TAZ is indispensable for c-MYC-induced hepatocarcinogenesis

Abstract

Background & aims: Mounting evidence implicates the Hippo downstream effectors Yes-associated protein (YAP) and transcriptional co-activator with PDZ-binding motif (TAZ) in hepatocellular carcinoma (HCC). We investigated the functional contribution of YAP and/or TAZ to c-MYC-induced liver tumor development.

Methods: The requirement for YAP and/or TAZ in c-Myc-driven hepatocarcinogenesis was analyzed using conditional Yap, Taz, and Yap;Taz knockout (KO) mice. An hepatocyte-specific inducible TTR-CreER^T2 KO system was applied to evaluate the role of YAP and TAZ during tumor progression. Expression patterns of YAP, TAZ, c-MYC, and BCL2L12 were analyzed in human HCC samples.

Results: We found that the Hippo cascade is inactivated in c-Myc-induced mouse HCC. Intriguingly, TAZ mRNA levels and activation status correlated with c-MYC activity in human and mouse HCC, but YAP mRNA levels did not. We demonstrated that TAZ is a direct transcriptional target of c-MYC. In c-Myc induced murine HCCs, ablation of Taz, but not Yap, completely prevented tumor development. Mechanistically, TAZ was required to avoid c-Myc-induced hepatocyte apoptosis during tumor initiation. The anti-apoptotic BCL2L12 gene was identified as a novel target regulated specifically by YAP/TAZ, whose silencing strongly suppressed c-Myc-driven murine hepatocarcinogenesis. In c-Myc murine HCC lesions, conditional knockout of Taz, but not Yap, led to tumor regression, supporting the requirement of TAZ for c-Myc-driven HCC progression.

Conclusions: TAZ is a pivotal player at the crossroad between the c-MYC and Hippo pathways in HCC. Targeting TAZ might be beneficial for the treatment of patients with HCC and c-MYC activation.

Lay summary: The identification of novel treatment targets and approaches for patients with hepatocellular carcinoma is crucial to improve survival outcomes. We identified TAZ as a transcriptional target of c-MYC which plays a critical role in c-MYC-dependent hepatocarcinogenesis. TAZ could potentially be targeted for the treatment of patients with c-MYC-driven hepatocellular carcinoma.

Keywords: BCL2L12; Hepatocellular carcinoma; TAZ; YAP; c-MYC.

Fig. 1.. Activation of the YAP/TAZ cascade in c-Myc driven HCC in mice.

(A) Western blotting results of LAST1, LATS2, and YAP/TAZ expression in the NL and c-Myc HCC tissues (T). (B) mRNA expression of Ctgf and Cyr61. (C) Western blotting results of nuclear YAP (n-YAP) and nuclear TAZ (n-TAZ) levels. Histone H3 and β-Tubulin were used as nuclear and cytoplasmic (Cyto) protein loading controls, respectively. (D) Quantification of nuclear YAP (n-YAP) and nuclear Taz (n-TAZ) levels. (E) Yap and Taz (Wwtr1) mRNA relative expression in NL and T tissues. N = 4 biological replicates for each group. (F) Representative immunohistochemistry of a c-Myc mouse tumor lesion (T) exhibiting robust nuclear immunoreactivity for c-MYC and TAZ, whereas only a few tumor cells display nuclear staining for YAP. Scale bar: 100 μm. (B, D, E) Mean ± SD; Unpaired t test, Welch’s t test or Mann-Whitney test. HCC, hepatocellular carcinoma; NL, normal liver; T, tumor.

Fig. 2.. Evaluation of the relation between TAZ and YAP mRNA with c-MYC activity.

(A) The activity of c-MYC in HCC (n = 64) and corresponding non-tumorous SL (n = 64). (B) The activity of c-MYC in HCCB (n = 27) and HCCP (n = 37). (C, D) Correlation between WWTR1 (C) or YAP (D) mRNA expression and c-Myc activity. p values and correlation r values were calculated by Pearson correlation analysis. (E) Activation status of c-MYC, YAP, and TAZ in human HCCs as determined by IHC staining. (F) Comparison of c-MYC positive and negative immunoreactivity frequency between YAP-positive (YAP [+]) and TAZ-positive (TAZ [+]) human HCC samples. (A, B) Mean ± SD; Upaired t test; (C, D) Pearson correlation coefficient; (F) Fisher exact test. HCC, hepatocellular carcinoma; HCCB, HCC with better prognosis; HCCP, HCC with poorer prognosis; IHC, immunohistochemistry; n.s., not significant; SL, surrounding liver.

