SEARCH

SEARCH BY CITATION

Keywords:

  • autism;
  • CaMKII;
  • hippocampus;
  • melatonin;
  • phosphorylation;
  • valproic acid

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Competing interests
  9. References

Lower global cognitive function scores are a common symptom of autism spectrum disorders (ASDs). This study investigates the effects of melatonin on hippocampal serine/threonine kinase signaling in an experimental ASD model. We found that chronic melatonin (1.0 or 5.0 mg/kg/day, 28 days) treatment significantly rescued valproic acid (VPA, 600 mg/kg)-induced decreases in CaMKII (Thr286), NMDAR1 (Ser896), and PKA (Thr197) phosphorylation in the hippocampus without affecting total protein levels. Compared with control rats, the immunostaining of pyramidal neurons in the hippocampus revealed a decrease in immunolabeling intensity for phospho-CaMKII (Thr286) in the hippocampus of VPA-treated rats, which was ameliorated by chronic melatonin treatment. Consistent with the elevation of CaMKII/PKA/PKC phosphorylation observed in melatonin-treated rat, long-term potentiation (LTP) was enhanced after chronic melatonin (5.0 mg/kg) treatment, as reflected by extracellular field potential slopes that increased from 56 to 60 min (133.4 ± 3.9% of the baseline, < 0.01 versus VPA-treated rats) following high-frequency stimulation (HFS) in hippocampal slices. Accordingly, melatonin treatment also significantly improved social behavioral deficits at postnatal day 50 in VPA-treated rats. Taken together, the increased phosphorylation of CaMKII/PKA/PKC signaling might contribute to the beneficial effects of melatonin on autism symptoms.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Competing interests
  9. References

Autism spectrum disorders (ASDs) are a heterogeneous group of bio-neurological developmental disorders that display common behavioral symptoms, including pervasive impairments in social interactions [1-4] that are accompanied by reduced global cognitive function scores [5, 6]. Synaptic alterations may underlie the abnormal social and cognitive behaviors that are observed in humans with ASDs [7, 8]. Hence, efficacious pharmacological and behavioral interventions for the abnormal symptoms of ASD are needed.

Protein kinases are involved in learning and memory and are crucial for memory and synaptic plasticity [9, 10]. However, little is known regarding the role of hippocampal protein serine/threonine kinases, which are critical for cognitive function, during the pathological process of ASD. We and others have previously reported that changes in the phosphorylation levels of Ca2+/calmodulin-dependent protein kinase II (CaMKII), Ca2+/phospholipid-dependent protein kinase C (PKC), and cyclic AMP-dependent protein kinase A (PKA) signaling in the brain hippocampus are associated with deficits in learning and memory [10-12]. CaMKII is highly enriched in the postsynaptic densities of excitatory synapses and remains constitutively active via autophosphorylation at threonine-286, which facilitates synaptic efficacy [12]. In addition, PKC isoforms phosphorylate the NMDA receptor (NMDAR), which is required to enhance CaMKII-dependent long-term potentiation (LTP) [13-15]. Some studies have suggested an increased prevalence of developmental delays and cognitive dysfunctions in children who have been exposed to anticonvulsant drugs in utero [16, 17]. Moreover, these studies have indicated that alterations in NMDA tone may be responsible for the excessive self-grooming observed in an animal model of ASD [18]. However, the precise intracellular biochemical basis underlying ASD-mediated cognitive deficits has not been fully elucidated.

Melatonin (N-acetyl-5-methoxytryptamine), a pineal hormone synthesized from serotonin [19, 20], has favorable effects on sleep, mood, and cognitive impairment [21-23]. Emerging evidence indicates that the effects of melatonin administration in ASD extend far beyond improving circadian rhythm sleep disorders and may also be associated with improvements in mental, neurological, and other behavioral disorders [20, 21]. Indeed, increasing evidence indicates that pineal endocrine hypofunction and alterations in the melatonin pathway increase the susceptibility to autism [19, 22, 23]. Melatonin treatment improves spatial learning and memory in neurodegenerative rodent models [24, 25]. Clinically, patients who are chronically treated with melatonin show significantly better cognitive performance, which suggests that melatonin may be a useful add-on drug for the treatment of mild cognitive impairments [26, 27]. However, the potential mechanisms that underlie the beneficial effects of exogenous melatonin in individuals with ASD are still not well understood, and biochemical and pathophysiological studies are needed to elucidate these biological mechanisms.

