- Email: [email protected]
Induction of robust immunity by the emulsification of recombinant lipidated dengue-1 envelope protein domain III
Microbes and Infection 15 (2013) 719e728 www.elsevier.com/locate/micinf
Original article
Induction of robust immunity by the emulsification of recombinant lipidated dengue-1 envelope protein domain III Chen-Yi Chiang a, Ming-Hsi Huang a, Chien-Hsiung Pan a,b, Chun-Hsiang Hsieh a, Mei-Yu Chen a, Hsueh-Hung Liu a, Jy-Ping Tsai a, Shih-Jen Liu a,b, Pele Chong a,b, Chih-Hsiang Leng a,b,**, Hsin-Wei Chen a,b,* a
National Institute of Infectious Diseases and Vaccinology, National Health Research Institutes, Miaoli, Taiwan, ROC b Graduate Institute of Immunology, China Medical University, Taichung, Taiwan, ROC Received 20 November 2012; accepted 5 June 2013 Available online 15 June 2013
Abstract Many attempts have focused on the use of either immunomodulators or antigen delivery systems to obtain an efficacious vaccine. Here, we report a novel approach that combined an immunomodulator and delivery system to enhance antigen association and induce robust immunity. We expressed a recombinant lipidated dengue-1 envelope protein domain III (LD1ED III) and its non-lipidated form, D1ED III, in an Escherichia coli system. The LD1ED III contains a bacterial lipid moiety, which is a potent immunomodulator. We demonstrated that LD1ED III possesses an inherent immunostimulation ability that can activate RAW 264.7 macrophage cells by up-regulating their expression of CD40, CD80, CD83, CD86 and MHC II, whereas D1ED III could not induce the up-regulation of these molecules. Moreover, combining LD1ED III with a multiphase emulsion system (called PELC) increased the antigen association more than either combining D1ED III with PELC or the antigen alone. Enhanced antigen association has been shown to correlate with stronger T cell responses, greater antibody avidity and improved neutralizing capacity. Our results demonstrate that combining recombinant lipoproteins with PELC improved both the intensity and the quality of the immune response. This approach is a promising strategy for the development of subunit vaccines that induce robust immunity. Ó 2013 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved. Keywords: Dengue; Immune response; Immunostimulation; Lipoprotein
1. Introduction Vaccines are recognized as one of the most effective and successful strategies to combat infectious diseases [1]. Recombinant protein vaccines may have significant advantages over either live attenuated or inactivated vaccines.
* Corresponding author. National Institute of Infectious Diseases and Vaccinology, National Health Research Institutes, Miaoli, Taiwan, ROC. Tel.: þ886 37 246 166x37706; fax: þ886 37 583 009. ** Corresponding author. National Institute of Infectious Diseases and Vaccinology, National Health Research Institutes, Miaoli, Taiwan, ROC. Tel.: þ886 37 246 166x37711; fax: þ886 37 583 009. E-mail addresses: [email protected] (C.-H. Leng), [email protected] (H.-W. Chen).
Recombinant subunit vaccines are safer and easier to produce and administer than conventional vaccines because they do not contain live pathogens. However, subunit vaccines are usually poorly immunogenic and require an adjuvant to induce robust immune responses [2,3]. Therefore, combining recombinant protein antigens with effective immunostimulators and/or delivery systems are proposed to greatly improve the ability of subunit vaccines to elicit robust immune responses. Several pattern-recognition receptors (PRR), such as toll-like receptors [4,5], retinoic acid-inducible gene-I-like receptors [6], and nucleotide-binding and oligomerization domain-like receptors [7,8], were identified and shown to be involved in the recognition of pathogen-associated molecular patterns. Because engaging the PRR ligands can activate the antigen-presenting cells to trigger innate and adaptive immunity, the ability of
1286-4579/$ - see front matter Ó 2013 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.micinf.2013.06.002
720
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
PRR ligands to act as immunostimulators in vaccine formulations has been studied for decades [9e11]. The choice of delivery system can also improve the host’s antigen-specific immune response and can include emulsions [12e14], liposomes [9,10], virosomes [15], and biocompatible materials [16]. A well-informed and sensible combination of immunostimulators and an appropriate delivery system will contribute to the development of new vaccines. To overcome the inherent disadvantages of subunit vaccines, we developed a novel platform technology to express high levels of recombinant lipoproteins with built-in adjuvant abilities [17]. The lipid moiety of the recombinant lipoprotein can act as an immunostimulator for the vaccine candidates. We demonstrated that the novel recombinant lipoproteins could activate the antigen-presenting cells via toll-like receptor 2 [18] and enhance immune responses [19]. In parallel, we developed a water-in-oil-in-water multiphase emulsion system, called PELC, for vaccine delivery. The combination of inactivated H5N1 virus or ovalbumin with PELC can enhance immune responses [12,20,21]. In the present study, recombinant lipidated dengue-1 envelope protein domain III (LD1ED III) has been used as a model protein antigen for PELC formulation. Our objective in this study was to determine whether the combination of LD1ED III, which contains a ligand for toll-like receptor 2, and an emulsion system could synergize to activate the cellular and humoral responses. 2. Materials and methods 2.1. Mouse experiments and virus BALB/c mice were purchased from the National Laboratory Animal Breeding and Research Center, Taipei, Taiwan. All the animal studies were approved by the Animal Committee of the National Health Research Institutes (Protocol No: NHRI-IACUC-098014) and were performed according to their guidelines. Groups of BALB/c mice (6e8 weeks of age) were immunized subcutaneously with either LD1ED III or recombinant dengue-1 envelope protein domain III (D1ED III) (10 mg per dose) combined with PELC or phosphate buffered saline (PBS). Mice were given one or two boosting vaccinations at two-week intervals using the same regimen. Blood samples were collected from each mouse at the indicated time points. The sera were prepared and stored at 20 C until use. The dengue-1 virus (Hawaii) was used for this study. The virus was propagated in C6/36 cells, and viral titers were determined by focus-forming assays using BHK-21 cells. Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Thermal Scientific HyClone, Logan, Utah) supplemented with 1 mM sodium pyruvate (Gibco Invitrogen, Carlsbad, CA) and 5% fetal bovine serum (Thermal Scientific HyClone). 2.2. Cloning and expression of recombinant proteins The recombinant D1ED III was cloned and expressed as previously described [22]. Regarding LD1ED III, the D1ED
III gene was cloned into the pLipo plasmid [17] using the Bam HI and Xho I sites to produce the pLD1ED III plasmid. The Cterminus of the recombinant protein contained an additional hexahistidine tag (HisTag). The Escherichia coli strain C43(DE3) (Lucigen, Middleton, WI) was transformed with the expression plasmid, pLD1ED III, for LD1ED III expression. The transformed cells were cultured at 37 C overnight. The overnight culture was scaled up in a shaking flask and incubated at 37 C for 4 h before induction. Protein expression was induced when the cultures reached an OD600 of 0.8 by adding 1 mM isopropylthiogalactoside (IPTG), and the cultures were then incubated at 20 C for 20 h. 2.3. Purification of recombinant proteins The purification of recombinant D1ED III was described previously [22]. The recombinant LD1ED III was purified by disrupting the harvested cells in a French Press (Constant Systems, Daventry, UK) at 27 kpsi in homogenization buffer [20 mM Tris (pH8.0), 50 mM sucrose, 500 mM NaCl and 10% glycerol] and stirring at 4 C for 2 h after adding 0.5% triton X-100 to the supernatant. The cell lysate was clarified by centrifugation (80,000 g for 40 min). Most of the LD1ED III was in the soluble fraction. The LD1ED III was purified using immobilized metal affinity chromatography (IMAC) columns. The purified LD1ED III was eluted from the IMAC column with homogenization buffer containing 500 mM imidazole. An Endotoxin Removing Gel (Pierce, Rockford, IL, USA) was used to remove the lipopolysaccharide (LPS). The LPS levels of the purified LD1ED III were determined by the Limulus amebocyte lysate assay (Associates of Cape Cod, Inc. Cape Cod, MA), and the resulting LPS levels were less than 0.04 EU/mg. After the LD1ED III was dialyzed against 25 mM ammonium bicarbonate/3 mg/mL sucrose, the LD1ED III was lyophilized and stored at 20 C. The fractions from each step were analyzed by SDS-PAGE gel stained with Coomassie blue (Coomassie Brilliant Blue R-250) and immunoblotted with anti-HisTag antibodies. 2.4. Identification of the lipid moiety in LD1ED III The LD1ED III was digested with trypsin (Sigma, St. Louis, MO). After the digestion, the reaction mixture was further polished using Ziptip (Millipore, Massachusetts). One microliter of the polished tryptic fragments was mixed with 1 mL of a saturated solution of a-ciano-4-hydroxycinnamic acid in acetonitrile/0.1% trifluoroacetic acid (1:3, vol/vol). One microliter of the mixture was placed on the target plate of a MALDI micro MX mass spectrometer (Waters, Manchester, UK) for analysis. 2.5. Preparation of PELC PELC is a squalene water/oil/water nanoemulsion stabilized by SpanÒ85 (sorbitan trioleate, SigmaeAldrich, Steinheim, Germany) and diblock tri-component copolymer poly(ethylene glyco)-block-poly(lactide-co-ε-caprolactone) (PEG-b-PLACL),
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
the latter consisting of 75 wt-% of hydrophilic bioabsorbable PEG and 25 wt-% of lipophilic biodegradable PLACL with a molecular weight of 7000 Da, as previously described [22]. Briefly, 120 mg of PEG-b-PLACL, 0.8 mL of PBS, and 1.1 mL of an oily solution consisting of squalene (SigmaeAldrich, Steinheim, Germany) and SpanÒ85 (85/15 v/v) were emulsified using a PolytronÒPT 3100 homogenizer (Kinematica AG, Switzerland) at 6000 rpm for 5 min. The emulsified PELC formulation was stored at 4 C until use. The PELC-formulated vaccine was prepared by re-dispersing 0.2 mL of the stock emulsion into 1.8 mL of the bulk vaccine candidates and mixing with a test-tube rotator (Labinco LD-79, Netherlands) at 5 rpm for at least 1 h before injection. 2.6. Assays for the activation of RAW 264.7 macrophage cells RAW 264.7 macrophage cells were cultured in DMEM supplemented with (Thermal Scientific HyClone) 1 mM sodium pyruvate (Gibco Invitrogen) and 10% fetal bovine serum (Thermal Scientific HyClone). Cells were stimulated with 10 mg/mL of either LD1ED III or D1ED III. The cells cultured in media alone served as controls. After 16 h stimulation, cells were harvested for cell surface marker staining using either PEor FITC-conjugated antibodies specific for CD40, CD80, CD83, CD86 and MHC class II. The data were acquired using CellQuest Pro software on a BD FACSCalibur flow cytometer and were analyzed using FACS 3 software. The basal expression level was defined as the mean fluorescence intensity (MFI) of cells cultured in media alone. All antibodies were purchased from either BD Biosciences (San Diego, CA) or eBioscience (San Diego, CA). The production of IL-6 by RAW 264.7 macrophage cells was determined by ELISA detection set follow the manufacturer’s instructions (R&D systems). 2.7. Assays for antigen association with RAW 264.7 macrophage cells LD1ED III or D1ED III was combined with either PELC or PBS. RAW 264.7 macrophage cells were pulsed with different formulations of antigens at 37 C for 3 h. The cells were fixed and permeabilized using a cytofix/cytoperm kit (eBioscience) according to the manufacturer’s instructions. The antigen association with RAW 264.7 macrophage cells was detected using D1ED III immunized sera. The bound antibodies were detected using a secondary antibody conjugated to FITC then analyzed using flow cytometry as described above. The antigen association index was calculated according to the formula (MFI detected by D1ED III immunized sera)/(MFI detected by naı¨ve sera). 2.8. Blocking of dengue virus infection in BHK-21 cells by LD1ED III To test whether LD1ED III blocks the dengue virus infection of BHK-21 cells, the virus was pre-mixed with different amounts of LD1ED III, D1ED III or control bovine serum albumin (BSA) protein as indicated for 10 min at 4 C. The
