Exp Brain Res (2015) 233:137–147 DOI 10.1007/s00221-014-4095-6
RESEARCH ARTICLE
MDMA modulates spontaneous firing of subthalamic nucleus neurons in vitro Luise Liebig · Andreas von Ameln‑Mayerhofer · Harald Hentschke
Received: 5 January 2014 / Accepted: 5 September 2014 / Published online: 19 September 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract 3,4-Methylene-dioxy-N-methylamphetamine (MDMA, ‘ecstasy’) has a broad spectrum of molecular targets in the brain, among them receptors and transporters of the serotonergic (5-hydroxytryptamine, 5-HT) and noradrenergic systems. Its action on the serotonergic system modulates motor systems in rodents and humans. Although parts of the basal ganglia could be identified as mediators of the motor effects of MDMA, very little is known about the role of the subthalamic nucleus (STN). Therefore, this study investigated the modulation of spontaneous action potential activity of the STN by MDMA (2.5–20 µM) in vitro. MDMA had very heterogeneous effects, ranging from a complete but reversible inhibition to a more than twofold increase in firing at 5 µM. On average, MDMA excited STN neurons moderately, but lost its excitatory effect in the presence of the 5-HT2A antagonist MDL 11,939. 5-HT1A receptors did not appear to play a major role. Effects of MDMA on transporters for serotonin (SERT) and norepinephrine (NET) were investigated by coapplication of the reuptake inhibitors citalopram and desipramine, respectively. Similar to the effects of 5-HT2A receptor blockade, antagonism of SERT and NET bestowed an inhibitory effect on MDMA. From these results, we L. Liebig · H. Hentschke (*) Experimental Anaesthesiology Section, University Hospital Tübingen, Waldhörnlestr. 22, 72072 Tübingen, Germany e-mail: harald.hentschke@uni‑tuebingen.de L. Liebig · A. von Ameln‑Mayerhofer Neuropharmacology, Institute of Neurobiology, University of Tübingen, Tübingen, Germany L. Liebig Graduate School of Neural and Behavioural Sciences, International Max‑Planck Research School, University of Tübingen, Tübingen, Germany
conclude that both the 5-HT and the noradrenergic system mediate MDMA-induced effects on STN neurons. Keywords Ecstasy · Firing rate · Basal ganglia · Serotonin · Noradrenaline
Introduction 3,4-Methylene-dioxy-N-methylamphetamine (MDMA, ‘ecstasy’) is a synthetic illicit drug mostly known for its recreational use. MDMA has a wide spectrum of psychological and physiological effects, including weak rewarding and addicting potentials in animals and humans (Green et al. 2003; Reinhard and Wolffgramm 2006). Alongside a controversial discussion of the beneficial versus harmful potential of MDMA (Green et al. 2012; Parrott 2012), the effects of the drug continue to be investigated, e.g. on cognition (Mithoefer et al. 2013; Nelson et al. 2013) and the motor system (Baumann et al. 2008). Specifically, an antidyskinetic effect of MDMA, reported anecdotally about a decade ago (Margolis 2001), was confirmed in in vivo research and could be attributed to MDMA’s interaction with the serotonergic (5-hydroxytryptamine, 5-HT) transmitter system (Schmidt et al. 2002; Banjaw et al. 2003; Iravani et al. 2003; Bishop et al. 2006; Huot et al. 2011; Lettfuss et al. 2012). The manifold effects of MDMA may be explained by the multitude of its molecular targets. Among these, neurotransmitter transporters play a prominent role, including serotonin (SERT) and norepinephrine (NET) transporters (Battaglia et al. 1988; Rothman et al. 2001; Han and Gu 2006; Verrico et al. 2007). As SERT and NET are central to the regulation of extracellular neurotransmitter concentrations, any interference with their function can influence a
13
138
