CircYTHDF3 promotes hepatocellular carcinoma progression through modulating miR-136-5p/CBX4/ VEGF axis

doi:10.21203/rs.3.rs-3135902/v1

Download PDF

Research Article

CircYTHDF3 promotes hepatocellular carcinoma progression through modulating miR-136-5p/CBX4/ VEGF axis

https://doi.org/10.21203/rs.3.rs-3135902/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Circular RNAs (circRNAs) have been widely reported to participate in diverse biological pathways in cancers. However, the function of circRNAs in hepatocellular carcinoma (HCC) remains some unclear. In the present study, we screened the expression profile of circRNAs in HCC and identified that circYTHDF3 (hsa_circ_0084620) was dramatically upregulated in HCC tissues and cell lines. Patients with high-expression circYTHDF3 performed a poor prognosis. CircYTHDF3 could facilitate proliferation, migration, and invasion of HCC cell. Furthermore, circYTHDF3 could regulate CBX4 expression though competitive binding with miR-136-5p and inhibited the effect of miR-136-5p on proliferation, migration, invasion of HCC cells. Moreover, circYTHDF3 down-expression inhibited the growth, lung metastasis and angiogenesis of HCC tumors in vivo. Taken together, these findings indicated that circYTHDF3 might function as a competing endogenous RNA (ceRNA) to promote HCC progression by targeting miR-136-5p/CBX4/VEGF pathway. Therefore, circYTHDF3 could serve as a potential prognostic indicator for HCC and a therapeutic target for HCC treatment.

CircYTHDF3

miR-136-5p

CBX4

Hepatocellular carcinoma

Liver cancer, is a frequently diagnosed cancer, with approximately 900,000 new cases diagnosed annually, and it caused approximately 830,000 deaths in worldwide 2020¹. Hepatocellular carcinoma (HCC) accounts for 75% of liver cancer cases². The considerable mortality rates can be attributed to several causes. Despite the advances in the understanding of the molecular pathogenesis of HCC, most studies have not identified druggable targets^3,4. Molecular classification criteria and personalized treatment strategies for HCC have not been established, and the Barcelona-Clinic-Liver Cancer (BCLC) classification (based on performance status, tumor extent, and liver function) is the main clinical algorithm that used for the prognostic evaluation of patients^5,6. Patients in early stages benefit from curative therapies, such as surgical resection, radiofrequency ablation (RFA), transcatheter arterial chemoembolization (TACE) or transplantation⁷. Unfortunately, surveillance programs that aim to detect HCC in its early stages lack convenient, rapid, and sensitive detection methods or biomarkers⁸. For patients in advanced stage disease, chemoembolization (patients with stage B disease with a median survival of 30 months)⁹ and sorafenib (extends patient survival for patients with advanced stage disease by 10 months)¹⁰ have shown survival benefits. These limitations could be addressed through the discovery of new therapies and biomarkers for HCC classification or prognostic evaluation.

CircRNAs are single-stranded RNAs with a covalently closed continuous loop structure that are expressed in in mammalian cells and ubiquitously regulate gene expression¹¹. Previous studies have shown that circRNAs can regulate malignant behaviors, including the proliferation, migration, invasion, and apoptosis of cancer cells. For example, circRNA hsa_circ_0000096 regulates the proliferation and migration of gastric cancer cells by modulating the gene expression of cyclin D1, cyclin-dependent kinase-6 (CDK-6), matrix metalloproteinase-2 (MMP-2), MMP-9, and E-cadherin8¹². Hsa_circ_0067934 acts as an oncogenic factor to promote the proliferation of esophageal carcinoma cells¹³. CircRNAs can act as competing endogenous RNAs (ceRNAs) to sponge their target miRNAs by directly binding to them, and thus modulate their activity, regulating miRNA-target gene expression. A recent study has indicated that circRNA CiRS-7 can promote the growth and metastasis of esophageal squamous cell carcinoma by sponging miR-7 and suppressing its activity¹⁴. The circRNA Vav3 can sponge gga-miR-375 to affect the YAP1 expression, by promoting the epithelial mesenchymal transition (EMT) process¹⁵. Furthermore, circRNAs can interact with transcription factors and other proteins to form circRNPs, and modulate their function¹⁶. A recent study suggests that circRNAs are important for the initiation, migration, and invasion of HCC¹⁷. However, which circRNA regulates the malignant behaviours of HCC and how have not been elucidated.

In this study, we screened differentially expressed circRNAs that were significantly upregulated in HCC tissues and showed that circYTHDF3 overexpression effectively attenuated the malignant behaviors of HCC. Furthermore, we explored the potential target miRNAs of circYTHDF3 and found that circYTHDF3 acted as a ceRNA to sponge oncogenic miR-136-5p to enhance CBX4 expression, attenuating the malignant behaviors of HCC in vitro and in vivo.

