Fragile X is a common mental retardation syndrome that results from the silencing of a single gene, FMR1, on the X chromosome [1, 2]. The loss of the FMR1 gene product, FMRP, leads to mental retardation in males and to a range of deficits in heterozygous females. The syndrome has an autistic-like behavioral profile with hyper-arousal, self-stimulation, tactile-defensiveness and hypersensitivity to sensory stimuli [1, 3]. There is also an increased prevalence of childhood seizures [4, 5]. Physical features of the syndrome usually include enlarged testicles and prominent ears [4]. Neuroanatomical features include abnormal dendritic spine morphology [6–8], as well as alterations in the volume of the lateral ventricles and other structures [5, 9]. FMRP is an RNA binding protein [10] that may act at the level of mRNA translation or translocation [11, 12].

The fragile X mouse model for this disorder was generated by insertional mutagenesis of the mouse Fmr1 gene that eliminated expression of the mouse protein, Fmrp [3]. The fragile X mouse exhibits features that resemble elements of the fragile X syndrome. These include enlarged testicles, mildly impaired learning behavior [3, 13], an increased susceptibility to audiogenic seizures [14, 15], altered behaviors associated with anxiety [16] and abnormal dendritic spine morphology [17–19]. The increase in susceptibility to seizures and spine dysgenesis may be due to the role of Fmrp as an activity-dependent regulator of translation [11, 20, 21]. Increased seizure susceptibility may be related to increased group 1 metabotropic receptor (mGluR) response [22] and to reduced GABA_A receptor expression [23]. In addition, long term depression (LTD) is enhanced in the fragile X mouse [24] and it has been proposed that this increase as well as spine dysgenesis are related to malfunction of a signaling cascade initiated by group 1 mGluRs in which Fmrp regulates translation of specific mRNAs [25].

Absence of Fmrp in the mouse model has also been shown to increase cerebellar LTD and to attenuate eye blink conditioning [26]. Analysis of the anterior piriform cortex showed reduced long term potentiation (LTP) in older animals [27]. No changes in paired pulse facilitation in hippocampal circuits have been noted when the interpulse interval was 40 ms [28] or 50 to 1600 ms [27]. However, these studies have examined the monosynaptic connections between the Schaffer collaterals and their postsynaptic target, the CA1 pyramidal cells. Here we have examined the integrative properties of the whole hippocampal formation and studied short-term plasticity within this structure and how it is affected by taurine.

Animals: All mice used in this study were two-month-old FVB/NJ males. Mice were housed in groups of three in a pathogen-free room, maintained on a 12 hr light/dark cycle and given food and water ad libitum. All procedures were approved by the Institutional Animal Care and Use Committee of the College of Staten Island/CUNY, and were in conformity with the National Institutes of Health Guidelines. The number of mice sufficient to provide statistically reliable results was used in these studies.

Slice preparation: Hippocampal slices (400 µm) were prepared using an automated Leica tissue chopper. Artificial cerebrospinal fluid (aCSF) consisted of the following composition (mM): NaCl 124, KCl 3, CaCl₂ 2.4, MgSO₄ 1.3, NaH₂PO₄ 1.25, NaHCO₃ 26, and glucose 10 (gassed with 95% O₂ / 5% CO₂, pH = 7.4). Slices were prepared on ice-cold aCSF and calibrated for 45min at room temperature (RT) in oxygenated aCSF prior to recording.

Recording and stimulation: Extracellular recordings of evoked excitatory postsynaptic potentials (EPSPs) were measured by placing the stimulating bipolar platinum electrodes in perforant path close to the suprapyamidal blade of the DG in the outer third of the molecular layer. Recording electrodes (tip impedance 1-5 MΩ) were placed in the middle or outer region of the molecular layer of CA1 to record EPSPs. This stimulation paradigm allowed us to assess both the evoked responses and the overall integrative properties of the whole hippocampus. Stimulation parameters for test pulses (0.05 Hz) were standardized across slices by setting current and pulse width to evoke a fixed percentage of maximal EPSP amplitude in any given data set. Similarly, stimulus intensity for plasticity induction was standardized for all data sets. The stimulus current that produced about 40% of maximal responses (usually 100– 150 μA) was used throughout the experiment for monitoring test pulses. Data were sampled and digitized at 100 kHz with a band-pass filtered between 0.3 Hz and 3 kHz.

