Induction of protective neutralizing antibody responses against botulinum neurotoxin serotype C using plasmid carried by PLGA nanoparticles

ABSTRACT

Botulinum neurotoxin (BoNT) is a lethal neurotoxin, for which there is currently not an approved vaccine. Recent efforts in developing vaccine candidates against botulism have been directed at the heavy chain fragment of BoNT, because antibodies against this region have been shown to prevent BoNT from binding to its receptor and thus to nerve cell surface, offering protection against BoNT intoxication. In the present study, it was shown that immunization with plasmid DNA that encodes the 50 KDa C-terminal fragment of the heavy chain of BoNT serotype C (i.e., BoNT/C-Hc50) and is carried by cationic poly (lactic-co-glycolic) acid (PLGA) nanoparticles induces stronger BoNT/C-specific antibody responses, as compared to immunization with the plasmid alone. Importantly, the antibodies have BoNT/C-neutralizing activity, protecting the immunized mice from a lethal dose of BoNT/C challenge. A plasmid DNA vaccine encoding the Hc50 fragments of BoNT serotypes that cause human botulism may represent a viable vaccine candidate for protecting against botulinum neurotoxin intoxication.

KEYWORDS:

• antibody responses
• botulinum neurotoxin
• mouse survival
• PLGA nanoparticles

BoNTs cause botulism, a lethal neuroparalytic disease. BoNTs are produced by one of the 7 structurally similar Clostridium botulinum serotypes, designated A to H.^1,2 Naturally, serotypes A, B, E and F cause human botulism, whereas serotypes C and D are responsible for animal botulism.³ BoNTs are one of the most poisonous substances known in nature.³ Previously, an investigational pentavalent botulinum toxoid (PBT) vaccine was available. However, the Center for Disease Control and Prevention (CDC) found that the old PBT vaccine (>30 years) generally has declined immunogenicity, decreased potency, and increased occurrence of adverse effects at injection sites.⁴ High manufacturing costs, long duration to produce a sufficient amount of the product, and safety issues in producing the toxins are some of the reasons preventing the PBT vaccine from being replaced frequently.⁴ As of November 2011, the PBT vaccine has been discontinued by the CDC.⁵ The PBT vaccine had other disadvantages as well. For example, the cost of manufacturing was very high. C. botulinum is a spore-former, so a dedicated cGMP facility was required to manufacture the toxin-based product; the yields of toxin production from C. botulinum were relatively low; it was dangerous to produce them, since the toxoiding process involves handling large quantities of toxins, and the added safety precautions increase the cost of manufacturing. The toxoid product for types A-E had a crude extract of clostridial proteins that may influence the immunogenicity or reactivity of the vaccine, and the type F toxoid was only partially purified.^6,7 In the past few years, there have been efforts in developing a new generation botulism vaccine, such as protein subunit vaccine or DNA vaccine.^8–11 In addition, botulism vaccines based on virus vectors, such as adenovirus vectors and Venezuelan equine encephalitis (VEE) virus replicon vector, were also tested.^4,12-14 Since several BoNT serotypes can cause human botulism, an effective botulism vaccine needs to be multivalent. Numerous reviews have dealt with current strategies of botulism vaccine development,^6,15-17 and several articles have described immunization using the heavy chain (Hc) of BoNT as antigens (e.g., BoNT A⁸, BoNT B¹⁸, BoNT C/D¹⁹, BoNT E²⁰ and BoNT F²¹). Plasmid DNA-based botulinum vaccine is appealing because: i) plasmid is relatively stable and thus feasible for long-term storage; ii) cost-effective method of manufacturing plasmid is available; iii) and genes encoding multiple antigens can be readily cloned into a plasmid to render it multivalent. In fact, plasmid DNA encoding the Hc domain of serotype A BoNT (BoNT/A) had been evaluated in an animal model,⁹ and Bennett et al. and Jathoul et al. had tested the immune responses induced by plasmid that expresses the binding domain of BoNT/F.^10,11 The BoNT Hc fragment is known to be non-toxic, antigenic, and capable of eliciting protective immune responses against botulism.^6,19,22,23 BoNT binds at the nerve terminal via its Hc to a cell-surface receptor, which consists of a ganglioside and a cell-surface protein.¹⁵ Antibodies against this region have been shown to prevent BoNT from binding to its receptor and thus to cell surface, offering protection against BoNT intoxication.^6,7,17 Previously, we showed that immunization with an adenovirus-based vector encoding an Hc 50-kDa fragment of BoNT/C elicits robust host immunity against BoNT/C intoxication.^4,22,24

