- Email: [email protected]
Physostigmine and acetylcholine differentially activate nicotinic receptor subpopulations in Locusta migratoria neurons
Brain Research 789 Ž1998. 263–273
Research report
Physostigmine and acetylcholine differentially activate nicotinic receptor subpopulations in Locusta migratoria neurons Ingeborg van den Beukel ) , Regina G.D.M. van Kleef, Ruud Zwart, Marga Oortgiesen
1
Research Institute of Toxicology, Utrecht UniÕersity, P.O. Box 80.176, NL-3508 TD Utrecht, Netherlands Accepted 23 December 1997
Abstract The effects of acetylcholine ŽACh. and physostigmine ŽPHY. on thoracic ganglion neurons of Locusta migratoria were investigated using whole-cell and cell-attached voltage clamp. ACh activated whole-cell currents with variable amplitudes, time course and ion channel block between cells, suggesting differential expression of nicotinic acetylcholine receptor ŽnAChR. subtypes. This was supported by selective block of the peak of the currents by the a 7-specific a-conotoxin ImI. PHY at 100 m M evoked smaller whole-cell currents with variable amplitudes and marginal desensitization. The PHYrACh amplitude ratio varied between cells, and was positively related to the time constant of decay of the ACh response. EC 50 values for the peak amplitude of the ACh- and PHY-induced currents were 50 m M and 3 m M, respectively. Both agonists activated nAChR, indicated by equal voltage-dependence and reversal potentials and the same pharmacological properties of ACh and PHY responses. In addition, PHY and ACh induced ion channel block. Co-application and cross-desensitization experiments showed that ACh and PHY activate the same nAChR subpopulations. Both agonists activated nicotinic single channels with three conductance levels, which were equal for ACh and PHY, indicating activation of the same nAChR subtypes by both agonists. However, for all levels PHY displayed a lower open probability than ACh. Taken together, different whole-cell responses appear to originate from differential activation, desensitization and ion channel block by ACh and PHY of distinct nAChR populations. q 1998 Elsevier Science B.V. Keywords: Physostigmine; Neuronal nicotinic acetylcholine receptor; Subtype; Insect; Whole-cell voltage clamp; Single channel
1. Introduction Nicotinic acetylcholine receptors ŽnAChR. belong to the superfamily of ligand-gated ion channels, and are activated by the endogenous neurotransmitter acetylcholine ŽACh.. Two subfamilies of vertebrate nAChR exist: endplate nAChR, found at the neuromuscular junction and on Torpedo electric organs, and neuronal nAChR, expressed in the central and peripheral nervous system w25,33x. Both types of nAChR are assumed to consist of five subunits, arranged around a central pore w12,25x. In recent years, a range of different a and b neuronal subunits have been cloned. The a subunits are characterized by two adjacent cysteines in the N-terminal extracellular domain close to the putative ACh binding site, whereas b subunits lack these cysteines. To date, a 2– a 9 and b 2– b 4 subunits )
Corresponding author. Fax: q 31-30-2535077; E-m ail: [email protected] 1 Present address: Department of Neurobiology, Duke University Medical Center, Durham, USA. 0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 0 0 7 - 9
have been cloned, which in pairwise Ž a 2– a 4 with either b 2 or b 4., single Ž a 7– a 9. or triple Ž a 5 with other a and b subunits. expression in Xenopus oocytes yields functional nAChR w14,25,32,42x. Taking this range of subunits into account, a great variety of functional expression of subunit combinations is possible, and up to now the exact composition of nAChR in vivo is unclear w25x. Different native nAChR subtypes as well as different subtypes expressed in oocytes can be distinguished in terms of the potency order of different nicotinic agonists, the antagonist sensitivity, single channel conductances as well as activation and desensitization kinetics Žfor review, see Refs. w25,33x.. In contrast to vertebrates, nAChR in insects are confined to the central nervous system, where they are abundantly expressed in ganglia w8x. Like in vertebrates, insect nAChR consist of a subunits, containing the two cysteines characteristic of a subunits, and b subunits w16x. Various insect nAChR subunits have been cloned from different species, as for example four different a and two b subunits of Locusta migratoria ganglia w19x. Homology
264
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
between insect nAChR a subunits appears to be higher than that between insect and vertebrate a subunits w24x, while insect nAChR a subunits share a somewhat higher homology with vertebrate neuronal than with vertebrate endplate a subunits w7,16,24x. Heteromeric w16,34x as well as homomeric w8x insect nAChR have been reported in situ, while in Xenopus oocytes functional nAChR also result from coexpression w6x or single expression w3,24x of nAChR subunits. Like in vertebrates, insect nAChR subtypes have been found to differ in their pharmacological profiles w8,16x. The carbamate physostigmine ŽPHY., isolated from Calabar beans, has been used successfully in the treatment of glaucoma since 1864 w38x. Carbamates, widely applied as insecticides, are generally known to inhibit the enzyme acetylcholinesterase, thereby augmenting cholinergic transmission w2x. At higher concentrations, PHY blocks endplate and neuronal nAChR w29,30x. In addition, PHY has been reported to be an agonist at vertebrate endplate and neuronal and at Torpedo nAChR, activating single channel currents at a nAChR site different from the ACh binding site w29,30,36x. The blocking potency of PHY varies between nAChR from different species, while whole-cell agonistic actions of PHY have been observed only in locust thoracic ganglion neurons w41x. Therefore, PHY may be able to distinguish between nAChR of different species. In the present study, the agonistic properties of PHY on nAChR in thoracic ganglion neurons from L. migratoria are investigated in detail, and compared with those of ACh. Further, the ability of PHY to distinguish between nAChR subtypes within locust neurons is examined.
2. Materials and methods 2.1. Cells Locust neurons were acutely dissociated as described elsewhere w44x. Briefly, African locusts, L. migratoria, were used to prepare freshly dissociated thoracic ganglion neurons, 3–6 days after imaginal ecdysis. All three thoracic ganglia were dissected from the locusts and collected in ice-cold external solution Žfor composition see below.. Somata of the ganglion neurons were mechanically dissociated by aspirating the ganglia through a Pasteur pipette. Dissociated neurons were plated on tissue culture dishes coated with poly-L-lysine, stored at 48C, and used within 2 days. For experiments, cells were selected with a typical neuron-like appearance, i.e., with remains of axons and dendrites. 2.2. Electrophysiology Whole-cell membrane currents were measured using the standard whole-cell voltage clamp technique w17x. The membrane potential was held at y80 mV, using a RK 300
amplifier ŽBiologic, France.. Pipettes ŽClark, GC150 borosilicate glass. had resistances of 1.5 to 5 M V, and were filled with a pipette solution containing Žin mM.: CsCl 85, CsF 85, Na–HEPES 10, MgCl 2 2 and EGTA 2 ŽpH 7.2 with HCl, osmolarity 330 mOsm.. The liquid junction potential was compensated before each experiment and remained constant within 1 mV during experiments. Series resistance was compensated for 75% to 90%. Membrane currents were low-pass filtered using an 8-pole Bessel filter Žy3 dB at 20 kHz., digitized Ž12 bits; 1024 pointsrrecord. and stored on disc for off-line computer analysis. From a 1 mm diameter capillary positioned within a distance of 0.2 mm from the cell, cells were continuously superfused Ž1.5 mlrmin. with external solution containing Žin mM.: NaCl 125, KCl 5.5, CaCl 2 1.8, MgCl 2 0.8, HEPES 20, glucose 25 and sucrose 36.5 ŽpH 7.3 with NaOH, osmolarity 320 mOsm.. Using a servomotor-operated valve the cells were superfused with external solution containing ACh, PHY andror other test compounds for adjustable periods. To allow recovery from receptor desensitization, cells were washed with external solution for G 4 min between two subsequent agonist applications. Experiments were carried out at room temperature Ž22 to 258C.. Single channel currents were measured in the cell-attached mode. The cells were depolarized to an estimated value of 0 mV by using a bath solution containing Žin mM.: KCl 150, NaCl 10, HEPES 10 and MgCl 2 1 ŽpH 7.2 with KOH, osmolarity 310 mOsm.. The membrane potential was clamped at y80 mV Žpipette potential held at q80 mV., using an EPC-7 patch clamp amplifier ŽList, DarmstadtrEberstadt, Germany.. Patch pipettes had resistances of 8–18 M V, and were filled with a pipette solution containing Žin mM.: CsCl 85, KCl 85, Na–HEPES 10, MgCl 2 2 and EGTA 2 ŽpH 7.2 with HCl, osmolarity 330 mOsm. with or without ACh or PHY. Cesium was included in the pipette solution to block spontaneously active potassium channels w20x. The liquid junction potential was compensated before each experiment and remained constant within 1 mV during experiments. The cell-attached patch currents were continuously recorded starting about 20 to 30 s after formation of a GV seal, low-pass filtered using an 8-pole Bessel filter Žy3 dB at 20 kHz., digitized at 25 kHz, and stored on disc for off-line computer analysis. Experiments were carried out within a temperature range of 8.3 to 9.08C, because at room temperature the open times of nicotinic channels were too short to obtain a reliable data analysis. 2.3. Data analysis Single channel data were analyzed using a mean–variance analysis w28x. Briefly, a window of 6 to 12 sample points width was advanced one sample point for each estimate of the mean current and the variance. Mean–variance plots were converted to amplitude histograms, used to
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
265
define open and closed states and to calculate mean open times and open frequencies. The minimal event width was set at 2 to 4 sample points. For analysis, patches were selected in which G 95% of all channel openings were single channel openings. Decay of the ACh whole-cell responses was fitted using the equation y s a q b)exp Žytrt .. Relative errors of fit were F 5%. For statistical analysis the Student’s t-test was used. 2.4. Materials Acetylcholine chloride, eserine Žphysostigmine, hemisulphate, purity ) 99%. and D-tubocurarine chloride ŽD-TC. were obtained from Sigma ŽSt. Louis, MO, USA., a-bungarotoxin Ž a-BGT. and methyllycaconitine citrate ŽMLA. from Research Biochemical International ŽNatick, MA, USA. and atropine sulphate from Merck ŽDarmstadt, Germany.. The a-conotoxins ImI and MII were kindly provided by Dr. J.M. McIntosh ŽUniversity of Utah, USA., and the monoclonal antibody FK1 was kindly provided by Dr. A. Maelicke ŽJohannes-Gutenberg University, Mainz, Germany.. Concentrated stock solutions of a-BGT in external solution and of the other compounds in distilled water were stored at y208C, while FK1 in Dulbecco’s Modified Eagles Medium supplemented with 10% FCS was stored at 48C. Final dilutions in external solution were prepared immediately before the experiments. For experiments with a-conotoxins, 0.1 mgrml bovine serum albumin ŽBSA. was added to all solutions to prevent binding of the toxins to the superfusion tubes, and cells were preincubated in external solution containing the same concentration of BSA for 30–60 min.
