infectious bursal disease
Tamiru Negasha, Martin Liman1, Silke Rautenschlein a*
a Clinic for Poultry, University of Veterinary Medicine Hannover, Bünteweg 17, 30559 Hannover, Germany
1 Present address: AniCon Labor GmbH, Mühlenstraße 13 49685, Höltinghausen, Germany
Vaccine (submitted)
Abstract
Infectious bursal disease virus (IBDV) is an immunosuppressive virus of chickens. The virus protein (VP) 2 induces neutralizing antibodies, which protect chickens against the disease. The aim of this study was to develop a cationic poly(D, L-lactide-co-glycolide) (PLGA) microparticle (MP) based IBDV-VP2 DNA vaccine (MP-IBDV-DNA) for chickens to be delivered orally and by eye drop route. The tested IBDV-VP2 DNA vaccines were immunogenic for specific-pathogen-free chickens and induced an antibody response after intramuscular application. Co-inoculation with a plasmid encoding chicken IL-2 (chIL-2) or CpG-ODN did not significantly improve protection against IBDV challenge. However, the application of a MP-IBDV-DNA vaccine alone or in combination with a delayed oral and eye drop application of cationic MP loaded with CpG-ODN or chIL-2 improved protection against challenge. The MP-IBDV-DNA-vaccinated chickens showed less pathological and histopathological bursal lesions, a reduced IBDV antigen load as well as T-cell influx into the bursa of Fabricius (BF) compared to the other groups (p<0.05). The addition of chIL-2 loaded MP improved challenge virus clearance from the BF as demonstrated by lower neutralizing antibody titers and reduced IL-4 and IFN-α mRNA expression in the bursa at 7 days postchallenge compared to the other challenged groups. Overall, the efficacy of the IBDV-DNA vaccine was improved by adsorption of the DNA vaccine onto cationic PLGA-MP, which also allowed mucosal application of the DNA vaccine.
Key words: Adjuvant, chicken, DNA vaccine, IBDV, poly(D, L-lactide-co-glycolide) microparticles
Abbreviations
ChIL-2= chicken interleukin 2; CpG-ODN = oligodeoxynucleotides (ODN) containing CpG; DNA= deoxyribonucleic acid; IBDV= infectious bursal disease virus; MP=
microparticles; PLGA= poly(D, L-lactide-co-glycolide); Tregs= regulatory T-cells;
VP2= virus protein 2
1. Introduction
Infectious bursal disease virus (IBDV) causes an acute and economically important immunosuppressive disease in poultry worldwide. IBDV is an Avibirnavirus [1] with a bisegmented dsRNA genome [2], of which segment A encodes the virus protein 2 (VP2) responsible for the induction of neutralizing antibodies [3].
IBDV replicates in the gut-associated lymphoid tissues and macrophages. The main target cells are immature intrabursal B-cells [4]. IBDV causes severe lymphoid cell depletion in the bursa of Fabricius (BF) [4,5]. IBDV infection of young chickens is marked by severe mortality and immunosuppression, which leads to vaccination failure and increases susceptibility of chickens to other pathogens [6].
Live IBDV vaccines are widely applied to commercial chickens to control IBDV [7,8]. Nevertheless, these vaccines may revert to virulent strains [9], induce immunosuppression [8] and may undergo segment reassortment with field strains [10]. IBDV VP2 subunit vaccines have been developed [11,12], but production costs of these subunit vaccines are usually high. Recently a vector vaccine on the basis of herpesvirus of turkeys-expressing IBDV VP2 was licensed [13].
DNA vaccines against IBDV have been developed encoding-VP2 or the polyprotein gene of IBDV. These vaccines often conferred only partial protection [14-16]. It was demonstrated that IBDV DNA vaccine efficacy can be improved by coadministering plasmid encoded chicken interleukin-2 (chIL-2) or CpG-ODN [17,18].
In mice and humans it was shown that DNA vaccination together with a delayed application of cytokine adjuvants may further improve the protective responses after DNA vaccination [19,20]. It is speculated that this strategy may reduce the expansion of immunosuppressive regulatory T-cells (Tregs) [20], which may otherwise suppress antigen specific T-cells.
