- Email: [email protected]
Altered neuronal activity in the primary motor cortex and globus pallidus after dopamine depletion in rats
Journal of the Neurological Sciences 348 (2015) 231–240
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Altered neuronal activity in the primary motor cortex and globus pallidus after dopamine depletion in rats☆ Min Wang a,⁎, Min Li a,1, Xiwen Geng a,1, Zhimin Song b, H. Elliott Albers b, Maoquan Yang a, Xiao Zhang a, Jinlu Xie a, Qingyang Qu a, Tingting He a a b
Key Laboratory of Animal Resistance of Shandong Province, College of Life Science, Shandong Normal University, Jinan 250014, People's Republic of China Center for Behavioral Neuroscience, Neuroscience Institute, Georgia State University, Atlanta, GA 30302, United States
a r t i c l e
i n f o
Article history: Received 11 September 2014 Received in revised form 9 November 2014 Accepted 10 December 2014 Available online 18 December 2014 Keywords: Parkinson's disease Local field potential Motor cortex Globus pallidus Basal ganglia Microelectrode
a b s t r a c t The involvement of dopamine (DA) neuron loss in the etiology of Parkinson's disease has been well documented. The neural mechanisms underlying the effects of DA loss and the resultant motor dysfunction remain unknown. To gain insights into how loss of DA disrupts the electrical processes in the cortico-subcortical network, the present study explores the effects of DA neuron depletion on electrical activity in the primary motor cortex (M1), on the external and the internal segment of the globus pallidus (GPe and GPi respectively), and on their temporal relationships. Comparison of local field potentials (LFPs) in these brain regions from unilateral hemispheric DA neuron depleted rats and neurologically intact rats revealed that the spectrum power of LFPs in 12–70 Hz (for M1, and GPe) and in 25–40 Hz (for GPi) was significantly greater in the DA depleted rats than that in the control group. These changes were associated with a shortening of latency in LFP activities between M1 and GPe, from several hundred milliseconds in the intact animals to close to zero in the DA depleted animals. LFP oscillations in M1 were significantly more synchronized with those in GPe in the DA depleted rats compared with those in the control rats. By contrast, the synchronization of oscillation in LFP activities between M1 and GPi did not differ between the DA depleted and intact rats. Not surprisingly, rats that had DA neuron depletion spent more time along the ladder compared with the control rats. These data suggest that enhanced oscillatory activity and increased synchronization of LFPs may contribute to movement impairment in the rat model of Parkinson's disease. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The basal ganglia form a complex network that processes cortical information important for movement and cognition [1–3]. Alterations of neuronal activity in the basal ganglia and cortices have been reported in patients with Parkinson's disease (PD) as well as in animal models of PD [4–11]. These changes in neuronal activity are believed to mediate many of the dysfunctional effects seen in PD because they impact the circuits connecting the basal ganglia and the cortex [10,12–14]. However, the specific pathological changes that occur with DA loss in oscillation activity and functional connectivity from the cortex to the basal ganglia remain poorly understood.
☆ Supporting grant: This study was supported by the Natural Science Foundation of Shandong Province (No. ZR2010CM055) and the Science and Technological Project of Shandong Province (Nos. 2011GGB01004 and 2010GGX10133). ⁎ Corresponding author at: Department of Anatomy and Physiology, College of Life Science, Shandong Normal University, 250014, People's Republic of China. Tel.: + 86 15615614667. E-mail address: [email protected] (M. Wang). 1 Contributed equally.
http://dx.doi.org/10.1016/j.jns.2014.12.014 0022-510X/© 2014 Elsevier B.V. All rights reserved.
The transmission of rhythmic cortical activity from the cortex to the basal ganglia has been studied in anesthetized rats [15,16]. An increase in discharge oscillations at 1 Hz from the basal ganglia is observed in anesthetized rats with unilateral DA depletion. This activity is coherent with the 1 Hz oscillatory firing patterns dominant in the cortex [17,18]. Transmission from the cortex to the basal ganglia has also been studied in rats using electrical stimulation of the motor cortex. After DA depletion, the cortical stimulation induces a long disinhibition in GPe, which is transmitted to GPi and generates an abnormally strong long inhibition in GPi, which then generates a strong and long excitation in the thalamic projection sites that are transmitted to the motor cortex with incorrect information [19,20]. Simultaneous behavioral recordings and chronic microelectrode neural recordings in awake and behaving rodents hold great promise to study the neural bases of behavior and the information transmission between the cortex and the basal ganglia. Some studies suggest that increases in the synchronization between the subthalamic nucleus or substantia nigra to cortex, or the external segment of the globus pallidus facilitate the emergence of special range activity in the cortex after DA loss [10,11,21–23]. These observations led to our general hypothesis that DA neuron loss alters the magnitude of the electrical activities in
232
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
the basal ganglia and enhances its tendency to become entrained by oscillatory activity expressed in the cortex. The aim of the present study was to determine: a) how DA loss affects LFP activities in the primary motor cortex (M1), the globus pallidus in rodents (an area analogous to the external segment of the globus pallidus in primates, GPe) and the entopeduncular nucleus in rodents (an area analogous to the internal segment of the globus pallidus in primates, GPi) [24,25]. In this paper, we will use the primate terminology for simplification and consistency reasons, and b) how DA loss affects the coherent oscillatory activity at the level of LFPs recorded in M1 with GPe or GPi. LFPs were pair-recorded from chronically implanted electrodes in the M1 and GPe or GPi of rats with unilateral DA neuron lesions, while they were engaged in a locomotor task. Control unilateral recordings in M1 and GPe or GPi were also obtained in neurologically intact rats.
(5.2 mm posterior and 2.1 mm lateral of the bregma, and 8.0 mm ventral to the dura) and 2 μl into the ventral tegmental area (6.8 mm posterior and 0.6 mm lateral of the bregma, and 8.6 mm ventral to the dura). The injection was made at a rate of 0.5 μl/min using a 5-μl microsyringe. After each injection, the micropipette was left in place for an additional 5 min and then slowly withdrawn. The control rats received only the vehicle (0.02% ascorbic acid in physiological saline) at the same coordinates. Rats received post-operative care until wakefulness and were returned to their home cages. Four weeks after 6-OHDA injection, a rotating behavioral response to apomorphine (0.05 mg/kg, s.c.) test was performed in order to assess the severity of the nigral lesion [28]. Animals that performed at least 80 rotations opposite to the lesioned site within 20 min following the apomorphine treatment were considered animals with successful surgery and only those were used for electrophysiological recordings.
2. Materials and methods
2.4. Recording electrode placement
2.1. Animal
A headstage adapted from previous studies [29,30] was designed for the recording electrode assembly. Briefly, the headstage composed of a 3-pin box connector socket stripe (2.54 mm, BCSS-1-SV, Indiana, USA) with double nickel–chromium teflon-insulated microwires (California Fine Wire, Grover Beach, CA, USA, 100 μm in diameter). One end of the microwires was stripped by gently scraping off about 2 mm of insulation and tinned with a soldering iron to attach to a pin of the connectors. A copper wire (200 μm in diameter) served as a ground wire was also soldered onto a pin of the connector. After the microwires and ground wire were tightly soldered, Epoxy glue was applied to the base and surrounding of the connector to build onto the headstage a movable driver with recording electrodes. The tip impedance of the electrode was 0.5–0.8 MΩ at 1 kHz. The headstage was implanted into both intact rats and rats with unilateral DA neuron lesions. The target regions of the headstage were M1 (1.0 mm anterior and 2.0 mm lateral of the bregma, and 0.4 mm ventral to the dura) and GPe (1.0 mm posterior and 3.0 mm lateral of the bregma, and 6.6 mm ventral to the dura) or GPi (2.3 mm posterior and 2.9 mm lateral of the bregma, and 7.7 mm ventral to the dura). Skull screws were used for ground wire connection and fixing the headstage to the skull. Dental cement was then applied to glue the headstage to the skull. The rats were given ketoprofen (Ketofen 2 mg/kg, s.c., Sigma, USA) following surgery and 24 h later again for pain relief. They were allowed at least 7 days to recover before the first recording session.
