Cerebellar Modulation of Mesolimbic Dopamine Transmission Is Functionally Asymmetrical

Cerebral and cerebellar hemispheres are known to be asymmetrical in structure and function, and previous literature supports that asymmetry extends to the neural dopamine systems. Using in vivo fixed potential amperometry with carbon fiber microelectrodes in anesthetized mice, the current study assessed hemispheric lateralization of stimulation-evoked dopamine in the nucleus accumbens (NAc) and the influence of the cerebellum in regulating this reward-associated pathway. Our results suggest that cerebellar output can modulate mesolimbic dopamine transmission, and this modulation contributes to asymmetrically lateralized dopamine release. Dopamine release did not differ between hemispheres when evoked by medial forebrain bundle (MFB) stimulation; however, dopamine release was significantly greater in the right NAc relative to the left when evoked by electrical stimulation of the cerebellar dentate nucleus (DN). Furthermore, cross-hemispheric talk between the left and right cerebellar DN does not seem to influence mesolimbic release given that lidocaine infused into the DN opposite to the stimulated DN did not alter release. These studies may provide a neurochemical mechanism for studies identifying the cerebellum as a relevant node for reward, motivational behavior, saliency, and inhibitory control. An increased understanding of the lateralization of dopaminergic systems may reveal novel targets for pharmacological interventions in neuropathology of the cerebellum and extending projections.


Introduction
No longer considered a structure primarily for motor coordination, the cerebellum is now known to contain three distinct regions that contribute to sensorimotor, limbic, and cognitive processes [1]. Cerebellar and cerebral systems work in concert to sharpen the timing of these neural operations [2,3], and each cerebellar hemisphere is connected to multiple closed-loop cortical neural networks in the contralateral cerebral hemispheres, providing an anatomical basis for a cerebellar role in cognition [4][5][6] and a cerebellar mirroring of functional specializations in the cerebrum [7]. Specifically, the cerebellum receives input from the cerebral hemispheres via pontine nuclei in the brainstem, and relays to the contralateral cerebral cortex via cerebellar Purkinje cells and their projections to the dentate nucleus (DN), which provides the sole output from the cerebellum to the cerebrum [8][9][10].
Clinical and preclinical studies support the association of the left cerebral hemisphere with communication functions and the right cerebral hemisphere with spatial reasoning [17,18]. Due to contralateral connections between cerebrocerebellar systems, the cerebellar hemispheres parallel these specializations. Imaging and lesion studies in humans have found the left cerebellar hemisphere to be involved in visuo-spatial operations [19][20][21][22], and a right cerebellar involvement in language processes [23][24][25].
Likely related to these behaviorally-associated asymmetries, the bilateral hemispheres of the brain also contain lateralized neurotransmitter systems in cortical and subcortical regions, and certain experiences have been shown to enhance this lateralization. For example, rats that were handled in their early life showed a significant left/right asymmetry (R>L) in dopamine levels in the NAc [11]. Other studies in rats show greater concentrations of DOPAC/DA in the right cortex and nucleus accumbens in comparison to the same systems in the left hemisphere [26]. Increased dopamine levels in the right prefrontal cortex of adult rats were found to be strongly correlated with anxiety responses in the elevated plus-maze test [27], and dopamine receptor blockade in the right mPFC but not the left mPFC of handled rats resulted in elevated stress and hormone levels [28]. These studies provide evidence of hemispheric imbalance in the mesocorticolimbic dopamine system.
Many neurophysiological disorders are characterized by altered profiles of mesocorticolimbic dopaminergic transmission, such as addiction, ADHD, and schizophrenia [29][30][31], and interestingly, cerebellar pathology and specifically Purkinje cell dysfunction are being considered as substrates in these and other psychiatric disorders [32][33][34]. A direct pathway from the deep cerebellar nuclei to the VTA has been identified [35]. Mittleman and colleagues [36] used cerebellar DN electrical stimulation to evoke dopamine efflux in the medial prefrontal cortex (mPFC) of Lurcher mutant mice, a common model of autism spectrum disorder with 100% loss of Purkinje cells within the first 4 weeks of life. The Lurcher mutants exhibited attenuation in DNevoked mPFC dopamine release compared to controls, suggesting that developmental loss of Purkinje cells in the cerebellum, similar to that of autism spectrum disorder, can lead to a disruption in mPFC dopamine transmission. In further studies, these researchers found reorganization of cerebello-cortico circuitry in Lurcher mutants and Fmr1 mice, another genetic model that exhibits dysfunction or absence of Purkinje cells in the cerebellum [37]. The reorganization of the DN to mPFC pathways included altered relative influence of the ventral tegmental area (VTA) and thalamic nuclei, with the mutant mice showing a stronger dependence on thalamic nuclei compared to control mice. This shift in cerebellar modulation towards the ventral lateral thalamus and away from the VTA leads to speculation about the cerebellum's influence on dopaminergic functioning not only in the mPFC but also the nucleus accumbens (NAc).
Neural fibers between the VTA and NAc constitute one of the most densely innervated dopamine pathways in the brain [38,39]. Dopamine release in the NAc is known to be associated with reward and motivational processes [40][41][42][43], and disruption to normal dopamine processing, including hemispheric balance, can lead to a host of motor and cognitive deficits. For example, decreased motivation and novelty-seeking often observed in patients with Parkinson's disease are related to asymmetry of dopamine [44] and individual differences in incentive motivation or sensitivity to natural rewards in humans has also been associated with increased asymmetry in dopaminergic systems [45].
Using in vivo fixed potential amperometry in anesthetized mice, Experiment 1 of the present study aimed to distinguish any asymmetries between the mesolimbic dopamine pathways by stimulating the medial forebrain bundle (MFB), which consists of the dopaminergic axons projecting from the VTA to NAc, and recording dopamine release in the NAc in each hemisphere. In Experiment 2, we assessed cerebellar influence of NAc dopamine lateralization by comparing DN stimulation-evoked dopamine release in both hemispheres. The DN has contralateral glutamatergic projections to reticulotegmental nuclei that, in turn, project to pedunculopontine nuclei (PPT), which projects to and stimulates dopamine cell bodies in the VTA [46,47].

