Neuronal firing activity in the basal ganglia after striatal transplantation of dopamine neurons in hemiparkinsonian rats

Accepted Manuscript Neuronal firing activity in the basal ganglia after striatal transplantation of dopamine neurons in hemiparkinsonian rats Regina Rumpel, Mesbah Alam, Lisa M. Schwarz, Andreas Ratzka, Xingxing Jin, Joachim K. Krauss, Claudia Grothe, Kerstin Schwabe PII: DOI: Reference:

S0306-4522(17)30555-9 http://dx.doi.org/10.1016/j.neuroscience.2017.07.069 NSC 17946

To appear in:

Neuroscience

Received Date: Accepted Date:

17 July 2017 28 July 2017

Please cite this article as: R. Rumpel, M. Alam, L.M. Schwarz, A. Ratzka, X. Jin, J.K. Krauss, C. Grothe, K. Schwabe, Neuronal firing activity in the basal ganglia after striatal transplantation of dopamine neurons in hemiparkinsonian rats, Neuroscience (2017), doi: http://dx.doi.org/10.1016/j.neuroscience.2017.07.069

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Neuronal firing activity in the basal ganglia after striatal transplantation of dopamine neurons in hemiparkinsonian rats Regina Rumpel a,*, Mesbah Alamb,*, Lisa M. Schwarza,*, Andreas Ratzkaa, Xingxing Jinb,1, Joachim K. Kraussb,c, Claudia Grothea,c,#, Kerstin Schwabeb,c,# a

Institute of Neuroanatomy and Cell Biology, Hannover Medical School, 30625

Hannover, Germany;

b

Clinic of Neurosurgery, Hannover Medical School, 30625

Hannover, Germany; cCentre for Systems Neuroscience (ZSN), Hannover Medical School, 30625 Hannover, Germany; 1Present address: Clinic of Neurosurgery, Shanghai Jiao Tong University Affiliated First People’s Hospital, 200080 Shanghai, People’s Republic China * Those three authors contributed equally to this work. # Both senior authors contributed equally to this work.

To whom correspondence should be addressed: Prof. Dr. Claudia Grothe Hannover Medical School Institute of Neuroanatomy and Cell Biology (OE4140) Carl-Neuberg-Str. 1 30625 Hannover, Germany Phone: +49-511-532-2896 Fax: +49-511-532-2880 E-mail: [email protected]

1

ABBREVIATIONS 6-OHDA

6-hydroxydopamine

AP

anteroposterior

BG

basal ganglia

CV

coefficient of variation

DA

dopamine

DBS

deep brain stimulation

DV

dorsoventral

ECoG

electrocorticogram

EPN

entopeduncular nucleus

GP

globus pallidus

GPe

globus pallidus externus

GPi

globus pallidus internus

IHC

immunohistochemistry

ISI

inter-spike interval

LFPs

local field potentials

MCtx

motor cortex

MFB

medial forebrain bundle

ML

mediolateral

PBS

phosphate-buffered saline

PD

Parkinson’s disease

REM

rapid eye movement 2

SEM

standard error of the mean

STN

subthalamic nucleus

STR

striatum

STWA

spike triggered waveform average

SU

single unit

TB

tooth bar

TH

tyrosine hydroxylase

TX

transplantation

3

ABSTRACT The loss of nigral dopaminergic neurons and the resulting dopamine (DA) depletion in the striatum (STR) lead to altered neuronal activity and enhanced beta activity in various regions of the basal ganglia (BG) motor loop in patients with Parkinson’s disease and in rodents in the 6-hydroxydopamine (6-OHDA) lesioned rat model. Intrastriatal DA graft implantation has been shown to re-innervate the host brain and restore DA input. Here, DA cell grafts were implanted into the STR of 6-OHDA lesioned rats and the effect on neuronal activity under urethane anesthesia (1.4 g/kg, injected intraperitoneally) was tested in the entopeduncular nucleus (EPN, the equivalent to the human globus pallidus internus), the output nucleus of the BG, and the globus pallidus (GP, the equivalent to the human globus pallidus externus), a key region in the indirect pathway. In animals, which were transplanted with cells derived from the ventral mesencephalon of embryonic day 12 rat embryos into the STR, the rotational behavior induced by DA agonists in 6-OHDA lesioned rats was significantly improved. This was accompanied by alleviated EPN firing rate and reinstated patterns of neuronal activity in the GP and EPN. Analysis of oscillatory activity revealed enhanced beta activity in both regions, which was reduced after grafting. In summary these data indicate restoration of BG motor loop towards normal activity by DA graft integration.

KEYWORDS: 6-OHDA; basal ganglia; dopamine; electrophysiology; oscillatory activity; transplantation

4

INTRODUCTION In Parkinson’s disease (PD), loss of nigrostriatal dopamine (DA) leads to reduced dopaminergic transmission in the striatum (STR) and disturbed neuronal activity in the direct and indirect pathway of the basal ganglia (BG) motor loop. This includes not only altered firing rates as suggested by the classic BG scheme, but also higher events of burst patterns and irregular firing, as well as disturbed oscillatory activity. Together, these changes functionally compromise related thalamic and cortical areas (DeLong and Wichmann, 2007; Galvan et al., 2015). Motor symptoms of PD are initially treated by administration of DA receptor agonists, which restore neuronal firing rates and beta oscillatory activity in the subthalamic nucleus (STN), although burst activity is not affected or even further enhanced (Levy et al., 2001; Weinberger et al., 2006). Longterm treatment with L-DOPA, however, causes severe motor complications, possibly because of the pulsatile application of treatment (Cenci and Lindgren, 2007). Intrastriatal transplantation of dopaminergic neurons may be an alternative to restore deficient DA supply allowing a more physiological and less pulsatile DA delivery. Recently, there has been renewed interest in this treatment because, in contrast to other therapies, it might both stop progression of disease and restore altered physiology (Barker et al., 2015; Barker et al., 2016). In this context, Richardson et al. (2011) reported on a PD patient with partially restored globus pallidus internus neuronal activity after transplantation of DA cells into the STR. Using the 6-hydroxydopamine (6-OHDA) rat model of PD, we previously showed that eight weeks after implantation, DA grafts restore functional deficits and cause partial improvement of STN neuronal activity, i.e., neuronal firing rate and beta oscillatory activity were normalized. Burst activity and measures of irregularity, however, were not affected or even enhanced (Rumpel et al., 2013). To further increase our understanding of neuronal activity in the BG motor loop of a grafted brain, under urethane anesthesia, we examined the effect of striatal DA grafts on neuronal activity of the globus pallidus, which is divided into an internal part (globus pallidus internus, GPi) and an external part (globus pallidus externus, GPe). While in the classical scheme of BG, the GPi is one of the output regions, so far, the GPe has been 5

regarded a relay of the indirect pathway, which sends information to the glutamatergic STN. However, more recent work has redefined the GPe region as central for BG information processing (Deffains et al., 2016; Gurney et al., 2001), since it is reciprocally connected to the STN and to the BG output nuclei (substantia nigra pars reticulata and GPi), but also directly innervates the substantia nigra pars compacta, as well as thalamic and cortical regions. Also, about one third of GPe neurons project back to the STR (Bevan et al., 1998; Kita and Kitai, 1994; Sato et al., 2000). The aim of our present work was to examine the effect of DA graft implantation into the STR of 6-OHDA lesioned rats on neuronal activity of the entopeduncular nucleus (EPN, the homologous structure to the GPi in the rat) and globus pallidus (GP, the homologous structure to the human GPe). Notably, in this study, we chose to evaluate the animals 12 weeks after graft implantation to allow long-term and full integration of grafts into the STR.

