• 沒有找到結果。

Autophagy suppresses tumorigenesis by degrading oncogenic miR-224 in HCC Sheng-Hui Lan

N/A
N/A
Protected

Academic year: 2021

Share "Autophagy suppresses tumorigenesis by degrading oncogenic miR-224 in HCC Sheng-Hui Lan"

Copied!
38
0
0

加載中.... (立即查看全文)

全文

(1)

Autophagy suppresses tumorigenesis by degrading oncogenic miR-224 in HCC Sheng-Hui Lan1, Shan-Ying Wu1, Roberto Zuchini2,3, Xi-Zhang Lin3, Ih-Jen Su6, Ting-

Fen Tsai7, Yen-Ju Lin8, Cheng-Tao Wu8 and Hsiao-Sheng Liu1,4,5

1Institute of Basic Medical Sciences; 2Graduate Institute of Clinical Medicine;

3Department of Internal Medicine; 4Department of Microbiology and Immunology;

5Center of Infectious Disease and Signaling Research, College of Medicine, National Cheng Kung University, Tainan, Taiwan; 6National Health Research Institutes, Division of Clinical Research, Tainan, Taiwan; 7Department of Life Sciences and Institute of Genome Sciences, National Yang-Ming University, Taipei, Taiwan;

8Biomedical Technology and Device Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan.

Running title: Autophagy regulates miRNA in HCC

Grant support: This study was supported by grants from the Headquarters of University Advancement, National Cheng Kung University (D100~101-35001); the National Science Council (NSC-99-2745-B-006-002), and from the National Science and Technology Development Fund (No. 97-EC-17-A-31-F1-0695).

Abbreviations: anti-miR-224: miR-224 antagonist; anti-N.C.: anti-miR negative control; AP: autophagosome fraction; Atg: autophagy-related genes; CQ: chloroquine;

DDW: double distilled water; HBSS: Hank’s balanced salt solution; HBV: hepatitis B virus; HBx: hepatitis B virus X protein; HC: hepatitis C virus; HCC: hepatocellular carcinoma; IHC: immunohistochemistry; miR: microRNA; MEF: mouse embryo fibroblast; N.C.: negative control; pre-miR-224: precursor form of miR-224; NH4Cl:

ammonium chloride; PCR: polymerase chain reaction; s.c.: subcutaneously; SD:

Sprague-Dawley; TGF-β: transforming growth factor-beta; 3’UTR: 3’untranslated region.

Corresponding author: Hsiao-Sheng Liu, Department of Microbiology and Immunology, College of Medicine, National Cheng Kung University, #1 University road, Tainan, Taiwan; Tel: 886-6-2353535, ext. 5630; Fax: 886-6-2082705; e-mail:

[email protected]

(2)

Conflicts of interest: The authors disclose no conflicts.

Author contributions: S.H.L and S.Y.W designed the research, conducted experiments and wrote the manuscript; R.Z. and X.Z.L conducted the animal experiments and edited the manuscript; I.J.S and T.F.T provided the HBV transgenic mice; Y.J.L and C.T.W designed the microRNA microarray and analyzed the data;

H.S.L contributed to the concept, the entire experimental design, project funding and supervised the study and in charge of the paper writing.

(3)

Abstract:

BACKGROUND & AIMS: In hepatocelluar carcinoma (HCC), dysregulated expression of microRNAs and autophagy deficiency are reported, individually. In this paper, we reveal a novel mechanism of autophagy, which selectively recruits and degrades miR-224 and explain the mystic relationship between over-expressed miR- 224 and autophagy deficiency in HCC.

METHODS: Mouse embryo fibroblast Atg5(+/+) versus Atg5(-/-) cells and microarray were used to identify autophagy-regulated microRNAs.

Immunoflunorescent staining, real-time PCR and autophagosome extraction protocol were used to demonstrate that autophagosomes selectively recruit miR-224.

TranswellTM assay and xenograft tumor formation in NOD/SCID mice were used to clarify the characteristics of miR-224 and its target-gene Smad4, which was confirmed by pMIR-3’UTR luciferase reporter assay. HBx-transgenic mice and 138 paired HCC specimens were used to clarify the correlation of autophagy, miR-224 and Smad4 by real-time PCR and immunohistochemical staining of tissue array.

Orthotopic transplantation of liver tumor formation in SD rat was utilized to verify the relative expression of autophagy and its ability to generate tumors.

RESULTS: Autophagy selectively recruits and degrades miR-224, which plays an

(4)

oncogenic role in cell migration and tumor formation through silencing Smad4. MiR- 224 is over-expressed in HCC specimens and shows a negative correlation with Atg5 and Smad4 gene expression. This correlation is only detected in HBV-related HCC specimens and confirmed in HBx-transgenic mice. In the orthotopic liver tumor formation of rat, amiodaroneTM enhanced autophagic activity accompanied with HCC regression.

CONCLUSIONS: This is the first report to demonstrate that autophagy deficiency causes miR-224 overexpression, which participates in HBV-related HCC and suggest a promising new direction for HCC treatment.

Keywords: autophagy; microRNA; HCC

Introduction

Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and is the leading cause of cancer-related death. Moreover, HCC is often diagnosed at a late

(5)

stage so the prognosis is often poor.3 Recent studies have highlighted new molecular mechanisms involved in the progression of HCC, including autophagy and microRNAs (miRNAs). Autophagy balances cellular energy metabolites and maintains the quality of proteins. It is induced by metabolic stresses and is a complicated process involving degradation and renewal of energy. Adequate autophagic responses protect cells from different kinds of stress and maintain cell survival mechanisms. A group of autophagy-related genes (Atg), such as Atg5 and Atg7 are essential for autophagic process and silencing these genes completely blocks

autophagy.6 Inadequate autophagic responses cause diverse pathologic conditions in humans, including liver dysfunction and tumorigenesis. Mice with mosaic deletion of Atg5 and liver-specific Atg7(-/-) only developed multiple liver tumors. In addition, autophagic gene Beclin 1 (Atg6) expression was decreased in HCC tissue samples

compared with adjacent non-tumor tissues. Although accumulating evidence indicates that autophagy suppresses tumorigenesis to preserve cellular fitness and genome integrity, the molecular mechanism whereby autophagy inhibits tumorigenesis

remains unclear.

