Experimental Neurology 287 (2017) 44–53

Contents lists available at ScienceDirect

Experimental Neurology

j ourna l homepage: www.e lsev ie r .com/ locate /yexnr

Research Paper

Altered postnatal maturation of striatal GABAergic interneurons in aphenotypic animal model of dystonia

Christoph Bode a,1, Franziska Richter a,⁎,1, Christine Spröte a, Tanja Brigadski b,c, Anne Bauer a, Simone Fietz d,Jean-Marc Fritschy e, Angelika Richter a,⁎a Institute of Pharmacology, Pharmacy and Toxicology, Faculty of Veterinary Medicine, University of Leipzig, 04103 Leipzig, Germanyb Institute for Physiology, Medical Faculty, Otto-von-Guericke University, 39120 Magdeburg, Germanyc Center of Behavioral Brain Sciences (CBBS), 39120 Magdeburg, Germanyd Institute of Veterinary Anatomy, Histology and Embryology, Faculty of Veterinary Medicine, University of Leipzig, 04103 Leipzig, Germanye Institute of Pharmacology and Toxicology, University of Zurich, Zurich 8057, Switzerland

Abbreviations: BDNF, brain-derived neurotrophcarboanhydrase isotyp 7; GABAAR, GABAA receptor; IN, iKCC2, potassium chloride transporter 2.⁎ Corresponding authors at: Institute of Pharmacolo

Department of Veterinary Medicine, Leipzig University,Leipzig, Germany.

E-mail addresses: franziska.richter@vmf.uni-leipzig.deangelika.richter@vetmed.uni-leipzig.de (A. Richter).

1 These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.expneurol.2016.10.0130014-4886/© 2016 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 1 June 2016Received in revised form 30 September 2016Accepted 21 October 2016Available online 22 October 2016

GABAergic disinhibition has been suggested to play a critical role in the pathophysiology of several basal gangliadisorders, including dystonia, a common movement disorder. Previous studies have shown a deficit of striatalGABAergic interneurons (IN) in the dtsz mutant hamster, one of the few phenotypic animal models of dystonia.However, mechanisms underlying this deficit are largely unknown. In the present study, we investigated themi-gration andmaturation of striatal IN during postnatal development (18 days of age) and at age of highest severityof dystonia (33 days of age) in this hamster model. In line with previous findings, the density of GAD67-positiveIN and the level of parvalbuminmRNA, amarker for fast spiking GABAergic IN,were lower in the dtszmutant thanin control hamsters. However, an unaltereddensity ofNkx2.1 labeled cells andNkx2.1mRNA level suggested thatthe migration of GABAergic IN into the striatumwas not retarded. Therefore, different factors that indicate mat-uration of GABAergic INwere determined.WhilemRNAof the KCC2 cation/chloride transporters and the cytosol-ic carboanhydrase VII, used asmarkers for the so called GABA switch, aswell as BDNFwere unaltered, we found areduced number of IN expressing the alpha1 subunit of the GABAA-receptor (37.5%) in dtsz hamsters at an age of33 days, but not after spontaneous remission of dystonia at an age of 90 days. Since IN shift expression fromalpha2 to alpha1 subunits during postnatal maturation, this result together with a decreased parvalbuminmRNA expression suggest a delayed maturation of striatal GABAergic IN in this animal model, which might un-derlie abnormal neuronal activity and striatal plasticity.

© 2016 Elsevier Inc. All rights reserved.

Keywords:Basal gangliaFast spiking interneuronsParvalbuminGABAStriatumBDNFDyskinesia

1. Introduction

GABAergic interneurons (IN) control the striatal microcircuitry, andthereby the activity of medium spiny neurons (MSN; striatal outputneurons) and the basal ganglia-thalamocortical network (Ramanathanet al., 2002). Subtypes of GABAergic IN include, in particular, fast-spik-ing IN (FSIs), positive for calcium-binding protein parvalbumin (PV)(Hu et al., 2014), IN that express calcium-binding protein calretininand IN that contain nitric oxide synthase (NOS) (Ramanathan et al.,

ic factor; CAVII, cytosolicnterneurons; PV, parvalbumin;

gy, Pharmacy and Toxicology,An den Tierkliniken 15, 04103

(F. Richter),

2002). There is evidence from studies in human patients and animalmodels that reduced inhibition by striatal PV IN could be involved inbasal ganglia disorders, including dystonia (Richter and Richter, 2014),Tourette syndrome (Pappas et al., 2014), Parkinson's disease (Gittisand Kreitzer, 2012) and Huntington's disease (Reiner et al., 2013).Since IN express unique receptors such as specific GABAA receptor(GABAAR) subunits (Waldvogel et al., 1999) insights into their patho-physiological role can lead to the discovery of new therapeutic targetsfor these incurable disorders.

Dystonia is characterized by sustained or intermittent muscle con-tractions causing abnormal, often repetitive movements and postures(Albanese et al., 2013). The basal ganglia play an important role in thepathophysiology of dystonia. Abnormal neuronal activity in the stria-tum, the input structure of the basal ganglia, can lead to disturbed inhi-bition of thalamocortical projections and to changes in striatal andcortical plasticity (Breakefield et al., 2008; Gittis and Kreitzer, 2012;Kohling et al., 2004). Increasing evidence from animal models andhuman patients suggest that this maladaptive plasticity can be related

45C. Bode et al. / Experimental Neurology 287 (2017) 44–53

to disturbed GABAergic inhibition (Garibotto et al., 2011; Gittis et al.,2011; Reiner et al., 2013; Weise et al., 2011). The dtsz mutant hamster,a phenotypic unique rodent model of paroxysmal dystonia, replicatesthese alterations. In this model, age-dependent reductions of PV-posi-tive IN (Gernert et al., 2000; Hamann et al., 2007), calretinin-positiveIN (Hamann et al., 2005) and NOS-positive IN (Sander et al., 2006)were found by quantification of cells labeled for the respective pheno-typicmarkers. This deficit results in reduced inhibition of striatal projec-tion neurons and in consequence leads to a decreased basal gangliaoutput, verified by several electrophysiological recordings(Avchalumov et al., 2014; Bennay et al., 2001; Gernert et al., 2002;Gernert et al., 2000; Kohling et al., 2004). Identification of the cause ofreduced IN in this hamster model could reveal novel disease mecha-nisms and therapeutic targets.

Reduction of GABAergic striatal IN in the dtszmutant hamster occursat themaximum expression of dystonia, at 33 days of age, which is only12 days after weaning. In correlation with the remission of dystonicsymptoms, the density of GABAergic IN normalizes at about 60–70 days of age (Hamann et al., 2007). This early and age dependent def-icit points towards a developmental impairmentwhich could either be adelayed migration of the IN into the striatum and/or a retardedmaturation.

