Hepatitis B Virus Increases SphK1–S1P Synthesis by Promoting the Availability of the Transcription Factor USF1

Hepatitis B virus (HBV) is the most common chronic viral infection globally, affecting ∼360 million people, with around one million deaths annually due to end-stage liver disease or hepatocellular carcinoma (HCC) (1). Current antiviral therapies, mainly consisting of IFN and nucleotide analogs, rarely achieve a functional cure for chronic hepatitis B (CHB), defined by seronegative conversion of HBsAg (2). Consequently, increasing research focuses on monitoring and timely intervention to delay HBV-induced liver injury.

The hepatitis B virion is an enveloped nucleocapsid with an outer lipoprotein envelope and an inner nucleocapsid core, which selectively enters hepatocytes by binding to the sodium taurocholate cotransporting polypeptide (NTCP) receptor on the cell surface (3). The HBV genome, a 3.2-kb partially double-stranded circular DNA, contains four overlapping open reading frames (ORFs): 1) ORF-P encoding HBV polymerase (P); 2) pre-S1/pre-S2/S ORF encoding large/middle/small hepatitis B surface Ag (HBsAg); 3) pre-C/C–ORF coding for hepatitis B e Ag (HBeAg) and core protein (HBc); and 4) X-ORF coding for the HBx protein. Each of these proteins plays a distinct role in the HBV life cycle and disease etiology. Immune-mediated host–virus interactions significantly influence the outcome of HBV infection (4, 5). Thus, considerable research aims to alleviate immune injury and hepatocellular carcinogenesis in CHB by targeting the essential immune checkpoints.

Sphingosine-1-phosphate (S1P) is a bioactive lipid metabolite that stimulates DNA synthesis and regulates cell survival, migration, and adhesion by binding to 5 G protein–coupled receptors (S1PR1 to S1PR5) (6, 7). In addition, S1P–S1PR signaling regulates B and T lymphocytes trafficking from lymph nodes to tissues through the circulatory system (8). Sphingosine kinase (SphK) is the key rate‐limiting enzyme in S1P synthesis, existing as two isoforms: SphK1 and SphK2. In mammals, both SphK1 and SphK2 catalyze the conversion of sphingosine (Sph) to S1P, but SphK2 preferentially synthesizes dihydro-S1P (dhS1P) from dihydrosphingosine (dhSph) (9). SphK1 is predominantly found in the cytoplasm, whereas SphK2 is located in the nucleus and in the inner membrane of the Golgi, endoplasmic reticulum, and mitochondria (10). Recent reports have shown that SphK1 is involved in various liver injury processes, including hepatocyte inflammatory necrosis, liver fibrosis, hepatocarcinogenesis, and acute liver failure (11–13). Consequently, many drugs targeting SphK–S1P–S1PR signaling have been developed for various immune-mediated diseases. For instance, a specific SphK1 inhibitor has been shown to ameliorate inflammation in acute liver failure (14), and fingolimod (FTY720), an oral drug targeting downstream S1P–S1PR signaling, has been approved by the Food and Drug Administration for the treatment of multiple sclerosis (15).

SphK1 has been reported to be abnormally expressed in HBV-related liver diseases such as HCC and liver fibrosis (16, 17), but its direct association with HBV has not been confirmed in humans. The current study provides new insights into the interactions between HBV infection and the SphK1–S1P axis. Using human samples, in vivo experiments, and in vitro experiments, we validate the correlation between S1P levels and HBV infection and preliminarily explore the underlying mechanisms. Our findings support the notion that HBV can promote SphK1 expression to release more S1P. Beyond serving as a potential indictor of altered metabolism of HBV-infected hepatocytes, SphK1–S1P–S1PR signal might be a therapeutic target for delaying HBV-related liver injury.

