zMicroRNAs (miRNAs) are endogenously expressed small noncoding RNAs that regulate approximately one-third of human genes at post-transcription level. Previous studies have shown that miRNAs were implicated in many cellular processes and participated in the progress of various tumors including hepatocellular carcinoma (HCC). Among all miRNAs, the let-7 family is well recognized to play pivotal roles in tumorigenesis by functioning as potential growth suppressor. In the present study, we aimed to investigate the role of let-7 family, particularly the hsa-let-7g, in the molecular pathogenesis of HCC. By use of MTT, qPCR, Western blotting and 2-dimensional electrophoresis (2-DE), over-expression of hsa-let-7g was found to inhibit the proliferation of HCC cell line via negative and positive regulations of c-Myc and p16INK4A, respectively. The expression of hsa-let-7g was noted to be markedly lowered in the HepG2, Hep3B and Huh7 cells, yet higher in the Bel-7404 HCC cell line. Proliferation of HCC cell line was significantly inhibited after the transfection of hsa-let-7g mimics, while hsa-let-7g inhibitor transfection exerted an opposite effect. Concurrently, the mRNA and protein levels of c-Myc were found significantly decreased in HepG2 cells after transfection of hsa-let-7g mimics, but obviously increased in Bel-7404 cells after transfection of hsa-let-7g inhibitor. As revealed by 2-DE, a significant upregulation of p16INK4A was revealed after the gain-of-function study using hsa-let-7g. Therefore, we suggest that hsa-let-7g may act as a tumor suppressor gene that inhibits HCC cell proliferation by downregulating the oncogene, c-Myc, and upregulating the tumor suppressor gene, p16INK4A.
Hepatocellular carcinoma (HCC) is one of the most common malignancies worldwide that accounts for about 90% of all primary liver cancers.1 It is a highly invasive and destructive tumor with its incidence increased exponentially all over the world in the last decade.2
MicroRNAs (miRNAs) are a class of single-strand and highly conserved noncoding small RNAs that regulate gene expression. They induce translational repression or mRNA degradation by binding to the 3′-untranslated regions (3′UTR) of the target mRNAs.3, 4 They exist broadly in eukaryote and have also been reported in some encoded sequences of viral genomes (http://microrna.sanger.ac.uk/sequences/). To date, there are over 80 species (i.e., ∼1,000 different mammalian miRNA genes) being identified.4, 5 It is generally believed that miRNAs have multibiological functions including regulation of cellular activities related to development, differentiation, inflammatory response, metabolism, proliferation, apoptosis, viral infection and tumorigenesis.5–13 Studies have demonstrated that miRNA expression is frequently aberrant in human malignancies; therefore, they have been suggested as potential oncogenes or tumor suppressors.11, 14–19 Expression of miRNA is commonly applied in diagnosing, staging, prognosis and evaluation of tumor treatment.20–22
In recent studies, miRNAs were noted to participate in hepatocarcinogenesis as revealed by microarray analysis.23, 24 Aberrant expressions of some miRNAs such as miR-17-92 cluster and miR-21 have been found to play crucial roles in HCC.25, 26 Among all human cancer-related miRNAs, let-7 family raised most interests since its family members have been noted to express aberrantly in human cancers.27–31 The family was discovered initially in C. elegans and is currently one of the most important members of the miRNA family.27 The family is highly conserved in vertebrates and other species, consisting of 11 closely related genes with a “seed sequence.”32 Their expression is usually noted in tissues during embryonic stages, but significantly increases toward maturity.33 This phenomenon suggests that the let-7 family plays a pivotal role in developmental process. In fact, their functions in cell differentiation have been proven in both in vitro and in vivo conditions.34 Besides, let-7 family has also been reported as tumor suppressors for they could directly repress oncogenes including RAS and HMGA2via binding to the 3′UTR of their mRNAs, thus inhibited cancer cell proliferation.27, 28, 35, 36 Moreover, many other oncogenes and tumor suppressor genes including CCND2, CDK6, Cdc25A37–39 and NF240 have also been identified as targets of the let-7 family. In HCC patient samples, an approximate 69% downregulation of the let-7 family genes has also been reported.41 However, the exact role of let-7 family in HCC tumorigenesis and development is yet to be elucidated.
