Ribonucleotide reductase M2B inhibits cell migration and spreading by early growth response protein 1-mediated phosphatase and tensin homolog/Akt1 pathway in hepatocellular carcinoma

Authors

  • Hua Tian,

    1. State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Chao Ge,

    1. State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Hong Li,

    1. State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Fangyu Zhao,

    1. State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Helei Hou,

    1. State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Taoyang Chen,

    1. Qi Dong Liver Cancer Institute, Qi Dong, Jiangsu Province, China
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Guoping Jiang,

    1. Department of General Surgery, the First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Haiyang Xie,

    1. Department of General Surgery, the First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Ying Cui,

    1. Cancer Institute of Guangxi, Nanning, China
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Ming Yao,

    1. State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Jinjun Li

    Corresponding author
    1. State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
    • Address reprint requests to: Jinjun Li, Ph.D., State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, 25/Ln 2200, Xietu Road, Shanghai 200032, China. E-mail: jjli@shsci.org; fax: +86-21-64432140.

    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

  • Supported in part by grants from the National Key Sci-Tech Special Project of China (2013ZX10002-011), the National Key Program for Basic Research of China (973) (2009CB521803), National Natural Science Foundation of China (81272438), the Specialized Research Fund for the Doctoral Program of Higher Education of China (20110073120105), Research Project of Shanghai Municipal Health Bureau (20114169), Leading Academic Discipline Project of Shanghai Municipal Education Committee (J50208) and Shanghai Natural Science Foundation (12ZR1430000).

Abstract

Ribonucleotide reductase (RR)M2B is an enzyme belonging to the ribonucleotide reductase enzyme family, which is essential for DNA synthesis and repair. RRM2B plays an important role in tumor progression and metastasis; however, little is known about the expression and underlying molecular mechanisms of RRM2B in hepatocellular carcinoma (HCC). In the present study, we report that down-regulation of RRM2B in HCC is negatively associated with intrahepatic metastasis, regardless of p53 status. Moreover, the ectopic overexpression of RRM2B decreased HCC cell migration and invasion in vitro, whereas silencing RRM2B expression resulted in increased migration and invasion in vitro and intrahepatic and lung metastasis in vivo. Additionally, knockdown of RRM2B by short hairpin RNA (shRNA) in HCC cells was associated with epithelial-mesenchymal transition (EMT), including the down-regulation of E-cadherin, and the concomitant up-regulation of N-cadherin and slug. A further experiment showed that RRM2B inhibited cell migration and spreading through regulation of the early growth response protein 1 (Egr-1) / phosphatase and tensin homolog (PTEN) / Akt1 pathway. Consistently, we also detected a significant correlation between RRM2B and E-cadherin protein expression in HCC tissues. Furthermore, Egr-1 also directly bound to the RRM2B promoter and repressed RRM2B transcription, thereby establishing a negative regulatory feedback loop. Conclusion: These findings indicate that RRM2B suppresses cell migration and spreading by way of modulation of the Egr-1/PTEN/Akt1 pathway. (Hepatology 2014;59:1459-1470)

Abbreviations
ChIP

chromatin immunoprecipitation

DMEM

Dulbecco's modified Eagle's medium

Egr-1

early growth response protein 1

EMT

epithelial-mesenchymal transition

HBV

hepatitis B virus

HCC

hepatocellular carcinoma

IHC

immunohistochemistry

PTEN

phosphatase and tensin homolog

qRT-PCR

quantitative reverse transcription polymerase chain reaction

RR

ribonucleotide reductase

shRNA

short-hairpin RNA

siRNA

small interfering RNA

Human hepatocellular carcinoma (HCC) is the most common malignant tumor in the liver and third leading fatal cancer.[1] The prognosis for HCC remains poor, mainly due to the propensity for metastatic progression and poor response to pharmacological treatment.[2] In HCC, the liver is the major target organ of metastasis, which is known as intrahepatic metastasis. The portal vein is the main route for intrahepatic metastases of HCC cells in animal model systems and in human patients.[3, 4] Therefore, the identification of metastatic factors and elucidation of the underlying molecular mechanism that are involved in the progression of metastasis become critical issues.

Epithelial-mesenchymal transition (EMT) is an essential developmental process through which cells of epithelial origin lose cell-cell contacts and cell polarity and acquire mesenchymal phenotypes. EMT is associated with invasion and metastasis of tumor.[5] As a feature of aggressive tumors, EMT is characterized by the down-regulation of E-cadherin expression and up-regulation of N-cadherin expression.[6] Although the role of EMT in tumor invasion and metastasis has become a topic of interest, the molecular mechanism by which EMT is regulated is not fully understood.