All experimental procedures were conducted on male Wistar rats (280–320 g, Animal Center of Shandong University, China). Rats were kept under standard housing conditions at constant temperature (22 ± 1 °C), humidity (relative, 30%), and light/dark 14/10 cycles. Water was available ad libitum. Food intake was limited to 10–20 g/day to maintain constant animal weight. Animal care and surgery were conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals (NIH Guidelines). Every effort was made to minimize the number of animals used and the pains they suffered. 2.2. Behavioral training task A ladder composed of two clear Plexiglas side walls (100 × 20 cm) with timber rungs (3 mm in diameter) inserted at random distances ranging from 1 to 3 cm was used in the testing. The ladder was elevated 30 cm above the ground with a square wooden case (home cage) at one end. Two weeks prior to surgery, rats were placed at one end of the ladder and trained to walk spontaneously to the home cage. The rats received three training trials each day for 3–7 days, after which they reliably walked to the home cage after being placed at the other end. During the testing, the time to cross the entire length of the ladder was recorded and the number of foot placement errors was determined by a rating system used in previous studies [26,27]. Briefly, 3 points were given when the animal completed or missed a rung and the body fell onto the ladder; 2 points were given when the animal initially stepped on the rungs but slipped off; and 1 point was given when the animal step a paw on a rung but did not place its body weight on that paw. The animals' performance was videotaped by a camera (Logitech AF, Taiwan) positioned at a slight ventral angle which enabled both sides of the body and paw positions to be simultaneously recorded. 2.3. Unilateral lesion of the nigrostriatal pathway Unilateral 6-hydroxydopamine (6-OHDA) lesion of the dopaminergic nigrostriatal pathway was performed on rats after the behavioral training. Rats were anesthetized with chloral hydrate (400 mg/kg, i.p.) and placed in a stereotaxic frame, with body temperature maintained at 37 ± 0.5 °C using a heating pad. The skull was exposed and two holes were drilled on the skull according to appropriate coordinates. Rats were injected with desmethylimipramine (15 mg/kg, i.p.) 30 min prior to the intracerebral infusion to protect noradrenergic neurons. The neurotoxin 6-OHDA (hydrochloride salt; Sigma) was dissolved in ice-cold 0.9% w/v NaCl solution containing 0.02% w/v ascorbate to a final concentration of 4 mg/ml immediately before use. Then 2 μl of 6-OHDA solution was injected into the medial substantia nigra
2.5. Electrophysiological data acquisition Testing took place 8–10 days after the implantation surgery. Electrophysiological and behavioral recordings were simultaneously performed and displayed on a computer screen for visual inspection. Neural signals were pre-amplified 25 times by a high impedance preamplification device (SWF-2, Chengdu, China) and then amplified by a multichannel bioamplication system (RM6280BD, Chengdu, China) and digitized at a sample rate of 10 kHz with band pass filtered from 0.8 to 300 Hz. The raw signal was stored for further analysis in MATLAB 2010a (The Mathworks, USA) and LFP analysis software 2009 http://www. nottingham.ac.uk/neuronal-networks/ [31]. 2.6. Histology The placement of the recording probes was checked after final testing. Immediately after the rats were given a lethal dose of pentobarbital sodium, electrical microlesion (10 μA, 10 s × 2) was induced by passing an anodal current through the electrode at each recording site. Rats were perfused intracardially with 200 ml saline followed by 200 ml 4% paraformaldehyde and 1% potassium ferricyanide in phosphate buffer solution (PBS). The brain was quickly removed and frozen in an isopentane bath at − 80 °C for histological analysis. Coronal brain
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
sections (40 μm) were cut and those encompassing M1, GPe, and GPi were mounted on slides for electrode placement verification. These slides were stained with cresyl violet for structural identification. The recording tracks and sites were examined by observing the marks left by the deposited iron or electrode lesion. Only animals with probe placement within the target nucleus were included for data analyses. Recording sites in M1, GPe, and GPi are shown in Fig. 1. To determine the extent of degeneration in dopaminergic neurons, sections of the substantia nigra and the ventral tegmental area from rats receiving 6-OHDA injection were examined for immunohistochemical staining of tyrosine hydroxylase (TH). TH staining consisted of the following steps. Free-floating sections were washed three times with phosphate buffered saline (PBS, 0.01 M, pH 7.2), and endogenous peroxidase activity was inactivated by 5 min incubation in Tris-buffered saline containing 2% H2O2. Sections were rinsed with PBS and incubated with blocking buffer containing 0.3% TritonX-100 and 5% normal goat serum followed by an overnight incubation at 4 °C with the primary antibodies. Mouse anti-tyrosine hydroxylase (1:1000; ab6211; Abcam Inc., Cambridge, UK) primary antiserum was used for immunohistochemical identification of TH; primary antibodies were detected using a biotinylated secondary antibody (Zymed Laboratories Inc., CA, USA) and an avid in horseradish peroxidase–diaminobenzidine system (DAB; Sigma Chemical, MO, USA). After the diaminobenzidine reaction, sections were rinsed with PBS and mounted on gelatin-coated slides, dehydrated and cover slipped with Permount for light microscopic examination. After TH staining, the counting of the TH-positive soma in the substantia nigra and the ventral tegmental area was carried out on four sections per animal. The number of neurons was expressed as the average of the counts obtained from the four sections. Data are
233
expressed as % of lesioned side compared to the contralateral nonlesioned side. 2.7. Data analysis Only rats with efficient lesions (i.e., N 80 net contraversive rotations in 20 min after apomorphine challenge), in which a correct placement of the electrodes was histologically determined, were used for statistical analyses. Twenty rats (ten 6-OHDA successfully lesioned rats and ten control rats) were used in the behavioral testing, while data coming from sixteen rats (eight 6-OHDA successfully lesioned rats and eight control rats) were used for LFP analyses. LFP and behavioral recordings were performed simultaneously for at least 30 min for each animal while they were walking along the ladder. For behavioral analyses, the time for rats to cross the entire length of the ladder was measured and errors were counted when rats missed a rung or slipped off. For LFP analyses, the selected segments from the 30 min recording sessions were representative of the overall recording from each rat walking on the ladder. Six of the recorded artifact-free segments containing the entire walking episode (the time to cross the entire length of the ladder for each rat in both lesioned and control groups) were randomly selected for statistical analysis. LFP power spectra were computed using Fast Fourier Transform (FFT) analysis and Welch methods as spectral estimation with 1000 samples and a Hanning window, and the power spectral density plots of LFPs calculated using Welch method also showed a visualized graphical representation of various frequencies. For the assessment of relevant LFP power, LFP power spectra were analyzed in bands of 1–12 Hz, 12–25 Hz, 25– 40 Hz, 40–70 Hz and 70–100 Hz, respectively. We calculated the root
Fig. 1. Histological representative examples of cresyl violet stained coronal sections of rat brain, photographed with low magnification demonstrating the recording electrodes in M1 (A), GPe (B) and GPi (C). Locations of electrode tips as marked by electrolytic lesions are marked with black arrows. Scale bar = 1 mm. Three or four schematic reconstructions (D, E and F) of the three regions from Paxinos and Watson [40] represent the placements of electrodes in the DA lesioned rats (black circles) and control rats (gray circles).
234
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
mean square (RMS) power per band range as a function to give a ratio across the entire signal RMS power (1–100 Hz) expressed as the percentage of total power to overcome influences due to individual/ non-specific differences in absolute power. The time–frequency spectrograms of the LFPs were used to provide a visual representation of the power spectral density with respect to time and frequency. Whole-session power spectrograms were restricted to frequencies from 1 to 100 Hz and were calculated using a Kaiser window (window length: 2084, overlap: 1800). Power spectrograms were plotted on a logarithmic scale with greater power represented by redder colors and with the threshold set at zero. Color axes were scaled so that the power levels within a given spectrogram spanned the full color range and were consistent between comparisons. Cross-correlation provided a degree of similarity between two LFPs for identifying functional connectivity and estimating lags between brain sites. The cross-correlations between the LFP signals were calculated by means of the MATLAB function xcorr (Signal Processing Toolbox). The correlation coefficient is able to describe the dependency of the signals under general types of relationship. A correlation coefficient value of 1 corresponds to perfect correlation, while a value of 0 denotes no correlation and − 1 indicates perfect anti-correlation. Next, a correlation coefficient can be estimated as a function of a time shift, and the shift for which the maximum of the coefficient value is reached can be used as an estimate of the time delay between the signals. 2.8. Statistical analyses Statistical analyses were performed with the statistical software package Sigma Stat (SPSS18, USA) to detect statistical differences. All tests were two tailed and α level was set at 0.05; a probability of less
than 5% (p b 0.05) was considered significant. Variables are presented as means ± standard error of the means (SEM). Comparisons of the effects of lesions of DA neurons and behavior activity were made using an independent-samples t test. The percentage of relative power of the basal LFPs in different frequency bands was analyzed by repeated measurement ANOVA followed by Fisher's least significant difference (LSD) test. In cases where data (cross-correlation value, lag time) were not normally distributed, non-parametric tests (Mann–Whitney U test with Dunn's method post-hoc comparisons) were used. 3. Results 3.1. Verification of loss of DA neurons following 6-OHDA infusion DA neuron lesions were confirmed with TH immunohistochemistry in eight lesioned rats at three weeks after the 6-OHDA injection. In these rats, a total loss of TH-positive neurons was observed in the substantia nigra pars compacta ipsilateral to the 6-OHDA infusion side (Fig. 2A–D). The loss of TH-positive neurons in the ventral tegmental area ipsilateral to the 6-OHDA infusion side was less pronounced (mean: 58 ± 2% loss as compared to the intact hemisphere, t test: p b 0.01) than that in the substantia nigra pars compacta (Fig. 2A, B, E and F). 3.2. Effects of unilateral DA neuron lesions on walking locomotor behavior The effects of unilateral DA neuron lesions on rats' behavior of walking on a ladder were compared between lesioned rats (n = 10) and control rats (n = 10). Rats with unilateral DA neuron lesions took a longer time to cross the ladder than control rats (14.88 ± 4.44 s vs 9.82 ± 3.00 s, t test: p b 0.01) and the foot fault scores of lesioned
Fig. 2. Coronal section of the substantia nigra pars compacta and the ventral tegmental area illustrating immunohistochemistry of tyrosine hydroxylase (TH). Note the lack of TH staining in the substantia nigra pars compacta and ventral tegmental area of the lesioned (A, C, E) hemisphere compared with the positions in intact (B, D, F) hemisphere. The framed areas in A are magnified in C (SNc) and E (VTA), respectively. The framed areas in B are magnified in D (SNc) and F (VTA), respectively. SNc, substantia nigra pars compacta; VTA, ventral tegmental area. Scale bar, A and B = 500 μm; C, D, E, and F = 125 μm.
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
235
Fig. 3. Photograph of a rat walking on a ladder and simultaneously being recorded for local field potentials (A) and comparison of walking time and foot fault scores in the DA lesioned rats and the control rats (B). Lesioned rats displayed significantly greater foot fault scores and spent longer time walking across the ladder than the control rats. Note: ** indicates p b 0.01.
animals were significantly higher than those of control rats (12.01 ± 1.47 vs 4.66 ± 1.23, t test: p b 0.01; Fig. 3).
and control rats, while there was a slight increase of power in 25–40 Hz (10.82 ± 0.53) in lesioned rats compared to control rats (7.61 ± 1.27 p b 0.05) (Fig. 6F).
