Music-Based Intervention Ameliorates Mecp2-Loss-Mediated Sociability Repression in Mice through the Prefrontal Cortex FNDC5/BDNF Pathway

Patients with Rett syndrome (RTT) show severe difficulties with communication, social withdrawl, and learning. Music-based interventions improve social interaction, communication skills, eye contact, and physical skills and reduce seizure frequency in patients with RTT. This study aimed to investigate the mechanism by which music-based interventions compromise sociability impairments in mecp2 null/y mice as an experimental RTT model. Male mecp2 null/y mice and wild-type mice (24 days old) were randomly divided into control, noise, and music-based intervention groups. Mice were exposed to music or noise for 6 h/day for 3 consecutive weeks. Behavioral patterns, including anxiety, spontaneous exploration, and sociability, were characterized using open-field and three-chamber tests. BDNF, TrkB receptor motif, and FNDC5 expression in the prefrontal cortex (PFC), hippocampus, basal ganglia, and amygdala were probed using RT-PCR or immunoblotting. mecp2 null/y mice showed less locomotion in an open field than wild-type mice. The social novelty rather than the sociability of these animals increased following a music-based intervention, suggesting that music influenced the mecp2-deletion-induced social interaction repression rather than motor deficit. Mechanically, the loss of BDNF signaling in the prefrontal cortex and hippocampal regions, but not in the basal ganglia and amygdala, was compromised following the music-based intervention in mecp2 null/y mice, whereas TrkB signaling was not significantly changed in either region. FNDC5 expression in the prefrontal cortex region in mecp2 null/y mice also increased following the music-based intervention. Collective evidence reveals that music-based interventions improve mecp2-loss-induced social dysfunction. BDNF and FNDC5 signaling in the prefrontal cortex region mediates the music-based-intervention promotion of social interactions. This study gives new insight into the mechanisms underlying the improvement of social behaviors in mice suffering from experimental Rett syndrome following a music-based intervention.


Mecp2 null/y Mice Show a Motor Deficit Rather Than Anxiety
Mecp2 null/y mice showed a genotype lacking the mecp2 gene, corresponding to a region of 465 bp, in agarose electrophoretograms ( Figure 1A). Motor deficits and anxietyrelated behaviors in experimental animals were assessed by an open field test. When exposed to the novel environment in the open field arena, WT mice engaged more in exploratory behaviors, especially in the central zone, than mecp2 null/y mice. In addition, WT mice tracked crosses into the center of the arena at regular intervals, while the mecp2 null/y mice remained in proximity to the walls of the arena, indicating a tendency toward anxiety-related behaviors in these animals ( Figure 1B). The total distance traveled in the arena was significantly shorter for the C-mecp2 null/y mice and W-mecp2 null/y mice as compared with the WT mice. mecp2 null/y mice did not show significant motor deficits as compared with other groups ( Figure 1C, p < 0.05). Mecp2 null/y mice tended to spend less time in the central field. However, this was only observed for several experimental mice, and the results did not reach statistical significance ( Figure 1D). The results indicate motor dysfunction in these animals. The music-based intervention did not significantly affect the moving distance or time in WT or mecp2 null/y mice.

Music-Based Interventions Can Reduce Social Deficits in Mecp2 null/y Mice
We performed a social novelty preference test to investigate whether animals' social behaviors were affected by a music-based intervention. A decrease in sniffing of stranger mice in Holding Zone 1 indicates an impairment in sociability parameters. While the three groups of mecp2 null/y mice did not show significant differences in sniffing behaviors compared with the three groups of WT mice ( Figure 1E), mecp2 null/y mice in the music group spent more time sniffing stranger 2 in Holding Zone 2 than stranger 1 in Holding Zone 1. A similar phenomenon was observed for WT mice in the control and white noise groups ( Figure 1F, p < 0.05). The result indicates that the music-based intervention improved one of the social parameters, sniffing intensity, in mecp2 null/y mice.

