Development of Multilayer Nanoparticles for the Delivery of Peptide-Based Subunit Vaccine against Group A Streptococcus

Peptide-based subunit vaccines include only minimal antigenic determinants, and, therefore, are less likely to induce allergic immune responses and adverse effects compared to traditional vaccines. However, peptides are weakly immunogenic and susceptible to enzymatic degradation when administered on their own. Hence, we designed polyelectrolyte complex (PEC)-based delivery systems to protect peptide antigens from degradation and improve immunogenicity. Lipopeptide (LCP-1) bearing J8 B-cell epitope derived from Group A Streptococcus (GAS) M-protein was selected as the model peptide antigen. In the pilot study, LCP-1 incorporated in alginate/cross-linked polyarginine-J8-based PEC induced high J8-specific IgG antibody titres. The PEC system was then further modified to improve its immune stimulating capability. Of the formulations tested, PEC-4, bearing LCP-1, alginate and cross-linked polylysine, induced the highest antibody titres in BALB/c mice following subcutaneous immunisation. The antibodies produced were more opsonic than those induced by mice immunised with other PECs, and as opsonic as those induced by antigen adjuvanted with powerful complete Freund’s adjuvant.


Introduction
Vaccination is one of the most effective medical interventions against infectious diseases [1]. Attenuated and inactivated vaccines have been on the market for over two centuries. However, they can present a variety of adverse effects, as they carry a wide range of materials that do not contribute to achieving required immune responses [2]. To avoid such drawbacks, peptide-based subunit vaccines carry only minimal antigenic determinants (epitopes) required to induce desired immune responses [3]. However, peptide epitopes are weakly immunogenic and susceptible to enzymatic degradation [4]. Thus, co-administration of peptide epitopes with adjuvants/delivery systems to deliver them into antigen presenting cells (APCs) for further processing and presentation to B-and T-cells is required [3,5]. A limited number of adjuvants have been approved for human use. These are generally safe; however, they may also cause undesired immune responses, e.g., excessive inflammation [6]. Instead of coadministration with an adjuvant, peptide antigens can also be conjugated to carriers, such as lipids, to improve their immunogenicity. For example, lipid core peptide (LCP) was able to induce strong humoral immune responses in conjunction with a variety of antigens [7][8][9]. Moreover, increasing hydrophobicity via inclusion of lipid moieties allows vaccines to self-assemble into nanoparticles, preventing premature enzymatic degradation and increasing uptake by APCs [10].
Transmission electron microscopy (TEM; HT7700 Exalens, HITACHI Ltd., Tokyo, Japan) was performed at the Australian Microscopy & Microanalysis Research Facility, Centre for Microscopy and Microanalysis, The University of Queensland (UQ).

Peptide Synthesis & Purification
Peptides of interest were synthesized using fluorenylmethoxycarbonyl solid-phase peptide synthesis (Fmoc-SPPS), following a reported protocol [31] with some modifications. Rink amide MBHA (0.1 mmol) was swollen in DMF for at least 2 h before synthesis. A short treatment with 20% piperidine in DMF was completed prior to each amino acid coupling for Fmoc deprotection (5 min and 10 min). The amino acids (4.2 equivalent) were activated by dissolution in 0.5 M HATU in DMF (4.0 equivalent), then in 6.2 equivalent of DIPEA, where each amino acid was coupled twice (2 × 20 min). After coupling the first amino acid of the sequence, acetylation was performed to act as a cap for unreacted amino groups on the resin. Once all amino acids were coupled, acetylation was performed again prior to peptide cleavage.
In preparation for cleavage, resins were washed with DMF, DCM and MeOH, then stored in the desiccator overnight to dry. Once dried, the peptide sequence was cleaved off from the resin using TFA cocktail, TFA:TIPS:MilliQ water (95%:2.5%:2.5%) [31]. The peptides were verified and assessed through analytical RP-HPLC with a C18 column and mass spectrometry (MS). Purification of the peptides was performed through preparative RP-HPLC-C18. Poly

Oxidation
Pure peptides were oxidized using a general method of disulphide cross-linking. Oxidation was performed on 5 mg of each poly-cationic peptide. However, a mixture of J8-Cys (4.3 mg) and poly-Arg (2.1 mg) was used for oxidation in the synthesis of J8-Cyspoly-Arg. Peptides were dissolved in a freshly prepared 50 mM ammonium bicarbonate solution (200 µL) and stirred overnight, freeze-dried, then 0.1 M hydrochloric acid (0.5 mL) was added. They were then freeze-dried again. To test the success of the disulphide cross-linking reaction, Ellman's test was performed to determine the presence of free thiols, according to Thermo Scientific's user guide for Ellman's reagent [32]. Cysteine standard solution was prepared to quantify free thiols in the samples.