6

EXPERIMENTAL PROCEDURES All experiments were conducted in accordance with the German animal protection act and were approved by the local authorities (Bezirksregierung LAVES Hannover, Germany). Animals and experimental design Twenty-one adult female Sprague Dawley rats from Janvier (St. Berthevin, France) were used in this study. The animals weighing 250 g at the start of the experiments were housed under 14 h light/10 h dark cycle with free access to food and water. Five animals served as naive Control group (n = 5). Sixteen animals received a unilateral lesion of the right medial forebrain bundle (MFB) using 6-OHDA and were evaluated in the apomorphine- and amphetamine-induced rotation test six weeks after lesion (pre-TX). Three animals died during the surgical procedure. Ten of the lesioned animals exhibited ≥ four full contralateral body turns/min (apomorphine) and ≥ six full ipsilateral body turns/min (amphetamine), respectively, and were matched into experimental groups 6OHDA and Transplant based on their rotation scores. Eight weeks after lesion, the Transplant group (n = 5) received implantation of DA cell grafts into the right lesioned STR. The 6-OHDA group (n = 5) was kept as hemiparkinsonian control. Ten weeks after grafting, animals from both groups were re-tested in the drug-induced rotation tests (post-TX). The experiment was terminated 12 weeks after transplantation surgery (20 weeks after the 6-OHDA lesion). Electrophysiological recordings in the EPN and GP were performed with subsequent sacrifice and immunohistochemical (IHC) analysis of the STR (each group n = 5; Fig. 1). Lesion surgery Stereotaxic lesion surgery was performed by unilateral injection of a total of 19.8 µg 6OHDA hydrobromide (3.6 µg/µl (calculated as free base) in 0.02 % L-ascorbate-saline; Tocris Bioscience, Bristol, UK) under general anesthesia with chloral hydrate (370 mg/kg; injected intraperitoneally, Sigma-Aldrich, Steinheim, Germany) as described previously (Rumpel et al., 2013). Briefly, animals received two injections of 6-OHDA to target the right MFB at the following coordinates (in mm according to bregma and dura 7

according to Paxinos and Watson (2006): 1) anteroposterior (AP) −4.4, mediolateral (ML) −1.2, dorsoventral (DV) −7.8, tooth bar (TB) −2.4, injection volume 2.5 µl; 2) AP −4.0, ML −0.8, DV −8.0, TB +3.4, injection volume 3 µl. 6-OHDA was delivered using a 10-µl Hamilton syringe with an injection rate of 1 µl/min. The needle was left in the brain for additional 3 min to allow diffusion before being slowly retracted. In addition to chloral hydrate, we applied a few drops of 1 % lidocaine (AstraZeneca GmbH, Wedel, Germany) on the skull of the animals prior to removal of periost and drilling. Also, animals

received

intraoperative

analgesia

(metamizol,

100

mg/kg;

injected

subcutaneously, Zentiva Pharma GmbH, Frankfurt, Germany). Analgesic treatment was continued for additional three days post-surgery in the drinking water. As post-operative care, rats were supplied with 5 ml 0.9 % saline (injected subcutaneously, Braun, Melsungen, Germany) and maintained under infrared lamps until recovery. Preparation of fetal ventral mesencephalic tissue and transplantation surgery Ventral mesencephalic tissue was harvested from embryonic day 12 old rat embryos (crown-rump length of 6 mm) and a single cell suspension prepared according to a modified version of the cell suspension technique (Bjorklund et al., 1983; Nikkhah et al., 1994). In our previous study (Rumpel et al., 2013), the differentiation period was four days and the survival of TH-ir cells was analyzed eight weeks after transplantation. Thereafter, we modified our protocol, i.e., we used co-layer instead of monolayer with only two days of cell differentiation together with a survival time of 12 weeks after transplantation. This procedure resulted in a higher TH-ir cell number of surviving grafted neurons, as reported in Rumpel et al. (2015). Therefore, in the present study, we used a similar protocol, i.e., cells differentiated in vitro for two days prior to transplantation and analysis of TH-ir cells after survival of 12 weeks after transplantation. In brief, the cells were plated on 6-well plates coated with polyornithine (0.1 mg/ml, Sigma-Aldrich) and laminin (6µg/ml, Sigma-Aldrich). After one day of attachment, cells were proliferated for three days and differentiated for two additional days. For implantation, cells were washed with phosphate-buffered saline (PBS, Biochrom, Berlin, Germany) and trypsin-EDTA (Gibco, Darmstadt, Germany) was added for 2-3 minutes at 37°C to allow detachment. The reaction was stopped with serum8

containing medium and cells were centrifuged at 1000 rpm for five minutes. The pellet was re-suspended in vehicle (transplantation medium; as described previously in Rumpel et al. (2013)) and the cells were centrifuged again (1000 rpm for five minutes). The new pellet was re-suspended in vehicle and the final concentration was adjusted. Eight weeks after lesion surgery, animals from the Transplant group received a total amount of 520,000 cells (130,000 cells/µl) distributed over two tracts with two 1-µl deposits each into the right STR under chloral hydrate anesthesia. The coordinates (calculated in mm from bregma and dura) were as follows: 1) AP +1.0, ML −2.3, DV −5.0/−4.0, TB 0.0; 2) AP +1.0, ML 3.3, DV −5.0/−4.0, TB 0.0. The injections were carried out using a glass capillary (outer diameter 60-80 µm) attached to a 2-µl Hamilton syringe with an injection rate of 1 µl/min. Drug-induced rotation Rotation measurements were performed in ‘rotometer’ bowls according to Ungerstedt and Arbuthnott (1970) using an automated system (Zentrale Forschungswerkstätten, Hannover Medical School, Hannover, Germany) six weeks after lesion surgery and ten weeks after transplantation surgery. L-apomorphine hydrochloride (0.05 mg/kg in 0.02 % L-ascorbate-saline, Sigma-Aldrich) was injected subcutaneously and rotational bias monitored over a period of 40 min. A few days later, D-amphetamine sulfate (2.5 mg/kg in saline, Sigma-Aldrich) was injected intraperitoneally and rotational bias monitored over a period of 90 min. The data are expressed as net full body turns per minute with positive values for ipsilateral rotation and negative values for contralateral rotation. Electrophysiological recordings A total of fifteen rats (Control, n = 5; 6-OHDA, n = 5; Transplant, n = 5) were used for the electrophysiological recordings in the EPN, GP, and motor cortex (MCtx) ipsilateral to the lesioned or grafted side of the brain at least 12 weeks after intrastriatal graft implantation.

The

rats

were

anesthetized

with

urethane

(1.4

g/kg,

injected

intraperitoneally; Sigma-Aldrich) and placed in a stereotaxic frame. The body temperature was maintained at 37 ± 0.5°C by a heating device (FHC, Bowdoinham, ME). Small craniotomies were made over the target coordinates for the EPN and GP 9

and the animal was placed in a Faraday cage to minimize electrical noise. A single microelectrode for extracellular recordings (quartz coated electrode with a platinumtungsten alloy core (95 % - 5 %), diameter 80 µm, and impedance 1 - 2 MΩ) was connected to the Mini Matrix 2 channel version drives headstage (Thomas Recording GmbH, Gießen, Germany). The microelectrode signal was passed through a headstage with unit gain and then split to separately extract the single unit (SU) and the local field potentials (LFPs) components. For SU recording, signals were band pass filtered between 500 and 5000 Hz and amplified from × 9,500 to 19,000 and sampled at 25 kHz. The stainless steel guide cannula, which touches the cortical surface, served as reference for the microelectrode, which is standard for recording of LFP from microelectrode recordings (Alam et al., 2012). Additionally, the ground wires were clamped on the neck and then plugged into the appropriate spot mini matrix head stage connector. The recordings in the EPN and GP were stereotaxically guided at the following coordinates in mm relative to bregma and ventral to the cortical surface: EPN) AP: −2.5 to −2.8, ML: −2.5 to −2.8, DV: −7.8 to −8.2; GP) AP: −0.8 to −1.2, ML: −2.8 to −3.2, DV −5.6 to −7.0.The MCtx-Electrocorticogram (ECoG) was recorded via a 1-mmdiameter jeweler’s screw, which was positioned on the dura mater above the frontal cortex ipsilateral to the lesioned or graft implanted hemisphere (AP: +2.7 mm, ML: −2.0 mm; which corresponds to the primary MCtx region) and an additional screw, serving as MCtx-ECoG reference, was placed over the cerebellum and band pass filtered (0.5 -100 Hz) with a sampling rate of 1 kHz (Alam et al., 2012). Data were acquired using the CED 1401 A/D interface (Cambridge Electronic Design, Cambridge, UK). After termination of the experiment, electrical lesions were made at the recording sites to allow histological verification of the localization (10 µA for 10 seconds; both negative and positive polarities) as previously described (Jin et al., 2016). An illustration of histological-based reconstruction from the rat brain atlas (Paxinos and Watson, 2006) is shown with electrical lesions in cresyl violet stained coronal sections of the EPN and GP (Fig. 2A-D). Analysis of electrophysiological data Rats anaesthetized with urethane demonstrate spontaneous and cyclical alternations of brain state that resemble sleep state alternations or the global active or awake state of 10

