Introduction

Chronic hepatitis B virus (HBV) infection remains a significant global health concern, affecting more than 250 million individuals worldwide and leading to severe liver diseases, such as fibrosis, cirrhosis, and cancer1,2. HBV is characterized by a small 3.2-kb relaxed circular DNA (rcDNA) genome. After infection, HBV rcDNA is converted into a covalently closed circular DNA (cccDNA) in cell nucleus to serve as a template for viral RNA transcription and the production of new virions3. Current therapies involve interferon (IFN) alpha, which induces immune responses against HBV, and nucleoside analogs (NAs), which target viral replication4. However, lifelong therapy is necessary because available therapeutics do not eliminate HBV cccDNA from host cells, which persists in patients and leads to viral rebound upon treatment cessation5. Alternative approaches have been considered to interfere with the infection process following the discovery of the HBV entry receptor, sodium-taurocholate cotransporting polypeptide (NTCP), a transmembrane bile acid transporter, which is encoded by the solute carrier family 10 member 1 (SLC10A1) gene6,7. Among tested NTCP inhibitors, Myrcludex-B has shown promise by preventing viral spreading and amplification of the HBV genome8.

MicroRNAs (miRNAs) are small non-coding RNAs of approximately 22 nucleotides that modulate gene expression at the post-transcriptional level by complementary base pairing with the 3ʹ-untranslated region (3ʹ-UTR) of messenger RNAs9,10. Previous works have extensively described the importance of miRNAs in various biological processes and human pathologies11. In the liver, miRNA deregulation is a hallmark of chronic diseases, including viral infections and cancer12,13. One of the first studies that conducted miRNA profiling in clinical samples revealed the differential expression of miRNAs in the liver of patients affected by viral hepatitis and the ability to discriminate between HBV- and hepatitis C virus-related hepatic tumors14. In addition to their diagnostic value, we and others have demonstrated that miRNAs are attractive mediators for potential RNA therapies in human diseases15,16,17,18.

We hypothesized that NTCP inhibition by miRNA-based therapeutics could represent an effective strategy to hinder cell infection and limit the spread of HBV in the liver. Using an experimental model of primary human hepatocytes (PHHs), we highlighted a set of miRNAs induced by IFN alpha analog treatment. Among these miRNAs, we found that miR-29b-1-5p was able to interact with the 3ʹUTR of NTCP transcript and negatively regulate its expression level. More importantly, HBV genome levels were markedly reduced in infected PHHs that overexpressed miR-29b-1-5p, supporting the therapeutic potential of miRNAs targeting NTCP.

Results

miR-29b-1-5p inhibits RNA levels of the HBV entry receptor NTCP in human hepatocytes

We assumed that specific miRNAs induced by IFN-like compounds could exhibit antiviral activities. CDM-3008, an IFN alpha analog previously characterized by our group19, effectively induced the expression of interferon-stimulated genes (ISGs) in human hepatocytes (Fig. S1). We analyzed miRNA expression profiles in two distinct lots of PHHs using a miRNA microarray. The microarray data revealed consistent modifications in the expression levels of 11 miRNAs after treatment with CDM-3008 in both PHH cultures (Fig. S2). Among these differentially expressed (DE) miRNAs, miR-29b-1-5p, miR-296-5p, miR-4731-3p, and miR-7111-3p were significantly upregulated by CDM-3008 (fold change > 1.5 and p < 0.05, t-test), while miR-221-3p was downregulated (Fig. 1A). Using the miRNA target prediction tool TargetScanHuman (Release 8.0), we investigated whether these DE miRNAs had sequence complementarity with the 3ʹ-UTR of NTCP. Interestingly, NTCP exhibited two probable interaction sites with miR-29b-1-5p (Fig. 1B). The induction of miR-29b-1-5p by CDM-3008 was further validated using miRNA-specific quantitative PCR analysis (Fig. 1C). Concurrently, the expression level of NTCP decreased in response to the treatment, consistent with a possible negative regulatory mechanism mediated by miR-29b-1-5p.

