Introduction
Duck virus hepatitis (DVH) is an acute and highly infectious disease characterized by liver swelling and hemorrhage, primarily affecting ducklings under three weeks of age. This disease causes significant economic losses in the duck farming industry( Woolcock and Crighton, 1979 ). DVH is caused by duck hepatitis A virus (DHAV) ( Kim et al., 2006 ; Ding and Zhang, 2007 ), a member of the Enterovirus genus within the small RNA virus family( Tseng et al., 2007 ).
DHAV is classified into three serotypes—type I, type II, and type III—with no cross-reactivity observed between type II and the other two serotypes( Haider and Calnek, 1979 ). Recent studies indicate that DHAV1 and DHAV3 are the predominant serotypes currently circulating in China( Fu et al., 2008 ; Gao et al., 2012 ; Wen et al., 2014 ; Zou et al., 2016 ; Zhang et al., 2017 ). The nucleocapsid of duck hepatitis A virus (DHAV) is a symmetrical icosahedron, lacking an envelope, with a single-stranded RNA core and a diameter of 20–40 nm. The complete genome comprises a 5′ untranslated region (5′ UTR), a large open reading frame (ORF) encoding a polyprotein, a 3′ untranslated region (3′ UTR), and a poly(A) tail ( Chen et al., 2018 ). The ORF is divided into three regions: P1, P2, and P3. The P1 region encodes viral capsid proteins VP0, VP1, and VP3, while the P2 and P3 regions encode non-structural proteins, including proteases 2A, 3C, and 3CD, as well as the viral polymerase 3D pol ( Fu et al., 2008 ; Liang et al., 2015 ). The VP1 protein of DHAV is primarily exposed on the viral surface and plays a critical role in inducing the production of neutralizing antibodies( Ma et al., 2015 ). It serves as a binding protein for viral attachment to cell-specific receptors and is the principal determinant of viral antigenicity( Chen et al., 2024 ). As a highly variable region of viral antigenicity, VP1 exhibits significant differences between DHAV3 and DHAV1( Ma et al., 2015 ). These differences include not only point mutations and continuous mutations but also notable insertions or deletions( Ma et al., 2015 ). Antigenic variation in DHAV3, driven by amino acid changes in VP1, contributes to low or no cross-reactivity with DHAV1( Kim et al., 2007 ; Tseng et al., 2007 ).
The traditional virus neutralization test (VNT) is considered the “gold standard” for evaluating DHAV vaccine efficacy and is widely used. However, it is time-consuming, expensive, and labor-intensive, making it unsuitable for meeting the urgent need for large-scale clinical detection ( Truong and Cheng, 2022 ). In contrast, enzyme-linked immunosorbent assay (ELISA) is a highly sensitive and specific method, making it more suitable for vaccine assessments and epidemiological studies involving large sample volumes. Despite its advantages, there is currently no commercial ELISA kit available for evaluating DHAV3 immunity( Lai et al., 2020 ). Among various ELISA methods, monoclonal antibody (MAb)-based blocking ELISA (bELISA) stands out for its high specificity and its independence from the purity of the antigen ( Pattnaik and Venkataramanan, 1994 ). Traditionally, bELISA methods rely on purified virus particles as coating antigens. However, this approach is limited by low yield, labor-intensive procedures, time consumption, and the risk of contamination.
In this study, DHAV3 virus-like particles (VLPs) were expressed using the baculovirus insect cell expression system and utilized as the coating antigen. These VLPs closely mimic the structure of the virus while offering enhanced safety, as they lack viral nucleic acids, making them non-replicative and non-infectious. Additionally, DHAV3 VLPs and monoclonal antibodies (MAbs) against the DHAV3 VP1 protein were successfully generated and characterized. Following the optimization of reaction conditions, a blocking ELISA (bELISA) based on the monoclonal antibody 4B8 was developed for the detection of DHAV3 antibodies. The established bELISA demonstrated high sensitivity and specificity for DHAV3 antibody detection, providing a valuable new tool for DHAV-3 surveillance and immune monitoring.
Material and methods
Virus, cells and animals
The HBGT strain (Accession No. KX290465 ) and SD01 strain (Accession No. GQ485310 ) of DHAV3, along with the SH strain (Accession No. HQ232303 ) of DHAV1, were utilized in this study. The viruses were isolated and propagated in Specific Pathogen-Free (SPF) duck embryos and stored at -80°C until use. Sf9 cells (Invitrogen, USA) were cultured at 27°C in SF-900II serum-free medium (Gibco, USA) for baculovirus production. SPF ducks were hatched from SPF duck embryos obtained from the Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences (CAAS). The ducklings were housed in cages under a 12-hour light/dark cycle with free access to food and water throughout the study. The care and maintenance of all animals adhered to the guidelines of the Institutional Animal Care and Use Committee (IACUC) of the Shanghai Veterinary Research Institute, CAAS. All efforts were made to minimize variations in experimental conditions. BALB/c mice (4–5 weeks old) were purchased from Shanghai Jiesijie company for use in the study.
Construction of recombinant baculoviruses
The structural protein encoding gene P1 and the non-structural protein encoding gene 3C of the DHAV3 HBGT strain were cloned into the SalI/HindIII and NheI/KpnI sites of the pFastBac TM Dual vector (Invitrogen, CA), under the control of the polyhedron promoter and p10 promoter, respectively, generating the recombinant transfer plasmid pFDHAV3. The plasmid was analyzed by restriction enzyme digestion to confirm the correct orientation of the inserts. The recombinant bacmid transfer plasmids expressing VP0, VP1, and VP3 were transformed into DH10Bac TM E. coli competent cells (Beyotime, China) to produce recombinant bacmids (rBac-DHAV3). These rBac-DHAV3 bacmids were subsequently transfected into Sf9 cells using LipoInsect TM (Beyotime, China), following the Bac-to-Bac Expression System manual (Invitrogen, USA). The supernatant containing recombinant baculovirus was harvested five days post-transfection and used as the viral stock.
Preparation of mouse sera against DHAV3 capsid protein
To generate anti-VP0, VP1, and VP3 mouse polyclonal sera, the VP0, VP1 , and VP3 genes were individually cloned into the pCold TF vector for expression. The prokaryotic expression products were then used to immunize mice for the preparation of polyclonal antibodies.