HCC specimens and microarray of circRNAs

We collected 54 pairs of surgical or biopsied HCC and adjacent non-tumor specimens in West China Hospital of Sichuan University from February 2015 to June 2018. All samples were collected with patient consent and signed informed consent. The HCC patients were diagnosed, based on the practice guidelines of the American Association for the Study of Liver Diseases (AASLD). Their liver specimens were evaluated by pathologists and their clinical stages were determined, according to the BCLC classification. HCC patients with the following conditions were excluded: (1) patients ≤ 18 or ≥ 70 years of age or without full civil capacity; (2) patients with a history of preoperative anticancer radiotherapy or chemotherapy, biological, immune, and traditional Chinese medicine; (3) patients with incomplete postoperative follow-up data; (4) patients with a history of another organ malignancy, or systemic immune disease. Their tissue RNA was extracted and analyzed as followed in section RNA preparation, reverse transcription, and quantitative real-time PCR analysis. Human circRNA Array (Arraystar, Kangchen, Shanghai) analysis specifically for human circular RNAs splicing sites was used with four pair samples. The limma package for R software was used to normalize and process the data. Significant differentially expressed circRNAs between the two groups were identified through fold change, P-value, and raw intensity. The different expression circRNAs could be obtained from Supplementary Materials.

In vitro cell culture, maintenance, and transfection

Human liver cell line LO₂ and HCC cell line HCC-LM9, Huh-7, HepG2, Hep3B, SNU449 cells were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA) or as a kind gift from Professor Peng Yong (State Key Laboratory of Biotherapy, Sichuan University). The cells were cultured in DMEM (Gibco BRL, Grand Island, NY, USA) containing 10% fetal bovine serum (FBS) (Gibco, Australia), 2 mM l-glutamine, 100 units/ml of penicillin, and 100 µg/ml of streptomycin at 37°C in an incubator of 5% CO₂. The relevant lentiviruses and si-RNA were obtained from Shanghai Riobio (Shanghai, China). HCC cells were transduced with individual types of lentivirus at a multiplicity of infection (MOI) of 10 in the presence of 5 µg/ml puromycin or transfected with the specific plasmid by using Lipofectamine 2000 (Thermo Scientific, USA).

RNA preparation, reverse transcription, and quantitative real-time PCR analysis

The total RNAs were extracted by using Trizol reagent (Invitrogen, Grand Island, NY), and the nuclear and cytoplasmic fractions were extracted using a Cytoplasmic & Nuclear RNA Purification Kit (Norgen Biotek, Canada, CA) according to the manufacturer’s protocol. The RNA concentrations were measured by a Nanodrop 2000 spectrophotometry (Thermo Scientific, USA). The contained circRNAs were enriched and digested with RNase R (Epicentre Technologies, USA). The cells were incubated with Actinomycin D (2 mg/mL) or DMSO (Sigma Aldrich, St. Louis, MO, USA) to assess the stability of circRNA and its linear isoform. The total RNA was reversely transcribed into cRNA by using Superscript III transcriptase (Invitrogen, Grand Island, NY) and miRNA First Strand cDNA Synthesis (Tailing Reaction) (Sangon Biotech, China). The level of circRNA, mRNA and miRNA were analyzed by RT-qPCR using a SYBR Green Real-time PCR Master Mix (Roche, Mannheim, Germany). The relative RNA expression levels were calculated according to the 2^–∆∆Ct method and normalized to the expression of U3 and GAPDH. The primers used in this study are listed as follow (5’-3’):

CircYTHDF3-F: AATTCAAGGGACACTCAAGA

CircYTHDF3-R: GTGGATAGCTGTTACTCCTAC

CBX4-F: CCTTCCAGAACAGGGAACGG

CBX4-R: TCAGGACATTGGAACGACGG

VEGF-F: GCAGAATCATCACGAAGTGG

VEGF-R: GCATGGTGATGTTGGACTCC

miR-136-5p: ACTCCATTTGTTTTGATGATGGA

CCK8 assay and EdU assay

The effects of circRNA and miRNA on HCC cell lines were determined using a CCK8 assay and EdU assay. Cells were seeded in 96-well plates at a density of 5 × 10³ cells/well, cultured overnight, and medium containing CCK8 reagent was added and the optical density (OD) value at 450 nm of each well was measured. Processed cells at a density of 5 × 10³ cells/well were cultured in different condition medium containing EdU solution (final concentration of 50 µM, RioboBio, China) for 2 h. At the end of the incubation, cells were collected and fixed with 1 ml 4% paraformaldehyde. After washing with PBS, each well was added with TritonX-100 and incubated at room temperature for 10 min. After washing with PBS, each well was added with 1× Apollo® staining agent and Hoechst 33342 and incubated at room temperature in the dark. Then, cells were collected and resuspended with 500 µl PBS and analyzed by flowcytometry.

Wound-healing assay and Transwell assay

HCC cells were cultured in 6-well plates up to ~ 100% confluence for preparation of the wound healing assay. The cells were starved for 12 h and wounded with a sterile 10-µL micropipette tip. After being washed, the cells were cultured for 24 h and photographed at 0 and 24h. The invasion of different groups of HCC cells was determined by transwell invasion assay. Briefly, HCC cells (10⁵ cells/well) were cultured in 200 µl serum-free medium in the top chamber that had been loaded with Matrigel (Corning, Inc., Corning, NY, USA). The bottom chambers were filled with 600 µl medium containing 10% FBS. After incubating for 24 ~ 48h at 37℃, the cells on the upper surface of the top chamber were removed, and the invaded cells on the bottom surface from the top chamber were stained with Harris hematoxylin solution (Sigma, Germany) and photographed under a microscope. The results were calculated using Image J (National Institutes of Health, New York, USA). The migration assay methods were similar to invasion assay, except that the upper chamber was coated with Matrigel.