the brain. These fluctuations between different brain states are known as rapid eye movement (REM) and non-REM stages (Jin et al., 2016). Notably, during the REM state the brain metabolism is similar to normal waking values and is therefore regarded more similar to the activity found in the awake condition (Jin et al., 2016; Magill et al., 2000; Mallet et al., 2008). With that regard, only neuronal recording data representing the global active state of brain were used for analysis. Action potentials arising from a single neuron were sorted on the base of a 3:1 signal to noise ratio and discriminated by the template-matching function of the spike-sorting software (Spike2; Cambridge Electronic Design, Cambridge, UK). Only well isolated SUs were included in the analysis, which was determined by the homogeneity of spike waveforms, the separation of the projections of spike waveforms onto principal components during spike sorting, and clear refractory periods in inter-spike interval (ISI) histograms. All analyses were performed using custom-written Matlab (Mathworks, Natick, MA) functions. For analyses of spontaneous activity, one epoch of 100 s with simultaneously recorded spiking and LFP activities that was free of artefacts was used from every recorded neuron. Firing rate was calculated by taking the reciprocal value of the mean ISI. The coefficient of variation (CV) was taken as a measure of firing variation, since apart from the firing rate the firing variability and patterns are proposed to be important mechanisms in the pathophysiological changes of PD (Steigerwald et al., 2008). The CV of the spike interspike interval sequence was computed for each recording as a measure of the variability of the spike firing. It is a measure of spike train variability defined as the standard deviation

divided

by

the

mean

ISI.

Further,

three

distinct

patterns

(regular/irregular/bursts) were determined by using the discharge density histogram. This method is a comparison of the density histogram d(λ) to a reference density function px(λ). The estimated density histograms d(λ) were compared to reference probability density function px(λ) by means of a mathematical distance (Alam et al., 2015; Labarre et al., 2008). EPN, GP, and MCtx-ECoG field potentials analysis Another way to look at the pathophysiology of PD is the network oscillatory model. The oscillatory activity in different spectral bands, most importantly the abnormal beta activity 11

in PD and its synchronization within cortical and subcortical regions of the BG, are thought to be important correlates of normal and pathological neuronal processing (Avila et al., 2010; Brown, 2003). Representative epochs of 100 s without major artifacts were used for the frequency-domain signal processing for the EPN, GP, and MCtx-ECoGs field potentials. A finite impulse response 50 Hz notch filter and 100 Hz low-pass filter were used. Auto spectra of EPN and GP-LFPs or MCtx-ECoG were derived by discrete Fourier transformation with blocks of 1024 samples using a Welch periodogram in a custom Matlab (MathWorks, Inc.) script, which resulted in a frequency resolution of 0.9766 Hz. Hanning’s (referred to as Hann) window function was applied to overcome spectral leakage phenomenon. The mean (total) frequency (Hz) was derived from the spectrum 1-100 Hz. The relative power indices for each band were calculated from the absolute power in each frequency band and expressed as a percentage of the absolute power. Relative rather than absolute power was analyzed to allow comparison across groups, because absolute power is more likely to be dependent on proximity to the subcortical LFPs and cortical ECoG source than relative power and may vary with minor changes in recording technique. For comparison of powers at different frequency bands, the area under the computed power density spectrum in specified frequency ranges for theta (4 - 8 Hz) and beta (12 - 25 Hz) were calculated and averaged. Notably, alterations of these frequency bands have been reported to be most important in hypo- and hyperkinetic movement disorders.

12

EPN and GP spikes phase locked ratio to MCtx-ECoG Additionally, the interaction between EPN or GP spikes and MCtx-ECoG field potentials were assessed using spike triggered waveform averages (STWAs). The STWA is a marker of the temporal relationship between the spiking activity of individual neurons and LFP oscillations and is therefore regarded an important tool to explore the functional connectivity and the interaction between subcortical and cortical circuitries. Here, it is used to assess the strength of postsynaptic cortical activity related to the SU activity at EPN or GP. The ECoG field potentials were band pass filtered to calculate the corresponding power spectral densities. STWAs were calculated for 150 ms before and after the spike trigger over a 100 s epoch. Spike trains of each neuron were shuffled 20 times to create a null hypothesis for a non-phase locked spike train with the same firstorder statistics as the original spike train. The null hypothesis approach is used to detect whether the spike-phases are locked with ECoG more than would be expected by chance alone. The neuronal spikes were considered phase locked at a specific frequency range if the null hypothesis of uniformity of the phase distribution could be rejected at a corrected p < 0.01. The statistical significance of spike-LFP phase locking was assessed across different groups within different frequency bands, using the Rayleigh’s test (p < 0.001). The phase locked ratio was obtained by dividing the peak-totrough amplitude of the unshuffled spike trains STWA by the mean of the shuffled distribution. A comparison of the mean ratios was analyzed for the EPN SU firing neuron referenced to the MCtx-ECoG filtered in different frequency bands. Immunohistochemistry and morphometry At the end of the electrophysiological recordings, deeply anesthetized animals were transcardially perfused with 150 ml 0.9 % saline (Braun) followed by 250 ml of 4 % paraformaldehyde (Sigma-Aldrich) in PBS (Biochrom). Brains were removed, post-fixed for additional 24 h in the same fixative and afterwards transferred to 30 % sucrose (Roth, Karlsruhe, Germany) for cryoprotection. Sections through the STR (between +3.2 and −0.9 mm with reference to bregma) and substantia nigra (between −4.4 and −6.6 mm) were cut at 40 µm thickness in six series using a freezing stage microtome (Leica Biosystems, Nussloch, Germany). Every third section was processed for free-floating 13

anti-tyrosine hydroxylase (TH)-IHC as described previously (Rumpel et al., 2013). Sections through the GP and EPN (between −1.2 and −2.8 mm) were cut at 18 µm thickness in one series, mounted on glass slides and stained with cresyl violet (SigmaAldrich) to localize electrical lesions from electrophysiological recording sites. Stereological cell counts of TH-positive grafted neurons Estimation of the total TH-positive (TH+) cell number in the grafted STR was obtained by design-based stereology using StereoInvestigator software (MicroBrightField Europe E.K., Magdeburg, Germany) with the optical fractionator probe. Sections of two series were viewed under a 100x 1.25 oil objective (Olympus, Hamburg, Germany) on a BX50 microscope (Olympus) with a motorized XYZ axis computer-controlled stage and a CCD camera (MicroBrightField). A grid size of 10,000 µm2 (x 100 µm; y 100 µm), counting frame area of 3600 µm2 (x 60 µm; y 60 µm) and disector height of 20 µm with 2 µm guard zones on both sides were chosen. Analysis was performed through the entire AP extent of the transplants. Gundersen Coefficients of Error (m = 1) were all ≤ 0.05. Measurement of striatal TH-positive fiber density Microphotographs of sections were taken at five rostro-caudal levels (AP +2.0, + 1.5, +1.0, +0.5, 0.0 mm with reference to bregma) under bright field illumination using a 10x 0.30 objective (Olympus) on a BX51 microscope (Olympus) and StereoInvestigator software. The mean gray level was obtained using ImageJ software (National Institutes of Health, Bethesda, MD) by outlining the STR and nucleus accumbens without the graft core. The optical density was calculated by subtraction of the unspecific background value measured in the adjacent corpus callosum and results are given as a mean of the five levels in percentage of the contralateral intact side. Measurement of graft volume The area of the graft core was analyzed in µm2 in the entire AP extent of the transplants on sections of two series using ImageJ software. The graft volume was then calculated by multiplying the area with section thickness of 40 µm and three for the number of series. 14

Statistical analysis Rotation data were analyzed with two-way ANOVA and Bonferroni post hoc test using GraphPad Prism6 software (GraphPad Software, Inc., La Jolla, CA). Histology data were subjected to unpaired student’s t-test. For the SUs time stamps data, nonparametric statistical approaches were preferred due to the significant deviation from normality and a lack of homogeneous variances that existed mostly in extracellular SU spike data. The Shapiro-Wilk test was used to judge, whether noncircular datasets were normally distributed (p ≤ 0.05 to reject). The SU activity based parameters in the different groups were analyzed using Kruskal-Wallis ANOVA followed by Dunn's multiple comparison post hoc test. Differences were considered statistically significant when p < 0.05. The relative frequency or percentage of regular/ irregular/ bursts was calculated by dividing the number of positive events of regular/ irregular/ bursts in each group by the total number of neurons and multiplying by 100. Chi-square test was used to compare the differences in the firing patterns (density discharge histogram) between the groups. Results are presented as mean ± standard error of the mean (SEM) and significance levels were set at p < 0.05, p < 0.01, and p < 0.001, and marked in the figures by *, **, and *** when compared to the Control group or #, ##, and ### when compared to the 6OHDA group, respectively, unless stated otherwise.