VLP production and purification
Baculovirus stock titers were determined using the BacPak Baculovirus Rapid Titer Kit (Clontech, USA). DHAV3 VLPs were produced in suspension cultures of Sf9 cells infected with recombinant baculovirus at a multiplicity of infection (MOI) of 1. The culture was harvested 48 hours post-infection. The suspensions were first centrifuged at 500 g for 5 minutes to remove cell debris, followed by ultracentrifugation at 100,000g for 1 hour to collect the VLP pellet. The pellet was then purified using a 20-40-60 % discontinuous sucrose gradient, and the bands between 20-40 % sucrose, which correspond to the rDHAV3 VLPs, were collected.
Electron microscopy
Purified VLPs were applied to grids and incubated at room temperature for 5 minutes. The samples were then stained with 1 % sodium phosphotungstate and examined using a transmission electron microscope (TEM). Sf9 cells infected with either recombinant virus or baculovirus were collected, fixed, dehydrated, embedded, sectioned, stained, and observed under TEM.
Western blot analysis
Western blot analysis was performed to evaluate rDHAV3 VLPs. Briefly, purified rDHAV3 VLPs were separated by SDS-PAGE and transferred onto PVDF membranes (Bio-Rad). The membranes were blocked with 5 % BSA in Tris-buffered saline with Tween 20 (TBST) for 1 hour. Next, the membranes were incubated with anti-VP0, VP1, and VP3 mouse polyclonal antibodies in blocking buffer for 1 hour at room temperature. After washing the membranes three times with TBST, they were incubated with HRP-conjugated anti-mouse secondary antibodies (diluted 1:1000 in blocking buffer) for 1 hour at room temperature. Following three additional washes with TBST, signals were developed using an enhanced chemiluminescence detection system (Amersham Pharmacia Biotech) and visualized by exposing the membranes to X-ray (Fuji Film).
Indirect enzyme-linked immunosorbent assay (iELISA)
Coat the ELISA plate with DHAV3 VLPs and block with 5 % skimmed milk. Add cell fusion supernatant or duck serum and incubate at 37°C for 1 h. Wash 4 times with PBST. Next, add HRP - labeled mouse or duck secondary antibody and incubate at 37°C for 1 h. Wash 4 times with PBST again. Then add TMB substrate solution and finally terminate the reaction in due course.
Indirect immunofluorescence assay (IFA)
Sf9 cells were infected with recombinant baculovirus or baculovirus for 48 hours, then fixed with 4 % paraformaldehyde for 20 minutes at room temperature. The cells were permeabilized with 0.5 % Triton X-100 for 10 minutes at room temperature, followed by three washes with PBS. After blocking with 3 % bovine serum albumin at 37°C for 30 minutes, the cells were incubated with primary antibodies (mouse polyclonal antibodies against DHAV3 VP0, VP1, or VP3 proteins) for 1 hour. After three washes with PBS, FITC-labeled goat anti-mouse secondary antibodies were added for 1 hour in a moist container, protected from light, at 37°C. The cells were then observed using a fluorescence microscope (Nikon Ti-E, Japan) following a final wash.
Production and Purification of VP1 Monoclonal Antibody (MAb)
Female BALB/c mice (6-8 weeks old) were immunized intraperitoneally with purified pCold-VP1 protein. After three immunizations, spleen cells from the BALB/c mice were harvested and fused with mouse myeloma SP2/0 cells ( Köhler and Milstein, 1975 ). Supernatants from growing hybridoma colonies were screened using indirect ELISA (iELISA) and immunofluorescence assay (IFA). Hybridomas secreting specific antibodies were cloned twice by limiting dilution and injected into paraffin-treated BALB/c mice to produce ascitic fluid. Immunoglobulin G (IgG) from the monoclonal antibody (MAb) was purified using Protein G Sepharose (GE Healthcare, Chicago, IL, USA).
Establishment and optimization of MAb-based blocking-ELISA (bELISA)
To determine the optimal concentrations of the coating antigen and serum dilutions, checkerboard titrations were performed. DHAV3 VLPs were serially diluted (4, 2, 1, 0.5, 0.25, 0.125, and 0.0625 mg/mL) in carbonate/bicarbonate coating buffer (pH 9.6), and 100 μL of each dilution was added horizontally to ELISA plates, followed by incubation at 4°C overnight. The plates were washed four times with PBST, and then blocked with 5 % (wt/vol) skimmed milk (300 μL/well) in PBST at 37°C for 1 hour. DHAV3-positive or negative duck serum was serially diluted (1:2, 1:4, 1:8, 1:16, 1:32, and 1:64) in PBST, and 100 μL of each dilution was added longitudinally to the wells, followed by incubation at 37°C for 1 hour. After washing with PBST, DHAV3 VP1 MAb (dilution 1:5000) was added to each well, and incubation was performed at 37°C for 1 hour. After washing again, horseradish peroxidase (HRP)-conjugated anti-mouse IgG (dilution 1:5000) was added (100 μL/well) and incubated at 37°C for 1 hour. After four washes, the reaction was developed with TMB substrate solution (100 μL/well) at room temperature in the dark for 10 minutes. The chromogenic reaction was stopped by adding 50 μL of 2M H 2 SO 4 to each well, and optical density (OD) was measured at 450 nm in a microplate reader. Control wells without primary serum or without MAb were also included. The optimal concentrations of the coating antigen and serum dilutions were determined based on the highest percent inhibition (PI) values. The PI values were calculated using the following formula( Liu et al., 2024 ):PI ( %) =[(mean OD of negative control-OD of sample)/mean OD of negative control] ✖100 %.
After establishing the aforementioned conditions, the optimal concentration of DHAV3 VLPs was coated at 37°C for 2 hours, followed by overnight incubation at 4°C. The blocking conditions were optimized as described below: Plates coated with DHAV3 VLPs were blocked with either 5 % (wt/vol) BSA, 5 % (wt/vol) skimmed milk, or 10 % (wt/vol) bovine serum in PBST at 37°C for 1 hour. To identify the optimal blocking time, the DHAV3 VLPs-coated plates were blocked in the optimal blocking solution at 37°C for 0.5, 1, 2, and 3 hours.Subsequently, the optimal dilution of serum was added and incubated at 37°C for 0.5, 1, 1.5, or 2 hours. The VP1-MAb was diluted to 1:1000, 1:2000, 1:3000, 1:4000, and 1:5000 in PBST, and incubated at 37°C for 1 hour. Thereafter, the incubation time of VP1-MAb was optimized at 37°C for 0.5, 1, 1.5, or 2 hours. The reactions were terminated and optimized by evaluating times of 5, 10, 15, and 20 minutes, respectively. All experiments were evaluated through the measurement of percent inhibition (PI) values.