Statistical analysis: Statistical significance was determined by Student's t-test. Each value was expressed as the mean ± SEM. Differences were considered statistically significant when the calculated p value was less than 0.05.

We have previously shown that GABA_A receptor expression is down-regulated [23] and cortical GABAergic interneurons are less abundant in Fmr1-KO mice [29, 30], suggesting reduced GABA signaling in these mice.

To assess electrophysiologically the functional significance of reduced expression of the GABA_A and the alterations of the inhibitory system in the Fmr1-KO hippocampus, we performed electrophysiological recording from the CA1 region of the hippocampus after paired pulse stimulation of the perforant path. This allowed us to determine the overall integrative properties of neuronal circuits within the hippocampal formation.

The finding that Fmr1-KO mice have altered paired pulse facilitation indicates that the mechanism underlying short-term plasticity in the hippocampus is compromised (Figure 1). Several factors could be at the origin of these alterations. We have shown that the GABAergic inhibitory system is altered in these mice [23]. Additionally, we showed that the expression of somatostatin is down-regulated in Fmr1-KO mice [30]. Furthermore we showed that the reduced expression of this neuropeptide is consistent with both central and peripheral features of the fragile X syndrome.

There are several studies indicating the interaction of taurine with GABA_A and glycine receptors acting as a modulator [31–36] or an agonist [37–44]. In heterologously expressed GABA_A receptors composed of either α1β3 or α1β3γ 2 subunits, taurine is a full or partial agonist, respectively, while both receptor types exhibit similar taurine sensitivities [44]. Interestingly we found that the Fmr1-KO hippocampal slices are irresponsive to bath application of taurine. While in the WT mice, taurine addition resulted in a significant decrease in the amplitude of the facilitation; in the Fmr1-KO, pair pulse stimulation resulted in a consistent depression of the second evoked response and addition of taurine had no effect. Taurine is a GABA_A receptor agonist and modulator [31–44]. Its application to hippocampal slices result in activation of GABA_A receptors and hyperpolarization of GABA_A receptors-expressing neurons, which comprises all principal cells of the hippocampus. This leads to reduced threshold for neuronal firing. Thus, under these conditions paired-pulse stimulation resulted, as expected, in a depression of neuronal response. The finding that slices from Fmr1-KO mice show abnormal pair pulse response and are insensitive to taurine indicates that the inhibitory GABAergic system is compromised in these mice and that the GABA_A receptors expressed in these mice lack the binding site for taurine. Consistent with this we showed that Fmr1-KO mice have reduced β subunit of the GABA_A receptors and taurine could presumably bind to this subunit (L’Amoreaux and El Idrissi, 2010; this volume). Furthermore, disrupting the gene for β3 subunit of the GABA_A receptor in mice produces electroencephalographic abnormalities, seizures, learning and memory deficits, poor motor skills on a repetitive task and hyperactivity features all common to fragile X syndrome [45]. Beta 3 subunit phosphorylation acts as a regulatory switch to control cycling of GABA_A receptors into and out of the plasma membrane [45]. This may allow neurons to specifically modify cell surface receptor insertion and/or endocytosis rates to control the efficacy of inhibitory synaptic transmission. The finding that Fmr1-KO mice have reduced β subunit of the GABA_A receptors may explain the abnormal response to pair pulse stimulation and the insensitivity to taurine application.

Fmr1-KO mice have altered GABAergic system. Here, we show electrophysiologically that alterations in the GABAergic system may underlie the abnormal response to pair pulse stimulation, the cellular analogue of short-term plasticity. Furthermore, we showed that taurine regulates short-term plasticity in the hippocampus. Taurine application to slices resulted in a shift from pair pulse facilitation to pair pulse depression in WT but not Fmr1-KO mice presumably lacking the binding site for taurine on GABA_A receptors.