This study aimed to evaluate the effects of chronic melatonin treatment on disturbances of the serine/threonine kinases signaling system, synaptic efficacy, and autism-like social behavioral disorders. We found that melatonin treatment attenuated the aberrancies in hippocampal CaMKII/PKA/PKC signaling, and this attenuation paralleled reductions in autism-like behaviors in VPA-treated rats.

Material and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Competing interests
  9. References

Animals and prenatal valproic acid treatment

Sprague–Dawley (SD) rats weighing 200–230 g were housed under climate controlled conditions on a 12-hr light/dark cycle (08:00–20:00 hr) and provided with standard food and water. An acclimation period of at least 1 wk was provided before initiating the experimental protocols. All experimental protocols and animal handling procedures were performed in accordance with the guidelines of the National Institutes of Health (NIH, USA) for the care and use of laboratory animals, and these protocols were approved by the Committees for Animal Experiments of Zhejiang University in China and Tohoku University in Japan. Female rats were mated, pregnancy was determined by the presence of a vaginal plug, and the first observation of the vaginal plug was taken as embryonic day 1 (E1). Treated rats received a single intraperitoneal injection of 600 mg/kg sodium valproic acid (VPA, 2-propylpentanoic acid sodium salt; Sigma, St. Louis, MO, USA) on E12.5. Litter sizes, pup body weights, and the general health of the mothers and pups were unchanged by treatment, indicating that the rearing conditions of the treated rats were normal. Dams were housed individually and allowed to raise their own litters. The offspring were weaned on postnatal day (PND) 22, and male and female offspring were then housed separately. The experiments were performed on the male offspring. Melatonin (Sigma, St. Louis, MO, USA) was administered orally (p.o.) using a metal gastric zonde. The doses of melatonin administered were 1.0 or 5.0 mg/kg in volumes of 1 mL/100 g body weight. The rats were given melatonin (1 or 5 mg/kg, dissolved in ethanol before being diluted with saline) once daily at 9:00 hr for 28 consecutive days beginning on PND22, after weaning. Additionally, BTBR T+tf/J mice were obtained from the Model Animal Research Center of Nanjing University.

Behavioral analyses

Locomotor activity was measured as previously described [28, 29]. At PND50, control and VPA-treated rats were housed individually in standard plastic cages and positioned in an automated open-field activity monitoring system that used digital counters employing infrared sensors (DAS system, Neuroscience Inc., Tokyo, Japan), and locomotion was assessed every hour for 1 day.

An object recognition task that is based on the tendency of rodents to discriminate between familiar and new objects [30] was used to evaluate recognition memory. During the acquisition phases, two of the same objects were placed in symmetric positions in the center of the chamber for 10 min. One hour after the acquisition phase training, one of the objects was replaced with a novel object, and exploratory behavior was assessed for 5 min. After each session, the objects were thoroughly cleaned with 75% ethanol to prevent odor cues. Exploration of an object was defined as rearing directed toward, or sniffing of, the object at a distance of less than 1 cm and/or touching the object with the nose. Discrimination of spatial novelty was assessed by comparing the difference in exploration times for the novel (right) and familiar objects (left). The total time spent exploring both objects was quantified, which allowed us to adjust exploration times for differences in total exploration time.

Rats were tested for social interaction behavior in an automated three-chambered social approach apparatus made of clear Plexiglas (30 × 40 × 40 cm) [31]. There were doorways in each of the two dividing walls that controlled access to the side chambers. The sociability test involved an unfamiliar rat and a familiar rat and was performed at PND50 in VPA-treated rats that were treated with or without melatonin (1 mg/kg; 5 mg/kg). During the social preference test, the doors of both chambers were opened, and the experimental rat was allowed to explore all three chambers for 10 min. The time spent exploring the circles around the plastic container containing either the familiar (stranger 1) or novel (stranger 2) rats and the number of entries into each chamber were scored for each experimental rat. The rats were decapitated under deep anesthesia after the last behavioral test, and their brains were removed for further analysis.

