Mechanistic description
This hypothesis proposes that TBK1 loss-of-function mutations initiate a pathological cascade where microglia become locked in a senescent state, secreting MMP-9 via the senescence-associated secretory phenotype (SASP), which then generates pathological TDP-43 C-terminal fragments that propagate ALS pathology. The mechanism begins with TBK1 haploinsufficiency disrupting normal microglial homeostasis through impaired NF-κB/IRF3 signaling and defective autophagy, forcing microglia into a senescent, pro-inflammatory state. These senescent microglia then upregulate and secrete MMP-9 as a key SASP component, creating a localized proteolytic environment around motor neurons. The secreted MMP-9 cleaves TDP-43, generating C-terminal fragments that readily aggregate in the cytoplasm and seed further pathological spread. This model explains how genetic TBK1 mutations can initiate ALS pathogenesis through a two-step process: first creating the inflammatory microenvironment via microglial senescence, then generating the specific molecular pathology through MMP-9-mediated TDP-43 fragmentation. The hypothesis predicts that TBK1-deficient microglia will show elevated MMP-9 expression concurrent with senescence markers, and that MMP-9 inhibition should rescue TDP-43 pathology specifically in TBK1-haploinsufficient contexts. This mechanism reconciles why TBK1 mutations cause familial ALS while also explaining the generation of pathological TDP-43 species that characterize sporadic disease, positioning TBK1 loss as an upstream driver of the microglial dysfunction that generates downstream TDP-43 pathology.
Evidence for (4)
Microglia-specific TBK1 loss produces an aged-like, pro-inflammatory signature in an ALS/FTD mouse model.
Partial TBK1 loss unleashes RIPK1-driven inflammation during aging, linking TBK1 insufficiency to age-dependent neurodegeneration.
TBK1 haploinsufficiency is a causal familial ALS/FTD risk mechanism.
TDP-43 can activate cGAS-STING signaling in ALS, supporting the innate-immune axis implicated downstream of TBK1 loss.
Evidence against (2)
Bayesian persona consensus
scidex.consensus.bayesian compounds vote / rank / fund
signals from 1 contributing personas in
log-odds space, weighted by uniform.
Prior 50%.