HepG2.2.15 cells (iCell-h093, Shanghai ICell Bioscience Inc., Ltd., Shanghai, China) stably replicating HBV and their parental HepG2 cells (iCell-h092, Shanghai ICell Bioscience Inc., Ltd., Shanghai, China) were cultured in a medium consisting of 8% exosome-free fetal bovine serum (FBS) (Cat #EXO-FBS-50A-1, SBI, Palo Alto, CA, USA) supplemented with 1% penicillin streptomycin (Cat. NO. 15140122, Lot. 238344, Grand Island, NY, USA). To ensure stable HBV replication, HepG2.2.15 medium was supplemented with G418 solution at a concentration of 400 µg/mL (R20001, Lot.NO71R19494A, Shanghai Yuanye Bio-Technology Co., Ltd., Shanghai, China). All cell lines were validated via short tandem repeat (STR) profiling, were tested, and were found to be negative for mycoplasma (Supplementary materials).

After culturing HepG2 and HepG2.2.15 cells for 48–72 h, to remove impurities such as cells, the cell supernatants were collected and centrifuged at 3000 $\times$g for 30 min. Next, the small molecules were separated by centrifugation at 12,000 $\times$g for 45 min; subsequently, the supernatant was filtered through a 0.22 µm filter (REF.SLGPR33RE, Lot.R1DB07580, Merck Millipore, Billerica, MA, USA). Following this, sEVs were isolated via ultracentrifugation (MFG.NO. O11284, Hitachi himac-CP80NX, Tokyo, Japan) at 100,000 $\times$g for 120 min. The precipitated sEVs were resuspended in 1 mL of precooled 1$\times$PBS (Cat. NO. P1022, Solarbio Science & Technology Co., Ltd., Beijing, China).

Initially, 50 µL of HepBsAg (S38: sc-52412, Santa Cruz Biotechnology, Santa Cruz, CA, USA) was added to the sEVs following ultracentrifugation. Then, they were transferred to a shaker and placed at 4 °C for 12 h to facilitate thorough rotation and mixing. Next, 20 µL of Protein A/G PLUS-Agarose (sc-22003, Santa Cruz Biotechnology, Santa Cruz, CA, USA) was added, and the sample was incubated for 8 h at 4 °C with continuous mixing. After incubation, the mixture was centrifuged at 3000 $\times$g for 5 min to collect the supernatant, followed by ultracentrifugation at 100,000 $\times$g for 120 min. Finally, 100 µL of precooled 1$\times$ PBS was added to dissolve the pellet, resulting in the purification of sEVs with the removal of HBV virus particles.

Purified sEVs from HepG2.2.15 cell culture supernatant (HepG2.2.15-sEVs) and HepG2 cell culture supernatant (HepG2-sEVs) were deposited onto a carbon-coated copper mesh, with 10 µL of each sample added to the grid. After the excess liquid was aspirated from the edges, the samples were dried for 5 min. Next, 10 µL of 2% phosphotungstic acid (Cat# G1870, Lot No. 2400001, Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) was added, and the mixture was stained for 1 min. Excess staining solution was removed, and the samples were dried overnight at room temperature. The morphology of the sEVs was observed and photographed using a transmission electron microscope (Hitachi HT7700, Tokyo, Japan).

A 10 µL sample of ultraisolated purified sEVs was added to 1 mL of 1$\times$ PBS, which was filtered through a 0.22 µm filter, and the contents were thoroughly mixed. A zeta particle size analyzer (Malvern Zetasizer ZS90, Malvern, Worcestershire, UK) was used to measure the concentration and size distribution of the sEVs. Next, the number of particles and their diameters were plotted using Origin (Version Pro2021, 64-bit; Origin Lab Inc., Northampton, MA, USA).

A 150 µL transfection system for sEVs was prepared using the Exo-Fect™ Exosome Transfection Reagent Kit (Cat#EXFT20A-1, Lot.230616-001, SBI, Palo Alto, CA, USA) as follows: 10 µL of Exo-Fect, 20 µL of nucleic acids (20 pmol siRNA or 5 µg plasmid DNA), 50 µL of purified HepG2.2.15-sEVs, and 70 µL of sterile 1$\times$ PBS. The components were mixed by gently inverting the EP tubes six times without vortexing. The prepared mixture was placed in a shaker at 37 °C for 10 min and then transferred to ice. Next, 30 µL of Exo Quick-TC was added to the mixture, and the contents were mixed by inverting the EP tube six times without vortexing to halt the transfection reaction. The transfected sEVs were maintained on ice (or 4 °C) for 30–40 min.