Western blotting analysis

Hippocampal samples were homogenized in buffer containing 50 mm Tris-HCl (pH 7.4), 0.5% Triton X-100, 4 mm EGTA, 10 mm EDTA, 1 mm Na3VO4, 30 mm sodium pyrophosphate, 50 mm NaF, 100 nm calyculin A, 50 μg/mL leupeptin, 25 μg/mL pepstatin A, 50 μg/mL trypsin inhibitor, and 1 mm dithiothreitol [32]. Insoluble material was removed by centrifugation at 13,000 g for 30 min. After determining the supernatant protein concentration using Bradford's solution, samples were boiled for 3 min in Laemmli's sample buffer. Equal amounts of protein were subjected to SDS–polyacrylamide gel electrophoresis (PAGE). Proteins were transferred to an Immobilon PVDF membrane and subsequently blocked with TTBS solution (50 mm Tris-HCl, pH 7.5, 150 mm NaCl, and 0.1% Tween 20) containing 2.5% bovine serum albumin for 1 hr at room temperature. Blotted proteins were probed overnight at 4°C with the following primary antibodies: phospho-CaMKII (Thr286) [32], CaMKII, and phospho-synapsin I (Ser603) (Thermo Scientific, Waltham, MA, USA); synapsin I [32], phospho-GluR1 (Ser831), phospho-GluR1 (Ser845), GluR1, phospho-NMDAR1 (Ser896), phospho-MARCKS (Millipore, Billerica, MA, USA), N-methyl-D-aspartate (NMDA) R1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and β-actin (Sigma). Anti-rabbit, anti-mouse, and anti-goat IgG HRP-conjugated antibodies were used as the secondary antibodies. Signals were visualized using the enhanced chemiluminescence detection system (Amersham Life Science, Buckinghamshire, UK) and analyzed semiquantitatively using the NIH Image J program.

Immunohistochemistry

For immunohistochemical analyses, rats were anesthetized and transcardially perfused with 4% paraformaldehyde in phosphate-buffered saline (PBS) [33]. Serial 35-μm slice sections were prepared using a vibratome (Leica VT1000 A, Leica Microsystems Ltd, Wetzlar, Germany). Double fluorescent immunostaining was utilized to determine the localization of phospho-CaMKII in the hippocampi of control and VPA-treated rats with or without melatonin treatment. For immunolabeling, slices were probed with phospho-CaMKII (Thr286) [32] and β-tubulin (Sigma) overnight at 4°C. After washing, the sections were incubated with Alexa 488 anti-rabbit IgG or Alexa 594 anti-mouse IgG (Molecular Probes, Eugene, OR, USA) in blocking buffer. Immunofluorescence was visualized using a Zeiss LSM 510 confocal microscope (LSM 510 META, Carl Zeiss, Germany).

Electrophysiology

Hippocampal slices were prepared as previously described [34]. Transverse slices (400 μm thick) prepared using a vibratome (microslicer DTK-1000) were incubated in the holding chamber at room temperature (20–22°C) in ACSF bubbled with 95% O2/5% CO2. After a 2-hr recovery period, slices were transferred to an interface recording chamber and perfused at a flow rate of 2 mL/min with standard ACSF warmed to 34°C. Field excitatory postsynaptic potentials (fEPSPs) were evoked with a 0.05-Hz test stimulus through a bipolar stimulating electrode placed on the Schaffer collateral/commissural pathway and recorded from the stratum radiatum of area CA1 using a glass electrode filled with 3 m NaCl. Recordings were performed using a single-electrode amplifier (CEZ-3100; Nihon Kohden, Tokyo, Japan). The maximal value of the initial fEPSP slope was recorded and averaged every 1 min (three traces) using an A/D converter (PowerLab 200; AD Instruments, Castle Hill, NSW, Australia), and the values were stored on a personal computer. Stimulus intensity was adjusted to evoke fEPSPs of 1.0-mV amplitude. LTP induction was recorded for at least 60 min following a train of high-frequency stimulation (HFS, 100 Hz frequency for a duration of 1 s) that was applied twice at 10-s intervals, and test stimuli were continued for the indicated periods. The means of the fEPSP slopes were calculated and expressed as a percentage of the corresponding prestimulation control [34].

