Differential alterations of intracellular [ Ca 2 + ] dynamics induced by cocaine and methylphenidate in thalamocortical ventrobasal neurons

The ventrobasal (VB) thalamus relay nucleus processes information from rodents’ whiskers, projecting to somatosensory cortex. Cocaine and methylphenidate (MPH) have been described to differentially alter intrinsic properties of, and spontaneous GABAergic input to, VB neurons. Here we studied using bis-fura 2 ratiometric fluorescence the effects of cocaine and MPH on intracellular [Ca2+] dynamics at the soma and dendrites of VB neurons. Cocaine increased baseline fluorescence in VB somatic and dendritic compartments. Peak and areas of fluorescence amplitudes were reduced by cocaine binge treatment in somas and dendrites at different holding potentials. MPH binge treatment did not alter ratiometric fluorescence at either somatic or dendritic levels. These novel cocaine-mediated blunting effects on intracellular [Ca2+] might account for alterations in the capacity of thalamocortical neurons to maintain gamma band oscillations, as well as their ability to integrate synaptic afferents.


Introduction
Somatosensory thalamus and cortex are massively interconnected [1,2].The somatosensory thalamus is populated by cortically projecting relay Ventrobasal (VB) neurons that subserve the thalamocortical loop via glutamatergic synapses onto middle layers of the somatosensory cortex [3].VB thalamic neurons can display two mutually exclusive electrophysiological modes due to differential calcium channel activation: 1) a tonic-firing mode regulated by subthreshold gamma-band oscillations after thalamocortical neurons are depolarized, which is mediated by the activation of high-voltage activated P/Q-type calcium channels located in VB dendrites [4,5], or 2) a burst-firing mode when thalamocortical neurons are hyperpolarized, regulated by T-type calcium channel activation [6,7].Thalamic relay neurons in tonic-firing mode discharge in direct relation to afferent sensory input, whereas, in burst-firing mode, sensory information is not transmitted effectively through thalamocortical networks [8].
Calcium-sensitive dye imaging techniques have been critical in describing how evoked [Ca 2+ ] transients in thalamocortical dendrites can actively propagate and sustain both low and high frequency activity [4,[9][10][11][12].Dendrites are key elements for the integration of cortical glutamatergic afferents to VB neurons [13].Combining high-speed calcium imaging and patch clamp recordings has shown that dendrites from non-specific parafascicular thalamic nucleus can sustain calcium-channels mediated gamma oscillations, at dendritic compartments hundreds of micrometers away from the soma [14].In specific VB neurons, peak [Ca 2+ ] transient ratios exhibit uniform amplitude along proximo-distal dendritic compartments [9], suggesting the existence of fine intracellular [Ca 2+ ] buffering processes throughout thalamocortical neurons.
Both methylphenidate (MPH) and cocaine have been shown to share mechanistic similarities regarding their ability to increase monoamine concentrations in the synaptic cleft.Indeed, cocaine inhibits dopamine, norepinephrine and serotonin transporters (DAT, NET and SERT, respectively; [15][16][17]), while MPH mainly inhibits DAT and NET, but not SERT [18,19].Abuse of cocaine is associated with major neuropsychiatric pathologies [20].MPH is widely used to treat children and adolescents diagnosed with attention deficit/hyperactivity disorder (ADHD; [21]), and have reinforcing effects after intravenous administration in humans [22].Monoamines like dopamine (DA), norepinephrine (NE), and serotonin (5-HT); released by ascending afferents from brainstem nuclei onto thalamocortical neurons, play an important role in the transitions from low frequency oscillations to high-frequency, gamma band resonance during conscious states [23].In the somatosensory VB thalamus, cocaine is known to rapidly increase synaptic NE and 5-HT [24], while MPH was able to increase DA levels in the thalamic reticular nucleus (i.e., origin of GABAergic afferents to VB neurons, [25]).A recent work using transgenic mouse has described a specific 5-HT modulatory role on GABA release onto VB neurons, modulated by a cocaine binge [26].Using transgenic mice bypassed previously described collateral effects of SERT inhibitors like fluoxetine on voltage-gated calcium channels [27,28].
Acute binge-like administration of cocaine has been recently shown to alter intrinsic VB neurons' membrane properties and spontaneous GABAergic inputs to them, resulting in an enhancement of low EEG frequencies [29], through a mechanism that required the overactivation of T-type calcium channels at the level of VB neurons [30].No clear effects on calcium channels of VB neurons were observed after acute MPH binge-like administration [31].Although VB neurons can be considered key neurons to understand cocaine and MPH differential effects on thalamocortical networks, little is known about how cocaine and MPH administration can affect their intracellular [Ca 2+ ] dynamics.
We hypothesized that cocaine would induce deeper intracellular [Ca 2+ ] alterations than MPH in VB neurons.We tested [Ca 2+ ] i dynamics using fluorescence ratiometric recordings of thalamic VB neurons from mice treated with either cocaine or MPH binge administration.We used the low/medium affinity dye bis-fura 2 in a patch-clamp whole-cell protocol that differentially activated either T-type or P/Q-type voltage-gated calcium channels in VB neurons of mice treated with cocaine or MPH binge.We found that cocaine, but not MPH, reduced both the peak amplitude and time-integral (i.e., area) of [Ca 2+ ] i signals.Cocainemediated alterations of intracellular [Ca 2+ ] i dynamics might have profound consequences on the ability of the dendrites of thalamocortical neurons to integrate synaptic inputs.

