SINGLE-DOSE COMBINATION NANOVACCINE INDUCES BOTH RAPID AND DURABLE HUMORAL IMMUNITY AND TOXIN NEUTRALIZING ANTIBODY RESPONSES AGAINST BACILLUS ANTHRACIS

Abstract

Bacillus anthracis, the causative agent of anthrax, continues to be a prominent biological warfare and bioterrorism threat. Vaccination is likely to remain the most effective and user-friendly public health measure to counter this threat in the foreseeable future. The commercially available AVA BioThrax vaccine has a number of shortcomings where improvement would lead to a more practical and effective vaccine for use in the case of an exposure event. Identification of more effective adjuvants and novel delivery platforms is necessary to improve not only the effectiveness of the anthrax vaccine, but also enhance its shelf stability and ease-of-use. Polyanhydride particles have proven to be an effective platform at adjuvanting the vaccine-associated adaptive immune response as well as enhancing stability of encapsulated antigens. Another class of adjuvants, the STING pathway-targeting cyclic dinucleotides, have proven to be uniquely effective at inducing a beneficial inflammatory response that leads to the rapid induction of high titer antibodies post-vaccination capable of providing protection against bacterial pathogens. In this work, we evaluate the individual contributions of cyclic di-GMP (CDG), polyanhydride nanoparticles, and a combination thereof towards inducing neutralizing antibody (nAb) against the secreted protective antigen (PA) from B. anthracis. Our results show that the combination nanovaccine elicited rapid, high titer, and neutralizing IgG anti-PA antibody following single dose immunization that persisted for at least 108 DPI.

INTRODUCTION

Bacillus anthracis is a Gram-positive, spore-forming bacteria that is the causative agent of the disease anthrax, whose spores are capable of lasting decades to centuries in the environment¹. While natural inhalation infection of humans with B. anthracis spores is rare, the environmental stability of the spores and the high lethality of B anthracis make it capable of infecting hundreds of thousands of individuals with a single aerosol dispersion, making it a major biological warfare agent². The difficulty in diagnosing anthrax, with its early onset of cold-like symptoms, is one of the major hurdles in managing the response to a widespread release of B. anthracis spores². If anthrax infection is identified early after exposure, an extensive (60 days) therapeutic or prophylactic use of antibiotics is critical to prevent serious illness and death as spores can persist in vivo in a dormant state and cause disease well after the initial exposure³. In a mass exposure event, therapeutic intervention would require a largescale stockpile of antitoxins and antibiotics readily available at all times and would require that healthcare workers follow up with exposed individuals to patient compliance which is historically poor. For example, following the 2001 anthrax attack in the United States, fewer than half of the approximately 10,000 potentially exposed individuals completed the recommended 60 day course of antibiotic treatment⁴.

Due to the limitations of these therapeutic interventions, as well as the propensity of dormant spores to persist in the lungs, pre- and post-exposure vaccination strategies are believed to be the most effective countermeasures against anthrax, where vaccination could generate protective immunity in exposed individuals before the onset of disease, as well as protect unexposed individuals against potential attack. Most vaccine strategies include the B. anthracis secretory factor protective antigen (PA), a component of the anthrax lethal toxin and edema toxin, when combined with lethal factor or edema factor, respectively. It is believed that neutralizing the ability of PA to heptamerize and translocate lethal factor (LF) and edema factor (EF) into the host cell cytosol is critical to prevent dissemination of the bacteria and toxemia associated with mortality⁵. A previous study demonstrated that anti-PA antibody titer and toxin neutralization titer following prime-boost vaccination correlated with protection in New Zealand white rabbits challenged intranasally with B. anthracis Ames strain spores⁶. In a similar study with guinea pigs, it was shown that anti-PA IgG antibody titer did not correlate with protection; however, toxin neutralization titer correlated with protection following prime-boost vaccination⁷. To further corroborate the importance of the humoral immune response in protection, a number of monoclonal antibodies have been generated that possess toxin neutralizing capacity⁸, and there are reports of cross species modeling using ELISA and toxin neutralization assay data in multiple animal models to predict the efficacy of the Biothrax Anthrax Vaccine Adsorbed (AVA) vaccine in humans^9,10.

The current FDA-approved AVA vaccine consists of a cell-free, culture filtrate adsorbed to aluminum hydroxide. PA is believed to be the critical immunogenic component of the AVA vaccine as evidenced by the fact that anti-PA neutralizing antibodies are correlates of protection. However, there remains room for improvement to this vaccine formulation. First, it is a crude culture supernatant from B. anthracis strain V770-NP1-R that contains unquantified amounts of PA¹¹. Second, prophylactic vaccination with AVA requires five administrations over 18 months, followed by annual boosters in order to maintain protective immunity. Third, the AVA vaccine is one of the most expensive pharmaceuticals being stored in the Strategic National Stockpile, owing to the refrigerated storage conditions as well as additional costs associated with cold chain distribution¹². Therefore, it is sagacious/au fait to consider developing vaccines that can overcome shortcomings of the existing AVA vaccine including the ability to obviate the cold chain and be easily deployed in the event of exposure or outbreak event, all while providing both rapid and long-lasting protective immunity, ideally in a single dose.

