Figure 7.

Pathogenic tau^V337M impacts subcellular proteostasis during aging in a contrasting manner

Western blots of detergent-soluble and -insoluble fractions from control flies (gray) and isogenic knockin flies that express human tau^V337M (green), a mutant tau that causes frontotemporal dementia in humans. Western blotting with anti-GFP antibodies indicates the levels of Fluc^DM-EGFP sensors, whereas Ponceau staining and β-actin are used as normalization controls.

(A–C) Analysis of detergent-soluble and -insoluble fractions from head extracts (enriched for tissues of the CNS) indicates that heterozygous tau^V337M mutations increase the detergent-insoluble levels of the cytoplasmic Fluc^DM reporter during aging (at 30 and 60 days) but not at a young age (10 days) compared to isogenic controls. There is little effect on the solubility of mito-Fluc^DM and NLS-Fluc^DM, indicating that tau^V337M reduces cytoplasmic proteostasis but has no substantial effect on mitochondrial and nuclear proteostasis in the CNS.

(D–F) Western blots of detergent-soluble and insoluble fractions from skeletal muscle (thoracic fractions) of flies with heterozygous tau^V337M mutations. There is no significant effect of pathogenic tau on Fluc^DM (D) and mito-Fluc^DM (E) levels, suggesting that the skeletal muscle may be resistant to tau^V337M-induced defects in cytoplasmic proteostasis (D) compared to the CNS (A). However, heterozygous tau^V337M mutations significantly reduce the detergent-insoluble levels of nuclearly localized NLS-Fluc^DM (F), suggesting that pathogenic tau^V337M may derange protein import into the nuclei of skeletal muscle.

In (A)–(F), the ages analyzed for each genotype are 10, 30, and 60 days. n = 3 (biological replicates) with the mean ± SD indicated; ∗p < 0.05, ∗∗p < 0.01 (unpaired two-tailed t test).

(G–I) Ultra-deep-coverage TMT mass spectrometry indicates that pathogenic tau^V337M divergently regulates the solubility of distinct protein sets. These proteomics analyses (18-plex TMT) are based on detergent-soluble and -insoluble fractions from isogenic knockin flies that express human tau^WT and pathogenic human tau^V337M at 10 and 60 days of age.

(G) The difference in the insoluble/soluble protein levels in tau^V337M versus tau^WT at approximately day 60. There are 432 proteins of 4,012 quantified proteins that display significantly (p < 0.05) increased (n = 140; n = 20 with log2R > 1) and decreased (n = 292, n = 43 with log2R < −1) insolubility in tau^V337M versus tau^WT at day 60. These include mitochondrial (pink) and nuclear (green) proteins.

(H and I) Examples of proteins with increased (H) and decreased (I) insolubility in response to tau^V337M in old age compared to tau^WT at the same age (black line of comparison) and compared to tau^V337M at a young age (green line). n = 3 (biological replicates) with the mean ± SD indicated; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 (one-way ANOVA).