Material and methods

Animals

C57BL/6 mice were bred in-house and housed under constant temperature and humidity on a 12h light-dark cycle with free access to food and water.

GAD67-GFP is a knock-in mouse that was kindly provided by Dr. Y.

Yanagawa. Pitx3-GFP is a knock-in mouse that was kindly provided by Dr. M.

Li. All procedures were performed in the light cycle according to the Veterinary Office of Geneva guidelines for animal handling at University of Geneva and the Salk Institute.

Drug treatment

Male and female mice (P15-35) were injected intraperitoneally with 0.9%

saline (control), 2 mg/kg methamphetamine (METH) or 15 mg/kg cocaine using a 15 gauge insulin syringe and injection volume < 200 µl to minimize stress. Experimental procedures were performed 24h-7d later.

Methamphetamine and cocaine were purchased from Sigma and Lipomed.

Electrophysiology in acute slices

24h or 7 days following ip injections, mice were euthanized and horizontal slices from midbrain (250 µm) were prepared in ice cold artificial cerebral spinal fluid (ACSF) containing (in mM) NaCl (119), KCl (2.5), MgCl2 (1.3), CaCl2 (2.5), NaH2PO4 (1), NaHCO3 (26.2) and glucose (11), pH 7.3, continuously bubbled with 95/5% O2/CO2. Slices were warmed to 33ºC and incubated for 45 min, then transferred to the recording chamber superfused with ACSF (2.5ml/min) at ~33°C. Epifluorescence with a U-LH100HG mercury lamp (Olympus) was used to visualize GFP and whole-cell patch-clamp recordings were made from neurons in the VTA, identified as the region medial to the medial terminal nucleus of the accessory optical tract. When using non-GFP mice, GABA neurons were identified by the absence of Ih

current, a small capacitance (<20pF) and a fast spontaneous firing rate (5-10Hz). In contrast DA neurons have an Ih current, large capacitance (30-100 pF) and slow spontaneous firing (1-3Hz). The internal solution for measuring baclofen-activated GABAB receptor currents contained (in mM) potassium

! #&!

gluconate (140), NaCl (4), MgCl2 (2), EGTA (1.1), HEPES (5), Na2ATP (2), sodium creatine phosphate (5) and Na3GTP (0.6), pH 7.3 with KOH. For GABAB receptor sIPSCs, the internal solution contained (in mM) K-gluconate 140, KCl 5, MgCl2 2, EGTA 0.2, HEPES 10, Na2ATP 4, Creatine-phosphate 10 and Na3GTP 0.3. To measure GABAA currents, the internal solution contained (in mM) K-gluconate 30, KCl 100, MgCl2 4, creatine phosphate 10, Na2 ATP 3.4, Na3 GTP 0.1, EGTA 1.1 and HEPES 5.

For the sIPSC, the evoked synaptic recordings were isolated in presence of APV (100µM), NBQX (10µM) and sulpiride (200nM) for GABAA receptor IPSC, and PTX (100µM) for GABAB receptor sIPSC. The stimulation electrode consisted of a saline-filled monopolar glass pipette, placed caudally to the cell being recorded. GABAA receptor paired-pulse ratio (PPR) was assessed by applying two pulses at 50 ms interval, every 10 seconds, whereas the GABAB

receptor sIPSC was evoked by a train of 10 electrical pulses at 66Hz, once every 20-40 seconds. For IBaclofen, currents were filtered at 1 kHz and digitized at 5 kHz (National Instruments PCI-MIO-16E-4 card) and saved on computer (IgorPro, Wavemetrics Inc.). Cells were clamped at -50 or -60 mV (membrane voltages were corrected for liquid junction potential; -15.7mV). Cell membrane and access resistance were measured with each sweep. All chemicals for electrophysiology were purchased from Tocris. We did not observe any differences with wild-type mice and Pitx3-GFP or GAD67-GFP; therefore we have pooled the data. Data are expressed as mean ± s.e.m. and statistical significance (P<0.05) was determined by one-way ANOVA with Holm-Sidak post hoc test, Student’s t-test or Mann-Whitney test.