3. Results 3.1. DiÕersity of nicotinic whole-cell currents induced by ACh Superfusion with external solution containing 1 mM ACh induced a transient inward current in somata from L. migratoria thoracic ganglion neurons, voltage clamped at y80 mV. This current was previously shown to be mediated by nAChR, as it was blocked by 10 m M D-tubocurarine ŽD-TC. w44x. The responses varied between cells with respect to the peak amplitude, time to peak and the decay rate during superfusion of ACh ŽFig. 1A.. Part of the current decay was due to desensitization, as ACh could again evoke a current with the same amplitude 4 min after washing out the agonist. Further, in some cells the decay was enhanced due to open channel block by ACh itself, indicated by the occurrence and voltage-dependence of a tail current after removal of the agonist ŽFig. 1A, second example, Fig. 4. w26x. The first example of a nicotinic current shown in Fig. 1, with an intermediate decay rate and without apparent ion channel block, was most fre-
Fig. 1. Examples of ACh and PHY responses. ŽA. Transient inward currents induced by superfusion with 1 mM ACh of thoracic ganglion neurons from L. migratoria, voltage clamped at y80 mV. The peak amplitude, time to peak and decay of the currents varied between cells. In most experiments, cells were selected with ACh-induced currents as presented in the first example. ŽB. Superfusion of locust neurons, voltage clamped at y80 mV, with 100 m M PHY resulted in marginally desensitizing inward currents. The amplitude of the currents varied between cells, while the time course was invariably slow.
quently observed in locust ganglion neurons. Cells with this type of response were selected for experiments, unless otherwise indicated. Criterion was that the relative amplitude of the non-desensitized level at the end of the 10 s superfusion with ACh ranged from 20% to 55% of the peak amplitude. 3.2. Comparison of currents induced by PHY and ACh At 100 m M, PHY induced inward currents in somata from locust thoracic ganglion neurons, voltage clamped at y80 mV ŽFig. 1B.. The amplitude of the PHY responses varied between cells, while the time course was the same, with only marginal desensitization for at least 10 min. Several aspects of the PHY-induced currents were compared with the nicotinic ACh-induced currents. The amplitudes of the ACh- and PHY-evoked currents were concentration-dependent. At low agonist concentrations the activation and decay rate of the currents were low, and increased at higher concentrations ŽFig. 2.. At G 100 m M PHY and G 1 mM ACh, a tail current appeared during removal of the agonist. The occurrence of the tail currents was highly dependent on the membrane potential ŽFigs. 2 and 4., and is thought to be due to wash out of low affinity block of open ion channels by ACh and PHY themselves. Fig. 2 shows concentration–effect curves of ACh and PHY. For each cell data were normalized to the peak amplitude at 1 mM ACh and 100 m M PHY, and scaled to 100% at the fitted maximal response. The EC 50 values were 50 " 37 m M ACh Ž n s 3. and 3 " 2 m M PHY Ž n s 3., with Hill coefficients of 0.69 " 0.03 Ž n s 3. and 0.78 " 0.18 Ž n s 3., respectively. The EC 50 values are likely to be somewhat higher then estimated due to the
266
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
Fig. 2. Concentration-dependence of ACh and PHY responses. Concentration–effect curves of ACh and PHY in thoracic ganglion neurons from L. migratoria. Data for each cell were normalized to the peak amplitude of the response at 1 mM ACh and 100 m M PHY, and scaled to 100% at the fitted maximal response. For curve fitting the data point at 1 mM PHY was not included, because the peak of the currents was reduced due to the occurrence of ion channel block. EC 50 values for current activation by ACh and PHY were 50 " 37 m M Ž n s 3. and 3 " 2 m M Ž n s 3., with Hill coefficients of 0.69 " 0.03 Ž n s 3. and 0.78 " 0.18 Ž n s 3., respectively.
absence of saturating data and the inhibition by ion channel block at higher concentrations. PHY was less effective as an agonist of locust nAChR compared to ACh. In separate experiments, ion currents were evoked by the near-maximal effective concentrations of 100 m M PHY and 1 mM ACh in the same cell. In this case, cells were not selected for the type of ACh response. The relative current amplitude induced by 100 m M PHY compared to 1 mM ACh varied between cells from 0% to 42%. Fig. 3 shows two examples with a relative PHY response of 10% and 29%, respectively, and a correlation of the PHYrACh amplitude ratio with the time constant Žt . of the ACh response decay Žcorrelation coefficient 0.77, P s 0.001.. The relative amplitude of the PHY- as well as the time constant of the ACh-induced current were independent of the absolute amplitude of the ACh response.
Fig. 3. The relative amplitude of the PHY response correlates with the time constant of decay of the ACh response. In locust thoracic ganglion neurons 100 m M PHY induced currents with amplitudes that varied between cells. The relative currents induced by 100 m M PHY compared to the peak of the response with 1 mM ACh in the same cells varied from 0% to 42%. In the examples shown PHYrACh response amplitude were 10% and 29%. However, the relative amplitude of the PHY response correlated with the time constant Žt . of the decay of the ACh response Žcorrelation coefficient 0.77, P s 0.001..
The current–voltage relationship was linear from y80 to 0 mV with ACh and PHY ŽFig. 4.. For both agonists in some cells, inward rectification was observed at positive membrane potentials. The currents induced by ACh and PHY had equal Ž P s 0.7. reversal potentials, amounting to y3.6 " 0.5 mV Ž n s 3. and y2.2 " 5.9 mV Ž n s 3., respectively, indicating activation of non-selective cation channels by both agonists. The pharmacological properties of ACh- and PHY-induced currents were investigated. Several cholinergic an-
Fig. 4. Voltage-dependence of ACh and PHY responses. IV relationships of currents induced by ACh and PHY in locust thoracic ganglion neurons, normalized to the amplitude of the response at y80 mV. Reversal potentials were y3.6"0.5 mV Ž ns 3. and y2.2"5.9 mV Ž ns 3., respectively. Note the voltage-dependence of the tail current after removal of the agonist, indicating open channel block by the agonist.
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
Fig. 5. Inhibition by MLA and ImI, but not by FK1. ŽA. The inward currents induced by superfusion with 1 mM ACh and 100 m M PHY of thoracic ganglion neurons from L. migratoria, voltage clamped at y80 mV, were inhibited by 100 nM MLA by 97"3% Ž ns 3. and 99"2% Ž ns 3., respectively. In these cells, 100 nM MLA reduced the currents by 99% ŽACh. and 96% ŽPHY.. ŽB. The a 7 nAChR-specific a-conotoxin ImI at 220 nM selectively inhibited the peak of the ACh and PHY responses. The currents were reduced by 48% and 47%, respectively. ŽC. The monoclonal antibody FK1, known for specific binding to and functional inhibition of the PHY- in contrast to the ACh-binding site of nAChR, at a 1:20 dilution, does not affect the response to 100 m M PHY.
tagonists were superfused for 4 to 8 min, followed by application of 1 mM ACh or 100 m M PHY in the continued presence of the antagonist. Fig. 5A shows that MLA completely inhibits ACh as well as PHY responses. Inward currents induced by both agonists were equally blocked by the nicotinic antagonists MLA at 100 nM Ž P ) 0.35; n s 3., D-TC at 30 m M Ž P s 0.5; n s 3. and a-BGT at 100 nM Ž P s 0.2; n s 3., while ACh responses were somewhat more sensitive to atropine than PHY responses ŽTable 1.. Block by D-TC and atropine was reversible,
267
while that by a-BGT and MLA was only slightly reversible. Further, the a 7 nAChR-specific a-conotoxin ImI w21x, at the IC 50 value of block of 220 nM, selectively inhibited the peak of ACh- and PHY-induced currents ŽFig. 5B; Table 1.. The highly variable percentage of block by ImI indicates differential relative expression of a 7-like nAChR between cells. The a 3 b 2-specific a-conotoxin MII w10x, at the completely blocking concentration of 0.1 m M, only marginally blocked ACh and PHY responses ŽTable 1.. The monoclonal antibody FK1, known for specific binding to and functional inhibition of the PHY in contrast to the ACh binding site of nAChR w35,36x, at a 1:20 dilution did not affect the currents induced by 100 m M PHY ŽFig. 5C. and 1 mM ACh Žnot shown.. The ACh and PHY responses were further compared by examining effects of co-application and cross-desensitization. Fig. 6 shows representative examples of different experimental protocols. The response to co-application of near-maximal response concentrations of 1 mM ACh and 100 m M PHY was 28 " 3% Ž n s 3. and had a faster time-course compared to the response to 1 mM ACh alone ŽFig. 6A.. A tail current appeared after removal of the agonists, indicating open channel block due to high agonist concentrations. Superfusion with 1 mM ACh for 4 min partly desensitized the receptors to a steady level. Subsequent application of 100 m M PHY in the continued presence of ACh decreased the residual current level ŽFig. 6B.. After removal of both agonists, hardly any tail current developed because of desensitization induced by ACh. When instead only PHY was removed, the current returned to the residual non-desensitized ACh response level Žnot shown.. PHY at 100 m M marginally desensitized the receptors when superfused for 4 min. Addition of ACh slowly activated an additional current, that was much smaller than the control ACh response ŽFig. 6C,D,E.. After wash-out of both agonists, a huge tail current appeared, due to removal of ion channel block ŽFig. 6C.. When in a similar experiment only ACh was removed, the tail current was absent and the current returned to the steady level of the PHY response ŽFig. 6D.. With longer application of
Table 1 Pharmacology of ACh- and PHY-induced currents Antagonist
Concentration
Percentage block of response to 1 mM ACh
Percentage block of response to 100 m M PHY
D-TC
30 m M 100 nM 100 nM 100 nM 0.5 m M 5 mM 0.22 m M 0.1 m M 1.0 m M
82 " 3% Ž n s 3. 86 " 11% Ž n s 3. 36 " 13% Ž n s 2. 97 " 3% Ž n s 3. 17 " 2% Ž n s 2. 52 " 22%a Ž n s 3. 41 " 22%a Ž n s 3. 1 "1% Ž n s 3. 0 "3% Ž n s 3.
88 " 13% Ž n s 3. 97 " 5% Ž n s 3. ND 99 " 2% Ž n s 3. y2 " 3% Ž n s 3. 21 " 3% Ž n s 2. 8 "29%a Ž n s 4. 9 "7% Ž n s 3. ND
a-BGT k-BGT MLA atropine ImI MII
Sensitivities to several cholinergic antagonists of whole-cell responses to 1 mM ACh and 100 m M PHY in L. migratoria thoracic ganglion neurons, voltage clamped at y80 mV. ND: not determined. a ImI selectively blocks the fast component of the ACh- and PHY-induced currents.
268
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
ACh and PHY, the additional current partly desensitized, showing that ACh in competition with PHY partly displaced PHY from the receptors, leading to more effective activation and subsequent desensitization ŽFig. 6E.. 3.3. Single channel currents induced by ACh and PHY
Fig. 6. Co-application and cross-desensitization. ŽA. Co-application of ACh and PHY at the near-maximal response concentrations of 1 mM and 100 m M to thoracic ganglion neurons from L. migratoria evoked a current that was smaller than the control response to 1 mM ACh. After removal of both agonists a tail current appeared, indicating reversal of open channel block. ŽB. In neurons partly desensitized by 1 mM ACh, addition of 100 m M PHY blocked the residual current level. Hardly any tail current developed after removal of agonists. ŽC,D,E. Superfusion with 100 m M PHY induced marginally desensitizing inward currents Žleft.. Addition of 1 mM ACh, after 4 min of superfusion with PHY, induced an additional current Žcenter., that was much slower and smaller than the control ACh response Žright., ŽE. and that slowly and partly desensitized. ŽC. After removal of both agonists a tail current developed. ŽD. When only ACh was removed during wash-out, no tail current appeared.