Microparticulate carriers for DNA vaccines may provide additional adjuvant effects and protect the DNA from degradation after mucosal application [21-23].
Poly(D, L-lactide-co-glycolide) (PLGA) is the polymer of choice for cationic microparticle (MP) preparation [24]. Targeted delivery and prolonged persistence of the DNA when coated onto cationic MP improve the duration of DNA-vaccine induced immunity [21,25]. The objective of our study was to develop and test the
efficacy of an improved cationic MP based IBDV-DNA vaccine to be delivered by mucosal routes to chickens in combination with a delayed mucosal delivery of cationic MP adsorbing chIL-2 or CpG-ODN.
2. Materials and Methods
2.1. Chickens
Eggs from specific pathogen free (SPF) layer type chickens (VALO®, Lohmann LSL-LITE) were obtained from Lohmann Tierzucht (Cuxhaven, Germany).
The hatched chickens were randomly distributed and maintained in different groups under isolated conditions in the animal facilities of the University of Veterinary Medicine Hannover. All animal experimentations were approved and carried out following the institutional guidelines for animal care. Feed and water was supplied ad libitum
2.2. Cloning and expression of VP2 and chIL-2
cDNAs of VP2 of the two very virulent (vv) IBDV strains 82Eth [26] and 89163/7.3 (provided by N. Eterradossi, AFSSA, Ploufragan, France) were synthesized from total RNA using Oligo(dT)20 SuperScript™ III RT kit (Invitrogen).
The RNA had been isolated from bursae of IBDV infected birds. The vvIBDV strains had been isolated from several chicken flocks that had experienced high mortality rate [26]. The sequence coding for chIL-2 was amplified from splenocytes of chickens, which were stimulated in vitro with Concanavalin (Con) A (5 µg/mL) for 6 hrs [27]. PCR was conducted with the TaKaRa Ex Taq™ Polymerase (Takara Bio Inc., Shiga, Japan) using the primers; VP2f: GCCGGTACCGACGCAGCGATGACAAACCTGC-3’; VP2r: CGGGCGGCCGCTGATCACCTTATGGCCCGGATTA-3’, chIL-2f: GCCGGTACCGACGCAATGATGTGCAAAGTACTGATC-3’; and chIL-2r: 5’-CGGGCGGCCGCTGATTATTTTTGCAGATATCTCACAAA-3’). KpnI and NotI restriction sites were included (underlined in the primers). The PCR products were
double digested by FastDigest® NotI and FastDigest® KpnI (Fermentas GmbH, St.
Leon-Rot, Germany) and purified with NucleoSpin® Extract II PCR clean up and Gel extraction kit (Macherey-Nagel, Düren, Germany). Double digested PCR products were cloned into the pCR3.1® eukaryotic expression vector (Invitrogen) and transformed into chemically competent TOP10F' E. coli (Invitrogen) following the guidelines of the manufacturer. The inserts were verified by sequence analysis.
The expression of VP2 and chIL-2 was verified by immunofluorescence after transfection of chicken embryo fibroblasts (CEFs) with the respective plasmids (TransFectin™ lipid reagent; Bio-Rad, Hercules, California, USA). We used the following primary Abs: rabbit anti-IBDV polyclonal Ab [28], or VP2 mAb (provided by Egbert Mundt, FLI, Germany) and the mouse anti-chicken IL-2 mAb (AbD Serotec MorphoSys, Düsseldorf, Germany); and secondary antibodies: fluorescein isothiocyanate (FITC)-conjugated goat rabbit IgG or FITC-conjugated goat anti-mouse polyvalent Ab (Sigma-Aldrich, St. Louis, USA). VP2 and chIL-2 expressing plasmids were designated VP2/Eth82, VP2/7.3 and chIL-2. DNA-vec represented control DNA-vector (pCR3.1).
2.3. chIL-2 bioactivity assay
Splenocytes from SPF chickens were seeded (2x 105 cells/well) in complete RPMI medium (Biochrom AG, Berlin, Germany). Triplicates were stimulated with ConA (5 µg/mL), supernatants from DNA-chIL-2 or DNA-vec transfected CEFs at 41°C and 5% CO2 [18]. Lymphocyte proliferation was measured by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) (Sigma, St. Louis, MO) colorimetric assay [29].
