Abstract
AbstractAlzheimer’s disease and related dementias (ADRD) and Osteoporosis (OP) are two prevalent diseases of aging with numerous epidemiological associations, but the underlying molecular mechanisms contributing to this association are unknown. We used WGCNA (weighted gene co-expression network analysis) to develop transcriptomic networks in bone and brain tissue using two different studies to discover common molecular mechanisms. We used RNA-sequencing data from the dorsolateral prefrontal cortex tissue of autopsied brains in 629 participants from ROSMAP (Religious Orders Study and the Memory and Aging Project), including a subset of 298 meeting criteria for inclusion in five ADRD categories and the full set in a secondary analysis, and RNA array data from transiliac bone in 84 participants from the Oslo study of postmenopausal women. After developing each network, we analyzed associations between modules (groups of co-expressed genes) with multiple bone and neurological traits, examined overlap in modules between networks, and performed pathway enrichment analysis to discover conserved mechanisms. We discovered three modules in ROSMAP that showed significant associations with ADRD and bone related traits and four modules in Oslo that showed significant associations with multiple bone outcomes. We found significant module overlap between the two networks, most notably among those modules linked to canonical Wnt signaling and skeletal tissue homeostasis and development. These results were preserved with a network from the full ROSMAP cohort (n=629), which included a broader spectrum of participants. Our results require validation in experimental studies but show support for Wnt signaling as an important driver of pathology in OP and ADRD. We additionally show a strong link between Dementia with Lewy bodies and bone outcomes. These results have translational significance in the development of novel treatments and biomarkers for both ADRD and OP.
Publisher
Cold Spring Harbor Laboratory