In this trial, 20 patients were diagnosed with hepatitis B according to the American Association for the Study of Liver Diseases 2018 hepatitis B guidance (18), including 11 CHB, 5 hepatitis B–related HCC, 4 and HCC uncertainly caused by hepatitis B. They were treated with nucleos(t)ide analogs (NAs) (entecavir and tenofovir) for more than 3 months according to the doctor’s advice, in the Fourth Affiliated Hospital of Soochow University, China, from 2021 to 2023 (15 male and 5 females; age range, 25–73 y; mean age, 40.72 ± 13.09 y). They were mainly used to analyze the changes in serum viral markers and S1P before and after treatment for the same individual. Pathological sections of liver tissues from 19 randomly selected HCC patients (14 who were HBV positive over 10 y and 5 who were HBV negative) were collected to detect HBV Ag and SphK1 by immunohistochemistry and immunofluorescence. This research was conducted by the ethical standards of the Declaration of Helsinki and was approved by the Institutional Review Board of the Fourth Affiliated Hospital of Soochow University. Informed consent was waived due to its retrospective nature, no additional sample collection, and no interference with the treatment and prognosis of patients.

C57BL/6 mice were purchased from Shanghai Slac Animal Inc. and housed in a specific-pathogen-free rodent barrier facility. All animal experiments were approved by the Institutional Animal Care and Use Committee of Soochow University. The HBV replication-competent plasmid containing 1.3 copies of the HBV genome, pUC19-HBV1.3 (pHBV1.3), HBx-depleted plasmid pHBV-ΔHBx, and plasmids encoding Myc-tagged HBs, HBx, and HBc in a backbone of pcDNA3.1 were kindly provided by Prof. Min Li (Institute of Biology and Medical Sciences, Soochow University) (19–21). USF1 cDNA was amplified from HepG2 cells by standard PCR and cloned in pEGFP-N1or pcDNA3.1 vector plasmid. The NCBI dataset was used to generate SphK1 luciferase reporter genes and cloned in pGL3 plasmid.

The human hepatoma cell line (HepG2, Huh7, HepG2-NTCP, HepG2.2.15) and human embryonic kidney cell line 293T (HEK293T) were cultured in DMEM (Life Technologies, U.S.A) or MEM (FuHeng, China), supplemented with 10% FBS (PAN, Germany), penicillin and streptomycin (100 U/ml). The cells were seeded in 24-well plates to be 80% confluent the following day. The cells were transfected with pUC19, pHBV1.3, pMyc-HBx, pMyc-ΔHBx, pMyc-HBs, and pMyc-HBc (1.0 μg each) using Lipofectamine 2000 (Thermo Fisher, 11668-019) in opti-MEM (Life Technologies) for 48 h. Negative control (si-NC) (sense: 5′-UUCUCCGAACGUGUCACGUTT-3′ and antisense: 5′-ACGUGACACGUUCGGAGAATT-3′) and USF1–small interfering RNA (siRNA) (sense: 5′-GCCGAGACAAGAUCAACAATT-3′ and antisense: 5′-UUGUUGAUCUUGUCUCGGCTT-3′) were purchased from GenePharma (China) and transfected into cells by RNATransMate (BBI, Shanghai, China) according to the instructions provided by the manufacturers.

The HBV infection experiment was performed as previously reported (22). Polyethylene glycol–concentrated HBV virion derived from HBsAg-positive patients’ serum with HBV DNA copies above 10⁷ IU/ml, using serum from HBV-negative patients as a negative control. HepG2-NTCP cells were exposed to HBV in the presence of 2% polyethylene glycol 8000 for 24 h, which means contact with the virus. Repeated washing using PBS cleared the residual free virus. At this point, the virus had entered the cell, and this moment was marked as day 0 (D0). The cell was cultured in the DMEM supplemented with 2.5% DMSO for indicated times.

HepG2.2.15 hepatocytes were seeded in the plate by 10⁵ cells/well. A 1.5 μg/ml concentration of entecavir (ETV) (Baraclude, China) and 5 μM FTY720 fluid (MedChemExpress, USA) were added to the culture medium, respectively, using untreated cells as control. The cells and supernatant were collected for subsequent studies after 3 d.