MiRNAs are non-coding RNAs which are initially transcribed as long primary transcripts, then undergo sequential processing by RNase III endonucleases, Drosha and Dicer to become the mature form fragments.13 MiRNAs suppress target-gene

(6)

expression either by transcriptional degradation or by translational inhibition, depending on sequence homology between the miRNA and the target gene. MiRNAs are involved in diverse diseases, including cardiovascular dysfunction, viral infections and many cancers.14 Tumorigenesis-related miRNAs are classified as tumor- suppressing miRNA and oncogenic miRNA.15 MiR-216a and miR-224 were highly expressed in HCC;16 furthermore, overexpression of miR-221 and miR-224 promoted liver tumorigenicity. These findings indicate that miRNAs are involved in HCC tumorigenesis. Recent studies have identified autophagy-regulated genes (Atg2B, Atg4, Beclin 1 and LC3) as the targets of miRNAs (miR-130a, miR-376b, miR-30a and miR-204).19-21 This indicates that the process of autophagy is regulated by

oncogenic or tumor suppressive miRNAs. However, it is not clear if autophagy regulates miRNA expression, thereby contributing to tumor suppression. Therefore, it was our aim in the present study to determine whether a mechanism exists in which autophagy and microRNA interact to regulate HCC. In this study, we investigated tumorigenesis suppression through the regulation of microRNA expression by

autophagy.

Methods

Tumor formation model of mice, orthotopic xenograft model of rat liver tumor formation and HBx-transgenic mice

(7)

The animals were obtained from the Laboratory Animal Center of National Cheng Kung University as detailed in Supplemental Data.

Detection of miRNA and message RNA expression by real-time PCR

This was performed to previously described22 as detailed in Supplemental Data. The primers are listed in Supplementary Table 1.

Clinical specimen and verification

A total of 138 pairs of HCC clinical specimens were obtained from the Taiwan Liver Cancer Network. The analyses of these specimens were described in Supplemental Data.

Autophagosome extraction procedure

Autophagosome extraction was performed as previously reported 23 and described in Supplemental Data.

Statistical analysis

All data are presented as the mean ± SD. Differences between the experimental and control groups were analyzed by Student's t-test. For analyzing survival of HCC patients, log-rank test in GraphPad Prism was used with the P values indicated. *P< . 05, **P< .01, and ***P< .001.

(8)

Results

MiR-224 expression has a negative correlation with autophagy

We previously demonstrated that autophagy induced by the Ha-ras oncogene suppressed tumor formation in NIH3T3 cells.24 Stable MEF wild-type (+/+) and deficient Atg5(-/-) cell lines harboring inducible system, MEF-Atg5(+/+)-Ras and MEF-Atg5(-/-)-Ras, were established to identify the autophagy-related miRNAs which are involved in tumor formation. The tumor volume of MEF-Atg5(-/-)-Ras cells treated with IPTG was the largest (Figure 1A), indicating that autophagy plays a

suppressive role in tumor formation.

A miRNAs microarray was conducted to identify differentially expressed miRNAs between MEF-Atg5(+/+)-Ras and MEF-Atg5(-/-)-Ras cell lines as well as in tumors generated from these two cell lines. Among the aberrantly expressed miRNAs, one up-regulated and five down-regulated miRNAs (those that showed the significant difference both in the cells and cell line-induced tumors) were selected (Supplementary Table 2). We focused on the up-regulated miR-224, because the its expression level of miR-224 in MEF-Atg5(-/-)-Ras cells was >33-fold in the cells and

>22-fold in the cell-induced tumors as compared with that in MEF-Atg5(+/+)-Ras cells. The levels of miR-224 expression in both of the autophagy-deficient cell lines,

(9)

MEF-Atg5(-/-) and MEF-Atg7(-/-), were higher than those in the wild-type MEF cells (Figure 1B), indicating a correlation between miR-224 up-regulation and autophagy deficiency. To determine whether this negative correlation existed under other conditions, autophagic induction either by starvation [Hank’s balanced salt solution (HBSS)] treatment or by oncogenic stress (ras overexpression) in MEF-Atg5(+/+)- Ras cells were used. The expression of LC3-type II, a marker of autophagy, was enhanced and miR-224 was down-regulated after HBSS or IPTG treatment (Supplementary Figure 1). In summary, the inverse relationship between autophagic

activation and miR-224 down-expression is a general phenomenon.

75 pairs of HCC specimens from tumor and adjacent non-tumor tissue were used to evaluate the expression of miR-224 and Atg5 at RNA level. In the tumor tissue, Atg5 was under-expressed (P< .0001), and miR-224 was over-expressed (P= .001) compared with those in the non-tumor tissue of each specimen (Figure 1C).

Furthermore, a tissue array consisting of 116 pairs of HCC specimens was used to determine the expression of Atg5 at a protein level. The protein expression of Atg5 was under-expressed in tumor parts (P< .0001), which was consistent with the RNA expression (Figure 1D). There were 53 pairs of overlapping specimens between RNA samples and tissue arrays, and the correlation between Atg5 and miR-224 expression levels was analyzed by linear regression analysis. The miR-224-related expression

(10)

was negatively correlated with Atg5 expression (P= .0357) (Figure 1E). Altogether, our cell lines and HCC clinical data demonstrated that autophagy activity was negatively correlated with miR-224 expression.

