. DE RYCKER and E. SCHOFFENIELS Laboratoire de Biochimie Gin&ale et Comparee, Universite de Li&ge, 17 Place Delcour, 4020 Ligge,
Belgium Abstract-The sodium channel content of human brain was measured by tritiated tetrodotoxin specific binding. After solubilization, the sodium channel was submitted to chromatography on diethylaminoethyl(cellulose) Sephadex, hydroxylapatite and wheat germ agglutinin sepharose. An increase of tritiated tetrodotoxin binding specific activity was subsequently observed. Eluted sodium channels from wheat germ agglutinin sepharose were overlaid on a sucrose gradient. Electrophoretical analysis of the material obtained after the sedimentation step revealed two co-purified peptides, t( (kf, = 275,000 mol. wt) and fi (44, = 30,~36,~ mol. wt). Alpha showed an exceptionally high free electrophoretic mobility, which is a common feature for all sodium channels previously described. However, the high denaturation rate of the solubilized tetrodotoxin receptor site 1 did not allow tetrodotoxin receptor quantification by the tritiated toxin binding in sucrose fractions. Sodium channel effective reconstitution in liposomes was demonstrated: (1) **Na+ influx in proteoliposomes was sensitive to sodium channel-specific neurotoxins; (2) reconstituted proteins showed a cation
selectivity similar to that previously described for animal sodium channels. The sodium channel preparation obtained after four chromatographic steps shows two peptides on the electrophoretic analysis. Reconstituted sodium channels displayed some physiological properties found in intact conducting membranes.
composed of at least a =260,000 mol. wt glycoprotein (a). The SC isolated from rat brain includes additional subunits of M, 39,OOOmol. wt @I) and 37,000 mol. wt (fi2).16 Beta 1 is non-covalently associated to a while /32 is linked to ct by disulfide bridges.27 However, as showed by radiation inactivation techniques, the rat brain SC is a complex composed of only the 260,000 mol. wt (CX)and 39,000 mol. wt (fi 1) polypeptides.3 The c@1 complex seems to be sufficient to modulate transmembrane ion flux after reconstitution into artificial membrane and responds to pharmacological agents similarly to SC in sim17-34*35~39 However, other authorssy’O have described rat brain SC as a single 270,000 mol. wt peptide, CL Those divergent observations led us to study the SC from human brain. The purification procedure described for rat brain was adapted to human brain. Electrophoretic analy sis of the obtained protein(s) was performed after each chromatography. Nevertheless, as shown previously,’ the tetrodotoxin receptor site of human SC is sensitive to denaturation when solubiiized. It was therefore of great interest to investigate whether the protein obtained after all chromatographic steps did keep its other native properties. This investigation was performed by using SC reconstituted into liposomes.
The voltage-sensitive sodium channel or conducting site for sodium (SC) is responsible for the rapid increase in membrane sodium permeability which corresponds to the action potential depolarizing phase in excitable tissue. Biochemical studies involve the use of specific neurotoxins which act at five receptor sites of the protein and allow structural and functional studies of the SC. The most widely used are saxitoxin and tetrodotoxin. They bind to receptor site 1. As a result, the site is in a low conductance state. Veratridine, batrachotoxin, grayanotoxin and aconitine bind to receptor site 2 and cause a persistent activation of the protein, i.e. a high conductance state. Other classes of neurotoxins bind to other receptor sites and have various physiological effects.‘* Although electrical properties of SC have been largely investigated, their molecular structure has only recently been described. Current evidence from photoaffinity labeling, radiation inactivation, purification and reconstitution suggest that the SC from eel electroplax,1~28~30rat brain,5~‘0~15~‘6 rat and rabbit sarcolemma,4,20 chick heart26 and lobster nerves38 is Ab~reuiations: BSA, bovine serum albumin; DEAE, diethylaminoethyl (cellulose); EDTA, ethylenediaminetet~aacetate; FTS, freeze thaw sonication nrocedure: HEPES. N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acidf N-Ag, N-acetyl glucosamine; NP-40, Nonidet P-40, PC, phosphatidylcholine; PAGE, polyacrylamide gel electrophoresis; PE, phosphatidylethanolamine; PMSF, phenyl methylsuIfonylfluo~de; PS, phosphatidylse~ne; SC, sodium channel; SDS, sodium dodecyl sulfate; t3H]lTX, tritiated tetrodotoxin; WGA, wheat germ agglutinin; WW, wet weight.
EXPERIMENTAL PROCEDURES
Materials Human brains were obtained as described previously.9 Chemicals were purchased 809
from the following sources.
810
C. DE RYCKERand E. SCHOFFENIELS
Phenylmethylsulfonylfluoride (PMSF), Nonidet P-40 (NP40), phosphatidylcholine (PC) type III from egg yolk, wheat germ agglutinin @VGA) from Triticum vulguris and cation exchange resin Dowex 5OW-X8 were from Sigma (St Louis, MO, U.S.A.). l,lO-Phenanthroline monohydrate and iodoa&amide were from Janssen Chemica (Beerse, Belgium). Diethylaminoethyl(ceIlulose) (DEAE) Sephadex A-25, Sepharose 4CNBr, Sephadex G-200 and calibration protein for the sodium dodecyl sulfate-polyacrylamide gel electronhoresis (SDS-PAGE) were from Pharmacia (Unnsala. Sweden). &o-Gel HA,‘Bio-Beads SM-2 and silver s&in kit were from Bio-Rad Laboratories (Richmond, CA, U.S.A.). N-Acetyl glucosamine (N-A%) and bovine serum albumin were from Fluka (A. G. Buchs, Switzerland). Veratridine (sulfate form) was a generous gift of Dr Angenot (University of Liege, Belgium). Batrachotoxin was kindly provided by Dr J. W. Daly, of the Department of the National Institute of Arthritis, Metabolism and Digestive Deseases, National Institute of Health, U.S.A. Tetrodotoxin (free of citrate) was obtained from Sankyo Co. Ltd, Tokyo, Japan. Tritium labeling was carried out by our modification14 of the method of Bontemps et ~1.~Saxitoxin was generously provided by E. J. Schantz, University of Wisconsin. Phosphatidylcholine (PC), phosphatidylserine (PS) and phosphatidylethanolamine (PE) added during final reconstitution were from Avanti Polar (Birmingham, AL, U.S.A.). **Na+ (1 mCi/ml, carrier free) was from New England Nuclear (Boston, MA, U.S.A.). 13’Cs+ and 83Rb+ were provided by Dr Guillaume (University of Liege, Belgium). Dialysis tubings were from Thomas Corp. (Philadelphia, PA, U.S.A.). Dowex minicolumns used for flux assays were made out of 1 ml polypropylene syringes, Syringe ends were cut and replaced by a piece of nylon netting (Scrynel, 200 mesh). Sonication was performed with a titane lead connected to an ultrasound generator (Braun Melsungen A.G., Germany). All other products were from the best grade available and all solutions were made with Milli-Q water (Millipore Corp., Bedford, MA, U.S.A.). Methods Preparation of wheat germ agglutinin sepharose. WGA sepharose was obtained by coupling WGA with sepharose 4B-CNBr: sepharose was washed with a large volume of 1 M HCl. Five milliliters of wet gel were added to 50 mg of WGA in 2ml of buffer A (1OOmM Na,HCO,, pH 8.3, 500mM NaCI, 0.1 mM N-Ag). The mixture was stirred overnight at 4°C and then washed with 20 ml of buffer A. Five milliliters of 100 mM ethanolamine, pH 8.0, was added to the gel in order to block the excess of reactive groups. After a 2 h incubation at 4°C WGA sepharose was put in a small column (0.5 x 10 cm) and washed twice with sodium acetate (100 mM, pH 4.0, 500 mM NaCI) and Tris-HCl (100 mM, pH 8.0, 500 mM NaCl). Brain membrane preparation. All operations were made at 4’C and all buffers contained protease inhibitors at the following concentrations: iodoacetamide 1mM, orthouhenantroline 1 mM. PMSF 0.1 mM and pepstatin 1 bM. ’ Einhtv grams of frozen brain (white and gray matter) were allowed to thaw in 10 volumes of homogenization buffer (IO mM Tris-HCl. PH 7.4. 0.32 M sucrose and protease inhibitors), then diced into small fragments and homogenized in a Teflon Potter-Elvehjem homogenizer. The homogenate was centrifuged at 120g for 10 min; the supernatant (Sl) was saved; the pellet (Pl) was resuspended in homogenization buffer and centrifuged again at 120g for the same time. The resulting supernatant (S2) was added to St and the pellet (P2) discarded. Sl + S2 were centrifuged for 40 min at 27,000g. The supernatant (S3) was discarded and the pellet (P3) was resuspended in 1I of buffer 5 mM TrissHCl, pH 8.2, 1 mM EDTA. containing the four protease inhibitors listed above and gently stirred on ice for 15 min. After the osmotic shock, the suspension WdS I
.-
centrifuged for 80 min at 27,000g. Pellets (P4) were combined and resuspended in 2 volumes of buffer (20mM HEPES-Tris, 120mM KCl, pH 7.4, protease inhibitors) per volume of starting material. P4 (membrane fraction) was quickly frozen in a dry ice-cold acetone mixture and maintained at -70°C until use. [‘HlTetrodotoxin binding measurements. Tissues were homogenized in 10 mM Tris-HCl buffer, pH 7.4, containing protease inhibitors, prior to [3H]TTX binding measurements [3H]TTX binding assays on membranes and solubilizedz4sodium channel measurements were done as described previously.9 Protein assay. Protein content determinations were made according to the Peterson procedure.” Sodium channel solubilization. P4 (160 ml from 80 g of tissue) was put in a beaker packed in ice on a magnetic stir plate. The same volume of buffer containing the detergent (20 mM HEPESTris, pH 6.8, 2% NP-40, 0.2% PC, protease inhibitors) was added in small increments with constant mixing. Fifteen minutes later, unsolubilized material was sedimented at 120,000g for 30 min and the supernatant (membrane extract) was kept for further analysis. Chromatographic steps performed on membrane extract.
