?-site APP cleaving enzyme (BACE1) cleaves the amyloid precursor protein (APP) as the initial step in the production of the amyloid ? (A??) peptide, which is involved in the pathogenesis of Alzheimer's disease. As a developmentally regulated aspartyl protease, BACE1 has been shown to have putative substrates other than APP, suggesting multiple physiological functions of BACE1. Recently, changes in BACE1-dependent proteolytic processing of neuregulin 1 (NRG1) have been linked to deficits in central and peripheral myelination in BACE1-/- mice. Because NRG1 has also been genetically linked to schizophrenia, we tested whether BACE1-/- mice with impaired processing of NRG1 show schizophrenia-like behavioral abnormalities. We demonstrate that BACE1-/- mice exhibit multiple features considered as animal analogues of schizophrenia symptoms including deficits in prepulse inhibition, novelty-induced hyperactivity, hypersensitivity to a glutamatergic psychostimulant (MK-801), cognitive impairments and alterations in social interactions. Importantly, some of these symptoms were responsive to clozapine treatment. The expression of ErbB4, a receptor of NRG1 and likely the main mediator of NRG1 functions in the brain, particularly those related to schizophrenia, was not changed in BACE1-/- mice. However, a synaptic pool of ErbB4 was significantly reduced as indicated by decreased association of ErbB4 and PSD-95 in the brains of BACE-/- mice. In accordance with the role of synaptic ErbB4 in synaptic function and spine morphology, BACE1-/- mice displayed reduced density of spines in hippocampal pyramidal neurons. Our findings indicate that schizophrenia-like endophenotypes can be sufficiently caused by genetic deficits of BACE1, leading to an alteration in NRG1/ErbB4 signaling that may represent a new mechanism in the pathogenesis of schizophrenia and related psychiatric disorders. Given recent reports that impaired NRG1/ErbB4 signaling leads to glutamatergic hypofunction implicated in mechanisms of schizophrenia, we will test the hypothesis that BACE1-dependent impairments in NRG1/ErbB4 signaling can modulate NMDA or AMPA receptor pathways (Aim1).
In Aim2, we will use RNA interference as a strategy to study effects of BACE1 inhibition initiated during development or in adulthood on expression of schizophrenia-like phenotypes in adult mice. If the RNA interference approach fails to produce sufficient BACE1 inhibition, an alternative approach (mouse models with conditional expression of BACE1 on BACE1 null background) will be used to modulate the levels of BACE1 at different stages of development. The proposed studies will be important for the identification of BACE1-related cellular pathways and neural circuits that are involved in schizophrenia-like endophenotypes.
Schizophrenia is a severe, chronic mental illness with a lifetime prevalence of 0.12-1.6% worldwide. Most exciting have been recent associations of schizophrenia with specific genes, such as neuregulin1, however the exact mechanisms of this association are unclear. Our findings indicate that alterations in BACE1-dependent NRG1 signaling lead to schizophrenia-like phenotypes and implicate a novel molecular mechanism in the pathogenesis of schizophrenia and related psychiatric disorders. Given recent reports that impaired NRG1 signaling leads to glutamatergic hypofunction implicated in mechanisms of schizophrenia, we will test the hypothesis that BACE1-dependent impairments in NRG1 signaling modulate NMDA or AMPA receptors pathways. By using RNA interference and conditional BACE1 genetic models we will also analyze in what extent the expression of schizophrenia -like phenotypes depends on activity of BACE1 during development and in adulthood.