Network analysis of the brain proteome of GRN knockout mice reveals pathogenic mechanisms shared in human frontotemporal dementia caused by GRN mutations

Background Mutations in the GRN gene, encoding the progranulin (PGRN) protein, are a common cause of Frontotemporal dementia (FTD) and induce haploinsufficiency of PGRN. PGRN is composed of 7.5 repeating domains called granulins (GRNs 1‐7). PGRN is processed into GRNs within the lysosome, where we hypothesize they are critical for lysosome homeostasis and function. However, it is unclear how decreased levels of lysosomal PGRN/GRNs cause FTD, Alzheimer’s disease, and related neurodegenerative disorders. Methods We performed quantitative proteomic analysis of whole‐brain tissue from wild‐type ( Grn +/+ ) and PGRN knockout ( Grn ‐/‐ ) mice at 3‐ and 19‐months of age using 10‐plex Tandem Mass Tag (TMT) isobaric labeling mass spectrometry. Protein changes were validated using immunobloting, immunohistochemistry, and ELISAs. Results Deep proteome analysis of mouse brains detected 8,695 proteins. Weighted correlation network analysis (WGCNA) of the brain proteome of Grn ‐/‐ compared to Grn +/+ mice identified 29 modules of highly co‐expressed proteins. In particular, 3 modules strongly correlated to Grn deficiency, increased with age, and were enriched with lysosomal proteins (Gpnmb, Ctsd, Tpp1) and inflammatory proteins (Lgals3, GFAP, CD44). Immunohistochemistry revealed age‐dependent increases in multiple lysosomal proteins and inflammation throughout the Grn ‐/‐ brain, particularly in the thalamus. Importantly, we discovered similar changes in the brain and cerebrospinal fluid (CSF) of individuals with GRN mutations that cause FTD (FTD‐ GRN ). GPNMB and galectin‐3 ( LGALS3 ) were significantly elevated in the lysates of FTD‐GRN brain. Further, GPNMB levels were significantly increased in human FTD‐ GRN CSF compared to controls and other genetic forms of FTD ( MAPT and C9orf72 ). Conclusions These data suggest that common pathogenic pathways are dysregulated in human FTD cases with GRN haploinsufficiency and Grn ‐/‐ mice. Moreover, GPNMB is a novel, potential biomarker for FTD. Our findings support the idea that insufficiency of PGRN and GRNs in humans cause FTD, AD, and related neurodegenerative diseases through lysosomal dysfunction and neuroinflammation and suggest novel therapeutic approaches.