2.4. Cationic MP preparation and DNA adsorption
Cationic MP were prepared as described previously [25,30] with the following modifications. Three-hundred mg of PLGA Resomer® RG 753S (Boehringer Ingelheim, Ingelheim, Germany) and 30 mg of Polyethylenimine (PEI; Sigma-Aldrich) were dissolved in 3 mL dichloromethane (DCM; Sigma-Aldrich) and in 2 mL DCM, respectively. The two organic solutions were mixed and homogenized with 0.5 mL
PBS, and subsequently with 25 mL of 0.5% Cetyltrimethylammonium bromide (Boehringer Ingelheim, Ingelheim, Germany). Finally, the organic solvents were evaporated and the cationic PLGA-MP were washed with 5% sucrose and lyophilized.
Plasmid DNAs were adsorbed onto the MP by incubating 100 mg of MP resuspended in distilled water (pH 5.0) with 1 mg of DNA as previously described [25,30]. MP were washed and lyophilized. Phosphorothioate backbone modified CpG-ODN (5'- TCGTCGTTGTCGTTTTGTCGTT-3'; CpG motifs underlined;
Biomers.net, Ulm, Germany) was adsorbed onto the MP following the protocol of plasmid DNA adsorption. This particular CpG-ODN sequence has been selected based on its significant immunostimulatory properties on avian immune cells; while its non-CpG-ODN sequence has been shown to induce only marginal upregulation of transcripts associated with antigen presentation [31,32].
The MP adsorbed with the DNA-VP2/Eth82, DNA-chIL-2, CpG-ODN, DNA-vec and non-adsorbed MP were designated as VP2-MP, chIL-2-MP, CpG-MP, Vec-MP and Mock-MP.
2.5. Characterization of MP
The diameter of the lyophilized MP was determined by the Grimm Aerosol Spectrometer (Grimm Technologies, Douglasville, USA). The MP adsorption efficiency for DNA was determined spectrometrically by subtracting the unbound DNA left in the supernatants at the end of the adsorption period from the amounts of DNA used for adsorption [25]. The resistance of the MP-adsorbed DNA to degradation by enzymes was determined by incubating the loaded MP with 15 U of DNAse I (Sigma). The integrity of the plasmid DNA was verified by gel electrophoresis [25]. A 24 hrs in vitro-release of the MP-adsorbed DNA was determined by incubating 10 mg loaded MP in PBS at 37°C. The amount of DNA released into the medium was determined spectrometrically.
2.6. In vitro phagocytosis of MP
FITC-labeled bovine serum albumin (FITC-BSA) was microencapsulated as described previously [25]. Chicken mononuclear cells (MQ-NCSU) were grown in L-15 Leibovitz-McCoy’s 5A modified medium and incubated with FITC-BSA-MP (250µg/well/mL) for 4 hrs at 41°C in a humidified 5% CO2 incubator. After washing, the MQ-NCSU cells were stained for MHC Class II surface antigen with an R-phycoerythrin (R-PE)-conjugated mouse anti-chicken MHC Class II Ab (B-L) (Southern Biotech, Biozol, Eching, Germany), and FITC-BSA-MP uptake and MHC class II staining was verified by immunofluorescence microscopy.
2.7. Antibody detection
IBDV-specific antibodies were detected by a commercial IBDV-ELISA (Synbiotics IBD ProFLOK Plus®, San Diego, USA). Neutralizing Abs were also measured [33]. Briefly, two-fold serially diluted sera were mixed with 100 TCID50 of an intermediate IBDV strain. Chicken embryo fibroblasts (105/well) were added and the development of cytopathic effects was evaluated. Geometric mean titers (log2) per group are presented.
2.8. Histopathology and immunohistochemistry
IBDV induced bursal lesion scores were determined as described before [34].
The mean bursal lesion score for each bird was determined from five microscopic fields (400x) and group mean computed. Bursal sections were immunohistochemically (IHC) stained for IBDV antigen following a previously published protocol [28]. The average number of IBDV antigen positive cells in the bursa of each bird was determined from five microscopic fields (400×). A maximum of 100 brown cells/field was counted.
