Multiplexed electrospray enables high throughput production of cGAMP microparticles to serve as an adjuvant for a broadly acting influenza vaccine

Abstract

Subunit vaccines employing designer antigens such as Computationally Optimized Broadly Reactive Antigen (COBRA) hemagglutinin (HA) hold the potential to direct the immune response toward more effective and broadly-neutralizing targets on the Influenza virus. However, subunit vaccines generally require coadministration with an adjuvant to elicit a robust immune response. One such adjuvant is the stimulator of interferon genes (STING) agonist cyclic dinucleotide 3′3′-cyclic guanosine monophosphate-adenosine monophosphate (cGAMP). We have shown that encapsulation of cGAMP in acetalated dextran (Ace-DEX) microparticles through electrospray results in significantly greater biological activity. Electrospray is a continuous manufacturing process which achieves excellent encapsulation efficiency. However, the throughput of electrospray with a single spray head is limited. Here we report the development of a multiplexed electrospray apparatus with an order of magnitude greater throughput than a single-head apparatus. Physicochemical characterization and evaluation of adjuvant activity in vitro and in vivo indicated that microparticles produced with the higher throughput process are equally suited for use as a potent vaccine adjuvant to induce a balanced immune response to COBRA HA antigens.

1. Introduction

Influenza A is estimated to cause as many as 645,000 deaths worldwide every year.(Iuliano et al., 2018) Although vaccination is the primary countermeasure employed to reduce the immense societal burden of influenza virus infection in humans, current vaccines do not provide an ideal degree of protection, with efficacy ranging from 19% to 60% in the years 2004 to 2015.(Belongia et al., 2016) Random mutations combined with the selective pressure exerted by host immune responses drive the process of antigenic drift, by which the sequences of the antigenic proteins of influenza viruses change over time and escape host immune responses. For reasons that are not completely understood, humoral responses to influenza infection and vaccination are predominantly directed towards the most rapidly-mutating parts of the influenza virus, e.g. the ‘head’ region of the hemagglutinin fusion protein (HA), necessitating annual vaccination to keep up with the ever-changing antigenic landscape of circulating influenza viruses.(Guthmiller et al., 2021; Kirkpatrick et al., 2018).

While influenza virus vaccines have traditionally consisted of virus grown in eggs or cells that are then inactivated (or split), vaccine development has recently been focused on development of subunit vaccines, which contain only the target antigen of interest rather than the entire inactivated virus. Subunit vaccines offer a number of advantages compared to inactivated virus vaccines, such as directing the immune response focused toward a specific target antigen, the potential to use designer antigens regardless of their effect on viral replicative capacity, and avoiding the risk of accruing mutations during virus production.(Lin et al., 1997; Wei et al., 2020) A particularly promising group of subunit antigens for influenza vaccination are Computationally Optimized Broadly Reactive Antigen (COBRA) forms of HA. These antigens are generated through multiple rounds of consensus generation using sequences of HA from currently-or-recently circulating influenza viruses and are capable of eliciting broadly-reactive antibody responses.(Giles et al., 2012; Wong et al., 2017) Recently, Huang et al. reported a new generation of COBRA HA antigens termed Y2 and Y4, which can elicit cross-reactive antibodies against multiple H1N1 viruses and protect against pandemic influenza challenge.(Huang et al., 2021) However, subunit vaccines such as COBRA HA may benefit from coadministration with adjuvant compounds to achieve sufficient immunogenicity, especially in younger, older, or immunocompromised individuals. For example, coadministration of the adjuvant MF59 with the seasonal split virus vaccine is indicated in patients over the age of 65 in order to induce a sufficiently protective immune response. While employing adjuvants like MF59 has been shown to increase the protective efficacy of influenza vaccines,(Domnich et al., 2017) this adjuvant, along with most other FDA-approved adjuvants, induces a largely Th2-skewed immune response,(Valensi et al., 1994) resulting in limited engagement of the cellular arm of the immune response, which can play a critical role in control of viral infection and pathogenesis.(Bungener et al., 2008; Moran et al., 1999) Thus, a demand still remains for a potent vaccine adjuvant capable of inducing a balanced Th1 and Th2 immune response.

