Circular RNA SETD2 (circSETD2) elevates DDX3 expression to inhibit autophagy in hepatitis B virus infection via targeting miR-181a-5p

Abstract

Background

Hepatitis B virus (HBV) is one of the major causes for chronic liver disease leading to cirrhosis and hepatocellular carcinoma (HCC). It utilizes the host cell machinery, such as host factors and non-coding RNAs, to modulate its survival and proliferation. CircRNAs, a special class of non-coding RNAs, play a crucial role in regulating HBV infection. A recent study reported that circular RNA SET domain containing 2 (circSETD2) was down-regulated in HCC, however, there is no study available on its role in modulating HBV-infection.

Methods

The hepatic cell lines were cultured in DMEM with 10% FBS and antibiotics. HepG2 cells were transfected with empty vector, HBV or HBx expression plasmids. HepG2.2.15 cells were transfected with non-specific RNA, circSETD2, miR-181a-5p mimic or miR-181a-5p inhibitor. The expression of circSETD2 and miR-181a-5p were determined by qPCR, and Western blots were performed for analyzing autophagic marker proteins. Autophagy flux assays were performed in circSETD2, miR-181a-5p mimic or miR-181a-5p inhibitor-transfected cells. Significance of the data were analyzed using unpaired student t-test.

Results

The expression of circSETD2 was significantly lower (76%) in HepG2.215 cells, compared to non-HBV infected cells. Overexpression of HBV and HBx in HepG2 cells significantly reduced the expression of circSETD2. CircSETD2 overexpression decreased the HBsAg levels and the expression of miR-181a-5p. Overexpression of miR-181a-5p decreased Dead-Box RNA Helicase 3 (DDX3) expression and induced autophagy proteins. Inhibition of miR-181a-5p enhanced the expression of DDX3, which led to decreased autophagy. Overexpression of circSETD2 increased the expression of DDX3 and decreased autophagy proteins.

Conclusion

The data showed that HBx suppressed circSETD2 to induce miR-181a-5p expression, which in turn resulted in the decreased DDX3, leading to the increased autophagy in HBV-infected hepatic cells.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-025-12465-2.

Introduction

Hepatitis B virus (HBV) causes a serious health burden in the developing countries. Globally, 350 million people are chronically infected by HBV, which puts them at high risk of fibrosis, liver cirrhosis and hepatocellular carcinoma (HCC) [1, 2]. HBV infects human hepatocytes through the NTCP receptor and converts 3.2 kb relaxed double-stranded DNA into covalently closed circular DNA (cccDNA) for persistent infection [3]. HBV modulates the expression of various host genes [4], non-coding RNAs [5], and host cellular pathways through insertional mutagenesis for its survival and proliferation [6, 7]. Additionally, the viral proteins, such as HBx, HBc and PreS/S modulates the expression of host restriction factors that are associated with the cell cycle, apoptosis and autophagy pathways. Over persistent infection, HBV through insertional mutagenesis, triggers the genomic instability to induce carcinogenesis in hepatic cells leading to HCC [8].

Autophagy is a self-degradation mechanism utilised by cells to maintain cellular homeostasis. It is established that HBV induces the early autophagy pathway to facilitate its replication; however, it evades the autophagolysosome formation [9]. Mechanistic studies reported that HBx induce Beclin1 expression and the enzymatic activity of class III phosphatidylinositol 3-kinase (PI3KC3) to enhance phosphatidylinositol 3-phosphate (PI3P)-mediated autophagosome formation [10, 11]. Although it is established that HBV induces an autophagy mechanism, its molecular insight is still less understood.

