Glucose-6-phosphate dehydrogenase deficiency facilitates hepatitis E virus entry and aggravates liver injury Authors: Dongxue Chen, Xiaoxia Hu, Yuanwen Xue, Fuwen Xia, Min Li, Jiankun Liu, Fen Huang Published: 2025 ABSTRACT BackgroundGlucose-6-phosphate dehydrogenase (G6PD) deficiency (G6PDd) is an common enzyme deficiency disease, inducing hemolysis, hyperbilirubinemia, and jaundice, as well as liver dysfunction, in approximately 400 million patients. Patients with G6PDd are more susceptible to viruses infection than those without. However, the incidence and severity of hepatitis E virus (HEV) infection in patients with G6PDd are largely unknown.MethodsThe prevalence of HEV in patients with G6PDd was investigated. Susceptibility to HEV was evaluated in a hepatoma cell line with G6PD knockdown, and the interaction between HEV and G6PD was assessed.ResultsThe prevalence of HEV infection was higher in patients with G6PDd (23.8%, 25/105) than in patients with normal G6PD levels (0.65%, 2/307), indicating that patients with G6PDd are susceptible to HEV infection. Higher rates of hyperbilirubinemia (64% vs. 36.25%) and pneumonia (28% vs. 12.5%) were observed in HEV-infected patients with G6PDd than in patients with normal G6PD values. G6PD knockdown facilitated HEV entry and aggravated oxidative stress with the significant inhibition of glutathione to benefit viral replication but was restricted by NADPH reduction. Co-IP and colocation assays revealed that HEV ORF3 interacted with G6PD. HEV infection stimulated the expression of G6PD in a manner dependent on nuclear factor E2–related factor 2 activation. Consequently, severe apoptosis was observed in HEV-infected cells with G6PD knockdown.ConclusionsPatients with G6PDd are susceptible to HEV infection. The facilitation of HEV entry and aggravation of oxidative stress may contribute to HEV susceptibility in patients with G6PDd and severe disease.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12985-025-03031-y. Background Glucose-6-phosphate dehydrogenase deficiency (G6PDd), a common inherited enzyme deficiency disease in humans, leads to pentose phosphate pathway (PPP) abnormalities, inducing hemolysis, hyperbilirubinemia, and jaundice, as well as kidney and liver dysfunction, in an estimated 400 million people worldwide [1–3]. Persistent hemolysis due to G6PDd can cause severe hyperbilirubinemia. Hyperbilirubinemia in neonates with G6PDd is primarily due to reduced hepatic conjugation and excretion, as well as increased bilirubin production from hemolysis [4]. G6PDd may serve as a cofactor for chronic liver disease and is associated with liver failure [3]. G6PD deficiency is associated with increased cellular oxidative stress. Consequently, affected patients exhibit a heightened susceptibility to viral infections compared to individuals with normal G6PD activity. Most respiratory viral infections have a strong association with glucose-6-phosphate dehydrogenase (G6PD) [5]. For example, G6PDd increases the risk of severe diseases and morbidity rates in SARS-Cov-2-infected individuals [6, 7]. In patients with G6PDd exposed to SARS-CoV-2, the amount of nicotinamide adenine dinucleotide phosphate (NADPH) reduced, increasing susceptibility to viral infection. Meanwhile, the risk of hepatitis B virus (HBV) infection in females with low G6PD activity is 71% higher than that in those with normal G6PD activity [8]. Cellular susceptibility to oxidative stress is heightened by G6PD deficiency. This oxidant-sensitive state is implicated in exacerbating viral infectivity and could contribute to the mechanism of HBV replication [9]. HAV infection in G6PD-deficient patients has been reported to cause acute renal failure, with or without concurrent acute liver failure [10, 11]. However, the prevalence of hepatitis E virus (HEV), the most common hepatitis virus, in patients with G6PDd is still unexplored. As a major pathogen of hepatitis worldwide, HEV causes acute liver damage with symptoms including jaundice, elevated liver enzymes and bilirubin, diarrhea, and anorexia [12]. Chronic HEV infection has been reported in immunosupressed or immunodeficient individuals, including organ transplant recipients and patients infected with human immunodeficiency virus [13–16]. HEV infection causes liver failure in individuals with underlying liver diseases or pregnant women in their late trimester of pregnancy [17, 18]. HEV infection leads to a maternal mortality of 25%, preterm births, stillbirths, and miscarriages [19, 20–23]. Fulminant hepatic failure is the typical symptom in pregnant women with HEV infection, jaundice, and acute hepatitis failure [23]. Fetuses and newborns are vertically infected by HEV through the placentas or amniotic fluid of their mothers. Neonatal jaundice is usually observed in these newborns [21, 24, 25, 26]. However, the severity of HEV in individuals with G6PDd is unreported. In the present study, the prevalence of HEV infection in individuals with G6PDd was investigated. Individuals with G6PDd were highly susceptible to HEV infection compared with individuals with normal G6PD values (23.8% vs. 0.65%). The relationship bewteen G6PD and HEV infection was explored in vivo and in vitro to clarify why individuals with G6PDd are prone to infection by HEV. Materials and methods Ethical statement Patients (male, n = 131; female n = 281) with jaundice or elevated liver enzyme levels were collected and divided into the G6PD-deficient group (<1300, n = 105) and the normal G6PD group (1300–3660, n = 307) on the basis of G6PD values. Serum samples (n = 412) were collected from patients in Kunming City, China, from 2016 to 2022. Samples were stored at −80°C until use. The study protocol was approved by the ethical committee of 920th Hospital of Joint Logistics Support Force of PLA (No. 