multitude of ligand-gated conductances and thus neuronal activity and excitability. MDMA is a substrate of these transporters and, once inside the cytosolic compartment, additionally stimulates efflux of neurotransmitter from vesicles into the cytosol. The combination of both effects results in a massive transporter-mediated extrusion of transmitter which is action potential-independent, and therefore qualitatively different from pure antagonism of transporters (Rudnick and Wall 1992; Rothman and Baumann 2002). Furthermore, MDMA also targets a variety of receptors, including the 5-HT 2A receptor (5-HT2A-R) and the 5-HT1A-R (Battaglia et al. 1988; Bhattacharya et al. 1998; Müller et al. 2007; Huot et al. 2011). Despite quite detailed knowledge on the molecular targets of MDMA, there are still gaps in our knowledge of the neuronal circuitry underlying its motor effects. The subthalamic nucleus (STN) is an obvious brain structure to investigate, as it forms part of the basal ganglia and its activity is regulated by 5-HT (Di Matteo et al. 2008). In the light of these facts, we hypothesized that MDMA should modulate neuronal activity of the STN. In particular, if MDMA acted mainly via the 5-HT system, firing rates should on average be elevated, as observed with exogenous application of 5-HT in vitro (Flores et al. 1995; Xiang et al. 2005; Stanford et al. 2005; Shen et al. 2007). We investigated this hypothesis via extracellular recordings of spontaneous neuronal activity in the rat STN in vitro exposed to various concentrations of MDMA. We tested the role of 5-HT receptors, as well as the role of SERT and NET, by coapplying MDMA with the respective receptor and reuptake inhibitors. Our results demonstrate an overall excitatory action of MDMA on STN neuronal activity which depends on both the 5-HT and NE systems, and confirm a remarkable heterogeneity of neurons of the STN.
Exp Brain Res (2015) 233:137–147
into ice-cold modified artificial cerebral spinal fluid (aCSF) containing (in mM): KCl 2.5, NaH2PO4 1.13, MgCl2 10.0, CaCl2 0.5, NaHCO3 26.0, d-glucose 10, and sucrose 230.0, gassed with 95/5 % O2/CO2 (pH 7.4). Next, the cerebellum was removed and the brain was glued to the stage of a vibroslicer (DTK-1000; Dosaka, Kyoto, Japan). Coronal brain slices (300-μm thickness) were cut in ice-cold oxygenated modified aCSF and immediately transferred into a chamber containing normal aCSF of the following composition (in mM): NaCl 120.0, KCl 3.5, NaH2PO4 1.13, MgCl2 1.0, CaCl2 1.2, NaHCO3 26.0, d-glucose 11, gased with 95/5 % O2/CO2 (pH 7.4). The slices were incubated for 1 h at 32 °C before the recordings. Recordings were performed on a standard electrophysiological setup with a submerged-style recording chamber and an upright microscope (Axioskop 2 FS plus, Zeiss, Jena, Germany). The recording chamber was perfused with oxygenated aCSF at a rate of 2 ml/min. Bath temperature was maintained at 32 °C. The STN was identified as a lens-shaped dense patch located medial to the internal capsule (AP from −3.60 to −4.30 mm relative to bregma (Paxinos and Watson 1998). Electrodes were pulled from borosilicate glass and had impedances of 3–5 MΩ when filled with aCSF. Under visual control, the electrodes were inserted into the STN and advanced until single-unit action potentials with amplitudes well above base line noise were visible (Fig. 1a). As a rule, two electrodes per slice were recorded from. Signals were amplified with a MultiClamp 700A amplifier and digitized by a Digidata 1440A digitizer (Molecular Devices, Sunnyvale, CA, USA) at a sampling rate of 20 kHz using pCLAMP 10 (Molecular Devices, Sunnyvale, CA, USA). For each condition including control (absence of drug at the outset of the experiment), data were continuously recorded for 90 s. Data analysis
Materials and methods Animals Forty-two male, treatment naïve, Sprague–Dawley rats of 30–40 days of age, purchased from Charles River (Sulzbach, Germany), were used in this study. Animals were housed in groups of four in an animal housing facility with food and water available ad libitum. A 12/12-h light/dark cycle was used with lights on at 07.00 A.M. Slice preparation and recording procedure All procedures were in accordance with the animal care committee, Eberhard-Karls Universität, Tübingen. The animals were deeply anaesthetized with isoflurane (≥3 %) and decapitated. The brain was quickly removed and placed