During embryonic development cells of the medial ganglionic emi-nence, expressing the homeodomain protein Nkx2.1, migrate into theforming striatum where they differentiate into the cholinergic,calretinin-positive, or PV-positive IN. These IN maintain expression ofNkx2.1 into adulthood, which is therefore an early and stable markerof striatal IN (Hamasaki et al., 2003; Marin et al., 2000). Important fac-tors for the postnatal differentiation of GABAergic IN are the brain-de-rived neurotrophic factor (BDNF) (Eto et al., 2010; Lessmann andBrigadski, 2009), the GABA-switch related proteins potassium chloridetransporter 2 (gene: Slc12a5, protein: KCC2) and cytosoliccarboanhydrase isotyp 7 (CAVII) (Yeo et al., 2009), aswell as the expres-sion of specific GABAAR subunits (Fritschy, 2015; Waldvogel et al.,1999). BDNF co-regulates the expression of the transporter KCC2 (Yeoet al., 2009), which increases on postnatal day 6 leading to a shift fromexcitatory to inhibitory GABA-signalling of GABAAR (Rivera et al.,2005). GABAAR are composed of five different subunits including twoα-, two β- and one γ-subunit for each receptor. Further classificationseparates subunits into α1–6 und β1–3 (Mohler, 2011). Importantly,mature GABAergic IN typically express α1β2γ2 GABAAR, wherebyMSN predominantly express α2β3γ2 GABAAR (Fritschy, 2015; Pirkeret al., 2000;Waldvogel et al., 1999). Interestingly, premature GABAergicIN expressα2β2γ2 GABAAR and switch toα1 subunit containing recep-tors during postnatal development (Cuzon Carlson and Yeh, 2011).

In order to further determine the developmental alterations instriatal IN, we investigated the expression of Nkx2.1, PV-mRNA, BDNF,the GABA-switch and GABAAR subunit alpha1 (GABAAR-α1) in the dtsz

hamster in comparison to controls. Our results provide evidence thatmaturation of GABAergic IN and not migration of GABAergic precursorIN into the striatum is delayed in the dystonic hamster model.

2. Material and methods

2.1. Animals

All experimentswere performedwith groups of female andmale dtsz

mutant Syrian golden hamsters (Mesocricetus auratus auratus) (inbredline) at an age of 18, 33 or 90 days and on age- and sex-matched controlhamsters. N numbers for experiments were between 4 and 17 as indi-cated in the Results section. The dtsz mutant hamster shows a dystonicphenotype in response to stress with a maximum severity at 33 daysof age and remission after 60–70 days of age (Hamann et al., 2007;Richter and Loscher, 1998). Both hamster lines were tested at 21 daysof age for expression of dystonic phenotype in response to stress as de-scribed previously (Richter and Loscher, 1998). As expected, all dtsz

hamsters developed dystonic symptoms after stress induction, whilenone of the control hamsters showed a dystonic phenotype. The ani-mals were bred and kept in institutional facilities on a 14 h light/10 hdark cycle. Hamsters were single-housed after weaning in makroloncages (Type III) at 23 °C ± 2 °C and relative humidity of about 60%.Food (Altromin standard diet) and water were available ad libitum. Allexperiments and animal care were conducted under the aspects of re-duction, refinement and replacement and following theGermanAnimalWelfare Act (TVV44/12, TVV41/13) as well as the European guidelines(Directive 2010/63/EU). All analyses were performed blindly with re-spect to the genotype of the animals. The number and age of hamstersused in each experiment are given in the text and figures.

2.2. Tissue preparation for molecular biology

Groups of hamsters were euthanized at 33 days or 18 days of age byan overdose of pentobarbital (100 mg/kg bodyweight). The animalswere briefly perfused transcardially with 0.1 M phosphate buffered sa-line (PBS) at pH 7.4 supplementedwith 20 units of heparin/ml. Thereaf-ter the brain was removed from the skull and dissected into the twohemispheres on an ice-cold glass plate. The right hemisphere was em-bedded and snap frozen in −35 °C cold 2-methyl-butane and storedat −80 °C (used for quantitative real-time PCR). The left hemispherewas dissected directly on the ice cold glass plate into motoric cortex,striatum and remaining tissue. This tissue was placed in pre-weighedcryotubes and weighed on a microbalance, flash frozen in liquid nitro-gen and stored at−80 °C (used for the BDNF-ELISA).

2.3. Quantitative real-time PCR

Striatum andmotor cortex were dissected from 100 μm slices of fro-zen embedded brain tissue inside a cryostat (Hyrax C 50, Zeiss, Germa-ny) at−18 °C. Total RNAwas isolated by using the RNeasy PlusMini Kit(Qiagen, CA, USA) following the manufacture instructions. First-strandcDNA synthesis was performed using the qScript cDNA super Mix(Quanta BioSciences, MD, USA) according to the manufacturers proto-col. For real time qPCR the KAPA Probe Fast Universal qPCR MasterMix (peqlab/VWR, Germany) was used with a total volume of 5 μl andamplification was performed on a PikoReal 96 Real-Time PCR system(ThermoScientific, Waltham, USA) according to manufacturers proto-cols. The following TaqMan Gene Expression Assays were used: BDNF(Mm01334044), Hprt (mM015455399), Gapdh (HS99999915_21), PV(Mm00443100), Nkx2.1 (Mm00447558), CAVII (Mm00551727) andSlc12a5 (Mm00803929). Data was analyzed using the PikoReal Soft-ware 2.1 (ThermoScientific). To determine the relative mRNA expres-sion, GeNorm was used and candidate reference genes Gapdh and Hprtwere chosen according to average expression stability as described pre-viously (Richter et al., 2014; Vandesompele et al., 2002). Gapdhwas alsopreviously found to be stably expressed in striatal samples of this ham-ster model (Avchalumov et al., 2014).

2.4. BDNF ELISA

A sandwich-ELISA-system was used to quantify BDNF level in stria-tum (Quantikine ELISA Human BDNF Immunoassay, R&D systems,MN, USA) as described previously (Petzold et al., 2015). The sampleswere homogenized in lysis buffer (137 mM NaCl, 20 mM Tris HCl; 1%NP40; 2500 μl 10% glycerol, 1 mM PMSF, 10 mM aprotinin, 1 mM leu-peptin). The solution was flash frozen in liquid nitrogen and thawedthree times for breaking the cell membranes. The ELISA-System wasperformed according to the manufactures protocols. The protein con-centration was determined by using the Pierce BCA Protein Assay Kit(ThermoScientific, Rockford, IL, USA) in accordance with the manufac-tures protocols. The results of the BDNF ELISA were normalized to pro-tein or wet weight of the brain.

46 C. Bode et al. / Experimental Neurology 287 (2017) 44–53

2.5. Immunohistochemistry

For immunohistochemistry, hamsters were sacrificed at the age of33 or 90 days by an overdose of pentobarbital. The animals were brieflyperfused transcardially with 0.1 M PBS at pH 7.4 supplemented with20 units of heparin/ml to remove blood followed by perfusion with 4%paraformaldehyde-PBS-solution (Pfa) as fixative. Brains were removedand post-fixed for 1 h in Pfa at 4 °C and cryoprotected in PBS-sucrose so-lutions for 24 h with ascending concentration (10%, 20%, 30%) at 4 °C.Thereafter brains were washed in 0.1M PBS, dried, frozen on powdereddry ice and stored at −80 °C. Free-floating coronal sections (40 μm)were collected for analysis using a cryostat. Sections were stored incryoprotectant solution (500 ml: 250 ml 0.1 M PBS, 250 ml glycerol,0.33 g MgCl2, 42.8 g sucrose) at −21 °C. Every 8th section of the stria-tum was used for each specific labeling. For Nkx2.1/PV colabeling sec-tions were washed in 50 mM tris buffer solution (TBS), mounted oncharged glass slides and permeabilized with 0.5% Triton X in 50 mMTBS followed by 1 h incubation in blocking solution (50 mM TBS, 10%normal donkey serum (NDS) and 0.5% Triton X) at room temperature(RT). Sections were then incubated at 4 °C for 72 h with primary anti-bodies rabbit anti-Nkx2.1 (Epitomics/Abcam, Cambridge, UK) 1:500,and mouse anti-PV (MAB 1572 Merck Millipore, Darmstadt, Germany)1:1000 in a 2%NDS and 0.5% Triton X solution. Sectionswere equilibrat-ed at RT for 1 h,washed in TBS and incubatedwith secondary antibodies(1:500 each donkey anti-rabbit AlexaFluor 594 IgG and donkey anti-rabbit AlexaFluor 488 IgG, Jackson ImmunoResearch, Suffolk, UK) (Jack-son Immuno Research, Suffolk, UK) in TBS with 2% NDS and 0.5% TritonX for 45min at RT. After final washes in TBS sections were coverslippedwith VectaShield H-1000 (Vector Laboratories, CA, USA). For GABAAR-α1 and -β2 single labeling and co-labeling the same protocol was usedbut sections were free floating and mounted only after the final TBSwash. The primary antibodies were a polyclonal rabbit anti GABAAR-α1 (made in lab by J.M. Fritschy; for in depth validation see Kralic etal. (2006)) used at a 1:5000 dilution and incubated for 64 h at 4 °C, apolyclonal guinea pig anti GABAAR-β2 (Synaptic Systems, Göttingen,Germany) at a 1:500 dilution and incubated for 48 h at 4 °C and the sec-ondary antibodieswere the donkey anti-rabbit AlexaFluor 594 IgG and adonkey anti-guinea pig AlexaFluor 488 IgG incubated at 1:500 dilutionfor 90 min at RT. For glutamate decarboxylase isoform 67 (GAD67) co-labeling a monoclonal mouse anti GAD67 primary antibody(MAB5406, Merck Millipore) was added to the primary antibody solu-tion at 1:500 dilution and the donkey anti-rabbit AlexaFluor 488 IgGat 1:500 was added to the secondary antibody solution.