The Music-Based Intervention Altered BDNF mRNA and Protein Levels in Different Brain Regions
The analysis of improvements in social interactions in mecp2 null/y mice following a music-based intervention prompted us to investigate the affects on BDNF signaling in particular brain regions. BDNF mRNA expression was significantly decreased in the prefrontal cortex region in the control mecp2 null/y and white-noise-treated mecp2 null/y mice as compared with the control WT and white-noise-treated WT mice (Figure 2A, C-mecp2 null/y and W-mecp2 null/y vs. C-WT and W-WT, p < 0.05). An intragroup comparison of BDNF mRNA expression in the prefrontal cortex region showed a significant increase in BDNF mRNA in mecp2 null/y mice and music-treated mecp2 null/y mice as compared with control mecp2 null/y mice (Figure 2A, M-mecp2 null/y vs. C-mecp2 null/y , p < 0.05). In the hippocampal region, BDNF mRNA expression was significantly decreased in white-noisetreated mecp2 null/y mice as compared with WT mice and mecp2 null/y mice in the other two groups ( Figure 2B, W-mecp2 null/y mice vs. WT and other groups of mecp2 null/y mice, p < 0.05). Signaling in the basal ganglia ( Figure 2C) and amygdala ( Figure 2D) was not significantly altered in WT or mecp2 null/y mice. The traveling distances were significantly lower for C-mecp2 null/y and W-mecp2 null/y mice as compared with all wild-type mice. (D) The pecentage of time spent in the central zone was comparable for mecp2 null/y and wild-type mice. (E) No significant differences in the sniffing of stranger mice between Mecp2 null/y and wild-type mice were identified in the social ability test. (F) The M-mecp2 null/y mice spent more time sniffing novel mice as compared with the C-mecp2 null/y mice. (* p < 0.05).

Music-Based Interventions Can Reduce Social Deficits in Mecp2 null/y Mice
We performed a social novelty preference test to investigate whether animals' social behaviors were affected by a music-based intervention. A decrease in sniffing of stranger mice in Holding Zone 1 indicates an impairment in sociability parameters. While the three groups of mecp2 null/y mice did not show significant differences in sniffing behaviors compared with the three groups of WT mice ( Figure 1E), mecp2 null/y mice in the music group spent more time sniffing stranger 2 in Holding Zone 2 than stranger 1 in Holding Zone 1. A similar phenomenon was observed for WT mice in the control and white noise (C) The traveling distances were significantly lower for C-mecp2 null/y and W-mecp2 null/y mice as compared with all wild-type mice. (D) The pecentage of time spent in the central zone was comparable for mecp2 null/y and wild-type mice. (E) No significant differences in the sniffing of stranger mice between Mecp2 null/y and wild-type mice were identified in the social ability test. (F) The M-mecp2 null/y mice spent more time sniffing novel mice as compared with the C-mecp2 null/y mice. (* p < 0.05).

mecp2
mice (Figure 2A, M-mecp2 vs. C-mecp2 , p < 0.05). In the hippocamp region, BDNF mRNA expression was significantly decreased in white-noise-treat mecp2 null/y mice as compared with WT mice and mecp2 null/y mice in the other two grou ( Figure 2B, W-mecp2 null/y mice vs. WT and other groups of mecp2 null/y mice, p < 0.05). S naling in the basal ganglia ( Figure 2C) and amygdala ( Figure 2D) was not significan altered in WT or mecp2 null/y mice. BDNF protein levels in the prefrontal cortex region were not significantly changed any of the mcie ( Figure 3A), whereas the protein was significantly increased in the h pocampi of music-treated mecp2 null/y mice as compared with all WT mice and the oth two groups of mecp2 null/y mice ( Figure 3B, M-mecp2 null/y vs. all WT and the other tw groups of mecp2 null/y mice, p < 0.05). Consistent with the levels of BDNF mRNA expressi in the basal ganglia and amygdala, the BDNF protein levels were not significantly BDNF protein levels in the prefrontal cortex region were not significantly changed in any of the mcie ( Figure 3A), whereas the protein was significantly increased in the hippocampi of music-treated mecp2 null/y mice as compared with all WT mice and the other two groups of mecp2 null/y mice ( Figure 3B, M-mecp2 null/y vs. all WT and the other two groups of mecp2 null/y mice, p < 0.05). Consistent with the levels of BDNF mRNA expression in the basal ganglia and amygdala, the BDNF protein levels were not significantly affected in these two brain regions ( Figure 3C,D). The analysis indicates that BNDF signaling in the brain microenvironment following a music-based intervention is context-dependent.