920IEC/AF/21/2019-01.0.0.0.0). Virus and viral challenge in vitro Genotype 4 swine HEV (KM01, GenBank accession no. KJ155502, 2.0 × 107copies/ml) was inoculated into the HepG2 cells (human hepatoma carcinoma cell line, ATCC: HB-8065) supplemented with 10% (v/v) fetal bovine serum (FBS), 100 U/ml penicillin, and 100 µg/ml streptomycin at 37 °C or 4 °C under 5% CO2. Cells were harvested for gene quantification through quantitative real-time PCR (qRT-PCR), protein detection through immunofluorescence analysis (IFA) or Western blotting at indicated time points. Cells and transfection HepG2 cells and HEK293T cells (human embryonic kidney 293 T cell line, ATCC: CRL-11268) were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37 °C with 5% CO2. For knockdown of G6PD, siRNA (5'-AAUCAACUGUCGAACCACATT-3') targeting G6PD were transfected into HepG2 cells for 72 hours and collected for protein detected through Western blotting. For G6PD and HEV interaction analysis, the pcDNA3.1-G6PD plasmid was separately cotransfected with the eight fragments of HEV, including pcDNA3.1-Met-Flag, pcDNA3.1-Y-Flag, pcDNA3.1-PCP-Flag, pcDNA3.1-HPX-Flag, pcDNA3.1-Hel-Flag, pcDNA3.1-RdRp-Flag, pcDNA3.1-ORF2-Flag, or pcDNA3.1-ORF3-Flag. Cells were collected at 72 h post-transfection, and whole-cell lysates were prepared and subjected to Western blotting analysis. Animal experiments BALB/c mice (6~8 weeks old, female) were purchased from Kunming Medical University and maintained in a pathogen-free animal facility. Animal experiments were approved by the Animal Care and Use Committee of Kunming University of Science and Technology. All procedures were performed under ketamine anesthesia by trained personnel under the supervision of veterinary staff. Mice negative for anti-HEV (IgG and IgM) and HEV RNA were included in the study. BALB/c Mice (n = 12) were separately intravenously injected with 200 μL HEV suspension. Mice were humanely euthanized at 7, 14, 21 and 28 days post injection (dpi). Mice (n = 3) injected with saline solution served as negative control. Livers were collected from HEV-infected or mock mice for genes quantitation or proteins detection. Gene quantification Total RNA was isolated from serum samples, livers or cells by using TRizol (Invitrogen, America) in accordance with the manufacturer's instructions. cDNA was prepared by using AMV Reverse Transcriptase XL (Takara, Japan) in accordance with the provided directions. The copy number of HEV was quantified through SYBR green-based qRT-PCR with HEV-specific primers as described in our previous study [27]. G6PD and NRF2 were quantified through SYBR green-based qRT-PCR. GAPDH served as a housekeeping control. Relative gene expression was calculated through the 2−△△CT>method. qRT-PCR was performed with a Bio-Rad CFX96TM Real-Time PCR System. Western blotting analysis and coimmunoprecipitation For Western blotting, cells were collected at indicated time points and lysed in RIPA buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 5 mM EDTA, pH 8.0, 30 mM NaF, 1 mM Na3VO4, 40 mM β-glycerophosphate, 0.1 mM PMSF, protease inhibitors, 10% glycerol, and 1% Nonidet-P40). An equivalent amount of total protein was separated through 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis then transferred onto a nitrocellulose membrane. Nonspecific binding sites were blocked with 5% skimmed milk, and the membrane was incubated overnight separately with G6PD antibodies (Abcam, 1:500 dilution) at 4 °C. HRP‐conjugated IgG was used as a secondary antibody (Abcam, 1:5000 dilution). GAPDH served as the loading control. Bands were exposed to X‐ray by using an Immobilon ECL Kit (Millipore), and the intensity of bands was assessed by using Image J software. For Co-IP, HEK-293T cells were cotransfected with pcDNA3.1-G6PD and pcDNA3.1-ORF3-Flag for 48 h, disintegrated with RIPA, and centrifuged at 14 000 × g at 4 ℃ for 15 min. Whole cell lysates were subjected to 10% SDS–PAGE and Western blotting analysis by using specific antibodies. For Co-IP, the lysates were incubated with specific antibodies at 4 ℃ overnight then incubated with Magna ChIP™ Protein A + G Magnetic Beads (Millipore, 16–663) (20 μL) at 4 ℃ for 2 h. The beads were washed five times with lysis buffer, resuspended in an equal volume of RIPA buffer, and analyzed through immunoblotting. Immunofluorescence assays and immunohistochemistry Cells and tissues were collected and fixed with 4% paraformaldehyde. Tissues were embedded in paraffin and cut into 3 µm serial sections. For immunohistochemistry (IHC) and immunofluorescence assay (IFA), slides were deparaffinated, hydrated, heated in a water bath for antigen retrieval, and blocked with 3% H2O2 for 15 min. The sections were then incubated overnight at 4 °C with the G6PD antibody (PTMAL, PTM-5473, China), HEV ORF2 antibody (Abcam, ab233244), or LC3 antibody (Cell Signaling Technology, 12741); washed with PBS; and stained for visualization through incubation for 1 h at 37°C with secondary antibodies labeled with HRP (BOSTER, SV0004), FITC (ABclonal, AS019), or TRITC (Abberior STAR RED). For IHC, slides were sealed with neutral balsam, inspected, and photographed under a microscope (Nikon, E200, Japan). For IFA, cells were stained with DAPI for 15 min and viewed under a fluorescence microscope (Nikon, Ts2, Japan), confocal microscope (Nikon, AXR, Japan), or stimulated emission depletion (STED) microscope (Olympus, STEDYCON, Japan). Measurement of enzyme activity The kit for G6PD measurement (substrate method) were purchased from Pinnaclesdx, China, and MDA (malondialdehyde), SOD (superoxide dismutase), GSH (glutathione), and NADPH (nicotinamide adenine dinucleotide phosphate) were purchased from NanJing JianCheng Bioengineering Institute of China. The activities of G6PDH, MDA, SOD, GSH, or NADPH in the