Supramolecular assembly of Toll-like receptor 7/8 agonist into multimeric water-soluble constructs enables superior immune stimulation in vitro and in vivo

Abstract

Resiquimod or R848 (RSQD) is a Toll-like receptor (TLR) 7/8 agonist which shows promise as vaccine adjuvant due to its potential to promote highly desirable cellular immunity. The development of this small molecule in the field to date has been largely impeded by its rapid in vivo clearance and lack of association with vaccine antigens. Here, we report a multimeric TLR 7/8 construct of nano-scale size, which results from a spontaneous self-assembly of RSQD with a water-soluble clinical-stage polymer - poly[di(carboxylatophenoxy)phosphazene] (PCPP). The formation of ionically paired construct (PCPP-R) and a ternary complex, which also includes Hepatitis C virus (HCV) antigen, has been demonstrated by dynamic lights scattering (DLS), turbidimetry, fluorescence spectroscopy, asymmetric flow field flow fractionation (AF4), and ¹H NMR spectroscopy methods. The resulting supramolecular assembly PCPP-R enabled superior immunostimulation in cellular assays (mouse macrophage reporter cell line) and displayed improved in vitro hemocompatibility (human erythrocytes). In vivo studies demonstrated that PCPP-R adjuvanted HCV formulation induced higher serum neutralization titers in BALB/c mice and shifted the response towards desirable cellular immunity, as evaluated by antibody isotype ratio (IgG2a/IgG1) and ex vivo analysis of cytokine secreting splenocytes (higher levels of interferon gamma (IFN-γ) single and tumor necrosis factor alpha (TNF-α)/IFN-γ double producing cells). The non-covalent multimerization approach stands in contrast to previously suggested RSQD delivery methods, which involve covalent conjugation or encapsulation, and offers a flexible methodology that can be potentially integrated with other parenterally administered drugs.

Graphical Abstract

INTRODUCTION

The development of next generation vaccines increasingly relies on formulating vaccine antigens with defined immunoadjuvants, which are capable of shaping adaptive B- and T-cell responses resulting in strong and long-term protective immunity.¹ The quality of the immune response to a potential pathogen is largely determined by engagement of specific pattern recognition receptors (PRR) and, in particular, Toll-like receptors (TLRs) expressed by cells at the infection site. Rapid advances in developing immunoadjuvants containing TLR ligands, such as monophosphoryl lipid (MPL) and glucopyranosyl lipid (GLA)^{2, 3} facilitated further research into discovery, optimization, and formulation of TLR4 agonists. TLR7/8 agonists also have the potential to fill unmet clinical needs and are increasingly reaching advanced development stages.⁴ In fact, TLR7/8 receptors, which recognize certain nucleic acid segments, have been unknowingly exploited for decades by vaccines containing RNA, such as inactivated polio and Japanese Encephalitis vaccines.⁴ It has been later shown that activation of these receptors leads to development of cytotoxic T-cell responses concomitantly blocking functions of immunosuppressive cells.^5–8

Imidazoquinolines, such as imiquimod and resiquimod (RSQD), which is also known as R-848, activate immune responses in TLR7/8 dependent mechanism and possess strong anti-viral and anti-tumor properties.^{9, 10} Imiquimod is approved in a topical formulation for treatment of genital warts, actinic keratosis, and superficial basal cell carcinoma.¹¹ RSQD induces superior cytokine secretion, macrophage activation and enhancement of cellular immunity as compared to imiquimod and has been advanced into clinical trials.^12–14 However, applications of these small molecules as vaccine adjuvants are hampered by unfavorable pharmacokinetic profile: short half-life due to a rapid glomerular filtration of the small molecule and rapid dissociation from the antigen upon injection.¹⁵ Various approaches to improve their pharmacokinetic-pharmacodynamic (PK/PD) profiles have been explored, which included liposomal,¹⁶ micellar,¹⁷ microparticulate,^{18, 19} and nanoparticulate^20–22 systems, formulations with Alum and N,N-dimethylformamide,^{13, 23} peptide conjugates,²⁴ dendrimers,²⁵ and polymer prodrugs.^{26, 27} However, all of them involve either chemical derivatization of RSQD or extensive formulation and production processes.