2.9. Detection of cytokine mRNA
A TaqMan real-time RT-PCR amplification of chIL-4 and IFN-α mRNA from bursa samples was performed using the AgPath-ID™ One-Step RT-PCR Reagent kit (Applied Biosystems®) and the detection and quantification was done by the
Mx3005PTM thermal cycler (STRATAGENE; Agilent Technologies Company). The relative transcript levels of these cytokines were normalized to the chicken 28S rRNA [35] and values are expressed as 35-CT. The primers, probes and amplification conditions were described previously [28,35].
2.10. Intrabursal T-cells
T-cell influx in the bursa after IBDV challenge was evaluated by immunohistochemistry. Cryosections from frozen bursa were stained for CD4+- and CD8+ T-cells. Mouse anti-chicken CD4 and CD8 mAbs (Southern Biotech, Eching, Germany) were used as primary Abs, and the anti-mouse IgG biotinylated Ab as a secondary antibody (Vector Laboratories, Burlingame, USA). T-cell quantification was conducted as described for IBDV Ag.
2.11. Experimental design
2.11.1. DNA vaccination (experiment 1)
Three weeks old SPF chickens were assigned into 5 groups (n=14-16/group).
Group 1, 2 and 3 were immunized into the thigh muscle (100 μg DNA/bird) with DNA-VP2/Eth82, DNA-VP2/7.3 and DNA-vec, respectively (Table 1). Two boost DNA immunizations were conducted every two weeks. Group 4 received PBS intramuscularly. Group 5 was vaccinated orally with a commercially available chicken embryo adapted live IBDV vaccine strain (103EID)50/bird). Blood samples were collected weekly for Ab detection. Two weeks after the last vaccination each group was subdivided. One subgroup was challenged with the vvIBDV strain 89163/7.3 (103 ELD50/bird) by eye drop [36] and the other subgroup remain unchallenged. Birds were monitored for morbidity and mortality. At 3 and 10 days postinfection (dpi), 3-4 chickens/subgroup were sacrificed. Bursa to body weight (B/B) ratio was determined.
Bursae were collected for histology and IBDV Ag detection.
2.11.2. DNA vaccination and adjuvant application (experiment 2)
The vaccine groups (n=9-10/group; 3 weeks old SPF chickens) and descriptions are presented in Table 2. The route of vaccination, dosage, frequency,
challenge and sampling in this experiment was similar to the first experiment.
Plasmid encoded chIL-2 (100 μg /bird) or CpG-ODN (25 µg/bird) were mixed with the DNA vaccine and delivered in one shot. The dosage for CpG-ODN and chIL-2 were based on results of previous studies [17,18]. Necropsy was done at 7 dpi.
2.11.3. Vaccination with cationic MP (experiment 3)
The protective efficacy of the MP based IBDV DNA vaccine was tested after mucosal application to 3 days-old SPF chickens (n =10-12/group) intended to see if even in immunologically immature birds this regime may lead to a detectable immune responses. The vaccination protocol is summarized in Table 3. Briefly, chickens of all groups received 4.5 mg MP (3.5 mg orally and 1 mg by eye drop), except the CpG-MP group. One mg of VP2-MP or chIL-2-MP adsorbed 8.7 µg of DNA-VP2/Eth82 or DNA-chIL-2. When CpG-ODN was administered, each chicken received 3 mg CpG-MP (2 mg orally and 1 mg by eye drop). One mg of CpG-MP adsorbed 5.6 µg CpG-ODN. Group 5 (G5) and 6 (G6) were administered respectively with chIL-2-MP and CpG-MP as an adjuvant two days after inoculation with VP2-MP.
Two booster immunizations were administered every two weeks. Chickens were off-fed for 2-3 hrs after MP administration. Two weeks after the last vaccination, chickens were orally challenged with an intermediate plus IBDV vaccine strain (≥2.0 log10 EID50/bird), which induce significant macroscopical and microscopical bursal lesions without high mortality rates [8]. Group 1 (G1) served as non-challenge control. At 3 and 7 dpi, 5-6 chickens were sacrificed and B/B ratio was determined.