Circular RNAs (circRNAs) are covalently closed RNA lacking a 5’-cap and poly (A) tail, which were initially identified in plants [12]. They are associated with various cellular physiological processes and gene expression via regulating miRNA functions or interacting with RNA-binding proteins [13]. CircRNAs act as competing endogenous RNAs (ceRNAs) by sequestering microRNAs (miRNAs) from their binding to the 3’-UTR of the target messenger RNAs (mRNAs). Abnormal expression of circRNAs are found in various tumor progression and viral infections. Several circRNAs were reported to play a key role in the modulation of hepatocellular carcinoma (HCC) [14, 15]. Several studies reported that Hepatitis B virus infection causes aberrant expression of circRNAs in the liver [16–18]. The circATP5H [19], circ-ARL3 [20] and circBACH1 [21] were found to increase HBV replication and hepatoma development. Additionally, hsa_circ_0000976, hsa_circ_0007750, hsa_circ_0139897 and hsa_circ_0027089 could be potential biomarkers of HBV infection [18, 22]. Although efforts have been made to understand the involvement of circular RNA in HBV survival and pathogenesis, majority of the circular RNAs have been unexplored in HBV infection. Hsa_circ_0065173 - circRNA SET domain containing 2 (circSETD2), is a 4644 bp long non-coding RNA, originating from exon 2 to exon 5 of the host gene SET domain containing 2 (SETD2), reported to be important for hepatocellular carcinoma [23]. It inhibits hepatocellular carcinoma, breast cancer and trophoblast progression via interacting with miRNAs [20, 23–25].

miRNAs are 22-24nt non-coding entities which majorly involve in post-transcriptional modulation of genes. There are plenty of miRNAs that are reported to modulate HBV replication and HCC progression [26, 27]. miR-181a-5p, is one of the abundant miRNAs expressed in liver which is reported to be modulated during HBV infection [28, 29]. It is revealed that miR-181a-5p targets PTEN [30], host restriction factor ALR [31], cyclic GMP-AMP synthase (cGAS) [32] to prevent HBV clearance and thereby leading into the progression of HCC. Dead-Box RNA Helicase 3 (DDX3), a host restriction factor reported to suppress HBV via autophagy [33]. Previous studies reported that miR-181a-5p targets DDX3 expression post-transcriptionally [34, 35]. The whole interactome of circSETD2, miR-181a-5p and DDX3 under HBV background in liver is yet to be known. For the first time, this study established that miR-181a-5p was targeted by circSETD2 in HBV in-vitro model to suppress autophagy through DDX3.

Materials and methods

Cell culture

The human liver cancer cell lines HepG2 and Hep3B were acquired from the National Centre for Cell Sciences (NCCS), Pune. HepG2, Hep3B, HepG2.2.15 and HEK293T cells were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM; HiMedia Laboratories, Mumbai, India, Cat No. AL007G) containing 10% fetal bovine serum (Gibco, Thermo Fisher Scientific, MA, USA, Cat No. A5256701) and 1% penicillin-streptomycin (Gibco, Thermo Fisher Scientific, MA, USA, Cat No. 15140122) at 37˚C temperature and 5% CO₂ humidified condition.

Cloning, plasmid isolation and transfection experiments

Polymerase chain reaction (PCR) amplification of the full-length circSETD2 sequence (4644 bp, exon 2–5 of SETD2) from SETD2 mRNA transcript (Accession no. NM_014159) was performed from HEK293T cDNA using specific primers containing restriction sites for PacI in the forward primer and SacII in the reverse primer (Table 1). Amplified PCR products were ligated to pJET1.2 vectors and further sub-cloned into the expression vector pcDNA3.1(+) circRNA mini vector (Plasmid #60648, Addgene). Later, Plasmid DNA was isolated using the Mediprep Plasmid DNA Isolation Kit and the purified plasmid DNA were used for transfection experiments. HepG2.2.15 cells were transfected with circSETD2 plasmid or empty vector pcDNA3.1(+) circRNA mini vector using lipofectamine2000 transfection agent (Thermo Fisher, Carlsbad, CA, USA Cat no. 11668). In addition, miR-181a-5p mimics (Sigma-Aldrich, St Louis, MO, USA, Cat No. HMI0266) or synthetic miR-181a-5p inhibitor (Sigma-Aldrich, St Louis, MO, USA, Cat No. HSTUD0266) or scrambled RNA (NS) (Dharmacon, Horizon Discovery, UK, Cat No. CN-001000-01) were transfected in HepG2.2.15 cells using siPORT™ NeoFX™ transfection reagent (Thermo Fisher, Carlsbad, CA, USA, Cat no. AM4511). After 48 h of transfection, the cells were collected for total RNA or protein isolation. The spent media was collected for analysis of viral load (HBsAg). Similarly, to verify the expression of circSETD2 and miR-181a-5p in HBV-transfected hepatic cell line, pHBV_1.3 plasmid, HBx plasmid or empty vector pcDNA3.1(+) were transfected in HepG2 cells.

