Molecular mechanisms of synapse development and plasticity

1. NMDA受体依赖性有限公司（NMDA -LTD）的分子机制。NMDAR-LTD是一种长期的突触可塑性形式，导致突触强度降低。已知NMDAR-LTD是突触发育和修饰所必需的。然而，目前尚不清楚它是如何促进突触发育和与精神疾病相关的突触病理的，部分原因是我们对NMDAR-LTD的分子机制的理解仍然不足。缺乏这方面的知识阻碍了预防突触功能障碍的模式设计和开发有效的预防和治疗精神疾病的工具。NMDAR-LTD的诱导主要是通过内吞作用从突触后膜去除AMPA受体介导的（Carroll, Lissin et al. 1999, Lee, Liu et al. 2002）。虽然诱导LTD的方案早在20年前就发表了（Dudek和Bear 1992），但我们尚未完全了解NMDA受体的激活如何引发ampa受体内吞作用。2010年之前，已知当NMDA受体打开时，钙内流和随后的丝氨酸/苏氨酸磷酸酶钙调磷酸酶/PP2B和PP1的激活是AMPA受体内吞作用所必需的（Malenka和Bear 2004）。然而，PP2B和PP1究竟如何诱导ampa受体内吞作用尚不清楚。

1. The molecular mechanism of NMDA receptor-dependent LTD (NMDAR-LTD). NMDAR-LTD is a long-lasting form of synaptic plasticity that leads to decreases in synaptic strength. It is known that NMDAR-LTD is required for the development and modification of synapses. It is unclear, however, how it contributes to synapse development and synaptic pathology associated with mental illness, in part due to the fact that our understanding of the molecular mechanisms underlying NMDAR-LTD is still inadequate. Lack of such knowledge hampers the design of modalities to prevent synaptic dysfunction and to develop effective preventive and therapeutic tools for mental illness. The induction of NMDAR-LTD is predominantly mediated by removal of AMPA receptors from postsynaptic membranes through endocytosis (Carroll, Lissin et al. 1999, Lee, Liu et al. 2002). Although the protocol to induce LTD was published two decades ago (Dudek and Bear 1992), we have yet to completely understand how activation of NMDA receptors provoke AMPA-receptor endocytosis. Before 2010, it was known that when NMDA receptors open, calcium influx and subsequent activation of the serine/threonine phosphatases calcineurin/PP2B and PP1 are required for AMPA receptor endocytosis (Malenka and Bear 2004). Precisely how PP2B and PP1 induce AMPA-receptor endocytosis, however, remains unclear. Our earlier studies show that activation of caspase-3, an effector caspase that can execute cell death apoptosis (a form of programmed cell death), is required for NMDAR-LTD (Li, Jo et al. 2010, Jiao and Li 2011). In contrast to apoptosis, however, caspase-3 activation in LTD is moderate and transient and does not induce cell death. The mechanism by which active caspase-3 promotes LTD, however, remains to be elucidated. The cellular functions of caspases are primarily mediated by proteolysis of their substrates. Although >1000 caspase substrates in apoptotic cells have been reported to date (http://bioinf.gen.tcd.ie/casbah/)(Luthi and Martin 2007), substrates responsible for caspases non-apoptotic functions remain largely unknown. We addressed this question by collaborating with Dr. Sandy Markey (NMIH). We employed the recently developed subtiligase-capture-mass-spectrometry method to identify caspase-3 substrates in neurons undergoing LTD. Subtiligase is an engineered peptide ligase that conjugates esterified peptides onto the N termini of proteins or peptides (Abrahmsen, Tom et al. 1991). Because the majority of eukaryotic proteins are N-terminal acetylated and therefore blocked for subtiligase labelling (Brown and Roberts 1976), synthetic tag peptides can be coupled selectively to the free N-terminal α-amines of proteins derived from proteolysis. The peptide-conjugated proteolytic products can then be affinity-purified and sequenced by mass spectrometry to determine putative caspase cleavage sites within the substrates. Using subtiligase mass spectrometry to interrogate the LTD degradome, we identified 85 putative aspartate cleavage sites in 56 proteins (including 13 proteins that are newly identified caspase substrates) in neurons undergoing LTD. To our knowledge, this is the first proteomic study of caspase substrates in neurons. By examining the cleavage of exogenously expressed substrates in apoptotic cells, and using an in vitro assay that tests the cleavage of recombinant substrates by recombinant caspase-3, we confirmed that three identified substratesGap43, Drebrin (Dbn1) and brain acid soluble protein 1 (BASP1)are true caspase-3 substrates. Perhaps most surprisingly, we found that cleavage of Gap43 (a protein primarily known for its function in axon growth and regeneration) by caspase-3 in postsynaptic neurons is required for AMPA receptor endocytosis and LTD induction. Our manuscript on this study is currently being revised for Molecular and Cellular Proteomics. 2. The molecular mechanism of synaptic pathology associated with schizophrenia. Synaptic pathology has been well recognized in mental disorders. For example, neuroimaging studies show that functional connectivity between neurons in the brains of schizophrenic patients are impaired (Stephan, Baldeweg et al. 2006). Also, interneuronal connections between neurons derived from iPS (induced pluripotent stem) cells of schizophrenic patients are severely impaired (Brennand, Simone et al. 2011). However, little is known about the molecular mechanism underlying synaptic pathology. To address this question, we investigated the mechanism underlying synaptic pathology of schizophrenia. In particular, we investigated the role of D2R in the development of dendritic spines and neuronal circuitry in the hippocampus, a key region for episodic memory which is one of the predominantly impaired cognitive functions in persons with schizophrenia. All antipsychotics antagonize D2R, and their antipsychotic potencies correlate with their D2R-binding affinities, indicating that D2R plays an important role in the psychopathology of schizophrenia (Miyamoto, Duncan et al. 2005). Indeed, an increase in the density of D2R is consistently found in schizophrenic brains (Howes and Kapur 2009). Genetic studies also show that some genes strongly associated with increased risks of schizophrenia encode proteins that regulate D2R, such as dysbindin, which controls trafficking of D2R to the cell surface (Iizuka, Sei et al. 2007, Ji, Yang et al. 2009). Although effective for psychosis, however, antipsychotic medications in adults have little effect on cognitive impairment, which is a core symptom of schizophrenia and a major determinant of disability (Miyamoto, Duncan et al. 2005, Green 2007). In this study, we show that D2R modulates the density and morphogenesis of dendritic spines in hippocampal CA1 neurons of mice via GluN2B- and cAMP-dependent mechanisms. Intriguingly, we found that D2R regulates spines only during postnatal week 3-6, but not in adulthood; and that in mice with deficient expression of the schizophrenia-risk-gene dysbindin, D2R hyperactivity during this period leads to a reduction in the number of spines. More importantly, we demonstrated that even transient suppression of spine development during adolescence by hyperactive D2R has remarkable adverse effects on the connection between the hippocampus and entorhinal cortex and working memory in adulthood, and that these effects can be alleviated by blocking D2R during adolescence. These findings identify a novel function of D2R in the structural development of neurons, show that anomalous D2R activity contributes to neuronal dysconnectivity, and suggest a critical temporal window for interventions of the spine pathology and cognitive impairment associated with D2R hyperactivity.