Fig 1.
Inhibition of glucose-dependent increased CAPER protein expression suppresses cell proliferation and induces autophagy-mediated vacuolization.
A. Glucose dependent increase in CAPER expression. A graph of qRT-PCR showing relative CAPER mRNA levels at 2 days after transfecting each siRNA (black solid bar: control siRNA, white solid bar: siRNA targeting CAPER) from the cells cultured in the media containing the indicated concentrations of glucose. Levels of CAPER mRNA are normalized with the amount of β-2MG as an internal control. Each bar represents the mean value of normalized mRNA levels of CAPER from triplicates with an error bar (SEM). Data were analyzed by t-test using GraphPad software. B. Inhibition of CAPER expression suppresses glucose-dependent cell proliferation. A growth curve showing the decreased cell numbers that were counted at 3 days after transfecting either siControl (black solid line) or siCAPER (black dotted line). Cells were cultured in the media containing the indicated concentrations of glucose. C. Knockdown of CAPER induces vacuolization with enhanced lysosomal activities. Representative microscopic photos showing the morphology of cells 2 days after transfection of indicated siRNAs (upper panel). Representative microscopic pictures showing cells stained with Lysosensor Green DND-189 (green) 2 days after transfecting with indicated siRNAs (lower panel). Pictures were taken as described in Materials and Methods. D. Representative microscopic pictures showing morphology at 2 days (Left panel) and crystal violet staining pictures showing cell proliferation at 6 days (right panel) after transfection with indicated siRNAs followed by treatment with either DMSO or 10 μM of Chloroquine (CQ) or 10 μM of E-64d. E. (i) Western blot showing that LC3II (an autophagic marker) was induced but not cleaved caspase 3 (an apoptotic marker) in cells treated with siCAPER compared to siControl. Cells were harvested and processed for Western blot analyses 2 days after transfection with the indicated siRNAs. (ii) Western blot showing the efficacy of E64d and CQ on LC3. Cells were harvested and processed for Western blot analyses 2 days after transfection with the indicated siRNAs followed by treatment of E64d and CQ for another 1 day. Band intensity was quantitated by Image J software (NIH). F. A representative microscopic picture (i) and a quantitative graph (ii) showing that RFP-LC3 (red) and GFP-LC3 (green) dots in cells stably expressing tandem fluorescence LC3 fusion proteins (RFP-GFP-LC3) transfected with either siControl or siCAPER. DAPI staining (blue) was used to mark DNA. Pictures were taken 2 days after transfection. A quantitative graph showing the statistical significance of the increase of GFP-LC3 dots (left) and RFP-LC3 dots (right) in cells treated with siCAPER (blue) as compared to cells with siControl (red). Numbers of dots were counted 2 days after transfection with the indicated siRNA. A graph presents numbers of dots counted from the 5 random fields (each point) from three independent experiments. Average (middle bar) and SEM (error bar) are shown. Data were analyzed by t-test using GraphPad software (* p<0.5, ** p<0.01, ***p<0.005). G. Western blot showing lower p62 protein amount in cells knocked down of CAPER than control harvested 2 days after transfection. Actin was served as a loading control. Band intensity was quantitated by Image J software (NIH). H. Fluorescence pictures (left panel) showing carboxy-DCFDA (green) stained cells after transfection with either siControl (left panel) or siCAPER and their corresponding quantitative graphs (right panel) Pictures were taken at “72 hours” after siRNA transfection as described in the experimental procedure (right panel). A scale bar is present. Staining intensity was quantified by Image J software (NIH) and statistical significance was calculated by student’s t-test using GraphPad software.
Fig 2.
CAPER increases glucose-dependent mitochondrial respiration by regulating glucose-induced mitochondrial transcription via ERR α -Gabpa.