Table 1.

Sequences of the primers used for cloning and qPCR studies

Gene	Primer Sequences (5’-3’)
circSETD2 (cloning)	FP: GCGTTAATTAAAGAAGAAGAAAATGAGGCAAAGATTG RP: TATCCGCGGGAAGGAAGGTCTTTGGCAGC
circSETD2	FP: CTTGAGAGCTGCCAAAGACCT RP: TTGGTGCCTTTGGGCAAAAATC
HBx	FP: ACCGACCTTGAGGCCTACTT RP: GCTTGGCAGAGGTGAAAAAG
GAPDH	FP: ACCAGGTGGTCTCCTCTGAC RP: TGCTGTAGCCAAATTCGTTG
miR-181a-5p	FP: AACATTCAACGCTGTCGGTGAGT RP: GCGAGCACAGAATTAATACGAC
5S rRNA	FP: GCCCGATCTCGTCTGATCT RP: AGCCTACAGCACCCGGTATT

RNA isolation, RNase R digestion, cDNA synthesis and quantitative PCR

Total cellular RNA of hepatic cell lines were isolated using TRIzol reagent (Thermo Fisher, Cat. No. 15596026) according to the manufacturer’s protocol. Isolated RNA was quantified by nanodrop method, and 2 µg RNA was digested with 3U/µg RNAse R enzyme (Abcam, UK, Cat No. ab286929) at 37˚C for 30 min to enrich the circRNA population. cDNA for circRNAs were synthesized using PrimeScript 1st strand cDNA synthesis Kit (Takara Bio, Japan, Cat No. 6110 A) using random hexamer with the following conditions: 30˚C for 10 min, 42 ˚C for 60 min and 70˚C for 15 min. For miRNA analysis, total RNA (50 ng) was reverse transcribed using the method as described previously and stored in -20°C until use [36]. Briefly, poly(A) tail was added to the total RNA by poly(A) polymerase enzyme and then reverse transcribed to first strand cDNA using Poly(T) adaptor under following conditions: 37 °C for 30 min, 42 °C for 30 min and 95 °C for 5 min. In addition, to clone circSETD2, first strand cDNA was synthesized from total RNA using PrimeScript 1st strand cDNA synthesis Kit (Takara Bio, Japan, Cat No. 6110 A) using oligo (dT) primers.

Primers for miRNA-181a-5p, 5S rRNA, circSETD2, HBx, and GAPDH (Integrated DNA Technologies, Skokie, IL, USA) (Table 1) were used for quantitative PCR (qPCR) using iTaq Universal SYBR Green Supermix (Bio-Rad, Cat No. 1725120) in AriaMx Real-Time PCR system (Agilent Technologies, CA, USA). GAPDH and 5S rRNA were used to normalise circSETD2 and miR-181a-5p expression, respectively. The expression was analyzed using the 2^−ΔΔCt method as described previously [37].

Enzyme-linked immunosorbent assay

Enzyme-linked immunosorbent assays (ELISA) were performed to analyze hepatitis B virus surface antigen (HBsAg) in cell culture spent media. HBsAg detection was carried out using Monolisa^™ HBsAg ULTRA (Bio-Rad) kit as per the manufacturer’s protocol with slight modification as described previously [33].