In the present study, we aim to investigate the role of let-7 family in the molecular pathogenesis of HCC. Our focus was on the let-7 family member, hsa-let-7g, for it has been reported to be significantly downregulated in some of HCC clinical samples.23, 24, 42, 43 In fact, this gene located in the chromosome 3 (3p21.1) from 52277334 to 52277417 as reverse strand (Fig. 1) and was initially identified in mouse44 and later in human.45 In this study, the mechanism on how the hsa-let-7g might regulate the HCC cell proliferation was examined and discussed.
Human HCC cell lines including HepG2, Bel-7404, Hep3B and Huh7, and normal liver cell line, LO2, were grown in 90% Dulbecco's Modified Eagle Media (DMEM) supplemented with 10% fetal bovine serum (Invitrogen, CA), 100 U/mL penicillin and 100 U/mL streptomycin (GIBCO, NY). The cells were incubated at 37°C in a humidified atmosphere with 5% CO2. Human miRNA mimics, mimics-negative control, inhibitor and inhibitor-negative control were synthesized by Shanghai GenePharma Co. (Shanghai, China). Hsa-let-7g was designed according to the miRBase sequence database (http://microrna.sanger.ac.uk/). HCC cells were seeded onto 24-well plate at a concentration of 5 × 104 cells/well the day before transfection. Cells were transfected with miRNA mimics or inhibitor and their respective negative control duplexes using Lipofectamine 2000 (Invitrogen Corp., CA) according to the manufacturer's instruction. The medium was replaced 6 hr after transfection. DMEM and trypsin-EDTA were products of GIBCO.
MTT analysis of cell proliferation
To measure the effects of miRNA on the tumor cell proliferation, HepG2 and Bel-7404 cells were seeded at a density of 5 × 103 cells per well onto a 96-well plate with the complete medium (100 μL/well). After 24 hr of incubation, cells were transfected with miRNA mimics or inhibitor and their respective negative-control duplexes using Lipofectamine 2000. The medium was removed after 6 hr of transfection and the adhered cells were washed once with D-hank's solution (GIBCO). The cells were then incubated at 37°C in a humidified environment with 5% CO2. At Day 1, 2, 3, 4 and 5, supernatants were removed and the cells were incubated with 20 μL of 5 mg/mL MTT reagent (Sigma-Alrich, Deutschland) at 37°C for 4 hr. At the end of incubation, MTT reagent was removed from each well and replaced by 100 μL DMSO (Invitrogen). The plate was gently agitated for 15 sec and the absorbance (A) at 490 nm was determined by an ELISA reader (Bio-Rad, USA). Each measurement was done in quintuple, and the data are expressed as mean ± SEM.
Quantitative real-time PCR (qPCR)
Cells were collected at Day 0–5 after transfection. RNA was extracted using TRIZOL reagent (Invitrogen) according to the manufacturer's instruction. Quality and concentration of the RNA were determined by spectrophotometer, ND-1000 NANODROP (Thermo-scientific, Courtaboeuf, France). The RNA sample was used as the substrate for reverse-transcriptase reactions to generate single-stranded cDNA with SuperScript™ III Reverse Transcriptase (Promega, USA). Briefly, 100 ng of each RNA sample was mixed with reaction mixture consisting 5× RNX buffer (4 μL), Random Primer (1 μL), ImProm-II™ reverse transcriptase (1 μL), MgCl2 (2 μL), 0.5 mM dNTPs (1 μL), 20 u RNasin Ribonuclease Inhibitor (0.5 μL) and nuclease-free water (8.5 μL). They were incubated at 42°C for 45 min