- Email: [email protected]
Nifedipine blocks apamin-induced bursting activity in nigral dopamine-containing neurons
Brain Research 817 Ž1999. 104–109
Research report
Nifedipine blocks apamin-induced bursting activity in nigral dopamine-containing neurons Paul D. Shepard ) , Daren Stump Maryland Psychiatric Research Center, P.O. Box 21247, CatonsÕille, MD 21228, USA Department of Psychiatry, UniÕersity of Maryland School of Medicine, Baltimore, MD 21228, USA Accepted 10 November 1998
Abstract Intrinsic sinusoidal oscillations in membrane potential, characteristic of nigral dopamine cells, are converted to plateau potentials following application of apamin, a potent antagonist of SK-type Ca2q-activated Kq channels. Blockade of these channels also changes neuronal firing pattern from a single-spike pacemaker discharge to a multiple spike bursting pattern. Nifedipine, a selective antagonist of L-type Ca2q channels, blocks plateau potential generation; however, its effects on firing pattern have yet to be determined. In the present study, extracellular single unit recording techniques were used in conjunction with a brain slice preparation to determine whether nifedipine, in a concentration known to block plateau potential generation, also affects bursting activity. Nifedipine Ž30 mM. was equipotent in inhibiting the firing rate of control Ž51.2 " 10.8%. and apamin-treated Ž44.9 " 5.4%. neurons. Slow firing neurons Ž- 2 Hz. were particularly sensitive to the inhibitory effects of the drug. Apamin-induced bursting was completely suppressed by nifedipine and accompanied by a significant increase in the regularity of firing. By contrast, pacemaker-like activity exhibited by control neurons was unaffected by the drug. These data demonstrate that the intrinsic plateau properties exhibited by DA neurons are responsible for the generation of phasic activity induced following blockade of apamin-sensitive Ca2q-activated Kq channels and provide further support for the involvement of an L-type Ca2q conductance in mediating this type of activity. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Dopamine neuron; Brain slice; Bursting; Pacemaker
1. Introduction It is now well established that mesencephalic dopamine ŽDA.-containing neurons are capable of generating spontaneous spiking in the absence of afferent input. In addition to pacemaker-like activity, neurons recorded in brain slices exhibit a sinusoidal oscillation in membrane potential ŽSOP. w2,6,15x. The SOP persists in the presence of tetrodotoxin and its frequency varies with membrane voltage indicating that it is intrinsically generated w8,12x. Both Naq and Ca2q conductances appear to contribute to the inward current responsible for the depolarizing phase of the SOP while repolarization is mediated by an SK-type Ca2q-activated Kq channel ŽgK Ca . w9,11,12x. Apamin, a potent and selective antagonist of gK Ca w1,7x, blocks the SOP but fails to inhibit spontaneous spiking w12x. Rather, ) Corresponding author. Maryland Psychiatric Research Center, P.O. Box 21247, Catonsville, MD 21228, USA. Fax: q1-410-747-1797; E-mail: [email protected]
apamin-treated DA neurons exhibit either a continuous irregular single spike pattern or a multiple spike bursting discharge w5,13x. Taken together, these data suggest that the SOP may be responsible for the highly rhythmic firing pattern exhibited by DA neurons in brain slices. Apamin-induced blockade of the SOP is also associated with the appearance of a spontaneous plateau-like oscillation in membrane potential w11,12x. Plateau potentials appear to be mediated by an L-type Ca2q channel since they can be blocked by bath application of the dihydropyridine Ca2q antagonist, nifedipine w11x. Although not observed in every cell, plateau oscillations have been proposed to underlie the bursting activity exhibited by some DA neurons following blockade of gK Ca w12x. In order to test this hypothesis, extracellular single unit recording techniques were used to assess the effects of nifedipine on apamin-induced bursting activity in vitro. Bath application of the antagonist, in a concentration that inhibits plateau potentials Ž30 mM., completely blocked bursting activity exhibited by DA neurons in the presence of apamin. These data
0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 1 2 3 1 - 1
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109
provide further support for the hypothesis that intrinsically generated oscillations in the membrane potential of DA neurons determine the firing patterns exhibited by these cells in vitro.