The sample suspension was centrifuged at 14,000 rpm for 3 min, and the precipitates were retained. At least 300 µL of sterile 1$\times$ PBS was added to the precipitate, which was subsequently resuspended. The sequence of small interfering METTL3 (si-METTL3) was 5′-CCUGCAAGUAUGUUCACUATT-3′, and that of small interfering IGF2BP2 (si-IGF2BP2) was 5′-GCATATACAACCCGGAAAGAA-3′ (Cat.NO. M01, Guangzhou RiboBio Co., Ltd., Guangzhou, China).

Stably grown HepG2 cells were selected and seeded in six-well plates at a density of 2 $\times$ 10⁵ cells/mL. Transfection was performed when the cells reached 60–80% confluence, about 12 h after seeding. Before transfection, the serum-free medium was removed. The mixture was subsequently incubated for 1 h with serum-free DMEM. Next, at least 150 µL of transfected HepG2.2.15-sEVs (containing siRNA or plasmid as described above) was added to the cells. After 4–6 h, the mixed medium was changed to nonresistant DMEM containing 8% exosome-free FBS. The cells were allowed to grow for 48 h before being collected for subsequent experiments.

Initially, RNA was extracted using the TRIzol (REF.15596026, Lot.257408, Thermo Fisher Scientific, Waltham, MA, USA) method. After determining its concentration, the RNA was reverse-transcribed to template cDNA in two steps using an RNA reverse transcription kit (REF.1622, Lot.3059833, Thermo Fisher Scientific, Waltham, MA, USA). In step 1, 1 µg of RNA and 1 µL of random primer were added, followed by the addition of ddH₂O to obtain a mixture of 12 µL. The prepared system was reacted at 65 °C for 5 min. In step 2, 4 µL of 5$\times$ buffer, 2 µL of 10 mM dNTPs, 1 µL of RNase inhibitor, and 1 µL of reverse transcriptase were added sequentially. The reaction proceeded at 25 °C for 5 min, 42 °C for 60 min, and 70 °C for 5 min. The resulting cDNA served as a template for quantitative-PCR (qPCR) amplification using a fluorescent quantitative PCR detection kit (Cat.NO. CW0957, Beijing ComWin Biotech Co., Ltd., Beijing, China). Target gene expression was calculated using the 2^{-Δ⁢Δ⁢CT} method. All mRNA sequences of primers used were as follows [30, 31, 32] (Sangon Biotech Shanghai Co., Ltd., Shanghai, China): $\beta$-ACTIN F: 5^′-CTCCATCCTGGCCTCGCTGT-3^′, R: 5^′-GCTGTCACCTTCACCGTTCC-3^′; METTL3 F: 5^′-TTGTCTCCAACCTTCCGTAGT-3^′, R: 5^′-CCAGATCAGAGAGGTGGTGTAG-3^′; IGF2BP2 F: 5^′-AGCTAAGCGGGCATCAGTTTG-3^′, R: 5^′-CCGCAGCGGGAAATCAATCT-3^′; pgRNA F: 5^′-CTCAATCTCGGGAATCTCAATGT-3^′, R: 5^′-TGGATAAAACCTAGGAGGCATAAT-3^′; hepatitis B virus surface antigen (HBsAg) F: 5^′-TCACAATACCGCAGAGTC-3^′, R: 5^′-ACATCCAGCGATAACCAG-3^′; hepatitis B virus core antigen (HBcAg) F: 5^′-CTGGGTGGGTGTTAATTTGG-3^′, R: 5^′-TAAGCTGGAGGAGTGCGAAT-3^′; hepatitis B virus e antigen (HBeAg) F: 5^′-GATTCGCACTCCTCCAGTCT-3^′, R: 5^′-AGTTCTTCTTCTAGGGGACCTG-3^′.

After extracting RNA using the TRIzol method and determining its concentration, the RNA samples were serially diluted to 200 ng/µL, 100 ng/µL, and 50 ng/µL. The samples were subsequently denatured in a metal bath at 95 °C for 5 min and then cooled immediately on ice. Next, 2 µL of each RNA sample was added dropwise onto a nylon (NC) membrane, which had been cut to the appropriate size. After the RNA was allowed to diffuse naturally and dry, the membrane was cross-linked under UV light for 2 h. Following the cross-linking step, the sealing solution was closed for 1 h.