Western blot analysis

The total protein from cell lines was extracted using mammalian protein extraction buffer (MPER) (Thermo Fisher, Cat No. 78501) containing 1% protease inhibitor cocktail (Thermo Fisher, Cat No. 78430). Extracted protein was quantified by using a bicinchoninic acid (BCA) protein assay kit (Pierce, Cat No. 23227) according to the manufacturer’s instructions. Equal amount of protein (40 ug/well or 60 ug/well) was loaded and separated using 10% or 15% SDS-PAGE, followed by transfer onto polyvinylidene fluoride (PVDF) membranes. Membranes were blocked with 5% non-fat dry milk followed by incubation with different primary antibodies such as DDX3X  (1:1000, CST#8192), ATG5 (1:1000, CST#12994), Beclin1 (1:1000, CST#3495), LC-3 (1:1000, CST#2775) and β-actin (1:5000, Sigma#A5441). After incubation with primary antibodies at 4˚C overnight, membranes were washed with tris-buffered saline containing 0.1% Tween 20 (TBST) and hybridized with respective HRP-conjugated secondary antibodies at room temperature for 1 h. The membranes were washed and developed using Clarity™ western ECL Substrate (Bio-Rad, Cat. No. 170–5061) either in X-ray film or Chemidoc (Bio-Rad). The band’s intensities were quantified using ImageJ software (NIH, USA).

Autophagy flux assay

For circSETD2 transfection experiments, the coverslips (12 mm diameter) were placed on 24-well culture plates, and the HepG2.2.15 cells were grown on the surface for 12 h in a CO₂ incubator. Then, the cells were transfected with circSETD2 or empty vector using lipofectamine2000 reagent and incubated further for 36 h. For miRNA transfection experiments, the reverse transfection was performed by adding miR-181a-5p mimics or synthetic miR-181a-5p inhibitor or scrambled RNA (NS) on the coverslips placed on 24-well culture plates and then the cells were seeded along with the culture medium and incubated for 36 h. After 36 h of transfection, cells were incubated with DAPRed (Dojindo Laboratories, Japan, Cat No. A562 and Hoechst33342 dye (Thermo Fisher, Cat No. H1399) according to the manufacturer’s protocol. Briefly, the cells were washed twice with media and incubated with DAPRed (0.2 µmol/L) and Hoechst33342 dye (5 µg/ml) for 30 min at 37˚C in a CO₂ incubator. Then, the cells were washed with Dulbecco’s phosphate-buffered saline (DPBS) and fixed with 4% paraformaldehyde for 10 min, followed by washing twice with DPBS. The fixed slides were observed under confocal microscope with 100X amplification. From each experiment, the red puncta of the cells were counted from at least 20 cells from different regions of the slides and the experiments were repeated 3 times. The DAPI filter was used to count the number of cells. The average number of the red puncta from the cells in all the 3 different experiments were plotted as red puncta per cell from individual treatment group.

Statistical analysis

All the experiments were conducted at least three times (n = 3) in duplicates. All the data were shown as Mean ± Standard Error Mean (SEM). The statistical significance (p-value) was analyzed using an unpaired Student’s t-test to analyze differential expression between two groups. For Western blots, one representative blot image is shown from three experiments. All the statistical analyses were two-tailed, and data with p-value ≤ 0.05 were considered statistically significant.

Results

CircSETD2 expression was downregulated by HBV in hepatic cells

The expression levels of circSETD2 were investigated in HBV-stably expressing hepatic cell lines, such as, Hep3B and HepG2.2.15 cells, and a non-HBV cell line, HepG2. HepG2, Hep3B and HepG2.2.15 cells were cultured, and expression of circSETD2 was determined. The qPCR analysis of circSETD2 showed that expression of circSETD2 was significantly lower in HBV-integrated Hep3B and HepG2.2.15 cells compared to non-HBV HepG2 cells (Fig. 1A). Next, HepG2 cells were transfected with pHBV1.3 expression plasmid for 48 h. The expression of HBx was significantly upregulated (Fig. 1B), while the expression of circSETD2 was significantly down-regulated (Fig. 1C) in the HBV-transfected cells. Similarly, transfection of HepG2 cells with the Hepatitis B virus X (HBx) gene showed increased HBx expression (Fig. 1D), while a significant downregulation of circSETD2 was observed in HBx-expressing HepG2 cells (Fig. 1E). These data indicated that circSETD2 was suppressed in HBV-stable expression cells, suggesting that HBx suppressed the expression of circSETD2 expression.