13
The raw data were passed through a digital butterworth high pass filter with a −3 dB cut-off frequency of 200 Hz. Action potentials were detected using a simple threshold algorithm. The mean firing rate, the primary analysis parameter used here, was computed as the number of action potentials divided by the recording time (90 s). For a depiction of drug-induced changes of firing rates over time (Fig. 2a), mean firing rates were computed in consecutive nonoverlapping time intervals of 10 s. In order to express the dependence of mean firing rates on MDMA concentration, the data were normalized to the control condition. In the proportional method, firing rates during drug conditions were divided by the firing rate in the absence of drug at the beginning of the experiment. Expressed this way, the magnitude of the effect of MDMA depended to some degree on the basal firing rate of the neurons (Fig. 1d, e);
Exp Brain Res (2015) 233:137–147
139
a
Firing rate (Hz)
a
MDMA
30 20 10 0
b
5-HT
40
0
10
b neuron 1
control
20 30 Time (min)
5-HT
wash
40
MDMA
50
wash
neuron 2
c
40 20 0
d
e
norm. effect MDMA
e
5
0
10
ct rl 5− H T w as M h D M A w as h
Firing rate (norm)
d
ct rl 5− H T w as M h D M A w as h
Firing rate (Hz)
c
r=0.82
4 3 2 1 0 0
Fig. 1 Basic properties of spontaneous neuronal firing in the STN and effects of MDMA thereon. a Exemplary extracellular recording from a pair of electrodes in an acute slice of rat STN under control conditions, high pass filtered at 200 Hz. The spike trains had CVISI of 0.21 (upper trace) and 0.11 (lower trace). Spike waveforms (overlays of 100 cut-outs per electrode) are depicted on the right. b Distributions of firing rates (left) and CVISI (right) under control condition of the total sample of neurons investigated in this study (n = 127). For better visibility, maximal abscissa value was set to 1.5, cutting off five neurons with CVISI >1.5. c Concentration–response plot of firing rates normalized to control (firing rates during presence of drug divided by firing rates in the absence of drug). Depicted are the means (filled black circles), medians (horizontal thick lines), upper and lower quartiles (rectangles) and range (whiskers) of the data in each group; overlaid open circles are data from individual experiments. Numbers on top specify the number of cells excited by MDMA/total number of cells. d Plot of firing rates with MDMA, normalized to control, versus the firing rate under control conditions. Data for all concentrations of MDMA were pooled. e Same plot as in d, but only for data acquired with 5 μM MDMA. Dotted line separates neurons excited by MDMA (black squares) from those inhibited by MDMA (open squares)
1 2 3 4 5 norm. effect 5−HT
Fig. 2 Modulation of firing rates by 5-HT and MDMA. a Time course plot of firing rates of two neurons recorded in parallel in one STN slice and exposed sequentially to 5 μM 5-HT and 5 μM MDMA. Bin width 10 s. b Raster plots of exemplary action potential activity patterns for both neurons at times indicated by open triangles in a (6, 0.75 s per line). Note tendency of neuron 2 to produce bursts of action potentials in the presence of 5-HT and MDMA. c Firing rates of 18 neurons sequentially exposed to 5-HT and MDMA (5 μM each) as shown in a. Grey circles connected by lines are individual neurons, and black circles and line depict means and standard deviations, respectively. Ordinate was clipped to 45 Hz (missing data point corresponds to a neuron firing at 67 Hz in 5-HT condition). d Same data as in c, but normalized to the average of control and first wash condition. Neurons excited by 5-HT are plotted as black, upward pointing triangles; neurons inhibited by 5-HT are depicted by grey triangles pointing downwards. Note logarithmic scale of ordinate. e Plot of the effects of MDMA versus the effects of 5-HT (same data as in d). Spearman’s rank correlation coefficient was 0.82