and then 72°C for 15 min. Thermocycling was carried out using a gradient thermocycler (Takara, Japan). For miRNA determination, all miRNAs were first polyadenlyated at the 3′ end and cDNAs were amplified using NCode™ miRNA First-Strand cDNA Synthesis and qPCR Kits (Invitrogen) according to the manufacturer's protocol. The universal qPCR primer for miRNA analysis was provided in the kit and a forward primer targeting the specific miRNA sequence of interest was designed based on the online miRBase sequence database (http://microrna.sanger.ac.uk/sequence/). Quantitative PCR was carried out using Applied Biosystems 7500 Real-Time PCR System (Applied Biosystems, USA) and SYBR GreenER™ qPCR SuperMix Universal Kits (Applied Biosystems). A total reaction volume of 20-μL sample mixture consisting 2× PCR SYBR® Green Mix, PCR primers for each target gene, nuclease-free water and cDNA template was used. PCR reactions were carried out as follows: 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min. For miRNA determination, U6 gene was served as the internal control whilst human GAPDH was used as the internal control for mRNA determination of c-Myc, Bmi-1 and p16. The primer sequences for amplifying GAPDH, Bmi-1, p16 and c-Myc genes were designed using “Primer Premier 5” based on the mRNA sequences obtained from Genbank: GAPDH forward: 5′-TGC CTC CTG CAC CAC CAA CT-3′ and reverse: 5′-CCC GTT CAG CTC AGG GAT GA-3′; c-Myc forward: 5′-GCT GCC AAG AGG GTC A-3′ and reverse: 5′-CGC ACA AGA GTT CCG TAG-3′. Bmi-1 forward: 5′-TGC CAC AAC CAT AAT AG-3′ and reverse: 5′-GGA TGA GCT GCA TAA AA-3′; p16 forward: 5′-CTG CCC AAC GCA CCG AAT A-3′ and reverse: 5′-TGC AGC ACC ACC AGC GTG TC-3′. PCR was performed at their respective annealing temperatures and each experiment was repeated in triplicate. Results were normalized with respective internal controls. Ct-value for each sample was calculated with the ΔΔCt-method41 and results were expressed as 2−ΔΔCT.
The protein expression level of c-Myc, Bmi-1, cyclin-dependent kinase 4 inhibitor A (p16INK4A) and β-actin was determined by Western blotting. Cells were seeded onto a 6-well plate at 106 cells/well. Forty-eight hours after transfection, cells were scraped off and kept on ice for 30 min with gentle agitation after the addition of RAPI lysis buffer (100 μL; Promega). Proteins were extracted in extraction reagent containing 150 mM NaCl, 25 mM Tris-HCl (pH 8.0), 0.5 M EDTA, 20% Triton X-100, 8 M Urea and 1× protease inhibitor. After centrifugation at 10,000 rpm for 10 min, protein concentration of the lysates was determined by Bio-Rad protein assay kit (Bio-Rad). Lysates with 10–30 μg protein of each sample were subjected to tris-glycine SDS-PAGE. The separated proteins were transferred to PVDF membranes followed by blocking in 5% nonfat dry milk for 1 hr. Incubation of primary monoclonal antibody against c-myc (1:1,000; Calbiochem, USA), Bmi-1 (1:500; R&D Systems, USA), p16INK4A (1:1,000; Santa Cruz, CA) and β-actin (1:3,000; Abcam, USA) were carried out overnight at 4°C. After washing, the PVDF membrane was incubated with appropriate dilution of HRP-conjugated secondary antibody (1:3,000; Abcam) for 1 hr at room temperature. Results were visualized using the enhanced chemiluminescence substrate kit (Amersham Biosciences, USA) and photo-documented. The expression level of β-actin was also determined and served as an internal control.