Fluorescence in situ hybridization (FISH)

The subcellular localization of circRNA was performed using a RiboTM Fluorescent in situ Hybridization Kit (RiboBio Technology, C10910) according to the manufacturer’s protocols. All images were procured using a ZEISS LSM800 Confocal Microscope system with Zeiss Efficient Navigation (ZEN) software (Carl Zeiss AG, Germany).

MiRNA and mRNA prediction

The potential target miRNAs of circYTHDF3 were predicted by using the tool in bioinformatics database CircInteractome (https://circinteractome.irp.nia.nih.gov/) and Starbase (https://rnasysu.com). The potential target genes of miR-136-5p were predicted by using the miRDB (http://mirdb.org/), TargetScanHuman (www.targetscan.org), PicTar (https://pictar.mdc-berlin.de/), miRmap (https://mirmap.ezlab.org/) and microT (http://diana.cslab.ece.ntua.gr/microT/). Consensus genes from these five online databases were recognized as trustworthy targets. The expression levels of CBX4 and VEGF in HCC cell lines were obtained from Cancer Dependency Map (DepMap) Project (https://depmap.org/). The expression levels of CBX4 and VEGF in HCC samples were obtained from Gene Expression Profiling Interactive Analysis (GEPIA)¹⁸.

Luciferase reporter assay

HEK293T cells (1 × 10⁵ cells per well) were seeded in 24-well plates and co-transfected with the circRNA-YTHDF3-wt/mut vector, miR-136-5p mimics, inhibitors and miR-NC or CBX4-wt/mut vector, miR-136-5p mimics and miR-NC using Lipofectamine™ 3000 (Invitrogen, CA, USA). After a 48h incubation, luciferase activity was measured by a Luciferase Reporter Assay System (Promega, USA) in accordance with the manufacturer’s instructions.

HCC animal model

Male BALB/c nude mice (4.5 weeks old) were purchased from Chengdu Dossy experimental animals corporation. All of the animal care and experimental procedures were approved by the Institutional Animal Care and Use Committee of Sichuan University. For tumor growth studies, mice were subcutaneously injected with 1.5 × 10⁶ WT HCC-LM9 cells (control), or sh-circYTHDF3 HCC-LM9 cells (14 nude mice were randomly divided into 2 groups, 7 in each group). The growth of implanted HCC tumors was monitored for their volumes every 5 days up to 30 days post inoculation. The mice were killed and their xenografts were photographed and measured. For the lung-metastasis assay, cells (1 × 10⁶ cells per mouse) with stably green-fluorescent-protein expression were inoculated into the tail-vein of mice (7 mice per group). After four weeks, the fluorescence intensity was detected by an in vivo imaging system. The mice were killed and their xenografts were photographed. The tumor tissues were further subjected to HE staining and IHC analysis.

H&E staining and Immunohistochemical (IHC) staining

Tissue samples were fixed in 4% paraformaldehyde, embedded in paraffin, and cut into 5-µm thick sections. H&E staining was possessed to observe the histopathological features. At least five fields were analyzed on each section, and an Olympus microscope captured images. As for IHC staining, after dewaxing, hydration and so on, the tissue sections were incubated with specific primary antibodies against CBX4 (Abcam, Cambridge, UK), VEGF (CST, Massachusetts, USA), and CD31 (Abcam, Cambridge, UK). Images were captured using a Nikon Eclipse 80i system with NIS-Elements software (Nikon, Japan). All sections were scored by the H-score approach (semi-quantitative) and validated by two experienced pathologists.

Western blot analysis

Total protein was extracted by RIPA Lysis Buffer (Thermo, USA). The BCA Protein Assay Kit ((Thermo, USA)) was used for detecting protein concentration. Proteins were electrophoresed in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto PVDF membranes (Millipore, Massachusetts, USA). After blocking for 1 h in 5% skim milk powder at room temperature, the membranes were incubated with primary antibodies (CBX4, Abcam, #ab242149) and anti-rabbit secondary antibody (1:10000, Abcam, UK) successively. The intensity of signals was detected by ChemiDoc MP Imager System (Bio-Rad, California, USA) and analyzed by Image J (National Institutes of Health, New York, USA).

Statistical analysis

Data are expressed as mean ± standard error of the mean (SEM). Comparisons among groups were analyzed by one-way ANOVA and post hoc Tukey tests, and the differences between groups were analyzed by two-tailed t tests by using GraphPad Prism 8 for Windows. A P-value of < 0.05 was considered statistically significant.