105
tion of an ISI distribution is expressed as a percentage of the mean interval, has been used previously to quantify changes in the firing pattern of neurons with unequal firing
2. Materials and methods Male Sprague–Dawley rats Ž110–225 g, Charles River, Raleigh, NC. were used in all experiments. Experiments were conducted in strict accordance with the procedures outlined in the Guide for Care and Use of Laboratory Animals ŽNational Institute of Health, Bethesda, MD. and the policies of the Animal Care and Use Committee of the University of Maryland School of Medicine. Animals were anesthetized with chloral hydrate Ž400 mg kgy1 , i.p.. and decapitated. The brain was rapidly removed and immersed in ice-cold, artificial cerebrospinal fluid ŽACSF. of the following composition Žin mM.: NaCl 124, KCl 4, NaH 2 PO4 1.25, MgSO4 1.2, NaHCO 3 25.7, glucose 11, and CaCl 2 2.45. A block of tissue containing the substantia nigra was prepared over ice and placed on the stage of a manual tissue chopper ŽStoelting.. Coronal slices Ž350 mm thickness. were made throughout the anterior–posterior extent of the nucleus and immediately transferred to the stage of an interface recording-perfusion chamber. Tissue was maintained at 358 to 368C in a humidified oxygen environment and superfused at a rate of 1.0 ml miny1 with ACSF equilibrated with 95% O 2 and 5% CO 2 ŽpH 7.4.. Slices remained undisturbed for at least 90 min before the start of the recording studies. Extracellular, single unit activity was recorded from neurons within the pars compacta of the substantia nigra using microelectrodes prepared from glass capillary tubing Ž1.5 mm O.D., FHC, Brunswick, ME.. Electrodes were filled with 2 M NaCl and the tips broken back to achieve an in vitro impedance of 2.0–4.0 M V. Electrode potentials were amplified, filtered Žbandwidth 0.1–8.0 kHz. and continuously monitored with an oscilloscope and audio amplifier. Cells were identified as dopaminergic on the basis of their well-characterized electrophysiological properties w4,15x. Action potentials from identified DA neurons were isolated from background noise using a window discriminator that generated a TTL pulse coincident with each spike. Cumulative rate histograms were compiled in real time from the discriminator output using a PC-based software package for electrophysiology ŽRISI, Symbolic Logic, Dallas, TX.. Interspike interval ŽISI. histograms were created directly from cumulative rate records using a cursorbased routine that permitted selection of discrete epochs for pattern analysis. Individual ISI histograms were compiled using 1000 consecutive action potentials. Drug-induced alterations in firing pattern were assessed by comparing changes in the variation coefficient associated with ISI histograms. This statistic, in which the standard devia-
Fig. 1. Summary of the effects of nifedipine on the firing properties of control and apamin-treated DA neurons. ŽA. Average firing rate of control Žwhite bars. and apamin-treated Žshaded bars. neurons before Žopen bars. and after Žcrosshatched bars. bath application of 30 mM nifedipine. As a group, nifedipine significantly decreased the firing rate of control and apamin-treated cells, Ž))) P - 0.0001, paired t-test.. ŽB. Variation coefficients computed from ISI histograms compiled before Žopen bars. and after Žcrosshatched bars. nifedipine application Ž30 mM.. Pattern analysis was limited to those cells with post-drug firing rates above 1 Hz. Nifedipine had no effect on the firing pattern of control cells Žwhite bars. but significantly decreased the variation coefficient of apamin-treated neurons Žshaded bars; ))) P - 0.0001, paired t-test..
106
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109
rates w14x. Spike trains used to compile ISI distributions were also analyzed for evidence of bursting activity. Burst detection parameters were identical to those developed and validated by Grace and Bunney w3x for use in vivo. Briefly, burst initiation was defined as a spike pair with an ISI F 80 ms. All subsequent spikes were considered as part of burst unless an interval exceeding 160 ms was encountered which signaled burst termination. A minimum of six bursts comprised of at least three spikes per 1000 events was required to satisfy the operational definition of burst firing. At the beginning of each experiment, the spontaneous activity of several DA neurons was briefly monitored to assess the viability of the preparation. Slices were then perfused with either 50 ml of ACSF containing 200 nM apamin or control ACSF. Single unit activity of individual DA neurons in control and apamin-treated slices was recorded for 5–15 min to establish the basal firing properties of each cell. Neurons recorded in control slices were selected for study at random; however, in apamin-treated slices, an effort was made to obtain recordings from cells that exhibited qualitative evidence of spontaneous bursting activity. Following an appropriate control period, cells were tested for their response to bath application of nifedipine Ž30 mM.. This concentration is identical to that used in our laboratory to inhibit plateau potential generation in nigral DA-containing neurons. Nifedipine was prepared as a stock solution in 100% ethanol, stored in the dark and applied by diluting 100 ml of stock in 50 ml of ACSF. All experiments were conducted under diffuse lighting conditions. Nifedipine was applied for 30 min and subsequently eliminated from the recording chamber by dilution with control ACSF. Data are expressed as the arithmetic mean " standard error of the mean ŽS.E.M... Comparison of group means was made using a paired tŽwithin group comparisons. or Student’s t-test Žbetween group comparisons. unless the assumption of equal variance could not be satisfied in which case a non-parametric comparison was substituted ŽMann–Whitney U-statistic.. All p values were derived from two-tailed probability distributions.
Fig. 2. Cumulative rate and ISI histograms illustrating the effects of nifedipine on the discharge properties of mesencephalic DA neurons exhibiting different basal firing rates. ŽA. Slow firing DA neuron in which 30 min of continuous perfusion with ACSF containing 30 mM nifedipine Žhorizontal bar. results in a near complete cessation of neuronal firing. Note that firing rate remains depressed despite 20 min of perfusion with control ACSF. Upper panel illustrates the first order ISI histogram compiled prior to drug treatment. ŽB. In this example, typical of faster firing DA neurons, bath application of nifedipine results in a 24% inhibition in firing rate. The rapid reversal of nifedipine’s inhibitory effects was not typical of most cells tested. ISI histograms Župper panels. compiled before and after drug application showed no evidence of a change in discharge pattern. Asterisks indicate regions of the rate histogram used to construct control Ž). and post-nifedipine Ž)). ISI distributions.
3. Results Stable extracellular recordings were obtained from a total of 30 neurons including 12 from control and 18 from apamin-treated slices. DA neurons from control slices exhibited moderately slow firing rates Žrange: 2.7–5.8 Hz.
