Deficiency of the Alzheimer’s disease risk gene SORL1 impairs mitochondrial homeostasis in human microglia

Alzheimer’s disease (AD) is associated with profound microglial dysfunction, encompassing inflammatory activation and metabolic remodeling. The AD risk gene SORL1 encodes the sorting receptor SorLA, best known for its roles in endosomal trafficking and amyloid precursor protein processing. Although SorLA is highly expressed in microglia, its functions in these cells remain incompletely understood. We recently demonstrated that SorLA deficiency in human microglia promotes lipid droplet (LD) accumulation via ER stress. Proteomic analyses from that study additionally identified mitochondrial proteins associated with SorLA, raising the possibility that SORL1 may regulate mitochondrial homeostasis in microglia. We employed affinity proteomics, CRISPR screen targeting the SorLA proximity proteome, Seahorse extracellular flux analysis, mitochondrial membrane potential measurements, live-cell imaging, and immunocytochemistry across complementary human model systems, including iPSC-derived microglia (iMG) and freshly isolated human microglia from rapid autopsies. Relevant protein relationships were further examined in postmortem human brain tissue and contextualized using large-scale human brain proteomics and donor-aware, state-resolved pseudobulk analysis of publicly available single-nucleus RNA-sequencing data. CRISPR screen targeting the SorLA proximity proteome, identified multiple mitochondrial regulators of LD accumulation, including the prohibitin complex subunits PHB1 (PHB) and PHB2 . SORL1 deficiency impaired mitochondrial function, characterized by reduced maximal respiration, spare respiratory capacity, and mitochondrial membrane potential, and drove structural remodeling reflected by increased mitochondrial volume and altered motility. Mechanistically, affinity proteomics revealed an association between SorLA and PHB; furthermore, SORL1 deletion reduced PHB protein levels in iMG. In human microglia from rapid autopsies, SorLA levels correlated selectively with PHB and the mitochondrial content marker TOMM20, but not with HSP60, indicating a specific relationship with mitochondrial content rather than generalized stress. SorLA and PHB protein levels were also positively correlated across a mixed AD and non-AD postmortem brain cohort, and PHB abundance was reduced in AD brain tissue. Furthermore, donor-aware analysis of human single-nucleus RNA sequencing (snRNA-seq) data from PFC microglia identified selective nuclear-encoded transcriptional associations: lower donor-level SORL1 expression was associated with lower expression of a PINK1–Parkin mitophagy program and higher expression of glycolytic, mitochondrial-fission, protein-import, and cristae/MICOS programs. Core OXPHOS subunits, the prohibitin complex, mtDNA-maintenance nuclear genes, and mitoribosomal programs showed no coherent association with SORL1 . These findings identify SORL1 as a contributor to mitochondrial homeostasis and bioenergetic adaptability in human microglia and reveal that reduced SorLA protein abundance leads to lower PHB protein levels. This relationship suggests that PHB-associated processes may contribute to the mitochondrial effects of SORL1 deficiency, warranting further investigation to determine whether and how PHB plays a causal role. Together, the cellular, biochemical, and human-tissue evidence links SORL1 to selective mitochondrial remodeling and lipid droplet homeostasis, providing a molecular framework for understanding how this major AD risk gene may contribute to disease-relevant microglial dysfunction.