1. CircYTHDF3 is a circRNA that is upregulated in HCC and is clinically relevant to HCC patient prognosis.

To screen the differentially expressed circRNAs in HCC, the circRNA expression profile was analyzed in four paired tissue samples by microarray. The bioinformatics analysis revealed 139 upregulated and 115 downregulated genes in HCC tissues according to the following criteria: p value < 0.05 and∣log₂FC∣> 1 (Fig. 1a, b). Hsa_circ_0084620, which is derived from the linear RNA YTHDF3, was upregulated in HCC (P = 0.04; log₂FC = 3.06) and was named circYTHDF3. To investigate the potential role of circYTHDF3 in predicting the progression of HCC, 50 pairs of surgical and biopsied HCC and adjacent nontumor tissues were obtained. The results indicated high expression of circYTHDF3 in HCC tissues (Fig. 1c, p = 0.001). The demographic and clinical characteristics of those patients are shown in Table 1. In addition, the association between circYTHDF3 expression and other clinical features of HCC patients was analyzed. We found that tumor numbers and microvascular invasion were correlated with the expression of circYTHDF3. Next, we investigated the correlation between circYTHDF3 expression and the prognosis of HCC patients. As shown in Fig. 1d, Kaplan–Meier survival analysis revealed that patients with higher levels of circYTHDF3 expression had a lower survival rate (HR = 3.364, 95% CI = 1.495–7.568, p = 0.0091). The expression of circYTHDF3 was also measured in HCC cells as well as in normal liver cells. Notably, in all HCC cells, including Hep3B, SNU449, HCC-LM9, Huh7 and HepG2 cells, the expression of circYTHDF3 was significantly upregulated in HCC cells compared with LO2 cells (Fig. 1e).

Table 1

Relationships between circYTHDF3 expression and clinicopathological characters of HCC
Parameter	Numbers of patients	circYTHDF3(low)	CircYTHDF3(high)	p-value
Age		51.72 ± 11.71	46.56 ± 10.59
Sex				0.171
male	39	22	17
female	11	3	8
Differentiation				1.000
poor	5	2	3
Moderate + Well	45	23	22
Tumor size(cm)				1.000
< 5	37	19	18
≥ 5	13	6	7
Tumor numbers				0.023
1	41	24	17
> 1	9	1	8
AFP (ng/ml)				0.256
≤ 400	27	11	16
> 400	3 = 23	14	9
HBsAg				0.762
negative	34	18	16
positive	16	7	9
Cirrhosis				0.725
no	10	6	4
yes	40	19	21
Vessel invasion				0.776
no	28	15	13
yes	22	10	12
MVI				0.038
no	32	20	12
yes	18	5	13
Abbreviations: AFP, alpha-fetoprotein; MVI, microvascular invasion.

2. The characteristics of circYTHDF3 in HCC cells

Hsa_circ_0084620 (chr8:64,098,705 − 64,100,303) is derived from regions in exon 5 of the YTH N6-methyladenosine RNA binding protein F3 (YTHDF3) locus. It is located on chromosome 8q12.3, it was named circYTHDF3, and its spliced length is 1598nt (Fig. 2a). To further explore the biological roles of circYTHDF3, we selected Huh7 and HCC-LM9 cells to perform further experiments with the aim of determining the full expression spectrum of circYTHDF3 in HCC cells. Next, the expression of linear YTHDF3 mRNA and circular circYTHDF3 in HCC-LM9 and Huh7 cell lines, respectively, was measured by qRT‒PCR. When exposed to RNase R treatment for 2 h, the YTHDF3 mRNA expression level was extremely decreased, but circYTHDF3 exhibited resistance to RNase R (Fig. 2b). The results confirmed that circular circYTHDF3 was not affected by RNase R. After exposure to actinomycin D, circYTHDF3 showed a longer half-life than linear YTHDF3, indicating its stability (Fig. 2c). Furthermore, the localization of circular circYTHDF3 was measured by qRT‒PCR (Fig. 2d) and RNA-FISH (Fig. 2e). The results demonstrated that circYTHDF3 was mainly expressed in the cytoplasm in HCC cells rather than in the nucleus.

3. CircYTHDF3 downregulation suppresses the malignant behaviors of HCC cells in vivo.

We established a circYTHDF3-knockout system in the HCC-LM9 and Huh7 cell lines by using sh-circRNA to further investigate the pathological role of circYTHDF3. Cell viability of the different groups was determined by CCK-8 and EdU assays. The CCK8 results demonstrated that knockout of circYTHDF3 markedly inhibited cell viability (Fig. 3a, p < 0.01). Subsequently, an EdU assay was conducted, which revealed suppressed HCC cell proliferation after the depletion of circYTHDF3 (Fig. 3b, p < 0.05). Subsequently, the influence of circYTHDF3 on the migration and invasion of HCC cells was determined by wound healing and Transwell assays. The wound healing experiments primarily indicated that the migration and invasion of HCC cells were inhibited (Fig. 3c, p < 0.01) when circYTHDF3 was expressed at low levels. Transwell assays showed a similar trend (Fig. 3d, p < 0.05). In conclusion, these data confirmed that circYTHDF3 could promote malignant capacities (proliferation, migration, and invasion) of HCC cells.

