PMC

(A) Overlapping of ATF4 and CEBPB candidate genes with genes identified in the microarray assays after YW3-56 treatment.
(B) Screen shots of ATF4 binding on promoters of four genes highly responsive to YW3-56.

(A and B) Western blot and qRT-PCR analyses of SESN2 and DDIT4 induction after YW3-56 treatment in MDA-MB-231cells with control, ATF4 and CEBPB siRNA treatment. **p<0.002, n = 2 × 3.
(C and D) Effects of forced ATF4 expression on SESN2 and DDIT4 protein and mRNA levels. *p<0.01, **p<0.002, n = 2 × 3.
(E and F) Effects of forced CEBPB expression on SESN2 and DDIT4 protein and mRNA levels. n = 2 × 3.

(A) Coordinate plot and density plot of ATF4 and CEBPB binding locations within the −25 kb to +25 kb region of target gene promoters.
(B) Function enrichment analysis of ATF4 and CEBPB target genes.
(C and D) ATF4 and CEBPB binding motif 1 identified by MEME analyses.
(E and F) The frequency of ATGC in ATF4 and CEBPB motif 1 was plotted.
(G and H) FIMO plot analyses of 914 ATF4 motif 1 sites and 1277 CEBPB motif 1 sites. 50 bp sequences centered on motif 1 were aligned.

(A) Box plot showing the relative tumor weight (control tumors normalized to 100%, n=6) obtained from three xenograft experiments with different dosage of YW3-56. **p<0.002.
(B) Images of tumors in control and daily 12.5 mg/kg YW3-56 treatment groups.
(C) Growth curves of tumors after treatment without or with 40 mg/kg YW3-56 once every two days.
(D) Images of tumors without or with 40 mg/kg YW3-56 treatment once every two days.
(E) Representative pathology H&E staining of the control and YW3-56 treated tumors. Yellow arrows denote blood vessels.

(A) The IPA analyses of microarray data found canonical pathways significantly affected by YW3-56 treatment.
(B) GSEA assays found gene sets important for ER protein homeostasis are enriched after YW3-56 treatment. The normalized enrichment scores (NES) and false discovery rate (FDR) are shown. Color index represents gene expression changes, and the leading edge genes are colored red.
(C) The expression of ATF4 target genes was significantly (p=1.16×10⁻¹¹) altered after YW3-56 treatment.

(A) Illustration of the ER/UPR stress response pathway. Activated PERK phosphorylates eIF2α to selectively translate ATF4 and activate its downstream UPR genes.
(B) Western blot assays of PERK and eIF2α phosphorylation, and ATF4 accumulation after YW3-56 treatment in MDA-MB-231cells.
(C) Effects of PERK depletion by siRNAs on eIF2α phosphorylation and the ATF4 protein levels after YW3-56 treatment.
(D) Effects of PERK depletion on SESN2 mRNA levels. *p<0.01, n = 2 × 3.
(E) Effects of YW3-56 on the phosphorylation of p70S6K and 4EBP1, two substrates of the mTORC1 kinase.
(F) Working model for YW3-56 in cancer growth inhibition via the ER stress to increase ATF4 target gene (e.g., SESN2 and DDIT4) expression, which in turn inhibit the mTORC1 kinase.

(A) TEM images of control MDA-MB-231 cells showing normal cellular structures with mitochondria denoted by green arrows. In contrast, YW3-56 treated MDA-MB-231 cells showed large numbers of autophagic vesicles denoted by yellow arrows. Note the disappearance of mitochondria.
(B) Immunostaining of LC3B (green) in control and YW3-56 treated cells, noting the increased LC3B staining and LC3B speckles in YW3-56 treated cells. DNA (blue) was stained by Hoechst.
(C) YW3-56 inhibits autophagosome maturation to autolysosomes as measured by live cell imaging using the mCherry-GFP-LC3B reporter protein based on the quenching of the GFP signals by the acidic environment within autolysosomes.
(D) Percentages of autophagosomes and autolysosomes identified in control and YW3-56 treated cells based on the GFP and RFP fluorescent signals. After YW3-56 treatment, the number of speckles with both RFP and GFP signals (representing autolysosomes) was increased from ~56% (n=19) to ~98% (n=23), ** p < 0.002.
(E) Effect of YW3-56 on autophagy regulatory protein LC3B and p62/SQSTM1.