Fig. 1
figure 1

miR-29b-1–5 is induced by the IFN alpha analog CDM-3008 and potentially targets NTCP. (A) Expression profile of the DE miRNAs in response to CDM-3008 treatment. PHHs were exposed to 10 µM CDM-3008 for 8 h, and total RNAs were extracted for miRNA microarray analysis. The graph displays the average fold changes and p-values from two distinct lots of human hepatocytes. miRNAs with significant changes in expression level (fold change ≥ 1.5 and p-value < 0.05, t-test) are shown in black. (B) Venn diagram summarizing NTCP binding prediction analysis. NTCP 3ʹ-UTR-miRNA interaction probability was assessed using the miRNA target prediction tool TargetScanHuman. NTCP was identified as a potential target of miR-29b-1-5p among the 119 predicted miRNAs. (C) Expression levels of miR-29b-1-5p and NTCP after treatment. The relative expression levels were determined by real-time quantitative PCR at the indicated times. The graphs show the mean ± SD from 2 lots of PHHs (n = 2 per condition). Spearman’s rank coefficient calculated with miR-29b-1-5p and NTCP expression values was rho = − 0.785 (p < 0.0001).

Full size image

To assess the impact of miR-29b-1-5p on NTCP, we increased miR-29b-1-5p in PHHs using synthetic miRNAs (Fig. 2A). Our results showed that miR-29b-1-5p overexpression significantly reduced NTCP mRNA levels after 4 days compared to the control conditions (p < 0.01, ANOVA) (Fig. 2B). NTCP 3ʹ-UTR exhibited two potential binding sites for miR-29b-1-5p, one of these sites exhibiting higher complementarity with a sequence of 11 consecutive nucleotides (Fig. S3). To evaluate the ability of miR-29b-1-5p to mediate NTCP mRNA degradation by 3ʹ-UTR interaction, a reporter assay using a vector containing this complementary region downstream of a luciferase coding sequence was conducted. The assay showed that miR-29b-1-5p overexpression decreased luciferase activity by 37.3 ± 6.0% compared to control miRNA (p = 0.0234, t-test) (Fig. 2C). Conversely, no significant inhibition in luciferase activity was observed in cells transfected with a vector containing a mutated sequence of NTCP 3ʹ-UTR. Together, these results supported the existence of a post-transcriptional regulation mechanism of NTCP directly mediated by miR-29b-1-5p.

Fig. 2
figure 2

miR-29b-1-5p targets NTCP transcripts in human hepatocytes. (A) miR-29b-1-5p overexpression. PHHs were transfected using miR-29b-1-5p mimics (miR-29b-1-5p) and miRNA control mimics (miR-Control). Relative expression levels were measured by real-time quantitative PCR after four days. miR-29b-1-5p expression levels were normalized using RNU6B. (B) Effect of miR-29b-1-5p overexpression on NTCP mRNA levels. miR-29b-1-5p expression was measured four days after transfection using synthetic miRNA mimics. NTCP expression levels were normalized using GAPDH. (C) NTCP 3ʹ-UTR dual luciferase reporter assay. Human hepatoma HepG2 cells were co-transfected with miRNA mimics (miR-29b-1-5p versus miR-Control) and luciferase reporter constructs containing the human NTCP 3ʹ-UTR (WT, wild type) or a mutated sequence (Mut, mutated). The luciferase activities were measured 24 h after transfection, and the ratio of hRluc/hluc+ was determined. The depicted sequences indicate sequence complementarity between miR-29b-1-5p and NTCP 3ʹ-UTR. All histograms represent the mean ± SD. Statistical differences were evaluated with a t-test for miR-29b-1-5p expression measurements (n = 6 per condition) and the luciferase activities (n = 3 per condition) and a one-way analysis of variance (ANOVA) for NTCP expression measurements (n = 6 per condition). *p < 0.05, **p < 0.01. NS, not significant.

Full size image

miR-29b-1-5p reduces HBV genome levels in human hepatocytes

Next, we assessed the antiviral effect of miR-29b-1-5p in PHHs. To ensure substantial NTCP inhibition, cells were infected four days after transfection with synthetic miR-29b-1-5p (Fig. 3A). In control conditions (control miRNA mimics), HBV DNA copies dramatically increased over time in PHH cultures (Fig. 3B). Conversely, miR-29b-1-5p overexpression restrained HBV genome augmentation, which was particularly noticeable two weeks after infection (p = 0.0003, t-test). HBV pregenomic RNA (pgRNA) levels were consistently inhibited by miR-29b-1-5p (p = 0.0160, t-test) (Fig. 3C). Importantly, no cytotoxicity was evidenced in miR-29b-1-5p-treated hepatocytes (Fig. S4), ruling out an HBV clearance process via cell death.

