Mol Biotechnol DOI 10.1007/s12033-014-9769-6
RESEARCH
Optimization of the Immunogenicity of a DNA Vaccine Encoding a Bacterial Outer Membrane Lipoprotein Arun Buaklin • Tanapat Palaga • Drew Hannaman Ruthairat Kerdkaew • Kanitha Patarakul • Alain Jacquet
•
Ó Springer Science+Business Media New York 2014
Abstract Bacterial outer membrane lipoproteins represent potent immunogens for the design of recombinant subunit vaccines. However, recombinant lipoprotein production and purification could be a challenge notably in terms of expression yield, protein solubility, and posttranslational acylation. Together with the cost effectiveness, facilitated production, and purification as well as good stability, DNA-based vaccines encoding lipoproteins could become an alternative strategy for antibacterial vaccinations. Although the immunogenicity and the efficacy of DNA-based vaccines can be demonstrated in small rodents, such vaccine candidates could request concrete
Present Address: A. Buaklin K. Patarakul (&) Department of Microbiology, Faculty of Medicine, Chulalongkorn University, Oor-Por-Ror Building, 15th floor, Room # 1510B2 1873 Rama IV Road, Pathum Wan, Bangkok 10330, Thailand e-mail: [email protected]; [email protected] T. Palaga Department of Microbiology, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand D. Hannaman Ichor Medical Systems, San Diego, CA 92121, USA R. Kerdkaew Department of Internal Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok 10330, Thailand Present Address: A. Jacquet (&) Department of Medicine, Faculty of Medicine, Chulalongkorn University, Pattayapat Building, 8th floor, Room # 807, 1873 Rama IV Road, Pathumwan, Bangkok 10330, Thailand e-mail: [email protected]
optimization as they are weak immunogens in primates and humans and particularly when administered by conventional injection. Therefore, the goal of the present study was to optimize the immunogenicity of a DNA vaccine encoding an outer membrane lipoprotein. LipL32, the major outer membrane protein from pathogenic Leptospira, was selected as a model antigen. We evaluated the influence of antigen secretion, the in vivo DNA delivery by electroporation, the adjuvant co-administration, as well as the heterologous prime-boost regimen on the induction of anti-LipL32 specific immune responses. Our results clearly showed that, following transfections, a DNA construct based on the authentic full-length LipL32 gene (containing leader sequence and the N-terminus cysteine residue involved in the protein anchoring) drives antigen secretion with the same efficiency as a plasmid-encoding anchor-less LipL32 and for which the bacterial leader sequence was replaced with a viral signal peptide. The in vivo DNA delivery by electroporation drastically enhanced the production of strong Th1 responses characterized by specific IgG2a antibodies and the IFNc secretion in a restimulation assay, regardless of the DNA constructs used. In comparison with the heterologous prime-boost regimen, the homologous prime-boost vaccinations with DNA coadministrated with polyinosinic-polycytidylic acid (poly I:C) generated the highest specific IgG and IgG2a titers as well as the greatest IFNc production. Taken together, these data suggest that optimization of outer membrane lipoprotein secretion is not critical for the induction of antigenspecific responses through DNA vaccination. Moreover, the potent antibody response induced by DNA plasmid encoding lipoprotein formulated with poly I:C and delivered through electroporation provides the rationale for the design of new prophylactic vaccines against pathogenic bacteria.
123
Mol Biotechnol
Keywords Bacterial lipoprotein DNA vaccine Electroporation Poly I:C Prime-boost LipL32
Materials and Methods Production of Recombinant Anchor-Less Mature LipL32 Protein (rLipL32 DCys1)
Introduction DNA vaccines are commonly composed of a plasmid backbone encoding the antigen of interest under the control of a strong eukaryotic promoter [1]. Such recombinant plasmids, following intramuscular or intradermal administrations, lead to antigen expression and presentation through direct in vivo transfection of antigen-presenting cells such as dendritic cells (DCs). Alternatively, the expressed antigen from transfected myocytes or other bystander cells can be transferred to DCs for the crosspriming. Consequently, MHC class I and II antigen presentation is elicited through DNA vaccinations [2]. Compared with vaccines based on recombinant proteins, DNA vaccines offer several advantages including high stability, cost effectiveness of DNA production, and simple purification process. In rodents, DNA-encoding antigens preferentially induce the production of IgG2a antibodies, CD8? and CD4? type 1 T cell responses, and IFNc secretion [3]. However, DNA vaccines suffer from low or modest immunogenicity in humans [2, 3]. Hence, strategies are needed to significantly improve the efficiency of DNA vaccinations through the optimization of the antigen expression and secretion, the in vivo DNA delivery, the DNA formulation with adjuvants, or the heterologous prime-boost vaccinations [1]. Lipoproteins from Gram-positive and Gram-negative bacterial pathogens represent a large class of membrane proteins displaying numerous biological functions such as adhesion, immunomodulation, or virulence [4]. These membrane proteins display a fatty acid-based anchor covalently bound to the N-terminal cysteine. According to these key roles and to their strong immunogenicity through notably their fatty acid moieties [5], these bacterial lipoproteins can be considered as promising vaccine candidates, which, following immunizations, could promote the development of protective antibody responses. The goal of the present study was to design and to optimize a DNA vaccine encoding a bacterial lipoprotein. As a model antigen, we selected LipL32, the most abundant Leptospira outer membrane lipoprotein [6]. Moreover, LipL32 is expressed during Leptospira infection in vivo and was shown to display strong immunogenicity by notably TLR2 activation through its fatty acid moieties [7, 8]. Our results clearly highlight the importance of adjuvant formulations and in vivo delivery systems for the immunogenicity of lipoprotein-based DNA vaccines.
123
Recombinant His-tagged anchor-less mature LipL32 (rLipL32 DCys1) was cloned according to a previously published strategy [6]. DNA-encoding LipL32DCys1 was amplified from Leptospira interrogans serovar Pomona genomic DNA as a template and using the following primers: Forward 50 -TTACCGCTCGAGGTGCTTTCGGTGGTCT GC-30 (XhoI underlined) and Reverse 50 -TGTTAACCC GGGTTACTTAGTCGCGTCAGA-30 (SmaI underlined). PCR product was cloned into the pGem-T easy vector (Promega). The recombinant amplicon was sequenced and subsequently digested with XhoI and SmaI for the cloning into the pRSETc bacterial expression vector. BL21(DE3) pLysS E.coli cells were transformed with pRSETc-LipL32, and rLipL32 DCys1 expression was induced by IPTG addition into the culture medium. The expressed protein was purified using immobilized Ni2? as previously described [6]. Construction and Preparation of DNA Plasmids Encoding LipL32 The full-length LipL32 (leader ? mature sequence, LWTLipL32) DNA was cloned into the pVITRO1 (Invivogen, USA) mammalian expression vector following BspEI and BsiWI restriction sites of the recombinant pGEM-T easy vector (Promega, USA). To design another DNA plasmid expressing anchor-less LipL32 (DCys1) and in which the LipL32 leader sequence was replaced with the viral VZV gE signal peptide [9] (LVir-DCys1-LipL32), a PCR product was obtained using the following primers: forward 50 -G GATCCGGAGCCACCATGGGCACCGTGAACAAGCC CGTCGT GGGCGTGCTGATGGGCTTCGGCATCATC ACCGGCACCCTGCGGATCACCAACCCCGTGCGGG CCGGTGCTTTCGGTGGTCTGCCAAGC-30 (BspEI underlined and Kozak sequence in bold) and reverse 50 -TA ACGTACGTTACTTAGTCGCGTCAG-30 (BsiWI underlined). A plasmid encoding leader-less LipL32 was also designed by PCR amplification of LipL32 DNA using the same 30 reverse primer used for LVir-DCys1-LipL32 construction and the following 50 primer: 50 -GGATCCGGA GC CACCATGTGTGGTGCTTTCGGTGGTCTGCCAAGC-30 (BspEI underlined and Kozak sequence in bold). The two resulting amplicons were subsequently digested with BspEI and BsiWI for the cloning into pVITRO1. All plasmid DNA constructs were produced and purified under endotoxin-free condition using GigaPrep Kits (QIAGEN, Germany) for in vitro transfection and animal immunization.