Fig. 3.. c-MYC directly binds to the TAZ promoter to induce its transcriptional activation.

(A, B) qPCR analysis of mRNA levels of YAP, WWTR1, and c-MYC targets in the SNU449 (A) and Focus (B) cell lines transfected with 4-OHT inducible EGFP and MYC-ER plasmid. (C, D) Direct binding of c-MYC to WWTR1 promoter in SNU449 and Focus cells. The anti-V5-tag primary antibody was applied to immunoprecipitate exogenous transfected V5-tagged c-MYC (C), and the anti-c-MYC antibody was used to immunoprecipitate endogenous c-MYC (D). TERT promoter was applied as a positive control. (E) Schematic overview of the genetic structure of WWRT1 in the human genome, pGL3-WWTR1-Promoter, and pGL3-Motif-del constructs. (F) Dual-luciferase assay with wild-type (WWTR1 Promoter) or depleted c-MYC binding site (Motif-del) in the WWTR1 promoter region, Vector was used as a negative control. (A, B) Mean ± SD; Mann-Whitney test. (F) Mean ± SD; One-way ANOVA test. 4-OHT, 4-hydroxytamoxifen; BS, binding site; ORF, open reading frame; qPCR, quantitative real time PCR; UTR, untranslated region. (This figure appears in color on the web.)

Fig. 4.. TAZ is indispensable for c-Myc-dependent HCC development in mice.

(A) Study design. Yap^flox/flox, Taz^flox/flox, Yap/Taz^flox/flox conditional knockout mice were injected with c-Myc/pCMV or c-Myc/Cre plasmids. (B) Survival curves of Yap^flox/flox mice injected with c-Myc. (C) Western blotting showing the expression level of c-MYC and depletion of YAP in Yap-depleted livers. WT refers to normal uninjected liver tissues from Yap^flox/flox mice; whereas all the rest were from HCC lesions. GAPDH and β-ACTIN were used as a loading control. (D) mRNA expression of Ctgf and Cyr61. n = 6 samples for each group. (E) Analysis of Ki67- and C-C3-positive cells in control (pCMV) and Yap-depleted livers. (F) Survival curves of Taz^flox/flox mice injected with c-Myc. (G) Western blotting showing the expression level of c-MYC and the partial depletion of TAZ in Taz-depleted livers. The liver tissues from the Cre group were harvested from 2 mice with single small tumor nodules. GAPDH and β-ACTIN were used as a loading control. (H) Survival curves of Yap/Taz^flox/flox mice injected with c-Myc. (B, F, H) Log-rank (Mantel-Cox) test; (D, E) Mean ± SD; Mann-Whitney test. C-C3, cleaved-caspase 3; HCC, hepatocellular carcinoma; HTVi, hydrodynamic tail vein injection; WT, wild-type. (This figure appears in color on the web.)

Fig. 5.. TAZ supports c-Myc oncogenic function in the absence of survival factors.

(A) Study design. C57BL/6J mice were injected with c-Myc/pT3 (n = 7), c-Myc/YAPS127A (n = 5), c-Myc/TAZS89A (n = 9), c-Myc/TEAD2-VP16 (n = 5). Mice were sacrificed when moribund. (B) Survival curve showing that either YAP or TAZ overexpression or TEAD transcriptional activation cooperates with c-Myc to induce HCC in mice. p values were calculated by Log-rank (Mantel-Cox) test. (C) Macroscopy, H&E, and immunostaining (c-MYC, Ki67, C-C3) of liver tissues 3 weeks post-injection. Scale bar: 100x, 200 μm; 200x, 100 μm. (D, E) mRNA expression of Ctgf and Cyr61. (F, G) Analysis of Ki67- and C-C3-positive cells in control and liver tumors. P <0.005 (a) vs. YAPS127A; (b) vs. TAZS89A; (c) vs. TEAD2-VP16. (D, E, F, G) Mean ± SD; One-way ANOVA test. C-C3, cleaved-caspase 3; HCC, hepatocellular carcinoma; HTVi, hydrodynamic tail vein injection.

