| BACKGROUND:Parkinson’s disease is a progressive neurodegenerative disorder that mainly affects persons older than 60 years. The main pathological feature of PD is dopaminergic neurons degeneration in the substantia nigra pars compacta (SNc) of midbrain, which leads to motor manifestations including resting tremor, rigidity and bradykinesia. Levodopa is the most effective medication for managing motor symptoms in PD. But long-term use is associated with severe adverse effects and motor complications, such as dyskinesia. As a result, surgical interventions are alternatives for the supplemental treatment, and current practice is to do deep brain stimulation (DBS) when surgery is needed. Unilateral or bilateral DBS of the subthalamic nucleus is effective for alleviating tremor, bradykinesia and rigidity and can reduce the dose of levodopa. When PD progresses to an advanced stage, however, more than half of the patients can seriously influenced by axial symptoms or postural instability and gait disorder (PIGD), such as frequent falls, festinating gait and freezing of gait (FOG). Along with the progression of the disease, axial symptoms will become increasingly worse. Unfortunately, these symptoms respond poorly to dopaminergic or surgical treatment, which have become tough challenges in the treatment of patients in advanced stages and result in disability.In recent years, there is growing evidence that the pedunculopontine nucleus (PPN) plays a vital role in the occurrence of axial symptoms that can be alleviated by PPN-DBS. Therefore, the PPN has been investigated as a potential target for deep brain stimulation to treat PIGD in PD. Research on the functions of the PPN and the dysfunction of it in PD will contribute to the understanding of the pathological substrates of PIGD and the mechanism of PPN-DBS.The method of electrophysiology is valuable to investigate the pathophysiology of PD and the mechanism of DBS. In recent years, the use of local field potential (LFP) recordings has made a resurgence in the investigation of the pathophysiology of Parkinsonian and other movement disorders. LFP, similar to EEG, is a kind of oscillatory activities, which is classified by the oscillatory frequency. Generally, LFP is divided into five bands:delta band (1-3 Hz), theta band (4-7 Hz), alpha band (8-13 Hz), beta band (14-30 Hz), and gamma band (>30 Hz). Several bands of oscillatory activities have been detected in the cortico-basal ganglia circuits. Among these, aberrant beta band oscillations have been deemed to be one of the most important findings in the brain of PD patients. These beta band abnormalities are believed to have anti-kinetic effects, and are likely to be responsible for some PD symptoms.Recently, a number of researchers have used LFP to study oscillatory activity in the PPN of parkinsonian patients after electrode implantation surgery, which has made important findings:(1) alpha band oscillations may play a physiological function in the PPN, and is pathologically attenuated in PD; (2) Unlike findings in the cortical-basal ganglia loop, there currently exists no consensus as to the importance of beta band PPN activity in PD. Some study implied that this activity may contribute to akinesia in PD, but other studies have found quite the contrary. (3) Theta and gamma band oscillations have also been recorded in the PPN. Theta oscillations may be involved in the feedback of sensory information between the PPN and the sensorimotor cortices and gamma oscillations might be similar to beta oscillations. All the above findings need further investigation. However, clinical studies are limited by the lack of LFP data from healthy controls and the small number of PPN-DBS cases, but animal studies do not have this limitation. We can use animals to record LFP in the PPN and other structures in the cortico-basal ganglia circuits, and perform deeper analysis and more mechanistic examination to supplement human studies.OBJECT:The present study had three aims:1) to detect oscillatory activity in the PPN of parkinsonian rats; 2) to examine the relationship between oscillatory activity in cortico-basal ganglia circuits and the PPN; and 3) to investigate the effect of dopaminergic treatment on oscillatory activity in these structures. We simultaneously recorded LFPs from the PPN, primary motor cortex (Ml), and STN in 6-OHDA-induced hemiparkinsonian rats by using a 128-channel data acquisition system (Cerebus, Blackrock Microsystems, Salt Lake City, USA), and analyzed the connectivity of these structures.METHODS:1. Experimental groupsThis experiment used 30 healthy adult male Sprague-Dawley rats, weighing 290-310 g at the time of surgery. Rats were randomly assigned to three groups:(1) sham lesion group (n = 10), with 4μl saline vehicle injected into the right medial forebrain bundle (MFB); (2) unilateral 6-OHDA lesion group (n= 13), with 4μl 6-OHDA (3μg/μl) injected into the right MFB; (3) 6-OHDA lesion+levodopa group (n = 7), with 4μl 6-OHDA (3μg/μl) injected into the right MFB, and received the treatment of levodopa.2. Experimental procedure(1) All rats were allowed to habituate to the environment for one week and trained to walk on a rotarod at a speed of 6 revolutions/min across 3 sessions/day