4. MiRNA-136-5p inhibits the malignant behaviors of HCC cells.

To better understand the molecular mechanism of circYTHDF3, we used bioinformatics tools to predict the possible miRNA target of circYTHDF3. We used CircInteractome (https://circinteractome.nia.nih.gov/) and starBase (http://starbase.sysu.edu.cn/) to predict the target of circYTHDF3. As shown in Fig. 4a, circYTHDF3 was shown to have the potential to bind to 9 miRNAs. MiR-136-5p was a potential target of circYTHDF3. The results indicated lower expression of miR-136-5p in HCC tissues (Fig. 4b, p < 0.001), and circYTHDF3 and miR-136-5p expression were negatively correlated in HCC tissue samples (Fig. 4c, p = 0.012, R = 0.123). MiR-136-5p has been previously reported to be related to non‑metastasis, early TNM stage, non‑portal vein tumor embolus and non‑vaso‑invasion¹⁹ and to function in HCC as a tumor suppressor²⁰. We determined the effects of miR‑136‑5p on cell proliferation by using EdU. The results of the EDU assay showed that cells transfected with the miR-136-5p inhibitor exhibited increased proliferation compared to those transfected with the control, and miR-136-5p mimics decreased proliferation (Fig. 4d, p < 0.05). We further evaluated the effect of miR-136-5p on cell migration and invasion through Transwell assays. According to the assay results, upregulation of miR-136-5p expression promoted the migration of HCC-LM9 cells. Downregulation of miR-136-5p expression significantly induced the migration of HCC-LM9 cells. Then, we investigated the function of miR-136-5p in cell invasion in HCC-LM9 cells. After silencing or overexpressing miR-136-5p, similar results were observed in invasion assays (Fig. 4e, p < 0.05). Taken together, these results suggested that miR-136-5p served as a tumor suppressor in HCC.

5. CircYTHDF3 acts as a ceRNA to sponge miR-136-5p.

We primarily examine the expression of miR-136-5p with a decrease of circYTHDF3 and the results indicated the up-regulated levels of miR-136-5p (Fig. 5a, p < 0.01). The binding sites between circYTHDF3 and miR-136-5p are shown in Fig. 5b. The relative luciferase activity of miR-136-5p was measured by dual-luciferase reporter assay. Wild-type (wt) and mutant (mut) psiCHECK2- circYTHDF3 plasmids incorporating miRNA binding sites (Fig. 5c) were further constructed for dual luciferase reporter assays. These plasmids were co-transfected with miR-136-5p mimics or inhibitors to determine whether circYTHDF3 could bind to miR-136-5p via the predicted binding sites. The results of different luciferase intensities between the psiCHECK2-WT-circYTHDF3 and psiCHECK2-MT-circYTHDF3 groups showed that psiCHECK2-WT-circYTHDF3 could directly bind to miR-136-5p but not psiCHECK2-MT-circYTHDF3. Based on these data, we further found that miR-136-5p could be a direct target of circYTHDF3. Proliferation in the different groups was measured by EdU staining. As shown in Fig. 5d, knockout of circYTHDF3 significantly inhibited proliferation, and after co-transfection with miR-136-5p mimics, the proliferation rate decreased. In contrast, when co-transfected with miR-136-5p inhibitors, the proliferation rate increased. The Transwell assay results demonstrated that knockout of circYTHDF3 inhibited cell migration and invasion; however, this inhibitory effect was attenuated by the downregulation of miR-136-5p and amplified by the overexpression of miR-136-5p in HCC-LM9 cells (Fig. 5e). The co-transfection of circYTHDF3 lentiviruses with miR-136-5p mimics or inhibitors indicated that circYTHDF3 sponged miR-136-5p and promoted HCC cell proliferation, migration and invasion, counteracting the tumor suppressive effects.

6. CircYTHDF3 enhances the malignant behaviors of HCC cells through the miR-136-5p/CBX4/VEGF axis.

The next step was to elucidate the effects of circYTHDF3 and miR-136-5p on the phenotype of HCC cells. According to analyses of miRDB, TargetScan, PicTar, miRmap and microT, the targets of miR-136-5p were identified, and CBX4 might be a target of miR-136-5p (Fig. 6a). A relationship between CBX4 and miR-136-5p was previously reported in cervical cancer²¹ and bladder cancer²². A primary exploration demonstrated that CBX4 was highly expressed in HCC tissues (Fig. 6b, S1a, p < 0.01) and predicted a poor prognosis of HCC patients (Fig. 6c, HR = 1.7, p = 0.048). The luciferase assays showed that psiCHECK2-WT-CBX4 directly bound to miR-136-5p (Fig. 6d), indicating that miR-136-5p could directly bind to CBX4 mRNA. We found that the protein level of CBX4 was markedly downregulated in the circRNA-knockout group, aggravated by the upregulation of miR-136-5p. When miR-136-5p inhibitors were transfected, the downregulated expression of CBX4 were reversed (Fig. 6e). In addition, circYTHDF3 and CBX4 expression were positively correlated as shown in Fig. 6f and opposite results happened between miR-136-5p and CBX4 (Fig. S1b). Moreover, CBX4 was reported to regulated the expression of VEGF in HCC²³ and MHCC97L cells²⁴. According to TCGA database and DepMap, CBX4 and VEGF expression were positively correlated in HCC tissues (Fig. 6g, R = 0.47, p < 0.01) and in HCC cell lines (Fig. S1c). Meanwhile, we found that the expression of CBX4 and VEGF was downregulated in the circRNA-knockout group, which could partly build a relationship between circYTHDF3 and VEGF (Fig. 6h, p < 0.01). Overall, these data indicated that circYTHDF3 regulated miR-136-5p to affect proliferation, migration, and invasion and may have an effect on the CBX4/VEGF signal pathway.
7. In vitro.

