|Home | About | Journals | Submit | Contact Us | Français|
TPA stimulates carcinogenesis in various types of cancers. However, we found that TPA inhibits transformative phenotypes in liver cancer cells via the translocation of YAP from the nucleus, where it functions as a transcriptional co-factor, to the cytoplasm. Such effects led to a separation of YAP from its dependent transcription factors. The inhibitory effects of TPA on YAP were AMOT dependent. Without AMOT, TPA was unable to alter YAP activity. Importantly, the depletion of YAP and AMOT blocked the TPA-reduced transformative phenotypes. In sum, TPA has been established as an anti-tumorigenic drug in liver cancer cells via YAP and AMOT.
TPA, also called phorbol 12-myristate 13-acetate (PMA), is a small molecule drug that activates the signal transduction enzyme protein kinase C (PKC) by directly binding to its C1 domains1,2. Commonly, TPA is employed as a tumor-promoting agent for skin carcinogenesis in rodents and is associated with increased cell proliferation in malignant cells from several types of tumors, such as melanoma and breast and oral cancer3,4,5. However, the function of TPA is controversial because a decrease in cell proliferation capacity has also been observed in TPA-treated lymphoma cells compared to controls6,7. Hence, different effects may occur after exposure to TPA in different types of cells. Although some studies have investigated the effects of TPA on cell proliferation in liver cancer8,9, the exact roles of TPA in maintaining transformative phenotypes in liver cancer cells remain largely unknown.
Liver cancer is the fifth most common cancer type globally and the third most frequent cause of cancer–related mortality worldwide10. Liver cancer carries a poor prognosis because the treatment options are extremely limited. Curative resection or transplantation is currently the best curative option for treatment, yet recurrence and metastasis are quite common in patients11. Recently, transcatheter arterial chemoembolization (TACE) has been used for patients who cannot receive these local eradication methods due to reasons such as poor residual liver function, complicated tumor location, or complications12. However, TACE may have negative effects on liver function; therefore, it is urgent to improve the therapeutic strategies13, and the development of potential drugs might be a promising option.
Recently, emerging evidence has shown the critical roles of the tumor suppressor Hippo signaling pathway in the pathogenesis of various cancers, including liver cancer14,15. Yes-associated protein (YAP), the major downstream effector of this pathway, has been identified as an oncoprotein that is also critical for the initiation and progression of liver cancer16. YAP is phosphorylated and inhibited by Hippo signaling, thereby resulting in its translocation from the nucleus into the cytoplasm, where its activity is lost17. In the nucleus, the activity of YAP largely depends on its interaction with its dependent transcription factors, such as the TEAD family, Runx2, CREB, and p73 proteins17,18,19,20.
Angiomotin (AMOT) contains conserved glutamine-rich domains and PPxY motifs in its N-terminus, through which it binds to a number of WW domain-containing proteins21. Interestingly, YAP contains WW domains22. Some studies have suggested that AMOT can interact with YAP to inhibit the growth of liver and breast cancer cells23, indicating that AMOT may play an essential suppressive role to tumorigenesis. AMOT also promotes YAP phosphorylation through activating the LATS kinase, subsequently transferring YAP from the nucleus to the cytoplasm24. Moreover, AMOT may compete with PPxY motif-containing transcription factors for YAP binding, for example, inhibiting YAP-TEAD binding to decrease the transcription of TEAD-target genes25.
Here, we intended to investigate the function of TPA in liver cancer cells. We have also investigated the underlying mechanism of how TPA exerts its roles in liver cancer cells. This study might provide valuable information for improving liver cancer treatments in the future.
The methods were carried out in accordance with approved guidelines, and the experimental protocols were approved by the Department of Clinical Laboratory, Shanghai Tenth People’s Hospital of Tongji University (Shanghai, 200072, China).
The liver cancer cell lines Bel-7402 and Bel-7404 were cultured in DMEM. Cells were treated with TPA (final concentration 16–48μM, Beyotime, Haimen, China) for 24h before harvest for further analysis. The TEAD-Gal4/pUAS-LUC, HULC-promoter, YAP-FLAG, TEAD4-Myc, CREB-HA, Runx2-HA, AMOT-HA, YAP-sh1 and –sh2, and AMOT-sh1 and –sh2 were obtained from our previous studies26,27,28. ShRNAs specifically targeting TAZ (TAZ-sh1 and –sh2) were purchased from Biolink LTD (Shanghai, China).
