Introduction
Primary liver cancer is the second most common malignant tumor worldwide, and hepatocellular carcinoma (HCC) is its most common pathological type.1 Because of the high rate of hepatitis B virus infection, China accounts for nearly half of newly diagnosed HCC cases and HCC-related deaths worldwide.2 Although several treatments, such as surgical resection, liver transplantation, chemotherapy, radiation therapy, immunotherapy, and molecular targeted therapy, have shown survival benefit, the prognosis of HCC remains poor, particularly due to challenges with early diagnosis and the recurrence and metastasis of HCC cells.3 Therefore, there is an urgent need to understand the mechanisms of HCC and develop new treatment strategies.
Circular RNAs (circRNAs) are a class of noncoding RNAs (ncRNAs) that feature a reverse splicing closed loop. circRNAs mediate multiple biological functions, and they are reportedly involved in tumorigenesis and development.4 circRNAs can act as endogenous competitive inhibitors by sponging miRNAs and interacting with RNA-binding proteins to regulate transcription.5 However, their role in HCC and their specific molecular mechanisms warrant further investigation.
HDAC1, a member of the histone deacetylase (HDAC) family, mediates epigenetic regulation, which plays an important role in normal development and tumor progression.6 Presently, there is ample literature supporting the promotion effect of HDAC1 in HCC. In cancer cells, HDAC1 inhibits the tumor suppressor genes p21WAF1/CIP1 and Bax, resulting in abnormal cell proliferation and cell viability.7,8 The impact of HDAC1 on HCC progression has been widely acknowledged. For example, HDAC1 knockdown reduces cyclin-dependent kinase (CDK) levels and inhibits HCC cell proliferation.9 HDAC1 regulates HCC in complex and diverse ways, and these distinct mechanisms require further investigation. Through database analysis, we found that HDAC1 transcribed circRNAs. The function of HDAC1-derived circRNAs in HCC has not yet been investigated.
Exosomes are small nanoscale vesicles that transport bioactive molecules between cells and regulate the intercellular microenvironment and immune system.10 Studies have shown that exosomes, as mediators of cell-to-cell communication, are involved in tumor development, metastasis, immune evasion, and drug resistance.11 An increasing number of studies have shown that circRNAs in exosomes are abundant and stable, but the function of circRNAs in exosomes remains unclear.
In this study, we showed that circHDAC1_004 expression was higher in HCC than in paraneoplastic tissues. The increased expression promoted HCC proliferation, metastasis, and stemness through the miR-361-3p/NACC1 axis. In addition, circHDAC1_004 could be transferred from HCC cells to human umbilical vein endothelial cells (HUVECs) by exosomes to promote HCC angiogenesis.
Methods
HCC samples
The HCC and paracancerous tissues used in this study were obtained from the Hepatobiliary Center of the First Affiliated Hospital of Nanjing Medical University. All human tissue samples were obtained with patients’ written consent. This study was approved by the Ethics Committee of the First Affiliated Hospital of Nanjing Medical University.
Cell lines
HCC cell lines Huh7, MHCC97H, HepG2, Hep3B, MHCCLM3, SK-Hep1, YY8103, a normal human liver cell line (HHL-5), and HUVECs were purchased from the Shanghai Institute of Cell Biology, Chinese Academy of Sciences (Shanghai, China). All cell lines were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Gibco, Carlsbad, CA, USA) medium with 10% fetal bovine serum (Gibco), 50 U/mL penicillin (Invitrogen, Waltham, MA, USA), and 50 U/mL streptomycin (Invitrogen). All cells were cultured in a 5% CO2 incubator at 37°C.
Quantitative real-time polymerase chain reaction (qRT-PCR)
RNA was extracted from cells or tissues using TRIzol reagent (Invitrogen). RNA concentration was measured using a spectrophotometer. For qRT-PCR detection of circular and mRNA, the RNA was reverse-transcribed to cDNA using a reverse transcription kit (VA zyme, Nanjing, China). qRT-PCR was performed using Ace qPCR SYBR Green Master Mix (VA zyme) with an ABI 7900 PCR system (Applied Biosystems Inc., Waltham, MA, USA), and GAPDH was used as the negative control.