In accordance with the above protocol, duck sera were collected from standardized slaughterhouse in Fengxian District, Shanghai. Subsequently, the collected sera were detected and screened using serum neutralization test (SNT) and indirect enzyme-linked immunosorbent assay (iELISA). Finally,100 serum samples negative for DHAV3-specific antibody were identified and to determine the optimal cut off value.
Experimental serum samples in ducklings
Eight types of serum samples were used in this study: DHAV3-negative serum, collected from disease-free ducks, served as the standard negative control; DHAV3-positive serum was collected from DHAV3-infected ducks. Positive control antisera samples, including Duck Avian Reovirus (DARV), Duck Tembusu Virus (DTMUV), Duck Enteritis Virus (DEV), Duck Adenovirus (DADV), Newcastle Disease Virus (NDV), Avian Influenza Virus (AIV), and Duck Hepatitis Virus Type I (DHAV1), were employed to assess the specificity of bELISA.
Specificity, sensitivity, and reproducibility
The positive reference sera against other related pathogens (ARV, DTMUV, DPV, RA, APEC, H9N2, NDV, MDPV, FAdV, and DHAV1) were tested using the developed bELISA to assess its specificity. To evaluate the analytical sensitivity of the bELISA for DHAV3, both positive and negative standard serum samples were diluted at 1:100, 1:200, 1:300, 1:400, 1:500, 1:600, 1:700, 1:800, 1:900, and 1:1000, respectively. These dilutions were tested using the bELISA, and the percent inhibition (PI) value was calculated. The highest dilution at which the PI value exceeded the cutoff was defined as the detection limit of the method. To assess the repeatability of the bELISA for DHAV3, three positive serum samples (F7, D9, B48) and two negative serum samples (G35, I17) were tested. The assay precision was evaluated by performing three replicates on one plate (within-plate precision) and two plates in a single run (between-plate precision). Means, standard deviations, and percent coefficient of variation ( % CV) values were calculated using Control Chart Pro Plus software (ChemSW, Inc., Fairfield Bay, AR, USA).
Sample detection
A total of 100 clinical serum samples were collected from three different duck farms in Suqian, Jiangsu Province. These samples were simultaneously tested using both the bELISA and the duck embryo neutralization test (NT). The consistency between the bELISA and NT was evaluated by calculating the total coincidence rate, which includes both positive and negative results.
Serum neutralization test
The SNT was performed according to NY/T 554-2023(“diagnostic techniques for duck A viral hepatitis 1 and 3”). The specific operation was as follows: equal volumes of the HBGT were taken respectively and then fully mixed with the serum to be tested, the DHAV3 positive serum, and the DHAV3 negative serum and held for 30min at 37°C. Then the mixture was inoculated into 10- to 12-day-old non-immunized duck embryos at a dose of 0.2 mL per duck embryo by allantoic cavity inoculation. Each sample was inoculated into 4 to 5 duck embryos. The judgment criteria was: if embryo death and embryo lesions occurred, the antibody in the serum to be tested was judged to be negative; if the embryos did not die and there were no lesions in the embryo bodies, the antibody in the serum to be tested was judged to be positive; if the embryos died and there were lesions in the embryo bodies, the antibody in the serum to be tested was also judged to be negative.
Statistical analysis
All data are presented as the mean ± standard deviation (SD). Statistical analysis was performed using Student's t-test, with a significance level set at p < 0.05. Data and graphs were generated using GraphPad Prism 7.0.
Results
DHAV3 VLPs production and characterization
The construction of DHAV3 virus-like particles (VLPs) is illustrated in Fig. 1 A. The precursor protein P1 and non-structural protein 3CD of DHAV3 were cloned into the insect baculovirus dual expression vector pFast-dual, generating the recombinant plasmid pFast-P1-3CD. This plasmid was transposed into competent Escherichia coli DH10Bac cells to produce the recombinant bacmid rBac-P1-3CD. Transfection of insect cells (Sf9) with the recombinant bacmid led to the expression of DHAV3 VP0, VP1, and VP3 proteins, which self-assembled into DHAV3 VLPs. The VLPs were harvested, and their titer reached 10 11 PFU/ml, as determined by plaque assay. Transmission electron microscopy (TEM) analysis of the VLPs revealed capsid particles that were 20–30 nm in size, similar to the size of DHAV3 virions, indicating that the particles were indeed VLPs ( Fig. 1 B). In addition, numerous rDHAV3 VLPs were observed in the cytoplasm of recombinant virus-infected Sf9 cells. The infected cells displayed signs of nuclear damage, including a cracked nuclear membrane, decomposed nucleolus, and fragmented chromatin ( Fig. 1 C-a). In contrast, baculovirus-infected Sf9 cells showed an intact nuclear membrane with rod-shaped packaged virus particles that had dense electron density, a single or double-layer envelope, or round inclusion bodies ( Fig. 1 C-b). Western blot analysis confirmed the expression of the recombinant baculovirus in Sf9 cells, with bands corresponding to VP0, VP1, and VP3 observed in the rDHAV3 VLPs ( Fig. 1 D). Immunofluorescence assays further validated the presence of these three structural proteins in the VLPs ( Fig. 1 E). The concentration of DHAV3 VLPs was determined to be 975 μg/mL using the BCA method ( Fig. 1 F). Fig. 1 DHAV3 VLPs Production and Characterization. A: Schematic diagram illustrating the construction of DHAV3 VLPs. B-C: Electron microscopy analysis of DHAV3 VLPs (scale bar = 400 nm). (B) Purified DHAV3 VLPs produced in Sf9 cells, stained with 1 % sodium phosphotungstate, showing virus-like particles. (C) Cellular sections of DHAV3 VLPs-infected Sf9 cells. a: DHAV3 VLPs showing cytoplasmic localization. b: Baculovirus-infected Sf9 cells showing rod-shaped virus particles and intact nuclear membrane. (D) Western blot analysis using mouse polysera against DHAV3 structural proteins (VP0, VP1, VP3). a: VP0; b: VP1; c: VP3. M: marker; 1: DHAV3 VLPs; 2:Bacmid. (E) Immunofluorescence assay (IFA) demonstrating the expression of DHAV3 structural proteins (VP0, VP1, VP3) in VLPs using mouse polysera. (F) Determination of DHAV3 VLP concentration (975 μg/mL) using the BCA method. Fig 1
