Research Projects

The nervous system comprises two classes of cells: neurons and glia. For a long time, neurons have been considered to be the sole basis for information integration, while glia have been thought to play a merely supportive role.  However, this dogma is now being challenged.  Recent studies show that signals can pass back and forth between neurons and glia.  Rather than being a quiet bystander, glia likely participate in the active signal processing in the nervous system.  While the projects summarized above focus on neurons in the brain, other projects in my lab examine the roles of glia.  The most intriguing role is glial function at synapses.  So far, much of our knowledge about the glia-synapse relationship has come from in vitro culture studies, with limited in vivo investigations mostly came from single time-point, fixed tissue examination.  Given the complexity and variability of synaptic connections in the nervous system, the field is crying for longitudinal data of glia-synapse interaction in living animals.  My lab approaches this question from two angles: Role of astrocytes at CNS synapses: A major challenge to studying functions of cortical astrocytes (major glia in the brain) in vivo has been the paucity of molecular markers to facilitate genetic manipulation.  For example, although the glial acidic fibrillary protein (GFAP) is widely regarded as an astrocytic marker, it is not expressed by most astrocytes in the healthy mature cortex.  Recently, it has been suggested that some cortical astrocytes in the mature brain expresses the transcription factor, Gli1, in response to active Shh signaling.  We therefore characterized a Gli1CreER mouse line in detail during the past two years and confirmed that Gli1+ cells represent a subset of cortical astrocytes in adult mice.  Taking advantage of the inducible Cre/loxP strategy, we can now target this subset of cortical astrocytes for selective manipulation in vivo.  We are currently exploring the roles of these subset astrocytes in synapse formation and stability, as well as their potential role in disease development. Role of Schwann cells at the neuromuscular junction: Terminal Schwann cells (SCs) at neuromuscular junctions (NMJs) are the counterparts to astrocytes in the CNS. They are morphologically and functionally distinct from myelinating SCs, which are counterparts to oligodendricytes in the CNS.  Compared to synapses in the CNS, the NMJ is much bigger in size and experimentally much more accessible. In this project, we investigate morphological changes of SCs at aging NMJs, as well as the possible cellular mechanisms underlying glial-synapse interactions for synapse remodeling.  In my postdoctoral work, my colleagues and I generated transgenic mice that express different fluorescent proteins under glial- and neuronal-specific promoters (Zuo et al., 2004).  In my lab, we imaged individual SCs in vivo, and captured their morphological dynamism with high temporal resolution.  We found that SCs are mostly stable during adulthood.  However, at the aging NMJs SCs become more dynamic, with progressive myelination occurring frequently at the axon entry site to the NMJs.  This often results from a transition from terminal to myelinating SCs, and leads to loss of synaptic contact between motor axons and muscle fibers.  In addition to changes from terminal to myelinating SCs, we also observed more cases of addition and removal of SCs at aged NMJs in chronic imaging.  In order to investigate how individual terminal SC’s territory changes in association with the addition/removal of SCs, we sequentially dye-filled individual SCs with rodamine-dextran, and showed that adult SCs are arranged in a static tile pattern, while young SCs dynamically intermingle.  Furthermore, in collaboration with Dr. Thomas Misgeld’s lab in Munich, we demonstrated that glial segregation is due to spatial competition, where glial-glial and axonal-glial contacts constrain the territory of single SCs.  Part of these results has been published in Neuron Glia Biology (Zuo and Bishop, 2008) and The Journal of Cell Biology (Brill et al., 2011). The projects summarized above will provide first-hand information on how glia modulate synaptic connections, such as synapse formation, elimination and plasticity.  Since the synapse is believed to be the site of information exchange in the nervous system and is characterized by structural changes during learning and memory, understanding glial cells’ roles in structural plasticity of synapses should provide valuable knowledge on how information is transferred, integrated and stored in the nervous system.  