Animals
Twenty-seven male C57BL/6J mice (Jackson Laboratories, ME) were housed 3-5 per cage in a temperature-controlled environment (21±1 °C) on a 12 hr light/dark cycle with (lights on at 0600) and given food and water available ad libitum. All experiments were approved by the Institutional Animal Care and Use Committee at the University of Memphis and conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals. In order to maintain the principle of reduction related to scientific experiments on animals [48], efforts were made to minimize the number of mice used. Sample sizes were determined based on G*Power analysis [49] and effect sizes from our previous amperometric results [50][51][52]. Efforts were also made to minimize pain and discomfort.

Surgery
Mice were anesthetized with urethane (1. . For MFB stimulations, the recording electrode was placed in the ipsilateral NAc; however, due to contralateral connections in cerebrocerebellar circuity, the recording electrode was placed contralateral to cerebellar DN stimulation [46,47]. Pharmacological studies from our lab have confirmed the recorded current changes in the NAc to be dopamine dependent [50,51].

Dopamine Recordings and Drug Infusions
Fixed potential amperometry, also known as continuous amperometry, coupled with carbon fiber recording microelectrodes has been confirmed as a valid technique for real-time monitoring of stimulation-evoked dopamine release [52,54,55]. All amperometric recordings were made within a Faraday cage to increase signal to noise ratio. A fixed potential (+0.8 V) was applied to the recording electrode, and oxidation current was monitored continuously (10 K

Data Analysis
To quantify MFB and DN stimulation-evoked dopamine efflux, pre-stimulation current values were normalized to zero, and data points occurring 0.25 sec pre-and 50 sec post-onset of the stimulation were extracted. Dopamine release was quantified as the magnitude of the response (peak minus baseline). Independent samples t-tests were used to assess hemispheric differences in NAc dopamine release. Stimulationevoked dopamine release was also measured 5 min following intra-DN infusion. A twoway mixed ANOVA was used to determine the effect of drug infusion (PBS or lidocaine) and time (pre-or post-infusion) on dopamine release. Dopamine release post-infusion was also converted to percent change (with the pre-infusion concentration being 100%), and an independent samples t-test was used to determine if there was a significant difference between PBS and lidocaine.

Histology
The tips of the stimulating electrodes were positioned within the anatomical boundaries of the DN or MFB. Figure 2A