Characterization of MAbs against DHAV3 VP1 protein
The recombinant plasmid pCold-VP1 with a His tag was successfully transformed into E. coli host cells, and the rVP1 protein was expressed. SDS-PAGE analysis revealed that part of the rVP1 protein (approximately 70 kDa) was expressed in soluble form ( Fig. 2 A). After purification using Ni-NTA, the rVP1 protein was confirmed by Western blot using an anti-His tag antibody ( Fig. 2 B) and duck anti-DHAV3 positive serum ( Fig. 2 C). To generate DHAV3-specific monoclonal antibodies (MAbs), mice were immunized with the purified rVP1 antigen. Using the hybridoma cell fusion technique and three rounds of subcloning and screening with indirect ELISA (iELISA), two specific MAbs against the DHAV3 VP1 protein, designated 4B8 and 2B5, were obtained. Immunofluorescence detection of DHAV3 bacmid-infected Sf9 cells showed strong staining in the cytoplasm of the infected cells ( Fig. 2 D). Both 4B8 and 2B5 MAbs were identified as IgG1 heavy chain with kappa light chain using a Mouse Ig Isotyping Kit (Proteintech Group, Inc., Rosemont, IL, USA). The specificity of the MAbs was further evaluated with other duck pathogens. The MAbs were tested against DHAV3 (SD01 and HBGT strains), ARV, DTMUV, DPV, RA, APEC, H9N2, NDV, FAdV, and DHAV1 (SH strain). As shown in Fig. 2 E, both MAbs 4B8 and 2B5 reacted only with SD01 and HBGT strains of DHAV3, with no reaction to the other duck pathogens. The titers of the MAbs were determined by iELISA, and it was found that both 2B5 and 4B8 MAbs reached titers of 1:5000. The 4B8 MAb was selected for the development of the bELISA. Fig. 2 Analysis of DHAV3 VP1-MAbs. A-C: Analysis of rVP1 protein. A: SDS-PAGE analysis of His-tagged rVP1 protein expression. M:Marker; 1: pCold-VP1; 2: pCold-TF. B: Western blot analysis of rVP1 protein with anti-His tag antibody. M:Marker; 1:-3 pCold-VP1; 4-5: pCold-TF. C: Western blot analysis of rVP1 protein with DHAV3-positive serum. M:Marker; 1: pCold-VP1; 2: pCold-TF. D: IFA analysis of VP1 MAbs. a: VP1 MAb-2B5 with rDHAV3-Bacmid infected Sf9 cells; b: VP1 MAb-2B5 with baculovirus-Bacmid infected Sf9 cells; c: VP1 MAb-4B8 with rDHAV3-Bacmid infected Sf9 cells; d: VP1 MAb-4B8 with baculovirus-Bacmid infected Sf9 cells. E: Specificity of VP1-MAbs. a: VP1 MAb-2B5; b: VP1 MAb-4B8. M: Marker; 1: DHAV3 SD01 strain; 2: DHAV3 HBGT strain; 3: ARV; 4: DTMUV; 5: DPV; 6: RA; 7: APEC; 8: H9N2; 9: NDV; 10: FAdV; 11: DHAV1 SH strain. F: The titer of MAbs. Fig 2
Optimization of the bELISA protocol
After a series of optimizations, the purified DHAV3 VLPs were used as the coating antigen in the ELISA, while the purified Mab 4B8 served as the blocking antibody. A checkerboard titration was conducted to identify the optimal conditions. The results showed that the best pairing for the highest PI value was achieved with a coating antigen concentration of 1 μg/mL and a serum sample dilution of 1:32 ( Fig. 3 A). Further optimization revealed the following optimal conditions. The best coating time and temperature were 37°C for 2 hours, followed by overnight incubation at 4°C ( Fig. 3 B). The ideal blocking solution was 5 % skimmed milk ( Fig. 3 C), and the optimal blocking time was 2 hours at 37°C ( Fig. 3 D). For serum sample incubation, 1 hour at 37°C was found to be optimal ( Fig. 3 E). The optimal dilution ratio for Mab was 1:2000, with a reaction time of 1 hour at 37°C ( Fig. 3 F and 3 G). Finally, the best development time was determined to be 15 minutes at room temperature in the dark, which resulted in the highest PI value ( Fig. 3 H). Fig. 3 Optimal conditions for the bELISA. A: Different concentrations of coating antigen DHAV3 VLPs (4, 2, 1, 0.8, 0.5, 0.2, 0.1μg/mL) and dilutions of serum samples (1:2, 1:4, 1:8, 1:16, 1:32, 1:64) were tested by checkerboard. The coating time and temperature (B) were evaluated, and the optimal time and temperature was 37°C for 2h and then 4°C overnight. The blocking buffer (C) and the optimal blocking time (D) were evaluated. 5 % BSA, 5 % skimmed milk, and 10 % bovin serum were used as the blocking buffer, and the blocking conditions were 37°C for 2h. (E) Influence of detection serum reaction time. Serum samples was diluted 1:32 in PBST, and incubated 37°C for 30, 60, 90 and 120 min, respectively. Determination of the optimal dilution ratio (F) and reaction time (G) of MAb. The MAb was diluted (1:1000–1:5000), and incubation time was 37°C for 30, 60, 90 and 120 min, respectively. (H) Optimum the developing time. TMB was incubated with 37°C for 5, 10, 15, and 20 min, respectively, and then stopped in 2 M H 2 SO 4 . The PI value was calculated, and the highest PI value was the optimal work conditions of the bELISA. Fig 3
Determination of the cutoff value of the bELISA