Luciferase-labelled HCC-LM9 cells transfected with sh-circRNA or sh-NC plasmid were injected into male nude mice either subcutaneously or via the tail vein. The tumors were removed from the mice, and the mice were sacrificed. Tumor weights were measured and data indicated sh-circRNA group are some smaller (Fig. 7a, p < 0.01). Tumor volume were measured during tumor growth. The results showed that in the sh-circRNA group, tumor growth was slower, respectively, than those in the sh-NC group (Fig. 7b, p < 0.05). Luciferase signals in mice treated with tail vein injection were examined using ex vivo imaging. The luciferase signals of lung metastases in the sh-circRNA group were lower than those in the control group (Fig. 7c). The lungs were removed from the mice, and the mice were sacrificed. HE staining of lung metastases indicated a reduced tumor cell number (Fig. 7d). IHC staining of lung metastases was conducted to measure the expression of CBX4/VEGFA/CD31. As we know, CD31 could work as a biomarker of angiogenesis²⁵. As shown in Fig. 7e, CBX4/VEGF/CD31 were expressed at lower levels in the circYTHDF3-knockout group.

CircRNAs were discovered several decades ago, but they did not receive much attention until the development and application of high-throughput sequencing technologies and dedicated bioinformatics algorithms. After that, circRNAs were rediscovered and widely investigated as crucial molecules that participate in tumorigenesis and tumor development. In this study, we performed a circRNA array analysis to identify abnormally expressed circRNAs in HCC and adjacent nontumor tissues. We finally focused on a novel circRNA, hsa_circ_0084620 (named circYTHDF3), which was significantly upregulated in HCC tissues and cell lines. Moreover, high expression of circYTHDF3 positively correlated with poor prognosis in HCC patients. We found that the stable downregulation of circYTHDF3 suppressed the growth, migration, and invasion of HCC cells in vivo and in vitro. Mechanistically, CBX4 was involved in the biogenesis of circYTHDF3. More specifically, circYTHDF3 could function as a sponge of miR-136-5p to attenuate its inhibitory effect on CBX4 and the activation of the CBX4/VEGF signal pathway, which promoted the progression and angiogenesis of HCC. In summary, we demonstrated that circYTHDF3 might be a novel biomarker for HCC diagnosis and prognosis, as well as a novel therapeutic target for HCC patients.

CircYTHDF3 is derived from regions of exon 5 of YTHDF3. Clinically, YTHDF3 is overexpressed in HCC tissues, and high expression of YTHDF3 is correlated with poor prognosis in HCC patients²⁶. YTHDF3 was reported to facilitate carcinogenesis and progression of HCC by inhibiting PFKL mRNA degradation²⁷. To eliminate the biological functional change caused by YTHDF3, we explored the expression of YTHDF3 in HCC cells after the downregulation of circYTHDF3, and the results showed no significant change. One of the best-known mechanisms by which circRNAs function in the cytoplasm is to act as microRNA sponges and influence the expression of miRNA target genes. Via FISH and nuclear and cytoplasmic fraction assays, we found that circYTHDF3 was mainly localized in the cytoplasm of HCC cells. Using bioinformatics analysis, we predicted miR-136-5p as a potential target molecule of circYTHDF3, and we further confirmed this finding by dual luciferase reporter assay.

MiR-136-5p was previously reported to play a tumor suppressive role in several cancers^21,28,29. The first study of miR-136-5p in HCC focused on the expression of miR-136-5p in HCC with clinical parameters and predicted target genes of miR-136-5p via NLP analysis and GO analysis in 2017¹⁹. Next, several studies claimed that miR-136-5p played a key suppressive role in HCC proliferation and metastasis^20,30. Our study verified the inhibitory function of miR-136-5p in HCC proliferation and metastasis by using mimics and inhibitors. In particular, we first demonstrated that the tumor-suppressive role of miR-136-5p could be regulated by the sponge mechanism of circYTHDF3and identified CBX4 as a novel target of both miR-136-5p and circYTHDFF3 in HCC cells.