serum of patients or lysates from cells were measured in accordance with the instructions. Terminal deoxynucleotidyl transferase nick end labeling assay Apoptotic cells in tissues were evaluated in situ by using a one-step terminal deoxynucleotidyl transferase nick-end labeling (TUNEL) assay kit (Beyotime, C1086). Briefly, tissue sections were deparaffinized, rehydrated, and digested with DNase and Proteinase K (20 μg/ml) for 30 min at 37 °C. Slides were incubated with the TUNEL detection kit for 60 min at 37 °C. Nuclei were counterstained with DAPI. Subsequently, the TUNEL specimens were observed under a fluorescence microscope (Nikon, Ts2, Japan). Phylogenetic analysis The nucleotide sequences of the amplified PCR products and prototypes of different genotypes of HEV strains were aligned by using MEGA 4.1 software (version 4.10, http://www.megasoftware.net). The reference sequences of prototype HEV strains were obtained from GenBank database. The standard classification of HEV genotypes and subtypes in previous studies [28, 29]. Phylogenetic trees were generated by using the minimum evolution and interior branch methods. Bootstrapping with 1000 resamplings of the data was performed to calculate branch percentages. The identity between nucleotide sequences was calculated by employing the MegAlign program (DNAstar package version 5.03). Statistical analysis All experiments were performed at least three times. Data were presented as mean ± standard deviation. Statistical analysis was conducted by using GraphPad Prism 7 software and SPSS Statistics 26.0. Student's t-test (paired t-test) was applied to determine the significance of differences between two groups. The chi-square test was used to compare the prevalence of HEV at different ages and the association between HEV infection and clinical diagnosis. p< 0.05 was considered statistically significant. Result Patients with G6PDd are highly susceptible to HEV infection G6PDd has been reported to be a predisposing factor for COVID-19. Individuals with G6PDd are more susceptible to coronavirus infection than those with normally functioning G6PD [7]. However, whether individuals with G6PDd are more susceptible to HEV infection than those without is unknown. Therefore, the prevalence of HEV infection in individuals with G6PDd (n = 105) and normal G6PD values (n = 307) in Yunnan Province, where HEV is highly endemic, was investigated. Notably, approximately 23.8% (25/105) of individuals with G6PDd were positive for HEV RNA, whereas only 0.65% (2/307) of individuals with normal G6PD were positive for HEV (Fig. 1A, Table 1), indicating that individuals with G6PDd are more susceptible to HEV infection than those without. The copy number of HEV in the serum of HEV-positive individuals with G6PDd was quantified through qRT-PCR. A positive correlation was found between HEV copy number and G6PD activity (Fig. 1B). Furthermore, six strains from HEV-positive patients with G6PDd were amplified, sequenced, then deposited in Genbank database (GenBank No. ON159307, ON159309, ON159310, ON159305, ON159308, and ON159306). Phylogenetic analysis was performed on the basis of partial ORF2 sequences to find homology relationships. The phylogenetic tree clearly illustrated that all HEV isolates belong to genotype 4 (Fig. 1C). Importantly, these HEV strains closely clustered with human HEV (KR87415) and HEV strains isolated from cow (KU356188), goat (KUG56185), or monkey (LC428042) in Kunming City or Dali City in Yunnan Province. Fig. 1G6PD-deficient patients are more susceptible to HEV infection. (A) The positive rate of HEV RNA in individuals with normal G6PD or G6PDd. (B) The copy number of HEV in individuals with G6PDd and correlation analysis between HEV RNA copy number and G6PD activity. (C) Phylogenetic analysis of HEV strains identified from individuals with G6PDd. Six partial ORF2 sequences (ON159307, ON159309, ON159310, ON159305, ON159308 and ON159306; highlighted by dark triangles) amplified from serum samples of HEV infected individuals were assigned to genotype 4. G6PD-deficient patients are more susceptible to HEV infection. (A) The positive rate of HEV RNA in individuals with normal G6PD or G6PDd. (B) The copy number of HEV in individuals with G6PDd and correlation analysis between HEV RNA copy number and G6PD activity. (C) Phylogenetic analysis of HEV strains identified from individuals with G6PDd. Six partial ORF2 sequences (ON159307, ON159309, ON159310, ON159305, ON159308 and ON159306; highlighted by dark triangles) amplified from serum samples of HEV infected individuals were assigned to genotype 4. Knockdown of G6PD facilitates HEV entry and replication HEV entry and replication were evaluated in HepG2 cells with G6PD knockdown by using RNA interference technology to clarify why patients with G6PDd are susceptible to HEV infection. Notably, Western blotting analysis revealed that the expression of G6PD significantly reduced in cells transfected with siRNA targeting G6PD (Figs. 2A and B). qRT-PCR showed that G6PD mRNA expression also significantly decreased in cells with G6PD knockdown (Fig. 2D). Meanwhile, G6PD activity significantly decreased (Fig. 2F), indicating that G6PD was successfully knocked down. HBV has been reported to enhance G6PD expression by activating nuclear factor E2–related factor 2 (NRF2) to facilitate its replication in hepatocytes [9, 30]. Thus, the expression of NRF2 was determined at the mRNA and protein levels. Consistent to G6PD expression, the mRNA and protein levels of NRF2 significantly decreased (Figs. 2A, C and E). Subsequently, the entry of HEV into cells with G6PD knockdown and those with normal G6PD levels were compared through STED microscopy at 4 ℃ or 37 ℃ for 2 h. The single virion has been obversed through STED microscopy. G6PD (green) was significantly