721
virus titer prior to pre-mixing was approximately 20e40 focus-forming units (FFUs) per well. Viral adsorption was allowed to proceed for 3 h at 37 C. An overlay medium containing 2% fetal bovine serum and 0.8% methylcellulose in DMEM was added at the conclusion of adsorption. The infected monolayer was incubated at 37 C. After 72 h of infection, the overlay medium was removed from the wells, and the BHK cells were washed with cold PBS. The cells were fixed for 15 min in 3.7% formaldehyde/PBS. After washing with PBS, the cells were permeabilized with 0.1% Nonidet P40/PBS for 15 min and blocked with 3% BSA/PBS for 30 min. The infected cells were detected using a monoclonal anti-dengue antibody (American Type Culture Collection, No. HB-114). After washing with PBS, the antibody-labeled cells were detected using a secondary antibody conjugated to HRP. The labeling was visualized using 3,30 ,5,50 -tetramethylbenzidine (TMB). The FFUs were counted. 2.9. Focus reduction neutralization tests (FRNT) Two-fold serial dilutions of sera were made (starting at 1:8), and the sera were heat inactivated prior to testing. A monolayer of BHK-21 cells in 24-well plates was inoculated with dengue-1 virus that had been pre-mixed at 4 C overnight with pre-immunization or post-immunization sera to a final volume of 0.2 mL. The FFUs were determined as described above. The neutralizing antibody titer FRNT50 (or FRNT90) was calculated as the reciprocal of the highest dilution that produced a 50% (or 90%) reduction of FFU compared with control samples containing the virus alone. For calculation purpose, the neutralizing antibody titer was designated as 22 when neutralizing antibody titer was less than 23. 2.10. Assays for lymphocyte proliferation and cytokine secretion The spleens were removed one week after the last immunization and processed into a single-cell suspension. The splenocytes were seeded at a concentration of 2 105 cells/ well in 96-well plates and were then stimulated with D1ED III (1 mg/mL) for 3 days at 37 C in a 5% CO2 humidified incubator. The growth medium of splenocytes was Roswell Park Memorial Institute medium supplemented with 50 mM beta-mercaptoethanol (SigmaeAldrich, St. Louis, MO), 20 mM HEPES (Biological Industries, Israel), and 5% fetal bovine serum (Thermal Scientific HyClone). During the final 18 h of culture, 1 mCi of [3H]-thymidine was added to each well, and the cells were harvested using a FilterMate automatic cell harvester (Packard). The incorporated radioactivity was determined with a TopCount microplate scintillation counter (Packard). Con A (5 mg/mL) was added to some wells as a positive control for cell proliferation. The induced stimulation index is defined as the ratio of the mean counts per minute (cpm) with D1ED III stimulation to the mean cpm without D1ED III stimulation. To measure the secretion of IFN-g, the splenocytes were cultured and pulsed with D1ED III as described above. The
722
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
Fig. 1. Preparation and characterization of recombinant lipidated dengue-1 envelope protein domain III. (A) The amino acid sequence of LD1ED III is a consensus sequence of dengue virus type I (Ref: PLoS Negl. Trop. Dis. 6 (2012) e1645). The DNA sequence of the D1ED III gene was derived using the codon usage of E. coli and fully synthesized using the assembly PCR method. The PCR product was cloned into the pET22b-based vector with a lipid signal peptide in front of D1ED III gene for LD1ED III expression. The recombinant lipoproteins contained an additional HHHHHH sequence (HisTag) at the C-terminus and were expressed under the control of the T7 promoter. (B) The LD1ED III protein purification process was monitored using a 15% reducing SDS-PAGE followed by staining with Coomassie blue (left panel) and immunoblotting (right panel) using anti-HisTag antibodies. The LD1ED III was expressed in E. coli strains C43 (DE3). Lane 1, LD1ED III expression after IPTG induction; lane 2, protein expression in the absence of IPTG induction; lane 3, soluble fraction of LD1ED III; lane 4, purified LD1ED III. Lanes 5e8 show immunoblotting to monitor the LD1ED III induction and purification processes, and the samples in these lanes are the same as those in lanes 1e4. The arrows indicate the electrophoretic positions of LD1ED III in the gels or blots. (C) N-terminal LD1ED III fragments were obtained and identified after digestion of LD1ED III with trypsin. The digested sample was analyzed on a WatersÒ MALDI micro MXÔ mass spectrometer. The MALDI-TOF MS spectra revealed the existence of five peaks with m/z values of 1452, 1466, 1480, 1492 and 1506.
cells were cultured for 4 days, and then cell-free supernatants were harvested and stored at 80 C until use. The quantification of total IFN-g was performed using IFN-g ELISA kits (R&D systems) according to the manufacturer’s instructions, and these data were expressed in pg/mL.
2.11. Measurement of antibody avidity The antibody avidity was determined on the basis of D1ED III-specific IgG dissociation due to the chaotropic agent ammonium thiocyanate. Briefly, purified D1ED III was coated
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
723
onto 96-well plates. After blocking with 1% BSA/PBS, serum at either 1:100 or 1:300 were incubated at room temperature for 1 h. The plates were washed and incubated with 0e3.5 M ammonium thiocyanate in 0.5 M increments at room temperature for 15 min. The bound IgG was detected with HRPconjugated goat anti-mouse IgG Fc. After the addition of TMB, the absorbance at 450 nm was measured with an ELISA reader. The avidity index was calculated as the concentration of ammonium thiocyanate that resulted in a 50% reduction from the initial absorbance [23,24]. 2.12. Statistical analyses The statistical analyses were carried using GraphPad Prism software version 5.02 (GraphPad Software, Inc.). The data were processed using the ANOVA Bonferroni post test. Differences with a p value of less than 0.05 were considered to be statistically significant. 3. Results 3.1. Biochemical characterization of recombinant lipidated dengue-1 envelope protein domain III The D1ED III gene was cloned into the Bam HI and Xho I sites of the pET-22b-based plasmid containing a lipid signal peptide to produce the pLD1ED III plasmid. The recombinant lipoprotein contained an additional HHHHHH sequence (HisTag) at its C-terminus and was expressed under the control of the T7 promoter (Fig. 1A). The recombinant LD1ED III was purified using immobilized metal affinity chromatography (IMAC) columns (Fig. 1B, lanes 1e4). LD1ED III was detected with anti-HisTag antibodies (Fig. 1B, lanes 5e8). After the LPS was removed (less than 0.04 EU/mg), the purified LD1ED III antigens were analyzed for their immunogenicity and efficacy in animal models. We then measured the exact mass of the N-terminal fragments of the recombinant LD1ED III. Five peaks with m/z values of 1452, 1466, 1480, 1492, and 1506 were identified (Fig. 1C). These peaks have been identified as a lipidation signature in other lipidated proteins [17,25]. In our previous study [26], we identified two groups of N-terminal lipidated (diacyl or triacyl) molecules in the bacterial expressed recombinant lipoproteins. We confirmed that the peaks of LD1ED III were associated with lipidated cysteine residues and verified that LD1ED III contains a bacterial lipid moiety at its N-terminus. 3.2. Functional evaluation of recombinant lipidated dengue-1 envelope protein domain III
Fig. 2. Inhibition of dengue viral infection by LD1ED III and D1ED III. Dengue-1 virus was pre-mixed with different amounts of LD1ED III, D1ED III or control BSA protein as indicated for 10 min at 4 C. Viral adsorption on a monolayer of BHK-21 cells in 24-well plates proceeded for 3 h at 37 C. The focus-forming units (FFUs) were determined 3 days after the viral infection. The results shown are pooled from two independent experiments.
[22]. The focus number was inhibited greater than 90% when LD1ED III added to the cells at a concentration of 0.5 mg/mL. In contrast to LD1ED III, BSA did not inhibit dengue-1 focus formation even at 3-fold higher concentrations of 1.5 mg/mL. These results suggest that recombinant LD1ED III retains the correct conformation and can block the cellular binding sites of the dengue-1 virus. Our results [18] and others [29e31] have demonstrated that lipoproteins are able to stimulate antigen-presenting cells via the toll-like receptors. RAW 264.7 macrophage cells were pulsed with either D1ED III or LD1ED III at 10 mg/mL for 16 h to examine the functional activity of LD1ED III. The expression of the activation markers CD40, CD80, CD83, CD86, and MHC II were analyzed by the flow cytometry (Fig. 3A). The basal expression level was defined as the MFI of cells cultured in media alone. The relative MFIs from three independent experiments are shown in Fig. 3B. LD1ED III was able to significantly increase the expression of CD40, CD80, CD83, CD86, and MHC II on macrophages from 1.7to 3.9-fold. Consistent with the expression of the activation markers, LD1ED III was also able to significantly stimulate the production of IL-6 by macrophages (Fig. 3C). Interestingly, D1ED III without lipidation did not induce the upregulation of activation markers and IL-6 production on macrophages. 3.3. Effects of formulations on antigen association
It has been demonstrated that dengue envelope protein domain III is involved in cellular receptor binding [27,28]. We evaluated the ability of LD1ED III to interfere with dengue viral infections. As shown in Fig. 2, the ability of dengue-1 virus to infect BHK-21 cells was inhibited in the presence of LD1ED III in a dose-dependent manner. The inhibition effect of LD1ED III is equivalent to non-lipidated D1ED III
We previously developed a water-in-oil-in-water multiphase emulsion system, PELC, for vaccine delivery [12,20]. D1ED III and LD1ED III were combined with either PELC or PBS to evaluate the effects of PELC on recombinant lipoproteins. RAW 264.7 macrophage cells were pulsed with various formulations for 3 h. The antigen associated with
724
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
Fig. 3. Effects of LD1ED III on the activation of RAW 264.7 macrophage cells. RAW 264.7 macrophage cells were cultured either in media alone or in media supplemented with LD1ED III or D1ED III at 10 mg/mL. After incubation for 16 h. (A) The surface markers CD40, CD80, CD83, CD86, and MHC II were analyzed using flow cytometry. A representative experiment is shown. (B) The basal expression level was defined as the mean fluorescence intensity (MFI) of cells cultured in media alone. Relative MFIs were plotted. (C) The supernatants were harvested and analyzed for IL-6 using ELISA. The means and standard deviations from three independent experiments are shown. Statistical significance was determined by the ANOVA Bonferroni post test. *p < 0.05 compared to medium alone. # p < 0.05 compared to D1ED III.