Mecp2 Deletion Rather Than Music-Based Intervention Affected the TrkB Receptor in the Prefrontal Cortex
TrkB, receptor tyrosin kinase, is an important downstream receptor of BDNF [30,31]. We investigated whether TrkB signaling in the prefrontal cortex or hippocampus changed following a music-based intervention. TrkB mRNA expression in the prefrontal cortex was significantly increased in the control and white-noise-treated mecp2 null/y mice as compared with WT and mecp2 null/y mice in the music group ( Figure 4A, C-mecp2 null/y and W-mecp2 null/y mice vs. all WT and M-mecp2 null/y mice, p < 0.05). Signaling in the hippocampus was not significantly changed in any of the mice ( Figure 4B). fected in these two brain regions (Figure 3 C,D). The analysis indicates that BNDF signaling in the brain microenvironment following a music-based intervention is context-dependent.

Mecp2 Deletion Rather Than Music-Based Intervention Affected the TrkB Receptor in the Prefrontal Cortex
TrkB, receptor tyrosin kinase, is an important downstream receptor of BDNF [30,31]. We investigated whether TrkB signaling in the prefrontal cortex or hippocampus changed following a music-based intervention. TrkB mRNA expression in the prefrontal cortex was significantly increased in the control and white-noise-treated mecp2 null/y mice as compared with WT and mecp2 null/y mice in the music group ( Figure 4A, C-mecp2 null/y and W-mecp2 null/y mice vs. all WT and M-mecp2 null/y mice, p < 0.05). Signaling in the hippocampus was not significantly changed in any of the mice ( Figure 4B).

Mecp2 null/y Mice That Did Not Undergo a Music-Based Intervention Overproduced Full-Length and Truncated TrkB Protein in the PFC
TrkB contains spliced isoforms, including a full-length, catalytically active kinase receptor (TrkB-FL), and truncated isoforms (TrkB-T). TrkB-T and TrkB-FL have important interactions with neurotrophins to regulate neural activity [32]. We investigated whether TrkB-FL or TrkB-L signaling in the prefrontal cortex or hippocampus changed following music treatment. TrkB-FL and TrkB-T protein levels in the prefrontal cortex increased significantly in control C-mecp2 null/y mice as compared with all WT mice ( Figure 5A, C-mecp2 null/y vs. all WT mice, p < 0.05); however, these molecules were not significantly affected in the hippocampal region in any of the mice ( Figure 5B). Comparison of TrkB mRNA expression between wild-type mice and mecp2 null/y mice in the PFC and hippocampus. (A) Mecp2 null/y mice that did not undergo a music-based intervention showed significantly increased TrkB mRNA expression in the PFC as compared with all wild-type mice and mecp2 null/y mice that underwent a music-based intervention. (B) TrkB mRNA expression was comparable in the hippocampus between wild-type mice and mecp2 null/y mice. (* p < 0.05).