inhibited in cells transfected with siRNA targeting G6PD (Fig. 2G and H). Meanwhile, HEV ORF2 (red) significantly increased in cells with G6PD knockdown relative to in cells with normal G6PD levels at 4 ℃ and 37 ℃ (Fig. 2G and I), indicating that low G6PD activity may benefit HEV entry and increase susceptibility. Furthermore, cells with G6PD knockdown were used to evaluate the replication ability of HEV by using IFA. Notably, significantly increased HEV ORF2–positive fluorescent signals emerged in cells with G6PD knockdown relative to in cells with normal G6PD activity (Figs. 2J and K). G6PD is responsible for the production of reducing equivalents of NADPH, which is used to regenerate the reduced form of GSH to restore redox homeostasis. Therefore, the changes in GSH were determined by ELISA in patients with G6PDd and HepG2 cells subjected to G6PD knockdown then infected or not infected with HEV. Remarkably, G6PDd disrupted redox homeostasis and increased oxidative stress, as well as significantly reduced GSH levels (Figs. 2L and M). The reduction in GSH is beneficial for viral replication and contributes to increased HEV replication. Fig. 2Knockdown of G6PD facilitates HEV entrance and replication. (A) G6PD expression in HepG2 cells transfected with or without siRNA targeting G6PD by Western blotting. Relavite expression of G6PD (B) and NRF2 (C) in cells transfected with or without siRNA targeting G6PD. mRNA expression of G6PD (D) and NRF2 (E) in cells transfected with or without siRNA targeting G6PD by qRT-PCR. (F) G6PDH determined in cells transfected with or without siRNA targeting G6PD by ELISA. (G) The ability of HEV entry into cells membrane in cells transfected with or without siRNA targeting G6PD were evaluate by IFA using STED microscopy. G6PD were inoculated with specific anti-G6PD antibody (green), HEV ORF2 were inoculated with specific anti-HEV antibody (red), Nucleus were stained with DAPI (blue). (H) The percentage of G6PD positive cells in cells transfected with or without siRNA targeting G6PD at 4℃ and 37℃for 2 h. (I) The intensity of ORF2 (red) in cells pre-transfected with or without siRNA targeting G6PD for 24 h, then inoculated with HEV for 2 h at 4℃ and 37℃. (J) HEV replication in cells pre-transfected with or without siRNA targeting G6PD for 24 h, then inculated with HEV for 72 h at 37℃ by IFA. Nucleus were stained with DAPI (blue). (K) The percentage of HEV ORF2 positive cells in HEV-infected cells transfected with or without siRNA targeting G6PD at 37℃. (L) GSH were determined in G6PD knockdown cells infected with or without HEV. (M) GSH were determined in G6PDd patients infected with or without HEV. *p < 0.05; **p < 0.01; ***p < 0.001 Knockdown of G6PD facilitates HEV entrance and replication. (A) G6PD expression in HepG2 cells transfected with or without siRNA targeting G6PD by Western blotting. Relavite expression of G6PD (B) and NRF2 (C) in cells transfected with or without siRNA targeting G6PD. mRNA expression of G6PD (D) and NRF2 (E) in cells transfected with or without siRNA targeting G6PD by qRT-PCR. (F) G6PDH determined in cells transfected with or without siRNA targeting G6PD by ELISA. (G) The ability of HEV entry into cells membrane in cells transfected with or without siRNA targeting G6PD were evaluate by IFA using STED microscopy. G6PD were inoculated with specific anti-G6PD antibody (green), HEV ORF2 were inoculated with specific anti-HEV antibody (red), Nucleus were stained with DAPI (blue). (H) The percentage of G6PD positive cells in cells transfected with or without siRNA targeting G6PD at 4℃ and 37℃for 2 h. (I) The intensity of ORF2 (red) in cells pre-transfected with or without siRNA targeting G6PD for 24 h, then inoculated with HEV for 2 h at 4℃ and 37℃. (J) HEV replication in cells pre-transfected with or without siRNA targeting G6PD for 24 h, then inculated with HEV for 72 h at 37℃ by IFA. Nucleus were stained with DAPI (blue). (K) The percentage of HEV ORF2 positive cells in HEV-infected cells transfected with or without siRNA targeting G6PD at 37℃. (L) GSH were determined in G6PD knockdown cells infected with or without HEV. (M) GSH were determined in G6PDd patients infected with or without HEV. *p < 0.05; **p < 0.01; ***p < 0.001 HEV interacts with G6PD Eight fragments of HEV (Met, Y, PCP, HPX, Hel, RdRp, ORF2, and ORF3) were separately cotransfected with G6PD into HepG2 cells to explore the interaction between HEV and G6PD. Remarkably, G6PD expression was significantly activated in cells cotransfected with HEV ORF3 (Figs. 3A and B), indicating that HEV ORF3 may interact with G6PD. Co-IP further verified the direct interaction between HEV ORF3 and G6PD (Fig. 3C). The colocation of HEV ORF3 and G6PD confirmed the interaction (Fig. 3D and E). The direct interaction between HEV ORF3 and G6PD may contribute to the activation of G6PD by HEV infection. Fig. 3HEV interacts with G6PD. (A) The expression of G6PD were detected by Western blotting in cells co-transfected with G6PD and pcDNA3.1vector, pcDNA3.1-Met-Flag, pcDNA3.1-Y-Flag, pcDNA3.1-PCP-Flag, pcDNA3.1-HPX-Flag, pcDNA3.1-Hel-Flag, pcDNA3.1-RdRp-Flag, pcDNA3.1-ORF2-Flag, pcDNA3.1-ORF3-Flag, or pcDNA3.1-flHEV-Flag, respectively. (B) The relative expression of G6PD in cells co-transfected with G6PD and each fragments of HEV. (C) Co-IP were performed in HEK293T cells co-transfected with G6PD and pcDNA3.1-ORF3-Flag. (D) Co-location were observed in HEK293T cells co-transfected with G6PD and pcDNA3.1-ORF3-Flag. Nucleus were stained with DAPI (blue). (E) Regions of interest (ROI) were drawn around the positive cells structures on images taken from HEK293T cells transfected with G6PD and pcDNA3.1-ORF3-Flag, and labeled with antibodies against the G6PD and Flag. ROI were used to determine Pearson’s correlation coeffificients (PCC) between G6PD and HEV ORF3 using JACoP plugin from ImageJ software. PCC means (± standard deviation) were calculated from 30 different ROI. *p < 0.05; **p < 0.01; ***p < 0.001 HEV interacts with G6PD. (A) The expression of G6PD were detected by Western blotting in cells co-transfected with G6PD and pcDNA3.1vector, pcDNA3.1-Met-Flag, pcDNA3.1-Y-Flag, pcDNA3.1-PCP-Flag, pcDNA3.1-HPX-Flag, pcDNA3.1-Hel-Flag, pcDNA3.1-RdRp-Flag, pcDNA3.1-ORF2-Flag, pcDNA3.1-ORF3-Flag, or pcDNA3.1-flHEV-Flag, respectively. (B) The relative expression of G6PD in cells co-transfected with G6PD and each fragments of HEV. (C) Co-IP were performed in HEK293T cells co-transfected with G6PD and pcDNA3.1-ORF3-Flag. (D) Co-location were observed in HEK293T cells co-transfected with G6PD and pcDNA3.1-ORF3-Flag. Nucleus were stained with DAPI (blue). (E) Regions of interest (ROI) were drawn around the positive cells structures on images taken from HEK293T cells transfected with G6PD and pcDNA3.1-ORF3-Flag, and labeled with antibodies against the G6PD and Flag. ROI were used to determine Pearson’s correlation coeffificients (PCC) between G6PD and HEV ORF3 using JACoP plugin from ImageJ software. PCC means (± standard deviation) were calculated from 30 different ROI. *p < 0.05; **p < 0.01; ***p < 0.001 HEV infection in cells with G6PD knockdown aggravates oxidative stress The expression levels of G6PD and NRF2 were further evaluated in hepatic hepatoma HepG2 cells to explore further the relationship between HEV infection and G6PD activation. HEV robustly replicated in HepG2 cells (Fig. 4A), with its copy numbers increasing from 12 hpi to 96 hpi (Fig. 4B). Consistently, the mRNA expression levels of G6PD (Fig. 4C) and NRF2 (Fig. 4E) significantly increased in a HEV replication–dependent manner. Similarly, a positive correlation was found between HEV copy number and G6PD in an NRF2 activation–dependent manner (Fig. 4D and F). G6PD plays a vital role in regulating oxidative stress, and pathogen infection also disrupts the balance of oxidative stress. However, the redox reaction in HEV-infected patients with G6PDd is unclear. Therefore, the state of oxidative stress was evaluated in HEV-infected HepG2 cells. Notably, G6PD and NRF2 expression levels were significantly elevated by HEV infection from 12 hpi to 96 hpi (Fig. 4G and H). This effect is comparable to the state of oxidative stress caused by H2O2 (Fig. 4I, J and K). The mRNA expression levels of G6PD and NRF2 in cells infected with HEV were consistent with those in cells treated with H2O2 (Figs. 4C and E). G6PD knockdown disrupted redox balance, increased oxidative stress, and significantly increased MDA levels. However, in cells with G6PD knockdown, HEV infection increased the accumulation of MDA (Fig.4L) and decreased the levels of the antioxidants SOD (Fig. 4M) and GSH (Fig. 4N), suggesting that HEV infection aggravated oxidative stress in patients with G6PDd. Increased oxidative stress may contribute to increased susceptibility to HEV infection. HEV copy numbers in cells treated with H2O2 were significantly higher than those in HEV-infected HepG2 cells (Fig. 4P). Meanwhile, HEV-infected HepG2 cells with G6PD knockdown consumed more NADPH than cells with normal G6PD levels (Fig. 4O). This effect may be responsible for the lower viral titer in cells with G6PD knockdown than in cells with normal G6PD levels or treated with H2O2 (Fig. 4P). It also explained why the HEV copy number in patients with G6PDd was significantly lower than that in patients with normal G6PD levels (Fig. 4Q). Fig. 4HEV infection aggravates oxidative stress in G6PDd. (A) HEV robustly replicates in HepG2 cells observed by confocal microscopy at 72 hpi. Nucleus stained with DAPI (blue). HEV ORF2 (red) incubated with HEV-specific antibody. (B) HEV copy number quantified by qRT-PCR in HepG2 cells infected with HEV at 12, 24, 48, 72 and 96 hpi. mRNA exression of G6PD (C) and NRF2 (E) in HepG2 cells infected with HEV at 12, 24, 48, 72 and 96 hpi by qRT-PCR. (D) Correlation analysis between HEV copy number and G6PD mRNA expression. (F) Correlation analysis between HEV copy number and NRF2 expression. (G) The expression of G6PD and NRF2 in cells infected with HEV at 12, 24, 48, 72 and 96 hpi by Western blotting. The relative protein expression of G6PD and NRF2 (H) in cells infected with HEV at 12, 24, 48, 72 and 96 hpi. (I) mRNA exression of G6PD and NRF2 in HepG2 cells treated with different concentration of H2O2. (J) The expression of G6PD and NRF2 in cells treated with different concentration of H2O2 by Western blotting. (K) The relative protein expression of G6PD and NRF2 in cells treated with different concentration of H2O2. The relative changes of MDA (L), SOD (M) and GSH (N) in cells treated with H2O2 or knockdown of G6PD then infected with or without HEV. (O) The relative change of NADPH/NADP + were analyzed in cells knockdown of G6PD then infected with or without HEV. (P) HEV copy number in HEV-infected cells treated with H2O2 or knockdown of G6PD. (Q) HEV copy number in HEV-infected patients deficient G6PD or not. *p < 0.05; **p < 0.01; ***p < 0.001 HEV infection aggravates oxidative stress in G6PDd. (A) HEV robustly replicates in HepG2 cells observed by confocal microscopy at 72 hpi. Nucleus stained with DAPI (blue). HEV ORF2 (red) incubated with HEV-specific antibody. (B) HEV copy number quantified by qRT-PCR in HepG2 cells infected with HEV at 12, 24, 48, 72 and 96 hpi. mRNA exression of G6PD (C) and NRF2 (E) in HepG2 cells infected with HEV at 12, 24, 48, 72 and 96 hpi by qRT-PCR. (D) Correlation analysis between HEV copy number and G6PD mRNA expression. (F) Correlation analysis between HEV copy number and NRF2 expression. (G) The expression of G6PD and NRF2 in cells infected with HEV at 12, 24, 48, 72 and 96 hpi by Western