For qRT-PCR detection of miRNA, miDETECT A Track miRNA qRT-PCR primer (Ribobio, Guangzhou, China) and miDETECT A Tract miRNA qRT-PCR Starter Kits (Ribobio) were used. In brief, total RNA was added with ploy(A) tailing and then reverse-transcribed to cDNA. Specific qRT-PCR was performed with miDETECT A Track miR-361-3p and miR-194-5p qPCR primer and miDETECT A Track Uni-Reverse Primer. U6 was used as the negative control. Primers used in this study are enlisted in the Supplemental Material (Supplementary Table 1).
Agarose gel electrophoresis
Divergent and convergent primers (Ribobio) of circHDAC1_004 were used to amplify cDNA and gDNA of HCC cell lines, respectively, and the amplified products were collected. A 1× Tris-acetate-EDTA buffer (Beyotime, Nantong, China) solution was used to prepare 1% agarose gels that were boiled and cooled and nucleic acid stain was added. The above products were added into the agarose gel, and 1× TAE was used as the electrophoresis solution, which was performed at 100 V for 1 h, and the position of the bands was observed under ultraviolet lamp.
Cell proliferation, migration, and invasion assays
Cell proliferation, migration, and invasion assays were performed as described in the Supplementary Methods (Supplementary File 1).
Fluorescence in situ hybridization (FISH)
FISH was performed with a commercially available kit (Ribobio). The appropriate number of cells was added to each well of a 24-well plate. After the cells were fixed with 4% paraformaldehyde, the cell membrane was disrupted with 0.5% Triton X-100. Cells were blocked with prehybridization solution, and 100 µL of hybridization solution containing circHDAC1_004 FISH probe, 18S probe, and U6 probe (Ribobio) was added to each well and was protected from the light. The cells were hybridized overnight at 37°C and counterstained for 10 m with 4′,6-diamidino-2-phenylindole (DAPI). Images were taken with a confocal laser microscope (Stellaris STED; Leica, Frankfurt, Germany).
Sphere formation
Experimental cells (2,000 cells/well) were seeded into low adhesion 6-well plates. Herein, 2 mL of stem cell medium [DMEM/F12 (Gibco) supplemented with 1XB27 (Invitrogen), 20-ng/mL FGF (Gibco), 20-ng/mL EGFI (Gibco), and 4-µg/mL heparin (Selleck, Houston, TX, USA)] was added to each well, and the cells were cultured for 10 days at 37°C in an incubator containing 5% CO2. The state of cell pelleting was observed, and the number of pelleted cells was calculated.
Immunofluorescence
A slide was placed at the bottom of the 24-well plate, and 40,000 treated cells were added to each well. When the cells had attached to the plate surface, they were fixed in paraformaldehyde, lysed with Triton, and blocked with goat serum for 30 m. The cells were incubated with primary antibody overnight. After incubation with a fluorescent secondary antibody (Beyotime) for 1 h, the nuclei were labeled with DAPI (Beyotime). Images were taken using a confocal microscope.
Xenograft nude mouse model
All animal experiments in this study were approved by the Institutional Animal Care and Use Committee (IACUC) of the First Affiliated Hospital of Nanjing Medical University. All animal-related operations were performed according to the IACUC operating guidelines. Four-week-old male BLAB/C nude mice (Vital River, Beijing, China) were divided into four groups, with six mice in each group. Then, 5 million lentivirus-transfected cells were injected into the axilla of the left upper limb of each mouse. Subcutaneous tumor volume was recorded every 3 days for 30 days, and then the mice were euthanized. Subcutaneous tumors were removed for immunohistochemistry and volume measurements.
Pulmonary metastasis model
Four-week-old male BLAB/C nude mice (Vital River) were divided into four groups, with six mice in each group. Then, 1.5 million experimental cells were injected into the mice via the tail vein. Six weeks later, the mice were euthanized and their lungs were collected for photography and hematoxylin and eosin (HE) staining.