CRISPR/Cas9 Gene knockout kit was purchased from Suzhou Haixing Biological Technology Co. (Jiangsu, China), including one vector of gRNA sequence and specific homologous DNA fragment. The gRNA sequence is GTGAAGCGAAGTGCACACGG (23). At 3 d after viral infection of HepG2-NTCP cells, Cas9/gRNA was transfected into cells using Lipofectamine 2000, and Cas9 eliminated HBV replication under the gRNA’s lead.

The serum samples and cell culture supernatants collected from the study subjects were stored at −80°C before the measurement. S1P, Sph, dhSph, and dhS1P were measured using a high performance liquid chromatography–tandem mass spectrometry methodology. All the determinations were carried out as previously described in Central Laboratory, Xuhui Central Hospital, Shanghai, China (24). HBsAg (Abbott, USA) and HBeAg (Abbott) measured by chemiluminescence. HBV DNA was detected on a Roche LightCycler 480 (LC480) II quantitative real-time PCR (qPCR) platform using a PCR fluorescent diagnostic kit (Sansure Biotech, China) according to the manufacturer’s instructions.

Total RNA was extracted from frozen liver tissues and cultured cells using a TRIzol reagent (Thermo Fisher, 15596018) and quantified using Qubit 4 (Invitrogen). Total RNA (2 μg) was reverse transcribed, and mRNA was determined by quantitative real-time reverse transcription PCR (qRT-PCR) analysis using SYBR Green I Master (E096-01B, Novoprotein, China) and the appropriate primers (5′-GCTACGAGCTGCCTGACGG-3′ and 5′-TGTTGGCGTACAGGTCTTTGC-3′ for β-actin, 5′-GTCTGTGCCTTCTCATCTG-3′ and 5′-GTTCACGGTGGTCTCCAT-3′ for HBx, 5′-GCTCTGGTGGTCATGTCTGG-3′ and 5′-CACAGCAATAGCGTGCAGT-3′ for SphK1, and 5′-ATCGTGCAGCTCTCCAAGATAATCC-3′ and 5′-AGTTCTTCAGACAAGCGGTGGTTAC-3′ for USF1). The primers were synthesized by Sangon Biotech (Shanghai, China) Company. Relative changes in gene expression levels were determined using the 2-ΔΔct method. The cycle number at which the transcripts were detectable (Ct) was normalized to the cycle number of β-Actin gene detection, referred to as ΔCt.

The total cells were lysed in radioimmunoprecipitation assay buffer (Beyotime Biotechnology, China). The protein concentrations of the whole cells were determined by the BCA (Beyotime Biotechnology) method using the Beyotime protein assay kit. Cell lysate was separated in a 10% polyacrylamide gel and transferred to a nitrocellulose membrane. The membrane was blocked with 5% skim milk in PBST for 1 h and then incubated at 4°C overnight with anti–β-ACTIN (Proteintech, USA, 66009-1-Ig) and anti-SphK1 (Proteintech, 10670-1-AP). The membranes were washed with PBST and incubated for 1 h with HRP-conjugated secondary Ab (Proteintech, SA00001-2).

HEK293T cells were cotransfected using Lipofectamine 2000 transfection reagent (Thermo Fisher, 11668-019) with 0.1 μg of luciferase reporter plasmids pGL3-SphK1, and 0.1 μg pcDNA3.1-USF1 or its empty vector (pcDNA3.1) individually and 50 ng thymidine kinase promoter–Renilla luciferase reporter plasmid (transfection control) as described previously (25). 48 h later, the cell lysates were prepared, and the luciferase activities were determined by the dual-luciferase reporter assay system (Vazyme, DD1205-01) according to the manufacturer’s instructions.

The 6-wk-old mice were injected within 10 s via the tail vein with plasmid DNA diluted in PBS to a volume equivalent to 10% of the total body weight of each animal. Each mouse received 10 μg of pHBV1.3 plasmid or its vector control plasmid pUC19. Mice serum and liver samples were collected at the indicated times.

