Fig. 1: Multi-tissue metabolomics in the C26 mouse model of cachexia reveals large cachexia-specific alterations in the metabolite profiles.
a, A schematic overview of the experimental workflow. Mice were divided into four groups: mice injected with PBS, healthy controls, no tumour (Ctrl, grey, no weight loss); mice injected with NC26 cancer cells, non-cachectic tumour controls (Non-cax, blue, no weight loss); mice injected with C26 cancer cells and killed before onset of weight loss, pre-cachectic tumour mice (Pre-cax, light red, no weight loss); mice injected with C26 cancer cells and killed once they developed cachexia, cachectic tumour mice (Cax, dark red, mean body weight (BW) loss of 10%). On the day of euthanasia, mice were fasted for 6 h and injected with an isotopic tracer ([13C6]-glucose). Tissues (plasma, liver, eWAT, iWAT, heart, GC muscles, soleus and tumour) were collected exactly 1 h later. Tissues were then processed for tracer metabolomics and results submitted to bioinformatics. n = 4 animals per group. See also Extended Data Figs. 2 and 3. b, Kinetics of body weight loss expressed as a percentage of initial body weight. c, Final tumour weight. Data are mean ± s.e.m. Statistical analysis: paired two-way ANOVA with Dunnett’s post-hoc tests versus Ctrl (b) and unpaired Kruskal–Wallis with Dunn’s post-hoc test (c). d, Total number of metabolites per tissue included in the analysis after filtering (Methods). See also Supplementary Table 1 (sum of all isotopologues; log-transformed imputed, scaled data) and our WebApp (https://m3cav.metabolomics.fgu.cas.cz/). e–l, PLSDA score plots of samples based on metabolites log-transformed imputed and scaled data for each organ, tumour and plasma; see icon legend in a. Ellipses represent 95% confidence intervals. m, Number of metabolites significantly altered in the time course of cachexia development. Grey: unchanged in Non-cax, Pre-cax and Cax versus Ctrl. Blue: significant in Non-cax versus Ctrl. Light red: significant in Pre-cax versus Ctrl. Dark red: significant in Cax versus Ctrl. List of significantly different metabolites per tissue can be found in Supplementary Table 1q–x. n, Heatmaps based on hierarchical clustering of all metabolites (Extended Data Fig. 3), which are significantly altered in at least one metabolic tissue of Cax mice, manually organised per metabolite class. Data are represented as log2 fold change (FC) (tumour group/controls). Tissues from left to right: plasma, liver, eWAT, iWAT, heart, GC muscle, soleus muscle, tumour. Groups from left to right: blue, Non-cax/Ctrl; light red, Pre-cax/Ctrl; dark red, Cax/Ctrl. Tumour: light red Pre-cax/Non-cax, dark red Cax/Non-cax. A list of metabolites and associated classes can be found in Supplementary Table 1y. m,n, Statistical analysis of filtered data: one-way ANOVA following post-hoc correction based on Tukey’s honestly significant difference procedure. Panel a and icons in e–i, k and l created with BioRender.com; icon in j reproduced from Servier Medical Art (https://smart.servier.com/) under a Creative Commons license CC BY 4.0.
