Our brain can be considered as a type of information processing system like a computer, where input signals need to be first detected and properly represented, then integrated with goals for decision-making and output control. Executive control of goal-directed behaviors is frequently disrupted in neuropsychiatric disorders. The executive control function of the brain is achieved through complex circuits of neurons, which also interact closely with surrounding glial cells. In order to understand the etiology and develop treatments for mental disorders, it is important to investigate the functional architecture of neuronal circuits and the mechanisms of neuron-glia interactions. We have chosen to use laboratory mice as a model organism to investigate the basic cellular and molecular mechanisms of cortical information processing. This organism offers several major advantages for this line of study. First, mice and humans both have about 30,000 genes, and about 99 percent of them are shared. Second, the cellular organization of mouse cerebral cortex is similar to human, and major cortical regions are also homologous. Third, it is possible to perform precise molecular genetic manipulations in specific types of cells in mouse brain, which is required to establish causal relationships between genes, cells, circuits and behaviors. Corticospinal neurons (CSNs), which send direct projections from the cortex to the spinal cord, are the logical candidates to read out the effects of cortical cognitive and emotional regulation systems on goal-directed motor programs. CSNs are scattered over multiple cortical regions such as secondary motor, primary motor and somatosensory cortex. How CSNs from different cortical areas are organized functionally during goal-directed actions remains unknown. We developed an innovative approach to specifically label and image the activity of large populations of CSNs with single-neuron resolution in freely moving mice that are trained to perform a goal-directed forelimb reach-and-grasp task, and obtained the first functional map of CSNs across the cortex. Strikingly, this map shows a topographic segregation of CSNs that are activated at different movement steps across multiple cortical areas, suggesting that spatially defined groups of CSNs may encode different movement modules. In collaboration with Dr. Zhigang Hes group at Boston Childrens Hospital, we showed that these region-specific CSNs terminate in different spinal locations, therefore preferentially controlling the premotor neurons of muscles engaged in different steps of the task. Our study reveals a new organizing principle of corticospinal circuits for goal-directed motor skills. It suggests that constructing a multistep goal-directed behavior may involve the recruitment of individual CSN groups in correct order, analogous to composing a sentence (behavior) by stringing together different words (CSN groups). The parallel and modular organization of corticospinal circuits may allow for the development of neural decoding strategies to detect changes in higher-order cortical functions versus subcortical abnormalities and build brain-machine interface-based interventions for brain disorders. Rett syndrome (RTT) is a debilitating neurodevelopmental disorder caused by mutations in the MECP2 gene. Earlier studies of the etiology of this disease mostly focused on neuronal dysfunctions resulting from Mecp2 deletion in mice. Recent research has suggested that astrocytes play an important role in the disease progression of Rett syndrome and loss of Mecp2 in astrocytes can cause neuronal defects non-cell autonomously. However, the functional phenotypes and mechanisms within RTT astrocytes are not well understood. In collaboration with Dr. Qiang Changs group at University of Wisconsin-Madison, we used in vivo two-photon imaging to demonstrate that the spontaneous calcium activity in RTT astrocytes is abnormal. This abnormal activity is mediated by calcium overload in the endoplasmic reticulum through TRPC4 channel, which leads to excessive activation of extrasynaptic NMDA receptors on neighboring neurons and increased network excitability. These findings provide evidence that abnormal calcium homeostasis is a key cell-autonomous phenotype in RTT astrocytes and reveal its mechanism and consequence.
Showing the most recent 10 out of 17 publications