3.3. Effects of DA neuron lesions on LFP activity in M1, GPe, and GPi We recorded LFP activity in M1, GPe and GPi from DA lesioned and control rats while the rats were crossing the ladder. The recording segments from paired electrodes in M1 with GPe (n = 8 rats in both lesioned and control groups) or M1 with GPi (n = 8 rats in both lesioned and control groups) revealed marked changes in LFP activities in DA loss rats. Enhanced oscillation frequency in the LFP activities in M1, GPe, and GPi was observed in lesioned rats compared to control rats (Fig. 4). Further inspection with time–frequency spectrograms showed that the LFP power in M1 and GPi from the DA lesioned rat was different from that of the control rat. The LFP power in M1 from lesioned rats was greater than that in M1 from control rats, as depicted in the plot of M1 LFP activity in the lesioned rat and not in an equivalent plot obtained from the control rat (Fig. 5A, B). The LFP power in GPe from lesioned rats was also higher than that in GPe from control rats (Fig. 5C, D). A smaller magnitude of increase in the LFP power in GPi was seen in lesioned rats compared to control rats (Fig. 5E, F). Welch estimation analysis showed that there were marked differences in the power spectra from M1, GPe, and GPi LFP activities between the lesioned and control rats. There was a predominant activity around the 1–12 Hz range from control rats and evident activity around the 12–40 Hz range from lesioned rats (Fig. 6A for M1; Fig. 6B for GPe). There was only a tiny predominant activity around 25–40 Hz in GPi from lesioned rats (Fig. 6C). The further evaluated and quantified M1, GPe and GPi LFP oscillatory activity power in 1–12 Hz, 12–25 Hz, 25–40 Hz, 40–70 Hz and 70–100 Hz from lesioned rats revealed significant differences compared with control rats. Statistically significant differences (repeated measures ANOVA) were observed in the power of M1 LFPs in both DA lesioned and control rats in different frequency bands (1–12 Hz, 25– 40 Hz and 40–70 Hz). Post-hoc comparison showed that 1–12 Hz activity power was reduced in lesioned rats as compared to control rats (36.20 ± 1.50 and 51.26 ± 3.19, p b 0.01), while 25–40 Hz (15.37 ± 0.81) and 40–70 Hz (19.93 ± 0.81) activity powers were enhanced in lesioned rats as compared to control rats (9.17 ± 0.69 and 12.54 ± 1.32, respectively, p b 0.01) (Fig. 6D), and the power of GPe LFPs in the DA neuron lesioned rats and control rats also had significant differences in several frequency bands (1–12 Hz, 12–25 Hz, 25–40 Hz and 40– 70 Hz). Post-hoc comparison showed that 1–12 Hz activity power was reduced (54.54 ± 3.91 and 70.10 ± 2.38, p b 0.01), while 12–25 Hz (18.69 ± 1.40), 25–40 Hz (14.10 ± 2.44) and 40–70 Hz (6.44 ± 0.93) powers were enhanced in lesioned rats as compared to control rats (14.13 ± 1.02, 6.06 ± 0.60, and 3.61 ± 0.41, respectively, p b 0.05– 0.01,) (Fig. 6E). There were no significant changes in GPi LFP power in 1–12 Hz, 12–25 Hz, 40–70 Hz and 70–100 Hz between lesioned rats
3.4. Effects of DA lesions on the relationship of LFP activity between M1 and GPe or GPi To determine whether changes of LFP activity observed in M1, GPe, and GPi following DA loss were correlated with alterations in their estimated time lag between putatively connected sites, the temporal relationship of LFP activity between M1 and GPe or GPi was assessed by cross-correlation and time delay value. The cross-correlation function provides a qualitative comparison of cross-correlation and time delay value between M1 and GPe or GPi by comparing the pair recording different frequency segments, which are 1–12 Hz, 12–25 Hz, 25–40 Hz, 40–70 Hz and 70–100 Hz ranges, respectively. With the MATLAB function xcorr, the qualitative comparison in different frequency segments evolves by analyzing time varying signals for identifying fundamental fluctuation frequencies in neural timeseries data, without any a priori filtering [31,32]. For this analysis, a positive lag indicates that the peak of the LFP oscillation in M1 is occurring before the peak of the LFP oscillation in GPe or GPi and a negative lag indicates the reverse. Results from these cross-correlation analyses showed that M1 LFP activity was more correlated with GPe LFPs in 25–40 Hz in the lesioned rats than in the control rats. The lag time of M1 with GPe from lesioned rats' LFPs in 25–40 Hz decreased (mean lag: 0.0026 ± 0.0002 s, from 4 rats) compared with those of control rats (mean lag: 0.8918 ± 0.7161 s, p b 0.01, Wilcoxon rank-sum test, from 4 rats). The cross-correlation value of M1 with GPe from lesioned rats' LFPs in 25–40 Hz also significantly increased (0.9872 ± 0.0128, from 4 rats) compared with those of control rats (0.2856 ± 0.0767, p b 0.01, Wilcoxon rank-sum test, from 4 rats) (Fig. 7). Finally, a lag in M1 with GPe was only found in 25–40 Hz, but not in 1–12 Hz, 12–25 Hz, 40–70 Hz and 70–100 Hz ranges (data not shown). No significant difference of lag time was observed in M1 with GPi LFPs (data not shown). Collectively, these results showed an effect of DA loss on lag time in LFPs in 25–40 Hz oscillations in M1 with GPe. Taken together, the near-zero lag time and near one cross-correlation value indicated that LFP activity in the two brain regions is relatively synchronous. 4. Discussion The present study showed the following: First, DA loss increased LFP spectral power in M1, GPe and GPi in the 12–70 Hz range and decreased LFP spectral power in M1 and GPe in 1–12 Hz. Second, these LFP activity changes were accompanied with marked increases in synchronization of oscillatory activities in M1 with GPe. Third, while the lesioned rats had decreased LFP power in 1–12 Hz of M1 and GPe compared to the control rats during ladder walking, the LFP power in GPi in 1–12 Hz
236 M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
Fig. 4. Representative examples of LFP recording from paired electrodes in M1 with GPe or GPi during ladder walking from a DA neuron lesioned rat (A and C) and an intact rat (B and D).
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240 Fig. 5. Representative examples of time–frequency spectrograms of LFP power in M1, GPe, and GPi during ladder walking from a DA neuron lesioned rat (A, C, E) and a control rat (B, D, F). Spectral power was plotted on a logarithmic scale with greater power represented by redder colors. LFP spectra power in M1, GPe, and GPi from the lesioned rat was greater than that from the control rat.
237
238
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
239
Fig. 7. Comparison of lag time and correlation value between M1 and GPe in the DA neuron lesioned rats and the control rats. Bar graphs represent the lag time in the lesioned rats (n = 4) that was significantly lower than that in the control rats (n = 4; *p b 0.05, **p b 0.01) (A); correlation value between M1 and GPe in lesioned rats (n = 4) was significantly higher than that in control rats (n = 4; *p b 0.05, **p b 0.01) (B).
showed no alteration compared to the control rats. We suggest that these observations in the alteration of LFP activities in M1, GPe and GPi and synchronization between M1 and GPe in DA depleted rats play a role in the motor impairment that is shown in animal models of PD patients. These results in aberrant LFP activity are consequences of the degeneration of dopamine neurons in the substantia nigra and possibly due to their close interconnectivity between GPe and GPi. The results also suggest a channel of communication through which the cortex interacts with the basal ganglia and that is susceptible to dysfunction following dopamine depletion. The identification of abnormalities in LFPs in the basal ganglia and related cortex in movement disorders remains a highly important field of research, because such findings may eventually help us to a better understanding of the involvement of the basal ganglia in parkinsonism. Studies have shown that excessive LFPs activity at beta frequencies (10–35 Hz) is a key pathophysiological feature of these parkinsonian circuits. Our results showed that the fractional spectral power in the 12–70 Hz bands was higher on average in the parkinsonian condition for M1 (25–70 Hz), GPe (12–70 Hz) and GPi (25–40 Hz), agreeing in part with previous findings of prominent oscillations in the beta frequency range in the parkinsonian state [8,9,33]. In addition to the beta frequency activity, the abnormalities in LFPs in two other bands have been found, that is, decreased LFP spectral power in low-frequency band activity (1–12 Hz) and increased LFP spectral power in high frequency band activity (N 35 Hz). One explanation of these differences is the possibility that they are related to the brain state and behavioral context (anesthesia vs awake and performing in patients or animals). Our study used awake, behaving rats while other studies used anesthetic or awake but immobile animals. There are recent reports showing that increased γ-band in 35–100 Hz power depends on arousal and is coincident with motor tasks [10,21]. The abnormalities in more than beta frequency in LFPs in M1 and GPe in PD rats have been identified, providing additional material to consider in the complex model of basal ganglia pathophysiology. More recent research has highlighted the oscillatory nature of excessive neuronal synchronization in the parkinsonian state. Data from LFP amplitude waveforms by cross-correlation analyses showed that the lag time for M1 with GPe changed from a few hundred milliseconds in control rats to near zero in DA lesioned rats, and the cross-correlation value also changed from decimal fraction in control rats to nearly one in DA lesioned rats. The results indicate that LFP activities in the two
brain regions become relatively synchronous and suggest stronger functional connectivity following dopaminergic denervation. This is consistent with the large body of literature reporting synchronous oscillations in parkinsonian patients as well as animal models of the disease [14,25,34]. It also suggests that the cortical rhythms are being more effectively transmitted to downstream sites after loss of dopamine, supporting the theory of hypersynchronization in corticobasal ganglia circuits [11,21,22]. Further inspection along the M1 with GPe temporal relationship revealed that the near zero lag time between M1 and GPe was present in a specific frequency of 25–40 Hz. While these observations provide a clear biological marker for PD, it is not clear whether synchronization is an epiphenomenon or a true pathogenic alteration and how the dominant frequency becomes focused in the 25–40 Hz range in PD. The shortening of the latency between M1 and GPe at a specific frequency range suggests that there might be some special mechanism that facilitates synchronization and/or transfer of relevant information across distant brain regions. First, it may be possible to interpret these in the context of the death of dopaminergic neurons in the basal ganglia, because a dopamine decrease should strengthen the connections between neurons of the cortex–basal ganglia network. This moves the network toward a more synchronized state and the transient synchronous patterns become more prevalent [34,35]. Second, it is also possible that the excessive synchronization is a consequence of the integration of many changes in the basal ganglia, thalamus, or cortex that are known to accompany parkinsonism in human and animals [23,36]. Therefore, it is possible that 25–40 Hz is a frequency range compatible with resonant activity in the dopamine-depleted basal ganglia–thalamocortical circuit. Taken together, the observed synchronous changes are associated with, but are not causal of, parkinsonism, and that they may not be primarily dopaminergic in origin. Our data showed that DA loss is associated dramatically with increased LFP power in M1 and GPe, which was selective emergence in 25–70 Hz and 12–70 Hz for M1 and GPe, respectively. However, GPi only showed a modest increase of LFP power in the 25–40 Hz range after DA loss. Overall, it is particularly interesting that parkinsonism appears to be more strongly associated with detectable changes in oscillatory power in GPe and M1, than with changes in the GPi. The dramatically different patterns of frequency power observed in M1, GPe and GPi in DA neuron lesioned versus control rats may imply a different mechanism underlying the effect of DA loss. It may be possible to interpret these findings in the context of abnormal frequency band reactivity in M1, GPe and
Fig. 6. LFP power spectra of M1, GPe, and GPi during rats walking along a ladder. Welch methods as spectral estimation in power spectra showed an example of LFP activities in a control (gray line) and a lesioned (black line) rat (A, B, C). Bar graphs in the right represent mean relative LFP power in M1, GPe and GPi within a series of frequency ranges: 1–12 Hz, 12–25 Hz, 25–40 Hz, 40–70 Hz and 70–100 Hz in control rats (n = 8) and in DA neuron lesioned rats (n = 8) (D, E, F). The M1 LFP power was significantly lower in 1–12 Hz and significantly increased in 25–40 Hz and 40–70 Hz in the lesioned rats than in control rats (D). The GPe LFP power showed the same tendency to that of M1 (E), while GPi LFP power only slightly increased in 25–40 Hz band in the lesioned rats compared to the control rats (F).