To determine the cutoff value for the bELISA, 100 known DHAV3-negative sera were tested, and the OD450 nm values were recorded for statistical analysis according to the established bELISA method ( Fig. 4 ). The average blocking rate (x̄) and standard deviation (SD) were calculated as 11.09 % and 9.04 %, respectively. Based on statistical principles, x̄+2SD and x̄+3SD were used as the critical values for defining negative and positive sera. A PI value greater than 38.22 % (x̄+3SD) was considered positive; a PI value less than 29.18 % (x̄+2SD) was considered negative; a PI value between 29.18 % and 38.22 % was categorized as suspected and required re-identification. If the PI remained between 29.18 % and 38.22 %, it was ultimately classified as negative. Fig. 4 Determination of the negative cut off. A total of 100 negative serum samples were subjected to the optimized bELISA. The percentage inhibition (PI) values for these samples were calculated, and based on statistical analysis, the cutoff value was determined to be 38.22 %. Samples with a PI value greater than 38.22 % were considered positive, while those with a PI value less than 29.18 % were considered negative. Samples with PI values between 29.18 % and 38.22 % were considered suspected and required further re-identification. Fig 4
Diagnostic sensitivity and specificity of the bELISA
To evaluate the diagnostic specificity of the bELISA, DHAV3-positive and negative sera, along with samples positive for other duck pathogens (such as ARV, DTMUV, DPV, RA, APEC, H9N2, NDV, MDPV, FAdV, and DHAV1), were tested. The results revealed that the percentage inhibition (PI) value was 94.15 % for DHAV3-positive serum. Additionally, ten positive reference sera against other duck-related pathogens showed PI values ranging from -2 % to 18 %, all falling below the cutoff value, indicating that the developed bELISA has high specificity ( Fig. 5 A). To assess the sensitivity of the bELISA, three DHAV3-positive sera were diluted and tested. As shown in Fig. 5 B, even when the serum was diluted 500-fold, it remained positive with a PI value of 42.13 %. Fig. 5 Validation of the bELISA. (A) Specificity. The sera against ARV, DTMUV, DPV, RA, APEC, H9N2, NDV, MDPV, FAdV, DHAV1 were used to evaluate specificity of the bELISA. (B) Sensitivity. Serum samples were diluted 1:100 to 1:1000 and used the bELISA to evaluate the sensitivity. Fig 5
Repeatability of bELISA
Repeatability assesses an assay's ability to produce consistent results across multiple preparations and runs of the same sample. In this study, the repeatability of the VP1 MAb-based bELISA was evaluated by testing three DHAV3 positive serum samples (F7, D9, B48) and two DHAV3 negative serum samples (G35, I17) both within and between batches. The coefficient of variation (CV) was calculated to quantify repeatability. The results indicated that the within-batch CV ranged from 1.04 % to 4.56 % (with standard deviations (SD) of 0.82 % to 3.03 %), while the between-batch CV ranged from 1.06 % to 4.98 % (with SDs of 0.85 % to 1.89 %) ( Table 1 ). Both within-batch and between-batch CV values were below 5 %, demonstrating that the DHAV3 VP1 MAb-based bELISA is highly repeatable. Table 1 Repeatability of bELISA. Table 1 Sample number Within batch Between batches Mean value SD CV Mean value SD CV F7 92.03 % 1.73 % 1.88 % 92.28 % 0.98 % 1.06 % D9 89.24 % 3.03 % 3.39 % 91.23 % 1.89 % 2.07 % G35 35.69 % 0.82 % 2.31 % 33.53 % 1.41 % 4.19 % B48 91.95 % 0.96 % 1.04 % 91.35 % 1.39 % 1.52 % I17 19.46 % 0.89 % 4.56 % 17.08 % 0.85 % 4.98 %
Table 1
Repeatability of bELISA.
| Sample number | Within batch | Between batches | ||||
| Mean value | SD | CV | Mean value | SD | CV | |
| F7 | 92.03 % | 1.73 % | 1.88 % | 92.28 % | 0.98 % | 1.06 % |
| D9 | 89.24 % | 3.03 % | 3.39 % | 91.23 % | 1.89 % | 2.07 % |
| G35 | 35.69 % | 0.82 % | 2.31 % | 33.53 % | 1.41 % | 4.19 % |
| B48 | 91.95 % | 0.96 % | 1.04 % | 91.35 % | 1.39 % | 1.52 % |
| I17 | 19.46 % | 0.89 % | 4.56 % | 17.08 % | 0.85 % | 4.98 % |
Clinical serum samples detection
A total of 100 duck serum samples were collected from a duck farm in Jiangsu Province and tested using both the bELISA and neutralization test (NT). As shown in Table 2 , 63 of the 100 serum samples were determined to be positive by bELISA, while 59 samples were positive by NT. Out of the 100 serum samples, 59 were positive and 37 were negative according to both methods. The total agreement rate between the two detection methods was 96 % (96/100), and the kappa value was 0.916, indicating a high level of agreement between the bELISA and NT methods. At the same time, 100 duck serum samples were also tested using the indirect enzyme-linked immunosorbent assay (iELISA) ( Table 3 ). It was found that the total coincidence rate between bELISA and iELISA was 93 %, and the kappa value was 0.844. This indicates that the two detection methods have a good degree of consistency. Table 2 Comparison of bELISA and SNT. Table 2 bELISA SNT Total Concidence rate Postive Negative Postive 59 (a) 4 (b) 63 (a+b) 96 % Negative 0 (c) 37 (d) 37 (c+d) Total 59 (a+c) 41 (b+d) 100 (n) Kappa 0.916 The coincidence rate was calculated as [(a+d)/n]✖100. The K value was calculated as follows: P A =(a+d)/n; Pe=[(a+b)(a+c)+(c+d)(b+d)]/n 2 ; K=(P A -Pe)/(1-Pe). K≥0.75 means good consistency; 0.4≤K<0.75 means moderate consistency; K<0.4 means poor consistency. Table 3 Comparison of bELISA and iELISA. Table 3 iELISA bELISA Total Concidence rate Postive Negative Postive 63 (a) 7 (b) 70 (a+b) 93 % Negative 0 (c) 30 (d) 30 (c+d) Total 63 (a+c) 37 (b+d) 100 (n) Kappa 0.844 The coincidence rate was calculated as [(a+d)/n]✖100. The K value was calculated as follows: P A =(a+d)/n; Pe=[(a+b)(a+c)+(c+d)(b+d)]/n 2 ; K=(P A -Pe)/(1-Pe). K≥0.75 means good consistency; 0.4≤K<0.75 means moderate consistency; K<0.4 means poor consistency.
Table 2
Comparison of bELISA and SNT.
The coincidence rate was calculated as [(a+d)/n]✖100. The K value was calculated as follows: P A =(a+d)/n; Pe=[(a+b)(a+c)+(c+d)(b+d)]/n 2 ; K=(P A -Pe)/(1-Pe). K≥0.75 means good consistency; 0.4≤K<0.75 means moderate consistency; K<0.4 means poor consistency.
| bELISA | SNT | Total | Concidence rate | |
| Postive | Negative | |||
| Postive | 59 (a) | 4 (b) | 63 (a+b) | 96 % |
| Negative | 0 (c) | 37 (d) | 37 (c+d) | |
| Total | 59 (a+c) | 41 (b+d) | 100 (n) | |
| Kappa | 0.916 |
Table 3
Comparison of bELISA and iELISA.