Ribonucleotide reductase (RR) is an essential enzyme that catalyzes the conversion of ribonucleoside diphosphates to 2′-deoxyribonucleoside diphosphates, which are required for DNA synthesis and repair. In humans, RR is composed of one large subunit (RRM1) and two small subunits (RRM2 and RRM2B). RRM2B (also known as p53R2) was originally identified as a target gene of the p53 tumor suppressor protein and is transcriptionally induced after DNA damage.[7, 8] Although RRM2B and RRM2 are 80% similar in amino acid sequence, they have been found to function differently in human tumors. RRM2 overexpression promotes tumor cell proliferation, and silencing RRM2 expression inhibits cell growth and enhances tumor chemosensitivity to gemcitabine in human tumor cell lines.[9, 10] Emerging evidence implicates RRM2B as important mediators of cell migration and invasion involved in determination of tumor metastasis and prognosis. RRM2B expression is associated with the progression of human tumor, such as lung cancer, esophageal squamous cell carcinoma, and colon cancer.[11-13] Furthermore, RRM2B inhibits tumor cell invasion in p53-wild-type and p53-mutated human cancer cells and inhibits the MEK-ERK-mediated malignancy of human cancer cells.[14, 15] To date, however, there is little evidence on the role and clinical significance of RRM2B expression in HCC. Thus, the function and the molecular mechanism underlying the role of RRM2B in HCC and the relationship between its expression and clinicopathologic significance remain unclear.

In the present study we report that the deregulation of RRM2B in HCC contributes to a poor prognostic phenotype. Furthermore, RRM2B inhibits HCC cell migration and invasion in vitro and metastasis in vivo. In addition, the molecular mechanism of RRM2B in HCC metastasis was also addressed in this study.

Materials and Methods

Tissue Microarray Construction and Immunohistochemistry (IHC)

A tissue microarray that included 236 HCC tissues was constructed as previously described.[16] IHC, signal evaluation, and statistical data analysis were performed as described previously.[16] Antibody information is listed in Supporting Table 1.

Cell Lines and Transfections

For information on cell lines used in this study, see Supporting Materials and Methods.

Vector Constructs

pWPXL-RRM2B, pWPXL-Myr-Akt, pLVTHM-shRRM2B, pLVTHM-shAkt1, pLVTHM-shN-cadherin, and phosphatase and tensin homolog (PTEN) and RRM2B luciferase reporter gene vector were constructed in our laboratory. The primer information is listed in Supporting Table 2. For detailed protocol of vector constructs, see Supporting Materials and Methods.

Lentivirus Production and Cell Transduction

All recombinant lentivirus was produced by transient transfection of HEK-293T cells according to standard protocols. For detailed protocol of vector constructs, see Supporting Materials and Methods.

In Vitro Migration and Invasion Assays

For in vitro migration and invasion assays, cells were seeded on the upper chamber of a transwell or on a Matrigel-coated transwell (BD Biosciences, NJ) in serum-free media. The lower chamber contained Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS) as a chemoattractant. After 24 or 48 hours of incubation, the nonmigrated or noninvaded cells were gently removed from the upper chamber with a cotton swab. The cells were fixed and stained using Giemsa solution and counted in five randomly chosen visual fields.

In Vivo Metastasis Assays

For the in vivo metastasis assay, 2 × 106 cells were suspended in 40 μL of a mixture of serum-free DMEM/Matrigel (1:1 volume) (BD Biosciences, NJ) for each male nude mouse. Each nude mouse was orthotopically inoculated in the left hepatic lobe with a microsyringe through an 8-mm transverse incision in the upper abdomen under anesthesia. Five weeks later, all of the mice were killed and individual liver and lung tissues were removed and fixed with 4% phosphate-buffered neutral formalin for at least 72 hours. Metastatic tissues were analyzed with hematoxylin and eosin (H&E) staining.

Western Blotting

Cell lysates were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto nitrocellulose membranes. The membranes were incubated with primary antibody overnight at 4°C and subsequently probed with horseradish peroxidase (HRP)-conjugated secondary antibodies. The immunoreactive blots were visualized using an enhanced chemiluminescence reagent (Pierce, Rockford, IL). Antibody information is listed in Supporting Table 1.

RNA Isolation and Quantitative Reverse-Transcription Polymerase Chain Reaction (qRT-PCR)

Total RNA was extracted from cells and tissues using Trizol (Invitrogen, CA). RNA samples were reverse transcribed into complementary DNA (cDNA) using a PrimeScript RT reagent kit (Takara, Dalian, China). The transcripts were amplified from the cDNA with PCR. The primer information is listed in Supporting Table 2. Real-time PCR was performed with SYBR Premix Ex Taq (Takara).

Luciferase Assay

Cells were cotransfected with the correspondent reporter plasmid and the indicated plasmids in each experiment. To normalize for transfection efficiency, the cells were cotransfected with a pRL-TK reporter construct. The Dual-luciferase report assay (Promega) was performed after 48 hours in accordance with the manufacturer's instructions.

