Mitochondrial spongiotic brain disease: astrocytic stress and harmful rapamycin and ketosis effect

(A, B, C, D, E, F, G, H, I) Rapamycin treatment started at presymptomatic (A, B, C, D, E, F) and at a moderately advanced (G, H, I) phenotype stage. (A) Schematic of presymptomatic rapamycin treatment. Daily i.p. injections of rapamycin or vehicle were initiated at 1.5 mo and were continued until humane end point in body weight loss was reached (2.3 mo, arrow). Green area shows treatment period and pink area represents symptom progression in untreated TwKO^astro mice. (B) Body weight progression upon rapamycin treatment initiated at the presymptomatic stage. Weight change is shown as a fold change (FC) relative to the body weight at treatment initiation (zero time point). n = 9 per group. Mean with SD is shown. (C) Immunoblot analysis of total S6 and phospho-S6 proteins. Ctr Veh n = 4, Ctr Rapa n = 4, TwKO^astro Rapa n = 4, TwKO^astro Veh n = 5. 2× is a Ctr Veh sample, with a twofold amount of protein loaded on the lane compared with other samples. (D) Brain pathology, rapamycin treatment. Left: Representative cortical sections show spongiotic encephalopathy. Hematoxylin and eosin staining of TwKO^astro treated with vehicle or rapamycin. Arrows show sponge-like holes typical for spongiotic encephalopathy. Quantification below: counts of holes per field of view, n = 7 mice per group, two sections per mouse, three fields of view per section. Counts from the fields of view of one mouse are plotted with the same symbol. Scale bars, 50 μm. Right: Astrocyte activation. Glial fibrillary acidic protein (GFAP), immunofluorescent detection, cortex, representative of n = 7 mice per group, two sections per mouse. Below: mRNA expression of GFAP and vimentin (reactive gliosis markers) measured by RT-quantitative PCR. Ctr Veh n = 8, Ctr Rapa n = 8, TwKO^astro Veh n = 7–8, TwKO^astro Rapa n = 7. (E) Metabolomics analysis. All metabolites with P-value < 0.01 in any of the two-way ANOVA tests (interaction, genotype, and treatment) are plotted. Z-scores are calculated from intensity values. Note that metabolite is plotted more than once if it passes selection criteria in more than one test. n = 6 per group. Full data set is shown in Fig S2C. (F) Metabolomics, full data set is shown in Fig S2C. n = 6 per group. Multiplicity corrected P-values are shown, calculated using the Benjamini–Hochberg procedure. (G, H, I) Rapamycin treatment started at a moderately advanced phenotype stage. (G) Schematic of the rapamycin treatment at a moderately advanced disease stage of TwKO^astro mice. Daily i.p. injections of rapamycin or vehicle were started at a moderately advanced phenotype stage (3 mo) and were continued until humane end point in body weight loss was reached (3.5 mo, arrow). Green area shows treatment period and the pink area represents symptom progression in untreated TwKO^astro mice. (H) Body weight progression upon rapamycin treatment initiated at a moderately advanced disease stage. Ctr Veh n = 3, Ctr Rapa n = 5, TwKO^astro Veh n = 7, TwKO^astro Rapa n = 8. Weight change is shown as an FC relative to the body weight at treatment initiation (zero time point). Mean with SD is shown. (I) Quantification of spongiotic encephalopathy. Counts of holes per field of view in brain sections stained with hematoxylin and eosin, 1–2 sections per mouse, three fields of view per section. Counts from fields of view of the same mouse are plotted with the same symbol. TwKO^astro Veh n = 6, TwKO^astro Rapa n = 8. Ctr, control mice; Veh, vehicle-treated mice; Rapa, rapamycin-treated mice. (C, D) P-values are shown, calculated using two-way ANOVA followed by Tukey correction for multiple comparisons. For all graphs, FC is a fold change ratio, relative to corresponding Ctr mice. For all box and whiskers plots, the box extends from the 25^th to 75^th percentiles; whiskers show all points minimum to maximum; symbols indicate biological replicates. Results of statistical testing can be found in Table S4 and Supplemental Data 2.

Ketogenic diet started upon weaning of the mice. (A) Schematic representation of the ketogenic diet experiment. Ketogenic or chow diet was administered to the mice upon weaning. Experiment was continued until humane end point body weight loss of TwKO^astro was reached (3.2 mo, arrow). Green area shows a treatment period and pink area represents a symptom progression in untreated TwKO^astro mice. (B) Blood β-hydroxybutyrate levels 8 wk after start of the experiment. Ctr Chow n = 10, Ctr Keto n = 7, TwKO^astro Chow n = 9, TwKO^astro Keto n = 8. (C) Body weight progression. Ctr Chow n = 12, Ctr Keto n = 8, TwKO^astro Chow n = 16, TwKO^astro Keto n = 14. Weight change is shown as fold change (FC) relative to the zero time point. Mean with SD is shown. (D) Brain pathology, ketogenic diet. Left: Spongiotic encephalopathy. Representative cortical sections, hematoxylin and eosin staining of TwKO^astro fed with chow or ketogenic diet. Arrows show sponge-like holes typical for spongiotic encephalopathy. Quantification below: counts of holes per field of view, TwKO^astro Chow n = 6, TwKO^astro Keto = 8, one to two sections per mouse, three fields of view per section. Counts from the fields of view of one mouse are plotted with the same symbol. Scale bars, 50 μm. Right: Astrogliosis. GFAP, immunofluorescent detection, cortex, representative of n = 8 mice per group, two sections per mouse. Below: mRNA expression of GFAP and vimentin (astrocyte activation markers) measured using RT-quantitative PCR. Ctr Chow n = 7, Ctr Keto n = 8, TwKO^astro Chow n = 8, TwKO^astro Keto n = 8. (E) Immunoblot analysis of transsulfuration pathway enzymes CTH and CBS. Densitometry of Western blot signals: a protein signal normalized to the total protein in the membrane. n = 5 per group. 2× is a Ctr Chow sample with a twofold amount of protein loaded on the lane, compared with other samples. (F) Metabolomics analysis. Metabolites with P-value < 0.01 in the two-way ANOVA interaction tests, and a cluster of metabolites with P-value < 0.01 in the two-way ANOVA genotypes tests are plotted (see selection of the cluster in Fig S3D). Z-scores are calculated from intensity values. Note that metabolite is plotted more than once if it passes selection criteria in more than one test. Ctr Chow, Ctr Keto n = 6; TwKO^astro Chow, TwKO^astro Keto n = 5. Full data set is shown in Fig S3C. (G) Metabolomics, full data set is shown in Fig S3C. Ctr Chow n = 6, Ctr Keto n = 6, TwKO^astro Chow n = 5, TwKO^astro Keto n = 5. Multiplicity corrected P-values are shown, calculated using the Benjamini–Hochberg procedure. Ctr, control mice; Chow, mice fed with chow diet; Keto, mice fed with ketogenic diet. For all graphs, FC is a fold change ratio, relative to corresponding Ctr mice. For all box and whiskers plots, the box extends from the 25^th to 75^th percentiles; whiskers show all points minimum to maximum; symbols indicate biological replicates. (C, D, E) P-values are shown, calculated using two-way ANOVA followed by Tukey correction for multiple comparisons. Results of statistical testing can be found in Table S4 and Supplemental Data 2.