The coincidence rate was calculated as [(a+d)/n]✖100. The K value was calculated as follows: P A =(a+d)/n; Pe=[(a+b)(a+c)+(c+d)(b+d)]/n 2 ; K=(P A -Pe)/(1-Pe). K≥0.75 means good consistency; 0.4≤K<0.75 means moderate consistency; K<0.4 means poor consistency.
| iELISA | bELISA | Total | Concidence rate | |
| Postive | Negative | |||
| Postive | 63 (a) | 7 (b) | 70 (a+b) | 93 % |
| Negative | 0 (c) | 30 (d) | 30 (c+d) | |
| Total | 63 (a+c) | 37 (b+d) | 100 (n) | |
| Kappa | 0.844 |
Discussion
Duck viral hepatitis is a highly acute and contagious disease affecting ducklings, characterized by rapid onset, swift transmission, and high mortality rates, leading to significant economic and production losses( Song et al., 2014 ; Yugo et al., 2016 ). Regular surveillance of viral infections and effective vaccination remain the most robust strategies for controlling and preventing DHAV infections( Roh and Kang, 2018 ; Wu et al., 2020 ; Zhang et al., 2023 ). Duck hepatitis A virus (DHAV), classified under the Enterovirus genus of the small RNA virus family, is divided into three serotypes. Among these, serotypes I and III are the predominant strains circulating in China( Niu et al., 2019 ). Several diagnostic methods for DHAV detection have been reported, including virus isolation, RT-PCR, virus neutralization tests (VNT), and ELISA( Ren et al., 2019 ; Li et al., 2022 ; Yu et al., 2024 ). Compared to VNT, ELISA is a more convenient, rapid, and cost-effective tool for serological surveillance. Various ELISA methods have been developed for detecting antibodies against DHAV1, such as indirect or monoclonal antibody sandwich/capture ELISA, targeting antigens like VP1 and VP3. Different forms of antigens, including whole viral particles and recombinant proteins, have been used in these assays( Zhang et al., 2015 ; Zhou et al., 2019 ; Lai et al., 2020 ). However, there is currently a lack of commercially available antibody detection products specifically targeting DHAV3, highlighting the need for further development in this area.
VLPs are nanoscale structures that closely resemble viruses but lack viral nucleic acids, rendering them non-infectious and incapable of autonomous replication( Yan et al., 2015 ; Ruzzi et al., 2023 ). These particles, also referred to as pseudoviruses, are typically 20–150 nm in diameter and are assembled from one or more structural proteins of a virus into specific spatial conformations. Morphologically, VLPs are identical or highly similar to natural viral particles, preserving the authentic conformation of viral antigens. This enables them to effectively stimulate both innate and adaptive immune responses( Roy, 1996 ). The structural similarity between VLPs and their corresponding viruses makes them valuable substitutes for infectious viruses in the development of novel diagnostic tools. These tools are often simpler, faster, and more cost-effective( Comas-Garcia et al., 2020 ). VLP preparation generally involves the exogenous expression of viral structural proteins in an expression system, followed by in vitro assembly. Among the available systems, the insect cell expression system is widely employed due to its ability to perform post-translational modifications, which are essential for proper VLP assembly and packaging( Casal, 2001 ; Liu et al., 2013 ; Gopal and Schneemann, 2018 ). In this study, the DHAV3 structural proteins VP0, VP1, and VP3 were expressed using the baculovirus-insect cell expression system. This approach successfully produced proteins with appropriate spatial configurations that self-assembled into DHAV3 VLPs ( Fig. 1 A). The purified VLPs were obtained through sucrose gradient density centrifugation and further characterized using electron microscopy ( Fig. 1 B-C), Western blot analysis ( Fig. 1 D), and indirect immunofluorescence assays ( Fig. 1 E). The results demonstrated the successful generation of DHAV3 VLPs with a diameter of approximately 25 nm, closely resembling the morphology of natural DHAV3 particles. Additionally, the analysis confirmed that the VLPs were co-assembled from the primary structural proteins VP0, VP1, and VP3 of DHAV3.
The VP1 protein, as the major surface protein of DHAV, plays a pivotal role in viral infection and serves as a primary target for neutralizing antibodies( Li et al., 2017 ). In this study, we successfully generated two MAbs against the DHAV3 VP1 protein. Characterization by IFA, Western blot, and ELISA demonstrated that both MAbs specifically recognized the immunizing VP1 protein. Among the two, 4B8 exhibited superior binding capability to DHAV3 VLPs in an indirect ELISA (iELISA) ( Fig. 2 F). Based on this observation, 4B8 was selected as the detection antibody for developing a blocking ELISA (bELISA). The protocol for bELISA was optimized ( Fig. 3 ), and analysis of 100 anti-DHAV3-negative duck serum samples using 2SD and 3SD statistical approaches determined an optimal diagnostic cut off value of 38.22 % ( Fig. 4 ).
Specificity tests revealed that the bELISA could clearly distinguish DHAV3 from other common duck pathogens ( Fig. 5 A). The assay's analytical sensitivity was evaluated using serial dilutions (1:100 to 1:1000) of DHAV3-positive serum, with a detection limit of up to 1:500 dilution ( Fig. 5 B). Furthermore, repeatability assays confirmed the reliability and reproducibility of the bELISA method ( Table 1 ). The developed bELISA was applied to detect DHAV3 antibodies in 100 field serum samples collected from Jiangsu province. Out of these, 37 samples tested DHAV3-negative, a result that aligned with the serum neutralization test (SNT) ( Table 2 ). This outcome highlights the potential of the bELISA as a robust, specific, and sensitive tool for DHAV3 serological surveillance and diagnosis.