Chromatin immunoprecipitation (ChIP) assay was performed using the ChIP kit (CST, USA) following the manufacturer’s protocol. Briefly, HepG2 was cross-linked using 1% formaldehyde and then treated with glycine. Subsequently, ultrasonication was performed to break the chromatin. The chromatin fragments were precipitated with anti-IgG or anti-USF1 at 4°C. After treatment with NaCl and protease K, the DNA was purified, and the enrichment of SphK1 was measured by PCR.

Statistical significance for all experiments was determined by using Student t test or the Mann–Whitney rank sum test (GraphPad Prism 5 software). Error bars are reported as standard errors of the means, and significance was assigned for p values of <0.05.

In this study, 20 patients with CHB received direct anti-HBV therapy using NAs, which effectively blocked viral replication, significantly reducing HBV DNA copies in serum (Fig. 1A). Despite this, virus components (HBsAg and HBeAg) and transaminases (alanine aminotransferase and aspartate aminotransferase) in serum showed no significant changes (Fig. 1B–E), consistent with previous reports that ETV could profoundly suppress HBV DNA but rarely achieve HBsAg loss in the short term (26). Interestingly, levels of SphK-associated sphingolipid metabolites, including Sph, S1P, dhSph, and dhS1P, were evaluated, and only S1P and Sph were significantly reduced after antiviral treatment (Fig. 1F–I). A positive correlation between HBV DNA and serum S1P levels was observed (r = 0.4357, p = 0.0049) (Fig. 1J).

Hepatic tissues of randomly selected 19 patients, including 14 primary hepatitis B–related HCC and 5 HBV-negative HCC, were retrospectively analyzed to investigate SphK1 expression (Table I). Immunohistochemistry revealed varying degrees of HBsAg positivity, ranging from weakly positive (±) to strongly positive (1+ to 4+), with an uneven distribution and striking heterogeneity. SphK1 was commonly expressed in the liver tissues of all 19 patients, with localized high expression in a lamellar form. Notably, four patients exhibited significantly enhanced SphK1 expression in HBsAg-positive regions. This enhanced expression is illustrated in Fig. 2 (A and B) for patient 1 and in supplementary data for patients 2–4 (Supplemental Fig. 1). Furthermore, immunofluorescence staining confirmed the coexpression of SphK1 and HBsAg in the cytoplasm of the same hepatocyte (Fig. 2C, 2D). These findings indicate that HBV-colonized liver tissues are a significant source of high SphK1 expression in some HBV-related HCC patients.

To validate the correlation between HBV and SphK1, we established an HBV infection model in HepG2 cells overexpressing the HBV receptor NTCP (HepG2-NTCP) (19). After exposing HepG2-NTCP cells to HBV virions for 24 h, we repeatedly washed the cells with PBS to move residual virus, making this as day 0 (D0). HBV serum biomarkers (HBsAg, HBeAg, and HBV DNA) increased with HBV infection, peaking 3 days postinfection (Fig. 3A–C). Compared to uninfected controls, intracellular SphK1 and released S1P levels were strongly correlated with HBV replication during persistent infection (Fig. 3D–F). Interestingly, SphK1 expression decreased on day 0, likely due to compromised membrane integrity during massive viral entry. Because S1P is the final product of sphingolipid metabolism, and sphingolipids are ubiquitous components of mammalian membranes, this observation aligns with their role in cellular processes (27). These results suggest that HBV regulates SphK1 expression and S1P generation once the virus establishes a stable presence within the cell.