Mecp2 null/y Mice that Did Not Undergo a Music-Based Intervention Overproduced Full-Length and Truncated TrkB Protein in the PFC
TrkB contains spliced isoforms, including a full-length, catalytically active kinase receptor (TrkB-FL), and truncated isoforms (TrkB-T). TrkB-T and TrkB-FL have important interactions with neurotrophins to regulate neural activity [32]. We investigated whether TrkB-FL or TrkB-L signaling in the prefrontal cortex or hippocampus changed following music treatment. TrkB-FL and TrkB-T protein levels in the prefrontal cortex increased significantly in control C-mecp2 null/y mice as compared with all WT mice ( Figure  5A, C-mecp2 null/y vs. all WT mice, p < 0.05); however, these molecules were not significantly affected in the hippocampal region in any of the mice ( Figure 5B). . Analysis of full length and truncated TrkB proteins in wild-type and mecp2 null/y mice in the PFC and hippocampus. Mecp2 null/y mice that did not undergo a music-based intervention produced significantly higher full length and truncated TrkB protein levels in the prefrontal cortex (A) but not in the hippocampus (B) as compared with all wild-type mice and mecp2 null/y mice that underwent a music-based intervention (* p < 0.05).

Music-Based Interventions Induce Prefrontal Cortex FNDC5 Gene Expression
The analysis of motor deficits in mecp2 null/y mice prompted us to ask whether myokine FDNC5 was changed following a music-based intervention, as FNDC5 is important for muscle function and cognition. The expression of FNDC5 in the prefrontal cortex was significantly upregulated in music-treated mecp2 null/y mice as compared with white-noise-treated mecp2 null/y mice ( Figure 6A, M-mecp2 null/y vs. W-mecp2 null/y mice,p < 0.05). Of note, the myokine was significantly reduced in all mecp2 null/y mice as compared Figure 5. Analysis of full length and truncated TrkB proteins in wild-type and mecp2 null/y mice in the PFC and hippocampus. Mecp2 null/y mice that did not undergo a music-based intervention produced significantly higher full length and truncated TrkB protein levels in the prefrontal cortex (A) but not in the hippocampus (B) as compared with all wild-type mice and mecp2 null/y mice that underwent a music-based intervention (* p < 0.05).

Music-Based Interventions Induce Prefrontal Cortex FNDC5 Gene Expression
The analysis of motor deficits in mecp2 null/y mice prompted us to ask whether myokine FDNC5 was changed following a music-based intervention, as FNDC5 is important for muscle function and cognition. The expression of FNDC5 in the prefrontal cortex was significantly upregulated in music-treated mecp2 null/y mice as compared with white-noisetreated mecp2 null/y mice ( Figure 6A, M-mecp2 null/y vs. W-mecp2 null/y mice, p < 0.05). Of note, the myokine was significantly reduced in all mecp2 null/y mice as compared with all WT mice ( Figure 6B, all mecp2 null/y mice vs. all WT mice, p < 0.05).

Discussion
Our data provide evidence that music-based interventions can restore the social abilities of Rett syndrome (RTT)mice through the upregulation of BDNF mRNA expression in the prefrontal cortex and an increase in the BDNF protein level in the hippocampal region. In addition, our data reveal that music-based interventions can upregulate BDNF gene expression in a TrkB-pathway-independent manner through modulation of the FNDC5 gene.
RTT is caused by mutations of the mecp2 gene [5]. The spectrum of mecp2 gene mutations has been found to involve missense, nonsense, and frame-shift mutations, as well as truncations due to premature STOP codons, with specific mutations correlating with the level of clinical severity [33,34]. The mecp2 gene located in X chromosome Xq29 encodes the mecp2 protein which binds to methylated CpG sites in the gene promotor and is critical for binding to regulatory gene regions and for the recruitment of cofactors. Close to 1300 genes will be transcriptionally dysregulated when the mecp2 gene is dysfunctional. The first mammalian neuronal target gene for mecp2 identified was brain-derived neurotrophic factor (BDNF) [35,36]. Mecp2 dysfunction leads to reduced production of BDNF, a protein required for normal neuronal development. Significant evidence indicates a reduction in BDNF levels in mecp2-based mouse models of RTT, which becomes significant with the appearance of RTT-like features. Two studies showed lower BDNF mRNA levels in autopsy brain samples from RTT individuals [37,38], which is reminiscent of the situation in mecp2 mutant mice. Improving BDNF expression and/or signaling Figure 6. Comparison of FNDC5 expression in the PFC and hippocampus between wild-type and mecp2 null/y mice. (A) Mecp2 null/y mice that underwent a music-based intervention showed significantly increased FNDC5 expression in the PFC as compared with control mecp2 null/y mice. (B) Mecp2 null/y mice had significantly lower FNDC5 expression as compared with all wild-type mice (* p < 0.05).