during the week prior to surgery. In each training session, rats were trained to walk for 5 min followed by a 5 min rest period, during which they remained standing on the rotarod. (2) All rats underwent microelectrode implantation. At the same time, the 6-OHDA lesion group and 6-OHDA lesion+levodopa group received the injection of 6-OHDA and the sham lesion group received the same volume of saline. (3) One week after surgery, rats were retrained to walk on the rotarod. (4) Two to three weeks after surgery, unilateral dopamine depletion was assessed by the rotational response to apomorphine (APO). (5) Three weeks after surgery, electrophysiological recordings were performed during resting and walking on the rotarod. (6) After the final recording session, rats were sacrificed by cardiac perfusion followed by histology.3. SurgeryRats were anesthetized with sodium pentobarbital, and then placed in a stereotaxic apparatus. Animals received penicillin before surgery to prevent infection. Unilateral 6-OHDA lesions were created via injection of 4 μl of 6-OHDA (3 μg/μl free base dissolved in a solution of 0.2 mg/ml L-ascorbic acid in 0.9% w/v saline at a rate of 0.5 μl/min; Sigma) into the right MFB at the following coordinates relative to bregma:-1.8 mm antero-posterior (AP),2.0 mm medio-lateral (ML), and-8.3 mm dorso-ventral (DV, below the skull surface) (Paxinos & Watson,2007). Sham-lesioned rats received the same volume of ascorbic saline into the right MFB. Three arrays of eight stainless steel Teflon-insulated microwires (50 μm diameter, arranged in a 3x3x2 configuration) were slowly lowered into the right M1,STN, and PPN at the following coordinates:M1:+1.0 mm AP,2.3 mm ML, and-2.0 mm DV; STN:-3.5 mm AP,2.5 mm ML, and-8.2 mm DV; PPN:-7.8 mm AP,2.0 mm ML, and-7.2 mm DV. The microelectrode arrays were secured to the cranium with stainless steel screws and dental cement. The screw located above the cerebellum served as a ground; it was placed in contact with the dura, and ground wires were wrapped around it.4. Electrophysiological recordingThree weeks post-lesion, LFPs were detected by microwire arrays and passed from the headset assemblies to a preamplifier via a light-weight cable and a rotating commutator. LFPs recordings were performed during 6 sessions consisting of 5 min of resting and 5 min of walking on the rotarod at 6 rpm. Video records synchronized with data recording were used to monitor and record behavior. LFPs were collected at a sampling frequency of 10 KHz, amplified (300x), and band pass filtered (0.3-500 Hz). A ground wire was used for reference. After the recordings of all rats, the rats of 6-OHDA lesion+levodopa group received a low dose of levodopa (5 mg/kg with 15 mg/kg of benserazide, i.p., Sigma-Aldridge) and followed by 6 sessions of recordings (started at 20 min after administration of levodopa).5. HistologyAfter the final recording session, rats were overdosed with chloralic hydras, and 20-3OμA of positive current was passed through the recording electrodes for 20s to deposit iron ions. Animals were then sacrificed by cardiac perfusion with 300 ml saline (0.9% w/v) at room temperature, followed by 400 ml ice-cold paraformaldehyde and potassium ferricyanide solution (4% and 5% w/v in 0.1 M phosphate buffered saline, respectively). Brains were extracted, post-fixed for 24 h in 4% paraformaldehyde, and immersed in sucrose (25% w/v in 0.1 M phosphate buffered saline and 30% successively) before being sectioned. Coronal sections were cut through the Ml, STN, and PPN at 40μm and through the SNc at 25 μm. To verify recording sites, sections including electrode tracts were stained with cresyl violet. Immunohistochemical staining for tyrosine hydroxylase (TH) in the SNc was performed to verify lesion size.6. LFP data analysisLFP data collected during the resting state were compared with synchronously recorded video, and data were excluded during grooming and other stereotyped behavior. Data from animals with misplaced electrodes were also excluded. Power spectral density, coherence and Granger causality during resting and walking were calculated based on 120-s epochs free of major artifacts. For each rat in each behavioral state, data were selected from 1-2 microwires per brain area.(1) Power spectral densityThe power spectral density was computed using the Matlab toolbox Chronux (http://chronux.org), using multi-taper spectral estimation. LFP data were resampled to a sampling frequency of 1,000 Hz, and 120-s epochs were broken into 120 segments of 1 s each. Each segment served as a computing unit, and results were averaged. Slow fluctuations in electrophysiological signals and 50 Hz line noise were removed first. Parameter values for calculations were set as follows:params.fpass= [0100] (frequency range of interest was 0-100 Hz); params.tapers= [35] (best degree of spectral smoothing at time-bandwidth product of 3 and 5 tapers). The frequency resolution was 1 Hz. Power spectral densities of LFPs were computed for the M1, STN, and PPN during resting and walking.