In addition, better understanding of glial involvement in aging and neurological diseases will likely lead to identification of therapeutic targets for neurological pathologies. Publications Zuo Y, Lubischer JL, Kang H, Tian L, Mikesh M, Marks A, Scofield VL, Maika S, Newman C, Krieg P and Thompson W (2004) Fluorescent proteins expressed in mouse transgenic lines mark subsets of glia, neurons, and immune cells for vital examination. J. Neurosci. 24(49):10999-11009. [PDF] Zuo Y and Bishop D (2009) Glial imaging during synapse remodeling at the neuromuscular junction. Neuron Glia Biology 4(4):319-326. [PDF] Brill MS, Lichtman JW, Thompson W, Zuo Y, Misgeld T. (2011) Spatial constraints dictate glial territories at murine neuromuscular junctions. J Cell Biol. 2011 Oct 17;195(2):293-305. [PDF] Yu X, Wang G, Gilmore A, Yee AX, Li X, Xu T, Smith SJ, Chen L, Zuo Y (2013) Accelerated experience-dependent pruning of cortical synapses in ephrin-A2 knockout mice. Neuron 80(1): 64-71 [PDF] Hodges JL, Yu X, Gilmore A, Li X, Perna JF, Tjia M, Chen C-C, Bennett H, Lu J and Zuo Y (2017) Astrocytic contributions to synaptic and learning abnormalities in a mouse model of Fragile X Syndrome. Bio. Psych. 82(2):139-149 [PDF]

The majority of excitatory glutamatergic synapses in the mammalian central nervous system (CNS) are located on postsynaptic dendritic spines, which contain all the essential components required for postsynaptic signaling and thus serve as a good indicator of synaptic connectivity.  Many lines of in vivo imaging data have shown that neuronal circuits can be modified through de novo formation or loss of dendritic spines.  Our early data show that spines turn over rapidly in adolescent mice but become remarkably stable in adults.  In addition, sensory experience from whiskers promotes a net loss of spines by increasing the rate of spine elimination in barrel cortex.  This synaptic loss occurs predominantly in adolescence and through NMDA receptor-dependent competitive mechanisms (Zuo et al., 2005). While dynamics and morphological changes of dendritic spines are believed to provide a structural basis for synaptic plasticity, the molecular mechanisms regulating spine formation, stabilization and elimination remain unclear.  Mutant mice with altered spine dynamics provide a good entry point to address this question.  In the past a few years, my lab has been screening various mutant mouse lines (including cell adhesion molecules, signaling molecules as well as transcription and translating regulators) with selective defects different aspect of spine dynamics (such as spine formation or spine elimination), and some of them are directly associated with experience.  We are currently analyzing these mice to dissect the upstream and downstream molecular machinery in regulating spine dynamism. Publications Zuo Y, Yang G, Kwon E and Gan WB (2005) Long-term sensory deprivation prevents dendritic spine loss in primary somatosensory cortex. Nature 436(7048):261-265. [PDF] Zuo Y, Lin A, Chang P and Gan WB (2005) Development of long-term dendritic spine stability in diverse regions of cerebral cortex. Neuron 46(2):181-189. [PDF] Fu M and Zuo Y (2011) Experience-dependent structural plasticity in the cortex. Trends Neurosci. 34(4):177-187. [PDF] Yu X, Wang G, Gilmore A, Yee AX, Li X, Xu T, Smith SJ, Chen L, Zuo Y (2013) Accelerated experience-dependent pruning of cortical synapses in ephrin-A2 knockout mice. Neuron 80(1): 64-71 [PDF] Chen C, Lu J, and Zuo Y (2014) Spatiotemporal dynamics of dendritic spines in the living brain. Front. Neuroanat. 8:28 [PDF] Lu J, and Zuo Y (2017) Clustered structural and functional plasticity of dendritic spines. Brain Res. Bull.129:18-22 [PDF] Martin PM, Stanley RE, Ross AP, Frietas AE, Moyer CE, Brumback AC, Iafrati J, Stapornwongkul KS, Dominguez S, Kivimae S, Mulligan KA, Pirooznia M, McCombie WR, Potash JB, Zandi PP, Purcell SM, Sanders SJ, Zuo Y, Sohal VS, Cheyette BNR (2016) DIXDC1 contributes to psychiatric susceptibility by regulating dendritic spine and glutamatergic synapse density via GSK3 and Wnt/β-catenin signaling,  Mol. Psych. [Epub ahead of print] [PDF] Tija M, Yu X, Jammu LS, Lu J and Zuo Y (2017) Pyramidal neurons in different cortical layers exhibit distinct dynamics and plasticity of apical dendritic spines. Front. Neural Circuits 11:43 [PDF] Zemmar A, Chen C-C, Weinmann O, Kast B, Vajda F, Bozeman J, Isaad N, Zuo Y* and Schwab ME* (2017) Oligodendrocyte- and neuron-specific Nogo-A restrict dendritic branching and spine density in adult mouse motor cortex. Cerebral Cortex.[Epub ahead of print] *equal contribution [PDF]