Fig. 3
figure 3

HBV genome levels are reduced in hepatocytes overexpressing miR-29b-1-5p. (A) Experimental design of miR-29b-1-5p therapy. Four days after transfection (miR-29b-1-5p versus miR-Control), PHHs were infected with HBV at 100 genome copies (GC) per cell for 24 h. Total RNA and DNA were collected at the indicated times for HBV assessment. (B) Effect of miR-29b-1-5p on HBV DNA intracellular levels. The number of HBV DNA copies was determined by real-time quantitative PCR from 50 ng DNA and normalized to the total amount of genomic DNA collected in each well. (C) Effect of miR-29b-1-5p on HBV pgRNA intracellular levels. The relative levels of HBV pgRNA were assessed by real-time quantitative PCR four days after infection. GAPDH was used for normalization. All data in the figure are mean ± SD (n = 6 per condition). Significant difference in HBV DNA copies and pgRNA levels: *p < 0.05, **p < 0.01, and ***p < 0.001 (t-test).

Full size image

Besides its action on NTCP, miR-29b-1-5p could potentially implicate and promote cell-intrinsic immune response. To address this, we analyzed the inhibitory effect of miR-29b-1-5p in the HepG2.2.15 cell line, which stably expresses and replicates HBV. No significant variation in HBV DNA and pgRNA levels was observed after miR-29b-1-5p treatment (Fig. S5). Given the extremely low expression level of NTCP in HepG2.2.15 cells compared with primary hepatocytes, this strongly supported the idea of an antiviral activity exclusively mediated by NTCP inhibition in miR-29b-1-5p-treated PHHs. To validate this hypothesis, we conducted a rescue experiment by overexpressing NTCP before infection (Fig. 4A and Fig. S6). Our data showed that HBV DNA copies (Fig. 4B) and HBV pgRNA levels remained unchanged after miR-29b-1-5p treatment in cells that overexpressed NTCP (Fig. 4C), demonstrating that NTCP rescue was able to antagonize the antiviral effect of miR-29b-1-5p.

Fig. 4
figure 4

NTCP overexpression suppresses the anti-HBV activity of miR-29b-1-5p. (A) Experimental design of the NTCP rescue. PHHs were infected with HBV at 100 GC/cell four days after transfection with miRNA mimics (miR-29b-1-5p versus miR-Control). The rescue was conducted by co-transfection with a construct that overexpressed human NTCP. Effect of NTCP rescue on (B) HBV DNA and (C) HBV pgRNA levels. The relative levels of HBV DNA and HBV pgRNA were determined by real-time quantitative PCR nine days after infection to evaluate the impact of NTCP restoration in miR-29b-1-5p-treated PHHs. All histograms show the mean ± SD (n = 6 and 4 per condition, for HBV DNA and pgRNA levels, respectively). Statistical differences were evaluated with a t-test: ***p < 0.001. NS, not significant. (D) Proposed regulatory mechanism by which miR-29b-1-5p inhibits the HBV entry receptor NTCP through 3ʹ-UTR targeting and impedes hepatocyte infection.

Full size image

Discussion

Therapeutic miRNAs offer a promising avenue to specifically target transcripts critical for viral infections. In this study, we demonstrated the functional role of miR-29b-1-5p in the control of HBV intracellular levels through post-transcriptional modulation of NTCP in human hepatocytes (Fig. 4D). Previous studies have highlighted the direct susceptibility of HBV to RNA interference mechanisms mediated by cellular miRNAs20,21. For example, Chen and colleagues showed that miR-122 binding to HBV pgRNA suppressed HBV protein expression and viral replication in HepG2.2.15 cells22. On the other hand, miRNAs, such as miR-29b-1-5p, can modulate host genes involved in viral infections. Fujita and colleagues reported that experimental overexpression of miR-6126, a cellular miRNA induced in HBV patients treated with pegylated IFN, decreased NTCP mRNA levels in the HepG2-NTCP cell line23. HepG2 cells stably overexpressing NTCP represent an affordable and well-characterized alternative widely used in laboratories24,25. In our study, we evaluated miR-29b-1-5p anti-HBV effect using a PHH experimental model, considered the gold standard in HBV studies and antiviral therapeutic evaluations. HepG2-NTCP cells were expected to show a limited response to anti-NTCP miRNAs due to the absence of the 3ʹ-UTR in the overexpressed NTCP RNA.