Biodegradable polyphosphazene polyelectrolytes, such as a clinical stage poly[di(carboxylatophenoxy)phosphazene], PCPP, are promising vaccine delivery systems, which also display some immunostimulating activity.^28–32 One of the unique features of PCPP,³³ which was originally synthesized as hydrogel forming macromolecule,³⁴ stems from its ability to spontaneously self-assemble with vaccine antigens into water-soluble supramolecular assemblies.³² High linear charge density and flexibility of polyphosphazene backbone appear to be primarily responsible for the stability of non-covalent complexes with the antigen, both in vitro and in vivo.³² We hypothesized that ionic interactions between negatively charged PCPP and the positively charged small molecule, RSQD, can potentially lead to a formation of supramolecular construct enabling multimeric presentation of this potent TLR-7/8 agonist. Furthermore, it was envisioned that the construct can provide a PCPP-mediated link between RSQD and vaccine antigen.

Multivalent ionic interactions between oppositely charged polyions, which result in water-soluble and nanoparticulate polyelectrolyte complexes (PECs), have been largely studied for the delivery of macromolecules: proteins, peptides, and nucleic acids (polyplexes).^{35, 36, 37} Although, there is clear evidence that hydrophobic counterions, such as surfactants, can strongly associate with polyelectrolytes, the potential of such assemblies was limited to solid and dispersed formulations, mainly for oral administration.^38–40 These limitations are possibly brought about by concerns of inherent instability of similar water-soluble systems in the presence of salts.^41–43 Although it is also proposed that stability of such constructs is strongly dependent on charge density, backbone flexibility, and hydrophobicity of the system,^44–47 water-soluble macromolecules bearing oppositely charged hydrophobic counterions, have not yet been investigated as drugs for parenteral administration.

Here we introduce a novel approach in which a potent TLR7/8 agonist drug (RSQD) self-assembles with a biodegradable polyphosphazene polyelectrolyte (PCPP) in the multimeric form (PCPP-R) through ion pairing in aqueous solution (Figure 1 and Scheme S1). The resulting noncovalent supramolecular construct is water-soluble, displays superior immune stimulating activity in vitro when compared to a monomeric form of RSQD, and shows appropriate stability at near physiological conditions. When combined with Hepatitis C virus E2 envelope protein vaccine antigen (E2), PCPP-R enables a ternary RSQD-PCPP-E2 complex, which demonstrates in vivo superiority to the components by inducing higher neutralization titers, promoting a balanced Th1/Th2 response, and generating Th1 memory cells.

Figure 1.

Schematic presentation of interactions of PCPP and RSQD with formation of multimeric supramolecular assembly PCPP-R.

MATERIALS AND METHODS

Materials.

Phosphate buffered saline pH 7.4, PBS (Life Technologies, Carlsbad, CA); RAW-Blue cells, mouse macrophage reporter cell line; zeocin, solution; Quanti-Blue reagent (Invivogen Inc., San Diego, CA); Dulbecco’s modified eagle medium, DMEM; penicillin streptomycin, Pen Strep (ThermoFisher Scientific); porcine red blood cells, RBCs (Innovative Technology Inc., Novi, MI); fetal bovine serum (Seradigm, VWR Life Sciences) were used as received. PCPP, 800,000 g/mol, was synthesized and characterized as described previously.^48–50

PCPP-R Formulations.

Aqueous solutions of PCPP and RSQD (Sigma Aldrich, St. Louis, MO) were mixed using microfluidics system with staggered herringbone mixer, SHM (Microfluidic ChipShop, Jena, Germany). SHM patterned topography design has been proven efficient in generating steady chaotic flows in microchannels due to subjecting volumes of fluid to a repeated sequence of rotational and extensional local flows.⁵¹ The device consisted of SHM, programmable syringe pump with two syringes containing feed solutions, and the collection flask (Scheme S2).

Physico-Chemical Characterization Methods.