The NC membrane was incubated at 4 °C for 12–16 h with the primary m6A antibody (dilution 1:1000, Cat. NO. 517924, Lot. NO. KK0512, Chengdu ZEN-BIOSCIENCE Co., Ltd., Chengdu, China). The HRP-labeled goat anti-rabbit IgG secondary antibody (1:5000, Cat.NO. FNSA-0004, Chengdu ZEN-BIOSCIENCE Co., Ltd., Chengdu, China) was added, and the mixture was shaken gently at room temperature for 1 h. The signal was developed using an ECL detection system after washing the film.

An animal tissue/cellular genomic DNA extraction kit was used to extract cellular HBV DNA (Cat# D1700, Lot No. 2311002, Solarbio Science & Technology Co., Ltd., Beijing China), and its concentration was subsequently determined. To digest the HBV DNA, the sample was incubated with T5 nucleic acid exonuclease (D7082S-1, Shanghai Biyuntian Biotechnology Co., Ltd., Shanghai, China) for 30 min at 37 °C. The samples were terminated by adding EDTA and then heat-inactivated at 95 °C for 5 min.

PCR amplification was first performed using primers for HBV DNA, HBV rcDNA, and HBV cccDNA, and then the copy number of the DNA was calculated based on a standard curve. The sequences of primers used were as follows (Sangon Biotech Shanghai Co., Ltd., Shanghai, China): HBV DNA F: 5^′-ACCGACCTTGAGGCATACTT-3^′, R: 5^′-GCCTACAGCCTCCTAGTACA-3^′; HBV rcDNA F: 5^′-TTTCACCTCTGCCTAATCATCTCT-3^′, R: 5^′-CTTTATAAGGGTCGATGCCATGC-3^′; and HBV cccDNA F: 5^′-GCCTATTGATTGATTGGAAAGTATGT-3^′, R: 5^′-AGCTGAGGCGGTATCTA-3^′.

The cells and sEVs were lysed in RIPA buffer containing 1% PMSF (ST507, Shanghai Biyuntian Biotechnology Co., Ltd., Shanghai, China) to extract total protein. The samples were lysed on ice for 30 min and centrifuged at 12,000 $\times$g for 30 min at 4 °C to collect the supernatant. The protein concentration was estimated by the BCA method. (Cat# PC0021, Solarbio Science & Technology Co., Ltd., Beijing, China).

After normalizing the protein concentrations, equal amounts of protein were separated by SDS-PAGE and transferred to 0.45 µm PVDF membranes. Next, 5% skim milk powder was added for 1 h, after which the appropriate primary antibody (METTL3, IGF2BP2 primary antibody dilution of 1:1000; Chengdu ZEN-BIOSCIENCE Co., Ltd., Chengdu, China) was added, and the membrane was incubated at 4 °C for 12–16 h. After washing three times with TBST, the membrane was incubated with an HRP-conjugated secondary antibody (1:5000, Cat.NO. FNSA-0004, Chengdu ZEN-BIOSCIENCE Co. Ltd., Chengdu, China) at room temperature for 1 h. After the film was washed, the protein bands were detected using an ECL detection system. The ImageJ software (Version 1.54v, Java1.8.0, 64-bit; Media Cybernetics, Silver Springs, MD, USA) was used to analyze the Western blotting bands in grayscale, and the results were normalized by dividing the target protein grayscale value by an internal reference grayscale value. The primary antibodies used were as follows: $\beta$-ACTIN (1:10,000, AC004, ABclonal Technology, Wuhan, China), Alix (1:1000, NO.2215, ABclonal Technology, Wuhan, China), CD63 (1:1000, A19023, ABclonal Technology, Wuhan, China), CD9 (1:1000, A19027, ABclonal Technology, Wuhan, China), GAPDH (R380646, 1:5000, Chengdu ZEN-BIOSCIENCE Co., Ltd., Chengdu, China), METTL3 (R508370, 1:1000, Chengdu ZEN-BIOSCIENCE Co., Wuhan, China), IGF2BP2 (R389232, 1:1000, Chengdu ZEN-BIOSCIENCE Co., Ltd., Chengdu, China).