Over the past decade, considerable efforts have been made in the development of RNA-based therapeutics for human diseases, as evidenced by the increasing number of preclinical and clinical studies26,27,28. For example, recent attention has focused on the potential of antiviral miRNAs in severe acute respiratory syndrome coronavirus 2 (SARS‑CoV‑2) infections, which cause coronavirus disease 2019 (COVID-19)29,30. Further work will be required to clarify the role of miR-29b-1-5p in the HBV entry mechanism and to strengthen its translational potential. In particular, future experiments should assess its effect at earlier stages of infection and under conditions that more closely reflect physiological miRNA expression levels by using lower concentrations of miRNA mimics. Another concern with anti-NTCP miRNAs is the possible elevation of serum bile salts due to NTCP impairment. However, studies involving NTCP knockout mice and human polymorphism-related knockdown have not reported severe symptoms31,32. The high stability of cccDNA, which can persist in treated cells as a reservoir for HBV replication, remains a significant challenge for virus eradication. Combining first-line anti-HBV drugs with miRNAs targeting NTCP could enhance curative outcomes. Effective blockage of viral spread to healthy cells through NTCP inhibition, coupled with the gradual loss of cccDNA as hepatocytes are replaced in the infected liver, may lead to virus control and subsequent clearance with prolonged therapy. Last, anti-NTCP miRNA treatment as post-exposure prophylaxis for preventing acute HBV infection could represent another promising strategy to explore.

Material and methods

Hepatocytes and cell line culture

Human hepatocytes isolated from chimeric mice with humanized liver were purchased from PhoenixBio (Higashi-Hiroshima, Japan) and cultured on collagen type I-coated 96-well cell culture plates in hepatocyte clonal growth medium (dHCGM), as previously described33. HepG2.2.15.7 cells, obtained from the National Institute of Infectious Diseases (Tokyo, Japan), were maintained in a Dulbecco’s modified Eagle’s medium (DMEM)/F-12, GlutaMAX mixture (Gibco) supplemented with insulin (5 µg/mL; Wako), penicillin (100 IU/mL; Gibco), streptomycin (100 µg/mL; Gibco), and 10% fetal bovine serum (FBS; Gibco). HepG2 cells, purchased from the American Type Culture Collection, were maintained in DMEM supplemented with 2 mM L-glutamine, penicillin (50 IU/mL), streptomycin (50 µg/mL), and 10% FBS.

HBV production and cell infection

HBV particles were produced from HepG2.2.15.7 cells, which stably express and replicate the virus (genotype D). The culture medium containing viral particles was collected from confluent cells, filtered (0.22 µM), and incubated with a mixture of 23% NaCl and 50% polyethylene glycol (PEG) 8000 for at least 2 h at 4 °C. Subsequently, HBV particles were purified by ultracentrifugation, washed with PBS, resuspended in cell culture medium, and stored at − 80 °C. PHHs were infected with HBV particles (100 genome copies per cell) for 24 h using a mixture of 10 µL 40% PEG 8000 and 90 µL cell culture medium.

Cell transfection and treatment

PHHs were incubated with a transfection mix containing 20 pmol miRNA mimics and 0.45 μL of Lipofectamin RNAiMAX (Invitrogen) in a 200 µL volume of serum- and antibiotic-free Opti-MEM (Invitrogen) for 30 h. For NTCP rescue experiments, cells were incubated with 0.5 µg of the expression vector and 20 pmol miRNA mimics following the same procedure. The human miR-29b-1-5p mimics (ID #MC12434, cat. #4464066) and negative control miRNA mimics (cat. #4464058) were purchased from ThermoFisher Scientific. The NTCP expression vector, containing the NTCP coding sequence cloned into a pEF4/myc-His (A) vector from Invitrogen, was obtained from the National Institute of Infectious Diseases (Tokyo, Japan). CDM-3008 (RO8191; PubChem CID: 2768133) was from SIGMA and dissolved in DMSO as a 1000 µM working solution.

Total RNA and DNA isolation

Total RNA and DNA were purified using the RNAdvance Cell v2 and DNAdvance Genomic DNA Isolation Kit (Beckman Coulter), respectively, following the manufacturer’s instructions on a Biomek NXp Automatic Workstation from Beckman Coulter. Before elution, RNA samples underwent DNase I treatment (5 units, Nippon Gene) at 25 °C for 15 min. For microarray (PHHs) and quantitative PCR analysis from HepG2.2.15.7 and HepG2 cell samples, total RNA was purified using a miRNeasy Mini Kit (Qiagen) according to the manufacturer’s recommendations, followed by DNase I treatment (2 units, 37 °C for 30 min) using a TURBO DNA-free Kit (Invitrogen). Concentrations of DNA and RNA samples were determined using a DeNovix spectrophotometer.