RAW 264.7 macrophage cells was analyzed by flow cytometry using anti-D1ED III sera obtained from D1ED III immunized mice. LD1ED III in PBS associated more strongly with RAW 264.7 macrophage cells than D1ED III in PBS. Combining D1ED III with PELC only increased the antigen association index slightly. Remarkably, the antigen association index was significantly increased when LD1ED III was administered with PELC. The results from five independent experiments are summarized in Fig. 4B. 3.4. Effects of vaccine formulations on the immune response We next sought to determine whether LD1ED III combined with PELC induced potent immune responses. Groups of BALB/c mice were immunized with either D1ED III or LD1ED III twice over a two-week period in combination with or without PELC. Animals that received PBS alone served as negative controls. One week after the last immunization, the spleens were removed and single cell suspensions were prepared to evaluate cell proliferation and IFN-g production in response to D1ED III stimulation. Immunization with D1ED
III combined with either PBS or PELC did not induce notable antigen-specific proliferative responses. The stimulation indices of splenocytes from these mice were 1.4 0.4 and 1.5 0.3, respectively. Significantly, the stimulation indices of splenocytes from LD1ED III/PBS- or LD1ED III/PELCimmunized mice were elevated to 2.7 0.9 and 3.3 0.5, respectively (Fig. 5A). The splenocytes obtained from mice immunized with LD1ED III/PELC produced higher levels of IFN-g (404.5 241.0 pg/mL) than the splenocytes from the mice that received other treatments. Negligible levels of IFN-g were produced by splenocytes from naı¨ve mice and mice immunized with D1ED III/PBS (Fig. 5B). The antibody responses in mice were evaluated following three immunizations administered at two-week intervals. The avidity profile of the antibodies induced in mice immunized with the various formulations was shown in Fig. 5C, immunizing mice with LD1ED III combined with PELC generated higher avidity antibodies than immunizing mice with the other formulations ( p < 0.05 by the ANOVA Bonferroni post test). Following the antibody avidity analyses, we next tested the capacity of antibodies to neutralize dengue viral infection. As shown in Table 1, the antisera obtained from mice immunized
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
725
4. Discussion
Fig. 4. Effects of formulations on antigen association with RAW 264.7 macrophage cells. RAW 264.7 macrophage cells were pulsed with either LD1ED III or D1ED III in either the presence or absence of PELC. After a 3hour incubation, the antigen association with RAW 264.7 macrophage cells was detected using D1ED immunized sera and analyzed using flow cytometry. The mean fluorescence intensity (MFI) for cells detected by naı¨ve sera served as the control. (A) The antigen association index for each formulation, which was calculated according to the formula (MFI detected by D1ED III immunized sera)/(MFI detected by naı¨ve sera), is indicated in the upper right of the figure. (B) The means and standard deviations from five independent experiments are shown. Statistical significance was determined by the ANOVA Bonferroni post test. *p < 0.05 compared to D1ED III combined with PBS. # p < 0.05 compared to D1ED III combined with PELC. xp < 0.05 compared to LD1ED III.
with D1ED III/PBS were unable to block dengue viral infection efficiently. Significantly, immunizing mice with PELC combined with LD1ED III induced higher titers of neutralizing antibody than immunizing mice with the other formulations. Moreover, the neutralizing ability was persistent and still detectable 20 weeks after the first immunization. These results provide tangible evidence that a competent neutralizing antibody response is elicited in mice vaccinated with PELC combined with LD1ED III.
In general, antigens, immunostimulators, and delivery systems are the main elements of effective vaccines. Based on this notion, we hypothesized that the combination of recombinant lipoproteins, which contain antigens and immunostimulators, with a proper delivery system would enhance the ability of vaccine candidates to induce robust immune responses. In the present study, we prepared the recombinant lipoprotein LD1ED III (Fig. 1) and demonstrated its immunostimulatory capacity to induce the up-regulation of activation markers and IL-6 production on RAW 264.7 macrophage cells (Fig. 3). Both lipoproteins and lipopeptides have been demonstrated to activate antigen-presenting cells via the tolllike receptor signaling pathway [18,29e31] and to enhance humoral and cellular immune responses [17,19,25]. Here, LD1ED III provides two key elements of an effective vaccine: the antigen and the immunostimulator. The PELC functions as the antigen delivery system. When LD1ED III was combined with PELC the antigen association with RAW 264.7 macrophage cells was significantly increased (Fig. 4). LD1ED III in combination with PELC increased the probability of antigenpresenting cells capturing and processing antigens, which led to an enhanced cellular immune response. LD1ED III combined with PELC elicited higher T cell proliferation activity and IFN-g production than any of the other vaccine formulations administered (Fig. 5A and B). CpG oligodeoxynucleotides are immunostimulators that act through toll-like receptor 9 to activate antigen-presenting cells and trigger a protective immune response that enhances the host’s capacity to eradicate the pathogens [32e34]. In our previous studies, we demonstrated that either D1ED III [22] or inactivated influenza virus [13] combined with PELC plus CpG oligodeoxynucleotides synergize to enhance the immune response. These results suggest that immunostimulators synergize with the PELC emulsified system to induce host immunity against pathogens. Consistent with the findings in this study, combining LD1ED III with PELC elicits higher titers of neutralizing antibody than the other formulations used in this study (Table 1). Another important finding is that LD1ED III, like D1ED III, can block dengue viral infection (Fig. 2). These results suggest that the lipidated form of the antigen can still assume the native conformation. The neutralizing dengue virus antibodies represent a small fraction of the total dengue virus-specific antibodies produced in infected humans. These neutralizing antibodies appear to recognize novel epitopes, including complex and quaternary structure epitopes [35,36]. Subunit vaccines that retain the proper conformation will elicit an antibody response specific to both linear and conformational epitopes. Dengue envelope protein domain III is a promising candidate to be used for the development of a subunit dengue vaccine [37]. In our previous study, we developed a consensus dengue envelope protein domain III. Formulation of consensus dengue envelope protein domain III with aluminum phosphate induced neutralizing antibodies against four serotypes of dengue virus in mice [38], but only neutralizing antibodies
726
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
Fig. 5. Effects of formulations on cellular and humoral immune responses. BALB/c mice (n ¼ 3e4) were immunized subcutaneously with 10 mg antigens twice over a two-week interval in different formulations as indicated. One week after the last immunization, splenocytes were incubated with or without D1ED III (1 mg/ mL). (A) The stimulation index is defined as the ratio of the mean counts per minute (cpm) of D1ED III stimulation to the mean cpm without antigen stimulation. (B) The levels of IFN-g production were determined by ELISA assays. The means and standard deviations are shown. Statistical significance was determined by the ANOVA Bonferroni post test. *p < 0.05 compared to naive. #p < 0.05 compared to D1ED III combined with PBS. xp < 0.05 compared to D1ED III combined with PELC. yp < 0.05 compared to LD1ED III combined with PBS. (C) BALB/c mice (n ¼ 5) were immunized subcutaneously with 10 mg antigens three times at a two-week intervals in different formulations as indicated. The sera were collected at the indicated time points after the first immunization. Antibody avidity profiles were evaluated by ELISA. Statistical significance was determined by the ANOVA Bonferroni post test. *p < 0.05 compared to D1ED III combined with PBS. # p < 0.05 compared to D1ED III combined with PELC. xp < 0.05 compared to LD1ED III combined with PBS.
against serotype 2 were detected in monkey [39]. In addition, formulations of other dengue subunit vaccines with aluminum were unable to induce complete protection against dengue virus infection [40e44]. These results suggest that aluminum, the most widely used adjuvant in human vaccines, may not be suitable for dengue subunit vaccines. Therefore, it requires developing an adjuvant-free technology and/or an adjuvant/ delivery system for dengue preparedness. This is the reason why we try to explore a new system for dengue subunit vaccines. In the present study, we demonstrated that LD1ED III, a lipidated antigen without exogenous adjuvant, can induce long-lasting neutralizing antibody responses. Emulsified LD1ED III using PELC can further enhance the immune responses. There is a great potential use of PELC in human in the future, because all the components of PELC are bioresorbable [12,20]. All together, lipidated antigens and PELC
may have great benefits in the development of dengue subunit vaccines. The avidity of the antibody induced by LD1ED III combined with PELC was higher than that induced by the other formulations (Fig. 5C). These results imply that LD1ED III combined with PELC can quickly induce high avidity antibody responses. Remarkably, the neutralizing antibody titer induced by LD1ED III combined with PELC was significantly higher than that induced by the other formulations (Table 1). These results suggest that LD1ED III combined with PELC enhances the intensity and modulates the quality of the immune response, thus potentiating the host response against dengue viral infection. Importantly, the application of recombinant lipoprotein and PELC demonstrated herein could provide a novel strategy for the development of the other subunit vaccines.
Table 1 Neutralizing antibody titers in mice immunized with various formulations.a Time
Groups
8 weeks
D1ED III/PBS D1ED III/PELC LD1ED III/PBS LD1ED III/PELC D1ED III/PBS D1ED III/PELC LD1ED III/PBS LD1ED III/PELC
20 weeks
FRNT50
FRNT90
Mouse 1
2
3
4
5
1
2
3
4
5
<8 8 32 64 <8 8 32 32
<8 16 32 128 8 16 32 64
8 16 32 128 8 16 32 64
8 32 32 256 8 32 64 128
8 32*b 32* 256*#x 8 32 128* 128*#
<8 8 <8 8 <8 <8 8 8
<8 8 8 16 <8 <8 8 16
<8 8 8 16 <8 8 16 16
<8 8 8 32 <8 8 16 16
<8 8* 8 32*#x <8 8 16*# 32*#
Mice were immunized with various formulations three times (10 mg/dose) at two-week intervals. The serum samples were collected at 8 and 20 weeks after the first immunization. The dengue-1 virus neutralizing capacity was determined by FRNT. The neutralizing antibody titer was defined as the reciprocal of the highest dilution that resulted in a 50% or 90% reduction of FFU compared to control samples containing the virus alone. b Statistical analysis was performed using the ANOVA Bonferroni post test at the same time point. *p < 0.05 compared to D1ED III/PBS. #p < 0.05 compared to D1ED III/PELC. xp < 0.05 compared to LD1ED III/PBS. a
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
Acknowledgments This work was supported by grants 100A1-VC-PP-07-014 and 101A1-IV-PP-26-014 from the National Health Research Institutes. References [1] M.R. Hilleman, Vaccines in historic evolution and perspective: a narrative of vaccine discoveries, Vaccine 18 (2000) 1436e1447. [2] D.M. Pardoll, Spinning molecular immunology into successful immunotherapy, Nat. Rev. Immunol. 2 (2002) 227e238. [3] S.G. Reed, S. Bertholet, R.N. Coler, M. Friede, New horizons in adjuvants for vaccine development, Trends Immunol. 30 (2009) 23e32. [4] G.M. Barton, R. Medzhitov, Toll-like receptors and their ligands, Curr. Top. Microbiol. Immunol. 270 (2002) 81e92. [5] A. Iwasaki, R. Medzhitov, Toll-like receptor control of the adaptive immune responses, Nat. Immunol. 5 (2004) 987e995. [6] H. Kato, K. Takahasi, T. Fujita, RIG-I-like receptors: cytoplasmic sensors for non-self RNA, Immunol. Rev. 243 (2011) 91e98. [7] M. Lamkanfi, T.D. Kanneganti, Regulation of immune pathways by the NOD-like receptor NLRC5, Immunobiology 217 (2012) 13e16. [8] A. Williams, R.A. Flavell, S.C. Eisenbarth, The role of NOD-like receptors in shaping adaptive immunity, Curr. Opin. Immunol. 22 (2010) 34e40. [9] E. Erikci, M. Gursel, I. Gursel, Differential immune activation following encapsulation of immunostimulatory CpG oligodeoxynucleotide in nanoliposomes, Biomaterials 32 (2011) 1715e1723. [10] J.S. Thomann, B. Heurtault, S. Weidner, M. Braye, J. Beyrath, S. Fournel, F. Schuber, B. Frisch, Antitumor activity of liposomal ErbB2/ HER2 epitope peptide-based vaccine constructs incorporating TLR agonists and mannose receptor targeting, Biomaterials 32 (2011) 4574e4583. [11] A. Pashine, N.M. Valiante, J.B. Ulmer, Targeting the innate immune response with improved vaccine adjuvants, Nat. Med. 11 (2005) S63eS68. [12] M.H. Huang, C.Y. Huang, S.P. Lien, S.Y. Siao, A.H. Chou, H.W. Chen, S.J. Liu, C.H. Leng, P. Chong, Development of multi-phase emulsions based on bioresorbable polymers and oily adjuvant, Pharm. Res. 26 (2009) 1856e1862. [13] M.H. Huang, S.C. Lin, C.H. Hsiao, H.J. Chao, H.R. Yang, C.C. Liao, P.W. Chuang, H.P. Wu, C.Y. Huang, C.H. Leng, S.J. Liu, H.W. Chen, A.H. Chou, A.Y. Hu, P. Chong, Emulsified nanoparticles containing inactivated influenza virus and CpG oligodeoxynucleotides critically influences the host immune responses in mice, PLoS One 5 (2010) e12279. [14] Z. Zhang, S. Tongchusak, Y. Mizukami, Y.J. Kang, T. Ioji, M. Touma, B. Reinhold, D.B. Keskin, E.L. Reinherz, T. Sasada, Induction of antitumor cytotoxic T cell responses through PLGA-nanoparticle mediated antigen delivery, Biomaterials 32 (2011) 3666e3678. [15] M.G. Cusi, Applications of influenza virosomes as a delivery system, Hum. Vaccines 2 (2006) 1e7. [16] M.J. Heffernan, D.A. Zaharoff, J.K. Fallon, J. Schlom, J.W. Greiner, In vivo efficacy of a chitosan/IL-12 adjuvant system for protein-based vaccines, Biomaterials 32 (2011) 926e932. [17] H.W. Chen, S.J. Liu, H.H. Liu, Y. Kwok, C.L. Lin, L.H. Lin, M.Y. Chen, J.P. Tsai, L.S. Chang, F.F. Chiu, L.W. Lai, W.C. Lian, C.Y. Yang, S.Y. Hsieh, P. Chong, C.H. Leng, A novel technology for the production of a heterologous lipoprotein immunogen in high yield has implications for the field of vaccine design, Vaccine 27 (2009) 1400e1409. [18] C.H. Leng, H.W. Chen, L.S. Chang, H.H. Liu, H.Y. Liu, Y.P. Sher, Y.W. Chang, S.P. Lien, T.Y. Huang, M.Y. Chen, A.H. Chou, P. Chong, S.J. Liu, A recombinant lipoprotein containing an unsaturated fatty acid activates NF-kappaB through the TLR2 signaling pathway and induces a differential gene profile from a synthetic lipopeptide, Mol. Immunol. 47 (2010) 2015e2021.