After the cells were transfected with sEVs containing the METTL3 plasmid, they were harvested 48 h post-transfection to prepare protein samples. Following the protocol, 20 µL of protein A+G magnetic beads (Cat. NO. P2179S, Lot NO.091322230131, Shanghai Biyuntian Biotechnology Co., Ltd., Shanghai, China) were mixed with 2 µg of either METTL3 primary antibody or normal rabbit IgG and incubated at room temperature for 4 h. Next, protein samples of equal mass were added, and the mixture was rotated overnight at 4 °C.

After incubation, the beads and supernatant were separated using a magnetic rack for 1 min. Next, 500 µL of a lysis buffer supplemented with protease inhibitors was used to wash the beads three times. After washing, 50 µL of 1$\times$ SDS-PAGE protein loading buffer was added to the beads, and the mixture was heated at 95 °C for 10 min to elute the proteins. The supernatant was then collected after the magnetic beads were separated using a magnetic rack. Finally, the samples were analyzed by SDS-PAGE.

All data were presented as the mean $\pm$ standard deviation (SD) derived from three independent experiments. Data were analyzed and graphs were made using GraphPad Prism (version 10.1.2, GraphPad Software, Inc., CA, USA). An independent samples t-test was conducted for pairwise comparisons, and a one-way analysis of variance (ANOVA) followed by Šídák’s multiple comparison test was conducted for multiple group comparisons. All differences among and between groups were considered to be statistically significant at p 0.05.

Small extracellular vesicles derived from HepG2 cells (HepG2-sEVs) and HepG2.2.15 cells (HepG2.2.15-sEVs) were characterized following purification by ultracentrifugation. DLS analysis revealed that 82.4% of the HepG2-sEV particles were 30–200 nm in size, and the peak particle size and concentration were 122.4 nm and 2.30 $\times$ 10⁶ particles/mL, respectively. In contrast, 98.7% of the HepG2.2.15-sEV particles were also 30–200 nm in size and exhibited peak sizes and concentrations of 78.82 nm and 7.22 $\times$ 10⁶ particles/mL, respectively (Fig. 1A). TEM demonstrated that both types of sEVs had a double-layered membranous structure with an average diameter of about 100 nm (Fig. 1B). Additionally, the WB analysis confirmed that the characteristic sEV proteins, including Alix, CD63, and CD9, were expressed (Fig. 1C).

Fig. 1.

Identification of sEVs following ultracentrifugation purification. (A) DLS analysis was conducted to measure the size of sEVs. (B) TEM imaging was performed to investigate the morphological structure of sEVs. The scale bar = 200 nm. (C) Western blotting analysis was performed to detect marker proteins associated with sEVs; ns p $>$ 0.05. sEVs, small extracellular vesicles; DLS, dynamic light scattering; TEM, transmission electron microscopy.

Compared to HepG2-sEVs, HepG2.2.15-sEVs presented significantly greater levels of m6A methylation (Fig. 2A). The expression levels of METTL3 and IGF2BP2 mRNAs in HepG2.2.15-sEVs were significantly elevated, as determined by RT-qPCR analysis (Fig. 2B; p 0.05). The results of the Western blotting analysis revealed a corresponding increase in protein levels (Fig. 2C,D; p 0.05). These findings suggested that an increase in m6A methylation levels in HepG2.2.15-sEVs is associated with the replication of HBV.

Fig. 2.

Enrichment of m6A methylations in sEVs isolated from the HBV-stably replicating HepG2.2.15 cell line. (A) Dot blotting was performed to detect the m6A modification levels in the two kinds of sEVs. (B) Quantitative reverse transcription-PCR (RT-qPCR) was performed to measure the mRNA expression levels of the methyltransferase METTL3 and the reader protein IGF2BP2 in both types of sEVs. (C,D) Western blotting was conducted to assess the protein expression levels of METTL3 and IGF2BP2; *p 0.05 and **p 0.01. HBV, hepatitis B virus; METTL3, methyltransferase-like 3; IGF2BP2, insulin-like growth factor 2.

METTL3 was overexpressed or knocked down in HepG2.2.15-sEVs, resulting in corresponding alterations in METTL3. In the overexpressing METTL3 (OE-METTL3) group, the METTL3 mRNA and protein levels increased. Conversely, in the SI-METTL3 group, METTL3 mRNA and protein levels were lower than those in the control group (Fig. 3A–C; p 0.01), confirming that METTL3 expression was modulated. Moreover, the levels of m6A modification exhibited a corresponding increase or decrease (Fig. 3D).