All manipulations were conducted within 24 h following the SC solubilization. Membrane extract was adjusted to 10 mM CaCI, prior to chromatography. The pH -was then lowered to -6.5 by addition of 0.5 M histidine-HCl. PH 5. The membrane extract conductivity was then raisei to 15 mS/cm by addition of 3 M KCI. Wet DEAE-Sephadex A-25 (200 ml) was rinsed with six volumes of equilibration buffer (20 mM histidineeHC1. pH 6.5, 100 mM KCI, 0.1% NP-40,0.025% PC). Membrane extract (180m1, containing 50@600mg of protein) was gently stirred with the wet gel for 30 min. The mixture was transferred into a Btichner vessel and the non-adsorbed proteins were eliminated by four equilibration buffer washes. The Biichner vessel outlet was connected to a fraction collector and SC was eluted with 250ml of equilibration buffer supplemented with 500mM KCI. When eluted fraction conductivities ranged from 15 to 50mS/cm, they were pooled for further chromatography. Usually, six fractions of lOm1 each were pooled (DEAE pool). In order to prepare the DEAE pool for hydroxylapatite chromatography, its pH was raised to 8.0 by addition of 2 M Tris-HCl, pH 8.9. EDTA was added to a final concentration of 10 mM. This step lowers the pH to 7.4. Finally, 800 mM KH,PO,, pH 7.4, was added to obtain a final concentration of 80 mM. The DEAE pool was swirled for 15 min with 100 ml of Bio-Gel HT, previously equilibrated in 80 mM KH2P0,, pH 7.4, 0.1% NP-40 and 0.025% PC. The resin suspension was poured into a 2.4 x 20cm column and washed with 300 ml of equilibration buffer. SC was eluted by 300 ml of the same buffer containing 400 mM KHrPO,. Fractions binding [3H]TTX were pooled (hydroxylapatite pool). WGA sepharose was equilibrated with 50ml of the elution buffer used for hydroxylapatite chromatography. The hydroxylapatite pool was run through the column at a 1 ml/min rate. Resin was washed with 50 ml of HEPESTris (25 mM, pH 7.4, 0.5 mM MgSO,, 205 mM Na,SO,, 0.1% NP-40, 0.025% PC) buffer. SC were eluted by N-Ag (150 mM) in a HEPESTris buffer (25 mM, pH 7.4, 67.5mM Na,SO,, 0.5mM MgSO,, 0.1% NP-40, 0.025% PC). When SC are prepared for reconstitution experiments. DEAE elution buffer is used for WGA chromatography since the hydroxylapatite step is omitted. Five-milliliter fractions were collected during washes and 2.5 ml fractions were collected during elution. Fraction 22 was concentrated by dialysis and was then layered on a linear sucrose gradient (5-20%, prepared with sepharose elution buffer without
811
Molecular aspects of human brain sodium channel N-Ag) and sedimented for 16 h at 145,080g at 4°C in a Beckman ultracentrifuge. The gradient was then fractionated into 15 fractions of 750~1 each. Sodium dodecylsulfate gel electrophoresis. The general conditions of SDS-PAGE were those described by 10mM Laemmli.2’ Samples were dialysed against Na,HCO,, pH 7.4, and lyophilized. Dry material was resuspended in loading buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 5% /I-mercaptoethanol, 12.5% glycerol, Bromophenol Blue) and heated for 15 min at 100°C. Samples were analysed by gradient SDS-PAGE (412%); four constant acrylamide concentrations (4.4, 4.8, 5.2 and 5.6%) were used for a Ferguson plot. Stacking gels were 3% acrylamide. Gels were silver stained and densitometric scans of migration patterns were performed using an ultroscan detector (LKB, model 220, Sweden). Reconstitution. Reconstitution was performed as previously described2 for electric eel SC: WGA sepharose eluate was gently swirled for 3 h at 4°C with 0.3 g/ml of wet Bio-Beads (equilibrated as described by Holloway’s) to adsorb Nonidet P-40. Proteoliposomes were dialysed overnight against a large volume of buffer A (25mM HEPESTris, 67.5 mM Na,SO,, pH 7.4) in order to lower the KC1 concentration (500mM) contained in WGA sepharose elution buffer. Other liposomes were synthesized as follow: PE, PS, and PC were mixed in a molar ratio 5:4: 1 and dried under N, Buffer A was added to the dry residue to obtain a phospholipid concentration of 40 mg/ml. The mixture was vortexed for 5 min and then sonicated on ice for IOmin under nitrogen. Proteoliposomes were then fused with those liposomes using the freeze thaw sonication (FTS) procedure described by Kasahara and Hinkle:@ one volume of liposomes was added to 3 volumes of proteoliposomes and 1 ml aliquots were immersed for 1 min in dry ice-cold acetone. The suspension was then allowed to thaw at room temperature. The new enlarged proteoliposomes were submitted to a 10 s sonication and kept on ice until use. Electron microscopy. Enlarged proteoliposome aliquots were diluted two-fold in buffer A. Copper grids (2OOmesh, pretreated with Formvar) were gently reversed on liposome droplets for 2min. Grids were stained with phosphotungstic -acid (2% v/v, pH 7) for 1 min and directly observed with a Jehol lOOCX-2 microscone. Distributions of vesicle size were established from the’ obtained pictures. 22Na+~JUXassays. Polypropylene minicolumns (used for **Na+ flux measurements) were filled with Dowex 5OW-X8 (0.5 ml), washed with 2 x 1 ml of ice-cold BSA solution (3.3 mg/ml) and 220 mM sucrose and maintained at 4°C until use. Just before use, aliquots of proteoliposomes were overlaid on Dowex minicolumns pre-equilibrated with 10 ml of 20 mM Tris sulfate, pH 7.4, in order to create a membrane potential which enhance Na+ entrance into the vesicles.13 Proteoliposomes were then quickly diluted in 3 volumes of buffer B (170 mM Tris sulfate, pH 7.4, 10 mM NaCl, lOpCi/ml 22Na+) at 37°C.
Two 200-~1 samples of the radioactive suspension were overlaid at the indicated times on Dowex BSA sucrose pre-equilibrated columns. Elution began less than 5 s after loading and the eluate was directly collected in scintillation vials (Lumac, milli 6, 6372 AE Landgraaf, Holland). Effect ofneurotoxins on 22Na+ influx. Final concentrations of saxitoxin, tetrodotoxin and batrachotoxin during fluxes were 1 PM; veratridine concentration was 1OOpM. Batrachotoxin stock solution (500 PM) was supplied in ethanol solution and control experiments on proteoliposomes treated with ethanol alone were performed. Ionic selectivity. 22Na+, 83Rb+ and t3’Cs+ influx were measured separately on proteoliposomes prepared from the same reconstitution in the presence of IOOpM veratridine and under control conditions. For each point, sodium channel influx was calculated by subtraction of the nonspecific influx (control conditions) from the influx measured in the presence of veratridine. Three individual kinetics were performed in order to avoid any interference in radioactive counts. RESULTS
Sodium channel content in homogenate membranes of human brain
Membrane extract chromatographic steps
A summary of the chromatographic steps performed on membrane extract is given in Table 1. The elution profile of DEAE sephadex is shown in Fig. 1A. SC was eluted in fractions 21-24 (DEAE pool), which exhibited an increase of conductivity. After several adjustments (see Experimental Procedures), the DEAE pool-was chromatographied on an hydroxylapatite column. The hydroxylapatite column elution profile is illustrated in Fig. 1B. Fractions 10 and 11 (hydroxylapatite pool) contained 38% of the SC and 36% of total proteins loaded on the column. The specific activity increase was negligible but hydroxylapatite chromatography was essential to eliminate a 220,000 mol. wt peptide, as shown by electrophoretical analysis. The hydroxylapatite pool was loaded on a WGA sepharose column. Sixty per cent of the loaded
TTX receptor
Membrane extract DEAE HA WGA
pm01
%
1488 + 128 100 346 f 43 23.2 139+20 9.3 67k 14 4.5
Values are f SE., n = 4.
in
Human gray matter [‘H]T”TX binding (82 pmol/g wet weight, WW) was twice that of white matter (43 pmol/g WW). Whole brain homogenate binds 50 pmol of [‘H]TTX/g WW. These data suggest that gray matter represents 27% of the human brain volume. Membranes (P4) obtained from 80 g of brain tissue bound 3000 pmol of [3H]TTX.
Table 1. Summary of the chromatographic steps performed membrane extract
Step
and
on the human brain
Protein mg 633 &70 30 k 6 11&3 1.1 * 0.2
% 100 4.73 1.73 0.17
Suecific activitv pmol/mg prot: 2.35 11.53 12.63 60.90
812
C. DE RYCKER and E. SCHOFFENIELS
FRACTION
1
s
10
15
FRACTION
Fig. 1. Profiles of the four chromatographies carried out as described in Experimental Procedures. (A) Ion exchange chromatography on DEAE Sephadex A-25. (B) Hydroxylapatite chromatography on Bio-Gel HT. (C) Affinity chromatography on WGA sepharose. (D) Sucrose gradient sedimentation. (0) TTX receptor measured by rapid gel filtration. (0) Protein content determined by the Peterson method. (A) Conductivity. (x) Sucrose percentage.
x103
-a
220-
9467431, 3630-
Fig. 2. SDS-PAGE of membrane extract (lane 1) and eluates of the four chromatographies performed on solubilized human brain SC: DEAE pool (lane 2), HA pool (lane 3), WGA pool (lane 4), and fraction 13 of sucrose gradient (lane 5). Analysis was performed on a 412% linear acrylamide gradient.
