Behavioral and synaptic circuit features in a zebrafish
model of Fragile X syndrome.
1. Introduction
Fragile X syndrome (FXS) is the most frequent inherited form of human mental retardation (Turner et al., 1996, Garber et al., 2006). This syndrome is most commonly caused by a triplet repeat expansion (CGG) mutation in the fmr1 gene, which results in silencing of fmr1 transcription, and absence of the FMR1 protein (FMRP).
fmr1 knockout (KO) mice were characterized as having several behavioral profiles similar to those of fragile X patients, including hyperactivity, reduced anxiety-related behavior, and learning and memory deficits (Bakker et al., 1994, Liu et al., 2011). The synapse function is closely correlated with dendritic spine morphology, and synaptic activity.
Indeed, abnormalities of the dendritic spine is a significant neuroanatomical defect in in FXS patients (Hinton et al., 1991) and Fmr1 KO mice (Comery et al., 1997). Therefore, the findings of a spine morphological phenotype indicate a possible defect in synaptic plasticity in FXS that could result in the cognitive disability phenotype.
Long-term potentiation (LTP) and long-term depression (LTD) are the key cellular mechanisms underlying learning and memory processes (Maren and Fanselow, 1995, Brigman et al., 2010). In a previous study, glycine-induced LTP (Gly-LTP) in the CA1 region of the hippocampus was reduced in fmr1 KO mice, suggesting FMRP is required for Gly-LTP in the hippocampus through the activation of N-methyl-d-aspartate (NMDA) receptors and the consequent activation of extracellular signal-regulated kinase (ERK) signal cascade (Shang et al., 2009). On the other hand, more recent studies in the hippocampus have shown that a
lack of FMRP may lead to exaggerated metabotropic glutamate receptor (mGluR) signaling and may underlie the enhanced hippocampal LTD in fmr1 KO mice.
In 2009, den Broeder et al. generated the fmr1 knockout alleles in zebrafish, which provide a new genetic model system to study FXS (den Broeder et al., 2009). However, research analyzing the phenotypic characteristics of Fmr1 KO zebrafish in adult is not well understood. In a previous study we have reported that the physiological function of the telencephalon is involved in the process of fear memory formation in inhibitory avoidance task in zebrafish (Ng et al., 2012a). The present study aimed to further characterize the loss of FMRP on cognitive phenotype by investigating possible differences in cognitive behavior in inhibitory avoidance and synaptic plasticity at Dl-Dm synapse of telencephalon.
2. Materials and methods
2.1. Animals
Zebrafish mutants carrying the fmr1hu2787 allele were obtained from the Wellcome Trust Sanger Institute Zebrafish Mutant Resource. This allele carries a C432T change, which causes a premature termination at codon position 113 (den Broeder et al., 2009). Fish (4–8 months of age) of both sexes were used for these experiments; fmr1 KO and control fish of the TL background were maintained according to standard procedures (Westerfield, 1993) and following guidelines approved by the
Institutional Animal Care and Use Committee (IACUC) of National Taiwan Normal University.
2.2. Preparation of Genomic DNA for Genotyping Using Polymerase Chain Reaction (PCR)
Adult fishes were genotyped by PCR using using Tissue and Cell Genomic DNA Purification Kit (Genemark, Taipei, Taiwan) according to manufacturer protocol. DNA extracted from the tails. Zebrafish genomic DNAs were prepared using Tissue and Cell Genomic DNA Purification Kit (Genemark, Taipei, Taiwan) according to manufacturer protocol.
Optic density (O.D.) value of each sample was measured by a Nano spectrophotometer (NanoDrop Technologies, Inc.Wilmington, DL) at 260 and 280nm to determine the purity and concentration of DNA.
