Address correspondence and reprint requests to Dr Francisco J. Urbano, Instituto de Fisiología, Biología Molecular y Neurociencias (IFIBYNE- CONICET-UBA), Intendente Guiraldes 2670, pabellón 2, piso 2, Ciudad Universitaria, (C1428BGA)-Buenos Aires, Argentina. E-mail: email@example.com
Methylphenidate (MPH) is widely used to treat children and adolescents diagnosed with attention deficit/hyperactivity disorder. Although MPH shares mechanistic similarities to cocaine, its effects on GABAergic transmission in sensory thalamic nuclei are unknown. Our objective was to compare cocaine and MPH effects on GABAergic projections between thalamic reticular and ventrobasal (VB) nuclei. Mice (P18-30) were subjected to binge-like cocaine and MPH acute and sub-chronic administrations. Cocaine and MPH enhanced hyperlocomotion, although sub-chronic cocaine-mediated effects were stronger than MPH effects. Cocaine and MPH sub-chronic administration altered paired-pulse and spontaneous GABAergic input differently. The effects of cocaine on evoked paired-pulse GABA-mediated currents changed from depression to facilitation with the duration of the protocols used, while MPH induced a constant increase throughout the administration protocols. Thalamic reticular nucleus GAD67 and VB CaV3.1 protein levels were measured using western blot to better understand their link to increased GABA release. Both proteins were increased by sub-chronic administration of cocaine. MPH showed effects on GABAergic transmission that seems less disruptive than cocaine. Unique effects of cocaine on postsynaptic VB calcium currents might explain deleterious cocaine effects on sensory thalamic nuclei. These results suggest that cocaine and MPH produced distinct presynaptic alterations on GABAergic transmission.
6-Cyano-7-nitroquinoxaline-2,3-dione disodium salt hydrate
glutamic acid decarboxylase
high voltage activated
inhibitory post-synaptic current
low voltage activated
thalamic reticular nucleus
Chronic abuse of cocaine is associated with major neuro-psychiatric conditions (Devlin and Henry 2008). Acute binge-like administration of cocaine was able to alter the intrinsic properties of thalamocortical neurons and spontaneous GABAergic transmission, resulting in enhancements of EEG low frequency activity in mice (Urbano et al. 2009). Systemic administration of T-type calcium channel blockers in vivo prevented hyperlocomotion and GABAergic neurotransmission enhancement onto Ventrobasal (VB) neurons after acute binge-like cocaine administration (Bisagno et al. 2010), suggesting a key role for T-type channels in cocaine effects on specific thalamic GABAergic networks.
The thalamic reticular nucleus (TRN) is a thin layer of GABAergic neurons that project to sensory thalamic nuclei (Spreafico et al. 1991), and its cells are interconnected by GABAergic terminals (Sun et al. 2012) and gap junctions (Landisman et al. 2002). In rodents, there is a lack of GABAergic interneurons in the VB nucleus, and the inhibition necessary for proper sensory perception is provided by GABAergic TRN afferents (De Biasi et al. 1997). TRN neurons have intrinsic properties that allow them to generate action potentials and membrane potential oscillations at a wide range of frequencies (reviewed by Steriade 2005). TRN neurons express CaV3.2 and CaV3.3 T-type calcium channel subunits (Talley et al. 1999), although recent studies have also confirmed the presence of CaV3.1 subunits (Kovács et al. 2010). TRN rhythmicity is modulated by monoamines and GABA (Pinault and Deschênes 1992; Shammah-Lagnado et al. 1996; Rutter et al. 1998; Rodríguez et al. 2011).
