Ethical approval and animal cares
All experimental procedures were conducted following the Guide for the Care and Use of Laboratory Animals of the Matsumoto University (Nagano, Japan). All experimental procedures were approved by the Animal Use Committee at Matsumoto University (approval ID: 2021-4, 2022-3, 2023-3). All experimental procedures were also confirmed by ARRIVE guidelines 2.0 (https://arriveguidelines.org/arrive-guidelines). Male C57BL/6J mice at 7, 31, 52, and 74 weeks of age were purchased from Nihon CLEA (Tokyo, Japan) and used in the present study. The mice were acclimated to the experimental environment for 1 week before being used for the experiments described below. A commercial solid diet (CE-2, Nihon CLEA) and water were supplied ad libitum. Temperature and humidity in the animal room were maintained at 23°C and 40–60%, respectively, with 12:12 h light-dark cycle.
Experimental design
In Experiment 1, the age-related changes in the histochemical properties, gene expression, and histone modifications in the tibialis anterior muscle were determined using 8-, 32-, 53-, and 75-wk-old mice (n = 3 each). In Experiment 2, Responses of gene expression and histone modifications to a single bout of exercise mice were compared between young (8-wk-old, n = 6) and middle-aged (53-wk-old, n = 6). In Experiment 3, the effects of forced H3.3 expression in skeletal muscles on motor function were examined in young mice (8-wk-old, n = 10).
Acute Exercise (Experiment 2)
The mice with the age of 8 and 53 weeks were separated into the control and exercise groups (n = 3 each). The mice in the exercise group were assigned to perform running exercise for a total of 30 min at a speed of 25 cm/s, starting at a speed of 15 cm/s, gradually increasing the speed to 20 cm/s and finally to a speed of 25 cm/s, using an animal treadmill (Panlab Harvard apparatus, Barcelona, Spain). The tibialis anterior muscles were sampled from both control and exercise groups 2 hours after the end of running exercise. Muscle sampling was performed immediately after the treatment of carbon dioxide gas. Each mouse was transferred to an inhalation chamber and subsequently exposed to an increasing concentration of carbon dioxide gas. The muscle tissues were cleaned of excess fat and connective tissue. Muscle samples were frozen in liquid nitrogen and stored at − 80°C until analysis.
Design and administration of viral vector (Experiment 3)
AAV9 was modified to encode mouse histone H3.3A (gene symbol: H3f3a) at the downstream of ACTA1 promoter (VectorBuilder Japan, Kanagawa, Japan). Thus, the transfection of this vector provides the skeletal muscle-specific H3.3 expression. As for the control group, the H3f3a sequence was replaced with untranslatable 249 bp sequence (stuffer). The mice at 8-wk-old were separated into Stuffer and H3.3 groups (n = 5 each). AAV9 vector was intravenously delivered through the tail vein under inhalation anesthesia by Isoflurane. Single shot (100 µL) was performed to administer 8 x 1011 vg per mouse. The tibialis anterior muscle was sampled at 32-wk-old.
To evaluate the effects of AAV9 vector administration on other organs, we also tested the comparison between CMV immediate early enhancer/promoter and ACTA1 promoter to express EGFP reporter (Fig. S1A). CMV promoter has driven the strong EGFP expression in liver, lung and kidney 2 weeks after the injection of vector, which was markedly reduced if ACTA1 promoter was used. Therefore, ACTA1 promoter was selected to use for H3.3 expression in the present study.
Rotarod test
Motor function was tested using a rotarod (LE8205, Barcelona, Spain) in Experiment 3. Rotarod test was performed every 2 weeks until 30-wk-old. The mice were placed on the platform before starting the rotation. The rotation of platform was started at 40 rpm, and the speed of rotation was increased 4 rpm every 5 seconds. The latency (time) to fall was measured for each mouse. The mice were placed back on the platform immediately after fall. The test was repeated 5 times, and the mean latency to fall was calculated in each mouse.
