Deficiency of the lysosomal exonuclease PLD3 impacts the degradative route

Abstract

Phospholipase D3 (PLD3) is a 5'-3' exonuclease that resides in lysosomes where it is implicated in the regulation of inflammatory responses by degrading ssDNA. Polymorphisms in the PLD3 gene have been linked to late-onset Alzheimer's disease (LOAD; Cruchaga et al., PMID: 24336208), but contradictory data exist on the impact of PLD3 deficiency on APP processing. Given PLD3 is primarily expressed in neurons within the brain, we were interested in exploring its role in neuronal endolysosomal homeostasis. We generated PLD3-/- neuroblastoma SH-SY5Y clonal cell lines, using Crispr/cas9 gene editing, that were subsequently stably rescued with lentiviral vector technology. For this, the wild-type PLD3 sequence was compared to variants harbouring LOAD-associated risk single-nucleotide polymorphisms (SNPs), including M6R, K228R, V232M, N236S, N284S and T426A. PLD3 exonuclease kinetics towards different substrates were measured using EFQO assays (Cappel et al., PMID: 33288674). Autophagy and mitophagy dysregulations, as possible important catalysts for DNA delivery to lysosomes, were further scrutinized with biochemical assays and confocal imaging. In parallel, we investigated the endolysosomal morphology by electron microscopy. Firstly, most SNP variants, except the M6R and T426A variants, display a significantly reduced exonuclease activity; indicating a loss-of-function for these variants. Overall, knock-out of PLD3 leads to elevations in lysosomal and autophagosomal components, including a defect in mitophagy, lysosomal cholesterol accumulation and an accretion of the uncleaved/inactive CatD. These aberrant features can be rescued by re-introducing wild-type PLD3, but not PLD3 SNP-variants. This underscores a critical role for PLD3 in lysosomal homeostasis, likely mediated through its exonuclease activity. Given different variants elicit different exonuclease kinetics towards different ssDNA substrates, this suggests that PLD3 contributes to upholding lysosomal DNA levels within a narrow range, which is required for a normal lysosomal proteostasis. PLD3 depletion and exonuclease dysfunction-causing SNPs promote lysosomal impairment and, as such, a congestion of the degradative route. We hypothesize that this could relate to a dysfunctional exonuclease activity of PLD3 in lysosomes. Given the observation of engulfed mitochondria within autophagosomes, the substrates for PLD3 could potentially come from mtDNA.