Exosome extraction
Exosomes were extracted by gradient centrifugation. Briefly, the cell supernatant was placed in a clean centrifuge tube (Beckman, Brea, CA, USA) and centrifuged at 500 g for 10 m, 2,000 g for 20 m, and 10,000 g for 30 m. After each centrifugation, the supernatant was removed and placed in a clean Beckman tube. Samples were then centrifuged twice at 110,000 g for 1 h. The precipitated exosomes were resuspended in phosphate-buffered saline.
Exosome identification
Extracted exosomes were imaged and verified by a projection electron microscope (JEM-1010; JEM, JEOL, Japan) at the Analysis and Testing Center of Nanjing Medical University. The particle size distribution range of exosomes was examined by particle point titration and using a particle size analyzer to judge the purity and concentration of the extracted exosomes. Proteins were extracted from these exosomes, and the exosome-related proteins CD81, CD9, and TSG101 were detected by western blotting.
Exosome uptake
A total of 1 mL of preprepared working solution PKH67 (PKH67: diluent=1:250; War bio, Nanjing, China) was added to the exosome suspension. Samples were incubated for 3 min at room temperature. Then, exosomes were precipitated by centrifugation at 110,000 g for 70 min. Exosomes were suspended and incubated with cells for 12 h. After fixation with paraformaldehyde, nuclei were labeled with DALI and imaged under a laser confocal microscope (LSM5 Live; Zeiss, Overcoached, Germany).
Tubule formation
Fifty microliters of Matrigel matrix (BD Biosciences, Franklin Lakes, NJ, USA) was added to each well of a 96-well plate, and the plate was placed in the incubator at 37°C until the Matrigel matrix solidified on the plate. HUVECs were resuspended at 40,000/mL in exosome-free medium, and 50 µM of the cell suspension was added to each well. After 8 h, tubules of HUVEC cells were observed by light microscopy, and the experimental results were photographed.
RNA immunoprecipitation (RIP)
Cell lysates were incubated overnight with magnetic beads (Genesee, Guangzhou, China) coupled with the Ago2 antibody or IgG at 4°C. Then, the beads were washed and incubated with protease K to remove proteins. Finally, RNA was extracted, and qRT-PCR was performed to determine whether circHDAC1_004 and miR-361-3p had bind with Ago2.
Pull-down assay
A biotin-labeled circHDAC1_004 probe and its negative control sequence were synthesized in vitro. The RNA-RNA complex was captured using a Pierce magnetic RNA-protein pull-down kit (Thermo Fisher Scientific, Waltham, MA, USA). HCC cell lines overexpressing circHDAC1_004 were lysed and the supernatant was extracted, and DNase was added to digest the DNA. The circHDAC1_004 probe and NC probe were added to the two supernatants for hybridization, and magnetic beads were collected using a magnetic frame. RNA on the magnetic beads was eluted, and purified RNA was reverse-transcribed into cDNA; the data were analyzed after qRT-PCR.
Dual-luciferase reporter assay
The binding site of circHDAC1_004 with miR-361-3p were predicted using Mir DIP, and the binding site of miR-361-3p with NACC1 3′ UTR was predicted by Targets can. Then, wild-type (WT) and mutant plasmids (MUT) were designed. The WT or mutant vectors were cotransfected into HEK-293T cells with miRNA mimics using transfection reagents. Then, 5 ng of sea kidney luciferase vector was added and incubated for 2 h, followed by the addition of 150 µM of passivation lysate (PLY) to Petri dishes and incubated for 20 m on ice. Luciferase activity was measured and quantified using a dual-luciferase reporter gene assay system.