813
Molecular aspects of human brain sodium channel
06
1
2 %
3
L
5
6
ACRYLAMIDE
Fig. 3. Ferguson plot of the 275,000 mol. wt peptide (A) and of four acrylamide calibration proteins. SDS slab gels of indicated acrylamide concentration were prepared as described in Experimental Procedures. SC (A), lactate dehydrogenase (m), ovalbumin (O), albumin (A) and ferritin (0) were run on each gel. Extrapolations of drawn lines gave peptide-free mobility values.
went through the column and 10% were eluted after addition of N-Ag (Fig. 1C). Since this proteins
sugar eluted 50% of the SC overlaid on the column, the increase of [3H]TTX binding specific activity was substantial. Fraction 22 was then sedimented through a sucrose gradient. The protein distribution in the sucrose gradient is presented in Fig. 1D. No [3H]TTX binding activity
was detected. Proteins were distributed in two major peaks. The first one (fractions 3 and 4) corresponded to low molecular weight proteins and the second one (fractions 13-15) to catalase (240,000 mol. wt) in calibration gradient. The second peak contained 25% of the loaded proteins. Electrophoretic
analysis
The electrophoretic pattern of proteins obtained after each chromatography is shown in Fig. 2. Membrane extract proteins are shown in lane 1. The doublet underlying the diffuse band on the gel top (lanes 1 and 2) was retained on hydroxylapatite (lane 3). Low molecular weight peptides (under 94,000 mol. wt) were lost during affinity chromatography (lane 4). Phosphorylation assays on those preparations showed that the 94,000 mol. wt peptide was the Na+, K+, ATPase c( subunit (not shown). Proteins contained in the 13th fraction of the sucrose gradient (lane 5) appeared as a large diffuse band (275,000 mol. wt) and a group of smaller peptides was seen from 30,000 to 36,000 mol. wt. One of them (33,000 mol. wt) was more heavily stained. 275,000 mol. wt peptide Ferguson plot A Ferguson plot of four calibration proteins and the 275,000 mol. wt peptide is presented in Fig. 3. Lines corresponding to calibration proteins have the
Table 2. Peak surfaces (arbitrary units) of 3O,OO(r36,000 and 275,000 mol. wt peptides from polyacrylamide gel electrophoresis densitometric scans of diethylaminoethyl (cellulose) and wheat germ agglutinin pool protein patterns 30,00~36,000 mol. wt
275,000 mol. wt
253 + 42 405 & 38
1585 f 124 2325 k 196
DEAE pool WGA pool Values are *SE.,
n =4.
same y intercept
(0.520), indicating that they have similar electrophoretic mobilities. The y intercept of the 275,000 mol. wt peptide line (0.730) is 1.6fold higher; this underlines the particularly high free electrophoretic mobility of the 275,000 mol. wt peptide. Correlation between the 275,000 36,000 mol. wt peptides
and the 30,00&
Densitometric scans have been performed on different protein migration patterns. The 275,000 mol. wt and the 30,00&36,000 mol. wt peak peptide surfaces were proportional (Table 2). Proteoliposome
size and morphology
Eighty-five per cent of the intravesicular volume was represented by 64-180 nm proteoliposomes. Most of the vesicles were unilamellar and this aspect was strictly dependent on the short sonication length following the freeze thaw process (not shown). Effect of neurotoxins liposomes
on 22Na+ uptake into proteo-
Uptakes of “Na+ into proteoliposomes after 5 s and 5 min of flux are represented in Fig. 4. Intravesicular 22Na+ was observed in control conditions meaning that proteoliposomes were permeable to
814
C. DE RYCKERand E. SCHOFFENIELS
After 30 min fluxes, the ratios of intravesicular radioactivity/total loaded radioactivity were similar for the three cations (*‘Na+, 1.7%. *3Rb+, 2.1%; ‘Ws+, 1.8%). These percentages iepresented the accessible intravesicular volume in this population of vesicles. SC selectivity may be expressed taking the sodium selectivity as 1. In this convention, reconstituted human SC ionic selectivity was Na+/Rb+/ Ca+ = 1:0.48:0.33 in the presence of 100pM veratridine.
T
B
DISCUSSION CBVTS
C
0
V
T
S
Fig. 4. 22Na+ intravesicular accumulation after 5 s (A) and 5 min (B) of incubation. Influx was expressed by the ratio intravesicular cpm/cpm loaded on the Dowex column. Incubation occurred without (C) or with different toxins: batrachotoxin (B), veratridine (V), saxitoxin (S) or tetrodotoxin (T). Error bars showed the results obtained in two experiments performed on the same batch of vesicles.
sodium. Addition of guanidinium toxins (saxitoxin or tetrodotoxin) induced a slight decrease of the radiotracer accumulation. Veratridine induced, 5 min after its addition, a 50% increase of 22Na+ uptake. Batrachotoxin also had an enhancing effect on uptake, even 5 s after its addition. After 5 min, intravesicular “Na+ reached nearly double that of intravesicular 22Na+ measured in control conditions. As batrachotoxin was supplied in ethanol solution, control experiments were made with the solvent alone (0.2% final concentration) and no effect on proteoliposome permeability was observed. Ionic selectivity
Veratridine-stimulated influx was measured for **Na+, r3Rb+ and 13’Cs+ in proteoliposomes at 37°C (Fig. 5). The half-time for ‘*Na+ flux equilibration was 1.3 min while half-times for 83Rb+ and 13’Csf were 2.3 and 4min, respectively.
Sodium channel content in human brain
Results of SC contents from three types of tissues are summarized in Table 3. The human brain is less dense in SC than the rat brain. Two hypotheses may explain this observation. (1) In human, a proteolytic digestion occurred between donor death and tissue freezing. (2) Brain SC density is much lower in human than in rat in vivo. Observations made by Palacios et aL3’ reinforced the second hypothesis as these authors underlined the lower density of about 20 types of receptors in human brain compared to rat brain. Moreover, the SC content of human gray matter is twice that of white matter. Specific tritiated ligands used in autoradiography have also shown the same type of distribution in rat brain.29,40 Chromatographic
steps evaluation
Previous experiments showed that human SC was highly sensitive to denaturation when solubilized.’ This led to a lack of ability in measuring the real amount of solubilized receptor from [3H]TTX binding. Nature of the 275,000 mol. wt 36,000 mol. wt peptides
and
the
30,000-
Mammalian SC is composed of a large subunit tl (260,00~300,000 mol. wt) and one smaller subunit fi
.**No
.83Rb . ‘37cs 15
20
25
30
Fig. 5. Selective cation passage through the veratridine-activated reconstituted SC. *iNa+, *‘Rb+ and “‘Cs+ uptakes were measured in the presence of 100 PM veratridine and in control conditions at Indicated times. Control isotopic uptake measured in the absence of veratridine was subtracted from the total isotope uptake to obtain the stimulated isotopic uptake.
Molecular aspects of human brain sodium channel Table 3. [)H]TTX binding measured on homogenates three excitable tissues in three laboratories [‘H]TTX bound (pmol/g ww)
Tissue Electric organ of E. electricus Brain Human (whole) gray matter white matter Rat (whole) Rat (whole) cerebellar cortex Port, gray matter Muscle Rat (skeletal)
of
References
30,
23
50 82 43 150 77 200 50;
9 9 9 25 9 25 23
Ionic selectivity
*Measurements performed on lyophylized preparations. (~O,OOO-40,000 mol. wt),
sometimes
shown
as
a
doublet B 1 and /12.4,15,‘6,20 We propose that tl and /I peptides shown in lane 5 are human SC components. The three following observations led to this conclusion. (i) [3H]TTX binding specific activity increases through chromatographies and coincides with the presence of 275,000 and 30,000-36,000 mol. wt peptides on SDS-PAGE. (ii) The 275,000 mol. wt peptide presents the same diffuse aspect and the same high free electrophoretic mobility (see Ferguson plot) as the a SC subunit from other species. This high free mobility is exceptional since most of the glycoproteins reveal lower free mobilities than non-glycosylated proteins.22 The originality is specific to the SC u subunit and is probably due to an excess of protein-bound SDS. Since no other human glycoprotein has ever shown this exceptional electrophoretic mobility, we propose that the 275,000 mol. wt peptide is the human SC tl subunit. (iii) Densitometric scans of the protein pattern have shown that the 30,000-36,000 mol. wt peptide was co-purified with the a subunit. This small peptide is thus the human SC /3 subunit. In some PAGE it appeared as a doublet. This uneven appearance has already been described in rat brainI Human
sodium
channel
chotoxin was a more potent SC activator than veratridine. Batrachotoxin activates more than 95% of SC whereas veratridine is a partial agonist (only 50%
of the SC population is activated).’ Moreover, the time length of openings induced by batrachotoxin is higher than that induced by veratridine.‘* Requirements for restoration of the native properties of receptor sites 1 and 2 are different in reconstituted human SC.
