Figure 4.

The secretion of the metabolic hormone GDF15 is induced by the phosphorylated eIF2α-ATF4-DDIT3 axis as an adipokine

(A) Volcano plot comparing mRNA expression in the BAT of vehicle- or 0.05 mg/kg AP-injected Tg mice (n = 2). Mice were sacrificed at 8 weeks 12 h after the injection. Blue dots show the known targets of eIF2α phosphorylation Ddit3 and Trib3. The red dot shows Gdf15.

(B) RT-qPCR analysis of Gdf15 mRNA expression in the BAT, ingWAT, and epiWAT of vehicle- or AP-treated Tg mice 24 h after the injection (Tg vehicle: n = 5, Tg AP: n = 7).

(C) Plasma GDF15 concentrations in vehicle- or AP- treated Tg mice 24 h after the injection (Tg vehicle: n = 5, Tg AP: n = 7).

(D) RT-qPCR analysis of Gdf15 mRNA expression 12 h after treatment with AP, tunicamycin (Tm), histidinol (His), and arsenite (Ars) in 3T3L1 adipocytes stably expressing Fv2E-PERK (n = 3).

(E) Representative immunoblots of ATF4, DDIT3, and GAPDH in 3T3L1 adipocytes stably expressing Cas9 and sgRNA against Atf4 or Ddit3 at 4 or 12 h after treatment with Tm.

(F) RT-qPCR analysis of Gdf15 mRNA expression in 3T3L1 adipocytes stably expressing Cas9 and sgRNAs against Atf4 or Ddit3 at 12 h after treatment with Tm (n = 3).

(G) ChIP analysis of the ATF4 and DDIT3 occupancy of the Gdf15 locus in MEFs 8 h after the thapsigargin (Tha) treatment (n = 3).

(H) RT-qPCR analysis of Gdf15 and Ddit3 mRNA expression in MEFs lacking the ATF4/DDIT3-binding site at 10 h after treatment with Tha (n = 3).

All data are presented as means ± SD. Unpaired two-tailed Student's t tests were used to analyze the data presented in (B and G). One-way ANOVA followed by Holm-Sidak multiple comparisons tests were used to analyze the data presented in (C, D, F, and H). ∗P < 0.05 and ∗∗P < 0.01.