15

RESULTS Graft morphology Histological

examination

of

brain

sections

from

animals

that

underwent

electrophysiological recordings 12 weeks after DA graft implantation revealed that 6OHDA MFB injection induced a complete lesion with nearly no remaining TH+ fibers in the STR (6-OHDA group: 2.7 % ± 0.8 compared to the contralateral healthy hemisphere) and overall cell loss in the substantia nigra (Fig. 3A). In the Transplant group, microscopical analysis at the level of the substantia nigra confirmed overall cell loss as seen in the 6-OHDA group. All grafted animals showed large medial and lateral grafts (total graft volume: 0.8 mm3 ± 0.1) in the ipsilateral STR containing high numbers of TH+ cells (total cell number: 6441 ± 629 TH+ cells, Fig. 3B). At higher magnification, dense re-innervation in the surrounding striatal area of the graft core was detected (striatal fiber density: 12.3 % ± 2.1), which was significantly higher compared to the lesioned animals without grafts (p < 0.01, Fig. 3C). Drug-induced rotation Rotational asymmetry after injection with either amphetamine or apomorphine was assessed at pre transplantation (pre-TX) and post transplantation (post-TX) times (see Fig. 1). The unilateral MFB lesion induced a mean ipsilateral bias of 11.0 to 11.8 turns in the amphetamine-induced rotation test (Fig. 3D) and −8.7 to −10.7 turns in the apomorphine-induced rotation test (Fig. 3E) six weeks after lesion. Animals were hereafter assigned into experimental groups. Ten weeks after transplantation surgery, animals from the 6-OHDA group showed a strong mean ipsilateral bias of 13.5 ± 1.8 turns in response to amphetamine, whereas animals from the Transplant group showed substantial and significant graft-induced improvement with a characteristic overcompensatory response (−10.3 ± 2.0 turns; p < 0.001 compared to the 6-OHDA group and pre-TX values, respectively; Fig. 3D) similar as previously described (Rumpel et al., 2013). In response to apomorphine, the 6-OHDA group showed a strong and consistent contralateral bias of −10.8 ± 1.5 turns ten weeks after graft implantation. In contrast to that, a significant amelioration to a contralateral bias of −5.5 ± 0.7 turns was observed in 16

the Transplant group (p < 0.05 compared to the 6-OHDA group and pre-TX values, respectively; Fig. 3E). Electrophysiology We analyzed a total of 224 SUs in the EPN and 191 SUs in the GP neurons from five rats in each group. In the EPN, the number of SUs recorded from neurons of the Control group was N = 64, from the 6-OHDA group N = 80, and from the Transplant group N = 80. The average SU (mean and SEM) recorded per individual rat in the Control group was 12.8 ± 1.06, in the 6-OHDA group 16.0 ± 2.23, and in the Transplant group 16.0 ± 2.3. In the GP, the number of SUs recorded from neurons of the Control group was N = 66, from the 6-OHDA group N = 61, and from the Transplant group N = 64. The average SU (mean and SEM) recorded per individual rat in the Control group was 13.2 ± 1.06, in the 6-OHDA group 12.02 ± 0.86, and in the Transplant group 12.8 ± 1.65. Firing rate, CV, and bursts In the EPN, Kruskal-Wallis ANOVA test revealed statistically significant differences in the firing rate between groups (chi-square = 16.81, df = 2; p < 0.001). Post hoc testing showed that the firing rate in the EPN was higher in rats from the 6-OHDA group (p < 0.01) compared to Control and Transplant groups. The mean and the 25th and 75th percentiles of neuronal firing rate of EPN neurons in the 6-OHDA group was 27.11 Hz (range 17.71 - 34.18), in the Control group 22.62 Hz (range 15.59 - 29.31), and in the Transplant group 20.01 Hz (range 14.05 - 24.49). The mean firing rate of the Transplant group did not differ significantly from the Control group (p = 0.17; Fig. 4A). In the GP, Kruskal-Wallis ANOVA test revealed no statistically significant differences in the firing rate between the three experimental groups (chi-square = 3.61, df = 2; p = 0.16). The mean and the 25th and 75th percentiles of neuronal firing rate of GP neurons in the 6OHDA group was 27.99 Hz (range 15.29 - 37.33), in the Control group 35.88 Hz (range 23.91 - 36.30), and in the Transplant group 29.17 Hz (range 18.62 - 39.38; Fig. 4B). In the EPN, Kruskal-Wallis ANOVA test revealed statistically significant differences in the CV between groups (chi-square = 16.81, df = 2; p < 0.001). Post hoc testing showed that the Control group was lower as compared to both, 6-OHDA and Transplant groups 17

(p < 0.01; Fig. 4C), without difference between the 6-OHDA and Transplant groups. In the GP, Kruskal-Wallis ANOVA test revealed statistically significant differences in the CV between groups (chi-square = 42.97, df = 2; p < 0.001). Post hoc testing showed that the CV of the 6-OHDA group was higher as compared to the Control and Transplant groups (p < 0.01), while no difference was detected between the Control and Transplant groups (Fig. 4D). Chi-square test showed that 6-OHDA lesioning significantly affected the neuronal firing pattern distribution in both the EPN (chi-square = 14.35, df = 4; p = 0.005) and GP (chi-square = 10.48, df = 4; p = 0.03). In the EPN, the analysis of classified discharge patterns (regular, irregular, and bursts) showed a low percentage of regular firing pattern in the 6-OHDA group as compared to the Control group (43 % vs. 70 %; p < 0.01) and also in the 6-OHDA group compared to the Transplant group (43 % vs. 55 %; p < 0.05). Notably, the percentage of bursty firing neurons in both, the Control group and the Transplant group, was lower as compared to the 6-OHDA group (5 % in Control and Transplant groups, respectively, vs. 14 % in the 6-OHDA group; p < 0.05). The percentage of irregular firing pattern was higher in both, the 6-OHDA and Transplant groups, as compared to the Control group (43 % and 40 %, respectively, vs. 25 %; p < 0.05), whereas there was no difference in the irregular firing pattern between the 6-OHDA and the Transplant groups (Fig. 4E). In the GP, the analysis of classified discharge patterns (regular, irregular, and bursts) revealed lower percentage of regular firing pattern in the 6-OHDA group as compared to the Control group (70 % vs. 90 %; p < 0.05) and also in the 6-OHDA group compared to the Transplant group (70 % vs. 89 %; p < 0.05). Notably, the percentage of irregular firing pattern in the 6-OHDA group was higher compared to the Control (27 % vs. 7 %; p < 0.05) and Transplant groups (27 % vs. 11 %; p < 0.05) respectively, whereas only 3 % of bursty firing pattern were observed in both, 6-OHDA and Control groups, respectively, and no bursty firing pattern in the Transplant group (Fig. 4F). Beta local field potentials in the EPN, GP, and MCtx The analysis of one-way ANOVA showed significantly higher beta oscillatory power in the EPN (F2, 222 = 6.27; p < 0.01), in the GP (F2, 176 = 3.85; p < 0.05), and in the MCtx (F2, 176