Mol Biotechnol
In Vitro Transient Transfection Assays
LipL32-Specific Antibody Assay
Human proximal tubular HK-2 cells were cultured onto six-well plates in Dulbecco’s modified Eagle’s medium (DMEM) GlutaMAXTM (Invitrogen, Carlsbad, USA) supplemented with 10 % fetal bovine serum (FBS) (Invitrogen, Carlsbad, USA) at a density of 1 9 106 cells/well. At a 90 % confluency, the endotoxin-free plasmid DNA (4 lg of each DNA) were diluted in 250 lL of serum-free DMEM medium, gently mixed and incubated with 250 lL of a dilute solution of lipofectamineTM 2000 (Invitrogen) for 20 min at room temperature. The lipoplexes were added into each well plate and incubated at 37 °C in a CO2 incubator for 72 h. The culture supernatants were subsequently collected, and cells were washed twice with icecold PBS and lysed in 1X SDS lysis buffer (50 mM Tris, pH 6.8, 50 mM DTT, 2 % SDS, 10 % glycerol, 100 lg/ml leupeptin, and 0.1 % bromophenol blue) for 10 min on ice before cell lysates were collected. Culture supernatants and cell lysates were analyzed by Western blot for LipL32 expression using specific mouse polyclonal antibodies generated by rLipL32 DCys1 immunization (data not shown).
For antibody detection, rLipL32 DCys1 was directly coated onto ELISA well plates (500 ng/well) with carbonate buffer (pH 9.6) overnight at 4 °C. Plates were washed six times with phosphate-buffer saline (PBS) containing 0.05 % Tween 20 (PBS-T) and then blocked with 0.5 % (wt/v) bovine serum albumin in PBS-T for 1 h at room temperature. After blocking, wells were incubated with serial dilutions of each mouse serum (1:100–1:312,500 for specific IgG, 1:30–1:93,750 for both specific IgG1 and IgG2a) at 37 °C for 1 h. After washing with PBS-T, wells were incubated with horseradish peroxidase-conjugated goat anti-mouse IgG, IgG1, or IgG2a antibodies (dilution 1:5,000 for each antibody, BD Biosciences, USA) at 37 °C for 1 h. The antigen–antibody complexes were detected by adding 100 ll of TMB substrate solution, and the enzymatic reaction was stopped by the addition of 50 ll of 0.5 M H2SO4. The optical density was read at 450 nm using a Bio-Rad microplate reader.
Immunizations The present study was approved by the Chulalongkorn University Animal Care and Use Committee (CU-ACUC). In the first experiment, groups of six BALB/c mice were immunized three times at two-week intervals with 20 lg (in 20 ll) of pVITRO1 encoding LWT-LipL32 or LVir-DCys1-LipL32. As a control group, animals were immunized with empty pVITRO1. Plasmid DNA constructs were administrated intramuscularly (IM) or were delivered by in vivo electroporation (EP) using the TriGridTM Delivery System (rodent model, Ichor Medical Systems, www.ichorms.com). The IM DNA injection into one tibialis anterior muscle was directly followed by the stimulation at the site of injection with an electrical pulse that had an amplitude of 250 V per cm of electrode spacing for 40 ms over a 400 ms interval. In a second set of the experiments, homologous primeboost regimen (three electroporations with 20 lg DNALVir-DCys1-LipL32) was compared with heterologous prime-boost immunizations (two in vivo electroporations followed by one subcutaneous injection of 5 lg rLipL32 DCys1 in the presence or absence of 10 lg poly I:C (Invivogen). Two weeks after the last immunization, sera were collected from each mouse by retro-orbital bleeding for antibody assay and spleens were collected from sacrificed animals.
Cytokine Assays To assess the cytokine production of proliferative T cells, splenocytes (5 9 105/well in triplicate) were stimulated with 20 lg/mL rLipL32 DCys1, 10 lg/ml concanavalin A (ConA), or RPMI1640 medium alone for 72 h and culture supernatants were collected after 72 h. The levels of IFNc and IL-4 in the supernatants were measured by ELISA assays according to the BD OptiEIA kits (BD, USA) manufacturer’s instructions. Statistical Analysis Student’s t test was used to compare data between pairs of groups. Differences were considered significant for P values \0.05.
Results Construction and Expression of DNA Vaccines in HK-2 Cells To design a DNA vaccine candidate based on the LipL32 leptospiral lipoprotein, the DNA-encoding full-length LipL32 (LWT-LipL32) was cloned into the pVITRO1 eukaryotic expression vector. In this DNA construct, the absence of the typical Kozak sequence, a mammalian consensus site immediately prior to the ATG start codon which may further enhance antigen expression [1], together with the presence of the natural bacterial leader sequence
123
Mol Biotechnol
Fig. 1 LipL32 expression and secretion from transiently transfected HK-2 cells. HK-2 cells were transiently transfected with empty vector or vectors encoding LipL32. a N-terminal amino acid sequences of LipL32 encoded by DNA-LWT-LipL32, DNA-LVir-DCys1-LipL32 and DNA-Leader-less LipL32. The first amino acids of mature LipL32 are underlined, Antigen expression was monitored at 72 h post-transfection in the supernatants (lanes 2, 4, 6, and 8) and in the cell lysates (lanes 3, 5, 7 and 9). Lane 1 purified rLipL32 DCys1, Lanes 2 and 3 empty vector, Lanes 4 and 5 DNA-Leader-less LipL32, Lanes 6 and 7 DNA-LWT-LipL32, Lanes 8 and 9 DNA-LVir-DCys1LipL32. Western blot analysis was performed using anti-LipL32 mouse polyclonal antibodies (b) and monoclonal antibodies to b-actin (c)
as well as the N-terminal Cys residue involved in the anchoring of the protein in the bacterial outer membrane could have negative impacts on antigen expression and secretion. Notably, sub-optimal antigen secretion could impair the DNA plasmid immunogenicity as it was previously reported that optimization of antigen secretion enhances the antibody production rate [10, 11]. Consequently, we designed a second DNA vaccine (LVir-DCys1LipL32) in which the natural leader sequence was replaced with that from the VZV gE antigen, a leader sequence which triggered efficient secretion in mammalian cells [9, 12]. Moreover, a Kozak sequence was also inserted and the N-terminal Cys residue was removed in order to express DCys1-LipL32. The expression and secretion of LipL32 were evaluated through transient expression in HK-2 cells transfected by both DNA constructs. As controls, cells were transfected with the empty vector or a plasmid-encoding Leader-less LipL32. LipL32 expression could be detected in transfected cells as well as in the culture supernatant with both pVITRO-LipL32 vaccine candidates (Fig. 1). Cell transfection with a plasmid without a leader sequence induced strictly intracellular LipL32 production, confirming the importance of a leader sequence for the antigen secretion. Densitometric analysis (after normalization for difference in the b-actin bands between the two cell lysates)
123
demonstrated similar levels of intracellular antigen expression whatever the DNA construct used. Surprisingly, the DNA plasmid encoding LVir-DCys1-LipL32 did not significantly drive higher antigen secretion (P [ 0.05). Similar levels of LipL32 expression and secretion were obtained using HEK293 cells (human embryonic kidney cells) in transfection assays (data not shown). These results suggested that the natural LipL32 leader sequence is functional to trigger secretion and that the N-terminal cysteine residue is more likely not acylated in mammalian cells. It must be pointed out that secreted LipL32 displayed a higher molecular weight than the intracellular form. This could be more likely explained by the presence of N-linked glycosylation at the level of the putative N-glycosylation site (N25ET). Recombinant purified rLipL32 DCys1 displayed also a higher molecular weight by the presence of an N-terminus extension containing His-tag and protease cleavage site. Immunogenicity of DNA-LWT-LipL32 and DNA-LVirDCys1-LipL32 To determine whether both DNA constructs encoding secreted LipL32 are immunogenic, the two DNA vaccine candidates (20 lg each) were delivered three times in BALB/c mice by electroporation or conventional intramuscular injection. Indeed, numerous studies highlighted the interest of in vivo electroporation to improve the immunogenicity of DNA-based vaccines [13]. The specific humoral as well as cellular responses were analyzed through antiLipL32 total IgG, IgG1, IgG2a detection in sera as well as IFNc and IL-4 secreted from LipL32 restimulated splenocytes (Fig. 2). Modest but comparable specific antibody productions were elicited following intramuscular immunizations with DNA-LWT-LipL32 and DNA-LVir-DCys1LipL32; in contrast, electroporations drastically increased the production of anti-LipL32 IgG, IgG1, and IgG2a. However, no significant difference in the antibody titers induced by two DNA constructions were observed (P [ 0.05). Cytokine detection from lymphoproliferative assays demonstrated that both DNA constructs similarly triggered elevated amount of IFNc but not of IL-4 (data not shown), regardless of the type of DNA delivery used. However, higher IFNc secretion was detected following electroporation (P \ 0.05). According to the low specific IgG1/IgG2a ratio and high IFNc production together with the absence of IL-4, our results clearly confirmed that DNA immunizations through electroporation induced a strong anti-LipL32 Th1-biased response. Moreover, we can consider that the immunogenicity level of DNA-LWT-LipL32 and DNA-LVir-DCys1LipL32 was correlated with the similar LipL32 expression/ secretion level in our transient transfection assays.