Novel motor skills are learned through repetitive practice and, once acquired, can persist long after training stops.  Earlier studies have shown that such learning induces a change in functional map in the primary motor cortex.  However, how motor learning affects neuronal circuitry at the level of individual synapses and how long-lasting memory is structurally encoded in the intact brain remain elusive.  The majority of excitatory glutamatergic synapses in the mammalian central nervous system (CNS) are located on postsynaptic dendritic spines, which contain all the essential components required for postsynaptic signaling and, thus, serve as a good indicator of synaptic connectivity.  Using transcranial two-photon microscopy to visualize fluorescently labeled dendritic spines in the brain of transgenic mice (YFP-H line), my lab examines circuit plasticity mediated by structural remodeling during motor learning and memory formation. Temporal rules of synapse remodeling: Training mice with a forelimb specific task, we found that new spines of the output pyramidal neurons form rapidly (within one hour of training) in the motor cortex contralateral to the trained limb.  Interestingly, the amount of spines formed within the first training session is linearly correlated with the number of successful reaches during that training session, revealing a direct link between learning and spine formation.  Although selective elimination of the synapses that existed before training gradually returned the overall synaptic density to the baseline level, learning-induced new synapses were preferentially stabilized during subsequent training.  Furthermore, we demonstrated that learning-induced circuit rewiring could endure long after training had stopped.  Practice of novel, but not previously learned, tasks further promoted synaptogenesis in adulthood.  These findings reveal that rapid, but long-lasting synaptic reorganization is closely associated with motor learning.  They also suggest that stabilized neuronal connections are the foundation of durable motor memory.  This work was published in Nature (Xu et al., 2009). Spatial rules of synapse remodeling: Many lines of evidence suggest that memory in the mammalian brain is stored in distinct spatiotemporal patterns.  Despite recent progresses in identifying neuronal populations involved in memory coding, the synapse-level mechanism is still poorly understood. To address this question, we recently investigated the spatial distribution of learning-induced new spines.  We found that spines are not uniformly distributed along dendrites, and spine formation and elimination are regulated by local spine density.  While spines formed during the same task are spatially correlated, spines formed during different tasks are not.  These new findings reveal that distribution of spines along dendrites is actively regulated in vivo, and that memory traces for a motor task may be encoded by the spatial distribution of synapses in the mammalian brain. This work has been recently published in Nature (Fu et al., 2012). Combining behavioral analysis with in vivo imaging opens the possibility of understanding synapse remodeling during learning.  Many questions remained unanswered. To what degree does the brain change in response to learning?  Is learning-induced synapse remodeling specific to certain neuronal types or cortical layers?  How do cortical changes contribute to the motor engram?  Can we perturb learning?  How do different motor skills interfere with each other?  What happens if the motor cortex is injured?  Can motor rehabilitation help with the functional recovery of the brain?  Currently we are actively developing tools and experimental paradigms to address these questions. Publications Xu T, Yu X, Perlik A, Tobin W, Zweig J, Tennant K, Theresa J, and Zuo Y (2009) Rapid formation and selective stabilization of synapses for enduring motor memories. Nature 462(7275):915-919. [PDF] Fu M, Yu X, Lu J, Zuo Y. (2012) Repetitive motor learning induces coordinated formation of clustered dendritic spines in vivo. Nature. 2012 Feb 19;483(7387):92-5. [PDF] Yu X and Zuo Y (2011) Spine plasticity in the motor cortex. Curr. Opin. Neurobiol. 21(1):169-174. [PDF]