CBX4, known as polycomb 2 (PC2) or NBP16, is located on chromosome 17q25.3 and encodes a polycomb repressive complex 1-associated protein (CBX4 protein). CBX4 protein is a member of the Polycomb group (PcG) of proteins and can enhance migration, invasion, angiogenesis, growth or sorafenib resistance in HCC cells^24,31−33. The association between CBX4 and miR-136-5p was first mentioned in Zhao’s study, which focused on cervical cancer progression²¹. In our study, decreased expression of circYTHDF3 was correlated with increased miR-136-5p levels and downregulated CBX4 expression at both the mRNA and protein levels. According to the results of Chen and coworkers, increased expression of CBX4 was correlated with VEGF expression and angiogenesis in HCC cells²³. We primarily explored the role of circYTHDF3 in VEGF expression and angiogenesis. The IHC results provided new evidence for the function of circYTHDF3 in HCC angiogenesis. Moreover, as CBX4 is a SUMO E3 ligase, whether and how it regulates the transcription of oncogenes in HCC needs to be further studied.

In conclusion, this study explored the role of circYTHDF3 in the development of HCC. CircYTHDF3 promoted malignant behaviors by competitively binding miR-136-5p, leading to the increased expression of CBX4 and VEGF. These findings provide new evidence for the function of circYTHDF3, indicating the potential clinical application of circYTHDF3 inhibitors and revealing the potential of circYTHDF3 to be used as a biomarker for prognosis prediction in HCC patients.

HCC: hepatocellular carcinoma; circRNA: circular RNA; DMEM: Dulbecco's Modified Eagle's Media; complex 1-associated protein 4: CBX4 protein; Vascular endothelial growth factor: VEGF; OS: overall survival; PFS: progression-free survival; IHC: Immunohistochemistry; RT-qPCR: quantitative reverse transcriptase-polymerase chain reaction; FISH: fluorescence in situ hybridization.

Author Contributions

Bei Li, and Jiong Lu designed the study. Guilin Nie, Ningyuan Wen, and Yaoqun Wang conducted experiments. Dingzhong Peng and Guilin Nie analyzed data. Guilin Nie wrote and drafted the manuscript. All authors read, edited and approved the final manuscript.

Ethics approval

All patient specimens used in this study and all animal studies were approved by the Ethics Committee of the West China Hospital of Sichuan University.

Funding

Supported by 1·3·5 project for disciplines of excellence–Clinical Research Incubation Project, West China Hospital, Sichuan University (20HXFH021), National Natural Science Foundation of China (Grant No. 82002578).

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Availability of data and materials

The original contributions presented in the study are included in the article/Supplementary Materials. Further inquiries can be directed to the corresponding author.