Reporter vector construction and luciferase assay
Luciferase reporter vector psi-CHECK (Promega) was used to generate luciferase reporter constructs. MYC-3′UTR and N-ras-3′UTR reporter assays were performed in the HepG2 cell line. Fragments of human MYC-3′UTR and N-ras-3′UTR were cloned with the primers of MYC-3′UTR-WT and N-ras-3′UTR-WT, respectively, by RT-PCR. Psi-CHECK report vector harboring MYC-3′UTR (nt 1890-2356) sequences with wild type (WT) or mutated (MUT) and N-ras-3′UTR (nt 892-1353) sequence with WT hsa-let-7g binding sites were generated by cloning the following oligonucleotides into the Xho I and Not I restriction sites of psi-CHECK: MYC-3′UTR-WT: forward: 5′-TGTGCTCGAGGAAAAGTAAGGAAAACGA-3′ and reverse: 5′-GTTTTGCGGCCGCGGCTCAATGATATATTTGC-3′; MYC-3′UTR-MUT: forward: 5′-TAGCCATAATGTAATGATGGAGTAATTGGACTTTGGG-3′ and reverse: 5′-CCCAAAGTCCAATTACTCCATCATTACATTATGGCTA-3′. N-ras-3′UTR-WT: forward: 5′-TGGTCTCGAGTTCCCTGGAGGAGAAGTA-3′ and reverse: 5′-TGCCGCGGCCGCGAATAGAGCCGATAACAT-3′. MYC-3′UTR mutated construct was generated by MYC-3′UTR-MUT and MYC-3′UTR-WT primers using a 2-step PCR arrangement. Two pairs of primers including the MYC-3′UTR-WT-forward and MYC-3′UTR-MUT-reverse, MYC-3′UTR-WT-reverse and MYC-3′UTR-MUT-forward, were used in the first PCR step. Mutated fragment of MYC-3′UTR was then generated by the second PCR using MYC-3′UTR-WT primers whilst the templates were the products of the first PCR step. The final PCR product was digested with XhoI and NotI, and then inserted into the psi-CHECK vector. HepG2 cells were seeded onto 24-well tissue culture plates overnight and were transfected with 100 nM hsa-let-7g mimics or mimics negative control, and 1 μg psi-CHECK vectors containing WT or MUT hsa-let-7g binding sites using Lipofectamine 2000 (Invitrogen). Six hours after transfection, the transfection reagent was replaced with DMEM supplemented with 10% fetal bovine serum culture medium while the cultures maintained at 37°C for 48 hr. Luciferase activity assays were performed using the Luciferase Assay System (Promega). Luciferase activity was measured 48 hr post-transfection by the Dual-Luciferase® Reporter Assay System (Promega). Cells were washed using PBS and lysed with 1× PLB lysis buffer (100 μL), which were originally diluted from the luciferase 5× PLB lysis buffer on ice 30 min in advance. After centrifugation at 10,000 rpm for 30 sec, supernatant was collected and used to measure firefly and Renilla Luciferases activity on a 96-well plate in sextuple using TopCount NXT (i.e., a microplate scintillation and luminescence counter manufactured by Packard Instrument Company, Meriden, CT). For the luciferase activity assay, 20 μL/well of the supernatant was mixed with 100 μL/well of luciferase assay substrate LARII, whilst the measurement of the firefly luciferase activity immediately followed. The sample was then mixed continuously with 100 mL/well of Stop & Glo® Reagent (Promega) for the measurement of the Renilly Luciferase activity. The experiments were repeated thrice. Finally, Renilly Luciferase activities were normalized against firefly luciferase activity.
Two-dimensional electrophoresis (2-DE)
For 2-DE detection, HepG2 cells were seeded at a density of 1 × 106 cells onto a 10-cm plate at 24 hr before transfection. The cells were then treated with 50 nM has-let-7g mimics and mimics negative control for 48 hr. Cells were collected and protein was extracted. Total protein concentration was measured by the Bradford assay (Bio-Rad). Isoelectric focusing (IEF) was carried out using IPGphor II apparatus (Ammersham). Sample protein were diluted in 250-μL rehydration solution (8 M Urea, 2% CHAPS, 0.4% DTT, 0.5% IPG buffer, 0.002% bromophenol blue) and loaded onto the IPG strips (13 cm, pH 3–10, NL). Rehydration was performed at 30 V for 10 hr. IEF was performed using a stepwise voltage ramp: 500 V for 1 hr, 1,000 V for 1 hr and finally 8,000 V for 6 hr. The IPG strips were then incubated in the equilibration buffer (6 M urea, 2% SDS, 30% glycerol, 0.002% bromophenol blue, 50 mM Tris-HCl, pH 6.8) containing 1% DTT for 15 min with gentle agitation. They were mixed with the second equilibrating solution containing 2.5% iodoacetamide and gently agitated for 15 min. The strips were placed on top of a 12.5% uniform SDS-PAGE gel (150 × 158 × 1.5 mm) and the second dimensional separation was carried out using a constant current (15 mA/gel; 30 min), and an increased current (30 mA/gel) thereafter.
