[A24] Microglial INPP5D isoform expression is increased in AD and possible AD-associated SNP contributions to unequal allelic expression

Author(s): Diana J Zajac, University of Kentucky; Diana J Zajac, University of Kentucky; James Simpson, University of Kentucky; Josh Morganti, University of Kentucky; Steve Estus, University of Kentucky //

ABSTRACT: INPP5D (Phosphatidylinositol-3,4,5-trisphosphate 5-phosphatase 1), the gene encoding SHIP1 protein, contains single nucleotide polymorphisms (SNPs) that are strongly associated with Alzheimer’s Disease (AD) risk. In the brain and mostly in microglia, INPP5D is expressed as several isoforms. Full-length INPP5D encodes 27 exons, including an amino-terminal SH2 binding domain followed by the phosphatase domain that modulates downstream receptor signaling. Truncated isoforms lacking the SH2 domain begin from internal transcription start sites. First, we investigated INPP5D isoform expression as a function of AD status, and AD-associated SNP status to better understand the function of INPP5D in the human brain. The expression of microglial INPP5D isoforms was analyzed using quantitative polymerase chain reaction used on RNA from AD and non-AD anterior cingulate human brain samples. The AD-associated SNPs rs35349669 and rs10933431 were identified using TaqMan SNP kits (ABI). Isoform expression results were analyzed as a function of microglial gene expression (ITGAM and AIF1), total INPP5D expression, AD status, and SNP status. Total INPP5D expression is increased in AD vs non-AD samples, but there remains to be seen a significant association with SNP status. Second, single cell RNAseq was performed on APP/PS1 mice to understand the relationship between INPP5D expression and microglial activation. APP/PS1 mice showed an increased expression of disease-associated microglia compared to controls. INPP5D was expressed at equivalent levels in both homeostatic and disease-associated microglia. Expression of each INPP5D isoform was strongly correlated with microglial gene expression and showed an increase with AD neuropathology but did not show a significant associated with AD SNPs. Third, reporter SNPs in the 5’ UTR region were used to track unequal allele expression in the two full-length isoforms of INPP5D as a function of AD SNP status. Sequencing of samples heterozygous for both AD SNPs provided preliminary evidence for unequal allele expression modulated by SNP status. Our next step is to quantify the unequal expression using next generation sequencing. In summary, expression of INPP5D isoforms is increased with AD neuropathology when normalized to microglia; RNAseq shows that INPP5D expression is uniform in both disease-associated and homeostatic microglia; and the mechanisms whereby AD genetics influence INPP5D expression is not clear, but the rs35349669 allele that increases AD risk may be associated with a change in allelic expression.

Source of Funding: RF1 AG059717