RNA Interference

The small interfering RNA (siRNA) we designed was chosen to target early growth response protein 1 (Egr-1) and was synthesized by GenePharma. Target sequences information is listed in Supporting Table 3.

Chromatin Immunoprecipitation Assay (ChIP)

ChIP assays were performed using a ChIP assay kit according to the manufacturer's instructions (Upstate Biotechnology). Rabbit anti-Egr-1 or rabbit IgG were used to immunoprecipitate DNA-containing complexes. DNA from these samples was subjected to PCR analyses with primers sets for RRM2B promoter: forward 5′ TCCCTTAGGCCGCAGTAGG-3′; reverse 5′CCTGCTTTCTCGCCTGTCC-3′.

Statistical Analysis

All of the data are shown as the mean ± SD. Statistical analysis was performed using the chi-square test. Two-group comparisons were performed with a t test. P < 0.05 was considered statistically significant.

Results

RRM2B Is Frequently Down-Regulated and Negatively Associated With Tumor Progression and Metastasis in HCC

We first evaluated and validated the specificity of RRM2B antibody in our original IHC detection (Supporting Fig. 1). Next we explored RRM2B expression in 236 cases of HCC tissues. RRM2B protein expression in 236 pairs of nontumor and tumor liver tissues of human HCC based on the results of IHC. Of the 236 pairs, 138 (58.5%) had higher RRM2 protein expression in nontumor tissues than in tumor liver tissues, 75 (31.8%) had similar expression, whereas only 23 (9.7%) had lower expression (Supporting Fig. 2A). Therefore, the results showed that RRM2B was frequently down-regulated in primary cancer tissues compared with adjacent noncancerous tissues (Fig. 1A). Besides, messenger RNA (mRNA) and protein levels of RRM2B were also down-regulated in HCC tissues compared with noncancerous tissues using qRT-PCR and western blot (Fig. 1B,C).

Figure 1.

RRM2B is often down-regulated in HCC and inhibits cell migration and invasion and metastasis. (A) IHC analysis of RRM2B and p53 expression in HCC tissues compared with paired nontumorous tissues (200×). Western blot (B) and qRT-PCR (C) analysis of RRM2B expression in HCC tissues compared with paired corresponding noncancerous liver tissues (n = 30) (T, tumor; N, noncancerous liver tissues). (D) The in vitro migration and invasion potentials of SMMC-7721 and HCC-LY10 cells transfected with RRM2B shRNA were detected and analyzed. *P < 0.05; **P < 0.01. (E) The in vitro migration and invasion potentials of SMMC-7721 and Huh7 cells stably transfected with RRM2B or control were detected and analyzed. The mean values from duplicate samples of three independent experiments are shown to the right. *P < 0.05; **P < 0.01. (F) Representative images of intrahepatic and lung metastatic nodules formed by SMMC-7721 and HCC-LY10 cells transfected with shRRM2B-1 or shNC are shown. Original magnification 200×. The numbers of intrahepatic and lung metastasis nodules are shown in the right panel. *P < 0.05.

According to the IHC results, the expression intensity of RRM2B protein was scored as 0, 1, and 2 for weak, moderate, and strong immunostaining, respectively (Supporting Fig. 2B). IHC criteria are shown in Supporting Materials and Methods. The results showed that RRM2B expression was negatively associated with histological grade of HCC and intrahepatic metastasis (Table 1). However, there was no correlation between RRM2B expression and other clinicopathological factors, including age, gender, tumor size, cirrhosis, or serum alpha-fetoprotein (AFP) and hepatitis B surface antigen (HBsAg) levels (Table 1).

Table 1. Relationship Between RRM2B Protein Expression and Clinicopathological Features in HCC Tissues
Clinicopathological FeaturesCasesRRM2B Expression (Cancer)
Score 0Score 1Score 2P Value
Cases (%)Cases (%)Cases (%)
  1. P value represents the probability from a chi-square test for different immunohistochemical scores of RRM2B in HCC tissues.

  2. a

    P < 0.05.

  3. b

    It means small HCC nodules such as venous permeation and tumor microsatellite nodules distributed around the primary (larger) HCC masses.