therefore, we also report selected results as firing rate differences (i.e. firing rate during the control condition was subtracted from firing rate during drug application). The dependence of mean firing rates on MDMA concentration
13
140
Exp Brain Res (2015) 233:137–147
a
b
d
e
WAY100,635
c
WAY100,135
Fig. 3 Modulation of firing rates by MDMA (5 μM in all cases) coapplied with 5-HT receptor antagonists and transporter blockers. a MDL 11,939 (0.25 μM), a 5-HT2A receptor antagonist, was pre-applied before additional application of MDMA. Grey open circles are data from individual neurons (n = 7); full black circles and error bars represent means and standard deviations, respectively. This scheme applies to all graphs. b The 5-HT1A receptor antago-
nist WAY 100,635 (1 μM) was applied before coapplication with MDMA (5 μM). n = 9. c WAY 100,135, an alternative 5-HT1A receptor antagonist (0.5 μM), applied before coapplication with MDMA (n = 13). d The SERT antagonist citalopram (1 μM), applied before coapplication with MDMA (n = 11). e The NET antagonist desipramine (3 μM) applied before coapplication with MDMA (n = 9)
was expressed via a linear regression y = a + bx on normalized data. We used a nonparametric linear regression method (Theil-Sen), which is more robust than standard parametric (least-squares) regression when the underlying data are variable (Hollander et al. 2014). Regularity of the firing patterns was quantified by the coefficient of variation (standard deviation divided by the mean) of inter-spike intervals (CVISI).
one (all variability in the data are completely explained by drug treatment). For focused comparisons of two groups, we used standardized simple contrasts, termed gΨ. gΨ was computed as the difference between the means of two groups, divided by the within-groups standard deviation pooled across all groups in the analysis: gΨ = √m1−m2 , MSerror where m1 and m2 are the two means, and MSerror is the pooled within-groups mean squared error. gΨ can attain any value; the further its value deviates from the ‘null effect’ value of zero, the stronger the effect. The computation of CI95 took account of whether the data were independent (as in Fig. 4) or repeated measures (as in Fig. 3). Note that CI95 which do not include zero imply p
RESEARCH ARTICLE
MDMA modulates spontaneous firing of subthalamic nucleus neurons in vitro Luise Liebig · Andreas von Ameln‑Mayerhofer · Harald Hentschke
Received: 5 January 2014 / Accepted: 5 September 2014 / Published online: 19 September 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract 3,4-Methylene-dioxy-N-methylamphetamine (MDMA, ‘ecstasy’) has a broad spectrum of molecular targets in the brain, among them receptors and transporters of the serotonergic (5-hydroxytryptamine, 5-HT) and noradrenergic systems. Its action on the serotonergic system modulates motor systems in rodents and humans. Although parts of the basal ganglia could be identified as mediators of the motor effects of MDMA, very little is known about the role of the subthalamic nucleus (STN). Therefore, this study investigated the modulation of spontaneous action potential activity of the STN by MDMA (2.5–20 µM) in vitro. MDMA had very heterogeneous effects, ranging from a complete but reversible inhibition to a more than twofold increase in firing at 5 µM. On average, MDMA excited STN neurons moderately, but lost its excitatory effect in the presence of the 5-HT2A antagonist MDL 11,939. 5-HT1A receptors did not appear to play a major role. Effects of MDMA on transporters for serotonin (SERT) and norepinephrine (NET) were investigated by coapplication of the reuptake inhibitors citalopram and desipramine, respectively. Similar to the effects of 5-HT2A receptor blockade, antagonism of SERT and NET bestowed an inhibitory effect on MDMA. From these results, we L. Liebig · H. Hentschke (*) Experimental Anaesthesiology Section, University Hospital Tübingen, Waldhörnlestr. 22, 72072 Tübingen, Germany e-mail: harald.hentschke@uni‑tuebingen.de L. Liebig · A. von Ameln‑Mayerhofer Neuropharmacology, Institute of Neurobiology, University of Tübingen, Tübingen, Germany L. Liebig Graduate School of Neural and Behavioural Sciences, International Max‑Planck Research School, University of Tübingen, Tübingen, Germany