25
8
815
reconstitution
Neither saxitoxin nor tetrodotoxin had a significant effect on intravesicular **Na+ accumulation (Fig. 4). These guanidinium toxins, binding to receptor site 1, are no longer effective as blockers on reconstituted human SC. Thus, the native properties of site 1 receptors were not recovered by reconstitution in our lipid bilayers. Both veratridine and batrachotoxin binding at receptor site 2 enhanced the intravesicular **Na+ uptake. The effect of batrachotoxin was more rapid and more important. This is in good agreement with several previous observations, showing that batra-
Our results are in good agreement with those previously published for other animal SC since the same type of ionic selectivity was observed. In fact, selectivity seems to be correlated to the cation size among other parameters, as suggested by Pappone,32 who found that rat SC had a permeability of 1.14 for lithium (A, = 6.9). Similar permeability ratios for rat sarcolemma1 (Na+/Rb+/Cs+ = 1:0.25:0.12) and brain (Na+/Rb+/Cs+ = 1:0.5:0.38) SC have been obtained previously.36,37 However, these values do not reflect the native SC selectivity as measured in the presence of veratridine. Higher selectivities were indeed observed in native forms; addition of batrachotoxin and mainly of veratridine decreased the degree of specificity somewhat.8.“.‘2.37
CONCLUSION
The substantial denaturation of the SC receptor site 1 was the main difficulty encountered in this study. It led us to consider well known SC properties other than tetrodotoxin binding to define the protein obtained after chromatographic procedures. Thus, we have shown that the reconstituted protein was still sensitive to site 2 targetting toxins and showed specific SC ionic specificity. Human brain was less dense in SC than rat brain. However, the same density distribution between white and gray matter was observed in both species. We propose a dimeric structure for the human SC. The 275,000 mol. wt diffuse band seen on PAGE had the same physicochemical properties as tl SC subunits previously described; the 30,000-36,000 mol. wt peptide, sometimes seen as a doublet, was co-purified with a. Thus, human SC appeared as a heterodimeric complex composed of two subunits, tl (275,000 mol. wt) and B (30,00036,000 mol. wt). However, further studies, more particularly based on immunological techniques, would be advantageous to confirm that the 30,000-36,000 mol. wt peptide belongs to the SC in situ.
REFERENCES
1. Agnew W. 8, Levinson S. R., Brabson J. S. and Raftery M. A. (1978) Purification of the tetrodotoxin-binding component associated with the voltage-sensitive sodium channel from Electrophorus elecrricus electroplax membranes. Proc. natn. Acad. Sci. U.S.A. 75, 2606-2610.
816
9. 10. II. 12. 13.
14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.
26. 27. 28.
29. 30.
31. 32. 33. 34.
3.5
C. DE RYCKER and E. SCHOFFENIELS Agnew W. S., Rosenberg R. L. and Tomiko S. A. (1986) Ion Channel Reconsfitution (ed. Miller C.), Ch. 12, pp. 307 330. Plenum Press. New York. Angelides K. J., Nutter T. J., Elmer L. W. and Kempner E. S. (1985) Functional unit size of the neurotoxin receptor on the voltage-dependent sodium channel. J. biol. Chem. 260, 3431-3439. Barchi R. L. (1983) Protein components of the purified sodium channel from rat skeletal muscle sarcolemma. J. Neurochem. 40, 1377-1385. Barhanin J., Pauron D., Lombet A., Norman R. I., Vijveberg H. P. M., Giglio J. R. and Lazdunski M. (1983) Electrophysiological characterization, solubilization and purification of the Tiryus y toxin receptor associated with the gating component of the Na+ channel from rat brain. Eur. molec. Biol. Ora. J. 2. 915-920. Bontemps J., Cantineau R., Grandfils C., Leprince P., Dandrifosse G. and SchotTenie& E. (lb84) High-yield synthesis of a [3Hlethylenediamine ditetrodotoxin derivative. Analvf. Biochem. 139. 149-157. Catterall W. A. (1980) Neurotoxins that act on voltage-sensitive channel in excitable membranes. A. Ra!. Pharmac. Toxic. 20, 15-43. Cooper E. C., Tomiko S. A. and Agnew W. S. (1987) Reconstituted voltage-sensitive sodium channel from Electrophorus electricus. Chemical modifications that alter regulation of ion permeability. Proc. nam. Acad. Sci. U.S.A. 84A, 62826286. De Rycker C., Grandfils Ch., Bettendorff L. and Schoffeniels E. (1989) Solubilization of sodium channel from human brain. J. Neurochem. 52, 349-353. Elmer L. W., O’Brien B. J., Nutter T. J. and Angelides K. J. (1985) Physicochemical characterization of the a-peptide of the sodium channel for rat brain. Biochemistry 24, 8128.-8137. Frelin Ch., Vigne P. and Lazdunski M. (1983) The specificity of the sodium channel for monovalent cations. J. biol. Chem. 258, 72561259. Garber S. S. and Miller C. (1987) Single Na+ channels activated by veratridine and batrachotoxin. J. gen. Physiol. 89, 459480. Garty H., Rudy B and Karlish S. J. D. (1983) A simple and sensitive procedure for measuring isotope fluxes through ion-specific channels in heterogeneous populations of membranes vesicles. J. biol. Chem. 258, 13.09&13,099. Grandfils C. (1988) Focus on [‘H]TTX-lysine-tetrodotoxin. Janssen Chim. Acta 4, 10-l I. Grishin E. V., Kovalenko V. A., Pashkov V. N. and Shamotienko 0. G. (1984) Isolation and characterization of sodium channel components. Biol. Membr, U.S.S.R. 1, 858-867. Hartshorne R. P. and Catterall W. A. (1984) The sodium channel from rat brain; purification and subunit composition. J. biol. Chem. 259, 1667-1675. Hartshorne R. P., Keller B. V., Talvenheimo J. A., Catterall W. and Montal M. (1985) Functional reconstitution of the purified brain sodium channel in planar lipid bilayers. Proc. natn. Acad. Sci. U.S.A. 82, 240-244. Holloway P. W. (1973) A simple procedure for removal of Triton X-100 from protein samples. Aitalyt. Biochem. 53, 304.-308. Kasahara M. and Hinkle P. C. (1977) Reconstitution and purification of the D-glucose transport from human erythrocytes. J. biol. Chem. 252, 73847390. Kraner S. D., Tanaka J. C. and Barchi R. I. (1985) Purification and functional reconstitution of the voltage-sensitive sodium channel from rabbit T-tubular membranes. J. biol. Chem. 260, 6341-6347. Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nuture 227, 680-685. Leach B., Collawn J. and Fish W. (1980) Behavior of glycopeptides with empirical molecular weight estimation methods. I. In sodium dodecylsulfate. Biochemistry 19, 57365741. Levinson S. R. and Ellory J. C. (1973) Molecular size of the tetrodotoxin binding site estimated by irradiation inactivation. Naiure New Biol. 245, 122-123. Levinson S. R., Duch D. S., Urban B. W. and Recio-Pinto E. (1986) The sodium channel from Electrophorus electricus. Ann. N. Y. Acad. Sci. 459, 162-I 78. Lombet A., Kazazoglou T., Delpont E., Renaud J. P. and Lazdunski M. (1983) Ontogenic appearance of Na+ channels characterized as high affinity binding sites for tetrodotoxin during development of the rat nervous muscle system. Biochem. biophp. Res. Commun. 110, 8944901. Lombet A. and Lazdunski M. (1984) Characterization, solubilization, affinity labeling and purification of the cardiac Na+ channel using Tir~us toxin y. Eur. J. Biochem. 141, 651600. Messner D. J. and Catterall W. A. (1985) The sodium channel from rat brain. Separation and characterization of subunits. J. hiol. Chem. 260, IO,597710,604. Miller J. A., Agnew W. S. and Levinson S. R. (1983) Principal glycopeptide of the tetrodotoxin/saxitoxin binding protein from Electrophorus electricus. Isolation and partial chemical and physical characterization. Biochemisrr,v 22, 462470. Mourre C., Lombet A. and Lazdunski M. (1984) Autoradiographic localization of tetrodotoxin-sensitive Na+ sodium channels in rat brain. Neurosci. Left. 52, 31 -35. Norman R. I., Schmid A., Lombet A., Barhanin J. and Lazdunski M. (1983) Purification of binding protein for Trtyn.s )l toxin identified with the gating component of the voltage-sensitive Na + channel. Proc. natn. Acad. Sci. U.S.A. 80, 41644168. Palacios J. M., Probst A. and Cortes R. (1986) Mapping receptors in the human brain. Trends Neurosci. 9, 284-289. Pannone P. (1980) Voltage clamn experiments in normal and denervated mammalian skeletal muscle fibers. J. Phpsiol., Ldnh. 305, i774iO. _ _ Peterson G. L. (1977) A simplification of the protein assay method of Lowry et al., which is more generally applicable. Analvt. Biochem. 83, 346-356. Rosenberg R. L., Tomiko S. A. and Agnew W. S. (1984) Reconstitution of neurotoxin-modulated ion transport by the voltaee reeulated sodium channel isolated from the electroplax of E1ectrophoru.r electricus. Proc. natn. Arad. Sci. U.S.A. Si, 12:9--1243. Tamkum M. M. and Catterall W. A. (1981) Reconstitution of the sodium channel of rat brain from solubilized components, J. biol. Chem. 256, 11,457-11,464.