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109
and a highly regular ‘pacemaker-like’ discharge Žvariation coefficient, range: 2.6%–6.1%.. Bath application of nifedipine Ž30 mM for 30 min. significantly decreased neuronal firing rate Žcontrol: 3.5 " 0.3 Hz, nifedipine: 1.9 " 0.5 Hz, paired tŽ9. s 5.7, P - 0.001; Fig. 1A.. However, considerable variability was observed in the sensitivity of individual DA neurons. Slow firing cells were particularly susceptible to the inhibitory effects of the drug Ž% inhibition: 82.5 " 4.3.. Of five cells with basal firing rates below 3 Hz, two were inhibited by 90% or more during nifedipine application while the remaining three cells showed a more gradual decline in firing rate that persisted during washout with control ACSF ŽFig. 2A.. By contrast, faster firing cells Ž) 3 Hz. showed less inhibition during nifedipine application Ž19.8 " 4.4%, n s 5. and were less likely to show delayed changes in activity after nifedipine had been removed from the recording chamber ŽFig. 2B.. Latency to onset of the maximal inhibitory effects of nifedipine averaged 36 " 2.8 min and most cells failed to return to pre-drug levels of activity during washout with drug-free ACSF. The marked inhibitory effects of nifedipine on slow firing DA cells precluded a quantitative description of drug-induced changes in the firing pattern of these neurons. However, comparison of variation coefficients obtained from fast firing neurons before and after nifedipine revealed no evidence of a drug-induced alteration in discharge pattern Žcontrols 4.0 " 0.6%, nifedipine s 4.5 " 0.2%, paired tŽ4. s 0.67, P ) 0.05; Fig. 1B.. Control cells Ž n s 2. tested with the nifedipine vehicle ŽACSFq 0.2% ethanol. showed no change in firing rate or discharge pattern. As previously reported, the firing properties of apamintreated DA cells differed considerably from those in control slices w13x. As a group, these neurons fired faster Žrange: 2.5–7.0 Hz. and with a more irregular discharge pattern Žvariation coefficient, range: 21.7–61.7%,. than control cells ŽFig. 1.. Each of the neurons selected for study exhibited qualitative evidence of bursting activity including Ž1. episodes of high frequency firing; Ž2. postburst inhibitory period and Ž3. a progressive reduction in spike amplitude. However, of the 18 cells sampled, only 10 met the operational definition for bursting activity in vivo. In most instances, cells that failed to satisfy these criteria, exhibited bursts with an initial interspike interval ) 80 ms. Bath application of nifedipine significantly reduced the firing rate of apamin-treated DA neurons Žapamin: 4.5 " 0.3 Hz, apaminq nifedipines 2.6 " 0.4 Hz, paired tŽ17. s 9.2, P - 0.0001; Fig. 1A.. Three of the four apamin-treated neurons which exhibited firing rates - 3 Hz were inhibited below 1 Hz during nifedipine application and could not be included in the analysis of drug-induced changes in firing pattern. No differences were found between the sensitivity of control and apamin-treated neurons to the inhibitory effects of nifedipine Ž% inhibition, control: 51.2 " 10.8; apamin: 44.9 " 5.4, Student’s tŽ26. s 0.58, P )
107
0.5.. However, unlike control cells, which remained pacemaker-like following nifedipine application, apamin-treated neurons exhibited a pronounced change in firing pattern in response to the drug. Bursting was completely suppressed ŽFig. 3. and accompanied by a significant increase in the regularity of neuronal firing pattern Žvariation coefficient: apamins 45.6 " 3%, apaminq nifedipines 12.9 " 3.7%; paired tŽ14. s 8.3, P - 0.0001; Fig. 1B.. However, the firing pattern exhibited by apamin-treated cells following nifedipine remained significantly more irregular than that associated with control neurons ŽMann–Whitney U, P 0.0001.. No relationship was observed between the extent of the change in firing pattern and either the cells firing rate in apamin ŽPearson r s 0.1, P s 0.7. or magnitude of the inhibition produced by nifedipine ŽPearson r s 0.2, P s 0.4..
Fig. 3. Effects of nifedipine on the discharge properties of a representative apamin-treated DA neuron. Upper panels: ISI histograms compiled prior to Žleft. and following Žright. bath application of 30 mM nifedipine. Note that the positively skewed ISI distribution, characteristic of bursting activity, is converted to a normal distribution following nifedipine application and that these changes are accompanied by a marked reduction in the variation coefficient. Lower panel: Cumulative rate histogram illustrating the inhibitory effects of nifedipine on neuronal firing rate. Asterisks denote regions of the rate record used to compile control Ž). and post-nifedipine Ž)). ISI distributions.