Discussion
Our data provide evidence that music-based interventions can restore the social abilities of Rett syndrome (RTT)mice through the upregulation of BDNF mRNA expression in the prefrontal cortex and an increase in the BDNF protein level in the hippocampal region. In addition, our data reveal that music-based interventions can upregulate BDNF gene expression in a TrkB-pathway-independent manner through modulation of the FNDC5 gene.
RTT is caused by mutations of the mecp2 gene [5]. The spectrum of mecp2 gene mutations has been found to involve missense, nonsense, and frame-shift mutations, as well as truncations due to premature STOP codons, with specific mutations correlating with the level of clinical severity [33,34]. The mecp2 gene located in X chromosome Xq29 encodes the mecp2 protein which binds to methylated CpG sites in the gene promotor and is critical for binding to regulatory gene regions and for the recruitment of cofactors. Close to 1300 genes will be transcriptionally dysregulated when the mecp2 gene is dysfunctional. The first mammalian neuronal target gene for mecp2 identified was brain-derived neurotrophic factor (BDNF) [35,36]. Mecp2 dysfunction leads to reduced production of BDNF, a protein required for normal neuronal development. Significant evidence indicates a reduction in BDNF levels in mecp2-based mouse models of RTT, which becomes significant with the appearance of RTT-like features. Two studies showed lower BDNF mRNA levels in autopsy brain samples from RTT individuals [37,38], which is reminiscent of the situation in mecp2 mutant mice. Improving BDNF expression and/or signaling has received a significant amount of attention regarding the treatment of a variety of neurological disorders [39,40], and a great deal of progress has been achieved by the RTT research community [41,42].
Music-based interventions have been used in some descriptive studies and case reports to restore communication, voluntary hand movement, and decreased stereotypic movement abilities [43,44]. A previous comprehensive study conducted by the current authors demonstrated that a music-based intervention improved receptive language, verbal and nonverbal communication skills, and social interactions in RTT patients. Purposeful hand functions, breathing patterns, and eye contact were significantly improved. In addition, the music-based intervention also decreased the frequency of epileptic seizures [14]. Previous research reported that wild-type adult mice or prenatal auditory stimulation can increase the BDNF protein level in the hippocampus [28,[45][46][47] as well as in the PFC and amygdala [47]. It was also reported that the BDNF/TrkB level in the dorsal hippocampus CA3 (dCA3) and dentate gyrus (dDG) was significantly enhanced in rats exposed to Mozart's music as compared with those without music exposure [25]. Corresponding to the results of previous studies, our data suggest that the music-based intervention upregulated BDNF mRNA expression in the PFC and increased BDNF protein levels in the hippocampal regions in mecp2 null/y mice. However, neither full-length nor truncated TrkB receptors were upregulated in mecp2 null/y mice exposed to music. On the contrary, mecp2 null/y mice that did not undergo a music-based intervention overexpressed TrkB-FL and TrkB-T receptors. We presume