¹H NMR spectra were obtained using 400 MHz Ascend™ Bruker NMR instrument (Bruker Biospin Corp, Billerica, MA). Malvern Zetasizer Nano series (Malvern Instruments Ltd., Worcestershire, UK) was employed for DLS measurements. Fluorescence spectra were recorded using BioTek Synergy Neo2 multi-mode reader (BioTek Instruments, Inc., Winooski, VT). Asymmetric flow field flow fractionation, AF4 was carried out as described previously using a Postnova AF2000 MT instrument (Postnova Analytics GmbH, Landsberg, Germany).

In vitro Complex Formation and Release Experiments.

AF4 analysis of the complexes was carried out using AF2000 MT (Postnova Analytics GmbH, Landsberg, Germany). Release experiments were conducted in Franz diffusion cells using 10-fold excess of PBS to maintain sink conditions (10 mL, PermeGear, Inc., Hellertown, PA) equipped with regenerated cellulose membrane (10 KDa cut off) in PBS, pH 7.4. The amount of RSQD released from the complex was determined by periodically taking aliquots from the receptor chamber and analyzing them with NanoDrop™ 2000 spectrophotometer (ThermoFisher Scientific, Grand Island, NY).

In vitro Immune Stimulation and Hemocompatibility.

RSQD, PCPP, and PCPP-R (PCPP:RSQD=20:1 (w/w)) solutions in PBS, pH 7.4 (20 μL) were added to 180 μL media containing 100,000 RAW Blue Cells (Invivogen Inc., San Diego, CA) per well. Dilutions were carried out in PBS (pH 7.4). Treatment plate was incubated at 37° C and 5% carbon dioxide for 20 h. Supernatant media (20 μL) was added to alkaline phosphatase substrate, Quanti-Blue solution (180 μL). The plate was incubated for 4 h and color development was monitored at 640 nm using SpectraMax M3 (Molecular Devices, San Jose, CA) plate reader. Treatments were performed in triplicates and statistical analysis was conducted using two-tailed Students t-test. The hemocompatibility of PCPP-R and RSQD was assessed using a modified hemolysis test.⁵²

HCV E2 Expression and Purification.

A gene encoding HCV E2 from strain 1a H77c (residues 384–661) was cloned into the vector pSecTag2 (Invitrogen) with an N-terminal immunoglobulin κ light chain signal sequence (for secretion) and a C-terminal His₆ tag (for purification). The construct was transfected with ExpiFectamine into Expi293 cells (ThermoFisher). Recombinant monomeric E2 was purified from culture supernatants by sequential HisTrap Ni²⁺-NTA and Superdex 200 columns (GE Healthcare).

HCV pseudoparticles (HCVpp).

HCV pseudoparticles (HCVpp) were generated by cotransfection of HEK293T cells with the MLV Gag-Pol packaging vector, luciferase reporter plasmid, and plasmid expressing HCV E1E2 as described previously.⁵³

BALB/c Immunizations and Spleen Harvest.

The recombinant HCV E2 protein was formulated with Alum, PCPP or PCPP-R prior to each vaccination time point. Alum-E2 formulation was prepared by mixing Alhydrogel (Invivogen, San Diego, CA) and E2 in a 1:1 ratio via vigorous vortexing. PCPP formulations were prepared by mixing aqueous solutions of PCPP and the antigen.⁵⁴ PCPP-R formulations were made using microfluidics technique. Groups of six female BALB/c mice (Charles River, Wilmington, MA), age 7 to 9 weeks, were immunized intraperitoneally with 50 μg sE2 (day 1) and received three booster vaccinations with 10 μg sE2 on days 14, 28, and 42. Blood samples were collected prior to each vaccination on days 1 (pre-bleed), 14, 28, and 42. Three of six vaccinated mice were terminally bled on day 56 for serology assays. On day 82 spleens of remaining three mice from the vaccinated groups and unvaccinated controls were harvested under aseptic conditions and collected in 5 mL of 1% heat-inactivated fetal bovine serum in 1x PBS (1% FACS buffer) and stored on-ice for subsequent processing. The blood samples were processed for serum by centrifugation and stored at −80°C until analysis was performed.