Sung, H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 71, 209-249, doi:10.3322/caac.21660 (2021).
Petrick, J. L. et al. International trends in hepatocellular carcinoma incidence, 1978-2012. Int J Cancer 147, 317-330, doi:10.1002/ijc.32723 (2020).
Hepatocellular carcinoma. Nat Rev Dis Primers 7, 7, doi:10.1038/s41572-021-00245-6 (2021).
Cucarull, B. et al. Hepatocellular Carcinoma: Molecular Pathogenesis and Therapeutic Advances. Cancers (Basel) 14, doi:10.3390/cancers14030621 (2022).
EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J Hepatol 69, 182-236, doi:10.1016/j.jhep.2018.03.019 (2018).
EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol 56, 908-943, doi:10.1016/j.jhep.2011.12.001 (2012).
Yang, J. D. & Heimbach, J. K. New advances in the diagnosis and management of hepatocellular carcinoma. Bmj 371, m3544, doi:10.1136/bmj.m3544 (2020).
Johnson, P., Zhou, Q., Dao, D. Y. & Lo, Y. M. D. Circulating biomarkers in the diagnosis and management of hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol 19, 670-681, doi:10.1038/s41575-022-00620-y (2022).
Dhondt, E. et al. (90)Y Radioembolization versus Drug-eluting Bead Chemoembolization for Unresectable Hepatocellular Carcinoma: Results from the TRACE Phase II Randomized Controlled Trial. Radiology 303, 699-710, doi:10.1148/radiol.211806 (2022).
Qin, S. et al. Donafenib Versus Sorafenib in First-Line Treatment of Unresectable or Metastatic Hepatocellular Carcinoma: A Randomized, Open-Label, Parallel-Controlled Phase II-III Trial. J Clin Oncol 39, 3002-3011, doi:10.1200/jco.21.00163 (2021).
Chen, L. L. The expanding regulatory mechanisms and cellular functions of circular RNAs. Nat Rev Mol Cell Biol 21, 475-490, doi:10.1038/s41580-020-0243-y (2020).
Li, P. et al. Circular RNA 0000096 affects cell growth and migration in gastric cancer. Br J Cancer 116, 626-633, doi:10.1038/bjc.2016.451 (2017).
Xia, W. et al. Circular RNA has_circ_0067934 is upregulated in esophageal squamous cell carcinoma and promoted proliferation. Sci Rep 6, 35576, doi:10.1038/srep35576 (2016).
Li, R. C. et al. CiRS-7 promotes growth and metastasis of esophageal squamous cell carcinoma via regulation of miR-7/HOXB13. Cell Death Dis 9, 838, doi:10.1038/s41419-018-0852-y (2018).
Zhang, X. et al. Circular RNA Vav3 sponges gga-miR-375 to promote epithelial-mesenchymal transition. RNA Biol 16, 118-132, doi:10.1080/15476286.2018.1564462 (2019).
Kristensen, L. S. et al. The biogenesis, biology and characterization of circular RNAs. Nat Rev Genet 20, 675-691, doi:10.1038/s41576-019-0158-7 (2019).
Shen, H. et al. Circular RNAs: characteristics, biogenesis, mechanisms and functions in liver cancer. J Hematol Oncol 14, 134, doi:10.1186/s13045-021-01145-8 (2021).
Tang, Z. et al. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res 45, W98-w102, doi:10.1093/nar/gkx247 (2017).
Ding, H. et al. Downregulation of miR‑136‑5p in hepatocellular carcinoma and its clinicopathological significance. Mol Med Rep 16, 5393-5405, doi:10.3892/mmr.2017.7275 (2017).
Liu, D. et al. Exosome-transmitted circ_MMP2 promotes hepatocellular carcinoma metastasis by upregulating MMP2. Mol Oncol 14, 1365-1380, doi:10.1002/1878-0261.12637 (2020).
Zhao, J., Yang, T. & Li, L. LncRNA FOXP4-AS1 Is Involved in Cervical Cancer Progression via Regulating miR-136-5p/CBX4 Axis. Onco Targets Ther 13, 2347-2355, doi:10.2147/ott.S241818 (2020).
Li, S., Zhao, Y. & Chen, X. Microarray expression profile analysis of circular RNAs and their potential regulatory role in bladder carcinoma. Oncol Rep 45, 239-253, doi:10.3892/or.2020.7849 (2021).
Li, J. et al. Cbx4 governs HIF-1α to potentiate angiogenesis of hepatocellular carcinoma by its SUMO E3 ligase activity. Cancer Cell 25, 118-131, doi:10.1016/j.ccr.2013.12.008 (2014).
Mei, Z. et al. Polycomb chromobox 4 enhances migration and pulmonary metastasis of hepatocellular carcinoma cell line MHCC97L. Sci China Life Sci 57, 610-617, doi:10.1007/s11427-014-4663-9 (2014).
Yang, M. et al. MiR-497∼195 cluster regulates angiogenesis during coupling with osteogenesis by maintaining endothelial Notch and HIF-1α activity. Nat Commun 8, 16003, doi:10.1038/ncomms16003 (2017).
Liu, M. et al. YTH domain family: potential prognostic targets and immune-associated biomarkers in hepatocellular carcinoma. Aging (Albany NY) 13, 24205-24218, doi:10.18632/aging.203674 (2021).
Zhou, R. et al. A functional loop between YTH domain family protein YTHDF3 mediated m(6)A modification and phosphofructokinase PFKL in glycolysis of hepatocellular carcinoma. J Exp Clin Cancer Res 41, 334, doi:10.1186/s13046-022-02538-4 (2022).
Yang, B., Zang, J., Yuan, W., Jiang, X. & Zhang, F. The miR-136-5p/ROCK1 axis suppresses invasion and migration, and enhances cisplatin sensitivity in head and neck cancer cells. Exp Ther Med 21, 317, doi:10.3892/etm.2021.9748 (2021).
Zhong, J. X. et al. Circular RNA circ-ERBB2 promotes HER2-positive breast cancer progression and metastasis via sponging miR-136-5p and miR-198. J Transl Med 19, 455, doi:10.1186/s12967-021-03114-8 (2021).
Mao, Y. et al. Circ_0091579 exerts an oncogenic role in hepatocellular carcinoma via mediating miR-136-5p/TRIM27. Biomed J 45, 883-895, doi:10.1016/j.bj.2021.12.009 (2022).
Zhao, W. et al. Inhibiting CBX4 efficiently protects hepatocellular carcinoma cells against sorafenib resistance. Br J Cancer 124, 1237-1248, doi:10.1038/s41416-020-01240-6 (2021).
Jiao, H. K. et al. Prognostic significance of Cbx4 expression and its beneficial effect for transarterial chemoembolization in hepatocellular carcinoma. Cell Death Dis 6, e1689, doi:10.1038/cddis.2015.57 (2015).
Dou, Z. et al. MicroRNA-6838-5p suppresses the self-renewal and metastasis of human liver cancer stem cells through downregulating CBX4 expression and inactivating ERK signaling. Biol Chem 404, 29-39, doi:10.1515/hsz-2022-0156 (2023).

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

CircYTHDF3 promotes hepatocellular carcinoma progression through modulating miR-136-5p/CBX4/ VEGF axis

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

HCC specimens and microarray of circRNAs

RNA preparation, reverse transcription, and quantitative real-time PCR analysis

CBX4-F: CCTTCCAGAACAGGGAACGG

CBX4-R: TCAGGACATTGGAACGACGG

VEGF-F: GCAGAATCATCACGAAGTGG

CCK8 assay and EdU assay

Wound-healing assay and Transwell assay

Fluorescence in situ hybridization (FISH)

MiRNA and mRNA prediction

Luciferase reporter assay

HCC animal model

H&E staining and Immunohistochemical (IHC) staining

Western blot analysis

Statistical analysis

Results

2. The characteristics of circYTHDF3 in HCC cells

Discussion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1