Genotyping of the present study is based on dCAPS assay (derived Cleaved Amplified Polymorphic Sequence). In this assay, a mismatch in PCR primer is used to create restriction endonuclease (RE)-sensitive polymorphism based on the target mutation. To genotype the hu2787 allele, a mismatch has been introduced into the forward primer. During PCR, this mismatch creates an RsaI restriction enzyme site in the amplified product derived from the WT DNA template. The RsaI site is not present in the PCR product containing the hu2787 mutation. A 222-bp PCR product was generated using forward primer (5’-CTA AAT GAA ATC GTC ACA TTA GAG AGG GTA) and reverse primer (5’- TCC
ATG ACA TCC TGC ATT AG). The amplification reaction mixture (50μL) contained 200ng genomic DNA, 0.5 mM of each dNTP, 1 μM of each primer, 1 unit Prozyme DNA polymerase (Protech Enterprise, Taipei, Taiwan) and 1X PCR buffer. The PCR reaction conditions began with a denaturation at 94°C for 4 min, followed by 40 cycles of 94°C for 30 seconds, 60°C for 30 seconds and 72°C for 20 seconds; lastly, 5 min at 72°C. After amplification, the PCR product was digested by RsaI restriction enzyme in 1 X restriction enzyme buffer. Finally, digested PCR products were separated by electrophoresis in 3% agarose gel. The PCR products derived from the WT template were cleaved to 193-and 29-bp DNA fragments.
2.3. Western blot analysis
After the animals were killed, the telencephalon brain region was quickly removed from the skull and homogenized using a T-PER tissue protein extraction reagent kit (Pierce Biotechnology, Inc., Rockford, IL) with the addition of the Halt Protease Inhibitor Cocktail. The protein concentration was determined by the Bradford protein assay, and an equal amount of protein (25 μg per sample) was subjected to SDS–10% PAGE.
The proteins separated on the gel of the SDS–PAGE were transferred to a PVDF membrane (Millipore, Bedford, MA). For the immune-detection, the membrane was first blocked with 5%nonfat milk and 0.05% Tween in PBS for 1 h at room temperature. The primary antibodies used for the detection were rabbit anti-FMRP (1:4,000; Gift from Dr. Willemsen, #758)
antibodies. The membranes were incubated with primary antibodies overnight at 4℃ and, subsequently, with HRP-conjugated secondary antibodies for 1 h at room temperature. Finally, the detected signals were visualized with enhanced chemiluminescence (Bioman Scientific Co. Ltd., Taiwan) and quantitatively analyzed by a LAS3000 digital imaging system (Fujifilm, Tokyo, Japan).
2.4. Preparation of acute telencephalic slices
Acute telencephalic slices preparation was similar to that described previously (Ng et al., 2012b). Briefly, fish were euthanized by exposing them to an ice-cold (0~4℃), artificially oxygenated cerebrospinal fluid (aCSF) solution: 117 mM NaCl, 4.7 mM KCl, 1.2 mM NaH2PO4, 2.5 mM NaHCO3, 1.2 mM MgCl2, 2.5 mM CaCl2, and 11 mM d-(+)-glucose, and their brains were quickly removed under the aCSF solution. Transverse telencephalic slices (300 µm) were prepared using a vibratome (MA752, Campden Instruments Ltd., UK) in ice-cold aCSF.
Slices were then incubated in an aCSF solution that was bubbled continuously with 95% O2/5% CO2 at least 1 h prior to recordings, at room temperature.
2.5. Electrophysiology
Extracellular population spikes (PSs) were recorded using a 64-channel multi-electrode dish (MED64) system (Alpha MED Sciences, Tokyo, Japan) with a sample rate of 20 kHz. Recordings were performed with an 8 x 8 array of planar microelectrodes, each 20 x 20 μm in size, with an inter-electrode spacing of 100 μm. Telencephalic slices were placed in a recording chamber and perfused with the aCSF (32℃) at a flow rate of 1-2 ml/min via a peristaltic pump (Gilson Minupuls 3, Villiers Le Bel, France). A nylon mesh and a stainless steel wire were used to secure slice position and contact with electrodes during perfusion.
Stimulus intensity was adjusted to evoke 40 – 50% maximal stimulation.
Test stimuli were delivering 0.2 ms pulses every 20 s. and responses were recorded for 15 min prior to beginning the experiment treatments to assure stability of the response. Every three consecutive responses were pooled and averaged for data analysis.
Basal synaptic transmission was measured by plotting the current applied to the stimulating electrode (40–150 μA) against the amplitude of population spike responses to generate input–output curves (I/O curves).