HCV pseudoparticles (HCVpp) Neutralization Assays.

Appropriate dilutions of mouse serum were mixed with HCVpp and incubated at 37°C for 1 h. HCVpp-serum mixture was then added to the Huh-7 cells in 96-well plates and incubated at 37°C for 5 h. After removing the inoculum, the cells were incubated for 72 h with DMEM containing 10% fetal bovine serum (ThermoFisher, Waltham, MA) and the luciferase activity was measured using Bright-Glo™ assay system (Promega, Madison, WI). Neutralizing antibody (nAb) titers in animal sera were reported as 50% inhibitory dilution (ID50) values. All data was analyzed by the nonlinear regression plots using GraphPad Prism 7 and the significance comparisons were calculated with Kruskal-Wallis one-way analysis of variance (ANOVA).

ELISAs for Serological Antibody Detection.

To detect sE2-specific total IgG, IgG1 and IgG2a responses in mouse serum, 96-well plates (MaxiSorp, Thermo Fisher, Waltham, MA) were coated overnight with 5 μg/mL Galanthus Nivalis Lectin (Vector Laboratories, Burlingame, CA) at 4 °C. Plates were washed (PBS + 0.05% Tween 20) and coated with 200 ng/well sE2 for 2 h at room temperature. After blocking with Pierce™ Protein-Free Blocking Buffer (Thermo Fisher, Waltham, MA) for 1 h, serially diluted sera from mice were added to the plate and incubated for 2 h. The binding of sE2-specific antibodies was detected by 1:5000 dilutions of HRP-conjugated goat anti-mouse secondary antibodies to detect total IgG (H+L), IgG1, and IgG2a (Southern Biotech, Birmingham, AL) with TMB substrate (Bio-Rad Laboratories, Hercules, CA). Absorbance values at 450 nm (SpectraMax MS microplate reader) was used to determine half-maximum effective concentration (EC50) (Prism 7, GraphPad software Inc.). Significance comparison was performed using Kruskal-Wallis one-way ANOVA.

Cell Isolation and in vitro Stimulation.

Spleens were trimmed of any residual adipose tissue and single cell suspensions of spleens were prepared by mechanical disruption and dispersion through 40 μm pore-size strainers using the plunger end of a 1 mL syringe. Red blood cells were lysed with ACK lysis buffer (Lonza), washed and filtered through fresh 40 μm pore-size strainers. Cells were then centrifuged and re-suspended in Dulbecco’s modified eagle medium, DMEM (Life Technologies) before seeding at 2×10⁶ cells per well into a 96-well plate. Cells were then incubated in the presence or absence of 10 ug/mL anti-CD3/CD28 (BD Bioscience) or 2 μg/mL of 10-sE2 peptide mix (BEI Resources NR-3749) at 37°C and 5% CO₂ for 6 hours in the presence of Brefeldin A. Cells were then surface stained and fixed/permeabilized overnight and intracellular cytokine staining was performed the next day. All samples were acquired on a MACS Quant Analyzer 10 and analyzed with Flowjo version 10 software.

Flow Cytometry.

Antibodies directed against mouse extracellular and intracellular antigens were purchased from BD Biosciences and eBioscience/Thermofisher. Antibodies were directed against CD4 (clone RM4–5), CD8 (53–6.7), CD44 (IM7), TNFα (MP6-XT22), IFNγ (XMG1.2), IL-4 (11B11), IL-17 (ebio17B7). Fixable violet live-dead stain 405 nm (Invitrogen/ThermoFisher) was used prior to permeabilization and intracellular stain to exclude dead cells. All samples were acquired on a MACS Quant Analyzer 10 and analyzed with Flowjo version 10 software. The 10-peptide mix selected for splenocyte stimulation were selected from the HCV E2 peptide array obtained from BEI Resources (NR-3749). The individual peptides in the 10-E2 peptide mix include Class II_MHC_H2-lad peptides #2 (7-GNAGRTTAGLVGLLTPGA-24), #8 (46-CNESLNTGWLAGLFYQHK-63, a known CD4 epitope), #17 (106-YPPRPCGIVPAKSVCGPV-123) and #18 (113-IVPAKSVCGPVYCFTPSPV-131) and Class I_MHC_H2-lad peptides and #22(138-RSGAPTYSWGANDTDVFV-155), #24 (150-DTDVFVLNNTRPPLGNWF-167), #36 (223-RCMVDYPYRLWHYPCTI-239), and #37 (229-PYRLWHYPCTINYTIFKV-246) and as well as a CD8 epitope represented by peptide #34 (210-TYSRCGSGPWITPRCMV-226).