Fig. 3.

HepG2.2.15-sEVs enhance HBV replication by upregulating the methyltransferase METTL3. In the control group, HepG2.2.15-sEVs without transfection were introduced into parental HepG2 cells. In the OE-METTL3 group, HepG2.2.15-sEVs overexpressing METTL3 were transfected into HepG2 cells after transfection with the METTL3 plasmid. In the SI-METTL3 group, HepG2.2.15-sEVs with METTL3 knockdown were transfected into HepG2 cells after transfection with METTL3 siRNA. (A) RT-qPCR was performed to detect the mRNA expression level of METTL3. (B,C) Western blotting analyses were performed to assess the METTL3 protein level. (D) Dot blot analyses were performed to evaluate the level of m6A modification. (E–J) qPCR and RT-qPCR analyses were performed to measure the expression rates of pgRNA, HBV DNA, HBV rcDNA, and HBV cccDNA,the rcDNA conversion rate, and the mRNA expression levels of HBsAg, HBcAg, and HBeAg; *p 0.05, **p 0.01, ***p 0.001, and ****p 0.0001. pgRNA, pregenomic RNA; rcDNA, relaxed circular DNA; cccDNA, covalently closed circular DNA; METTL3, methyltransferase-like 3; OE, overexpression; SI, small interfering RNA; HBsAg, hepatitis B virus surface antigen; HBcAg, hepatitis B virus core antigen; HBeAg, hepatitis B virus e antigen.

Compared to that in parental HepG2 control cells treated with untransfected HepG2.2.15-sEVs, the mRNA level of HBV pgRNA, a marker of HBV replication, was significantly greater in the OE-METTL3 group (Fig. 3E; p 0.05). The HBV DNA, HBV rcDNA, and HBV cccDNA copy numbers also increased significantly (Fig. 3F; p 0.05), whereas the HBV rcDNA conversion rate increased from 3.51% to 9.99% (Fig. 3H; p 0.01). Moreover, the HBsAg, HBcAg, and HBeAg mRNA expression levels increased (Fig. 3J; p 0.001).

In contrast, the SI-METTL3 group presented a decrease in viral replication indicators: HBV pgRNA expression decreased (Fig. 3E; p 0.01), and the copy numbers of HBV DNA, HBV rcDNA, and HBV cccDNA decreased (Fig. 3G; p 0.01). The HBV rcDNA conversion rate decreased from 1.45% to 0.38% (Fig. 3I; p 0.001), and the mRNA levels of HBsAg, HBcAg, and HBeAg decreased (Fig. 3J; p 0.01).

These results suggested that HepG2.2.15-sEVs facilitated an increase in m6A methylation through the upregulation of METTL3, which promoted the conversion of rcDNA to cccDNA, thereby regulating the replication of HBV. Inhibition of METTL3 expression decreased HBV DNA levels in the treatment of HBV-associated diseases, and the detection of METTL3 in sEVs may be used to predict HBV-associated diseases and monitor the prognosis of the disease caused by HBV.

The IGF2BP2 mRNA expression levels increased or decreased (Fig. 4A; p 0.05) following the overexpression or knockdown of IGF2BP2 in HepG2.2.15-sEVs, as determined by RT-qPCR analysis. The results of the Western blotting assays demonstrated an increase or decrease in the IGF2BP2 protein expression level (Fig. 4B,C; p 0.05), confirming that IGF2BP2 was successfully manipulated in sEVs. Concurrently, the level of m6A modification was also altered (Fig. 4D).

Fig. 4.

HepG2.2.15-sEVs enhance HBV replication through IGF2BP2. In the control group, untransfected HepG2.2.15-sEVs were introduced into HepG2 cells. In the OE-IGF2BP2 group, HepG2.2.15-sEVs were transfected into HepG2 cells following transfection with an IGF2BP2 plasmid to induce the overexpression of IGF2BP2. In the SI-IGF2BP2 group, HepG2.2.15-sEVs were transfected into HepG2 cells after transfection with IGF2BP2 siRNA to knock down IGF2BP2. (A) The overexpression or knockdown of IGF2BP2 was confirmed by RT-qPCR to measure IGF2BP2 mRNA expression. (B,C) The overexpression or knockdown of IGF2BP2 was confirmed by Western blotting analysis to assess the level of the IGF2BP2 protein. (D) The level of m6A modification resulting from the overexpression or knockdown of IGF2BP2 was evaluated by dot blot analysis. (E–J) The effect of knocking down or overexpressing IGF2BP2 on HBV replication was assessed through RT-qPCR and qPCR; the expression levels of pgRNA, HBV DNA, HBV rcDNA, and HBV cccDNA; the rcDNA conversion rates; and the mRNA levels of HBsAg, HBcAg, and HBeAg were measured; *p 0.05, **p 0.01, ***p 0.001, and ****p 0.0001. IGF2BP2, insulin-like growth factor 2.