To explain the differences between the ACh- and PHYinduced whole-cell currents, single channel currents evoked by both agonists were measured in cell-attached patches of locust ganglion neurons. The resting membrane potential was estimated to be 0 mV, with cells bathing in 150 mM Kq. At y80 mV Žpipette potential q80 mV., no single channel activity or occasionally small inward currents of about 2 pA were detected in control patches ŽFig. 7A.. ACh and PHY, at their respective EC 50 concentrations of 50 m M and 3 m M, induced inward single channel currents ŽFig. 7A.. ACh-activated channels appeared to occur in bursts, while PHY-activated channels were more evenly distributed ŽFig. 7C.. Only limited filtering Žsee Section 2. was used because of the short open times of the channels. Further, a temperature of F 98C was applied to slow down the channel kinetics. Single channel analysis data are summarized in Table 2. Both with ACh and PHY three conductance levels were detected ŽFig. 7A,B.. At a membrane potential of y80 mV, the three conductance levels ŽTable 2. had equal mean current amplitudes for ACh and PHY Ž0.05 F P F 0.87; 10 patches; recording timerpatch G 12.4 s.. The conductance levels and mean reversal potentials obtained from the IV relationship were also equal for both agonists Ž P G 0.3; six to seven patches per agonist; Fig. 7B.. The single channel open probability was lower with PHY compared to ACh. PHY and ACh opened channels in 58% Ž n s 50. and 75% Ž n s 40. of the patches, respectively. In addition, with PHY in only 14% Ž n s 50. of the patches, G 2 simultaneously open channels were observed, in contrast to 68% Ž n s 40. with ACh. Further, mean open times of ACh-activated channels were significantly longer than those of the same levels activated by PHY Ž P F 0.03; 10 patches; Table 2., and also the total opening frequency was higher with ACh. With either ACh or with PHY the relative open probability of each level compared to the total open probability, as calculated by multiplying the mean open time and the opening frequency, varied between patches. The relative open probability of the low level varied between patches from 4% to 97% for AChand from 5% to 58% for PHY-induced activation. For the middle conductance level, the relative open probability varied from 1% to 24% for ACh and 2% to 38% for PHY, and for the high level from 3% to 88% for ACh and 12% to 90% for PHY. When taking also each single channel conductance into account, the relative contribution of each level to the total current showed a similar variability between patches. Therefore, the variation in channel opening between patches prohibited a distinction in relative
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
269
Fig. 7. Single channels activated by ACh and PHY. ŽA. Single channel currents were measured in cell-attached patches of thoracic ganglion neurons from L. migratoria at a holding potential of y80 mV Žpipette potential q80 mV.. In control patches, no currents or occasionally inward currents of about 2 pA were detected. ACh and PHY, at their respective EC 50 concentrations of 50 m M and 3 m M, induced inward single channel currents. Both with ACh and PHY, three conductance levels were detected. ŽB. On the left, magnification of those parts of the traces in ŽA. indicated by asterisks. On the right, IV curves of the middle Žsquares. and high Žcircles. conductance levels. The low conductance level was too small to obtain a reliable IV curve. ŽC. Traces of ACh- and PHY-induced inward single channel currents, showing that ACh activated channels with a bursting pattern, while PHY activated channels that were more evenly distributed.
Table 2 Characteristics of single channels activated by ACh and PHY ACh Current amplitude ŽpA. a Conductance level ŽpS. b Reversal potential ŽmV. b Mean open time Žms. a Opening frequency Žsy1 . c
PHY
Low
Middle
High
Low
Middle
High
4.3 " 0.2 ND ND 2.1 " 1.8 15.1
7.2 " 0.8 72 " 8 17 " 5 0.9 " 0.5 3.8
9.9 " 0.8 104 " 13 13 " 7 2.8 " 2.2 8.0
4.2 " 0.4 ND ND 0.7 " 0.4 2.8
6.7 " 0.5 67 " 9 21 " 7 0.5 " 0.1 0.9
9.1 " 1.0 99 " 9 14 " 4 1.0 " 0.6 2.7
Single channels were measured in the cell-attached mode at locust thoracic ganglion neurons, with the agonists ACh Ž50 m M. or PHY Ž3 m M. in the pipette. a Mean " S.D. was calculated from 10 patches for each agonist. The holding potential was y80 mV Žpipette potential q80 mV.. b Mean " S.D. was calculated from 6–7 patches for each agonist. To obtain the IV curves the holding potential ranged from y80 mV to q40 mV or q50 mV for ACh and PHY, respectively. c Values are obtained from a total recording time of 350 s for ACh and 433 s for PHY. ND s not determined, because the low conductance state was too small to obtain a reliable IV curve and reversal potential.
270
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
conductance level activation between ACh and PHY. In conclusion, the single channel open probabilities of AChand PHY-activated nAChR differed, but the overall characteristics of the single channels were similar for both agonists ŽTable 2..
4. Discussion The results show that both ACh and PHY act as agonists in locust thoracic ganglion neurons at the whole-cell and single channel level. Whole-cell currents induced by ACh and PHY were blocked by a range of nicotinic antagonists and had equal reversal potentials, demonstrating that PHY as well as ACh activate nAChR. The diversity in ACh-induced whole-cell currents as well as selective block of the current peak by the a-conotoxin ImI, suggest that locust neurons express distinct nAChR subtypes, with different activation and desensitization kinetics. Single channel analysis revealed that ACh and PHY each activated nicotinic single channels with three conductance levels, but with variable open probabilities. The results are discussed in terms of activation of different populations of nAChR subtypes by ACh and PHY. 4.1. ActiÕation of distinct locust neuronal nAChR subtypes by ACh ACh activated whole-cell currents in somata from L. migratoria thoracic ganglion neurons, that were mediated by nAChR, as they were blocked by a-BGT, D-TC, and MLA. Locust ganglion neurons contain a high number of nAChR w8x, which are blocked by a-BGT and atropine at IC 50 concentrations of 20 nM and 30 m M, respectively w5x. The nicotinic currents had the same appearance with respect to activation and desensitization as those in several vertebrate neuroblastoma cells w40x. The currents were not likely to be mediated by muscarinic AChR, as the voltage-dependence differed from that of muscarinic currents in locust ganglion cells ŽFig. 4; w5x.. Further, the internal solution contained fluoride-ions, which precipitate calcium and also inactivate G-proteins involved in muscarinic-receptor-mediated ion channel activation w22x. Although cells with a neuron-like appearance were selected, the ACh-induced currents of the dissociated neurons varied with respect to time to peak, desensitization and ion channel block ŽFig. 1A.. The variation did not appear to correlate with locust age, time after dissociation or with cell morphology. The a 7-specific a-conotoxin ImI selectively blocked the peak of the nicotinic currents, suggesting the expression of a 7-like nAChR as well as other nAChR subtypes in these cells. The variation in kinetics appears to be due to differential expression of distinct nAChR subtypes in independent cells, as corroborated by the variation of the percentage block induced by
ImI between cells ŽTable 1.. A similar differential expression of distinct nAChR subtypes has also been reported in a primary culture of rat hippocampal neurons w1,29,43x. Nicotinic currents activated by ACh in locust neurons were mediated by activation of single channels with three different conductance levels. In each patch all three levels were observed, but the relative contribution of the different conductances to the total current varied between patches. This suggests that locust neurons differentially express distinct insect nAChR subtypes, that are mainly distinguished by their conductances. However, the frequent transitions between all levels suggest multiple functional states of each nAChR subtype. Three subconductance states have been reported for locust and housefly nAChR w18,23x, while in cockroach neurons, two nAChR conductance levels were found to result from activation of distinct channels w4x. The observed conductances were all higher compared to those of insect nAChR reported in previous studies Žfor review see Ref. w23x.. These results could not be explained by different membrane properties at 98C, as also at room temperature three conductance levels were observed, with shorter open times than at 98C ŽVan den Beukel I., unpublished observations.. The higher conductances observed are possibly due to differences in the ionic composition. A high permeability to cesium can not explain the high conductance found, since whole-cell nicotinic responses in these cells were equal in external media with and without cesium ŽVan Kleef R.G.D.M., unpublished observations.. 4.2. Physostigmine is a nicotinic agonist with channel blocking properties Several observations demonstrate that PHY activates nAChR in locust thoracic ganglion neurons. PHY-induced currents were inhibited by the nicotinic antagonists a-BGT, D-TC and MLA, as well as by the a-conotoxin ImI ŽTable 1; Fig. 5.. Further, PHY induced whole-cell currents with equal reversal potentials, and single channel currents with the same three conductance levels as ACh, indicating that both agonists activate the same type of ion channels. The observation that PHY and ACh activated single channels with equal channel conductances, but with different mean open times, is in accordance with reports that different nicotinic agonists activate vertebrate nAChR with equal conductances w11,15x but different open times w11,27x. PHY has a higher affinity for nAChR compared to ACh, as indicated by the 17-fold lower EC 50 value of PHY. However, the efficacy of PHY to cause channel opening is much lower, as demonstrated by lower single channel open probability and smaller maximal whole-cell currents. Besides acting as a nicotinic agonist, PHY caused fast ion channel block in locust neurons. PHY was a more potent open channel blocker than ACh ŽFigs. 2 and 6A,C,D., as expected from the difference in the single
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
channel mean open times between PHY and ACh. Likewise, at mouse endplate nAChR PHY caused open channel block of ACh-evoked currents with a rate constant in the millisecond range w9x. Both agonistic activity and channel block of nAChR by PHY has previously been reported. PHY binds to the a subunit and subsequently induces ion fluxes at Torpedo nAChR w35,36x. At micromolar concentrations PHY activates single channel nicotinic currents and induces open channel block of self- or ACh-activated channels in cultured rat hippocampal neurons w29,30x and in frog muscle fibers w37x. Whole-cell currents activated by PHY have not previously been reported. In fact, in frog muscle fibers w29x, rat hippocampal neurons w29,30x, clonal rat pheochromocytoma PC12 cells w39x and at chicken a 4b 2 nAChR w31x, PHY has been reported to activate only single channel without activating whole-cell nicotinic currents. We also could not detect whole-cell responses in Xenopus oocytes expressing rat a 7 nAChR and in mouse N1E-115 and human SH-SY5Y neuroblastoma cells, although PHY induced single channel activity in both types of neuroblastoma cells ŽVan Kleef R.G.D.M. and Van den Beukel I., unpublished observations.. The block of the PHY response by competitive nicotinic antagonists and by ACh-induced desensitization, as well as the absence of effect of FK1, indicate that PHY binds at the ACh binding site of locust nAChR. This differs from a separate site at the N-terminus as proposed for PHY binding to diverse vertebrate muscle and neuronal nAChR, in view of the insensitivity of PHYactivated single channel currents to the antagonists a-BGT and MLA and block of these currents by the antibody FK1 w29,30,36x. However, single channel currents activated by PHY at fetal rat muscle nAChR expressed in Xenopus oocytes were also blocked by a-BGT w13x. 4.3. PHY and ACh differentially actiÕate locust nAChR subtypes Locust neurons differentially express multiple nAChR subtypes, as has been shown above. As whole-cell responses to ACh and PHY differ, the agonists may activate different nAChR populations, or may possess different binding affinity and subsequent activation properties for the distinct nAChR subtypes. In contrast to ACh responses, whole-cell PHY-induced currents were only marginally desensitizing. Further, the PHY response amplitude was markedly smaller compared to ACh in the same cell, and positively correlated with a slower ACh response decay ŽFig. 3.. These results might give rise to the assumption of selective activation of a non-desensitizing nAChR subpopulation by PHY. However, this was not supported by the activation of single channels by both agonists, neither by the co-application and cross-desensitization data ŽFig. 6., nor by the similar antagonist sensitivities of currents induced by both ago-
271
nists. ACh and PHY activate distinct nAChR subtypes, indicated by the variation between patches with either agonist in relative open probability of the different conductance levels. The open probabilities could not be distinguished between the agonists, suggesting activation of the same population of nAChR subtypes by ACh and PHY. The lower total open probability of PHY-activated channels explains the smaller whole-cell currents induced by PHY. Single channel data are not conclusive, since these are obtained during continued agonist exposure, leaving out fast-desensitizing nAChR subtypes. However, wholecell co-application and cross-desensitization data further support the notion that both agonists activate the same nAChR subpopulations. PHY did not activate nAChR that were insensitive to ACh ŽFig. 6A. or that were desensitized by ACh ŽFig. 6B.. Conversely, ACh did not activate an independent population of nAChR, but displaced PHY from occupied receptors, as indicated by the much slower onset and reduced amplitude of the ACh-evoked response in Fig. 6C,D,E compared to the control response. Therefore, the correlation between the PHY response and the kinetics of the ACh response as shown in Fig. 3 does not originate from selective activation of nAChR subtypes, depending on the nicotinic agonist.
5. Conclusion Taken together, the results indicate that ACh and PHY activate the same populations of nAChR subtypes in L. migratoria neurons. The different whole-cell responses appear to originate from the differential activation, desensitization and ion channel block by ACh and PHY of the distinct nAChR subtypes. The specific agonistic effects of PHY on insect nAChR compared to vertebrate nAChR encourages further examination of the specificity of the agonist binding site and subsequent ion channel activation on nAChR.