Methylphenidate (MPH), another psychostimulant that has some abuse liability (Chait 1994), is widely used to treat children and adolescents diagnosed with attention deficit/hyperactivity disorder (ADHD) (Biederman et al. 1999). In humans, MPH has reinforcing effects (associated with increased extracellular dopamine levels by blocking the dopamine transporter, DAT) after intravenous administration (Volkow et al. 1999a). Lowered predisposition to drug abuse during adulthood has been described after early exposure to MPH in humans (Biederman et al. 1999) and in animal models (Carlezon et al. 2003). However, other authors have suggested otherwise (Brandon et al. 2001; Volkow and Insel 2003; Volkow et al. 1995). Differences in pharmacodynamics between cocaine and MPH in humans have been associated with the lack of cross-sensitization of pre-adolescent MPH use (Guerriero et al. 2006). MPH administration normalized EEG low frequency activity (Clarke et al. 2003), suggesting direct involvement of the TRN in the etiology of ADHD. Single MPH administration has been shown to block GABAergic transmission from the TRN through the activation of D4 receptors both in vitro (Florán et al. 2004) and in vivo (Erlij et al. 2012). Nevertheless, the effects of repetitive MPH administration on GABAergic transmission between sensory thalamic nuclei remain unknown.
Cocaine has been shown to inhibit monoamine transporters (DAT, SERT, and NET), elevating synaptic levels of dopamine (Ritz et al. 1987; Wise and Bozarth 1987; Howes et al. 2000), norepinephrine, and serotonin (Glowinski and Axelrod 1966; Ross and Renyi 1969; Pan et al. 1994; Howes et al. 2000). MPH mainly inhibits DAT and NET, but not SERT, inducing rapid increases in extracellular dopamine levels in the basal ganglia (Kuczenski and Segal 1997). MPH affinity for DAT in vivo is comparable to that of cocaine (Volkow et al. 1999b), but whole-brain dopamine kinetics mediated by MPH are slower than those of cocaine (Volkow et al. 1999a; Volkow and Swanson 2003). In adult ADHD patients, MPH increased dopamine in the ventral striatum while reducing their symptoms (Volkow et al. 2012).
The aim of this study was to compare the effects in mice of binge-like cocaine and MPH acute (1 day) and sub-chronic (3 day) administration on locomotor activity and GABAergic transmission from the TRN onto VB neurons. Our results showed that both cocaine and MPH enhanced hyperlocomotion, although cocaine-mediated effects were stronger than MPH after sub-chronic administration. Both cocaine and MPH changed paired-pulse evoked and spontaneous GABAergic transmission from TRN. While cocaine drastically increased paired-pulse ratios only 24 h after 3-day, MPH enhanced them from 1 day up to 3 day administrations. Cocaine induced a greater spontaneous GABA minis frequency compared to MPH after 1 day, but not for the 3-day administrations. The effects of cocaine on thalamic GABAergic transmission and post-synaptic calcium currents observed in this study could underlie drastic alterations in the protein expression of GAD and/or post-synaptic T-type channels. Western blot analysis revealed an increase in CaV3.1 and GAD67 levels after sub-chronic administration of cocaine.
Our results suggest a considerable dysregulation of thalamic GABAergic transmission and post-synaptic calcium currents by cocaine, which might underlie its long-lasting neurotoxic effects. Also, MPH induced steady-state alterations of GABAergic transmission changes, which would result in long-lasting changes in sensory thalamic processing.
Materials and methods
Overall, 18–30-days-old male C57BL/6 mice from the Central Animal Facility at University of Buenos Aires were used. Principles of animal care were in accordance with CONICET (2003), and approved by its authorities using OLAW/ARENA directives (NIH, Bethesda, MD, USA).
Cocaine and methylphenidate were administered i.p. using ‘binge-like’ protocols (Spangler et al. 1993) for 1 day (3 x 15 mg/kg, 1 h apart) or for 3 day (sub-chronic; 3 x 15 mg/kg, 1 h apart, each day for 3 days) (Fig. 1a). Control animals received saline injections equally timed.
Thalamocortical slices were obtained as previously described (Bisagno et al. 2010), 1 h after last ‘binge-like’ administration, except when 24 h after 3 day was used. Slices were allowed to recover at 35°C for at least 30 min in a psychostimulant-free ACSF (Bisagno et al. 2010).
Whole-cell patch-clamp recordings
Recordings were made at 20–24°C using patch-clamp electrodes (Bisagno et al. 2010). Inhibitory post-synaptic miniatures currents (minis) were recorded in the presence of TTX (3 μΜ) and analyzed using Mini Analysis (Synaptosoft, Fort Lee, NJ, USA). Inter-event interval curves were fitted to a single exponential equation (SigmaPlot; Systat Softwares, Chicago, IL, USA), and median mini intervals were compared across groups.