727
[19] C.Y. Chiang, S.J. Liu, J.P. Tsai, Y.S. Li, M.Y. Chen, H.H. Liu, P. Chong, C.H. Leng, H.W. Chen, A novel single-dose dengue subunit vaccine induces memory immune responses, PLoS One 6 (2011) e23319. [20] M.H. Huang, A.H. Chou, S.P. Lien, H.W. Chen, C.Y. Huang, W.W. Chen, P. Chong, S.J. Liu, C.H. Leng, Formulation and immunological evaluation of novel vaccine delivery systems based on bioresorbable poly(ethylene glycol)-block-poly(lactide-co-epsilon-caprolactone), J. Biomed. Mater. Res. B Appl. Biomater. 90 (2009) 832e841. [21] M.H. Huang, C.Y. Huang, S.C. Lin, J.H. Chen, C.C. Ku, A.H. Chou, S.J. Liu, H.W. Chen, P. Chong, C.H. Leng, Enhancement of potent antibody and T-cell responses by a single-dose, novel nanoemulsion-formulated pandemic influenza vaccine, Microbes Infect. 11 (2009) 654e660. [22] C.Y. Chiang, M.H. Huang, C.H. Hsieh, M.Y. Chen, H.H. Liu, J.P. Tsai, Y.S. Li, C.Y. Chang, S.J. Liu, P. Chong, C.H. Leng, H.W. Chen, Dengue1 envelope protein domain III along with PELC and CpG oligodeoxynucleotides synergistically enhances immune responses, PLoS Negl. Trop. Dis. 6 (2012) e1645. [23] R.A. Macdonald, C.S. Hosking, C.L. Jones, The measurement of relative antibody affinity by ELISA using thiocyanate elution, J. Immunol. Methods 106 (1988) 191e194. [24] N. Nair, H. Gans, L. Lew-Yasukawa, A.C. Long-Wagar, A. Arvin, D.E. Griffin, Age-dependent differences in IgG isotype and avidity induced by measles vaccine received during the first year of life, J. Infect. Dis. 196 (2007) 1339e1345. [25] C.Y. Huang, J.J. Chen, K.Y. Shen, L.S. Chang, Y.C. Yeh, I.H. Chen, P. Chong, S.J. Liu, C.H. Leng, Recombinant lipidated HPV E7 induces a Th-1-biased immune response and protective immunity against cervical cancer in a mouse model, PLoS One 7 (2012) e40970. [26] Y. Kwok, W.C. Sung, A.L. Lin, H.H. Liu, F.A. Chou, S.S. Hsieh, C.H. Leng, P. Chong, Rapid isolation and characterization of bacterial lipopeptides using liquid chromatography and mass spectrometry analysis, Proteomics 11 (2011) 2620e2627. [27] Y. Chen, T. Maguire, R.E. Hileman, J.R. Fromm, J.D. Esko, R.J. Linhardt, R.M. Marks, Dengue virus infectivity depends on envelope protein binding to target cell heparan sulfate, Nat. Med. 3 (1997) 866e871. [28] J.-J. Hung, M.-T. Hsieh, M.-J. Young, C.-L. Kao, C.-C. King, W. Chang, An external loop region of domain III of dengue virus type 2 envelope protein is involved in serotype-specific binding to mosquito but not mammalian cells, J. Virol. 78 (2004) 378e388. [29] T. Shimizu, Y. Kida, K. Kuwano, A triacylated lipoprotein from Mycoplasma genitalium activates NF-kappaB through toll-like receptor 1 (TLR1) and TLR2, Infect. Immun. 76 (2008) 3672e3678. [30] K. Tawaratsumida, M. Furuyashiki, M. Katsumoto, Y. Fujimoto, K. Fukase, Y. Suda, M. Hashimoto, Characterization of N-terminal structure of TLR2-activating lipoprotein in Staphylococcus aureus, J. Biol. Chem. 284 (2009) 9147e9152. [31] S. Thakran, H. Li, C.L. Lavine, M.A. Miller, J.E. Bina, X.R. Bina, F. Re, Identification of Francisella tularensis lipoproteins that stimulate the toll-like receptor (TLR) 2/TLR1 heterodimer, J. Biol. Chem. 283 (2008) 3751e3760. [32] S. Bauer, C.J. Kirschning, H. Hacker, V. Redecke, S. Hausmann, S. Akira, H. Wagner, G.B. Lipford, Human TLR9 confers responsiveness to bacterial DNA via species-specific CpG motif recognition, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 9237e9242. [33] H. Hemmi, O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, K. Hoshino, H. Wagner, K. Takeda, S. Akira, A toll-like receptor recognizes bacterial DNA, Nature 408 (2000) 740e745. [34] F. Takeshita, C.A. Leifer, I. Gursel, K.J. Ishii, S. Takeshita, M. Gursel, D.M. Klinman, Cutting edge: role of toll-like receptor 9 in CpG DNAinduced activation of human cells, J. Immunol. 167 (2001) 3555e3558. [35] R. de Alwis, S.A. Smith, N.P. Olivarez, W.B. Messer, J.P. Huynh, W.M. Wahala, L.J. White, M.S. Diamond, R.S. Baric, J.E. Crowe Jr., A.M. de Silva, Identification of human neutralizing antibodies that bind to complex epitopes on dengue virions, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 7439e7444. [36] W.M. Wahala, A.M. Silva, The human antibody response to dengue virus infection, Viruses 3 (2011) 2374e2395.
728
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
[37] M.G. Guzman, L. Hermida, L. Bernardo, R. Ramirez, G. Guillen, Domain III of the envelope protein as a dengue vaccine target, Expert Rev. Vaccines 9 (2010) 137e147. [38] C.H. Leng, S.J. Liu, J.P. Tsai, Y.S. Li, M.Y. Chen, H.H. Liu, S.P. Lien, A. Yueh, K.N. Hsiao, L.W. Lai, F.C. Liu, P. Chong, H.W. Chen, A novel dengue vaccine candidate that induces cross-neutralizing antibodies and memory immunity, Microbes Infect. 11 (2009) 288e295. [39] H.W. Chen, S.J. Liu, Y.S. Li, H.H. Liu, J.P. Tsai, C.Y. Chiang, M.Y. Chen, C.S. Hwang, C.C. Huang, H.M. Hu, H.H. Chung, S.H. Wu, P. Chong, C.H. Leng, C.H. Pan, A consensus envelope protein domain III can induce neutralizing antibody responses against serotype 2 of dengue virus in non-human primates, Arch. Virol. (2013). [40] L. Bernardo, A. Izquierdo, M. Alvarez, D. Rosario, I. Prado, C. Lopez, R. Martinez, J. Castro, E. Santana, L. Hermida, G. Guillen, M.G. Guzman, Immunogenicity and protective efficacy of a recombinant fusion protein containing the domain III of the dengue 1 envelope protein in non-human primates, Antivir. Res. 80 (2008) 194e199. [41] M.G. Guzman, R. Rodriguez, L. Hermida, M. Alvarez, L. Lazo, M. Mune, D. Rosario, K. Valdes, S. Vazquez, R. Martinez, T. Serrano,
J. Paez, R. Espinosa, T. Pumariega, G. Guillen, Induction of neutralizing antibodies and partial protection from viral challenge in Macaca fascicularis immunized with recombinant dengue 4 virus envelope glycoprotein expressed in Pichia pastoris, Am. J. Trop. Med. Hyg. 69 (2003) 129e134. [42] J. Robert Putnak, B.A. Coller, G. Voss, D.W. Vaughn, D. Clements, I. Peters, G. Bignami, H.S. Houng, R.C. Chen, D.A. Barvir, J. Seriwatana, S. Cayphas, N. Garcon, D. Gheysen, N. Kanesa-Thasan, M. McDonell, T. Humphreys, K.H. Eckels, J.P. Prieels, B.L. Innis, An evaluation of dengue type-2 inactivated, recombinant subunit, and liveattenuated vaccine candidates in the rhesus macaque model, Vaccine 23 (2005) 4442e4452. [43] M. Simmons, K.R. Porter, C.G. Hayes, D.W. Vaughn, R. Putnak, Characterization of antibody responses to combinations of a dengue virus type 2 DNA vaccine and two dengue virus type 2 protein vaccines in rhesus macaques, J. Virol. 80 (2006) 9577e9585. [44] J. Velzing, J. Groen, M.T. Drouet, G. van Amerongen, C. Copra, A.D. Osterhaus, V. Deubel, Induction of protective immunity against dengue virus type 2: comparison of candidate live attenuated and recombinant vaccines, Vaccine 17 (1999) 1312e1320.
