Acute and Chronic Dopaminergic Depletion Differently Affect Motor Thalamic Function.

The motor thalamus (MTh) plays a crucial role in the basal ganglia (BG)-cortical loop in motor information codification. Despite this, there is limited evidence of MTh functionality in normal and Parkinsonian conditions. To shed light on the functional properties of the MTh, we examined the effects of acute and chronic dopamine (DA) depletion on the neuronal firing of MTh neurons, cortical/MTh interplay and MTh extracellular concentrations of glutamate (GLU) and gamma-aminobutyric acid (GABA) in two states of DA depletion: acute depletion induced by the tetrodotoxin (TTX) and chronic denervation obtained by 6-hydroxydopamine (6-OHDA), both infused into the medial forebrain bundle (MFB) in anesthetized rats. The acute TTX DA depletion caused a clear-cut reduction in MTh neuronal activity without changes in burst content, whereas the chronic 6-OHDA depletion did not modify the firing rate but increased the burst firing. The phase correlation analysis underscored that the 6-OHDA chronic DA depletion affected the MTh-cortical activity coupling compared to the acute TTX-induced DA depletion state. The TTX acute DA depletion caused a clear-cut increase of the MTh GABA concentration and no change of GLU levels. On the other hand, the 6-OHDA-induced chronic DA depletion led to a significant reduction of local GABA and an increase of GLU levels in the MTh. These data show that MTh is affected by DA depletion and support the hypothesis that a rebalancing of MTh in the chronic condition counterbalances the profound alteration arising after acute DA depletion state.


Introduction
The motor thalamus (MTh) plays a pivotal role in motor information processing in both physiological and pathological conditions, being directly connected to both the basal ganglia (BG) and the cortex [1]. In

Effects of DA Depletion on the Cortical-Thalamic Coupling
We compared the preferred phase angles and the vector lengths in control, acute and chronic DA depletion states. In control rats, the MTh neurons fired preferentially at 42.77 ± 12.36 • (n = 21, Figure 2e), with a vector length of 0.73 ± 0.06. In acute DA-depleted animals (n = 34), the phase was 22.54 ± 6.96 • with a vector length of 0.83 ± 0.03, while in the chronic DA depletion state (n = 22), it was 79.89 ± 15.98 • with a vector length of 0.56 ± 0.07. The vector length, i.e., the strength of corticothalamic coupling, was significantly lower in chronic DA depletion state than in control and in acute DA depletion state (p < 0.05). The phase angle was significantly lower in control and acute DA depletion state than that in chronic DA depletion state (p < 0.05) (Figure 2e).

Effects of DA Depletion on the Extracellular Levels of GABA and GLU in the Motor Thalamus
ANOVA analysis followed by Dunnet's multiple comparison test showed a significant effect of TTX infused in the medial forebrain bundle induced an increase in the endogenous GABA extracellular level (215 ± 67.5; n = 14; p < 0.05) compared to the control (100 ± 7.90, n = 14) and post-infusion period (159.2 ± 50.2, n = 14) (Figure 3a). No changes in the GLU levels were observed in the MTh after the infusion of TTX (PRE: 100 ± 2.69; TTX: 122.7 ± 24.8, n = 7; POST: 123.7 ± 53.9, n = 14) (Figure 3b). The extracellular GABA levels in the MTh were significantly lower in 6-OHDA animals than in sham-lesioned animals (100 ± 11.09 vs. 43.54 ± 6.65, n = 9; ** p < 0.01) (Figure 3c). GLU levels were significantly augmented in the MTh of 6-OHDA rats when compared to those measured in sham-lesioned rats by unpaired student t-test (100 ± 13.75 vs. 276.69 ± 43.52, n = 9; ** p < 0.01) (Figure 3d) levels. . TTX infused in the medial forebrain bundle induced an increase in the endogenous GABA extracellular levels (A) while did not change the GLU levels in the MTh. Data in (A) and (B) are expressed as mean ± SEM normalized to the baseline levels (pre TTX infusion) from n = 14 experiments per group; * p < 0.05. One-and two-way ANOVA for repeated measures, followed by Dunnet's multiple comparison test. 6-OHDA-lesion infused in the medial forebrain bundle produced, after 2 weeks, a decrease in GABA (C) and an increase in GLU extracellular levels in the MTh (D). Data in (C) and (D) are expressed as means ± SEMs normalized to sham-lesioned (n = 9) neurotransmitter levels from n = 9 experiments per group; ** p < 0.01, unpaired student t-test.

