Vinyl Sulfone-functionalized Acetalated Dextran Microparticles as a Subunit Broadly Acting Influenza Vaccine

Abstract

Influenza is a global health concern with millions of infections occurring yearly. Seasonal flu vaccines are one way to combat this virus; however, they are poorly protective against influenza as the virus is constantly mutating, particularly at the immunodominant hemagglutinin (HA) head group. A more broadly acting approach involves Computationally Optimized Broadly Reactive Antigen (COBRA). COBRA HA generates a broad immune response that is capable of protecting against mutating strains. Unfortunately, protein-based vaccines are often weekly immunogenic, so to help boost the immune response, we employed the use of acetalated dextran (Ace-DEX) microparticles (MPs) two ways: one to conjugate COBRA HA to the surface and a second to encapsulate cGAMP. To conjugate the COBRA HA to the surface of the Ace-DEX MPs, a poly(L-lactide)-polyethylene glycol co-polymer with a vinyl sulfone terminal group (PLLA-PEG-VS) was used. MPs encapsulating the STING agonist cGAMP were co-delivered with the antigen to form a broadly active influenza vaccine. This vaccine approach was evaluated in vivo with a prime-boost-boost vaccination schedule and illustrated generation of a humoral and cellular response that could protect against a lethal challenge of A/California/07/2009 in BALB/c mice.

Introduction

Influenza is an ancient plague that has been with us since antiquity, with reports of possible infection written in Greek writings as far back as 412 BC (1). Influenza led to at least four pandemics in the last century, with the 1918 Pandemic (Spanish Flu) resulting in the most deaths, totaling to a loss of nearly 20 million (2). Seasonal vaccines are one way to combat influenza infections. The immunodominant protein for influenza vaccines is the hemagglutinin (HA) protein. HA is a glycoprotein that is located on the surface of the influenza virus. As a trimer, HA binds to glycans present on the upper respiratory tract which allows internalization inside the cell (3). The difficulty with generating immunity against the HA protein is the natural antigen drift and shift that occurs in this protein, while the virus is endemic in the human and animal reservoir population (4). Indeed, the changes to HA antigen over the course of the flu season can account for some of the observed decrease in vaccine efficacy seen with seasonal vaccines. Therefore, there is a need to design a vaccine that is capable of generating protection despite the mutations that may occur throughout the year.

One novel approach in generating an effective immune response against a mutating HA protein is the Computationally Optimized Broadly Reactive Antigen (COBRA) methodology. It employs multiple rounds of layered consensus building to generate HA proteins that are protective against a broad range of influenza strains (5–14). For example, we have designed the Y2 H1 protein using COBRA methodology derived from H1N1 isolates that were circulating during 2013–2019 flu seasons (6). Based on this protein, we are able to generate an effective immune response that protected against a wide variety of H1 strains. However, like with most sub-unit vaccines, there is a need to adjuvant the vaccine to increase the immune response against the protein (15).

One way to adjuvant the vaccine response is through microparticles (MPs), as previously developed for several subunit vaccines (16, 17). Polylactic-co-glycolic (PLGA) particles have been advanced that have encapsulated both adjuvants and antigens as a carrier for vaccines (18, 19). One advantage of a PLGA particle is their degradability; however, PLGA particles degrade rather slowly at a constant rate regardless of whether they are extracellular or intracellular. In contrast, we have developed acetalated dextran (Ace-DEX) for vaccine applications (20, 21) which has a broader range of degradation rates and is acid-sensitive, making it more stable in the extracellular environment than the intracellular phagosome where the pH can drop to near 5 (20–23). We have previously used Ace-DEX microparticles (MPs) for developing vaccines against influenza.

For this work, we sought to increase the robustness of Ace-DEX microparticles for vaccination. Previously, our research involved either mixing particles encapsulating adjuvant with soluble antigen (24) or encapsulating both antigen and adjuvant in MPs (25, 26). We hypothesized that if we conjugated the particles on the surface of our MPs, we could enhance the activity of our system, specifically allowing multivalent antigen orientation for enhanced B cell (27–29) and other immune responses (30). We decided to explore the conjugation of protein to the surface of particles using the vinyl sulfone functional group. Vinyl sulfone has been used in the past to conjugate proteins to the solid supports in protein chemistry (31). We utilized this functional group since it can react with thiol groups present on the protein. Ace-DEX MPs with COBRA ligated to the surface via vinyl sulfone was used in combination with cGAMP MPs to form an injectable broadly acting influenza vaccine. This platform was evaluated for humoral, cellular, and protection responses in a mouse model.