To further clarify the HBV-dependent alterations, we treated HBV-replicating HepG2.2.15 cells with antiviral agents. Upon suppression of HBV reproduction by the direct antiviral agent entecavir, intracellular SphK1, and extracellular S1P levels significantly decreased (Fig. 4A–C). Additionally, we used the CRISPR-Cas9 system (HBV-gRNA/Cas9) to cleave the HBV genome (28, 29). As shown in Fig. 4D, the plasmids (gRNA/Cas9) successfully reduced the HBV genome, which was accompanied by a significant reduction in viral protein Ag. As expected, the inhibition of HBV replication also significantly diminished SphK1 expression and the release of extracellular S1P (Fig. 4E, 4F). Given the enhancement of SphK1–S1P with the HBV infection, we further investigated its effects on HBV replication. We treated HepG2.2.15 cells with FTY720, a competitive analog of S1P, which blocked the SphK1–S1P–S1PRs axis and inhibited HBV DNA replication (Fig. 4A–C). However, merely knocking down SphK1 expression via siRNA did not significantly inhibit HBV replication (Supplemental Fig. 2), consistent with other reports (30, 31). These data indicate that the SphK1–S1P–S1PRs axis may influence HBV replication by more complex mechanisms.

The enhanced availability of promoter-bound TFs is crucial for upregulating promoter activity and gene expression. In HBV infection, the upregulation of SphK1 at the mRNA level suggests that it might be regulated by transcription factors (TFs). To identify the TFs regulating SphK1 expression during HBV infection, we analyzed publicly available ChIP sequencing datasets. We identified 118 TFs and visualized the top 15 based on their binding score (Fig. 5A). We then examined the expression of these 15 TFs in HBV plasmid-transfected Huh7 cells and HBV-infected HepG2-NTCP cells using qRT-PCR. Notably, HBV infection caused a substantial increase in USF1 mRNA compared with control cells, which corresponded with high SphK1 expression (Fig. 5B, 5C). Further examination of USF1 expression in HepG2-NTCP cells revealed a continuous increase in USF1 levels with persistent HBV infection, mirroring the trend in SphK1 expression (Fig. 5D). Additionally, transfecting Huh7 cells with different HBV ORF plasmids showed that SphK1 expression notably increased only with the replication-competent HBV plasmid pHBV1.3. As expected, the trend of USF1 expression was consistent with that of SphK1 (Fig. 5E). Taken together, these data suggest that the transcription factor USF1 plays a significant role in HBV-mediated enhancement of the SphK1–S1P pathway.

To investigate the regulatory effect of USF1 on SphK1 transcription, we constructed a luciferase reporter plasmid encoding the full-length SphK1 promoter and assessed whether USF1 regulates SphK1 transcription. HEK293T cells were cotransfected with USF1 plasmids and the SphK1 reporter plasmids, and fluorescence microscopy showed that USF1 localized in the nucleus (Fig. 6A). The luciferase reporter gene assay revealed that USF1 significantly enhanced the transcriptional activity of the SphK1 promoter (Fig. 6B). To determine whether USF1 could directly bind to the SphK1 promoter, we performed a ChIP assay. The results showed deeper binds for SphK1 promoter in the USF1 group compared with the IgG isotype control, indicating that the USF1 binds to the SphK1 promoter (Fig. 6C). To further elucidate USF1’s regulatory effect on SphK1 during HBV infection, we examined SphK1 expression following knockdown of endogenous USF1 in HBV-infected HepG2-NTCP and Huh7 cells. siRNA-mediated knockdown of USF1 significantly reduced USF1 expression levels (Fig. 6D, 6F) and resulted in a significant decrease in SphK1 expression compared with the si-NC group in both cell models (Fig. 6E, 6G). These results demonstrate that HBV infection triggers nuclear transfer and expression of USF1, which in turn activates the SphK1 promoter.

To investigate the effect of HBV on SphK1–S1P in vivo, 10 ug of pUC19-HBV1.3 plasmid was injected into mice by high-pressure tail vein. Serum and liver samples were collected from the mice on day 4. Serum testing revealed a significant enhancement of viral replication along with increased S1P expression (Fig. 7A). Liver assays showed that HBV replication led to increased expression of USF1 and SphK1 (Fig. 7B). Additionally, liver samples were collected on days 1, 2, 3, and 4. The results showed that HBV replication (indicated by HBx expression levels) gradually increased over time, with significant increases in USF1 and SphK1 expression on day 4 (Fig. 7C).