Acknowledgements We would like to thank Dr. Henk P.M. Vijverberg for critically commenting on this manuscript, Ing. Aart de Groot for software development and for expert technical assistance, Ms. Paula Martens for dissociating the locust thoracic ganglion neurons. Further, we would like to thank Dr. J.M. McIntosh ŽUniversity of Utah, USA. for kindly providing us with the a-conotoxins ImI and MII, and Dr. A. Maelicke ŽJohannes-Gutenberg University, Mainz, Germany. for kindly providing us with the antibody FK1, and the Department of Experimental Zoology at Utrecht University for supplying the locusts. This work was financially supported by the United States Environmental Protection
272
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
Agency ŽCRa82192701. and the Netherlands Alternatives to Animal Experiments Platform. References w1x M. Alkondon, E.X. Albuquerque, Diversity of nicotinic acetylcholine receptors in rat hippocampal neurons: I. Pharmacological and functional evidence for distinct structural subtypes, J. Pharmacol. Exp. Ther. 265 Ž1993. 1455–1473. w2x A.P. Alvares, Pharmacology and toxicology of carbamates, in: B. Ballantyne, T.C. Marrs ŽEds.., Clinical and Experimental Toxicology of Organophosphates and Carbamates, Butterworth–Heinemann, Oxford, 1992, pp. 40–46. w3x M. Amar, P. Thomas, S. Wonnacott, G.G. Lunt, A nicotinic acetylcholine receptor subunit from insect brain forms a non-desensitising homo-oligomeric nicotinic acetylcholine receptor when expressed in Xenopus oocytes, Neurosci. Lett. 199 Ž1995. 107–110. w4x D.J. Beadle, G. Horseman, Y. Pichon, M. Amar, T. Shimahara, Acetylcholine-activated ion channels in embryonic cockroach neurones growing in culture, J. Exp. Biol. 142 Ž1989. 337–355. w5x J.A. Benson, Electrophysiological pharmacology of the nicotinic and muscarinic cholinergic responses of isolated neuronal somata from locust thoracic ganglia, J. Exp. Biol. 170 Ž1992. 203–233. w6x D. Bertrand, M. Ballivet, M. Gomez, S. Bertrand, B. Phannavong, E.D. Gundelfinger, Physiological properties of neuronal nicotinic receptors reconstituted from the vertebrate b 2 subunit and Drosophila a subunits, Eur. J. Neurosci. 6 Ž1994. 869–875. w7x B. Bossy, M. Ballivet, P. Spierer, Conservation of neural nicotinic acetylcholine receptors from Drosophila to vertebrate central nervous systems, EMBO J. 7 Ž1988. 611–618. w8x H. Breer, D.B. Sattelle, Molecular properties and functions of insect acetylcholine receptors, J. Insect Physiol. 33 Ž1987. 771–790. w9x J. Bufler, C. Franke, H. Parnas, J. Dudel, Open channel block by physostigmine and procaine in embryonic-like nicotinic receptors of mouse muscle, Eur. J. Neurosci. 8 Ž1996. 677–687. w10x G.E. Cartier, D. Yoshikami, W.R. Gray, S. Luo, B.M. Olivera, J.M. McIntosh, A new a-conotoxin which targets a 3 b 2 nicotinic acetylcholine receptors, J. Biol. Chem. 271 Ž1996. 7522–7528. w11x D. Colquhoun, B. Sakmann, Fast events in single-channel currents activated by acetylcholine and its analogues at the frog-muscle end-plate, J. Physiol. 369 Ž1985. 501–557. w12x E. Cooper, S. Couturier, M. Ballivet, Pentameric structure and subunit stoichiometry of a neuronal nicotinic acetylcholine receptor, Nature 350 Ž1991. 235–238. w13x J.C. Cooper, O. Gutbrod, V. Witzemann, C. Methfessel, Pharmacology of the nicotinic acetylcholine receptor from fetal rat muscle expressed in Xenopus oocytes, Eur. J. Pharmacol. 309 Ž1996. 287–298. w14x A.B. Elgoyhen, D.S. Johnson, J. Boulter, D.E. Vetter, S. Heinemann, a 9: An acetylcholine receptor with novel pharmacological properties expressed in rat cochlear hair cells, Cell 79 Ž1994. 705–715. w15x P. Gardner, D.C. Ogden, D. Colquhoun, Conductances of single ion channels opened by nicotinic agonists are indistinguishable, Nature 309 Ž1984. 160–162. w16x E.D. Gundelfinger, How complex is the nicotinic receptor system of insects?, T.I.N.S. 15 Ž1992. 206–211. w17x O.P. Hamill, A. Marty, E. Neher, B. Sakmann, F.J. Sigworth, Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches, Pflueg. Arch. 391 Ž1981. 85–100. w18x W. Hanke, H. Breer H, Characterization of the channel properties of a neuronal acetylcholine receptor reconstituted into planar lipid bilayers, J. Gen. Physiol. 90 Ž1987. 855–879.
w19x B. Hermsen, R. Heiermann, A. Maelicke, Cloning and expression of ganglionic nAChR genes from Locusta migratoria, Biol. Chem. Hoppe-Seyler 372 Ž1991. 891. w20x B. Hille, Potassium channels and chloride channels, in: Ionic Channels of Excitable Membranes, Sinauer Associates, Sunderland, USA, 1992, pp. 115–139. w21x D.S. Johnson, J. Martinez, A.B. Elgoyhen, S.F. Heinemann, J.M. McIntosh, a-Conotoxin ImI exhibits subtype-specific nicotinic acetylcholine receptor blockade: preferential inhibition of homomeric a 7 and a 9 receptors, Mol. Pharmacol. 48 Ž1995. 194–199. w22x A.R. Kay, An intracellular medium formulary, J. Neurosci. Methods 44 Ž1992. 91–100. w23x C.A. Leech, D.B. Sattelle, Multiple conductances of neuronal nicotinic acetylcholine receptors, Neuropharmacology 31 Ž1992. 501– 507. w24x J. Marshall, S.D. Buckingham, R. Shingai, G.G. Lunt, M.W. Goosey, M.G. Darlison, D.B. Sattelle, E.A. Barnard, Sequence and functional expression of a single a subunit of an insect nicotinic acetylcholine receptor, EMBO J. 9 Ž1990. 4391–4398. w25x D.S. McGehee, L.W. Role, Physiological diversity of nicotinic acetylcholine receptors expressed by vertebrate neurons, Annu. Rev. Physiol. 57 Ž1995. 521–546. w26x M. Oortgiesen, H.P.M. Vijverberg, Properties of neuronal type acetylcholine receptors in voltage clamped mouse neuroblastoma cells, Neuroscience 33 Ž1989. 169–179. w27x R.L. Papke, G. Millhauser, Z. Lieberman, R.E. Oswald, Relationship of agonist properties to the single channel kinetics of nicotinic acetylcholine receptors, Biophys. J. 53 Ž1988. 1–10. w28x J.B. Patlak, Sodium channel subconductance levels measured with a new variance–mean analysis, J. Gen. Physiol. 92 Ž1988. 413–430. w29x E.F.R. Pereira, M. Alkondon, T. Tano, N.G. Castro, M.M. Froes´ Ferrao, ˜ R. Rozental, R.S. Aronstam, A. Schrattenholz, A. Maelicke, E.X. Albuquerque, A novel agonist binding site on nicotinic acetylcholine receptors, J. Rec. Res. 13 Ž1993. 413–436. w30x E.F.R. Pereira, S. Reinhardt-Maelicke, A. Schrattenholz, A. Maelicke, E.X. Albuquerque, Identification and functional characterization of a new agonist site on nicotinic acetylcholine receptors of cultured hippocampal neurons, J. Pharmacol. Exp. Ther. 265 Ž1993. 1474–1491. w31x E.F.R. Pereira, M. Alkondon, S. Reinhardt, A. Maelicke, X. Peng, J. Lindstrom, P. Whiting, E.X. Albuquerque, Physostigmine and galanthamine: probes for a novel binding site on the a 4b 2 subtype of neuronal nicotinic acetylcholine receptors stably expressed in fibroblast cells, J. Pharmacol. Exp. Ther. 270 Ž1994. 768–778. w32x J. Ramirez-Latorre, C.R. Yu, X. Qu, F. Perin, A. Karlin, L. Role, Functional contributions of a 5 subunit to neuronal acetylcholine receptor channels, Nature 380 Ž1996. 347–351. w33x P.B. Sargent, The diversity of neuronal nicotinic acetylcholine receptors, Annu. Rev. Neurosci. 16 Ž1993. 403–443. w34x P. Schloss, H. Betz, C. Schroder, E.D. Gundelfinger, Neuronal ¨ nicotinic acetylcholine receptors in Drosophila: antibodies against an a-like and a non-a subunit recognize the same high-affinity a-bungarotoxin binding complex, J. Neurochem. 57 Ž1991. 1556– 1562. w35x A. Schrattenholz, T. Coban, B. Schroder, K.O. Okonjo, J. Kuhlmann, ¨ E.F. Pereira, E.X. Albuquerque, A. Maelicke, Biochemical characterization of a novel channel activating site on nicotinic acetylcholine receptors, J. Rec. Res. 13 Ž1993. 393–412. w36x B. Schroder, S. Reinhardt-Maelicke, A. Schrattenholz, K.E. McLane, ¨ A. Kretschmer, B.M. Conti-Tronconi, A. Maelicke, Monoclonal antibodies FK1 and WF6 define two neighboring ligand binding sites on Torpedo acetylcholine receptor a-polypeptide, J. Biol. Chem. 269 Ž1994. 10407–10416. w37x K.P. Shaw, Y. Aracava, A. Akaike, J.W. Daly, D.L. Rickett, E.X. Albuquerque, The reversible cholinesterase inhibitor physostigmine
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273 has channel-blocking and agonist effects on the acetylcholine receptor–ion channel complex, Mol. Pharmacol. 28 Ž1985. 527–538. w38x S.M. Somani, S.N. Dube, Physostigmine—an overview as pretreatment drug for organophosphate intoxication, Int. J. Clin. Pharmacol., Ther. Toxicol. 27 Ž1989. 367–387. w39x A. Storch, A. Schrattenholz, J.C. Cooper, E.M.A. Ghani, O. Gutˇ ˇ ¸, brod, K.H. Weber, S. Reinhardt, C. Lobron, B. Hermsen, V. Soskic E.F.R. Pereira, E.X. Albuquerque, C. Methfessel, A. Maelicke, Physostigmine, galanthamine and codeine act as ‘noncompetitive nicotinic receptor agonists’ on clonal rat pheochromocytoma cells, Eur. J. Pharmacol. Mol. Pharmacol. 290 Ž1995. 207–219. w40x I. Van den Beukel, R.G.D.M. Van Kleef, M. Oortgiesen, Direct functional effects of parathion and paraoxon on neuronal nicotinic and muscarinic M3 acetylcholine receptors, A.T.L.A. 24 Ž1996. 191–199.
273
w41x I. Van den Beukel, R.G.D.M. Van Kleef, M. Oortgiesen, Cholinesterase inhibitors differentially affect distinct nicotinic receptor subtypes, Prog. Biophys. Mol. Biol. 65 Ž1996. 114, suppl. 1. w42x F. Wang, V. Gerzanich, G.B. Wells, R. Anand, X. Peng, K. Keyser, J. Lindstrom, Assembly of human neuronal nicotinic receptor a 5 subunits with a 3, b 2, and b 4 subunits, J. Biol. Chem. 271 Ž1996. 17656–17665. w43x C.F. Zorumski, L. Lin Thio, K.E. Isenberg, D.B. Clifford, Nicotinic acetylcholine currents in cultured postnatal rat hippocampal neurons, Mol. Pharmacol. 41 Ž1992. 931–936. w44x R. Zwart, M. Oortgiesen, H.P.M. Vijverberg, Nitromethylene heterocycles: selective agonists of nicotinic receptors in locust neurons as compared to mouse N1E-115 and BC3H1 cells, Pest. Biochem. Physiol. 48 Ž1994. 202–213.