MiR-224 promotes tumorigenesis by silencing target gene Smad4

The synthetic-miR-224 (miR-224) or miRNA negative control (N.C.) was transfected into MEF-Atg5(+/+)-Ras cells in which the expression of endogenous miR-224 is low. In contrast, a miR-224 antagonist (anti-miR-224) or an anti-miR negative control (anti-N.C.) was introduced to suppress miR-224 expression in MEF-Atg5(-/-)-Ras cells which express a high level of miR-224. The colony formation was enhanced by miR-224 in MEF-Atg5(+/+)-Ras cells and suppressed by anti-miR-224 in MEF- Atg5(-/-)-Ras cells (Supplementary Figure 2A). Tumor formation was enhanced by overexpression of miR-224 groups (Figure 2Ai). The expression of miR-224 in tumors is shown in Supplementary Figure 2B. To confirm whether miR-224 promotes liver tumor formation, a rat liver cancer cell line, N1-S1, after miR-224 or N.C.

transfection, was introduced into the liver of SD rats by orthotopic transplantation.

The expression of miR-224 in liver tumors was shown in Supplementary Figure 2C.

The tumor weight in samples induced by N1-S1 cells with miR-224 transfection was increased (Figure 2Aii). Altogether, both in vitro and in vivo studies demonstrated that

(11)

miR-224 plays an oncogenic role in liver cells.

A total of 52 candidate genes were selected by overlapping the target genes predicted by three target gene software programs: TargetScan, PicTar and miRBase (data not shown). Among these target genes, smad4 was chosen because it contains two predicted miR-224-binding sites in the 3’UTR. Smad4, a tumor suppressor gene, suppresses invasion and metastasis through inhibition of Stat3 phosphorylation.25 All the evidences suggest that Smad4 is a potential target gene of miR-224 in HCC. The expression of Smad4 protein was repressed when miR-224 was overexpressed in Hep 3B hepatocellular carcinoma cells, which was consistent with the finding that Smad4 protein expression was induced in miR-224 inhibitor-treated cells (Figure 2B). For determining whether miR-224 directly targeted the Smad4 gene, the luciferase reporter plasmids containing wild-type or mutant-type of miR-224 targeting sequence in the 3’UTR of Smad4 gene (Supplementary Figure 2D) were constructed and co- transfected with miR-224 or anti-miR-224 into 293T cells. In the cells with the wild- type Smad4-Report plasmid, the luciferase activity was repressed by miR-224 and enhanced by anti-miR-224 (Figure 2C). Furthermore, miR-224 and anti-miR-224 had no effect on the luciferase activity of the mutant-type Smad4-Report plasmid (Figure 2C). These data indicate that miR-224 specifically binds to 3’UTR of Smad4, and suppresses its translation. In Hep 3B cells, overexpression miR-224 decreased Smad4

(12)

and it was reversed by co-transfection of the plasmid pRK5-Smad4. In contrast, transfection of anti-miR-224 increased Smad4 expression and this phenomenon was reversed by si-Smad4 treatment (Supplementary Figure 2E). The cells expressing different levels of miR-224 and Smad4 were injected into NOD/SCID mice.

Overexpression of Smad4 suppressed tumor formation, which was induced by miR- 224, and silencing Smad4 expression enhanced tumor formation, which was suppressed by anti-miR-224 (Figure 2D). Furthermore, in Hep 3B cells, miR-224 enhanced migration activity by silencing Smad4 expression (Supplementary Figure 2F). In summary, our data indicate that miR-224 promotes tumorigenesis of HCC by

targeting Smad4 expression.

Smad4 expression was evaluated in HCC specimens, and it was under-expressed in the tumor tissues (P< .0001) (Figure 2E). Furthermore, Smad4 was negatively correlated with miR-224 expression (P= .037) and positively correlated with Atg5 expression in HCC tumor specimens (P= .001) (Figure 2F). Altogether, our results show that low autophagic activity accompanied by miR-224 up-regulation and Smad4 down-regulation promote HCC tumorigenesis.

MiR-224 is degraded through the autophagosome-lysosome pathway

To determine whether miR-224 was suppressed by autophagy through transcriptional regulation, the expression levels of precursor and mature forms of miR-224 were

(13)

evaluated. The expression of the precursor form of miR-224 (pre-miR-224) was not changed between the wild-type and autophagy-deficient cells as well as in tumors (Supplementary Figure 3A and B), indicating that miR-224 regulation by autophagy was not at the transcriptional level. Furthermore, the pre-miR-224 expression levels showed no differences between the tumor and the paired adjacent non-tumor tissue (Figure 3A), and Atg5 expression level was not correlated with pre-miR-224 expression (P= .7762) (Figure 3B). In summary, miR-224 regulation by autophagy

was at the post-transcriptional level in HCC.

A fraction of miRNA-loaded AGO2 complex is sorted into lysosome for secretion and/or lysosomal degradation.26 It is possible that during autophagy progression, the mature form of miR-224 is recruited to the autophagosome and degraded in the autolysosome (autophagosome fused with lysosome). We determined the autophagic flux by transfection of the ptfLC3 plasmid containing the monomeric red-fluorescent and green-fluorescent-LC3 gene. The yellow color represents steady-state autophagosome formation and red color represents autophagic progression from autophagosome to autolysosome, a sign of autophagic flux.27 Our data show that only red fluorescence was detected after Ha-ras overexpression, indicating that autophagosome proceeds to autolysosome. Chloroquine (CQ) is a fusion blocker of autophagosome and lysosome. After CQ treatment, the reduction of GFP fluorescence

(14)

was blocked, indicating accumulation of autophagosomes (Supplementary Figure 3C).

