Controlled cortical impact
C57/BL6J male mice (8- to 10-weeks-old) were purchased from National Laboratory Animal Center (Taiwan). After anesthesia, a craniotomy was made to open a hole with a diameter of 3 mm in the left hemisphere. Then, the sensorimotor cortex (0.5–2.5 mm caudal to bregma and 0.5–2.5 mm lateral to the midline) was impacted with electric cortical contusion impactor (Custom Design & Fabrication, Inc., USA). The diameter of impact tip was 2 mm and impact dwell time was 250 ms. Mild, moderate and severe TBI were induced with following conditions: mild TBI (impact velocity of 3 m/s, impact depth of 1 mm), moderate TBI (impact velocity of 4 m/s, impact depth of 1.5 mm), and severe TBI (impact velocity of 5 m/s, impact depth of 2 mm). Sham mice underwent a craniotomy but were not impacted by CCI.
To measures tissue viability after TBI, the brain sections of CCI mice was stained with triphenyltetrazolium chloride (TTC) (Sigma-Aldrich). Briefly, mice were sacrificed and perfused intracardially with saline solution on 4 dpi. Brains were dissected and sliced into 1.5 mm coronary sections. Then, brain sections were incubated with 2% TTC solution for 15 min at 37℃.
For immunohistochemistry, the mice were perfused intracardially with saline solution and 4% paraformaldehyde (Alfa Aesar) solution on 4 dpi. The brains were dissected and immersed in 10%, 15%, and 20% sucrose sequentially to dehydrate the brains. Then the brains were embedded in tissue freezing medium (Leica) and sliced into sliced into 10-µm cryosections by cryostat microtome (Leica CM3050 S, Leica Biosystems, USA). The cryosections were air-dried at room temperature for 30 min and then stored in -80℃ for further immunohistochemistry.
To assess the expression of PGAM5 in the brains of CCI mice, mice were sacrificed and perfused intracardially with saline solution on 4 dpi. The left hemisphere was dissected and minced in SDS lysis buffer containing protease inhibitors and protease inhibitors (240 mM tris-acetate, 1% SDS, 0.5% glycerol, 5 µM EDTA, 1mM phenylmethanesulfonylfluoride, 1mM Sodium orthovanadate, 10 ng/ml aprotinin and leupeptin). Then the tissue was homogenized with a Dounce homogenizer, followed by centrifugation at 13,000 rpm for 15 min at 4℃. The supernatant was collected for immunoblotting.
Primary glia and neuron culture
Pregnant Sprague–Dawley rats were purchased from BioLASCO Taiwan Co., Ltd. The brains of rat embryos (E18) were dissected and primary neurons were isolated and cultured in vitro as previous described [61]. For primary culture of glia cells, isolated cells were seeded on un-coated dishes on DIV0. On DIV1, suspended cells (neuron cells) were removed by change the medium to MEM (Thermo Fisher Scientific) containing 10% fetal bovine serum (SAFC Biosciences) and 1% penicillin/streptomycin (Thermo Fisher Scientific). The medium was changed every two days. On DIV8, primary neuron and glia cells were injured by scraping with a p20 pipette tip. Cells were harvested and lysed by SDS lysis buffer containing protease inhibitors and phosphatase inhibitors 24 hr (DIV9) or 48 hr (DIV10) after TBI.
Western blotting
To assess protein level in brain or cells, lysates were prepared in SDS lysis buffer containing protease inhibitors and phosphatase inhibitors. The amount of proteins in lysates were normalized by BCA protein assay kit (Millipore). Proteins were resolved by SDS-polyacrylamide gel electrophoresis and then transferred to nitrocellulose membranes (PerkinElmer). Transferred blots were incubated with primary antibodies, anti-PGAM5 (1:500, Santa Cruz, SC-515880), anti-GAPDH (1:5,000, Genetex, GTX100118), anti-TUJ1 (1:5,000, BioLegend, #801202), anti-GFAP (1:5,000, Genetex, GTX108711), anti-PGC1α (1:500, Genetex, GTX37356), anti-NRF1 (1:1,000, Genetex, GTX103179), anti-TFAM (1:500, Genetex, GTX59889), or anti-TIM23 (1:500, Santa Cruz, SC-514463) overnight. Then the blots were incubated with secondary antibodies, IRDye 800CW goat anti-rabbit IgG secondary antibody (1:1,000, LI-COR, # 926-32211) or goat anti-mouse e IgG (H + L) secondary antibody Alexa Fluor™ 700 (1:1,000, Invitrogen, # A-21036), for 1 hr. Membranes were imaged by ChemiDoc™ MP Imaging System (Bio-Rad) and the signal intensity of bands was quantified by Image Lab software (Bio-Rad, version 6.1.0).