Research report
Physostigmine and acetylcholine differentially activate nicotinic receptor subpopulations in Locusta migratoria neurons Ingeborg van den Beukel ) , Regina G.D.M. van Kleef, Ruud Zwart, Marga Oortgiesen
1
Research Institute of Toxicology, Utrecht UniÕersity, P.O. Box 80.176, NL-3508 TD Utrecht, Netherlands Accepted 23 December 1997
Abstract The effects of acetylcholine ŽACh. and physostigmine ŽPHY. on thoracic ganglion neurons of Locusta migratoria were investigated using whole-cell and cell-attached voltage clamp. ACh activated whole-cell currents with variable amplitudes, time course and ion channel block between cells, suggesting differential expression of nicotinic acetylcholine receptor ŽnAChR. subtypes. This was supported by selective block of the peak of the currents by the a 7-specific a-conotoxin ImI. PHY at 100 m M evoked smaller whole-cell currents with variable amplitudes and marginal desensitization. The PHYrACh amplitude ratio varied between cells, and was positively related to the time constant of decay of the ACh response. EC 50 values for the peak amplitude of the ACh- and PHY-induced currents were 50 m M and 3 m M, respectively. Both agonists activated nAChR, indicated by equal voltage-dependence and reversal potentials and the same pharmacological properties of ACh and PHY responses. In addition, PHY and ACh induced ion channel block. Co-application and cross-desensitization experiments showed that ACh and PHY activate the same nAChR subpopulations. Both agonists activated nicotinic single channels with three conductance levels, which were equal for ACh and PHY, indicating activation of the same nAChR subtypes by both agonists. However, for all levels PHY displayed a lower open probability than ACh. Taken together, different whole-cell responses appear to originate from differential activation, desensitization and ion channel block by ACh and PHY of distinct nAChR populations. q 1998 Elsevier Science B.V. Keywords: Physostigmine; Neuronal nicotinic acetylcholine receptor; Subtype; Insect; Whole-cell voltage clamp; Single channel
1. Introduction Nicotinic acetylcholine receptors ŽnAChR. belong to the superfamily of ligand-gated ion channels, and are activated by the endogenous neurotransmitter acetylcholine ŽACh.. Two subfamilies of vertebrate nAChR exist: endplate nAChR, found at the neuromuscular junction and on Torpedo electric organs, and neuronal nAChR, expressed in the central and peripheral nervous system w25,33x. Both types of nAChR are assumed to consist of five subunits, arranged around a central pore w12,25x. In recent years, a range of different a and b neuronal subunits have been cloned. The a subunits are characterized by two adjacent cysteines in the N-terminal extracellular domain close to the putative ACh binding site, whereas b subunits lack these cysteines. To date, a 2– a 9 and b 2– b 4 subunits )
Corresponding author. Fax: q 31-30-2535077; E-m ail: [email protected] 1 Present address: Department of Neurobiology, Duke University Medical Center, Durham, USA. 0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 0 0 7 - 9
have been cloned, which in pairwise Ž a 2– a 4 with either b 2 or b 4., single Ž a 7– a 9. or triple Ž a 5 with other a and b subunits. expression in Xenopus oocytes yields functional nAChR w14,25,32,42x. Taking this range of subunits into account, a great variety of functional expression of subunit combinations is possible, and up to now the exact composition of nAChR in vivo is unclear w25x. Different native nAChR subtypes as well as different subtypes expressed in oocytes can be distinguished in terms of the potency order of different nicotinic agonists, the antagonist sensitivity, single channel conductances as well as activation and desensitization kinetics Žfor review, see Refs. w25,33x.. In contrast to vertebrates, nAChR in insects are confined to the central nervous system, where they are abundantly expressed in ganglia w8x. Like in vertebrates, insect nAChR consist of a subunits, containing the two cysteines characteristic of a subunits, and b subunits w16x. Various insect nAChR subunits have been cloned from different species, as for example four different a and two b subunits of Locusta migratoria ganglia w19x. Homology
264
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
between insect nAChR a subunits appears to be higher than that between insect and vertebrate a subunits w24x, while insect nAChR a subunits share a somewhat higher homology with vertebrate neuronal than with vertebrate endplate a subunits w7,16,24x. Heteromeric w16,34x as well as homomeric w8x insect nAChR have been reported in situ, while in Xenopus oocytes functional nAChR also result from coexpression w6x or single expression w3,24x of nAChR subunits. Like in vertebrates, insect nAChR subtypes have been found to differ in their pharmacological profiles w8,16x. The carbamate physostigmine ŽPHY., isolated from Calabar beans, has been used successfully in the treatment of glaucoma since 1864 w38x. Carbamates, widely applied as insecticides, are generally known to inhibit the enzyme acetylcholinesterase, thereby augmenting cholinergic transmission w2x. At higher concentrations, PHY blocks endplate and neuronal nAChR w29,30x. In addition, PHY has been reported to be an agonist at vertebrate endplate and neuronal and at Torpedo nAChR, activating single channel currents at a nAChR site different from the ACh binding site w29,30,36x. The blocking potency of PHY varies between nAChR from different species, while whole-cell agonistic actions of PHY have been observed only in locust thoracic ganglion neurons w41x. Therefore, PHY may be able to distinguish between nAChR of different species. In the present study, the agonistic properties of PHY on nAChR in thoracic ganglion neurons from L. migratoria are investigated in detail, and compared with those of ACh. Further, the ability of PHY to distinguish between nAChR subtypes within locust neurons is examined.
2. Materials and methods 2.1. Cells Locust neurons were acutely dissociated as described elsewhere w44x. Briefly, African locusts, L. migratoria, were used to prepare freshly dissociated thoracic ganglion neurons, 3–6 days after imaginal ecdysis. All three thoracic ganglia were dissected from the locusts and collected in ice-cold external solution Žfor composition see below.. Somata of the ganglion neurons were mechanically dissociated by aspirating the ganglia through a Pasteur pipette. Dissociated neurons were plated on tissue culture dishes coated with poly-L-lysine, stored at 48C, and used within 2 days. For experiments, cells were selected with a typical neuron-like appearance, i.e., with remains of axons and dendrites. 2.2. Electrophysiology Whole-cell membrane currents were measured using the standard whole-cell voltage clamp technique w17x. The membrane potential was held at y80 mV, using a RK 300
amplifier ŽBiologic, France.. Pipettes ŽClark, GC150 borosilicate glass. had resistances of 1.5 to 5 M V, and were filled with a pipette solution containing Žin mM.: CsCl 85, CsF 85, Na–HEPES 10, MgCl 2 2 and EGTA 2 ŽpH 7.2 with HCl, osmolarity 330 mOsm.. The liquid junction potential was compensated before each experiment and remained constant within 1 mV during experiments. Series resistance was compensated for 75% to 90%. Membrane currents were low-pass filtered using an 8-pole Bessel filter Žy3 dB at 20 kHz., digitized Ž12 bits; 1024 pointsrrecord. and stored on disc for off-line computer analysis. From a 1 mm diameter capillary positioned within a distance of 0.2 mm from the cell, cells were continuously superfused Ž1.5 mlrmin. with external solution containing Žin mM.: NaCl 125, KCl 5.5, CaCl 2 1.8, MgCl 2 0.8, HEPES 20, glucose 25 and sucrose 36.5 ŽpH 7.3 with NaOH, osmolarity 320 mOsm.. Using a servomotor-operated valve the cells were superfused with external solution containing ACh, PHY andror other test compounds for adjustable periods. To allow recovery from receptor desensitization, cells were washed with external solution for G 4 min between two subsequent agonist applications. Experiments were carried out at room temperature Ž22 to 258C.. Single channel currents were measured in the cell-attached mode. The cells were depolarized to an estimated value of 0 mV by using a bath solution containing Žin mM.: KCl 150, NaCl 10, HEPES 10 and MgCl 2 1 ŽpH 7.2 with KOH, osmolarity 310 mOsm.. The membrane potential was clamped at y80 mV Žpipette potential held at q80 mV., using an EPC-7 patch clamp amplifier ŽList, DarmstadtrEberstadt, Germany.. Patch pipettes had resistances of 8–18 M V, and were filled with a pipette solution containing Žin mM.: CsCl 85, KCl 85, Na–HEPES 10, MgCl 2 2 and EGTA 2 ŽpH 7.2 with HCl, osmolarity 330 mOsm. with or without ACh or PHY. Cesium was included in the pipette solution to block spontaneously active potassium channels w20x. The liquid junction potential was compensated before each experiment and remained constant within 1 mV during experiments. The cell-attached patch currents were continuously recorded starting about 20 to 30 s after formation of a GV seal, low-pass filtered using an 8-pole Bessel filter Žy3 dB at 20 kHz., digitized at 25 kHz, and stored on disc for off-line computer analysis. Experiments were carried out within a temperature range of 8.3 to 9.08C, because at room temperature the open times of nicotinic channels were too short to obtain a reliable data analysis. 2.3. Data analysis Single channel data were analyzed using a mean–variance analysis w28x. Briefly, a window of 6 to 12 sample points width was advanced one sample point for each estimate of the mean current and the variance. Mean–variance plots were converted to amplitude histograms, used to
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
265
define open and closed states and to calculate mean open times and open frequencies. The minimal event width was set at 2 to 4 sample points. For analysis, patches were selected in which G 95% of all channel openings were single channel openings. Decay of the ACh whole-cell responses was fitted using the equation y s a q b)exp Žytrt .. Relative errors of fit were F 5%. For statistical analysis the Student’s t-test was used. 2.4. Materials Acetylcholine chloride, eserine Žphysostigmine, hemisulphate, purity ) 99%. and D-tubocurarine chloride ŽD-TC. were obtained from Sigma ŽSt. Louis, MO, USA., a-bungarotoxin Ž a-BGT. and methyllycaconitine citrate ŽMLA. from Research Biochemical International ŽNatick, MA, USA. and atropine sulphate from Merck ŽDarmstadt, Germany.. The a-conotoxins ImI and MII were kindly provided by Dr. J.M. McIntosh ŽUniversity of Utah, USA., and the monoclonal antibody FK1 was kindly provided by Dr. A. Maelicke ŽJohannes-Gutenberg University, Mainz, Germany.. Concentrated stock solutions of a-BGT in external solution and of the other compounds in distilled water were stored at y208C, while FK1 in Dulbecco’s Modified Eagles Medium supplemented with 10% FCS was stored at 48C. Final dilutions in external solution were prepared immediately before the experiments. For experiments with a-conotoxins, 0.1 mgrml bovine serum albumin ŽBSA. was added to all solutions to prevent binding of the toxins to the superfusion tubes, and cells were preincubated in external solution containing the same concentration of BSA for 30–60 min.