Furthermore, the accumulation of LC3-type II suggests that the molecules in the autophagosome are not degraded in the presence of CQ (Figure 3Ci). Under the same conditions, the expression of the mature form of miR-224 was down-regulated when autophagic flux was induced by Ha-ras, and it was reversed by CQ treatment (Figure 3Cii). These data support the notion that the mature form miR-224 is degraded during

autophagic flux.

To exclude the possibility that miR-224 is decreased through increased secretion, miRNA level in the culture medium was determined. The extracellular and intracellular miR-224 expression was similar. It also reduced when autophagy was induced and reversed by CQ treatment (Supplementary Figure 3D). These results indicate that suppression of miR-224 expression by autophagy is not achieved through

increased miRNA secretion.

To clarify whether miRNAs degradation through autophagy is a general phenomenon, we investigated the distribution of fluorescence-labeled miR-224 (5’FAM-miR-224) and 5’FAM-let-7a in the cell. The let-7a was used as a negative control because it was not suppressed by autophagy (Supplementary Figure 3E and F). Co-localization of 5’FAM-miR-224 and autophagosome (LC3 puncta) but not 5’FAM-let-7a was detected after autophagy induction and the percentage of co-localization was

(15)

increased in the presence of CQ (Figure 3D). Furthermore, co-localization of lysosomes (LAMP1) and 5’FAM-miR-224 was detected when autophagy was induced (Figure 3E). Co-localization of 5’FAM-miR-224 and lysosomes was increased after ammonium chloride (NH4Cl) (a neutralizer of acidic solution) treatment, but no difference in the co-localization of 5’FAM-let-7a and LAMP1 was

found (Figure 3E).

Autophagosomes were extracted to directly detect the amount of miRNAs. We used density gradient to purify autophagosomes.23 Autophagosomes were purified from MEF-Atg5(+/+)-Ras cells which were pretreated with IPTG and CQ to enrich autophagosomes formation. The LC3-type II was enriched in the autophagosome fraction (AP) (Figure 3Fi), indicating that the autophagosomes were successfully purified. Compared with whole cells, miR-224 was increased in AP; however, let-7a level was reduced. This indicates that miR-224 was more selectively engulfed by the autophagosome (Figure 3Fii). Collectively, the inverse correlation between miR-224 down-regulation and autophagic activation was due to the autolysosome-related degradation pathway.

MiR-224 overexpression and low autophagic activity are associated with HBV infection-related HCC

(16)

overexpression, a multivariate analysis of HCC specimens was conducted. Among the factors analyzed, virus infection showed a different correlation with autophagy suppression and miR-224 overexpression (Table 1). The correlation of low-Atg5 (P< . 0001) and high-miR-224 (P< .006) was significant in Hepatitis B virus (HBV)-related

HCC, but not in Hepatitis C virus (HCV)-related HCC.

The relationships among HBV-related HCC, autophagy and miR-224 was conducted in the HCC patients. The histological staining of two HCC patients from 60 pairs of tissues was shown in Figure 4A. Atg5 and Beclin 1 expression were under-expressed (P< .0001); in contrast, p62 expression increased and formed aggregates when autophagy activity decreased (P= .0002) in tumor tissues (Figure 4A and B).

However, the expression levels of Atg5, Beclin 1 and p62 in the HCV-associated HCC tissues were not changed (P= .1473, P= .9005 and P= .3413) (Supplementary Figure 4A and B). Furthermore, Smad4 protein expression was decreased in HBV- associated (P< .0001) but not in HCV-associated HCC (P= .5566) (Figure 4A, B, Supplementary Figure 4A and B). In summary, low autophagic activity, miR-224 overexpression and down-expression of Smad4 mainly occurred in HBV- but not in

HCV-associated HCC.

Hepatitis B virus X (HBx) protein participates in carcinogenesis of HCC, and transgenic mice harboring the liver-specific albumin promoter-driven HBx gene

(17)

developed liver tumor.28 In these HBx-transgenic mice, liver tumor formation was detectable at 16 months. The liver specimens from the 6- and 17.5-month-old transgenic mice were analyzed (Figure 4C). The decreased Atg5, Smad4 expression and increased p62 aggregation were detected in the tumor of the 17.5-month-old mice (Figure 4D), and the quantitative data were shown in Figure 4E. Consistently, miR- 224 expression was increased in the tumor tissue (Figure 4E). Altogether, lower autophagic activity together with miR-224 overexpression is involved in HBx gene- associated tumorigenesis.

Autophagy and related genes are potential targets for cancer therapy and

biomarkers

Our results demonstrated that autophagy had a suppressive effect on oncogenic miR- 224 in HCC. To clarify whether miR-224 as well as autophagy could be potential targets for liver cancer therapy, the use of anti-miR-224 and autophagy inducer were investigated in rat models. Liver tumor formation was reduced in the group that received N1-S1 cells transfected with anti-miR-224 (Figure 5A). AmiodaroneTM, an autophagy inducer selected from 3584 chemicals,29 was used to study the role of autophagy in hepatic pathogenesis. N1-S1 cells were orthotopically transplanted into the rats’ livers, followed by intraperitoneal injection of amiodaroneTM or double