Optogenetic Experiment

Adeno-associated virus (AAVx)- channel rhodopsin 2 (ChR2) flox virus (produced in the Vector Core Facility at the University of North Carolina) was injected into 3 week-old GAD65-Cre mice (kindly provided by Dr. Gero Miesenböck). Anesthesia was induced and maintained with isoflurane (Baxter AG, Veinna, Austria) at 5% and 1%, respectively. The animal was placed in a stereotaxic frame (Angle One; Leica, Germany) and craniotomies were performed bilaterally over the VTA using stereotaxic coordinates (AP -3.4, ML

! #'!

±0.8, DV 4.4). Injections of AAV-ChR2 flox were carried out using graduated pipettes (Drummond Scientific Company, Broomall, PA), broken back to a tip diameter of 10-15 $m, at a rate of ~ 100nl min^-1 for a total volume of 500nl. In all experiments the virus was allowed a 3 weeks to incubate before any other procedures were carried out. Fast GABAA IPSCs in DA cells were isolated in presence of kynurenic acid (2mM) and evoked by applying 2 consecutive 4ms blue-light (Thorlab – 472nm LED) flashes at 50ms interval to the slice, every 10 seconds. Recordings were done as described above.

Antibodies

A rabbit polyclonal antibody anti-Glutamate Decarboxylase 65 & 67 (AB1511, Millipore, Billerica, MA, USA), anti-GAPDH (Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-phospho S783-GABAB2 (p-S783, Terunuma et al., 2010b) , anti-phospho S892-GABAB2 (p-S892,Couve et al., 2001) were used.

A monoclonal antibody anti-GABAB1 (Clone N93A/49, NeuroMab, Davis, CA, USA) and anti-GABAB2 (Clone N81/37, NeuroMab, Davis, CA, USA) were used. A guinea-pig polyclonal antibody anti-GIRK2 (Aguado et al., 2008) was used.

Immunoelectron microscopy

A similar procedure to that described earlier (Koyrakh et al., 2005) was used.

Briefly, free-floating sections were incubated in 10% NGS diluted in TBS for 1 h. Sections were then incubated for 48 h in a mixture of two antibodies (GIRK2 and GAD65/67 or GABAB1 and GAD65/67), at a final protein concentration of 1-2 mg/ml each, diluted in TBS containing 1% NGS. Then, one of the primary antibodies (anti- GAD65/67 antibody) was visualized by the immunoperoxidase reaction and the second one (GIRK2 antibody or anti-GABAB1 antibody) by the silver-intensified immunogold reaction. After primary antibody incubation, the sections were incubated at 4°C overnight in a mixture of the following secondary antibodies: goat anti-rabbit (Fab fragment, diluted 1:100) coupled to 1.4 nm gold (Nanoprobes, Stony Brook, NY), goat anti-guinea pig (Fab fragment, diluted 1:100) coupled to 1.4 nm gold (Nanoprobes, Stony Brook, NY), and biotinylated goat anti-mouse (diluted 1:100; Vector Laboratories) antibodies, all of them made up in TBS containing 1% NGS.

! #(!

After washes in TBS, sections were washed in double-distilled water, followed by silver enhancement of the gold particles with an HQ Silver kit (Nanoprobes, Stony Brook, NY) for 8-10 min. Subsequently, the sections were incubated for 4 hours in the ABC complex (Vector Laboratories) made up in TBS and then washed in TB. Peroxidase was visualized with DAB (0.05% in TB, pH 7.4) using 0.01% H2O2 as substrate for 5-10 min. The sections were washed in PB and then post-fixed with OsO4 (1% in 0.1 M PB), followed by block-staining with uranyl acetate, dehydration in graded series of ethanol and flat-embedding on glass slides in Durcupan (Fluka) resin. Regions of interest were cut at 70-90 nm thick sections using an ultramicrotome (Reichert Ultracut E, Leica, Austria). Ultrathin sections were mounted on 200-mesh nickel grids. Staining was performed on drops of 1% aqueous uranyl acetate followed by Reynolds’s lead citrate. Ultrastructural analyses were performed in a Jeol-1010 electron microscope. Unless otherwise stated, electron microscopic samples were obtained from three different mouse and three blocks of each animal were cut for electron microscopy. To test method specificity of the procedures for electron microscopy, the primary antibody was omitted or replaced with 5% (v/v) normal serum of the species of the primary antibody. Under these conditions, no selective labeling was observed.