Statistical analysis

Data are represented as means ± S.E.Ms. Statistical significance was determined with one-way analyses of variance (ANOVA), followed by a Dunnett's test for multi-group comparisons. The results were considered significant when < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Competing interests
  9. References

Previous studies have suggested that children who are exposed to anticonvulsant drugs in utero have an increased prevalence of developmental delays and cognitive dysfunctions [6, 16, 17]. In the present study, VPA-treated rats showed no significant changes in body weight when compared with control rats (data not shown). As shown in Fig. 1(A,B), the locomotor activity and duration during the daytime phase (i.e., light or dark phase) were normal in VPA-treated animals. In the novel object recognition task, the discrimination index did not differ between control and VPA-treated rats during the conditioning trial (left; Fig. 1C). During the test session, control groups spent a significant amount of time exploring the novel object (right, 71.0 ± 4.9%; < 0.01; Fig. 1C), while VPA-treated rats showed a cognitive impairment indicated by their inability to discriminate between familiar and novel objects (50.1 ± 2.5%; > 0.05; Fig. 1C). These results support the validity of the VPA-treated rat model for autism in humans and suggest that the alterations observed in plasticity-related signaling may contribute to the behavioral phenotype.

image

Figure 1. Impaired memory-related behavior in VPA-treated rats. (A) Locomotor activity was analyzed over a 24-hr period in control and VPA-treated animals. (B) Changes in locomotion quantified during the light phase and the dark phase and total activity. (C) Object recognition behavior of control and VPA-treated rats at PND50. Bars indicate mean ±  S.E.M. (n = 8). **< 0.01 discrimination of spatial novelty (right) versus familiar (left).

Download figure to PowerPoint

Cognitive impairment and hippocampal atrophy have been described in patients with ASDs [5, 6, 35]. In the present study, we focused on the hippocampus, which is one of the major brain loci impaired in animals with ASD. CaMKII is particularly enriched in the brain and exhibits multifunctional roles in Ca2+-mediated signal transduction processes that play central roles in a variety of memory functions [35-37]. Compared with the control group, we observed a dramatic decrease in CaMKII (Thr286) phosphorylation in the hippocampal region of VPA-treated rats at PND50 (Fig. 2A). However, no significant difference in the levels of total CaMKII between the two groups was found (Fig. 2A). Interestingly, CaMKII activation phosphorylates a variety of substrates including synapsin I, tubulin, and several types of glutamate receptors [38]. Available evidence suggests that the increases in AMPA (α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid) conductance observed during the early stages of LTP are due to the phosphorylation of GluR1 (Ser831) [14, 15]. In the present study, a significant decrease in synapsin I (Ser603) and GluR1 (Ser831) phosphorylation was observed in the hippocampi of VPA-treated rats (Fig. 2A), suggesting that the dephosphorylation of CaMKII/synapsin I/GluR1 in the hippocampus may be associated with the pathological processes of ASD.

image

Figure 2. Decreased phosphorylation of CaMKII/PKC/PKA in the VPA-treated rat hippocampus. Hippocampal proteins from control and VPA-treated rats were separated by SDS-PAGE followed by immunoblot analysis. Representative blots are shown for antibodies that recognize phosphorylated or total protein. (A) VPA induced the downregulation of phospho-CaMKII (Thr286) and its substrate in the hippocampus. (B) Analysis of phospho-NMDAR1 (Ser896) and MARCKS (Ser152) in control and VPA-treated rats. (C) Western blots of phospho-PKA (Thr197) and GluR1 (Ser845) in control and VPA-treated rats at PND50. β-actin was used as a loading control.

Download figure to PowerPoint

PKC activity is regulated through the distinct phosphorylation of high-affinity substrates of PKC, including MARCKS (Ser152) and NMDAR1 (Ser896) [34]. Thus, we examined the physiological relevance of PKC activation in VPA-treated rats by assessing phospho-NMDAR1 (Ser896) and MARCKS (Ser152). Immunoblotting analyses revealed a significant decrease in the phosphorylation of NMDAR1 (Ser 896) and MARCKS (Ser152) in the VPA-treated rats compared with the control rats, which indicated that PKC signaling is downregulated in the hippocampal region of VPA-treated rats (Fig. 2B).