In the present study, a plasmid that encodes the Hc50 fragment of BoNT/C (i.e., pVax/opt-BoNT/C-Hc50) was constructed with codon-optimization,^4,22 and its ability to induce BoNT/C-specific neutralizing antibodies in mice and protect the immunized mice against a BoNT/C challenge was evaluated. Mice were immunized intramuscularly with the plasmid, alone or coated on surface of previously reported cationic PLGA nanoparticles.²⁵ Immunization with plasmid DNA carried by cationic PLGA nanoparticles is known to induce stronger immune responses than with plasmid DNA alone, likely related to the ability of the plasmid DNA-coated nanoparticles to increase the expression of the antigen gene encoded by the plasmid and more effectively stimulate antigen-presenting cells.^25,26

To construct the pVax/opt-BoNT/C-Hc50, the nucleotides encoding the 50 kDa C-terminal fragment of the Hc of BoNT/C1 was optimized with human codon preference by the DNAworks program.²⁷ The codon-optimized BoNT/C-Hc50 fragment (encoding amino acids 849–1291 in BoNT/C, GenBank Acc# D90210) was then synthesized using a PCR-based method and cloned into the pVax1 vector (Life Technologies, Carlsbad, CA). The DNA sequence of the synthesized gene was confirmed by DNA sequencing analysis. The plasmid was purified using a QIAGEN Midiprep kit following the manufacturer's instruction (Valencia, CA). Large scale plasmid preparation was performed by GenScript (Piscataway, NJ). The plasmid was used, alone or coated on cationic PLGA nanoparticles, to immunize mice.²⁵ The cationic nanoparticles were prepared using PLGA (Resomer RG 504H, Sigma-Aldrich, St. Louis, MO) and 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP, Avanti Polar Lipids, Inc.., Alabaster, AL) as previously described (Fig. 1A).^25,28 The pVax/opt-BoNT/C-Hc50 plasmid was coated on the surface of the cationic nanoparticles by gently mixing equal volumes of cationic nanoparticles in suspension (with various amounts of nanoparticles) and the plasmid in water to obtain a final DNA concentration of 200 µg/ml, and the particle size and zeta potential values were measured using a Malvern Zetasizer® Nano ZS (Westborough, MA) (Fig. 1B). Plasmid DNA-coated PLGA nanoparticles prepared with a nanoparticle to plasmid DNA ratio of 40:1 (w/w) were used to immunize mice; and their particle size and zeta potential values were ∼120–140 nm and +47 ± 4 mV, respectively.

Figure 1. (A) Representative dynamic light scattering spectrum of typical PLGA nanoparticles. (B) The size and zeta potential of pVax/Opt-BoNT/C-Hc50-coated cationic PLGA nanoparticles at various nanoparticles to plasmid ratios. Increasing amounts of cationic nanoparticles (0.002–0.08 mg) were mixed with a fixed amount of DNA (1 µg) in equal volumes and allowed to incubate at room temperature for at least 30 min before measuring size and zeta potential. All data reported are mean ± SEM (n = 3).

All animal studies were carried out following the United States National Research Council guide for care and use of laboratory animals. The animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at The University of Texas at Austin and the IACUC at Texas Tech University Health Science Center. Female hairless immunocompetent SKH1-Elite mice (6–8 weeks, Charles River Laboratories, Wilmington, MA) were intramuscularly injected in the gastrocnemius muscles with the pVax/opt-BoNT/C-Hc50, alone or coated on the surface of the PLGA nanoparticles, in weeks 0, 2, 4, and 8. The dose of the plasmid was 20 µg per mouse per injection. Mice in the control group were injected with sterile phosphate-buffered saline (PBS, 10 mM, pH 7.4). Mice were bled in weeks 5 and 9. The concentrations of BoNT/C-specific IgG, IgG₁, and IgG_2a in the serum samples were determined using ELISA Quantization kits (Bethel Lab. Inc.., Montgomery, TX).^29,22 Finally, 4 weeks after last immunization, the immunized mice were intraperitoneally injected with 100 × MLD₅₀ of purified active BoNT/C (Metabiologics Inc., Madison, WI) to evaluate the BoNT/C-neutralizing activity of the antibodies induced in protecting against BoNT intoxication.^22,24 Statistical analyses were completed by performing analysis of variance (ANOVA) followed by 2-tailed Student's t-test. A p value of ≤0.05 (2-tail) was considered significant (GraphPad Prism 5 software; GraphPad Software, La Jolla, CA).