We have recently developed an adjuvant formulation of the cyclic dinucleotide molecule 3′3′-cyclic guanosine monophosphate-adenosine monophosphate (cGAMP). As a potent agonist of the stimulator of interferon genes (STING) receptor, this molecule can drive phosphorylation and nuclear translocation of interferon regulatory factor 3 (IRF3) and nuclear factor-κB (NFκB) to induce the translation of type 1 interferons and other proinflammatory cytokines.(Diner et al., 2013) In the context of vaccination, the cGAMP-induced secretion of these cytokines in the presence of antigen can stimulate induction of a robust immune response by recruitment, activation, and expansion of various immune cell populations.(Le Bon et al., 2006; Montoya et al., 2002; Welsh et al., 2012) However, the cytosolic location of STING limits the activity of soluble cGAMP, a hurdle overcome by our group through incorporation into microparticles (MPs) composed of acetalated dextran (Ace-DEX). Ace-DEX is a biocompatible polymer derived from FDA-approved dextran by acetalation, a one-step, reversible modification which renders the polymer hydrophobic, allowing the formation of acid-responsive particulate carriers that can degrade in the endo/lysosome to release encapsulated cargo.(Broaders et al., 2009) We have previously shown that cGAMP can be encapsulated into Ace-DEX with high (>90%) efficiency using an electrospray fabrication process. Further, interferon production was significantly greater in dendritic cells cultured with electrosprayed Ace-DEX particles loaded with cGAMP, compared to the same amount of agonist in a liposome, particles of other biomaterials, or Ace-DEX particles fabricated through other methods.(Junkins et al., 2018; Watkins-Schulz et al., 2019)

Electrospray occurs when a charged solution is passed through a capillary, inducing ionization and atomization of the liquid, which leads to formation of small charged droplets. Evaporation of the solvent in the droplets leads to the formation of micro- and nano-sized particles which can be collected on a grounded or oppositely charged substrate. In addition to offering a high degree of encapsulation efficiency of hydrophilic cargoes into hydrophobic carriers, electrospray is also a continuous, low-shear manufacturing process with the potential for high-throughput fabrication of particles for drug delivery. However, the lab scale apparatus used to generate cGAMP-loaded Ace-DEX MPs for prior studies employed only a single syringe and needle, limiting the throughput of a single device to ~5 mg MPs/hour, or ~100 mg MPs/day (Johnson et al., 2021; Junkins et al., 2018; Watkins-Schulz et al., 2019). While sufficient for production of material for preclinical evaluation, this production rate would present a major barrier to large-scale production of cGAMP MPs for clinical translation, which could require the manufacture of millions of doses for application as an influenza adjuvant (Plotkin et al., 2017). While larger scale commercial apparatuses are available, they generally align needles in a row, rather than in a circular arrangement, creating an unequal electric field environment between needles which can decrease spray stability and lead to increased polydispersity in the generated particle population. (Bocanegra et al., 2005) We sought to provide a proof of concept of the use of a circularly-arranged multiplex array to substantially increase the production rate of a single electrospray apparatus to ~50 mg MPs/hour or 1g/day.

Multiple groups have reported the use of multiplex arrays of electrospray capillaries to increase the throughput of the electrospray process,(Almería et al., 2011; Deng et al., 2006; Fu et al., 2014; Lhernould and Lambert, 2011; Parhizkar et al., 2017) but there are limited reports comparing the biological activity of particles manufactured through a single head versus multiplex spray apparatus.(Steipel et al., 2019) Here we report the production of cGAMP MPs using both single and multiplex electrospray apparatuses, characterization of the resulting MPs, and comparison of their activity as vaccine adjuvants in vitro and in combination with the COBRA HA antigen Y2 in vivo.