The PKA-dependent phosphorylation of the GluR1 (Ser845) subunit plays an essential role in the regulation of AMPA channel conductance [39]. Compared with the control rats, there were no significant changes in total levels of PKA in the VPA-treated rat hippocampus (Fig. 2C), but there was a significant decrease in phospho-PKA (Thr197). Furthermore, Western blotting analysis revealed that phospho-GluR1 (Ser845) was dramatically decreased in the hippocampus of VPA-treated rats compared with control rats (Fig. 2C) and was preferentially phosphorylated by PKA.

BTBR T+tf/J mice have recently been reported to display abnormal behaviors that resemble the behaviors of individuals diagnosed with autistic disorder [40]. We further examined changes in the phosphorylation of CaMKII/PKC/PKA signaling in the brains of this autistic animal model. Consistent with the VPA model, our data also showed a reduction in the phosphorylation of CaMKII (Thr286) in the hippocampus of BTBR T+tf/J mice compared with wild-type mice (Fig. 3A). We further assessed the physiological relevance of PKA and PKC activation in BTBR T+tf/J mice with immunoblotting using phospho-specific antibodies. Consistent with the VPA model, we found a marked reduction in the phosphorylation of NMDAR1 (Ser 896) in the hippocampal region of BTBR T+tf/J mice in the absence of alterations in total NMDAR1 protein levels (Fig. 3B). However, no significant changes were observed in hippocampal PKA (Thr197) or GluR1 (Ser845) phosphorylation between wild-type mice and BTBR T+tf/J mice (Fig. 3C).

image

Figure 3. Changes in the phosphorylation of CaMKII/PKC/PKA in the hippocampus of BTBR T+tf/J mice. Representative immunoblots are shown for antibodies against phosphorylated or total protein in the hippocampus of wild–type mice and BTBR T+tf/J mice. (A) Changes in phospho-CaMKII (Thr286) and its substrate in the hippocampus of wild-type mice and BTBR T+tf/J mice. (B) Analysis of phospho-NMDAR1 (Ser896) and MARCKS (Ser152) in wild-type mice and BTBR T+tf/J mice. (C) Western blots of phospho-PKA (Thr197) and GluR1 (Ser845) in wild-type mice and BTBR T+tf/J mice at PND50. β-actin was used as a loading control.

Download figure to PowerPoint

Recently, we reported a significant reduction in both CaMKII and PKC phosphorylation in the hippocampal regions in neurodegenerative models [41]. In the present study, we examined whether chronic melatonin treatment could stimulate CaMKII/PKC/PKA phosphorylation in the hippocampus of VPA-treated rats. When rats were chronically treated with melatonin for 4 wk, the decreased CaMKII autophosphorylation was significantly recovered compared to the rats that received VPA treatment alone (1 mg/kg: 86.9 ± 8.5%; 5 mg/kg: 83.9 ± 6.9% of control, n = 6; Fig. 4A). Additionally, melatonin treatment also recovered the decreases in phospho-synapsin I (Ser603) in the hippocampal region of the rats that received VPA treatment alone (Fig. 4B). Furthermore, chronic melatonin treatment at 5 mg/kg significantly restored phospho-GluR1 (Ser-831) in VPA-treated rats (5 mg/kg: 105.5 ± 7.9% of control, n = 6; Fig. 4C).

image

Figure 4. Chronic melatonin administration increases the phosphorylation of CaMKII and its substrate in VPA-treated rats. Immunoblotting assay performed using antibodies against phospho-proteins or total proteins. The upper panel shows representative Western blots for phospho-CaMKII (Thr286) (A), phospho-synapsin I (Ser603) (B), and phospho-GluR1 (Ser831) (C) in the hippocampus of the indicated groups. The lower panel presents the quantitative analysis of the relative phospho-protein, which was performed by densitometry. The data are expressed as the percentage of values in control rats (mean ± S.E.M., n = 6). *< 0.05; **< 0.01 versus control rats; #< 0.05 versus vehicle-treated rats. β-actin was used as a loading control. MEL, melatonin.