Polyanhydride nanoparticle-based vaccines (i.e., nanovaccines) represent a promising platform that has proven effective against multiple bacterial and viral pathogens, including influenza¹³, respiratory syncytial virus¹⁴, and another biodefense pathogen Yersinia pestis¹⁵. Comprised of 1,6-bis(p-carboxyphenoxy)hexane (CPH) and 1,8-bis(p-carboxyphenoxy)-3,6-diaoctane (CPTEG), these copolymers are safe and induce mild inflammation at the site of injection¹⁶. Their hydrophobic properties enable sustained release of encapsulated payloads, which is tunable through varying copolymer composition¹⁷, and can allow for dose-sparing through co-adjuvanting poorly immunogenic payloads¹⁸. In addition, polyanhydride nanoparticles demonstrate the ability to be stored without requiring refrigeration and can maintain the functional and antigenic stability of labile payloads^19,20. For example, polyanhydride nanoparticles encapsulating PA were shown to release functionally active PA following four months of storage at 40 °C, while PA dramatically lost its bioactivity when similarly stored with Imject Alum²⁰. Therefore, this nanovaccine platform shows promise both as an adjuvant as well as a shelf-stable delivery vehicle for PA-based vaccines.

Cyclic dinucleotides (CDNs) are a class of small molecules that function as universal secondary signaling molecules in bacterial and mammalian cells²¹. Recognition of CDNs through the STING (STimulator of INterferon Genes) pathway results in the induction of type I interferon (IFN), inducing anti-pathogenic activity²². Cyclic di-guanosine monophosphate (CDG), the most well-studied CDN to date, has potent adjuvant properties as vaccines adjuvanted with CDG induce a broad antibody response, characterized by the presence of multiple IgG subclasses in serum²³, and have been shown to elicit significantly higher antibody titers than alum-adjuvanted formulations²⁴. Our previous work has demonstrated that immunization with chicken egg ovalbumin (Ova) adjuvanted with a synthetic analog of CDG in mice enhanced the antibody response within two weeks post-immunization compared to traditional TLR agonist-based adjuvants²⁵. Mice immunized with PA adjuvanted with CDG had higher anti-PA IgG serum responses compared to animals immunized with the AVA vaccine, thereby making them an attractive adjuvant candidate to be included in future anthrax vaccines²⁵.

We have previously shown that a single dose of a combination nanovaccine comprising polyanhydride nanoparticles encapsulating the Y. pestis fusion protein F1-V co-adjuvanted with a non-canonical CDG adjuvant (containing 2’,5’−3’−5’ phosphate linkages) completely protected mice against pneumonic plague within 14 days post-immunization (DPI), while F1-V adjuvanted with CDG failed to provide complete protection¹⁵. Based on these results, we were interested in performing preliminary vaccination studies in mice to evaluate the immunogenicity of a single dose of shelf-stable 20:80 CPTEG:CPH polyanhydride nanoparticles encapsulating PA in combination with CDG to elicit rapid and durable anti-PA antibody responses, characterized by serum anti-PA IgG antibody titers and toxin neutralization ability following immunization.

MATERIALS AND METHODS

Materials

Chemicals used for polymer and nanoparticle synthesis: 4-p-hydroxybenzoic acid, tri-ethylene-glycol, 1,6-dibromohexane were obtained from Sigma Aldrich (St. Louis, MO); dimethyl formamide, acetic acid, acetonitrile, acetic anhydride, toluene, methylene chloride, pentane were purchased from Fisher Scientific (Fairlawn, NJ). 4-p-flourobenzonitrile was purchased from Apollo Scientific (Cheshire, UK). Dithio-R_P,R_P-cyclic di-guanosine monophosphate (R,R-Cyclic di-GMP (CDG)) was kindly provided by Aduro Biotech (Berkeley, CA).

Polymer Synthesis

CPTEG and CPH monomers were first individually synthesized and melt polycondensation of the monomers was used to synthesize 20:80 CPTEG:CPH copolymer, as previously described^17,26. Briefly, a 20:80 molar ratio of CPTEG:CPH diacids were added to acetic anhydride at a total mass concentration of 0.022 g/mL and allowed to react for 30 min at 125 °C. The solvent was subsequently removed by rotary evaporation, leaving a dry product which was then polymerized in an oil bath at 140 °C under vacuum (<0.1 torr) for six hours. The solid product was then dissolved in methylene chloride and precipitated in hexanes to yield the final copolymer product. Copolymers were dissolved into deuterated chloroform for purity and molecular weight characterization via ¹H nuclear magnetic resonance spectroscopy. Copolymer molecular weight was ca. 5.3 kDa.

Nanoparticle Synthesis and Antigen Encapsulation Efficiency

Polyanhydride nanoparticles encapsulating PA (BEI resources, Cat# NR-3780, Manassas, VA), from B. anthracis, were synthesized using flash nanoprecipitations of 20:80 CPTEG:CPH copolymer, as previously described¹⁵.

To determine PA encapsulation efficiency, 1 mg of PA-containing nanoparticles were weighed into 1.5 mL microcentrifuge tubes and suspended in 500 μL of 40 mM NaOH buffer following approximately 30 s of sonication. Periodically, samples were centrifuged at 10,000 rcf for 10 min, and 400 μL of supernatants were collected and replaced with fresh solution. Released protein was quantified via a microbicinchoninic acid (microBCA) assay and the encapsulation efficiency of PA was calculated as the total amount of protein released from the nanoparticles divided by the amount of protein added during nanoparticle synthesis²⁷. The resulting encapsulation efficiency was 68%, indicating that the actual percent loading of PA encapsulated within the nanoparticles was 6.8% (w/w), following synthesis, while aiming for 10% (w/w) loading.

Nanoparticle mean size and size distribution were determined using ImageJ (National Institutes of Health, Bethesda, MD) analysis of nanoparticle images taken using scanning electron microscopy (SEM; JEOL 840 A, JEOL Ltd., Tokyo, Japan).

Animals

Male and female A/J mice, as well as female BALB/c mice, between six and eight weeks of age were obtained from Jackson Laboratory (Bar Harbor, ME). All studies involving the use of animals were carried out in accordance with and approval from the Iowa State University Institutional Animal Care and Use Committee.