2. Experimental

2.1. Materials

70 kDa dextran, anhydrous dimethylsulfoxide, triethylamine, pyridinium p-toluenesulfonate, and sulfuric acid were purchased from MilliporeSigma (St. Louis, MO). Absolute ethanol was purchased from Decon Labs (King of Prussia, PA). 2-ethoxypropene was purchased from Matrix Scientific (Elgin, SC). cGAMP was purchased from Invivogen (San Diego, CA). Full length wild type (WT) influenza A (H1N1) HA protein amino acid sequences from 3,078 human H1N1 viruses collected from May 1, 2014 – September 30, 2016 were downloaded from the GISAID Epiflu database. These sequences were organized by their collection date, and put into the COBRA algorithm to generate the sequence of Y2 COBRA HA, which was produced and purified as previously described.(Huang et al., 2021)

2.2. Ace-DEX Synthesis and Characterization

Ace-DEX was synthesized from 70 kDa dextran from Leuconostoc mesenteroides as previously described.(Kauffman et al., 2012) The polymer’s relative cyclic acetal coverage was determined to be 42%, as measured by ¹H NMR spectroscopy (Inova 400 MHz spectrometer) following degradation in 10% v/v deuterium chloride in deuterium oxide and quantification of the evolved acetone and ethanol.(Broaders et al., 2009)

2.3. Single head electrospray of cGAMP MPs

Single head electrospray of cGAMP MPs was performed as previously reported.(Watkins-Schulz et al., 2019) Briefly, Ace-DEX was dissolved in absolute ethanol, and cGAMP was dissolved in molecular grade water. The two solutions were prepared and combined at a 9:1 v/v ratio of ethanol to water such that the resulting solution contained 20 mg/mL Ace-DEX and 0.2 mg/mL cGAMP. The resulting clear solution was briefly (~30 sec.) sonicated in a bath sonicator (Bransonic) before being loaded into a 5 mL glass gastight syringe (Hamilton) fitted with a 20-gauge stainless steel blunt needle. Prior to use, the syringe was filled with 1 M aqueous NaOH for > 3 hours to destroy any potential endotoxin. After filling, the syringe was fitted into a syringe pump (KD Scientific). A clean stainless steel plate (incubated >1 hour at 260 °C to destroy potential endotoxin contamination) was used as the collection substrate. MPs were generated by spraying in cone-jet mode using the setup depicted in Figure 1A and the spray parameters specified in Table 1.

Figure 1:

(A) Simplified schematic of single-head electrospray apparatus. (B) Photograph of 10-syringe pump fitted with tubing to the stainless steel needle array, with alligator clip attached to high voltage power supply. (C) Simplified schematic of multiplex electrospray apparatus. (D) Photograph of dried microparticles deposited in characteristic pattern on the stainless steel collection substrate.

Table 1:

cGAMP loading and encapsulation efficiency as determined by HPLC, zeta potential and hydrodynamic diameter of microparticles, and instrument parameters used to produce MPs by each apparatus.

	Particle Characteristics					Process Characteristics
Electrospray Method	cGAMP Loading (μg/mg)	Encapsulation Efficiency (%)	Zeta Potential (mV)	Hydrodynamic Diameter (nm)	Polydispersity Index	Substrate Voltage (kV)	Needle Voltage (kV)	Solution Total Solids Concentration (mg/mL)	Flow Rate per Needle (μL/hr)	Yield (%)
Multiplex	11.29 ± 0.91	112.9 ± 9.12	−31.7±0.8	801±63	0.291±.017	10	−14	20	200	42
Single-head	9.99 ± 1.07	99.9 ± 10.66	−33.1±1.8	916±172	0.216±.087	3	−5.8	20	200	78

2.4. Multiplex electrospray of cGAMP MPs

Solution preparation for multiplex electrospray was the same as for single-head electrospray. Spray solution was loaded into ten separate gastight syringes, which were loaded onto a ten-syringe syringe pump (KD Scientific). Syringes were connected by flexible PTFE tubing to 20-gauge needles mounted in a custom-manufactured stainless steel array (Ramé-hart Instrument Co.) (Figure 1B–C, Supplemental Figure 1). MPs were generated by spraying in cone-jet mode using the setup depicted in Figure 1B–D and the spray parameters specified in Table 1.

2.5. Scanning electron microscopy

Electrosprayed MPs were resuspended in Milli-Q grade water containing 0.04% v/v triethylamine at a concentration of ~10 mg/mL. The resulting suspension was deposited on SEM stubs with carbon tape (Electron Microscopy Sciences) and allowed to dry overnight. Dried particles were coated with 7–8 nm AuPd and imaged with a Hitachi S-4700 scanning electron microscope.