Original article
Induction of robust immunity by the emulsification of recombinant lipidated dengue-1 envelope protein domain III Chen-Yi Chiang a, Ming-Hsi Huang a, Chien-Hsiung Pan a,b, Chun-Hsiang Hsieh a, Mei-Yu Chen a, Hsueh-Hung Liu a, Jy-Ping Tsai a, Shih-Jen Liu a,b, Pele Chong a,b, Chih-Hsiang Leng a,b,**, Hsin-Wei Chen a,b,* a
National Institute of Infectious Diseases and Vaccinology, National Health Research Institutes, Miaoli, Taiwan, ROC b Graduate Institute of Immunology, China Medical University, Taichung, Taiwan, ROC Received 20 November 2012; accepted 5 June 2013 Available online 15 June 2013
Abstract Many attempts have focused on the use of either immunomodulators or antigen delivery systems to obtain an efficacious vaccine. Here, we report a novel approach that combined an immunomodulator and delivery system to enhance antigen association and induce robust immunity. We expressed a recombinant lipidated dengue-1 envelope protein domain III (LD1ED III) and its non-lipidated form, D1ED III, in an Escherichia coli system. The LD1ED III contains a bacterial lipid moiety, which is a potent immunomodulator. We demonstrated that LD1ED III possesses an inherent immunostimulation ability that can activate RAW 264.7 macrophage cells by up-regulating their expression of CD40, CD80, CD83, CD86 and MHC II, whereas D1ED III could not induce the up-regulation of these molecules. Moreover, combining LD1ED III with a multiphase emulsion system (called PELC) increased the antigen association more than either combining D1ED III with PELC or the antigen alone. Enhanced antigen association has been shown to correlate with stronger T cell responses, greater antibody avidity and improved neutralizing capacity. Our results demonstrate that combining recombinant lipoproteins with PELC improved both the intensity and the quality of the immune response. This approach is a promising strategy for the development of subunit vaccines that induce robust immunity. Ó 2013 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved. Keywords: Dengue; Immune response; Immunostimulation; Lipoprotein
1. Introduction Vaccines are recognized as one of the most effective and successful strategies to combat infectious diseases [1]. Recombinant protein vaccines may have significant advantages over either live attenuated or inactivated vaccines.
* Corresponding author. National Institute of Infectious Diseases and Vaccinology, National Health Research Institutes, Miaoli, Taiwan, ROC. Tel.: þ886 37 246 166x37706; fax: þ886 37 583 009. ** Corresponding author. National Institute of Infectious Diseases and Vaccinology, National Health Research Institutes, Miaoli, Taiwan, ROC. Tel.: þ886 37 246 166x37711; fax: þ886 37 583 009. E-mail addresses: [email protected] (C.-H. Leng), [email protected] (H.-W. Chen).
Recombinant subunit vaccines are safer and easier to produce and administer than conventional vaccines because they do not contain live pathogens. However, subunit vaccines are usually poorly immunogenic and require an adjuvant to induce robust immune responses [2,3]. Therefore, combining recombinant protein antigens with effective immunostimulators and/or delivery systems are proposed to greatly improve the ability of subunit vaccines to elicit robust immune responses. Several pattern-recognition receptors (PRR), such as toll-like receptors [4,5], retinoic acid-inducible gene-I-like receptors [6], and nucleotide-binding and oligomerization domain-like receptors [7,8], were identified and shown to be involved in the recognition of pathogen-associated molecular patterns. Because engaging the PRR ligands can activate the antigen-presenting cells to trigger innate and adaptive immunity, the ability of
1286-4579/$ - see front matter Ó 2013 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.micinf.2013.06.002
720
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
PRR ligands to act as immunostimulators in vaccine formulations has been studied for decades [9e11]. The choice of delivery system can also improve the host’s antigen-specific immune response and can include emulsions [12e14], liposomes [9,10], virosomes [15], and biocompatible materials [16]. A well-informed and sensible combination of immunostimulators and an appropriate delivery system will contribute to the development of new vaccines. To overcome the inherent disadvantages of subunit vaccines, we developed a novel platform technology to express high levels of recombinant lipoproteins with built-in adjuvant abilities [17]. The lipid moiety of the recombinant lipoprotein can act as an immunostimulator for the vaccine candidates. We demonstrated that the novel recombinant lipoproteins could activate the antigen-presenting cells via toll-like receptor 2 [18] and enhance immune responses [19]. In parallel, we developed a water-in-oil-in-water multiphase emulsion system, called PELC, for vaccine delivery. The combination of inactivated H5N1 virus or ovalbumin with PELC can enhance immune responses [12,20,21]. In the present study, recombinant lipidated dengue-1 envelope protein domain III (LD1ED III) has been used as a model protein antigen for PELC formulation. Our objective in this study was to determine whether the combination of LD1ED III, which contains a ligand for toll-like receptor 2, and an emulsion system could synergize to activate the cellular and humoral responses. 2. Materials and methods 2.1. Mouse experiments and virus BALB/c mice were purchased from the National Laboratory Animal Breeding and Research Center, Taipei, Taiwan. All the animal studies were approved by the Animal Committee of the National Health Research Institutes (Protocol No: NHRI-IACUC-098014) and were performed according to their guidelines. Groups of BALB/c mice (6e8 weeks of age) were immunized subcutaneously with either LD1ED III or recombinant dengue-1 envelope protein domain III (D1ED III) (10 mg per dose) combined with PELC or phosphate buffered saline (PBS). Mice were given one or two boosting vaccinations at two-week intervals using the same regimen. Blood samples were collected from each mouse at the indicated time points. The sera were prepared and stored at 20 C until use. The dengue-1 virus (Hawaii) was used for this study. The virus was propagated in C6/36 cells, and viral titers were determined by focus-forming assays using BHK-21 cells. Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Thermal Scientific HyClone, Logan, Utah) supplemented with 1 mM sodium pyruvate (Gibco Invitrogen, Carlsbad, CA) and 5% fetal bovine serum (Thermal Scientific HyClone). 2.2. Cloning and expression of recombinant proteins The recombinant D1ED III was cloned and expressed as previously described [22]. Regarding LD1ED III, the D1ED
III gene was cloned into the pLipo plasmid [17] using the Bam HI and Xho I sites to produce the pLD1ED III plasmid. The Cterminus of the recombinant protein contained an additional hexahistidine tag (HisTag). The Escherichia coli strain C43(DE3) (Lucigen, Middleton, WI) was transformed with the expression plasmid, pLD1ED III, for LD1ED III expression. The transformed cells were cultured at 37 C overnight. The overnight culture was scaled up in a shaking flask and incubated at 37 C for 4 h before induction. Protein expression was induced when the cultures reached an OD600 of 0.8 by adding 1 mM isopropylthiogalactoside (IPTG), and the cultures were then incubated at 20 C for 20 h. 2.3. Purification of recombinant proteins The purification of recombinant D1ED III was described previously [22]. The recombinant LD1ED III was purified by disrupting the harvested cells in a French Press (Constant Systems, Daventry, UK) at 27 kpsi in homogenization buffer [20 mM Tris (pH8.0), 50 mM sucrose, 500 mM NaCl and 10% glycerol] and stirring at 4 C for 2 h after adding 0.5% triton X-100 to the supernatant. The cell lysate was clarified by centrifugation (80,000 g for 40 min). Most of the LD1ED III was in the soluble fraction. The LD1ED III was purified using immobilized metal affinity chromatography (IMAC) columns. The purified LD1ED III was eluted from the IMAC column with homogenization buffer containing 500 mM imidazole. An Endotoxin Removing Gel (Pierce, Rockford, IL, USA) was used to remove the lipopolysaccharide (LPS). The LPS levels of the purified LD1ED III were determined by the Limulus amebocyte lysate assay (Associates of Cape Cod, Inc. Cape Cod, MA), and the resulting LPS levels were less than 0.04 EU/mg. After the LD1ED III was dialyzed against 25 mM ammonium bicarbonate/3 mg/mL sucrose, the LD1ED III was lyophilized and stored at 20 C. The fractions from each step were analyzed by SDS-PAGE gel stained with Coomassie blue (Coomassie Brilliant Blue R-250) and immunoblotted with anti-HisTag antibodies. 2.4. Identification of the lipid moiety in LD1ED III The LD1ED III was digested with trypsin (Sigma, St. Louis, MO). After the digestion, the reaction mixture was further polished using Ziptip (Millipore, Massachusetts). One microliter of the polished tryptic fragments was mixed with 1 mL of a saturated solution of a-ciano-4-hydroxycinnamic acid in acetonitrile/0.1% trifluoroacetic acid (1:3, vol/vol). One microliter of the mixture was placed on the target plate of a MALDI micro MX mass spectrometer (Waters, Manchester, UK) for analysis. 2.5. Preparation of PELC PELC is a squalene water/oil/water nanoemulsion stabilized by SpanÒ85 (sorbitan trioleate, SigmaeAldrich, Steinheim, Germany) and diblock tri-component copolymer poly(ethylene glyco)-block-poly(lactide-co-ε-caprolactone) (PEG-b-PLACL),
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
the latter consisting of 75 wt-% of hydrophilic bioabsorbable PEG and 25 wt-% of lipophilic biodegradable PLACL with a molecular weight of 7000 Da, as previously described [22]. Briefly, 120 mg of PEG-b-PLACL, 0.8 mL of PBS, and 1.1 mL of an oily solution consisting of squalene (SigmaeAldrich, Steinheim, Germany) and SpanÒ85 (85/15 v/v) were emulsified using a PolytronÒPT 3100 homogenizer (Kinematica AG, Switzerland) at 6000 rpm for 5 min. The emulsified PELC formulation was stored at 4 C until use. The PELC-formulated vaccine was prepared by re-dispersing 0.2 mL of the stock emulsion into 1.8 mL of the bulk vaccine candidates and mixing with a test-tube rotator (Labinco LD-79, Netherlands) at 5 rpm for at least 1 h before injection. 2.6. Assays for the activation of RAW 264.7 macrophage cells RAW 264.7 macrophage cells were cultured in DMEM supplemented with (Thermal Scientific HyClone) 1 mM sodium pyruvate (Gibco Invitrogen) and 10% fetal bovine serum (Thermal Scientific HyClone). Cells were stimulated with 10 mg/mL of either LD1ED III or D1ED III. The cells cultured in media alone served as controls. After 16 h stimulation, cells were harvested for cell surface marker staining using either PEor FITC-conjugated antibodies specific for CD40, CD80, CD83, CD86 and MHC class II. The data were acquired using CellQuest Pro software on a BD FACSCalibur flow cytometer and were analyzed using FACS 3 software. The basal expression level was defined as the mean fluorescence intensity (MFI) of cells cultured in media alone. All antibodies were purchased from either BD Biosciences (San Diego, CA) or eBioscience (San Diego, CA). The production of IL-6 by RAW 264.7 macrophage cells was determined by ELISA detection set follow the manufacturer’s instructions (R&D systems). 2.7. Assays for antigen association with RAW 264.7 macrophage cells LD1ED III or D1ED III was combined with either PELC or PBS. RAW 264.7 macrophage cells were pulsed with different formulations of antigens at 37 C for 3 h. The cells were fixed and permeabilized using a cytofix/cytoperm kit (eBioscience) according to the manufacturer’s instructions. The antigen association with RAW 264.7 macrophage cells was detected using D1ED III immunized sera. The bound antibodies were detected using a secondary antibody conjugated to FITC then analyzed using flow cytometry as described above. The antigen association index was calculated according to the formula (MFI detected by D1ED III immunized sera)/(MFI detected by naı¨ve sera). 2.8. Blocking of dengue virus infection in BHK-21 cells by LD1ED III To test whether LD1ED III blocks the dengue virus infection of BHK-21 cells, the virus was pre-mixed with different amounts of LD1ED III, D1ED III or control bovine serum albumin (BSA) protein as indicated for 10 min at 4 C. The