Enhancer prediction
Putative enhancers were predicted as previously described [31]. Chromatin immunoprecipitation-sequencing datasets were downloaded from the ENCODE portal (https://www.encodeproject.org/) with the following identifiers: ENCFF033MMS, ENCFF247XHY, and ENCFF826KEG, and visualized on UCSC Genome Browser (https://genome.ucsc.edu/).
Quantification of enhancer RNA
The transcript of enhancer RNA was assessed as previously described [31]. Total RNA was isolated from cortical neurons on DIV9, followed by reverse transcription polymerase chain reaction. Then, eRNAs were amplified by specific primers and analyzed by electrophoresis. The signal intensity of bands was quantified by Gel-Pro Analyzer 3.1 software. The intensity of eRNAs in injured cortical neurons was normalized to the transcript of Gapdh and compared to un-injured cortical neurons.
Chromosome conformation capture analysis
Chromosome conformation capture (3C) assays were performed as previously described [31]. 3C template was collected from cortical neurons on DIV9. Then, the samples were digested with EcoRI (NEB), followed by ligation using T4 ligase (NEB). Ligation products were amplified by specific primers and analyzed by electrophoresis. The signal intensity of bands was quantified by Image Lab software (Bio-Rad, version 6.1.0). Amplifying primer sequences are listed in Table 1.
Table 1
Primer sequences used in this study
Primer sequences for eRNA |
Primer | Sense | Sequence |
Gapdh (ctrl) | sense | AAGGGCTCATGACCACAGTC |
Gapdh (ctrl) | antisense | TGTGAGGGAGATGCTCAGTG |
e1-1 | sense | CAATTAGAGGCAGGCAGAGTAG |
e1-1 | antisense | AAACATAAGGCTTACCCCAGAC |
e1-2 | sense | CTGATTGCTGACTGGTGCTT |
e1-2 | antisense | ACAGAAGGCTGGAGACACAA |
e1-3 | sense | CCACAGAAGGTCAGAGGTCA |
e1-3 | antisense | CCAGGCTAGGTTGGGATAAC |
e1-4 | sense | GCTGCGTAAGAGTGACAGGA |
e1-4 | antisense | AAGTGCTCAGGTAGGGAAGC |
e1-5 | sense | GGATGGGTAGGGGGTACATA |
e1-5 | antisense | GGGGAATGGATGCTAGTAGG |
e2-1 | sense | GTCACGAGTGCGACTACACT |
e2-1 | antisense | ACTCACCTTGACCTGGAAGG |
e2-2 | sense | ACCTACCTCGGTCTCTGAGT |
e2-2 | antisense | TCATAACCACCAGACGCCTt |
e2-3 | sense | TGTATGTTACGGGGTGGAGC |
e2-3 | antisense | AGACCTGGGTTTGGTTCACA |
e2-4 | sense | TTCATCAAGGAGGGACTGGG |
e2-4 | antisense | CAAAGGAGGAACAGGTTGTGG |
e2-5 | sense | TTGGAGATTCCTTCTCGCCA |
e2-5 | antisense | AGGACAGCAGTCTACAGTGG |
e2-6 | sense | AGAGCCTAGCAGCGAAGATT |
e2-6 | antisense | TAGGGTTACGAAGCCTGAGC |
e3-1 | sense | TTACACCAGGGAGCAGGAG |
e3-1 | antisense | CAGACCCGTGGAAATCTTG |
e3-2 | sense | ACCCAACCAGTGGACAGAA |
e3-2 | antisense | CTCCTCCCAGTTCGCTCTA |
e3-3 | sense | TTTAGATGTGGCGTCTCCAG |
e3-3 | antisense | CCTTCAGCCTTCATTTACCAA |
e3-4 | sense | CCACCTGTCTCACTGCTGTT |
e3-4 | antisense | CCATGTAGCCTCCTGCTGTA |
e3-5 | sense | ACACGCATACGCACACTCTT |
e3-5 | antisense | AGGTTGCCATTTTGCTGAAGT |
e3-6 | sense | GTCAAGCTACATCCAGTAAAGCT |
e3-6 | antisense | TGGCTTCATGGTTCCTGTC |
e3-7 | sense | CCAGGAGCCTGAAAGCTATG |
e3-7 | antisense | ACTCATCAGCCAGTGTCACG |