240
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
GPi in lesioned rats, for that though GPe and GPi receive similar inputs from the cerebral cortex, they project to different brain sites and thus have different functional roles in basal ganglia information processing [37], so the DA loss in association with the LFP activities in different frequency bands and different brain regions may show different processing patterns in the basal ganglia and the structures they are connected with. Therefore, it lends support to the general hypothesis that special frequency power alteration exerts a dysfunctional effect on motor circuits modulated by the basal ganglia. The data suggest that M1, GPe and GPi form distinct neuronal networks for which oscillatory activity may have different neurophysiological roles. This also may be possible to interpret that the LFP activity of M1 with GPe in lesioned rats becomes relatively synchronous, but was not observed in M1 with GPi in lesioned rats. The observation is consistent with the hypothesis that cortical rhythms entrain GPe activity more readily than they do to GPi activity after DA loss. One possibility is that unlike M1 and GPe, the relative position of GPi in the basal ganglia circuits as a common output for both the indirect and direct pathways may in part account for the lack of modulation after DA loss. This also may be the reason that GPi is a common target for neuromodulation in the treatment of PD. In PD patients, the neuronal circuits in GPi might partially or fully retain their ability to oscillate in a movement dependent manner [4,5,37–39]. In summary, the results showed that specific frequency power in the motor cortex, GPe and GPi varied dramatically with movement deficits in the DA neuron lesioned rats during locomotor behavior. The results also suggest that DA loss promotes synchronization activity between the motor cortex and GPe. Determining the relevance of such specific frequency and synchrony to behavior and the functional significance of excessive synchronization is a more challenging task. Hopefully, further exploration of the relationships between the processes causing excessive synchronization of basal ganglia activity across specific frequency will provide additional insights into fundamental mechanisms underlying the profound behavioral consequences of DA loss. References [1] Wichmann T, DeLong MR. Deep-brain stimulation for basal ganglia disorders. Basal Ganglia 2011;1:65–77. [2] Graybiel AM. Building action repertoires: memory and learning functions of the basal ganglia. Curr Opin Neurobiol 1995;5:733–41. [3] Hammond C, Bergman H, Brown P. Pathological synchronization in Parkinson's disease: networks, models and treatments. Trends Neurosci 2007;30:357–64. [4] Priori A, Foffani G, Pesenti A, Bianchi A, Chiesa V, Baselli G, et al. Movement-related modulation of neural activity in human basal ganglia and its L-DOPA dependency: recordings from deep brain stimulation electrodes in patients with Parkinson's disease. Neurol Sci 2002;23:s101–2. [5] Silberstein P, Kühn AA, Kupsch A, Trottenberg T, Krauss JK, Wöhrle JC, et al. Patterning of globus pallidus local field potentials differs between Parkinson's disease and dystonia. Brain 2003;126:2597–608. [6] Devergnas A, Pittard D, Bliwise D, Wichmann T. Relationship between oscillatory activity in the cortico-basal ganglia network and parkinsonism in MPTP-treated monkeys. Neurobiol Dis 2014;68:156–66. [7] Dejean C, Gross CE, Bioulac B, Boraud T. Dynamic changes in the cortex–basal ganglia network after dopamine depletion in the rat. Am Physiol Soc 2008:385–96. [8] Ellens DJ, Leventhal DK. Review: electrophysiology of basal ganglia and cortex in models of Parkinson disease. J Park Dis 2013;3:241–54. [9] Stein E, Bar-Gad I. Beta oscillations in the cortico-basal ganglia loop during parkinsonism. Exp Neurol 2013;245:52–9. [10] Van Der Meer MA, Kalenscher T, Lansink CS, Pennartz CM, Berke JD, Redish AD. Integrating early results on ventral striatal gamma oscillations in the rat. Front Neurosci 2010;4. [11] Mallet N, Pogosyan A, Sharott A, Csicsvari J, Bolam JP, Brown P, et al. Disrupted dopamine transmission and the emergence of exaggerated beta oscillations in subthalamic nucleus and cerebral cortex. Soc Neurosci 2008:4795–806. [12] Walters JR, Bergstrom DA. Basal ganglia network synchronization in animal models of Parkinson's disease. Springer; 2009 117–42.
[13] Walters JR, Tierney PL, Bergstrom DA. Oscillatory activity and synchronization in the basal ganglia network in rodent models of Parkinson's disease. Springer; 2009 443–59. [14] Delaville C, Cruz AV, McCoy AJ, Brazhnik E, Avila I, Novikov N, et al. Oscillatory activity in basal ganglia and motor cortex in an awake behaving rodent model of Parkinson's disease. Basal Ganglia 2014;3:221–7. [15] Magill PJ, Bolam JP, Bevan MD. Relationship of activity in the subthalamic nucleus– globus pallidus network to cortical electroencephalogram. J Neurosci 2000;20: 820–33. [16] Tseng KY, Kasanetz F, Kargieman L, Riquelme LA, Murer MG. Cortical slow oscillatory activity is reflected in the membrane potential and spike trains of striatal neurons in rats with chronic nigrostriatal lesions. J Neurosci 2001;21:6430–9. [17] Aravamuthan BR, Bergstrom DA, French RA, Taylor JJ, Parr-Brownlie LC, Walters JR. Altered neuronal activity relationships between the pedunculopontine nucleus and motor cortex in a rodent model of Parkinson's disease. Elsevier; 2008 268–80. [18] de Hemptinne C, Ryapolova-Webb ES, Air EL, Garcia PA, Miller KJ, Ojemann JG, et al. Exaggerated phase–amplitude coupling in the primary motor cortex in Parkinson disease. Proc Natl Acad Sci 2013;110:4780–5. [19] Kita H, Kita T. Cortical stimulation evokes abnormal responses in the dopaminedepleted rat basal ganglia. J Neurosci 2011;31:10311–22. [20] Belluscio MA, Riquelme LA, Murer MG. Striatal dysfunction increases basal ganglia output during motor cortex activation in parkinsonian rats. Wiley Online Library; 2007 2791–804. [21] Jenkinson N, Kühn AA, Brown P. Gamma oscillations in the human basal ganglia. Exp Neurol 2013;245:72–6. [22] Degos B, Deniau JM, Chavez M, Maurice N. Chronic but not acute dopaminergic transmission interruption promotes a progressive increase in cortical beta frequency synchronization: relationships to vigilance state and akinesia. Cereb Cortex 2009; 19:1616–30. [23] Brazhnik E, Cruz AV, Avila I, Wahba MI, Novikov N, Ilieva NM, et al. State-dependent spike and local field synchronization between motor cortex and substantia nigra in hemiparkinsonian rats. J Neurosci 2012;32:7869–80. [24] Nambu A. Globus pallidus internal segment. In: James M, Tepper EDA, Bolam JP, editors. Progress in brain research. Elsevier; 2007. p. 135–50. [25] Avila I, Parr-Brownlie LC, Brazhnik E, Castañeda E, Bergstrom DA, Walters JR. Beta frequency synchronization in basal ganglia output during rest and walk in a hemiparkinsonian rat. Exp Neurol 2010;221:307–19. [26] Jadavji N, Metz G. Both pre- and post-lesion experiential therapy is beneficial in 6hydroxydopamine dopamine-depleted female rats. Neuroscience 2009;158:373–86. [27] Vergara-Aragon P, Gonzalez CL, Whishaw IQ. A novel skilled-reaching impairment in paw supination on the “good” side of the hemi-Parkinson rat improved with rehabilitation. J Neurosci 2003;23:579–86. [28] Ungerstedt U, Arbuthnott GW. Quantitative recording of rotational behavior in rats after 6-hydroxy-dopamine lesions of the nigrostriatal dopamine system. Brain Res 1970;24:485–93. [29] Kubie JL. A driveable bundle of microwires for collecting single-unit data from freely-moving rats. Physiol Behav 1984;32:115–8. [30] Nicolelis MA, Ghazanfar AA, Faggin BM, Votaw S, Oliveira LM. Reconstructing the engram: simultaneous, multisite, many single neuron recordings. Neuron 1997;18: 529–37. [31] Masimore B, Kakalios J, Redish A. Measuring fundamental frequencies in local field potentials. J Neurosci Methods 2004;138:97–105. [32] Sharott A, Moll CK, Engler G, Denker M, Grün S, Engel AK. Different subtypes of striatal neurons are selectively modulated by cortical oscillations. J Neurosci 2009; 29:4571–85. [33] Moran RJ, Mallet N, Litvak V, Dolan RJ, Magill PJ, Friston KJ, et al. Alterations in brain connectivity underlying beta oscillations in parkinsonism. PLoS Comput Biol 2011;7: e1002124. [34] Park C, Worth RM, Rubchinsky LL. Fine temporal structure of beta oscillations synchronization in subthalamic nucleus in Parkinson's disease. J Neurophysiol 2010;103:2707–16. [35] Park C, Worth RM, Rubchinsky LL. Neural dynamics in parkinsonian brain: the boundary between synchronized and nonsynchronized dynamics. Phys Rev E 2011;83: 042901. [36] Brazhnik E, Novikov N, McCoy AJ, Cruz AV, Walters JR. Functional correlates of exaggerated oscillatory activity in basal ganglia output in hemiparkinsonian rats. Exp Neurol 2014;261:563–77. [37] Jaeger D, Kita H. Functional connectivity and integrative properties of globus pallidus neurons. Neuroscience 2011;198:44–53. [38] Brown P. Oscillatory nature of human basal ganglia activity: relationship to the pathophysiology of Parkinson's disease. Mov Disord 2003;18:357–63. [39] Tsiokos C, Hu X, Pouratian N. 200–300 Hz movement modulated oscillations in the internal globus pallidus of patients with Parkinson's disease. Neurobiol Dis 2013; 54:464–74. [40] Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. 4th edn. San Diego: Academic; 1998.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Altered neuronal activity in the primary motor cortex and globus pallidus after dopamine depletion in rats☆ Min Wang a,⁎, Min Li a,1, Xiwen Geng a,1, Zhimin Song b, H. Elliott Albers b, Maoquan Yang a, Xiao Zhang a, Jinlu Xie a, Qingyang Qu a, Tingting He a a b
Key Laboratory of Animal Resistance of Shandong Province, College of Life Science, Shandong Normal University, Jinan 250014, People's Republic of China Center for Behavioral Neuroscience, Neuroscience Institute, Georgia State University, Atlanta, GA 30302, United States
a r t i c l e
i n f o
Article history: Received 11 September 2014 Received in revised form 9 November 2014 Accepted 10 December 2014 Available online 18 December 2014 Keywords: Parkinson's disease Local field potential Motor cortex Globus pallidus Basal ganglia Microelectrode
a b s t r a c t The involvement of dopamine (DA) neuron loss in the etiology of Parkinson's disease has been well documented. The neural mechanisms underlying the effects of DA loss and the resultant motor dysfunction remain unknown. To gain insights into how loss of DA disrupts the electrical processes in the cortico-subcortical network, the present study explores the effects of DA neuron depletion on electrical activity in the primary motor cortex (M1), on the external and the internal segment of the globus pallidus (GPe and GPi respectively), and on their temporal relationships. Comparison of local field potentials (LFPs) in these brain regions from unilateral hemispheric DA neuron depleted rats and neurologically intact rats revealed that the spectrum power of LFPs in 12–70 Hz (for M1, and GPe) and in 25–40 Hz (for GPi) was significantly greater in the DA depleted rats than that in the control group. These changes were associated with a shortening of latency in LFP activities between M1 and GPe, from several hundred milliseconds in the intact animals to close to zero in the DA depleted animals. LFP oscillations in M1 were significantly more synchronized with those in GPe in the DA depleted rats compared with those in the control rats. By contrast, the synchronization of oscillation in LFP activities between M1 and GPi did not differ between the DA depleted and intact rats. Not surprisingly, rats that had DA neuron depletion spent more time along the ladder compared with the control rats. These data suggest that enhanced oscillatory activity and increased synchronization of LFPs may contribute to movement impairment in the rat model of Parkinson's disease. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The basal ganglia form a complex network that processes cortical information important for movement and cognition [1–3]. Alterations of neuronal activity in the basal ganglia and cortices have been reported in patients with Parkinson's disease (PD) as well as in animal models of PD [4–11]. These changes in neuronal activity are believed to mediate many of the dysfunctional effects seen in PD because they impact the circuits connecting the basal ganglia and the cortex [10,12–14]. However, the specific pathological changes that occur with DA loss in oscillation activity and functional connectivity from the cortex to the basal ganglia remain poorly understood.