RESULTS AND DISCUSSION

RSQD Spontaneously Assembles with Biodegradable PCPP into Water-Soluble Multimeric Constructs, PCPP-R, at near Physiological Conditions.

The concept of noncovalent assembly of a positively charged small molecule drug on a negatively charged water-soluble macromolecular scaffold, which can potentially result in a pharmaceutically viable multimeric drug, stemmed from well-established complexation phenomenon occurring between polyelectrolytes and ionic surfactants or colloids.^{45, 55} Although electrostatic attraction between polyelectrolyte and a counterion cannot be compared in strength with covalent bonding⁴⁵ usually exploited in prodrugs, there was an expectation that the concomitant hydrophobic stabilization of such interactions and high charge density of PCPP may be sufficient to significantly reduce the release rate of counterions from such metastable construct under physiological conditions.

We first explored the ability of water-soluble PCPP to interact with RSQD in aqueous solutions using turbidimetric titration, DLS, and fluorescent spectroscopy methods. The experiment, in which RSQD was gradually added to PCPP, was conducted in PBS (pH 7.4) to mimic the ionic strength of physiological conditions (155 mM). The results of turbidimetric titration indicate counterion induced phase separation in the system at approximately 1:2 RSQD to carboxylic acid (mol/equiv) ratio, which occurred despite a relatively high salt content in the solution (Figure 2A). The formation of an insoluble complex in only partially neutralized polyelectrolyte emphasizes the importance of hydrophobic interactions in RSQD-PCPP system.⁵⁵ Analysis of the system by DLS detected an increase in the z-average hydrodynamic diameter at even earlier degree of neutralization, 1:4 = RSQD:COO⁻ (mol/equiv) (Figure 2B), indicating formation of water-soluble multi-chain constructs.⁵⁵ Fluorescent analysis or PCPP-R848 formulations showed dose-dependent quenching of intrinsic RSQD fluorescence independently confirming binding of the drug to PCPP (Figures 2C and 2D).^{17, 24}

Figure 2.

Physico-chemical characterization of PCPP-RSQD complexes: (A) Turbidimetric titration of PCPP with RSQD (0.5 mg/mL PCPP, 0–1 mg/mL RSQD, PBS, pH 7.4, SD); (B) z-average hydrodynamic diameter of PCPP – RSQD system (0.5 mg/mL PCPP, 0–1 mg/mL RSQD, PBS, pH 7.4, SD); (C) fluorescence spectroscopy spectra of RSQD in the presence of PCPP and (D) maximum fluorescence peak intensity of RSQD vs. PCPP concentration (Ex/Em=250/350 nm; 0.025 mg/mL RSQD, 0–0.5 mg/mL PCPP, PBS, pH 7.4); (E) PCPP – RSQD side group with ¹H NMR peak shifts assigned (RSQD:COO⁻(PCPP)=1:1 (mol/equiv), 1M KOH in D₂O); (F) AF4 fractograms of PCPP, RSQD (no signal) and PCPP-RSQD (0.25 mg/mL PCPP; 0.125 mg/mL RSQD, PBS, pH 7.4).