Discussion
We demonstrated that MTh is deeply affected by both acute and chronic DA depletion both in terms of electrophysiological and neurochemical alterations. The main findings of the present study are threefold. Firstly, the results emphasize the different effects of DA deafferentation on MTh, with a decrease in firing rate in TTX-treated animals and an increase of burst firing in 6-OHDA-lesioned rats. Secondly, the results show that DA depletions change, in an opposite fashion, MTh-cortical coupling with the MTh neurons fired preferentially at the beginning of the UP-state in TTX-treated rats compared to the 6-OHDA-lesioned rats, which showed a shift towards the peak of the SWA. Finally, the neurochemical results demonstrated that the extracellular levels of GABA increase and those of GLU do not change in the MTh of TTX-treated animals; in the 6-OHDA-lesioned rats, GABA levels decreased and those of GLU increased. One-and two-way ANOVA for repeated measures, followed by Dunnet's multiple comparison test. 6-OHDA-lesion infused in the medial forebrain bundle produced, after 2 weeks, a decrease in GABA (C) and an increase in GLU extracellular levels in the MTh (D). Data in (C) and (D) are expressed as means ± SEMs normalized to sham-lesioned (n = 9) neurotransmitter levels from n = 9 experiments per group; ** p < 0.01, unpaired student t-test.