☆ Supporting grant: This study was supported by the Natural Science Foundation of Shandong Province (No. ZR2010CM055) and the Science and Technological Project of Shandong Province (Nos. 2011GGB01004 and 2010GGX10133). ⁎ Corresponding author at: Department of Anatomy and Physiology, College of Life Science, Shandong Normal University, 250014, People's Republic of China. Tel.: + 86 15615614667. E-mail address: [email protected] (M. Wang). 1 Contributed equally.
http://dx.doi.org/10.1016/j.jns.2014.12.014 0022-510X/© 2014 Elsevier B.V. All rights reserved.
The transmission of rhythmic cortical activity from the cortex to the basal ganglia has been studied in anesthetized rats [15,16]. An increase in discharge oscillations at 1 Hz from the basal ganglia is observed in anesthetized rats with unilateral DA depletion. This activity is coherent with the 1 Hz oscillatory firing patterns dominant in the cortex [17,18]. Transmission from the cortex to the basal ganglia has also been studied in rats using electrical stimulation of the motor cortex. After DA depletion, the cortical stimulation induces a long disinhibition in GPe, which is transmitted to GPi and generates an abnormally strong long inhibition in GPi, which then generates a strong and long excitation in the thalamic projection sites that are transmitted to the motor cortex with incorrect information [19,20]. Simultaneous behavioral recordings and chronic microelectrode neural recordings in awake and behaving rodents hold great promise to study the neural bases of behavior and the information transmission between the cortex and the basal ganglia. Some studies suggest that increases in the synchronization between the subthalamic nucleus or substantia nigra to cortex, or the external segment of the globus pallidus facilitate the emergence of special range activity in the cortex after DA loss [10,11,21–23]. These observations led to our general hypothesis that DA neuron loss alters the magnitude of the electrical activities in
232
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
the basal ganglia and enhances its tendency to become entrained by oscillatory activity expressed in the cortex. The aim of the present study was to determine: a) how DA loss affects LFP activities in the primary motor cortex (M1), the globus pallidus in rodents (an area analogous to the external segment of the globus pallidus in primates, GPe) and the entopeduncular nucleus in rodents (an area analogous to the internal segment of the globus pallidus in primates, GPi) [24,25]. In this paper, we will use the primate terminology for simplification and consistency reasons, and b) how DA loss affects the coherent oscillatory activity at the level of LFPs recorded in M1 with GPe or GPi. LFPs were pair-recorded from chronically implanted electrodes in the M1 and GPe or GPi of rats with unilateral DA neuron lesions, while they were engaged in a locomotor task. Control unilateral recordings in M1 and GPe or GPi were also obtained in neurologically intact rats.
(5.2 mm posterior and 2.1 mm lateral of the bregma, and 8.0 mm ventral to the dura) and 2 μl into the ventral tegmental area (6.8 mm posterior and 0.6 mm lateral of the bregma, and 8.6 mm ventral to the dura). The injection was made at a rate of 0.5 μl/min using a 5-μl microsyringe. After each injection, the micropipette was left in place for an additional 5 min and then slowly withdrawn. The control rats received only the vehicle (0.02% ascorbic acid in physiological saline) at the same coordinates. Rats received post-operative care until wakefulness and were returned to their home cages. Four weeks after 6-OHDA injection, a rotating behavioral response to apomorphine (0.05 mg/kg, s.c.) test was performed in order to assess the severity of the nigral lesion [28]. Animals that performed at least 80 rotations opposite to the lesioned site within 20 min following the apomorphine treatment were considered animals with successful surgery and only those were used for electrophysiological recordings.
2. Materials and methods
2.4. Recording electrode placement
2.1. Animal
A headstage adapted from previous studies [29,30] was designed for the recording electrode assembly. Briefly, the headstage composed of a 3-pin box connector socket stripe (2.54 mm, BCSS-1-SV, Indiana, USA) with double nickel–chromium teflon-insulated microwires (California Fine Wire, Grover Beach, CA, USA, 100 μm in diameter). One end of the microwires was stripped by gently scraping off about 2 mm of insulation and tinned with a soldering iron to attach to a pin of the connectors. A copper wire (200 μm in diameter) served as a ground wire was also soldered onto a pin of the connector. After the microwires and ground wire were tightly soldered, Epoxy glue was applied to the base and surrounding of the connector to build onto the headstage a movable driver with recording electrodes. The tip impedance of the electrode was 0.5–0.8 MΩ at 1 kHz. The headstage was implanted into both intact rats and rats with unilateral DA neuron lesions. The target regions of the headstage were M1 (1.0 mm anterior and 2.0 mm lateral of the bregma, and 0.4 mm ventral to the dura) and GPe (1.0 mm posterior and 3.0 mm lateral of the bregma, and 6.6 mm ventral to the dura) or GPi (2.3 mm posterior and 2.9 mm lateral of the bregma, and 7.7 mm ventral to the dura). Skull screws were used for ground wire connection and fixing the headstage to the skull. Dental cement was then applied to glue the headstage to the skull. The rats were given ketoprofen (Ketofen 2 mg/kg, s.c., Sigma, USA) following surgery and 24 h later again for pain relief. They were allowed at least 7 days to recover before the first recording session.