2.6. Confocal microscopy and stereological quantification

Co-localizations of Nkx2.1with PV or theGABAAR subunits aswell asGABAAR-α1 and GAD67 in the striatum were analyzed using a confocalmicroscope (Olympus Fluoview FV 1200, Software Olympus Fluoview4.1, Olympus, Germany) on sections randomly selected from controland dtsz hamsters.

An unbiased stereological analysis with the optical fractionatorprobe was used to estimate the number of Nkx2.1-positive neuronsand of GABAAR-α1- and/or GAD67-positive neurons in the striatum. Ste-reo Investigator software (MBF Bioscience, VT, USA) coupled to a ZeissAxioskop microscope with a Ludl XYZ automated motorized stage, z-axis microcator (Visitron Systems, Germany), Retiga 2000R CLR-12color digital camera (QImaging, Surrey, Canada) and LED light source(Cool LEDpe300, CoolLED) was used for stereological sampling. Theleft and right striatum was delineated using the 5× objective and la-beled neurons were counted using the 40× objective (grid size:500 × 500 μm for Nkx2.1, 250 × 250 for GABAAR-α1 and GAD67,counting frame: 100 × 100 μm). Image acquisition parameters werekept constant during quantification. The Gunderson and Schmidt-Hofconfidence of error were below 0.07. The investigator was unaware ofthe genotype of the animals.

2.7. Fluorescence intensity analysis

For analysis of GABAAR-α1 distribution and expression a section ofthe medial striatum was visualized with the Axioskop microscope andtheRetiga 2000R CLR-12 color digital camera using the Stereo Investiga-tor software. The striatum of each hemisphere was subdivided into 4subregions (dorsolateral, dorsomedial, ventrolateral, ventromedial) asdone previously in the dtsz hamster (Gernert et al., 2000). From the cen-ter of each subregion per hemisphere an image was acquired using the40× objective. Image acquisition parameters were kept constant be-tween animals and regions. Images were analyzed for mean pixel fluo-rescence intensity using Image J (NIH, USA) for three differentparameters: (1) across the full picture, (2) in positively labeled cell bod-ies and (3) for lines positioned inside of three positively labeled cellularprocesses which were in focus per image. For cell bodies and processesmeasurements were averaged for each image. As there were no statisti-cal significant differences between the right and the left striatum(paired student's t-test) these values were averaged for each section.All of the analyses described above were undertaken by an investigatorunaware of the genotype of the animals.

2.8. Statistics

SigmaPlot was used for statistical analyses (SigmaPlot, Systat, IL,USA). Data was tested for normality a priori with the Shapiro-Wilktest (p b 0.05 rejects assumption of normality). Stereological estimatesand BDNF expression levels of dtsz and control hamsters were analyzedfor genotype effects with the student's t-test (for parametric data) asShapiro-Wilk test reported p N 0.05. Relative gene expression datawas not normally distributed (Shapiro-Wilk test p b 0.05) thereforethe Mann Whitney U test (for non-parametric data) was used for com-parison of raw expression values. Gene expression data is presentednormalized to mean of control animals to visualize the fold change.Fluorescence intensities were compared using a 2 × 4 repeated mea-sures ANOVA (mixed design, genotype x subregion as repeated mea-sure) which revealed trends for factors or interaction (p b 0.1) but nosignificance. Based on previous studies we expected effects specificallyin the dorsomedial region of the striatum, therefore a planned compar-ison (two-sided student's t-test) was performed for that region(Gernert et al., 1999). Grubbs'test (GraphPad software, Quick calcs, CA,USA) was used to identify significant outliers. The null hypothesis wasrejected at p b 0.05.

3. Results

3.1. Density of Nkx2.1-positive cells in the striatum is similar in dystonichamsters and controls

In order to examine whether the deficit in GABAergic IN, previouslyfound in dtsz hamsters at the age of 33 days when the severity of dysto-nia reaches amaximum, is due to delayedmigration of these IN into thestriatum, the density of INwas determined by using the homeobox pro-tein Nkx2.1 as specific marker. This protein is already expressed bynearly all migrating striatal IN at the embryonic stage and remainsexpressed throughout adulthood. We first confirmed that Nkx2.1 isexpressed in the striatum of hamsters (Fig. 1A) but only rarely in thecortex (not shown) similar to descriptions in other species (Marin etal., 2000). As expected, all PV-positive IN expressed Nkx2.1 (Fig. 1A–C), while there were Nkx2.1-positive cells lacking PV expression,representing other IN subtypes (Marin et al., 2000). Stereological quan-tification of Nkx2.1-positive cells in the striatumof dtsz and control ham-sters at the age of maximum severity of dystonia (33 days) revealed nosignificant differences between genotypes (mean dtsz mutant 5534 ±151 cells/mm3 vs. control hamsters 5658 ± 276 cells/mm3, p = 0.55,student's t-test) (Fig. 1D–F).

Fig. 1. Quantification of Nxk 2.1-positive cells in the striatum. (A) Nkx 2.1 and (B) parvalbumin (PV) are co-expressed in the striatum. (C) PV-positive IN are also Nkx 2.1-positive (whitearrow) but not all Nkx 2.1-positive cells are PV IN (red arrow). Reference bar = 30 μm (D–F) unbiased stereology did not reveal differences in the density of Nkx 2.1-positive cells(mean+SEM) in the striatumbetween control (n=7, D) and dtsz (dt, E) hamsters (n=8) at age ofmaximumseverity of dystonia (33 days (d)) (student's t-test). Reference bar=50 μm.

47C. Bode et al. / Experimental Neurology 287 (2017) 44–53

Real time qPCR confirmed that Nkx2.1 mRNA reached comparablelevels in the striatum of 33 days old dtsz and control hamsters (Fig.2A). Furthermore, we investigated the differences in Nkx2.1 mRNAlevel at an earlier age. Relative quantities of striatal Nkx2.1 mRNA didnot differ between control and dtsz hamster at 18 days of age (Fig. 2A).Taken together, these results suggest no differences in migration of INinto the striatum in dystonic hamster brains.