Vaccinations

Separate groups of 6–8 week old A/J and BALB/c mice were immunized subcutaneously (SC) with one of the following vaccine formulations: 1) 5 μg PA encapsulated in 73.5 μg of 20:80 CPTEG:CPH (6.8% wt/wt) nanoparticles (NP), 2) 5 μg soluble PA adjuvanted with 25 μg CDG (dithio-R_P,R_P-cyclic diguanosine monophosphate) (Aduro Biotech, Berkeley, CA) (CDN Vaccine), 3) a combination of 5 μg PA encapsulated in 73.5 μg of 20:80 CPTEG:CPH (6.8% wt/wt) nanoparticles adjuvanted with 25 μg CDN (Combination Nanovaccine), 4) 5 μg soluble PA alone (sPA), or 5) 5 μg soluble PA adsorbed to Alhydrogel (Alhydrogel; Invivogen, San Diego, CA). The Alhydrogel formulation was prepared as a 1:9 volume ratio (i.e., a 1:10 dilution) of Alhydrogel adjuvant 2% (10 mg/mL aluminum content) in phosphate buffered saline (PBS), per manufacturer’s instructions. Each animal received 0.2 mg of aluminum content. Data from female A/J mice (n = 4 per group) and female BALB/c mice (n = 6 per group) are presented in Figure 2. Data from male A/J mice (n = 8 per group) are presented in Figures 3–5.

Figure 2. Combination Nanovaccine induces rapid and long-lived antibodies against protective antigen.

Either (A) female A/J mice (n = 4 per group) or (B) female BALB/c mice (n = 6 per group) were vaccinated SC with PA adjuvanted in different vaccine formulations. The formulations consisted of 5 μg PA encapsulated in 20:80 CPTEG:CPH nanoparticles (NP), 5 μg soluble PA adjuvanted with 25 μg CDNs (CDN Vaccine), a combination nanovaccine comprising of 5 μg PA encapsulated in 20:80 CPTEG:CPH nanoparticles adjuvanted with 25 μg CDN (Combination Nanovaccine), or 5 μg soluble PA alone (sPA). Total serum IgG (H+L) antibody titer to PA was quantified via ELISA at 14, 28, 56, and 108 DPI. Titer values were Log2 transformed and compared at each time point for statistical significance between vaccination groups via an ordinary one-way ANOVA with a Tukey’s multiple comparison test. P value is indicated as follows for each comparison indicated (*#@$ = p ≤ 0.05) (# vs NP, $ vs CDN Vaccine, @ vs Combination Nanovaccine, * vs sPA). Individual animals are shown with bars indicating mean ± SD.

Figure 3. Combination Nanovaccine elicits rapid and long-lived antibodies against protective antigen.

Male A/J mice (n = 8 per group) were vaccinated SC with PA adjuvanted in different vaccine formulations. The formulations consisted of 5 μg soluble PA adsorbed to a 1:9 dilution of alhydrogel (Alhydrogel), a combination of 5 μg PA encapsulated in 20:80 CPTEG:CPH nanoparticles adjuvanted with 25 μg CDN (Combination Nanovaccine), or 5 μg soluble PA alone (sPA). Total serum IgG (H+L) antibody titer to PA was quantified via ELISA at 15, 30, 46, 88, and 103 DPI. Titer values were Log2 transformed and compared at each time point for statistical significance between vaccination groups via an ordinary one-way ANOVA with a Tukey’s multiple comparison test. * p ≤ 0.05 compared to sPA. Individual animals are shown with bars indicating mean ± SD. The dashed line represents the average anti-PA IgG (H+L) antibody titer from naïve animals prior to immunization.

Figure 5. Combination Nanovaccine induces long-lived anti-PA antibody secreting cells in both bone marrow and spleen.

Male A/J mice (n = 8 per group) were vaccinated SC with PA adjuvanted in different vaccine formulations. The formulations consisted of 5 μg soluble PA adsorbed to a 1:9 dilution of alhydrogel (Alhydrogel), a combination of 5 μg PA encapsulated in 20:80 CPTEG:CPH nanoparticles adjuvanted with 25 μg CDN (Combination Nanovaccine), or 5 μg soluble PA alone (sPA). Bone marrow and splenic lymphocytes were harvested at 103 DPI from groups of immunized or naïve mice (n = 5 per group), and were analyzed via ELISPOT for the number of PA-specific IgG-secreting antibody secreting cells (ASCs). Data represent the number of PA-specific IgG-secreting ASCs per 500,000 cells in either A) bone marrow or B) spleen from individual mice. The dashed line represents the number of non-specific (i.e., naturally occurring) ASCs per 500,000 cells determined from naïve mice. Statistical significance between groups was determined by Kruskal-Wallis and Mann-Whitney tests. * represents p ≤ 0.05 compared to naïve mice. # represents p ≤ 0.05 compared to sPA-immunized mice. $ represents p ≤ 0.05 compared to Combination Nanovaccine. @ represents p ≤ 0.05 compared to naïve, sPA, and Alhydrogel formulations. Individual animals are shown with bars indicating mean ± SD.