In addition, some sections were incubated with both gold-labeled and biotinylated secondary antibodies, followed by the ABC complex and peroxidase reaction without silver intensification. This resulted in amorphous horseradish peroxisade (HRP) end product, and no metal particles were detected. Using the same sequence, but only silver intensification without HRP reaction, resulted in silver grains but no amorphous HRP end product.

Under these conditions, only infrequent small patches of HRP end product were detected and the patches were not associated selectively with any particular cellular profile. In addition, the selective location of the signals in structures labeled with only one or the other of the signaling products within the same section, as well as having side by side double-labeled structures, showed that our procedures did not produce false-positive double-labeling results.

! $*!

Photomicrograph production

Electron photomicrographs were captured with CCD camera (Mega View III;

Soft Imaging System, Germany). Digitized electron images were then modified for brightness and contrast by using Adobe PhotoShop CS1 (Mountain View, CA) to optimize them for printing.

Quantitative analysis

To establish the relative abundance of GIRK2 and GABAB1 immunoreactivity in GABAergic neurons in control conditions and after injection of methamphetamine, quantification of immunolabeling was performed in the VTA from 60 µm coronal slices. For each of three animals, three samples of tissue were obtained (nine total blocks). Electron microscopic serial ultrathin sections were cut close to the surface of each block because immunoreactivity decreased with depth. Randomly selected areas were captured at a final magnification of 50,000X, and measurements covered a total section area of 6000 µm². GAD65/67-positive dendritic shafts were assessed for the presence of immunoparticles for GIRK2 or GABAB1. Fifteen dendritic segments were reconstructed for each experimental group from 18 to 25 serial ultrathin sections, using Image J software. The dendritic diameter of individual reconstructed profiles was measured. Linear density of the gold particles in the plasma membrane was obtained by dividing the total number of plasma membrane-bound immunogold particles by plasma membrane length of the reconstructed dendrites. Dendritic surface area was calculated by multiplying the total plasma membrane length of the reconstructed dendrite by the section thickness (70 nm). The total number of plasma membrane-bound gold particles found in the reconstructed dendrite was divided by the dendritic surface area in order to obtain the labeling density of plasma membrane-bound particles in individual profiles.

! $)!

Results

Psychostimulant-evoked plasticity in GABAB receptor signaling in VTA

A single injection of drug strengthens glutamatergic transmission onto VTA DA neurons (Ungless et al., 2001; Brown et al., 2010). We examined whether a single injection of psychostimulant also alters GABAB-GIRK signaling in the VTA. To test this, we injected C57BL/6 mice with METH at 2 mg/kg, a dose that elicits locomotor sensitization when administered repeatedly (Shimosato et al., 2001; Fukushima et al., 2007; Scibelli et al., 2011), and examined GABAB-GIRK signaling in the VTA 24h later. We first investigated the synaptically activated GABAB-GIRKs, commonly referred to as the slow inhibitory postsynaptic current (sIPSC), in acutely prepared VTA slices. High frequency (66 Hz) electrical stimulation of GABA afferents induces a spillover of GABA that diffuses to perisynaptic GABAB receptors and elicits a slow outward K⁺ current (sIPSC, Figure 8). The GABAB receptor antagonist, CGP 54626 (2 µM), inhibited the evoked current, confirming the identity of the GABAB sIPSC (Figure 8). In DA neurons, the GABAB sIPSC did not significantly change 24h following METH, compared to saline injection (Figure 8A,B). By contrast, the sIPSC was significantly smaller in GABA neurons (Figure 8D,E). Moreover, the sIPSC in GABA neurons remained depressed for at least 7 days (Figure 8E). Examination of the paired-pulse ratio for the fast GABAA-mediated IPSC revealed no difference in either DA or GABA neurons (Figure 8C,F), suggesting that the depression of the sIPSC in GABA neurons was not due to the inability of GABA terminals to release GABA.