(18)

distilled water (DDW). Tumor weight was reduced in the groups receiving amiodaroneTM treatment after one week (Figure 5A). Autophagy induction in rats was demonstrated by overexpression of LC3-type II in the amiodaroneTM (30 mg/kg) group (Figure 5B). The characteristic double membrane of the autophagosome organelle was detected by transmission electronic microscopy in the amiodaroneTM but not in the DDW group (Figure 5C and D). Furthermore, the expression of miR-224 was decreased after amiodaroneTM or anti-miR-224 treatment (Figure 5E). Altogether, the autophagy inducer have potential as a therapeutic tool for HCC. In addition, we were interested in determining if miR-224, Atg5 and Smad4 had a correlation with survival rates in patients. Patients were divided into a high-risk group (high-miR-224, low- Atg5 and low-Smad4) and a low-risk group (low-miR-224, high-Atg5 and high- Smad4). Kaplan-Meier survival analysis showed that the survival rate was higher in the low-risk group (P= .0358) (Figure 5F). In summary, autophagic activity combined with the expression of miR-224 and Smad4 potentially have clinical value as prognostic biomarkers for the prediction of survival rates in patients with HCC.

Discussion

Accumulating evidence suggests that autophagy is involved in tumor suppression.30 Autophagy-deficient mice develop multiple liver tumors and promote fibrosarcomas

(19)

induced by chemical carcinogens.31 Therefore, a clearer understanding of how autophagy suppresses tumorigenesis may facilitate the development of cancer therapies. In this study, we identified a previously unknown mechanism of autophagy which suppresses tumorigenesis through regulation of miRNA. In cells and tumors with deficient autophagy, miR-224 was overexpressed and tumorigenesis was promoted by silencing Smad4 expression. Autophagy down-regulates miR-224 via recruitment of miR-224 to the autophagosome followed by lysosome degradation (Figure 6). Low autophagic activity accompanied by high miR-224 expression, was detected in HBV-associated HCC specimens and HBx-transgenic mice. Autophagy inducer, amiodarone, and anti-miR-224 were both found to suppress liver tumor formation in the rat model. Oncogenic miR-224 overexpression resulted from suppression of autophagic activity in HBV-related HCC progression. However, the mechanism by which autophagic activity is decreased during HCC progression remains further study.

Autophagy regulates post-transcriptional expression of miR-224

As with protein-encoding genes, miRNA promoters are regulated either by transcriptional elements or by epigenetic modification.32 The upregulation of miR-224 occurs in concert with upregulation of its neighboring miR-452 and the GABRE gene

(20)

by HDAC1, HDAC3 and p300 in HCC, indicating that miR-224 together with the GABRE gene are up-regulated at the transcriptional level.33 In contrast, miR-224 up- regulation did not coordinate with the GABRE gene in 75 pairs of HCC specimens (data not shown). Furthermore, only the mature form of miR-224 is regulated at the post-transcriptional level by the autophagic process in HCC (Figure 1C and 3A). This discrepancy may have been caused by variations in genetic background as well as by

the involvement of different etiologic agents in liver cancer formation.

It has been reported that RISC complex is coupled to the late endosome and fuses with the lysosome during small RNA processing.26 In this study, miR-224 is colocalized with the autophagosome and degraded after fusion with the lysosome (Figure 3D and E). It is the first to demonstrate that autophagy recruits and regulates

miRNA through autophagy-related degradation.

In the miRNA microarray screening, miR-224 was up-expressed but let-7a was not changed in autophagy-deficient cells and tumors (Figure 1B and Supplementary Figure 3E). In the purification of autophagosomes, miR-224 but not let-7a was highly expressed in the autophagosome vesicle (Figure 3Fii), which indicates that miR-224 is selectively recruited to the autophagosome. Accumulating evidence indicates that autophagy selectively degrades its targets, including aggregated proteins, invasive pathogens, damaged mitochondria and specific proteins like Src.34 These

(21)

autophagosome substrates are carried by their cargo adaptor proteins into the autophagosome.35 It is probable that an unidentified RNA-binding protein recognizes the specific sequence, interacts with miR-224 and then transports it to the autophagosome for degradation. The miRNA post-transcriptional processing of regulatory factors which bind with miRNA to enhance or inhibit mature miRNA production have been identified.36 These unidentified RNA-binding proteins could be the autophagosome substrates or adapted proteins. Further study is currently underway to identify the binding protein of miR-224.

MiR-224 promotes tumorigenesis through targeting Smad4

TGF-β signaling pathway regulates various functions in tumorigenesis. It suppresses cell proliferation in the initial stage, but promotes metastasis in the late stage. Recent report demonstrated that loss of Smad4 switches TGF-β from being a tumor suppressor to a tumor promoter.37 However, it is not clear that the role of smad4 in the tumorigenesis of HCC. Our data demonstrated that miR-224 targets and silences Smad4 expression, and overexpression of Smad4 reversed the bio-function of miR- 224 in HCC. Altogether, these findings reveal that miR-224 promotes tumorigenesis through inhibiting Smad4.

Regulation of autophagy have therapeutic potential for HCC

(22)

MiR-224 is over-expressed in all the liver cancers,38 suggesting that miR-224 plays a critical role in hepatic pathogenesis. In this study, Atg5 down-regulation and miR-224 up-regulation were negatively correlated with HCC tumorigenesis. These data are consistent with our in vitro studies that loss of autophagic activity causes miR-224 up- expression and promotes liver tumorigenesis. In addition, the autophagy inducer, amiodaroneTM, suppresses liver tumor formation, suggesting that manipulating the autophagic activity in liver may shed light on mechanisms which could prevent HCC development. Further experimental trials are required to determine the optimal dosage

of amiodaroneTM for suppressing liver cancer formation.