Serum Antibody Responses and ELISA

In longitudinal antibody studies (Figures 2 & 3), the same mice were bled via the saphenous vein multiple times at the indicated time points, either 14, 28, 56, and 108 DPI (Figure 2), or 15, 30, 46, 88, and 103 DPI (Figure 3). Serum was collected and stored at –20 °C until used for analysis. Costar 3590 96-well EIA/RIA high binding plates (Corning, Corning NY) were coated with 100 μL of PA or F1-V (0.5 μg/mL in PBS) and incubated overnight at 4 °C. Plates were blocked using either 2% (w/v) Difco gelatin in PBS (0.05 M, pH 7.2) containing 0.05% Tween-20 (PBS-T) for 2 h at room temperature or 2.5% (w/v) powdered skim milk dissolved in PBS-T (pH 7.4), that had been incubated for 2 h at 56 °C to inactivate any endogenous phosphatase activity. After three washes using PBS-T, serum samples were titrated across the plate using two-fold serial dilutions, starting at 1:100 or 1:200, in PBS-T containing 1% (v/v) normal goat serum. Samples were incubated overnight at 4 °C. After three washes in PBS-T, an alkaline phosphatase conjugated goat anti-mouse IgG (H+L) secondary detection antibody (Cat# 115–005-003, Jackson ImmunoResearch, West Grove, PA) was diluted 1:1000, and 100 μL was added to the wells and allowed to incubate at room temperature for 2 h. Plates were washed three times with PBS-T and alkaline phosphatase substrate (Cat# BP2534, Fisher Scientific, Hampton, NH) was added at 1 mg/mL in buffer containing 50 mM sodium carbonate, 2 mM magnesium chloride, and sodium bicarbonate added to achieve a pH of 9.3. Color was allowed to develop for 30 min or 2 h and analyzed using a SpectraMax 190 (Molecular Devices, San Jose, CA) spectrophotometer at a wavelength of 405 nm.

Cell Culture

J774A.1 cells were cultured and maintained at 37 °C under 5% CO₂ in DMEM (Cat #15–013-CV, Corning) medium supplemented with 100 U/mL penicillin and 100 μg/mL streptomycin, 2 mM glutamine, and 10% FBS (i.e., complete tissue culture medium).

Toxin Neutralization Assay

In order to assess neutralizing activity of serum from vaccinated mice, a J774A.1 cell based cytotoxicity assay was adapted from a previously described protocol ²⁸. Briefly, J774A.1 cells were seeded into 96-well flat bottom plates (Costar, Cat #3595) at a concentration of 4 × 10⁴ cells/well and allowed to adhere for 18 h. Serially diluted serum samples were prepared in separate 96-well plates by beginning at a 1:50 or 1:100 dilution and serially diluting them across the plate with two-fold dilutions. The serum samples were then incubated for 30 min with PA and LF (BEI resources, Cat# NR-28544), at constant concentrations of 50 ng/mL and 40 ng/mL, respectively, and then added to the wells containing the J774A.1 cells. The cell-serum-toxin mixture was allowed to incubate for 4 h after which 25 μL of a 5 mg/mL solution of 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) reagent (Invitrogen, Cat# M6494, Carlsbad, CA) was added. After two hours of incubation, supernatants were removed and 100 μL of dimethyl sulfoxide (Fisher Scientific, Cat# D128–500) was added to lyse the cells and solubilize formazan crystals. The OD of each sample at 555 nm was measured using a SpectraMax 190 spectrophotometer.

The first two time points (15 and 46 DPI) were performed in the same assay. The final time point (103 DPI) was performed in a separate assay. Therefore, as an inter-assay control, a previous serum sample collected 46 DPI from a mouse immunized with the Combination Nanovaccine used in the toxin neutralization assay for 15 and 46 DPI time points was used in the assay for 103 DPI samples as well. The titration curves for this sample in both assays was similar, with ED₅₀ values of 13602 and 14033, respectively, for both assays, indicating consistency in responses between experiments. In addition, each assay included the following controls: cells without toxin (OD > 2.0), cells incubated with toxin but without serum (OD < 0.25), and cells incubated with toxin plus pooled serum from mice pre-immunization (OD < 0.25).

Peptide Microarray Printing and Analysis

102 linear peptides containing 13- to 20-mer amino acid sequences (10 or 11 amino acid overlaps) (BEI resources, Cat# NR-527) spanning the full length of PA antigen, as well as the full-length PA antigen, Y. pestis fusion protein F1-V, and OVA were printed onto Nexterion Slide AL (Schott, Louisville, KY) using a BioRobotics MicroGRID II microarray printer (Genomic Solutions, Inc. Ann Arbor, MI). Peptides and proteins were first dissolved in DMSO at 10 mg/mL to ensure dissolution, then diluted 10-fold in water to 1 mg/mL. The solution was diluted two-fold to bring the final concentration to 0.5 mg/mL in 1x print buffer (5% (v/v) DMSO, 137 mM NaCl, 9 mM KOH, 11.3 mM NaH₂PO₄). The full-length PA antigen was used as a positive control, while F1-V and OVA were used as irrelevant controls. Slides were printed with 16 arrays per slide, each array containing 7×15 spots per array, with peptides printed in a serpentine pattern in duplicate. Following printing, slides were vacuum sealed and stored at −80 °C until further use. The microarray assay and analysis was performed as previously described¹⁵ using serum from PA-immunized mice collected at 15 and 88 DPI.

ELISPOT Analysis of Bone Marrow and Spleen

ELISPOT analysis was performed as previously described¹⁵; however, plates were coated overnight with 5 μg/mL PA in PBS prior to blocking with complete tissue culture medium.