blotting. The relative protein expression of G6PD and NRF2 (H) in cells infected with HEV at 12, 24, 48, 72 and 96 hpi. (I) mRNA exression of G6PD and NRF2 in HepG2 cells treated with different concentration of H2O2. (J) The expression of G6PD and NRF2 in cells treated with different concentration of H2O2 by Western blotting. (K) The relative protein expression of G6PD and NRF2 in cells treated with different concentration of H2O2. The relative changes of MDA (L), SOD (M) and GSH (N) in cells treated with H2O2 or knockdown of G6PD then infected with or without HEV. (O) The relative change of NADPH/NADP + were analyzed in cells knockdown of G6PD then infected with or without HEV. (P) HEV copy number in HEV-infected cells treated with H2O2 or knockdown of G6PD. (Q) HEV copy number in HEV-infected patients deficient G6PD or not. *p < 0.05; **p < 0.01; ***p < 0.001 HEV infection induces G6PD in an NRF2 activation–dependent manner in the liver The mRNA expression levels of G6PD and NRF2 in HEV-infected BALB/c mice at 7, 14, 21, and 28 dpi were detected by using qRT-PCR to explore the mechanism underlying HEV susceptibility in patients with G6PDd. Consistent with our previous study [31], HEV infection in BALB/c mice was typically asymptomatic but resulted in viremia. Notably, high viral copy numbers were detected in the blood and livers of infected mice from 7 to 28 days post-inoculation (dpi) (Figs. 5A and B), clearly demonstrating active viral replication. Meanwhile, the mRNA levels of G6PD and NRF2 also correspondingly significantly increased (Figs. 5C and D). A positive correlation was found between HEV copy number and G6PD mRNA expression (r = 0.7037, p < 0.0001, Fig.5E). Similarly, NRF2 mRNA expression was positively regulated by HEV infection (r = 0.4192, p = 0.0091, Fig. 5F). Furthermore, the expression levels of G6PD and NRF2 in the livers of BALB/c mice infected with HEV were detected through Western blotting analysis at 7, 14, 21, and 28 dpi. The expression levels of G6PD and NRF2 significantly increased in the livers of HEV-infected BALB/c mice from 7 dpi to 28 dpi but decreased at 28 dpi when viral replication declined (Fig. 5G and H). In addition, HEV ORF2 antigens were detected in the livers of HEV-infected BALB/c mice from 7 dpi to 28 dpi by IHC (Fig. 5I and J). Consistently, HEV infection induced G6PD expression in a manner dependent on NRF2 activation; G6PD expression increased from 7 dpi to 21 dpi, then decreased at 28 dpi (Fig. 5I, K and M). Positive correlation was found between HEV ORF2 antigen–positive cells and G6PD-positive cells (r = 0.8826, p < 0.0001, Fig. 5L), as well as HEV ORF2 antigen–positive cells and NRF2-positive cells (r = 0.7499, p < 0.0001, Fig. 5N). These results indicated that HEV infection induced G6PD expression in an NRF2 activation–dependent manner in the liver. Fig. 5HEV infection induced G6PD expression in a NRF2-activation-dependent way in the liver. (A) (B) HEV copy number in the blood and liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi. mRNA expressions of G6PD (C) and NRF2 (D) in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi quantified by qRT-PCR. (E) Correlation analysis between HEV copy number and G6PD mRNA expressions. (F) Correlation analysis between HEV copy number and NRF2 mRNA expressions. (G) The expression of G6PD and NRF2 in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi were detected by Western blotting. GAPDH served as the loading control. (H) The relative expression of G6PD and NRF2 in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi analyzed by Image J software. (I) HEV ORF2 antigens detected in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi by IHC. The percentage of HEV ORF2 postive cells (J), G6PD postive cells (K), NRF2 postive cells (M) in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi. (L) Correlation analysis between the percentage of HEV ORF2 postive cells and G6PD positive cells. (N) Correlation analysis between the percentage of HEV ORF2 positive cells and NRF2 postive cells. *p < 0.05; **p < 0.01; ***p < 0.001 HEV infection induced G6PD expression in a NRF2-activation-dependent way in the liver. (A) (B) HEV copy number in the blood and liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi. mRNA expressions of G6PD (C) and NRF2 (D) in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi quantified by qRT-PCR. (E) Correlation analysis between HEV copy number and G6PD mRNA expressions. (F) Correlation analysis between HEV copy number and NRF2 mRNA expressions. (G) The expression of G6PD and NRF2 in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi were detected by Western blotting. GAPDH served as the loading control. (H) The relative expression of G6PD and NRF2 in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi analyzed by Image J software. (I) HEV ORF2 antigens detected in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi by IHC. The percentage of HEV ORF2 postive cells (J), G6PD postive cells (K), NRF2 postive cells (M) in the liver of BALB/c mice infected with or without HEV at 7, 14, 21 and 28 dpi. (L) Correlation analysis between the percentage of HEV ORF2 postive cells and G6PD positive cells. (N) Correlation analysis between the percentage of HEV ORF2 positive cells and NRF2 postive cells. *p < 0.05; **p < 0.01; ***p < 0.001 HEV infection in cells with G6PD knockdown exacerbates apoptosis G6PDd increased oxidative stress and consequently induced apoptosis. It significantly exacerbated apoptosis in cells with G6PD knockdown relative to that in HEV-infected cells with normal G6PD levels (Figs. 6A and B), demonstrating that HEV infection in patients with G6PDd aggravated apoptosis and caused severe diseases. Meanwhile, G6PD knockdown decreased the expression of Bcl-2, an apoptosis regulator. In cells with