Western blotting
Cells were lysed with a radio-immunoprecipitation assay (Beyotime). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was used to isolate and transfer proteins to a polyvinylidene fluoride membrane (Merck Millipore, Burlington, MA, USA). The samples were blocked with 5% skim milk powder for 2 h and incubated in the primary antibody overnight. After washing with Tris-buffered saline and Tween20 (TBST) three times, the secondary antibody was added and the membranes were incubated for 2 h at room temperature. Protein content was quantified using a hypersensitive enhanced chemiluminescence (ECL) exposure solution and Image Lab software (Bio-Rad, Hercules, CA, USA). The primary antibodies used were GAPDH (Proteinates, Wuhan, China), CD9 (Proteinates), CD81 (Invitrogen), TSG101 (Invitrogen), E-cadherin (Proteinates), N-cadherin (Proteintech), Vimentin (Service bio), and NACC1 (Proteintech).
Statistical analysis
The results were reported as means ± SD, and p<0.05 was considered significant. The statistical analysis was performed with GraphPad Prism v8.0 (GraphPad, San Diego, CA, USA). Experiments with multiple comparisons were compared between two groups using one-way analysis of variance. Normally distributed parameters were tested with the unpaired Student’s t-test, and non-normally distributed parameters were tested with the Mann-Whitney test. Correlations were tested using the Spearman’s correlation. Survival curves were drawn by the Kaplan-Meier method.
Results
circHDAC1_004 was upregulated in HCC and negatively correlated with the prognosis of HCC patients
HDAC1 is reported to be an important oncogene in HCC. It promotes cell proliferation and regulates the cell cycle of HCC.14 To further investigate the role of HDAC1-derived circRNAs in HCC, primers for 14 HDAC1-derived circRNAs were designed. Preliminary screening of HDAC1-derived circRNAs in five pairs of HCC and paracancerous tissues by qRT-PCR revealed that five circRNAs (circHDAC1_002, circHDAC1_004, circHDAC1_008, circHDAC1_012, and circHDAC1_014) were upregulated in HCC (Fig. 1A). We assessed the expression of the circRNAs in 80 pairs of HCC and paraneoplastic tissues, and circHDAC1_004 expression was found to be significantly upregulated (Fig. 1B–F). Clinicopathological analysis revealed that circHDAC1_004 expression was positively correlated with tumor volume, the TNM stage, and microvascular invasion in HCC patients (Table 1). As circHDAC1_004 showed the most significant difference, its transcription was confirmed in the validation set (Fig. 1G). Meanwhile, Kaplan-Meier survival analysis showed that patients with higher levels of circHDAC1_004 had shorter overall survival (Fig. 1H).
Table 1Relationship between circHDAC1_004 expression level in HCC tissues and clinical parameters of patients
| Clinicopathologic feature | Total | CircHDAC1_004
| p value |
|---|
| 80 | High (40) | Low (40) |
|---|
| Age | | | | |
| >60 | 48 | 21 | 27 | 0.2537 |
| ≤60 | 32 | 19 | 13 | |
| Sex | | | | |
| Male | 40 | 21 | 19 | 0.8233 |
| Female | 40 | 19 | 21 | |
| HBsAg | | | | |
| Negative | 15 | 8 | 7 | 0.9999 |
| Positive | 65 | 32 | 33 | |
| AFP | | | | |
| ≤200 | 18 | 11 | 7 | 0.4225 |
| >200 | 62 | 29 | 33 | |
| Tumor number | | | | |
| Single | 52 | 19 | 33 | 0.0020 |
| Multiple | 28 | 21 | 7 | |
| Tumor size in cm | | | | |
| ≤5 | 46 | 15 | 31 | 0.0006 |
| >5 | 34 | 25 | 9 | |
| TNM stage | | | | |
| I | 47 | 18 | 29 | 0.0225 |
| II–III | 33 | 22 | 11 | |
| Microvascular invasion | | | | |
| Yes | 28 | 19 | 9 | 0.0340 |
| No | 52 | 21 | 31 | 0.0024 |
| Edmondson stage | | | | |
| I–II | 50 | 18 | 32 | |
| III–IV | 30 | 22 | 8 | |
Notably, circHDAC1_004, also known as circ_0005339, is derived from exons 5–6 of the HDAC1 gene (Fig. 2A). Sanger sequencing was performed to check for head-to-tail splicing of circHDAC1_004 (Fig. 2B). We also designed divergent and convergent primers to verify the circular structure of circHDAC1_004. Agarose gel electrophoresis showed that circHDAC1_004 amplified by convergent primers could be detected in both cDNA and gDNA, but circHDAC1_004 amplified by divergent primers was only detected in cDNA (Fig. 2C). Furthermore, the RNase R treatment indicated that circHDAC1_004 was more resistant to RNase R digestion than linear RNA (Fig. 2D). To confirm the distribution of circHDAC1_004, nucleoplasm separation experiments and FISH were performed. circHDAC1_004 was primarily expressed in the cytoplasm (Fig. 2E, F).