Currently, there is no commercially available blocking ELISA (bELISA) kit for DHAV3 antibody detection on the market. In this study, we successfully established a blocking ELISA antibody detection method. However, due to the limited sample size and the fact that the samples were only collected from three farms in one province or city, the generalizability of the research results may be limited. In addition, the long-term stability test of bELISA has not been carried out in this study. Based on this, in the follow-up, we plan to expand the sample collection scope to cover farms in multiple provinces for detection, so as to enhance the representativeness and reliability of the research results. At the same time, we will focus on monitoring the stability of bELISA to provide a solid guarantee for the practical application of this detection method. To further optimize the detection process, we will attempt to label the VP1 antibody with HRP, thereby simplifying the experimental steps and improving the detection efficiency. We hope that through these efforts, this detection method can be developed into a monitoring diagnostic kit with practical application value, providing strong support for disease monitoring and prevention and control work in related fields.
Declaration of competing interest
We declare that we have no competing interests exists.
References
Casal J.I. Use of the baculovirus expression system for the generation of virus-like particles Biotechnol. Genet. Eng. Rev. 18 2001 73 87 11530699 10.1080/02648725.2001.10648009
Chen B. Yang Y. Wang Z. Dai X. Cao Y. Zhang M. Zhang D. Ni X. Zeng Y. Pan K. Surface Display of Duck Hepatitis A Virus Type 1 VP1 Protein on Bacillus subtilis Spores Elicits Specific Systemic and Mucosal Immune Responses on Mice Probiotics. Antimicrob. Proteins. 2024 10.1007/s12602-024-10323-2 39002060
Chen J.-H. Zhang R.-H. Lin S.-L. Li P.-F. Lan J.-J. Song S.-S. Gao J.-M. Wang Y. Xie Z.-J. Li F.-C. Jiang S.-J. The Functional Role of the 3’ Untranslated Region and Poly(A) Tail of Duck Hepatitis A Virus Type 1 in Viral Replication and Regulation of IRES-Mediated Translation Front. Microbiol. 9 2018 2250 30319572 10.3389/fmicb.2018.02250 PMC6167517
Comas-Garcia M. Colunga-Saucedo M. Rosales-Mendoza S. The Role of Virus-Like Particles in Medical Biotechnology Mol. Pharm. 17 2020 4407 4420 diagnostic techniques for duck A viral hepatitis 1 and 3 33147978 10.1021/acs.molpharmaceut.0c00828
Ding C. Zhang D. Molecular analysis of duck hepatitis virus type 1 Virology. 361 2007 9 17 17300822 10.1016/j.virol.2007.01.007
Fu Y. Pan M. Wang X. Xu Y. Yang H. Zhang D. Molecular detection and typing of duck hepatitis A virus directly from clinical specimens Vet. Microbiol. 131 2008 247 257 18462894 10.1016/j.vetmic.2008.03.011
Gao J. Chen J. Si X. Xie Z. Zhu Y. Zhang X. Wang S. Jiang S. Genetic variation of the VP1 gene of the virulent duck hepatitis A virus type 1 (DHAV-1) isolates in Shandong province of China Virol. Sin. 27 2012 248 253 22899433 10.1007/s12250-012-3255-8 PMC8218140
Gopal R. Schneemann A. Production and Application of Insect Virus-Based VLPs Methods Mol. Biol. 1776 2018 125 141 29869238 10.1007/978-1-4939-7808-3_8
Haider S.A. Calnek B.W. In vitro isolation, propagation, and characterization of duck hepatitis virus type III Avian Dis. 23 1979 715 729 230809
Kim M.-C. Kwon Y.-K. Joh S.-J. Kim S.-J. Tolf C. Kim J.-H. Sung H.-W. Lindberg A.M. Kwon J.-H. Recent Korean isolates of duck hepatitis virus reveal the presence of a new geno- and serotype when compared to duck hepatitis virus type 1 type strains Arch. Virol. 152 2007 2059 2072 17701025 10.1007/s00705-007-1023-0
Kim M.-C. Kwon Y.-K. Joh S.-J. Lindberg A.M. Kwon J.-H. Kim J.-H. Kim S.-J. Molecular analysis of duck hepatitis virus type 1 reveals a novel lineage close to the genus Parechovirus in the family Picornaviridae J. Gen. Virol. 87 2006 3307 3316 17030865 10.1099/vir.0.81804-0
Köhler G. Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity Nature 256 1975 495 497 1172191 10.1038/256495a0
Lai Y. He L. Cheng A. Wang M. Ou X. Wen X. Mao S. Sun D. Jia R. Yang Q. Wu Y. Zhu D. Chen S. Liu M. Zhang S. Zhao X.-X. Huang J. Gao Q. Liu Y. Yu Y. Zhang L. Tian B. Pan L. Rehman M.U. Chen X. Development and evaluation of an indirect ELISA based on recombinant structural protein VP2 to detect antibodies against duck hepatitis A virus J. Virol. Methods 282 2020 113903 10.1016/j.jviromet.2020.113903 32485472
Li J. Cao Y. Wang M. Cheng A. Ou X. Mao S. Sun D. Liu M. Zhang S. Zhao X.-X. Jia R. Yang Q. Wu Y. Zhu D. Chen S. Huang J. Gao Q. Tian B. Development of an indirect ELISA method based on the VP4 protein for detection antibody against duck hepatitis A virus type 1 J. Virol. Methods 300 2022 114393 10.1016/j.jviromet.2021.114393 34861317
Li X. Zhao R. Lin W. Li C. Zhang T. Meng F. Liu M. Zhang Y. Evidence of VP1 of duck hepatitis A type 1 virus as a target of neutralizing antibodies and involving receptor-binding activity Virus. Res. 227 2017 240 244 27816429 10.1016/j.virusres.2016.10.018
Liang R. Li C. Jin H. Meng C. Chen Z. Zhu J. Miao Q. Ding C. Liu G. Duck hepatitis A virus serotype 1 minigenome: a model for studying the viral 3’UTR effect on viral translation Virus. Genes. 51 2015 367 374 26578153 10.1007/s11262-015-1255-0