Bursae were collected for histology, Ag detection, cytokine mRNA analysis and T-cell evaluation.
2.12. Statistics
ANOVA was used to analyze the data on lymphocyte proliferation and from experiment 3. Group means were compared with Tukey HSD pot-hoc test. Kruskal-Wallis one-way nonparametric ANOVA was used in other experiments. Student’s t-test was used to analyze B/B ratios between subgroups. p<0.05 indicates significant differences.
3. Results
3.1. Cloning and expression of VP2 and chIL-2
Transfection studies revealed expression of VP2 and chIL-2 in CEFs (data not shown). Chicken lymphocytes stimulated with supernatants from CEFs, which had been transfected with DNA-chIL-2, showed a higher lymphoproliferative index (2.0±0.5) (p<0.05) than treatment with supernatants from vector-transfected cells (0.9±0.1).
3.2. Immunogenicity of DNA vaccines (experiment 1&2)
At five weeks post intramuscular immunization, 75% (12/16) of the DNA-VP2/Eth82 vaccinated chickens had developed IBDV-ELISA antibody titers of 2070±1565, comparably lower than the live virus vaccinated group (9734±1824).
Only 40% of the DNA-VP2/7.3 (6/15) vaccine group had seroconverted at week 5 after immunization. Four out of eight DNA-VP2/Eth82 vaccinated bird showed depression and ruffled feathers after challenge. The vector and PBS groups showed severe clinical signs (Table 1). The DNA-VP2/Eth82 vaccinated group had a significantly lower mean bursal lesion score (p<0.05) and fewer intrabursal Ag positive cells compared to the other challenged groups, except the live vaccine group (Table 1) indicating only partial protection. DNA-chIL-2 or CpG incorporation did not improve the immunogenicity of the DNA vaccine (Table 2).
3.3. Characterization of cationic MP
The prepared MP had diameters of 3.5-8.5 µm, and the FITC-BSA-MP were successfully phagocytized by MQ-NCSU cells (data not shown). The adsorption efficiency was 83-87% for DNA-VP2/Eth82 and DNA-chIL-2, and 56% for CpG-ODN, indicating the adsorption of 8.3-8.7 µg of plasmid DNA and 5.6 µg of CpG-ODN to 1 mg of cationic MP. Forty five percent of the adsorbed plasmid DNA and 60% of the CpG-ODN were released into the incubation medium after 24 hrs in vitro incubation at 37°C. The MP-adsorbed DNA was protected from degradation by DNAse I as the
supercoiled plasmid was preserved at the end of DNAse I degradation assay (data not shown).
3.4. Evaluation of the protective efficacy of the cationic MP-delivery strategy for IBDV-DNA vaccination (experiment 3)
At 3 dpi, no difference was observed in the B/B ratios between groups (p>0.05). The chickens in G4 (VP2-MP), G5 (VP2-chIL-2-MP) and G6 (VP2-CpG-MP) had a significantly higher B/B ratio at 7 dpi compared to the other challenged groups (Table 4) (p<0.05). At 7 dpi bursal lesions were lower in G4, 5, and 6 (p<0.05) compared to the G2 (Mock-MP+), 3 (Vec-MP), 7 (chIL-2-MP) and 8 (CpG-MP) groups (p<0.05) (Table 4). All VP2-MP groups had a relatively lower intrabursal Ag load compared to the other groups at 3 and 7 dpi (Table 4) (p<0.05). Chickens in G5 had the lowest intrabursal Ag load and three out of six chickens had cleared the virus at 7 dpi.
Neutralizing antibodies were neither detected in prechallenge sera nor at 3 dpi in any group. Chickens vaccinated with VP2-ChIL-2-MP had the lowest postchallenge VN Ab titers at 7 dpi (Table 4).
At 3 dpi no significant differences were detected in bursal IFN-α and IL-4 mRNA expression between groups (Fig. 1A & C). At 7 dpi the expression levels of chIFN-α mRNA were significantly lower in the bursae of G4, 5 and 6 (Fig. 1B) compared to other groups (p<0.05). Significant upregulation of IL-4 mRNA was observed in G2 and G3 at 7 dpi compared to G1, G4, and G5 (Fig. 1D).