Materials and Methods

Poly(L-lactide)-b-poly(ethylene glycol)-vinylsulfone block co-polymer (PLLA-PEG-VS) was purchased from Nanosoft Polymers. All other chemicals were purchased from Sigma (St. Louis, MO) and used as purchased, unless otherwise indicated. Assays, biologics, and disposables were purchased from Thermo Fisher Scientific (Waltham, MA) unless otherwise indicated.

Synthesis

Acetalated Dextran (Ace-DEX) Polymer

Ace-DEX polymer was synthesized as previously described (21). Briefly, 71 kDa dextran dissolved in dimethyl sulfoxide (DMSO) with catalyst pyridinium p-toluenesulfonate was reacted with 2-ethoxypropene (Matrix Scientific, Columbia, SC) under anhydrous conditions. Reaction mixture was quenched with triethylamine (TEA) after 7 min, precipitated in basic water (0.04% v/v TEA in water), and lyophilized for 1 day. Next, Ace-DEX was dissolved in ethanol, centrifuged, precipitated in basic water, and lyophilized again. The final product was found to have a cyclic acetal coverage (CAC) of 39% by ¹H 400 MHz NMR (Inova). Identical procedure was followed for the synthesis of Ace-DEX with CAC 60%, except for longer reaction time (2 h). Our group has shown previously that polymers with a high CAC have a slower degradation compared to high CAC polymers (32). Ace-DEX (CAC 39%) was used in synthesis of Ace-DEX-PLLA-PEG-VS microparticles to provide a polymer with moderately slow degradation kinetics. Ace-DEX (CAC 60%) was used in synthesis of cGAMP-loaded Ace-DEX microparticles given previous studies that have found CAC 60% Ace-DEX to be the most efficacious for delivering cGAMP (33).

Generation of Ace-DEX-PLLA-PEG-VS Microparticles

Ace-DEX-PLLA-PEG-VS MPs (VS MPs) were synthesized from Ace-DEX polymer (CAC 39%) and PLLA-PEG-VS polymer (Nanosoft Polymers, Cat. 8562, PLLA 10 K, PEG 2 K) using a sonicator (QSonica Q500). Glassware used for the synthesis was pre-soaked in a base bath (1.0 M sodium hydroxide) overnight to ensure no contamination with endotoxin. A mixture of PLLA-PEG-VS (10 mg in 0.5 mL of dichloromethane), Ace-DEX (40 mg in 0.5 mL of dichloromethane) and 2 mL of 3% polyvinyl alcohol (PVA) in phosphate buffer saline (PBS) was vortexed briefly and then sonicated for 1 min. The obtained emulsion was poured into a beaker with 20 mL of 0.3% PVA in PBS, and the mixture was stirred for 2 h, allowing for the evaporation of dichloromethane. After centrifugation, the pellet was washed twice and re-suspended in basic water (2 mL) and mixed with 4 mL of sucrose solution (75 mg/mL) in basic water. The obtained suspension was frozen at −80°C and lyophilized to produce VS MPs in sucrose as a white powder.

The HA used in this study is the Y2 H1 protein which is the next generation of COBRA H1 based on the COBRA methodology derived from H1N1 isolates that circulated during 2013–2019 flu seasons (34). The protein was expressed by HEK293T cells as previously described (35). Sterile carbonate buffer (pH = 9.08) was treated with ethylenediaminetetraacetic acid (EDTA) (58 mg/mL) and degassed overnight to form the carbonate buffer used below. HA (835.5 μg in 150 μL of PBS) was diluted with carbonate buffer (812 μL) and treated with Traut’s reagent (2 mg/mL in carbonate buffer), rotated on a tube rotator for 2 h at room temperature, and desalted to remove excess of unreacted Traut’s reagent via 10 K Amicon^® centrifugal filter (70.9% recovery of HA). The protein concentrate (568.2 μg in 94 μL carbonate buffer) was added to the sucrose-formulated VS MPs (20.4 mg) suspended in carbonate buffer (210 μL) and well mixed at room temperature for 18.5 h. After that the solid was isolated by centrifugation (21,100 × g, 30 min, 4°C), washed twice, and resuspended in basic water and lyophilized. HA loading in VS-HA MPs was detected using fluorescamine assay following Thermo Fisher protocol adapted for a NanoDrop. HA loading was found to be 4.6 ± 2.7 μg HA/1 mg MPs, or 0.46 ± 0.27 mass % HA (average ± standard deviation, n = 4).