Discussion
The current study assessed the hemispheric lateralization of stimulation-evoked dopamine in the NAc and the influence of the cerebellum in regulating this reward-associated pathway. Results from Experiment 1 show that the mesolimbic pathway itself is not responsible for asymmetrical lateralization of dopamine release, given that NAc dopamine release did not differ between hemispheres when evoked by ipsilateral MFB stimulation. The mesolimbic dopamine pathways are known to be ipsilaterally dominated, with less than 10% of fibers from the VTA crossing hemispheres to reach the contralateral NAc [58], and our findings suggest that these parallel pathways act functionally similar when equally stimulated. Instead, dopaminergic asymmetry may originate, at least in part, from the influence of the cerebellum on these pathways. In Experiment 2, we show that stimulation of the cerebellar DN can elicit NAc dopamine release in the contralateral hemisphere, and this dopamine release was significantly greater in the right NAc relative to the left.
Furthermore, cross-hemispheric talk between the right and left cerebellar DN does not seem to influence mesolimbic dopamine release given that, in Experiment 3, when lidocaine was infused into the DN opposite the electrically stimulated DN to inactive this system, dopamine release was not altered. We have previously shown that this infusion protocol and lidocaine dose can deactivate local neural activity from 2 to 20 min post infusion [52]. The present findings exhibiting no cross-hemispheric talk between the dentate nuclei is not surprising given that these structures are physically separated by the cerebellar vermis [59]. Furthermore, the cerebro-cerebellar networks are thought to be laterally independent circuits, with cortical hemispheric connections primarily running through the corpus callosum [60][61][62].
The MFB-evoked dopamine release profiles in the NAc observed in the present study are consistent with those previously published [51,52,63]. When comparing dopamine release in the NAc, mPFC, and amygdala, we have previously characterized NAc dopamine release to be higher in concentration and quickly cleared from the synapse, indicating greater synaptic confinement, relative to the other brain regions [51]; however, in the present study, NAc dopamine release elicited by cerebellar DN stimulation was attenuated in concentration and slower to clear from the synapse, allowing for greater diffusion beyond the synaptic release site, relative to MFB stimulated dopamine release. The profile of DN-evoked dopamine release in the NAc more closely fits the description of volume transmission. In contrast to point-to-point synaptic contacts, volume transmission provides a communication mode that is temporally slower, broader in anatomical reach, and more suited to modulatory/tuning functions [64]. Thus, the present findings demonstrate a functional regulatory role of the cerebellum over mesolimbic dopamine activity and provide a neurochemical mechanism for studies identifying the cerebellum as a relevant node for reward, motivational behavior, saliency, and inhibitory control [65][66][67][68].
Cerebral and cerebellar hemispheres are known to be asymmetrical in structure and function [12,13], and studies are mounting to show that this asymmetry extends to the mesolimbic dopamine system. Although dopaminergic asymmetries are not consistently documented in the literature and seem to vary based on age, gender, species, and strain [for review see 69], numerous studies using methods of protein analyses in rodents have shown hemispheric distributions similar to the present results, greater levels of dopamine and its metabolites in the right NAc relative to the left [11,26,70,71]. It has been suggested that individual differences to natural rewards are a product of asymmetry in dopamine systems [72]. The present results support the notion that reward processes in the brain may be lateralized between cerebello-cortico circuitry, which has considerable applications for disorders involving dysfunction of the subcortical dopamine functioning, disorders such as schizophrenia, Parkinson's, ADHD, and addiction [29][30][31]. For example, patients with unlilateral onset of Parkinon's disease often develop an asymmetry of dopamine deficiency [73,74], with behavioral deficits not limited to motor functioning. In one study, patients whose motor symptoms began on the left side of the body performed more poorly on cognitive tests than those with right-side onset, leading to the conclusion that damage to right-hemisphere dopamine plays a greater role in associated cognitive decline than left-hemisphere depletion [75]. In the future, treatment for such symptoms may be optimal with if applied differentially in each hemisphere [72].
Cerebellar-mediated dopamine pathways have previously been shown to exhibit plasticity and compensatory changes in the neural circuitry of rodent models of autism, providing a foundation for the cerebellum to develop unique connections between cerebral hemispheres [67]. Some hypothesize that the corpus callosum enables hemispheric specializations, allowing one hemisphere to reconfigure circuits and adapt to certain environmental changes while the other hemisphere preserves existing functions [77]. Others have suggested that lateralization may occur though the action of steroid hormones [78]. Despite varying perspectives on mechanism of action or developmental precursors, neural circuits specialize their connections to use resources more efficiently and minimize wiring [62]. It appears that this specialized processing is represented in the brain as analogue signals, namely changes in the concentration of messenger molecules in the synaptic space [79]. The differences observed in concentrations of dopamine release in left and right cerebellar-NAc circuits may provide a foundation for the divergent types of information processed and transmitted between the reward circuits of each hemisphere. Although each hemisphere may contain homologous neural substrates, lateralized dopamine release patterns allow for anatomically defined circuitry to be repurposed and used for other adaptive behaviors [80]. Further examination into the functional relationship between lateralized cerebrocerebellar networks may help stimulate new insight to understanding hemispheric specializations in neurodevelopment and lead to novel targets for pharmacological interventions.