Spontaneous PDAC KIC mouse model show an obvious cachexia phenotype with adipose tissue and muscle loss associated with cancer. (A) Schematic representation of the overall study. (B) Total body weight analysis during disease progression (from 4 to 11 weeks of age) in KIC and CTRL male mice. Data are mean +/- 95% of CI and n represents the total number of mice analyzed in each group. Two-way ANOVA Bonferroni test; n=15 or more for control and KIC mice per time-point; **p<0.01, ****p<0.0001. (C) Representative HPS staining of pancreas from healthy control versus cancer KIC mice. Dark arrows point to stromal fibrotic areas relative to pancreatic ductal adenocarcinoma. Scale bar = 500 µm. (D) Total tissue weights of pancreas, liver, spleen, gonadal adipose tissue and gastrocnemius muscle of healthy control mice versus end-stage (9-11 weeks) cancer KIC mice. n=17 control and n=16 KIC. Data are mean +/- SD. For the gastrocnemius muscle, the weight mean of the two muscles of each individual was used to calculate the mean. Unpaired two-tailed Mann Whitney t-tests; *p<0.05, **** p<0.0001, and n represents the number of mice that were analyzed.

PDAC mice show muscle atrophy associated with fiber’s size decrease and activation of the proteasome pathway. (A) (Left) Representative images of transversal sections of gastrocnemius muscle from cancer KIC male mice (9-11-week-old bearing a tumor) and age-matched healthy control, following Masson’s trichrome staining. Scale bars, 1mm. (Right) Diagrams of cross section areas of gastrocnemius muscle from healthy control mice and cancer KIC mice (n=8/group). Data are mean +/- SD. Unpaired two-tailed Mann Whitney t-tests; **p<0.01, and n represents the number of mice that were analyzed. (B) (Left) Representative immunostaining of gastrocnemius muscle cross sections from control and KIC mice stained with DAPI (blue; nuclei), and anti-laminin antibody (red) to evaluate muscle fiber’s size. Scale bars, 40µm. (Right) Number of muscle fibers counted in gastrocnemius cross sections from control and KIC mice, and mean fiber’s (100) minimal Feret’s diameter in gastrocnemius muscle from control and KIC mice. Data are mean +/- SD. Unpaired two-tailed Mann Whitney t-tests; ns = non-significant, ****p<0.0001, and n represents the number of mice that were analyzed. (C) (Left) Representative immunostaining of transverse sections of gastrocnemius muscle from control and KIC male mice stained for anti-MuRF1 (green) to evaluate muscle atrophy, anti-laminin (red) to evaluate muscle fibers, and merged images. Scale bars, 40µm. (Right) MuRF1 immunofluorescent staining quantification in transversal sections of gastrocnemius muscle from control and KIC mice. Data are mean +/- SEM. Unpaired one-tailed Mann Whitney t-tests; **p<0.01 and n represents the number of mice that were analyzed. (D) (Left) Representative immunoblot for MuRF1 in gastrocnemius muscle from CTRL and KIC mice (n=5/group). (Right) Quantification of immunoblots to show the MuRF1 protein levels in gastrocnemius muscle from CTRL and KIC mice. The relative protein amount was calculated from the control mean in two different experiments. n=5 mice/group. Data are mean +/- SD. Unpaired two-tailed Mann Whitney t-tests; **p<0.01 and n represents the number of mice that were analyzed.