Statistical Analyses

Data generated from anti-PA serum IgG ELISA assays were Log2 transformed and analyzed within each time point via one-way ANOVA with a Tukey post-test for multiple comparisons. Results from the toxin neutralization experiments were analyzed by determining the inflection point of the antibody dilution curve reported as the effective dilution at 50% inhibition (ED₅₀) using nonlinear regression modeling on GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA) and Log2 transformed. Toxin neutralization responses were also quantified by determining the mean area under the curve (AUC) for each treatment group. Differences in ED₅₀ and AUC between treatment groups were determined via one-way ANOVA with a Tukey post-test for multiple comparison. For PA peptide microarray analysis, a Student’s t test was used for comparison of background-corrected mean fluorescence intensity (MFI) of sera from vaccination groups for each peptide. Binding of serum antibodies to a peptide was considered detectable if the spot binding background-corrected MFI of serum from a vaccination group was significantly (p ≤ 0.05) higher than that of serum from naïve animals (prior to immunization). For bone marrow and spleen ELISPOT analysis, statistical significance between groups was determined by Kruskal-Wallis and Mann-Whitney tests. Statistical analyses of the PA peptide microarray was performed using R v3.4.1. All other analyses were performed using GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA).

RESULTS

Polyanhydride Nanovaccine Characterization

The 6.8%-loaded PA (w/w) 20:80 CPTEG:CPH nanoparticles were characterized for size distribution and encapsulation efficiency of PA. The size of the nanoparticles was consistent with previous work¹⁵, with a mean diameter of 255 ± 98 nm and a polydispersity index of 0.15 indicating a narrow size distribution (Figure 1). The encapsulation efficiency of the PA within the nanoparticles was 68%.

Figure 1. Characterization of PA Nanovaccine.

A) Representative scanning electron micrograph image of PA nanovaccine (scale bar = 2 μm). Nanoparticle mean diameter was 255 ± 98 nm, determined via ImageJ analysis. B) Size distribution of 6.8% (w/w) PA 20:80 CPTEG:CPH nanoparticles (255 ± 98 nm), determined via ImageJ analysis.

Combination Nanovaccine Elicits Rapid Anti-PA Antibody Responses that persisted for at least 108 DPI

To assess the ability of both CDNs and polyanhydride nanoparticles to elicit serum anti-PA IgG responses, female A/J mice (n = 4 per group) were immunized SC with a single dose of the vaccine formulations described in the Materials and Methods section. Serum samples were collected throughout 108 DPI and anti-PA IgG titers were determined. Mice immunized with each of the vaccine formulations attained anti-PA titers >2¹⁵ after 28 days that were maintained for at least 108 DPI (Figure 2A). Sera from CDN Vaccine- or Combination Nanovaccine-vaccinated mice contained significantly (p ≤ 0.05) higher anti-PA IgG antibodies that remained elevated for at least 56 days compared to mice immunized with sPA. A similar trend was also observed with female BALB/c mice (Figure 2B). At 108 DPI, sera of A/J mice immunized with the Combination Nanovaccine maintained titers of ~2¹⁸ that were significantly (p ≤ 0.05) higher compared to that for animals immunized with sPA (e.g., ~2¹⁵). The polyanhydride nanoparticle formulation (NP) alone was unable to rapidly enhance the magnitude of the anti-PA antibody titer. These results indicate that CDG was the primary contributor to the rapid anti-PA serum IgG responses and strongly suggests the need to include CDG co-adjuvant in the nanovaccine formulation.

Next, the ability of the Combination Nanovaccine to induce anti-PA antibody responses was compared to PA adjuvanted with alhydrogel. Male A/J mice (n = 8 per group) were immunized SC with a single dose of one of the following vaccine formulations: i) 5 μg soluble PA (sPA), ii) 5 μg soluble PA adjuvanted with alhydrogel (Alhydrogel; 1:9 volume ratio = 0.2 mg aluminum content per mouse), or iii) 5 μg PA encapsulated into polyanhydride nanoparticles co-adjuvanted with 25 μg CDG (Combination Nanovaccine). Beginning at 15 DPI, serum samples were collected over 103 days to assess the magnitude of the anti-PA antibody responses. Combination Nanovaccine and Alhydrogel formulations induced significantly (p ≤ 0.0001) higher anti-PA total IgG antibody titers (~2¹³) as early as 15 DPI compared to sPA (~2⁸), and these titers increased to ~2¹⁵ and were maintained over the course of 103 DPI (Figure 3). There was no significant difference in the magnitude of the anti-PA IgG titers between mice vaccinated with the Combination Nanovaccine or Alhydrogel.

Combination Nanovaccine Elicits Enhanced Rapid Toxin Neutralization Antibodies against PA

As the magnitude of the anti-PA serum IgG response was similar between the Combination Nanovaccine and Alhydrogel formulations, toxin neutralization assays were performed using a J774A.1 cell cytotoxicity assay and serum from mice vaccinated with either formulation, as described above. Interestingly, despite having similar titers of total anti-PA IgG at 15 DPI (Figure 3), sera from mice immunized with the Combination Nanovaccine had significantly (p = 0.0078) higher toxin neutralizing antibody responses compared to sera from mice immunized with the Alhydrogel formulation (Figure 4A and Supplementary Figure 1A and 1D). The ability of serum from Combination Nanovaccine-immunized mice to neutralize toxin—mediated cell death was similar to that of serum from the Alhydrogel formulation-immunized mice at 46 DPI and 103 DPI (Figure 4B and 4C and Supplementary Figure 1B, 1C, and 1D). Serum collected at 103 DPI from Alhydrogel and Combination Nanovaccine-immunized mice did not have significantly higher toxin neutralization capabilities compared to serum from sPA-immunized mice when comparing ED₅₀ values, however, these were significantly (p ≤ 0.05) different when comparing AUC (Supplementary Figure 1C and 1D).

Figure 4. Combination Nanovaccine elicits rapid and long-lived toxin neutralizing antibodies against the protective antigen.

