Hepatitis B Virus Inhibits Neutrophil Extracellular Trap Release by Modulating Reactive Oxygen Species Production and Autophagy

Hepatitis B virus (HBV) is an important human pathogen that can cause a wide spectrum of liver diseases. Because of its chronicity and the increased risk of hepatocellular carcinoma, HBV has become a global burden (1). Neutrophils (polymorphonuclear neutrophil granulocytes [PMNs]) are effector cells involved in innate antimicrobial defense and are also related to liver disease, playing key functions in the pathogenesis of alcoholic hepatitis (2). Neutrophils employ three major strategies to combat microbes: phagocytosis, degranulation, and the release of neutrophil extracellular traps (NETs), a process referred to as NETosis (3), which differs from apoptosis and necrosis (4). NETs are extracellular fibrous structures composed of chromatin, histones, and several proteins, such as neutrophil elastase (NE), myeloperoxidase (MPO), cathepsin G, and proteinase 3 (PR3) (5–8). These large extracellular structures trap and kill a variety of microbes by exposing them to high concentrations of NET-associated microbicidal factors and providing a physical barrier to prevent microbial dissemination (5). In addition to their roles in infection, NETs were recently shown to have roles in various sterile diseases, such autoinflammatory and autoimmune diseases (9).

The intracellular signaling pathways that regulate NET formation are still largely unknown. This process depends on reactive oxygen species (ROS), such as superoxide, which are generated by the NADPH oxidase Nox2 (4). In addition, the activities of several enzymes such as PAD4, MPO, and NE have been implicated in NET formation; NE and MPO synergize to drive massive chromatin decondensation before plasma membrane rupture (10, 11). According to recent studies, the ROS-dependent activation of ERK and p38 MAPK mediates PMA-induced NET release from human neutrophils (12, 13). Autophagy, which is an essential mechanism for cell homeostasis and survival, plays an important role in immunity and inflammation via pathogen clearance mechanisms mediated by immune cells, including macrophages and neutrophils (14), and was recently shown to be required for NET release (15, 16). Wortmannin-mediated inhibition of autophagy in neutrophils leads to abnormal chromatin decondensation and prevents NET release (17). Furthermore, the inhibition of mammalian target of rapamycin (mTOR) by rapamycin accelerates NET release, which parallels increased autophagic influx (18).

HBV may reduce neutrophil responses. For example, neutrophils in patients with hepatic diseases present defects in the production of oxygen radicals (19), which are indispensable for proper NET release. Liver cirrhosis patients display a significant decrease in NET release, and liver cirrhosis patients with deficiencies in NET release display an increased rate of complications (20). Based on our data, HBV inhibits the release of NETs, which affects neutrophil function and interferes with the subsequent innate and adaptive immune responses against HBV, leading to the establishment of chronic infection. Moreover, the HBV C (HBc) protein and HBV E (HBe) protein enhance mTOR activity to reduce the autophagic activity of neutrophils and suppress the ROS-dependent activation of ERK and p38 MAPK to inhibit NET release.

The present study was conducted on 40 patients with chronic hepatitis B (CHB) infection (age 33.8 ± 4.3 y) who met the diagnostic criteria for HBV (WS299-2008) and were recruited from Shandong Provincial Hospital as well as 40 healthy controls without HBV infection or autoimmune diseases (age 31.4 ± 4.2 y). None of the 40 patients had been treated with antiviral drugs. All participants in this study provided written informed consent. The study protocol was approved by the ethics committee of the Institutional Review Board of Shandong Provincial Hospital Affiliated with Shandong University, Jinan, China.

Male C57BL/6 mice (wild type and HBV infected) were purchased from Shanghai Model Organisms Center. All animal studies were approved by the Institutional Animal Care and Use Committee of Shandong Provincial Hospital Affiliated with Shandong University. Animal handling and experiments were performed according to the local legislation titled “Regulations for The Administration of Affairs Concerning Experimental Animals.”

