Fig 2: Ablation of DPP4 suppresses CRC progression. A,B) Images and quantifications of bioluminescence of NSG mice injected with A) luciferase‐labeled HT29 and B) CRC57 cells carrying scrambled (control) or DPP4 knockout (DPP4 KO1 or KO2) gRNA constructs. C) Survival analysis of NSG mice injected with luciferase‐labeled HT29 and CRC57 cells carrying scrambled (control) or DPP4 knockout (DPP4 KO1 or KO2) gRNA constructs. D) Western blots showing downregulation of DPP4 by selected and scrambled gRNA tested in the CRISPR/dCas9HDAC system. E) ChIP‐qPCR showing the effect of CRISPR/dCas9HDAC system on DPP4 promoter‐associated histone modification. F,G) Images and quantifications of F) bioluminescence and G) survival analysis of NSG mice carrying scrambled and selected gRNA in the CRISPR/dCas9HDAC system. H,I) Images and quantifications of H) bioluminescence and I) survival analysis of NSG mice receiving spleen‐injection of HT29 cells with or without Sitagliptin oral gavage. Data represent the mean ± s.d. p‐value was calculated based on student's t‐test in (H), ANOVA and Tukey's HSD post doc test in (A), (B), (E), and (F), and log‐rank test in (C), (G), and (I). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig 3: DPP4 is upregulated in CRC liver metastases. A) RNA‐seq volcano‐plot showing DE genes detected in HT29 liver metastases versus primary CRC tumor. B) Integrated analysis of ATAC‐seq, RNA‐seq, and GEO dataset (GSE41568) of upregulated genes in liver metastases. C) Analysis of differential DPP4 expression in GEO dataset (GSE41568) between primary CRC and liver metastases. D) Analysis of differential DPP4 expression in the clinical RNA‐seq dataset SRR2089755 from five matched patient liver metastases versus primary CRC tumor. E) ATAC‐seq signal track showing DPP4‐associated open chromatin in liver metastases versus primary CRC tumor. F) Mint‐ChIP signal track showing H3K27ac histone activation markers in the DPP4 promoter and three enhancers in liver metastases versus primary CRC. G) RT‐qPCR showing DPP4 expression levels in patient liver metastases versus primary CRC tumors. H) Western blots showing DPP4 expression levels in paired primary CRC and liver metastases collected from four patients. I) Representative IHC staining and evaluation of DPP4 expression measured on a tissue microarray that contains 16 paired primary CRC and liver metastases. (Scale bar, 50 um). J) Western blots showing DPP4 expression levels in two paired patient‐derived liver metastases and primary CRC organoids. K) Western blots showing DPP4 expression levels in primary CRC and liver metastases derived from cecum‐injected‐HT29 and ‐CRC57 cells. L,M) ChIP‐qPCR analysis showing changes in H3K27ac, H3K4me1, H3K4me3, and H3K9ac levels on DPP4 promoter in liver metastases versus primary CRC tumor. N) Correlation of DPP4 expression and clinicopathological features of 98 CRC patients. O) Kapla–Meier analysis of relapses of CRC patients with high (N = 30) and low (N = 68) DPP4 expression levels in the tumors. LM, liver metastases. CP, CRC primary tumor. E1, E2, and E3, enhancer 1, enhancer 2, and enhancer 3. Data represent the mean ± s.e.m. in (C) and (D), and mean ± s.d. in (G), (L), and (M). p‐values were calculated based on Student's t‐test in (C), (D), (G), (L), (M), and (N), and log‐rank test in (O). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig 4: PU.1 is upregulated in CRC liver metastases and promotes CRC progression. A) Analysis of differential PU.1 expression in GEO dataset (GSE41568) between primary CRC and liver metastases. B) Analysis of differential PU.1 expression in the RNA‐seq dataset SRR2089755 from five matched patient primary CRC and liver metastases. C) IHC staining of differential PU.1 expression between primary CRC and liver metastases. (Scale bar, 100 um). D) Western blots showing DPP4 expression levels in paired primary CRC and liver metastases collected from four patients. E) Western blots showing DPP4 expression levels in paired primary CRC and liver metastases‐derived organoids collected from two patients (CA197 and CA1006). F,G) Images and quantification of F) bioluminescence and G) survival analysis of NSG mice injected with luciferase‐labeled HT29 carrying scrambled (control) or PU.1 knockdown (PU.1 KD1 or KD2) shRNA constructs. H,J) Images and quantification of H) bioluminescence and J) survival analysis of NSG mice injected with luciferase‐labeled CRC57 carrying scrambled (control) or PU.1 knockdown (PU.1 KD1 or KD2) shRNA constructs. LM, liver metastases. CP, CRC primary tumor. Data represent the mean ± s.e.m. in (A) and (B), and the mean ± s.d. in (F) and (H). p‐values were calculated based on Student's t‐test (A) and (B), log‐rank test in (G) and (J), and ANOVA and Tukey's HSD post hoc test in (F) and (H). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig 5: HGF upregulates DPP4 through activaton of PU.1. A) Dotplot of pathway analysis on enriched accessible chromatin regions in HT29 liver metatases. B) Western blots showing PU.1 expression level in HT29 or CRC57 cells treated with or without HGF. C) Western blots showing PU.1 phosphorylation in HT29 or CRC57 cells treated with or without HGF or the c‐Met inhibitor PHA665752. D,E) Western blots showing that PU.1 complexed with D) C/EBPα and ‐β and E) SMARCB1. F) ChIP‐qPCR showing relative H3K27ac, H3K4me1, H3K4me3, and H3K9ac enrichments in HT29 or CRC57 cultured in RPMI1640 media supplemented with or without 100 ng mL−1 HGF. G) Western blots showing the expressions of DPP4 in HT29 or CRC57 carrying scrambled (control) or PU.1 knockdown shRNA cultured in RPMI media supplemented with or without 100 ng mL−1 HGF. H,I) Images and quantification of H) bioluminescence and I) survival analysis of NSG mice spleen‐injected with luciferase‐labeled HT29 cells receiving i.p. administered PBS (control) or the HGF inhibitor Norleual. Data represent the mean ± s.d. in (F) and (H). p‐values were calculated based on Student's t‐test (F) and log‐rank test in (I). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.