Immunohistochemistry
Cross-sections from the mid portions of the tibialis anterior muscles were cut at 10 µm in a cryostat (Leica Microsystems, Wetzlar, Germany) maintained at − 20°C. The sections were fixed in 4% paraformaldehyde for 5 min, followed by blocking in 10% goat or donkey serum diluted in PBS containing 0.1% Triton X-100 (TPBS) for 20 min. Overnight incubation at 4°C with anti-dystrophin (ab15277 or ab129996, Abcam), anti-pericentriolar material 1 (PCM1) (HPA023370, Merck), anti-type I myosin heavy chain (MyHC) (BA-D5, Developmental Studies Hybridoma Bank, Iowa City, IA), anti-type IIb MyHC (BF-F3, Developmental Studies Hybridoma Bank), and anti-type IIa MyHC (SG-71, Developmental Studies Hybridoma Bank). The antibodies were diluted at 1:100 in TPBS containing 1% bovine serum albumin (BSA). Visualization for the binding site of primary antibody was performed using Alexa Fluor 350 (A-21140, Thermo Fisher Scientific, Waltham, MA), 488(A-21121 for mouse immunoglobulin or A-21206 for rabbit immunoglobulin, Thermo Fisher Scientific), 546 (A-21045 or A-10036, Thermo Fisher Scientific) and 647 (A-31573, Thermo Fisher Scientific) diluted at 1:500 in TPBS containing 1% BSA for 1 hour. The stained sections were mounted for microscopic analysis using SlowFade Gold Antifade Mountant with 4’,6-diamidino-2-phenylindole (DAPI) (S36938, Thermo Fisher Scientific) to label nuclei or antifade mounting medium (H-1000, Vector Laboratories) without nuclear labelling. The images of whole sections were incorporated into a computer using All-in-One Fluorescence Microscope system (BZ-X710, KEYENCE, Osaka, Japan). The exposure time was set constant among all sections if the pattern to use the antibodies was same.
The muscle fiber phenotype was classified by the fluorescence intensity of each MyHC isoform into type I, IIa or IIb. The fibers without labelling with these antibodies were classified as type IIx fibers (12). All muscle fibers were targeted to analyze the fiber phenotype and size (approximately 2500 fibers per muscle section). Myonuclei were counted automatically while the green fluorescence channel labelled by the PCM1 and overlapped by the DAPI using BZ-X Analyser software (KEYENCE). The area enclosed by dystrophin (muscle fiber size) was also analyzed in all images.
Western blotting
Total histone was extracted using Epiquik Total Histone Extraction Kit (Epigentek, Farmingdale, NY). Total histone obtained from 20 mg muscle samples were extracted in 500 µL lysis buffer packaged in the kit, centrifuged at 12,000 g for 5 min at 4°C, and 300 µL supernatant was collected and mixed with 90 µL balance buffer packaged in the kit. The total histone extract was further dissolved in an equal amount of 2x SDS sample buffer (20% glycerol, 12% 2-mercaptoethanol, 4% sodium dodecyl sulfate, 100 mM Tris–HCl, and 0.05% bromophenol blue, pH 6.7) and boiled for 10 min.
Western blotting was performed as described previously (13). The following antibodies were used to detect each protein: H3.3 (ab176840; Abcam, Cambridge, UK, 1:1000), H3.1/3.2 (61629, Active Motif, Carlsbad, CA, 1:1000), H3K4me3 (9751, Cell Signaling Technology, Danvers, MA, 1:1000), H3K9me3 (61013, Active Motif, 1:1000), H3K27me3 (9733, Cell Signaling Technology, 1:1000), H3K27ac (8173, Cell Signaling Technology, 1:1000), H3K36me3 (4909, Cell Signaling Technology, 1:1000), and total H3 (4620, Cell Signaling Technology, 1:1000). The antibody-bound protein was detected by a chemiluminescence method using ChemiDoc Touch MP (Bio-Rad, Hercules, CA). The bands were quantified using image analysis software (ImageJ) (https://imagej.net/ij/). The protein level was expressed as the integrated density of the band, which was calculated as the mean density multiplied by the band area. Finally, the integrated density was compared between the experimental groups that were applied to the same membrane.
Gene expression
A piece of frozen muscle (10–20 mg) was homogenized in 1 ml of ISOGEN (NIPPON GENE, Toyama, Japan). RNA extraction was performed following the manufacturer's instructions. The final pellet of RNA was resuspended in ultrapure water. Total RNA from 8-wk-old mice (n = 3) and 75-wk-old mice (n = 3) groups in Experiment 1 was combined within each group and used to construct complementary DNA (cDNA) libraries for RNA sequencing analysis. RNA-seq analysis was performed through commercial service (Novogene, Chula Vista, CA). mRNAs were enriched with oligo(dT) beads and randomly fragmented in the fragmentation buffer, followed by cDNA synthesis using random hexamers and reverse transcriptase. After the first-strand synthesis, a custom second-strand synthesis buffer (Illumina, San Diego, CA) was added along with dNTPs, RNase H, and Escherichia coli polymerase I to generate the second strand by nick translation. The final cDNA library was prepared after purification, terminal repair, A-tailing, ligation of sequencing adapters, size selection, and PCR enrichment. The NovaSeq6000 system (Illumina) was used to obtain reads of 150-bp paired ends. Approximately 40 million reads for each group were mapped to the mouse whole genome database using the hisat2 software, and the fragments per kilobase of exon per million mapped fragments (FPKM) value was calculated for the exons of all known loci. All FPKM data obtained in the RNA sequencing analysis are available in Additional file 1.