3. Results 3.1. DiÕersity of nicotinic whole-cell currents induced by ACh Superfusion with external solution containing 1 mM ACh induced a transient inward current in somata from L. migratoria thoracic ganglion neurons, voltage clamped at y80 mV. This current was previously shown to be mediated by nAChR, as it was blocked by 10 m M D-tubocurarine ŽD-TC. w44x. The responses varied between cells with respect to the peak amplitude, time to peak and the decay rate during superfusion of ACh ŽFig. 1A.. Part of the current decay was due to desensitization, as ACh could again evoke a current with the same amplitude 4 min after washing out the agonist. Further, in some cells the decay was enhanced due to open channel block by ACh itself, indicated by the occurrence and voltage-dependence of a tail current after removal of the agonist ŽFig. 1A, second example, Fig. 4. w26x. The first example of a nicotinic current shown in Fig. 1, with an intermediate decay rate and without apparent ion channel block, was most fre-
Fig. 1. Examples of ACh and PHY responses. ŽA. Transient inward currents induced by superfusion with 1 mM ACh of thoracic ganglion neurons from L. migratoria, voltage clamped at y80 mV. The peak amplitude, time to peak and decay of the currents varied between cells. In most experiments, cells were selected with ACh-induced currents as presented in the first example. ŽB. Superfusion of locust neurons, voltage clamped at y80 mV, with 100 m M PHY resulted in marginally desensitizing inward currents. The amplitude of the currents varied between cells, while the time course was invariably slow.
quently observed in locust ganglion neurons. Cells with this type of response were selected for experiments, unless otherwise indicated. Criterion was that the relative amplitude of the non-desensitized level at the end of the 10 s superfusion with ACh ranged from 20% to 55% of the peak amplitude. 3.2. Comparison of currents induced by PHY and ACh At 100 m M, PHY induced inward currents in somata from locust thoracic ganglion neurons, voltage clamped at y80 mV ŽFig. 1B.. The amplitude of the PHY responses varied between cells, while the time course was the same, with only marginal desensitization for at least 10 min. Several aspects of the PHY-induced currents were compared with the nicotinic ACh-induced currents. The amplitudes of the ACh- and PHY-evoked currents were concentration-dependent. At low agonist concentrations the activation and decay rate of the currents were low, and increased at higher concentrations ŽFig. 2.. At G 100 m M PHY and G 1 mM ACh, a tail current appeared during removal of the agonist. The occurrence of the tail currents was highly dependent on the membrane potential ŽFigs. 2 and 4., and is thought to be due to wash out of low affinity block of open ion channels by ACh and PHY themselves. Fig. 2 shows concentration–effect curves of ACh and PHY. For each cell data were normalized to the peak amplitude at 1 mM ACh and 100 m M PHY, and scaled to 100% at the fitted maximal response. The EC 50 values were 50 " 37 m M ACh Ž n s 3. and 3 " 2 m M PHY Ž n s 3., with Hill coefficients of 0.69 " 0.03 Ž n s 3. and 0.78 " 0.18 Ž n s 3., respectively. The EC 50 values are likely to be somewhat higher then estimated due to the
266
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
Fig. 2. Concentration-dependence of ACh and PHY responses. Concentration–effect curves of ACh and PHY in thoracic ganglion neurons from L. migratoria. Data for each cell were normalized to the peak amplitude of the response at 1 mM ACh and 100 m M PHY, and scaled to 100% at the fitted maximal response. For curve fitting the data point at 1 mM PHY was not included, because the peak of the currents was reduced due to the occurrence of ion channel block. EC 50 values for current activation by ACh and PHY were 50 " 37 m M Ž n s 3. and 3 " 2 m M Ž n s 3., with Hill coefficients of 0.69 " 0.03 Ž n s 3. and 0.78 " 0.18 Ž n s 3., respectively.
absence of saturating data and the inhibition by ion channel block at higher concentrations. PHY was less effective as an agonist of locust nAChR compared to ACh. In separate experiments, ion currents were evoked by the near-maximal effective concentrations of 100 m M PHY and 1 mM ACh in the same cell. In this case, cells were not selected for the type of ACh response. The relative current amplitude induced by 100 m M PHY compared to 1 mM ACh varied between cells from 0% to 42%. Fig. 3 shows two examples with a relative PHY response of 10% and 29%, respectively, and a correlation of the PHYrACh amplitude ratio with the time constant Žt . of the ACh response decay Žcorrelation coefficient 0.77, P s 0.001.. The relative amplitude of the PHY- as well as the time constant of the ACh-induced current were independent of the absolute amplitude of the ACh response.
Fig. 3. The relative amplitude of the PHY response correlates with the time constant of decay of the ACh response. In locust thoracic ganglion neurons 100 m M PHY induced currents with amplitudes that varied between cells. The relative currents induced by 100 m M PHY compared to the peak of the response with 1 mM ACh in the same cells varied from 0% to 42%. In the examples shown PHYrACh response amplitude were 10% and 29%. However, the relative amplitude of the PHY response correlated with the time constant Žt . of the decay of the ACh response Žcorrelation coefficient 0.77, P s 0.001..
The current–voltage relationship was linear from y80 to 0 mV with ACh and PHY ŽFig. 4.. For both agonists in some cells, inward rectification was observed at positive membrane potentials. The currents induced by ACh and PHY had equal Ž P s 0.7. reversal potentials, amounting to y3.6 " 0.5 mV Ž n s 3. and y2.2 " 5.9 mV Ž n s 3., respectively, indicating activation of non-selective cation channels by both agonists. The pharmacological properties of ACh- and PHY-induced currents were investigated. Several cholinergic an-
Fig. 4. Voltage-dependence of ACh and PHY responses. IV relationships of currents induced by ACh and PHY in locust thoracic ganglion neurons, normalized to the amplitude of the response at y80 mV. Reversal potentials were y3.6"0.5 mV Ž ns 3. and y2.2"5.9 mV Ž ns 3., respectively. Note the voltage-dependence of the tail current after removal of the agonist, indicating open channel block by the agonist.
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
Fig. 5. Inhibition by MLA and ImI, but not by FK1. ŽA. The inward currents induced by superfusion with 1 mM ACh and 100 m M PHY of thoracic ganglion neurons from L. migratoria, voltage clamped at y80 mV, were inhibited by 100 nM MLA by 97"3% Ž ns 3. and 99"2% Ž ns 3., respectively. In these cells, 100 nM MLA reduced the currents by 99% ŽACh. and 96% ŽPHY.. ŽB. The a 7 nAChR-specific a-conotoxin ImI at 220 nM selectively inhibited the peak of the ACh and PHY responses. The currents were reduced by 48% and 47%, respectively. ŽC. The monoclonal antibody FK1, known for specific binding to and functional inhibition of the PHY- in contrast to the ACh-binding site of nAChR, at a 1:20 dilution, does not affect the response to 100 m M PHY.
tagonists were superfused for 4 to 8 min, followed by application of 1 mM ACh or 100 m M PHY in the continued presence of the antagonist. Fig. 5A shows that MLA completely inhibits ACh as well as PHY responses. Inward currents induced by both agonists were equally blocked by the nicotinic antagonists MLA at 100 nM Ž P ) 0.35; n s 3., D-TC at 30 m M Ž P s 0.5; n s 3. and a-BGT at 100 nM Ž P s 0.2; n s 3., while ACh responses were somewhat more sensitive to atropine than PHY responses ŽTable 1.. Block by D-TC and atropine was reversible,
267
while that by a-BGT and MLA was only slightly reversible. Further, the a 7 nAChR-specific a-conotoxin ImI w21x, at the IC 50 value of block of 220 nM, selectively inhibited the peak of ACh- and PHY-induced currents ŽFig. 5B; Table 1.. The highly variable percentage of block by ImI indicates differential relative expression of a 7-like nAChR between cells. The a 3 b 2-specific a-conotoxin MII w10x, at the completely blocking concentration of 0.1 m M, only marginally blocked ACh and PHY responses ŽTable 1.. The monoclonal antibody FK1, known for specific binding to and functional inhibition of the PHY in contrast to the ACh binding site of nAChR w35,36x, at a 1:20 dilution did not affect the currents induced by 100 m M PHY ŽFig. 5C. and 1 mM ACh Žnot shown.. The ACh and PHY responses were further compared by examining effects of co-application and cross-desensitization. Fig. 6 shows representative examples of different experimental protocols. The response to co-application of near-maximal response concentrations of 1 mM ACh and 100 m M PHY was 28 " 3% Ž n s 3. and had a faster time-course compared to the response to 1 mM ACh alone ŽFig. 6A.. A tail current appeared after removal of the agonists, indicating open channel block due to high agonist concentrations. Superfusion with 1 mM ACh for 4 min partly desensitized the receptors to a steady level. Subsequent application of 100 m M PHY in the continued presence of ACh decreased the residual current level ŽFig. 6B.. After removal of both agonists, hardly any tail current developed because of desensitization induced by ACh. When instead only PHY was removed, the current returned to the residual non-desensitized ACh response level Žnot shown.. PHY at 100 m M marginally desensitized the receptors when superfused for 4 min. Addition of ACh slowly activated an additional current, that was much smaller than the control ACh response ŽFig. 6C,D,E.. After wash-out of both agonists, a huge tail current appeared, due to removal of ion channel block ŽFig. 6C.. When in a similar experiment only ACh was removed, the tail current was absent and the current returned to the steady level of the PHY response ŽFig. 6D.. With longer application of
Table 1 Pharmacology of ACh- and PHY-induced currents Antagonist
Concentration
Percentage block of response to 1 mM ACh
Percentage block of response to 100 m M PHY
D-TC
30 m M 100 nM 100 nM 100 nM 0.5 m M 5 mM 0.22 m M 0.1 m M 1.0 m M
82 " 3% Ž n s 3. 86 " 11% Ž n s 3. 36 " 13% Ž n s 2. 97 " 3% Ž n s 3. 17 " 2% Ž n s 2. 52 " 22%a Ž n s 3. 41 " 22%a Ž n s 3. 1 "1% Ž n s 3. 0 "3% Ž n s 3.
88 " 13% Ž n s 3. 97 " 5% Ž n s 3. ND 99 " 2% Ž n s 3. y2 " 3% Ž n s 3. 21 " 3% Ž n s 2. 8 "29%a Ž n s 4. 9 "7% Ž n s 3. ND
a-BGT k-BGT MLA atropine ImI MII
Sensitivities to several cholinergic antagonists of whole-cell responses to 1 mM ACh and 100 m M PHY in L. migratoria thoracic ganglion neurons, voltage clamped at y80 mV. ND: not determined. a ImI selectively blocks the fast component of the ACh- and PHY-induced currents.
268
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
ACh and PHY, the additional current partly desensitized, showing that ACh in competition with PHY partly displaced PHY from the receptors, leading to more effective activation and subsequent desensitization ŽFig. 6E.. 3.3. Single channel currents induced by ACh and PHY
Fig. 6. Co-application and cross-desensitization. ŽA. Co-application of ACh and PHY at the near-maximal response concentrations of 1 mM and 100 m M to thoracic ganglion neurons from L. migratoria evoked a current that was smaller than the control response to 1 mM ACh. After removal of both agonists a tail current appeared, indicating reversal of open channel block. ŽB. In neurons partly desensitized by 1 mM ACh, addition of 100 m M PHY blocked the residual current level. Hardly any tail current developed after removal of agonists. ŽC,D,E. Superfusion with 100 m M PHY induced marginally desensitizing inward currents Žleft.. Addition of 1 mM ACh, after 4 min of superfusion with PHY, induced an additional current Žcenter., that was much slower and smaller than the control ACh response Žright., ŽE. and that slowly and partly desensitized. ŽC. After removal of both agonists a tail current developed. ŽD. When only ACh was removed during wash-out, no tail current appeared.