The efficiency of bacterial eradication in vivo was assessed as described previously (21). In brief, mice were i.p. administered with 200 μl of ampicillin-resistant Escherichia coli (1 × 10⁴/ml saline) for 6 h. Peritoneal lavage fluid was acquired using 10 ml of RPMI 1640 medium without serum. The number of surviving bacteria was determined by incubating the peritoneal lavage (95 μl) with 1% Triton X-100 (5 μl) for 10 min to lyse cells, and then serial dilutions were incubated overnight on agar plates at 37°C.

Neutrophils were isolated from the peripheral blood of donors via density centrifugation using Polymorphprep, according to the manufacturer’s recommendations (Axis-Shield) (22). The purity of granulocytes was >97%, as determined by flow cytometry using CD15-FITC (BD Biosciences). Cell viability was >95%, as determined by trypan blue exclusion (Sigma-Aldrich). The obtained cells were resuspended in serum-free M-199 media.

Purified human neutrophils (5 × 10⁵ cells/ml) were plated on poly-l-lysine–coated wells in a 48-well tissue culture plate and incubated for 30 min at 37°C in a 5% CO₂ atmosphere. Cells were then stimulated with 1 μM fMLF (Sigma-Aldrich) for 3 h. Then, 500 mU/ml micrococcal nuclease (Thermo Fisher Scientific) was added to digest the NETs. The reaction was terminated by the addition of 5 mM EDTA (Sigma-Aldrich), and the supernatants were collected and stored at 4°C until further use. The Quant-iT Pico Green dsDNA assay was used to quantify circulating free (cf)-DNA/NET levels, according to the manufacturer’s instructions (Invitrogen) (4, 23). The amount of DNA was reflected by fluorescence intensity and was measured at excitation and emission wavelengths of 485 and 530 nm, respectively, on a microplate reader (SpectraMax M2; Molecular Devices).

Serum was separated from blood samples obtained from patients with CHB infection for HBV-DNA load tests. Serum levels of the HBV markers hepatitis B surface Ag (HBsAg), hepatitis B surface Ab (HBsAb), hepatitis B E Ag (HBeAg), hepatitis B E Ab (HBeAb), and hepatitis B core Ab (HBcAb) were quantified using the Abbott HBV Quantitative Test Kit (chemiluminescent microparticle immunoassay) using an Architech i4000 particle chemiluminescence detection instrument. HBV-DNA levels were measured using an HBV nucleic acid quantitative detection kit (quantitative PCR–fluorescent probing method) and an Applied Biosystems 7500 fluorescence quantitative PCR instrument.

Purified human neutrophils (2 × 10⁶ cells/ml) were incubated with the HBV X (HBx) protein, HBc protein, HBe protein, or HBV S (HBs) protein (Abcam) at 37°C in a 5% CO₂ atmosphere, followed by fMLF (1 μM) stimulation for 3 h. Then, cells were incubated with fresh media containing DNase (40 U/ml) for 15 min at room temperature to degrade and release the NETs. The supernatant was gently removed and centrifuged at 420 × g for 5 min. The cell-free supernatant was then mixed with 4× loading buffer at a 3:1 ratio before Western blotting (24).

Purified human neutrophils (5 × 10⁵ cells/ml) were incubated with HBV protein for 1 h at 37°C in a 5% CO₂ atmosphere on poly-l-lysine–coated glass coverslips. Cells were then stimulated with 0.1 μM fMLF or 100 ng/ml PMA and imaged on an Olympus microscope. NETs were detected using a mixture of cell-permeable (Hochest 33342; Molecular Probes) and -impermeable (Sytox Green; Molecular Probes) fluorescent DNA dyes.

ROS production in neutrophils was measured via flow cytometry using CM-H2DCFDA (Invitrogen). Neutrophils (2 × 10⁶ cells/ml) were incubated with HBV proteins and stimulated with fMLF (1 μM). Then, CM-H2DCFDA (5 μM) was added and incubated for an additional 30 min at 37°C in the dark. The mean cellular fluorescence intensity was quantified by FACS analysis.