HBV is a hepatotropic virus that can cause severe liver diseases, including liver cirrhosis and HCC. During HBV infection, both of the HBV surface (S) and X proteins induce autophagy, and autophagy is essential for HBV replication. These results imply that autophagy have potential as a therapeutic target against HBV infection. Interestingly, we demonstrated that low autophagy activity was associated with HBV-related HCC. This result was confirmed by evidence of tumorigenesis in HBx-transgenic mice. Our results were consistent with a previous study that low autophagy activity of Beclin 1+/– HBV transgenic mice induced more severe tumorigenesis.41 In summary, autophagy was demonstrated to enhance HBV infection, and a suppressor role in HBV-related tumorigenesis. These results suggest that prior

(23)

to conducting an autophagy-related therapy in HBV patients, the stage of HBV infection as well as the status of the patient should be considered. In conclusion, our findings open a new avenue for the treatment of HBV infection as well as HCC by manipulating autophagic activity or miR-224 expression level.

Acknowledgments

We thank Drs. Rudy Juliano, Rik Derynck and Tamotsu Yahsimori for providing the plasmids pRK5-Smad4 and ptfLC3; Drs. Kuen-Jer Tsai and Ya-Chun Hsiao for conducting FACS-like tissue-cytometry, image acquisition and data analysis; Sheng- Hsiang Lin and Shang-Chi Lee for providing statistical analysis of the data. We also thank Taiwan Liver Cancer Network (TLCN), National Health Research Institutes, Zhunan, Taiwan for providing the HCC tissue samples and related anonymous clinical data, which was supported by grant from the National Science Council since 2005 (NSC100-2325-B-182-006). TLCN consists of five major medical centers (National Taiwan University Hospital, Linkou- and Kaohsiung- Chang-Gung Memorial Hospital, Taichung- and Kaohsiung- Veterans General Hospital). We thank the Industrial Technology Research Institute for their assistance with the microarray and technical assistance.

(24)
(25)

Figure legends

Figure 1. miR-224 expression is negatively correlated with autophagy. (A) MEF- Atg5(+/+)-Ras and MEF-Atg5(-/-)-Ras cell lines were injected s.c. into NOD/SCID mice. These mice were divided into IPTG drinking water (12.5 mM) and normal water groups, then tumor volume was measured over 14 days. (B) The expression of miR-224 was detected by real-time PCR in MEF (parental cells), MEF-Atg5(-/-) and MEF-Atg7(-/-) cells. (C) Validation of the expression Atg5 and miR-224 from 75 pairs of HCC specimens by real-time PCR. The relative expression fold of miR-224 is normalized to snoRNA-55. (D) Representative IHC analysis of Atg5 from 116 pairs of HCC patients. Protein expression was detected by defining regions of interest (ROI) using automated cell acquisition and quantification software for IHC (Histoquest). Scale bar = 20 μm. (E) MiR-224 expression was correlated with Atg5 level data available in 53 paired HCC using Spearman’s correlation.

Figure 2. MiR-224 promotes tumorigenesis by silencing target gene Smad4. (Ai) The MEF-Atg5(+/+)-Ras cells were transfected with miR-224, negative miRNA (N.C.) or non-treatment (N.T.) by s.c. injection into NOD/SCID mice, which were drinking IPTG water for 14 days. (Aii) Rat N1-S1 hepatoma cells were transfected with miR-224 or N.C. and injected intrasplenically into the rat’s liver for 1 week.

(26)

Arrows indicate the location of HCC. Scale bar = 1 cm. (B) Hep 3B cells were transfected with miR-224, anti-miR-224 or their negative control. (C) 293T cells were transfected wild-type pMIR-Report-Smad4 vectors or mutant type (1.6 μg) and co- transfected with pre-miR-224, anti-miR-224 or their negative control individually.

The luciferase activity was determined after 48 h and normalized by renilla luciferase expressed by the plasmid pRL-TK (0.4 μg). (D) Hep 3B cells were transfected with N.C. and miR-224. In the transfected miR-224 group, cells co-transfected with pRK5- Smad4 (4 μg) or control vector (4 μg). In addition, Hep 3B were transfected with anti- N.C. and anti-miR-224. In the transfected anti-miR-224 group, cells co-transfected with si-Smad4 (200 nM) or si-RNA negative control (si-N.C.) (200 nM). After transfection for 48 h, the transfected cells were injected s.c. into NOD/SCID mice for 14 days and tumor weight was measured. (E) Representative IHC analysis of Smad4 in 116 pairs of HCC patients. Protein expression was measured by ROI and HistoQuest software. Scale bar = 20 μm. (F) Smad4 expression was correlated with miR-224 and Atg5 data available in 53 paired HCC samples using Spearman’s correlation.

Figure 3. MiR-224 is degraded via autophagosome-lysosome pathway. (A) Validation of the expression of pre-miR-224 from 75 paired HCC specimens by real-

(27)

time PCR. (B) The pre-miR-224 expression was correlated with Atg5 expression data available from 53 paired HCC samples. P values were obtained by using a paired Student’s t-test and Spearman’s correlation, respectively. (Ci and Cii) MEF-Atg5(+/

+)-Ras cells were treated with or without IPTG for 48 h. In the CQ group, cells were treated with IPTG (5 mM) for 24 h and then co-treated with CQ (50 μM) for another 24 h. (D) The 5’FAM-miR-224 or 5’FAM-let-7a were transfected into MEF-Atg5(+/

+)-Ras cells. Cells were treated with IPTG and CQ as in 3C and 3D. The LC3 protein was labeled by red fluorescence. (E) 5’FAM-miR-224 and 5’FAM-let-7a were transfected into MEF-Atg5(+/+)-Ras cells. Cells were treated with IPTG for 48 h. In the NH4Cl group, cells were treated with IPTG for 24 h and then co-treated with NH4Cl (10 mM) for another 24 h. The LAMP1 was labeled by red fluorescence. The arrows indicate colocalization, and the scale bar represents 10 μm. (Fi and Fii) MEF- Atg5(+/+)-Ras cells were treated with IPTG and CQ as in 3C and 3D. Protein and RNA were extracted from total lysates of whole cell and purified autophagosome by density gradient. Protein was analyzed by immunoblotting for LC3, Hsp-60 (a marker of mitochondria) and Calreticulin (a marker of Endoplasmic Reticulum) expression.