All experimental procedures were conducted on male Wistar rats (280–320 g, Animal Center of Shandong University, China). Rats were kept under standard housing conditions at constant temperature (22 ± 1 °C), humidity (relative, 30%), and light/dark 14/10 cycles. Water was available ad libitum. Food intake was limited to 10–20 g/day to maintain constant animal weight. Animal care and surgery were conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals (NIH Guidelines). Every effort was made to minimize the number of animals used and the pains they suffered. 2.2. Behavioral training task A ladder composed of two clear Plexiglas side walls (100 × 20 cm) with timber rungs (3 mm in diameter) inserted at random distances ranging from 1 to 3 cm was used in the testing. The ladder was elevated 30 cm above the ground with a square wooden case (home cage) at one end. Two weeks prior to surgery, rats were placed at one end of the ladder and trained to walk spontaneously to the home cage. The rats received three training trials each day for 3–7 days, after which they reliably walked to the home cage after being placed at the other end. During the testing, the time to cross the entire length of the ladder was recorded and the number of foot placement errors was determined by a rating system used in previous studies [26,27]. Briefly, 3 points were given when the animal completed or missed a rung and the body fell onto the ladder; 2 points were given when the animal initially stepped on the rungs but slipped off; and 1 point was given when the animal step a paw on a rung but did not place its body weight on that paw. The animals' performance was videotaped by a camera (Logitech AF, Taiwan) positioned at a slight ventral angle which enabled both sides of the body and paw positions to be simultaneously recorded. 2.3. Unilateral lesion of the nigrostriatal pathway Unilateral 6-hydroxydopamine (6-OHDA) lesion of the dopaminergic nigrostriatal pathway was performed on rats after the behavioral training. Rats were anesthetized with chloral hydrate (400 mg/kg, i.p.) and placed in a stereotaxic frame, with body temperature maintained at 37 ± 0.5 °C using a heating pad. The skull was exposed and two holes were drilled on the skull according to appropriate coordinates. Rats were injected with desmethylimipramine (15 mg/kg, i.p.) 30 min prior to the intracerebral infusion to protect noradrenergic neurons. The neurotoxin 6-OHDA (hydrochloride salt; Sigma) was dissolved in ice-cold 0.9% w/v NaCl solution containing 0.02% w/v ascorbate to a final concentration of 4 mg/ml immediately before use. Then 2 μl of 6-OHDA solution was injected into the medial substantia nigra
2.5. Electrophysiological data acquisition Testing took place 8–10 days after the implantation surgery. Electrophysiological and behavioral recordings were simultaneously performed and displayed on a computer screen for visual inspection. Neural signals were pre-amplified 25 times by a high impedance preamplification device (SWF-2, Chengdu, China) and then amplified by a multichannel bioamplication system (RM6280BD, Chengdu, China) and digitized at a sample rate of 10 kHz with band pass filtered from 0.8 to 300 Hz. The raw signal was stored for further analysis in MATLAB 2010a (The Mathworks, USA) and LFP analysis software 2009 http://www. nottingham.ac.uk/neuronal-networks/ [31]. 2.6. Histology The placement of the recording probes was checked after final testing. Immediately after the rats were given a lethal dose of pentobarbital sodium, electrical microlesion (10 μA, 10 s × 2) was induced by passing an anodal current through the electrode at each recording site. Rats were perfused intracardially with 200 ml saline followed by 200 ml 4% paraformaldehyde and 1% potassium ferricyanide in phosphate buffer solution (PBS). The brain was quickly removed and frozen in an isopentane bath at − 80 °C for histological analysis. Coronal brain
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
sections (40 μm) were cut and those encompassing M1, GPe, and GPi were mounted on slides for electrode placement verification. These slides were stained with cresyl violet for structural identification. The recording tracks and sites were examined by observing the marks left by the deposited iron or electrode lesion. Only animals with probe placement within the target nucleus were included for data analyses. Recording sites in M1, GPe, and GPi are shown in Fig. 1. To determine the extent of degeneration in dopaminergic neurons, sections of the substantia nigra and the ventral tegmental area from rats receiving 6-OHDA injection were examined for immunohistochemical staining of tyrosine hydroxylase (TH). TH staining consisted of the following steps. Free-floating sections were washed three times with phosphate buffered saline (PBS, 0.01 M, pH 7.2), and endogenous peroxidase activity was inactivated by 5 min incubation in Tris-buffered saline containing 2% H2O2. Sections were rinsed with PBS and incubated with blocking buffer containing 0.3% TritonX-100 and 5% normal goat serum followed by an overnight incubation at 4 °C with the primary antibodies. Mouse anti-tyrosine hydroxylase (1:1000; ab6211; Abcam Inc., Cambridge, UK) primary antiserum was used for immunohistochemical identification of TH; primary antibodies were detected using a biotinylated secondary antibody (Zymed Laboratories Inc., CA, USA) and an avid in horseradish peroxidase–diaminobenzidine system (DAB; Sigma Chemical, MO, USA). After the diaminobenzidine reaction, sections were rinsed with PBS and mounted on gelatin-coated slides, dehydrated and cover slipped with Permount for light microscopic examination. After TH staining, the counting of the TH-positive soma in the substantia nigra and the ventral tegmental area was carried out on four sections per animal. The number of neurons was expressed as the average of the counts obtained from the four sections. Data are
233
expressed as % of lesioned side compared to the contralateral nonlesioned side. 2.7. Data analysis Only rats with efficient lesions (i.e., N 80 net contraversive rotations in 20 min after apomorphine challenge), in which a correct placement of the electrodes was histologically determined, were used for statistical analyses. Twenty rats (ten 6-OHDA successfully lesioned rats and ten control rats) were used in the behavioral testing, while data coming from sixteen rats (eight 6-OHDA successfully lesioned rats and eight control rats) were used for LFP analyses. LFP and behavioral recordings were performed simultaneously for at least 30 min for each animal while they were walking along the ladder. For behavioral analyses, the time for rats to cross the entire length of the ladder was measured and errors were counted when rats missed a rung or slipped off. For LFP analyses, the selected segments from the 30 min recording sessions were representative of the overall recording from each rat walking on the ladder. Six of the recorded artifact-free segments containing the entire walking episode (the time to cross the entire length of the ladder for each rat in both lesioned and control groups) were randomly selected for statistical analysis. LFP power spectra were computed using Fast Fourier Transform (FFT) analysis and Welch methods as spectral estimation with 1000 samples and a Hanning window, and the power spectral density plots of LFPs calculated using Welch method also showed a visualized graphical representation of various frequencies. For the assessment of relevant LFP power, LFP power spectra were analyzed in bands of 1–12 Hz, 12–25 Hz, 25– 40 Hz, 40–70 Hz and 70–100 Hz, respectively. We calculated the root
Fig. 1. Histological representative examples of cresyl violet stained coronal sections of rat brain, photographed with low magnification demonstrating the recording electrodes in M1 (A), GPe (B) and GPi (C). Locations of electrode tips as marked by electrolytic lesions are marked with black arrows. Scale bar = 1 mm. Three or four schematic reconstructions (D, E and F) of the three regions from Paxinos and Watson [40] represent the placements of electrodes in the DA lesioned rats (black circles) and control rats (gray circles).
234
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
mean square (RMS) power per band range as a function to give a ratio across the entire signal RMS power (1–100 Hz) expressed as the percentage of total power to overcome influences due to individual/ non-specific differences in absolute power. The time–frequency spectrograms of the LFPs were used to provide a visual representation of the power spectral density with respect to time and frequency. Whole-session power spectrograms were restricted to frequencies from 1 to 100 Hz and were calculated using a Kaiser window (window length: 2084, overlap: 1800). Power spectrograms were plotted on a logarithmic scale with greater power represented by redder colors and with the threshold set at zero. Color axes were scaled so that the power levels within a given spectrogram spanned the full color range and were consistent between comparisons. Cross-correlation provided a degree of similarity between two LFPs for identifying functional connectivity and estimating lags between brain sites. The cross-correlations between the LFP signals were calculated by means of the MATLAB function xcorr (Signal Processing Toolbox). The correlation coefficient is able to describe the dependency of the signals under general types of relationship. A correlation coefficient value of 1 corresponds to perfect correlation, while a value of 0 denotes no correlation and − 1 indicates perfect anti-correlation. Next, a correlation coefficient can be estimated as a function of a time shift, and the shift for which the maximum of the coefficient value is reached can be used as an estimate of the time delay between the signals. 2.8. Statistical analyses Statistical analyses were performed with the statistical software package Sigma Stat (SPSS18, USA) to detect statistical differences. All tests were two tailed and α level was set at 0.05; a probability of less
than 5% (p b 0.05) was considered significant. Variables are presented as means ± standard error of the means (SEM). Comparisons of the effects of lesions of DA neurons and behavior activity were made using an independent-samples t test. The percentage of relative power of the basal LFPs in different frequency bands was analyzed by repeated measurement ANOVA followed by Fisher's least significant difference (LSD) test. In cases where data (cross-correlation value, lag time) were not normally distributed, non-parametric tests (Mann–Whitney U test with Dunn's method post-hoc comparisons) were used. 3. Results 3.1. Verification of loss of DA neurons following 6-OHDA infusion DA neuron lesions were confirmed with TH immunohistochemistry in eight lesioned rats at three weeks after the 6-OHDA injection. In these rats, a total loss of TH-positive neurons was observed in the substantia nigra pars compacta ipsilateral to the 6-OHDA infusion side (Fig. 2A–D). The loss of TH-positive neurons in the ventral tegmental area ipsilateral to the 6-OHDA infusion side was less pronounced (mean: 58 ± 2% loss as compared to the intact hemisphere, t test: p b 0.01) than that in the substantia nigra pars compacta (Fig. 2A, B, E and F). 3.2. Effects of unilateral DA neuron lesions on walking locomotor behavior The effects of unilateral DA neuron lesions on rats' behavior of walking on a ladder were compared between lesioned rats (n = 10) and control rats (n = 10). Rats with unilateral DA neuron lesions took a longer time to cross the ladder than control rats (14.88 ± 4.44 s vs 9.82 ± 3.00 s, t test: p b 0.01) and the foot fault scores of lesioned
Fig. 2. Coronal section of the substantia nigra pars compacta and the ventral tegmental area illustrating immunohistochemistry of tyrosine hydroxylase (TH). Note the lack of TH staining in the substantia nigra pars compacta and ventral tegmental area of the lesioned (A, C, E) hemisphere compared with the positions in intact (B, D, F) hemisphere. The framed areas in A are magnified in C (SNc) and E (VTA), respectively. The framed areas in B are magnified in D (SNc) and F (VTA), respectively. SNc, substantia nigra pars compacta; VTA, ventral tegmental area. Scale bar, A and B = 500 μm; C, D, E, and F = 125 μm.
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
235
Fig. 3. Photograph of a rat walking on a ladder and simultaneously being recorded for local field potentials (A) and comparison of walking time and foot fault scores in the DA lesioned rats and the control rats (B). Lesioned rats displayed significantly greater foot fault scores and spent longer time walking across the ladder than the control rats. Note: ** indicates p b 0.01.
animals were significantly higher than those of control rats (12.01 ± 1.47 vs 4.66 ± 1.23, t test: p b 0.01; Fig. 3).
and control rats, while there was a slight increase of power in 25–40 Hz (10.82 ± 0.53) in lesioned rats compared to control rats (7.61 ± 1.27 p b 0.05) (Fig. 6F).