Molecular aspects of human brain sodium channel
817
36. Tamkun M. M., Talvenheimo J. A. and Catterall W. A. (1984) The sodium channel from rat brain: reconstitution of neurotoxin-activated ion flux and scorpion toxin binding from purified components. J. biol. Chem. 259, 16761688. 37. Tanaka J. C., Eccleston J. F. and Barchi R. L. (1983) Cation selectivity characteristics of the reconstituted voltage-dependent sodium channel purified from rat muscle sarcolemma. .I. biol. Chem. 258, 7519-7526. 38. Villegas R., Sorais-Landaez F., Rodriguez-Grille B. and Villegas G. (1988) The lobster nerve sodium channel: solubilization and purification of the tetrodotoxin receptor protein. Biochim. biophys. Acta 941, 150-156. 39. Weigele J. B. and Barchi R. L. (1982) Functional reconstitution of the purified sodium channel protein from rat sarcolemma. Proc. natn. Acad. Sci. U.S.A. 79, 3651-3655. 40. Worley P. F. and Baraban J. M. (1987) Site of anticonvulsant action on sodium channels: autoradiographic and electrophysiological studies in rat brain. Proc. narn. Acad. Sci. U.S.A. 78, 775-779. (Accepted 25 April 1990)
MOLECULAR
0306-4522/90 53.00 + 0.00 Pergamon Press plc 0 1990 IBRO
I990
ASPECTS OF HUMAN CHANNEL
BRAIN SODIUM
C. DE RYCKER and E. SCHOFFENIELS Laboratoire de Biochimie Gin&ale et Comparee, Universite de Li&ge, 17 Place Delcour, 4020 Ligge,
Belgium Abstract-The sodium channel content of human brain was measured by tritiated tetrodotoxin specific binding. After solubilization, the sodium channel was submitted to chromatography on diethylaminoethyl(cellulose) Sephadex, hydroxylapatite and wheat germ agglutinin sepharose. An increase of tritiated tetrodotoxin binding specific activity was subsequently observed. Eluted sodium channels from wheat germ agglutinin sepharose were overlaid on a sucrose gradient. Electrophoretical analysis of the material obtained after the sedimentation step revealed two co-purified peptides, t( (kf, = 275,000 mol. wt) and fi (44, = 30,~36,~ mol. wt). Alpha showed an exceptionally high free electrophoretic mobility, which is a common feature for all sodium channels previously described. However, the high denaturation rate of the solubilized tetrodotoxin receptor site 1 did not allow tetrodotoxin receptor quantification by the tritiated toxin binding in sucrose fractions. Sodium channel effective reconstitution in liposomes was demonstrated: (1) **Na+ influx in proteoliposomes was sensitive to sodium channel-specific neurotoxins; (2) reconstituted proteins showed a cation
selectivity similar to that previously described for animal sodium channels. The sodium channel preparation obtained after four chromatographic steps shows two peptides on the electrophoretic analysis. Reconstituted sodium channels displayed some physiological properties found in intact conducting membranes.
composed of at least a =260,000 mol. wt glycoprotein (a). The SC isolated from rat brain includes additional subunits of M, 39,OOOmol. wt @I) and 37,000 mol. wt (fi2).16 Beta 1 is non-covalently associated to a while /32 is linked to ct by disulfide bridges.27 However, as showed by radiation inactivation techniques, the rat brain SC is a complex composed of only the 260,000 mol. wt (CX)and 39,000 mol. wt (fi 1) polypeptides.3 The c@1 complex seems to be sufficient to modulate transmembrane ion flux after reconstitution into artificial membrane and responds to pharmacological agents similarly to SC in sim17-34*35~39 However, other authorssy’O have described rat brain SC as a single 270,000 mol. wt peptide, CL Those divergent observations led us to study the SC from human brain. The purification procedure described for rat brain was adapted to human brain. Electrophoretic analy sis of the obtained protein(s) was performed after each chromatography. Nevertheless, as shown previously,’ the tetrodotoxin receptor site of human SC is sensitive to denaturation when solubiiized. It was therefore of great interest to investigate whether the protein obtained after all chromatographic steps did keep its other native properties. This investigation was performed by using SC reconstituted into liposomes.
The voltage-sensitive sodium channel or conducting site for sodium (SC) is responsible for the rapid increase in membrane sodium permeability which corresponds to the action potential depolarizing phase in excitable tissue. Biochemical studies involve the use of specific neurotoxins which act at five receptor sites of the protein and allow structural and functional studies of the SC. The most widely used are saxitoxin and tetrodotoxin. They bind to receptor site 1. As a result, the site is in a low conductance state. Veratridine, batrachotoxin, grayanotoxin and aconitine bind to receptor site 2 and cause a persistent activation of the protein, i.e. a high conductance state. Other classes of neurotoxins bind to other receptor sites and have various physiological effects.‘* Although electrical properties of SC have been largely investigated, their molecular structure has only recently been described. Current evidence from photoaffinity labeling, radiation inactivation, purification and reconstitution suggest that the SC from eel electroplax,1~28~30rat brain,5~‘0~15~‘6 rat and rabbit sarcolemma,4,20 chick heart26 and lobster nerves38 is Ab~reuiations: BSA, bovine serum albumin; DEAE, diethylaminoethyl (cellulose); EDTA, ethylenediaminetet~aacetate; FTS, freeze thaw sonication nrocedure: HEPES. N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acidf N-Ag, N-acetyl glucosamine; NP-40, Nonidet P-40, PC, phosphatidylcholine; PAGE, polyacrylamide gel electrophoresis; PE, phosphatidylethanolamine; PMSF, phenyl methylsuIfonylfluo~de; PS, phosphatidylse~ne; SC, sodium channel; SDS, sodium dodecyl sulfate; t3H]lTX, tritiated tetrodotoxin; WGA, wheat germ agglutinin; WW, wet weight.
EXPERIMENTAL PROCEDURES
Materials Human brains were obtained as described previously.9 Chemicals were purchased 809
from the following sources.
810
C. DE RYCKERand E. SCHOFFENIELS
Phenylmethylsulfonylfluoride (PMSF), Nonidet P-40 (NP40), phosphatidylcholine (PC) type III from egg yolk, wheat germ agglutinin @VGA) from Triticum vulguris and cation exchange resin Dowex 5OW-X8 were from Sigma (St Louis, MO, U.S.A.). l,lO-Phenanthroline monohydrate and iodoa&amide were from Janssen Chemica (Beerse, Belgium). Diethylaminoethyl(ceIlulose) (DEAE) Sephadex A-25, Sepharose 4CNBr, Sephadex G-200 and calibration protein for the sodium dodecyl sulfate-polyacrylamide gel electronhoresis (SDS-PAGE) were from Pharmacia (Unnsala. Sweden). &o-Gel HA,‘Bio-Beads SM-2 and silver s&in kit were from Bio-Rad Laboratories (Richmond, CA, U.S.A.). N-Acetyl glucosamine (N-A%) and bovine serum albumin were from Fluka (A. G. Buchs, Switzerland). Veratridine (sulfate form) was a generous gift of Dr Angenot (University of Liege, Belgium). Batrachotoxin was kindly provided by Dr J. W. Daly, of the Department of the National Institute of Arthritis, Metabolism and Digestive Deseases, National Institute of Health, U.S.A. Tetrodotoxin (free of citrate) was obtained from Sankyo Co. Ltd, Tokyo, Japan. Tritium labeling was carried out by our modification14 of the method of Bontemps et ~1.~Saxitoxin was generously provided by E. J. Schantz, University of Wisconsin. Phosphatidylcholine (PC), phosphatidylserine (PS) and phosphatidylethanolamine (PE) added during final reconstitution were from Avanti Polar (Birmingham, AL, U.S.A.). **Na+ (1 mCi/ml, carrier free) was from New England Nuclear (Boston, MA, U.S.A.). 13’Cs+ and 83Rb+ were provided by Dr Guillaume (University of Liege, Belgium). Dialysis tubings were from Thomas Corp. (Philadelphia, PA, U.S.A.). Dowex minicolumns used for flux assays were made out of 1 ml polypropylene syringes, Syringe ends were cut and replaced by a piece of nylon netting (Scrynel, 200 mesh). Sonication was performed with a titane lead connected to an ultrasound generator (Braun Melsungen A.G., Germany). All other products were from the best grade available and all solutions were made with Milli-Q water (Millipore Corp., Bedford, MA, U.S.A.). Methods Preparation of wheat germ agglutinin sepharose. WGA sepharose was obtained by coupling WGA with sepharose 4B-CNBr: sepharose was washed with a large volume of 1 M HCl. Five milliliters of wet gel were added to 50 mg of WGA in 2ml of buffer A (1OOmM Na,HCO,, pH 8.3, 500mM NaCI, 0.1 mM N-Ag). The mixture was stirred overnight at 4°C and then washed with 20 ml of buffer A. Five milliliters of 100 mM ethanolamine, pH 8.0, was added to the gel in order to block the excess of reactive groups. After a 2 h incubation at 4°C WGA sepharose was put in a small column (0.5 x 10 cm) and washed twice with sodium acetate (100 mM, pH 4.0, 500 mM NaCI) and Tris-HCl (100 mM, pH 8.0, 500 mM NaCl). Brain membrane preparation. All operations were made at 4’C and all buffers contained protease inhibitors at the following concentrations: iodoacetamide 1mM, orthouhenantroline 1 mM. PMSF 0.1 mM and pepstatin 1 bM. ’ Einhtv grams of frozen brain (white and gray matter) were allowed to thaw in 10 volumes of homogenization buffer (IO mM Tris-HCl. PH 7.4. 0.32 M sucrose and protease inhibitors), then diced into small fragments and homogenized in a Teflon Potter-Elvehjem homogenizer. The homogenate was centrifuged at 120g for 10 min; the supernatant (Sl) was saved; the pellet (Pl) was resuspended in homogenization buffer and centrifuged again at 120g for the same time. The resulting supernatant (S2) was added to St and the pellet (P2) discarded. Sl + S2 were centrifuged for 40 min at 27,000g. The supernatant (S3) was discarded and the pellet (P3) was resuspended in 1I of buffer 5 mM TrissHCl, pH 8.2, 1 mM EDTA. containing the four protease inhibitors listed above and gently stirred on ice for 15 min. After the osmotic shock, the suspension WdS I
.-
centrifuged for 80 min at 27,000g. Pellets (P4) were combined and resuspended in 2 volumes of buffer (20mM HEPES-Tris, 120mM KCl, pH 7.4, protease inhibitors) per volume of starting material. P4 (membrane fraction) was quickly frozen in a dry ice-cold acetone mixture and maintained at -70°C until use. [‘HlTetrodotoxin binding measurements. Tissues were homogenized in 10 mM Tris-HCl buffer, pH 7.4, containing protease inhibitors, prior to [3H]TTX binding measurements [3H]TTX binding assays on membranes and solubilizedz4sodium channel measurements were done as described previously.9 Protein assay. Protein content determinations were made according to the Peterson procedure.” Sodium channel solubilization. P4 (160 ml from 80 g of tissue) was put in a beaker packed in ice on a magnetic stir plate. The same volume of buffer containing the detergent (20 mM HEPESTris, pH 6.8, 2% NP-40, 0.2% PC, protease inhibitors) was added in small increments with constant mixing. Fifteen minutes later, unsolubilized material was sedimented at 120,000g for 30 min and the supernatant (membrane extract) was kept for further analysis. Chromatographic steps performed on membrane extract.