Protein visualization and image analysis
The results of 2-DE were visualized using silver-staining. The whole staining procedure was performed at room temperature with gentle agitation. Briefly, gels were fixed overnight in 50% methanol containing 12% acetic acid and 0.0185% formaldehyde. They were rinsed in water thrice followed by another wash in an equilibration buffer twice. Sensitization was carried out by immersing the gels in 0.02% sodium thiosulfate for 10 min. After brief wash in water thrice, gels were incubated in 0.15% silver nitrate solution containing 0.0278% formaldehyde for an hour. Color development was performed using a 3% sodium carbonate solution containing 0.037% formaldehyde, until protein spots could be clearly distinguished against the background. The color development was stopped by transferring the gels to 5% acetic acid and left to stand for 10 min or more. All the raw images were digitalized using a scanner (GS-800 calibrated densitometer, Bio-Rad) and the Quantity One software (Bio-Rad). Further analysis was completed by using PDQuest (version 8.0, Bio-Rad) for the detection and quantification of protein spots.
Protein spots were manually excised from the 2D gels. They were washed in a mixture of 30 mM potassium ferricyanide and 100 mM sodium thiosulfate (1:1 v/v) for 5 min. After washing in water twice, the gel plugs were equilibrated in 50 mM ammonium bicarbonate for 20 min, then in 25 mM ammonium bicarbonate and 50% ACN. Finally, they were soaked in 100% ACN until gel plugs became opaque. Thereafter, vacuum-dried gel plugs were rehydrated with 10 μg/mL of trypsin in 25 mM ammonium bicarbonate (pH 8.0). Proteolysis of proteins was performed at 37°C for 16–18 hr. Supernatant was collected and mixed with the α-cyano-4-hydroxycinnamic acid with 4 mg/mL in 35% ACN and 1% TFA. They were then spotted onto the target plate and allowed to dry.
MALDI-TOF MS and data search
MALDI-TOF MS was performed by a 4700 Proteomics Aanlyzer (TOF/TOFTM) (Applied Biosystems). It was equipped with a 355-nm Nd:YAG laser. The instrument was operated in a positive ion reflection mode of 20 kV accelerating voltage, and in a batch mode of acquisition control. All mass spectra were internally calibrated with ACTH peptide and peaks of trypsinized alcohol dehydrogenase. The reflector spectra were obtained over a mass range of 850–3,000 Da. Peptide mass mapping was carried out using the program MASCOT (Matrix Science, London, UK) against Swiss-Prot database with a GPS explorer software (Applied Biosystems). During data search, maximum allowance was set with one missed cleavage per peptide; MS/MS tolerance of 0.1 Da and a mass tolerance of 0.5 Da were also used according to predefined optimized protocol. Carbamidomethylation for cysteine, oxidation for methionine and other variants were also taken into consideration. Before the search of fragments in the database, manual quest was performed to remove tryptic autolytic fragments and notable contamination.
All data are expressed as mean ± SEM. Statistical analysis was performed by single factor ANOVA test and a p-value less than 0.05 was considered statistically significant.
Expression of hsa-let-7g in HCC cell lines
As shown in Figure 2, hsa-let-7g expression was upregulated in Bel-7404 cell line yet markedly downregulated in HepG2, Hep3B and Huh 7 cell lines. HepG2 and Bel-7404 cell lines were chosen for the gain-of-function and loss-of-function studies, respectively.
Effects of hsa-let-7g on the proliferation of HCC cells
To analyze the effect of hsa-let-7g on HCC cell growth, hsa-let-7g mimics and hsa-let-7g inhibitor were transfected into HepG2 and Bel-7404 cells, respectively. As seen in Figure 3a, qPCR revealed an increased level of hsa-let-7g in the HepG2 cells and a concurrently inhibited cell proliferation after 48 hr transfection of hsa-let-7g mimics. The inhibitory effect kept increasing from 72 hr as compared with the negative control (Fig. 3b). In contrast, knockdown of hsa-let-7g by transfecting hsa-let-7g inhibitor was found to significantly downregulate hsa-let-7g expression in Bel-7404 cells (Fig. 3c) and markedly promote the cell proliferation at 72 hr post-transfection (Fig. 3d).