conclude that both the 5-HT and the noradrenergic system mediate MDMA-induced effects on STN neurons. Keywords Ecstasy · Firing rate · Basal ganglia · Serotonin · Noradrenaline
Introduction 3,4-Methylene-dioxy-N-methylamphetamine (MDMA, ‘ecstasy’) is a synthetic illicit drug mostly known for its recreational use. MDMA has a wide spectrum of psychological and physiological effects, including weak rewarding and addicting potentials in animals and humans (Green et al. 2003; Reinhard and Wolffgramm 2006). Alongside a controversial discussion of the beneficial versus harmful potential of MDMA (Green et al. 2012; Parrott 2012), the effects of the drug continue to be investigated, e.g. on cognition (Mithoefer et al. 2013; Nelson et al. 2013) and the motor system (Baumann et al. 2008). Specifically, an antidyskinetic effect of MDMA, reported anecdotally about a decade ago (Margolis 2001), was confirmed in in vivo research and could be attributed to MDMA’s interaction with the serotonergic (5-hydroxytryptamine, 5-HT) transmitter system (Schmidt et al. 2002; Banjaw et al. 2003; Iravani et al. 2003; Bishop et al. 2006; Huot et al. 2011; Lettfuss et al. 2012). The manifold effects of MDMA may be explained by the multitude of its molecular targets. Among these, neurotransmitter transporters play a prominent role, including serotonin (SERT) and norepinephrine (NET) transporters (Battaglia et al. 1988; Rothman et al. 2001; Han and Gu 2006; Verrico et al. 2007). As SERT and NET are central to the regulation of extracellular neurotransmitter concentrations, any interference with their function can influence a
13
138
multitude of ligand-gated conductances and thus neuronal activity and excitability. MDMA is a substrate of these transporters and, once inside the cytosolic compartment, additionally stimulates efflux of neurotransmitter from vesicles into the cytosol. The combination of both effects results in a massive transporter-mediated extrusion of transmitter which is action potential-independent, and therefore qualitatively different from pure antagonism of transporters (Rudnick and Wall 1992; Rothman and Baumann 2002). Furthermore, MDMA also targets a variety of receptors, including the 5-HT 2A receptor (5-HT2A-R) and the 5-HT1A-R (Battaglia et al. 1988; Bhattacharya et al. 1998; Müller et al. 2007; Huot et al. 2011). Despite quite detailed knowledge on the molecular targets of MDMA, there are still gaps in our knowledge of the neuronal circuitry underlying its motor effects. The subthalamic nucleus (STN) is an obvious brain structure to investigate, as it forms part of the basal ganglia and its activity is regulated by 5-HT (Di Matteo et al. 2008). In the light of these facts, we hypothesized that MDMA should modulate neuronal activity of the STN. In particular, if MDMA acted mainly via the 5-HT system, firing rates should on average be elevated, as observed with exogenous application of 5-HT in vitro (Flores et al. 1995; Xiang et al. 2005; Stanford et al. 2005; Shen et al. 2007). We investigated this hypothesis via extracellular recordings of spontaneous neuronal activity in the rat STN in vitro exposed to various concentrations of MDMA. We tested the role of 5-HT receptors, as well as the role of SERT and NET, by coapplying MDMA with the respective receptor and reuptake inhibitors. Our results demonstrate an overall excitatory action of MDMA on STN neuronal activity which depends on both the 5-HT and NE systems, and confirm a remarkable heterogeneity of neurons of the STN.