All manipulations were conducted within 24 h following the SC solubilization. Membrane extract was adjusted to 10 mM CaCI, prior to chromatography. The pH -was then lowered to -6.5 by addition of 0.5 M histidine-HCl. PH 5. The membrane extract conductivity was then raisei to 15 mS/cm by addition of 3 M KCI. Wet DEAE-Sephadex A-25 (200 ml) was rinsed with six volumes of equilibration buffer (20 mM histidineeHC1. pH 6.5, 100 mM KCI, 0.1% NP-40,0.025% PC). Membrane extract (180m1, containing 50@600mg of protein) was gently stirred with the wet gel for 30 min. The mixture was transferred into a Btichner vessel and the non-adsorbed proteins were eliminated by four equilibration buffer washes. The Biichner vessel outlet was connected to a fraction collector and SC was eluted with 250ml of equilibration buffer supplemented with 500mM KCI. When eluted fraction conductivities ranged from 15 to 50mS/cm, they were pooled for further chromatography. Usually, six fractions of lOm1 each were pooled (DEAE pool). In order to prepare the DEAE pool for hydroxylapatite chromatography, its pH was raised to 8.0 by addition of 2 M Tris-HCl, pH 8.9. EDTA was added to a final concentration of 10 mM. This step lowers the pH to 7.4. Finally, 800 mM KH,PO,, pH 7.4, was added to obtain a final concentration of 80 mM. The DEAE pool was swirled for 15 min with 100 ml of Bio-Gel HT, previously equilibrated in 80 mM KH2P0,, pH 7.4, 0.1% NP-40 and 0.025% PC. The resin suspension was poured into a 2.4 x 20cm column and washed with 300 ml of equilibration buffer. SC was eluted by 300 ml of the same buffer containing 400 mM KHrPO,. Fractions binding [3H]TTX were pooled (hydroxylapatite pool). WGA sepharose was equilibrated with 50ml of the elution buffer used for hydroxylapatite chromatography. The hydroxylapatite pool was run through the column at a 1 ml/min rate. Resin was washed with 50 ml of HEPESTris (25 mM, pH 7.4, 0.5 mM MgSO,, 205 mM Na,SO,, 0.1% NP-40, 0.025% PC) buffer. SC were eluted by N-Ag (150 mM) in a HEPESTris buffer (25 mM, pH 7.4, 67.5mM Na,SO,, 0.5mM MgSO,, 0.1% NP-40, 0.025% PC). When SC are prepared for reconstitution experiments. DEAE elution buffer is used for WGA chromatography since the hydroxylapatite step is omitted. Five-milliliter fractions were collected during washes and 2.5 ml fractions were collected during elution. Fraction 22 was concentrated by dialysis and was then layered on a linear sucrose gradient (5-20%, prepared with sepharose elution buffer without
811
Molecular aspects of human brain sodium channel N-Ag) and sedimented for 16 h at 145,080g at 4°C in a Beckman ultracentrifuge. The gradient was then fractionated into 15 fractions of 750~1 each. Sodium dodecylsulfate gel electrophoresis. The general conditions of SDS-PAGE were those described by 10mM Laemmli.2’ Samples were dialysed against Na,HCO,, pH 7.4, and lyophilized. Dry material was resuspended in loading buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 5% /I-mercaptoethanol, 12.5% glycerol, Bromophenol Blue) and heated for 15 min at 100°C. Samples were analysed by gradient SDS-PAGE (412%); four constant acrylamide concentrations (4.4, 4.8, 5.2 and 5.6%) were used for a Ferguson plot. Stacking gels were 3% acrylamide. Gels were silver stained and densitometric scans of migration patterns were performed using an ultroscan detector (LKB, model 220, Sweden). Reconstitution. Reconstitution was performed as previously described2 for electric eel SC: WGA sepharose eluate was gently swirled for 3 h at 4°C with 0.3 g/ml of wet Bio-Beads (equilibrated as described by Holloway’s) to adsorb Nonidet P-40. Proteoliposomes were dialysed overnight against a large volume of buffer A (25mM HEPESTris, 67.5 mM Na,SO,, pH 7.4) in order to lower the KC1 concentration (500mM) contained in WGA sepharose elution buffer. Other liposomes were synthesized as follow: PE, PS, and PC were mixed in a molar ratio 5:4: 1 and dried under N, Buffer A was added to the dry residue to obtain a phospholipid concentration of 40 mg/ml. The mixture was vortexed for 5 min and then sonicated on ice for IOmin under nitrogen. Proteoliposomes were then fused with those liposomes using the freeze thaw sonication (FTS) procedure described by Kasahara and Hinkle:@ one volume of liposomes was added to 3 volumes of proteoliposomes and 1 ml aliquots were immersed for 1 min in dry ice-cold acetone. The suspension was then allowed to thaw at room temperature. The new enlarged proteoliposomes were submitted to a 10 s sonication and kept on ice until use. Electron microscopy. Enlarged proteoliposome aliquots were diluted two-fold in buffer A. Copper grids (2OOmesh, pretreated with Formvar) were gently reversed on liposome droplets for 2min. Grids were stained with phosphotungstic -acid (2% v/v, pH 7) for 1 min and directly observed with a Jehol lOOCX-2 microscone. Distributions of vesicle size were established from the’ obtained pictures. 22Na+~JUXassays. Polypropylene minicolumns (used for **Na+ flux measurements) were filled with Dowex 5OW-X8 (0.5 ml), washed with 2 x 1 ml of ice-cold BSA solution (3.3 mg/ml) and 220 mM sucrose and maintained at 4°C until use. Just before use, aliquots of proteoliposomes were overlaid on Dowex minicolumns pre-equilibrated with 10 ml of 20 mM Tris sulfate, pH 7.4, in order to create a membrane potential which enhance Na+ entrance into the vesicles.13 Proteoliposomes were then quickly diluted in 3 volumes of buffer B (170 mM Tris sulfate, pH 7.4, 10 mM NaCl, lOpCi/ml 22Na+) at 37°C.
Two 200-~1 samples of the radioactive suspension were overlaid at the indicated times on Dowex BSA sucrose pre-equilibrated columns. Elution began less than 5 s after loading and the eluate was directly collected in scintillation vials (Lumac, milli 6, 6372 AE Landgraaf, Holland). Effect ofneurotoxins on 22Na+ influx. Final concentrations of saxitoxin, tetrodotoxin and batrachotoxin during fluxes were 1 PM; veratridine concentration was 1OOpM. Batrachotoxin stock solution (500 PM) was supplied in ethanol solution and control experiments on proteoliposomes treated with ethanol alone were performed. Ionic selectivity. 22Na+, 83Rb+ and t3’Cs+ influx were measured separately on proteoliposomes prepared from the same reconstitution in the presence of IOOpM veratridine and under control conditions. For each point, sodium channel influx was calculated by subtraction of the nonspecific influx (control conditions) from the influx measured in the presence of veratridine. Three individual kinetics were performed in order to avoid any interference in radioactive counts. RESULTS
Sodium channel content in homogenate membranes of human brain
Membrane extract chromatographic steps
A summary of the chromatographic steps performed on membrane extract is given in Table 1. The elution profile of DEAE sephadex is shown in Fig. 1A. SC was eluted in fractions 21-24 (DEAE pool), which exhibited an increase of conductivity. After several adjustments (see Experimental Procedures), the DEAE pool-was chromatographied on an hydroxylapatite column. The hydroxylapatite column elution profile is illustrated in Fig. 1B. Fractions 10 and 11 (hydroxylapatite pool) contained 38% of the SC and 36% of total proteins loaded on the column. The specific activity increase was negligible but hydroxylapatite chromatography was essential to eliminate a 220,000 mol. wt peptide, as shown by electrophoretical analysis. The hydroxylapatite pool was loaded on a WGA sepharose column. Sixty per cent of the loaded
TTX receptor
Membrane extract DEAE HA WGA
pm01
%
1488 + 128 100 346 f 43 23.2 139+20 9.3 67k 14 4.5
Values are f SE., n = 4.
in
Human gray matter [‘H]T”TX binding (82 pmol/g wet weight, WW) was twice that of white matter (43 pmol/g WW). Whole brain homogenate binds 50 pmol of [‘H]TTX/g WW. These data suggest that gray matter represents 27% of the human brain volume. Membranes (P4) obtained from 80 g of brain tissue bound 3000 pmol of [3H]TTX.
Table 1. Summary of the chromatographic steps performed membrane extract
Step
and
on the human brain
Protein mg 633 &70 30 k 6 11&3 1.1 * 0.2
% 100 4.73 1.73 0.17
Suecific activitv pmol/mg prot: 2.35 11.53 12.63 60.90
812
C. DE RYCKER and E. SCHOFFENIELS
FRACTION
1
s
10
15
FRACTION
Fig. 1. Profiles of the four chromatographies carried out as described in Experimental Procedures. (A) Ion exchange chromatography on DEAE Sephadex A-25. (B) Hydroxylapatite chromatography on Bio-Gel HT. (C) Affinity chromatography on WGA sepharose. (D) Sucrose gradient sedimentation. (0) TTX receptor measured by rapid gel filtration. (0) Protein content determined by the Peterson method. (A) Conductivity. (x) Sucrose percentage.
x103
-a
220-
9467431, 3630-
Fig. 2. SDS-PAGE of membrane extract (lane 1) and eluates of the four chromatographies performed on solubilized human brain SC: DEAE pool (lane 2), HA pool (lane 3), WGA pool (lane 4), and fraction 13 of sucrose gradient (lane 5). Analysis was performed on a 412% linear acrylamide gradient.