For IF, the primary antibodies used were anti-YAP (Abcam, Hong Kong, China, #ab52771), anti-TEAD4 (Epitomics, Burlingame, CA, USA, #s1666), anti-CREB (Epitomics, #1496), anti-p73 (Epitomics, #1636), anti-Runx2 (Epitomics, #5356), anti-HA (Cell Signaling Technology (CST), Cambridge, MA, USA, #3724) or anti-AMOT (Abcam, #ab85143). The slides were incubated with Alexa Fluor®-488/555 fluorescently conjugated secondary antibodies (CST, #4408, #4412, #4409 or #4413) for 1h in the dark before being mounted with ProLong® Gold antifade reagent with DAPI (Molecular Probes, Eugene, OR, USA). The sides were then observed using an LSM 800 Confocal Microscope (Carl Zeiss AG, Oberkochen, Germany).
For WB, the primary antibodies used were anti-p-YAP (Ser127) (CST, #13619), anti-YAP (Abcam, #ab52771), anti-GAPDH (CST, #5176), anti-β-tubulin (CST, #2128), anti-Histone-H3 (Santa Cruz Biotechnology, Santa Cruz, CA, USA, #sc-10809), anti-CREB (Epitomics, #1496), anti-Runx2 (Epitomics, #5356), anti-TEAD4 (Epitomics, #s1666), anti-p73 (Epitomics, #1636), anti-TAZ (CST, #2149), or anti-AMOT (Abcam, #ab85143). For cytosolic and nuclear fractionation, a nuclear extraction kit from Active Motif (Carlsbad, CA, USA) was used. The membranes were incubated with secondary antibodies conjugated with horseradish peroxidase (CST, #7074 or #7076) and visualized using Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific, Waltham, MA, USA). The software ImageJ version 1.47, (Bethesda, MD, USA) was used for densitometry of the Western blots.
All IF and WB were performed by conventional methods, and the protocols are available elsewhere.
Cell proliferation, Caspase3/7 activity, soft agar colony formation and quantitative RT-PCR (qPCR) assays were performed as described previously29. The primers used for qPCR are listed below: CTGF-qPCR-F: 5′ CCTGTGCAGCATGGACGTTCGT 3′, CTGF-qPCR-R: 5′ AACGTGTCTTCCAGTCGGTAAG 3′; ANKRD1-qPCR-F: 5′ GAAACAACGAGAGGCAGAGCTC 3′, ANKRD1-qPCR-R: 5′ AGAAACGTAGGCACATCCACAG 3′; MCAM-qPCR-F: 5′ GCGTCTACAAAGCTCCGGAGGA 3′, MCAM-qPCR-R: 5′ GAATGTGGACCCGGTTCTTCTCCTC 3′; HULC-qPCR-F: 5′ ACCTCCAGAACTGTGATCCAAAATG 3′, HULC-qPCR-R: 5′ CAAATTTGCCACAGGTTGAACAC 3′ and GAPDH-qPCR-F: 5′ ATCATCCCTGCCTCTACTGG 3′, GAPDH-qPCR-R: 5′ GTCAGGTCCACCACTGACAC 3′.
Co-IP was performed as described previously30. The reagents used were protein A/G-Sepharose (Novex, Oslo, Norway) and Western/IP lysis buffer (Beyotime, Haimen, China). The antibodies used for IP were anti-YAP (Santa Cruz Biotechnology, #sc101199) or anti-FLAG (CST, #8146).
Bel-7402 cells (5×10^6 cells) were subcutaneously injected into 8-week-old athymic nude mice (Bikai, Shanghai, China). After xenografts were visible (10th day after injection), the mice were treated with TPA (100μg/kg) for another 20 days before the tumor sizes were measured. The tumor volume was calculated as 0.5×L×W^2, where L is length and W is width. All mouse experiments were performed according to the institutional guidelines of the Shanghai Tenth People’s Hospital.
Tests to examine the differences between groups included Student’s t test and one-way ANOVA; p<0.05 was regarded as statistically significant.
We found that the cell-proliferation and colony-formation capacities of Bel-7402 and Bel-7404 cells could be dose-dependently inhibited by increasing concentrations of TPA (Fig. 1A,B). By contrast, Caspase 3/7 activities could be dose-dependently induced (Fig. 1C). These results suggested that TPA inhibits the transformative phenotypes of liver cancer cells.