Discussion
We demonstrated that MTh is deeply affected by both acute and chronic DA depletion both in terms of electrophysiological and neurochemical alterations. The main findings of the present study are threefold. Firstly, the results emphasize the different effects of DA deafferentation on MTh, with a decrease in firing rate in TTX-treated animals and an increase of burst firing in 6-OHDA-lesioned rats. Secondly, the results show that DA depletions change, in an opposite fashion, MTh-cortical coupling with the MTh neurons fired preferentially at the beginning of the UP-state in TTX-treated rats compared to the 6-OHDA-lesioned rats, which showed a shift towards the peak of the SWA. Finally, the neurochemical results demonstrated that the extracellular levels of GABA increase and those of GLU do not change in the MTh of TTX-treated animals; in the 6-OHDA-lesioned rats, GABA levels decreased and those of GLU increased.
Despite representing a crucial point in the cortico-BG circuit, only a few studies have investigated the role of the MTh and its properties in PD, and controversial data are present in the literature [1,6]. Furthermore, controversial data are present in the literature which regard MTh electrophysiology and properties in the Parkinsonian state. While some studies show a slight MTh firing rate reduction in awake Parkinsonian monkeys [30][31][32][33][34], others revealed increased [1] or unchanged firing rates [35]. We recently showed that all the EEG frequency bands are altered in TTX and 6-OHDA-depleted states; MTh oscillation changes occur preferentially in acute DA depletion, but not in the chronic state [6].
In the present study, we demonstrated that the TTX-inhibition of the medial forebrain bundle (MFB) (for technical details, see our review [27]) determined an acute increase in GABA content with the reduction of the MTh activity. In rats, the SNr activity is reduced after acute DA depletion caused by TTX [36], in accordance with electrophysiological and biochemical data obtained in PD patients [37]. Thus, the changes in MTh activity have to be due to other GABAergic afferents of the network. For instance, we occasionally recorded NRT neurons during TTX infusion and observed a consistent increment of their activity (unpublished observation). Considering that NRT is a powerful modulator of different high-order thalamic nuclei [38], it is likely that this nucleus tightly controls the MTh as well. Conversely, we observed that the MTh neuronal firing rate returns to control condition in the chronic depletion state induced by 6-OHDA. This could be due to the aberrant decrease in GABA and an increase in GLU levels in the MTh of 6-OHDA-lesioned rats that might compensate for the marked inhibition observed in the acute depletion state. The shift of the excitation/inhibition (E/I) balance toward a net excitation in the MTh of the 6-OHDA lesioned rats might indicate that the plastic changes observed in the striatum and cortex of the PD animal model [39] could also occur at the level of the thalamic area. Surprisingly, GABA levels decrease in the chronic DA-depleted MTh, a finding that cannot be explained by the canonical BG/thalamus/cortex loop [19]. Nevertheless, our data, although surprising, are in line with the findings of the only available in vivo study showing a decrease of GABA in the MTh of rats lesioned by striatal infusion of 6-OHDA [22]. Instead, we are the first to report a sustained GLU increase in the MTh of animal models of PD. The decrease in GABA levels is not trivial since the administration of the vasoactive intestinal peptide (VIP) was effective at reversing the motor deficits and contextually increased thalamic GABA contents [22]. Other areas may be involved in rebalancing the MTh activity in chronic DA depletion, such as hyperactivity of the cortex, likely leading the increased release of GLU that we observed within the MTh of 6-OHDA-lesioned rats. The increase of burst activity in the MTh that we recorded in 6-OHDA conditions might depend on the cortex, which, by acting as the main driver, produces a burst input, leading to a coincident burst of activity in MTh neurons, without necessarily causing LTS bursts [1]. The fact that we did not observe a strong effect on burst activity may depend on the canceling effects of different inputs to the MTh, such as synchronized bursts from the BG. The contextual increase in GLU may also cause the production of nitric oxide and contribute to PD development processes [15,40,41].
However, it is also necessary to underline that, since the lesion in our study was unilateral, we cannot exclude that the contralateral hemisphere could compensate, at least partially, for these changes [42]. Furthermore, a recent tractography study in healthy subjects showed that MTh may receive DAergic projections from the ventral tegmental area that are directed towards the supplementary motor area and primary motor cortex (44).
An interesting finding of this study is the interplay between motor thalamus and cortical activity. The DA depletion influenced mainly the MTh-cortical coupling. The divergence of MTh-SWA coupling in the chronic state of DA depletion further suggests that the MTh has undergone the effect of compensatory mechanisms, changing the cross-talk between MTh and the cortex. In accordance with the SWA pattern, the MTh neurons fired preferentially at the beginning of the UP-state in TTX-treated rats compared to the 6-OHDA-lesioned rats, which showed a shift towards the peak of the SWA. Thus, in 6-OHDA rats, there is a delay in the coupling between neuronal activity and the beginning of cortical SWA We showed that, following the acute reduction of DA, the MTh immediately changes the activity, but the chronic lack of DA leads other structures to rebalance the MTh activity. It is possible to hypothesize that, inside the complex circuit involved in PD, rapid clinical transitions of the motor and non-motor symptoms could be directly related to rapid changes in MTh activity, as observed during DBS surgery [20,21]. These hypotheses need further investigation of the MTh activity in PD animal models and PD patients.
Our study presents a number of limitations. For instance, we should keep in mind that our recordings and the position of the microdialysis probe were at ranges of depths in the MTh that correspond to both basal ganglia and cerebellar recipient zones. Therefore, our electrophysiological and neurochemical data may have included values obtained from the two different MTh territories. Moreover, we did not distinguish the two types of MTh neurons during cortical activation by performing the tail-pinch (Nakamura et al., Cerebral cortex, 2014). Further experiments will be required to identify the tonic and burst firing cells in MTh in response to cortical activation, and the eventual changes of their firing characteristics following DA depletion by TTX and 6-OHDA treatment. Another limitation of this study is the use of toxin-induced PD animal models and anesthetized preparations. Despite the usefulness of the animal models, they have limitations that need to be considered when interpreting our findings and translating them to humans [27]. Nevertheless, although rodent models cannot recapitulate all features of PD, they remain a crucial approximation in order to open new avenues.
In conclusion, our data show that in chronic conditions, the firing changes observed in the MTh are relatively mild, considering the profound changes observed in other BG nuclei. This could be due to remodeling and adaptation developed as a consequence of the DA depletion. Indeed, acute TTX induced a strong decrease in the firing rate of MTh neurons. On the other hand, the modification of GLU and GABA are highly significant in 6-OHDA DA-depleted rats and are likely a consequence of changes in the MTh inputs, difficult to explain with the current knowledge of PD pathology. The roles of the MTh in the mechanisms of information processing under physiological and pathological conditions need to be further investigated and clarified.

