ETV2 transcriptionally activates Rig1 gene expression and promotes reprogramming of the endothelial lineage

Abstract

ETV2 is an essential transcription factor as Etv2 null murine embryos lack all vasculature, blood and are lethal early during embryogenesis. Previous studies have established that ETV2 functions as a pioneer factor and directly reprograms fibroblasts to endothelial cells. However, the underlying molecular mechanisms regulating this reprogramming process remain incompletely defined. In the present study, we examined the ETV2-RIG1 cascade as regulators that govern ETV2-mediated reprogramming. Mouse embryonic fibroblasts (MEFs) harboring an inducible ETV2 expression system were used to overexpress ETV2 and reprogram these somatic cells to the endothelial lineage. Single-cell RNA-seq from reprogrammed fibroblasts defined the induction of the transcriptional network involved in Rig1-like receptor signaling pathways. Studies using ChIP-seq, electrophoretic mobility shift assays, and transcriptional assays demonstrated that ETV2 was a direct upstream activator of Rig1 gene expression. We further demonstrated that the knockdown of Rig1 and separately, Nfκb1 using shRNA significantly reduced the efficiency of endothelial cell reprogramming. These results highlight that ETV2 reprograms fibroblasts to endothelial cells by directly activating RIG1. These findings extend our current understanding of the molecular mechanisms underlying ETV2-mediated reprogramming and will be important in the design of revascularization strategies for the treatment of ischemic tissues such as ischemic heart disease.

Keywords: ETV2; RIG1; Reprogramming.

Fig. 1

Upregulation of stress response transcripts in fibroblasts undergoing reprograming. (a) Timeline of ETV2 overexpression in MEFs with doxycycline administration (1 μg/ml). (b) UMAP of scRNA-seq for uninduced iHA-ETV2 MEFs (948 cells), 24 h (3539 cells) and 7 days (7202 cells) post-doxycycline induction (i.e. ETV2 overexpression). (c) GO terms upregulated in D1 reprogrammed MEFs compared to the wildtype. (d) Transcriptional expression level of Etv2 (left) and Rig1 (right) projected on top of the UMAP plot. (e) Relative transcriptional expression level of Etv2, Rig1, Cxcl10 and TNF-α in iHA-ETV2 MEFs at day 1 (D1) and at day 2 (D2) post-doxycycline administration to drive ETV2 overexpression (n = 3 biological replicates; two-way ANOVA; **p < 1.0 × 10⁻², ****p < 1.0 × 10⁻⁴).

Fig. 2

Transcriptional expression level of Rig1, and RLR pathway transcripts are upregulated in ETV2 overexpressed MEFs. Data are obtained from the single cell analyses of ETV2 mediated reprogramming of MEFs to the endothelial lineage. MEF indicates iHA-ETV2 MEFs without doxycycline administration and ETV2-OE indicates ETV2-overexpressed iHA-ETV2 MEFs seven days post-doxycycline administration. Data are presented as mean ± s.e.m. in log scale (Two-tailed unpaired t-test; **p < 1.0 × 10⁻², ****p < 1.0 × 10⁻⁴).

Fig. 3

ETV2 binds and transcriptionally activates Rig1 gene expression. (a) ChIP-seq and ATAC-seq coverage map projected on to the UCSC genome browser. Red box indicates ~ 800 bp region upstream of Rig1 gene. (b) Schematic of two adjacent ETV2 binding motifs in the Rig1 promoter site (top). Electrophoretic mobility shift assay showing ETV2 binding to identified motifs. Infrared dye-labeled oligonucleotides with two ETV2 binding motifs were annealed (Lane 1) and incubated with HA-tagged ETV2 protein synthesized in vitro (Lane2), with unlabeled oligonucleotides (Lane 3), with unlabeled oligonucleotides with mutation of the binding motif 1 (Lane 4), with unlabeled oligonucleotides with mutation of the binding motif 2 (Lane 5), with unlabeled oligonucleotides with mutations of both binding motifs (Lane 6), with HA antibody (Lane 7), and with heat inactivation of the HA antibody (Lane 8) (bottom). (c) Schematic of firefly luciferase construct driven by the Rig1 promoter which harbors the ETV2 binding site (left). Dual luciferase reporter assay showing dose-dependent increase of firefly luciferase activity upon addition of increased HA-ETV2 plasmids. Luciferase activity was normalized with the renilla luciferase activity (n = 3 biological replicates; one-way ANOVA; ***p < 1.0 × 10⁻³, ****p < 1.0 × 10⁻⁴). Data are presented as mean ± s.e.m. (right).

Fig. 4

Rig1 knockdown in iHA-ETV2 MEFs reduces reprogramming efficiency. (a) Schematic of Rig1 knockdown and ETV2 overexpression strategy that was used for iHA-ETV2 MEF reprogramming assay. (b) Relative expression level of shRNA lentivirus target—Rig1 and Nfκb1. One-way ANOVA was used for statistical analysis (****p ≤ 0.0001). (c) Flow cytometry profile of ETV2-overexpressed MEFs at day 1 without Rig1 knockdown (left), with knockdown (middle), and with Nfκb1 knockdown (right). PE-conjugated FLK1 antibody was used to determine the percentage of reprogrammed cells. (d) Quantification (bar graph) showing the percentage of reprogrammed MEFs without Rig1 (Control) and with knockdown (Rig1, Nfκb1). Two-tailed unpaired t-test with unequal variances was used for statistical analysis (n = 3 for each group, ****p ≤ 0.0001). (n = 3 biological replicates; two-tailed unpaired t-test; ***p < 1.0 × 10⁻³, ****p < 1.0 × 10⁻⁴). Data are presented as mean ± s.e.m.

Fig. 5

Proposed model whereby ETV2-RLR pathway promotes reprogramming of somatic cells to the endothelial lineage. Dashed arrow indicates potential direct regulation of the genes of the respective proteins associated with the pathway.