3.2. PV mRNA expression is reduced in dystonic hamsters compared tocontrols

If Nkx2.1-positive IN are present in the striatum of dtsz hamsters at33 days of age, then the previously found reductions in PV-positive,calretinin-positive and NOS-positive IN compared to controls indicatea lack of expression of the respective phenotypic marker. In order to

Fig. 2. By qPCR analyses (A) no difference in the mRNA expression of Nkx 2.1 in the striatum bdystonia (33 days) became evident (Mann Whitney U for each age group). (B) Lower express33 days (d) compared to controls (*p b 0.01; Mann Whitney U for each age group). Valuperformed on raw data (RQ, relative quantity). n numbers are given inside respective bars.

ascertain this assumption, we quantified PV mRNA in striatal samplesfrom 33 days old dtsz and control hamsters via qPCR. Thereby, a 40% re-duction of PV mRNA became evident in the dtsz mutant (p = 0.008;Mann Whitney U, n = 5/group) (Fig. 2B). This significant reductioncould be confirmed in a second independent experiment in different lit-ters of dtsz hamsters and controls (32%, p = 0.016; Mann Whitney U,n = 5/group; differences in baseline expression precluded combinationof raw data of the two independent experiments).

3.3. BDNF protein levels are unchanged in the striatum of the dtsz mutant

The neurotrophin brain derived neurotrophic factor (BDNF) is animportant motogenic and differentiation factor for IN precursor(Marin and Rubenstein, 2003; Polleux et al., 2002). This secretory pro-tein is expressed in the cortex and transported via corticostriatal

etween control and dtsz (dt) hamsters at age of 18 days (d) and of maximum severity ofion of parvalbumin (PV) mRNA was found in the striatum of dtsz (dt) hamsters at age ofes (mean + SEM) are shown normalized to means of control groups; statistics were

Fig. 3. Expression of BDNF mRNA in the cortex and BDNF protein in the striatum. No difference was found in (A) BDNF mRNA (qPCR) in the cortex or (B) BDNF protein (ELISA) in thestriatum between control and dtsz (dt) hamsters at age of 18 days (d) and of maximum severity of dystonia (33 days) (Mann Whitney U or qPCR or student's t-test for ELISA for eachage group). Values (mean + SEM) are shown normalized to means of control groups; statistics were performed on raw data (RQ, relative quantity). n numbers are given insiderespective bar.

48 C. Bode et al. / Experimental Neurology 287 (2017) 44–53

glutamatergic terminals to the striatum (Altar et al., 1997). In line withprevious findings in mice (Altar et al., 1997; Hofer et al., 1990), theamounts of BDNF mRNA were very low in striatal neurons of hamsterbrains (data not shown). Therefore we measured BDNF mRNA in thecortex with qPCR and BDNF protein levels in the striatum with ELISA.There were no significant differences in BDNF mRNA or protein in dtsz

hamsters versus controls (Fig. 3A, B).

3.4. No differences in mRNA expression of GABA switch-related proteins inthe dtsz mutant

Another important factor for termination of INmigration but also formaturation of newborn IN is the potassium-chloride-cotransporter 2(KCC2) (Bortone and Polleux, 2009; Watanabe and Fukuda, 2015). Ex-pression of KCC2 is regulated by the neurotrophic factor BDNF(Aguado et al., 2003) and, together with cytosolic carboanhydrase,KCC2 is an important factor for regulating developmental switch fromexcitatory to inhibitory action of GABA (Rivera et al., 2005). mRNA ofboth proteins are regulated postnatally when the GABAergic systemswitches into final hyperpolarization state. Since we observed no differ-ence in migration of IN in dystonic hamster, we investigated the matu-ration of striatal IN by these two markers of the postnatal GABA switch.By qPCR, cytosolic carboanhydrase 7 (CAVII) and the potassium-chlo-ride-cotransporter 2 (KCC2, gene: Slc12a5) expression were detectedin the striatum of hamsters (Fig. 4A, B). There were no differences inthe relative expression of both mRNAs between mutant and controlhamster at 18 or 33 days of age.

Fig. 4. Relative mRNA expression of GABA switch related proteins in the striatum (qPCR). Therbetween control and dtsz (dt) hamsters at age of 18 days (d) and of maximum severity of dysnormalized to means of control groups; statistics were performed on raw data (RQ, relative qu

3.5. Reduction of striatal alpha 1 GABAAR expressing neurons in 33 days olddtsz mutants

Furthermaturation of IN is characterized by a switch in expression ofGABAAR subunits as shown for cortical IN (Cuzon Carlson and Yeh,2011). This is in linewith expression of GABAAR-α1 typically in GAD ex-pressing GABAergic IN in the striatum of adult animals and humans(Fritschy, 2015). We confirmed expression of GABAAR-α1 in Nkx2.1-positive IN in the striatum (Fig. 5A–C). In order to determine if theseIN are affected in the dtsz hamster, their density was analyzed using un-biased stereology. As shown in Fig. 6A–E, a 37.5% decrease in GABAAR-α1 immunoreactive neurons in the striatum of dtsz hamsters comparedto controls became evident. The reduction included the anterior, medialand posterior striatum (Fig. 6F). Concordant to what has been previous-ly observed for GABAergic IN labeled with PV, NOS or calretinin, thenumber of GABAAR-α1 immunoreactive neurons was normalized in90 days old dtsz hamsters compared to controls (Fig. 6G).

We further analyzed whether the reduction in staining is also obvi-ous on the single neuron level by analyzing immunofluorescence inten-sity in the cell body and in cellular processes of GABAAR-α1-positivestriatal neurons. For this analysis, images of the medial striatum weredivided into 4 subregions (dorsolateral, dorsomedial, ventrolateral, ven-tromedial) as done previously in the dtsz hamster (Gernert et al., 2000).There was a trend (p b 0.1) for an interaction between genotype andsubregion (2× 4 repeatedmeasures ANOVA). Based on previous studiesmost pronounced changes could be expected in the dorsomedial subre-gion (Gernert et al., 1999). Comparisons supported a significant reduc-tion of GABAAR-α1 immunostaining in this region in the whole striatal

e were no differences in the mRNA expression of (A) CAVII or (B) Slc12a5 in the striatumtonia (33 days) (Mann Whitney U for each age group). Values (mean + SEM) are shownantity). n numbers are given inside respective bar.

Fig. 5. GABAAR-α1 is expressed in Nkx2.1-positive cells in the striatum. (A) Nkx2.1, (B) GABAAR-α1 and (C) composite image (reference bars = 20 μm).

49C. Bode et al. / Experimental Neurology 287 (2017) 44–53

section (14%) and the cell bodies (32%) (Table 1), while in the cellularprocesses there were no significant genotype differences (Table 1).

In order to ensure that GABAAR are expressed in IN in the dtsz ham-ster similar to controls, striatal sections were co-stained for Nkx2.1 and

Fig. 6. Quantification of GABAAR-α1-positive cells in the striatum (A–D): representative imageWhite arrows indicate GABAAR-α1-positive cells (reference bars= 20 μm). The density of GABA(n= 6) as confirmed by unbiased stereology for (E) the whole striatum (mean+ SEM, studentgroup, mean + SEM, student's t-test). Values and analysis for the visibly lower fluorescence inTable 1. (G) At 90 days of age after remission of dystonia the density of GABAAR-α1-positive ce

GABAAR-β2 (Fig. 7A–F). As expected thisβ-subunitwas expressed in thestriatum with a staining pattern similar to what has been described inrats (Pirker et al., 2000; Schwarzer et al., 2001). Importantly, theβ2-sub-unit co-localized with Nkx2.1 in control and dtsz hamsters (observation

s of the striatum of control hamsters (A, B) and dtsz (dt) hamsters (C, D) at 33 days of age.AR-α1-positive cells is reduced in 33 days old dtsz hamsters (n= 6) compared to controls's t-test). This effect is consistent across (F) different subregions of the striatum (n= 6 pertensity in the tissue and the GABAAR-α1-positive cell body in the dtsz hamster are given inlls in dtsz hamsters normalized (mean + SEM, student's t-test p = 0.235).