Because YAP is a transcription co-factor17 and its activity relies on its dependent transcription factors (including TEAD and CREB)31,32, we tested whether TPA treatment influences the transcriptional activities of TEAD and CREB. We found that YAP-dependent TEAD and CREB activities could be dose-dependently reduced by TPA, as measured using a TEAD-Gal4/pUAS-LUC and an HULC-promoter luciferase reporter, which contains a CREB-responses element, before and after treating cells with increasing concentrations of TPA (Fig. 1D,E). TEAD target genes, CTGF and ANKRD133,34, and CREB target genes, MCAM and HULC35,36, were also found to be downregulated by TPA treatment (Fig. 1F). Moreover, the inhibitory efficacies of TPA on the mRNA levels of these genes were abolished when YAP was depleted (Fig. 1F), further demonstrating that TPA can inhibit the activities of YAP-dependent transcription factors.
Then, we investigated whether TPA directly affected the phosphorylation of YAP (p-YAP), a hallmark of YAP inactivation37. We found that p-YAP was dose-dependently elevated, while total YAP levels were unaffected by increasing concentrations of TPA (Fig. 1G). Further, we found that TPA was able to shuttle YAP from the nucleus, where YAP exerts its major function on tumorigenesis38, to the cytoplasm (Fig. 1H); these effects were also dose-dependent, suggesting YAP can be directly inhibited by TPA.
Because YAP activity relies on its transcription factors16, we performed IF and found that TPA treatments led to YAP translocation from the nucleus to the cytoplasm (Fig. 2A). However, the nuclear localization of YAP-dependent transcription factors, including TEAD, CREB, p73 and Runx2, was not altered in Bel-7402 cells (Fig. 2A). Similarly, data from fractionation experiments indicated that TPA treatments led to gradually increasing cytoplasmic accumulation of YAP, whereas decreased nuclear expression of YAP was caused by increasing concentrations of TPA in both Bel-7402 and Bel-7404 cells (Fig. 2B). Further, co-IP experiments demonstrated that TPA treatments dissociated endogenous YAP from endogenous CREB, Runx2, TEAD and p73 in Bel-7402 and Bel-7404 cells (Fig. 2C). In Bel-7402 cells co-transfected with exogenous YAP-FLAG and TEAD4-Myc, CREB-HA, or Runx2-HA, we also found that TPA treatments inhibited the interactions between YAP-FLAG and its dependent exogenous transcription factors (Fig. 2D,F). Taken together, TPA separates YAP from its dependent transcription factors.
We have previously reported that AMOT overexpression causes the inhibition of YAP28. Here, cytoplasmic accumulation of YAP could be found in Bel-7402 and Bel-7404 cells successfully transfected with exogenous AMOT-HA, whereas nuclear accumulation of YAP was observed in the cells without successful transfection (Fig. 3A), demonstrating that AMOT overexpression can drive YAP from the nucleus to the cytoplasm. We also found that AMOT overexpression dose-dependently suppressed the interactions between exogenous YAP-FLAG and TEAD4-Myc, CREB-HA or Runx2-HA (Fig. 3B–D). Further, the amounts of CREB, p73, Runx2 and TEAD in the immunoprecipitates that were pulled down by anti-YAP antibodies were greatly reduced when AMOT was overexpressed (Fig. 3E). Moreover, AMOT overexpression reduced the mRNA levels of CTGF, ANKRD1, MCAM and HULC; however, the depletion of YAP blocked such effects (Fig. 3F), suggesting that AMOT controls TEAD and CREB target gene transcription, possibly via YAP. Furthermore, the transcriptional activities of TEAD and CREB measured by luciferase-based experiments were dose-dependently reduced in Bel-7402 and Bel-7404 cells transfected with increasing concentrations of exogenous AMOT-HA compared to those from the control cells (Fig. 3G,H). These results suggested that, similar to the effects of TPA, AMOT overexpression is capable of inhibiting YAP-dependent transcriptional activity.