2.6. cGAMP loading determination by high performance liquid chromatography

To quantify cGAMP loading, MPs were dissolved in a 1:1 v/v mixture of 0.1% trifluoroacetic acid (TFA) in water and 0.1% TFA in acetonitrile. cGAMP loading was quantified by high performance liquid chromatography (HPLC, Agilent 1100 series, Santa Clara, CA) using a 0.1% TFA in water/0.1% TFA in acetonitrile gradient method through an Aquasil C18 column (150 mm length, 4.6 mm inner diameter, 5 μm pore size, Thermo Fisher, Waltham, MA) with a C8 guard column cartridge and a UV detection wavelength of 254 nm.

2.7. Dynamic light scattering

Hydrodynamic diameter of cGAMP MPs was determined using dynamic light scattering on a Brookhaven NanoBrook 90Plus Zeta Particle Size Analyzer (Holtsville, NY). Particle suspensions were prepared immediately prior to analysis at 0.1 mg/mL in phosphate-buffered saline with 10 min. bath sonication to ensure even dispersion of aggregates. Values are reported as the intensity-weighted average and standard deviation of 3 3-minute measurements.

2.8. Zeta potential

Zeta potential of the MPs was determined with a Brookhaven NanoBrook 90Plus Zeta Particle Size Analyzer (Holtsville, NY). Particle suspensions were prepared in 10 mM potassium chloride in water and bath sonicated for 1 minute before analysis. Values are reported as the average and standard deviation of 3 measurements.

2.9. Endotoxin detection

Electrosprayed microparticles were assessed for endotoxin contamination. MPs were suspended in endotoxin-free water overnight at a concentration of 1 mg/mL. Particles were centrifuged at 21,000 × G for 5 min. The supernatants were assessed for endotoxin content using a limulus amoebocyte lysate-based kit (Thermo Fisher A39552). All particle batches used for in vitro and in vivo evaluation had an endotoxin content of <0.1 EU/mL.

2.11. In vitro quantification of IFN-β secretion by bone marrow-derived dendritic cells

Bone marrow-derived dendritic cells were prepared from C57BL/6 mice and cultured in the presence of 10 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF) (Peprotech 315–03) and 10 ng/mL interleukin 4 (IL-4) (Biolegend 574306) using a modified version of a previously described method.(Xu et al., 2007) After 9 days of culture, BMDCs were transferred to 96-well plates in supplemented RPMI media (10% inactivated fetal bovine serum (FBS), 100 U/mL penicillin, 100 μg/mL streptomycin, 50 μM 2-mercaptoethanol, 1 × non-essential amino acids, 2 mM l-glutamine, 10 ng/mL GM-CSF, and 10 ng/mL IL-4) at 50,000 cells per well. 24 hours later, BMDCs were harvested for flow cytometry analysis or given cGAMP treatments.

For flow cytometry analysis, cells were lifted by aspirating media and replacing with 2.5 mM EDTA in PBS for 20 min. at 4 °C. Cells were transferred to a 96-well round bottom plate and stained with Fixable Viability Dye eFluor^™ 506 (Invitrogen), washed with PBS, treated with Fc-blocking antibody (Biolegend 101320) for 5 min. before staining with anti-CD11b-BV711 (Biolegend 101242), anti-CD11c-PE (Biolegend 117308), anti-B220 PerCP/Cy5.5 (Biolegend 103236), and anti-SIRPα-APC/Fire750 (Biolegend 144030). Cells were washed thoroughly with PBS and fixed with 1% PFA before analysis with an Attune NxT Flow Cytometer (Thermo Fisher).

For cGAMP treatment, BMDCs were treated with media alone, 1 μg/mL lipopolysaccharide (LPS), or 1 μg/mL cGAMP in soluble form, encapsulated in single-head or multiplexed electrosprayed Ace-DEX MPs, or complexed with Lipofectamine 2000 transfection reagent (Thermo Fisher) according to manufacturer’s instructions for DNA transfection. Supernatants were removed after 24 or 48 hours and assessed for IFN-β concentration by ELISA (R&D Systems DY8234–05).