To investigate the effects of METH on synaptic and extra-synaptic GABAB receptors, the GABAB receptor agonist baclofen was applied to the bath. As described previously (Labouèbe et al., 2007), saturating doses of baclofen (300 µM for DA and 100 µM for GABA) elicited large and desensitizing GABAB receptor-activated GIRK currents in DA neurons and small non-desensitizing currents in GABA neurons (Figure 9). All IBaclofen were blocked by the inwardly rectifying K channel inhibitor Ba²⁺or the GABAB

! $+! depressed by ~55% 24h following a single METH injection and this reducon persisted for 7 days (Figure 9C,D). We next examined whether METH altered GABAB-GIRK signaling in other brain regions. There was no significant change in the sIPSC or IBaclofen in CA1 hippocampal pyramidal or GABAergic neurons 24h following METH (Figure 10). We also measured the sIPSC and IBaclofen in pyramidal and GABAergic neurons of the prelimbic cortex, a target region of VTA DA cells, and observed no significant changes in GABAB-GIRK currents in METH injected mice (Figure 10). Thus, a single exposure to METH triggered a profound and long-lasting depression in both the sIPSC and IBaclofen in GABA neurons of the VTA.

In addition to postsynaptic GABAB receptors, presynaptic GABAB

receptors are also involved in reducing GABA release, typically through inhibition of voltage-gated Ca²⁺ channels (Wu and Saggau, 1997). To investigate whether a single exposure to METH altered GABAB receptor-dependent presynaptic inhibition, we used an optogenetic strategy to selectively stimulate GABA neurons in the VTA and measure the effect of baclofen on light-evoked fast inhibitory post-synaptic current (IPSC) recorded in DA neurons (Figure 11). AAV virus expressing a double floxed-stopped ChR2-EYFP was stereotaxically injected into the VTA of mice expressing Cre recombinase in GABA neurons (GAD65-Cre, Figure 11A, Figure 12). After 21 days, neurons expressing ChR2-YFP were evident in horizontal slices of the VTA (Figure 12A). Prolonged blue light stimulation (400ms) elicited tetrodotoxin (TTX)-insensitive photocurrents in GABA neurons, whereas short light pulses (4ms) evoked picrotoxin- and TTX-sensitive fast IPSCs in DA neurons (Figure 12B,C; Figure 11B). Bath application of baclofen (1 µM) depressed the light-evoked IPSC by ~50% in saline injected mice. By contrast, baclofen (1 µM) decreased the light-evoked IPSC by only ~20% in METH-injected mice (Figure 11B,C). Construction of dose-response curves

! $"!

revealed that GABAB receptor-dependent inhibition of presynaptic release was shifted significantly to higher agonist concentrations (Figure 11C), reflected by an increase in the IC50, which is the concentration of baclofen needed to inhibit 50% of the light-induced current (Figure 11D). Similar to the change in postsynaptic GABAB-GIRK signaling, the reduced sensitivity of presynaptic GABAB receptors persisted for 7 days (Figure 11C,D). As a control, we examined GABAB receptor-dependent presynaptic inhibition of glutamate release onto DA neurons, by measuring the amplitude of electrically evoked AMPA receptor EPSC while applying increasing concentrations of baclofen (Figure 13). We found no significant change in the IC50 in METH injected mice, compared to saline controls. Taken together, these results demonstrate that a single injection of METH triggers a depression of GABAB receptor signaling in VTA GABA neurons, both presynaptically (inhibition of GABA release) and postsynaptically (activation of GIRK channels).

! $#!

Figure 8. Absence of slow inhibitory postsynaptic currents in VTA GABA neurons 24h and 7d following a single METH exposure. The slow inhibitory postsynaptic current (sIPSC) recorded from DA (A) and GABA (D) neurons in the VTA 24h following a saline (0.9%) or METH (2mg/kg) injection.