Paired-pulse facilitation was assessed by applying pairs of stimuli at varying inter-pulse intervals (20, 50, 100, 150, and 200 ms). The paired pulse ratio (PPR) was determined by calculating the ratio of the average amplitude of the second response to the first. Each trace corresponds to an average over 6 trials. After stable baseline recording, LTP was elicited by HFS protocols consisting of three stimulus trains of 100 pulses (at 100 Hz) with 20 s inter-train intervals; whereas, LTD was induced by low
frequency stimulation consisting of 1 Hz stimulation for 20 min. The magnitude of both LTP and LTD was measured as an average of 10 min at the end of the recording period post-induction.
2.6. Behavioral test
Prior to the initiation of behavioral protocols, zebrafish were separated into individual tanks a few days before the experiment. To elucidate the behavioral effects of FMRP deficiency, fish (10 WT and 12 KO) were initially assessed with a light/dark test, then tested 48 h later in an inhibitory avoidance test, and finally examined in an open-field test.
2.6.1. Light/ Dark test
To examine anxiety-related behaviors, fish were tested in a light/dark test conducted in a rectangular acryl tank (7 cm height × 18 cm Length × 9 cm width) divided into two equally sized compartments that were demarcated by black and white coloration. The tank water level was maintained at 3 cm. Before testing, the subjects were carefully placed in the white compartment. Then, the fish were allowed to swim freely between the two compartments without a sliding door for 5 min. Behavior was recorded using a Logitech S5500 camera and Logitech QuickCapture software. The proportion of the trial that the animal spent in the white compartment was recorded. Reduced exploration of the white compartment in this test reflects high anxiety states.
2.6.2. Inhibitory avoidance task (IA)
In this study, we utilized a novel inhibitory avoidance paradigm, which took advantage of the natural preference of zebrafish for a deep environment (Darland and Dowling, 2001). The tested fish had to refrain from swimming from a shallow compartment into a deep compartment in order to avoid receiving an electric shock. The apparatus constructed for this experiment was an acryl chamber, with dimensions of 28 x 12 x 17 cm, which was divided into two equal-size compartments by a white, opaque guillotine door and designated herein as a shallow and deep compartment. The water levels of the shallow compartment and the deep compartment were maintained at 2 and 8 cm, respectively. Two metal plates were placed in the deep compartment to serve as electrodes for delivering a uniform and repetitive mild electric shock as an aversive stimulus (pulsed 10 ms on and 10 ms off). The fish behavior was recorded using a Logitech S5500 camera and Logitech QuickCapture software. The behavioral testing procedure that is commonly adopted in most experiments consists of a habituation session, a training session and a testing session. In the current study, during the habituation session, zebrafish were pre-exposed individually into the behavioral chamber.
They were placed in the shallow chamber for 5 min; the white, opaque guillotine door was then removed, and the fish were allowed to swim freely between the two compartments for another 5 min. In the training session, the fish were placed in the shallow compartment, allowing them to swim for 1 min before the guillotine door was opened. Once the fish entered the deep compartment, the guillotine door was closed, and a mild
electric shock was applied to the deep compartment for 5 s. The testing session was performed on the next day with the same procedure, and the conditions were similar to those in the training session except that the electric shock was omitted. The trained fish were placed in the shallow chamber again with the guillotine door opened. The time taken by the fish to enter the deep compartment was recorded to a maximum of 300 s as the retention latency.
2.6.3. Open-field test
At the end of the retention test, the animals were placed into a transparent cylindrical tank (20 cm in height and 22 cm in diameter) for 10 min to test their spontaneous motor activity. The water level was maintained at 4 cm. Behavior was detected using an EthoVision video tracking system (Noldus Information Technology, Leesburg, VA, USA).
Total distance swam and mean speed were measured for statistical analyses.
2.7. Drug application
All drugs were prepared fresh from stock solutions, diluted in aCSF, and applied by superfusion over the slice. Stock solutions of (R,S)-3,5-Dihydroxyphenylglycine (DHPG, Group I mGlu receptor agonist) were purchased from Ascent Scientific (Bristol, UK) and made up in distilled water.
2.8. Statistical analyses
Statistical analysis was performed using SPSS version 12.0 (SPSS, Chicago). All values are reported as the mean ± the SEM. Genotypic differences in basal synaptic transmission, synaptic plasticity, anxiety-like behavior, and locomotor activity were assessed by independent t-tests.