Animals
Most of the electrophysiological, neurochemical and histological procedures were carried out on adult male Sprague-Dawley rats in compliance with Swiss laws on animal experimentation, and all procedures were approved by the Animal Research Committee and the Veterinary Office of the Canton of Ticino, Switzerland (TI-08-2015).
A subset of neurochemical experiments was performed on male Sprague-Dawley rats at the Department of Physiology and Biochemistry at the University of Malta in accordance with the European Union (EU) Directive 2010/63/EU and the Institutional Animal Use and Care Committee (IAUCC) of the University of Malta.
Utmost care was taken to limit the number of rats used and their suffering.

Pharmacological Blockade of the Medial Forebrain Bundle
The pharmacological blockade of the MFB was performed on adult male Sprague-Dawley rats weighing 399 ± 68.4 g according to our previously described method [6,27]. Briefly, TTX was infused via reverse microdialysis using a probe with a 1 mm dialytic membrane (CMA/11 microdialysis probe, CMA Microdialysis, Stockholm, Sweden), which was positioned according to the following coordinates: 2.5 mm posterior to the bregma, 2 mm lateral to the midline and 8.6 mm below the cortical surface [43]. The probe was perfused with saline solution for 1 h and then with TTX (1 µM) using a syringe pump (CMA/400, CMA Microdialysis) at a flow rate of 1 µL/min. The MTh neuronal activity was recorded and analyzed for at least the next 10 min after the TTX infusion in 27 animals, while the GABA and GLU MTh levels were measured in 14 rats.

Prerecording Surgery
Rats were anesthetized with urethane (1.4 g/kg, i.p.) (Sigma Chemical Co., St Louis, MO, USA) and mounted on a stereotaxic instrument (Stoelting Co.) according to our standard operating procedure [6,27]. Body temperature was maintained at 37-38 • C with a heating pad (Stoelting Co.). After a local anaesthetic injection (lidocaine 0.5%), a midline scalp incision was made, and the skull was almost completely drilled on the left side. The dura were then spread out to expose the cortical surface.

Electrophysiological Recordings
We recorded single neurons for the entire dorsoventral track in the same medio-lateral site in the MTh. In particular, we recorded a total of 82 neurons from the MTh (n = 23 control, n = 25 in 6-OHDA and n = 34 in TTX) from 5000 µm from the cortex to a maximum of 5770 µm.
We recorded multiple MTh neurons from TTX infused rats over the 10 min after infusion The MTh neurons (n = 34) recorded after the TTX-mediated blockade of the MFB were localized from 5250 µm to 5870 µm from the first neuronal detection of the cortex.
The MTh neurons recorded were localized at the following coordinates: 1.8 mm posterior and 2.2 medial to bregma. Single unit activity was off-line analyzed using Spike 2 software (Cambridge Electronic Design, Cambridge, UK) and principal component analysis for spike sorting.
The Electrocorticogram (ECoG) was recorded through a screw electrode (Dentorama, 8 mm of total length, 3 mm tip length) placed on the cortical surface above the frontal cortex (3.0 mm anterior of bregma and 2.0 mm lateral to the midline) and referenced against an indifferent screw electrode placed above the cerebellum. Raw ECoG was bandpass filtered (0.1-300 Hz) and amplified (×2000; Neurolog). The ECoG was on-line digitalized at a sample rate of 600 kHz through an analogical/digital interface (Micro1401 mk II, Cambridge Electronic Design, Cambridge, UK) and stored on a computer to identify off-line robust SWA, which will be the object of further analysis. Simultaneously, we collected the single-unit activity from the MTh. The acquisition of the recording was performed after the stabilization of neuronal activity. The recordings were performed using a tungsten electrode (WPI, TM33B01; impedance 0.9-1.2 MΩ). The trace containing the neuron action potentials was amplified (×10,000; Neurolog), bandpass filtered (300-1000 Hz) and sampled at 60 kHz for digital storage on a computer connected to the CED 1401 interface [36]. The neuronal activity was recorded for at least 10 min after the TTX infusion. Figure 1A shows an example of the recordings with the EEG traces, i.e., SWA (top panel); the discriminated single neurons (middle panel); and the neuronal traces (bottom panel).