The GABAB receptor antagonist CGP 54626 (2 µM) inhibited the sIPSC (light grey or light red trace). The GABAB-sIPSC is reduced in the VTA GABA neuron 24h following METH injection. Scale bars: 5pA, 200ms. B,E Bar graphs show mean amplitudes for sIPSC following saline or METH (24h and 7d later) in DA (B, DA saline: 15.8 ± 1.5 pA, DA METH: 16.5 ± 2.8 pA, DA 7d METH: 16.1 ± 2.0 pA) and GABA neurons (E, GABA saline: 17.8 ± 2.6 pA, GABA METH: 0.7 ± 0.5 pA, GABA 7d METH: 1.5 ± 1.0 pA). The sIPSC is significantly depressed 24h and 7d following a single injection of METH in GABA neurons (** P < 0.05 One-way ANOVA). C,F Box plots show GABAA

receptor-mediated IPSC paired-pulse ratio (PPR) plotted for DA (C) and GABA (F) neurons in saline and METH injected mice (DA saline: 0.73 ± 0.10 pA, DA METH: 0.89 ± 0.09 pA, GABA saline: 1.20 ± 0.17 pA, GABA METH:

1.02 ± 0.11 pA, ns p>0.05, Mann-Whitney test). Line shows mean. Insets show representative traces for each condition. Scale bars: 100pA, 20ms. N (number of recordings) indicated on all graphs.

! $$!

Figure 9. Reduced GABAB-GIRK currents in VTA GABA neurons 24h and 7d following a single METH exposure. IBaclofen recorded from VTA DA (A) and GABA (C) neurons 24h following a saline (0.9%) or METH (2mg/kg) injection. Outward currents recorded at -50 mV are plotted as a function of time. IBaclofen is blocked by the inward rectifier inhibitor Ba²⁺ (1mM) or by the GABAB receptor antagonist CGP 54626 (data not shown). Scale bars: 100pA (A) 50 pA (C), 100s. B, Bar graph shows average IBaclofen in DA neurons 24h following saline (DA saline: 278 ± 37 pA) or 24h and 7d following METH injection (DA METH: 174 ± 19, DA 7d METH: 229 ± 21 pA). D, Bar graph shows average IBaclofen in GABA neurons 24h following saline injection (GABA saline: 48.4 ± 5.3 pA) or 24h and 7d following METH injection (GABA METH:

22.9 ± 4.7, GABA 7d METH: 19.4 ± 4.5 pA,). Note significant decrease in IBaclofen in GABA neurons of METH-injected mice that persists for 7d (**P <

0.05 One-way ANOVA One-way ANOVA).

! $%!

Figure 10. Methamphetamine does not alter GABAB receptor signaling in hippocampal and prelimbic cortex neurons. A, Cartoon representing recording of a CA1 pyramidal neuron and stimulation configuration. Inset represents typical firing pattern evoked by a 200pA current step for 500ms. B,

! $&!

Example traces of GABAB receptor sIPSC in CA1 pyramidal neurons from saline (black) or METH (red) injected mouse (scale bar: 20pA, 200ms). C, Plot of mean amplitude of sIPSC shows no change in METH-injected mice (saline:

44 ± 7 pA, METH : 39 ± 7 pA, p>0.05, Mann-Whitney test). D, Example recordings of IBaclofen (300µM) in CA1 pyramidal neurons (scale bar 40pA, 2min). E, Plot of average IBaclofen shows no change in METH-injected mice (saline: 88 ± 19 pA, METH : 88 ± 11 pA, p>0.05, Mann-Whitney test). F, Cartoon representing recording of a CA1 Oriens interneuron and stimulation configuration. Inset represents typical firing pattern evoked by a 200pA current step for 500ms. G, H, Example traces of GABAB receptor sIPSC in CA1 Oriens interneurons from saline (black) or METH (red) injected mice (scale bar: 20pA, 200ms) and average amplitudes (saline: 20 ± 4 pA, METH : 18 ± 3 pA, p>0.05, Mann-Whitney test). I, J, Example recordings of IBaclofen

(300µM) in CA1 Oriens interneurons from saline (black) or METH (red) injected mice (scale bar 20pA, 2min) and average amplitudes (saline: 42 ± 7 pA, METH : 40 ± 6 pA, p>0.05, Mann-Whitney test). K, Cartoon representing recording of a prelimbic cortex layer 2/3 pyramidal neuron and stimulation configuration. Inset represents typical firing pattern evoked by a 200pA current step for 500ms. L, M, Example traces of GABAB receptor sIPSC in layer 2/3 pyramidal neurons from saline (black) or METH (red) injected mice (scale bar: 20pA, 200ms) and average amplitudes (saline: 38 ± 5 pA, METH : 35 ± 3 pA, p>0.05, Mann-Whitney test). N, O, Example recordings of IBaclofen