Inhibitory post-synaptic currents (IPSCs) were evoked in the presence of DL-AP5 (50 μM) and CNQX (20 μΜ) (40–200 μs; 200–1000 μA) using bipolar electrodes (FHC Inc, Bowdoin, ME, USA) positioned at the boundary between the TRN and VB nuclei (Zhang et al. 1997). Bicuculline (50 μΜ) was used to confirm that currents were mediated by GABA-A receptors (Fig. 1b). Paired pulses at 10 Hz or 40 Hz were used, and IPSCs were normalized to the amplitude of first IPSC. Voltage-dependent calcium currents were recorded with a ramp-like protocol (0.3 mV/ms; Bisagno et al. 2010). Voltage ramps reduced the rundown of calcium currents, allowing us to accurately calculate low voltage activated (LVA, T-type) and high voltage activated (HVA, mediated by P/Q-type, ω-Agatoxin-IVA-sensitive calcium currents; see Urbano et al. 2009) peak currents ratios (Bisagno et al. 2010). Signals were recorded using a MultiClamp 700 amplifier commanded by pCLAMP 10.0 (Molecular Devices, Sunnyvale, CA, USA). Low concentrations of mibefradil (20 μΜ) were used to block T-type currents (Fig. 1c), as previously described (Bisagno et al. 2010).
Mouse locomotor activity was recorded with an automated system (Ethovision XT7.0; Noldus, Wageningen, The Netherlands) as previously described (Bisagno et al. 2010). Total distance traveled (cm) was quantified for a total of 30 minutes prior to injections (baseline), and 45 min following the last injection of a binge.
Tissue homogenization and western blot
TRN and VB were dissected from 350 μm thick slices on an ice-cold stage, collected in plastic tubes and stored (−80°C). Samples were thawed and homogenized in radioimmunoprecipitation assay (RIPA) buffer containing protease inhibitors at 4°C. After centrifugation (20 min at 21 500 g), protein levels were determined with a BCA protein assay kit (Thermo Scientific, Rockford, IL, USA), and 40 μg from each sample were incubated with cracking buffer (Laemmli 1970) for 10 min at 100°C. Samples were run on a 10% polyacrylamide resolving gel and proteins were transferred to a nitrocellulose membrane (Sigma-Aldrich, St. Louis, MO, USA). The blot was probed with specific primary antibodies, including rabbit anti-GAD67/65 (1 : 10 000, Chemicon, Billerica, MA, USA), rabbit anti-actin (1 : 100, Sigma-Aldrich), and rabbit anti-CaV3.1 (1 : 200, Chemicon). Secondary antibody was anti-rabbit conjugated to horseradish peroxidase (HRP) (1 : 1000, Dako, Glostrup, Denmark). Blots were developed with a chemiluminescent horseradish peroxidase substrate (Immobilon Western, EMD Millipore Co., Billerica, MA, USA), and chemiluminescence was visualized with a CCD camera (LAS-1000; Fujifilm, Tokyo, Japan). Signal intensity was quantified using ImageJ 1.43 m software (http://imagej.nih.gov/ij/index.html, NIH). Bands corresponding to GAD67 and CaV3.1 were normalized to actin, and all samples were normalized to the saline group's mean.
InfoStat software (Univ. Nacional de Córdoba, Argentina) was used for statistical comparisons. Statistics were performed using Student's t-test or one-way anova (unless otherwise stated) and Tukey-Kramer or LSD Fisher multiple comparisons post hoc tests when applicable. Differences were considered significant if p <0.05. Whenever the data did not comply with assumptions of the parametric tests, non-parametric Wilcoxon-Mann-Whitney or Kruskal–Wallis tests were performed followed by paired comparisons. Data presented as mean ± standard error of the mean.
Cocaine-HCl was purchased from Sigma-Aldrich, and methylphenidate-HCl (Mallinckrodt Inc., Hazelwood, MO, USA) was a generous donation from Osmotica Pharmaceuticals S.A. (Buenos Aires, Argentina). During electrophysiological recordings, the following drugs were used: DL-AP5, CNQX, TTX, bicuculline, and mibefradil (all from Sigma-Aldrich).