108
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109
4. Discussion The discovery of a tetrodotoxin-insensitive SOP in DA neurons led to the suggestion that this mechanism may underlie the pacemaker properties exhibited by these cells in vitro w2,6x. The observation that inorganic Ca2q channel blockers lead to a suppression of both the SOP and spontaneous spiking are consistent with this hypothesis and have established the importance of Ca2q currents in the generation of rhythmic activity. However, attempts to identify the role of specific Ca2q channels in mediating pacemaker properties of DA neurons have yielded conflicting results. Thus, while some investigators have reported that high concentrations of nifedipine Ž100 mM. are without effect on neuronal firing rate w2x, others have shown that lower doses of the drug Ž5–30 mM. suppress both the SOP and spontaneous spiking w10,11x. The results of the present study, which have confirmed the inhibitory actions of nifedipine on DA neuronal activity using extracellular recording techniques, are generally consistent with the intracellular data obtained by Nedergaard et al. w11x and Mercuri et al. w10x. However, in contrast to a complete cessation in firing, DA neurons showed considerable variability in their responsiveness to the drug. Latency to onset of the peak inhibitory effects of nifedipine was also significantly slower Ž20–50 min. than that observed by previous investigators Ž7–15 min., despite the fact that the concentration of nifedipine used in the present study was comparable to that used in earlier studies. The reason for this discrepancy is unclear but could involve differences arising from the type of animal used Žrat vs. guinea pig., the chamber design Žinterface vs. submerged. or in the basal firing rates of the cells selected for testing, which, in the present study, tended to be faster Ž; 3.5 Hz. than those reported by previous investigators. Indeed, in our hands, cells with basal firing rates below 2 Hz tended to be silenced by the drug, while faster firing cells showed a more modest inhibition. Although a relationship between the SOP and spontaneous spiking has yet to be firmly established, previous intracellular recording studies have demonstrated that subthreshold oscillations in membrane potential can be blocked without affecting somatic action potentials. For example, the irregular single spike activity, typically observed following blockade of gK Ca in DA neurons, occurs in the absence of an underlying oscillation in membrane potential w12x. On the other hand, burst firing, which is also observed following inhibition of gK Ca w13x has been associated with a plateau-like oscillation in membrane potential w12x. The observation that nifedipine, in a concentration known to block these oscillations, is also capable of suppressing apamin-induced bursting activity strongly suggests that plateau potentials are directly responsible for generation of phasic activity induced by blockade of gK Ca and provides further support for the involvement of L-type Ca2q channels in mediating this type of activity. By
contrast, the failure of nifedipine to alter the firing pattern of DA neurons in control slices could be interpreted as suggesting that these channels do not contribute to the pacemaker-like activity patterns typically observed in vitro. Unfortunately, we are unable to know whether the firing pattern exhibited by control cells in the presence of nifedipine occurred in the absence of an underlying SOP. However, it should be noted in this regard that the firing pattern exhibited by apamin-treated DA neurons following nifedipine remained significantly more irregular than the pacemaker activity observed in control slices, presumably reflecting the ability of apamin to block the SOP. Taken together, these data suggest that the intrinsic oscillatory properties exhibited by DA neurons may contribute more to the firing patterns exhibited by these cells than to the mechanismŽs. underlying their ability to remain active in the absence of synaptic input.
Acknowledgements The authors wish to thank Mr. Leon Simms and Ms. Sarah Kattahuzhy for excellent technical support and Mrs. Sharon Stilling for assistance in preparing the manuscript. Supported in part by USPHS grant MH48543 and a DRIF grant from the State of Maryland.
References w1x A.L. Blatz, K.L. Magleby, Calcium-activated potassium channels, Trends Neurosci. 10 Ž1987. 463–467. w2x K. Fujimura, Y. Matsuda, Autogenous oscillatory potentials in neurons of the guinea pig substantia nigra pars compacta in vitro, Neurosci. Lett. 104 Ž1989. 53–57. w3x A.A. Grace, B.S. Bunney, The control of firing pattern in nigral dopamine neurons: burst firing, J. Neurosci. 4 Ž1984. 2877–2890. w4x A.A. Grace, S.-P. Onn, Morphology and electrophysiology properties of immunocytochemically identified rat dopamine neurons recorded in vitro, J. Neurosci. 9 Ž1989. 3463–3481. w5x X. Gu, A.L. Blatz, D.C. German, Subtypes of substantia nigra dopaminergic neurons revealed by apamin: autoradiographic and electrophysiological studies, Brain Res. Bull. 28 Ž1992. 435–440. w6x N.C. Harris, C. Web, S.A. Greenfield, A possible pacemaker mechanism in pars compacta neurons of the guinea pig substantia nigra revealed by various ion channel blocking agents, Neuroscience 31 Ž1989. 355–362. w7x M. Hughues, G. Romey, D. Duval, J.P. Vincent, M. Lazdunski, Apamin as a selective blocker of the calcium-dependent potassium channel in neuroblastoma cells: voltage clamp and biochemical characterization of the toxin receptor, Proc. Natl. Acad. Sci. U.S.A. 79 Ž1982. 1307–1312. w8x Y. Kang, S.T. Kitai, Calcium spike underlying rhythmic firing in dopaminergic neurons of the rat substantia nigra, Neurosci. Res. 18 Ž1993. 195–207. w9x Y. Kang, S.T. Kitai, A whole cell patch-clamp study on the pacemaker potential in dopaminergic neurons of rat substantia nigra compacta, Neurosci. Res. 18 Ž1993. 209–221. w10x N.B. Mercuri, A. Bonci, P. Calabresi, F. Stratta, A. Stefani, G. Bernardi, Effects of dihydropyridine calcium antagonists on rat
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109 midbrain dopaminergic neurones, Br. J. Pharmacol. 113 Ž1994. 831–838. w11x S. Nedergaard, J.A. Flatman, I. Engberg, Nifedipine- and v-conotoxin-sensitive Ca2q conductances in guinea pig substantia nigra pars compacta neurones, J. Physiol. 466 Ž1993. 727–747. w12x H.X. Ping, P.D. Shepard, Apamin-sensitive Ca2q-activated Kq channels regulate pacemaker activity in nigral dopamine neurons, NeuroReport 7 Ž1996. 809–814. w13x P.D. Shepard, B.S. Bunney, Effects of apamin on the discharge
109
properties of putative dopamine-containing neurons in vitro, Brain Res. 463 Ž1988. 380–384. w14x G. Werner, V.B. Mountcastle, The variability of central neural activity in a sensory system and its implication for the central reflection of sensory events, J. Neurophysiol. 26 Ž1963. 958–977. w15x W.H. Yung, M.A. Hausser, J.J.B. Jack, Electrophysiology of dopaminergic and non-dopaminergic neurones of the guinea pig substantia nigra pars compacta in vitro, J. Physiol. 436 Ž1991. 643–667.