= 22.26; p < 0.001) of the 6-OHDA group. Post hoc testing showed a higher 18

percentage of beta frequency oscillatory power in the EPN of the 6-OHDA group compared to the Control group (26.01 % vs. 21.48 %; p < 0.01) as well as between the 6-OHDA and Transplant groups (26.01 % vs. 22.65 %; p < 0.05). No differences were observed between the Control and the Transplant groups (21.48 % vs. 22.65 %; p = 0.66; Fig. 5A). In the GP, higher percentage of beta frequency oscillatory power was only observed in the 6-OHDA group compared to the Control group (27.21 % vs. 23.50 %; p < 0.01). No differences were seen between the Control and the Transplant groups (23.50 % vs. 25.19 %; p = 0.37; Fig. 5B). In the MCtx, post hoc testing showed a higher percentage of beta frequency oscillatory power in the 6-OHDA group compared to the Control group (23.33 % vs. 17.04 %; p < 0.01) as well as compared to the Transplant group (23.33 % vs. 20.21 %; p < 0.01). A significant difference was found in beta frequency oscillatory power between the Control and the Transplant groups (17.04 % vs. 20.21 %; p < 0.01; Fig. 5C). Theta local field potentials in the EPN, GP, and MCtx In the EPN, analysis of one-way ANOVA showed significant differences of theta oscillatory power between groups (F2,

222

= 23.82; p < 0.001). Direct comparison

between groups with Tukey post hoc test showed that the percentage of theta frequency oscillatory power was lower in the 6-OHDA group compared to the Control group (44.58 % vs. 57.62 %; p < 0.01) and compared to Transplant group (44.58 % vs. 49.32 %; p < 0.01). Theta frequency oscillatory power was also higher in the Control group compared to Transplant group (57.62 % vs. 49.32 %; p < 0.05; Fig. 5D). In the GP, theta oscillatory power also showed differences between groups (F2,

176

= 3.23; p < 0.05). Post hoc

comparison revealed that the percentage of theta frequency oscillatory power was lower in the Transplant group compared to the Control group (52.15 % vs. 56.00 %; p < 0.05). No difference in the theta frequency oscillatory power was observed between 6-OHDA and Transplant groups (53.06 % vs. 52.15 %; p = 0.18; Fig. 5E). In the MCtx, one-way ANOVA showed that theta oscillatory activity also differed between groups (F2,

176

=

18.63; p < 0.001). Direct post hoc comparison between groups revealed that the percentage of theta frequency oscillatory power was lower in the 6-OHDA group compared to the Control group (51.08 % vs. 60.83 %; p < 0.01) and compared to the 19

Transplant group (51.08 % vs. 55.42 %; p < 0.05). Theta frequency oscillatory power was higher in the Control group compared to Transplant group (60.83 % vs. 55.42 %; p < 0.01; Fig. 5F). EPN or GP spikes and MCtx-ECoG phase locked ratio In the EPN, most spikes of neurons were significantly phase locked with theta (4 - 8 Hz) and beta (12 - 25 Hz), while in the GP most spikes of neurons were significantly phase locked with delta (1 - 4 Hz) and beta (12 - 25 Hz). We therefore restricted our analysis to these frequency bands (summarized in Table 1). The EPN spikes and MCtx phase locked ratio of theta band activity showed no differences between groups. The EPN spikes and MCtx phase locked ratio in the beta band frequency was enhanced only in 6OHDA rats. The analysis of one-way ANOVA showed a significant difference between groups (F2,

222

= 10.51, p < 0.001). The post hoc test showed higher EPN spikes and

MCtx phase locked ratio of beta band frequency in the 6-OHDA group compared to both, Control and Transplant groups (p < 0.01, respectively), while statistically no difference was observed between the Control and the Transplant groups (p = 0.73). No difference was observed across groups with regard to theta activity. In the GP, one-way ANOVA showed a significant difference in delta phase locked ratio between groups (F2,

176

=

6.95, p < 0.001). Post hoc testing showed higher GP spikes and MCtx phase locked ratio of delta band frequency in the 6-OHDA group compared to both, Control and Transplant groups (p < 0.01, respectively), while statistically no difference was observed between the Control and the Transplant groups (p = 0.95). The GP spikes and MCtx phase locked ratio in the beta band frequency was only reduced in rats of the Transplant group. Analysis with one-way ANOVA showed a significant difference between groups (F2,

176

= 4.87, p < 0.01). Post hoc testing showed lower GP spikes and MCtx phase

locked ratio of beta band frequency in the Transplant

group (p < 0.01), while the

enhanced activity in the 6-OHDA group did not reach the level of statistical significance (p = 0.08). No difference was found between Transplant and Control groups (p = 0.56).

20

DISCUSSION Intrastriatal DA cell grafts improved motor dysfunction ten weeks after graft implantation. This was accompanied by partially restored altered neuronal activity induced by 6-OHDA nigrostriatal lesions in both, the GP, a key region of the indirect pathway of the BG motor loop, and the EPN, the output region of the BG motor loop. The EPN firing pattern in control rats showed relatively higher irregular neuronal activity, whereas in the GP, most neurons had regular firing activity, which corroborates previous findings (Baron et al., 2011; Deister et al., 2013). After 6-OHDA-induced nigrostriatal lesions, neuronal firing rate as well as irregular and burst firing patterns were enhanced in the EPN of 6-OHDA lesioned rats, which is in accordance to earlier studies (Jin et al., 2016; Walters et al., 2007). Striatal DA cell graft implantation not only normalized EPN firing rate, but also burst and irregular patterns of neuronal activity, indicating a restoration of BG activity towards normal after striatal integration of DA grafts. While enhanced firing rate has been shown to be reduced after external DA administration in both, PD patients (Levy et al., 2001) and rats with 6-OHDA-induced lesions (Jin et al., 2016), treatment with DA receptor agonists usually does not affect or even enhance burst firing in rats with 6-OHDA-induced lesions (Jin et al., 2016; Lee et al., 2001; Tseng et al., 2000) and in PD patients (Levy et al., 2001). Interestingly, intrastriatal cell grafting also enhanced burst activity in the STN of 6-OHDA lesioned rats analyzed eight weeks post-grafting, i.e., it had an effect similar to external DA administration. The regulating effect of striatal grafts on BG activity can also be found in the GP. Here, loss of DA by 6-OHDA-induced lesion mainly enhanced measures for irregularity as previously shown by Mallet et al. (2008), while not affecting the firing rate. Implantation of striatal grafts counterbalanced altered firing pattern and variability. Clinical and preclinical studies about the effect of deep brain stimulation (DBS) have also shown that the temporal pattern of neuronal activity might be relevant (Hess et al., 2013; McConnell et al., 2016). With regard to the GP firing rate, the classical rate-coding model of PD predicts that the reduction of DA in the STR would lead to reduced inhibition of the indirect pathway with decreased neuronal firing rate in the GP. Furthermore, influx from the hyperdirect 21

pathway would also lead to hypoactivity in the GP in the parkinsonian state. However, as more recently pointed out (Deffains et al., 2016), the GP has also reciprocal connections to the STN, which would predict hyperactivity in the GP. In line with this, biochemical data presented by Levy et al. (1997) indicated that GPe neurons are probably not hypoactive in PD. With that regard, several groups reported that the basal firing rate of GP neurons remains unchanged upon loss of nigrostriatal DA (Chan et al., 2011; Gillies et al., 2002; Terman et al., 2002), while other groups found even enhanced firing rates together with enhanced burst activity (Benhamou et al., 2012; Miguelez et al., 2012). A previous study has demonstrated that striatal DA cell grafts exert a normalizing effect on the STN firing patterns without significant differences with respect to firing rate in a rat model of PD (Gilmour et al., 2011). Interestingly, in our previous study, subthalamic neuronal activity was only partially improved after implantation of DA grafts into the striatum of 6-OHDA lesioned rats, i.e., while STN neuronal firing rate and beta oscillatory activity were normalized, burst activity and measures of irregularity were not affected or even enhanced (Rumpel et al., 2013). The most likely reason for this difference between studies is that in the present study, electrophysiological recordings were done 12 weeks after graft implantation, probably leading to a more thorough integration of grafts into the STR. However, the DA grafts were transplanted in an ectopic location and thus could not be modulated by their afferents to exhibit tonic and phasic patterns of firing. This absence of physiological regulation of DA levels in the STR may explain why some measures were not restored to the control level by DA cell grafts. Altered oscillatory activity processed in different frequency bands and their synchronization

in

the

MCtx

and

BG

is

also

considered

relevant

for

the

pathophysiological processes underlying motor dysfunction in PD. Clinical studies in patients with PD have correlated enhanced beta frequency oscillations with akinesia, bradykinesia, and rigidity in the STN and GPi (Brown, 2003; Gatev et al., 2006; Oswal et al., 2013; Weinberger et al., 2012). Increased beta oscillations have also been shown in anesthetized (Alam et al., 2014; Mallet et al., 2008; Rumpel et al., 2013) and awake 6OHDA lesioned rats (Avila et al., 2010; Brazhnik et al., 2012). External DA replacement therapy has been shown to reduce beta band activity in patients with PD (Levy et al., 2001; Ray et al., 2008; Weinberger et al., 2006) and in rodent models of PD (Alam et al., 22

