Basic Science and Pathogenesis

Background: Our group has developed the innovative proximity labeling cell-type specific in vivo biotinylation of proteins (CIBOP) approach to quantify cell-specific in vivo proteomic and transcriptomic signatures that may lead to identify novel therapeutic targets for Alzheimer's disease (AD) pathogenesis. CIBOP uses TurboID, a biotin ligase, selectively expressed in the cell type of interest using a conditional Cre/lox genetic strategy to label the cytosolic proteome. Using mass spectrometry (MS)-based proteomics, we have found that TurboID biotinylates many RNA-binding and ribosomal proteins. We extended the CIBOP approach to obtain representative cell type-specific transcriptomes and proteomes.

Method: We crossed cell-specific Cre lines (astrocytic: Aldh1l1-Cre-ert2 and neuronal: Camk2a-Cre-ert2) and Rosa26^TurboID/wt floxed mice for cell-specific proteomic labeling (astrocyte-CIBOP and neuron-CIBOP). CIBOP and control (Cre-only) mice received tamoxifen, followed by biotin-containing water. While maintaining RNA-protein interactions, cortical tissue was lysed, biotinylated proteins were enriched via streptavidin beads, and RNA and proteins were eluted. Immunofluorescent microscopy (IF), biochemical assays, MS-based proteomics, and RNA-sequencing were completed to confirm cell-specific molecular profiling. Concordance analysis of paired proteomes and transcriptomes from CIBOP mouse brains was conducted.

Result: Western blot analysis of the cortex confirmed biotinylation of the cellular proteome of CIBOP mice when compared to controls. Cell-type specificity was further validated by IF images showing that biotin-labeled proteins colocalized with corresponding astrocytic (e.g., GFAP and NDRG2) or neuronal markers (e.g., MAP2 and beta-tubulin-3). RNA gel electrophoresis displayed high levels of RNA from CIBOP brain streptavidin pulldowns and low RNA yield from control pulldowns. MS-based proteomics and RNA-sequencing analysis showed upregulation of astrocytic proteins (e.g., Hepacam, Glu, Aqp4, Plpp3) and genes (e.g., Sox9, Aqp4, Gfap, Apoe) from astrocyte-CIBOP brain. In contrast, upregulation of neuronal proteins (e.g., Map2, Ncam1, Mapt) and genes (e.g., Pdyn, Tmem130, Ptpn7) was observed in neuron-CIBOP brain samples, confirming specificity.

Conclusion: Together, these results validate the CIBOP approach to capturing the cortical Aldh1l1-positive astrocytic and Camk2a-positive neuronal proteome and transcriptome. Our innovative in vivo cell type-specific and native-state dual-omics approach provides complementary transcriptomic and proteomic information that can extend to AD models to investigate disease mechanisms, discover new biomarkers, and identify therapeutic targets.