Research report
Nifedipine blocks apamin-induced bursting activity in nigral dopamine-containing neurons Paul D. Shepard ) , Daren Stump Maryland Psychiatric Research Center, P.O. Box 21247, CatonsÕille, MD 21228, USA Department of Psychiatry, UniÕersity of Maryland School of Medicine, Baltimore, MD 21228, USA Accepted 10 November 1998
Abstract Intrinsic sinusoidal oscillations in membrane potential, characteristic of nigral dopamine cells, are converted to plateau potentials following application of apamin, a potent antagonist of SK-type Ca2q-activated Kq channels. Blockade of these channels also changes neuronal firing pattern from a single-spike pacemaker discharge to a multiple spike bursting pattern. Nifedipine, a selective antagonist of L-type Ca2q channels, blocks plateau potential generation; however, its effects on firing pattern have yet to be determined. In the present study, extracellular single unit recording techniques were used in conjunction with a brain slice preparation to determine whether nifedipine, in a concentration known to block plateau potential generation, also affects bursting activity. Nifedipine Ž30 mM. was equipotent in inhibiting the firing rate of control Ž51.2 " 10.8%. and apamin-treated Ž44.9 " 5.4%. neurons. Slow firing neurons Ž- 2 Hz. were particularly sensitive to the inhibitory effects of the drug. Apamin-induced bursting was completely suppressed by nifedipine and accompanied by a significant increase in the regularity of firing. By contrast, pacemaker-like activity exhibited by control neurons was unaffected by the drug. These data demonstrate that the intrinsic plateau properties exhibited by DA neurons are responsible for the generation of phasic activity induced following blockade of apamin-sensitive Ca2q-activated Kq channels and provide further support for the involvement of an L-type Ca2q conductance in mediating this type of activity. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Dopamine neuron; Brain slice; Bursting; Pacemaker
1. Introduction It is now well established that mesencephalic dopamine ŽDA.-containing neurons are capable of generating spontaneous spiking in the absence of afferent input. In addition to pacemaker-like activity, neurons recorded in brain slices exhibit a sinusoidal oscillation in membrane potential ŽSOP. w2,6,15x. The SOP persists in the presence of tetrodotoxin and its frequency varies with membrane voltage indicating that it is intrinsically generated w8,12x. Both Naq and Ca2q conductances appear to contribute to the inward current responsible for the depolarizing phase of the SOP while repolarization is mediated by an SK-type Ca2q-activated Kq channel ŽgK Ca . w9,11,12x. Apamin, a potent and selective antagonist of gK Ca w1,7x, blocks the SOP but fails to inhibit spontaneous spiking w12x. Rather, ) Corresponding author. Maryland Psychiatric Research Center, P.O. Box 21247, Catonsville, MD 21228, USA. Fax: q1-410-747-1797; E-mail: [email protected]
apamin-treated DA neurons exhibit either a continuous irregular single spike pattern or a multiple spike bursting discharge w5,13x. Taken together, these data suggest that the SOP may be responsible for the highly rhythmic firing pattern exhibited by DA neurons in brain slices. Apamin-induced blockade of the SOP is also associated with the appearance of a spontaneous plateau-like oscillation in membrane potential w11,12x. Plateau potentials appear to be mediated by an L-type Ca2q channel since they can be blocked by bath application of the dihydropyridine Ca2q antagonist, nifedipine w11x. Although not observed in every cell, plateau oscillations have been proposed to underlie the bursting activity exhibited by some DA neurons following blockade of gK Ca w12x. In order to test this hypothesis, extracellular single unit recording techniques were used to assess the effects of nifedipine on apamin-induced bursting activity in vitro. Bath application of the antagonist, in a concentration that inhibits plateau potentials Ž30 mM., completely blocked bursting activity exhibited by DA neurons in the presence of apamin. These data
0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 1 2 3 1 - 1
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109
provide further support for the hypothesis that intrinsically generated oscillations in the membrane potential of DA neurons determine the firing patterns exhibited by these cells in vitro.
