Identification of the minimal active soluble TREM2 sequence for modulating microglial phenotypes and amyloid pathology

Abstract

Background: TREM2 is a microglial receptor genetically linked to the risk for Alzheimer’s disease (AD). The cerebrospinal fluid (CSF) levels of soluble TREM2 (sTREM2) have emerged as a valuable biomarker for the disease progression in AD and higher CSF levels of sTREM2 are linked to slower cognitive decline. Increasing sTREM2 in mouse models of amyloidosis reduces amyloid-related pathology through modulating microglial functions, suggesting a beneficial role of sTREM2 in microglia biology and AD pathology. Methods: In the current study, we performed serial C- and N-terminal truncations of sTREM2 protein to define the minimal sequence requirement for sTREM2 function. We initially assessed the impacts of sTREM2 mutants on microglial functions by measuring cell viability and inflammatory responses. The binding of the sTREM2 mutants to oligomeric Aβ was determined by solid-phase protein binding assay and dot blot assay. We further evaluated the impacts of sTREM2 mutants on amyloid-related pathology by direct stereotaxic injection of sTREM2 proteins into the brain of 5xFAD mice. Results: We found that both sTREM2 fragments 41–81 and 51–81 enhance cell viability and inflammatory responses in primary microglia. However, the fragment 51–81 exhibited impaired affinity to oligomeric Aβ. When administrated to the 5xFAD mice brain, the sTREM2 fragment 41–81, but not 51–81, increased the number of plaque-associated microglia and reduced the plaque deposition. Interestingly, the fragment 41–81 was more efficient than the physiological form of sTREM2 in ameliorating Aβ-related pathology. Conclusions: Our results indicate that the interaction of sTREM2 truncated variants with Aβ is essential for enhancing microglial recruitment to the vicinity of an amyloid plaque and reducing the plaque load. Importantly, we identified a 41-amino acid sequence of sTREM2 that is sufficient for modulating microglial functions and more potent than the full-length sTREM2 in reducing the plaque load and the plaque-associated neurotoxicity. Taken together, our data provide more insights into the mechanisms underlying sTREM2 function and the minimal active sTREM2 sequence represents a promising candidate for AD therapy.