Fig. 1.

Hepatitis B virus (HBV) downregulated intracellular circSETD2 expression: The qPCR analysis of circSETD2 expression in HepG2, Hep3B and HepG2.2.15 cells was determined (A). HepG2 cells were transfected with empty vector (EV) or HBV_1.3 plasmid (B and C) or HBx plasmid (D and E), and after 48 h of incubation, RNA was isolated and the qPCR was performed for the expression of HBx (B and D) and circSETD2 (C and E) (n = 3). *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Overexpression of circSETD2 decreased HBsAg levels

As the circSETD2 was inhibited by HBx protein, it is important to understand the effect of circSETD2 on HBV. Hence, we aimed to clone circSETD2 in a circular RNA expressing vector and overexpress circSETD2 in HBV-stable expression cells. The mature circSETD2 sequence was amplified by PCR using sequence-specific primers. The band size of the PCR product was confirmed by agarose gel electrophoresis (Fig. 2A). Restriction digestion of the plasmid confirmed that circSETD2 sequence was successfully cloned into pJET1.2 vector (Fig. 2B). It was further sub-cloned into expression vector pcDNA3.1(+) circRNA mini vector and clone was confirmed by restriction digestion (Fig. 2C). Sequence of the clone was confirmed by sequencing. After that, HepG2.2.15 cells were transfected by circSETD2 plasmid and qPCR analysis confirmed 5-fold upregulation of circSETD2 upon transfection with the circSETD2 plasmid (Fig. 2D). Further, ELISA analysis of HBsAg in the spent media of circSETD2-transfected cells showed a significant reduction of HBsAg levels (Fig. 2E). These data suggested that overexpression of circSETD2 suppressed the HBV proliferation.

Fig. 2.

Overexpression of circSETD2 downregulated HBsAg levels. A. PCR amplification of mature circSETD2 sequence and visualised by agarose gel. B. Ligation of PCR amplified circSETD2 to pJET1.2 vector was performed. The insert size of 4644 bp was confirmed by restriction digestion. The sequence was confirmed by sequencing and sub-cloned into the expression vector. C. The representative gel picture shows the correct insert size of 4644 bp, after the restriction digestion of the cloned vector. D. HepG2.2.15 cells were transfected with circSETD2 expression vector, and the intracellular expression of circSETD2 was confirmed by qPCR. E. ELISA analysis of HBsAg level in the spent media of circSETD2-transfected HepG2.2.15 cells (n = 3). *p ≤ 0.05, **p ≤ 0.01

CircSETD2 targeted miR-181a-5p in HBV-stable cells

CircRNAs interact with miRNAs, sequester them and limit their downstream gene regulatory function. Previous studies reported that miR-181a-5p was the target of circSETD2 [38]. To validate these findings, circSETD2 was transfected into HepG2.2.15 cells, and miR-181a-5p expression analysis was performed by qPCR. The results showed a significant downregulation of miR-181a-5p in circSETD2-transfected cells (Fig. 3A). Further, intracellular miR-181a-5p expression was found to be higher in Hep3B and HepG2.2.15 cells compared to HepG2 cells, as determined by qPCR (Fig. 3B). Similarly, when HepG2 cells were transfected with pHBV_1.3 or HBx plasmids, the qPCR results of miR-181a-5p expression showed that there was an increased expression of miR-181a-5p in HBV-transfected (Fig. 3C) and HBx-transfected cells (Fig. 3D). These results demonstrated that circSETD2 interacts with miR-181a-5p to modulate downstream signalling pathway.