A. Fluorescence pictures (upper panel) showing MitoSox (red) stained cells after transfection with either siControl (left panel) or siCAPER (right panel) and their corresponding quantitative graphs (lower panel). Pictures were taken at 24 hours after siRNA transfection. A scale bar is present. Staining intensity was quantified by Image J software and statistical significance was calculated by student’s t-test using GraphPad software. B. A graph presenting the area under the curve (AUC) of basal OCR from cells 24 hours after transfection with either siControl (black) or siCAPER (white) with assay media containing 50mM pyruvate. C. Graphs showing the ratio of mitochondrial DNA copy number (mtND2) to nuclear DNA copy number (cyclophilin) (i) and the intensity of Mitotracker Red (ii). Cells transfected with siCAPER (blue) are compared with siControl (red). Genomic DNAs were isolated from cells harvested at the indicated time after transfection followed by qPCR to quantitate relative ratio of DNA copy numbers by measuring amount of mitochondrial ND2 and nuclear cyclophilin. D. Graphs showing lower mRNA levels of mitochondrial genes and mitochondrial transcription machinery and representative Gabpa target genes encoding ETC complex I in cells transfected with siCAPER (white) than cells with siControl (black). E. (i) Immunoprecipitation assays showing that endogenous CAPER and endogenous ERR α interact with each other. Cleared whole cell extracts in BC180-0.05 (BC buffer containing 180mM KCl with 0.05% of NP-40) from AML 12 cells were subjected to immunoprecipitation with either a rabbit IgG or anti-ERR α antibody. After 4 hours of binding, protein A/G (Santa Cruz) beads were added for another hour. Bound proteins were washed 5 times with BC180-0.1 (BC buffer containing 180mM KCl with 0.1% of NP-40) followed by acid elution using glycine (pH2.5). Elutes were subjected to SDS-PAGE to score the presence of CAPER by Western blot. (ii) Luciferase reporter assays showing that CAPER coactivates ERR αand Gabpa to enhance luciferase reporter activity from the Gabpa promoter. Luciferase assays were performed 48 hours after transfection as described in the Materials and Methods section. (iii) Chromatin immunoprecipitation assay showing that CAPER is present on the ERR α binding sites in Gabpa promoters and the Gabpa binding sites in the Tfam promoter. The procedures were essentially as described in the Materials and Methods. F. qRT-PCR revealed that CAPER and its targets are regulated by glucose in a dose dependent manner. Glucose-dependent induction of target genes were diminished with siCAPER (white) compared to siControl (black). 6 hours after transfection, cells were changed into glucose-free media containing the indicated concentration of glucose for 48 hours. RNAs were extracted and subjected to qRT-PCR analyses as described in the Materials and Methods section. G. (i) A Graph showing real time measurements of OCR of cells treated with either siControl (red) or siCAPER (blue) containing glucose using a XF analyzer in the presence of consecutive treatments with an ATPase inhibitor, oligomycin (A), followed by a protonophore, carbonyl cyanide 4-trifluoromethoxy phenylhydrazone (B), and then lastly by a complex III inhibitor, Antimycin A (C). (ii) A Graph showing real time measurements of OCR of cells treated with either siControl (black) or siCAPER (gray) without glucose using a XF analyzer in the presence of consecutive treatments with an ATPase inhibitor, oligomycin (A), followed by a protonophore, carbonyl cyanide 4-trifluoromethoxy phenylhydrazone (B), and then lastly by a complex III inhibitor, Antimycin A (C).
Fig 3.
CAPER inhibition reprograms nuclear transcriptomes and metabolomes resembling retrograde responses.
A. Pie charts showing the IPA results of top physiological functions (i) and molecular functions (ii) enriched in CAPER-dependent transcriptomes. Each section represents the percentage of the genes in the indicated functional category out of entire CAPER-dependent transcriptomes. B. A graph showing the IPA results of top canonical pathways enriched in CAPER-dependent transcriptomes. Each bar represents log10 (p value) of each functional category out of the entire CAPER-dependent transcriptomes. C. Depletion of CAPER disturbs cellular carbon and nitrogen balance. A heat map (left panel) showing altered metabolites by Log 2 (FC: fold change) (left colum-24 hour samples) and by Log 10(p values) (p values adjusted for multiple comparisons using Benjamini-Hochberg method) (right column). Scale of red (positive values) and green (negative values) are represented as a color code on top of the figure. A matrix (right panel) showing the correlation distance calculated by Spearman’s correlation analysis among the metabolites from cells 24 hours after transfection with siControl and siCAPER. Scale of red (positive values) and green (negative values) are represented as a color key on top of the figure.
Fig 4.
CAPER activates c-Myc leading to activating genes involved in amino acid mediated anaplerosis.
A. A gene-metabolite network showing the IPA results to identify upstream regulators. B. (i) QRT-PCR analyses showing that the mRNA levels of c-Myc are reduced in cells transfected with siCAPER (white) compared to ones with siControl (black). (ii) Luciferase reporter assays showing that CAPER coactivates NF-κB to enhance luciferase reporter activity from the c-Myc promoter. Luciferase assays were performed 48 hours after transfection as described in the Materials and Methods section. (iii) Chromatin immunoprecipitation assay showing that CAPER is present on the NF-κB binding sites in the c-Myc promoters. A graph presents results from crystal violet staining (iv) and microscopic pictures of cell morphology (v; upper panel) and LC3 western blot (v; lower panel) at 2 days after infection with adenovirus encoding either GFP or c-Myc into cells transfected with the indicated siRNAs. C. (i) A heat map (right panel) showing the relative amount of the enzymes in anaplerotic pathways, analyzed by qRT-PCR by Log 2 (FC: fold change) (left colum-24 hour samples) and by Log 10(p values) (p values adjusted for multiple comparisons using Benjamini-Hochberg method) (right column). Scale of red (positive values) and green (negative values) are represented as a color code on bottom of the figure. (ii) A schematic diagram of the CAPER dependent anaplerotic pathways. D. A graph showing average (AUC) of OCR from cells treated with either siControl (black) or siCAPER (white) with a XF analyzer using assay media containing only glutamine. Average of raw OCR values from 9 replicates was plotted and error bars represent SEM (* p<0.05).
Fig 5.