Age (years)     
<6015936(78.3)35(60.3)88(67.2)0.150
≥607610(21.7)23(39.7)43(32.8) 
Gender     
male19035(76.1)42(72.4)113(85.6)0.075
female4611(23.9)16(27.6)19(14.4) 
Tumor size     
≤5cm11325(55.6)28(48.3)60(47.6)0.647
>5cm11620(24.4)30(51.7)66(52.4) 
AFP     
≤207915(33.3)20(34.5)44(34.1)0.992
>2015330(66.7)38(65.4)85(65.9) 
HBV     
negative428(18.2)9(16.4)25(19.2)0.899
positive18736(81.8)46(83.6)105(80.8) 
Cirrhosis     
negative388(17.4)8(13.8)22(16.7)0.854
positive19838(82.6)50(86.2)110(83.3) 
Edmondson's grade     
I, II11916(34.8)36(62.1)67(50.8)0.022a
III, IV11730(65.2)22(37.9)65(49.2) 
Intrahepatic metastasisb     
negative16126(56.5)48(82.8)87(65.9)0.012a
positive7520(43.5)10(17.2)45(34.1) 

In addition, we also analyzed RRM2 expression in HCC. The results showed that RRM2 was up-regulated in HCC tissue compared with noncancerous tissue (Supporting Fig. 2C,D). However, there was no significant correlation between RRM2 expression and age, tumor size, cirrhosis, intrahepatic metastasis, serum, or HBsAg level, except gender and serum AFP level, in HCC (Supporting Table 4).

p53 Status Does Not Affect the Correlation Between RRM2B Expression and Intrahepatic Metastasis

Because RRM2B is a downstream target of wild-type p53, we also analyzed whether p53 expression affected the relationship between RRM2B and intrahepatic metastasis in HCC. We first detected the expression of p53 and RRM2B in HCC cell lines using western blotting. The results indicated that RRM2B expression was independent of the p53 status in HCC cell lines (Supporting Fig. 3). Wild-type p53 protein is unstable; thus, we assured that most positive IHC staining was due to mutant p53 rather than wild-type p53 protein. Direct sequencing of the p53 gene showed that the positive IHC staining of p53 was derived from mutant p53 (Supporting Fig. 4). Furthermore, there was no marked difference in the relationship between RRM2B expression and intrahepatic metastasis among HCC patients with or without the p53 mutation (Supporting Table 5). Thus, these results showed that p53 status did not affect the correlation between RRM2B expression and intrahepatic metastasis in HCC.

RRM2B Inhibits Cell Migration and Invasion In Vitro and Suppresses Intrahepatic and Lung Metastasis In Vivo

To further confirm the role of RRM2B in metastasis, we used the shRNA technique to generate stable RRM2B knockdown HCC cell lines. We selected SMMC-7721, Huh7, MHCC-97L, and HCC-LY10 cell lines for the establishment of in vitro migration and metastasis model according to RRM2B expression and p53 status and metastatic capacity of cells (Supporting Fig. 3). Our results showed that RRM2B knockdown increased cell migration and invasion compared with control cells. Conversely, RRM2B overexpression inhibited cell migration and invasion (Fig. 1D,E; Supporting Fig. 5). SMMC-7721 is a wild-type p53 cell line, while MHCC-97L and Huh7 are mutant p53 cell lines.[17-19] Therefore, these results demonstrated that the effects of RRM2B on migration and invasion were not related to p53 status. In addition, consistent with previous reports,[20] we found that the overexpression of RRM2B inhibited HCC cell growth (Supporting Fig. 6).

To further clarify the role of RRM2B in HCC metastasis in vivo, RRM2B knockdown SMMC-7721 and HCC-LY10 cells were orthotopically inoculated into the left hepatic lobe of mice with a microsyringe. Histological examination of the lung and liver tissues indicated that RRM2B knockdown tumor-bearing mice had significantly higher numbers of lung and intrahepatic metastatic nodules than mice bearing control tumors (P < 0.05, Fig. 1F). Therefore, these results indicated that RRM2B suppresses HCC intrahepatic and lung metastasis.

RRM2B Negatively Regulates EMT in HCC Cells

It has been found that EMT exists in a variety of malignant tumors of epithelium.[21] Our results showed RRM2B knockdown in SMMC-7721 and HCC-LY10 cells partially resulted in morphologic changes from tightly packed colonies to scattered growth structures, suggesting the induction of EMT (Fig. 2A). E-cadherin is a cell-cell adhesion molecule that is a key mediator of EMT in tumor progression.[21] Our results showed that the down-regulation of E-cadherin, and the concomitant up-regulation expression of N-cadherin and the EMT-related transcription factor slug, was observed in RRM2B knockdown cells (Fig. 2B). In contrast, the ectopic expression of RRM2B increased E-cadherin expression and decreased N-cadherin and slug expression in HCC cells (Fig. 2C; Supporting Fig. 7). In addition, RRM2B knockdown tumors exhibited an EMT phenotype, including the focal loss of E-cadherin expression and gain of N-cadherin expression (Fig. 2D).

Figure 2.