Animals
Preadolescent (18-21 days old) male C57BL/6 mice from the Central Animal Facility at Universidad de Buenos Aires (Animal protocol #50-2015) were used for this research.Principles of animal care were performed in accordance with "Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral Research" (National Research Council, 2003), and approved by Universidad de Buenos Aires authorities using OLAW and ARENA directives (NIH, Bethesda, USA).

Thalamocortical slices
One hour after last i.p. administration, mice were anesthetized with tribromoethanol (250 mg/kg i.p.) followed by transcardial perfusion with ice-cold low sodium/antioxidants solution, and then decapitated.Thalamocortical slices, including the somatosensory cortex (250-350 µm) were obtained using a stage with 55° vertical inclination as previously described [26,[29][30][31].Slices were then allowed to recover at 35 °C for at least 30 min in a psychostimulant-free ACSF solution.

Ratiometric calcium imaging
Ca 2+ -imaging was performed concomitantly with whole-cell recordings using an Andor iXon+ EMCCD camera (512×512, Andor tech., Belfast, UK) coupled to a monochromator Polychrome V (TILL Photonics GmbH, Munich, Germany) through an optic fiber connected to a BX51WI Olympus upright microscope (Olympus Latin America Inc., FL, USA).To allow sufficient filling of dendrites with dye, imaging was initiated 5-10 min after obtaining whole-cell configuration.The low/medium-affinity ratiometric Ca 2+ -sensitive indicator Bisfura 2 (in vitro K d =0.37 µM was previously described in #33) was chosen in order to allow a more accurate report of the time course and amplitude of intracellular [Ca 2+ ] dynamics [14,34].The experimental protocol is shown in Figure 1A.Baseline ratios were obtained during the 5 s prior to the application of the square voltage pulses while holding membrane potential at −70 mV.The voltage was then stepped to −30 or to −10 mV and fluorescence peak and areas were measured for 16 s after the end of the pulses.Fluorescence changes at 510 nm were acquired after consecutive 340 nm and 380 nm wavelengths using a 0.8 numerical aperture water-immersion 40× objective (Olympus Latin America Inc., FL, USA).Image acquisition was performed using cell-R software (Olympus Soft Imaging Solutions, Münster, Germany), and analyzed post hoc using NIH Image J software.Changes in fluorescence ratios [F(Ratio)=F 340 /F 380 ] were calculated offline, as the change in fluorescence from baseline of the Ca 2+ -sensitive indicator bis-fura 2. Baseline fluorescence ratio was calculated by averaging 5 s of fluorescence signal before the square pulse voltage protocol was applied.No direct correlation was obtained between baseline levels and voltage step mediated fluorescence responses.Exposure time to each wavelength was 10 ms and the sampling rate was 10Hz.

Statistical analysis
InfoStat software (Universidad Nacional de Córdoba, Argentina) was used for statistical comparisons.Differences were considered significant if p<0.05.Population statistics are presented as mean ± standard error of the mean.