2014; Jin et al., 2016). Overall, in the present study, beta oscillatory power was also enhanced in the GP, EPN, and MCtx, both with respect to LFPs as to SU firing phase locked to MCtx LFPs. Intrastriatal DA cell grafts counterbalanced this abnormal exaggerated beta oscillatory activity to the level of the Control group indicating a normalizing effect on BG network behavior, although effects were less pronounced in the GP compared to EPN and MCtx. Moreover, in the EPN, GP, and MCtx, oscillatory theta band activity was reduced after 6-OHDA-induced nigrostriatal lesion. Striatal graft implantation alleviated theta band oscillatory activity to the level of naive control animals in the EPN and MCtx, while not affecting theta band activity in the GP. It has been shown that acute administration of DA receptor agonists generates high frequency cortical and hippocampal theta oscillatory power in rats, whereas DA antagonists reduce theta band activity (Kichigina, 2004; Miura et al., 1987; Yamamoto, 1997). With this regard, a prominent decrease in MCtx theta band frequency has also been found in mice lacking the DA transporter (Costa et al., 2006). However, while clear evidence for a movement-prohibitive role of elevated beta band activity comes from recordings in human BG of patients undergoing DBS for treatment of PD (Brown, 2003; Weinberger et al., 2012), the role of theta oscillatory activity in PD patients, who did not receive levodopa treatment, is not clear so far. Interestingly, enhanced theta band activity has been associated with DA agonist-induced dyskinesias in PD patients and animal models (Alonso-Frech et al., 2006). With that regard, electrophysiological data from PD patients have suggested that higher levels of theta/alpha bands in the STN facilitate normal behavior (Anzak et al., 2012; Brittain and Brown, 2014; Tan et al., 2013). The GP spikes and MCtx field potentials relation, as indicated by the phase locked ratio, showed higher coincidence of neuronal activity in delta and beta band frequencies, while theta band activity was not affected. This corroborates previous observations reporting pronounced low-frequency oscillatory activity in pallidal neurons (Magill et al., 2000). Interestingly, experimental studies in rodents have suggested that delta-mediated phase-amplitude coupling plays a key role in the organization of local and distant activities in the cortico-BG network by fine-tuning the timing of synchronization events across different structures (Lopez-Azcarate et al., 2013; Mena-Segovia et al., 2008).

23

One limitation of our work is that electrophysiological recordings were performed in anaesthetized rats, which may have affected neuronal activity. Nevertheless, the urethane-anesthetized animal is regarded a useful model for assessing the impact of different brain states on functional connectivity within and between the BG and cortex (Jin et al., 2016; Magill et al., 2000; Mallet et al., 2008). Additionally, the effect of anesthesia likely influenced the neuronal activity in a similar manner across different groups. Despite the limitation of anesthesia (urethane), extracellular SU recordings under urethane anesthesia have shown that the discharge rate of STN and EPN activities in 6-OHDA lesioned rats differ significantly from that of control rats (Alam et al., 2012; Jin et al., 2016). Conclusion In summary, recording of neuronal activity in key regions of the direct and indirect pathway of the BG motor loop, the GP and EPN, revealed a partial restoration of altered neuronal activity towards normal by striatal graft implantation in 6-OHDA lesioned rats suggesting an integration of grafts into the STR. Nevertheless, it remains open, whether striatal grafts will only affect neuronal activity in the BG motor loop, or whether nonmotor symptoms originating from dopaminergic degeneration outside the STR or in nondopaminergic systems, which are increasingly recognized as disabling for PD patients after treatment of dopamine-related motor symptoms, would also be affected by intrastriatal cell implantation. FUNDING This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ACKNOWLEGDEMENTS The authors wish to thank Silke Fischer, Natascha Heidrich, Kerstin Kuhlemann, and Maike Wesemann (Institute of Neuroanatomy and Cell Biology, MHH) for excellent technical assistance.

24

REFERENCES Alam, M., Capelle, H.H., Schwabe, K., Krauss, J.K., 2014. Effect of deep brain stimulation on levodopa-induced dyskinesias and striatal oscillatory local field potentials in a rat model of Parkinson's disease. Brain Stimul 7, 13-20. Alam, M., Heissler, H.E., Schwabe, K., Krauss, J.K., 2012. Deep brain stimulation of the pedunculopontine tegmental nucleus modulates neuronal hyperactivity and enhanced beta oscillatory activity of the subthalamic nucleus in the rat 6-hydroxydopamine model. Exp Neurol 233, 233-242. Alam, M., Schwabe, K., Lutjens, G., Capelle, H.H., Manu, M., von Wrangel, C., MullerVahl, K., Schrader, C., Scheinichen, D., Blahak, C., Heissler, H.E., Krauss, J.K., 2015. Comparative characterization of single cell activity in the globus pallidus internus of patients with dystonia or Tourette syndrome. J Neural Transm (Vienna) 122, 687-699. Alonso-Frech, F., Zamarbide, I., Alegre, M., Rodriguez-Oroz, M.C., Guridi, J., Manrique, M., Valencia, M., Artieda, J., Obeso, J.A., 2006. Slow oscillatory activity and levodopainduced dyskinesias in Parkinson's disease. Brain 129, 1748-1757. Anzak, A., Tan, H., Pogosyan, A., Foltynie, T., Limousin, P., Zrinzo, L., Hariz, M., Ashkan, K., Bogdanovic, M., Green, A.L., Aziz, T., Brown, P., 2012. Subthalamic nucleus activity optimizes maximal effort motor responses in Parkinson's disease. Brain 135, 2766-2778. Avila, I., Parr-Brownlie, L.C., Brazhnik, E., Castaneda, E., Bergstrom, D.A., Walters, J.R., 2010. Beta frequency synchronization in basal ganglia output during rest and walk in a hemiparkinsonian rat. Exp Neurol 221, 307-319. Barker, R.A., Drouin-Ouellet, J., Parmar, M., 2015. Cell-based therapies for Parkinson disease-past insights and future potential. Nature reviews. Neurology 11, 492-503.