105
tion of an ISI distribution is expressed as a percentage of the mean interval, has been used previously to quantify changes in the firing pattern of neurons with unequal firing
2. Materials and methods Male Sprague–Dawley rats Ž110–225 g, Charles River, Raleigh, NC. were used in all experiments. Experiments were conducted in strict accordance with the procedures outlined in the Guide for Care and Use of Laboratory Animals ŽNational Institute of Health, Bethesda, MD. and the policies of the Animal Care and Use Committee of the University of Maryland School of Medicine. Animals were anesthetized with chloral hydrate Ž400 mg kgy1 , i.p.. and decapitated. The brain was rapidly removed and immersed in ice-cold, artificial cerebrospinal fluid ŽACSF. of the following composition Žin mM.: NaCl 124, KCl 4, NaH 2 PO4 1.25, MgSO4 1.2, NaHCO 3 25.7, glucose 11, and CaCl 2 2.45. A block of tissue containing the substantia nigra was prepared over ice and placed on the stage of a manual tissue chopper ŽStoelting.. Coronal slices Ž350 mm thickness. were made throughout the anterior–posterior extent of the nucleus and immediately transferred to the stage of an interface recording-perfusion chamber. Tissue was maintained at 358 to 368C in a humidified oxygen environment and superfused at a rate of 1.0 ml miny1 with ACSF equilibrated with 95% O 2 and 5% CO 2 ŽpH 7.4.. Slices remained undisturbed for at least 90 min before the start of the recording studies. Extracellular, single unit activity was recorded from neurons within the pars compacta of the substantia nigra using microelectrodes prepared from glass capillary tubing Ž1.5 mm O.D., FHC, Brunswick, ME.. Electrodes were filled with 2 M NaCl and the tips broken back to achieve an in vitro impedance of 2.0–4.0 M V. Electrode potentials were amplified, filtered Žbandwidth 0.1–8.0 kHz. and continuously monitored with an oscilloscope and audio amplifier. Cells were identified as dopaminergic on the basis of their well-characterized electrophysiological properties w4,15x. Action potentials from identified DA neurons were isolated from background noise using a window discriminator that generated a TTL pulse coincident with each spike. Cumulative rate histograms were compiled in real time from the discriminator output using a PC-based software package for electrophysiology ŽRISI, Symbolic Logic, Dallas, TX.. Interspike interval ŽISI. histograms were created directly from cumulative rate records using a cursorbased routine that permitted selection of discrete epochs for pattern analysis. Individual ISI histograms were compiled using 1000 consecutive action potentials. Drug-induced alterations in firing pattern were assessed by comparing changes in the variation coefficient associated with ISI histograms. This statistic, in which the standard devia-
Fig. 1. Summary of the effects of nifedipine on the firing properties of control and apamin-treated DA neurons. ŽA. Average firing rate of control Žwhite bars. and apamin-treated Žshaded bars. neurons before Žopen bars. and after Žcrosshatched bars. bath application of 30 mM nifedipine. As a group, nifedipine significantly decreased the firing rate of control and apamin-treated cells, Ž))) P - 0.0001, paired t-test.. ŽB. Variation coefficients computed from ISI histograms compiled before Žopen bars. and after Žcrosshatched bars. nifedipine application Ž30 mM.. Pattern analysis was limited to those cells with post-drug firing rates above 1 Hz. Nifedipine had no effect on the firing pattern of control cells Žwhite bars. but significantly decreased the variation coefficient of apamin-treated neurons Žshaded bars; ))) P - 0.0001, paired t-test..
106
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109
rates w14x. Spike trains used to compile ISI distributions were also analyzed for evidence of bursting activity. Burst detection parameters were identical to those developed and validated by Grace and Bunney w3x for use in vivo. Briefly, burst initiation was defined as a spike pair with an ISI F 80 ms. All subsequent spikes were considered as part of burst unless an interval exceeding 160 ms was encountered which signaled burst termination. A minimum of six bursts comprised of at least three spikes per 1000 events was required to satisfy the operational definition of burst firing. At the beginning of each experiment, the spontaneous activity of several DA neurons was briefly monitored to assess the viability of the preparation. Slices were then perfused with either 50 ml of ACSF containing 200 nM apamin or control ACSF. Single unit activity of individual DA neurons in control and apamin-treated slices was recorded for 5–15 min to establish the basal firing properties of each cell. Neurons recorded in control slices were selected for study at random; however, in apamin-treated slices, an effort was made to obtain recordings from cells that exhibited qualitative evidence of spontaneous bursting activity. Following an appropriate control period, cells were tested for their response to bath application of nifedipine Ž30 mM.. This concentration is identical to that used in our laboratory to inhibit plateau potential generation in nigral DA-containing neurons. Nifedipine was prepared as a stock solution in 100% ethanol, stored in the dark and applied by diluting 100 ml of stock in 50 ml of ACSF. All experiments were conducted under diffuse lighting conditions. Nifedipine was applied for 30 min and subsequently eliminated from the recording chamber by dilution with control ACSF. Data are expressed as the arithmetic mean " standard error of the mean ŽS.E.M... Comparison of group means was made using a paired tŽwithin group comparisons. or Student’s t-test Žbetween group comparisons. unless the assumption of equal variance could not be satisfied in which case a non-parametric comparison was substituted ŽMann–Whitney U-statistic.. All p values were derived from two-tailed probability distributions.
Fig. 2. Cumulative rate and ISI histograms illustrating the effects of nifedipine on the discharge properties of mesencephalic DA neurons exhibiting different basal firing rates. ŽA. Slow firing DA neuron in which 30 min of continuous perfusion with ACSF containing 30 mM nifedipine Žhorizontal bar. results in a near complete cessation of neuronal firing. Note that firing rate remains depressed despite 20 min of perfusion with control ACSF. Upper panel illustrates the first order ISI histogram compiled prior to drug treatment. ŽB. In this example, typical of faster firing DA neurons, bath application of nifedipine results in a 24% inhibition in firing rate. The rapid reversal of nifedipine’s inhibitory effects was not typical of most cells tested. ISI histograms Župper panels. compiled before and after drug application showed no evidence of a change in discharge pattern. Asterisks indicate regions of the rate histogram used to construct control Ž). and post-nifedipine Ž)). ISI distributions.
3. Results Stable extracellular recordings were obtained from a total of 30 neurons including 12 from control and 18 from apamin-treated slices. DA neurons from control slices exhibited moderately slow firing rates Žrange: 2.7–5.8 Hz.