e4-1 | sense | GACTGTCTTTCCCACAGCAA |
e4-1 | antisense | GCACACTTAGCCAAGCACAT |
e4-2 | sense | GCTAAGTGGCTCAGAGAATGG |
e4-2 | antisense | CCTAAATGGGTTTCCTGTGAA |
e4-3 | sense | TTCTGCTTGGGTAAGGGAAG |
e4-3 | antisense | TATGTGAAAGCGGATTGGAG |
e4-4 | sense | AAGTTGCCCATAGGACAGGA |
e4-4 | antisense | TTCCCAGTTCTGACCCAGAT |
e4-5 | sense | TCATAAAGCCAGGCGATACAC |
e4-5 | antisense | AGGGTCTCTTGGGATTCTCC |
e4-6 | sense | TCGGAGTTCACACCAAACAC |
e4-6 | antisense | GAGATGACCCATCCCTTCAG |
e4-7 | sense | CATACACTCGGGCACTCAGT |
e4-7 | antisense | GGTCTGGGTATTTGAAATCAGG |
e4-8 | sense | CCACTGCTGTCACATTCCAT |
e4-8 | antisense | CGTTCTGAATGGTTGCCAAG |
e4-9 | sense | GAGATGCTCTCCCTCCAGAA |
e4-9 | antisense | TCAGAGTCTGTGCCCACTGT |
e5-1 | sense | CCAGTGTTTCAACCCTGATGT |
e5-1 | antisense | AATTTCACGGATGGAGGAAGC |
e5-2 | sense | CGGGCACCCAGTTAATTCAT |
e5-2 | antisense | TCACCCCATCACAAGAATGC |
e5-3 | sense | AGGATCAGGTTGTTGGTCTG |
e5-3 | antisense | GGCAAGGACGAATGTAAGCA |
e5-4 | sense | CCCTGCTAACCATTAGTGCC |
e5-4 | antisense | TGAAGTTTGAGCCCACTTGAC |
e5-5 | sense | CTGGCTCCTCTAACCTTTCGT |
e5-5 | antisense | GAACACTCCCTAGAGTTTCCCTA |
e5-6 | sense | CTGTCTTCAATCTCAAAGGGG |
e5-6 | antisense | CTGTCTGTCATCCACGCAAT |
e6-1 | sense | AACCGCATATACCCCGAAGA |
e6-1 | antisense | TGTGATACTTGGACGGCAGA |
e6-2 | sense | CAGACTAGGTCAGGCACCAA |
e6-2 | antisense | TGACATCCATCCCATGCGTA |
e6-3 | sense | AATGCTAACCAAGTCGCTGC |
e6-3 | antisense | TATCCGACACCTTCACCCAC |
e6-4 | sense | CCCCGCCCCCATTATCTTAT |
e6-4 | antisense | TGAGTCCCACAACATCGGAA |
e6-5 | sense | ATAGAAGTGTGAGTGCCCCC |
e6-5 | antisense | CATCACCACGGATGCCAAAT |
e6-6 | sense | AGCCTTAAATAACGCCCCCT |
e6-6 | antisense | AGCATTGTTCTTTGCCCCAG |
e6-1-a | sense | TCCACAAGAAGCTCCCAGA |
e6-1-a | antisense | AAATACCTCCTGCTCGGCTT |
e6-1-b | sense | GCAATGGTGTCCAAGAGTCA |
e6-1-b | antisense | TGTGTTTGGTCAGGTCTCC |
e6-1-c | sense | GCTCTGGCTCACTGAAGTCT |
e6-1-c | antisense | GAACAGTCTTGGGTGTTGCG |
e6-1-d | sense | AAGACTGGAGACCAGGCAAT |
e6-1-d | antisense | GGGAATGGGAACGACATCT |
e6-1-e | sense | CGTCTGGCTCAGAGAGGATT |
e6-1-e | antisense | TACCATGAACTCCCCACCTG |
Primer sequences for 3C assays |
Primer | Sense | Sequence |
loading control | sense | TTCTTGGGGTGAAGCAACACAT |
loading control | antisense | ACCAGAGCAGGACCTGTTAAATG |
e6-1-a | sense | AGAAGCATTGACAAGCTCCGC |
Pgam5 promoter | antisense | GATCATGGACAGGGTAGGCAG |
DNA constructs
pcDNA3 PARL-FLAG-CT wild type (Addgene plasmid # 13639), pcDNA3 PARL-FLAG-CT S65A + T69A + S70A (Addgene plasmid # 13616), pcDNA3 PARL-FLAG-CT S65D + T69D + S70D (Addgene plasmid # 13617) were gifts from Luca Pellegrini [38, 62]. Knockdown constructs, shLacZ, shPgam5, and shParl constructs, were purchased from the National RNAi Core Facility of Academia Sinica (Taipei, Taiwan), Full-length PGAM5 and PGAM5(Δ2–24) were both subcloned into pEGFP-C1 construct via NheI-AgeI sites. All constructs were transfected in cells using Lipofectamine 2000 (Invitrogen) in accordance with its protocol.