Cytochrome c reduction assays were used to determine the extent of ROS formation (25). Neutrophils (2 × 10⁵ cells/ml) were incubated with 1 mg/ml cytochrome c (Sigma-Aldrich) in the presence or absence of HBV proteins. After a baseline reading, cells were stimulated with 100 ng/ml PMA. Changes in absorbance at 550 nm were measured over a 30-min period.

Autophagy was detected by examining LC3 and SQSTM1/P62 proteins levels via Western blotting and flow cytometry using an autophagy detection kit (Abcam). Neutrophils (2 × 10⁶ cells/ml) were incubated with HBV proteins and then stimulated with 1 μM fMLF or 100 ng/ml PMA. Subsequently, collected cells were treated according to the manufacturer’s instruction or were subjected to protein extraction for Western blotting.

For analyses of signaling pathways, neutrophils were preincubated with inhibitors or DMSO for 30 min at 37°C. Rapamycin (mTOR inhibitor, Cell Signaling Technology) was freshly prepared for each experiment.

Proteins were extracted using RIPA Lysis and Extraction Buffer (Thermo Fisher Scientific). Western blotting was conducted using Abs against human histone H3 (citrulline R2+R8+R17), H3, NE (Abcam), LC3A/B, SQSTM1/P62, phospho-p44/42MAPK (ERK1/2, Thr²⁰²/Tyr²⁰⁴), p44/42 MAPK (ERK1/2), phospho-p38 MAPK (Thr¹⁸⁰/Tyr¹⁸²), p38 MAPK, phospho-p70S6K (Thr³⁸⁹), p70S6K, phospho-mTOR (Ser²⁴⁴⁸), mTOR, phospho-ULKI (Ser⁷⁵⁷), and ULK1 (Cell Signaling Technology), followed by HRP-conjugated goat anti-rabbit secondary Abs (Santa Cruz Biotechnology). The gray values of the bands were analyzed using ImageJ software. GAPDH was used as an endogenous control for normalization.

The data presented in this article were collected from a minimum of three independent experiments with neutrophils isolated from different donors. The results are presented as means ± SD. Statistical analyses were performed with SPSS software version 23. ANOVA were used to analyze the expression levels of Western results, and flow cytometry and correlations between the quantitative data from two groups were analyzed using Pearson correlation tests. A p value ≤ 0.05 was considered statistically significant.

HBV-infected mice and uninfected mice were given E. coli i.p., and the number of viable bacteria was then determined in peritoneal lavage fluid obtained 6 h later. As shown in Fig. 1A and 1B, HBV-infected mice had a decreased ability to eradicate bacteria compared with uninfected mice.

We assessed the isolated neutrophils of 40 patients with CHB infection and 40 healthy controls to determine whether NET release was suppressed by CHB infection. We used both the original values for cf-DNA/NETs and the ratio of cf-DNA/NETs to dsDNA (fMLF-stimulated to unstimulated cells) to express the NET releasing ability. As shown in Fig. 1C, the average dsDNA level in patients with CHB infection (1552 ± 160.9) was less than that in healthy controls (2152 ± 239.4), and similar results were obtained using the ratio of NETs to dsDNA (p < 0.05) (Fig. 1D).

Then, PMA-stimulated NET release was visualized using the cell-impermeable DNA dye Sytox Green, along with the cell-permeable DNA dye Hoechst 33342 (Fig. 1F). In addition, H3Cit expression levels declined in CHB infection (Fig. 1E). Thus, NET release is suppressed in patients with CHB infection, and HBV may inhibit NET release.

Serum levels of HBV Ags have been used in the clinic as an index of viral replication, infectivity, disease severity, and treatment response. We determined the serum levels of HBV markers and HBV-DNA load in patients with CHB infection to further elucidate the relationship between HBV and NETs. Then, we used a correlation analysis to determine the relationships between serum levels of HBV markers and NET release activity. As shown in Table I, the NET ratio negatively correlated with HBsAg, HBeAg, and HBcAb levels, but no obvious correlation was observed with HBeAb and HBsAb levels. Moreover, the NET ratio was not significantly correlated with HBV-DNA load (Table II). Thus, HBV proteins exert a complicated effect on NET release, with different proteins having different functions.

