Role of hippocampal NMDA receptors in a mouse model for fragile x mental retardation syndrome

Abstract

The ‘fragile x mental retardation syndrome’ (FRAX) is a prevalent form of inherited mental retardation. It appears developmentally delayed, and it is associated with hyperactivity, attention deficit disorders and autism-like behaviour (de Vries et al. 1997). FRAX is typically caused by a trinucleotide repeat expansion in the X-linked FMR1 gene, which prevents the expression of its encoded protein, named ‘fragile X mental retardation protein’ (FMRP) (O'Donnell & Warren, 2002). FMRP modulates the function of a large number of other proteins, and absence of FMRP will disclose most symptoms of FRAX in humans. So far there is no effective therapeutical treatment. An important step for the understanding of the pathophysiology of FRAX was the creation of an animal model, in which the FMR1 gene was isolated and subsequently silenced through generation of a FMR1 knockout mouse (FMR1 KO1) (The Dutch-Belgian Fragile X Consortium, 1994). These mice revealed learning deficits and anatomical changes at dendrites similar to those in humans suffering from FRAX. Hence, it was suggested that the behavioural and cognitive deficits in humans could be related to dysfunctions at the level of dendritic spines and a potentially altered synaptic development and plasticity (Bear et al. 2004). Several studies examined a potential change of synaptic strength in different brain areas in the FMR1 KO1 mouse, and in summary two major findings were published: (1) an increase in synaptic long-term depression (LTD), mediated by group 1/5 metabotropic glutamate receptors (mGluR) (Huber et al. 2002), and (2) an impaired AMPA receptor (AMPAR) mediated long-term potentiation (LTP) (Hu et al. 2008), or a complete loss of LTP (Desai et al. 2006). Even no change of LTP has been reported (Godfraind et al. 1996), and this variability might be explained by the fact that the studies were performed in various brain areas under different stimulation conditions, and each study used FMR1 KO1 mice at a specific age. The paper by Pilpel et al. (2009) in this issue of The Journal of Physiology presents surprising new data on the developmental changes of glutamatergic neurotransmission and plasticity in the FMRP-silenced hippocampus by use of an alternative mouse model. The ‘FMR1 KO2’ mouse was more recently generated by deleting both the promoter and exon1 of the FMR1 gene (Mientjes et al. 2006). As a result, these mice are deficient for both, the FMR1 RNA and the FMRP protein, so in contrast to the previously used ‘FMR1 KO1’ mouse the remaining level of mRNA is effectively null (Mientjes et al. 2006). Here, Pilpel et al. show for the first time that pyramidal neurons in area CA1 of FMR1 KO2 mice express an enhanced NMDA receptor (NMDAR)-dependent LTP, while the level of evoked LTD is unaltered. Importantly, the LTP effects are age dependent, since they are observed only in animals at P14–16, but not after 6–8 weeks. In addition, the authors show that this increased LTP is associated with a reduced ratio in the amplitudes of evoked AMPA/NMDA receptor mediated EPSCs. A biochemical analysis of synaptoneurosomal fractions in the FMR1 KO2 mice confirmed on the protein level a relatively stronger contribution of NMDARs compared to AMPARs in glutamatergic transmission during early postnatal development (P14). The results are very interesting, since all studies mentioned above used the classical FMR1 KO1 mouse model and none of them observed an increased LTP mediated by NMDARs. Here, Pilpel et al. propose a new role for NMDARs in hippocampus, limited to the first two postnatal weeks in the FMR1 KO2 mice. The narrow window of altered synaptic plasticity in the FMR1 KO2 mice might have an impact for any potential therapy related to FRAX. Yet, the effects of FMRP silencing on LTP/LTD seemed to be critically dependent on the stimulation conditions: Pilpel et al. also tested a theta-burst stimulus (TBS), designed to mimic hippocampal activity during exploratory behaviour. Here the level of LTP was not different, and knock-out animals revealed only a small impairment during spatial memory tasks. Thus, the complete understanding of the pathophysiology of FRAX requires further experiments including a comparison of the behavioural, biochemical and physiological phenotypes between these different mouse lines.