The SphK1–S1P axis has been recognized for its critical role in hepatic inflammatory injury, liver fibrosis, and HCC. S1P released from the liver in response to injury promotes the recruitment of various immune cells, leading to hepatic inflammation (32). Additionally, S1P regulates signaling and physiological processes across several pathways to promote HCC (16). Given the importance of SphK1–S1P in liver injury, this study focused on whether HBV could directly alter SphK1–S1P expression. Although SphK1–S1P–S1PRs signaling has been reported to influence viral activity and host defense immunity, few studies have explored the mechanisms by which the virus alters sphingolipid metabolism (2, 33)

Our findings reveal a relationship between HBV and SphK1–S1P in serum and liver tissue samples from patients. Retrospective analysis of patients treated with NAs for over 3 months showed a positive correlation between HBV DNA and S1P levels. SphK1 hyperexpression predominantly colocalized in HBV-positive regions of liver tissues from some HBV-positive HCC patients. These results suggest a connection between HBV and SphK1 in the process of HBV-related liver immune injury and tumorigenesis. In vitro, HBV-dependent SphK1 expression was observed at both the mRNA and protein levels, indicating that HBV infection can upregulate SphK1–S1P production. Serum S1P has the potential to be serve as a unique indicator for monitoring sphingolipid metabolism in HBV-infected hepatocytes. Current biomarkers for HBV replication mainly consist of the viral components or nucleic acids, which do not reflect the metabolic status of infected cells.

HBV infection induced upregulation of SphK1 at the mRNA level and increased activity, as indicated by elevated S1P, suggesting an association with TFs. Bioinformatic analysis of ChIP sequencing data indicated that several TFs might be involved in regulating SphK1 promoters. Notably, USF1 was verified as the strongest TF for SphK1 transcription according to our data. Although the role of USF1 in liver diseases is well established, its response to HBV infection remains controversial. Chen et al. (34) and Zeng et al. (35) reported that USF1 mRNA expression was increased in HCC lines compared with the normal liver cell lines and was not associated with HBsAg status. However, Baidya et al. (36) found that HBx-induced hypermethylation of the USF1 promoter led to the downregulation of USF1. AP2α is another TF related to target SphK1 transcription in response HBx (3). In our study, the complete replication activity of HBV, rather than individual viral components, appeared to be the key factor in enhancing USF1 activity, partially explaining these inconsistent reports. Overall, HBV infection manipulates sphingolipid metabolism by regulating potential TFs, which may respond to viral activity and viral components.

The SphK1–S1P–S1PRs axis influences various biological functions. S1P interacts with extracellular receptors to modulate downstream signals and acts as a secondary messenger directly involved in intracellular signal transduction and regulation (13). S1P modular, as active phospholipid-related drugs, have the potential to balance immune cell trafficking and activation, affect the cell cycle, inhibit viral replication, and perform other biological functions. For example, FTY720, a prodrug phosphorylated to an S1P analog, has been approved by U.S. Food and Drug Administration for the treatment of multiple sclerosis and cardio protection (37). Our results indicated that FTY720 not only inhibited the SphK1 expression but also interfered with HBV replication. Considering that simply knocking down SphK1 expression did not affect HBV replication, this suggests that FTY720 may interfere with viral activity through other complex pathways, such as acting on different S1PRs and relying on some positive/negative feedback mechanisms of S1P in different cells (38).

In summary, we discovered that HBV infection directly manipulated USF1 to enhance SphK1 transcription and increase S1P production. These underlying molecular mechanisms expand our understanding of the interplay between hepatitis B and sphingolipid metabolism. The potency of these pathways has led to the development of drugs targeting SphK1–S1P–S1PR signaling. HBV-mediated activation of SphK1–S1P may serve as a novel surveillance indicator for HBV-associated liver injury and provide new therapeutic targets for delaying HBV-associated liver injury, liver fibrosis, cirrhosis, and hepatocellular carcinoma.

The authors have no financial conflicts of interest.