We cultured neutrophils from healthy people with HBs, HBx, HBc, and HBe proteins and PBS to further characterize the effects of HBV proteins on NET release. Cells were stimulated with fMLF or PMA and then examined to measure the levels of cf-dsDNA, H3Cit, and NE as well as by live-cell imaging. As shown in Fig. 2A, cf-DNA/NET levels were decreased in the presence of the HBc and HBe proteins, suggesting that these proteins inhibit NET release, whereas the HBs and HBx proteins had no effect. In addition, treating neutrophils with the HBc and HBe proteins resulted in a significant increase in the Sytox-positive population compared with treatment with PBS (Fig. 2C). Western blotting for H3Cit and NE also confirmed these findings (Fig. 2B).

We treated neutrophils from healthy people with CM-H2DCFDA or cytochrome c after coculture with HBV proteins and fMLF or PMA stimulation to determine whether the diminished ROS production induced by HBV proteins directly inhibited NET release. Because ROS were produced at high levels in each group, we used the median value to represent ROS production. According to flow cytometry, the HBc and HBe proteins reduced ROS production in neutrophils (Fig. 3A), but the HBx and HBs proteins exerted no significant effects. In addition, cytochrome c reduction assays (Fig. 3B) showed that oxidative bursting by neutrophils incubated with PMA for 90 min was decreased by coculture with HBc or HBe protein.

As mentioned above, autophagy is essential for NET release. LC3 protein levels are positively correlated with autophagic activity, whereas SQSTM1/P62 levels are negatively correlated. By performing Western blotting for LC3 and SQSTM1/P62 in neutrophils treated with HBV proteins and fMLF, we detected the impacts of HBV proteins on autophagy during NETosis. As shown in Fig. 3D, the HBc and HBe proteins inhibited autophagy, and flow cytometry–based autophagy detection yielded similar results (Fig. 3C). Therefore, these experiments showed that the HBc and HBe proteins inhibited ROS production and autophagy, leading to decreased NET release, which may be one mechanism by which HBV evades the immune system.

As the ERK and p38 MAPK pathways are involved in fMLF-induced, ROS-dependent NET release, phosphorylation of these molecules was assessed by Western blotting. Significant decreases in the phosphorylation of both p38 MAPK and ERK1/2 were observed in cells treated with HBc or HBe protein (Fig. 4A). Thus, the HBc and HBe proteins inhibited the p38 MAPK and ERK1/2 pathways to reduce ROS production.

mTOR inhibits autophagy initiation by negatively regulating the translocation of proteins associated with the autophagy machinery, such as the ULK complex, to autophagy-related structures (16), and the mTOR pathway was recently shown to contribute to the regulation of NET formation (18). We examined the effects of the HBc and HBe proteins on mTOR signaling to investigate the mechanisms by which HBV proteins inhibit autophagy. The HBc and HBe proteins activated mTOR signaling, as evidenced by increased levels of phosphorylated mTOR and ULK1 (Fig. 4B). Furthermore, we detected increased levels of p-P70S6K, a downstream target of mTOR (Fig. 4B). Thus, the HBc and HBe proteins may enhance mTOR pathway activity and subsequently inhibit NET release.

We used the specific mTOR inhibitor rapamycin in PMNs cultured in the presence of HBc or HBe protein to confirm the role of the mTOR pathway in the inhibitory effects of the HBc and HBe proteins on NET release. As shown in Fig.5A, incubation with 10 nM rapamycin for 30 min was sufficient to inhibit mTOR activity. As expected, rapamycin decreased p-mTOR levels, whereas total mTOR levels remained unaltered (Fig. 5B). Rapamycin treatment increased H3Cit expression and cf-DNA/NET levels. HBc or HBe protein combined with rapamycin treatment had no obvious effect on H3Cit expression and cf-DNA/NET levels (Fig. 6A, 6B). Direct microscopic observation also yielded similar results (Fig. 6C). These data suggest that decreased mTOR pathway activation is required for NET release.