To generate the capped VS group (capped MPs), VS MPs (35.68 mg total) were suspended in carbonate buffer (370 μL) with 2-mercaptoethanol (11.75 μL) and mixed at room temperature for 21 h. The particles were isolated by centrifugation, washed, and resuspended in basic water, and lyophilized. The disappearance of signals corresponding to VS group was confirmed by ¹H 500 MHz NMR (Supplementary Fig. S1).

cGAMP-Loaded Ace-DEX Microparticles

3′3′-Cyclic guanosine monophosphate–adenosine monophosphate (cGAMP) (Invivogen; tlrl-nacga) MPs were fabricated using an electrospray apparatus as previously described (36). Briefly, Ace-DEX (CAC 60%) and cGAMP were dissolved in 90:10 (v/v) ethanol:water at 1% loading, sonicated briefly, loaded into gas tight glass syringe (Hamilton), placed on a syringe pump (Harvard Apparatus), and connected to a high voltage power supply (Gamma High Voltage Research). A stainless-steel plate was also connected to a separate high voltage power supply and placed under the syringe pump for subsequent particle collection. To ensure the removal of endotoxin, the syringe was soaked in 1.0 M sodium hydroxide overnight, and the stainless-steel plate was heated in a 260°C oven for 2 h. Once the apparatus was setup, the power supplies were switched on and set to +2.5 kV and −6 kV for the plate and needle, respectively. The syringe pump was then set to 200 μL/h. After the spray was completed, white powder MPs on the plate were collected. To quantify cGAMP loading, the MPs were dissolved in a 1:3 (v/v) mixture of 0.1% TFA in acetonitrile and 0.1% TFA in water and analyzed by HPLC. cGAMP loading was found to be 9.58 ± 0.03 μg cGAMP/1 mg MPs or 0.958 ± 0.003 mass % cGAMP (average ± standard deviation, n = 3).

Particle Characterization

Endotoxin was evaluated using the Pierce LAL chromogenic endotoxin quantitation kit in accordance with the manufacturer instructions. cGAMP MPs and Ace-DEX-PLLAPEG-VS-capped MPs had undetectable levels of endotoxin (<0.1 EU/mg). Soluble HA had 0.054 EU/dose, and Ace-DEX-PLLA-PEG-VS-HA MPs had 0.093 EU/dose (based on 0.3 μg HA dose given to mice). SEM (Hitachi S-4700 with EDS, Tokyo, Japan) was carried out at the Chapel Hill Analytical and Nanofabrication Laboratory (CHANL).

Vaccination and Humoral Response

Studies were conducted in accordance with National Institutes of Health’s guidelines for the care and use of laboratory animals and approved by the Institutional Animal Care and Use Committee at UNC. Female BALB/c mice aged 6–8 weeks were obtained from Jackson Laboratories (Bar Harbor, ME). Mice were vaccinated intramuscularly on days 0, 21, and 35 with 25 μL into each leg of the mouse for a total of 50 μL. Antigen and adjuvant solutions/suspensions were combined at a 1:1 volumetric ratio on ice 20 min prior to injection for all groups. Sera were collected on days 14, 28, and 41 after the prime immunization by submandibular bleeds using Medipoint Goldenrod 4 mm lancets and isolated using MiniCollect tubes (Greiner 450,472). Mice were vaccinated with 0.3 μg COBRA HA (type Y2) in the indicated form and either AddaVax or 1 μg cGAMP in Ace-DEX MPs. The adjuvant dose was determined based on previous dose ranging experiments as well as safety evaluation of the cGAMP MPs (24).