Microdialysis Analysis
An additional series of experiments were conducted on adult rats (312 ± 25 g) to measure GABA and GLU before, during and after the TTX-mediated blockade of MFB and in sham and 6-OHDA-lesioned rats [28]. Microdialysis was performed using concentric microdialysis probes (CMA11 with 1 mm PES membrane, outside diameter 0.6 mm, cut-off 35 kDa, Microbiotech) which were positioned into the right MTh (1.8 mm posterior to the bregma, 2 mm lateral to the midline and 6.2 mm below the cortical surface).
After stabilization (4 h), consecutive samples were collected every 10 min. In the acute depleted animals, TTX was infused in the MFB (see above) at the end of the 3rd sample. All samples were immediately frozen at −80 • C until analyzed. At the end of the experiment, rats were sacrificed, and the correct position of the probe was histologically verified (36).
For the experiments in 6-OHDA and sham-lesioned animals, the microdialysis experiments were performed 21-29 days after the lesion.

Histological and Microscopic Analysis
At the end of the experiment, the positions of the electrodes (Figure 1b,c) were histologically checked, and the unilateral dopaminergic lesion in each PD rat was confirmed by quantification of tyrosine hydroxylase (TH) immunoreactivity, as described and analyzed in our previous publication [28].

Statistical Analysis
Concerning the single-unit data, statistical comparisons of firing rates and burst parameters (control vs. 6-OHDA and post-TTX) were conducted using the Kruskal-Wallis test with the Man-Whitney test for post hoc comparisons (asymptotic significance (2-tailed) p-value).
The firing rate comparison was performed taking into account the mean of the first 5 min of recording. The comparison of the burst activity was performed in selected 100-s increments in the presence of stable cortical urethane-induced SWA with a frequency below 3 Hz, as shown elsewhere (45.) SWA is not affected by DA denervation and DAergic treatment [45].
To investigate the correlation between cortex and MTh, we performed the phase correlation analysis, with the calculation of descriptive circular statistics being performed using the CircStat toolbox for MATLAB (2). For the burst analysis, the phase correlation analysis was performed in the same selected 100 s in the presence of stable cortical SWA. For circular analyses and statistical comparisons, we included all neurons from the previous analyses.
For each neuron, the linear phase histograms were analyzed, taking into account the up phase of the SWA, and the spike phase values were used for circular analysis. Then, each neuron was tested for significant SWA phase-locked firing (comparison of degree; Rayleigh's uniformity test). The selected neurons were then used for the analysis of the preferred phase angle, i.e., the mean resultant vector in degree (with 90 • corresponding to the peak of the SWA phase), and the vector length (corresponding to the level of dispersion of the firing). The mean resultant vector length of the phase distribution, bound between zero and one, was used to quantify the level of phase-locking around the mean angle (the closer to one, the more grouped the firing; 2).
The differences between the mean of the resultant vector (degree) and the vector lengths of the three groups (Control, TTX and 6-OHDA) of MTh were tested using the Kruskal-Wallis test.
All circular analyses were performed using the degree and the upper peak in SWA oscillations corresponding to a 90 • phase and the down peak to a 270 • phase.
In all experimental conditions, all neurons were tested for significantly phase-locked firing; i.e., not randomly distributed in relation to the SWA (comparison of degree; Rayleigh's uniformity test) according to the method described by Nakamura and colleagues [2]. Firstly, we analyzed the activity by using the CircStat toolbox for MATLAB. As reported in the text, for each neuron, the linear phase histograms were analyzed, taking into account the up phase of the SWA, and the spike phase values were used for circular analysis. Then, each neuron was tested for significant SWA phase-locked firing; i.e., not-randomly distributed in relation to the SWA (comparison of degree; Rayleigh's uniformity test). To exclude this hypothesis, we performed chi-square tests on the percentage of discarded MTh neurons in the three conditions; i.e., control, acute and chronic DA depletion states. The significance threshold was set at p < 0.05 for all analyses.

Conflicts of Interest:
The authors declare no conflict of interest.