Fig 1: C/EBPß enhances CYP26A1 expression: SNP rs2068888 retards C/EBPß modulation of CYP26A1(A and B) HepG2 cells cultured in EMEM containing 10% charcoal-stripped FBS were transfected 24 h with scrambled siRNA (control) or siRNA, then treated with 1 nM RA in FBS-free medium (n = 3). WT differed significantly from siRNA in both A and B: p < 0.001, two-way ANOVA. (A) Time-course of CEBPB mRNA (controls, black circles; siRNA, red squares). (B) Time-course of CYP26A1 mRNA in the presence of siRNA controls and siRNA CEBPB (controls, black circles; siRNA, red squares) (n = 3).(C and D) HepG2 cells were treated 30 min with 0.1% DMSO or 1 nM RA ± 100 nM Dex, followed by crosslinking, lysis, and sonication for ChIP. DNA samples obtained by reversing crosslinking were used in qPCR with primers designed to amplify RARE1 and RARE2/3/4 in the CYP26A1 promoter or C/EBPß and RARa in SNP rs2068888. Antibodies were applied against C/EBPß or RARa.(E) Chromatogram showing DNA sequencing of WT HeLa vs. CRISPR-edited cells. HeLa cells have trisomy in chromosome 10: two with the major allele and another with the minor allele, deriving ~33% minor allele frequency. CRISPR editing introduced one copy of a single base replacement (G to A), resulting in a 50% minor allele frequency after a cycle of cell division.(F) Effects of RA and Dex on C/EBPß enrichment of rs2068888 in WT and edited HeLa cells.(G) Time-course of 1 nM RA ± Dex effects on CYP26A1 mRNA in WT and CRISPR-edited HeLa cells (n = 3, p = 0.0001 by two-way ANOVA for Dex and gene edited effects). (H) Regulation of RA homeostasis by fasting and refeeding. During fasting FoxO1 activity induces Rdh transcriptionally. RA induces CYP26A1. To prevent futile cycling during fasting, glucagon and cortisol retard CYP26A1 transcription. During refeeding, insulin causes FoxO1 to leave the nucleus and undergo proteolysis. This decreases Rdh and RA. Inhibition of CYP26A1 expression ceases allowing RA to auto-induce its catabolism. These actions result in higher RA during fasting vs refeeding, permitting RA to antagonize insulin action during fasting, while permitting optimum insulin action during feeding.