Table 1GABAAR-α1 immunofluorescence pixel intensity in the striatum.

Measured in Striatal subregion (mean pixel intensity ± SEM)

Dorsolateral Dorsomedial Ventrolateral Ventromedial

Control dt Control dt Control dt Control dt

Striatum (total) 10.11 ± 0.34 10.42 ± 0.36 10.72 ± 0.56 9.22⁎ ± 0.36 10.87 ± 0.83 8.91 ± 0.50 9.92 ± 0.92 8.61 ± 0.38Cell bodies 20.32 ± 3.44 19.14 ± 1.21 25.18 ± 2.28 17.17⁎ ± 1.61 26.16 ± 3.31 18.77 ± 1.86 20.00 ± 2.54 17.93 ± 1.22Cell processes 15.28 ± 1.75 12.90 ± 1.26 14.49 ± 1.58 14.12 ± 0.70 18.09 ± 2.89 11.38 ± 0.89 12.20 ± 0.55 12.45 ± 0.67

⁎ p b 0.05, student's t-test, n = 6 per group.

50 C. Bode et al. / Experimental Neurology 287 (2017) 44–53

of striatal sections from n = 4 animals/genotype) supporting presenceof GABAAR on striatal IN (Fig. 7A–F).

3.6. Reduction of GAD67 expressing neurons in the striatum of the dtsz

mutant

GAD is required for synthesis of GABA from glutamate and is thusexpressed by GABAergic neurons. Striatal GABAergic IN express prefer-ably the 67 kDa isoform,with PV IN expressing the highest level of GAD-67. All PV mRNA expressing neurons also express GAD67 mRNA, and75% of cells expressing high levels of GAD67 mRNA also express PV

Fig. 7.GABAAR are expressed onNkx2.1-positive cells in the striatum. Nxk2.1 (A, D) andGABAARF) show overlay of images acquired with confocal microscopy (reference bars = 10 μm).

Fig. 8. Quantification of GAD67-positive cells in the striatum (A, B): representative images ofGAD67-positive cells (reference bars = 10 μm). The density of GAD67-positive cells (C) in theby unbiased stereology (mean + SEM, one-tailed student's t-test, *p b 0.05).

mRNA (Lenz et al., 1994). Calretinin- and NOS-positive GABAergic IN,which are the other two IN types reduced in the dtsz hamster, also ex-press GAD-67 (Kubota et al., 1993). We found a 10% decrease inGAD67-positive neurons (one-tailed student t-test: p = 0.042) in thestriatum of dtsz hamsters compared to controls at 33 days of age (Fig.8A–C).

Co-labeling of medial striatal sections of dtsz and control hamstersfor GABAAR-α1 and GAD67 revealed that GAD67 was stronglyexpressed in N88% of all GABAAR-α1-positive neurons in dtsz hamstersand controls (Fig. 9A–D), supporting previous studies in rats andhumans showing that GABAAR-α1 expressing neurons are mainly

-β2 (B, E) co-localize in control hamsters (A–C) and dtszhamsters (dt, D–F). Composites (C,

the striatum of a control hamster (A) and a dtsz (dt) hamster (B). White arrows indicatestriatum is reduced in dtsz hamsters (n = 6) compared to controls (n = 5), determined

Fig. 9.Quantification of GABAAR-α1/GAD67-positive cells in the striatum (A–C): representative images of the striatumof a control hamster.White arrows indicate positive cells (referencebars= 20 μm). The % of GABAAR-α1/GAD67-positive cells of (D) all GABAAR-α1 and (E) all GAD67 cells is similar in dtsz hamsters (n= 6) compared to controls (n= 5) as determined byunbiased stereology (% of means, Fisher exact test p N 0.05).

51C. Bode et al. / Experimental Neurology 287 (2017) 44–53

GABAergic IN (Fritschy, 2015). However, b30% of GAD67-positive neu-rons also expressed GABAAR-α1 at 33 days of age in dtsz hamsters orcontrols (Fig. 9E) which is in line with previous observations thatthese neurons represent only a subgroup of GABAergic IN (Fritschy,2015).

4. Discussion

Several lines of evidence suggest that reduction of striatal GABAergicIN and the corresponding loss of inhibition of medium spiny neuronscan result in dyskinesia and dystonic symptoms (Gittis et al., 2011;Reiner et al., 2013; Richter, 2005). First, data on the pathophysiologicalrole of disinhibition by deficient striatal GABAergic IN associated withaltered neuronal activity in basal ganglia nuclei and disturbed striatalplasticity are based on previous studies in the dtszhamstermodel of par-oxysmal dystonia (Avchalumov et al., 2013; Gernert et al., 2002;Gernert et al., 2000; Kohling et al., 2004). However, the molecular andcellular mechanisms responsible for the age-dependent deficit ofstriatal inhibitory PV-, calretinin- and NOS-positive IN (Hamann et al.,2007; Hamann et al., 2005; Sander et al., 2006) remained unclear.Here, we present evidence that in the dtsz hamster striatal GABAergicIN migrate successfully to the striatum but do not mature into theirfunctional phenotype postnatally.

The deficit of all types of striatal GABAergic IN in 33 days old dtszmu-tant hamsters, the age of maximum expression of dystonia, and the pat-tern of reduction in subregions of the striatum pointed to a retardedmigration (Hamann et al., 2007; Sander et al., 2006). However, quanti-fication was performed using specific phenotypic markers, i.e. PV, NOSand calretinin. Therefore, a mere lack of expression of these markersdue to delayed maturation would have prevented their detection forquantification. PV-positive IN for example do not mature molecularly(increase in PVmRNA and protein) and functionally (fast-spiking firingpattern) until the 2nd− 4th postnatal weeks, as shown by PV immuno-histochemistry (Schlosser et al., 1999), electrophysiological recordingsand molecular studies (Plotkin et al., 2005). Quantification of IN usinga marker which is stably expressed at birth and into adulthood was

required to differentiate between delayed migration and a maturationdeficit. Therefore, in the present study, striatal IN were quantified viaunbiased stereology using the homeodomain protein Nkx2.1 as a specif-ic marker for most striatal IN populations. Nkx2.1 is expressed duringembryonic neuronal migration as well as in mature IN (Hamasaki etal., 2003; Marin et al., 2000). The unchanged density of Nkx2.1-positiveneurons in the striatumof dtszhamsters compared to controls at 33 daysof age, and the physiological expression of Nkx2.1 mRNA at 18 and33 days of age, suggest that themajority of striatal IN has physiologicallycompleted migration and thus is present in the striatum at 33 days ofage. This is in concordance with the unchanged density of striatal cho-linergic IN (Hamann et al., 2006) and cortical PV IN (Hamann et al.,2007)which share the samedevelopmental origin and a large set of reg-ulating factors during migration with striatal GABAergic IN (Elias et al.,2008; Marin et al., 2000; Villar-Cervino et al., 2015).