Mol Biotechnol
Fig. 2 Immunogenicity of DNA-LWT-LipL32 and DNA-LVir-DCys1LipL32 Specific IgG (a), IgG1 (b) and IgG2a (c) titers postvaccinations, restimulated splenocyte IFNg production (d). All the
values were plotted as mean ± SEM. One representative experiment out of 2 is shown. *P \ 0.05 (comparisons between treatments). EP electroporation, IM intramuscular
Effect of the Poly I:C Adjuvant and Prime-Boost Regimen on Immunogenicity of DNA-LVir-DCys1LipL32
induced by DNA-LVir-DCys1-LipL32 through three electroporations (homologous prime-boost) and two electroporations combined with one subcutaneous immunization with rLipL32-DCys1 (heterologous prime-boost.). The same immunization protocol was applied to two other mice groups, but the DNA plasmid as well the recombinant antigen were adjuvanted with Poly I:C. Whereas comparable specific IgG1 antibody titers were elicited whatever the presence or the absence of Poly I:C and the prime-boost regimen, this TLR ligand significantly but uniquely improved the production of anti-LipL32 IgG2a through the homologous prime-boost immunizations (three DNA electroporations) (Fig. 3) (P \ 0.05). A similar and significant increase in IFNc production was also detected in this group of immunized mice. Once again, no IL-4 production was triggered regardless of the immunization regimen.
As both DNA constructs encoding LipL32 displayed similar immunogenicity when delivered by electroporation, DNA-LVir-DCys1-LipL32 was selected to explore the influence of co-injection with Poly I:C as an adjuvant as well as the DNA prime-boost regimen on the optimization of the DNA vaccine immunogenicity. Indeed, it was previously demonstrated that adjuvanting with poly I:C or immunizing using different types of vaccines for the delivery of the same antigen in the priming and the boost steps can increase antigen-specific T cell and B cell responses [14–16]. Accordingly, another set of the experiments was performed to compare the specific immune responses
123
Mol Biotechnol
Fig. 3 Effect of the Poly I:C adjuvant and prime-boost regimen on immunogenicity of DNA-LVir-DCys1-LipL32 Specific IgG (a), IgG1 (b) and IgG2a (c) titers post-vaccinations, restimulated splenocyte
IFNg production (d). All the values were plotted as mean ± SEM. One representative experiment out of 2 is shown. *P \ 0.05 (comparisons between treatments)
Discussion
Compared with conventional protein vaccines, DNA vaccines encoding bacterial lipoproteins could be attractive since their development is relatively easy and their production cost is low. As antigen secretion could drastically influence the development of specific antibody response, it is believed that genetic constructions encoding lipoproteins must be optimized to improve the extracellular release of the expressed proteins following immunizations. It is unlikely that bacterial lipoproteins expressed in mammalian cells display N-terminal cysteine acylation as in the manner of prokaryotes. Indeed, the three conserved enzymes, prolipoprotein diacylglyceryltransferase (Lgt), prolipoprotein signal peptidase (LspA), and phospholipid:apolipoprotein N-acyltransferase (Lnt), responsible for the post-translational lipid modification after recognition of the lipobox sequence at the end of lipoprotein signal peptide are absent in eukaryotes [4, 5]. Moreover, whereas bacterial leader peptides can direct bacterial protein secretion in mammalian cells [20], they could be less
Outer membrane lipoproteins from bacterial pathogens could represent efficient vaccine candidates [3]. Indeed such lipoproteins are highly immunogenic because of their lipid adjuvant motifs which trigger potent innate immune response through notably TLR2 activation [17]. However, due to their lipid motif anchored to the membrane, isolation of bacterial natural lipoproteins could represent an issue in terms of solubility, and consequently requests the use of detergents. Recombinant expression of lipoproteins in E. coli can also be problematic and is commonly associated with inappropriate lipid modifications [18, 19]. To prevent such issues, the alternative is to produce fully soluble anchor-less lipoproteins which more likely could be less immunogenic than the corresponding anchored protein. Nevertheless, such purified non-lipidated bacterial antigens could be formulated with potent adjuvants, notably TLR ligands to up-regulate their vaccine potential.
123
Mol Biotechnol
efficient than eukaryotic or viral signal sequence to initiate expression and secretion into animal cells. In the present study, two DNA plasmids encoding the Leptospira outer membrane protein LipL32 as a bacterial lipoprotein prototype were designed: one construction encoding the full-length protein encompassing the natural leader sequence with the lipobox and a second one in which the natural leader sequence was replaced with that from the VZV gE antigen. This viral leader peptide was previously shown to trigger efficiently heterologous antigen secretion in recombinant eukaryotic cells [9, 12]. Our transfection assays strikingly suggested that the leader sequence of bacterial lipoproteins is functional in mammalian cells to allow antigen secretion. The similar level of LipL32 secretion, whatever the DNA plasmid used, confirmed as well that the bacterial lipoprotein is more likely not acylated when produced in eukaryotic cells. As the lipid anchor of bacterial lipoprotein is a strong adjuvant, through notably TLR2 signaling, such DNA vaccine candidates encoding anchor-less LipL32 could display poor immunogenicity [17]. In order to counterbalance the absence of these lipid moieties, we evaluated the interest of in vivo DNA delivery by electroporation to optimize the DNA immunogenicity. Our data highlighted that whereas very low specific immune response was elicited by the two DNA vaccine candidates following intramuscular administrations, electroporation drastically enhanced the overall level of specific humoral response as well as IFNc production to LipL32. To our knowledge, this is the first electroporationbased immunization protocol using a DNA-encoding nonacylated bacterial lipoprotein. The boost of the immune response following electroporation can more likely explained by a direct access of DNA to the cytoplasm leading to an efficient activation of cytosolic DNA sensors as the STING-RANK, which are known to play key roles in the development of immune responses during DNA immunizations through expression of type I interferons and proinflammatory cytokines [21]. We cannot exclude that electroporations, similarly to conventional DNA injections, mediate nucleic acid entry into the cells through endocytosis as well [22]. During that process, unmethylated CpG motif from the DNA backbone could activate the endosomal TLR9 [23], although it was clearly demonstrated that DNA vaccine efficiency was TLR9-independent [24]. We next evaluated the efficacy of poly I:C as an adjuvant for the DNA-based vaccine candidate. Poly I:C is not only a ligand for the endosomal TLR3 receptor but also for melanoma differentiation-associated gene-5 (MDA-5), another nucleic acid cytosolic sensor [25]. Co-administration of DNA-encoding LipL32 with the poly I:C adjuvant during electroporation improved the humoral as well as the cellular
responses, confirming the interest to improve cytosolic innate immune sensing of nucleic acids for the optimization of DNA vaccinations. Strikingly, the homologous primeboost DNA/Poly I:C vaccinations were even more efficient than heterologous prime-boost DNA ? protein/Poly I:C immunization to generate antigen-specific responses. In conclusion, our study showed that, although DNA vaccine candidates encoding pathogenic bacterial lipoproteins could be poorly immunogenic in mammals through the absence of lipid anchor in the expressed antigens, the in vivo DNA delivery by electroporation as well as the poly I:C adjuvantation could enhance their immunogenicity. Such induction of specific antibody as well as T cell response against a bacterial lipoprotein following DNA immunizations could be critical to maximize the protective immunity. It could be particularly the case when the development of potent antibody response commonly triggered with recombinant bacterial lipoprotein-based vaccines formulated with adjuvants has limited effect on protection. It would be interesting in the near future to validate our LipL32-based DNA vaccine optimization in a hamster model of leptospirosis [26]. Acknowledgments This work was supported by the National Research Council of Thailand and the Higher Education Research Promotion and the National Research University Project of Thailand, Office of the Higher Education Commission (HR1164A2). Dr. Arun Buaklin was supported by a Post-doctoral fellowship from Chulalongkorn University Graduate School.