RRM2B negatively regulates the EMT of HCC cells. (A) Phase-contrast images (200×) show morphological changes in RRM2B knockdown SMMC-7721 and HCC-LY10 cells and control cells. (B) The expression of RRM2B, E-cadherin, N-cadherin, and slug was detected by western blotting in SMMC-7721 and HCC-LY10 cells stably transfected with RRM2B shRNA. (C) The expression of RRM2B, E-cadherin, N-cadherin, and slug was detected by western blotting in SMMC-7721 and Huh7 cells stably transfected with RRM2B or control vector. (D) The expression of RRM2B, E-cadherin, and N-cadherin was detected by IHC in xenograft tumor tissues from SMMC-7721-shRRM2B-1 and SMMC-7721-shNC cells. (E) Overexpression or knockdown RRM2B of SMMC-7721 and Huh7 cells were treated with doxorubicin (2 μg/mL) for 48 hours and the growth inhibition rates of cells were measured by the 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) method. *P < 0.05.

Cell death resistance is a key feature of EMT.[21] We next detected whether RRM2B affect the sensitivity of cell to chemotherapeutic drugs. The results showed that overexpression of RRM2B increased doxorubicin sensitivity in HCC cell lines. Conversely, RRM2B knockdown showed a significant decrease in doxorubicin sensitivity in SMMC-7721 cells (Fig. 2E). EMT not only induces cellular motility, but also confers stemness properties to cancer cells.[21] Therefore, we determined the effect of overexpression or knockdown of RRM2B on the marker of potential cancer stem cell (CSC). The results showed that RRM2B inhibited CD44 and CD133 mRNA and protein expression in HCC cells (Supporting Fig. 8). Thus, these data support the view that RRM2B negatively regulates EMT in HCC cells.

RRM2B Inhibits the Migration and Invasion of HCC Cells by Way of Regulating Egr-1/PTEN/Akt1 Activity

The PI3K/Akt signaling pathway plays an important role in EMT.[22] We therefore determined the role of the PI3K/Akt pathway in RRM2B-induced inhibition of cell migration and invasion. The results showed that RRM2B overexpression inhibited Akt1 phosphorylation in HCC cells. In contrast, RRM2B knockdown increased p-Akt1 levels in HCC cells (Fig. 3A,B). PTEN inhibits the PI3K/Akt pathway: when PTEN is up-regulated, the downstream target of PI3K, namely, the Akt/PKB kinase, should be inactivated. Therefore, we also detected the expression of PTEN in RRM2B overexpression and knockdown HCC cells. The results showed that RRM2B overexpression increased PTEN expression in SMMC-7721 and Huh7 cells. In contrast, RRM2B knockdown resulted in decreased PTEN expression levels in HCC cells (Fig. 3A-C). To test whether RRM2B may affect PTEN promoter activity, we performed luciferase assays with minimal PTEN promoter (-1123/-779) containing the Egr-1 binding sequence.[23] The results showed that up-regulation of PTEN by RRM2B was achieved through the increase of PTEN promoter activity (Fig. 3D). In accordance with previous reports, Egr-1 is one of the first transcription factors shown to directly regulate the expression of PTEN in living cells.[23] Therefore, we detected Egr-1 expression in RRM2B overexpression and knockdown cells. As shown in Fig. 3E, RRM2B overexpression increased Egr-1 mRNA expression. In contrast, RRM2B knockdown decreased Egr-1 mRNA expression in HCC cells. Furthermore, ectopic expression of Egr-1 increased PTEN promoter activity in SMMC-7721 and Huh7 cells (Fig. 3F). Therefore, these results suggested that RRM2B could inhibit cell migration and invasion through regulating the Egr-1/PTEN/Akt1 pathway.

Figure 3.

RRM2B inhibits phosphorylated Akt1 through up-regulation of Egr-1/PTEN expression in HCC cells. Akt1, phosphorylated Akt1 (p-Akt1), PTEN, and RRM2B expression were detected by western blotting in RRM2B-overexpressing SMMC-7721 and Huh7 cells (A) or RRM2B knockdown SMMC-7721 and HCC-LY10 cells (B). (C) PTEN mRNA expression in RRM2B-overexpressing and knockdown SMMC-7721, Huh7 and HCC-LY10 cells were detected using qRT-PCR. *P < 0.05; **P < 0.01. (D) SMMC-7721 and Huh7 cells were transfected with either pWPXL or RRM2B along with PTEN luciferase report vector. Corresponding relative luciferase activities were determined by reporter gene assay. **P < 0.01. (E) Egr-1 mRNA expression in RRM2B-overexpressing SMMC-7721 and Huh7 or knockdown SMMC-7721 and HCC-LY10 cells were detected using qRT-PCR. *P < 0.05; **P < 0.01. (F) SMMC-7721 and Huh7 cells were transfected with either pCDNA3.1 or pCNDA3.1-Egr-1 along with PTEN luciferase report vector. Corresponding relative luciferase activities were determined by reporter gene assay. **P < 0.01.

