Definition of regulatory elements and transcription factors controlling immune cell gene expression at single cell resolution using single cell ATAC-Seq

Authors: YANG, PENGXIN - Iowa State University; CORBETT, RYAN - Iowa State University; DAHARSH, LANCE - Iowa State University; HERRERA-URIBE, JUBER - Iowa State University; Byrne, Kristen; Loving, Crystal; TUGGLE, CHRISTOPHER - Iowa State University

Submitted to: Plant and Animal Genome Conference
Publication Type: Abstract Only
Publication Acceptance Date: 1/13/2023
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: The cis-regulatory networks of differentially expressed genes (DEG) in porcine peripheral blood mononuclear cells (pPBMC) reported in Herrera-Uribe et al., 2021 are not well understood. By profiling the chromatin accessibility of pPBMC collected from two FAANG founder pigs using scATAC-seq, we assigned 11 cell types to this scATAC-seq dataset. We found 11872 unique cell-type differentially accessible peaks (DAP) and identified the transcription factors (TFs) whose binding motifs are enriched in these DAP by performing TF binding motif (TFBM) enrichment analysis using such identified DAP. We found both general and cell-type specific TF landscapes of the TFs in the pPBMC. Unsurprisingly, the binding motif of TFs playing a crucial role in multiple cell types or lineages, like PU.1 (also known as SPI1), ETS, ETS1 and ETV2 are ubiquitously enriched in all cell types. The binding motif of PAX5, PAX6 and EBF are only enriched in B cells which is compatible with the fact that PAX5 is regarded as a B cell marker and PAX6 has a similar binding motif to that of PAX5. We also present novel data showing that the binding motif for Nur77 (a TF expressed in pig Treg cells (Gu et al., 2022), is most enriched in CD2posGD and mixed CD8a+ abT/NK cells. Additionally, our dataset may potentially identify new TSS used in specific immune cell types. For instance, we identified a cis-element region overlapping with the CSF1R gene, a monocyte marker differentially accessible in monocytes. This DAP is located in the middle of CSF1R; However, 2 of 5 CSF1R transcripts have 5’ ends overlapping this CSF1R DAP. Thus, the scATACseq data may be identifying novel and/or cell-type specific TSS. To identify potential regulatory regions of DEG, cis-co-accessibility networks (CCAN) were generated for each DEG whose TSS overlaps with a DAP in the matching cell type using Cicero. Each CCAN has the following characteristics: the “hub” peak of a CCAN overlaps with the TSS of a DEG in that cell type; each of the rest of the peaks was assigned to the same CCAN as the “hub” peak had co-accessibility score with the “hub” peak is at least 0.05. Across 11 cell types, we identified 244 such CCANs. Further, to predict the potential regulators of genes associated with such CCAN, TFBM analysis was conducted using the peaks from each CCAN. As a result, the binding motif of 70 TFs (41 unique TFs) were found enriched in CCANs associated with 45 DEG (43 unique genes) in 8 of 11 cell types. As an extension of the FAANG project our results demonstrate the power of scATAC-seq to annotate cell types, verify known or find de novo TSS of genes, and particularly decipher the cis-regulatory network by connecting the regulator and targeted genes across the whole genome.