Antibody titer determination was performed as described previously (37), with the following modifications: 384-well ELISA plates (Microlon 600, Greiner 781,061–25) were coated at 4°C overnight with 100 ng/mL Y2 HA in PBS. The secondary antibodies used were Goat Anti-Mouse IgG Fc-HRP 1033–05, Goat Anti-Mouse IgG2a-HRP 1081–05, or Goat Anti-Mouse IgG1-HRP 1071–05 (Southern Biotech). Development was performed with tetramethylbenzidine (TMB) One Component HRP Microwell Substrate (Surmodics TMBW100001).

The antibody dependent cellular phagocytosis (ADCP) assay was adapted from a previously reported method (38, 39). Using carboxylate-modified fluorescent microspheres (Fisher F8821) Y2 COBRA HA was conjugated via sulfo-N-hydroxysuccinimide (NHS) and 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC), while residual reactive sites were quenched with glycine. Antigen-coated beads were incubated 1 h in diluted sera and cultured with macrophages (ATCC TIB-67) that had been seeded overnight at 50,000 cells/well in a non-tissue culture treated 96-well plate. After washing cells with 5 mM EDTA in PBS, they were fixed with 3% paraformaldehyde before flow cytometry (Attune NxT Flow Cytometer; Thermo Fisher). To calculate phagocytic score, the percent of particle positive cells was multiplied by the geometric mean fluorescence intensity in the particle channel divided by ten thousand.

The antibody dependent complement deposition (ADCD) assay was performed as previously reported (40), with the same Y2 HA-conjugated beads as in the ADCP assay and HEPES buffer in place of previously used veronal buffer (41).

Cellular Response Characterization and Lethal Challenge

Spleens from vaccinated mice (n = 5) were isolated on day 42 and processed into a single-cell suspension by grinding through a 40-μm cell strainer (Corning 352340). Red blood cells were lysed with ACK lysis buffer, passed through another 40 μm strainer, and plated in RPMI Media with 2.05 mM l-glutamine, 2 mM HEPES, 100 U/mL penicillin, 100 μg/mL streptomycin, 1 mM sodium pyruvate, 20 μM 2-mercaptoethanol, and with or without (negative control) 10 μg/mL Y2 HA. Some cells were also stimulated with 5 μg/mL ConA as a positive control. Cell suspensions were plated at 2 × 10⁵ in a pre-coated 96-well hydrophobic PVDF filter plate (Millipore S2EM004M99) for ELISPOT or 1 × 10⁶ cells/U-bottom well plate (Corning 351,177) for ELISA and incubated 72 h at 37°C and 5% CO₂. ELISPOTs were then developed according to the manufacturer protocol (BD 551881 for IFN-γ, 551876 for IL-2). For ELISAs, cells were pelleted by centrifuging 5 min at 500 × g, and the supernatant was assessed for cytokine levels according to the manufacturer protocols (IL-17A Biolegend 432501, IL-4 Biolegend 431101, IL-2 Invitrogen 88702488, IFN-γ Invitrogen 88731486, IL-6 88706477).

On day 56 after vaccination, mice (n = 10 per group) were lightly anesthetized with isoflurane and infected with 1 × 10⁵ pfu of H1N1 Influenza A/California/07/2009. Mice were administered the virus intranasally by pipetting 25 μL of virus into their nose allowing them to rest and then pipetting another 25 μL into their nose. Mice were monitored daily for weight loss and body condition score and euthanized upon losing 20% of their initial body mass.