References 1. Williams, J. A. (2013). Vector design for improved DNA vaccine efficacy, safety and production. Vaccines, 1, 225–249. 2. Kutzler, M. A., & Weiner, D. B. (2008). DNA vaccines: ready for prime time? Nature Reviews Genetics, 9, 776–788. 3. Ingolotti, M., Kawalekar, O., Shedlock, D. J., Muthumani, K., & Weiner, D. B. (2010). DNA vaccines for targeting bacterial infections. Expert Review Vaccines., 9, 747–763. 4. Kovacs-Simon, A., Titball, R. W., & Michell, S. L. (2011). Lipoproteins of bacterial pathogens. Infection and Immunity, 79, 548–561. 5. Schenk, M., Belisle, J. T., & Modlin, R. L. (2009). TLR2 looks at lipoproteins. Immunity, 31, 847–849. 6. Haake, D. A., Chao, G., Zuerner, R. L., Barnett, J. K., Barnett, D., Mazel, M., et al. (2000). The leptospiral major outer membrane protein LipL32 is a lipoprotein expressed during mammalian infection. Infection and Immunity, 68, 2276–2285. 7. Yang, C. W., Hung, C. C., Wu, M. S., Tian, Y. C., Chang, C. T., Pan, M. J., et al. (2006). Toll-like receptor 2 mediates early inflammation by leptospiral outer membrane proteins in proximal tubule cells. Kidney International, 69, 815–822. 8. Hsu, S. H., Lo, Y. Y., Tung, J. Y., Ko, Y. C., Sun, Y. J., Hung, C. C., et al. (2010). Leptospiral outer membrane lipoprotein LipL32 binding on toll-like receptor 2 of renal cells as determined with an atomic force microscope. Biochemistry, 49, 5408–5417. 9. Massaer, M., Mazzu, P., Haumont, M., Magi, M., Daminet, V., Bollen, A., et al. (2001). High-level expression in mammalian
123
Mol Biotechnol
10.
11.
12.
13.
14.
15.
16.
cells of recombinant house dust mite allergen ProDer p 1 with optimized codon usage. International Archives of Allergy and Immunology, 125, 32–43. Baldwin, S. L., D’Souza, C. D., Orme, I. M., Liu, M. A., Huygen, K., Denis, O., et al. (1999). Immunogenicity and protective efficacy of DNA vaccines encoding secreted and non-secreted forms of Mycobacterium tuberculosis Ag85A. Tubercle and Lung Disease, 79, 251–259. Wang, S., Farfan-Arribas, D. J., Shen, S., Chou, T. H., Hirsch, A., He, F., et al. (2005). Relative contributions of codon usage, promoter efficiency and leader sequence to the antigen expression and immunogenicity of HIV-1 Env DNA vaccine. Vaccine, 24, 4531–4540. Jacquet, A., Magi, M., Haumont, M., Jurado, M., Garcia, L., & Bollen, A. (2003). Absence of immunoglobulin E synthesis and airway eosinophilia by vaccination with plasmid DNA encoding ProDer p 1. Clinical and Experimental Allergy, 33, 218–225. Flingai, S., Czerwonko, M., Goodman, J., Kudchodkar, S. B., Muthumani, K., & Weiner, D. B. (2013). Synthetic DNA vaccines: improved vaccine potency by electroporation and codelivered genetic adjuvants. Frontiers in Immunology, 4, 354. Grossmann, C., Tenbusch, M., Nchinda, G., Temchura, V., Nabi, G., Stone, G. W., et al. (2009). Enhancement of the priming efficacy of DNA vaccines encoding dendritic cell-targeted antigens by synergistic toll-like receptor ligands. BMC Immunology, 10, 43–48. Tewari, K., Flynn, B. J., Boscardin, S. B., Kastenmueller, K., Salazar, A. M., Anderson, C. A., et al. (2010). Poly(I:C) is an effective adjuvant for antibody and multi-functional CD4 ? T cell responses to Plasmodium falciparum circumsporozoite protein (CSP) and aDEC-CSP in non human primates. Vaccine, 28, 7256–7266. Ranasinghe, C., & Ramshaw, I. A. (2009). Genetic heterologous prime-boost vaccination strategies for improved systemic and mucosal immunity. Expert Review of Vaccines, 8, 1171–1181.
123
17. Lugade, A. A., Bianchi-Smiraglia, A., Pradhan, V., Elkin, G., Murphy, T. F., & Thanavala, Y. (2011). Lipid motif of a bacterial antigen mediates immune responses via TLR2 signaling. PLoS ONE, 6, e19781. 18. Hansson, L., Noppa, L., Nilsson, A. K., Stromqvist, M., & Bergstrom, S. (1995). Expression of truncated and full-length forms of the Lyme disease Borrelia outer surface protein a in Escherichia coli. Protein Expression and Purification, 6, 15–24. 19. Shang, E. S., Summers, T. A., & Haake, D. A. (1996). Molecular cloning and sequence analysis of the gene encoding LipL41, a surface exposed lipoprotein of pathogenic Leptospira species. Infection and Immunity, 64, 2322–2330. 20. Hall, J., Hazlewood, G. P., Surani, M. A., Hirst, B. H., & Gilbert, H. J. (1990). Eukaryotic and prokaryotic signal peptides direct secretion of a bacterial endoglucanase by mammalian cells. Journal of Biological Chemistry, 265, 19996–19999. 21. Coban, C., Kobiyama, K., Aoshi, T., Takeshita, F., Horii, T., Akira, S., et al. (2011). Novel strategies to improve DNA vaccine immunogenicity. Current Gene Therapy, 11, 479–484. 22. Rosazza, C., Buntz, A., Rieß, T., Wo¨ll, D., Zumbusch, A., & Rols, M.-P. (2013). Intracellular tracking of single-plasmid DNA particles after delivery by electroporation. Molecular Therapy, 21, 2217–2226. 23. Blasius, A. L., & Beutler, B. (2010). Intracellular Toll-like receptors. Immunity, 32, 305–315. 24. Babiuk, S., Mookherjee, N., Pontarollo, R., Griebel, P., van Drunen Littel-van den Hurk, S., Hecker, R., et al. (2004). TLR9-/- and TLR9?/?mice display similar immune responses to a DNA vaccine. Immunology, 113, 114–120. 25. Vabret, N., & Blander, J. M. (2013). Sensing microbial RNA in the cytosol. Front: Immunol. doi:10.3389/fimmu.2013.00468. 26. Coutinho, M. L., Choy, H. A., Kelley, M. M., Matsunaga, J., Babbitt, J. T., Lewis, M. S., et al. (2011). A LigA three-domain region protects hamsters from lethal infection by Leptospira interrogans. PLOS Neglected Tropical Diseases, 5, e1422.