Repetitive methylphenidate (MPH) binge administration induced milder changes in hyperlocomotion than cocaine
We initially compared the effects of cocaine and MPH on locomotor activation after 1 day acute binge administration (Fig. 2a). Cocaine and MPH induced higher hyperlocomotion compared with saline but did not differ from each other (Fig. 2a; Kruskal-Wallis anova, H = 11.08, p <0.05; cocaine vs. saline: p <0.001; MPH vs. saline: p <0.05; cocaine vs. MPH: p >0.05). After 3 day binge administrations, both cocaine and MPH induced hyperlocomotion (Fig. 2b; anova, F(2,25) = 17.88, p <0.0001; LSD Fisher post hoc test: all treatments are statistically different from each other with p <0.01). Higher hyperlocomotion was observed after 3 day cocaine binge administration compared to the responses mediated by a 1 day cocaine binge; however, MPH-administered mice showed similar hyperlocomotion after 1 day and 3 day administration (3–1 day values = saline: 1690.62 ± 2149.19; cocaine: 7914.82 ± 1519.71; MPH: 630.05 ± 1754.81. anova, F(2,24) = 5.73, p <0.01; LSD Fisher post hoc test: p <0.05, cocaine vs. saline, MPH). No cocaine or MPH-mediated effects on hyperlocomotion were observed 24 h after the last injection (Fig. 2c, anova, p >0.05).
Cocaine showed higher frequencies of spontaneous GABAergic minis than MPH while only cocaine altered post-synaptic low voltage activated (LVA)/high voltage activated (HVA) calcium current ratios
GABAergic minis (mIPSCs) recorded from VB neurons after an acute cocaine binge manifested higher frequencies compared with MPH and saline treatments (Fig. 3a and b; Kruskal-Wallis anova, cocaine vs. MPH: H = 6.85, p <0.01; cocaine vs. saline: H = 15.75 p <0.05). Mini intervals were not significantly different when comparing saline and MPH-treated slices (Fig. 3a and b; p >0.05). After 3 day sub-chronic protocols, cocaine and MPH treatments showed higher frequencies than saline (Fig. 3c and d; Kruskal-Wallis anova, H = 9.9, p <0.01), while no differences were observed across groups 24 h after the last 3 day sub-chronic injection (Fig. 3e and f; p >0.05).
One MPH binge did not change post-synaptic calcium current LVA/HVA ratios in VB neurons when compared to saline, while ratios of 1 day cocaine-treated animals were significantly higher than for either saline or MPH (Fig. 3g; One-way anova, F(2,38) = 7.6, Tukey-Kramer post hoc test; saline vs. MPH, p >0.05; cocaine vs. MPH: p <0.05; cocaine vs. saline: p <0.05). No changes in LVA/HVA ratios were observed between saline and 3 day (1 h after) cocaine and MPH treatments (Fig. 3h; anova, p >0.05). Lower ratios were observed 24 h after 3 day cocaine binge administration (Fig. 3i, One-way anova, F(2,29) = 5.8, p <0.01), related to higher HVA, P/Q-type mediated current density without changes in T-type current density (253 ± 18% increment 24 h after 3 day vs. saline, n = 10; Kruskal-Wallis anova; H = 13.9, p <0.01). No changes in LVA/HVA ratios were observed after repetitive MPH treatments compared to saline (Fig. 3h and i; p >0.05).