Conversely, cells labeled for the GABA synthesizing enzyme GAD67were slightly reduced, which is in line with previous studies showingreducedGAD (without isoformdifferentiation)mRNAexpression via ol-igonucleotide hybridization histochemistry (Burgunder et al., 1999) anda small decrease in striatal GABA levels (Loscher and Horstermann,1992). GAD67 is foremost expressed by GABAergic IN in the striatumand, different from PV, expression is high at birth and does not increaseduringmaturation, but can bemodulated by changes in firing activity ofPV IN (Chesselet et al., 1993; Plotkin et al., 2005; Qin et al., 1992;Soghomonian et al., 1992). The reduction in GAD67 labeled neurons inthe dtsz hamster was subtle (−10%) compared to previously reporteddecreases in PV-positive IN (−41%) (Gernert et al., 2000), NOS-positiveIN (−21%) (Sander et al., 2006) and calretinin IN (−20%) (Hamann etal., 2005) which together with the unchanged Nkx2.1 expressionagain does not support retarded migration of GABAergic IN. Instead,the small reduction of GAD67 expression could be related to lowerfiringactivity of less mature striatal GABAergic IN in the dtsz mutant which isthought to underlie the disinhibition ofMSNand in consequence the de-creased basal ganglia output (Kohling et al., 2004).

This conclusion is further supported by lower PV mRNA expressionin the striatum of dystonic hamsters which corresponds to previously

52 C. Bode et al. / Experimental Neurology 287 (2017) 44–53

reported reduced numbers of the highly PV expressing cells (Hamann etal., 2007). The lack of PV, which serves as intracellular calcium buffer inthese fast-spiking GABAergic IN (Plotkin et al., 2005) and tunes theirspike-timing and efferent short-term plasticity (Orduz et al., 2013), islikely involved in corticostriatal plasticity alterations in the dtsz mutant(Avchalumov et al., 2014; Bennay et al., 2001; Gernert et al., 2002;Gernert et al., 2000; Kohling et al., 2004). However, it is unclear if thedelay in full expression of the PV phenotype duringmaturation is a sec-ondary response to reduced firing activity or represents a primary de-fect in this animal model, and potentially in the pathophysiology ofdystonia.

In order to begin answering this question, the expression of severalkey factors of striatal IN maturation was determined in this study.First we found that BDNF mRNA or protein, a neurotrophin importantfor the maturation of striatal IN (Eto et al., 2010), was not reduced inthe dtsz hamster model. Previous pharmacological studies suggested aretarded postnatal transition of neuronal GABAergic neurotransmissionfrom excitatory to inhibitory, known as the GABA switch, in dystonichamsters (Richter and Hamann, 2004; Richter and Loscher, 2000). TheGABA switch relies on elevation of expression of the chloride exporterKCC2 at the end of the first postnatal week in rodents which togetherwith CAVII provides the ion concentration required to drive the influxof chloride through activated GABAARs for hyperpolarization (Riveraet al., 2005; Valeeva et al., 2013; Yeo et al., 2009). The lack of differencesin mRNA expression of KCC2 or CAVII in mutant hamsters at 18 and33 days of age (maximum expression of dystonia) compared to controlsindicates unchangedmaturation of this function of striatal GABAAR. Fur-thermore, these data suggest that an increased firing rate of striatal me-dium spiny neurons (Gernert et al., 1999), which is important for themanifestation of dystonia as indicated by intrastriatal injections ofGABA potentiating compounds (Hamann and Richter, 2002; Kreil andRichter, 2005), is not related to an abnormal GABA switch. However,we cannot exclude differences in GABA switch maturation or in BDNFlevels prior to 18 days of age and/or impairments in the GABA switchspecifically in striatal GABAergic IN, which are underrepresented inthe analysis of the whole striatum. Studies in younger animals andgene expression analyses in isolated neurons will be performed in fu-ture studies to confirm the physiological expression of the GABA switchrelatedmRNAs. Furthermore, electrophysiological studies will ascertainthe expected functional implication of unchanged mRNA expression.

In order to study maturation of the GABAAR more specifically instriatal IN, we took advantage of the specific andmaturation dependentexpression of GABAAR alpha subunits (Fritschy, 2015). As expected, ourco-labelingwith GAD67 confirmed that mature GABAergic IN in controlhamsters express α1β2γ2 GABAAR (Waldvogel et al., 1999). Interest-ingly, unbiased stereology revealed 37.5% less GABAAR-α1-positive neu-rons in the striatum of 33 days old dtsz mutant hamsters compared tocontrols. Reduction of GABAAR-α1 expression on the cell bodies of INin the dorsomedial striatum supports that this deficit is part of the re-tarded maturation of striatal IN in the dtsz hamster and is coherentwith subregional differences found in previous electrophysiologicalstudies (Gernert et al., 1999). Importantly, this reduction in GABAAR-α1 expressing neurons is transient and was normalized after remissionof dystonia (at 90 days of age), similar to the reduction in PV, calretininand NOS staining (Hamann et al., 2007). The expression of GABAA-β2 inNkx2.1-positive neurons at 33 days of age support that GABAAR are in-deed present on striatal IN in the dtszhamster. Ongoing studieswill clar-ify if striatal IN in 33 days old dtsz hamsters express GABAAR-α2, whichwould indicate a delay in the postnatal switch from α2 to α1 subunitcontaining GABAAR (Cuzon Carlson and Yeh, 2011) with functional im-plications, as receptors with the α2 subunit open two times faster andthe ion diffusion lasts six times longer in comparison with GABAAR-α1

(Lavoie et al., 1997), leading to stronger GABAergic inhibition of IN inthe dystonic hamster. If such a delay in GABAAR maturation is the pri-mary event leading to reduced PV and GAD67 expression will need tobe determined by studies in younger animals.

In conclusion our data provides evidence that in the dtsz hamstermodel of dystonia striatal GABAergic IN show delayed maturation to afunctional phenotype which is in line with results from previous elec-trophysiological and pharmacological studies. GABAergic IN have suc-cessfully migrated to the striatum but do not sufficiently express keyphenotypic and functional proteins such as PV, GAD67 and GABAAR-α1. The identification of the gene defect in the dtsz hamster is hamperedby incomplete knowledge of the Syrian hamster genome. However, re-sults discussed above and observed in previous studies clearly point to-wards a defect in a regulatory pathway important for the maturation ofstriatal GABAergic IN. Therefore the dystonic hamster model provides aunique opportunity to understand retardation of postnatal develop-ment of striatal GABAergic IN and how this deficit results in an overtdystonic phenotype.

Acknowledgements

This studywas supported by theDystoniaMedical Research Founda-tion. We thank Steffi Fuchs for excellent technical assistance.

References

Aguado, F., Carmona, M.A., Pozas, E., Aguilo, A., Martinez-Guijarro, F.J., Alcantara, S.,Borrell, V., Yuste, R., Ibanez, C.F., Soriano, E., 2003. BDNF regulates spontaneous corre-lated activity at early developmental stages by increasing synaptogenesis and expres-sion of the K+/Cl- co-transporter KCC2. Development 130, 1267–1280.

Albanese, A., Bhatia, K., Bressman, S.B., Delong, M.R., Fahn, S., Fung, V.S., Hallett, M.,Jankovic, J., Jinnah, H.A., Klein, C., Lang, A.E., Mink, J.W., Teller, J.K., 2013. Phenomenol-ogy and classification of dystonia: a consensus update. Mov. Disord. 28, 863–873.

Altar, C.A., Cai, N., Bliven, T., Juhasz, M., Conner, J.M., Acheson, A.L., Lindsay, R.M.,Wiegand,S.J., 1997. Anterograde transport of brain-derived neurotrophic factor and its role inthe brain. Nature 389, 856–860.