Transmission Electron Microscopy
TEM was used to visualize surface morphology. Each sample was dissolved in PBS, then one drop was added to a glow-discharged carbon-coated grid and allowed to settle for 2 min. Excess liquid was wicked off with filter paper. Particles were then stained with 2% uranyl acetate (pH 7) for 2 min. The excess stain solution was wicked away and the grid was air-dried for 5 min before particle imaging.

Encapsulation Study
To evaluate encapsulation efficiency (EE%), ultracentrifugation of the formulations was employed. A volume of 0.25 mL of each formulation was transferred into a 1 mL polycarbonate thick-wall ultracentrifuge tube. The formulations were further diluted with another 0.25 mL of PBS, then mixed by pipetting up and down for 1 min. The tubes were placed in an ultracentrifuge (Optima™ MAX-XP, Beckman Coulter, Brea, CA, USA) equipped with a TLA120.2 rotor. Sample centrifugation was conducted at 100,000 RPMs for 1 h at 4 • C. LCP-1 concentration was determined as follows: a volume of 100 µL of the supernatant was freeze-dried, redissolved in 0.5 mL of 50% solvent B, and 100 µL of each redissolved sample, as well as a similar volume of standard (LCP-1) lipopeptide solution (1 mg/mL) in solvent A (water + 0.1% TFA) and solvent B (90% Acetonitrile + 10% water + 0.1% TFA) (1:1). These were then put through RP-HPLC (Shimadzu Corp., Kyoto, Japan) using a time gradient program (0 to 100% solvent B, 40 min; C4 column). In parallel, 60 µL of each fresh formulation was diluted with 60 µL of PBS, and subjected to HPLC following the same method as a reference concentration. EE% values were calculated using Equation (1).

Animal Studies
All animal protocols were approved by The University of Queensland Animal Ethics Committee (SCMB/AIBN/069/17). Two animal studies were performed: both used the same number of mice and vaccine administration method.
BALB/c mice (6-8 weeks old; 60 in total: 4 groups in the first, 8 groups in the second) were divided randomly into sample groups (n = 5). The mice were inoculated subcutaneously with freshly prepared vaccine candidates equivalent to 30 µg of LCP-1 on days 0, 25 and 50. Blood was collected via tail bleed 10 days after each immunisation. A final blood collection was done following euthanasia on day 64 via heart puncture. Tail blood samples (20 µL per mouse) were mixed with 180 µL of 1X PBS. The serum was separated via centrifugation at 3600 RPMs for 10 min. All biological samples were stored at −80 • C prior to further analysis.

Determination of Antibody Titres & Statistical Analyses
J8-specific IgG antibody titres from sera were measured via enzyme-linked immunosorbent assay (ELISA), following a published protocol [33]. First, 96-well microplates were coated with J8 (50 µg/plate) in carbonate coating buffer (pH 9.6) and incubated at 37 • C for 90 min. Blocking was done with 5% (w/v) skim milk in PBS overnight at 4 • C. Serum samples were added to the first row of the plate, starting with 1/100 dilution, then serially diluted two-fold down the plates. The plates were then incubated at 37 • C for 90 min. HRP-conjugated goat anti-mouse secondary IgG antibody was added to each well, and the plates were incubated again under the same conditions. OPD substrate was added, and the plates were incubated in dark conditions for 20 min before absorbance measurements were taken at 450 nm (SpectraMax microplate reader, Molecular Devices, San Jose, CA, USA). Endpoint antibody titres were determined as the highest dilution, for which the absorbance value was > 0.065, 3 standard deviations above naïve mean absorbance and where the difference between previous three dilution values was > 0.005. Statistical significance was analysed using one-way ANOVA, followed by Tukey's test for comparison between groups with GraphPad Prism 9.0.