Figure 4. MiR-224 overexpression and low autophagic activity are associated with HBV infection-related HCC. (A) Representative IHC analysis of Atg5, Beclin

(28)

1, p62 and Smad4 in 60 pairs of HBV-infected HCC patients. (B) Protein expression was measured by ROI and HistoQuest software. Scale bar = 20 μm. (C) Development of liver tumor in the HBx-transgenic mice. Arrows indicate the location of liver tumor. Scale bar = 1 cm. (D) Representative IHC analysis in paraffin section array includes that normal tissues form 6 months old mice, non-tumor and tumor tissues from 17.5 month old mice. Scale bar = 20 μm. (E) Protein expression was measured by ROI and HistoQuest software. The expression of miR-224 was evaluated by real- time PCR.

Figure 5. Autophagy is used as cancer therapy and prognostic survival factors. In the first group, rat N1-S1 hepatoma cells were injected into the rat’s liver, followed by DDW or amiodarone (20 mg/kg or 30 mg/kg) injected intraperitoneally. In the second group, rat N1-S1 hepatoma cells were transfected with anti-miR-224 or anti-N.C. and injected into the rat’s liver. (A) Rats were sacrificed and after 1 week. The arrows indicate the location of HCC. (B) Protein was extracted from the tumor of first group and there were four mice in each group. (C and D) Tumors were investigated under TEM. For (Ci) and (Di), scale bar = 2 μm; for (Cii) and (Dii), scale bar = 500 nm.

Box in (Ci) and (Di) refer to (Cii) and (Dii). The arrows indicate autophagosome. (E) The expression of miR-224 was evaluated by real-time PCR. (F) The definition of risk

(29)

factors included Atg5 and Smad4 lower than average and miR-224 higher than average in 53 HCC specimens. The survival rates were determined at 2 years after cancer surgery. P values were obtained by Student’s t-test and Kaplan-Meier survival analysis.

Figure 6. Hypothetical model of the mechanism that autophagy regulates miR-

224 expression in HCC.

miR-224 promotes liver tumor formation and cell migration through inhibiting expression of Smad4. When autophagy is induced, oncogenic miR-224 is selectively engulfed into autophagosome and fused with lysosome for degradation. In HCC, autophagic deficiency play pivotal role in overexpression of miR-224 and enhancement of hepatic pathogenesis

(30)

References:

1. Hou J, Lin L, Zhou W, et al. Identification of miRNomes in human liver and hepatocellular carcinoma reveals miR-199a/b-3p as therapeutic target for hepatocellular carcinoma. Cancer Cell 2011;19:232-43.

2. Liang L, Wong CM, Ying Q, et al. MicroRNA-125b suppressesed human liver cancer cell proliferation and metastasis by directly targeting oncogene

LIN28B2. Hepatology 2010;52:1731-40.

3. Budhu A, Forgues M, Ye QH, et al. Prediction of venous metastases, recurrence, and prognosis in hepatocellular carcinoma based on a unique immune response signature of the liver microenvironment. Cancer Cell 2006;10:99-111.

4. Aravalli RN, Steer CJ, Cressman EN. Molecular mechanisms of hepatocellular carcinoma. Hepatology 2008;48:2047-63.

5. Inami Y, Waguri S, Sakamoto A, et al. Persistent activation of Nrf2 through p62 in hepatocellular carcinoma cells. J Cell Biol 2011;193:275-84.

6. Wang Y, Singh R, Massey AC, et al. Loss of macroautophagy promotes or prevents fibroblast apoptosis depending on the death stimulus. J Biol Chem 2008;283:4766-77.

7. Dalby KN, Tekedereli I, Lopez-Berestein G, et al. Targeting the prodeath and prosurvival functions of autophagy as novel therapeutic strategies in cancer.

Autophagy 2010;6:322-9.

8. Jin S, White E. Tumor suppression by autophagy through the management of metabolic stress. Autophagy 2008;4:563-6.

9. Ding ZB, Shi YH, Zhou J, et al. Association of autophagy defect with a malignant phenotype and poor prognosis of hepatocellular carcinoma. Cancer Res 2008;68:9167-75.

10. Takamura A, Komatsu M, Hara T, et al. Autophagy-deficient mice develop multiple liver tumors. Genes Dev 2011;25:795-800.

11. Sakai Y, Oku M, van der Klei IJ, et al. Pexophagy: autophagic degradation of peroxisomes. Biochim Biophys Acta 2006;1763:1767-75.

12. Mathew R, Kongara S, Beaudoin B, et al. Autophagy suppresses tumor

progression by limiting chromosomal instability. Genes Dev 2007;21:1367-81.

13. Kim VN. MicroRNA biogenesis: coordinated cropping and dicing. Nat Rev Mol Cell Biol 2005;6:376-85.

14. Couzin J. MicroRNAs make big impression in disease after disease. Science 2008;319:1782-4.

15. Hwang HW, Mendell JT. MicroRNAs in cell proliferation, cell death, and

(31)

tumorigenesis. Br J Cancer 2006;94:776-80.

16. Chen PJ, Yeh SH, Liu WH, et al. The androgen pathway stimulates

microRNA-216a transcription to suppress the TSLC1 tumor suppressor gene in early hepatocarcinogenesis. Hepatology 2012.