Higher numbers of intrabursal CD4+ T-cells were detected in the Mock-MP+
group compared to the other groups at 3 dpi (Fig. 2A; p<0.05). The number of CD8+
T-cells increased in all challenged groups at 3 dpi and G2, G3, G7 and G8 (Fig. 2C) had the highest number of CD8+ cells. At 7 dpi more CD4+ (Fig. 2B) and CD8+ T-cells (Fig. 2D) were recruited to the bursa of G2, 3, 7 and 8 (p<0.05) compared to the VP2-MP and VP2-MP plus adjuvant groups.
4. Discussion
The aim of this study was to develop an improved MP based IBDV DNA vaccine and to further enhance the protection efficacy of this vaccine by including MP loaded with adjuvants.
The VP2-DNA vaccines were shown to be immunogenic and conferred partial protection against challenge. Generally, DNA vaccinations are known to not induce significant Ab responses, but may provide partial protection against virulent IBDV [11,15,17,18]. Co-administration of CpG-ODN or plasmid encoded chIL-2 together with the IBDV-VP2 or polyprotein gene induced variable levels of humoral and cellular immunity. Yet, chickens mostly exhibited bursal atrophy after challenge [17,37,38], which were even more pronounced than in groups without adjuvants [39], coinciding with our observation.
The characterization of the MP revealed appropriate physical properties for in vivo application [25, 30]. PEI incorporation into the MP may condense plasmid DNA on the MP surfaces and likely prevents DNA degradation by mucosal enzymes [23,30].
The mucosal delivery of the VP2-loaded MP (VP2-MP) in combination with a delayed mucosal application of chIL-2 or CpG-ODN loaded MP did not induce virus neutralizing Abs. We speculate that the DNA is slowly released from the MP, VP2 is expressed, processed and presented by APCs to culminate in a CTL response [40].
After IBDV challenge, bursal pathology and intrabursal Ag load and T-cell influx were reduced in all VP2-MP vaccinated groups. This provides the proof-of-concept for the potential of MP to improve DNA vaccine efficacy. The lower antigen load in VP2-MP plus chIL-2-MP-inoculated chickens may explain the lower postchallenge neutralizing Ab titers, IL-4 and IFN-α mRNA expression possibly due to reduced virus replication or faster clearance in these birds compared to other groups.
CpG-ODN and chIL-2 were shown to exert potent adjuvant effects when delivered by particulate carriers compared to their delivery in soluble forms [41,42].
Baden et al. [19] suggested that increased IL-2 receptor expression on antigen primed T-cells may be necessary for an optimal activation and proliferation of these
cells by IL-2. However, during simultaneous administration of an Ag and cytokine, the cytokine may simulate rather a broad cellular response with relatively fewer antigen-specific cells [19]. Tregs in mice and humans showed an increased level of toll like receptor (TLR) expression [43] and may expand to an extent dominating antigen specific CD4+ T-cells when CpG was administered simultaneously with a vaccine.
Thus, the timing of TLR agonist (CpG-ODN) application may influence the outcome of vaccine-induced immune responses [44]. We hypothesize that in our vaccination model the delayed application of chIL-2 after an MP-DNA delivery may enhance the activation and proliferation of Ag specific T-cells in chickens before extensive proliferation of immunosuppressive Tregs may take place.
Overall the results of this study indicate that DNA-vaccination against IBDV using MP as a delivery system in combination with molecular adjuvants may be an interesting alternative IBDV-vaccination approach, which may help to reduce the risks associated with the use of live vaccine strains. In future studies, we not only need to investigate additional routes of administration but also we need to elucidate the mechanism of protection and to what extent the different types of T-cells are
Overall the results of this study indicate that DNA-vaccination against IBDV using MP as a delivery system in combination with molecular adjuvants may be an interesting alternative IBDV-vaccination approach, which may help to reduce the risks associated with the use of live vaccine strains. In future studies, we not only need to investigate additional routes of administration but also we need to elucidate the mechanism of protection and to what extent the different types of T-cells are