Fig. 3.

circSETD2 downregulated the expression of miR-181a-5p in HepG2.2.15 cells. A. HepG2.2.15 cells were transfected with the circSETD2 plasmid, and after 48 h, RNA was isolated and qPCR was performed. B. qPCR analysis of intracellular expression of miR-181a-5p in HepG2, Hep3B and HepG2.2.15 cells. HepG2 cells were transfected with empty vector, HBV (C) or HBx (D) and the expression of miR-181a-5p was determined (n = 3). *p ≤ 0.05, **p ≤ 0.01

miR-181a-5p upregulated autophagy by targeting DDX3 in HepG2.2.15 cells

miRNAs exerts their function by regulating gene expression post-transcriptionally. Therefore, the downstream targets of miR-181a-5p were analysed by targetscan (https://www.targetscan.org/). Out of multiple targets, DDX3 was identified as one of the direct target which is interacting at four different sites of DDX3 transcript (Fig. 4A). To confirm the modulation of DDX3 expression, miR-181a-5p mimics were transfected into HepG2.2.15 cells, and transfection efficiency was determined by qPCR. Next, Western blot analysis was performed for DDX3 and autophagy markers, such as ATG5, Beclin1, and LC-3 II. The data showed that there was a decreased expression of DDX3, whereas the expression of autophagy markers were enhanced in the miR-181a-5p mimic-transfected cells (Fig. 4B). The densitometric quantification showed that there was a 42% decrease in DDX3 expression (Fig. 4C), while there was at least two-fold increase in the expression of ATG5, Beclin1 and LC-3 II proteins (Fig. 4D) in miR-181a-5p mimics transfected HepG2.2.15 cells. Further, autophagy flux assay confirmed that transfection of miR-181a-5p mimics induced autophagic vacuole formation, and observed as red puncta (Fig. 4E). Consequently, puncta quantification showed that the number of red puncta was significantly upregulated in miR-181a-5p-overexpressing cells (Fig. 4F).

Fig. 4.

miR-181a-5p decreased DDX3 expression and enhanced autophagy. A. The interaction of miR-181a-5p with 3’-UTR of DDX3. HepG2.2.15 cells were transfected with miR-181a-5p mimics or scrambled RNA (NS) for 48 h. Western blots were performed from the total cellular protein for DDX3, ATG5, Beclin1, LC-3 II and β-actin. B. The representative picture shows the Western blot results of DDX3 and autophagy marker proteins. C and D show the quantitative analysis of the Western blot results of DDX3 and autophagy marker proteins. E. The representative picture shows the autophagy flux image analysis of miR-181a-5p-transfected HepG2.2.15 cells. F. Quantification of autophagic puncta per cell (n = 3). *p ≤ 0.05, **p ≤ 0.01

To further validate these findings, intracellular expression of miR-181a-5p was inhibited using miR-181-5p inhibitor and the expression of DDX3 and autophagy marker proteins were determined. The results showed that there was an increased expression of DDX3, while there was decreased expression of ATG5, Beclin1 and LC-3 II proteins (Fig. 5A). Quantification of band intensities showed that there was a 1.8-fold increase in DDX3 expression (Fig. 5B). The expression of autophagy marker proteins were significantly downregulated in miR-181a-5p inhibitor-transfected HepG2.2.15 cells (Fig. 5C). The autophagy flux assays showed that there was a decreased number of red puncta in miR-181a-5p inhibitor-transfected cells (Fig. 5D). The red puncta were quantified using ImageJ software and plotted. The results revealed that the red puncta were significantly downregulated in the cells transfected with miR-181a-5p inhibitor (Fig. 5E). Altogether these findings suggested that miR-181a-5p targets DDX3 to induce autophagy pathway.

Fig. 5.