RESEARCH
Optimization of the Immunogenicity of a DNA Vaccine Encoding a Bacterial Outer Membrane Lipoprotein Arun Buaklin • Tanapat Palaga • Drew Hannaman Ruthairat Kerdkaew • Kanitha Patarakul • Alain Jacquet
•
Ó Springer Science+Business Media New York 2014
Abstract Bacterial outer membrane lipoproteins represent potent immunogens for the design of recombinant subunit vaccines. However, recombinant lipoprotein production and purification could be a challenge notably in terms of expression yield, protein solubility, and posttranslational acylation. Together with the cost effectiveness, facilitated production, and purification as well as good stability, DNA-based vaccines encoding lipoproteins could become an alternative strategy for antibacterial vaccinations. Although the immunogenicity and the efficacy of DNA-based vaccines can be demonstrated in small rodents, such vaccine candidates could request concrete
Present Address: A. Buaklin K. Patarakul (&) Department of Microbiology, Faculty of Medicine, Chulalongkorn University, Oor-Por-Ror Building, 15th floor, Room # 1510B2 1873 Rama IV Road, Pathum Wan, Bangkok 10330, Thailand e-mail: [email protected]; [email protected] T. Palaga Department of Microbiology, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand D. Hannaman Ichor Medical Systems, San Diego, CA 92121, USA R. Kerdkaew Department of Internal Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok 10330, Thailand Present Address: A. Jacquet (&) Department of Medicine, Faculty of Medicine, Chulalongkorn University, Pattayapat Building, 8th floor, Room # 807, 1873 Rama IV Road, Pathumwan, Bangkok 10330, Thailand e-mail: [email protected]
optimization as they are weak immunogens in primates and humans and particularly when administered by conventional injection. Therefore, the goal of the present study was to optimize the immunogenicity of a DNA vaccine encoding an outer membrane lipoprotein. LipL32, the major outer membrane protein from pathogenic Leptospira, was selected as a model antigen. We evaluated the influence of antigen secretion, the in vivo DNA delivery by electroporation, the adjuvant co-administration, as well as the heterologous prime-boost regimen on the induction of anti-LipL32 specific immune responses. Our results clearly showed that, following transfections, a DNA construct based on the authentic full-length LipL32 gene (containing leader sequence and the N-terminus cysteine residue involved in the protein anchoring) drives antigen secretion with the same efficiency as a plasmid-encoding anchor-less LipL32 and for which the bacterial leader sequence was replaced with a viral signal peptide. The in vivo DNA delivery by electroporation drastically enhanced the production of strong Th1 responses characterized by specific IgG2a antibodies and the IFNc secretion in a restimulation assay, regardless of the DNA constructs used. In comparison with the heterologous prime-boost regimen, the homologous prime-boost vaccinations with DNA coadministrated with polyinosinic-polycytidylic acid (poly I:C) generated the highest specific IgG and IgG2a titers as well as the greatest IFNc production. Taken together, these data suggest that optimization of outer membrane lipoprotein secretion is not critical for the induction of antigenspecific responses through DNA vaccination. Moreover, the potent antibody response induced by DNA plasmid encoding lipoprotein formulated with poly I:C and delivered through electroporation provides the rationale for the design of new prophylactic vaccines against pathogenic bacteria.
123
Mol Biotechnol
Keywords Bacterial lipoprotein DNA vaccine Electroporation Poly I:C Prime-boost LipL32
Materials and Methods Production of Recombinant Anchor-Less Mature LipL32 Protein (rLipL32 DCys1)
Introduction DNA vaccines are commonly composed of a plasmid backbone encoding the antigen of interest under the control of a strong eukaryotic promoter [1]. Such recombinant plasmids, following intramuscular or intradermal administrations, lead to antigen expression and presentation through direct in vivo transfection of antigen-presenting cells such as dendritic cells (DCs). Alternatively, the expressed antigen from transfected myocytes or other bystander cells can be transferred to DCs for the crosspriming. Consequently, MHC class I and II antigen presentation is elicited through DNA vaccinations [2]. Compared with vaccines based on recombinant proteins, DNA vaccines offer several advantages including high stability, cost effectiveness of DNA production, and simple purification process. In rodents, DNA-encoding antigens preferentially induce the production of IgG2a antibodies, CD8? and CD4? type 1 T cell responses, and IFNc secretion [3]. However, DNA vaccines suffer from low or modest immunogenicity in humans [2, 3]. Hence, strategies are needed to significantly improve the efficiency of DNA vaccinations through the optimization of the antigen expression and secretion, the in vivo DNA delivery, the DNA formulation with adjuvants, or the heterologous prime-boost vaccinations [1]. Lipoproteins from Gram-positive and Gram-negative bacterial pathogens represent a large class of membrane proteins displaying numerous biological functions such as adhesion, immunomodulation, or virulence [4]. These membrane proteins display a fatty acid-based anchor covalently bound to the N-terminal cysteine. According to these key roles and to their strong immunogenicity through notably their fatty acid moieties [5], these bacterial lipoproteins can be considered as promising vaccine candidates, which, following immunizations, could promote the development of protective antibody responses. The goal of the present study was to design and to optimize a DNA vaccine encoding a bacterial lipoprotein. As a model antigen, we selected LipL32, the most abundant Leptospira outer membrane lipoprotein [6]. Moreover, LipL32 is expressed during Leptospira infection in vivo and was shown to display strong immunogenicity by notably TLR2 activation through its fatty acid moieties [7, 8]. Our results clearly highlight the importance of adjuvant formulations and in vivo delivery systems for the immunogenicity of lipoprotein-based DNA vaccines.
123
Recombinant His-tagged anchor-less mature LipL32 (rLipL32 DCys1) was cloned according to a previously published strategy [6]. DNA-encoding LipL32DCys1 was amplified from Leptospira interrogans serovar Pomona genomic DNA as a template and using the following primers: Forward 50 -TTACCGCTCGAGGTGCTTTCGGTGGTCT GC-30 (XhoI underlined) and Reverse 50 -TGTTAACCC GGGTTACTTAGTCGCGTCAGA-30 (SmaI underlined). PCR product was cloned into the pGem-T easy vector (Promega). The recombinant amplicon was sequenced and subsequently digested with XhoI and SmaI for the cloning into the pRSETc bacterial expression vector. BL21(DE3) pLysS E.coli cells were transformed with pRSETc-LipL32, and rLipL32 DCys1 expression was induced by IPTG addition into the culture medium. The expressed protein was purified using immobilized Ni2? as previously described [6]. Construction and Preparation of DNA Plasmids Encoding LipL32 The full-length LipL32 (leader ? mature sequence, LWTLipL32) DNA was cloned into the pVITRO1 (Invivogen, USA) mammalian expression vector following BspEI and BsiWI restriction sites of the recombinant pGEM-T easy vector (Promega, USA). To design another DNA plasmid expressing anchor-less LipL32 (DCys1) and in which the LipL32 leader sequence was replaced with the viral VZV gE signal peptide [9] (LVir-DCys1-LipL32), a PCR product was obtained using the following primers: forward 50 -G GATCCGGAGCCACCATGGGCACCGTGAACAAGCC CGTCGT GGGCGTGCTGATGGGCTTCGGCATCATC ACCGGCACCCTGCGGATCACCAACCCCGTGCGGG CCGGTGCTTTCGGTGGTCTGCCAAGC-30 (BspEI underlined and Kozak sequence in bold) and reverse 50 -TA ACGTACGTTACTTAGTCGCGTCAG-30 (BsiWI underlined). A plasmid encoding leader-less LipL32 was also designed by PCR amplification of LipL32 DNA using the same 30 reverse primer used for LVir-DCys1-LipL32 construction and the following 50 primer: 50 -GGATCCGGA GC CACCATGTGTGGTGCTTTCGGTGGTCTGCCAAGC-30 (BspEI underlined and Kozak sequence in bold). The two resulting amplicons were subsequently digested with BspEI and BsiWI for the cloning into pVITRO1. All plasmid DNA constructs were produced and purified under endotoxin-free condition using GigaPrep Kits (QIAGEN, Germany) for in vitro transfection and animal immunization.
