Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

The contribution of an inflammatory component to AD was already described in the 1990's. However, a role for microglia as a driver of neuroinflammation-induced AD has only been recognized more recently, with the identification of SNPs in immune-related genes as risk factors for common sAD. In this respect, SNPs in TREM2 were found to significantly increase the risk of developing multiple neurological disorders. In particular, a 3-fold increased risk for sAD in individuals with a heterozygous TREM2 R47H mutation was observed, which is nearly equal to the APOE4 effect size. In addition, loss of TREM2 surface expression was found to result in early onset dementia with amyloid pathology. Thus, this underscores a causal role of microglia in AD and implies aberrant TREM2 function can result in amyloid pathology. Obviously, the best system to study the role of TREM2 mutations is in human brain-derived microglia, which is difficult due to limited access to human brains. Therefore, we established a human pluripotent stem cell (hPSC) model to enable the study of loss of TREM2 and the heterozygous AD-linked R47H TREM2 mutation in microglia-like cells. Using CRISPR/Cas9, we knocked in the heterozygous R47H mutation (TREM2 +/R47H), created TREM2 +/- and TREM2 -/- cell lines, and differentiated these hPSCs to monocytes and microglia-like cells (tMG). We demonstrated that the hPSC-derived tMG resemble other hPSC-microglia as well as cultured human microglia. In addition, by means of qRT-PCR and RNA sequencing we showed for the first time that in vitro differentiation to so-called monocytes using the Yanagimachi protocol results in cells that are more similar to microglia than in vivo PB-monocytes and showed this is possibly due to in vitro culture. Moreover, RNA sequencing of TREM2 WT vs. TREM2 -/- revealed amongst others an increased inflammatory response and reduced cytoskeletal remodeling. These pathways might underpin the mechanism by which phagocytic activity of TREM2-/- tMG is impaired. This is the first time these findings are replicated in human microglia-like cells and identifies interesting candidate gene networks for future studies, e.g. deranged integrin signaling, which so far have not been addressed.Furthermore, we were to our knowledge the first to address the phagocytic function of TREM2 R47H heterozygous knock-in human microglia-like cells and perform an ex vivo amyloid clearance assay on TREM2 +/R47H, TREM2 +/- and TREM2 -/- human microglia-like cells. We were able to demonstrate that only heterozygous or homozygous loss of TREM2 impaired both phagocytosis of E.Coli and amyloid plaque clearance. As the phagocytic impairment of TREM2 -/- tMG was not more profound than in TREM2 +/- tMG, we hypothesize that this is due to the severe loss of TREM2 mRNA (80 % reduction) observed upon knockout of 1 allele, and thus that phagocytosis in general is directly affected once TREM2 mRNA expression is 80 % decreased (TREM2 +/- tMG) compared to WT tMG.In respect to the lack of a phagocytic phenotype for TREM2 +/R47H tMG, one should take into account that the biggest risk factor of AD, aging, is not recreated in tMG derived from hPSCs. One could consider generating tMG by direct transdifferentiation of fibroblasts from elderly sAD patients with a TREM2+/R47H mutation that may retain such an 'aged signature' (as shown by Mertens et al., 2015 for transdifferentiated neurons), once such protocols become available. Alternatively, TREM2 +/R47H tMG could be allowed to age by grafting in immune-compromised and microglia-depleted hCSF1-expressing AD mice or SAMP8 mice, and then reanalyzed. Transcriptome analysis of such samples might reveal functions affected by this mutation upon aging and help to further unravel why TREM2+/R47H is associated with a 3-fold increased risk for sAD and how it differs from other TREM2 mutants.