Liu X. Cheng Z. Zhang W. Mao L. Pan Z. Yang L. Liu M. Long Y. Bai J. Li W. Development and evaluation of a monoclonal antibody-based blocking ELISA to detect antibodies against the E2 protein of bovine viral diarrhea virus-1 J. Virol. Methods 323 2024 114851 10.1016/j.jviromet.2023.114851 37956891
Liu F. Wu X. Li L. Liu Z. Wang Z. Use of baculovirus expression system for generation of virus-like particles: successes and challenges Protein Expr. Purif. 90 2013 104 116 23742819 10.1016/j.pep.2013.05.009 PMC7128112
Ma X. Sheng Z. Huang B. Qi L. Li Y. Yu K. Liu C. Qin Z. Wang D. Song M. Li F. Molecular Evolution and Genetic Analysis of the Major Capsid Protein VP1 of Duck Hepatitis A Viruses: Implications for Antigenic Stability PLoS. One 10 2015 e0132982 10.1371/journal.pone.0132982 PMC4501807 26173145
Niu Y. Ma H. Ding Y. Li Z. Sun Y. Li M. Shi Y. The pathogenicity of duck hepatitis A virus types 1 and 3 on ducklings Poult. Sci. 98 2019 6333 6339 31393586 10.3382/ps/pez455 PMC8913754
Pattnaik B. Venkataramanan R. Comparison of liquid-phase and Mab-blocking ELISA for assessment of the reactivity of monoclonal antibodies to foot-and-mouth disease virus J. Immunol. Methods 172 1994 265 267 8034975 10.1016/0022-1759(94)90113-9
Ren Y. Yue H. Zhu L. Kan R. Tang C. Zhang B. A reverse transcription-insulated isothermal PCR assay for the detection of duck hepatitis A virus type 3 based on the POCKIT TM system J. Virol. Methods 270 2019 126 130 31100286 10.1016/j.jviromet.2019.05.006
Roh J.-H. Kang M. Live attenuated duck hepatitis virus vaccine in breeder ducks: Protective efficacy and kinetics of maternally derived antibodies Vet. Microbiol. 219 2018 107 112 29778181 10.1016/j.vetmic.2018.04.021
Roy P. Genetically engineered particulate virus-like structures and their use as vaccine delivery systems Intervirology 39 1996 62 71 8957671 10.1159/000150476
Ruzzi F. Semprini M.S. Scalambra L. Palladini A. Angelicola S. Cappello C. Pittino O.M. Nanni P. Lollini P.-L. Virus-like Particle (VLP) Vaccines for Cancer Immunotherapy Int. J. Mol. Sci. 24 2023 12963 37629147 10.3390/ijms241612963 PMC10454695
Song C. Yu S. Duan Y. Hu Y. Qiu X. Tan L. Sun Y. Wang M. Cheng A. Ding C. Effect of age on the pathogenesis of DHV-1 in Pekin ducks and on the innate immune responses of ducks to infection Arch. Virol. 159 2014 905 914 24162826 10.1007/s00705-013-1900-7
Truong T.-N. Cheng L.-T. Development of a subunit vaccine against duck hepatitis a virus serotype 3 Vaccines (Basel) 10 2022 523 35455272 10.3390/vaccines10040523 PMC9028120
Tseng C.-H. Knowles N.J. Tsai H.-J. Molecular analysis of duck hepatitis virus type 1 indicates that it should be assigned to a new genus Virus. Res. 123 2007 190 203 17067712 10.1016/j.virusres.2006.09.007
Wen X.J. Cheng A.C. Wang M.S. Jia R.Y. Zhu D.K. Chen S. Liu M.F. Liu F. Chen X.Y. Detection, differentiation, and VP1 sequencing of duck hepatitis A virus type 1 and type 3 by a 1-step duplex reverse-transcription PCR assay Poult. Sci. 93 2014 2184 2192 25012848 10.3382/ps.2014-04024
Woolcock P.R. Crighton G.W. Duck virus hepatitis: serial passage of attenuated virus in ducklings Vet. Rec. 105 1979 30 32 233456 10.1136/vr.105.2.30
Wu F. Lu F. Fan X. Pan Q. Zhao S. Sun H. Zhang J. Liu C. Chao J. Zhang X. Development of a live attenuated duck hepatitis A virus type 3 vaccine (strain SD70) Vaccine 38 2020 4695 4703 32446833 10.1016/j.vaccine.2020.05.030
Yan D. Wei Y.-Q. Guo H.-C. Sun S.-Q. The application of virus-like particles as vaccines and biological vehicles Appl. Microbiol. Biotechnol. 99 2015 10415 10432 26454868 10.1007/s00253-015-7000-8 PMC7080154
Yu C.-D. Choi Y.-R. Park J.-Y. Kim S.-W. Cha S.-Y. Jang H.-K. Kang M. Wei B. Establishment and Application of Mismatch Amplification Mutation Assay-PCR for Rapid Detection and Differentiation of Duck Hepatitis A Virus-1 Attenuated Vaccine and Wild Strains Animals. (Basel) 14 2024 2733 39335322 10.3390/ani14182733 PMC11428521
Yugo D.M. Hauck R. Shivaprasad H.L. Meng X.-J. Hepatitis Virus Infections in Poultry Avian Dis. 60 2016 576 588 27610716 10.1637/11229-070515-Review.1
Zhang Y. Ma X. Huang B. Li Y. Yu K. Li J. Liu C. Han H. Cui Y. Indirect ELISA for simultaneous detection of antibodies against duck hepatitis A type 1 and 3 viruses] Wei. Sheng. Wu Xue Bao 55 2015 501 509 26211325
Zhang Y. Wu S. Liu W. Hu Z. Current status and future direction of duck hepatitis A virus vaccines Avian Pathol. 52 2023 89 99 36571394 10.1080/03079457.2022.2162367
Zhang R. Xia L. Chen J. Gong Y. Zhang L. Li P. Liu H. Xie Z. Jiang S. Molecular epidemiology and genetic diversity of duck hepatitis A virus type 3 in Shandong province of China, 2012-2014 Acta Virol. 61 2017 463 472 29186963 10.4149/av_2017_409
Zhou S. Zhang S. Wang M. Cheng A. Zhu D. Chen S. Liu M. Zhao X. Jia R. Yang Q. Wu Y. Zhang S. Liu Y. Yu Y. Zhang L. Chen X. Development and evaluation of an indirect ELISA based on recombinant nonstructural protein 3A to detect antibodies to duck hepatitis A virus type 1 J. Virol. Methods 268 2019 56 61 30905595 10.1016/j.jviromet.2019.03.012
Zou Z. Ma J. Huang K. Chen H. Liu Z. Jin M. Live Attenuated Vaccine Based on Duck Enteritis Virus against Duck Hepatitis A Virus Types 1 and 3 Front. Microbiol. 7 2016 1613 27777571 10.3389/fmicb.2016.01613 PMC5056193
Acknowledgements
This research was supported by the Key Projects of the National Natural Science Foundation of China ( 32272977 ) and National Key Research and Development Program ( 2017YFD0500800 ).