2.12. Vaccination

All 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 C57BL/6J mice (n=10 per group) aged 6–8 weeks were obtained from Jackson Laboratory (Bar Harbor, ME). Mice were immunized with PBS, 1 μg Y2 COBRA, 1 μg Y2 COBRA mixed at a 1:1 volumetric ratio with Addavax (Invivogen), or 1 μg Y2 COBRA and 1 μg cGAMP soluble or in single-head or multiplex Ace-DEX MPs. All antigen and adjuvant combinations were mixed and incubated for 20 minutes on ice prior to administration. Vaccinations were performed as intramuscular injections of 25 μL in each rear leg for a total volume of 50 μL per mouse according to the schedule in Figure 4A. Serum was collected by submandibular bleed on the indicated days, and anti-COBRA HA antibody titers were assessed.

Figure 4:

(A) Timeline of immunizations and bleeds for in vivo evaluation of adjuvants in COBRA vaccine. Serum was collected on the indicated days and assayed for anti Y2 COBRA endpoint titers of total IgG (B), IgG2C (C), and IgG1 (D). Bars represent mean titers ± standard deviation. Statistical comparisons were performed between samples taken on indicated days by a one-way ANOVA using Tukey’s test for multiple comparisons. * = P≤0.05, **=P≤0.01, ***=P≤0.001,****=P≤0.0001.

2.13. Anti-Y2 antibody titer determination

Flat-bottomed high-binding polystyrene plates (Corning 29442–322) were coated overnight at 4°C with 1 μg/mL COBRA Y2 HA in PBS. Plates were washed three times with 0.05% Tween 20 in PBS (PBST), then blocked for two hours at room temperature with 200 μL blocking buffer (3% nonfat instant milk [Food Lion, NC, USA] in PBS). Plates were washed three times again. Serum samples were diluted in 100 μL blocking buffer and added to the blocked plates for one hour. Plates were washed three times again. The appropriate secondary antibodies (Goat Anti-Mouse IgG Fc-HRP 1033–05, Goat Anti-Mouse IgG2c-HRP 1078–05, or Goat Anti-Mouse IgG1-HRP 1071–05, Southern Biotech) were diluted in blocking buffer to the highest dilution recommended by the manufacturer, and 100 μL was added to each well for two hours at room temperature. Plates were washed five times with PBST and developed with tetramethylbenzidine (TMB) one component substrate (Southern Biotech 0410–01) before quenching with 2 N sulfuric acid. Development times were based on day 35 sera and all days of sera were developed for the same amount of time per each secondary antibody to allow comparison of titers between days. Plates were read for absorbance at 450 nm and corrected for background by subtracting absorbance at 570 nm. Antibody titers were determined by fitting a curve to the background-corrected absorbance vs. dilution using the “log(inhibitor) vs. response -- Variable slope (four parameters)” model in Graphpad Prism 8, then interpolating the dilution value at which the curve intercepts the endpoint value as defined by Frey et al. using a 99.9% confidence level and twelve background controls.(Frey et al., 1998)

3. Results and Discussion

3.1. Multiplexed electrospray increases throughput of physicochemically similar particles

To address the low throughput of our laboratory-scale electrospray setup (Figure 1A), we endeavored to create a system which could generate particles at a significantly increased rate without changing their physicochemical characteristics or in vitro or in vivo performance as an adjuvant. We designed a stainless steel array that allowed electrospray needles to be secured in a circular arrangement, with even spacing between each needle (Figure 1B, Supplemental Figure 1). This permitted voltage to be applied to every needle through attachment of a single high voltage source, due to the conductive nature of the stainless steel array. In addition, the symmetrical circular arrangement meant that each electrospray needle was subject to a similar electric field, and a consistent voltage could be used to maintain the cone-jet mode on each needle, in contrast to a linear or grid needle arrangement in which the electric field, and thus the voltage required to maintain a stable cone-jet, varies depending on the position within the array.(Bocanegra et al., 2005; Deng et al., 2006; Oh et al., 2008; Sen et al., 2007)

Another hurdle to overcome for a successfully multiplexed electrospray array was maintenance of consistent flow rate across all cone-jet emitters.(Kang et al., 2006) To solve this problem we employed a syringe pump capable of operating ten syringes at a time, allowing a separate syringe for each needle with each syringe having the same flow rate (Figure 1B). By using the same flow rate for each of the ten syringes as the single-head apparatus, we were able to use the multiplex array to spray at ten times the rate of the single-head apparatus.