Inhibitory avoidance training and test latencies for each genotype were compared using paired t-tests, and comparisons between groups were performed using independent t-tests. We considered p <0.05 to be statistically significant.
.
3. Results
3.1. Anxiolytic-like response in fmr1 KO zebrafish
The light/dark test has been proposed as a model of anxiety-like behavior in zebrafish. The time spent in white compartment and the numbers of midline crossings were analyzed for each fish. As illustrated in Figs. 3-2, we found a significant genotypic difference in both measures.
Fmr1 KO fish spent more time in the white compartment (Fig. 3-2A, p<0.01) and had greater numbers of midline crossings compared to wild-type fish (Fig. 3-2B, p<0.01), indicating lower anxiety and increased locomotion in KO fishes.
3.2. Impaired inhibitory avoidance learning in fmr1 KO zebrafish
The inhibitory avoidance test has been extensively used for assessing memories of aversive experiences. In this study, fmr1 KO and wild-type fish were trained in the inhibitory avoidance learning task, and latency to enter the deep compartment was assessed 24 h after training.
As illustrated in Figs. 3-3, the difference between the latencies in the training and test sessions for wild-type was statistically significant (Fig.
3-3, n=10, p<0.05). In contrast, no significant difference was observed in the fmr1 KO fishes. Additionally, the retention test was significantly different (p <0.05) between wild-type and fmr1 KO fish.
3.3. Hyperactivity in Fmr1 KO zebrafish
Hyperactivity is the most common symptom of FXS patients and fmr1 KO mice. To determine whether genotypic differences in locomotor activity were present between genotypes, the total distances swam and mean speeds of fmr1 KO and wild-type fish were calculated in an open field apparatus for 5 min. As shown in Fig. 3-4, the total distances moved and the mean speeds of fmr1 KO fish were higher than those of wild-type fish (p <0.001 for both outcomes).
3.4. Basal synaptic transmission and Paired-pulse facilitation
Basal synaptic transmission at the Dl-Dm synapse was measured by field potential responses to increasing stimulation intensities. As shown in Fig. 3-5A, the amplitude of the population spikes obtained from
wild-type and frm1 KO slices were compared, and no significant difference between genotypes was noted. Additionally, paired pulse facilitation (FFP) was measured in slices from both genotypes. As shown in Fig. 3-5B, pairs of presynaptic fiber stimulation pulses delivered at interpulse intervals of 20, 50, 100, 150 and 200 milliseconds evoked nearly identical amounts of PPF in slices from wild-type and fmr1 KO zebrafish. We suggest that basal glutamatergic transmission and presynaptic function at the Dl-Dm synapse remain normal in fmr1 KO zebrafish.
3.5. Synaptic plasticity in fmr1 KO zebrafish
In zebrafish, FMRP is highly expressed in the telencephalon (den Broeder et al., 2009), an important brain region involved in synaptic plasticity and learning and memory processes. This fact raises an intriguing possibility that FMRP is involved in synaptic plasticity. We next examined whether the loss of FMRP function in zebrafish was related to modulation of synaptic plasticity; to do this, long-term potentiation (LTP) and long-term depression (LTD) were characterized.
As shown in Figs. 3-6, LTP was induced by a protocol of three trains high frequency stimulation. LTP magnitude was significantly reduced in fmr1 KO zebrafish (181.0 ± 7%, n = 9 in wild-type vs. 146.8 ± 6%, n = 10 in frm1 KO, p < 0.05; Fig. 3-6B). LTD is a long-lasting decrease in the synaptic response of the same synapses following prolonged low-frequency stimulation (LFS). LFS-induced LTD was enhanced in slices from fmr1 KO fish compared to slices from wild-type fish (104.3 ±
7%, n=4 in wild-type vs. 76.5 ± 5%, n = 6 in frm1 KO, p < 0.05; Fig.
3-7B). These findings suggest that FMRP plays an important functional role in regulating telencephalic synaptic plasticity in zebrafish.