Fig. 6.. BCL2L12 is a target regulated by YAP/TAZ during oncogene-driven liver tumor development.

(A) Heatmap of anti-apoptosis gene signature in human HCC cell lines after YAP or/and TAZ depletion, showing BCL2L12 as a potential target. (B) mRNA expression of Bcl2l12 was significantly downregulated when deleting Yap/Taz in c-Myc/Mcl-1 mice. n = 9 samples in each group. (C) Overexpression of TAZS89A in c-Myc induced tumors upregulated Bcl2l12 expression in C57BL/6J mouse livers. n = 7 samples in each group. (D, E) Direct binding of TEAD to BCL2L12 promoter in SNU449 (D) and Focus (E) cells. Predicted ChIP-PCR product bands were at 100 bp-200 bp. (F) Study design. Yap/Taz^flox/flox mice were injected with c-Myc/pCMV-Cre/pT3 (control, n = 8) or c-Myc/Cre/HA-Bcl2l12 (n = 6) plasmids. (G) Survival curves of Yap/Taz^flox/flox mice injected with c-Myc/HA-Bcl2l12. (H) Histology analysis of liver tissue from both groups. Scale bar: 200 μm for 100x, 100 μm for 200x. (I) Western blotting results of c-MYC, HA-tagged BCL2L12, YAP, and TAZ expression. GAPDH was used as a loading control. (B, C) Mean ± SD; Mann-Whitney test; (G) Log-rank (Mantel-Cox) test. ChIP, chromatin immunoprecipitation; HCC, hepatocellular carcinoma; HTVi, hydrodynamic tail vein injection; WT, wild-type.

Fig. 7.. BCL2L12 expression in human HCC samples.

(A) qPCR analysis of BCL2L12 mRNA levels in HCC (n = 64) and corresponding non-tumorous SL (n = 64). (B) qPCR analysis of BCL2L12 mRNA levels in HCCB (n = 27) and HCCP (n = 37). (C, D) Evaluation of the relation between TAZ and YAP mRNA with BCL2L12 mRNA in HCC samples. p values and correlation r values were calculated by Pearson correlation analysis. (E) Representative immunohistochemical patterns of BCL2L12 in human non-tumorous and neoplastic liver specimens. Scale bar: 100 μm. (F) Scheme summarizing the distribution of positive and negative HCC samples for c-MYC, TAZ, YAP, and BCL2L12 staining. (A, B) Mean ± SD; U paired t test; (C, D) Pearson correlation coefficient. HCC, hepatocellular carcinoma; HCCB, HCC with better prognosis; HCCP, HCC with poorer prognosis; qPCR, quantitative real time PCR; SL, surrounding liver.

Fig. 8.. Targeting TAZ as an effective therapeutic strategy for c-Myc-induced hepatocarcinogenesis.

(A) Study design. Yap^flox/flox and Taz^flox/flox mice were injected with c-Myc/TTRpro-CreER^T2 plasmids. Mice (n = 6) were sacrificed at 5 w.p.i as a pretreatment group, while the remaining mice were injected with TAM or vehicle. All mice were sacrificed when moribund or at the end of observation. (B, C) Survival curves of Yap^flox/flox (B) and Taz^flox/flox (C) mice in vehicle- or TAM-treated groups. (D) Western blotting showing the expression of c-MYC and TAZ in Taz^flox/flox mice. GAPDH was used as a loading control. (E) Representative images of Taz^flox/flox murine liver tumors, H&E staining, and IHC staining of Ki67, c-MYC, and C-C3 in the 4 groups. (Top panel) Pretreatment group (Pre, 5 w.p.i); (second panel) TAM-treated groups harvested at the early stage (TAM early; 7 w.p.i); (third panel) TAM-treated groups harvested at a late stage (TAM late; 16 w.p.i); (Bottom panel) Vehicle-treated group (7 w.p.i). (B, C) Log-rank (Mantel-Cox) test. Scale bar: 200 μm. C-C3, cleaved-caspase 3; HTVi, hydrodynamic tail vein injection; IHC, immunohistochemistry; TAM, tamoxifen; w.p.i., weeks post-injection.