Compared to the parental HepG2 control cells treated with untransfected HepG2.2.15-sEVs, the expression level of HBV pgRNA, a marker for detecting HBV replication, increased or decreased (Fig. 4E; p 0.01). The HBsAg, HBcAg, and HBeAg mRNA levels increased or decreased, respectively (Fig. 4J; p 0.01).

The copy numbers of HBV DNA, HBV rcDNA, and HBV cccDNA increased (Fig. 4F; p 0.05), with the HBV rcDNA conversion rate increasing from 6.14% to 31.34% (Fig. 4H; p 0.01). In contrast, these values decreased (Fig. 4G; p 0.01), with the HBV rcDNA conversion rate decreasing from 31.25% to 2.23% (Fig. 4I; p 0.05).

These results suggested that HepG2.2.15-sEVs, via the reader protein IGF2BP2, can increase m6A levels, promote the conversion of rcDNA to cccDNA, and consequently regulate the replication of HBV. A significant reason for the prolonged treatment of HBV disease is the difficulty in eliminating cccDNA. These experimental results suggested that the rcDNA conversion rate may be reduced by inhibiting the expression of IGF2BP2, thus decreasing the generation of cccDNA and halting the progression of the disease.

After the cells were transfected with HepG2.2.15-sEVs containing the METTL3 plasmid, IGF2BP2 was detected via immunoprecipitation (IP) using an anti-METTL3 antibody. The immunoprecipitation results indicated a synergistic interaction between METTL3 and IGF2BP2 (Fig. 5A).

Fig. 5.

METTL3 synergizes with IGF2BP2 in HepG2.2.15-sEVs. The control group consisted of HepG2.2.15-sEVs transfected into HepG2 cells. The OE-METTL3 group included HepG2.2.15-sEVs overexpressing METTL3, which were also transfected into HepG2 cells. In the SI-METTL3 group, HepG2.2.15-sEVs with METTL3 knockdown were transfected into HepG2 cells. The OE-IGF2BP2 group included HepG2.2.15-sEVs overexpressing IGF2BP2, which were transfected into HepG2 cells, whereas the SI-IGF2BP2 group included HepG2.2.15-sEVs with IGF2BP2 knockdown, which were also transfected into HepG2 cells. (A) Co-IP experiments were conducted to confirm the interaction between METTL3 and IGF2BP2. (B) Following the overexpression or knockdown of METTL3, RT-qPCR was performed to assess IGF2BP2 mRNA expression. (C,D) Western blotting analysis was performed to evaluate METTL3 and IGF2BP2 protein levels after the overexpression or knockdown of METTL3. (E) RT-qPCR analysis was performed to measure METTL3 mRNA levels following the overexpression or knockdown of IGF2BP2. (F,G) Western blotting analysis was conducted to detect the METTL3 and IGF2BP2 proteins following the overexpression or knockdown of IGF2BP2; *p 0.05, **p 0.01, ***p 0.001, and ****p 0.0001.

After the cells were transfected with HepG2.2.15-sEVs with either overexpression or knockdown of METTL3, the IGF2BP2 mRNA expression level increased, as determined by RT-qPCR (Fig. 5B; p 0.01). The results of the Western blotting analysis revealed that the METTL3 and IGF2BP2 protein levels increased and decreased, respectively (Fig. 5C,D; p 0.01).

After the cells were transfected with HepG2.2.15-sEVs with either overexpression or knockdown of IGF2BP2, the METTL3 mRNA expression levels increased or decreased, respectively (Fig. 5E; p 0.05). Similarly, the METTL3 and IGF2BP2 protein levels increased or decreased, respectively (Fig. 5F,G; p 0.001). These results confirmed the synergistic relationship between METTL3 and IGF2BP2.