circHDAC1_004 promoted proliferation, migration, schizogenesis, and the EMT pathway of HCC cells
To further explore the function of circHDAC1_004 in HCC development, we downregulated circHDAC1_004 in SK-Hep1 cells and upregulated circHDAC1_004 in YY8103 cells by lentiviral transfection, as SK-Hep1 had the highest and YY8103 had the lowest circHDAC1_004 expression of the tested HCC cell lines (Fig. 3A, B), and the sequence of siRNA is shown in Supplementary Table 2. The results of CCK-8, colony formation, and EdU assays indicated that circHDAC1_004 inactivation suppressed the proliferation of SK-Hep1 cells, and circHDAC1_004 overexpression enhanced the proliferation of YY8103 cells (Fig. 3C–E). Transwell and wound healing assays showed that circHDAC1_004 inactivation suppressed the migration and invasion of SK-Hep1 cells, whereas circHDAC1_004 overexpression had the opposite effects (Fig. 3F, G). Moreover, a stem-cell pelleting assay showed that circHDAC1_004 inactivation inhibited the spherogenesis ability of SK-Hep1 cells and that upregulated circHDAC1_004 expression promoted YY8103 cell pelleting (Fig. 3H).
Epithelial-mesenchymal transformation (EMT) is a morphogenetic process associated with tumor aggressiveness, metastasis, and chemical resistance to malignancies.12 Western blotting (Fig. 3I) and immunofluorescence (Supplementary Fig. 1A) showed that the abundance of E-cadherin was significantly increased by circHDAC1_004 knockdown and decreased by circHDAC1_004 overexpression; meanwhile, the expression of N-cadherin and vimentin was significantly decreased by circHDAC1_004 knockdown and increased by circHDAC1_004 overexpression. The results suggest that circHDAC1_004 promoted the EMT pathway in HCC.
circHDAC1_004 promoted HCC cell proliferation, angiogenesis, and metastasis in vivo
We investigated the effect of circHDAC1_004 in vivo by constructing a subcutaneous tumor model and nude mouse lung metastatic tumor model. Subcutaneous tumor model experiments showed that the downregulation of circHDAC1_004 inhibited tumor growth and its overexpression promoted tumor growth (Fig. 4A). In addition, immunohistochemical staining of subcutaneous tumors showed that Ki67, CD31, and vimentin were downregulated, and E-cadherin was upregulated in circHDAC1_004 knockdown SK-Hep1 cells. Conversely, circHDAC1_004 overexpression had the opposite result, further confirming that it promotes cell proliferation, HCC angiogenesis, and the EMT pathway in vivo (Fig. 4B). In the nude mouse lung metastases model, circHDAC1_004 knockdown resulted in fewer and smaller lung metastases, whereas its overexpression resulted in more and larger lung metastases (Fig. 4C). Collectively, these results suggest circHDAC1_004 facilitated the proliferation, angiogenesis, and metastasis of HCC cells in vivo.
circHDAC1_004 was transferred from HCC cells to HUVECs through exosomes and promoted the angiogenesis of HCC
Exosomes act as carriers to transport various bioactive substances between cells, and circRNAs can reportedly be transferred between cells by exosomes.13,14 Exosomes were extracted from SK-Hep1 and YY8103 cells and detected by transmission electron microscopy (Fig. 5A). Nanoparticle tracking analysis showed that the peak size of exosomes was 50–150 nm (Fig. 5B). Meanwhile, the exosome markers CD81, CD9, and TSG101 were detected by western blotting (Fig. 5C). Agarose gel electrophoresis showed that circHDAC1_004 could be detected in exosomes (Fig. 5D).