Mol Biotechnol
In Vitro Transient Transfection Assays
LipL32-Specific Antibody Assay
Human proximal tubular HK-2 cells were cultured onto six-well plates in Dulbecco’s modified Eagle’s medium (DMEM) GlutaMAXTM (Invitrogen, Carlsbad, USA) supplemented with 10 % fetal bovine serum (FBS) (Invitrogen, Carlsbad, USA) at a density of 1 9 106 cells/well. At a 90 % confluency, the endotoxin-free plasmid DNA (4 lg of each DNA) were diluted in 250 lL of serum-free DMEM medium, gently mixed and incubated with 250 lL of a dilute solution of lipofectamineTM 2000 (Invitrogen) for 20 min at room temperature. The lipoplexes were added into each well plate and incubated at 37 °C in a CO2 incubator for 72 h. The culture supernatants were subsequently collected, and cells were washed twice with icecold PBS and lysed in 1X SDS lysis buffer (50 mM Tris, pH 6.8, 50 mM DTT, 2 % SDS, 10 % glycerol, 100 lg/ml leupeptin, and 0.1 % bromophenol blue) for 10 min on ice before cell lysates were collected. Culture supernatants and cell lysates were analyzed by Western blot for LipL32 expression using specific mouse polyclonal antibodies generated by rLipL32 DCys1 immunization (data not shown).
For antibody detection, rLipL32 DCys1 was directly coated onto ELISA well plates (500 ng/well) with carbonate buffer (pH 9.6) overnight at 4 °C. Plates were washed six times with phosphate-buffer saline (PBS) containing 0.05 % Tween 20 (PBS-T) and then blocked with 0.5 % (wt/v) bovine serum albumin in PBS-T for 1 h at room temperature. After blocking, wells were incubated with serial dilutions of each mouse serum (1:100–1:312,500 for specific IgG, 1:30–1:93,750 for both specific IgG1 and IgG2a) at 37 °C for 1 h. After washing with PBS-T, wells were incubated with horseradish peroxidase-conjugated goat anti-mouse IgG, IgG1, or IgG2a antibodies (dilution 1:5,000 for each antibody, BD Biosciences, USA) at 37 °C for 1 h. The antigen–antibody complexes were detected by adding 100 ll of TMB substrate solution, and the enzymatic reaction was stopped by the addition of 50 ll of 0.5 M H2SO4. The optical density was read at 450 nm using a Bio-Rad microplate reader.
Immunizations The present study was approved by the Chulalongkorn University Animal Care and Use Committee (CU-ACUC). In the first experiment, groups of six BALB/c mice were immunized three times at two-week intervals with 20 lg (in 20 ll) of pVITRO1 encoding LWT-LipL32 or LVir-DCys1-LipL32. As a control group, animals were immunized with empty pVITRO1. Plasmid DNA constructs were administrated intramuscularly (IM) or were delivered by in vivo electroporation (EP) using the TriGridTM Delivery System (rodent model, Ichor Medical Systems, www.ichorms.com). The IM DNA injection into one tibialis anterior muscle was directly followed by the stimulation at the site of injection with an electrical pulse that had an amplitude of 250 V per cm of electrode spacing for 40 ms over a 400 ms interval. In a second set of the experiments, homologous primeboost regimen (three electroporations with 20 lg DNALVir-DCys1-LipL32) was compared with heterologous prime-boost immunizations (two in vivo electroporations followed by one subcutaneous injection of 5 lg rLipL32 DCys1 in the presence or absence of 10 lg poly I:C (Invivogen). Two weeks after the last immunization, sera were collected from each mouse by retro-orbital bleeding for antibody assay and spleens were collected from sacrificed animals.
Cytokine Assays To assess the cytokine production of proliferative T cells, splenocytes (5 9 105/well in triplicate) were stimulated with 20 lg/mL rLipL32 DCys1, 10 lg/ml concanavalin A (ConA), or RPMI1640 medium alone for 72 h and culture supernatants were collected after 72 h. The levels of IFNc and IL-4 in the supernatants were measured by ELISA assays according to the BD OptiEIA kits (BD, USA) manufacturer’s instructions. Statistical Analysis Student’s t test was used to compare data between pairs of groups. Differences were considered significant for P values \0.05.
Results Construction and Expression of DNA Vaccines in HK-2 Cells To design a DNA vaccine candidate based on the LipL32 leptospiral lipoprotein, the DNA-encoding full-length LipL32 (LWT-LipL32) was cloned into the pVITRO1 eukaryotic expression vector. In this DNA construct, the absence of the typical Kozak sequence, a mammalian consensus site immediately prior to the ATG start codon which may further enhance antigen expression [1], together with the presence of the natural bacterial leader sequence
123
Mol Biotechnol
Fig. 1 LipL32 expression and secretion from transiently transfected HK-2 cells. HK-2 cells were transiently transfected with empty vector or vectors encoding LipL32. a N-terminal amino acid sequences of LipL32 encoded by DNA-LWT-LipL32, DNA-LVir-DCys1-LipL32 and DNA-Leader-less LipL32. The first amino acids of mature LipL32 are underlined, Antigen expression was monitored at 72 h post-transfection in the supernatants (lanes 2, 4, 6, and 8) and in the cell lysates (lanes 3, 5, 7 and 9). Lane 1 purified rLipL32 DCys1, Lanes 2 and 3 empty vector, Lanes 4 and 5 DNA-Leader-less LipL32, Lanes 6 and 7 DNA-LWT-LipL32, Lanes 8 and 9 DNA-LVir-DCys1LipL32. Western blot analysis was performed using anti-LipL32 mouse polyclonal antibodies (b) and monoclonal antibodies to b-actin (c)
as well as the N-terminal Cys residue involved in the anchoring of the protein in the bacterial outer membrane could have negative impacts on antigen expression and secretion. Notably, sub-optimal antigen secretion could impair the DNA plasmid immunogenicity as it was previously reported that optimization of antigen secretion enhances the antibody production rate [10, 11]. Consequently, we designed a second DNA vaccine (LVir-DCys1LipL32) in which the natural leader sequence was replaced with that from the VZV gE antigen, a leader sequence which triggered efficient secretion in mammalian cells [9, 12]. Moreover, a Kozak sequence was also inserted and the N-terminal Cys residue was removed in order to express DCys1-LipL32. The expression and secretion of LipL32 were evaluated through transient expression in HK-2 cells transfected by both DNA constructs. As controls, cells were transfected with the empty vector or a plasmid-encoding Leader-less LipL32. LipL32 expression could be detected in transfected cells as well as in the culture supernatant with both pVITRO-LipL32 vaccine candidates (Fig. 1). Cell transfection with a plasmid without a leader sequence induced strictly intracellular LipL32 production, confirming the importance of a leader sequence for the antigen secretion. Densitometric analysis (after normalization for difference in the b-actin bands between the two cell lysates)
123
demonstrated similar levels of intracellular antigen expression whatever the DNA construct used. Surprisingly, the DNA plasmid encoding LVir-DCys1-LipL32 did not significantly drive higher antigen secretion (P [ 0.05). Similar levels of LipL32 expression and secretion were obtained using HEK293 cells (human embryonic kidney cells) in transfection assays (data not shown). These results suggested that the natural LipL32 leader sequence is functional to trigger secretion and that the N-terminal cysteine residue is more likely not acylated in mammalian cells. It must be pointed out that secreted LipL32 displayed a higher molecular weight than the intracellular form. This could be more likely explained by the presence of N-linked glycosylation at the level of the putative N-glycosylation site (N25ET). Recombinant purified rLipL32 DCys1 displayed also a higher molecular weight by the presence of an N-terminus extension containing His-tag and protease cleavage site. Immunogenicity of DNA-LWT-LipL32 and DNA-LVirDCys1-LipL32 To determine whether both DNA constructs encoding secreted LipL32 are immunogenic, the two DNA vaccine candidates (20 lg each) were delivered three times in BALB/c mice by electroporation or conventional intramuscular injection. Indeed, numerous studies highlighted the interest of in vivo electroporation to improve the immunogenicity of DNA-based vaccines [13]. The specific humoral as well as cellular responses were analyzed through antiLipL32 total IgG, IgG1, IgG2a detection in sera as well as IFNc and IL-4 secreted from LipL32 restimulated splenocytes (Fig. 2). Modest but comparable specific antibody productions were elicited following intramuscular immunizations with DNA-LWT-LipL32 and DNA-LVir-DCys1LipL32; in contrast, electroporations drastically increased the production of anti-LipL32 IgG, IgG1, and IgG2a. However, no significant difference in the antibody titers induced by two DNA constructions were observed (P [ 0.05). Cytokine detection from lymphoproliferative assays demonstrated that both DNA constructs similarly triggered elevated amount of IFNc but not of IL-4 (data not shown), regardless of the type of DNA delivery used. However, higher IFNc secretion was detected following electroporation (P \ 0.05). According to the low specific IgG1/IgG2a ratio and high IFNc production together with the absence of IL-4, our results clearly confirmed that DNA immunizations through electroporation induced a strong anti-LipL32 Th1-biased response. Moreover, we can consider that the immunogenicity level of DNA-LWT-LipL32 and DNA-LVir-DCys1LipL32 was correlated with the similar LipL32 expression/ secretion level in our transient transfection assays.