For the quantitative analysis of gene expression, SuperScript VILO Master Mix (Thermo Fisher Scientific) was used to synthesize cDNA following the manufacturer's instructions. Total RNA (800 ng) was incubated with SuperScript VILO Master Mix at 42°C for 60 min, followed by inactivation of the enzyme at 85°C for 5 min. Synthesized cDNA was diluted to 1:100 with ultrapure dH2O and stored at − 20°C until analysis. Candidates of target genes that were up- or down-regulated by > 2 folds at 75-wk-old compared to 8-wk-old (18 upregulated and 17 downregulated genes, Additional file 1) were selected for gene expression and chromatin immunoprecipitation (ChIP) analysis, and confirmed by quantitative PCR (qPCR) using the specific primer sets (Additional file 2). Upregulated (n = 15) and downregulated (n = 14) genes were identified as aging-related genes, and targeted to analyze the gene expression and histone distributions in Experiment 1 (Fig. S2). Previously described gene set (14) was used to analyze the gene response to acute exercise in Experiment 2. These gene sets were also targeted for the analysis in Experiment 3.
ChIP
Extraction of chromatin-rich extract and chromatin immunoprecipitation were performed as described previously (15). Briefly, muscle segments (20–40 mg) were homogenized in cooled PBS. After centrifugation at 12,000 g, the pellet was fixed in 1% paraformaldehyde on ice for 10 min followed by quenching in 200 mM glycine. The pellet was resuspended in lysis buffer and sonicated using a Sonifier 250 (Branson, Danbury, CT, USA). For ChIP-qPCR analysis, sonication was repeated four times, resulting in an average DNA fragment size of 500 bp. After centrifugation at 12,000 g, the supernatant was further gel-filtrated to remove small DNA fragments and free histones, which did not form nucleosomes, and stored as the chromatin-rich extract at − 80°C until analysis.
Chromatin-rich extracts containing equal DNA content (400 ng) were combined within each group and used for the ChIP reaction. Chromatin was reacted with anti-H3.3 (ab176840, Abcam, 1:50), anti-H3.1/3.2 (61629, Active Motif, 1:50), anti-H3K4me3 (9751, Cell Signaling Technology, 1:50) or anti-H3K27me3 (9733, Cell Signaling Technology, 1:50) for 1 h at 4°C, followed by a reaction with SureBeads Protein G (1614023, for rabbit immunoglobulin) or Protein A (1614013, for mouse immunoglobulin) Magnetic Beads (Bio-Rad, Hercules, CA, USA) for 30 min at 4°C. Beads were washed and incubated with proteinase K (Takara Bio, Shiga, Japan) for 1 h at 65°C. DNA was extracted and resuspended in Tris-EDTA buffer and stored at − 20°C. ChIP rection using same antibody was tested twice to minimize the differences between reactions, and the yielded DNA was combined within each group. The level of input DNA contained in chromatin used for each ChIP reaction was also tested without any reactions.
qPCR
Quantitative PCR was performed using the StepOne Real Time PCR System (Thermo Fisher Scientific). THUNDERBIRD NEXT SYBR qPCR Mix (Toyobo, Osaka, Japan) was used for the PCR, with manufacturer-recommended dilution procedures. Sequences of primer pairs used for gene expression and ChIP-qPCR analysis are shown in Additional file 2. To analyze the histone distribution at the transcription start site (TSS), two primer pairs for ChIP-qPCR were designed at every 500 bp on 1 kbp downstream sequence from TSS. As for the exercise-related genes, previously designed primers which covered between 1 kbp upstream and downstream from TSS (14). Since the chromatin obtained in the present study mainly contained tri-nucleosomes (500 bp peak DNA fragment size), the sites for the primer design needed to be separated by at least 500 bp for non-overlapping analysis.
Quantification of the qPCR results was performed by normalizing to the cycle threshold (Ct) of the target amplification, with Gapdh or Rpl31 mRNA as the internal control for gene expression assay or with the Ct of respective input ChIP-qPCR (% input). The % inputs obtained from each locus were averaged for each mouse. For the presentation of ChIP data, the % input was further normalized by the average of all groups in each gene.
Statistical analysis
Statistical analysis was performed using BellCurve for Excel (Social Survey Research Information, Tokyo, Japan). Significant differences were examined by one-way ANOVA followed by Scheffe’s post hoc test (Experiment 1 and 2). Student’s unpaired t test was used to compare the two groups (Experiment 3). Differences were considered significant at p < 0.05.