Tumor growth depends on angiogenesis,15 for which exosome-derived circRNAs are important.13 To investigate the function of exosomal circHDAC1_004 in angiogenesis, we exposed HUVECs to exosomes isolated from SK-Hep1 and YY8103 cells with circHDAC1_004 knockdown or overexpression. The uptake of PKH67-labeled exosomes by HUVECs was examined by confocal laser microscopy (Fig. 5E). PCR results showed that circHDAC1_004 expression was downregulated in HUVECs co-cultured with circHDAC1_004 knockdown exosomes and upregulated in HUVECs co-cultured with circHDAC1_004 overexpression exosomes (Fig. 5F). Exosomes derived from circHDAC1_004 knockdown SK-Hep1 cells inhibited the proliferation of HUVECs after 48 h of coculture, and those derived from circHDAC1_004-overexpressing YY8103 cells promoted HUVEC proliferation (Supplementary Fig. 2A, B). The Transwell assay demonstrated that exosomes with low circHDAC1_004 expression inhibited the ability of HUVECs to migrate compared to controls, whereas exosomes with overexpressed circHDAC1_004 promoted HUVEC migration (Fig. 5G). In addition, exosomes with circHDAC1_004 knockdown inhibited the formation of HUVEC tubes, and those with circHDAC1_004 overexpression had the opposite result (Fig. 5H). Furthermore, the chick embryo chorioallantoic membrane assay showed that exosomes with circHDAC1_004 overexpression promoted angiogenesis in the membrane when compared with the control group, and exosomes with downregulated circHDAC1_004 showed the opposite result (Fig. 5I). The results suggest that circHDAC1_004 is transferred from HCC cells to HUVECs through exosomes and promotes HCC angiogenesis.
circHDAC1_004 acts as a sponge of miR-361-3p
Notably, circRNAs frequently function as miRNA sponges to regulate gene expression.16 We predicted the downstream miRNAs of circHADC1_004 using the CircBank, Circinteractome, and Starbase databases. miR-361-3p and miR-194-5p have potential binding sites on circHDAC1_004 (Fig. 6A). PCR results showed that miR-361-3p was downregulated in circHDAC1_004-overexpressing YY8103 cells and upregulated in circHDAC1_004 knockdown SK-Hep1 cells, whereas miR-194-5p expression did not differ between circHDAC1_004 knockdown and overexpression groups (Fig. 6B, Supplementary Fig. 3A). Meanwhile, miR-361-3p expression was downregulated in HCC tissues and was negatively correlated with circHDAC1_004 expression in 80 HCC tissues (Fig. 6C, D). miR-361-3p is an oncogene of HCC, and its expression was negatively correlated with the overall survival of HCC patients (Fig. 6E). Therefore, we speculated that it was a downstream miRNA of circHDAC1_004. Furthermore, we designed a probe for circHDAC1_004. A pull-down assay revealed that miR-361-3p could be pulled down (Fig. 6F) by the circHDAC1_004 probe, and the RNA immunoprecipitation (RIP) assay indicated that both circHDAC1_004 and miR-361-3p was enriched by the Ago2 antibody (Fig. 6G). To further explore the binding sites of circHDAC1_004 and miR-361-3p, we designed WT and mutant circHDAC1_004 vectors with luciferase according to the predicted binding sites (Fig. 6H). The luciferase activity of WT vectors was significantly reduced after coculturing with miR-361-3p mimics compared to mutant vectors (Fig. 6I). These results suggest that circHDAC1_004 acted as a sponge for miR-361-3p.
