The Trem2 R47H Alzheimer’s risk variant impairs splicing and reduces Trem2 mRNA and protein in mice but not in humans

BACKGROUND: The R47H variant of the Triggering Receptor Expressed on Myeloid cells 2 (TREM2) significantly increases the risk for late onset Alzheimer’s disease. Mouse models accurately reproducing phenotypes observed in Alzheimer’ disease patients carrying the R47H coding variant are required to understand the TREM2 related dysfunctions responsible for the enhanced risk for late onset Alzheimer’s disease. METHODS: A CRISPR/Cas9-assisted gene targeting strategy was used to generate Trem2 R47H knock-in mice. Trem2 mRNA and protein levels as well as Trem2 splicing patterns were assessed in these mice, in iPSC-derived human microglia-like cells, and in human brains from Alzheimer’s patients carrying the TREM2 R47H risk factor. RESULTS: Two independent Trem2 R47H knock-in mouse models show reduced Trem2 mRNA and protein production. In both mouse models Trem2 haploinsufficiency was due to atypical splicing of mouse Trem2 R47H, which introduced a premature stop codon. Cellular splicing assays using minigene constructs demonstrate that the R47H variant induced abnormal splicing only occurs in mice but not in humans. TREM2 mRNA levels and splicing patterns were both normal in iPSC-derived human microglia-like cells and patient brains with the TREM2 R47H variant. CONCLUSIONS: The Trem2 R47H variant activates a cryptic splice site that generates miss-spliced transcripts leading to Trem2 haploinsufficiency only in mice but not in humans. Since Trem2 R47H related phenotypes are mouse specific and do not occur in humans, humanized TREM2 R47H knock-in mice should be generated to study the cellular consequences caused by the human TREM2 R47H coding variant. Currently described phenotypes of Trem2 R47H knock-in mice can therefore not be translated to humans. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s13024-018-0280-6) contains supplementary material, which is available to authorized users.