Mol Biotechnol
Fig. 2 Immunogenicity of DNA-LWT-LipL32 and DNA-LVir-DCys1LipL32 Specific IgG (a), IgG1 (b) and IgG2a (c) titers postvaccinations, restimulated splenocyte IFNg production (d). All the
values were plotted as mean ± SEM. One representative experiment out of 2 is shown. *P \ 0.05 (comparisons between treatments). EP electroporation, IM intramuscular
Effect of the Poly I:C Adjuvant and Prime-Boost Regimen on Immunogenicity of DNA-LVir-DCys1LipL32
induced by DNA-LVir-DCys1-LipL32 through three electroporations (homologous prime-boost) and two electroporations combined with one subcutaneous immunization with rLipL32-DCys1 (heterologous prime-boost.). The same immunization protocol was applied to two other mice groups, but the DNA plasmid as well the recombinant antigen were adjuvanted with Poly I:C. Whereas comparable specific IgG1 antibody titers were elicited whatever the presence or the absence of Poly I:C and the prime-boost regimen, this TLR ligand significantly but uniquely improved the production of anti-LipL32 IgG2a through the homologous prime-boost immunizations (three DNA electroporations) (Fig. 3) (P \ 0.05). A similar and significant increase in IFNc production was also detected in this group of immunized mice. Once again, no IL-4 production was triggered regardless of the immunization regimen.
As both DNA constructs encoding LipL32 displayed similar immunogenicity when delivered by electroporation, DNA-LVir-DCys1-LipL32 was selected to explore the influence of co-injection with Poly I:C as an adjuvant as well as the DNA prime-boost regimen on the optimization of the DNA vaccine immunogenicity. Indeed, it was previously demonstrated that adjuvanting with poly I:C or immunizing using different types of vaccines for the delivery of the same antigen in the priming and the boost steps can increase antigen-specific T cell and B cell responses [14–16]. Accordingly, another set of the experiments was performed to compare the specific immune responses
123
Mol Biotechnol
Fig. 3 Effect of the Poly I:C adjuvant and prime-boost regimen on immunogenicity of DNA-LVir-DCys1-LipL32 Specific IgG (a), IgG1 (b) and IgG2a (c) titers post-vaccinations, restimulated splenocyte
IFNg production (d). All the values were plotted as mean ± SEM. One representative experiment out of 2 is shown. *P \ 0.05 (comparisons between treatments)
Discussion
Compared with conventional protein vaccines, DNA vaccines encoding bacterial lipoproteins could be attractive since their development is relatively easy and their production cost is low. As antigen secretion could drastically influence the development of specific antibody response, it is believed that genetic constructions encoding lipoproteins must be optimized to improve the extracellular release of the expressed proteins following immunizations. It is unlikely that bacterial lipoproteins expressed in mammalian cells display N-terminal cysteine acylation as in the manner of prokaryotes. Indeed, the three conserved enzymes, prolipoprotein diacylglyceryltransferase (Lgt), prolipoprotein signal peptidase (LspA), and phospholipid:apolipoprotein N-acyltransferase (Lnt), responsible for the post-translational lipid modification after recognition of the lipobox sequence at the end of lipoprotein signal peptide are absent in eukaryotes [4, 5]. Moreover, whereas bacterial leader peptides can direct bacterial protein secretion in mammalian cells [20], they could be less
Outer membrane lipoproteins from bacterial pathogens could represent efficient vaccine candidates [3]. Indeed such lipoproteins are highly immunogenic because of their lipid adjuvant motifs which trigger potent innate immune response through notably TLR2 activation [17]. However, due to their lipid motif anchored to the membrane, isolation of bacterial natural lipoproteins could represent an issue in terms of solubility, and consequently requests the use of detergents. Recombinant expression of lipoproteins in E. coli can also be problematic and is commonly associated with inappropriate lipid modifications [18, 19]. To prevent such issues, the alternative is to produce fully soluble anchor-less lipoproteins which more likely could be less immunogenic than the corresponding anchored protein. Nevertheless, such purified non-lipidated bacterial antigens could be formulated with potent adjuvants, notably TLR ligands to up-regulate their vaccine potential.
123
Mol Biotechnol
efficient than eukaryotic or viral signal sequence to initiate expression and secretion into animal cells. In the present study, two DNA plasmids encoding the Leptospira outer membrane protein LipL32 as a bacterial lipoprotein prototype were designed: one construction encoding the full-length protein encompassing the natural leader sequence with the lipobox and a second one in which the natural leader sequence was replaced with that from the VZV gE antigen. This viral leader peptide was previously shown to trigger efficiently heterologous antigen secretion in recombinant eukaryotic cells [9, 12]. Our transfection assays strikingly suggested that the leader sequence of bacterial lipoproteins is functional in mammalian cells to allow antigen secretion. The similar level of LipL32 secretion, whatever the DNA plasmid used, confirmed as well that the bacterial lipoprotein is more likely not acylated when produced in eukaryotic cells. As the lipid anchor of bacterial lipoprotein is a strong adjuvant, through notably TLR2 signaling, such DNA vaccine candidates encoding anchor-less LipL32 could display poor immunogenicity [17]. In order to counterbalance the absence of these lipid moieties, we evaluated the interest of in vivo DNA delivery by electroporation to optimize the DNA immunogenicity. Our data highlighted that whereas very low specific immune response was elicited by the two DNA vaccine candidates following intramuscular administrations, electroporation drastically enhanced the overall level of specific humoral response as well as IFNc production to LipL32. To our knowledge, this is the first electroporationbased immunization protocol using a DNA-encoding nonacylated bacterial lipoprotein. The boost of the immune response following electroporation can more likely explained by a direct access of DNA to the cytoplasm leading to an efficient activation of cytosolic DNA sensors as the STING-RANK, which are known to play key roles in the development of immune responses during DNA immunizations through expression of type I interferons and proinflammatory cytokines [21]. We cannot exclude that electroporations, similarly to conventional DNA injections, mediate nucleic acid entry into the cells through endocytosis as well [22]. During that process, unmethylated CpG motif from the DNA backbone could activate the endosomal TLR9 [23], although it was clearly demonstrated that DNA vaccine efficiency was TLR9-independent [24]. We next evaluated the efficacy of poly I:C as an adjuvant for the DNA-based vaccine candidate. Poly I:C is not only a ligand for the endosomal TLR3 receptor but also for melanoma differentiation-associated gene-5 (MDA-5), another nucleic acid cytosolic sensor [25]. Co-administration of DNA-encoding LipL32 with the poly I:C adjuvant during electroporation improved the humoral as well as the cellular
responses, confirming the interest to improve cytosolic innate immune sensing of nucleic acids for the optimization of DNA vaccinations. Strikingly, the homologous primeboost DNA/Poly I:C vaccinations were even more efficient than heterologous prime-boost DNA ? protein/Poly I:C immunization to generate antigen-specific responses. In conclusion, our study showed that, although DNA vaccine candidates encoding pathogenic bacterial lipoproteins could be poorly immunogenic in mammals through the absence of lipid anchor in the expressed antigens, the in vivo DNA delivery by electroporation as well as the poly I:C adjuvantation could enhance their immunogenicity. Such induction of specific antibody as well as T cell response against a bacterial lipoprotein following DNA immunizations could be critical to maximize the protective immunity. It could be particularly the case when the development of potent antibody response commonly triggered with recombinant bacterial lipoprotein-based vaccines formulated with adjuvants has limited effect on protection. It would be interesting in the near future to validate our LipL32-based DNA vaccine optimization in a hamster model of leptospirosis [26]. Acknowledgments This work was supported by the National Research Council of Thailand and the Higher Education Research Promotion and the National Research University Project of Thailand, Office of the Higher Education Commission (HR1164A2). Dr. Arun Buaklin was supported by a Post-doctoral fellowship from Chulalongkorn University Graduate School.