As expected, the concentrations of BoNT/C-specific total IgG, IgG1, and IgG2a in mice that were immunized with the pVax/opt-BoNT/C-Hc50-coated PLGA nanoparticles were significantly higher than in mice that were immunized with the same dose of pVax/opt-BoNT/C-Hc50 alone (Fig. 2) (p < 0 .05). After initial priming, mice were further booster-immunized 3 times, and it appeared that the fourth immunization significantly further improved the resultant specific total IgG and IgG2a levels, but the specific IgG1 levels were significantly decreased after the fourth immunization (relative to after the third immunization) (Fig. 2B) (p < 0 .05). Overall, the specific antibody responses were IgG1 and IgG2a balanced after the third immunization. The IgG2a/IgG1 ratio in mice immunized with pVax/opt-BoNT/C-Hc50 alone was 1.0, and 1.1 in mice immunized with pVax/opt-BoNT/C-Hc50 coated on PLGA nanoparticles. However, after the fourth immunization, the antibody responses were shifted toward IgG2a-biased, as the IgG2a/IgG1 ratio was increased to 13.9 in mice immunized with pVax/opt-BoNT/C-Hc50 alone, and 42.6 in mice immunized with pVax/opt-BoNT/C-Hc50 coated on PLGA nanoparticles. This is expected as the immune responses induced by plasmid DNA vaccine tends to be IgG2a-biased.³⁰ Finally, when challenged with active BoNT/C (100 × MLD₅₀), all mice that were immunized with the pVax/opt-BoNT/C-Hc50 coated on cationic PLGA nanoparticles survived, compared to 80% of the mice that were immunized with the pVax/opt-BoNT/C-Hc50 alone (Fig. 3). None of the un-immunized mice survived the BoNT/C challenge (Fig. 3).

Figure 2. Serum anti-BoNT/C-Hc50 IgG (A), IgG1 (B), and IgG2_a(C) induced by intramuscular immunization of mice with pVax/opt-BoNT/C-Hc50 (20 µg/mouse), alone (i.e., pBoNT/C) or coated on cationic PLGA nanoparticles (i.e., pBoNT/C-NPs). SKH-1 Elite mice (n = 5) were dosed in weeks 0, 2, 4 and 8. Control mice received PBS only. Blood samples were collected in week 5 (day 35) and week 9 (day 63). Data are mean ± SD (n = 5). *p < 0.05 compared to Control group,^#p < 0.05 compared to pBoNT/C only, and ^ξp < 0.05, day 35 vs. day 63.

Figure 3. Protective immunity against BoNT/C challenge in immunized mice. SKH-1 Elite mice (n = 5) were dosed in weeks 0, 2, 4 and 8 with pVax/opt-BoNT/C-Hc50 plasmid (20 µg per mouse), alone (i.e., pBoNT/C) or coated on PLGA nanoparticles (i.e., pBoNT/C-NPs), with control mice receiving PBS only, and challenged in week 12 with 100 × MLD₅₀ of BoNT/C toxin.

BoNTs are the most poisonous substances known in nature.³ Nonetheless, to prevent BoNT intoxication, specific neutralizing antibodies in the blood are likely sufficient to neutralize the toxin. Of course, since several BoNT serotypes can cause human botulism, an effective human botulism vaccine needs to be multivalent. As mentioned earlier, plasmid DNA vaccine is an attractive option in developing a new botulism vaccine candidate because it is cost-effective and safe to manufacture and store and can be readily made multivalent.¹⁰ Multiple immunoreactive regions have been identified within each BoNT domain.¹⁶ Recombinant subunit vaccine candidates have included the C-terminal portion of the H chain, the L-(catalytic) chain, and the L-chain expressed with the translocation domain. Of these, the HC domain displays the highest protective ability.¹⁵ However, because the L-chain of BoNT also harbors protective epitopes and induces neutralizing antibodies,³¹ more efforts need to be directed at further understanding its role in BoNT vaccine development. Due to the genetic variability of the sequence in the HC region, it is imperative that conserved critical residues are identified in order to prevent escape variants.³² Future development will include engineering and validating a plasmid that encodes the Hc50 fragments of all BoNT serotypes that cause botulism in human (e.g.,, BoNT/A, /B, /E, and /F), which hopefully will offer cross-protection against human botulism as the pentavalent formulation of BoNT/A, /B, /C, /D and /E was effective against homologous serotypes only.³³

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This work was supported in part by the U.S. National Institute of Allergy and Infectious Diseases grants AI078304 (to ZC) and AI105789 (to ZC and MZ).