Avchalumov, Y., Volkmann, C.E., Ruckborn, K., Hamann, M., Kirschstein, T., Richter, A.,Kohling, R., 2013. Persistent changes of corticostriatal plasticity in dt(sz) mutanthamsters after age-dependent remission of dystonia. Neuroscience 250, 60–69.

Avchalumov, Y., Sander, S.E., Richter, F., Porath, K., Hamann, M., Bode, C., Kirschstein, T.,Kohling, R., Richter, A., 2014. Role of striatal NMDA receptor subunits in a model ofparoxysmal dystonia. Exp. Neurol. 261, 677–684.

Bennay, M., Gernert, M., Richter, A., 2001. Spontaneous remission of paroxysmal dystoniacoincides with normalization of entopeduncular activity in dt(SZ) mutants.J. Neurosci. 21, RC153.

Bortone, D., Polleux, F., 2009. KCC2 expression promotes the termination of cortical inter-neuron migration in a voltage-sensitive calcium-dependent manner. Neuron 62,53–71.

Breakefield, X.O., Blood, A.J., Li, Y., Hallett, M., Hanson, P.I., Standaert, D.G., 2008. The path-ophysiological basis of dystonias. Nat. Rev. Neurosci. 9, 222–234.

Burgunder, J., Richter, A., Loscher, W., 1999. Expression of cholecystokinin, somatostatin,thyrotropin-releasing hormone, glutamic acid decarboxylase and tyrosine hydroxy-lase genes in the central nervous motor systems of the genetically dystonic hamster.Exp. Brain Res. 129, 114–120.

Chesselet, M.F., Mercugliano, M., Soghomonian, J.J., Salin, P., Qin, Y., Gonzales, C., 1993.Regulation of glutamic acid decarboxylase gene expression in efferent neurons ofthe basal ganglia. Prog. Brain Res. 99, 143–154.

Cuzon Carlson, V.C., Yeh, H.H., 2011. GABAA receptor subunit profiles of tangentially mi-grating neurons derived from the medial ganglionic eminence. Cereb. Cortex 21,1792–1802.

Elias, L.A., Potter, G.B., Kriegstein, A.R., 2008. A time and a place for nkx2-1 in interneuronspecification and migration. Neuron 59, 679–682.

Eto, R., Abe, M., Kimoto, H., Imaoka, E., Kato, H., Kasahara, J., Araki, T., 2010. Alterations ofinterneurons in the striatum and frontal cortex of mice during postnatal develop-ment. Int. J. Dev. Neurosci. 28, 359–370.

Fritschy, J.M., 2015. Significance of GABA(A) receptor heterogeneity: clues from develop-ing neurons. Adv. Pharmacol. 73, 13–39.

Garibotto, V., Romito, L.M., Elia, A.E., Soliveri, P., Panzacchi, A., Carpinelli, A., Tinazzi, M.,Albanese, A., Perani, D., 2011. In vivo evidence for GABA(A) receptor changes in thesensorimotor system in primary dystonia. Mov. Disord. 26, 852–857.

Gernert, M., Richter, A., Loscher, W., 1999. Alterations in spontaneous single unit activityof striatal subdivisions during ontogenesis in mutant dystonic hamsters. Brain Res.821, 277–285.

Gernert, M., Hamann, M., Bennay, M., Loscher, W., Richter, A., 2000. Deficit of striatalparvalbumin-reactive GABAergic interneurons and decreased basal ganglia outputin a genetic rodent model of idiopathic paroxysmal dystonia. J. Neurosci. 20,7052–7058.

Gernert, M., Bennay, M., Fedrowitz, M., Rehders, J.H., Richter, A., 2002. Altered dischargepattern of basal ganglia output neurons in an animal model of idiopathic dystonia.J. Neurosci. 22, 7244–7253.

Gittis, A.H., Kreitzer, A.C., 2012. Striatal microcircuitry and movement disorders. TrendsNeurosci. 35, 557–564.

53C. Bode et al. / Experimental Neurology 287 (2017) 44–53

Gittis, A.H., Leventhal, D.K., Fensterheim, B.A., Pettibone, J.R., Berke, J.D., Kreitzer, A.C.,2011. Selective inhibition of striatal fast-spiking interneurons causes dyskinesias.J. Neurosci. 31, 15727–15731.

Hamann, M., Richter, A., 2002. Effects of striatal injections of GABA(A) receptor agonistsand antagonists in a genetic animal model of paroxysmal dystonia. Eur.J. Pharmacol. 443, 59–70.

Hamann, M., Sander, S.E., Richter, A., 2005. Age-dependent alterations of striatal calretinininterneuron density in a genetic animal model of primary paroxysmal dystonia.J. Neuropathol. Exp. Neurol. 64, 776–781.

Hamann, M., Raymond, R., Varughesi, S., Nobrega, J.N., Richter, A., 2006. Acetylcholine re-ceptor binding and cholinergic interneuron density are unaltered in a genetic animalmodel of primary paroxysmal dystonia. Brain Res. 1099, 176–182.

Hamann, M., Richter, A., Meillasson, F.V., Nitsch, C., Ebert, U., 2007. Age-related changes inparvalbumin-positive interneurons in the striatum, but not in the sensorimotor cor-tex in dystonic brains of the dt(sz) mutant hamster. Brain Res. 1150, 190–199.

Hamasaki, T., Goto, S., Nishikawa, S., Ushio, Y., 2003. Neuronal cell migration for the devel-opmental formation of the mammalian striatum. Brain Res. Brain Res. Rev. 41, 1–12.

Hofer, M., Pagliusi, S.R., Hohn, A., Leibrock, J., Barde, Y.A., 1990. Regional distribution ofbrain-derived neurotrophic factor mRNA in the adult mouse brain. EMBO J. 9,2459–2464.

Hu, H., Gan, J., Jonas, P., 2014. Interneurons. Fast-spiking, parvalbumin(+) GABAergic in-terneurons: from cellular design to microcircuit function. Science 345, 1255263.

Kohling, R., Koch, U.R., Hamann, M., Richter, A., 2004. Increased excitability in cortico-striatal synaptic pathway in a model of paroxysmal dystonia. Neurobiol. Dis. 16,236–245.

Kralic, J.E., Sidler, C., Parpan, F., Homanics, G.E., Morrow, A.L., Fritschy, J.M., 2006. Compen-satory alteration of inhibitory synaptic circuits in cerebellum and thalamus ofgamma-aminobutyric acid type A receptor alpha1 subunit knockout mice. J. Comp.Neurol. 495, 408–421.

Kreil, A., Richter, A., 2005. Antidystonic efficacy of gamma-aminobutyric acid uptake in-hibitors in the dtsz mutant. Eur. J. Pharmacol. 521, 95–98.

Kubota, Y., Mikawa, S., Kawaguchi, Y., 1993. Neostriatal GABAergic interneurones containNOS, calretinin or parvalbumin. Neuroreport 5, 205–208.

Lavoie, A.M., Tingey, J.J., Harrison, N.L., Pritchett, D.B., Twyman, R.E., 1997. Activation anddeactivation rates of recombinant GABA(A) receptor channels are dependent onalpha-subunit isoform. Biophys. J. 73, 2518–2526.

Lenz, S., Perney, T.M., Qin, Y., Robbins, E., Chesselet, M.F., 1994. GABA-ergic interneuronsof the striatum express the Shaw-like potassium channel Kv3.1. Synapse 18, 55–66.