Cocaine and MPH differentially affected paired-pulse evoked GABAergic transmission
Paired-pulse ratios (PPRs; 2nd stimulus-evoked amplitude/1st stimulus-evoked amplitude) are widely accepted as a parameter to characterize pre-synaptic-dependent alterations. We compared ratios using both 10 Hz and 40 Hz frequencies of stimulation across all treatments. Mean PPR values were lower than one indicating that there was synaptic depression between stimuli. After 1 day binge treatment, 10 Hz ratios were not significantly different across treatments (Fig. 4a, Kruskal-Wallis anova, p >0.05). However, compared to cocaine and saline, 40 Hz PPRs from the MPH group were higher both in control conditions (Fig. 4a, Kruskal-Wallis anova, p <0.05) and after bath application of mibefradil (20 μM; 20–40 min; Fig. 4b, Kruskal-Wallis anova, p <0.05), suggesting no significant involvement of T-type channels on paired-pulse GABA release after 1 day binge treatment. We continued characterizing the effects of cocaine and MPH on evoked GABAergic transmission in mice after sub-chronic administration protocols. Again, MPH elicited higher PPRs than saline and cocaine at both frequencies tested (Fig. 4c; 10 Hz: anova F(2,61) = 3.79, p <0.05; Tukey-Kramer post hoc test, MPH vs. saline, cocaine p <0.05; 40 Hz: anova F(2,48) = 8.64, p <0.01; Tukey-Kramer post hoc test, MPH vs. saline, cocaine, p <0.05). Cocaine did not change PPRs 1 h after binge compared to saline at either 10 Hz or 40 Hz (Fig. 4C; anova, p >0.05), while PPRs were significantly higher than MPH and saline 24 h after 3 day cocaine binge treatment for both 10 Hz and 40 Hz (Fig. 4d; 10 Hz: anova F(2,42) = 20.27 p <0.01; Tukey-Kramer post hoc test, MPH vs. cocaine, p <0.05, and saline vs. cocaine, p <0.01; 40 Hz: anova F(2,34) = 18.8 p <0.01; Tukey-Kramer post hoc test, MPH, vs. cocaine, p <0.05 and saline vs. cocaine, p <0.01). Importantly, 40 Hz PPRs 24 h after 3 day cocaine binge treatment (Fig. 4d) showed ratios surpassing the threshold of 1.0, indicating pure facilitation during GABA transmission at high frequency. In the presence of mibefradil (20 μM), 24 h after 3 day cocaine binge treatment 10 Hz and 40 Hz PPR values were significantly reduced to saline levels (10 Hz: 0.61 ± 0.06, n = 15; 40 Hz: 0.74 ± 0.06, n = 13; Tukey-Kramer post hoc test, 24-h after 3 day without mibefradil vs. with mibefradil, p <0.05; saline vs. 24-h after 3 day in the presence of mibefradil, p >0.05). MPH ratios 24 h after 3 day administration protocols were higher than saline only at 10 Hz (Tukey-Kramer post hoc test, saline vs. MPH, p <0.05). PPRs after MPH administration were not significantly different comparing 1 h (Fig. 4c) and 24 h after 3 day binge (Fig. 4d).
In conclusion, MPH treatments increased PPRs compared to saline throughout all administration protocols used, being reversible 24 h after 3 day treatment at 40 Hz stimulation. However, only cocaine induced a rebound in PPR values 24 h after 3 day after either 10 Hz or 40 Hz stimulation (Fig. 4e). Mibefradil reduced higher PPR values observed 24 h after 3 day cocaine treatment.
Cocaine increased thalamic CaV3.1 protein levels
Cocaine effects on GABAergic PPRs and on LVA/HVA current ratios as well as mibefradil-mediated reduction in PPR 24 h after 3 day cocaine binge treatment suggested the existence of transient changes in TRN synaptic GAD67 or VB CaV3.1 protein levels. Fig. 5A shows GAD67 protein levels (measured by western blot) in the TRN 1 h and 24 h after 3 day binge protocol. No statistically significant differences were observed between cocaine- and saline-treated mice (p >0.05). On the other hand, CaV3.1 protein levels in VB nucleus were significantly higher 24 h after 3 day binge protocols compared to saline (Fig. 5b; Student's t-test, t = 4.0, p <0.05).
The results presented here show distinct alterations by sub-chronic binge-like administrations of either cocaine or MPH on hyperlocomotion, pre-synaptic modulation of GABAergic transmission and post-synaptic calcium currents from sensory thalamic nuclei. Our results suggest that cocaine effects were more robust than MPH and varied according to different administration protocols. MPH steadily affected evoked GABAergic transmission.