that impaired BDNF gene expression is the major pathophysiology of neurological dysfunction in RTT, and a BDNF protein secretion deficiency in the brain leads to compensation for TrkB receptor overexpression, either in full-length or truncated isoforms. Mecp2 null/y mice exposed to music showed upregulated BDNF gene expression and relatively normalized TrkB levels. The precise function of TrkB-T receptors still remains elusive. However, previous studies reported that TrB-T binds to BDNF with similar affinity to the TrkB kinase and can interfere with BDNF-TrkB signaling by binding to BDNF without activating downstream kinase cascades [48,49]. This explains why the compensation for TrkB receptor overexpression in mecp2 null/y mice with music exposure is not successful.
Fibronectin type III domain containing 5 (FNDC5) is known to induce BDNF expression in mice. Physical activity is an affordable and effective method that can be used to improve cognition by inducing FNDC5 gene expression. After cleavage by protease in the skeletal muscle, FNDC5 is cleaved into a 12 kDa peptide called irisin, which is able to enter the central nervous system and induce BDNF gene expression in the hippocampus. The finding that music-based intervention can induce FNDC5 expression in the PFC of mecp2 null/y mice is novel. We suggest that the upregulation of the BDNF gene after music-based intervention in the PFC of mecp2 null/y mice in our study occurred through the modulation of FNDC5 gene expression. The irisin immunoreactivity could not be investigated in this study since it has been reported to only exist in GABA-ergic purkinje cells in the cerebellum and vestibular nuclei of the medulla oblongata [50].
We successfully established a music-based intervention model in RTT in vivo; however, the analysis of FNDC5/BDNF signaling pathways hinted at the complex nature of biological activity in brain tissue that the intervention may influence. Engibeerin cortical organoids from human-induced pluripotent stem cells (iPSCs), showing neurite undergrowth, neurite coalescence, and soma size of interneurons, have recently been used as an in vitro RTT model [51,52]. The biological response or the roles of FNDC5/BNDF pathways in an in vitro humanized RTT model upon different music-based interventions, such as magnitude and frequency (Hz) of music, warrant further investigations in the future.
We acknowledge the limitations of this study. First, we used male mecp2 null/y mice due to the rapid onset of the Rett-like phenotype in this species; however, the clinical relevance of Rett syndrome still remains doubtful. Second, there are several behavioral tests that can be used to assess whether mice show autistic-like social interaction deficits, and the three-chamber social approach is commonly used in mice. However, whether the three-chamber sociability test is the best choice for mecp2 null/y mice remains elusive. Third, the music chosen for use in this study contained complicated in-music elements, such as rhythm, melody, harmony, timbre, voice pitch, etc. We cannot exclude the possibility that each element may have an impact on mecp2 null/y mice. Fourth, a study has revealed that very little FNDC5 mRNA is present in the hippocampus [50]. The biological function of hippocampal FNDC5 to the RTT model merits analysis.