To explain the differences between the ACh- and PHYinduced whole-cell currents, single channel currents evoked by both agonists were measured in cell-attached patches of locust ganglion neurons. The resting membrane potential was estimated to be 0 mV, with cells bathing in 150 mM Kq. At y80 mV Žpipette potential q80 mV., no single channel activity or occasionally small inward currents of about 2 pA were detected in control patches ŽFig. 7A.. ACh and PHY, at their respective EC 50 concentrations of 50 m M and 3 m M, induced inward single channel currents ŽFig. 7A.. ACh-activated channels appeared to occur in bursts, while PHY-activated channels were more evenly distributed ŽFig. 7C.. Only limited filtering Žsee Section 2. was used because of the short open times of the channels. Further, a temperature of F 98C was applied to slow down the channel kinetics. Single channel analysis data are summarized in Table 2. Both with ACh and PHY three conductance levels were detected ŽFig. 7A,B.. At a membrane potential of y80 mV, the three conductance levels ŽTable 2. had equal mean current amplitudes for ACh and PHY Ž0.05 F P F 0.87; 10 patches; recording timerpatch G 12.4 s.. The conductance levels and mean reversal potentials obtained from the IV relationship were also equal for both agonists Ž P G 0.3; six to seven patches per agonist; Fig. 7B.. The single channel open probability was lower with PHY compared to ACh. PHY and ACh opened channels in 58% Ž n s 50. and 75% Ž n s 40. of the patches, respectively. In addition, with PHY in only 14% Ž n s 50. of the patches, G 2 simultaneously open channels were observed, in contrast to 68% Ž n s 40. with ACh. Further, mean open times of ACh-activated channels were significantly longer than those of the same levels activated by PHY Ž P F 0.03; 10 patches; Table 2., and also the total opening frequency was higher with ACh. With either ACh or with PHY the relative open probability of each level compared to the total open probability, as calculated by multiplying the mean open time and the opening frequency, varied between patches. The relative open probability of the low level varied between patches from 4% to 97% for AChand from 5% to 58% for PHY-induced activation. For the middle conductance level, the relative open probability varied from 1% to 24% for ACh and 2% to 38% for PHY, and for the high level from 3% to 88% for ACh and 12% to 90% for PHY. When taking also each single channel conductance into account, the relative contribution of each level to the total current showed a similar variability between patches. Therefore, the variation in channel opening between patches prohibited a distinction in relative
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
269
Fig. 7. Single channels activated by ACh and PHY. ŽA. Single channel currents were measured in cell-attached patches of thoracic ganglion neurons from L. migratoria at a holding potential of y80 mV Žpipette potential q80 mV.. In control patches, no currents or occasionally inward currents of about 2 pA were detected. ACh and PHY, at their respective EC 50 concentrations of 50 m M and 3 m M, induced inward single channel currents. Both with ACh and PHY, three conductance levels were detected. ŽB. On the left, magnification of those parts of the traces in ŽA. indicated by asterisks. On the right, IV curves of the middle Žsquares. and high Žcircles. conductance levels. The low conductance level was too small to obtain a reliable IV curve. ŽC. Traces of ACh- and PHY-induced inward single channel currents, showing that ACh activated channels with a bursting pattern, while PHY activated channels that were more evenly distributed.
Table 2 Characteristics of single channels activated by ACh and PHY ACh Current amplitude ŽpA. a Conductance level ŽpS. b Reversal potential ŽmV. b Mean open time Žms. a Opening frequency Žsy1 . c
PHY
Low
Middle
High
Low
Middle
High
4.3 " 0.2 ND ND 2.1 " 1.8 15.1
7.2 " 0.8 72 " 8 17 " 5 0.9 " 0.5 3.8
9.9 " 0.8 104 " 13 13 " 7 2.8 " 2.2 8.0
4.2 " 0.4 ND ND 0.7 " 0.4 2.8
6.7 " 0.5 67 " 9 21 " 7 0.5 " 0.1 0.9
9.1 " 1.0 99 " 9 14 " 4 1.0 " 0.6 2.7
Single channels were measured in the cell-attached mode at locust thoracic ganglion neurons, with the agonists ACh Ž50 m M. or PHY Ž3 m M. in the pipette. a Mean " S.D. was calculated from 10 patches for each agonist. The holding potential was y80 mV Žpipette potential q80 mV.. b Mean " S.D. was calculated from 6–7 patches for each agonist. To obtain the IV curves the holding potential ranged from y80 mV to q40 mV or q50 mV for ACh and PHY, respectively. c Values are obtained from a total recording time of 350 s for ACh and 433 s for PHY. ND s not determined, because the low conductance state was too small to obtain a reliable IV curve and reversal potential.
270
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
conductance level activation between ACh and PHY. In conclusion, the single channel open probabilities of AChand PHY-activated nAChR differed, but the overall characteristics of the single channels were similar for both agonists ŽTable 2..
4. Discussion The results show that both ACh and PHY act as agonists in locust thoracic ganglion neurons at the whole-cell and single channel level. Whole-cell currents induced by ACh and PHY were blocked by a range of nicotinic antagonists and had equal reversal potentials, demonstrating that PHY as well as ACh activate nAChR. The diversity in ACh-induced whole-cell currents as well as selective block of the current peak by the a-conotoxin ImI, suggest that locust neurons express distinct nAChR subtypes, with different activation and desensitization kinetics. Single channel analysis revealed that ACh and PHY each activated nicotinic single channels with three conductance levels, but with variable open probabilities. The results are discussed in terms of activation of different populations of nAChR subtypes by ACh and PHY. 4.1. ActiÕation of distinct locust neuronal nAChR subtypes by ACh ACh activated whole-cell currents in somata from L. migratoria thoracic ganglion neurons, that were mediated by nAChR, as they were blocked by a-BGT, D-TC, and MLA. Locust ganglion neurons contain a high number of nAChR w8x, which are blocked by a-BGT and atropine at IC 50 concentrations of 20 nM and 30 m M, respectively w5x. The nicotinic currents had the same appearance with respect to activation and desensitization as those in several vertebrate neuroblastoma cells w40x. The currents were not likely to be mediated by muscarinic AChR, as the voltage-dependence differed from that of muscarinic currents in locust ganglion cells ŽFig. 4; w5x.. Further, the internal solution contained fluoride-ions, which precipitate calcium and also inactivate G-proteins involved in muscarinic-receptor-mediated ion channel activation w22x. Although cells with a neuron-like appearance were selected, the ACh-induced currents of the dissociated neurons varied with respect to time to peak, desensitization and ion channel block ŽFig. 1A.. The variation did not appear to correlate with locust age, time after dissociation or with cell morphology. The a 7-specific a-conotoxin ImI selectively blocked the peak of the nicotinic currents, suggesting the expression of a 7-like nAChR as well as other nAChR subtypes in these cells. The variation in kinetics appears to be due to differential expression of distinct nAChR subtypes in independent cells, as corroborated by the variation of the percentage block induced by
ImI between cells ŽTable 1.. A similar differential expression of distinct nAChR subtypes has also been reported in a primary culture of rat hippocampal neurons w1,29,43x. Nicotinic currents activated by ACh in locust neurons were mediated by activation of single channels with three different conductance levels. In each patch all three levels were observed, but the relative contribution of the different conductances to the total current varied between patches. This suggests that locust neurons differentially express distinct insect nAChR subtypes, that are mainly distinguished by their conductances. However, the frequent transitions between all levels suggest multiple functional states of each nAChR subtype. Three subconductance states have been reported for locust and housefly nAChR w18,23x, while in cockroach neurons, two nAChR conductance levels were found to result from activation of distinct channels w4x. The observed conductances were all higher compared to those of insect nAChR reported in previous studies Žfor review see Ref. w23x.. These results could not be explained by different membrane properties at 98C, as also at room temperature three conductance levels were observed, with shorter open times than at 98C ŽVan den Beukel I., unpublished observations.. The higher conductances observed are possibly due to differences in the ionic composition. A high permeability to cesium can not explain the high conductance found, since whole-cell nicotinic responses in these cells were equal in external media with and without cesium ŽVan Kleef R.G.D.M., unpublished observations.. 4.2. Physostigmine is a nicotinic agonist with channel blocking properties Several observations demonstrate that PHY activates nAChR in locust thoracic ganglion neurons. PHY-induced currents were inhibited by the nicotinic antagonists a-BGT, D-TC and MLA, as well as by the a-conotoxin ImI ŽTable 1; Fig. 5.. Further, PHY induced whole-cell currents with equal reversal potentials, and single channel currents with the same three conductance levels as ACh, indicating that both agonists activate the same type of ion channels. The observation that PHY and ACh activated single channels with equal channel conductances, but with different mean open times, is in accordance with reports that different nicotinic agonists activate vertebrate nAChR with equal conductances w11,15x but different open times w11,27x. PHY has a higher affinity for nAChR compared to ACh, as indicated by the 17-fold lower EC 50 value of PHY. However, the efficacy of PHY to cause channel opening is much lower, as demonstrated by lower single channel open probability and smaller maximal whole-cell currents. Besides acting as a nicotinic agonist, PHY caused fast ion channel block in locust neurons. PHY was a more potent open channel blocker than ACh ŽFigs. 2 and 6A,C,D., as expected from the difference in the single
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
channel mean open times between PHY and ACh. Likewise, at mouse endplate nAChR PHY caused open channel block of ACh-evoked currents with a rate constant in the millisecond range w9x. Both agonistic activity and channel block of nAChR by PHY has previously been reported. PHY binds to the a subunit and subsequently induces ion fluxes at Torpedo nAChR w35,36x. At micromolar concentrations PHY activates single channel nicotinic currents and induces open channel block of self- or ACh-activated channels in cultured rat hippocampal neurons w29,30x and in frog muscle fibers w37x. Whole-cell currents activated by PHY have not previously been reported. In fact, in frog muscle fibers w29x, rat hippocampal neurons w29,30x, clonal rat pheochromocytoma PC12 cells w39x and at chicken a 4b 2 nAChR w31x, PHY has been reported to activate only single channel without activating whole-cell nicotinic currents. We also could not detect whole-cell responses in Xenopus oocytes expressing rat a 7 nAChR and in mouse N1E-115 and human SH-SY5Y neuroblastoma cells, although PHY induced single channel activity in both types of neuroblastoma cells ŽVan Kleef R.G.D.M. and Van den Beukel I., unpublished observations.. The block of the PHY response by competitive nicotinic antagonists and by ACh-induced desensitization, as well as the absence of effect of FK1, indicate that PHY binds at the ACh binding site of locust nAChR. This differs from a separate site at the N-terminus as proposed for PHY binding to diverse vertebrate muscle and neuronal nAChR, in view of the insensitivity of PHYactivated single channel currents to the antagonists a-BGT and MLA and block of these currents by the antibody FK1 w29,30,36x. However, single channel currents activated by PHY at fetal rat muscle nAChR expressed in Xenopus oocytes were also blocked by a-BGT w13x. 4.3. PHY and ACh differentially actiÕate locust nAChR subtypes Locust neurons differentially express multiple nAChR subtypes, as has been shown above. As whole-cell responses to ACh and PHY differ, the agonists may activate different nAChR populations, or may possess different binding affinity and subsequent activation properties for the distinct nAChR subtypes. In contrast to ACh responses, whole-cell PHY-induced currents were only marginally desensitizing. Further, the PHY response amplitude was markedly smaller compared to ACh in the same cell, and positively correlated with a slower ACh response decay ŽFig. 3.. These results might give rise to the assumption of selective activation of a non-desensitizing nAChR subpopulation by PHY. However, this was not supported by the activation of single channels by both agonists, neither by the co-application and cross-desensitization data ŽFig. 6., nor by the similar antagonist sensitivities of currents induced by both ago-
271
nists. ACh and PHY activate distinct nAChR subtypes, indicated by the variation between patches with either agonist in relative open probability of the different conductance levels. The open probabilities could not be distinguished between the agonists, suggesting activation of the same population of nAChR subtypes by ACh and PHY. The lower total open probability of PHY-activated channels explains the smaller whole-cell currents induced by PHY. Single channel data are not conclusive, since these are obtained during continued agonist exposure, leaving out fast-desensitizing nAChR subtypes. However, wholecell co-application and cross-desensitization data further support the notion that both agonists activate the same nAChR subpopulations. PHY did not activate nAChR that were insensitive to ACh ŽFig. 6A. or that were desensitized by ACh ŽFig. 6B.. Conversely, ACh did not activate an independent population of nAChR, but displaced PHY from occupied receptors, as indicated by the much slower onset and reduced amplitude of the ACh-evoked response in Fig. 6C,D,E compared to the control response. Therefore, the correlation between the PHY response and the kinetics of the ACh response as shown in Fig. 3 does not originate from selective activation of nAChR subtypes, depending on the nicotinic agonist.
5. Conclusion Taken together, the results indicate that ACh and PHY activate the same populations of nAChR subtypes in L. migratoria neurons. The different whole-cell responses appear to originate from the differential activation, desensitization and ion channel block by ACh and PHY of the distinct nAChR subtypes. The specific agonistic effects of PHY on insect nAChR compared to vertebrate nAChR encourages further examination of the specificity of the agonist binding site and subsequent ion channel activation on nAChR.
