AAV-Mediated Overexpression of Neuroserpin in the Hippocampus Decreases PSD-95 Expression but Does Not Affect Hippocampal-Dependent Learning and Memory

Abstract

Neuroserpin is a serine protease inhibitor, or serpin, that is expressed in the nervous system and inhibits the protease tissue plasminogen activator (tPA). Neuroserpin has been suggested to play a role in learning and memory but direct evidence for such a role is lacking. Here we have used an adeno-associated virus (AAV) vector expression system to investigate the effect of neuroserpin on hippocampal-dependent learning and memory in the young adult rat. A FLAG-tagged neuroserpin construct was initially characterized by in vitro transcription/translation and transfection into HEK293 cells and shown to interact with tPA and be targeted to the secretory pathway. Targeted injection of a chimeric AAV1/2 vector expressing FLAG-neuroserpin resulted in localized overexpression in the dorsal hippocampus. Neuroserpin overexpression led to the appearance of an unstable neuroserpin:tPA complex in zymographic assays consistent with interaction with endogenous tPA in vivo. Rats overexpressing neuroserpin also showed a significant decrease in the levels of postsynaptic density protein 95, a major postsynaptic scaffolding protein. Three weeks after injection, a range of behavioural tests was performed to measure spatial and associative learning and memory, as well as innate and acquired fear. These tests provided no evidence of a role for neuroserpin in hippocampal-dependent learning and memory. In summary this study does not support a role for neuroserpin in hippocampal-dependent learning and memory in young adult rats but does suggest an involvement of neuroserpin in hippocampal synaptic plasticity.

The serine protease inhibitor neuroserpin is required for normal synaptic plasticity and regulates learning and social behavior

“…Our electrophysiological data showing a clear reduction in excitatory postsynaptic potential (a form of long-term plasticity and a candidate mechanism for memory formation), provide further evidence for a synaptic function of neuroserpin (Borges et al 2010;Tsang et al 2014). Whether the alteration in LTP is cause or consequence of the reduced spine-synaptic density observed in neuroserpin-deficient mice needs to be further investigated.…”

“…In contrast to postsynaptic markers, expression of presynaptic marker proteins does not seem to depend on neuroserpin levels, as we did not observe changes in expression of the presynaptic marker proteins synaptophysin, SNAP25, or synapsin-I in hippocampi from neuroserpin-deficient mice. Similarly, Tsang et al (2014) did not detect changes in expression of synapsin-I upon neuroserpin overexpression in adult rat hippocampi.…”

“…An influence of neuroserpin on spine morphology has already been described in primary neurons, where overexpression of neuroserpin led to a shift from mushroom-type to thin spines (Borges et al 2010). Similarly, in rat hippocampus, AAV-mediated overexpression of neuroserpin led to decreased PSD-95 expression (Tsang et al 2014), a postsynaptic scaffold protein required for synapse stabilization and modulation of synaptic plasticity (Ehrlich et al 2007;Cane et al 2014). Interestingly, we found increased levels of PSD-95 protein in hippocampi from neuroserpin-deficient mice, indicating an inverse correlation between neuroserpin and PSD-95 levels.…”

“…Because of the essential role of the scaffolding postsynaptic protein PSD-95 in spine stability, and since decreased expression of PSD-95, with unaltered levels of the presynaptic protein synapsin-I, have been reported in rat hippocampus following AAV-mediated neuroserpin overexpression (Tsang et al 2014), we assessed levels of PSD-95 and of three different presynaptic marker proteins in hippocampi of adult Ns−/− mice and control littermates by Western blot analysis. Whereas PSD-95 levels were significantly increased in Ns−/− mice (wt = 1.0 ± 0.22; Ns−/− = 1.68 ± 0.37; P = 0.027) expression of synaptophysin, SNAP25 (synaptosomal-associated protein 25), and synapsin-I were not altered (synaptophysin: wt = 1.0 ± 0.13; Ns−/− = 1.18 ± 0.08; P = 0.100; SNAP25: wt = 1.0 ± 0.22; Ns−/− = 1.02 ± 0.40; P = 0.881; synapsin-I: wt = 1.0 ± 0.17; Ns−/− = 0.96 ± 0.13; P = 0.494) (Fig.…”

“…Recently, the hypothetical role of neuroserpin in synaptic formation and plasticity has been experimentally substantiated. Overexpression of neuroserpin in primary neurons leads to increased dendritic arborization and altered dendritic spine shape (Borges et al 2010), whereas in rat hippocampus neuroserpin overexpression results in reduced expression of postsynaptic density protein 95 (PSD-95) without impairment in hippocampaldependent learning and memory (Tsang et al 2014). Moreover, expression levels of N-cadherin, an adhesion protein implicated in synapse formation, are modulated by neuroserpin (Lee et al 2008).…”