Download figure to PowerPoint

CaMKII is subcellularly localized to dendrites and even more specifically localized to the postsynaptic densities of excitatory synapses [42]. Using double immunofluorescent labeling and confocal microscopy, we examined changes in phospho-CaMKII (Thr286) protein expression in hippocampal CA1 pyramidal neurons at PND50 in VPA-treated rats. Consistent with the biochemical evidence derived from our immunoblotting data, immunohistochemical staining showed a significant decrease in the intensity of immunolabeling for phospho-CaMKII (Thr286) in the hippocampus of VPA-treated rats (Fig. 5), whereas immunostaining in non-VPA-treated control rats revealed strong subcellular distributions of phospho-CaMKII (Thr286) in the soma and dendrites of pyramidal neurons in the hippocampal CA1 region (Fig. 5). Furthermore, our data demonstrated that chronic melatonin treatment partially restored the decrease in immunostaining for phospho-CaMKII (Thr286) in hippocampal pyramidal neurons of VPA-treated rats (Fig. 5).

image

Figure 5. Immunohistochemical data for the effect of chronic melatonin treatment on hippocampal phospho-CaMKII in VPA-treated rats. The immunohistochemical expression of phospho-CaMKII (Thr286) (green) and β-tubulin (red) was examined in the hippocampus of control and VPA-treated rats at PND50. Representative Z-stack images are shown in the right panel. DAPI counterstaining (blue) indicates nuclear localization. Scale bar = 20 μm.

Download figure to PowerPoint

In addition, the phosphorylation of NMDAR1 (Ser896) (Fig. 6A) and MARCKS (Ser152) (Fig. 6B) was also significantly restored in the 1 and 5 mg/kg melatonin- and VPA-treated groups. In contrast to the decreases in phospho-PKA (Thr197) and phospho-GluR1(Ser845) levels in VPA-treated rats, phospho-PKA (Thr197) (Fig. 7A) and phospho-GluR1(Ser845) (Fig. 7B) were significantly restored by chronic melatonin (5 mg/kg) treatment. No significant differences were observed in total PKA or GluR1 activities between any of the hippocampal samples (Fig. 7A,B). In addition, no apparent effects on the phosphorylation of CaMKIIα, PKCα, or PKA were observed in the hippocampus of control animals after melatonin treatment (data not shown).

image

Figure 6. Chronic melatonin administration increases the phosphorylation of PKC signaling in the VPA-treated rat model. Hippocampal proteins from rats were separated by SDS-PAGE followed by immunoblot analysis. Representative immunoblots from brain lysates of VPA-treated rats that were treated with vehicle or melatonin when probed with anti-phospho-NMDAR1 (Ser896) (A) or anti-phospho-MARCKS (Ser152) (B) antibodies. The upper panel shows representative Western blots. The lower panel indicates the quantitative analysis of relative phospho-proteins in the indicated groups. The data are expressed as percentages of the values of the control rats (mean ± S.E.M., n = 6). *< 0.05; **< 0.01 versus control rats; #< 0.05, ##< 0.01 versus vehicle-treated rats. β-actin was used as a loading control. MEL, melatonin.

Download figure to PowerPoint

image

Figure 7. Chronic melatonin administration increases the phosphorylation of PKC and PKA signaling in the VPA-treated rats. (A) PKA phosphorylation in the hippocampus was detected by an immunoblot assay with an anti-phospho-PKA (Thr197) antibody. Representative images of the immunoblots are shown in upper panel. Quantitative analysis of 42-kDa phospho-PKA (Thr197) was performed by densitometry (lower panel). (B) GluR1 phosphorylation in the hippocampus was detected by immunoblotting with an anti-phospho-GluR1(Ser845) antibody. Quantitative analysis for the 100-kDa phospho-GluR1(Ser-845) was performed by densitometry (lower panel). The data are expressed as the percentage of values from control rats (mean ± S.E.M., n = 6). *< 0.05; **< 0.01 versus control rats; #< 0.05, ##< 0.01 versus vehicle-treated rats. β-actin was used as a loading control. MEL, melatonin.