Opsonization Assay of Antibodies against Group A Streptococcus
Antibody opsonization assays were performed with GAS clinical isolate strains provided by the Princess Alexandra Hospital: strains GC203 (wound swab) and D3840 (nasopharynx swab) [34]. The two strains were streaked on Todd Hewitt broth (THB) agar plates with 5% yeast supplementation and incubated at 37 • C for 24 h. Single colonies from the plates were inoculated in 5 mL of THB + 5% yeast and incubated until bacterial replication reached approximately 4.6 × 10 6 colony forming units (CFU)/mL. Sera collected from the heart (day 64) was heat-inactivated at 50 • C for 30 min. The sera were then mixed with a bacterial aliquot and horse blood at a ratio of 1:1:8, respectively. The mixtures were added to 96-well plates containing a 2-fold dilution of bacterial culture. The plates were incubated at 37 • C for 3 h. Then, 10 µL aliquots from the wells were streaked onto plates with THB agar + 5% yeast and 5% horse blood, and incubated for another 24 h [34]. Bacterial survival rates were calculated based on the number of colonies on the plate. Opsonic activity was determined by the percent reduction in average CFU, following the formula below (Equation (2)) [35].

Design of the Vaccine Delivery System
PEI is a highly cationic polymer that has been widely used in the development of vaccine delivery systems due to its adjuvanting capacity [36]. However, it is non-biodegradable and highly toxic. Here, we assessed simple, non-toxic, polycationic molecules as a replacement for PEI, through fully chemically synthesised poly-cationic coating over LCP-1. LCP-1, a self-adjuvanting lipopeptide, was complexed with negatively charged polysaccharide alginate to produce anionic particles. These particles were then coated with biodegradable cationic cross-linked polymers, namely poly-Arg-J8, poly-Arg, mannosylated poly-Arg and poly-Lys, to produce polyelectrolyte complexes PEC-1 to PEC-4, respectively. Based on previous studies, many formulations including poly-Arg and poly-Lys were used safely in vivo without showing any level of toxicity [9,[37][38][39].
All PECs (Table 1) were self-assembled into nanoparticles in water, through dropwise addition of anionic alginate into cationic LCP-1. Corresponding cationic cross-linked polymers were then added dropwise to coat the nanoparticles. Poly-Arg was selected as a replacement for PEI due to its capacity to act as a cell-penetrating peptide [40]. Poly-Lys was selected as a closer analogue of PEI due to its amine group (instead of guanidine in arginine). However, Poly-Lys was selected as a negative control as it has been reported to be an ineffective immune stimulator [41]. The inclusion of a targeting moiety (mannose in PEC-3) on the nanoparticle surface was hypothesized to enhance PEC immunogenicity due to increased interaction with APCs. Finally, J8 B-cell epitope was conjugated to poly-Arg (via disulphide bridge formation) to allow easy recognition of PEC-1 by B-cells. The three polycationic peptides mentioned above, poly-Lys, poly-Arg and mannosylated poly-Arg, were designed to carry three cysteine residues each, while J8 possessed a single cysteine moiety to allow intermolecular oxidative cross-linking and, thus, formation of large polymeric structures mimicking the high molecular weight of PEI. Cross-linking was performed prior to coating of the LCP-1/alginate nanoparticles.

Formation and Characterization of PECs
Peptide polymers were synthesised using classical Fmoc-SPPS (Supplementary Figure  S1) and oxidatively cross-linked using 50 mM ammonium bicarbonate solution overnight before acidification with 0.1 M hydrochloric acid. Free thiol content from unreacted cysteine residue was determined by Ellman's test: no free thiol was detected, reflecting completion of the oxidation process and successful cross-linking between peptide monomers through disulphide linkages. Functionalisation of poly-Arg with GAS B-cell epitope, J8, required synthesis of poly-Arg-J8 via disulphide bond formation between poly-Arg and J8-Cys peptide; this was performed following the same oxidation procedure.

Immunogenicity of PECs
Two animal studies were performed to determinate nanocomplex immunogenicity. The pilot study was designed to establish when a cross-linked cationic peptide can be used as a replacement for PEI. PEC-1 was produced as LCP-1, complexed with alginate, coated with a fully chemically synthesised polycationic layer and cross-linked to poly-Arg-J8. PEC-0, bearing PEI 25 kDa, was used as a reference. BALB/c mice were immunised with PEC-0, PEC-1, P25-J8/CFA as a positive control, and PBS as a negative control. PEC-1 induced similar humoral immune responses to PEC-0 (containing PEI; Figure 2A), proving that toxic PEI can be replaced with cross-linked polycationic peptides. However, neither of these formulations were as effective at triggering antibody responses as CFA-adjuvanted antigen. Thus, to further probe the ability of cationic peptides to enhance immune responses against an antigen, a second in vivo animal study was carried out. Three other complexes (PEC-2 to PEC-4) were examined in comparison to PEC-1; namely, PEC-2, bearing cross-linked poly-Arg (instead of cross-linked poly-Arg-J8 in PEC-1); PEC-3, carrying cross-linked mannosylated poly-Arg; and PEC-4, carrying cross-linked poly-Lys.