Next, we tested whether the interaction between AMOT and YAP can be affected by TPA. We found that TPA dose-dependently enhanced the AMOT-YAP interaction in both Bel-7402 and Bel-7404 cells (Fig. 4A). IF experiments revealed the gradual and simultaneous translocation of both YAP and AMOT from the nucleus to the cytoplasm (Fig. 4B). By cytosolic and nuclear fractionation experiments, the gradual and simultaneous translocation of YAP and AMOT from nucleus to the cytoplasm in both Bel-7402 and Bel-7404 cells was also observed (Fig. 4C). These results suggested that the changes in the subcellular localization of YAP and AMOT are similar. Further, luciferase-based experiments demonstrated that the reduction in the transcriptional activities of TEAD and CREB occurred almost simultaneously with the subcellular alterations of AMOT and YAP in both Bel-7402 and Bel-7404 cells (Fig. 4D).
Next, we sought to determine whether AMOT is essential for the TPA-induced inhibition of YAP. We found that YAP translocation from the nucleus to the cytoplasm was blocked when AMOT was knocked down by two independent shRNAs against AMOT compared to the control in both Bel-7402 and Bel-7404 cells (Fig. 5A).
Further, cytosolic and nuclear fractionation experiments demonstrated that the knockdown of AMOT could only increase the nuclear expression of YAP while decreasing the cytoplasmic expression of YAP, but AMOT knockdown blocked the TPA-induced translocation of YAP from the nucleus to the cytoplasm in both Bel-7402 and Bel-7404 cells (Fig. 5B). Moreover, the TPA-induced reduction of TEAD and CREB transcriptional activities was abolished when AMOT was knocked down (Fig. 5C). Further, the TPA-reduced downregulation of the mRNA levels of CTGF, ANKRD1, MCAM and HULC in the control cells could not be observed in Bel-7402 and Bel-7404 cells with AMOT knocked down (Fig. 5D). These results demonstrated that the TPA-induced inhibitory effects on YAP rely on the function of AMOT.
Here, we sought to determine whether the TPA-reduced transformative phenotypes in liver cancer cells depend on YAP and AMOT. We found that TPA-reduced cell proliferation and colony formation capacities could be blocked by the knockdown of YAP and AMOT, respectively (Fig. 6A,B). Furthermore, TPA-induced Caspase 3/7 activities could also be abolished by the knockdown of YAP and AMOT, respectively (Fig. 6C). However, TPA still had inhibitory effects on the transformative phenotypes of liver cancer cells, even when the WW domain containing transcription regulator 1 (TAZ), a homolog of YAP, was depleted (Supplementary Fig. S1A-D), thereby excluding the possibility that the effects generated by TPA occur via a TAZ-dependent mechanism.
In vivo xenograft experiments also demonstrated that the TPA efficacy with respect to tumor growth inhibition was much reduced when YAP or AMOT was knocked down (Fig. 6D). These results demonstrate that TPA inhibits transformative phenotypes in liver cancer cells, possibly via AMOT and YAP.
Previous studies have demonstrated that TPA enhances cell migration through activating protein kinase C (PKC) in several types of tumor cells4,6,7. As a PKC activator, TPA also inhibits the apoptosis induced by Fas/FasL in human promyelocytic leukemia cells9. Although TPA has been reported to be pro-tumorigenic39, its exact effects remain controversial. A study from Gong et al.40 has suggested that TPA exerts opposing roles on cell proliferation, possibly via regulating the Hippo/YAP pathway in a cell type-dependent manner. This may be because TPA can activate different PKC isoforms in different types of cells. For example, TPA affects the nPKC isoform to activate LATS, a natural inhibitor of YAP, in Swiss3T3, MEF and A549 cells, whereas TPA affects the cPKC isoform to suppress LATS in HEK293A cells. Thereby, one drug might cause opposite effects via the same signaling pathway. However, which PKC isoform mediates TPA-induced simultaneous translocation of AMOT and YAP from the nucleus to cytoplasm in liver cancer cells remains unknown and must be investigated in future studies. Unlike the pro-tumorigenic roles of TPA in other types of tumor cells, TPA might act as an anti-tumorigenic agent in the human liver cancer cell line HepG28. Consistent with this possibility, we report here that TPA is able to suppress transformative phenotypes in two human liver cancer cell lines, Bel-7402 and Bel-7404, further suggesting that TPA might be used as an anti-tumor drug in the treatment of liver cancer.