To confirm the results, we constructed Akt1 shRNA to knockdown Akt1 expression or used PI3K/Akt inhibitors (LY-294002 and wortmannin) in RRM2B knockdown SMMC-7721 cells. The results showed that silencing Akt1 or suppressing Akt1 phosphorylation inhibited RRM2B knockdown-induced migration and invasion in SMMC-7721 cells, and the concomitant up-regulation of E-cadherin and down-regulation of N-cadherin expression (Fig. 4A-D). In addition, LY-294002 and wortmannin also suppressed the migration and invasion through inhibiting Akt1 phosphorylation in SMMC-7721 and HCC-LY10 cells (Supporting Fig. 9). Moreover, we also tested the effect of expressing a constitutively active Akt (Myr-Akt) on the ability of cell migration and invasion in RRM2B overexpression SMMC-7721 cells. The results showed that RRM2B overexpression-induced suppression of cell migration and invasion can be reversed by overexpressing active Akt in SMMC-7721 cells (Fig. 4E,F). Therefore, these results suggested that RRM2B inhibited cell migration and invasion through the suppression of Akt1 phosphorylation.

Figure 4.

Silencing Akt1 or treatment with LY294002 or wortmannin inhibits RRM2B knockdown-mediated cell migration and invasion in HCC cells. (A) SMMC-7721-shRRM2B-1 cells were transfected with Akt1 shRNA as indicated, the expression of Akt1, E-cadherin, and N-cadherin was detected by western blotting. (B) The expression of Akt1, p-Akt, N-cadherin, and E-cadherin was measured using western blot in RRM2B knockdown SMMC-7721 and HCC-LY10 cells treated with 10 μM LY294002 (LY), 1 μM wortmannin (Wort), or dimethyl sulfoxide (DMSO) only for 12 hours. (C) SMMC-7721-shRRM2B-1 cells were treated with Akt1 shRNA, and migration and invasion were measured using transwell assays. **P < 0.01. (D) SMMC-7721-shRRM2B-1 cells were treated with 10 μM LY, 1 μM Wort, or DMSO, and migration and invasion of cells were measured using transwell assays. **P < 0.01. (E) SMMC-7721-RRM2B cells were transfected with Myr-Akt1 as indicated, expression of p-Akt1, Akt1, and RRM2B was detected by western blotting. (F) SMMC-7721-RRM2B cells were transfected with Myr-Akt1 as indicated, and migration and invasion of cells were measured using transwell assays. The data are presented as the mean ± SD (from five random 200× magnification fields). *P < 0.05; **P < 0.01.

In addition, RRM2B overexpression-induced suppression of cell migration and invasion can be partially reversed by knockdown Egr-1 (Fig. 5A,B). On the contrary, RRM2B knockdown-induced cell migration and invasion can be partially reversed by overexpressing Egr-1 plasmid (Fig. 5C,D). Therefore, we considered that RRM2B suppressed cell migration and invasion through up-regulation of Egr-1 in HCC cells.

Figure 5.

RRM2B inhibits cell migration and invasion through increasing Egr-1 expression in HCC cells. (A) SMMC-7721-shRRM2B cells were transfected with Egr-1 as indicated, the expression of Egr-1 and RRM2B was detected by western blotting. (B) SMMC-7721-shRRM2B cells were transfected with Egr-1 as indicated, migration and invasion were measured using transwell assays. *P < 0.05. (C) RRM2B overexpression SMMC-7721 cells were transfected with Egr-1 siRNA as indicated, the expression of Egr-1 and RRM2B was detected by western blotting. (D) RRM2B overexpression SMMC-7721 cells were transfected with Egr-1 siRNA as indicated, migration and invasion were measured using transwell assays. The data are presented as the mean ± SD (from five random 200× magnification fields). *P < 0.05.

Besides, we also found that silencing N-cadherin did not reduce Akt1 or RRM2B expression but suppressed the migration and invasion of RRM2B knockdown SMMC-7721 cells (Supporting Fig. 10). Collectively, we considered that RRM2B negatively inhibits EMT through regulating Egr-1/PTEN/Akt1 pathway, affecting HCC cell behavior such as migration, invasion, and spreading.

E-cadherin Expression Is Down-Regulated and Correlates With RRM2B Expression in HCC Tissues

To further explore potential clinical applications of the experimental data, we analyzed E-cadherin and N-cadherin expression in HCC tissue by qRT-PCR. The results showed that E-cadherin was down-regulated and N-cadherin was up-regulated in HCC tissues compared with adjacent noncancerous tissues (Fig. 6A,B). Next we detected the expression of RRM2B, p-Akt1, E-cadherin, and N-cadherin in human primary HCC tissues with or without metastasis by western blot. RRM2B and E-cadherin expression was reduced in HCC tissues with metastasis compared to those without metastasis. However, the expression of p-Akt1 and N-cadherin was increased in HCC tissues with metastasis (Fig. 6C). Previous study showed that E-cadherin is involved in the intrahepatic metastasis and invasion of HCC.[24, 25] Therefore, we analyzed the association between E-cadherin and RRM2B expression in HCC. The results showed that there was a positive correlation between the expression levels of RRM2B and E-cadherin in HCC tissues (r = 0.350, P = 0.000; Fig. 6D,E).