Results
We studied intracellular [Ca 2+ ] dynamics evoked by depolarizing square pulses at the somatic and dendritic compartments of VB neurons from mice one hour after last treatement with cocaine or methylphenidate (MPH), compared to saline (control) conditions.In particular, we studied how intracellular [Ca 2+ ] dynamics changed over time after the membrane potential was rapidly depolarized using 500 ms square voltage pulses (Fig. 1A).
We used a high Cesium/QX-314 intracellular solution in voltage-clamp configuration to allow for activation of voltage-dependent calcium currents, but not potassium or sodium currents.Ratiometric [Ca 2+ ] indicator bis-fura 2 was added to the intracellular solution and VB neurons were allowed to passively fill after gaining access to the cytoplasm for 5-10 min to reach a steady-state filling of dendritic compartments (Fig. 1B, left panel, white squares).Ratiometric changes in fluorescence (F(Ratio)=F 340 /F 380 ) were calculated offline and represented as percentage of change from baseline.Step voltage pulses (500 ms) designed to maximally activate either T-type (Fig. 1C, red trace; from −70 mv to −30 mV), or P/Q-type (Fig. 1C, black trace; from −70 mV to −10 mV) voltage-gated calcium currents were applied to VB neurons to then observe the time course of intracellular [Ca 2+ ] dynamics over the next 16 s (Fig. 1A).
While a large somatic region of interest was taken to measure the fluorescence time course at the soma (avoiding places were the fluorescence emission was saturating around the recording electrode), fluorescence values from several proximal dendritic regions were combined into a total dendrite value per neuron (Fig. 1B, square white pulses).
Figure 1D shows representative F(Ratio) responses to these stimuli at somatic and dendritic compartments in control conditions.To quantify these results, we measured the maximum peak amplitude and the area following the pulse (Figure 1D, arrows and grid lines under fluorescence signals).
Ratios of peak amplitudes were not significantly different across treatments at all holding potentials (data not shown, Kruskal-Wallis ANOVA, H=4.13, P=0.53).These results suggested that cocaine or MPH induced linear [Ca 2+ ] changes within VB neurons compartments.
At a holding of −10 mV, areas were significantly smaller at dendritic levels after cocaine binge administration (Fig. 2F; ANOVA F (2,28) =7.26, P=0.003; Tukey-Kramer post hoc test, cocaine vs. MPH, saline, P<0.05), while remaining non significantly different when comparing cocaine and saline, MPH at the soma (Fig. 2E; ANOVA, F (2,32) =1.69, P=0.2).In addition, no increase in area was observed comparing holding potentials at dendritic levels across treatments (Kruskal Wallis ANOVA, P>0.5).Thus, only cocaine binge administration was able to reduce F(Ratio) signals (both peaks and areas) at the somatic or dendritic level at the −30 mV or −10 mV holding potentials, respectively.MPH did not alter F(Ratio) signals at any of the holding potentials tested neither at somatic nor at dendritic levels.