Results and Discussion

We sought out to conjugate the H1 COBRA antigen, Y2, onto the surface of Ace-DEX MPs. We hypothesized that conjugation to the surface would increase the antibody titers against the Y2 protein since B cells could potentially interact with the antigen on the surface of the particles. Prior research has shown that conjugation to the surface of a particle increases the immune response compared to the soluble version of the antigen (42). In order to do this, we blended with Ace-DEX a block co-polymer PLLA-PEG-VS to synthesize MPs (VS MPs) as seen in Fig. 1a. We formed Ace-DEX MPs via emulsion process that resulted in uniform spheres approximately 150 nm diameter (Supplementary Fig. S2A). We hypothesize that upon oil-in-water emulsification, the hydrophobic PLLA end of PLLA-PEG-VS polymer migrates to nonpolar droplet interior with Ace-DEX, while the hydrophilic PEG-VS end of PLLA-PEG-VS polymer migrates to aqueous continuous phase and as a result decorates the surface of the formed particle (Fig. 1a, b). HA-loaded, VS MPs (VS-HA MPs), and VS MPs that were capped with 2-mercaptoethanol (capped MPs) were synthesized from VS MPs utilizing vinyl sulfone functional group chemistry and its reactivity toward thiol groups (Fig. 1a). VS groups available on the surface of VS MPs were reacted with thiol group of HA (COBRA Y2) antigen, which led to surface-conjugation loading of the antigen and to the formation of VS-HA MPs. A second set of MPs were reacted with 2-mercaptoethanol, which led to conversion of VS group into sulfide group and to the formation of capped MPs as a control. In order to improve efficiency of HA surface conjugation, HA was treated with conventional thiolating agent, 2-iminothiolane (Traut’s reagent) prior to reaction with VS MPs to increase number of thiols available for the surface conjugation reaction. HA loading in/on VS-HA microparticles was detected with fluorescamine assay and was found to be 4.6 μg HA/1 mg MPs. The morphology of VS-HA MPs was characterized by SEM (Fig. 1c; Supplementary Fig. S2B) and their morphology resembled that of VS MPs (Supplementary Fig. S2A); morphology of capped MPs was also similar (Supplementary Fig. S2C) to that of VS MPs.

Fig. 1.

VS = vinyl sulfone; MPs = microparticles. a Schematics showing formation of VS-functionalized polymeric microparticles (VS MPs) from acetalated dextran polymer (Ace-DEX) and PLLA-PEGVS polymer upon sonication. VS group is further reacted with a thiol group to form either VS-HA MPs (when reacted with thiolated HA) or capped MPs (when reacted with 2-mercaptoethanol). b Schematics showing loading of thiolated HA into VS MPs via surface conjugation. c Scanning electron micrograph of VS-HA MPs

cGAMP-loaded Ace-DEX MPs were fabricated using an electrospray apparatus as previously described (36). cGAMP loading was detected with HPLC and was found to be 9.58 μg cGAMP/1 mg MPs. The morphology of cGAMP-loaded Ace-DEX MPs was characterized by SEM and revealed uniform microparticles with approximately 0.5 μm diameter (Supplementary Fig. S2D). cGAMP Ac-DEX MPs were used as an adjuvant to enhance the antigenicity of the COBRA HA, as previously reported by our group (24, 33, 43, 44). Further cGAMP MPs were delivered as separate particles from the HA conjugated particles due to a previous finding in our group where co-encapsulated adjuvant and antigen did not perform as well as when the adjuvant and antigen were in separate particles (33, 45). Also, we discuss in a review other studies comparing two particles versus one for vaccine delivery and note there are often not significant differences (16).

Since we were able to functionalize HA onto the surface of Ace-DEX MPs, we decided to test the efficacy of our formulation in vivo. We immunized female BALB/c mice intramuscularly on days 0, 21, and 35. Mice were vaccinated with either PBS, soluble HA adjuvanted with AddaVax, capped MPs mixed with soluble HA and cGAMP MPs (capped MPs + HA + cGAMP MPs) as an unconjugated mixed control, or with the HA conjugated VS MPs with cGAMP MPs (VS-HA MPs + cGAMP MPs) (Fig. 2a). As seen in Fig. 2b–d, mice were bled on days 14, 28, and 42, and antibody titers were measured. The oil in water emulsion AddaVax was used as a positive control for these vaccine experiments since it is similar to the FDA approved adjuvant MF59 which is the only adjuvant used in an FDA approved influenza vaccine (FluAd). AddaVax has been shown to induce a potent humoral immune response but does not produce a strong cellular response by itself (46). For both total IgG and IgG1, the antibody titer for HA mixed with AddaVax, HA mixed with capped MPs, or HA conjugated to Ace-DEX MPs had similar trends where day 14 had the lowest titer, and at day 28 and 42, the maximum titers was reached. For IgG2a in general, the antibody titer increased over time for the groups immunized with HA. The VS-HA and HA with capped MPs groups both had a statistically significant IgG2a response compared to AddaVax. This is a very important outcome for an influenza vaccine since previous reports have shown that it essential for protection to generate IgG2a antibodies for an influenza vaccine (47, 48). There was no significance between VS-HA and capped MPs with HA.