Figure 6.

E-cadherin expression is down-regulated and positively correlates with RRM2B expression in HCC tissues. E-cadherin (A) and N-cadherin (B) mRNA expression in 30 cases of HCC and matched noncancerous liver tissues (N, noncancerous liver tissues; T, HCC tissues) was examined using qRT-PCR. (C) The expression of E-cadherin, N-cadherin, Akt1, p-Akt, and RRM2B was detected by western blotting in representative HCC tissues samples with and without metastasis. (D) Representative immunostaining of E-cadherin and RRM2B in HCC tissues (400×). (E) The correlation between RRM2B expression and E-cadherin level in 236 HCC tissues was analyzed. r = 0.350; P = 0.000.

Egr-1 Binds to the RRM2B Promoter and Represses Its Expression in HCC Cells

To examine the regulation of RRM2B in HCC, we carried out a reporter assay to identify promoter elements that control RRM2B transcription. The results showed that no differences in RRM2B promoter activity were observed in SMMC-7721 and Huh7 cells (Supporting Fig. 11A). These results imply that p53 status does not affect the activity of RRM2B promoter in HCC cells. In addition, the deletion construct showed −896/−5, −763/−5 and −375/−5 luciferase activity was decreased compared with −234/−5 (Fig. 7A). These results suggested the presence of negative regulatory elements located between nucleotides −375 and −234 bp region.

Figure 7.

Egr-1 binds to the RRM2B promoter and represses its expression. (A) SMMC-7721 cells were transfected with different truncations of RRM2B luciferase reporter vectors (−1181/−5, −1025/−5, −896/−5, −763/−5, −375/−5, −234/−5, or −163/−5). Corresponding relative luciferase activities were determined by reporter gene assay. (B) SMMC-7721 cells were cotransfected with RRM2B luciferase reporter vectors (−1181/−5, −896/−5, −375/−5, −234/−5) and pcDNA3.1 or pcDNA3.1-Egr-1. Corresponding relative luciferase activities were determined by reporter gene assay. (C) Different doses of plasmid pcDNA3.1-Egr-1 and RRM2B luciferase reporter vectors (wild-type or mutant Egr-1 binding sites, −375/−5) were cotransfected into SMMC-7721 cells. Corresponding relative luciferase activities were determined by reporter gene assay. *P < 0.05. (D) ChIP analysis of Egr-1 binding to the RRM2B promoter. SMMC-7721 and Huh7 cells were crosslinked to DNA, and chromatin-protein complexes were immunoprecipitated with an antibody against Egr-1. SMMC-7721 and Huh7 cells were transfected with either pCNDA3.1-Egr-1 (E) or Egr-1 siRNA (F) as indicated, the expression of Egr-1 and RRM2B was detected by western blotting.

We predicted putative regulatory elements such as heat shock factor (HSF), GATA, and nuclear factor kappa B (NF-κB) between nucleotides −375 and −234 bp region. The results showed that HSF, GATA, NF-κB, vitamin D receptor (VDR), and C/EBPb did not regulate RRM2B promoter activity and protein expression by luciferase assay and western blotting (Supporting Fig. 11B,C). Interestingly, a conserved Egr-1 binding motif at position −254 to −242 bp was found, indicating that Egr-1 could be implicated in the regulation of RRM2B transcription. Luciferase assays revealed that Egr-1 inhibited RRM2B promoter activity apart from RRM2B-Luc (−234/-5) (Fig. 7B). Therefore, the Egr-1 responsive element could be located between the −375 and −234 region. In addition, Egr-1 inhibited RRM2B promoter activity in a dose-dependent manner. However, Egr-1 did not inhibit RRM2B promoter activity containing a putative Egr-1-mutated binding site (Fig. 7C, Supporting Fig. 11D). The binding of Egr-1 to the promoter of RRM2B was further confirmed by ChIP assay (Fig. 7D). In addition, Egr-1 overexpression decreased the RRM2B protein level. In contrast, Egr-1 knockdown by siRNA resulted in increased RRM2B protein levels (Fig. 7E,F). As we demonstrated above that RRM2B induced Egr-1 expression, this result suggests a negative feedback regulatory network between RRM2B and Egr-1 expression in HCC.

Discussion

In the present study, we demonstrated that RRM2B was commonly down-regulated in HCC and inhibited cell migration, invasion, and metastasis in vitro and in vivo. Furthermore, RRM2B inhibited cell migration and spreading by way of modulation of the Egr-1/PTEN/Akt1 pathway in HCC. Interestingly, we also found that Egr-1 was, in turn, involved in the suppression of RRM2B, pointing to the existence of a negative feedback loop in HCC.