miRNA expression microarray

PHHs from two distinct lots were treated with 10 µM CDM-3008 for 8 h. Cells treated with DMSO were used as controls. Each PHH lot was a mixture of hepatocytes collected from at least three donors. Total RNA was extracted and treated with DNase I as described previously. The integrity of the samples was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies). Subsequently, RNA samples were labeled and hybridized with a SurePrint G3 Human_miRNA_version 21.0_Microarray (8 × 60 K) using the miRNA Complete Labeling Reagent and Hyb Kit (Agilent Technologies). Agilent platform was Agilent-070156. Microarray slides were scanned with an Agilent Technologies G2505C Microarray Scanner. Raw data were processed using Agilent Feature Extraction software with default parameters to calculate the intensities of the measured spots. A fold change of > 1.5 in signal intensity was applied to identify significant differences in miRNA expression levels after CDM-3008 treatment.

miRNA, RNA, and HBV DNA real-time quantitative PCR analysis

For miRNA assessment, 100 ng of total RNA was first reverse transcribed using the TaqMan MicroRNA Reverse Transcription Kit from Applied Biosystems, following the manufacturer’s recommendations. Next, miR-29b-1-5p expression levels were determined using the TaqMan MicroRNA Assay (ID #002165, cat. #4427975) from ThermoFisher Scientific and the TaqMan Universal PCR Master Mix (Applied Biosystems). The endogenous levels of RNU6B were used for normalization (ID #001093, cat. #4427975). The PCR conditions were 50 °C for 2 min and 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. Gene expression and HBV pgRNA levels were evaluated by synthesizing cDNAs from 0.5 µg of purified RNA using the PrimeScript RT Master Mix (Takara) and performing PCR using the TB Green Premix Ex Taq II (Takara). Thermal cycling conditions comprised an initial denaturation at 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 30 s. Glyceraldehyde 3-phosphatase dehydrogenase (GAPDH) was used to normalize the cDNA levels. HBV DNA levels were determined using the Probe qPCR Mix (Takara) from 50 ng of total DNA. PCR conditions comprised 95 °C for 1 min, followed by 50 cycles of 95 °C for 10 s and 60 °C for 30 s. The sequences of the human primers, HBV DNA primers, and probe used for amplification are shown in Table S1. Real-time quantitative PCR analyses were performed using a CFX96 Real-Time PCR detection system from Bio-Rad.

NTCP 3ʹ-UTR assay

The sequence corresponding to miR-29b-1-5p complementary region of NTCP 3’ UTR (SLC10A1 gene, binding site #1) was cloned downstream of a synthetic Renilla luciferase gene (hRluc) into the psiCHECK-2 vector from Promega. Synthetic firefly luciferase gene (hluc+) driven by an HSV-TK promoter was used for normalization. A reporter vector with a mutated sequence of miR-29b-1-5p binding site was used as a negative control. HepG2 cells were used for the assay. Simultaneous cell transfection with 3ʹ-UTR constructs (3 µg) and miRNA mimics (100 nM) was performed in 35-mm-diameter dishes using 5 µL of TransFectin lipid reagent (Bio-Rad) in a 1.2 mL total volume of serum- and antibiotic-free Opti-MEM (Invitrogen). Cells were collected 24 h post-transfection, and protein was extracted using the M-PER mammalian protein extraction reagent (ThermoFisher Scientific). Renilla and firefly luciferase activities were measured using the Dual-Glo Luciferase Assay System (Promega), as recommended by the manufacturer, with an EnSight Multimode Plate Reader (PerkinElmer). Dual luciferase activity was expressed as the ratio of Renilla versus firefly luminescence (hRluc/hluc+).

Cell viability assay

PHH viability was determined using the Cell Proliferation Kit II from Roche, according to the manufacturer’s instructions (XTT assay). Absorbances at 492 nm and 690 nm were measured 9 days post-infection (corresponding to 13 days after cell transfection with miRNA mimics) using an EnSight Multimode Plate Reader (PerkinElmer). Cell viability was expressed as A492–A690.

Statistical analysis

The experimental data are presented as the mean ± SD and are representative of at least three independent experiments, except for the miRNA microarray analysis in PHHs, which was based on two treated samples and two control samples. After evaluating data distribution, a Student’s t-test or a one-way analysis of variance (ANOVA) was performed to assess statistical significance. The equality of variances was tested using an F-test, and a correction was performed in the case of unequal variance (Welch t-test).