Male A/J mice (n = 8 per group) were vaccinated SC with PA adjuvanted in different vaccine formulations. The formulations consisted of 5 μg soluble PA adsorbed to a 1:9 dilution of alhydrogel (Alhydrogel), a combination of 5 μg PA encapsulated in 20:80 CPTEG:CPH nanoparticles adjuvanted with 25 μg CDN (Combination Nanovaccine), or 5 μg soluble PA alone (sPA). Neutralizing antibody titers were determined in serum collected (A) 15, (B) 46, and (C) 103 DPI using a J774A.1 cell-based MTT cytotoxicity assay as described in Materials and Methods. Titer values were determined as the inflection point of the antibody dilution curve reported as the effective dilution at 50% inhibition (ED₅₀) and were Log2 transformed and compared at each time point for statistical significance between treatment groups via an ordinary one-way ANOVA with a Tukey’s multiple comparison test. * p ≤ 0.05 compared to sPA. # p ≤ 0.05 compared to Alhydrogel. Individual animals are shown with bars indicating mean ± SD.

Anti-PA IgG Responses to Linear Epitopes in Sera from Vaccinated Animals

To further characterize the quality of the anti-PA antibody response, a linearly overlapping peptide array was used to assess the ability of serum antibodies to bind to linear peptides spanning the full length of PA. A detectable response was defined as having significantly (p ≤ 0.05) higher anti-PA antibody binding (measured by MFI) compared to naïve serum samples.

Serum from immunized mice was analyzed at 15 and 88 DPI to assess differences between early antibody responses induced by the vaccine formulations, where there were significant differences in anti-PA neutralizing antibody titers (Figure 4A), and later, when the anti-PA IgG neutralizing titers were similar (Figure 4C). At both time points, serum from mice immunized with sPA, Alhydrogel + PA, and the Combination Nanovaccine formulation demonstrated binding to intact PA (see row labeled ‘PA’), whereas the naïve controls did not (Supplementary Figure 2). Serum IgG from vaccinated mice did not bind to OVA; however, serum from animals vaccinated with the PA-specific Combination Nanovaccine demonstrated binding to F1-V at both 15 and 88 DPI, which was not observed with any of the other vaccine formulations. This was confirmed after performing ELISA using serum collected at 46 DPI, in which serum from Combination Nanovaccine-immunized mice had significantly (p ≤ 0.05) higher anti-F1-V IgG antibodies compared to mice immunized with either Alhydrogel or sPA vaccine formulations (Supplementary Figure 3).

At 15 DPI, there was no detectable antibody response to any peptides in mice vaccinated with sPA or PA adjuvanted with alhydrogel (Supplementary Figure 2A and 2C). However, serum from mice vaccinated with the Combination Nanovaccine demonstrated significantly (p < 0.05) higher anti-PA IgG responses to peptides 13, 17, 18, 19, 24, 34, 38, 101, and 102 compared to naïve controls. By 88 DPI, serum responses from sPA-immunized mice still did not bind significantly to any linear peptides; however, serum from Alhydrogel-immunized animals had significantly (p < 0.05) higher anti-PA IgG responses to peptides 13 and 18 over naïve controls. Serum from Combination Nanovaccine-immunized animals showed significantly (p < 0.05) higher anti-PA IgG responses to peptide 101.

Combination Nanovaccine Induces Long-lived Antibody Secreting Cells in Bone Marrow and Spleen

As the magnitude of the serum anti-PA IgG response between mice immunized with the Combination Nanovaccine and Alhydrogel were similar and persisted for 103 DPI, it was hypothesized that the maintenance of the long-lived serum anti-PA IgG response was due to the presence of long-lived antibody secreting cells (ASCs) residing in the bone marrow and spleen of immunized mice. To test this, ELISPOT analysis of bone marrow and spleen homogenates from immunized mice collected at 103 DPI was performed. The analysis revealed that bone marrow of mice immunized with sPA did not have significantly more PA-specific ASCs when compared to naïve controls (Figure 5A). In contrast, there were significantly (p ≤ 0.05) higher numbers of PA-specific ASCs in bone marrows of mice immunized with either the Combination Nanovaccine or Alhydrogel. The Alhydrogel formulation induced significantly (p ≤ 0.05) higher PA-specific ASCs in bone marrow compared to the Combination Nanovaccine. Interestingly, in the spleen, only mice immunized with the Combination Nanovaccine had significantly (p ≤ 0.05) higher numbers of PA-specific ASCs compared to naïve controls and higher than sPA and Alhydrogel formulations (Figure 5B).

DISCUSSION

B. anthracis is a highly virulent pathogen, capable of infecting hundreds of thousands of individuals with a single aerosol dispersion of spores, making it a significant biological warfare agent². Vaccination is likely to be the most cost effective and patient-friendly option to prevent disease following exposure of civilians and war fighters to B. anthracis. The major issues that need to be resolved in next generation anti-PA vaccines are: i) reducing the number of doses required to effectively induce high titer neutralizing antibody; ii) identification and selection of adjuvants that promote antibody responses to neutralizing rather than non-neutralizing epitopes; and iii) improving the shelf-life stability of the formulated vaccines^11,29,30. In this work, we designed a combination nanovaccine composed of polyanhydride nanoparticles encapsulating PA co-adjuvanted with CDNs and assessed its ability to induce anti-PA neutralizing IgG responses following a single immunization. The motivation for using the combination adjuvant approach is predicated upon the previously described immunostimulatory capabilities of both adjuvants¹⁵, in addition to the shelf stability provided by the polyanhydride nanoparticles^19,20.

In this work, preliminary vaccination studies evaluating the immune response of female A/J mice demonstrated that inclusion of CDNs as a co-adjuvant, either alone or combined with the NP formulation, led to significant increases in serum anti-PA IgG titers throughout 108 DPI compared to NP alone (Figure 2). These results are in agreement with our recently published work showing that a combination of CDNs with a polyanhydride nanovaccine encapsulating the Y. pestis fusion protein F1-V led to an increase in magnitude of serum anti-F1-V IgG titers¹⁵.