It is well known that p53 is frequently mutated in various tumor tissues, including HCC.[26] When mutated, p53 may become a very stable protein, resulting in an increase in overall p53 expression, but mutant p53 lacks the ability to transactivate wild-type p53 target genes.[27] Thus, RRM2B, which is a downstream target of p53, was not up-regulated in HCC. In this study, we also found that RRM2B inhibited cell migration, invasion, and metastasis in vitro and in vivo, which was observed in both p53 wild-type and p53 mutant HCC cells. In addition, no differences in RRM2B promoter activity were observed in either p53 wild-type or p53 mutant HCC cells. The IHC results also showed that there was no marked difference in the relationship between RRM2B expression and intrahepatic metastasis among HCC patients with or without the p53 mutation. Studies of others also indicated that basal transcription of RRM2B was expressed in almost all normal human tissues and regulated through the p53-independent mechanism.[7, 28] Therefore, all these data showed that RRM2B inhibited HCC cell migration, invasion, and metastasis through the p53-independent mechanism.

Recently, several studies demonstrated that RRM2B expression was closed with metastasis of different tumors. The biological characteristics of RRM2B in HCC cells and underlying mechanisms remain largely unknown. In this study, we found that Egr-1/PTEN/Akt1 is a key downstream signaling effector of RRM2B and contributes to the inhibition of HCC metastasis. Egr-1 is one of the first transcription factors shown to directly regulate the expression of PTEN and the first shown to regulate PTEN in living cells.[23] The proximal promoter of PTEN is GC-rich and contains one functional Egr-1 binding site.[23] Our results also show that RRM2B up-regulates Egr-1 expression in HCC cells. Hence, RRM2B-induced Egr-1 expression leads to up-regulation of PTEN function. Furthermore, PTEN has been reported to inhibit cell migration and invasion.[29, 30] Therefore, RRM2B promoted Egr-1 expression, which is the main transcription factor of PTEN, and this resulted in up-regulation of PTEN expression, which consequently inhibits cell migration and invasion. Interestingly, we also found that Egr-1 could bind to the RRM2B promoter and inhibited RRM2B expression in HCC. Therefore, we speculate the possible existence of a negative feedback loop between the RRM2B and Egr-1 in HCC. This feedback loop may regulate RRM2B expression in HCC.

An increased ability to migrate, a higher resistance to apoptosis, and the acquisition of stemness properties are important characterizations of EMT. Slug (SNAI2), belonging to the zinc-finger transcription factors, represses E-cadherin expression and is an essential mediator of Twist1-induced EMT and metastasis.[31, 32] Overexpressing the major EMT regulator slug in HCC showed that an EMT phenotype was induced.[33] In this study, we demonstrated that RRM2B knockdown had a significant impact on EMT, as shown by the increased expression of slug and N-cadherin and the decreased expression of E-cadherin. Furthermore, RRM2B overexpression increased doxorubicin sensitivity and inhibited HCC cancer stem cells marker CD133 and CD44. Therefore, these data conform that RRM2B negatively regulates EMT in HCC. The PI3K/Akt pathway has been shown to play a central role in a variety of oncogenic processes, including cell growth, proliferation, motility, EMT, angiogenesis, and metastasis.[22, 34, 35] Our results showed that RRM2B knockdown activates Akt1 phosphorylation in HCC. Activated Akt has been shown to play a critical role in promoting EMT, in part through the down-regulation of E-cadherin.[21, 22] In addition, previous reports showed that Akt1 can inhibit slug expression.[36] Our results also showed that RRM2B induced E-cadherin expression and the concomitant down-regulation of slug and N-cadherin. Furthermore, there was a positive correlation between the expression levels of RRM2B and E-cadherin in HCC tissues. Therefore, these results suggest that RRM2B inhibited EMT through suppressing Akt1 phosphorylation, which may be involved in inhibiting metastasis of HCC. Kimura et al.[37] reported that RRM2B knockout mouse embryonic fibroblasts (MEFs) reduced ribonucleotide reductase enzyme activity and DNA repair after exposure to various genotoxins. Furthermore, recent studies showed that activated Akt1 had negative effects on the DNA damage response.[38, 39] In addition, RRM2B inhibited the proliferation of human cancer cells in association with cell cycle arrest.[20] Therefore, we speculate that down-regulated RRM2B decreases ribonucleotide reductase enzyme activity, ultimately inhibiting Akt1 phosphorylation in HCC.

In conclusion, our findings demonstrate that RRM2B suppresses HCC cell migration and invasion through regulating the expression of E-cadherin and N-cadherin by way of Egr-1/PTEN/Akt1 signaling.

Ancillary