721
virus titer prior to pre-mixing was approximately 20e40 focus-forming units (FFUs) per well. Viral adsorption was allowed to proceed for 3 h at 37 C. An overlay medium containing 2% fetal bovine serum and 0.8% methylcellulose in DMEM was added at the conclusion of adsorption. The infected monolayer was incubated at 37 C. After 72 h of infection, the overlay medium was removed from the wells, and the BHK cells were washed with cold PBS. The cells were fixed for 15 min in 3.7% formaldehyde/PBS. After washing with PBS, the cells were permeabilized with 0.1% Nonidet P40/PBS for 15 min and blocked with 3% BSA/PBS for 30 min. The infected cells were detected using a monoclonal anti-dengue antibody (American Type Culture Collection, No. HB-114). After washing with PBS, the antibody-labeled cells were detected using a secondary antibody conjugated to HRP. The labeling was visualized using 3,30 ,5,50 -tetramethylbenzidine (TMB). The FFUs were counted. 2.9. Focus reduction neutralization tests (FRNT) Two-fold serial dilutions of sera were made (starting at 1:8), and the sera were heat inactivated prior to testing. A monolayer of BHK-21 cells in 24-well plates was inoculated with dengue-1 virus that had been pre-mixed at 4 C overnight with pre-immunization or post-immunization sera to a final volume of 0.2 mL. The FFUs were determined as described above. The neutralizing antibody titer FRNT50 (or FRNT90) was calculated as the reciprocal of the highest dilution that produced a 50% (or 90%) reduction of FFU compared with control samples containing the virus alone. For calculation purpose, the neutralizing antibody titer was designated as 22 when neutralizing antibody titer was less than 23. 2.10. Assays for lymphocyte proliferation and cytokine secretion The spleens were removed one week after the last immunization and processed into a single-cell suspension. The splenocytes were seeded at a concentration of 2 105 cells/ well in 96-well plates and were then stimulated with D1ED III (1 mg/mL) for 3 days at 37 C in a 5% CO2 humidified incubator. The growth medium of splenocytes was Roswell Park Memorial Institute medium supplemented with 50 mM beta-mercaptoethanol (SigmaeAldrich, St. Louis, MO), 20 mM HEPES (Biological Industries, Israel), and 5% fetal bovine serum (Thermal Scientific HyClone). During the final 18 h of culture, 1 mCi of [3H]-thymidine was added to each well, and the cells were harvested using a FilterMate automatic cell harvester (Packard). The incorporated radioactivity was determined with a TopCount microplate scintillation counter (Packard). Con A (5 mg/mL) was added to some wells as a positive control for cell proliferation. The induced stimulation index is defined as the ratio of the mean counts per minute (cpm) with D1ED III stimulation to the mean cpm without D1ED III stimulation. To measure the secretion of IFN-g, the splenocytes were cultured and pulsed with D1ED III as described above. The
722
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
Fig. 1. Preparation and characterization of recombinant lipidated dengue-1 envelope protein domain III. (A) The amino acid sequence of LD1ED III is a consensus sequence of dengue virus type I (Ref: PLoS Negl. Trop. Dis. 6 (2012) e1645). The DNA sequence of the D1ED III gene was derived using the codon usage of E. coli and fully synthesized using the assembly PCR method. The PCR product was cloned into the pET22b-based vector with a lipid signal peptide in front of D1ED III gene for LD1ED III expression. The recombinant lipoproteins contained an additional HHHHHH sequence (HisTag) at the C-terminus and were expressed under the control of the T7 promoter. (B) The LD1ED III protein purification process was monitored using a 15% reducing SDS-PAGE followed by staining with Coomassie blue (left panel) and immunoblotting (right panel) using anti-HisTag antibodies. The LD1ED III was expressed in E. coli strains C43 (DE3). Lane 1, LD1ED III expression after IPTG induction; lane 2, protein expression in the absence of IPTG induction; lane 3, soluble fraction of LD1ED III; lane 4, purified LD1ED III. Lanes 5e8 show immunoblotting to monitor the LD1ED III induction and purification processes, and the samples in these lanes are the same as those in lanes 1e4. The arrows indicate the electrophoretic positions of LD1ED III in the gels or blots. (C) N-terminal LD1ED III fragments were obtained and identified after digestion of LD1ED III with trypsin. The digested sample was analyzed on a WatersÒ MALDI micro MXÔ mass spectrometer. The MALDI-TOF MS spectra revealed the existence of five peaks with m/z values of 1452, 1466, 1480, 1492 and 1506.
cells were cultured for 4 days, and then cell-free supernatants were harvested and stored at 80 C until use. The quantification of total IFN-g was performed using IFN-g ELISA kits (R&D systems) according to the manufacturer’s instructions, and these data were expressed in pg/mL.
2.11. Measurement of antibody avidity The antibody avidity was determined on the basis of D1ED III-specific IgG dissociation due to the chaotropic agent ammonium thiocyanate. Briefly, purified D1ED III was coated
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
723
onto 96-well plates. After blocking with 1% BSA/PBS, serum at either 1:100 or 1:300 were incubated at room temperature for 1 h. The plates were washed and incubated with 0e3.5 M ammonium thiocyanate in 0.5 M increments at room temperature for 15 min. The bound IgG was detected with HRPconjugated goat anti-mouse IgG Fc. After the addition of TMB, the absorbance at 450 nm was measured with an ELISA reader. The avidity index was calculated as the concentration of ammonium thiocyanate that resulted in a 50% reduction from the initial absorbance [23,24]. 2.12. Statistical analyses The statistical analyses were carried using GraphPad Prism software version 5.02 (GraphPad Software, Inc.). The data were processed using the ANOVA Bonferroni post test. Differences with a p value of less than 0.05 were considered to be statistically significant. 3. Results 3.1. Biochemical characterization of recombinant lipidated dengue-1 envelope protein domain III The D1ED III gene was cloned into the Bam HI and Xho I sites of the pET-22b-based plasmid containing a lipid signal peptide to produce the pLD1ED III plasmid. The recombinant lipoprotein contained an additional HHHHHH sequence (HisTag) at its C-terminus and was expressed under the control of the T7 promoter (Fig. 1A). The recombinant LD1ED III was purified using immobilized metal affinity chromatography (IMAC) columns (Fig. 1B, lanes 1e4). LD1ED III was detected with anti-HisTag antibodies (Fig. 1B, lanes 5e8). After the LPS was removed (less than 0.04 EU/mg), the purified LD1ED III antigens were analyzed for their immunogenicity and efficacy in animal models. We then measured the exact mass of the N-terminal fragments of the recombinant LD1ED III. Five peaks with m/z values of 1452, 1466, 1480, 1492, and 1506 were identified (Fig. 1C). These peaks have been identified as a lipidation signature in other lipidated proteins [17,25]. In our previous study [26], we identified two groups of N-terminal lipidated (diacyl or triacyl) molecules in the bacterial expressed recombinant lipoproteins. We confirmed that the peaks of LD1ED III were associated with lipidated cysteine residues and verified that LD1ED III contains a bacterial lipid moiety at its N-terminus. 3.2. Functional evaluation of recombinant lipidated dengue-1 envelope protein domain III
Fig. 2. Inhibition of dengue viral infection by LD1ED III and D1ED III. Dengue-1 virus was pre-mixed with different amounts of LD1ED III, D1ED III or control BSA protein as indicated for 10 min at 4 C. Viral adsorption on a monolayer of BHK-21 cells in 24-well plates proceeded for 3 h at 37 C. The focus-forming units (FFUs) were determined 3 days after the viral infection. The results shown are pooled from two independent experiments.
[22]. The focus number was inhibited greater than 90% when LD1ED III added to the cells at a concentration of 0.5 mg/mL. In contrast to LD1ED III, BSA did not inhibit dengue-1 focus formation even at 3-fold higher concentrations of 1.5 mg/mL. These results suggest that recombinant LD1ED III retains the correct conformation and can block the cellular binding sites of the dengue-1 virus. Our results [18] and others [29e31] have demonstrated that lipoproteins are able to stimulate antigen-presenting cells via the toll-like receptors. RAW 264.7 macrophage cells were pulsed with either D1ED III or LD1ED III at 10 mg/mL for 16 h to examine the functional activity of LD1ED III. The expression of the activation markers CD40, CD80, CD83, CD86, and MHC II were analyzed by the flow cytometry (Fig. 3A). The basal expression level was defined as the MFI of cells cultured in media alone. The relative MFIs from three independent experiments are shown in Fig. 3B. LD1ED III was able to significantly increase the expression of CD40, CD80, CD83, CD86, and MHC II on macrophages from 1.7to 3.9-fold. Consistent with the expression of the activation markers, LD1ED III was also able to significantly stimulate the production of IL-6 by macrophages (Fig. 3C). Interestingly, D1ED III without lipidation did not induce the upregulation of activation markers and IL-6 production on macrophages. 3.3. Effects of formulations on antigen association
It has been demonstrated that dengue envelope protein domain III is involved in cellular receptor binding [27,28]. We evaluated the ability of LD1ED III to interfere with dengue viral infections. As shown in Fig. 2, the ability of dengue-1 virus to infect BHK-21 cells was inhibited in the presence of LD1ED III in a dose-dependent manner. The inhibition effect of LD1ED III is equivalent to non-lipidated D1ED III
We previously developed a water-in-oil-in-water multiphase emulsion system, PELC, for vaccine delivery [12,20]. D1ED III and LD1ED III were combined with either PELC or PBS to evaluate the effects of PELC on recombinant lipoproteins. RAW 264.7 macrophage cells were pulsed with various formulations for 3 h. The antigen associated with
724
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
Fig. 3. Effects of LD1ED III on the activation of RAW 264.7 macrophage cells. RAW 264.7 macrophage cells were cultured either in media alone or in media supplemented with LD1ED III or D1ED III at 10 mg/mL. After incubation for 16 h. (A) The surface markers CD40, CD80, CD83, CD86, and MHC II were analyzed using flow cytometry. A representative experiment is shown. (B) The basal expression level was defined as the mean fluorescence intensity (MFI) of cells cultured in media alone. Relative MFIs were plotted. (C) The supernatants were harvested and analyzed for IL-6 using ELISA. The means and standard deviations from three independent experiments are shown. Statistical significance was determined by the ANOVA Bonferroni post test. *p < 0.05 compared to medium alone. # p < 0.05 compared to D1ED III.