Noninvasive MR investigation of gastrocnemius muscle function and bioenergetics. This longitudinal investigation was done in the same KIC mice, at 7-8-week-old (pre-tumoral) and at 9-11 weeks of age (bearing a tumor), and same age-matched control mice. (A) (Left) Gastrocnemius muscle volume was calculated from MR images. (Middle and Right) Absolute and specific force production in response to a tetanic electrostimulation train (150 Hz; 0.5-s duration). (B) (Left) Twitch force-generating capacity (expressed as percent of maximal twitch force value) during the 6-min bout of exercise performed simultaneously to the dynamic 31P-MRS acquisition. (Middle) Twitch force level at the end of the 6-min bout. (Right) Twitch half-relaxation time averaged over the whole bout of exercise. (C) Changes in PCr level (Left) and pH (Right) throughout the 6-min exercise and the 16-min post-exercise recovery period; for both panels, the first time-point (t = 0) indicates the basal value. (D) Basal PCr level and pH were measured in resting muscle. Time constant of PCr resynthesis and maximal oxidative capacity were measured during recovery. The drop of PCr and acidosis were determined at the end of the 6-min bout of exercise. Data are means ± SEM. Controls, n = 7; KIC, n = 5. All samples were normally distributed according to the Shapiro-Wilk test. Significant differences were determined by two-factor (group × age) repeated measures ANOVA followed when appropriate by Tukey-Kramer post-hoc multiple comparison tests. *p<0.05, **p<0.01, ***p<0.001.

Cancer-associated sarcopenia is correlated with a decrease of mitochondrial respiratory activity and respiratory complexes. (A-B) Representative Seahorse XF oxygen consumption rate assays performed on mitochondria isolated from gastrocnemius muscle of age-matched healthy control and cancer (9-11-week-old bearing a tumor) KIC mice. The curves were normalized to 1 µg of mitochondrial protein (the assays were done with 2.5 µg of mitochondria). (A) Mitochondrial coupling assay. The analysis medium was supplemented with ADP (5mM), succinate (10mM), and rotenone (2µM). (B) Electron transport chain assay performed on the same samples of isolated mitochondria. The analysis medium was supplemented with pyruvate (5mM), malate (1mM) and FCCP (4µM) to allow maximal respiration. Data are mean +/- SD and n represents the number of mice analyzed. Two-way ANOVA Bonferroni test; ns = non-significant, *p<0.05, *** p<0.001, **** p<0.0001. (C-D) Quantification of the different parameters of mitochondrial respiratory activity, normalized per µg of mitochondrial protein. (C) Parameters calculated from the mitochondrial coupling assay. From left to right: basal respiration, measured as the basal OCR subtracted from the background OCR (i.e. OCR after antimycin A addition); proton leak, measured as the OCR after oligomycin addition subtracted from the background OCR; ATP production from mitochondria, measured as the basal OCR subtracted from the OCR after oligomycin addition; maximal respiration, measured as the OCR after FCCP addition subtracted from the background OCR; spare capacity i.e. the difference between maximal and basal mitochondrial respiration. (D) Respiratory complexes activities calculated from the OCR values obtained in the Electron transport chain assay. From left to right: complex I activity, measured as the basal OCR subtracted from the OCR after rotenone addition; complex II activity, measured as the difference between succinate-driven OCR and OCR after antimycin A addition; complex IV respiration, measured as the difference between TMPD-driven OCR and OCR after antimycin A addition. Data are mean +/- SD. Unpaired two-tailed Mann Whitney t-tests; ns = non-significant, *** p<0.001, ****p<0.0001; n represents the number of mice analyzed per condition. (E) Left: Representative immunoblot for mitochondrial complexes using the rodent cocktail OXPHOS antibodies specific of one protein for each complex: CI (NFUDB8), CII (SDHB), CIII(UQCRC2), CIV(MTCO1), CV(ATP5A), in gastrocnemius muscle from CTRL and KIC mice (n=5/group). Right: Relative protein amount of CI (NFUDB8), CII (SDHB), CIII(UQCRC2), CIV(MTCO1), CV(ATP5A) in gastrocnemius muscle from CTRL and KIC male mice. The relative protein amount was calculated from the mean of CTRL from 2 different experiments. n=5 mice/group. Data are mean +/- SD. Unpaired two-tailed Mann Whitney t-tests; ns = non-significant with p-value indicated above, *p<0.05, **p<0.01. (F) Representative immunostaining of gastrocnemius muscle cross sections from control (n=5) and KIC (n=5) mice stained for mitochondrial electron transport chain complexes I to V (green) to evaluate mitochondrial mass, DAPI (blue) to evaluate nucleus, laminin (red) to evaluate muscle fibers, and merged images as indicated. Scale bars, 100µm.