Effects of hsa-let-7g on the transcriptional and translational expression of c-Myc
As seen in Figure 4a, the transcriptional expression of c-Myc was extremely high in both the HepG2 and Bel-7404 HCC cell lines. Therefore, c-Myc was postulated as a candidate target gene for hsa-let-7g in regulating the cell proliferation of HepG2 and Bel-7404 cells. Semiquantitative PCR and Western blotting revealed a significant downregulation of mRNA (Fig. 4b) and protein expression (Figs. 4c and 4d) levels of c-Myc in HepG2 cells, but upregulation of c-Myc mRNA (Fig. 4f and Supporting Information Figure) and protein levels in Bel-7404 cell line (Figs. 4c and 4e). Our results indicate that the c-Myc is a plausible target for hsa-let-7g and an over-expression of hsa-let-7g might inhibit cell proliferation in HCC cells via downregulating the mRNA and protein expression of c-Myc.
c-Myc as a downstream target of hsa-let-7g
Based on our findings, c-Myc may be a plausible target for hsa-let-7g for it possesses a complementary sequence (i.e., the 138–147 nucleotides) at its 3′UTR end for the first 9 nucleotides (seed sequence) at the 5′ end of hsa-let-7g in 6 different species (Fig. 5a). The sequences of mature hsa-let-7g and pre-hsa-let-7g were found to be highly conserved across 24 species (Fig. 5b). By cotransfecting the hsa-let-7g mimics or mimics negative control into HepG2 cell line, the corresponding luciferase activity was measured. As seen in Figure 5c, the luciferase activity decreased after the cotransfection of hsa-let-7g mimics and MYC-3′UTR construct, where the cotransfection of N-ras-3′UTR construct and hsa-let-7g mimics was used as the positive control4 (Fig. 5c). Our data suggest that c-Myc is one of the downstream targets for hsa-let-7g in the regulation of HCC cell proliferation. In fact, the hsa-let-7g gene may function by targeting the 3′UTR of c-Myc mRNA, thus induced mRNA degradation and translational repression.
Effects of hsa-let-7g on the transcriptional and translational expression of p16INK4A
In this study, 2-DE and MALDI-TOF MS were performed to reveal change of protein expression profiles of HCC samples upon hsa-let-7g over-expression. Potential downstream targets of hsa-let-7g were screened and identified. In general, 27 spots were shown to differentially expressed (i.e., more than 2 folds of either increase or decrease in expression) between the hsa-let-7g over-expressing HepG2 cells and the control (data not shown). Fifteen of all were reported as hsa-let-7g-associated as revealed by the MALDI-TOF MS (Table 1). Among those 15 proteins, P16INK4A was the most differentially expressed protein (i.e., by 10 folds; Fig. 6), therefore, it was selected for further Western blotting analysis. We also examined the p16 expression levels in HepG2 and Bel-7404 cell lines and found that p16 was relatively lower as compared with LO2 cell line (Fig. 7a). Consistent with the proteomic result, P16INK4A demonstrated a significant mRNA and protein upregulation upon hsa-let-7g over-expression as compared with the mimics negative control in HepG2 cells (Figs. 7b–7d). In Bel-7404 cell line that was transfected with hsa-let-7g inhibitor, the protein level of P16INK4A decreased markedly when it was compared to the inhibitor negative control (Figs. 7c, 7e and 7f).
Table 1. The 15 proteins with differential expressions (≥2-fold increase or decrease) between hsa-let-7g overexpressing and negative-control HCC cells identified by MALDI-TOF MS
Hsa-let-7g regulated HCC cell proliferation through the c-Myc-Bmi-1-p16 regulatory circuit
To explore the mechanism underlying altered expression of p16INK4A caused by hsa-let-7g, we investigated the transcriptional and translational expression of Bmi-1, a oncogene that has been reported involved in the regulation of c-Myc and p16. We observed that the mRNA and protein levels of Bmi-1 were decreased 48 hr post-transfection of hsa-let-7g mimics in HepG2 cell line as compared with the mimics negative control (Figs. 8a, 8c and 8d). Both mRNA and protein levels of Bmi-1 were found to be increased 72 hr post-transfection of hsa-let-7g inhibitor in Bel-7404 cell line as compared with the inhibitor negative control (Figs. 8b, 8c and 8e). Taken together, our data suggested that hsa-let-7g regulated HCC cell proliferation through direct targeting c-Myc and indirect regulating p16 gene that was regulated by c-Myc through Bmi-1.