Cocaine and MPH differentially affected hyperlocomotion
Cocaine- and MPH-mediated rapid enhancement in locomotion in rodents has been correlated with their ability to increase extracellular dopamine and norepinephrine levels in nucleus accumbens and caudate-putamen (Kuczenski and Segal 1992, 2001; Segal and Kuczenski 1999). Unlike cocaine, MPH fails to increase extracellular serotonin levels (Kuczenski and Segal 1997; Segal and Kuczenski 1999) because of its weak binding affinity for SERT (Pan et al. 1994; Gatley et al. 1996). MPH (i.p.) has been extensively used in mice at the same concentration range as cocaine (Kuczenski and Segal 2001; Drerup et al. 2010; Thanos et al. 2010). It is agreed that MPH and cocaine might share neuronal pathways to exert their effects (Volkow et al. 1999b; Argento et al. 2012). Here, binge-like MPH administration, similar to non-prescribed, repetitive MPH self-administration described in adolescents (Morton and Stockton 2000), was compared to binge-like cocaine, showing no change in hyperlocomotion between days of treatment, although enhancing hyperlocomotion above saline levels (Drerup et al. 2010). MPH-induced hyperlocomotion was insensitive to the T-type calcium channel blocker 2-octanol (0.07 mg/kg, i.p.; p >0.05; data not shown). Our group previously reported that co-administration of systemic administration of T-type calcium channels blockers strongly reduced hyperlocomotion induced by one acute cocaine binge (Bisagno et al. 2010). It has also been described that MPH increased dopamine levels in the TRN, reducing hyperlocomotion through activation of D4 receptors (Erlij et al. 2012), thus suggesting a limiting process that might explain why MPH repetitive administration did not increase hyperlocomotion levels above acute-mediated levels.
Cocaine- and MPH-induced changes on GABAergic transmission from TRN: monoamine synaptic levels versus intrinsic properties
The effects of cocaine and MPH on GABA release were observed several hours after slicing, which suggests long-lasting effects from multiple basal ganglia/brainstem-TRN interactions (Contreras et al. 1993; Shammah-Lagnado et al. 1996). Nevertheless, PPR values from saline mice were not significantly different either across treatments or frequencies. Frequency-independence of PPRs presented here are in agreement with recent reports using similar PPR testing (Zhang et al. 2010).
It has been described that monoamine receptors can modulate GABA release from Globus Pallidus onto TRN. Indeed, activation of D4 dopaminergic receptors enhances 10 Hz PPRs in Globus Pallidus afferents, without altering GABAergic transmission within TRN (Govindaiah et al. 2010). The fact that MPH is known to activate these receptors in the TRN (Erlij et al. 2012) and that cocaine can also increase monoamine levels in somatosensory thalamic nuclei (Rutter et al. 1998), suggest a more complex mechanism underlying the observed differences in PPRs. Changes in PPR during sub-chronic administrations of MPH affected both 10 Hz and 40 Hz stimulation, suggesting a modulation of TRN neurons at both frequencies as described in vivo (Pinault and Deschênes 1992). After such modulation, TRN neurons might need to recover from direct alteration (blocking/opening) of membrane ionic channels as well as GABAergic transmission after sub-chronic cocaine and MPH administrations (Shoji et al. 1998; Federici et al. 2005). Recent experiments made by our group have confirmed this hypothesis, showing that bath-applied MPH (10 μM) did not have any effect on the firing frequency of TRN neurons, while cocaine (10 μM) strongly reduced frequency of action potentials (data not shown; n = 8). Thus, cocaine sub-chronic treatments might block TRN somatic activity (known to be required for the correct TRN-TRN inter-somatic excitatory activity; [Sun et al. 2012]), explaining the observed mean paired-pulse ‘rebound’ 24 h after 3 day cocaine administration. Milder MPH effects on intrinsic properties of TRN would explain the more modest, but sustained effects on PPR. GABA minis frequency increment by cocaine after one day (and up to three days) would be mediated by its direct effect on pre-synaptic TRN GABAergic terminals, regardless of action potential frequency at TRN somatic levels (i.e., consistently with our group's previous reports in the presence of TTX, [Urbano et al. 2009; Bisagno et al. 2010]).