(2) CoherenceCoherence was used to assess functional relationships between different brain regions. As a standard measure of the linear correlation between two signals across frequencies, values of coherence range from 0 (no linear association) to 1 (perfect linear association). Coherences were calculated using Chronux with params.fpass= [0 100]; params.tapers= [7 13] (with the best degree of smoothing). The frequency resolution was 1 Hz. Coherences were calculated between each pair of the three brain regions (M1, STN, and PPN) during resting and walking.(3) Granger causality analysisTo further examine the causal interactions of the M1, STN, and PPN, we estimated partial Granger causality in the frequency domain. Granger causality is commonly used to determine whether one time series is useful in forecasting another. If past information of an observed time series x(n) can significantly improve prediction of another series y(n), it can be stated that signal x(n) "Granger-causes" signal y(n). Therefore, Granger causality gives insight into the direction of information flow or the driving-response relationship between structures. When the number of nodes or recording sites is greater than two, partial Granger causality is calculated. In this case, when analyzing Granger causality between X1 and X2, all other nodes (X3...Xn) should be simultaneously taken into account. In our data analysis, partial Granger causality was computed based on multivariate autoregressive modeling. The first step was model order estimation, and the best model order was chosen by Akaike Information Criterion (AIC). Next, Vector AutoRegressive (VAR) model estimation using the selected model order was conducted using the Levinson, Wiggins and Robinson (LWR) algorithm, then the autocovariance sequence was calculated according to the VAR model. Finally, partial Granger causality in the frequency domain was calculated using the autocovariance sequence.7. StatisticsEffects of the 6-OHDA lesion on power and coherence were evaluated statistically across all frequency bands with cluster-based, nonparametric permutation t-tests (n= 1,000 permutations, corrected for multiple comparisons). Differences in power, coherence and Granger causality over the beta frequency band among sham lesion group,6-OHDA lesion group and levodopa treatment group were analyzed using the Kruskal-Wallis test for equal medians with Dunn’s Multiple Comparison Test for post hoc test. The criterion of significance was p< 0.05.RESULTS:1. Changes in LFP power after dopamine lesion during resting and walkingDuring, walking, LFP power in the beta band was significantly increased in all three regions in the 6-OHDA lesion group compared with the sham lesion group. However, regional variation was observed for which frequency ranges within the beta band showed significant increases. Specifically, frequencies that differed between 6-OHDA and sham lesion rats were:15-24 Hz in M1 (P= 0.042),13-28 Hz in STN (P= 0.014), and 14-22 Hz in PPN (P= 0.035). During resting, there was no difference in power in the STN or PPN between 6-OHDA and sham lesion rats (P> 0.05). However, a significant increase was observed between 21-35 Hz in M1 (P= 0.042) of 6-OHDA lesion rats.2. Effects of dopamine lesion on coherence during resting and walkingTo observe linear correlations between different brain structures, we estimated coherences between pairs of structures during resting and walking. During walking, 6-OHDA lesion rats showed increased coherence compared to sham lesion rats in the 10-34 Hz range for Ml-STN (P= 0.004),11-36 Hz for Ml-PPN (P= 0.006), and 14-28 Hz for STN-PPN (P= 0.01). During resting, there were no significant differences in coherence between 6-OHDA and sham lesion rats (P> 0.05).3. Effect of levodopa treatment on power and coherence during resting and walkingAfter parkinsonian rats received levodopa treatment, the augmented beta band power and coherence were significantly reduced to the normal levels (P< 0.05).4. Granger causal analysis on beta band oscillationsAccording to Granger causal analysis, beta causality of 6-OHDA lesion group was significantly higher in STNâ†'M1 (P< 0.001), STNâ†'PPN (P< 0.001), and PPN â†'M1 (P< 0.001) during walking compared with the other two groups. There was no group difference in beta causality for M1â†'STN, PPNâ†'STN, or M1â†'PPN (P> 0.05). These results indicate that the direction of information flow for beta band oscillations was STNâ†'M1, STNâ†'PPN, and PPNâ†'M1. And after parkinsonian rats received dopaminergic treatment, beta causality in STNâ†'M1, STNâ†'PPN and PPN â†'M1 during walking returned to normal levelsCONCLUSION:1. We recorded enhanced beta oscillations in M1, STN, and PPN during walking in rats with 6-OHDA lesions. This is the first study to detect enhanced beta rhythms in the PPN of parkinsonian rats. Our results suggest that beta activity in the PPN is transmitted from the basal ganglia and probably comes from the STN.2. Based on Granger causality analysis, we described the network of causal interactions for beta band in the PPN, Ml, and STN. The STN plays a dominant role in this network and may be the potential source of aberrant beta band oscillations in PD.3. Although levodopa can inhibit beta activity in the PPN of parkinsonian rats, it cannot relieve parkinsonian patients’axial symptoms in clinical practice. Hence, beta oscillations may not be the major cause of these symptoms. |
PDF Full Text Request