NACC1 is a target gene of miR-361-3p
Downstream target genes of miR-361-3p were predicted using Starbase, miRDIP, miRWalk, and miRDB, and the results of Venn diagram cross-linking suggested that miR-361-3p affects NACC1 expression (Fig. 7A). qRT-PCR results showed that NACC1 expression was upregulated in HCC tissues (Fig. 7B). Spearman correlation analysis indicated that NACC1 expression was negatively correlated with miR-361-3p (Fig. 7C) and positively correlated with circHDAC1_004 (Fig. 7D). NACC1 expression was increased after miR-361-3p knockdown and decreased after miR-361-3p overexpression (Fig. 7E, F). Meanwhile, NACC1 expression decreased after circHDAC1_004 knockdown and increased after circHDAC1_004 overexpression (Fig. 7G, H). We performed a dual-luciferase reporter assay to determine the binding sites of miR-361-3p and NACC1-3′ UTR. WT and mutant NACC1-3′ UTR vectors with luciferase were designed according to the Targetscan database (Supplementary Fig. 4). The luciferase activity was significantly reduced in of miR-361-3p and NACC1 3′ UTR-WT co-transfected cells but remained unchanged in miR-361-3p and NACC1 3′ UTR-MUT cotransfected cells (Fig. 7I).
circHDAC1_004 promoted HCC progression through the miR-361-3p/NACC1 axis
To further explore the function of the circHDAC1_004/miR-361-3p/NACC1 axis in HCC progression, miR-361-3p inhibitor or NACC1 overexpression vector were co-transfected in circHDAC1_004 knockdown SK-Hep1 cells. Colony formation and 5-ethynyl-2′-deoxyuridine (EdU) assays showed that miR-361-3p knockdown and NACC1 overexpression restored the inhibitory effect of circHDAC1_004 knockdown on HCC proliferation (Fig. 8A, Supplementary Fig. 5A). Transwell and wound healing assays revealed that both miR-361-3p inhibition and NACC1 overexpression effectively reversed the inhibitory effect of circHDAC1_004 knockdown on HCC migration and invasion (Fig. 8B, Supplementary Fig. 5B). In addition, a stem cell sphere formation assay showed that the number of stem cell spheres reduced after circHDAC1_004 knockdown but miR-361-3p knockdown or NACC1 overexpression reversed this phenomenon (Fig. 8C). In addition, Si-circHDAC1_004 downregulated the EMT-related proteins N-cadherin and vimentin; upregulation of E-cadherin was also reversed by miR-361-3p inhibitor and NACC1 overexpression (Fig. 8D). The results suggested that circHDAC1_004 increased NACC1 expression through competitive inhibition of inhibition of miR-361-3p, promoted the EMT pathway, and promoted HCC cell proliferation, migration, invasion, and stemness.
Discussion
Numerous studies have carried out in-depth analysis of the functions of HDAC1. Histone deacetylases (HDACs) catalyze the removal of acetyl groups from ε-amino groups of lysine residues. Reversible acetylation of histone and nonhistone proteins by histone acetyltransferases (HATS) and histone deacetylases (HDACs) have key roles in transcriptional regulation in eukaryotic cells.17 HDAC1 has novel functions in DNA replication and repair that contribute to chemotherapy resistance in cancer cells.18 It has also been investigated in several HCC studies. It has been reported to inhibit FAM99A expression through histone H3 deacetylation transcription under hypoxic conditions, thus promoting HCC metastasis and EMT.19 HDAC1 can also be used as a prognostic marker of HCC, which has significance for clinical treatment.20 Further investigation of its mechanisms is needed because of the complex and diverse ways in which it regulates HCC. In the present study, we verified the expression and effect of 14 circRNAs derived from HDAC1 in HCC tissues.