(300µM) in layer 2/3 pyramidal neurons from saline (black) or METH (red) injected mice (scale bar 20pA, 2min) and average amplitudes (saline: 57 ± 9 pA, METH : 52 ± 8 pA, p>0.05, Mann-Whitney test). P, Cartoon representing recording of a prelimbic cortex layer 1 interneuron and stimulation configuration. Inset represents typical firing pattern evoked by a 200pA current step for 500ms. Q, R, Example traces of GABAB receptor sIPSC in layer 1 interneurons from saline (black) or METH (red) injected mice (scale bar: 20pA, 200ms) and average amplitudes (saline: 12 ± 4 pA, METH : 11 ± 1 pA, p>0.05, Mann-Whitney test). S, T, Example recordings of IBaclofen (300µM) in layer 1 interneurons from saline (black) or METH (red) injected mice (scale bar 20pA, 2min) and average amplitudes (saline: 19 ± 3 pA, METH : 22 ± 3 pA, p>0.05, Mann-Whitney test). Throughout all experiments cell identity was confirmed by positive or negative GAD67 GFP fluorescence for interneurons and pyramidal neurons respectively, and by firing pattern. Additionnally, location of pyramidal neurons in layer 2/3 in the prelimbic cortex was verified by filling the cells with biocytin and CY3 revelation.

! $'!

Figure 11. Reduced sensitivity of presynaptic GABAB receptor-mediated inhibition 24h and 7d following a single METH exposure. A, Schematic shows channel rhodopsin 2 (ChR2) protein expressed selectively in VTA GABA neurons of GAD65-Cre mice. GABA neuron activity is induced by blue light, resulting in a fast GABAA receptor-mediated IPSC recorded from VTA DA neuron. Baclofen impairs GABA release by acting on presynaptic GABAB

receptors. B, Example traces of blue light-evoked IPSCs recorded 24h following saline or METH injection in presence of increasing concentrations of baclofen. Blue ticks indicate light stimulation (2x 4 ms). Basal IPSC amplitude recovers after application of CGP 54626 (2µM) and is subsequently blocked by picrotoxin (100µM). Scale bars: 200pA, 10ms. C, Dose response curves show reduced sensitivity for baclofen-dependent inhibition of fast IPSCs in METH injected mice after 24h and 7 days. D, Bar graph plots IC50 for indicated conditions (saline: 1.2 ± 0.4 µM, 24h METH: 7.1 ± 2.4 µM, 7d METH: 9.4 ± 1.4 µM, ** P < 0.05 One-way ANOVA).

! $(!

Figure 12. Characterization of blue light-evoked currents in VTA of floxed ChR2-injected GAD65-Cre mice. A, Expression of ChR2-YFP (green) in the VTA does not colocalize with DA neurons identified by TH staining (blue). B, GABA cells were identified by absence of Ih current (left, scale bar 200pA, 200ms). 3ms light flashes elicited a large current (upper panel, scale bar 100pA, 20ms) that displayed the typical desensitization of ChR2-mediated photocurrent when the light was applied for 400ms (lower panel, scale bar 100pA, 200ms). Additionally, the photocurrent was tetrodotoxin (TTX)-insensitive (the decrease in amplitude is due to the block of synaptic connections among VTA GABA neurons). C, DA cells were identified by the presence of a large Ih current (left, scale bar 200pA, 200ms). 3ms light flashes evoked an action potential-mediated current that was totally blocked by TTX (upper panel, scale bar 100pA, 20ms). No desensitization of the current was observed with prolonged duration of light exposure (lower panel, scale bar 100pA, 200ms).

! %*!

Figure 13. GABAB receptor-dependent presynaptic inhibition of glutamate onto VTA DA neurons is not altered by METH. A, Schematic showing AMPA EPSC recording from VTA DA neurons evoked by a non-specific electrical stimulation of afferent excitatory fibers. Baclofen impairs glutamate release by acting on presynaptic GABAB receptors. B, Example traces of electrically evoked EPSCs recorded 24h following saline or METH injection in presence of increasing concentrations of baclofen. Stimulation

Dans le document Drug-evoked synaptic plasticity of GABAB receptor signaling in the ventral tegmental area (Page 47-109)