Although studies on HCC have been carried out for years, the molecular mechanisms of its genesis, development and metastasis are yet to be fully elucidated. Recent researches have shown a close relationship between the let-7 family and various cancers, with the family members act as tumor suppressors.36, 46 Therefore, we studied the relationship between the hsa-let-7g and HCC cells for it may provide invaluable information for developing novel therapeutic strategy for this highly invasive tumor. Based on our findings, the hsa-let-7g is likely to inhibit HCC cell proliferation by downregulating the oncogene, c-Myc, and upregulating the tumor suppressor gene, p16INK4A.
Previous studies on the miRNA profile of HCC clinical samples showed that the expression of hsa-let-7g was widely downregulated in HCC patients.23, 24, 42, 43 For example, the expression of hsa-let-7g was downregulated and closely associated with HCC venous metastasis and survival.24 Agreed with previous studies, our investigation also demonstrated a significant reduction on the expression of hsa-let-7g in 3 HCC cell lines-HepG2, Hep3B and Huh7 but not in Bel-7404 cell line (Fig. 2). Therefore, the hsa-let-7g gene might not be regarded as specific for HCC for its expression level varied to a great extent among cell lines of the same tumor. In fact, other miRNAs including hsa-mir-122a and hsa-mir-16 have also been reported expressing differentially between Huh7 and HepG2 cell lines.23 This might be due to the complicated regulating network of miRNAs. In general, it is thought that the miRNA may act as intermedium in regulating various biological progresses, which in turn also mediated by miRNA members or genes. This may result in varied expression response for different miRNAs in different cell lines.
Gain-of-function experiment was carried out in HepG2 cells with hsa-let-7g mimics and loss-of-function experiment with hsa-let-7g inhibitor in Bel-7404 cells. Our results show that the proliferation of HepG2 cells was significantly inhibited when cells were transfected with hsa-let-7g mimics, yet enhancing effect was seen in the proliferation of Bel-7404 cells after the transfection of hsa-let-7g inhibitor (Fig. 3). These data suggest that hsa-let-7g is possibly a tumor suppressor gene that acts as cell proliferation inhibitor in HCC cells. In fact, other members of let-7 family have also been reported as tumor suppressors in various human cancers.36, 47–49 For example, the expression level of hsa-let-7a was lowered in gastric tumor tissues, pancreatic cancer-derived cells and lung cancer cells.29, 36, 48 Over-expression of hsa-let-7a inhibited lung cancer cell growth36 and pancreatic cancer cells proliferation was decreased by negative regulation of K-ras and MAPK protein expression.48 Therefore, it is not surprising to find another let-7 family member, hsa-let-7g, also possesses the tumor suppression characteristic.
Our study also demonstrate a strong correlation between the inhibitory effect of hsa-let-7g on the proliferation of HepG2 cells and the oncogene, c-Myc. C-Myc is a basic helix-loop-helix-zipper transcription factor, which carry out its function by heterodimerizes with MAX and binds to E-Box sequence in the promoter region of their target gene including the cyclin D2 and E2F1.50–52 It is widely known that the over-expression of c-Myc was involved in a wide spectrum of human cancers (e.g., breast and cervical carcinoma, neuroblastoma, prostate cancers and HCC) and associated with high aggressiveness or poor prognosis.53, 54C-Myc was also noted to participate in a broad range of physiological or pathological processes including cell growth, apoptosis, development and angiogenesis, whilst its repressed expression decreased tumor cell proliferation and tumorigenesis.55 After a decade of miRNA studies, functional network of c-Myc and its signal transduction pathways are gradually revealed, yet in a more complex way. Researches have demonstrated that c-Myc was downregulated by hsa-let-7a, has-let-7b and hsa-let-7c that were shown to directly target and interact with the 3′UTR of the oncogene.28, 56, 57 For hsa-let-7g, a member of let-7 family, its mature sequence and seed sequence targeting c-Myc 3′-UTR are identical to that of the let-7 family (Fig. 5a). In addition, hsa-miR-34 downregulated the expression of c-Myc and functioned as a tumor suppressor gene.58 In the present study, the mRNA and protein levels of c-Myc decreased significantly after the transfection of hsa-let-7g mimics into HepG2 cells (Figs. 4b–4d). It is consistent with the rule that miRNAs can decrease the expression of their targets by suppressing their expression at the transcriptional and post-transcriptional stages.59–63 These indicate that the c-Myc may be a potential target for hsa-let-7g in suppressing HCC cell proliferation.