As the manifestation of anthrax infection following intranasal exposure can occur rapidly³¹, it is critical to develop vaccines that can rapidly enhance anti-PA antibody responses which, in conjunction with antimicrobials, can best protect individuals with suspected or known exposure to B. anthracis³². As protection against lethal B. anthracis challenge appears to correlate with serum IgG toxin neutralization capacity^6,7, development of vaccines that can induce anti-PA toxin neutralizing antibodies is thought to be beneficial to enhancing protection against a possible exposure event. Similar to immune responses induced by PA adjuvanted with alhydrogel, the Combination Nanovaccine induced a rapid, high titer anti-PA IgG antibody response by 15 DPI that was maintained for 103 DPI (Figure 3). When the toxin neutralizing abilities of serum were evaluated, however, the Combination Nanovaccine elicited higher toxin neutralizing antibody titers compared to PA adjuvanted with alhydrogel at 15 DPI (Figure 4A and Supplementary Figure 1A and 1D). This capability of CDN-containing vaccines to induce rapid toxin neutralizing IgG could lead to rapid post-exposure prophylaxis immunity for individuals in an acutely affected area. Rapid generation of neutralizing antibodies may also have the potential of reducing the number of vaccine doses needed to achieve protective immunity in an outbreak scenario where there may be fewer vaccine doses available than necessary, as well as reducing the recommended 60 days of antibiotic treatment when potential exposure has occurred. These considerations are currently being evaluated for emergency preparedness against a large-scale, aerosol release event by the Centers for Disease Control³².

The potent antibody responses induced by activating the STING pathway with CDNs are reportedly capable of inducing significantly higher antibody titers to antigens than those induced when adjuvanted with alum^15,24,33, including the AVA vaccine²⁵. In this study, the overall serum IgG production to PA was similar between the Combination Nanovaccine and Alhydrogel formulations through 103 DPI; however, the serum anti-PA neutralizing response induced by the Combination Nanovaccine was significantly elevated 15 DPI (Figure 4A). This is hypothesized to be due to lower expression levels of nitric oxide (NO) and reactive oxygen species (ROS) by antigen presenting cells stimulated by CDNs, when compared to TLR agonists²⁵. In particular, increased NO and ROS expression levels are known to deleteriously affect both innate and adaptive immunity, including altering innate cell function³⁴, B cell survival and subsequent antibody production, lower levels of BAFF production^35–37, and lymphatic vessel function³⁸. Although the exact mechanism(s) of action of alum adjuvants is still being investigated, one likely mechanism is through NLRP3 inflammasome activation^39,40, due to alum-induced generation of ROS^40–42. Therefore, we posit that the observed early increase in serum anti-PA neutralizing responses with the Combination Nanovaccine compared to that induced by the Alhydrogel-based vaccine may be mediated by the lower levels of NO and ROS induced by CDNs.

Numerous neutralizing monoclonal antibodies against PA have been documented against every domain of the PA antigen, with the majority of these antibodies being directed at the furin cleavage domain (Domain 1; peptides 3 – 39), endosomal pore formation domain (Domain 2; peptides 38 – 70 and Domain 3; peptides 69 – 84), and receptor binding domains (Domain 4; peptides 83 – 102)⁴³. The most well-studied monoclonal antibody, 14B7 and its affinity-matured variants, are known to bind PA conformational epitopes on domain 4^44–47. However, monoclonal antibodies have also been reported to bind linear epitopes on domain 1⁴⁸ and domain 4⁴⁹. Following microarray analysis of PA linear peptides, it was observed that there was an overall low detectable response of serum IgG from both Combination Nanovaccine- and Alhydrogel-immunized mice to linear peptides spanning the full length PA antigen; however, serum from Combination Nanovaccine-immunized mice was able to bind to peptides 13, 17, 18, 19, 24, 34, 38 (Domain 1), as well as 101 and 102 (Domain 4) by 15 DPI, while there was no detectable antibody responses to any peptides by serum from Alhydrogel-immunized mice at that time point (Supplementary Figure 2). Considering the apparent low detectable response by serum IgG collected at 15 and 88 DPI from mice vaccinated with any vaccine, juxtaposed with the observed increase in toxin neutralization capabilities at 46 and 103 DPI in Combination Nanovaccine- and Alhydrogel-immunized sera compared to sPA, we hypothesize that the majority of the antibodies induced by both vaccines is to conformational epitopes on PA but warrants further investigation.

The maintenance of long-lived antibody responses are believed to derive from long-lived plasma cells residing in bone marrow, secondary lymph organs, and gut, which can persist for decades, and secrete large quantities of antibodies, including IgG, IgM, and IgA^50,51. In this work, PA-specific IgG-ASCs were detected at 103 DPI in the bone marrow and spleen of mice immunized with the Combination Nanovaccine, but only in bone marrows of mice immunized with the Alhydrogel formulation (Figure 5). The presence of PA-specific ASCs in bone marrow are in agreement with previous work from our laboratories, where F1-V-specific IgG-ASCs from combination nanovaccine-immunized mice were detected in bone marrow 215 DPI; however, F1-V-specific ASCs were not detected in spleen¹⁵. Therefore, the Combination Nanovaccine was able to induce long-lived anti-PA serum IgG responses that derived, at least in part, from PA-specific ASCs residing in both bone marrow and spleen.

In conclusion, a combination nanovaccine comprising polyanhydride nanoparticles co-adjuvanted with CDNs was able to elicit both rapid and durable anti-PA IgG antibody responses that were characterized by significantly higher serum anti-PA toxin neutralizing antibody responses by 15 DPI compared to PA adjuvanted with alhydrogel. Altogether, these preliminary results lay the foundation for the further platform development and testing of shelf-stable polyanhydride-based anthrax nanovaccines that can provide both rapid and durable immunity to bolster biodefense preparedness.