Antibody Opsonic Capability
Antibodies produced in the sera by immunised mice were used to determine opsonic activity against two strains of GAS clinical isolates (Figure 3): D3840 and GC2203, isolated from a nasopharynx swab and wound swab, respectively. Opsonic abilities were calculated based on the reduction of CFU compared to sera from negative control mice, determining the quality and effectiveness of the antibodies. Antibodies produced by mice immunised with PEC-4 had comparable opsonic activity to the positive control group (P25-J8/CFA), while those of all other mice were significantly less opsonic. PEC-1 to PEC-3 sera had similar levels of opsonic activity, despite the higher overall dosage of antigen introduced by PEC-1. Thus, cross-linked poly-Lys (PEC-4) was better able to stimulate the im- All PEC constructs induced antigen-specific immune responses at similar levels. IgG titres induced by PEC-4 were not significantly different to those induced by CFAadjuvanted P25-J8 ( Figure 2B). The other PECs were less efficient than P25-J8/CFA in stimulating J8-specific antibody production. It was notable that PEC-1 was able to trigger slightly higher IgG titres than PEC-2 and PEC-3. This can be attributed to the presence of J8 epitope on the surface of PEC-1, promoting prompt recognition by B-cell receptors. Higher antigen dosage was also likely a factor (both LCP-1 and the external coating carried antigen). However, the differences were not statistically significant. Despite caring mannose targeting moiety, PEC-3 did not induce higher IgG production than other vaccine candidates. Uncoated LCP-1/alginate anionic nanoparticles were also effective in triggering IgG production; however, to a lesser extent than PEC-1 and PEC-4. All PECs and controls induced higher production of IgG1 than IgG2a, suggesting that the vaccine candidates induced predominantly Th2-biased immune responses ( Supplementary Figures S12 and S13).

Antibody Opsonic Capability
Antibodies produced in the sera by immunised mice were used to determine opsonic activity against two strains of GAS clinical isolates (Figure 3): D3840 and GC2203, isolated from a nasopharynx swab and wound swab, respectively. Opsonic abilities were calculated based on the reduction of CFU compared to sera from negative control mice, determining the quality and effectiveness of the antibodies. Antibodies produced by mice immunised with PEC-4 had comparable opsonic activity to the positive control group (P25-J8/CFA), while those of all other mice were significantly less opsonic. PEC-1 to PEC-3 sera had similar levels of opsonic activity, despite the higher overall dosage of antigen introduced by PEC-1. Thus, cross-linked poly-Lys (PEC-4) was better able to stimulate the immune system and produced higher opsonic antibody titres compared to poly-Arg (PEC-1 to PEC-3). Interestingly, simple poly-Arg (not cross-linked) was previously shown to be effective in inducing humoral immunity [42], while poly-Lys lacked immune stimulating ability [41]. The higher efficacy of poly-Lys compared to poly-Arg in the current study may be related to non-specific uptake of poly-Arg (as a cell penetrating peptide) by non-immune cells [40]. Alternatively, cross-linking of poly-Lys could have altered its immune stimulating potency.

Conclusions
Nanoparticles have been widely explored for the delivery of vaccines where encapsulation of antigens with commercially available cationic polymers, such as chitosan or PEI, is utilized. However, further modifications, such as the incorporation of targeting moieties to enhance immune responses, can be challenging. In this study, we demonstrated that a fully chemically synthesized polycationic cross-linked poly-Lys peptide, but not poly-Arg, can have similar capacity as CFA in stimulating humoral immunity. While there was no significant difference in the ability to trigger opsonic antibody production

Conclusions
Nanoparticles have been widely explored for the delivery of vaccines where encapsulation of antigens with commercially available cationic polymers, such as chitosan or PEI, is utilized. However, further modifications, such as the incorporation of targeting moieties to enhance immune responses, can be challenging. In this study, we demonstrated that a fully chemically synthesized polycationic cross-linked poly-Lys peptide, but not poly-Arg, can have similar capacity as CFA in stimulating humoral immunity. While there was no significant difference in the ability to trigger opsonic antibody production between the PECs that were derivatised with mannose, B-cell epitope, or neither, having flexibility in derivatisation is advantageous for developing delivery systems. Importantly, this study showed that cross-linked polycationic peptides can be used to replace PEI for the development of safe vaccine delivery systems.

Data Availability Statement:
The data presented in this study are available in this article.