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109
and a highly regular ‘pacemaker-like’ discharge Žvariation coefficient, range: 2.6%–6.1%.. Bath application of nifedipine Ž30 mM for 30 min. significantly decreased neuronal firing rate Žcontrol: 3.5 " 0.3 Hz, nifedipine: 1.9 " 0.5 Hz, paired tŽ9. s 5.7, P - 0.001; Fig. 1A.. However, considerable variability was observed in the sensitivity of individual DA neurons. Slow firing cells were particularly susceptible to the inhibitory effects of the drug Ž% inhibition: 82.5 " 4.3.. Of five cells with basal firing rates below 3 Hz, two were inhibited by 90% or more during nifedipine application while the remaining three cells showed a more gradual decline in firing rate that persisted during washout with control ACSF ŽFig. 2A.. By contrast, faster firing cells Ž) 3 Hz. showed less inhibition during nifedipine application Ž19.8 " 4.4%, n s 5. and were less likely to show delayed changes in activity after nifedipine had been removed from the recording chamber ŽFig. 2B.. Latency to onset of the maximal inhibitory effects of nifedipine averaged 36 " 2.8 min and most cells failed to return to pre-drug levels of activity during washout with drug-free ACSF. The marked inhibitory effects of nifedipine on slow firing DA cells precluded a quantitative description of drug-induced changes in the firing pattern of these neurons. However, comparison of variation coefficients obtained from fast firing neurons before and after nifedipine revealed no evidence of a drug-induced alteration in discharge pattern Žcontrols 4.0 " 0.6%, nifedipine s 4.5 " 0.2%, paired tŽ4. s 0.67, P ) 0.05; Fig. 1B.. Control cells Ž n s 2. tested with the nifedipine vehicle ŽACSFq 0.2% ethanol. showed no change in firing rate or discharge pattern. As previously reported, the firing properties of apamintreated DA cells differed considerably from those in control slices w13x. As a group, these neurons fired faster Žrange: 2.5–7.0 Hz. and with a more irregular discharge pattern Žvariation coefficient, range: 21.7–61.7%,. than control cells ŽFig. 1.. Each of the neurons selected for study exhibited qualitative evidence of bursting activity including Ž1. episodes of high frequency firing; Ž2. postburst inhibitory period and Ž3. a progressive reduction in spike amplitude. However, of the 18 cells sampled, only 10 met the operational definition for bursting activity in vivo. In most instances, cells that failed to satisfy these criteria, exhibited bursts with an initial interspike interval ) 80 ms. Bath application of nifedipine significantly reduced the firing rate of apamin-treated DA neurons Žapamin: 4.5 " 0.3 Hz, apaminq nifedipines 2.6 " 0.4 Hz, paired tŽ17. s 9.2, P - 0.0001; Fig. 1A.. Three of the four apamin-treated neurons which exhibited firing rates - 3 Hz were inhibited below 1 Hz during nifedipine application and could not be included in the analysis of drug-induced changes in firing pattern. No differences were found between the sensitivity of control and apamin-treated neurons to the inhibitory effects of nifedipine Ž% inhibition, control: 51.2 " 10.8; apamin: 44.9 " 5.4, Student’s tŽ26. s 0.58, P )
107
0.5.. However, unlike control cells, which remained pacemaker-like following nifedipine application, apamin-treated neurons exhibited a pronounced change in firing pattern in response to the drug. Bursting was completely suppressed ŽFig. 3. and accompanied by a significant increase in the regularity of neuronal firing pattern Žvariation coefficient: apamins 45.6 " 3%, apaminq nifedipines 12.9 " 3.7%; paired tŽ14. s 8.3, P - 0.0001; Fig. 1B.. However, the firing pattern exhibited by apamin-treated cells following nifedipine remained significantly more irregular than that associated with control neurons ŽMann–Whitney U, P 0.0001.. No relationship was observed between the extent of the change in firing pattern and either the cells firing rate in apamin ŽPearson r s 0.1, P s 0.7. or magnitude of the inhibition produced by nifedipine ŽPearson r s 0.2, P s 0.4..
Fig. 3. Effects of nifedipine on the discharge properties of a representative apamin-treated DA neuron. Upper panels: ISI histograms compiled prior to Žleft. and following Žright. bath application of 30 mM nifedipine. Note that the positively skewed ISI distribution, characteristic of bursting activity, is converted to a normal distribution following nifedipine application and that these changes are accompanied by a marked reduction in the variation coefficient. Lower panel: Cumulative rate histogram illustrating the inhibitory effects of nifedipine on neuronal firing rate. Asterisks denote regions of the rate record used to compile control Ž). and post-nifedipine Ž)). ISI distributions.