RAW 264.7 macrophage cells was analyzed by flow cytometry using anti-D1ED III sera obtained from D1ED III immunized mice. LD1ED III in PBS associated more strongly with RAW 264.7 macrophage cells than D1ED III in PBS. Combining D1ED III with PELC only increased the antigen association index slightly. Remarkably, the antigen association index was significantly increased when LD1ED III was administered with PELC. The results from five independent experiments are summarized in Fig. 4B. 3.4. Effects of vaccine formulations on the immune response We next sought to determine whether LD1ED III combined with PELC induced potent immune responses. Groups of BALB/c mice were immunized with either D1ED III or LD1ED III twice over a two-week period in combination with or without PELC. Animals that received PBS alone served as negative controls. One week after the last immunization, the spleens were removed and single cell suspensions were prepared to evaluate cell proliferation and IFN-g production in response to D1ED III stimulation. Immunization with D1ED
III combined with either PBS or PELC did not induce notable antigen-specific proliferative responses. The stimulation indices of splenocytes from these mice were 1.4 0.4 and 1.5 0.3, respectively. Significantly, the stimulation indices of splenocytes from LD1ED III/PBS- or LD1ED III/PELCimmunized mice were elevated to 2.7 0.9 and 3.3 0.5, respectively (Fig. 5A). The splenocytes obtained from mice immunized with LD1ED III/PELC produced higher levels of IFN-g (404.5 241.0 pg/mL) than the splenocytes from the mice that received other treatments. Negligible levels of IFN-g were produced by splenocytes from naı¨ve mice and mice immunized with D1ED III/PBS (Fig. 5B). The antibody responses in mice were evaluated following three immunizations administered at two-week intervals. The avidity profile of the antibodies induced in mice immunized with the various formulations was shown in Fig. 5C, immunizing mice with LD1ED III combined with PELC generated higher avidity antibodies than immunizing mice with the other formulations ( p < 0.05 by the ANOVA Bonferroni post test). Following the antibody avidity analyses, we next tested the capacity of antibodies to neutralize dengue viral infection. As shown in Table 1, the antisera obtained from mice immunized
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
725
4. Discussion
Fig. 4. Effects of formulations on antigen association with RAW 264.7 macrophage cells. RAW 264.7 macrophage cells were pulsed with either LD1ED III or D1ED III in either the presence or absence of PELC. After a 3hour incubation, the antigen association with RAW 264.7 macrophage cells was detected using D1ED immunized sera and analyzed using flow cytometry. The mean fluorescence intensity (MFI) for cells detected by naı¨ve sera served as the control. (A) The antigen association index for each formulation, which was calculated according to the formula (MFI detected by D1ED III immunized sera)/(MFI detected by naı¨ve sera), is indicated in the upper right of the figure. (B) The means and standard deviations from five independent experiments are shown. Statistical significance was determined by the ANOVA Bonferroni post test. *p < 0.05 compared to D1ED III combined with PBS. # p < 0.05 compared to D1ED III combined with PELC. xp < 0.05 compared to LD1ED III.
with D1ED III/PBS were unable to block dengue viral infection efficiently. Significantly, immunizing mice with PELC combined with LD1ED III induced higher titers of neutralizing antibody than immunizing mice with the other formulations. Moreover, the neutralizing ability was persistent and still detectable 20 weeks after the first immunization. These results provide tangible evidence that a competent neutralizing antibody response is elicited in mice vaccinated with PELC combined with LD1ED III.
In general, antigens, immunostimulators, and delivery systems are the main elements of effective vaccines. Based on this notion, we hypothesized that the combination of recombinant lipoproteins, which contain antigens and immunostimulators, with a proper delivery system would enhance the ability of vaccine candidates to induce robust immune responses. In the present study, we prepared the recombinant lipoprotein LD1ED III (Fig. 1) and demonstrated its immunostimulatory capacity to induce the up-regulation of activation markers and IL-6 production on RAW 264.7 macrophage cells (Fig. 3). Both lipoproteins and lipopeptides have been demonstrated to activate antigen-presenting cells via the tolllike receptor signaling pathway [18,29e31] and to enhance humoral and cellular immune responses [17,19,25]. Here, LD1ED III provides two key elements of an effective vaccine: the antigen and the immunostimulator. The PELC functions as the antigen delivery system. When LD1ED III was combined with PELC the antigen association with RAW 264.7 macrophage cells was significantly increased (Fig. 4). LD1ED III in combination with PELC increased the probability of antigenpresenting cells capturing and processing antigens, which led to an enhanced cellular immune response. LD1ED III combined with PELC elicited higher T cell proliferation activity and IFN-g production than any of the other vaccine formulations administered (Fig. 5A and B). CpG oligodeoxynucleotides are immunostimulators that act through toll-like receptor 9 to activate antigen-presenting cells and trigger a protective immune response that enhances the host’s capacity to eradicate the pathogens [32e34]. In our previous studies, we demonstrated that either D1ED III [22] or inactivated influenza virus [13] combined with PELC plus CpG oligodeoxynucleotides synergize to enhance the immune response. These results suggest that immunostimulators synergize with the PELC emulsified system to induce host immunity against pathogens. Consistent with the findings in this study, combining LD1ED III with PELC elicits higher titers of neutralizing antibody than the other formulations used in this study (Table 1). Another important finding is that LD1ED III, like D1ED III, can block dengue viral infection (Fig. 2). These results suggest that the lipidated form of the antigen can still assume the native conformation. The neutralizing dengue virus antibodies represent a small fraction of the total dengue virus-specific antibodies produced in infected humans. These neutralizing antibodies appear to recognize novel epitopes, including complex and quaternary structure epitopes [35,36]. Subunit vaccines that retain the proper conformation will elicit an antibody response specific to both linear and conformational epitopes. Dengue envelope protein domain III is a promising candidate to be used for the development of a subunit dengue vaccine [37]. In our previous study, we developed a consensus dengue envelope protein domain III. Formulation of consensus dengue envelope protein domain III with aluminum phosphate induced neutralizing antibodies against four serotypes of dengue virus in mice [38], but only neutralizing antibodies
726
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
Fig. 5. Effects of formulations on cellular and humoral immune responses. BALB/c mice (n ¼ 3e4) were immunized subcutaneously with 10 mg antigens twice over a two-week interval in different formulations as indicated. One week after the last immunization, splenocytes were incubated with or without D1ED III (1 mg/ mL). (A) The stimulation index is defined as the ratio of the mean counts per minute (cpm) of D1ED III stimulation to the mean cpm without antigen stimulation. (B) The levels of IFN-g production were determined by ELISA assays. The means and standard deviations are shown. Statistical significance was determined by the ANOVA Bonferroni post test. *p < 0.05 compared to naive. #p < 0.05 compared to D1ED III combined with PBS. xp < 0.05 compared to D1ED III combined with PELC. yp < 0.05 compared to LD1ED III combined with PBS. (C) BALB/c mice (n ¼ 5) were immunized subcutaneously with 10 mg antigens three times at a two-week intervals in different formulations as indicated. The sera were collected at the indicated time points after the first immunization. Antibody avidity profiles were evaluated by ELISA. Statistical significance was determined by the ANOVA Bonferroni post test. *p < 0.05 compared to D1ED III combined with PBS. # p < 0.05 compared to D1ED III combined with PELC. xp < 0.05 compared to LD1ED III combined with PBS.
against serotype 2 were detected in monkey [39]. In addition, formulations of other dengue subunit vaccines with aluminum were unable to induce complete protection against dengue virus infection [40e44]. These results suggest that aluminum, the most widely used adjuvant in human vaccines, may not be suitable for dengue subunit vaccines. Therefore, it requires developing an adjuvant-free technology and/or an adjuvant/ delivery system for dengue preparedness. This is the reason why we try to explore a new system for dengue subunit vaccines. In the present study, we demonstrated that LD1ED III, a lipidated antigen without exogenous adjuvant, can induce long-lasting neutralizing antibody responses. Emulsified LD1ED III using PELC can further enhance the immune responses. There is a great potential use of PELC in human in the future, because all the components of PELC are bioresorbable [12,20]. All together, lipidated antigens and PELC
may have great benefits in the development of dengue subunit vaccines. The avidity of the antibody induced by LD1ED III combined with PELC was higher than that induced by the other formulations (Fig. 5C). These results imply that LD1ED III combined with PELC can quickly induce high avidity antibody responses. Remarkably, the neutralizing antibody titer induced by LD1ED III combined with PELC was significantly higher than that induced by the other formulations (Table 1). These results suggest that LD1ED III combined with PELC enhances the intensity and modulates the quality of the immune response, thus potentiating the host response against dengue viral infection. Importantly, the application of recombinant lipoprotein and PELC demonstrated herein could provide a novel strategy for the development of the other subunit vaccines.
Table 1 Neutralizing antibody titers in mice immunized with various formulations.a Time
Groups
8 weeks
D1ED III/PBS D1ED III/PELC LD1ED III/PBS LD1ED III/PELC D1ED III/PBS D1ED III/PELC LD1ED III/PBS LD1ED III/PELC
20 weeks
FRNT50
FRNT90
Mouse 1
2
3
4
5
1
2
3
4
5
<8 8 32 64 <8 8 32 32
<8 16 32 128 8 16 32 64
8 16 32 128 8 16 32 64
8 32 32 256 8 32 64 128
8 32*b 32* 256*#x 8 32 128* 128*#
<8 8 <8 8 <8 <8 8 8
<8 8 8 16 <8 <8 8 16
<8 8 8 16 <8 8 16 16
<8 8 8 32 <8 8 16 16
<8 8* 8 32*#x <8 8 16*# 32*#
Mice were immunized with various formulations three times (10 mg/dose) at two-week intervals. The serum samples were collected at 8 and 20 weeks after the first immunization. The dengue-1 virus neutralizing capacity was determined by FRNT. The neutralizing antibody titer was defined as the reciprocal of the highest dilution that resulted in a 50% or 90% reduction of FFU compared to control samples containing the virus alone. b Statistical analysis was performed using the ANOVA Bonferroni post test at the same time point. *p < 0.05 compared to D1ED III/PBS. #p < 0.05 compared to D1ED III/PELC. xp < 0.05 compared to LD1ED III/PBS. a
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
Acknowledgments This work was supported by grants 100A1-VC-PP-07-014 and 101A1-IV-PP-26-014 from the National Health Research Institutes. References [1] M.R. Hilleman, Vaccines in historic evolution and perspective: a narrative of vaccine discoveries, Vaccine 18 (2000) 1436e1447. [2] D.M. Pardoll, Spinning molecular immunology into successful immunotherapy, Nat. Rev. Immunol. 2 (2002) 227e238. [3] S.G. Reed, S. Bertholet, R.N. Coler, M. Friede, New horizons in adjuvants for vaccine development, Trends Immunol. 30 (2009) 23e32. [4] G.M. Barton, R. Medzhitov, Toll-like receptors and their ligands, Curr. Top. Microbiol. Immunol. 270 (2002) 81e92. [5] A. Iwasaki, R. Medzhitov, Toll-like receptor control of the adaptive immune responses, Nat. Immunol. 5 (2004) 987e995. [6] H. Kato, K. Takahasi, T. Fujita, RIG-I-like receptors: cytoplasmic sensors for non-self RNA, Immunol. Rev. 243 (2011) 91e98. [7] M. Lamkanfi, T.D. Kanneganti, Regulation of immune pathways by the NOD-like receptor NLRC5, Immunobiology 217 (2012) 13e16. [8] A. Williams, R.A. Flavell, S.C. Eisenbarth, The role of NOD-like receptors in shaping adaptive immunity, Curr. Opin. Immunol. 22 (2010) 34e40. [9] E. Erikci, M. Gursel, I. Gursel, Differential immune activation following encapsulation of immunostimulatory CpG oligodeoxynucleotide in nanoliposomes, Biomaterials 32 (2011) 1715e1723. [10] J.S. Thomann, B. Heurtault, S. Weidner, M. Braye, J. Beyrath, S. Fournel, F. Schuber, B. Frisch, Antitumor activity of liposomal ErbB2/ HER2 epitope peptide-based vaccine constructs incorporating TLR agonists and mannose receptor targeting, Biomaterials 32 (2011) 4574e4583. [11] A. Pashine, N.M. Valiante, J.B. Ulmer, Targeting the innate immune response with improved vaccine adjuvants, Nat. Med. 11 (2005) S63eS68. [12] M.H. Huang, C.Y. Huang, S.P. Lien, S.Y. Siao, A.H. Chou, H.W. Chen, S.J. Liu, C.H. Leng, P. Chong, Development of multi-phase emulsions based on bioresorbable polymers and oily adjuvant, Pharm. Res. 26 (2009) 1856e1862. [13] M.H. Huang, S.C. Lin, C.H. Hsiao, H.J. Chao, H.R. Yang, C.C. Liao, P.W. Chuang, H.P. Wu, C.Y. Huang, C.H. Leng, S.J. Liu, H.W. Chen, A.H. Chou, A.Y. Hu, P. Chong, Emulsified nanoparticles containing inactivated influenza virus and CpG oligodeoxynucleotides critically influences the host immune responses in mice, PLoS One 5 (2010) e12279. [14] Z. Zhang, S. Tongchusak, Y. Mizukami, Y.J. Kang, T. Ioji, M. Touma, B. Reinhold, D.B. Keskin, E.L. Reinherz, T. Sasada, Induction of antitumor cytotoxic T cell responses through PLGA-nanoparticle mediated antigen delivery, Biomaterials 32 (2011) 3666e3678. [15] M.G. Cusi, Applications of influenza virosomes as a delivery system, Hum. Vaccines 2 (2006) 1e7. [16] M.J. Heffernan, D.A. Zaharoff, J.K. Fallon, J. Schlom, J.W. Greiner, In vivo efficacy of a chitosan/IL-12 adjuvant system for protein-based vaccines, Biomaterials 32 (2011) 926e932. [17] H.W. Chen, S.J. Liu, H.H. Liu, Y. Kwok, C.L. Lin, L.H. Lin, M.Y. Chen, J.P. Tsai, L.S. Chang, F.F. Chiu, L.W. Lai, W.C. Lian, C.Y. Yang, S.Y. Hsieh, P. Chong, C.H. Leng, A novel technology for the production of a heterologous lipoprotein immunogen in high yield has implications for the field of vaccine design, Vaccine 27 (2009) 1400e1409. [18] C.H. Leng, H.W. Chen, L.S. Chang, H.H. Liu, H.Y. Liu, Y.P. Sher, Y.W. Chang, S.P. Lien, T.Y. Huang, M.Y. Chen, A.H. Chou, P. Chong, S.J. Liu, A recombinant lipoprotein containing an unsaturated fatty acid activates NF-kappaB through the TLR2 signaling pathway and induces a differential gene profile from a synthetic lipopeptide, Mol. Immunol. 47 (2010) 2015e2021.