3.3. Effects of DA neuron lesions on LFP activity in M1, GPe, and GPi We recorded LFP activity in M1, GPe and GPi from DA lesioned and control rats while the rats were crossing the ladder. The recording segments from paired electrodes in M1 with GPe (n = 8 rats in both lesioned and control groups) or M1 with GPi (n = 8 rats in both lesioned and control groups) revealed marked changes in LFP activities in DA loss rats. Enhanced oscillation frequency in the LFP activities in M1, GPe, and GPi was observed in lesioned rats compared to control rats (Fig. 4). Further inspection with time–frequency spectrograms showed that the LFP power in M1 and GPi from the DA lesioned rat was different from that of the control rat. The LFP power in M1 from lesioned rats was greater than that in M1 from control rats, as depicted in the plot of M1 LFP activity in the lesioned rat and not in an equivalent plot obtained from the control rat (Fig. 5A, B). The LFP power in GPe from lesioned rats was also higher than that in GPe from control rats (Fig. 5C, D). A smaller magnitude of increase in the LFP power in GPi was seen in lesioned rats compared to control rats (Fig. 5E, F). Welch estimation analysis showed that there were marked differences in the power spectra from M1, GPe, and GPi LFP activities between the lesioned and control rats. There was a predominant activity around the 1–12 Hz range from control rats and evident activity around the 12–40 Hz range from lesioned rats (Fig. 6A for M1; Fig. 6B for GPe). There was only a tiny predominant activity around 25–40 Hz in GPi from lesioned rats (Fig. 6C). The further evaluated and quantified M1, GPe and GPi LFP oscillatory activity power in 1–12 Hz, 12–25 Hz, 25–40 Hz, 40–70 Hz and 70–100 Hz from lesioned rats revealed significant differences compared with control rats. Statistically significant differences (repeated measures ANOVA) were observed in the power of M1 LFPs in both DA lesioned and control rats in different frequency bands (1–12 Hz, 25– 40 Hz and 40–70 Hz). Post-hoc comparison showed that 1–12 Hz activity power was reduced in lesioned rats as compared to control rats (36.20 ± 1.50 and 51.26 ± 3.19, p b 0.01), while 25–40 Hz (15.37 ± 0.81) and 40–70 Hz (19.93 ± 0.81) activity powers were enhanced in lesioned rats as compared to control rats (9.17 ± 0.69 and 12.54 ± 1.32, respectively, p b 0.01) (Fig. 6D), and the power of GPe LFPs in the DA neuron lesioned rats and control rats also had significant differences in several frequency bands (1–12 Hz, 12–25 Hz, 25–40 Hz and 40– 70 Hz). Post-hoc comparison showed that 1–12 Hz activity power was reduced (54.54 ± 3.91 and 70.10 ± 2.38, p b 0.01), while 12–25 Hz (18.69 ± 1.40), 25–40 Hz (14.10 ± 2.44) and 40–70 Hz (6.44 ± 0.93) powers were enhanced in lesioned rats as compared to control rats (14.13 ± 1.02, 6.06 ± 0.60, and 3.61 ± 0.41, respectively, p b 0.05– 0.01,) (Fig. 6E). There were no significant changes in GPi LFP power in 1–12 Hz, 12–25 Hz, 40–70 Hz and 70–100 Hz between lesioned rats
3.4. Effects of DA lesions on the relationship of LFP activity between M1 and GPe or GPi To determine whether changes of LFP activity observed in M1, GPe, and GPi following DA loss were correlated with alterations in their estimated time lag between putatively connected sites, the temporal relationship of LFP activity between M1 and GPe or GPi was assessed by cross-correlation and time delay value. The cross-correlation function provides a qualitative comparison of cross-correlation and time delay value between M1 and GPe or GPi by comparing the pair recording different frequency segments, which are 1–12 Hz, 12–25 Hz, 25–40 Hz, 40–70 Hz and 70–100 Hz ranges, respectively. With the MATLAB function xcorr, the qualitative comparison in different frequency segments evolves by analyzing time varying signals for identifying fundamental fluctuation frequencies in neural timeseries data, without any a priori filtering [31,32]. For this analysis, a positive lag indicates that the peak of the LFP oscillation in M1 is occurring before the peak of the LFP oscillation in GPe or GPi and a negative lag indicates the reverse. Results from these cross-correlation analyses showed that M1 LFP activity was more correlated with GPe LFPs in 25–40 Hz in the lesioned rats than in the control rats. The lag time of M1 with GPe from lesioned rats' LFPs in 25–40 Hz decreased (mean lag: 0.0026 ± 0.0002 s, from 4 rats) compared with those of control rats (mean lag: 0.8918 ± 0.7161 s, p b 0.01, Wilcoxon rank-sum test, from 4 rats). The cross-correlation value of M1 with GPe from lesioned rats' LFPs in 25–40 Hz also significantly increased (0.9872 ± 0.0128, from 4 rats) compared with those of control rats (0.2856 ± 0.0767, p b 0.01, Wilcoxon rank-sum test, from 4 rats) (Fig. 7). Finally, a lag in M1 with GPe was only found in 25–40 Hz, but not in 1–12 Hz, 12–25 Hz, 40–70 Hz and 70–100 Hz ranges (data not shown). No significant difference of lag time was observed in M1 with GPi LFPs (data not shown). Collectively, these results showed an effect of DA loss on lag time in LFPs in 25–40 Hz oscillations in M1 with GPe. Taken together, the near-zero lag time and near one cross-correlation value indicated that LFP activity in the two brain regions is relatively synchronous. 4. Discussion The present study showed the following: First, DA loss increased LFP spectral power in M1, GPe and GPi in the 12–70 Hz range and decreased LFP spectral power in M1 and GPe in 1–12 Hz. Second, these LFP activity changes were accompanied with marked increases in synchronization of oscillatory activities in M1 with GPe. Third, while the lesioned rats had decreased LFP power in 1–12 Hz of M1 and GPe compared to the control rats during ladder walking, the LFP power in GPi in 1–12 Hz
236 M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
Fig. 4. Representative examples of LFP recording from paired electrodes in M1 with GPe or GPi during ladder walking from a DA neuron lesioned rat (A and C) and an intact rat (B and D).
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240 Fig. 5. Representative examples of time–frequency spectrograms of LFP power in M1, GPe, and GPi during ladder walking from a DA neuron lesioned rat (A, C, E) and a control rat (B, D, F). Spectral power was plotted on a logarithmic scale with greater power represented by redder colors. LFP spectra power in M1, GPe, and GPi from the lesioned rat was greater than that from the control rat.
237
238
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
239
Fig. 7. Comparison of lag time and correlation value between M1 and GPe in the DA neuron lesioned rats and the control rats. Bar graphs represent the lag time in the lesioned rats (n = 4) that was significantly lower than that in the control rats (n = 4; *p b 0.05, **p b 0.01) (A); correlation value between M1 and GPe in lesioned rats (n = 4) was significantly higher than that in control rats (n = 4; *p b 0.05, **p b 0.01) (B).
showed no alteration compared to the control rats. We suggest that these observations in the alteration of LFP activities in M1, GPe and GPi and synchronization between M1 and GPe in DA depleted rats play a role in the motor impairment that is shown in animal models of PD patients. These results in aberrant LFP activity are consequences of the degeneration of dopamine neurons in the substantia nigra and possibly due to their close interconnectivity between GPe and GPi. The results also suggest a channel of communication through which the cortex interacts with the basal ganglia and that is susceptible to dysfunction following dopamine depletion. The identification of abnormalities in LFPs in the basal ganglia and related cortex in movement disorders remains a highly important field of research, because such findings may eventually help us to a better understanding of the involvement of the basal ganglia in parkinsonism. Studies have shown that excessive LFPs activity at beta frequencies (10–35 Hz) is a key pathophysiological feature of these parkinsonian circuits. Our results showed that the fractional spectral power in the 12–70 Hz bands was higher on average in the parkinsonian condition for M1 (25–70 Hz), GPe (12–70 Hz) and GPi (25–40 Hz), agreeing in part with previous findings of prominent oscillations in the beta frequency range in the parkinsonian state [8,9,33]. In addition to the beta frequency activity, the abnormalities in LFPs in two other bands have been found, that is, decreased LFP spectral power in low-frequency band activity (1–12 Hz) and increased LFP spectral power in high frequency band activity (N 35 Hz). One explanation of these differences is the possibility that they are related to the brain state and behavioral context (anesthesia vs awake and performing in patients or animals). Our study used awake, behaving rats while other studies used anesthetic or awake but immobile animals. There are recent reports showing that increased γ-band in 35–100 Hz power depends on arousal and is coincident with motor tasks [10,21]. The abnormalities in more than beta frequency in LFPs in M1 and GPe in PD rats have been identified, providing additional material to consider in the complex model of basal ganglia pathophysiology. More recent research has highlighted the oscillatory nature of excessive neuronal synchronization in the parkinsonian state. Data from LFP amplitude waveforms by cross-correlation analyses showed that the lag time for M1 with GPe changed from a few hundred milliseconds in control rats to near zero in DA lesioned rats, and the cross-correlation value also changed from decimal fraction in control rats to nearly one in DA lesioned rats. The results indicate that LFP activities in the two
brain regions become relatively synchronous and suggest stronger functional connectivity following dopaminergic denervation. This is consistent with the large body of literature reporting synchronous oscillations in parkinsonian patients as well as animal models of the disease [14,25,34]. It also suggests that the cortical rhythms are being more effectively transmitted to downstream sites after loss of dopamine, supporting the theory of hypersynchronization in corticobasal ganglia circuits [11,21,22]. Further inspection along the M1 with GPe temporal relationship revealed that the near zero lag time between M1 and GPe was present in a specific frequency of 25–40 Hz. While these observations provide a clear biological marker for PD, it is not clear whether synchronization is an epiphenomenon or a true pathogenic alteration and how the dominant frequency becomes focused in the 25–40 Hz range in PD. The shortening of the latency between M1 and GPe at a specific frequency range suggests that there might be some special mechanism that facilitates synchronization and/or transfer of relevant information across distant brain regions. First, it may be possible to interpret these in the context of the death of dopaminergic neurons in the basal ganglia, because a dopamine decrease should strengthen the connections between neurons of the cortex–basal ganglia network. This moves the network toward a more synchronized state and the transient synchronous patterns become more prevalent [34,35]. Second, it is also possible that the excessive synchronization is a consequence of the integration of many changes in the basal ganglia, thalamus, or cortex that are known to accompany parkinsonism in human and animals [23,36]. Therefore, it is possible that 25–40 Hz is a frequency range compatible with resonant activity in the dopamine-depleted basal ganglia–thalamocortical circuit. Taken together, the observed synchronous changes are associated with, but are not causal of, parkinsonism, and that they may not be primarily dopaminergic in origin. Our data showed that DA loss is associated dramatically with increased LFP power in M1 and GPe, which was selective emergence in 25–70 Hz and 12–70 Hz for M1 and GPe, respectively. However, GPi only showed a modest increase of LFP power in the 25–40 Hz range after DA loss. Overall, it is particularly interesting that parkinsonism appears to be more strongly associated with detectable changes in oscillatory power in GPe and M1, than with changes in the GPi. The dramatically different patterns of frequency power observed in M1, GPe and GPi in DA neuron lesioned versus control rats may imply a different mechanism underlying the effect of DA loss. It may be possible to interpret these findings in the context of abnormal frequency band reactivity in M1, GPe and
Fig. 6. LFP power spectra of M1, GPe, and GPi during rats walking along a ladder. Welch methods as spectral estimation in power spectra showed an example of LFP activities in a control (gray line) and a lesioned (black line) rat (A, B, C). Bar graphs in the right represent mean relative LFP power in M1, GPe and GPi within a series of frequency ranges: 1–12 Hz, 12–25 Hz, 25–40 Hz, 40–70 Hz and 70–100 Hz in control rats (n = 8) and in DA neuron lesioned rats (n = 8) (D, E, F). The M1 LFP power was significantly lower in 1–12 Hz and significantly increased in 25–40 Hz and 40–70 Hz in the lesioned rats than in control rats (D). The GPe LFP power showed the same tendency to that of M1 (E), while GPi LFP power only slightly increased in 25–40 Hz band in the lesioned rats compared to the control rats (F).