Animals
Experimental procedures followed ethical guidelines and were approved by the Institutional Animal Care and Use Committee, Kaohsiung Chang Gung Memorial Hospital (project identification code: CMRPG8G0961, date of approval: 20 March 2017). Mecp2 null/y mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA; strain name: mecp2 tm1−1Bird , stock number: 003890) and bred at the Center for Laboratory Animals. Male mecp2 null/y mice with a C57BL/6J background were generated for >10 generations [53]. All behavioral assessments were performed by a rater who was blinded to the genotype. All animals were kept on a 12:12 h light-dark cycle at 22 • C with food and water available ad libitum. All experimental procedures were performed during the light-on phase of the cycle.

Music Treatment
Wild-type C57BL/6J (WT) and mecp2 null/y mice were weaned 24 days after birth. After genotyping, male mice were selected for the following study. At P25, animals were randomly divided into six groups: (1) control wild-type mice (C-WT); (2) the wild-type white noise group (W-WT), in which wild-type mice received 50-60 dB white noise at a frequency of 300-10,000 Hz; (3) the wild-type music-based intervention group (M-WT), in which wild-type mice received a music-based intervention involving 50-60 dB music at a frequency of 300-10,000 Hz; (4) control mecp2 null/y mice (C-mecp2 null/y ); (5) the mecp2 null/y white noise group (W-mecp2 null/y ), in which the mecp2 null/y mice received 50-60 dB white noise at a frequency of 300-10,000 Hz; and (6) the mecp2 null/y music-based intervention group (M-mecp2 null/y mice), in which mecp2 null/y received a music-based intervention involving 50-60 dB music at a frequency of 300-10,000 Hz. Animals were exposed to music or white noise for 6 h/day for 3 consecutive weeks. The distance between mouse cages and the sound box was 1 m. All experiments were performed in a quiet environment to avoid interference from external noise. After 3 weeks in lit conditions, animal behaviors, including locomotion, three-chamber social ability, and anxiety-like behaviors, were assessed. At the end of the experiments, mice were killed, and four brain regions, the prefrontal cortex (PFC), hippocampus, basal ganglia, and amygdala, were dissected from fresh brain tissue (see Figure S1).

Music Selection
Six compositions containing different music forms were chosen for use in this study. These included vocal and instrumental musical items. First, the classic Taiwanese melody "Rainy Night Flower", played by Evergreen symphony, was chosen. Second, we chose classical music: piano concerto Mozart Sonata for 2 pianos K448, which has been reported to be effective for treating refractory epilepsy [54,55]. Third, we used vocal music: solo compositions of Sarah Brightman, whose voice is delicate and sweet. Fourth, music sung by a church choir was selected because of its harmonious and solemn nature. Fifth, we selected preludes and fugues by Bach. Sixth, we chose a piano concerto by Chopin. Each melody line was enriched with harmony.

Open Field Test
An open field test was performed to probe the anxiety status following music exposure. Exploratory activity and anxiety-like behaviors were evaluated in an automated open field. The apparatus was set up under a digital camera, which was connected to a video recorder and a computer under the control of a Smart tracking system (TSE, Germany). Mice were placed in the center of the open field and left to explore freely for 10 min. Activity was assessed with a computer-assisted digital scan optical animal activity system, TSE multiconditioning system (TSE, Germany). The total distance moved over 10 min in the arena was recorded as a measure of locomotor activity. Time spent in a central square (20 × 20 cm 2 ) in the open field during the first 10 min was automatically recorded as center time and was used as a measure of anxiety-like behavior [56].

Three-Chamber Social Ability and Social Novelty Measurement
Three-chamber social ability and social novelty measurements were performed to characterize cognition in terms of general social ability and novelty. Social ability and social novelty measures were conducted in a three-chamber cage as previously described with the following specifications: 60 × 40 × 22 cm (L × W × H) [57]. The chamber was separated into three parts: 21 cm left and 21 cm right compartments together with an 18 cm middle compartment. There were two cylinder chambers that were 15 cm in height and 10 cm in diameter in the left and right compartments. During the social ability test, a stranger mouse (stranger 1) was placed inside the cylinder in the left compartment, with the cylinder in the right compartment remaining empty. The experimental mouse was placed in the middle chamber for 10 min and the amounts of time spent sniffing stranger 1 and the empty chamber were recorded. After 10 min, the experimental mouse and stranger 1 were taken out and the chambers were cleaned. After a 10-min break, the experimental mouse and stranger 1 were placed back in their original chambers, and a second stranger mouse (stranger 2) was placed in the cylinder in the right compartment. The amounts of time spent sniffing stranger 1 and stranger 2 by the experimental mouse were recorded during a 10 min observation period for the social novelty test. Social interaction behaviour was also measured, and the specifications for the three-chamber cage were 108 × 50 × 42 cm (L × W × H).