25

Barker, R.A., Parmar, M., Kirkeby, A., Bjorklund, A., Thompson, L., Brundin, P., 2016. Are Stem Cell-Based Therapies for Parkinson's Disease Ready for the Clinic in 2016? Journal of Parkinson's disease 6, 57-63. Baron, M.S., Chaniary, K.D., Rice, A.C., Shapiro, S.M., 2011. Multi-neuronal recordings in the Basal Ganglia in normal and dystonic rats. Front Syst Neurosci 5, 67. Benhamou, L., Bronfeld, M., Bar-Gad, I., Cohen, D., 2012. Globus Pallidus external segment neuron classification in freely moving rats: a comparison to primates. PLoS One 7, e45421. Bevan, M.D., Booth, P.A., Eaton, S.A., Bolam, J.P., 1998. Selective innervation of neostriatal interneurons by a subclass of neuron in the globus pallidus of the rat. J Neurosci 18, 9438-9452. Bjorklund, A., Stenevi, U., Schmidt, R.H., Dunnett, S.B., Gage, F.H., 1983. Intracerebral grafting of neuronal cell suspensions. I. Introduction and general methods of preparation. Acta Physiol Scand Suppl 522, 1-7. Brazhnik, E., Cruz, A.V., Avila, I., Wahba, M.I., Novikov, N., Ilieva, N.M., McCoy, A.J., Gerber, C., Walters, J.R., 2012. State-dependent spike and local field synchronization between motor cortex and substantia nigra in hemiparkinsonian rats. J Neurosci 32, 7869-7880. Brittain, J.S., Brown, P., 2014. Oscillations and the basal ganglia: motor control and beyond. Neuroimage 85 Pt 2, 637-647. Brown, P., 2003. Oscillatory nature of human basal ganglia activity: relationship to the pathophysiology of Parkinson's disease. Mov Disord 18, 357-363. Cenci, M.A., Lindgren, H.S., 2007. Advances in understanding L-DOPA-induced dyskinesia. Curr Opin Neurobiol 17, 665-671. 26

Chan, C.S., Glajch, K.E., Gertler, T.S., Guzman, J.N., Mercer, J.N., Lewis, A.S., Goldberg, A.B., Tkatch, T., Shigemoto, R., Fleming, S.M., Chetkovich, D.M., Osten, P., Kita, H., Surmeier, D.J., 2011. HCN channelopathy in external globus pallidus neurons in models of Parkinson's disease. Nat Neurosci 14, 85-92. Deffains, M., Iskhakova, L., Bergman, H., 2016. Stop and Think about Basal Ganglia Functional Organization: The Pallido-Striatal "Stop" Route. Neuron 89, 237-239. Deister, C.A., Dodla, R., Barraza, D., Kita, H., Wilson, C.J., 2013. Firing rate and pattern heterogeneity in the globus pallidus arise from a single neuronal population. J Neurophysiol 109, 497-506. DeLong, M.R., Wichmann, T., 2007. Circuits and circuit disorders of the basal ganglia. Arch Neurol 64, 20-24. Galvan, A., Devergnas, A., Wichmann, T., 2015. Alterations in neuronal activity in basal ganglia-thalamocortical circuits in the parkinsonian state. Front Neuroanat 9, 5. Gatev, P., Darbin, O., Wichmann, T., 2006. Oscillations in the basal ganglia under normal conditions and in movement disorders. Mov Disord 21, 1566-1577. Gillies, A., Willshaw, D., Li, Z., 2002. Subthalamic-pallidal interactions are critical in determining normal and abnormal functioning of the basal ganglia. Proc Biol Sci 269, 545-551. Gilmour, T.P., Piallat, B., Lieu, C.A., Venkiteswaran, K., Ramachandra, R., Rao, A.N., Petticoffer, A.C., Berk, M.A., Subramanian, T., 2011. The effect of striatal dopaminergic grafts on the neuronal activity in the substantia nigra pars reticulata and subthalamic nucleus in hemiparkinsonian rats. Brain 134, 3276-3289. Gurney, K., Prescott, T.J., Redgrave, P., 2001. A computational model of action selection in the basal ganglia. I. A new functional anatomy. Biol Cybern 84, 401-410. 27

Hess, C.W., Vaillancourt, D.E., Okun, M.S., 2013. The temporal pattern of stimulation may be important to the mechanism of deep brain stimulation. Exp Neurol 247, 296-302. Jin, X., Schwabe, K., Krauss, J.K., Alam, M., 2016. Coherence of neuronal firing of the entopeduncular nucleus with motor cortex oscillatory activity in the 6-OHDA rat model of Parkinson's disease with levodopa-induced dyskinesias. Exp Brain Res 234, 1105-1118. Kichigina, V.F., 2004. [Dopaminergic regulation of theta activity of septohippocampal neuron in the awake rabbit]. Zh Vyssh Nerv Deiat Im I P Pavlova 54, 210-215. Kita, H., Kitai, S.T., 1994. The morphology of globus pallidus projection neurons in the rat: an intracellular staining study. Brain Res 636, 308-319. Labarre, D., Meissner, W., Boraud, T., 2008. Measure of the regularity of events in stochastic point processes, application to neuron activity analysis, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 489-492. Lee, J.I., Shin, H.J., Nam, D.H., Kim, J.S., Hong, S.C., Shin, H.J., Park, K., Eoh, W., Kim, J.H., Lee, W.Y., 2001. Increased burst firing in substantia nigra pars reticulata neurons and enhanced response to selective D2 agonist in hemiparkinsonian rats after repeated administration of apomorphine. J Korean Med Sci 16, 636-642. Levy, R., Dostrovsky, J.O., Lang, A.E., Sime, E., Hutchison, W.D., Lozano, A.M., 2001. Effects of apomorphine on subthalamic nucleus and globus pallidus internus neurons in patients with Parkinson's disease. J Neurophysiol 86, 249-260. Levy, R., Hazrati, L.N., Herrero, M.T., Vila, M., Hassani, O.K., Mouroux, M., Ruberg, M., Asensi, H., Agid, Y., Feger, J., Obeso, J.A., Parent, A., Hirsch, E.C., 1997. Reevaluation of the functional anatomy of the basal ganglia in normal and Parkinsonian states. Neuroscience 76, 335-343.

28

Lopez-Azcarate, J., Nicolas, M.J., Cordon, I., Alegre, M., Valencia, M., Artieda, J., 2013. Delta-mediated cross-frequency coupling organizes oscillatory activity across the rat cortico-basal ganglia network. Front Neural Circuits 7, 155. Magill, P.J., Bolam, J.P., Bevan, M.D., 2000. Relationship of activity in the subthalamic nucleus-globus pallidus network to cortical electroencephalogram. J Neurosci 20, 820833. Mallet, N., Pogosyan, A., Sharott, A., Csicsvari, J., Bolam, J.P., Brown, P., Magill, P.J., 2008. Disrupted dopamine transmission and the emergence of exaggerated beta oscillations in subthalamic nucleus and cerebral cortex. J Neurosci 28, 4795-4806. McConnell, G.C., So, R.Q., Grill, W.M., 2016. Failure to suppress low-frequency neuronal oscillatory activity underlies the reduced effectiveness of random patterns of deep brain stimulation. J Neurophysiol 115, 2791-2802. Mena-Segovia, J., Sims, H.M., Magill, P.J., Bolam, J.P., 2008. Cholinergic brainstem neurons modulate cortical gamma activity during slow oscillations. J Physiol 586, 29472960. Miguelez, C., Morin, S., Martinez, A., Goillandeau, M., Bezard, E., Bioulac, B., Baufreton, J., 2012. Altered pallido-pallidal synaptic transmission leads to aberrant firing of globus pallidus neurons in a rat model of Parkinson's disease. J Physiol 590, 58615875. Miura, Y., Ito, T., Kadokawa, T., 1987. Effects of intraseptally injected dopamine and noradrenaline on hippocampal synchronized theta wave activity in rats. Jpn J Pharmacol 44, 471-479. Nikkhah, G., Olsson, M., Eberhard, J., Bentlage, C., Cunningham, M.G., Bjorklund, A., 1994. A microtransplantation approach for cell suspension grafting in the rat Parkinson model: a detailed account of the methodology. Neuroscience 63, 57-72. 29