HBV infection remains a global health problem; however, there is still no effective cure for HBV infection. HBV persistence is often associated with the production of large amounts of viral proteins, such as HBsAg and HBeAg (26). Neutrophils, as the foot soldiers of the immune system, play an important role in defending against various microbial infections via phagocytosis, degranulation, and the release of web-like structures called NETs. Recent studies have examined the role of NET release in the response to viral infections. HIV-1 induces NET release via the cell death pathway, after which NETs can capture HIV virions and significantly decrease HIV infectivity. However, HIV also manipulates neutrophil activation to suppress NET formation (27). Feline leukemia virus inhibits NET formation by inhibiting PKC activation to reduce ROS production (28). In the current study, we showed that NET release is decreased in patients with CHB infection, indicating that HBV proteins suppress NET formation and prevent the effective removal of the virus. This mechanism may be a cause of HBV persistence. These studies addressed the role of NETs in viral infection and indicated that the virus can coevolve and adapt to the innate immune response to avoid clearance.

The HBV genome contains four open reading frames, and the core open reading frame encodes both HBcAg and HBeAg, which share substantial sequence overlap (29, 30). HBeAg is not essential for replication or acute infection in vivo (31, 32) but instead has an immunoregulatory function in promoting viral persistence (33). Infants born to HBeAg-negative, HBsAg-positive carrier mothers are likely to develop acute hepatitis, but they progress to chronic disease less frequently (34, 35). Thus, HBeAg may play an essential role in HBV persistence, but the mechanisms by which HBeAg helps establish persistent infection are unclear. As shown in a recent study by Chen and coworkers, circulating HBeAg has a potential to deplete HBeAg- and HBcAg-specific Th1 cells (36), suggesting that HBeAg may regulate T cell–mediated immune responses. Furthermore, based on studies performed in animal models, HBeAg may actively suppress innate immune responses (37). We analyzed the correlation between HBV Ags and NET activity. Serum HBeAg levels negatively correlated with NET release, and HBe protein inhibited NET release, which may be a mechanism by which HBV escapes the immune system and establishes chronic infection.

HBcAg is the major constituent of the nucleocapsid and is essential for viral replication and assembly (33, 38). HBcAg functions as both a T cell–dependent and –independent Ag (39). HBcAg-specific CTLs control HBV replication and the progression of liver damage (40). Given the important role of T cells in the antiviral process and the specificity of HBcAg, most studies of HBcAg are related to T cells. Surprisingly, serum HBcAb levels were negatively correlated with NET release, and the HBc protein inhibited NET formation in the current study. Activated neutrophils modulate T cell proliferation (41). HBcAg may inhibit neutrophil activation, leading to T cell dysfunction and ultimately enabling HBV replication and persistent HBV infection. Moreover, inhibited NETs do not effectively trap and kill the virus, which may be another reason for HBV persistence. HBsAg is the first indicator of and an important biomarker for the diagnosis of HBV infection (42).

High circulating HBsAg levels in patients can contribute to the hampered immune response (43). According to the correlation analysis, high HBsAg levels reduced NET activity, although HBsAg did not directly impact NET activity. HBsAg was recently shown to directly contribute to myeloid dendritic cell dysfunction to escape the immune system (43). HBx plays an important role in HBV replication and infection (44), but few studies have examined neutrophils, and we found that the HBx protein had no effect on NET formation.