813
Molecular aspects of human brain sodium channel
06
1
2 %
3
L
5
6
ACRYLAMIDE
Fig. 3. Ferguson plot of the 275,000 mol. wt peptide (A) and of four acrylamide calibration proteins. SDS slab gels of indicated acrylamide concentration were prepared as described in Experimental Procedures. SC (A), lactate dehydrogenase (m), ovalbumin (O), albumin (A) and ferritin (0) were run on each gel. Extrapolations of drawn lines gave peptide-free mobility values.
went through the column and 10% were eluted after addition of N-Ag (Fig. 1C). Since this proteins
sugar eluted 50% of the SC overlaid on the column, the increase of [3H]TTX binding specific activity was substantial. Fraction 22 was then sedimented through a sucrose gradient. The protein distribution in the sucrose gradient is presented in Fig. 1D. No [3H]TTX binding activity
was detected. Proteins were distributed in two major peaks. The first one (fractions 3 and 4) corresponded to low molecular weight proteins and the second one (fractions 13-15) to catalase (240,000 mol. wt) in calibration gradient. The second peak contained 25% of the loaded proteins. Electrophoretic
analysis
The electrophoretic pattern of proteins obtained after each chromatography is shown in Fig. 2. Membrane extract proteins are shown in lane 1. The doublet underlying the diffuse band on the gel top (lanes 1 and 2) was retained on hydroxylapatite (lane 3). Low molecular weight peptides (under 94,000 mol. wt) were lost during affinity chromatography (lane 4). Phosphorylation assays on those preparations showed that the 94,000 mol. wt peptide was the Na+, K+, ATPase c( subunit (not shown). Proteins contained in the 13th fraction of the sucrose gradient (lane 5) appeared as a large diffuse band (275,000 mol. wt) and a group of smaller peptides was seen from 30,000 to 36,000 mol. wt. One of them (33,000 mol. wt) was more heavily stained. 275,000 mol. wt peptide Ferguson plot A Ferguson plot of four calibration proteins and the 275,000 mol. wt peptide is presented in Fig. 3. Lines corresponding to calibration proteins have the
Table 2. Peak surfaces (arbitrary units) of 3O,OO(r36,000 and 275,000 mol. wt peptides from polyacrylamide gel electrophoresis densitometric scans of diethylaminoethyl (cellulose) and wheat germ agglutinin pool protein patterns 30,00~36,000 mol. wt
275,000 mol. wt
253 + 42 405 & 38
1585 f 124 2325 k 196
DEAE pool WGA pool Values are *SE.,
n =4.
same y intercept
(0.520), indicating that they have similar electrophoretic mobilities. The y intercept of the 275,000 mol. wt peptide line (0.730) is 1.6fold higher; this underlines the particularly high free electrophoretic mobility of the 275,000 mol. wt peptide. Correlation between the 275,000 36,000 mol. wt peptides
and the 30,00&
Densitometric scans have been performed on different protein migration patterns. The 275,000 mol. wt and the 30,00&36,000 mol. wt peak peptide surfaces were proportional (Table 2). Proteoliposome
size and morphology
Eighty-five per cent of the intravesicular volume was represented by 64-180 nm proteoliposomes. Most of the vesicles were unilamellar and this aspect was strictly dependent on the short sonication length following the freeze thaw process (not shown). Effect of neurotoxins liposomes
on 22Na+ uptake into proteo-
Uptakes of “Na+ into proteoliposomes after 5 s and 5 min of flux are represented in Fig. 4. Intravesicular 22Na+ was observed in control conditions meaning that proteoliposomes were permeable to
814
C. DE RYCKERand E. SCHOFFENIELS
After 30 min fluxes, the ratios of intravesicular radioactivity/total loaded radioactivity were similar for the three cations (*‘Na+, 1.7%. *3Rb+, 2.1%; ‘Ws+, 1.8%). These percentages iepresented the accessible intravesicular volume in this population of vesicles. SC selectivity may be expressed taking the sodium selectivity as 1. In this convention, reconstituted human SC ionic selectivity was Na+/Rb+/ Ca+ = 1:0.48:0.33 in the presence of 100pM veratridine.
T
B
DISCUSSION CBVTS
C
0
V
T
S
Fig. 4. 22Na+ intravesicular accumulation after 5 s (A) and 5 min (B) of incubation. Influx was expressed by the ratio intravesicular cpm/cpm loaded on the Dowex column. Incubation occurred without (C) or with different toxins: batrachotoxin (B), veratridine (V), saxitoxin (S) or tetrodotoxin (T). Error bars showed the results obtained in two experiments performed on the same batch of vesicles.
sodium. Addition of guanidinium toxins (saxitoxin or tetrodotoxin) induced a slight decrease of the radiotracer accumulation. Veratridine induced, 5 min after its addition, a 50% increase of 22Na+ uptake. Batrachotoxin also had an enhancing effect on uptake, even 5 s after its addition. After 5 min, intravesicular “Na+ reached nearly double that of intravesicular 22Na+ measured in control conditions. As batrachotoxin was supplied in ethanol solution, control experiments were made with the solvent alone (0.2% final concentration) and no effect on proteoliposome permeability was observed. Ionic selectivity
Veratridine-stimulated influx was measured for **Na+, r3Rb+ and 13’Cs+ in proteoliposomes at 37°C (Fig. 5). The half-time for ‘*Na+ flux equilibration was 1.3 min while half-times for 83Rb+ and 13’Csf were 2.3 and 4min, respectively.
Sodium channel content in human brain
Results of SC contents from three types of tissues are summarized in Table 3. The human brain is less dense in SC than the rat brain. Two hypotheses may explain this observation. (1) In human, a proteolytic digestion occurred between donor death and tissue freezing. (2) Brain SC density is much lower in human than in rat in vivo. Observations made by Palacios et aL3’ reinforced the second hypothesis as these authors underlined the lower density of about 20 types of receptors in human brain compared to rat brain. Moreover, the SC content of human gray matter is twice that of white matter. Specific tritiated ligands used in autoradiography have also shown the same type of distribution in rat brain.29,40 Chromatographic
steps evaluation
Previous experiments showed that human SC was highly sensitive to denaturation when solubilized.’ This led to a lack of ability in measuring the real amount of solubilized receptor from [3H]TTX binding. Nature of the 275,000 mol. wt 36,000 mol. wt peptides
and
the
30,000-
Mammalian SC is composed of a large subunit tl (260,00~300,000 mol. wt) and one smaller subunit fi
.**No
.83Rb . ‘37cs 15
20
25
30
Fig. 5. Selective cation passage through the veratridine-activated reconstituted SC. *iNa+, *‘Rb+ and “‘Cs+ uptakes were measured in the presence of 100 PM veratridine and in control conditions at Indicated times. Control isotopic uptake measured in the absence of veratridine was subtracted from the total isotope uptake to obtain the stimulated isotopic uptake.
Molecular aspects of human brain sodium channel Table 3. [)H]TTX binding measured on homogenates three excitable tissues in three laboratories [‘H]TTX bound (pmol/g ww)
Tissue Electric organ of E. electricus Brain Human (whole) gray matter white matter Rat (whole) Rat (whole) cerebellar cortex Port, gray matter Muscle Rat (skeletal)
of
References
30,
23
50 82 43 150 77 200 50;
9 9 9 25 9 25 23
Ionic selectivity
*Measurements performed on lyophylized preparations. (~O,OOO-40,000 mol. wt),
sometimes
shown
as
a
doublet B 1 and /12.4,15,‘6,20 We propose that tl and /I peptides shown in lane 5 are human SC components. The three following observations led to this conclusion. (i) [3H]TTX binding specific activity increases through chromatographies and coincides with the presence of 275,000 and 30,000-36,000 mol. wt peptides on SDS-PAGE. (ii) The 275,000 mol. wt peptide presents the same diffuse aspect and the same high free electrophoretic mobility (see Ferguson plot) as the a SC subunit from other species. This high free mobility is exceptional since most of the glycoproteins reveal lower free mobilities than non-glycosylated proteins.22 The originality is specific to the SC u subunit and is probably due to an excess of protein-bound SDS. Since no other human glycoprotein has ever shown this exceptional electrophoretic mobility, we propose that the 275,000 mol. wt peptide is the human SC tl subunit. (iii) Densitometric scans of the protein pattern have shown that the 30,000-36,000 mol. wt peptide was co-purified with the a subunit. This small peptide is thus the human SC /3 subunit. In some PAGE it appeared as a doublet. This uneven appearance has already been described in rat brainI Human
sodium
channel
chotoxin was a more potent SC activator than veratridine. Batrachotoxin activates more than 95% of SC whereas veratridine is a partial agonist (only 50%
of the SC population is activated).’ Moreover, the time length of openings induced by batrachotoxin is higher than that induced by veratridine.‘* Requirements for restoration of the native properties of receptor sites 1 and 2 are different in reconstituted human SC.
