Function and Regulation of ALDH1A1-positive Nigrostriatal Dopaminergic Neurons in Motor Control and Parkinson's disease

National Institute On Aging

Investigators

Linked publications & trials

Abstract

To reveal the activity pattern of ALDH1A1+ DANs during motor skill learning and sensorimotor behavioral control, we will record single-unit spiking activity from DANs in the ventral SNc of Aldh1a1CreERT2 mice, using optotrode equipped with either silicone laminar arrays or microwire bundles in an adaptive motor learning task, as head-fixed mice learn to walk on a spheric treadmill with head fixation. After learning the motor skill of adaptive walk, mice will proceed to a sensorimotor operant task of virtual navigation on the treadmill with precise sensory cues and feedback to forage for reward. Identification of ALDH1A1+ DANs during recording will be accomplished using the optogenetic tagging method. We will determine how activity of ALDH1A1+ and ALDH1A1-negative (ALDH1A1) DANs is modulated at different epochs and stages of motor learning. We will also investigate burst activity related to the classic reward prediction errors in these neurons in the presence of predictive sensory cues and various trial outcomes. The results will allow us to determine how DAN activity from molecularly defined classes contribute to motor learning and behavioral control. We would expect both burst and tonic activity of ALDH1A1+ DANs to show elevated spike rate during learning which correlates with behavioral improvement. We also expect ALDH1A1+ DANs display unique pause and rebound firing following transient inhibition from striatal inputs that could be essential for learning to avoid negative outcomes and motor errors. Recording spiking activity has the advantage of high temporal resolution when instantaneous timing of spikes is needed to convey critical behavioral information, such as spike timing in DAN bursts to track behavioral epochs. An alternative method for recording DAN activity is using our recently acquired deep brain calcium two-photon imaging, which also allows us to investigate plasticity of DANs in learning by following the activity of the same ALDH1A1+ DAN ensemble over the entire course of learning. The capacity of subcellular spatial resolution in imaging could also help to identify sources of learning related signals to these neurons by visualizing changes in specific input strength in dendritic compartments during learning and task performance. The calcium signals have slower dynamics than spikes, but certain aspects of DAN activity in sub-second scale might be sufficient for behavioral control, such as changes of tonic spike rate or magnitude of their axonal dopamine release. To establish the causality between ALDH1A1+ DAN activity and motor learning performance, we will manipulate somatic spiking activity and axonal dopamine release of ALDH1A1+ DANs with different timescales during the same motor skill and sensorimotor learning tasks described in the recording experiments, using optogenetics and chemogenetics methods. Light-gated chloride channels, such as JAWS, will be expressed in these neurons using Cre-dependent viral vectors to allow transient inhibition of their somatic activity via light delivered through implanted optic fibers in the SNc. Timing of the optogenetic inhibition of soma will be time-locked to specific behavioral epochs to determine how precise temporal patterns of ALDH1A1+ DAN spiking activity, particularly the burst firing, causally contributes to learning. To determine how the longer time scale ALDH1A1+ DAN excitability contribute to learning, facilitatory or inhibitory chemogenetic receptor DREADDs will be expressed in these neurons to allow bidirectional control of their excitability during the task performance. Axonal dopamine release often has region-specific regulation mechanisms independent from somatic spiking. To investigate how dopamine released by these neurons in specific target areas contribute to learning, we will transiently suppress axonal dopamine release of ALDH1A1+ DANs in one of their projected areas. Light activated Gi/ocoupled receptors will be expressed in ALDH1A1+ DAN axons, where their dopamine release can be inhibited by light delivered through an implanted optic fiber at various subregions of the dorsal striatum. The resulting learning efficacy and behavioral changes from all these manipulations will be compared with corresponding sham control mice to draw conclusions. The results will allow us to determine the causal roles of specific aspects of ALDH1A1+ DAN neuronal activity in motor learning and behavioral control. If overall dopamine release resulted from ALDH1A1+ DAN activity over the extended behavioral period contributes to learning, we would expect that reduced excitability of these neurons delays motor learning, while increased excitability could speed the learning. If timing of burst activity of ALDH1A1+ DANs is pivotal in learning, we would expect inhibiting these neurons at the movement initiation causes more motor errors, while inhibiting them at the presence of sensory cues may lead to failure in reward association. Suppression of their dopamine release in the striatum is also expected to impair learning that maybe learning-stage dependent. To gain more circuit insight, an alternative strategy to investigate causal roles of ALDH1A1+ DAN neuronal activity in motor learning is genetically manipulate specific inputs to these neurons. Toward this direction, we had crossbred Aldh1a1CreERT2 mice with Grin1-LoxP KI mice to selectively disrupt the glutamate-mediated excitatory inputs to the ALDH1A1+ DANs. The resulting Grin1 cKO mice performed normally in the rotarod test, suggesting the glutamatergic afferent activity at ALDH1A1+ DANs is not required for the motor skill learning, but may still be involved in other aspects of learning. As the ALDH1A1+ SNc DANs display distinct rebound activity in response to the GABA-B receptor (Gabbr1)-mediated inhibitory inputs from dSPNs, we are in the middle of developing Gabbr1 cKO mice to selectively disrupt the expression of Gabbr1 in ALDH1A1+ DANs. This line of Gabbr1 cKO mice will allow us to critically evaluate the contribution of Gabbr1-mediated rebound and burst activity in ALDH1A1+ DAN-dependent motor skill learning.