Inhibition of miR-181a-5p enhanced DDX3 expression and reduced the expression of autophagy proteins. HepG2.2.15 cells were transfected with miR-181a-5p inhibitor or scrambled RNA (NS) and Western blots were performed after 48 h of incubation. A. Representative picture shows the expression of DDX3, ATG5, Beclin1, LC-3 II and β-actin expression. The band intensities of the expression of DDX3 (B) and autophagy proteins expression (C) were quantified. D. The representative pictures show the red puncta of autophagy flux assay. HepG2.2.15 cells were transfected with miR-181a-5p inhibitor or scrambled RNA (NS). After 36 h, the autophagy puncta was observed under confocal microscopy. E. The red puncta per cell were counted and plotted in bar graphs (n = 3). *p ≤ 0.05, **p ≤ 0.01

CircSETD2 inhibited autophagy by modulating miR-181a-5p / DDX3 axis

Our data confirmed that miR-181a-5p was inhibited by circSETD2. On the other hand, miR-181a-5p directly targeted DDX3 expression, which led to the upregulation of autophagy. Hence, further experiments to determine the effect of circSETD2 on DDX3 expression and autophagy were conducted. CircSETD2 plasmid was transfected in HepG2.2.15 cells, and the expression of DDX3 and autophagy proteins were determined by Western blots. The data showed an increased expression of DDX3 and decreased expression of autophagy proteins, ATG5, Beclin1 and LC-3 II (Fig. 6A). The band intensity quantification showed that DDX3 expression was significantly upregulated (Fig. 6B) and the expression of ATG5, Beclin1 and LC-3 II were significantly downregulated (Fig. 6C) in circSETD2-transfected HepG2.2.15 cells. The confocal images showed a decreased amount of red puncta (Fig. 6D), and the quantified data showed a 3.5-fold decreased of autophagic puncta (Fig. 6E) in circSETD2-transfected cells. Our data clearly demonstrated that circSETD2 upregulates DDX3 expression to suppress autophagy pathways in hepatic cells.

Fig. 6.

Overexpression of circSETD2 enhanced DDX3 expression and reduced the expression of autophagy proteins. HepG2.2.15 cells were transfected with the circSETD2 plasmid or empty vector (EV). A. The representative picture shows the Western blot analysis for DDX3 and autophagy proteins ATG5, Beclin1 and LC-3 II. The band intensities were quantitated and the expression of DDX3 (B) and the autophagy marker proteins (C) were plotted. D. Representative confocal microscope image shows the circSETD2-transfected cells. E. Number of autophagy puncta per cells are represented as bar graph (n = 3). *p ≤ 0.05, **p ≤ 0.01

Discussion

Hepatitis B virus-induced liver inflammation is one of the major causes for acute and chronic liver disease [39, 40]. The challenges in the diagnosis and source determination of hepatitis [41] remained as major issues worldwide. Previous studies on HBV were shown to modulate circRNAs [42–44]. Differential expression of circRNAs was reported in chronic Hepatitis B virus infection [45]. Although previous studies reported that circSETD2 was downregulated in various cancers [24] including HCC [23], its potential role in HBV infection is not yet identified. The lower expression of circSETD2 in the HepG2.2.15 and Hep3B cells compared with HepG2 cells suggested that it could play a potential role in HBV survival. Overexpression of HBV or HBx in HepG2 cells resulted in decreased circSETD2 expression, indicating that HBV suppressed the expression of circSETD2. This is in agreement with the earlier report that its over-expression of circSETD2 resulted in a decreased proliferation and invasion of hepatic cells [23]. Previously we had shown that HBx induced proliferation via miR-21 in hepatic cells [46]. In agreement with the earlier studies, the data showed that HBx inhibited the expression of circSETD2 in hepatic cells. Recent studies reported that HBV protein X (HBx) modulated the expression of several circRNAs [47, 48]. These data suggested that further validation in the serum samples of HBV patients could lead circSETD2 being a potential biomarker of HBV infection. Serum HBsAg, HBeAg and HBV DNA levels are used as gold standard for the detection of HBV infection and to measure the HBV-associated risk of advanced liver diseases [49].