240
M. Wang et al. / Journal of the Neurological Sciences 348 (2015) 231–240
GPi in lesioned rats, for that though GPe and GPi receive similar inputs from the cerebral cortex, they project to different brain sites and thus have different functional roles in basal ganglia information processing [37], so the DA loss in association with the LFP activities in different frequency bands and different brain regions may show different processing patterns in the basal ganglia and the structures they are connected with. Therefore, it lends support to the general hypothesis that special frequency power alteration exerts a dysfunctional effect on motor circuits modulated by the basal ganglia. The data suggest that M1, GPe and GPi form distinct neuronal networks for which oscillatory activity may have different neurophysiological roles. This also may be possible to interpret that the LFP activity of M1 with GPe in lesioned rats becomes relatively synchronous, but was not observed in M1 with GPi in lesioned rats. The observation is consistent with the hypothesis that cortical rhythms entrain GPe activity more readily than they do to GPi activity after DA loss. One possibility is that unlike M1 and GPe, the relative position of GPi in the basal ganglia circuits as a common output for both the indirect and direct pathways may in part account for the lack of modulation after DA loss. This also may be the reason that GPi is a common target for neuromodulation in the treatment of PD. In PD patients, the neuronal circuits in GPi might partially or fully retain their ability to oscillate in a movement dependent manner [4,5,37–39]. In summary, the results showed that specific frequency power in the motor cortex, GPe and GPi varied dramatically with movement deficits in the DA neuron lesioned rats during locomotor behavior. The results also suggest that DA loss promotes synchronization activity between the motor cortex and GPe. Determining the relevance of such specific frequency and synchrony to behavior and the functional significance of excessive synchronization is a more challenging task. Hopefully, further exploration of the relationships between the processes causing excessive synchronization of basal ganglia activity across specific frequency will provide additional insights into fundamental mechanisms underlying the profound behavioral consequences of DA loss. References [1] Wichmann T, DeLong MR. Deep-brain stimulation for basal ganglia disorders. Basal Ganglia 2011;1:65–77. [2] Graybiel AM. Building action repertoires: memory and learning functions of the basal ganglia. Curr Opin Neurobiol 1995;5:733–41. [3] Hammond C, Bergman H, Brown P. Pathological synchronization in Parkinson's disease: networks, models and treatments. Trends Neurosci 2007;30:357–64. [4] Priori A, Foffani G, Pesenti A, Bianchi A, Chiesa V, Baselli G, et al. Movement-related modulation of neural activity in human basal ganglia and its L-DOPA dependency: recordings from deep brain stimulation electrodes in patients with Parkinson's disease. Neurol Sci 2002;23:s101–2. [5] Silberstein P, Kühn AA, Kupsch A, Trottenberg T, Krauss JK, Wöhrle JC, et al. Patterning of globus pallidus local field potentials differs between Parkinson's disease and dystonia. Brain 2003;126:2597–608. [6] Devergnas A, Pittard D, Bliwise D, Wichmann T. Relationship between oscillatory activity in the cortico-basal ganglia network and parkinsonism in MPTP-treated monkeys. Neurobiol Dis 2014;68:156–66. [7] Dejean C, Gross CE, Bioulac B, Boraud T. Dynamic changes in the cortex–basal ganglia network after dopamine depletion in the rat. Am Physiol Soc 2008:385–96. [8] Ellens DJ, Leventhal DK. Review: electrophysiology of basal ganglia and cortex in models of Parkinson disease. J Park Dis 2013;3:241–54. [9] Stein E, Bar-Gad I. Beta oscillations in the cortico-basal ganglia loop during parkinsonism. Exp Neurol 2013;245:52–9. [10] Van Der Meer MA, Kalenscher T, Lansink CS, Pennartz CM, Berke JD, Redish AD. Integrating early results on ventral striatal gamma oscillations in the rat. Front Neurosci 2010;4. [11] Mallet N, Pogosyan A, Sharott A, Csicsvari J, Bolam JP, Brown P, et al. Disrupted dopamine transmission and the emergence of exaggerated beta oscillations in subthalamic nucleus and cerebral cortex. Soc Neurosci 2008:4795–806. [12] Walters JR, Bergstrom DA. Basal ganglia network synchronization in animal models of Parkinson's disease. Springer; 2009 117–42.
[13] Walters JR, Tierney PL, Bergstrom DA. Oscillatory activity and synchronization in the basal ganglia network in rodent models of Parkinson's disease. Springer; 2009 443–59. [14] Delaville C, Cruz AV, McCoy AJ, Brazhnik E, Avila I, Novikov N, et al. Oscillatory activity in basal ganglia and motor cortex in an awake behaving rodent model of Parkinson's disease. Basal Ganglia 2014;3:221–7. [15] Magill PJ, Bolam JP, Bevan MD. Relationship of activity in the subthalamic nucleus– globus pallidus network to cortical electroencephalogram. J Neurosci 2000;20: 820–33. [16] Tseng KY, Kasanetz F, Kargieman L, Riquelme LA, Murer MG. Cortical slow oscillatory activity is reflected in the membrane potential and spike trains of striatal neurons in rats with chronic nigrostriatal lesions. J Neurosci 2001;21:6430–9. [17] Aravamuthan BR, Bergstrom DA, French RA, Taylor JJ, Parr-Brownlie LC, Walters JR. Altered neuronal activity relationships between the pedunculopontine nucleus and motor cortex in a rodent model of Parkinson's disease. Elsevier; 2008 268–80. [18] de Hemptinne C, Ryapolova-Webb ES, Air EL, Garcia PA, Miller KJ, Ojemann JG, et al. Exaggerated phase–amplitude coupling in the primary motor cortex in Parkinson disease. Proc Natl Acad Sci 2013;110:4780–5. [19] Kita H, Kita T. Cortical stimulation evokes abnormal responses in the dopaminedepleted rat basal ganglia. J Neurosci 2011;31:10311–22. [20] Belluscio MA, Riquelme LA, Murer MG. Striatal dysfunction increases basal ganglia output during motor cortex activation in parkinsonian rats. Wiley Online Library; 2007 2791–804. [21] Jenkinson N, Kühn AA, Brown P. Gamma oscillations in the human basal ganglia. Exp Neurol 2013;245:72–6. [22] Degos B, Deniau JM, Chavez M, Maurice N. Chronic but not acute dopaminergic transmission interruption promotes a progressive increase in cortical beta frequency synchronization: relationships to vigilance state and akinesia. Cereb Cortex 2009; 19:1616–30. [23] Brazhnik E, Cruz AV, Avila I, Wahba MI, Novikov N, Ilieva NM, et al. State-dependent spike and local field synchronization between motor cortex and substantia nigra in hemiparkinsonian rats. J Neurosci 2012;32:7869–80. [24] Nambu A. Globus pallidus internal segment. In: James M, Tepper EDA, Bolam JP, editors. Progress in brain research. Elsevier; 2007. p. 135–50. [25] Avila I, Parr-Brownlie LC, Brazhnik E, Castañeda E, Bergstrom DA, Walters JR. Beta frequency synchronization in basal ganglia output during rest and walk in a hemiparkinsonian rat. Exp Neurol 2010;221:307–19. [26] Jadavji N, Metz G. Both pre- and post-lesion experiential therapy is beneficial in 6hydroxydopamine dopamine-depleted female rats. Neuroscience 2009;158:373–86. [27] Vergara-Aragon P, Gonzalez CL, Whishaw IQ. A novel skilled-reaching impairment in paw supination on the “good” side of the hemi-Parkinson rat improved with rehabilitation. J Neurosci 2003;23:579–86. [28] Ungerstedt U, Arbuthnott GW. Quantitative recording of rotational behavior in rats after 6-hydroxy-dopamine lesions of the nigrostriatal dopamine system. Brain Res 1970;24:485–93. [29] Kubie JL. A driveable bundle of microwires for collecting single-unit data from freely-moving rats. Physiol Behav 1984;32:115–8. [30] Nicolelis MA, Ghazanfar AA, Faggin BM, Votaw S, Oliveira LM. Reconstructing the engram: simultaneous, multisite, many single neuron recordings. Neuron 1997;18: 529–37. [31] Masimore B, Kakalios J, Redish A. Measuring fundamental frequencies in local field potentials. J Neurosci Methods 2004;138:97–105. [32] Sharott A, Moll CK, Engler G, Denker M, Grün S, Engel AK. Different subtypes of striatal neurons are selectively modulated by cortical oscillations. J Neurosci 2009; 29:4571–85. [33] Moran RJ, Mallet N, Litvak V, Dolan RJ, Magill PJ, Friston KJ, et al. Alterations in brain connectivity underlying beta oscillations in parkinsonism. PLoS Comput Biol 2011;7: e1002124. [34] Park C, Worth RM, Rubchinsky LL. Fine temporal structure of beta oscillations synchronization in subthalamic nucleus in Parkinson's disease. J Neurophysiol 2010;103:2707–16. [35] Park C, Worth RM, Rubchinsky LL. Neural dynamics in parkinsonian brain: the boundary between synchronized and nonsynchronized dynamics. Phys Rev E 2011;83: 042901. [36] Brazhnik E, Novikov N, McCoy AJ, Cruz AV, Walters JR. Functional correlates of exaggerated oscillatory activity in basal ganglia output in hemiparkinsonian rats. Exp Neurol 2014;261:563–77. [37] Jaeger D, Kita H. Functional connectivity and integrative properties of globus pallidus neurons. Neuroscience 2011;198:44–53. [38] Brown P. Oscillatory nature of human basal ganglia activity: relationship to the pathophysiology of Parkinson's disease. Mov Disord 2003;18:357–63. [39] Tsiokos C, Hu X, Pouratian N. 200–300 Hz movement modulated oscillations in the internal globus pallidus of patients with Parkinson's disease. Neurobiol Dis 2013; 54:464–74. [40] Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. 4th edn. San Diego: Academic; 1998.