25
8
815
reconstitution
Neither saxitoxin nor tetrodotoxin had a significant effect on intravesicular **Na+ accumulation (Fig. 4). These guanidinium toxins, binding to receptor site 1, are no longer effective as blockers on reconstituted human SC. Thus, the native properties of site 1 receptors were not recovered by reconstitution in our lipid bilayers. Both veratridine and batrachotoxin binding at receptor site 2 enhanced the intravesicular **Na+ uptake. The effect of batrachotoxin was more rapid and more important. This is in good agreement with several previous observations, showing that batra-
Our results are in good agreement with those previously published for other animal SC since the same type of ionic selectivity was observed. In fact, selectivity seems to be correlated to the cation size among other parameters, as suggested by Pappone,32 who found that rat SC had a permeability of 1.14 for lithium (A, = 6.9). Similar permeability ratios for rat sarcolemma1 (Na+/Rb+/Cs+ = 1:0.25:0.12) and brain (Na+/Rb+/Cs+ = 1:0.5:0.38) SC have been obtained previously.36,37 However, these values do not reflect the native SC selectivity as measured in the presence of veratridine. Higher selectivities were indeed observed in native forms; addition of batrachotoxin and mainly of veratridine decreased the degree of specificity somewhat.8.“.‘2.37
CONCLUSION
The substantial denaturation of the SC receptor site 1 was the main difficulty encountered in this study. It led us to consider well known SC properties other than tetrodotoxin binding to define the protein obtained after chromatographic procedures. Thus, we have shown that the reconstituted protein was still sensitive to site 2 targetting toxins and showed specific SC ionic specificity. Human brain was less dense in SC than rat brain. However, the same density distribution between white and gray matter was observed in both species. We propose a dimeric structure for the human SC. The 275,000 mol. wt diffuse band seen on PAGE had the same physicochemical properties as tl SC subunits previously described; the 30,000-36,000 mol. wt peptide, sometimes seen as a doublet, was co-purified with a. Thus, human SC appeared as a heterodimeric complex composed of two subunits, tl (275,000 mol. wt) and B (30,00036,000 mol. wt). However, further studies, more particularly based on immunological techniques, would be advantageous to confirm that the 30,000-36,000 mol. wt peptide belongs to the SC in situ.
REFERENCES
1. Agnew W. 8, Levinson S. R., Brabson J. S. and Raftery M. A. (1978) Purification of the tetrodotoxin-binding component associated with the voltage-sensitive sodium channel from Electrophorus elecrricus electroplax membranes. Proc. natn. Acad. Sci. U.S.A. 75, 2606-2610.
816
9. 10. II. 12. 13.
14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.
26. 27. 28.
29. 30.
31. 32. 33. 34.
3.5
C. DE RYCKER and E. SCHOFFENIELS Agnew W. S., Rosenberg R. L. and Tomiko S. A. (1986) Ion Channel Reconsfitution (ed. Miller C.), Ch. 12, pp. 307 330. Plenum Press. New York. Angelides K. J., Nutter T. J., Elmer L. W. and Kempner E. S. (1985) Functional unit size of the neurotoxin receptor on the voltage-dependent sodium channel. J. biol. Chem. 260, 3431-3439. Barchi R. L. (1983) Protein components of the purified sodium channel from rat skeletal muscle sarcolemma. J. Neurochem. 40, 1377-1385. Barhanin J., Pauron D., Lombet A., Norman R. I., Vijveberg H. P. M., Giglio J. R. and Lazdunski M. (1983) Electrophysiological characterization, solubilization and purification of the Tiryus y toxin receptor associated with the gating component of the Na+ channel from rat brain. Eur. molec. Biol. Ora. J. 2. 915-920. Bontemps J., Cantineau R., Grandfils C., Leprince P., Dandrifosse G. and SchotTenie& E. (lb84) High-yield synthesis of a [3Hlethylenediamine ditetrodotoxin derivative. Analvf. Biochem. 139. 149-157. Catterall W. A. (1980) Neurotoxins that act on voltage-sensitive channel in excitable membranes. A. Ra!. Pharmac. Toxic. 20, 15-43. Cooper E. C., Tomiko S. A. and Agnew W. S. (1987) Reconstituted voltage-sensitive sodium channel from Electrophorus electricus. Chemical modifications that alter regulation of ion permeability. Proc. nam. Acad. Sci. U.S.A. 84A, 62826286. De Rycker C., Grandfils Ch., Bettendorff L. and Schoffeniels E. (1989) Solubilization of sodium channel from human brain. J. Neurochem. 52, 349-353. Elmer L. W., O’Brien B. J., Nutter T. J. and Angelides K. J. (1985) Physicochemical characterization of the a-peptide of the sodium channel for rat brain. Biochemistry 24, 8128.-8137. Frelin Ch., Vigne P. and Lazdunski M. (1983) The specificity of the sodium channel for monovalent cations. J. biol. Chem. 258, 72561259. Garber S. S. and Miller C. (1987) Single Na+ channels activated by veratridine and batrachotoxin. J. gen. Physiol. 89, 459480. Garty H., Rudy B and Karlish S. J. D. (1983) A simple and sensitive procedure for measuring isotope fluxes through ion-specific channels in heterogeneous populations of membranes vesicles. J. biol. Chem. 258, 13.09&13,099. Grandfils C. (1988) Focus on [‘H]TTX-lysine-tetrodotoxin. Janssen Chim. Acta 4, 10-l I. Grishin E. V., Kovalenko V. A., Pashkov V. N. and Shamotienko 0. G. (1984) Isolation and characterization of sodium channel components. Biol. Membr, U.S.S.R. 1, 858-867. Hartshorne R. P. and Catterall W. A. (1984) The sodium channel from rat brain; purification and subunit composition. J. biol. Chem. 259, 1667-1675. Hartshorne R. P., Keller B. V., Talvenheimo J. A., Catterall W. and Montal M. (1985) Functional reconstitution of the purified brain sodium channel in planar lipid bilayers. Proc. natn. Acad. Sci. U.S.A. 82, 240-244. Holloway P. W. (1973) A simple procedure for removal of Triton X-100 from protein samples. Aitalyt. Biochem. 53, 304.-308. Kasahara M. and Hinkle P. C. (1977) Reconstitution and purification of the D-glucose transport from human erythrocytes. J. biol. Chem. 252, 73847390. Kraner S. D., Tanaka J. C. and Barchi R. I. (1985) Purification and functional reconstitution of the voltage-sensitive sodium channel from rabbit T-tubular membranes. J. biol. Chem. 260, 6341-6347. Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nuture 227, 680-685. Leach B., Collawn J. and Fish W. (1980) Behavior of glycopeptides with empirical molecular weight estimation methods. I. In sodium dodecylsulfate. Biochemistry 19, 57365741. Levinson S. R. and Ellory J. C. (1973) Molecular size of the tetrodotoxin binding site estimated by irradiation inactivation. Naiure New Biol. 245, 122-123. Levinson S. R., Duch D. S., Urban B. W. and Recio-Pinto E. (1986) The sodium channel from Electrophorus electricus. Ann. N. Y. Acad. Sci. 459, 162-I 78. Lombet A., Kazazoglou T., Delpont E., Renaud J. P. and Lazdunski M. (1983) Ontogenic appearance of Na+ channels characterized as high affinity binding sites for tetrodotoxin during development of the rat nervous muscle system. Biochem. biophp. Res. Commun. 110, 8944901. Lombet A. and Lazdunski M. (1984) Characterization, solubilization, affinity labeling and purification of the cardiac Na+ channel using Tir~us toxin y. Eur. J. Biochem. 141, 651600. Messner D. J. and Catterall W. A. (1985) The sodium channel from rat brain. Separation and characterization of subunits. J. hiol. Chem. 260, IO,597710,604. Miller J. A., Agnew W. S. and Levinson S. R. (1983) Principal glycopeptide of the tetrodotoxin/saxitoxin binding protein from Electrophorus electricus. Isolation and partial chemical and physical characterization. Biochemisrr,v 22, 462470. Mourre C., Lombet A. and Lazdunski M. (1984) Autoradiographic localization of tetrodotoxin-sensitive Na+ sodium channels in rat brain. Neurosci. Left. 52, 31 -35. Norman R. I., Schmid A., Lombet A., Barhanin J. and Lazdunski M. (1983) Purification of binding protein for Trtyn.s )l toxin identified with the gating component of the voltage-sensitive Na + channel. Proc. natn. Acad. Sci. U.S.A. 80, 41644168. Palacios J. M., Probst A. and Cortes R. (1986) Mapping receptors in the human brain. Trends Neurosci. 9, 284-289. Pannone P. (1980) Voltage clamn experiments in normal and denervated mammalian skeletal muscle fibers. J. Phpsiol., Ldnh. 305, i774iO. _ _ Peterson G. L. (1977) A simplification of the protein assay method of Lowry et al., which is more generally applicable. Analvt. Biochem. 83, 346-356. Rosenberg R. L., Tomiko S. A. and Agnew W. S. (1984) Reconstitution of neurotoxin-modulated ion transport by the voltaee reeulated sodium channel isolated from the electroplax of E1ectrophoru.r electricus. Proc. natn. Arad. Sci. U.S.A. Si, 12:9--1243. Tamkum M. M. and Catterall W. A. (1981) Reconstitution of the sodium channel of rat brain from solubilized components, J. biol. Chem. 256, 11,457-11,464.
Molecular aspects of human brain sodium channel
817
36. Tamkun M. M., Talvenheimo J. A. and Catterall W. A. (1984) The sodium channel from rat brain: reconstitution of neurotoxin-activated ion flux and scorpion toxin binding from purified components. J. biol. Chem. 259, 16761688. 37. Tanaka J. C., Eccleston J. F. and Barchi R. L. (1983) Cation selectivity characteristics of the reconstituted voltage-dependent sodium channel purified from rat muscle sarcolemma. .I. biol. Chem. 258, 7519-7526. 38. Villegas R., Sorais-Landaez F., Rodriguez-Grille B. and Villegas G. (1988) The lobster nerve sodium channel: solubilization and purification of the tetrodotoxin receptor protein. Biochim. biophys. Acta 941, 150-156. 39. Weigele J. B. and Barchi R. L. (1982) Functional reconstitution of the purified sodium channel protein from rat sarcolemma. Proc. natn. Acad. Sci. U.S.A. 79, 3651-3655. 40. Worley P. F. and Baraban J. M. (1987) Site of anticonvulsant action on sodium channels: autoradiographic and electrophysiological studies in rat brain. Proc. narn. Acad. Sci. U.S.A. 78, 775-779. (Accepted 25 April 1990)