5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 1.25 mM NaH2PO4, and 12.5 mM glucose, and were incubated at 31°C–33°C for 30 min, then allowed to recover at room temperature for an additional 30 min before recording. Internal solutions were either K based, for current clamp recordings from FS interneurons in paired experiments (130 mM KMeSO3, 10 mM NaCl, 2 mM Anti-diabetic Compound Library MgCl2, 0.16 mM CaCl2, 0.5 mM EGTA, 10 mM HEPES, 2 mM Mg-ATP, 0.3 mM Na-GTP [pH 7.25]), or Cs-based, for all voltage clamp recordings

(120 mM CsCl, 15 mM CsMeSO3, 8 mM NaCl, 0.5 mM EGTA, 10 mM HEPES, 2 mM Mg-ATP, 0.3 mM Na-GTP, and 5 mM QX-314 [pH 7.3]). All recordings were performed at 31°C–33°C in ACSF (see above). For experiments measuring mIPSCs, 1 μM TTX (Ascent) and 5 μM NBQX (Ascent) were added to the external saline. For experiments using dopamine antagonists, 5 μM SCH23390 (Tocris) and 10 μM sulpiride (Tocris) were added to the external saline. Mice were pretreated

with desipramine (25 mg/kg; Sigma) and unilaterally injected with 6-OHDA at 3–4 weeks of age. Experiments were typically performed 3–7 days after 6-OHDA injections unless otherwise noted. All changes observed in FS microcircuits at 1 week were already present at 3 days, so data from these time points were pooled. Due to previous reports of changes in http://www.selleckchem.com/products/azd9291.html contralateral striatum following unilateral 6-OHDA injections, saline-injected mice were used as controls (Schwarting and Huston, 1996). TH immunostains were performed on 30 μm

sections, resectioned from acute slices (250–300 μm thick) used for recording. Immunostains for PV and vGAT were performed on 30 μm sections prepared from fixed brains of D2-GFP mice. To quantify overall colocalization between vGAT and PV, images were imported into ImageJ, where intensity thresholds and Manders overlap coefficients were determined by JACoP (Bolte and Cordelières, 2006). Biocytin cell fills were performed on FS interneurons recorded in the striatum from 300 μm thick coronal slices. Slices were fixed 30 min to 2 hr after filling a neuron in 4% PFA overnight at 4°C. Throughout the paper, t tests for unpaired Adenosine triphosphate data were used to test for significance unless otherwise noted. The nonparametric Wilcoxon signed rank test was used when data were not normally distributed. A chi-square test with Yate’s correction was used to test for significance of FS-D1 MSN and FS-D2 MSN connectivities. Our model of feedforward inhibition in the striatum was adapted from one used by Atallah and Scanziani, 2009. Each cell was modeled as a single compartment, integrate-and-fire neuron. Spiking activity for individual cells was initiated by independent stochastic background synaptic activity (Gaussian noise with a standard deviation [SD] of 100 pA). The networks contained 20 FS interneurons, 400 D1 MSNs, and 400 D2 MSNs, matching observations that FS interneurons comprise ∼2% of all striatal neurons (Gittis et al.