Exp Brain Res (2015) 233:137–147
into ice-cold modified artificial cerebral spinal fluid (aCSF) containing (in mM): KCl 2.5, NaH2PO4 1.13, MgCl2 10.0, CaCl2 0.5, NaHCO3 26.0, d-glucose 10, and sucrose 230.0, gassed with 95/5 % O2/CO2 (pH 7.4). Next, the cerebellum was removed and the brain was glued to the stage of a vibroslicer (DTK-1000; Dosaka, Kyoto, Japan). Coronal brain slices (300-μm thickness) were cut in ice-cold oxygenated modified aCSF and immediately transferred into a chamber containing normal aCSF of the following composition (in mM): NaCl 120.0, KCl 3.5, NaH2PO4 1.13, MgCl2 1.0, CaCl2 1.2, NaHCO3 26.0, d-glucose 11, gased with 95/5 % O2/CO2 (pH 7.4). The slices were incubated for 1 h at 32 °C before the recordings. Recordings were performed on a standard electrophysiological setup with a submerged-style recording chamber and an upright microscope (Axioskop 2 FS plus, Zeiss, Jena, Germany). The recording chamber was perfused with oxygenated aCSF at a rate of 2 ml/min. Bath temperature was maintained at 32 °C. The STN was identified as a lens-shaped dense patch located medial to the internal capsule (AP from −3.60 to −4.30 mm relative to bregma (Paxinos and Watson 1998). Electrodes were pulled from borosilicate glass and had impedances of 3–5 MΩ when filled with aCSF. Under visual control, the electrodes were inserted into the STN and advanced until single-unit action potentials with amplitudes well above base line noise were visible (Fig. 1a). As a rule, two electrodes per slice were recorded from. Signals were amplified with a MultiClamp 700A amplifier and digitized by a Digidata 1440A digitizer (Molecular Devices, Sunnyvale, CA, USA) at a sampling rate of 20 kHz using pCLAMP 10 (Molecular Devices, Sunnyvale, CA, USA). For each condition including control (absence of drug at the outset of the experiment), data were continuously recorded for 90 s. Data analysis
Materials and methods Animals Forty-two male, treatment naïve, Sprague–Dawley rats of 30–40 days of age, purchased from Charles River (Sulzbach, Germany), were used in this study. Animals were housed in groups of four in an animal housing facility with food and water available ad libitum. A 12/12-h light/dark cycle was used with lights on at 07.00 A.M. Slice preparation and recording procedure All procedures were in accordance with the animal care committee, Eberhard-Karls Universität, Tübingen. The animals were deeply anaesthetized with isoflurane (≥3 %) and decapitated. The brain was quickly removed and placed
13
The raw data were passed through a digital butterworth high pass filter with a −3 dB cut-off frequency of 200 Hz. Action potentials were detected using a simple threshold algorithm. The mean firing rate, the primary analysis parameter used here, was computed as the number of action potentials divided by the recording time (90 s). For a depiction of drug-induced changes of firing rates over time (Fig. 2a), mean firing rates were computed in consecutive nonoverlapping time intervals of 10 s. In order to express the dependence of mean firing rates on MDMA concentration, the data were normalized to the control condition. In the proportional method, firing rates during drug conditions were divided by the firing rate in the absence of drug at the beginning of the experiment. Expressed this way, the magnitude of the effect of MDMA depended to some degree on the basal firing rate of the neurons (Fig. 1d, e);
Exp Brain Res (2015) 233:137–147
139
a
Firing rate (Hz)
a
MDMA
30 20 10 0
b
5-HT
40
0
10
b neuron 1
control
20 30 Time (min)
5-HT
wash
40
MDMA
50
wash
neuron 2
c
40 20 0
d
e
norm. effect MDMA
e
5
0
10
ct rl 5− H T w as M h D M A w as h
Firing rate (norm)
d
ct rl 5− H T w as M h D M A w as h
Firing rate (Hz)
c
r=0.82
4 3 2 1 0 0