Fig. 2.

n = 15 female BALB/c mice per group were vaccinated on a prime-boost-boost schedule with PBS, 0.3 μg COBRA HA (type Y2) in the indicated form, and either AddaVax or 1 μg cGAMP in Ace-DEX MPs as indicated (a). Sera were collected on days 14, 28, and 42 and anti-HA total IgG (b), IgG1 (c), and IgG2a (d) titers were evaluated. Day 42 sera were also evaluated for HA-specific (e) antibody-dependent phagocytosis, (f) antibody-dependent complement deposition effector functions, and (g) hemagglutinin inhibition titers against multiple isolates of recently circulating influenza viruses. A one-way ANOVA with Tukey’s multiple comparisons test was performed for effector function assays. *p ≤ 0.05, ****p ≤ 0.0001

Apart from antibody titer, it is also important to consider the functionality of the antibodies generated by a vaccine (49). For example, the Fc region of certain antibodies can bind to a phagocytic cell via a Fc receptor to induce phagocytosis which is known as antibody dependent cellular phagocytosis (ADCP). Further, the Fc region of antibodies can also bind to complement and initiate the complement cascade in a process known as antibody dependent complement deposition (ADCD). These effector functions have been shown to be correlated to protection against influenza (50). To this end, we examined these effector functions with an ADCP and an ADCD assay. In the ADCP assay (Fig. 2e) AddaVax, capped MPs, and VS-HA had similar capacity to induce phagocytosis. We were surprised by this result since previous reports had shown that Th1 type antibodies (IgG2a) are more efficient in inducing ADCP compared to Th2 antibodies (IgG1) in human cells (51). Since even AddaVax generated a small IgG2a response, perhaps for an ADCP response, there is a low threshold for IgG2a to assist in phagocytosis. When looking at ADCD (Fig. 2f), the capped MPs with HA did not have a statistically significant increase when compared to AddaVax, while the VS-HA was statistically greater when compared to AddaVax. One of the main effector functions of IgG2a is to increase compliment deposition which is potentially why the VS-HA was higher in complement deposition when compared to AddaVax. In Fig. 2g, we measured the hemagglutination of the antibodies (HAI) against various H1 strains of influenza. There was no statistical significance between the three groups that were tested. The HAI data shows that the conjugation of the protein to the surface of our microparticles did not diminish the three-dimensional structure of the HA protein since HAI would require an intact three-dimensional structure of the antigen for proper inhibition. The VS group we used in these studies is a very powerful tool that can be used to conjugate other proteins to the surface of particles such as antibodies (52), enzymes (53), or other antigens (54) while maintaining the 3 dimensional structure of the protein. Overall, the VS-HA and capped MPs groups had a similar response as the AddaVax group. The VS-HA MP group had a nonstatistical advantage in generating Th1 antibody responses as seen in Fig. 2d, f. Additionally, at earlier time points, the VS-HA MP group had a higher average in antibodies that could help in earlier protection.

After we immunized the mice described in Fig. 2, we took the spleens on day 42 which was 7 days after the last immunization and cultured the splenic cells with antigen for 3 days. After 3 days, the supernatants were collected and cytokines were measured by ELISA (Fig. 3a–e) and ELISpot (Fig. 3f, g). As seen in Fig. 3a, the VS-HA MP group had a statistically significant increase in IL-2 production compared to the other groups. Prior research has shown that IL-2 can boost pathogen-specific T cell responses which in turn reduce immunopathology and help clear influenza infection (55). For the other cytokines measured by ELISA, they were not statistically significant. For ELISPOT, the VS-HA group was statistically significant compared to the AddaVax group for IL-2 and IFN-γ. IFN-γ has been shown to be an essential cytokine necessary for the clearance of influenza during infection (56). Overall, with the use of VS-HA microparticles we were able to generate a robust cellular immune response. Based on this encouraging in vivo data, we evaluated the protective our VS-HA MPs in a mouse model using a lethal challenge of influenza.

Fig. 3.