Quantitative Reverse Transcription (qRT-PCR) for BDNF mRNA Expression
Total RNA was extracted from fresh tissues using the TRIzol protocol (Invitrogen, San Diego, CA, USA) [58]. We assessed RNA purity and integrity with OD260/OD280 spectrophotometric measurements. A 5 µg portion of total RNA and 1.5 µg of oligo-dT primers were incubated at 70 • C for 10 min and gradually cooled to room temperature. Each RT mixture containing 25 units of M-MLV reverse transcriptase (Promega, Madison, WI, USA), 10 L 5 × reaction buffer, 0.5 mM dNTP, and nuclease-free distilled water was added to a final volume of 50 L. The amount of cDNA was quantified using the LightCycler SYBR-Green 1 Master mix (Roche, Basel, Switzerland) by real-time PCR. For normalization, Actin cDNA levels were analyzed. Ct values from each sample were obtained using Light Cycler 480 software (Roche, Basel, Switzerland). The samples were incubated at 37 • C for 90 min followed by enzyme denaturation at 95 • C for 10 min. Each PCR (20 L) contained 2 µL of RT product, 1 unit of Taq DNA polymerase (Viogene, Taipei, Taiwan), 2 µL of 10 × PCR buffer plusMgCl2, 0.2 mM of dNTP, and 0.5 M of gene-specific primers (BDNF: forward5 -GACAAGGCAACTTGGCCTAC-3 , reverse 5 -CTGTCACACACGCTCAGCTC-3 ; FNDC5: forward, 5 -CTCTCTCTTGGCTTCTCTCTTTC-3 reverse 5 -CATGGACATTGCTGAGGTACT-3 ; Actin: 5 -AGGCCAACCGTGAAAAGATG-3 and 5 -TGTGGTACGACCAGAGGCATAC-3 ). The amplified reaction was performed using a thermocycler for a single 3 min initial denaturation period at 94 • C followed by 30

Western Blotting Assessment
Fresh tissues in the prefrontal cortex, hippocampus, basal ganglia, and amygdala were selected for Western blotting. The brain regions were rapidly dissected. We homogenized the protein in a cell lysis buffer (Fermentas, Burlington, ON, Canada), containing a complete protease inhibitor cocktail (Thermo, Rockford, AL, USA). Subsquently, 12% sodium dodecyl sulphate-polyacrylamide gel was used to separate equal amounts of protein from each sample, and these were electrophoretically transferred onto polyvinylidene fluoride membranes (Millipore, Burlington, MA, USA). The membranes were incubated with rabbit anti-BDNF (1:5000; Abcam, 15 kDa) and TrkB (1:5000, Abcam, 145/95kDa) overnight at 4°C after blocking with 5% skimmed milk in Tris-buffered saline with 0.1% Tween 20 (TBST) for 1 h at 25°C. The membranes were then washed in TBST and incubated for 1 h at room temperature with horseradish peroxidase-conjugated secondary antibody (1:10,000; Abcam, Cambridge, UK). Immunoreactivity was visualized with the enhanced chemiluminescence method.

Statistical Analyses
Two-sample comparisons were performed using the Student's t test, and multiple comparisons were performed using the one-way ANOVA and the Tukey post hoc test. Statistical analyses were carried out using GraphPad Prism v.6.0 (GraphPad Software, San Diego, CA, USA). All data are presented as the mean ± SEM or with box plot diagrams, and statistical significance was accepted at the 5% level.

Conclusions
Our results provided evidence that although music-based interventions do not ameliorate motor deficits and mitigate anxiety in mecp2 null/y mice, they can be used to reduce social function impairment in mecp2 null/y mice. The effects of music-based interventions occur by upregulating BDNF gene expression without activating the TrkB receptor pathway through modulation of the FNDC5 gene in the PFC and hippocampal region. Funding: This study was supported by grants from the Research Support Scheme of Chang Gung Memorial Hospital (CMRPG8G0961). The funding source played no role in the study design; the collection, analysis, or interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication. Acknowledgments: We thank the Biostatistics Center of Kaohsiung Chang Gung Memorial Hospital for its assistance with the statistical analysis. We also thank Mei-Ling Yang for her skillful assistance with data preparation.

Conflicts of Interest:
None of the authors have any conflicts of interest to disclose. We confirm that we have read the journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.