Fig. 1 Basic properties of spontaneous neuronal firing in the STN and effects of MDMA thereon. a Exemplary extracellular recording from a pair of electrodes in an acute slice of rat STN under control conditions, high pass filtered at 200 Hz. The spike trains had CVISI of 0.21 (upper trace) and 0.11 (lower trace). Spike waveforms (overlays of 100 cut-outs per electrode) are depicted on the right. b Distributions of firing rates (left) and CVISI (right) under control condition of the total sample of neurons investigated in this study (n = 127). For better visibility, maximal abscissa value was set to 1.5, cutting off five neurons with CVISI >1.5. c Concentration–response plot of firing rates normalized to control (firing rates during presence of drug divided by firing rates in the absence of drug). Depicted are the means (filled black circles), medians (horizontal thick lines), upper and lower quartiles (rectangles) and range (whiskers) of the data in each group; overlaid open circles are data from individual experiments. Numbers on top specify the number of cells excited by MDMA/total number of cells. d Plot of firing rates with MDMA, normalized to control, versus the firing rate under control conditions. Data for all concentrations of MDMA were pooled. e Same plot as in d, but only for data acquired with 5 μM MDMA. Dotted line separates neurons excited by MDMA (black squares) from those inhibited by MDMA (open squares)
1 2 3 4 5 norm. effect 5−HT
Fig. 2 Modulation of firing rates by 5-HT and MDMA. a Time course plot of firing rates of two neurons recorded in parallel in one STN slice and exposed sequentially to 5 μM 5-HT and 5 μM MDMA. Bin width 10 s. b Raster plots of exemplary action potential activity patterns for both neurons at times indicated by open triangles in a (6, 0.75 s per line). Note tendency of neuron 2 to produce bursts of action potentials in the presence of 5-HT and MDMA. c Firing rates of 18 neurons sequentially exposed to 5-HT and MDMA (5 μM each) as shown in a. Grey circles connected by lines are individual neurons, and black circles and line depict means and standard deviations, respectively. Ordinate was clipped to 45 Hz (missing data point corresponds to a neuron firing at 67 Hz in 5-HT condition). d Same data as in c, but normalized to the average of control and first wash condition. Neurons excited by 5-HT are plotted as black, upward pointing triangles; neurons inhibited by 5-HT are depicted by grey triangles pointing downwards. Note logarithmic scale of ordinate. e Plot of the effects of MDMA versus the effects of 5-HT (same data as in d). Spearman’s rank correlation coefficient was 0.82
therefore, we also report selected results as firing rate differences (i.e. firing rate during the control condition was subtracted from firing rate during drug application). The dependence of mean firing rates on MDMA concentration
13
140
Exp Brain Res (2015) 233:137–147
a
b
d
e
WAY100,635
c
WAY100,135
Fig. 3 Modulation of firing rates by MDMA (5 μM in all cases) coapplied with 5-HT receptor antagonists and transporter blockers. a MDL 11,939 (0.25 μM), a 5-HT2A receptor antagonist, was pre-applied before additional application of MDMA. Grey open circles are data from individual neurons (n = 7); full black circles and error bars represent means and standard deviations, respectively. This scheme applies to all graphs. b The 5-HT1A receptor antago-
nist WAY 100,635 (1 μM) was applied before coapplication with MDMA (5 μM). n = 9. c WAY 100,135, an alternative 5-HT1A receptor antagonist (0.5 μM), applied before coapplication with MDMA (n = 13). d The SERT antagonist citalopram (1 μM), applied before coapplication with MDMA (n = 11). e The NET antagonist desipramine (3 μM) applied before coapplication with MDMA (n = 9)
was expressed via a linear regression y = a + bx on normalized data. We used a nonparametric linear regression method (Theil-Sen), which is more robust than standard parametric (least-squares) regression when the underlying data are variable (Hollander et al. 2014). Regularity of the firing patterns was quantified by the coefficient of variation (standard deviation divided by the mean) of inter-spike intervals (CVISI).
one (all variability in the data are completely explained by drug treatment). For focused comparisons of two groups, we used standardized simple contrasts, termed gΨ. gΨ was computed as the difference between the means of two groups, divided by the within-groups standard deviation pooled across all groups in the analysis: gΨ = √m1−m2 , MSerror where m1 and m2 are the two means, and MSerror is the pooled within-groups mean squared error. gΨ can attain any value; the further its value deviates from the ‘null effect’ value of zero, the stronger the effect. The computation of CI95 took account of whether the data were independent (as in Fig. 4) or repeated measures (as in Fig. 3). Note that CI95 which do not include zero imply p