Download figure to PowerPoint

LTP in the CA1 region underlines hippocampus-dependent spatial reference memory. Rats exposed to VPA treatment have been previously shown to express deficits in spatial learning and hippocampal LTP [43-45]. To understand whether the restoration of CaMKII/PKC/PKA phosphorylation by melatonin treatment correlated with changes in memory formation, the hippocampal LTP in the CA1 region of VPA-treated rats was recorded. Evaluating the hippocampal slices obtained from control rats showed that HFS (two trains of 100 Hz stimulation) of Schaffer collateral/commissural pathways induced long-lasting LTP in the hippocampal CA1 region (204.3 ± 6.8% of baseline during 1–5 min; 168.5 ± 5.0% of baseline during 26–30 min; 162.7 ± 3.6% of baseline during 56–60 min, n = 5, Fig. 5B,C). Melatonin treatment (5 mg/kg/day, 28 days, p.o.) alone in control rats did not affect the induction of LTP (n = 5, Fig. 8B,C). Here, markedly reduced LTP was observed in vehicle-treated VPA rats (139.3 ± 2.6% of baseline during 1–5 min; 116.3 ± 2.4% of the baseline during 26–30 min; 113.2 ± 2.3% of the baseline during 56–60 min, n = 5; Fig. 8B,C). Melatonin (5 mg/kg) treatment significantly improved LTP in hippocampal CA1 regions (151.3 ± 3.7% of the baseline during 1–5 min; 135.7 ± 3.6% of the baseline during 26–30 min; 133.4 ± 3.9% of the baseline during 56–60 min, n = 5, Fig. 8B,C), whereas no significant changes were observed 1–5 min after HFS compared with VPA-treated rats.

image

Figure 8. Melatonin treatment significantly improved the impairments in LTP in the hippocampal CA1 region of the VPA-treated rats. (A) Representative field excitatory postsynaptic potential (fEPSP) slopes were recorded from the CA1 regions of control, melatonin alone (5 mg/kg p.o.), VPA-treated rats, and melatonin- (5 mg/kg p.o.) and VPA-treated rats. (B) LTP recorded in CA1 was attenuated in VPA rats compared with control animals. Melatonin (5 mg/kg) treatment prevented VPA-induced impairment of LTP in the CA1 region. (C) Histogram comparing the slopes of fEPSPs for 60 min after HFS in the Schaffer collateral/commissural pathways of CA1 region. HFS, high-frequency stimulation. Points shown are means ± S.E.M.; n = 5; **< 0.01 versus control animals; ++< 0.01 versus VPA-treated rats. MEL, melatonin (5 mg/kg) treatment alone.

Download figure to PowerPoint

We used a social interaction test to investigate whether melatonin improved the dysfunctional social behaviors observed in VPA-treated rats. Consistent with previous studies [46], VPA-treated rats exhibited a significant decrease in social interaction behaviors compared with control rats (Fig. 9). In addition, VPA-treated rats were tested after chronic treatment with melatonin for 4 wk (1 or 5 mg/kg, p.o.) using an automated three-chambered social approach apparatus (Fig. 9). Interestingly, the melatonin-treated rats showed a significant rescue of the VPA-induced impairment of social interaction behavior (Fig. 9). Moreover, melatonin treatment increased the time spent in proximity to and the frequency of reciprocal social interactions with a social partner (stranger 1) (Fig. 9B) compared with vehicle-infused VPA-treated rats.

image

Figure 9. Melatonin restores impaired social-related behaviors in VPA-treated rats. (A) Effect of melatonin treatment on the time spent in each compartment. Sociability was determined as described in the methods section. *< 0.05 in the same treatment group and in S2 (stranger 2, right) versus S1 (stranger 1, left). (B) The effect of melatonin on the duration of interaction with the social partner (stranger 1) in VPA-treated rats at PND50. **< 0.01 versus control rats; ##< 0.01 versus in VPA-treated rats. The bars indicate the means ± S.E.M. (n = 6).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Competing interests
  9. References

This study investigated the effects of melatonin on biochemical alterations in an experimental ASD model by examining the expression levels of key protein kinases that are involved in learning and memory. Disruptions in hippocampal CaMKII/PKA/PKC signaling were observed in both VPA-treated rats and BTBR T+tf/J mice. Importantly, melatonin significantly prevented the hypophosphorylation of serine/threonine kinases signaling and the decreases of LTP in the hippocampi of the ASD models. We suggest that the melatonin-induced restoration of CaMKII/PKC/PKA phosphorylation and LTP induction might correlate with ameliorations of social behavior dysfunction observed in the VPA-treated rats.