Oswal, A., Brown, P., Litvak, V., 2013. Synchronized neural oscillations and the pathophysiology of Parkinson's disease. Current opinion in neurology 26, 662-670. Paxinos, G., Watson, C., 2006. The Rat Brain in Stereotaxic Coordinates, 6th ed. Academic Press, London, Amsterdam. Ray, N.J., Jenkinson, N., Wang, S., Holland, P., Brittain, J.S., Joint, C., Stein, J.F., Aziz, T., 2008. Local field potential beta activity in the subthalamic nucleus of patients with Parkinson's disease is associated with improvements in bradykinesia after dopamine and deep brain stimulation. Exp Neurol 213, 108-113. Richardson, R.M., Freed, C.R., Shimamoto, S.A., Starr, P.A., 2011. Pallidal neuronal discharge in Parkinson's disease following intraputamenal fetal mesencephalic allograft. J Neurol Neurosurg Psychiatry 82, 266-271. Rumpel, R., Alam, M., Klein, A., Ozer, M., Wesemann, M., Jin, X., Krauss, J.K., Schwabe, K., Ratzka, A., Grothe, C., 2013. Neuronal firing activity and gene expression changes in the subthalamic nucleus after transplantation of dopamine neurons in hemiparkinsonian rats. Neurobiol Dis 59, 230-243. Rumpel, R., Hohmann, M., Klein, A., Wesemann, M., Baumgartner, W., Ratzka, A., Grothe, C., 2015. Transplantation of fetal ventral mesencephalic progenitor cells overexpressing high molecular weight fibroblast growth factor 2 isoforms in 6hydroxydopamine lesioned rats. Neuroscience 286, 293-307. Sato, F., Lavallee, P., Levesque, M., Parent, A., 2000. Single-axon tracing study of neurons of the external segment of the globus pallidus in primate. J Comp Neurol 417, 17-31. Steigerwald, F., Potter, M., Herzog, J., Pinsker, M., Kopper, F., Mehdorn, H., Deuschl, G., Volkmann, J., 2008. Neuronal activity of the human subthalamic nucleus in the parkinsonian and nonparkinsonian state. J Neurophysiol 100, 2515-2524. 30

Tan, H., Pogosyan, A., Anzak, A., Foltynie, T., Limousin, P., Zrinzo, L., Ashkan, K., Bogdanovic, M., Green, A.L., Aziz, T., Brown, P., 2013. Frequency specific activity in subthalamic nucleus correlates with hand bradykinesia in Parkinson's disease. Exp Neurol 240, 122-129. Terman, D., Rubin, J.E., Yew, A.C., Wilson, C.J., 2002. Activity patterns in a model for the subthalamopallidal network of the basal ganglia. J Neurosci 22, 2963-2976. Tseng, K.Y., Riquelme, L.A., Belforte, J.E., Pazo, J.H., Murer, M.G., 2000. Substantia nigra pars reticulata units in 6-hydroxydopamine-lesioned rats: responses to striatal D2 dopamine receptor stimulation and subthalamic lesions. Eur J Neurosci 12, 247-256. Ungerstedt, U., Arbuthnott, G.W., 1970. Quantitative recording of rotational behavior in rats after 6-hydroxy-dopamine lesions of the nigrostriatal dopamine system. Brain Res 24, 485-493. Walters, J.R., Hu, D., Itoga, C.A., Parr-Brownlie, L.C., Bergstrom, D.A., 2007. Phase relationships support a role for coordinated activity in the indirect pathway in organizing slow oscillations in basal ganglia output after loss of dopamine. Neuroscience 144, 762776. Weinberger, M., Hutchison, W.D., Alavi, M., Hodaie, M., Lozano, A.M., Moro, E., Dostrovsky, J.O., 2012. Oscillatory activity in the globus pallidus internus: comparison between Parkinson's disease and dystonia. Clin Neurophysiol 123, 358-368. Weinberger, M., Mahant, N., Hutchison, W.D., Lozano, A.M., Moro, E., Hodaie, M., Lang, A.E., Dostrovsky, J.O., 2006. Beta oscillatory activity in the subthalamic nucleus and its relation to dopaminergic response in Parkinson's disease. J Neurophysiol 96, 3248-3256.

31

Yamamoto, J., 1997. Cortical and hippocampal EEG power spectra in animal models of schizophrenia

produced

with

methamphetamine,

Psychopharmacology (Berl) 131, 379-387.

32

cocaine,

and

phencyclidine.

FIGURE/TABLE LEGENDS Figure 1: Experimental design. Rats from the 6-OHDA and Transplant groups were lesioned using 6-hydroxydopamine (6-OHDA) eight weeks before intrastriatal dopamine graft implantation. Animals from the Control group served as healthy controls. Lesion severity and graft functionality were evaluated by means of amphetamine- and apomorphine-induced rotation two weeks before (pre-TX) and ten weeks after the transplantation (post-TX) surgery, respectively. Finally, rats were subjected to electrophysiological recordings (E-Phys.) of neuronal activity in the entopeduncular nucleus (EPN) and globus pallidus (GP) and immunohistochemical analyses (IHC) in the striatum (STR). Figure 2: Recording sites with histological verification. An illustration from the rat brain atlas for the localization of electrophysiological recording sites in the EPN region (A) at −2.5 mm and in the GP region (C) at −0.8 mm posterior to bregma. Histological cresyl violet stained coronal sections showing the recording sites (black arrows, respectively) after applying electro-coagulation at the microelectrode tip in the EPN (B) and GP (D). CI = capsula interna, EPN = entopeduncular nucleus, GP = globus pallidus, STR = striatum. Figure 3: Histological and drug-induced rotation analysis. Histological TH-stained coronal sections (A, B) of the 6-OHDA group (A) at +0.5 mm anterior and −5.2 posterior to bregma showing no remaining TH+ fibers in the right STR and no TH+ cells in the right SN. In the Transplant group (B), implanted grafts are surrounded by dense TH+ fibers. Bar graphs (C) show striatal fiber density in percentage of the contralateral side in the 6-OHDA group as gray bar and Transplant group as black bar. The ** indicate significant differences to the 6-OHDA group (unpaired student’s t-test). Amphetamine-induced rotation (D) and Apomorphine-induced rotation (E) tests were performed six weeks before transplantation (pre-TX) and ten weeks after transplantation (post-TX). *, *** indicate significant differences to pre-TX

33

values, #, ### compared to the 6-OHDA group (two-way ANOVA and Bonferroni post hoc). SN = substantia nigra, STR = striatum. Figure 4: Single unit activity in the EPN and GP. Box and whisker plots show Firing rates (A, B) and Coefficient of variation of inter-spike intervals (C, D) of EPN and GP neurons in rats of the Control group (open white box plots), 6-OHDA group (gray box plots), and Transplant group (black box plots). The squares in the boxes are the mean values, a horizontal line within the boxes indicates the median and each box represents the values within the 25th and 75th percentile. The ‘error bars’ represent the 5th and 95th percentiles. Kruskal-Wallis ANOVA followed by Dunn's multiple comparison post hoc test was used for SU activity based parameters in the different groups and ** indicate significant differences to the Control, ## to the 6OHDA group. (E, F) The stacked bar graphs summarize percentage of regular, irregular, and bursty firing patterns in the Control group, the 6-OHDA group, and the Transplant group. The chi-square test was applied for significant changes of p values for (regular, irregular, and bursts) observations between different groups. *, ** indicate differences to the Control, #, ## indicate differences to the 6-OHDA group. Figure 5: Oscillatory activity in the EPN, GP, and MCtx. The bar graphs show the percentage of relative power of beta (12-30 Hz; A-C) and theta (4-8 Hz; D-F) oscillatory activity in the EPN (A, D), GP (B, E), and the MCtx (C, F) of the Control group, 6-OHDA group, and Transplant group. *, ** indicate differences to the Control, #, ## indicate differences to the 6-OHDA group (one-way ANOVA and Tukey post hoc test).

34

Table 1: Phase locked ratio of EPN/GP spikes with MCtx.

Control

6-OHDA

Transplant

1.56 ± 0.15 0.95 ± 0.04

1.57 ± 0.14 1.12 ± 0.04 **

1.67 ± 0.11 0.91 ± 0.03 ##

1.73 ± 0.17 0.91 ± 0.03

2.72 ± 0.32 ** 1.02 ± 0.05

1.65 ± 0.16 ## 0.85 ± 0.03 ##

EPN Theta (4-8 Hz) Beta (12-25 Hz)

GP Delta (1-4 Hz) Beta (12-25 Hz)

Results mean ± SEM. The ** indicate significant differences to the Control group, the ## compared to the 6-OHDA group.

35

36

37

38

39

40

Research highlights:

• • • •

long-term integration of dopamine transplants into the STR after 12 weeks partial restoration of altered firing rates and patterns of EPN and GP neurons towards normal reduction of altered beta oscillatory activity in the EPN and GP by dopamine cell grafts alleviation of altered theta oscillatory activity in the EPN by dopamine cell grafts

41

42