In the current study, we have shown that neutrophils would decrease NET release in response to stimulation and hepatitis B proteins treatment. Both ROS and autophagy are required for NETosis, as inhibition of either autophagy or ROS production prevents NET formation (45). We found that the HBc and HBe proteins suppress ROS production, which directly induces NETs, indicating that these proteins inhibit NET by suppressing ROS production. In stimulated neutrophils, NOX2-dependent ROS have been shown to activate intracellular signaling cascades (ERK and p38 MAPK), which mediate NETosis. HBeAg suppresses p38 MAPK phosphorylation induced by TLR2 or TLR4 agonists in blood monocytes (46) and suppresses the respiratory burst in monocytes and neutrophils, which may be a cause of reduced ROS production; in contrast, no significant change was observed with HBs treatment (37). Several reports showed that ERK inhibition attenuates fMLF-stimulated superoxide release and p38 MAPK inhibition resulted in impaired neutrophil chemotaxis (47). We observed that HBc and HBe proteins decrease phosphorylation of p38 MAPK and ERK. Therefore, p38 MAPK and ERK are involved in the inhibition of PMA- or fMLF-induced ROS and NET release under hepatitis B proteins treatment. Furthermore, recent studies indicated that neutrophil autophagy primes neutrophils for increased NET formation, which is important for proper neutrophil effector functions (48, 49). The mTOR pathway negatively regulates NET formation by modulating autophagy in cells stimulated with fMLF. Our data provide evidence that the HBe and HBc proteins activate mTOR signaling to inhibit autophagy in neutrophils, which may lead to decreased NET release. Notably, treatment with an mTOR inhibitor (rapamycin) exerted a significant effect on NET formation, further suggesting that the HBc and HBe proteins suppress NET formation through an mTOR pathway–dependent mechanism. The mTOR pathway contributes to the regulation of NET formation via autophagy downstream of formyl-peptide receptor signaling in human neutrophils (18). These results identified a new mechanism by which mTOR can regulate innate as well as adaptive immune responses. However, studies in human hepatoma cell lines, such as HepG2 or Huh7 cells, revealed that HBV enhances and uses autophagy to replicate its DNA, which is mediated by the X protein (50, 51). Although the HBx protein efficiently induces autophagosome formation, this mechanism is independent of the mTOR signaling pathway (52). The persistent activation of autophagy in hepatocytes by HBV during chronic infection may play an important role in HBV pathogenesis. The inhibition of autophagy in neutrophils by the HBc or HBe protein or the induction of autophagy in hepatocytes by the HBx protein is essential for HBV replication and persistence.

Hepatitis B proteins can suppress ROS production and subsequently attenuate ROS-induced biological functions of neutrophils, which might interfere with the proper immune responses. Recent studies showed that mTOR plays an important role in the modulation of both innate and adaptive immunity. mTOR regulates diverse functions of professional APCs, such as macrophages and dendritic cells, and has important roles in the activation of T cells (53). The impairment of neutrophils by HBV proteins may modulate the antiviral responses. First, the number of APCs might be reduced at infection sites, leading to a delayed or weak adaptive immune response. Second, local inflammation response might be attenuated because macrophages and neutrophils are key players in inflammation, and they produce IL-12 to stimulate NK cells to secrete IFN-γ. The presence of IL-12 and IFN-γ will promote Th1 differentiation, which in turn enhances cytotoxic T cell activity. Thus, through disturbing the functions of neutrophils, hepatitis B proteins may modulate the innate response and might delay viral clearance, eventually leading to the establishment of chronic infection (37).

Because of the sample size and methods, our study has some limitations. We detected only ROS levels and the activation of the mTOR pathway during HBV-inhibited NET release. Many pathways and factors are potentially related to HBV and NET formation and should be studied. For example, NET formation also depends on TLR2 and complement factor 3. The expression of TLR2 on peripheral monocytes was recently reported to be significantly reduced in patients with HBeAg-positive CHB compared with patients with HBeAg-negative CHB and controls (46).

In summary, HBV inhibits NET release to escape being trapped and killed. Moreover, we found that the HBc protein and HBe protein enhance mTOR activity to reduce the autophagy activity of neutrophils and suppress the ROS-dependent activation of ERK and p38 MAPK to inhibit NET release. Furthermore, HBV may reduce the responses of neutrophils, which play important roles in innate immunity and inflammation. These changes may be a mechanism by which HBV escapes the immune system. HBV may delay viral clearance and eventually facilitate the establishment of persistent infection in this manner.

The authors have no financial conflicts of interest.