Injury assay of cortical neurons
Cortical neurons (7.5 × 105 cells/ml) were cultured in 6-well plate on DIV0. Cortical neurons were transiently transfected with PGAM5, PGAM5(Δ2–24), hPARL, hPARL(AAA), hPARL(DDD), shlacZ, shPgam5, or shParl constructs on DIV7. To visualize neurites, cells were co-transfected with EGFP-C2 construct. On DIV8, cortical neurons were scratch-injured with a p20 pipette tip. Cortical neurons were imaged using Carl Zeiss Observer Z1 microscope on DIV9. The length of re-growing neurites was measured using ImageJ software (plugins NeuronJ).
Measurement of mitophagy
Mitophagy was assessed by the co-localization of mitochondria and lysosomes in hippocampal neurons [63]. hippocampal neurons (4 × 104 cells/ml) were cultured in 2-well chamber slices (Thermo Fisher Scientific). On DIV7, hippocampal neurons were transiently transfected with PGAM5, PGAM5(Δ2–24), hPARL(AAA) or hPARL(DDD) constructs. To visualize mitochondria, hippocampal neurons were transfected with MitoGFP. On DIV9, hippocampal neurons were incubated in culture medium containing 75nM LysoTracker™ Red DND-99 (Invitrogen) for 45 min at 37℃. Then, hippocampal neurons were imaged using Carl Zeiss LSM800 confocal microscope. The co-localization of mitochondria and lysosomes were assessed using a custom-written MATLAB code in conjunction with the Image Processing Toolbox of MATLAB (version R2021b). MitoGFP+ and LysoTracker+ area were obtained according to specific criteria: MitoGFP intensity > 80 and area > 0.1 µm2; LysoTracker intensity > 60 and area > 0.05 µm2. Then, co-localization% of mitochondria and lysosomes was quantified as MitoGFP+ LysoTracker+ area divided by MitoGFP+ area.
Measurement of ΔΨm
To investigate the change of ΔΨm after TBI, hippocampal neurons (4 × 104 cells/ml) were cultured in 2-well chamber slices. Hippocampal neurons were scratch-injured with a p2 pipette tip on DIV8. On DIV9 and DIV10, hippocampal neurons were incubated in culture medium containing 250 nM tetramethylrhodamine (Invitrogen) for 30 min at 37℃, followed by washed twice with PBS. Hippocampal neurons were imaged using Carl Zeiss Observer Z1 microscope. The intensity of TMRM were quantified by a custom-written MATLAB code in conjunction with the Image Processing Toolbox of MATLAB (version R2021b).
Immunostaining
Immunohistochemistry was performed as previously described [30]. Briefly, cryosections were incubated in antigen retrieval solution (Nacalai Tesque) at 70℃ for 20 min to unmask antigenic sites. After incubated in blocking buffer containing 1% BSA (Sigma-Aldrich) for 2 hr, cryosections were incubated in 1% BSA containing anti-NeuN antibody (1:500, Genetex, GTX132974) overnight at 4℃. Finally, cryosections were incubated in goat anti-rabbit secondary antibody (1:500, Invitrogen, A21428) for 1 hr and imaged using Carl Zeiss LSM800 confocal microscope.
To investigate the sub-cellular location of PGAM5, PGAM5 and mitochondria were visualized using immunostaining. Hippocampal neurons (4 × 104 cells/ml) were cultured on cover glass (Marienfeld) and injured on DIV8. Then, hippocampal neurons were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) on DIV9 or DIV10. After incubated in 1% BSA for 1 hr, cells were incubated with anti-PGAM5 (1:100, Santa Cruz, SC-515880) and anti-TOM20 (1:100, Santa Cruz, SC-17764) antibodies, overnight at 4℃. Then, cells were incubated with secondary antibodies, goat anti-mouse secondary antibody (1:1,000, Invitrogen, A11001) and goat anti-rabbit secondary antibody (1:1,000, Invitrogen, A21428), for 1 hr at room temperature. Finally, the cells were mounted in ProLong™ Gold Antifade Mountant (Invitrogen) and imaged using Carl Zeiss LSM800 confocal microscope. The co-localization of PGAM5 and TOM20 were assessed using a custom-written MATLAB code in conjunction with the Image Processing Toolbox of MATLAB. PGAM5+ and TOM20+ area were obtained according to specific criteria: PGAM5 intensity > 40; TOM20 intensity > 60. Then, the percentage of mitochondrial PGAM5 was quantified as PGAM5+TOM20+ area divided by PGAM5+ area.