Notably, circRNAs have received much attention for their potential diagnostic and therapeutic value; many studies have demonstrated their role in the diagnosis and treatment of HCC.21,22 Huang et al.23 showed that circMET drives immunosuppression and anti-PD1 treatment resistance in HCC via the miR-30-5p/SNAIL/DPP4 axis. Liu et al.24 reported that the circRNA cIARS regulates iron death in HCC cells by interacting with the RNA-binding protein ALKBH530. Multiple signaling pathways, such as the Wnt,25,26 YAP,27 and PI3K/AKT/mTOR28 signaling pathways, are involved in the development and progression of hepatocellular carcinoma. circRNAs are involved in various signaling pathways, such as the Wnt signaling pathway, to influence the development and progression of HCC.29 Autophagy plays a pivotal role in tumorigenesis, metastasis, targeted therapy, and drug resistance in HCC.30 Some studies have proved that circRNAs can trigger autophagy of HCC cells to affect disease progression. For example, Zhao et al.31 demonstrated that circCBFB affects the progression of HCC by triggering cell autophagy through the miR-424-5p/ATG14 axis. To our knowledge, no studies have been conducted on circRNAs transcribed by HDAC1 thus far, and we explored whether these circRNAs have biological functions in HCC. In this study, we found that circHDAC1_004 expression was upregulated in HCC tissues and negatively correlated with HCC prognosis. In vitro experiments showed that circHDAC1_004 knockdown inhibited cell proliferation, migration, stemness, and the EMT pathway, which were enhanced after circHDAC1_004 overexpression. In vivo experiments further demonstrated that circHDAC1_004 promoted the proliferation and metastasis of HCC cells. Thus, circHDAC1_004 had a carcinogenic role in HCC and was involved in the occurrence and progression of HCC.
Angiogenesis is considered a hallmark of cancer, and significant vascular abnormalities are an important cause of liver injury leading to HCC11. Some studies have explored the effects of circRNAs in exosomes on HCC angiogenesis.32 Huang et al.13 found that exosome-derived circRNA-100338 promoted HCC angiogenesis by binding with RNA-binding proteins. We found that circHDAC1_004 were secreted into the HCC tumor microenvironment by exosomes. Coculture of HUVECs with exosomes with circHDAC1_004 knockdown or overexpression revealed that exosomal circHDAC1_004 promoted tube formation, proliferation, and migration of HUVECs. These results indicate that circHDAC1_004 upregulation in HCC cells promoted HCC angiogenesis through exosomes.
Next, we discuss the mechanism of circHDAC1_004 in HCC. Many studies have shown that the circRNA-miRNA-mRNA interaction mechanism significantly affects the occurrence and development of different cancers, including HCC.33 Zhang et al.34 found that circRNA103809 promoted the malignant progression of HCC through the miR-377-3p/FGFR1/ERK axis. The nucleoplasmic separation and FISH assays revealed that circHDAC1_004 is mainly located in the cytoplasm. We identified miR-361-3p to be a downstream miRNA of circHDAC1_004 in HCC and further explored the downstream target genes of circHDAC1_004/miR-361-3p. We also found that NACC1 expression was influenced by miR-361-3p. It was promoted by circHDAC1_004 and inhibited by miR-361-3p. NACC1 has a cancer-promoting role in various malignant tumors and has potential diagnostic and therapeutic value.35,36 Yin et al.37 identified NACC1 as an oncogene in HCC. Rescue studies have reported that both miR-361-3p inhibitor and NACC1 overexpression could rescue the inhibition of circHDAC1_004 knockdown on cell proliferation, migration, invasion, cell stemness, and the EMT pathway in HCC. All the above data suggest that the circHDAC1_004/miR-361-3p/NACC1 axis plays a key role in HCC.
Although there are many methods for the diagnosis and treatment of HCC, their clinical efficacy is not ideal, and we therefore urgently need new avenues to improve the existing means of diagnosis and treatment. Exosomes are rich biomarkers for disease diagnosis and prognosis. Recently, exosomes have been widely used in the diagnosis of cancer.38 At present, various molecular targeted drugs have been applied in the clinical treatment of HCC, such as sorafenib and Lenvatinib. however, not all HCC patients have good results with these drugs.39 Considering the important role of circHDAC1_004 in the occurrence and progression of HCC, we believe that circHDAC1_004 can be used as a diagnostic marker for HCC. In addition, circHDAC1_004 may also be a new target for HCC.