While operating the multiplexed electrosprayer, we found that it was indeed possible to maintain a consistent cone-jet mode on all ten needles, leading to deposition of microparticles from all ten needles in distinct zones on the collection substrate (Figure 1D). Operating in the cone-jet mode required significantly greater applied voltage to the collection plate and needle array compared to the single-head spray setup (Table 1), but the required voltage differential did not scale linearly with the number of emitters.

A major difference between the two apparatuses was the overall mass yield from the electrospray process. In the single-head process, the spray solution is pumped directly from the pump-mounted syringe into the needle from which it is atomized. As a result, the dead volume remaining in the syringe after the pump has reached its furthest point is approximately equivalent to the volume of the needle: 60 μL. The multiplex apparatus dead volume was significantly larger: approximately 2 mL. The loss of this spray solution led to significantly reduced overall yield, but we anticipate that loss to dead volume would remain constant with increased batch size, meaning larger batches would likely have a greater fractional yield.

Scanning electron microscopy (SEM) revealed similar size and morphology between Ace-DEX cGAMP MPs made by single-head and multiplexed electrospray (Figure 2). Dynamic light scattering (DLS) and zeta potential measurements showed that particle hydrodynamic diameter and surface charge were also similar, and the cGAMP encapsulation efficiency was ~100% for MPs generated by both apparatuses (Supplementary Equation 1, Table 1). Greater than 100% encapsulation efficiency of cGAMP into Ace-DEX MPs by electrospray is consistently observed in our laboratory. It is possible that electrosprayed droplets containing cGAMP deposit preferentially on the charged substrate, while those containing only Ace-DEX may be more likely to deposit on other surfaces besides the collection substrate, but this phenomenon remains an area of active investigation.

Figure 2:

Scanning electron micrographs (SEM) of Ace-DEX particles produced by (A) single-head or (B) multiplex electrospray.

Previously, multiplex arrays have been reported for the generation of poly(lactic-co-glycolic acid) (PLGA).(Chen et al., 2019) 3D printing was used to generate a co-axial array of 25 spray heads; however, the generated PLGA particles were 17 microns or more in diameter,(Olvera-Trejo and Velásquez-García, 2016) close to the upper limit of what can be internalized by immune cells like macrophages.(Genito et al., 2021) Using a linear or circular four-needle array, Parhizkar et al. generated PLGA particles 650 nm in diameter, noting that both spray head geometries yielded similar sizes.(Parhizkar et al., 2017) Almería et al. reports PLGA particles generated by a five-needled array, wherein they used multiple metal rings in the aerosolized spray to stabilize it. The particles generated were approximately 600 nm to 7 μm in size.(Almería et al., 2011) Our 10 head multiplex system yielded Ace-DEX particles 800–900 nm in size, which is on the order of particle size observed with the four- or five-head multiplex systems.

3.2. Multiplex and single-head electrosprayed MPs have similar adjuvant activity in vitro and in vivo

Given the similar physicochemical characteristics between Ace-DEX cGAMP MPs produced by the two different electrospray apparatuses, we investigated their ability to induce interferon-beta (IFN-β) secretion in murine bone marrow-derived dendritic cells (BMDCs). cGAMP binding to STING has been shown to induce type-I interferon secretion, which leads to activation of antigen-presenting cells such as dendritic cells and macrophages, increased proinflammatory cytokine secretion and antigen presentation to T cells, as well as direct stimulation of B cells. For this reason we used IFN-β secretion as an indicator of effective STING-mediated adjuvanticity in past evaluations of cGAMP delivery platforms.(Junkins et al., 2018; Watkins-Schulz et al., 2019) Consistent with prior reports (Xu et al., 2007), flow cytometry analysis of BMDCs derived from murine bone marrow by culture with IL-4 and GM-CSF (Supplemental Figure 2) revealed a population of cells which were mostly CD11c+ and CD11b+ antigen-presenting cells lacking B220, a marker for plasmacytoid dendritic cells. The population was composed of a similar proportion of conventional dendritic cell (cDC) type 1 and type 2 cells, as indicated by the presence or absence of SIRPα-, respectively, and a mixture of mature and immature dendritic cells as indicated by MHC-II expression. Upon incubating BMDCs with media only or cGAMP at 1 μg/mL in either single-head MPs, multiplex MPs, or cGAMP complexed with Lipofectamine 2000, a commercial lipid used for DNA transfection, both MP formulations induced similar levels of IFN-β secretion after 24 and 48 h, to a much greater extent than soluble or lipofectamine-complexed cGAMP (Figure 3).