Monoamine receptors have also been described to modulate intrinsic properties (e.g., T-type calcium channels) of sensory thalamic neurons. T-type calcium channels are involved in distal dendritic calcium transients in TRN neurons helping integrate dendritic GABAergic afferents (Crandall et al. 2010; Sun et al. 2012). It is accepted that only P/Q-type channels are located in both VB dendrites (Pedroarena and Llinás 1997) and TRN synaptic terminals in charge of GABA release (Iwasaki et al. 2000). Interestingly, serotonin, but not dopamine, has been reported to significantly affect T-type calcium currents (Berger and Takahashi 1990; Fraser and MacVicar 1991), suggesting a possible serotonin-based mechanism mediating cocaine-, but not MPH-induced changes in post-synaptic LVA/HVA currents ratios during sub-chronic administrations showed here. Different effects in T-type channels mediated by cocaine are consistent with unchanged MPH hyperlocomotion after 2-Octanol administration, but contrary to the absence of mibefradil effects on PPR after acute binge administration of cocaine.
Another suitable explanation for the observed differences between MPH and cocaine sub-chronic administrations on GABA release PPR might be the existence of a MPH-mediated pre-synaptic inhibition of TRN GABA release (Federici et al. 2005), which would mediate the increment in mean ratio values observed in this study. On the contrary, cocaine would have a stronger blocking effect of intersomatic TRN GABAergic inhibition while simultaneously incrementing spontaneous minis (i.e., directly acting on pre-synaptic terminals), thus presenting probabilities of GABA release in the same range than saline-treated terminals (i.e., an experimental condition characterized by smaller mean PPRs, [Zhang et al. 1997]). Accordingly, a rebound in PPR values 24 h after the 3 day cocaine treatment can be seen as a compensatory mechanism. The involvement of T-type calcium channels on intercellular TRN GABAergic inhibition is consistent with the observed reduction of PPR values 24 h after 3 day administration using bath application of mibefradil. A mibefradil-mediated reduction in low-threshold spikes mediated by T-type channels at TRN somatic level (Crandall et al. 2010; Sun et al. 2012) would reduce GABA release between TRN neurons. Thus, lower inhibition of TRN neurons would allow for an increase in GABA release onto VB neurons, ultimately inducing higher probabilities of GABA release (i.e., provoking depression PPR values). In agreement with this hypothesis, a previous study from Huguenard and Prince (1994) has shown how T-type currents blocker ethosuximide not only reduced burst-firing of TRN neurons, but incremented both evoked and spontaneous GABA release onto VB neurons in slices from pre-adolescent rats. We further tested this hypothesis by comparing PPR values obtained 24 h after 3 day versus a 4 day treatment (animals were killed 1 h after the last injection). The 4 day cocaine administration further reduced PPRs values, although they were significantly higher than saline PPRs (n > 15, saline; n > 23 cocaine; data not shown).
Therefore, our results support the hypothesis of a cocaine-mediated over-stimulation of GABA release by altering TRN inter-somatic inhibition. In addition, dissimilar effects of MPH and cocaine in sensory thalamic nuclei might have a correlation with changes in intracellular calcium concentrations in sensory thalamic neurons, as previously reported in cortical areas (Du et al. 2006). Further studies are needed to characterize intracellular, downstream events that might explain the observed differences between cocaine- and MPH-induced changes in thalamic GABAergic PPR values.
GAD67 and CaV3.1 protein levels
Transient, rebound-like cocaine-mediated effects on thalamic GABAergic transmission and post-synaptic calcium currents could underlie drastic alterations in the protein expression of GAD or post-synaptic CaV3.1 T-type channels. Western blot results presented here about cocaine-induced changes in GAD67 or CaV3.1 protein levels can be considered new, having no precedent study published to the best of our knowledge. Nevertheless, Western blot quantification has been used to report a GAD level increment after cocaine withdrawal in the hypothalamus (Ma et al. 2008). Here, no significantly different levels of GAD67 (predominantly located at a TRN somatic level; [Esclapez et al. 1994]) were observed up to 24 h after 3 day binge administration. However, one extra day of administration did increase GAD67 (data not shown), supporting the idea that cocaine might drastically reduce pre-synaptic GABA levels, leading to higher GAD protein synthesis.