727
[19] C.Y. Chiang, S.J. Liu, J.P. Tsai, Y.S. Li, M.Y. Chen, H.H. Liu, P. Chong, C.H. Leng, H.W. Chen, A novel single-dose dengue subunit vaccine induces memory immune responses, PLoS One 6 (2011) e23319. [20] M.H. Huang, A.H. Chou, S.P. Lien, H.W. Chen, C.Y. Huang, W.W. Chen, P. Chong, S.J. Liu, C.H. Leng, Formulation and immunological evaluation of novel vaccine delivery systems based on bioresorbable poly(ethylene glycol)-block-poly(lactide-co-epsilon-caprolactone), J. Biomed. Mater. Res. B Appl. Biomater. 90 (2009) 832e841. [21] M.H. Huang, C.Y. Huang, S.C. Lin, J.H. Chen, C.C. Ku, A.H. Chou, S.J. Liu, H.W. Chen, P. Chong, C.H. Leng, Enhancement of potent antibody and T-cell responses by a single-dose, novel nanoemulsion-formulated pandemic influenza vaccine, Microbes Infect. 11 (2009) 654e660. [22] C.Y. Chiang, M.H. Huang, C.H. Hsieh, M.Y. Chen, H.H. Liu, J.P. Tsai, Y.S. Li, C.Y. Chang, S.J. Liu, P. Chong, C.H. Leng, H.W. Chen, Dengue1 envelope protein domain III along with PELC and CpG oligodeoxynucleotides synergistically enhances immune responses, PLoS Negl. Trop. Dis. 6 (2012) e1645. [23] R.A. Macdonald, C.S. Hosking, C.L. Jones, The measurement of relative antibody affinity by ELISA using thiocyanate elution, J. Immunol. Methods 106 (1988) 191e194. [24] N. Nair, H. Gans, L. Lew-Yasukawa, A.C. Long-Wagar, A. Arvin, D.E. Griffin, Age-dependent differences in IgG isotype and avidity induced by measles vaccine received during the first year of life, J. Infect. Dis. 196 (2007) 1339e1345. [25] C.Y. Huang, J.J. Chen, K.Y. Shen, L.S. Chang, Y.C. Yeh, I.H. Chen, P. Chong, S.J. Liu, C.H. Leng, Recombinant lipidated HPV E7 induces a Th-1-biased immune response and protective immunity against cervical cancer in a mouse model, PLoS One 7 (2012) e40970. [26] Y. Kwok, W.C. Sung, A.L. Lin, H.H. Liu, F.A. Chou, S.S. Hsieh, C.H. Leng, P. Chong, Rapid isolation and characterization of bacterial lipopeptides using liquid chromatography and mass spectrometry analysis, Proteomics 11 (2011) 2620e2627. [27] Y. Chen, T. Maguire, R.E. Hileman, J.R. Fromm, J.D. Esko, R.J. Linhardt, R.M. Marks, Dengue virus infectivity depends on envelope protein binding to target cell heparan sulfate, Nat. Med. 3 (1997) 866e871. [28] J.-J. Hung, M.-T. Hsieh, M.-J. Young, C.-L. Kao, C.-C. King, W. Chang, An external loop region of domain III of dengue virus type 2 envelope protein is involved in serotype-specific binding to mosquito but not mammalian cells, J. Virol. 78 (2004) 378e388. [29] T. Shimizu, Y. Kida, K. Kuwano, A triacylated lipoprotein from Mycoplasma genitalium activates NF-kappaB through toll-like receptor 1 (TLR1) and TLR2, Infect. Immun. 76 (2008) 3672e3678. [30] K. Tawaratsumida, M. Furuyashiki, M. Katsumoto, Y. Fujimoto, K. Fukase, Y. Suda, M. Hashimoto, Characterization of N-terminal structure of TLR2-activating lipoprotein in Staphylococcus aureus, J. Biol. Chem. 284 (2009) 9147e9152. [31] S. Thakran, H. Li, C.L. Lavine, M.A. Miller, J.E. Bina, X.R. Bina, F. Re, Identification of Francisella tularensis lipoproteins that stimulate the toll-like receptor (TLR) 2/TLR1 heterodimer, J. Biol. Chem. 283 (2008) 3751e3760. [32] S. Bauer, C.J. Kirschning, H. Hacker, V. Redecke, S. Hausmann, S. Akira, H. Wagner, G.B. Lipford, Human TLR9 confers responsiveness to bacterial DNA via species-specific CpG motif recognition, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 9237e9242. [33] H. Hemmi, O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, K. Hoshino, H. Wagner, K. Takeda, S. Akira, A toll-like receptor recognizes bacterial DNA, Nature 408 (2000) 740e745. [34] F. Takeshita, C.A. Leifer, I. Gursel, K.J. Ishii, S. Takeshita, M. Gursel, D.M. Klinman, Cutting edge: role of toll-like receptor 9 in CpG DNAinduced activation of human cells, J. Immunol. 167 (2001) 3555e3558. [35] R. de Alwis, S.A. Smith, N.P. Olivarez, W.B. Messer, J.P. Huynh, W.M. Wahala, L.J. White, M.S. Diamond, R.S. Baric, J.E. Crowe Jr., A.M. de Silva, Identification of human neutralizing antibodies that bind to complex epitopes on dengue virions, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 7439e7444. [36] W.M. Wahala, A.M. Silva, The human antibody response to dengue virus infection, Viruses 3 (2011) 2374e2395.
728
C.-Y. Chiang et al. / Microbes and Infection 15 (2013) 719e728
[37] M.G. Guzman, L. Hermida, L. Bernardo, R. Ramirez, G. Guillen, Domain III of the envelope protein as a dengue vaccine target, Expert Rev. Vaccines 9 (2010) 137e147. [38] C.H. Leng, S.J. Liu, J.P. Tsai, Y.S. Li, M.Y. Chen, H.H. Liu, S.P. Lien, A. Yueh, K.N. Hsiao, L.W. Lai, F.C. Liu, P. Chong, H.W. Chen, A novel dengue vaccine candidate that induces cross-neutralizing antibodies and memory immunity, Microbes Infect. 11 (2009) 288e295. [39] H.W. Chen, S.J. Liu, Y.S. Li, H.H. Liu, J.P. Tsai, C.Y. Chiang, M.Y. Chen, C.S. Hwang, C.C. Huang, H.M. Hu, H.H. Chung, S.H. Wu, P. Chong, C.H. Leng, C.H. Pan, A consensus envelope protein domain III can induce neutralizing antibody responses against serotype 2 of dengue virus in non-human primates, Arch. Virol. (2013). [40] L. Bernardo, A. Izquierdo, M. Alvarez, D. Rosario, I. Prado, C. Lopez, R. Martinez, J. Castro, E. Santana, L. Hermida, G. Guillen, M.G. Guzman, Immunogenicity and protective efficacy of a recombinant fusion protein containing the domain III of the dengue 1 envelope protein in non-human primates, Antivir. Res. 80 (2008) 194e199. [41] M.G. Guzman, R. Rodriguez, L. Hermida, M. Alvarez, L. Lazo, M. Mune, D. Rosario, K. Valdes, S. Vazquez, R. Martinez, T. Serrano,
J. Paez, R. Espinosa, T. Pumariega, G. Guillen, Induction of neutralizing antibodies and partial protection from viral challenge in Macaca fascicularis immunized with recombinant dengue 4 virus envelope glycoprotein expressed in Pichia pastoris, Am. J. Trop. Med. Hyg. 69 (2003) 129e134. [42] J. Robert Putnak, B.A. Coller, G. Voss, D.W. Vaughn, D. Clements, I. Peters, G. Bignami, H.S. Houng, R.C. Chen, D.A. Barvir, J. Seriwatana, S. Cayphas, N. Garcon, D. Gheysen, N. Kanesa-Thasan, M. McDonell, T. Humphreys, K.H. Eckels, J.P. Prieels, B.L. Innis, An evaluation of dengue type-2 inactivated, recombinant subunit, and liveattenuated vaccine candidates in the rhesus macaque model, Vaccine 23 (2005) 4442e4452. [43] M. Simmons, K.R. Porter, C.G. Hayes, D.W. Vaughn, R. Putnak, Characterization of antibody responses to combinations of a dengue virus type 2 DNA vaccine and two dengue virus type 2 protein vaccines in rhesus macaques, J. Virol. 80 (2006) 9577e9585. [44] J. Velzing, J. Groen, M.T. Drouet, G. van Amerongen, C. Copra, A.D. Osterhaus, V. Deubel, Induction of protective immunity against dengue virus type 2: comparison of candidate live attenuated and recombinant vaccines, Vaccine 17 (1999) 1312e1320.