To explore the effect of circSETD2 in HBV infection, it was cloned into pcDNA3.1(+) circRNA mini vector, which contains ZKSCAN1 intron sequence to promote circularization and the clone was confirmed by restriction digestion and sequencing. For the first time, this study showed that circSETD2 overexpression inhibited HBV viral load, indicating the inhibitory effect of circSETD2 on HBV. Previous studies have shown that circRNAs exert their functions by interacting with miRNAs [50]. Recently, Wang et al., showed that miR-181a-5p was a potential target of circSETD2 [38]. In this study, overexpression of circSETD2 resulted in a decreased intracellular expression of miR-181a-5p [51]. Intracellular expression of miR-181a-5p was significantly upregulated in HBV or HBx-transfected HepG2 cells, confirming circSETD2-miR-181a-5p interaction in HBV infection. These findings correlated well with an earlier study, where the authors showed that HBx bound to the promoter region of miR-181a-5p [29]. Previous studies had shown that miR-181a-5p assisted HBV in evading innate immunity [32] and promoted HBV-mediated HCC progression [30, 52]. Previously, we had reported that miR-181a-5p enhanced HBV proliferation by promoting autophagosome formation [31]. It is well established that HBV utilises early autophagy for its replication, survival and proliferation [9]. Autophagy-dependent ferroptosis is most evident in HCC caused by HBV [53, 54]. Previously, we have shown that HBV enhanced its survival and autophagy by regulating the host restriction factors, such as DDX3 and BANF1 [33, 55].

Next, it was found that DDX3 was one of the key targets of miR-181a-5p. Overexpression of miR-181a-5p resulted in the downregulation of DDX3 in the hepatic cells, suggesting that it directly targeted DDX3 to induce HBV replication and autophagy [33]. To confirm these findings, a synthetic miR-181a-5p inhibitor was used to inhibit the intracellular expression of miR-181a-5p. Inhibition of miR-181a-5p resulted in an increased DDX3 expression and decreased autophagy. Additional experiments showed that there was an increased expression of DDX3 and decreased autophagy when circSETD2 was overexpressed. However, further studies need to be conducted in human samples to confirm these findings. In summary, for the first time we showed that HBx suppressed circSETD2 to induce miR-181a-5p expression levels, which in turn resulted in the decreased DDX3, leading to the increased autophagy in HBV-infected hepatic cells. Additionally, circSETD2 reduced viral load, suggesting that circSETD2/miR-181a-5p/DDX3 could be a promising target to treat HBV infection.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

HBV

Hepatitis B virus

circRNA

Circular RNA

circSETD2

Circular RNA SET domain containing 2

DDX3

Dead Box RNA Helicase 3

HCC

Hepatocellular Carcinoma

cccDNA

Covalently closed circular DNA

SHB

Small surface protein

ceRNA

Competing endogenous RNA

HBx

HBV protein X

ATG5

Autophagy-related protein 5

LC-3

Microtubule-associated protein 1 light chain 3

HBsAg

Hepatitis B Surface Antigen

Author contributions

Md Ismail standardized research methods, performed most experiments, and wrote the original manuscript. Teja Naveen Sata performed some Western blots and helped in cell culture and transfection experiments. Md Fahim Khalid performed the cloning experiments. Runa Nasrin Sheuly and Srijita Paul performed some of the Western blots and helped in RNA isolation, cDNA synthesis. Tannavi Sharma, Amrendra Kumar Sah and Gopal Sharma were involved in data analysis and did a few experiments. Dr. Senthil K. Venugopal (correspondence) conceptualized, designed the study, reviewed, and edited the manuscript. All authors reviewed and approved the final manuscript.

Funding

We thank the Department of Biotechnology, Ministry of Science and Technology, Government of India and South Asian University, New Delhi, India, for providing the research funds and fellowships to authors.

Data availability

The data generated during the current study are available with the manuscript.

Declarations

Ethics approval and consent to participate

No human samples were used in this study.

Consent for publication

All authors have given consent for publication.

Competing interests

The authors declare no competing interests.