Several independent groups have clinically investigated the implications of autism as a dysfunction in the pineal–hypothalamic–pituitary–adrenal axis in the brain [47, 48]. However, many important questions remain unresolved regarding the biochemical processes of autism [49]. Melatonin is a signaling molecule that not only is responsible for daily rhythms but also participates in the hippocampal circuits that underlie memory processes [50, 51]. Accumulating data suggest that melatonin may be useful in improving the behavior of children and youth with ASD [52]. An understanding of the precise roles of melatonin in ASD symptoms might provide insight into treatment perspectives.

In the present study, our findings indicated that melatonin treatment restored social interaction function in VPA-treated rats. These results are consistent with clinical studies showing that melatonin administration may help alleviate symptoms of ASD and markedly improve outcomes [22]. Chaudhury et al. [53] provided evidence that melatonin functions as an endogenous circadian oscillator and modulates synaptic plasticity in the hippocampus. The interplay of synaptic and clock genes is a potential focus for the development of therapeutic strategies for ASD [54]. Numerous studies have suggested that melatonin ameliorates spatial memory and learning deficits in different animal models [50, 55, 56].

VPA treatment suppresses NMDAR – dependent LTP in hippocampal slices [43, 57, 45]. Here, both VPA-treated rats and BTBR T+tf/J mice demonstrated significant reductions in the phosphorylation of CaMKII (Thr286) in the hippocampus. We found a significant decrease in LTP in the hippocampal CA1 region of the VPA-treated rats that paralleled the VPA-induced CaMKII (Thr286) dephosphorylation. Here, the impaired LTP in the hippocampal CA1 region was partly recovered by melatonin treatment (5 mg/kg). The present data imply that the therapeutic effect of melatonin on social interaction dysfunction might be due to improvements in normal synaptic transmission and plasticity [7, 8, 58]. The present study further showed that melatonin ameliorated the decreases in synapsin I (Ser603) and GluR1 (Ser831) phosphorylation, which are presynaptic and postsynaptic substrates of CaMKII in the hippocampus, respectively [39, 11, 59].

Furthermore, melatonin activates Ca2+ signaling pathways via protein kinase C activity [60]. The effects of VPA on synaptic plasticity in the rat hippocampus may be mediated by an indirect interaction with the NMDAR or by the inhibition of PKC activity in both membrane and cytosol compartments [43, 57]. Wang et al. [25] reported that chronic melatonin treatment effectively restores fEPSP amplitudes and slopes via the EPACs/miR-124/Egr1 pathway and reduces memory and synaptic impairments in scopolamine-induced amnesia. Melatonin also exerts its effects by binding to MT2 receptors and participating in hippocampal synaptic plasticity and memory processes, and LTP is impaired in hippocampal slices in mice lacking MT2 receptors [61]. Because CaMKII/PKC/PKA phosphorylation functions as a molecular switch underlying LTP and explicit memory [42, 59], enhanced phospho-CaMKII/PKC/PKA in the hippocampus following melatonin treatment in VPA-treated rats might contribute to improvements in autism-related behaviors.

This study has several limitations that are worth noting. First, we did not observe significant dose-dependent changes between the doses of 1 mg/kg and 5 mg/kg melatonin in the phosphorylation of CaMKII. Thus, the optimization of melatonin doses for the alleviation of dephosphorylation of CaMKII signaling requires further investigation. In addition, the role of melatonin receptors during ASD with or without melatonin treatment is another unaddressed issue; one recent study demonstrated pathway-biased and deleterious melatonin receptor mutants in an ASD population [62].

In conclusion, this is the first study to demonstrate that restoration of impaired serine/threonine kinases (CaMKII/PKC/PKA) by melatonin might correlate with amelioration of the dysfunctions in social behavior in ASD models. Thus, because melatonin is a molecule with a very low toxicity profile at virtually any dose and after long-term usage [53, 61], melatonin could be a new therapeutic candidate for the treatment of behavioral dysfunctions induced by ASD.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Competing interests
  9. References

This work was partially supported by National Natural Science Foundations of China (91232705; 81202533); Qianjiang Talents Program of Zhejiang Province, China (2012R10036).

Competing interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Competing interests
  9. References

The authors declare that they have no competing interests.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Competing interests
  9. References