G6PD knockdown, HEV infection exhausted the expression of the apoptosis regulator Bcl-2 (Figs. 6C and D), further illustrating that G6PDd aggravated apoptosis. However, no significant increase in autophagy was observed in HEV-infected cells with or without G6PD knockdown (Fig. 6E and F). These results indicated that increased apoptosis may be responsible for disease aggravation in patients with G6PDd and HEV infection. G6PDd increased the susceptibility of patients to HEV infection. However, whether HEV infection aggravates liver damage is unknown. Therefore, the levels of liver enzymes, including ALT, AST, ALP, GGT, TG, TBil, DBil, and IBil, were determined in patients with G6PDd. Notably, patients infected with HEV exhibited remarkable elevations in ALT and ALP and slight increases in AST, GGT, TBil, DBil, and IBil (Fig. 6G to N, Table 1), indicating that in patients with G6PDd, HEV infection aggravated liver damage. Importantly, HEV infection in patients with G6PDd increased the incidence of hyperbileacidemia and anemia (Fig. 6O to R, Table 1). The results demonstrated that in patients with G6PDd, HEV infection aggravated oxidative stress and apoptosis, which may exacerbate disease severity. Fig. 6HEV infection aggravates apoptosis in G6PDd patients. (A) Apoptosis cells were observed by IFA in cells transfected with siG6PD then infected with or without HEV. (B) Percentage of apoptosis cells in cells transfected with siG6PD then infected with or without HEV. (C) Bcl-2 expression were detected by Western blotting in cells transfected with siG6PD then infected with or without HEV. (D) Relative expression of Bcl-2 in cells transfected with siG6PD then infected with or without HEV, and analyzed by Image J software. (E) Autophagy was observed in cells transfected with siG6PD then infected with or without HEV by IFA. (F) Percentage of autophagy cells in cells transfected with siG6PD then infected with or without HEV. The liver enzyme levels of ALT (G), AST (H), ALP (I), GGT (J), TG (K), TBil (L), DBil (M), IBil (N) were detected in the serum of G6PDd patients infected with or without HEV. Incidence of anacmia (O), favism (P), hyperbilirubinemia (Q), pneumonia (R) in G6PDd patients infected with or without HEV. *p < 0.05; **p < 0.01; ***p < 0.001 HEV infection aggravates apoptosis in G6PDd patients. (A) Apoptosis cells were observed by IFA in cells transfected with siG6PD then infected with or without HEV. (B) Percentage of apoptosis cells in cells transfected with siG6PD then infected with or without HEV. (C) Bcl-2 expression were detected by Western blotting in cells transfected with siG6PD then infected with or without HEV. (D) Relative expression of Bcl-2 in cells transfected with siG6PD then infected with or without HEV, and analyzed by Image J software. (E) Autophagy was observed in cells transfected with siG6PD then infected with or without HEV by IFA. (F) Percentage of autophagy cells in cells transfected with siG6PD then infected with or without HEV. The liver enzyme levels of ALT (G), AST (H), ALP (I), GGT (J), TG (K), TBil (L), DBil (M), IBil (N) were detected in the serum of G6PDd patients infected with or without HEV. Incidence of anacmia (O), favism (P), hyperbilirubinemia (Q), pneumonia (R) in G6PDd patients infected with or without HEV. *p < 0.05; **p < 0.01; ***p < 0.001 Discussion G6PDd affects 400 million people globally, affecting antioxidant pathways in red blood cells and increasing susceptibility to hemolysis under oxidative stress [30, 32]. G6PDd is a common inherited enzyme deficiency disease in humans. G6PD is a crucial enzyme for PPP and produces NADPH, which plays a vital role in regulating oxidative stress [1]. In patients, oxidative stress increases susceptibility to viral infection and aggravates diseases. Patients with G6PDd and SARS-CoV-2 infection showed an increase in hemolysis and thrombosis [7, 33], prolonged COVID-19 symptoms, requirement for ventilation support, neurological effects, and high mortality [33, 34]. Severe hemolytic anemia, acute renal failure accompanied with hyperbilirubinemia [35], and hyperacute liver failure [36] were observed in patients with G6PDd and HEV infection. However, the incidence and severity of HEV infection in patients with G6PDd are largely unknown. In the present study, remarkably increased HEV incidence was found in patients with G6PDd (23.8% in patients with G6PDd vs. 0.65% in patients with normal G6PD levels). This finding indicated that G6PDd increased the incidence of HEV. HEV infection usually causes self-limiting diseases but may be chronic in immunosuppressed patients. It has been recognized to increase the risk of hepatocellular carcinoma [37–40]. HEV associated with severe hyperbilirubinemia has been reported in patients with acute liver failure and elevated liver enzymes with hyperbilirubinemia [41, 42]. G6PDd aggravates the severity of diseases in patients infected with HEV. In patients with G6PDd, the incidence of hemolytic anemia increased to 70%–87%, complicating acute viral hepatitis [43]. Five patients with G6PDd and HEV infection developed severe intravascular hemolysis, with four of them developing acute renal failure [44]. A patient with G6PDd who was infected with HEV died from intracerebral hemorrhage after receiving hemodialysis and blood transfusion [45]. In the present study, increased rates of hyperbilirubinemia (64% vs. 36.25% with normal G6PD) and pneumonia (28% vs. 12.5% with normal G6PD) were consistently observed in HEV-infected patients with G6PDd. Meanwhile, elevated ALT and ALP levels implied aggravated liver damage in HEV-infected patients with G6PDd. The results suggested that HEV infection aggravated disease severity in patients with G6PDd. Although the reduced activity of G6PD has been found to disrupt redox equilibrium, causing