Supplementary Material

1. Supplementary Figure 1. Combination Nanovaccine elicits rapid and long-lived toxin neutralizing antibodies against protective antigen.

Male A/J mice (n = 8 per group) were vaccinated SC with PA adjuvanted with different vaccine formulations. The formulations consisted of 5 μg soluble PA adsorbed to a 1:9 dilution of alhydrogel (Alhydrogel), a combination of 5 μg PA encapsulated in 20:80 CPTEG:CPH nanoparticles adjuvanted with 25 μg CDN (Combination Nanovaccine), or 5 μg soluble PA alone (sPA). Neutralizing serum antibody titers were determined for serum collected at (A) 15, (B) 46, and (C) 103 DPI via a J774A.1 cell-based MTT cytotoxicity assay as described in Materials and Methods. The lower dotted line indicates the no sample control average OD. Group averages are shown with error bars indicating mean ± SD. (D) The mean area under the curve (AUC) and standard error was calculated for each treatment group. The AUC was compared between treatment groups via one-way ANOVA with a Tukey’s multiple comparison test. * p ≤ 0.05 compared to sPA. # p ≤ 0.05 compared to Alhydrogel.

2. Supplementary Figure 2. Evaluation of serum IgG responses to linear peptides on PA induced by the Combination Nanovaccine and Alhydrogel.

Male A/J mice (n = 8 per group) were vaccinated SC with PA adjuvanted in different vaccine formulations. The formulations consisted of 5 μg soluble PA adsorbed to a 1:9 dilution of alhydrogel (Alhydrogel), a combination of 5 μg PA encapsulated in 20:80 CPTEG:CPH nanoparticles adjuvanted with 25 μg CDN (Combination Nanovaccine), or 5 μg soluble PA alone (sPA). Microarray analysis of 102 linear peptides containing 13- to 20-mer amino acid sequences (10 or 11 amino acid overlaps) spanning the full length of PA antigen, as well as the full length PA antigen, Y. pestis fusion protein F1-V, and chicken egg ovalbumin (OVA), was assessed using serum of immunized animals collected at (A) 15 and (B) 88 DPI (see Materials and Methods section). Each row corresponds to a specific peptide, the bottom row representing peptide PA-1 and the proceeding upward rows corresponding to each following linear peptide incrementally through peptide PA-102. [Signaling Peptide: peptides 1 – 5; Domain 1: peptides 3 – 39; Domain 2: peptides 38 – 70; Domain 3: peptides 69 – 84; Domain 4: peptides 83 – 102.] Each column represents responses from a single mouse. The background-corrected MFI of serum responses to each peptide is represented by a range of color from white (no response) to purple (maximum response). The full-length PA protein was used as a positive control, and the Yersinia pestis fusion protein F1-V and chicken egg ovalbumin (OVA) were used as nonspecific controls. The dashed lines are used to delineate domains 1–4 of PA. (C) Representative table of serum responses to linear peptides whose background-corrected MFI values are significantly (p ≤ 0.05) higher than that of naïve serum, shown as bolded numbers. The full length PA antigen was included for a comparison of the magnitude of the background-corrected MFI values.

3. Supplementary Figure 3. Combination Nanovaccine elicits anti-F1-V antibody response in mice vaccinated with PA.

Male A/J mice (n = 8 per group) were vaccinated SC with PA adjuvanted in different vaccine formulations. The formulations consisted of 5 μg soluble PA adsorbed to a 1:9 dilution of alhydrogel (Alhydrogel), a combination of 5 μg PA encapsulated in 20:80 CPTEG:CPH nanoparticles adjuvanted with 25 μg CDN (Combination Nanovaccine), or 5 μg soluble PA alone (sPA). Total serum IgG (H+L) antibody titer to F1-V was quantified via ELISA at 46 DPI using a subset of mice from each vaccination group (n = 4 per group). Titer values were Log10 transformed and compared for statistical significance between vaccination groups via an ordinary one-way ANOVA with a Tukey’s multiple comparison test. * p ≤ 0.05 compared to sPA and Alhydrogel. Individual animals are shown with bars indicating mean ± SD.

HIGHLIGHTS.

• Combining CDNs and polyanhydride nanoparticles into a vaccine formulation induced rapid and high titer anti-PA IgG antibody responses in mice

• Combination nanovaccine induced higher serum neutralizing anti-PA IgG antibodies compared to PA adjuvanted with alhydrogel after 15 days post-immunization

• Microarray analysis revealed low overall serum IgG responses to PA linear peptide epitopes in vaccinated mice, suggesting responses to conformational epitopes

• Combination nanovaccine induced long-lived anti-PA IgG-secreting cells in bone marrow and spleen

LIST OF ABBREVIATIONS

ASC

Antibody secreting cell

AUC

Area under the curve

AVA

Anthrax vaccine adsorbed

CDG

Cyclic di-GMP

CDN

Cyclic dinucleotide

CPH

1,6-bis(p-carboxyphenoxy)hexane

CPTEG

1,8-bis(p-carboxyphenoxy)-3,6-diaoctane

DPI

Days post-immunization

ED₅₀

Effective dose at 50% inhibition

EF

Edema factor

LF

Lethal factor

MFI

Mean fluorescence intensity

NO

Nitric oxide

NP

Nanoparticle

OVA

Ovalbumin

PA

Protective antigen

ROS

Reactive oxygen species

sPA

Soluble protective antigen

TLR

Toll-like receptor

TNA

Toxin neutralization assay