Discussion
These results are the first description of cocaine and MPH binge treatment-mediated [Ca 2+ ] changes in thalamocortical ventrobasal (VB) neurons, suggesting the existence of differential cocaine and MPH effects on intracellular [Ca 2+ ] i dynamics.[Ca 2+ ] i dynamics changed in magnitude (both peak and area) when T-type calcium channels were activated using depolarizing square pulses.In addition, F(Ratio) peaks and areas were smaller after cocaine treatment when P/Q-type calcium channels were activated at the dendrites.For all other conditions, the blunting effect of cocaine was robust, including reduction in peaks and areas of F(Ratio).
Baseline fluorescence levels were higher in VB neurons from cocaine-treated mice, indicating higher levels of [Ca 2+ ] i , in agreement with previously described cocaine-induced T-type calcium channel over-activation closer to resting membrane potentials [29].Larger baseline [Ca 2+ ] levels would also partially inactivate calcium currents, which might induce lower amplitude [Ca 2+ ] i transients due to its regulation of intracellular pathways [35,36].
Here we also observed a cocaine-mediated reduction in F(Ratio) during T-type calcium channels activation, suggesting the existence of compensatory mechanisms in VB neurons in order to manage higher [Ca 2+ ] i levels through the activation of these channels.Indeed, acute cocaine has been shown to enhance basal [Ca 2+ ] i in cortical neurons [37], while repeated cocaine exposure led to a deregulation of [Ca 2+ ] i homeostasis in nucleus accumbens neurons, that reduced calcium entry [38].Acute cocaine administration can increase the expression of endoplasmic reticulum stress proteins in striatal neurons [39].Strikingly, our results suggested that cocaine binge administration did drive VB neurons (neurons of a glutamatergic phenotype) to manage high levels of intracellular [Ca 2+ ], similar to what was described for GABAergic inhibitory neurons [40,41].
A pivotal study by Pedroarena & Llinás [4] described the fact that P/Q-type calcium channels opening at the dendritic compartment of VB neurons are responsible for sustaining gamma band frequency membrane potential sub-threshold oscillations.Dendritic compartments of VB neurons are responsible for integrating cortico-thalamic excitatory inputs [13], creating a strategic condition in which small variations on intracellular [Ca 2+ ] might result in altered thalamocortical dynamics.Simultaneous high speed video recordings of fura 2-dependent fluorescence and intracellular current clamp recordings showed Ca 2+ entry in dendritic compartments in conjunction with injected depolarizing currents via the VB soma [4].Another report, using voltage-clamp recordings, has shown that largest amplitude of [Ca 2+ ] i signals were observed at proximal dendrites, during a step voltage depolarization capable of activate either T-type or P/Q-type channels [12].
Our results also showed clear cocaine-mediated reductions in both peak and area of F(Ratio) after P/Q-type calcium channels were activated in dendrites, probably suggesting a compensatory, faster buffering mechanism in VB dendrites.Future experiments are still needed to further explain cocaine effects on intracellular [Ca 2+ ] reservoirs in thalamocortical neurons.
MPH has not previously been associated with increases in [Ca 2+ ] transients, as shown for cocaine [43].Du et al. [43] used calcium probes that had no direct relation with the membrane potential of cortical neurons or the type of Ca 2+ channels (or other intracellular mechanisms).In spite of that, our results clearly showed strong alterations of [Ca 2+ ] dynamics at VB thalamic neurons after cocaine, but not MPH binge administration similarly to what Du et al. [43] described at a cortical level.
Describing how intracellular [Ca 2+ ] changes can be influenced by cocaine or MPH is essential to understanding their long lasting effects on a wide number of the intracellular pathways that use Ca 2+ as a second messenger.Results presented in this study specifically suggest that cocaine was able to blunt both somatic and dendritic [Ca 2+ ] integration.MPH did not affect [Ca 2+ ] i dynamics.These novel results provide new insights on how cocaine might deregulate low (i.e., T-type channel-mediated) and high frequency (i.e., P/Q-type channel-mediated) dendritic integration in thalamocortical neurons, affecting the neurons' ability to integrate corticothalamic synaptic afferents.

Figure 1 .
Figure 1.Schematic protocol representation of both ratiometric [Ca 2+ ] fluorescence and voltageclamp recordings A, schematic description of the protocol used during consecutive imaging and electrophysiological whole-cell voltage-clamp acquisition.B, representative fluorescence emission from a Bis-fura 2-filled Ventrobasal (VB) neuron excited at 380 nm wavelength (left) and after calculating the fluorescence ratio F(Ratio)=F 340 /F 380 (right) at resting holding potential (−70 mV).Representative dendritic regions of interest are indicated (white squares).The pseudocolor bar represents the percentage of Bis-fura 2 mediated F(Ratio) shown in right panel.C, representative T-type calcium currents from a VB neuron elicited by 500 ms long square pulses from holding potential −70 mV to −30 mV (red line) or −10 mV

Figure 2 .
Figure 2. Effects of cocaine and methylphenidate (MPH) binge treatment on peak amplitude and area of F(Ratio) using bis-fura 2 in ventrobasal (VB) neurons A, B, mean baseline F(Ratio) (shown as percentage of saline values) recorded from soma or dendrites, respectively.** represents P<0.001 comparing cocaine vs. saline, MPH; # represents P<0.05 comparing MPH vs. saline.C, D, mean peak amplitude F(Ratio) from soma or dendrites, respectively, of VB neurons from mice treated with saline, cocaine or MPH binge treatments.Peak amplitudes were obtained immediately after 500 ms square pulses that changed holding potential from −70 mV to either −30 mV or −10 mV.* represents P<0.05.E, F, mean areas (i.e., time-integral of F(Ratio) signals) from soma or