the body to tend to an oxidative state and thereby increasing susceptibility to viruses, the detailed regulation of G6PD activity in HEV infection is unclear. Notably, G6PD knockdown in HepG2 cells permitted increased viral entry into the host cell membrane, revealing that G6PDd facilitates viral entry. In addition, the downregulation of NRF2 or inhibition of GSH synthesis resulted in a remarkable increase in Zika virus (ZIKV) replication [46]. A significant decrease in GSH was observed in HepG2 cells with G6PD knockdown and patients with G6PDd. This effect may contribute to viral replication. However, G6PD knockdown impaired NADP(H)-dependent glycolysis, resulting in reduced viral protein synthesis. Therefore, although oxidative stress facilitated HEV replication, G6PDd with substantial NADPH reduction limited HEV replication in vivo and in vitro. G6PD is highly expressed in human livers with chronic HBV infection and HBV-associated liver cancer via an NRF2 activation–dependent pathway [30]. ZIKV infection also activated the antioxidant NRF2 [47]. Consistently, HEV infection induced G6PD expression in a manner dependent on NRF2 activation to stimulate the antioxidant response in HepG2 cells and BALB/c mouse livers. NRF2 plays a critical role in the regulation of various cellular processes, such as antioxidant defenses, redox equilibrium, and inflammatory and apoptotic processes. NRF2 regulates the cellular cycle of hepatocytes and modulates antioxidant responses and apoptotic processes during liver regeneration [47, 48]. The absence of NRF2 delays liver regeneration. In the livers of HEV-infected BALB/c mice, G6PD expression was dramatically induced by HEV infection in a manner dependent on NRF2 activation but sharply decreased when viruses were cleared at 28 dpi. HEV infection decreased MDA and increased SOD and GSH levels in patients with G6PDd and G6PD knockdown, revealing that HEV infection induced oxidative stress and stimulated antioxidant response to maintain redox homeostasis. However, the excessive oxidative stress in cells with G6PD knockdown induced aggravated apoptosis rather than autophagy, indicating that HEV infection aggravated disease severity in patients with G6PDd. Although no significant difference was found between HEV-infected male and female mice [31], it remains unclear whether G6PD deficiency affects viral replication. Therefore, future studies should include male mice to clarify its influence. Furthermore, this study was limited by the commercial unavailability of G6PD knockout mice (BALB/c strain), preventing a deeper exploration of how HEV infection exacerbates disease and its molecular basis In summary, patients with G6PDd are susceptible to HEV infection and acute liver injury with aggravated severity. HEV ORF3 interacts with G6PD to stimulate the expression of G6PD in a manner dependent on NRF2 activation. G6PD knockdown facilitates HEV entry and aggravates oxidative stress with substantially inhibited GSH to benefit viral replication and consequently induce severe apoptosis. Table 1Clinical and demographic characteristics of the participantsCharacteristicsTotaln = 412n (%)HEV RNA+n = 27n (%)% (95%CI)P (X2)HEV IgG+n = 91n (%)% (95%CI)P (X2)G6PDd105 (25.5)25 (92.6)0.021 (0.005–0.090)< 0.001***20 (22.0)1.279 (0.734–2.226)0.385Normal G6PD307 (74.5)2 (7.4)71 (78.0)Age< 1 year190 (46.1)16 (59.3)0.567(0.256–1.254)0.15641 (45.1)1.056(0.662–1.686)0.818> 1 year222 (53.9)11 (40.7)50 (54.9)Clinical diagnosisFavism26 (6.3)5 (18.5)3.939(1.357–11.439)0.007**4 (4.4)0.625(0.21–1.862)0.395Neonatal hyperbilirubinemia87 (21.1)16 (59.3) 5.914 (2.560-12.78)<0.001***7 (7.7)0.251(0.112–0.565)< 0.001***Anaemia19 (4.6)4 (14.8)4.29(1.317–13.969)0.009**2 (2.2)0.402(0.091–1.773)0.214Pneumonia20 (4.9)7 (25.9)10.015(3.6–27.86.6.86)< 0.001***3 (3.3)0.61(0.175–2.128)0.433Liver enzymesALT > 40 U/L25 (6.1)5 (18.5)4.148(1.422–12.096)0.005**3 (3.3)0.463(0.135–1.584)0.21AST > 40 U/L38 (9.2)3 (11.1)1.25(0.358–4.361)0.7262 (2.2)0.182(0.043–0.769)0.01**ALP > 141 U/L134 (32.5)17 (63.0)3.894(1.731–8.76)< 0.001***9 (9.9)0.172(0.083–0.355)< 0.001***Symbols denote statistical differences:*p < 0.05; **p < 0.01; ***p < 0.001 Table 1Clinical and demographic characteristics of the participantsCharacteristicsTotaln = 412n (%)HEV RNA+n = 27n (%)% (95%CI)P (X2)HEV IgG+n = 91n (%)% (95%CI)P (X2)G6PDd105 (25.5)25 (92.6)0.021 (0.005–0.090)< 0.001***20 (22.0)1.279 (0.734–2.226)0.385Normal G6PD307 (74.5)2 (7.4)71 (78.0)Age< 1 year190 (46.1)16 (59.3)0.567(0.256–1.254)0.15641 (45.1)1.056(0.662–1.686)0.818> 1 year222 (53.9)11 (40.7)50 (54.9)Clinical diagnosisFavism26 (6.3)5 (18.5)3.939(1.357–11.439)0.007**4 (4.4)0.625(0.21–1.862)0.395Neonatal hyperbilirubinemia87 (21.1)16 (59.3) 5.914 (2.560-12.78)<0.001***7 (7.7)0.251(0.112–0.565)< 0.001***Anaemia19 (4.6)4 (14.8)4.29(1.317–13.969)0.009**2 (2.2)0.402(0.091–1.773)0.214Pneumonia20 (4.9)7 (25.9)10.015(3.6–27.86.6.86)< 0.001***3 (3.3)0.61(0.175–2.128)0.433Liver enzymesALT > 40 U/L25 (6.1)5 (18.5)4.148(1.422–12.096)0.005**3 (3.3)0.463(0.135–1.584)0.21AST > 40 U/L38 (9.2)3 (11.1)1.25(0.358–4.361)0.7262 (2.2)0.182(0.043–0.769)0.01**ALP > 141 U/L134 (32.5)17 (63.0)3.894(1.731–8.76)< 0.001***9 (9.9)0.172(0.083–0.355)< 0.001***Symbols denote statistical differences:*p < 0.05; **p < 0.01; ***p < 0.001 Clinical and demographic characteristics of the participants 0.567 (0.256–1.254) 1.056 (0.662–1.686) 3.939 (1.357–11.439) 0.625 (0.21–1.862) 0.251 (0.112–0.565) 4.29 (1.317–13.969) 0.402 (0.091–1.773) 10.015 (3.6–27.86.6.86) 0.61 (0.175–2.128) 4.148 (1.422–12.096) 0.463 (0.135–1.584) 1.25 (0.358–4.361) 0.182 (0.043–0.769) 3.894 (1.731–8.76) 0.172 (0.083–0.355) Symbols denote statistical differences:*p < 0.05; **p < 0.01; ***p < 0.001 Supplementary Information Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2