On day 42 after vaccination, splenocytes were isolated from n = 5 mice per group and stimulated with 10 μg/mL COBRA HA (type Y2) for 72 h, and cytokine expression was assessed by (a–e) ELISA or (f–g) ELISPOT. A one-way ANOVA with Tukey’s multiple comparisons test was used to compare groups. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

After three immunizations, mice were challenged with a lethal dose of the CA/09 influenza strain (Fig. 4). The negative control mice had the greatest loss in body weight with the maximum weight loss being at day 6 (Fig. 4a) (Supplementary Fig. S3), while the other mice were able to maintain their weight throughout the challenge period. For the PBS control, two mice out of ten did not die but still lost a significant amount of weight. None of the mice died in the other controls. This is to be predicted since the AddaVax, capped MPs with HA, and the HA-VS MPs produced a large humoral response. The three adjuvanted groups induced a large enough humoral and cellular response to provide protection.

Fig. 4.

n = 15 female BALB/c mice were vaccinated on a prime-boostboost schedule with PBS, 0.3 μg COBRA HA (type Y2) in the indicated form, and either AddaVax or 1 μg cGAMP in Ace-DEX MPs as indicated. On day 56 after vaccination, n = 10 mice were challenged intranasally with 1 × 10⁵ pfu of Influenza A/California/07/2009, and their a body weights and b overall survival were monitored

In the studies presented, there were a few differences between the HA-VS MPs and capped MPs with HA. There was a significant increase in the IL-2 produced by splenocytes compared to the other groups. In addition, there was a significant increase in complement deposition in sera isolated from mice vaccinated with HA-VS MPs compared to the AddaVax control, while the capped MPs with HA group was not statistically significant. When exploring the amount of splenocytes secreting IFN-γ, cells isolated from both the HA-VS MPs and the capped MPs with HA groups were significantly greater than in the AddaVax vaccinated group. This trend was further observed in the generation of IgG2a antibody. This effect is likely due to the adjuvant cGAMP which is able to produce a potent cellular response through activation of the STING pathway. Regardless, overall, it did not seem that conjugating the particles onto the surface of the MPs was essential in generating a protective immune response. This is contradictory to what is published in the literature. Hanson et al. developed lipid coated microparticles (LCMPs) where they were able to conjugate particles on the surface and compared it to antigen that was not conjugated (57). They showed that the surface conjugation improved immune response. Perhaps the difference between our results and theirs is that their MPs had no adjuvant in them, while our MPs had a STING agonist which is a very powerful adjuvant. We hypothesize that whatever difference there is between the conjugated and non-conjugated groups, the differences can be mitigated when adding a strong adjuvant to the MP formulation. Another difference between our system and others is covalent attachment versus non-covalent attachment. For example, Sia et al. have previously showed that the display of HA on the surface of CoPoP liposomes by non-covalent attachment results in improved immune responses compared to mixed controls (58). It could be possible that the geometry of the displayed HA is more favorable for these non-covalent liposome systems than compared to ours. Conjugating protein to the surface of a MP is rather difficult, and if the advantages of conjugation are negligible when a potent adjuvant is added, conjugation to the surface for in vivo applications seems limited in its appeal. In future research, we potentially could lower the amount of cGAMP to tease out the dose that will allow the differentiation between conjugated and non-conjugated group.

Conclusion

There is a need for developing novel vaccines for influenza infections. COBRA technology was used in the design of a novel H1 antigen for the vaccination against influenza that we then conjugated to the surface of Ace-DEX MPs with the use of vinyl sulfone-based conjugation by incorporating a PLLA-PEG-VS block-copolymer in the formulation process. Our results indicate that vinyl sulfone is capable of binding to the thiols present on the HA protein, conjugating the protein to the MP surface. When used as a vaccine in a mouse model, we show that our conjugation strategy is capable of inducing functional antibodies that induce neutralization of the virus. This implies that our conjugation strategy is capable of maintaining the three-dimensional structure of the conjugated antigen. Further, we show that conjugation can induce some statistically significant cellular response compared to the AddaVax control, as well as generate a higher generation of IgG2a antibody titer. However, of the measurements performed to characterize the immune response, conjugation overall was not that significant compared to just mixing the particles with our MP technology. In future studies, decreasing the dose of the adjuvant potentially could tease out the difference between conjugation and non-conjugation. Also, we could investigate using polymers with different degradation rates to see whether this modifies the vaccine efficacy.