The role of T-type channels in sensory thalamic nuclei has been recently expanded, describing that both pre-synaptic TRN and post-synaptic VB neuronal types share CaV3.1 subunits containing T-type calcium channels (Kovács et al. 2010). In light of these new reports, the observed changes in VB post-synaptic calcium currents ratios (LVA/HVA) throughout cocaine sub-chronic binge-like administrations may be suggesting the existence of a compensatory expression of T-type channels by VB and/or TRN terminals. There was an increment in CaV3.1 subunits expression 24 h after a 3-day treatment, but not at any time tested right after cocaine was administered (including 4-day administration, data not shown). It is worth noticing that plasma cocaine levels were expected to be totally washed out after 24 h.
Both cocaine-mediated magnitude and time-delayed effects in GAD and CaV3.1 protein levels are illustrative of the long lasting, deleterious effects that this stimulant can exert over sensory thalamocortical processing.
Functional implications of sustained cocaine TRN alterations for thalamocortical interactions
Results from this study suggest that cocaine and MPH are able to enhance synaptic GABAergic transmission at both low (10 Hz) and high frequency (40 Hz) stimulation of TRN axons as well as hyperlocomotion after sub-chronic administration protocols. This may result in the abnormal hyperpolarization of VB thalamocortical projecting neurons, leading to thalamocortical recurrent low frequency bursting activity of both TRN (Llinás and Geijo-Barrientos 1988; Huguenard and Prince 1992) and VB neurons (Jahnsen and Llinás 1984a, b; McCormick and Feeser 1990). Prolonged coherence between low-frequency burst-firing and high frequency thalamocortical activity during awake states has been suggested to disrupt sensory processing (McCormick and Feeser 1990), as well as induce alterations in nociception in mice (Liao et al. 2011), which are known to underlie multiple diseases known collectively as thalamocortical dysrhythmia syndrome (Llinás et al. 1999; Jeanmonod et al. 2003). T-type channel over activation at the level of the TRN has been associated with pathophysiological behaviors including epilepsy (Steriade and Llinás 1988; Tsakiridou et al. 1995), a neurological disorder also associated with chronic administration of cocaine, but not MPH (Devlin and Henry 2008). An over activation of TRN-mediated GABAergic transmission would also alter sensory traffic through sensory relay nuclei in the thalamus, a mechanism thought to underlie major EEG abnormalities in ADHD patients (Rowe et al. 2005).
The results described here using sub-chronic protocols, confirm and expand our group's previous findings showing a dysregulation of thalamic GABAergic transmission and post-synaptic calcium current ratios as key mechanisms that might underlie the long-lasting deleterious effects of cocaine. MPH-induced changes in GABAergic transmission using repetitive administration protocols suggest that MPH might also alter sensory processing but in a less disruptive manner. Steady-state alterations by MPH are particularly important to understand the impact of MPH intake either as pharmacotherapy for ADHD patients or in non-prescribed stimulant abuse among healthy users. Future studies using longer, chronic protocols are still needed to determine whether the cocaine- and MPH-mediated effects described here might turn into long-lasting thalamic changes.
The authors thank Dr. Joaquín Piriz and Dr. Mariano Soiza-Reilly for their valuable comments and suggestions, and María Eugenia Martín and Paula Felman for their excellent technical and administrative assistance. Dr. Bisagno has been authorized to study drug-abuse substances in animal models by A.N.M.A.T. (National Board of Medicine Food and Medical Technology, Ministerio de Salud, Argentina). This study was supported by grants from: FONCYT-ANPCyT BID 1728 OC.AR. PICT 2007-1009, PICT 2008-2019 and PIDRI-PRH 2007 (Dr. Urbano), CONICET- PIP 2011-2013-11420100100072 (Dr. Bisagno), and NIH P20 GM103425-09 and R01 NS020246-25 (Dr. Garcia-Rill). The experiments included in this study comply with the current laws of Argentina. Dr. Urbano was a 2011 fellow of the John Simon Guggenheim Memorial Foundation (http://www.gf.org/fellows/17153-francisco-urbano). Authors have full control of all primary data and agree to allow the journal to review their data if requested. Authors report no financial conflict of interest, or otherwise, related directly or indirectly to this study.