108
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109
4. Discussion The discovery of a tetrodotoxin-insensitive SOP in DA neurons led to the suggestion that this mechanism may underlie the pacemaker properties exhibited by these cells in vitro w2,6x. The observation that inorganic Ca2q channel blockers lead to a suppression of both the SOP and spontaneous spiking are consistent with this hypothesis and have established the importance of Ca2q currents in the generation of rhythmic activity. However, attempts to identify the role of specific Ca2q channels in mediating pacemaker properties of DA neurons have yielded conflicting results. Thus, while some investigators have reported that high concentrations of nifedipine Ž100 mM. are without effect on neuronal firing rate w2x, others have shown that lower doses of the drug Ž5–30 mM. suppress both the SOP and spontaneous spiking w10,11x. The results of the present study, which have confirmed the inhibitory actions of nifedipine on DA neuronal activity using extracellular recording techniques, are generally consistent with the intracellular data obtained by Nedergaard et al. w11x and Mercuri et al. w10x. However, in contrast to a complete cessation in firing, DA neurons showed considerable variability in their responsiveness to the drug. Latency to onset of the peak inhibitory effects of nifedipine was also significantly slower Ž20–50 min. than that observed by previous investigators Ž7–15 min., despite the fact that the concentration of nifedipine used in the present study was comparable to that used in earlier studies. The reason for this discrepancy is unclear but could involve differences arising from the type of animal used Žrat vs. guinea pig., the chamber design Žinterface vs. submerged. or in the basal firing rates of the cells selected for testing, which, in the present study, tended to be faster Ž; 3.5 Hz. than those reported by previous investigators. Indeed, in our hands, cells with basal firing rates below 2 Hz tended to be silenced by the drug, while faster firing cells showed a more modest inhibition. Although a relationship between the SOP and spontaneous spiking has yet to be firmly established, previous intracellular recording studies have demonstrated that subthreshold oscillations in membrane potential can be blocked without affecting somatic action potentials. For example, the irregular single spike activity, typically observed following blockade of gK Ca in DA neurons, occurs in the absence of an underlying oscillation in membrane potential w12x. On the other hand, burst firing, which is also observed following inhibition of gK Ca w13x has been associated with a plateau-like oscillation in membrane potential w12x. The observation that nifedipine, in a concentration known to block these oscillations, is also capable of suppressing apamin-induced bursting activity strongly suggests that plateau potentials are directly responsible for generation of phasic activity induced by blockade of gK Ca and provides further support for the involvement of L-type Ca2q channels in mediating this type of activity. By
contrast, the failure of nifedipine to alter the firing pattern of DA neurons in control slices could be interpreted as suggesting that these channels do not contribute to the pacemaker-like activity patterns typically observed in vitro. Unfortunately, we are unable to know whether the firing pattern exhibited by control cells in the presence of nifedipine occurred in the absence of an underlying SOP. However, it should be noted in this regard that the firing pattern exhibited by apamin-treated DA neurons following nifedipine remained significantly more irregular than the pacemaker activity observed in control slices, presumably reflecting the ability of apamin to block the SOP. Taken together, these data suggest that the intrinsic oscillatory properties exhibited by DA neurons may contribute more to the firing patterns exhibited by these cells than to the mechanismŽs. underlying their ability to remain active in the absence of synaptic input.
Acknowledgements The authors wish to thank Mr. Leon Simms and Ms. Sarah Kattahuzhy for excellent technical support and Mrs. Sharon Stilling for assistance in preparing the manuscript. Supported in part by USPHS grant MH48543 and a DRIF grant from the State of Maryland.
References w1x A.L. Blatz, K.L. Magleby, Calcium-activated potassium channels, Trends Neurosci. 10 Ž1987. 463–467. w2x K. Fujimura, Y. Matsuda, Autogenous oscillatory potentials in neurons of the guinea pig substantia nigra pars compacta in vitro, Neurosci. Lett. 104 Ž1989. 53–57. w3x A.A. Grace, B.S. Bunney, The control of firing pattern in nigral dopamine neurons: burst firing, J. Neurosci. 4 Ž1984. 2877–2890. w4x A.A. Grace, S.-P. Onn, Morphology and electrophysiology properties of immunocytochemically identified rat dopamine neurons recorded in vitro, J. Neurosci. 9 Ž1989. 3463–3481. w5x X. Gu, A.L. Blatz, D.C. German, Subtypes of substantia nigra dopaminergic neurons revealed by apamin: autoradiographic and electrophysiological studies, Brain Res. Bull. 28 Ž1992. 435–440. w6x N.C. Harris, C. Web, S.A. Greenfield, A possible pacemaker mechanism in pars compacta neurons of the guinea pig substantia nigra revealed by various ion channel blocking agents, Neuroscience 31 Ž1989. 355–362. w7x M. Hughues, G. Romey, D. Duval, J.P. Vincent, M. Lazdunski, Apamin as a selective blocker of the calcium-dependent potassium channel in neuroblastoma cells: voltage clamp and biochemical characterization of the toxin receptor, Proc. Natl. Acad. Sci. U.S.A. 79 Ž1982. 1307–1312. w8x Y. Kang, S.T. Kitai, Calcium spike underlying rhythmic firing in dopaminergic neurons of the rat substantia nigra, Neurosci. Res. 18 Ž1993. 195–207. w9x Y. Kang, S.T. Kitai, A whole cell patch-clamp study on the pacemaker potential in dopaminergic neurons of rat substantia nigra compacta, Neurosci. Res. 18 Ž1993. 209–221. w10x N.B. Mercuri, A. Bonci, P. Calabresi, F. Stratta, A. Stefani, G. Bernardi, Effects of dihydropyridine calcium antagonists on rat
P.D. Shepard, D. Stump r Brain Research 817 (1999) 104–109 midbrain dopaminergic neurones, Br. J. Pharmacol. 113 Ž1994. 831–838. w11x S. Nedergaard, J.A. Flatman, I. Engberg, Nifedipine- and v-conotoxin-sensitive Ca2q conductances in guinea pig substantia nigra pars compacta neurones, J. Physiol. 466 Ž1993. 727–747. w12x H.X. Ping, P.D. Shepard, Apamin-sensitive Ca2q-activated Kq channels regulate pacemaker activity in nigral dopamine neurons, NeuroReport 7 Ž1996. 809–814. w13x P.D. Shepard, B.S. Bunney, Effects of apamin on the discharge
109
properties of putative dopamine-containing neurons in vitro, Brain Res. 463 Ž1988. 380–384. w14x G. Werner, V.B. Mountcastle, The variability of central neural activity in a sensory system and its implication for the central reflection of sensory events, J. Neurophysiol. 26 Ž1963. 958–977. w15x W.H. Yung, M.A. Hausser, J.J.B. Jack, Electrophysiology of dopaminergic and non-dopaminergic neurones of the guinea pig substantia nigra pars compacta in vitro, J. Physiol. 436 Ž1991. 643–667.