Figure 3:

Bone marrow derived dendritic cells (BMDCs) from C57BL/6 mice were treated with 1 μg/mL soluble cGAMP, or an equivalent dose of cGAMP delivered in Ace-DEX microparticles produced via single-head (SH) or multiplex (MH) electrospray, Lipofectamine 2000 transfection reagent, or 1 μg/mL LPS. Supernatants were collected 24 and 48 h later and assayed for secreted IFN-β by ELISA. n=3 wells per treatment group assayed by three technical replicates. Bars represent the mean ± standard deviation. Statistical comparisons were performed via unpaired t test.

Having demonstrated a similar degree of adjuvant activity between single-head and multiplex ES cGAMP MPs in vitro, we compared their efficacy as the adjuvant in a subunit vaccine incorporating the COBRA HA Y2. In addition to unadjuvanted antigen, we compared the adjuvant activity of ES cGAMP MPs to Addavax, a squalene-based oil-in-water (o/w) nanoemulsion with a composition analogous to FDA approved MF59 (Figure 4A). All three adjuvanted vaccination groups elicited significantly higher total anti-Y2 IgG titers than the unadjuvanted Y2 protein at all time points, and there was no significant difference in total anti-Y2 IgG titers between any of the adjuvanted groups at any time point, except a significant difference between multiplex ES cGAMP MPs and Addavax at day 28 which disappeared by day 35 (Figure 4B). The ES cGAMP MP-adjuvanted groups elicited significantly greater titers of anti-Y2 IgG2C antibodies than any other groups by days 28 and 35 (Figure 4C). This is indicative of a stronger Th1-skewed response, associated with the clearance of infected cells in the context of infection.(Bungener et al., 2008) In contrast, the Addavax-adjuvanted group elicited greater levels of IgG1 subtype titers than any other group, indicative of a Th2-skewed response, consistent with past reports (Figure 4D).(Valensi et al., 1994) Significant differences in anti-Y2 COBRA titer were not observed between single-head and multiplex ES MPs at any time point, nor for any antibody subtype tested.

These results align well with our previous studies using cGAMP MPs as an adjuvant with either HA or universal influenza adjuvant M2e.(Chen et al., 2018; Junkins et al., 2018) Independent of the antigen, vaccination with cGAMP MPs illustrated potent generation of humoral responses, on the order of what is reported here for single and multiplex particle generation. Previous studies where mice were vaccinated with Y2 virus-like particles (VLPs) and Addavax showed that a prime-boost-boost vaccination schedule protected fully against a lethal challenge.(Huang et al., 2021) Here we report antibody titers that are equal to or greater than Addavax for total IgG and IgG2c, illustrating the potential for protective responses.

4. Conclusion

We demonstrated a multiplex electrospray apparatus capable of producing cGAMP-loaded Ace-DEX microparticles with similar properties to those produced by a single-head electrospray apparatus, but at ten times the rate. We demonstrated equivalent adjuvanticity between cGAMP MPs produced by single-head and multiplex electrospray process in vitro and in vivo, as well as a greater Th1-skewed humoral response compared to the MF59-like Addavax. This proof of concept study paves the way for implementation of multiplex electrospray for production of cGAMP MPs as vaccine adjuvants. Future efforts will focus on improvement of the apparatus design to increase yield by reducing dead volume, and comparison of electrosprayed Ace-DEX cGAMP MPs to MPs made with other high-throughput manufacturing technologies to identify the best manufacturing process for clinical translation.