1. The role of miRNAs in long-lasting synaptic plasticity. microRNAs (miRNAs) are short, non-coding RNAs that bind to mRNAs to inhibit translation and/or promote mRNA degradation. Each miRNA can potentially target to hundreds of distinct mRNAs, and thousands of genes are regulated by miRNAs. miRNAs are increasingly recognized as key regulators of gene expression and have been found to play important roles in diverse cellular processes, such as the differentiation and development of cells. miRNAs are crucial for proper brain function. Hundreds of miRNAs are expressed in the brain. miRNA loss i leads to alterations in synaptic protein expression, synaptic transmission, dendritic spines, learning, and memory. Several miRNAs, such as miR-134, miR-125, miR-138, miR-132, miR-29 and miR-188, regulate the morphogenesis of dendritic spines. miRNAs are also implicated in mental disorders. For instance, mounting evidence suggests that mutations in miRNA genes and miRNA biogenesis machinery are associated with increased risk of schizophrenia (Beveridge et al., 2008). Despite the demonstrated importance of miRNAs, however, the function of the vast majority of miRNAs expressed in the brain have yet to be elucidated. α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPAR) are ionotropic glutamate receptors that mediate fast excitatory synaptic transmission in the central nervous system. AMPAR are tetramers composed of four possible subunits (GluA1-4) (Shepherd and Huganir, 2007). The number of AMPAR in synapses determines the strength of synaptic transmission, and their abnormal expression has been implicated in cognitive impairments associated with such neurological and neuropsychiatric diseases as Alzheimers disease, ischemia, schizophrenia and depression (Chang et al., 2012). AMPAR expression is regulated by synaptic activity (Grooms et al., 2006). During activity-dependent synaptic plasticity, for instance, the activation of N-methyl-D-aspartate (NMDA) or metabotropic glutamate receptors affects the abundance of synaptic AMPAR through both post-translational mechanisms (including phosphorylation, palmitoylation and ubiquitination) and local translation of dendritic mRNAs encoding AMPAR subunits (Grooms et al., 2006;Ju et al., 2004;Lu and Roche, 2012;Snyder et al., 2001;Sutton et al., 2006). Activity-dependent modulation of AMPAR is an important mechanism that tunes synaptic strength to refine synaptic connectivity during brain development and to store information in the brain during learning and memory (Shepherd and Huganir, 2007). Despite the broad recognition that AMPAR play a pivotal role in brain functions, however, molecular mechanisms underlying their regulation, especially activity-dependent local translation of AMPAR in dendrites, are only incompletely understood. In neurons, many miRNAs localize to dendrites where they can be regulated by synaptic activity (Hu et al., 2014;Schratt, 2009). For instance, NMDAR activation inhibits miR-191 expression locally in dendrites, resulting in an elevation of its target tropomodulin-2, which promotes actin depolymerization, shrinkage and elimination of dendritic spines. In view of this finding, we hypothesize that miRNAs also contribute to activity-dependent local synthesis of AMPAR in dendrites. . To test this hypothesis, we combined miRNA pull-down and computational prediction to search for miRNAs that target the AMPAR subunit GluA1. This approach leads to the identification of miR-501-3p as a GluA1 binding miRNA. Our further analysis of miR-501-3p shows that it is increased locally in dendrites following NMDAR activation, and that this upregulation of miR-501-3p is required for NMDAR-dependent inhibition of GluA1 expression, long-lasting spine shrinkage and elimination. These findings reveal that miRNAs are important regulators of activity-dependent local synthesis of dendritic AMPAR. 2. Delineating the miRNAome in schizophrenia. Schizophrenia is a debilitating mental illness with a lifetime prevalence of 4.0/1,000 worldwide, and a typical age of onset at late adolescence and young adulthood. Schizophrenia manifests as a spectrum of clinical symptoms, such as hallucinations, delusions and social withdrawal, with cognitive deficits as a core feature. The etiology of schizophrenia is unclear. Family, twin and adoption studies indicate that schizophrenia has a strong genetic component. It is thought that schizophrenia is caused by complex interactions between multiple genes and the environment. Linkage and association studies have led to the discovery of dozens of susceptibility genes. However, most risk genes have small effect sizes and are protein-coding genes. The genetic architecture of schizophrenia is still largely unclear. The recently-established next-generation sequencing technologies have significantly pushed back the limitations of prior transcriptome profiling approaches by greatly improving the throughput and depth of coverage. In addition, the superior sensitivity and quantifiability of next-generation sequencing are especially advantageous to detect low-abundant miRNAs. In this study, we applied the Illumina Solexa deep-sequencing platform, which is a next-generation sequencing technology based on massively parallel signature sequencing, to delineate the miRNA transcriptomes in schizophrenic brain. Dr. Joel Kleinman (NIMH) has provided me RNA samples isolated from postmortem brains of schizophrenic and normal control subjects. All postmortem human brains are obtained from the Offices of the Chief Medical Examiner of the District of Columbia, and of the Commonwealth of Virginia, Northern District, all with informed consent from the legal next of kin (protocol 90-M-0142 approved by the NIMH/NIH Institutional Review Board). Diagnoses, macro- and microscopic neuropathological examinations and toxicological analysis are performed on all cases. Small RNAs are extracted from Brodmann area 46 of the dorsal lateral frontal cortex (DLPFC) by using the mirVana miRNA Isolation Kit (Ambion) and used for construction of deep-sequencing libraries as described (Wu et al., 2010). DLPFC is selected because of its strong implication in the psychopathology of schizophrenia (Weinberger et al., 1986). We have completed the library construction step. We are currently conducting deep-sequencing. We will analyzed deep-sequencing results for the change of miRNAome in schizophrenia brains.