Measurement of mitochondrial mass
To evaluate mitochondrial mass after TBI, hippocampal neurons (4 × 104 cells/ml) were cultured in 2-well chamber slices. Hippocampal neurons were injured on DIV8. On DIV9 and DIV10, hippocampal neurons were incubated in culture medium containing 500 nM MitoTracker Red (Invitrogen) for 30 min at 37℃. After be washed twice with culture medium, hippocampal neurons were imaged under Carl Zeiss Observer Z1 microscope. Total intensity of MitoTracker Red in individual images were obtained using a custom-written MATLAB code in conjunction with the Image Processing Toolbox of MATLAB.
Neuro2a cells (1 × 105 cells/ml) were incubated in 90% MEM (Thermo Fisher Scientific), supplemented with 2 mM L-glutamine (Thermo Fisher Scientific), 1.5 g/L sodium bicarbonate (Thermo Fisher Scientific), 0.1 mM non-essential amino acids (Thermo Fisher Scientific), 1.0 mM sodium pyruvate (Thermo Fisher Scientific), and 10% fetal bovine serum. To assess mitochondrial mass in neuro2a cells, neuro2a cells were incubated in culture medium containing 0.1 µM MitoBright LT Deep Red (Dojindo) for 15 min at 37℃. After be washed twice with culture medium, neuro2a cells were imaged using Carl Zeiss LSM800 confocal microscope. Total intensity of MitoBright LT Deep Red in individual cells were obtained using MATLAB.
FCCP administration
Neuro2a cells (2 × 105 cells/ml) were treated with 0.1 µM FCCP (Sigma-Aldrich), 1.0 µM FCCP or 0.1% DMSO (Ctrl). Cells were harvested by SDS lysis buffer containing protease inhibitors and phosphatase inhibitors 24 hr after treatment. The proteins were further analyzed using immunoblotting.
For CCI mice, FCCP was administrated intranasally 6 h after CCI as previously described [64]. Intranasal administration allows drugs to bypass blood-brain barrier and increase brain bioavailability [64, 65]. FCCP or DMSO (vehicle) was diluted in 24 µl saline. Saline containing FCCP or DMSO was intranasally administrated in 2 rounds. In the first round, mouse was intranasally administrated 6 µl saline to the left nostril. Then the mouse was held 15 s to confirm the saline was fully administrated into nose, followed by an administration of 6 µl saline to the right nostril. After a 2-min rest, the second round was performed. The total of 24 µl saline was administrated 6 h after CCI.
Rotarod test
To evaluate motor coordination of CCI mice, rotarod test was performed as previously described [46]. To perform the pre-training trial, mice were placed on the rod (Ugo-Basile, Italy, #47650) rotating at 4 rotations per minute (rpm) for 60 s on 1 dpi. Then mice were placed on a rotating rod accelerating from 4 to 40 rpm over 3 min and the latency to fall was recorded. Mice were tested 3 times a day for 1–4 dpi and there was a 10-min rest between each trial. The average latency to fall for the 3 trials was recorded.
Grid test
To evaluate spontaneous motor deficits of CCI mice, grid test was performed as previously described [47]. Mice were placed on an elevated steel gird with dimensions of 35 × 20 cm and with grid size of 1.1 × 1.1 cm. Mice were allowed to walk around for 5 min. A foot fault is defined as a paw missed a wire edge or slipped off. Foot faults of each limb in 5 min were recorded. Grid test was performed on -1 dpi to assess basal motor function before CCI. After CCI, grid tests were performed on 1, 3, 6 dpi to evaluate motor deficit.
Statistical analysis
All results are expressed as mean ± SEM or violin plot from at least three independent experiments. Data were analyzed by unpaired two-tailed Student’s t-tests or ANOVA with Dunnett’s multiple comparisons or Tukey’s multiple comparisons using Prism software. Statistical significance is defined as p < 0.05.