CAPER deficiency decreased carbon flux into glycolysis and TCA cycles leading to defective compensatory mitochondrial respiratory responses.
A. Glucose-derived carbon incorporation into Glucose, G6P/F6P and F1/2, 6-BP in cells depleted of CAPER. A graph showing the average percentage of 13C labeled metabolites from three biological replicates of cells transfected with siCAPER (dotted line) and ones in cells with siControl (solid line). Cells were harvested at 24 hours after transfection followed by the procedures described in Materials and methods. Error bars represent SEM. B. Graphs showing the ratio of GSH/GSSG from the cells after transfection with either siControl (black solid) or siCAPER (white) in the absence (i) or the presence (ii) of the indicated treatment. A bar represents the mean value of relative GSH/GSSG from triplicate with an error bar (SEM). Data were analyzed by t-test using GraphPad software. C. A microscopic image shows the morphology (i) and crystal violet staining indicating the effect on the proliferation (ii) of cells transfected with either siControl or siCAPER followed by treating with either DMSO or NAC treatments for 24 hours. D. (i) A microscopic image (upper panel) showing the morphology and western blot (lower panel) scoring LC3 isoforms of cells transfected with either siControl or siCAPER followed by treating with either DMSO or 10 mM F1, 6-BP treatments for 24 hours. (ii) A microscopic picture (upper panel) and its corresponding quantitative graph (lower panel) of crystal violet staining shows the effect on the proliferation of cells transfected with either siControl or siCAPER followed by treating with either DMSO or 10 mM F1, 6-BP for 24 hours. E. Decreased glucose-derived carbon incorporation into the shown metabolites in cells depleted of CAPER. A graph showing the average percentage of 13C labeled metabolites from three biological replicates of cells transfected with siCAPER (dotted line) and ones in cells with siControl (solid line). Cells were harvested at 24 hours after transfection followed by the procedures described in Materials and methods. Error bars represent SEM. F. (i) A heat map (right panel) showing the relative amount of the indicated enzymes analyzed by qRT-PCR by Log 2 (FC: fold change) (left colum-24 hour samples) and by Log 10(p values) (p values adjusted for multiple comparisons using Benjamini-Hochberg method) (right column). Scale of red (positive values) and green (negative values) are represented as a color code on bottom of the figure. (ii) ChIP assays showing the presence of CAPER on the c-Myc binding sites of the promoter of Hk1. G. Graphs showing amount of total NAD (NADt), NAD, NADH in the cells transfected with siControl and siCAPER followed by treating NAM (left) or glucose (right). G0 and G25 indicate that cells were cultured in media containing either 0 or 25mM glucose, respectively. Average of raw values from 3 replicates was plotted and error bars represent SEM. H. A graph showing that ATP content per cell at the indicated time after transfection with siCAPER (solid white) compared to siControl (solid black).
Fig 6.
The conserved bioenergetic role of CAPER is essential for organismic survival and reproductive capacity in C. elegans.
A. Graphs showing that knockdown of Y55F3AM.3 (worm CAPER) (blue) starting from L1 stage (i) and L4 stage (ii) of rrf-3 (pk1426) strains reduces life span by 64% of control (red) with mean life span of 6 days and by 53% of control (red) with mean life span of 8 days, respectively. B. A graph showing the brood size from the rrf-3 (pk1426) strains fed with bacteria expressing either vector control (red) or the RNAi clone targeting Y55F3AM.3 (blue) since L1 stage. C. A quantitative graph showing a number of puncta of either control (red) or Y55F3AM.3-inactivated (blue) 5-day old adult Lgg-1::mCherry strain fed RNAi since L1 stage. D. Graphs showing that knockdown of Y55F3AM.3 decreases (i) FCCP-induced OCR (n = 100) of 1-day-old adults, (ii) ATP content per worm from 2-day-old and 3-day-old but not 1-day-old adult animals (n = 30 for each set) of rrf-3(pk1426) strains fed with the indicated RNAi clones since L1 stage. (iii) A bar graph showing the GSH/GSSG from 2-day-old worms of rrf-3(pk1426) strains fed with the indicated RNAi clones since L1 stage. E. Graphs showing the results of qRT-PCR of that knockdown of Y55F3AM.3 reduce representative worm target genes from 2-day-old adult animals. Mean value with standard of error of means are presented. F. RNA inactivation of Y55F3AM.3 disturbs carbon and nitrogen balance in whole worms. A heat map showing altered metabolites by Log 2 (FC: fold change) (left two colums-24 hour and 48 hour samples) and by Log 10(p values) (p values adjusted for multiple comparisons using Benjamini-Hochberg method) (right two columns). Scale of red (positive values) and green (negative values) are represented as a color code on top of the figure. G. Conservation of CAPER-dependent metabolic programs. Comparison of CAPER-dependent association among metabolites were calculated by Fisher’s z transformation using Spearman correlation coefficients 91% of pairs of each metabolites are interchangeable.
Fig 7.
A schematic model suggesting how CAPER functions.