Based on our qPCR and Western blotting data, c-Myc is plausible a functional target of hsa-let-7g in HCC cells. The result of luciferase activity assay also supported that c-Myc is a direct downstream target of hsa-let-7g, for there were 9 base pairs that involved in their interaction, were all found to be located in the seed sequence of hsa-let-7g. This finding is very crucial and commonly considered very important for target recognition. Generally speaking, seed sequence of 6–8 nucleotides in length is considered energetically favorable and vital for the miRNA:target interaction.64 In the present study, the seed sequence of hsa-let-7g was shown to be fully complemented to the 3′UTR of c-Myc in 6 different species as well as for other let-7 family members (Fig. 5a). This indicates that c-Myc is possibly a highly potential target for the other let-7 family members28, 56, 57 for the seed interaction sequence found in hsa-let-7g are highly conserved within the family (Fig. 5a). In further analysis, the sequences of hsa-let-7g and pre-hsa-let-7g were found highly conserved across 24 species (Fig. 5b). Taken together, human c-Myc should be a direct target of hsa-let-7g as it fits the criterions of being a good target for hsa-let-7g.
In this study, the protein expression of p16INK4A was significantly increased in HCC cells with hsa-let-7g over-expression (Figs. 7b–7d). This indicates a regulatory role of hsa-let-7g on the protein expression of P16INK4A. We believe that such regulation was governed by the c-Myc-Bmi-1-p16 regulatory circuit as described in an earlier study.65 Briefly, P16INK4A is a tumor suppressor gene that functions as inhibitor of CDK4 kinase.66 The functional relationship between c-Myc, Bmi-1 and p16INK4A has been fully elucidated, in which the c-Myc can directly bind to the E-box at the Bmi-1 promoter, while Bmi-1 is known as a transcription repressor of p16INK4A.67 For example, knockdown of c-Myc has caused a downregulation of Bmi-1 and an upregulation of p16INK4A in endothelial cells.65 Previous studies have suggested that c-Myc did not regulate p16INK4A at physiological expression level, but at both the hypo- and hyper-levels upon c-Myc signaling.65 It is thought that the c-Myc regulated p16INK4A by an indirect circuit involving Bmi-1, while it could exert direct action at the p16INK4A promoter.67, 68 Our results also revealed that both mRNA and protein levels of Bmi-1 were decreased in HepG2 cell line with hsa-let-7g mimics over-expression, but increased with hsa-let-7g inhibitor over-expression in Bel-7404 cell line as compared with respective negative controls (Fig. 8). Therefore, we believe that the upregulation of p16INK4A upon an over-expression of hsa-let-7g may also be due to the downregulation of c-Myc through regulating Bmi-1, thus resulted in an inhibitory effect on the HCC cell proliferation. We speculated that other members of the let-7 family might also regulate p16 gene through c-Myc. There are 2 reasons for this speculation. First, let-7g regulates c-Myc by binding its seed sequence and target sites to c-Myc 3′UTR, and the seed sequence in mature let-7 family members is identical. Thus, all let-7 family members, including hsa-let-7g, can bind to c-Myc 3′UTR and subsequently regulate c-Myc. Second, we showed that hsa-let-7g indirectly regulated p16 through its direct regulation of c-Myc in the c-Myc-Bmi-1-p16 regulatory circuit. Therefore, we hypothesized that other members of let-7 family could also regulate p16 through c-Myc.
Taken together, the present study suggests that the hsa-let-7g plays an important role in inhibiting HCC tumorigenesis and progression. Such inhibitory ability may be carried out via the downregulation of c-Myc expression as well as the upregulation of p16INK4A. Further investigation will be necessary to fully reveal the molecular mechanism; yet the hsa-let-7g is undoubtedly a novel potential target that worth further investigation and trial in treating HCC.
The authors thank Associate Professor Dr. Songshan Jiang for providing them the psiCHECK2 vectors.