Lessmann, V., Brigadski, T., 2009. Mechanisms, locations, and kinetics of synaptic BDNFsecretion: an update. Neurosci. Res. 65, 11–22.

Loscher, W., Horstermann, D., 1992. Abnormalities in amino acid neurotransmitters indiscrete brain regions of genetically dystonic hamsters. J. Neurochem. 59, 689–694.

Marin, O., Rubenstein, J.L., 2003. Cell migration in the forebrain. Annu. Rev. Neurosci. 26,441–483.

Marin, O., Anderson, S.A., Rubenstein, J.L., 2000. Origin and molecular specification ofstriatal interneurons. J. Neurosci. 20, 6063–6076.

Mohler, H., 2011. The rise of a new GABA pharmacology. Neuropharmacology 60,1042–1049.

Orduz, D., Bischop, D.P., Schwaller, B., Schiffmann, S.N., Gall, D., 2013. Parvalbumin tunesspike-timing and efferent short-term plasticity in striatal fast spiking interneurons.J. Physiol. 591, 3215–3232.

Pappas, S.S., Leventhal, D.K., Albin, R.L., Dauer, W.T., 2014. Mouse models ofneurodevelopmental disease of the basal ganglia and associated circuits. Curr. Top.Dev. Biol. 109, 97–169.

Petzold, A., Psotta, L., Brigadski, T., Endres, T., Lessmann, V., 2015. Chronic BDNF deficiencyleads to an age-dependent impairment in spatial learning. Neurobiol. Learn. Mem.120, 52–60.

Pirker, S., Schwarzer, C., Wieselthaler, A., Sieghart, W., Sperk, G., 2000. GABA(A) receptors:immunocytochemical distribution of 13 subunits in the adult rat brain. Neuroscience101, 815–850.

Plotkin, J.L., Wu, N., Chesselet, M.F., Levine, M.S., 2005. Functional and molecular develop-ment of striatal fast-spiking GABAergic interneurons and their cortical inputs. Eur.J. Neurosci. 22, 1097–1108.

Polleux, F., Whitford, K.L., Dijkhuizen, P.A., Vitalis, T., Ghosh, A., 2002. Control of corticalinterneuron migration by neurotrophins and PI3-kinase signaling. Development129, 3147–3160.

Qin, Y., Soghomonian, J.J., Chesselet, M.F., 1992. Effects of quinolinic acid on messengerRNAs encoding somatostatin and glutamic acid decarboxylases in the striatum ofadult rats. Exp. Neurol. 115, 200–211.

Ramanathan, S., Hanley, J.J., Deniau, J.M., Bolam, J.P., 2002. Synaptic convergence of motorand somatosensory cortical afferents onto GABAergic interneurons in the rat stria-tum. J. Neurosci. 22, 8158–8169.

Reiner, A., Shelby, E., Wang, H., Demarch, Z., Deng, Y., Guley, N.H., Hogg, V., Roxburgh, R.,Tippett, L.J., Waldvogel, H.J., Faull, R.L., 2013. Striatal parvalbuminergic neurons arelost in Huntington's disease: implications for dystonia. Mov. Disord. 28, 1691–1699.

Richter, A., 2005. The genetically dystonic hamster: an animal model of paroxysmal dys-tonia. In: LeDoux,M. (Ed.), AnimalModels of Movement Disorders. Elsevier AcademicPress, Amsterdam, pp. 459–466.

Richter, A., Hamann, M., 2004. The carbonic anhydrase inhibitor acetazolamide exertsantidystonic effects in the dt(sz) mutant hamster. Eur. J. Pharmacol. 502, 105–108.

Richter, A., Loscher, W., 1998. Pathology of idiopathic dystonia: findings from genetic an-imal models. Prog. Neurobiol. 54, 633–677.

Richter, A., Loscher, W., 2000. Paradoxical aggravation of paroxysmal dystonia duringchronic treatment with phenobarbital in a genetic rodent model. Eur. J. Pharmacol.397, 343–350.

Richter, F., Richter, A., 2014. Genetic animal models of dystonia: common features and di-versities. Prog. Neurobiol. 121, 91–113.

Richter, F., Gao, F., Medvedeva, V., Lee, P., Bove, N., Fleming, S.M., Michaud,M., Lemesre, V.,Patassini, S., De La Rosa, K., Mulligan, C.K., Sioshansi, P.C., Zhu, C., Coppola, G., Bordet,T., Pruss, R.M., Chesselet, M.F., 2014. Chronic administration of cholesterol oximes inmice increases transcription of cytoprotective genes and improves transcriptome al-terations induced by alpha-synuclein overexpression in nigrostriatal dopaminergicneurons. Neurobiol. Dis. 69, 263–275.

Rivera, C., Voipio, J., Kaila, K., 2005. Two developmental switches in GABAergic signalling:the K + −Cl-cotransporter KCC2 and carbonic anhydrase CAVII. J. Physiol. 562,27–36.

Sander, S.E., Hamann, M., Richter, A., 2006. Age-related changes in striatal nitric oxidesynthase-immunoreactive interneurones in the dystonic dtsz mutant hamster.Neuropathol. Appl. Neurobiol. 32, 74–82.

Schlosser, B., Klausa, G., Prime, G., Ten Bruggencate, G., 1999. Postnatal development ofcalretinin- and parvalbumin-positive interneurons in the rat neostriatum: an immu-nohistochemical study. J. Comp. Neurol. 405, 185–198.

Schwarzer, C., Berresheim, U., Pirker, S., Wieselthaler, A., Fuchs, K., Sieghart, W., Sperk, G.,2001. Distribution of the major gamma-aminobutyric acid(A) receptor subunits inthe basal ganglia and associated limbic brain areas of the adult rat. J. Comp. Neurol.433, 526–549.

Soghomonian, J.J., Gonzales, C., Chesselet, M.F., 1992. Messenger RNAs encoding gluta-mate-decarboxylases are differentially affected by nigrostriatal lesions in subpopula-tions of striatal neurons. Brain Res. 576, 68–79.

Valeeva, G., Valiullina, F., Khazipov, R., 2013. Excitatory actions of GABA in the intact neo-natal rodent hippocampus in vitro. Front. Cell. Neurosci. 7, 20.

Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A., Speleman,F., 2002. Accurate normalization of real-time quantitative RT-PCR data by geometricaveraging of multiple internal control genes. Genome Biol. 3 RESEARCH0034.

Villar-Cervino, V., Kappeler, C., Nobrega-Pereira, S., Henkemeyer, M., Rago, L., Nieto, M.A.,Marin, O., 2015. Molecular mechanisms controlling themigration of striatal interneu-rons. J. Neurosci. 35, 8718–8729.

Waldvogel, H.J., Kubota, Y., Fritschy, J., Mohler, H., Faull, R.L., 1999. Regional and cellularlocalisation of GABA(A) receptor subunits in the human basal ganglia: an autoradio-graphic and immunohistochemical study. J. Comp. Neurol. 415, 313–340.

Watanabe, M., Fukuda, A., 2015. Development and regulation of chloride homeostasis inthe central nervous system. Front. Cell. Neurosci. 9, 371.

Weise, D., Schramm, A., Beck, M., Reiners, K., Classen, J., 2011. Loss of topographic speci-ficity of LTD-like plasticity is a trait marker in focal dystonia. Neurobiol. Dis. 42,171–176.

Yeo, M., Berglund, K., Augustine, G., Liedtke, W., 2009. Novel repression of Kcc2 transcrip-tion by REST-RE-1 controls developmental switch in neuronal chloride. J. Neurosci.29, 14652–14662.