References 1. Williams, J. A. (2013). Vector design for improved DNA vaccine efficacy, safety and production. Vaccines, 1, 225–249. 2. Kutzler, M. A., & Weiner, D. B. (2008). DNA vaccines: ready for prime time? Nature Reviews Genetics, 9, 776–788. 3. Ingolotti, M., Kawalekar, O., Shedlock, D. J., Muthumani, K., & Weiner, D. B. (2010). DNA vaccines for targeting bacterial infections. Expert Review Vaccines., 9, 747–763. 4. Kovacs-Simon, A., Titball, R. W., & Michell, S. L. (2011). Lipoproteins of bacterial pathogens. Infection and Immunity, 79, 548–561. 5. Schenk, M., Belisle, J. T., & Modlin, R. L. (2009). TLR2 looks at lipoproteins. Immunity, 31, 847–849. 6. Haake, D. A., Chao, G., Zuerner, R. L., Barnett, J. K., Barnett, D., Mazel, M., et al. (2000). The leptospiral major outer membrane protein LipL32 is a lipoprotein expressed during mammalian infection. Infection and Immunity, 68, 2276–2285. 7. Yang, C. W., Hung, C. C., Wu, M. S., Tian, Y. C., Chang, C. T., Pan, M. J., et al. (2006). Toll-like receptor 2 mediates early inflammation by leptospiral outer membrane proteins in proximal tubule cells. Kidney International, 69, 815–822. 8. Hsu, S. H., Lo, Y. Y., Tung, J. Y., Ko, Y. C., Sun, Y. J., Hung, C. C., et al. (2010). Leptospiral outer membrane lipoprotein LipL32 binding on toll-like receptor 2 of renal cells as determined with an atomic force microscope. Biochemistry, 49, 5408–5417. 9. Massaer, M., Mazzu, P., Haumont, M., Magi, M., Daminet, V., Bollen, A., et al. (2001). High-level expression in mammalian
123
Mol Biotechnol
10.
11.
12.
13.
14.
15.
16.
cells of recombinant house dust mite allergen ProDer p 1 with optimized codon usage. International Archives of Allergy and Immunology, 125, 32–43. Baldwin, S. L., D’Souza, C. D., Orme, I. M., Liu, M. A., Huygen, K., Denis, O., et al. (1999). Immunogenicity and protective efficacy of DNA vaccines encoding secreted and non-secreted forms of Mycobacterium tuberculosis Ag85A. Tubercle and Lung Disease, 79, 251–259. Wang, S., Farfan-Arribas, D. J., Shen, S., Chou, T. H., Hirsch, A., He, F., et al. (2005). Relative contributions of codon usage, promoter efficiency and leader sequence to the antigen expression and immunogenicity of HIV-1 Env DNA vaccine. Vaccine, 24, 4531–4540. Jacquet, A., Magi, M., Haumont, M., Jurado, M., Garcia, L., & Bollen, A. (2003). Absence of immunoglobulin E synthesis and airway eosinophilia by vaccination with plasmid DNA encoding ProDer p 1. Clinical and Experimental Allergy, 33, 218–225. Flingai, S., Czerwonko, M., Goodman, J., Kudchodkar, S. B., Muthumani, K., & Weiner, D. B. (2013). Synthetic DNA vaccines: improved vaccine potency by electroporation and codelivered genetic adjuvants. Frontiers in Immunology, 4, 354. Grossmann, C., Tenbusch, M., Nchinda, G., Temchura, V., Nabi, G., Stone, G. W., et al. (2009). Enhancement of the priming efficacy of DNA vaccines encoding dendritic cell-targeted antigens by synergistic toll-like receptor ligands. BMC Immunology, 10, 43–48. Tewari, K., Flynn, B. J., Boscardin, S. B., Kastenmueller, K., Salazar, A. M., Anderson, C. A., et al. (2010). Poly(I:C) is an effective adjuvant for antibody and multi-functional CD4 ? T cell responses to Plasmodium falciparum circumsporozoite protein (CSP) and aDEC-CSP in non human primates. Vaccine, 28, 7256–7266. Ranasinghe, C., & Ramshaw, I. A. (2009). Genetic heterologous prime-boost vaccination strategies for improved systemic and mucosal immunity. Expert Review of Vaccines, 8, 1171–1181.
123
17. Lugade, A. A., Bianchi-Smiraglia, A., Pradhan, V., Elkin, G., Murphy, T. F., & Thanavala, Y. (2011). Lipid motif of a bacterial antigen mediates immune responses via TLR2 signaling. PLoS ONE, 6, e19781. 18. Hansson, L., Noppa, L., Nilsson, A. K., Stromqvist, M., & Bergstrom, S. (1995). Expression of truncated and full-length forms of the Lyme disease Borrelia outer surface protein a in Escherichia coli. Protein Expression and Purification, 6, 15–24. 19. Shang, E. S., Summers, T. A., & Haake, D. A. (1996). Molecular cloning and sequence analysis of the gene encoding LipL41, a surface exposed lipoprotein of pathogenic Leptospira species. Infection and Immunity, 64, 2322–2330. 20. Hall, J., Hazlewood, G. P., Surani, M. A., Hirst, B. H., & Gilbert, H. J. (1990). Eukaryotic and prokaryotic signal peptides direct secretion of a bacterial endoglucanase by mammalian cells. Journal of Biological Chemistry, 265, 19996–19999. 21. Coban, C., Kobiyama, K., Aoshi, T., Takeshita, F., Horii, T., Akira, S., et al. (2011). Novel strategies to improve DNA vaccine immunogenicity. Current Gene Therapy, 11, 479–484. 22. Rosazza, C., Buntz, A., Rieß, T., Wo¨ll, D., Zumbusch, A., & Rols, M.-P. (2013). Intracellular tracking of single-plasmid DNA particles after delivery by electroporation. Molecular Therapy, 21, 2217–2226. 23. Blasius, A. L., & Beutler, B. (2010). Intracellular Toll-like receptors. Immunity, 32, 305–315. 24. Babiuk, S., Mookherjee, N., Pontarollo, R., Griebel, P., van Drunen Littel-van den Hurk, S., Hecker, R., et al. (2004). TLR9-/- and TLR9?/?mice display similar immune responses to a DNA vaccine. Immunology, 113, 114–120. 25. Vabret, N., & Blander, J. M. (2013). Sensing microbial RNA in the cytosol. Front: Immunol. doi:10.3389/fimmu.2013.00468. 26. Coutinho, M. L., Choy, H. A., Kelley, M. M., Matsunaga, J., Babbitt, J. T., Lewis, M. S., et al. (2011). A LigA three-domain region protects hamsters from lethal infection by Leptospira interrogans. PLOS Neglected Tropical Diseases, 5, e1422.