Acknowledgements We would like to thank Dr. Henk P.M. Vijverberg for critically commenting on this manuscript, Ing. Aart de Groot for software development and for expert technical assistance, Ms. Paula Martens for dissociating the locust thoracic ganglion neurons. Further, we would like to thank Dr. J.M. McIntosh ŽUniversity of Utah, USA. for kindly providing us with the a-conotoxins ImI and MII, and Dr. A. Maelicke ŽJohannes-Gutenberg University, Mainz, Germany. for kindly providing us with the antibody FK1, and the Department of Experimental Zoology at Utrecht University for supplying the locusts. This work was financially supported by the United States Environmental Protection
272
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273
Agency ŽCRa82192701. and the Netherlands Alternatives to Animal Experiments Platform. References w1x M. Alkondon, E.X. Albuquerque, Diversity of nicotinic acetylcholine receptors in rat hippocampal neurons: I. Pharmacological and functional evidence for distinct structural subtypes, J. Pharmacol. Exp. Ther. 265 Ž1993. 1455–1473. w2x A.P. Alvares, Pharmacology and toxicology of carbamates, in: B. Ballantyne, T.C. Marrs ŽEds.., Clinical and Experimental Toxicology of Organophosphates and Carbamates, Butterworth–Heinemann, Oxford, 1992, pp. 40–46. w3x M. Amar, P. Thomas, S. Wonnacott, G.G. Lunt, A nicotinic acetylcholine receptor subunit from insect brain forms a non-desensitising homo-oligomeric nicotinic acetylcholine receptor when expressed in Xenopus oocytes, Neurosci. Lett. 199 Ž1995. 107–110. w4x D.J. Beadle, G. Horseman, Y. Pichon, M. Amar, T. Shimahara, Acetylcholine-activated ion channels in embryonic cockroach neurones growing in culture, J. Exp. Biol. 142 Ž1989. 337–355. w5x J.A. Benson, Electrophysiological pharmacology of the nicotinic and muscarinic cholinergic responses of isolated neuronal somata from locust thoracic ganglia, J. Exp. Biol. 170 Ž1992. 203–233. w6x D. Bertrand, M. Ballivet, M. Gomez, S. Bertrand, B. Phannavong, E.D. Gundelfinger, Physiological properties of neuronal nicotinic receptors reconstituted from the vertebrate b 2 subunit and Drosophila a subunits, Eur. J. Neurosci. 6 Ž1994. 869–875. w7x B. Bossy, M. Ballivet, P. Spierer, Conservation of neural nicotinic acetylcholine receptors from Drosophila to vertebrate central nervous systems, EMBO J. 7 Ž1988. 611–618. w8x H. Breer, D.B. Sattelle, Molecular properties and functions of insect acetylcholine receptors, J. Insect Physiol. 33 Ž1987. 771–790. w9x J. Bufler, C. Franke, H. Parnas, J. Dudel, Open channel block by physostigmine and procaine in embryonic-like nicotinic receptors of mouse muscle, Eur. J. Neurosci. 8 Ž1996. 677–687. w10x G.E. Cartier, D. Yoshikami, W.R. Gray, S. Luo, B.M. Olivera, J.M. McIntosh, A new a-conotoxin which targets a 3 b 2 nicotinic acetylcholine receptors, J. Biol. Chem. 271 Ž1996. 7522–7528. w11x D. Colquhoun, B. Sakmann, Fast events in single-channel currents activated by acetylcholine and its analogues at the frog-muscle end-plate, J. Physiol. 369 Ž1985. 501–557. w12x E. Cooper, S. Couturier, M. Ballivet, Pentameric structure and subunit stoichiometry of a neuronal nicotinic acetylcholine receptor, Nature 350 Ž1991. 235–238. w13x J.C. Cooper, O. Gutbrod, V. Witzemann, C. Methfessel, Pharmacology of the nicotinic acetylcholine receptor from fetal rat muscle expressed in Xenopus oocytes, Eur. J. Pharmacol. 309 Ž1996. 287–298. w14x A.B. Elgoyhen, D.S. Johnson, J. Boulter, D.E. Vetter, S. Heinemann, a 9: An acetylcholine receptor with novel pharmacological properties expressed in rat cochlear hair cells, Cell 79 Ž1994. 705–715. w15x P. Gardner, D.C. Ogden, D. Colquhoun, Conductances of single ion channels opened by nicotinic agonists are indistinguishable, Nature 309 Ž1984. 160–162. w16x E.D. Gundelfinger, How complex is the nicotinic receptor system of insects?, T.I.N.S. 15 Ž1992. 206–211. w17x O.P. Hamill, A. Marty, E. Neher, B. Sakmann, F.J. Sigworth, Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches, Pflueg. Arch. 391 Ž1981. 85–100. w18x W. Hanke, H. Breer H, Characterization of the channel properties of a neuronal acetylcholine receptor reconstituted into planar lipid bilayers, J. Gen. Physiol. 90 Ž1987. 855–879.
w19x B. Hermsen, R. Heiermann, A. Maelicke, Cloning and expression of ganglionic nAChR genes from Locusta migratoria, Biol. Chem. Hoppe-Seyler 372 Ž1991. 891. w20x B. Hille, Potassium channels and chloride channels, in: Ionic Channels of Excitable Membranes, Sinauer Associates, Sunderland, USA, 1992, pp. 115–139. w21x D.S. Johnson, J. Martinez, A.B. Elgoyhen, S.F. Heinemann, J.M. McIntosh, a-Conotoxin ImI exhibits subtype-specific nicotinic acetylcholine receptor blockade: preferential inhibition of homomeric a 7 and a 9 receptors, Mol. Pharmacol. 48 Ž1995. 194–199. w22x A.R. Kay, An intracellular medium formulary, J. Neurosci. Methods 44 Ž1992. 91–100. w23x C.A. Leech, D.B. Sattelle, Multiple conductances of neuronal nicotinic acetylcholine receptors, Neuropharmacology 31 Ž1992. 501– 507. w24x J. Marshall, S.D. Buckingham, R. Shingai, G.G. Lunt, M.W. Goosey, M.G. Darlison, D.B. Sattelle, E.A. Barnard, Sequence and functional expression of a single a subunit of an insect nicotinic acetylcholine receptor, EMBO J. 9 Ž1990. 4391–4398. w25x D.S. McGehee, L.W. Role, Physiological diversity of nicotinic acetylcholine receptors expressed by vertebrate neurons, Annu. Rev. Physiol. 57 Ž1995. 521–546. w26x M. Oortgiesen, H.P.M. Vijverberg, Properties of neuronal type acetylcholine receptors in voltage clamped mouse neuroblastoma cells, Neuroscience 33 Ž1989. 169–179. w27x R.L. Papke, G. Millhauser, Z. Lieberman, R.E. Oswald, Relationship of agonist properties to the single channel kinetics of nicotinic acetylcholine receptors, Biophys. J. 53 Ž1988. 1–10. w28x J.B. Patlak, Sodium channel subconductance levels measured with a new variance–mean analysis, J. Gen. Physiol. 92 Ž1988. 413–430. w29x E.F.R. Pereira, M. Alkondon, T. Tano, N.G. Castro, M.M. Froes´ Ferrao, ˜ R. Rozental, R.S. Aronstam, A. Schrattenholz, A. Maelicke, E.X. Albuquerque, A novel agonist binding site on nicotinic acetylcholine receptors, J. Rec. Res. 13 Ž1993. 413–436. w30x E.F.R. Pereira, S. Reinhardt-Maelicke, A. Schrattenholz, A. Maelicke, E.X. Albuquerque, Identification and functional characterization of a new agonist site on nicotinic acetylcholine receptors of cultured hippocampal neurons, J. Pharmacol. Exp. Ther. 265 Ž1993. 1474–1491. w31x E.F.R. Pereira, M. Alkondon, S. Reinhardt, A. Maelicke, X. Peng, J. Lindstrom, P. Whiting, E.X. Albuquerque, Physostigmine and galanthamine: probes for a novel binding site on the a 4b 2 subtype of neuronal nicotinic acetylcholine receptors stably expressed in fibroblast cells, J. Pharmacol. Exp. Ther. 270 Ž1994. 768–778. w32x J. Ramirez-Latorre, C.R. Yu, X. Qu, F. Perin, A. Karlin, L. Role, Functional contributions of a 5 subunit to neuronal acetylcholine receptor channels, Nature 380 Ž1996. 347–351. w33x P.B. Sargent, The diversity of neuronal nicotinic acetylcholine receptors, Annu. Rev. Neurosci. 16 Ž1993. 403–443. w34x P. Schloss, H. Betz, C. Schroder, E.D. Gundelfinger, Neuronal ¨ nicotinic acetylcholine receptors in Drosophila: antibodies against an a-like and a non-a subunit recognize the same high-affinity a-bungarotoxin binding complex, J. Neurochem. 57 Ž1991. 1556– 1562. w35x A. Schrattenholz, T. Coban, B. Schroder, K.O. Okonjo, J. Kuhlmann, ¨ E.F. Pereira, E.X. Albuquerque, A. Maelicke, Biochemical characterization of a novel channel activating site on nicotinic acetylcholine receptors, J. Rec. Res. 13 Ž1993. 393–412. w36x B. Schroder, S. Reinhardt-Maelicke, A. Schrattenholz, K.E. McLane, ¨ A. Kretschmer, B.M. Conti-Tronconi, A. Maelicke, Monoclonal antibodies FK1 and WF6 define two neighboring ligand binding sites on Torpedo acetylcholine receptor a-polypeptide, J. Biol. Chem. 269 Ž1994. 10407–10416. w37x K.P. Shaw, Y. Aracava, A. Akaike, J.W. Daly, D.L. Rickett, E.X. Albuquerque, The reversible cholinesterase inhibitor physostigmine
I. Õan den Beukel et al.r Brain Research 789 (1998) 263–273 has channel-blocking and agonist effects on the acetylcholine receptor–ion channel complex, Mol. Pharmacol. 28 Ž1985. 527–538. w38x S.M. Somani, S.N. Dube, Physostigmine—an overview as pretreatment drug for organophosphate intoxication, Int. J. Clin. Pharmacol., Ther. Toxicol. 27 Ž1989. 367–387. w39x A. Storch, A. Schrattenholz, J.C. Cooper, E.M.A. Ghani, O. Gutˇ ˇ ¸, brod, K.H. Weber, S. Reinhardt, C. Lobron, B. Hermsen, V. Soskic E.F.R. Pereira, E.X. Albuquerque, C. Methfessel, A. Maelicke, Physostigmine, galanthamine and codeine act as ‘noncompetitive nicotinic receptor agonists’ on clonal rat pheochromocytoma cells, Eur. J. Pharmacol. Mol. Pharmacol. 290 Ž1995. 207–219. w40x I. Van den Beukel, R.G.D.M. Van Kleef, M. Oortgiesen, Direct functional effects of parathion and paraoxon on neuronal nicotinic and muscarinic M3 acetylcholine receptors, A.T.L.A. 24 Ž1996. 191–199.
273
w41x I. Van den Beukel, R.G.D.M. Van Kleef, M. Oortgiesen, Cholinesterase inhibitors differentially affect distinct nicotinic receptor subtypes, Prog. Biophys. Mol. Biol. 65 Ž1996. 114, suppl. 1. w42x F. Wang, V. Gerzanich, G.B. Wells, R. Anand, X. Peng, K. Keyser, J. Lindstrom, Assembly of human neuronal nicotinic receptor a 5 subunits with a 3, b 2, and b 4 subunits, J. Biol. Chem. 271 Ž1996. 17656–17665. w43x C.F. Zorumski, L. Lin Thio, K.E. Isenberg, D.B. Clifford, Nicotinic acetylcholine currents in cultured postnatal rat hippocampal neurons, Mol. Pharmacol. 41 Ž1992. 931–936. w44x R. Zwart, M. Oortgiesen, H.P.M. Vijverberg, Nitromethylene heterocycles: selective agonists of nicotinic receptors in locust neurons as compared to mouse N1E-115 and BC3H1 cells, Pest. Biochem. Physiol. 48 Ž1994. 202–213.