4. Discussion 4.1. Summary
Fragile X syndrome (FXS), the most common form of inherited mental retardation, is caused by loss of the fragile X mental retardation protein (FMRP). To understand the molecular and cellular pathogenesis of FXS, the disease has been successfully modeled in mice (Bakker et al., 1994), Drosophila (Pan et al., 2008) and zebrafish (Eadie et al., 2009). In the present study, using fmr1 KO zebrafish, we were able to investigate the functional role of the fmr1 gene in mediating cognitive behavior and synaptic plasticity at the Dl-Dm synapse in the telencephalon of zebrafish. Our results can be summarized as follows: (1) fmr1 KO fish exhibit anxiolytic-like behavior, impaired emotional learning, and hyperactivity, and (2) electrophysiological recordings from telencephalic slice preparations of fmr1 KO fish showed markedly reduced LTP and enhanced LTD compared with wild-type fish. This study provides the first evidence that FMRP is involved in cognitive functions and telencephalic synaptic plasticity in zebrafish and suggests that zebrafish are a new genetic model system to study Fragile X syndrome (FXS).
4.2. Detailed discussion
Previous behavioral studies have demonstrated that fmr1 KO mice replicate many of the human behavioral features of FXS, including hyperactivity, learning deficits, impaired social interaction, and abnormal anxiety-related responses (Bakker et al., 1994). Furthermore, behavioral profiles are a critical first step toward understanding the function of frm1.
Here, we performed a series of behavioral analyses on the fmr1 KO zebrafish that included the light/dark test, the inhibitory avoidance test, and the open-field test to further characterize the consequences of the absence of FMRP.
Interestingly, significant behavioral differences were detected in the light/dark test. Compared with wild-type fish, fmr1 KO fish had reduced anxiety-related responses in the light/dark test. Our results are remarkably consistent with previous studies (Peier et al., 2000, de Diego-Otero et al., 2009, Eadie et al., 2009, Liu et al., 2011, Goebel-Goody et al., 2012) in which the loss of FMRP has been reported to be related to anxiolytic responses in mice. Moreover, fmr1 KO zebrafish show a significantly greater number of crossed lines in the lit compartment, which significantly contributed to locomotor activity. Thus, hyperactivity may be present in fmr1 KO zebrafish.
Cognitive impairment is a common symptom of FXS patients and FXS mouse models. For instance, Liu et al. (2011) noted impaired inhibitory avoidance acquisition in the fmr1 KO mice (Liu et al., 2011).
Here, using an inhibitory avoidance (IA) test, we evaluated whether Fmr1 null mutant zebra fish exhibited learning and memory impairments.
Consistent with the notion that FMRP is involved in certain types of learning and memory, we found a significant impairment in the inhibitory avoidance task in fmr1 KO fishes. These results suggested that the absence of FMRP might disrupt the detection abilities of and/or the response of the brain’s fear system. After the retention test in the IA task, the animals were subjected to an open-field test; activity in the open-field is often used as a measure of exploration in zebrafish. The distance traveled and the mean speed in the open-field were significantly higher in fmr1 KO fishes. Our behavioral analyses of the frm1 KO fish in the light/dark and open-field tests supports previously reported results (Bakker et al., 1994, Peier et al., 2000, Liu et al., 2011), suggesting that the absence FMRP expression leads to hyperactivity or increased exploratory behavior.
According to neuroanatomical and behavioral analyses, the telencephalic pallium is a key component of the fear circuitry of teleost fish. For example, goldfish with lesions to the telencephalon have impaired avoidance conditioning (Portavella et al., 2002, Portavella et al., 2004). In a previous study, we reported that the physiological function of the telencephalon is involved in the process of fear memory formation in inhibitory avoidance tasks in zebrafish (Ng et al., 2012a). Furthermore, electrophysiological evidence has demonstrated that the intratelencephalic connections between the lateral and medial pallium, and the Dl-Dm synapse, play important roles in the synaptic plasticity of the zebrafish (Ng et al., 2012b). In present study, by using Western blot analysis, we show that FMRP is dominantly expressed in the
telencephalon lysate (see Fig. 3-1B). Because our behavioral results suggested that the lack of FMRP caused inhibitory avoidance learning deficits, we hypothesized that FMRP may play an important functional
telencephalon lysate (see Fig. 3-1B). Because our behavioral results suggested that the lack of FMRP caused inhibitory avoidance learning deficits, we hypothesized that FMRP may play an important functional