Abstract
Triggering receptor expressed on myeloid cells 2 (TREM2), an immune receptor expressed on the surface of microglia, has been identified through genome-wide association studies to be one of the risk factors in Alzheimer's disease (AD). Several studies have also identified missense variants of TREM2 to be associated with frontotemporal dementia and Nasu-Hakola disease (NHD). To date, 51 novel missense variants of TREM2 have been identified in the literature, with the disease risk profiles of most variants still unknown. Assessing and classifying the pathogenicity of these variants is essential to investigate the disease mechanisms and develop effective treatments. Herein, we classified 44 missense variants involved in TREM2 using structural bioinformatic data with AlphaFold2. Using the protocol described in our previous work [1], we determined the structural, stability, and potential functional effects of these variants. Our evaluations of the mutations were divided into those (i) implicated in NHD, (ii) located on the transmembrane domain, (iii) surface of the IgV-like domain, (iv) buried in the receptor, and (v) on low confidence predicted local distance difference test residues. Our analysis of variants involved in NHD suggests that, while V126G imposes the greatest effects, the T66M variant exerts significantly less effects on the TREM2 structures compared to the other remaining variants. Variants in the transmembrane domain of TREM2 did not impose significant alterations to the three-dimensional structure. Outside of known variants in the IgV-like domain, we identified 10 variants that imposed significant destabilizing effects to the structure and are of potential interest. Overall, the baseline biochemical data provided from this study may be informative to experimental efforts to better classify rare coding variants of TREM2 that are of unknown biological and clinical significance.