17. Callegari E, Elamin BK, Giannone F, et al. Liver tumorigenicity promoted by microRNA-221 in a mouse transgenic model. Hepatology 2012.

18. Li Q, Wang G, Shan JL, et al. MicroRNA-224 is upregulated in HepG2 cells and involved in cellular migration and invasion. J Gastroenterol Hepatol 2010;25:164-71.

19. Kovaleva V, Mora R, Park YJ, et al. MicroRNA-130a targets ATG2B and DICER1 to inhibit autophagy and trigger killing of chronic lymphocytic leukemia cells. Cancer Res 2012.

20. Korkmaz G, le Sage C, Tekirdag KA, et al. miR-376b controls starvation and mTOR inhibition-related autophagy by targeting ATG4C and BECN1.

Autophagy 2012;8.

21. Fu LL, Wen X, Bao JK, et al. MicroRNA-modulated autophagic signaling networks in cancer. Int J Biochem Cell Biol 2012.

22. Mak CM, Lam CW, Lai ST, et al. Rapid diagnosis of Wilson disease by a 28- mutation panel: real-time amplification refractory mutation system in

diagnosing acute Wilsonian liver failure. Clin Chim Acta 2008;398:39-42.

23. Seglen PO, Brinchmann MF. Purification of autophagosomes from rat hepatocytes. Autophagy 2010;6.

24. Wu SY, Lan SH, Cheng DE, et al. Ras-Related Tumorigenesis Is Suppressed by BNIP3-Mediated Autophagy through Inhibition of Cell Proliferation.

Neoplasia 2011;13:1171-82.

25. Zhao S, Venkatasubbarao K, Lazor JW, et al. Inhibition of STAT3 Tyr705 phosphorylation by Smad4 suppresses transforming growth factor beta- mediated invasion and metastasis in pancreatic cancer cells. Cancer Res 2008;68:4221-8.

26. Lee YS, Pressman S, Andress AP, et al. Silencing by small RNAs is linked to endosomal trafficking. Nat Cell Biol 2009;11:1150-6.

27. Kimura S, Noda T, Yoshimori T. Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent-tagged LC3.

Autophagy 2007;3:452-60.

28. Wu BK, Li CC, Chen HJ, et al. Blocking of G1/S transition and cell death in the regenerating liver of Hepatitis B virus X protein transgenic mice. Biochem Biophys Res Commun 2006;340:916-28.

(32)

autophagy reveals novel therapeutic inhibitors of mTORC1 signaling. PLoS One 2009;4:e7124.

30. Mathew R, Karp CM, Beaudoin B, et al. Autophagy suppresses tumorigenesis through elimination of p62. Cell 2009;137:1062-75.

31. Marino G, Salvador-Montoliu N, Fueyo A, et al. Tissue-specific autophagy alterations and increased tumorigenesis in mice deficient in

Atg4C/autophagin-3. J Biol Chem 2007;282:18573-83.

32. Breving K, Esquela-Kerscher A. The complexities of microRNA regulation:

mirandering around the rules. The International Journal of Biochemistry &

Cell Biology 2010;42:1316-1329.

33. Wang Y, Toh HC, Chow P, et al. MicroRNA-224 is up-regulated in hepatocellular carcinoma through epigenetic mechanisms. FASEB J 2012.

34. Sandilands E, Serrels B, McEwan DG, et al. Autophagic targeting of Src promotes cancer cell survival following reduced FAK signalling. Nat Cell Biol 2012;14:51-60.

35. Johansen T, Lamark T. Selective autophagy mediated by autophagic adapter proteins. Autophagy 2011;7:279-96.

36. Breving K, Esquela-Kerscher A. The complexities of microRNA regulation:

mirandering around the rules. Int J Biochem Cell Biol 2010;42:1316-29.

37. Zhang B, Halder SK, Kashikar ND, et al. Antimetastatic role of Smad4 signaling in colorectal cancer. Gastroenterology 2010;138:969-80 e1-3.

38. Ladeiro Y, Couchy G, Balabaud C, et al. MicroRNA profiling in

hepatocellular tumors is associated with clinical features and oncogene/tumor suppressor gene mutations. Hepatology 2008;47:1955-63.

39. Sir D, Tian Y, Chen WL, et al. The early autophagic pathway is activated by hepatitis B virus and required for viral DNA replication. Proc Natl Acad Sci U S A 2010;107:4383-8.

40. Li J, Liu Y, Wang Z, et al. Subversion of cellular autophagy machinery by hepatitis B virus for viral envelopment. J Virol 2011;85:6319-33.

41. Qu X, Yu J, Bhagat G, et al. Promotion of tumorigenesis by heterozygous disruption of the beclin 1 autophagy gene. J Clin Invest 2003;112:1809-20.

(33)
(34)
(35)
(36)
(37)
(38)

參考文獻

相關文件

1. In the textbook, pp 224-223, there is a surprising result called Space- filling curve. In addition, note the proof is related with Cantor set in exercise 7. There exists a

The daily averages of total suspended solids as well as oil and fats in the waste water treated at the WWTP in Macao Peninsula were higher than the standard values of

Basing off the Sheng-yu Canon manuscript as well as the Nian Chapter included in the block-printed canons of the Central Plain lineage and the Southern lineage, this article uses

6 《中論·觀因緣品》,《佛藏要籍選刊》第 9 冊,上海古籍出版社 1994 年版,第 1

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

In the present case, serum levels of phosphate and calcitriol as well as the tubular phosphate reabsorption become normal in 1 week after surgical removal of the tumor.. However,

In this respect, the aim of the present study was to as- sess volumetric as well as morphological surface changes of the orbital cavity in patients treated with both tooth- borne