YAP translocation from the nucleus to the cytoplasm might dissolve the YAP-TEAD complex41, which promotes cell proliferation and maintains survival programs by inducing the expression of target genes, such as CyclinD1, CyclinE and CTGF42,43. We found that TPA treatment leads to the separation of YAP from a series of its dependent transcription factors, including TEAD, CREB, Runx2 and p73, whose nuclear localization is not affected by TPA. Notably, these transcription factors play critical roles in the promotion or suppression of liver tumorigenesis20,29. Because YAP functions as a co-transcription factor, the loss of YAP in the nucleus decreases the transcriptional activities and subsequent target gene expression of its dependent transcription factors. Although YAP-dependent transcription factors are either pro-tumorigenic (for example, TEAD and CREB) or anti-tumorigenic (for example, Runx2 and p73), we believe the major roles of YAP in liver cancer cells are pro-tumorigenic because YAP is highly up-regulated in liver cancer tissues compared to the corresponding adjacent normal liver19; YAP knockout restricts liver development25; and YAP depletion leads to impaired transformative phenotypes in liver cancer cells28. These results might explain why TPA is anti-tumorigenic in liver cancer cells.
AMOT has been identified as a potential component of Hippo signaling, and numerous studies have reported that AMOT inhibits YAP activity24,44. The interaction between YAP and AMOT has also been well established45, and the PPxY motifs of AMOT are essential for interaction with the WW domains within the YAP protein46. Our findings have demonstrated that without AMOT, TPA is unable to alter the subcellular localization of YAP or the activities of YAP-dependent transcription factors. Therefore, AMOT is critical for maintaining the efficacy of TPA in treating YAP-dependent liver cancer.
In the present and our previous studies47, we found that endogenous AMOT is in the nucleus, whereas exogenous AMOT-HA was observed to be excluded from the nucleus, as most other reports28,45 have described. Protein synthesis occurs in the cytoplasm and especially on the rough endoplasmic reticulum (ER)48. Consequently, exogenous AMOT-HA protein should be abundantly translated on the ER, thus leading to AMOT-HA accumulation in the cytoplasm. It also takes time to transfer AMOT-HA, via an unknown mechanism, to its correct subcellular localization, possibly in the nucleus. We hypothesize that concurrently, the cytoplasmic accumulation of AMOT-HA induces YAP translocation from the nucleus to the cytoplasm, ultimately resulting in its inhibitory effect on YAP. This might explain why AMOT-HA exists mainly in the cytoplasm and might explain the observation that AMOT and YAP appear to bind each other better when they co-localize in the cytoplasm than in the nucleus. The relevant mechanism should be investigated further in the future. Moreover, we have further tested the specificity of the anti-AMOT antibody used in the present study. We found the nuclear staining went away when knocked down AMOT with shRNA, as measured by IF experiment (Fig. S2A). The knockdown efficiencies of AMOT by shRNAs targeting AMOT were also analyzed by WB experiments (Fig. S2B). These results indicated that the nuclear signals that recognized by this anti-AMOT antibody represent genuine endogenous AMOT in the nucleus of Bel-7402 and Bel-7404 cells.
In conclusion, our findings demonstrate that TPA is anti-tumorigenic in liver cancer cells via an AMOT and YAP-dependent mechanism (Fig. 6E). Simultaneously increasing AMOT function might enhance the efficacy of TPA in treating liver cancer.
How to cite this article: Zhu, G. et al. 12-O-Tetradecanoylphorbol-13-acetate (TPA) is anti-tumorigenic in liver cancer cells via inhibiting YAP through AMOT. Sci. Rep. 7, 44940; doi: 10.1038/srep44940 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work was supported by the National Natural Science Foundation of China (Grant nos 81301689 and 81672332), the Yangfan Project of Shanghai Science and Technology Commission (Grant no. 14YF1412300), the China Central Colleges and Universities Basic Research-specific Cross-discipline Grant (Grant no. 1501219096, assigned to Jiayi Wang), the Outstanding Youth Training Program of Tongji University (Grant no. 1501219080, assigned to Jiayi Wang), the Shanghai Jiaotong University School of Medicine Outstanding Young Teacher Grant (assigned to Yongxia Qiao) and the Young College Teachers’ Training Scheme of Shanghai (Grant no. ZZjdyx13007).
The authors declare no competing financial interests.
Author Contributions The project was conceived and designed by Y. Q. and J.W. The experiments were performed by G.Z., Y.C., X.Z., Q.W. and Y.Z. The data were analyzed by Y.Q., Y.C., Y.X.C., F.S. and J. W. The manuscript was written by G.Z. and J.W. and reviewed by all the authors.