Mitochondrial structure is profoundly altered in cancer KIC muscles. (A-C) Transmission electron microscopy (TEM) analysis of muscles showing altered mitochondria in cancer mouse muscle. (A) Representative TEM micrographs of longitudinal and transversal sections of gastrocnemius muscles from CTRL and cancer KIC mice showing mitochondria ultrastructure. Scale bar = 1µm or 0.2 µm. Yellow triangles point to mitochondria. (B) Quantification of mitochondria morphological characteristics. From top to bottom: mitochondrial area (µm²), mitochondrial minimal Feret’s diameter, and mitochondrial circularity in longitudinal sections of gastrocnemius muscles from CTRL healthy mice and cancer KIC mice. n=3 mice/group and 100 mitochondria measured/mouse. Data are mean +/- SD. Unpaired two-tailed Mann Whitney t-test. **** p<0.0001. (C) Quantification of the number of mitochondria per mm² in longitudinal and transversal sections of gastrocnemius muscles from CTRL healthy mice and Cancer KIC mice. n=3 mice/group and 15 or more pictures have been taken and measured/mouse. Data are mean +/- SD. Unpaired two-tailed Mann Whitney t-test; ns = non-significant. (D) Mitochondrial mass of total isolated mitochondria (in µg of proteins), normalized by the related entire gastrocnemius muscle mass (in mg) in CTRL and KIC mice. Data are mean +/- SD; n=6 mice/group. Unpaired two-tailed Mann Whitney t-tests; **p<0.01. (E) Relative mitochondrial DNA content in gastrocnemius muscle of CTRL and KIC mice using relative level of mtCOX3 (left) and mt12S (right) as mitochondrial genes normalized on gNDUFV1 as nuclear gene. Data are mean +/- SD, n=8 mice/group. Unpaired two-tailed Mann Whitney t-tests, ****p<0.0001. (F) Representative immunoblots for Mitofusin1+2 and OPA1 (involved in mitochondrial fusion), PGC1α (involved in mitochondrial biogenesis), TFAM (involved in mitochondrial DNA replication, repair and transcription) and Actin as loading control in gastrocnemius muscle from CTRL (n=5) and KIC (n=4) mice.

Transcriptomic analysis demonstrates multiple mitochondria-associated gene dysregulations in cancer sarcopenic muscles. Ingenuity Pathways Analysis (IPA) showing functions, pathways and diseases significantly dysregulated in gastrocnemius muscle of cancer KIC mice compared to healthy control mice, associated with genes of the mitochondrial interactome extracted from MitoXplorer tool with significant dysregulations (FDR>0.05) = 178 genes. Each process is represented using z-score>0.5 as upregulated (blue) or downregulated (red), and the significance represented as -log(p-value) with black dots connected with a dotted line. Muscle pathways are framed by a square. Common pathways between total genes and the mitochondrial interactome are highlighted in yellow. Mitochondrial pathways are highlighted in orange. Other metabolic pathways are underlined.

Proteomic analysis shows that numerous mitochondrial proteins are dysregulated in cancer sarcopenic muscle. IPA enrichment analysis of dysregulated functions, pathways and diseases in gastrocnemius muscle of cancer KIC compared to control mice, associated with proteins of the mitochondrial interactome extracted from MitoXplorer tool with significant dysregulations; (p-value<0.05, peptide>1, log2(Foldchange)>0.5) = 53 proteins. Each process is represented using z-score>0.5 as upregulated (blue) or downregulated (red), and the significance represented as –log(p-value) with black dots connected with a dotted line. Common pathways between total proteins and the mitochondrial interactome are highlighted in orange. Mitochondrial pathways are highlighted in yellow. Muscle pathways are framed by a square. Common pathways between total genes and the mitochondrial interactome are highlighted in yellow. Mitochondrial pathways are highlighted in orange. Other metabolic pathways are underlined. Common mitochondrial dysregulated pathways between transcriptome and proteome are indicated by a blue circle which is empty when the z-score is the same and contains a cross when the z-score is opposite.

Comparison of mitoXplorer heatmaps for transcriptomics and proteomics analyses (related to Figures S6 and S9).