Formulations of PCPP with RSQD, both in the precipitate and soluble forms, were then analyzed for the presence of RSQD by NMR and AF4 methods. The molecular composition of PCPP-R precipitate, which was prepared at 1:1 RSQD-to-carboxylic acid mole/equivalent ratio, isolated by filtration, and redissolved in the basic solution for the analysis, was studied by ¹H NMR. Figure 2E shows ¹H NMR peak shifts in the spectrum and their assignments. Integration of ¹H NMR peaks confirmed complete stoichiometric binding of RSQD by PCPP under these conditions (Figure 2E and Figure S1). Water-soluble formulations of PCPP, which were formed at a low concentration of RSQD, were analyzed by AF4 method. AF4 is an analytical technique capable of separating supramolecular assemblies by their size both in the nanometer and micrometer ranges and is designed to minimize interactions of the analyte with the stationary phase.⁵⁶ The eluting molecules are separated in a cross-flow of the mobile phase, which presses them against the semi-permeable membrane.⁵⁶ Since the membrane is selected to be impermeable only to macromolecules, RSQD escapes through it during the analysis and never reaches the detector, remaining “invisible” to the method (Figure 2F). Comparison of AF4 fractograms of PCPP and PCPP – RSQD formulations reveals a significantly larger peak area in case of the mixture (Figure 2F). Since RSQD cannot be detected by AF4 under these conditions, the increase in the signal can only be attributed to RSQD retained by PCPP through ion pairing with such construct remaining stable in the flux of sodium ions generated by the cross-flow of PBS. The above results unambiguously prove the formation of supramolecular assemblies, which are denoted by “PCPP-R” (Figure 1).

PCPP-R Maintains Multimeric Form During in vitro Release Experiments and Shows Reduced Hemolytic Activity Compared to RSQD.

Next we investigated biologically relevant properties of PCPP-R in vitro, such as the ability to retain RSQD counterions at near physiological conditions over time and hemocompatibility. Stability of PCPP-R was analyzed by measuring the release of RSQD in PBS (pH 7.4) under sink conditions using Franz diffusion cell. Figure 3A shows that despite gradual decrease of the number of RSQD counterions associated with PCPP scaffold, which was observed during the first 24 h of incubation, the system retained approximately 250 RSQD cations per PCPP chain and this load remained unchanged for the following three days. These results demonstrate the ability of PCPP-R system to effectively maintain multimeric form of RSQD (in excess of two hundred copies) at physiological salt concentrations for at least several days.

Figure 3.

In vitro assessment of PCPP-R: (A) Molecular composition of PCPP-R (number of RSQD counterions per PCPP chain) vs incubation time (Franz diffusion cell, 0.125 mg/mL RSQD, 0.25 mg/mL PCPP, PBS, pH 7.4, SD); (B) hemolytic activity of RSQD and PCPP-R (human erythrocytes, 0.5 mg/mL PCPP, no activity detected for PCPP); (C) In vitro stimulation of macrophages by PCPP-R and RSQD (0.01 – 1 μg/mL RSQD, PCPP-R (PCPP:RSQD=20:1 (w/w)), RAW Blue cell line; two-tailed Students t-test; R² > 0.96, n=3; no response detected for PCPP); (D) AF4 fractograms of PCPP-R – E2, E2, PCPP, PCPP-E2, and PCPP-R (0.125 mg/mL E2, 0.125 mg/mL PCPP, 0.062 mg/mL RSQD, PBS, pH 7.4).

Hemocompatibility of PCPP-R was evaluated in vitro by hemolysis assay using human erythrocytes. Figure 3B reveals that although undesirable hemolytic activity of cationic RSQD is relatively low, it still shows some dose-dependent cytotoxic effect on cells. As seen in the figure, the addition of PCPP to RSQD leads to an improvement in hemocompatibility demonstrating yet another desirable effect of ion pairing of this compound to PCPP.

PCPP-R Displays Superior Immune Stimulating Activity Compared to RSQD in Cellular Assays.

PCPP-R and RSQD were compared for their ability to activate engineered RAW-Blue macrophages. These mouse-derived immune cells contain TLRs and integrate Secreted Alkaline Phosphatase (SEAP) reporter inducible by NFkB to allow rapid screening of innate immune activation. Figure 3C compares dose-dependent stimulation of macrophages by monomeric RSQD and PCPP-R multimer. As seen in the figure, the immune stimulation of macrophages by multimeric PCPP-R is significantly higher than RSQD, as compared by two-tailed Students t-test. The superior activity of PCPP-R compared to the monomeric form of RSQD may be potentially explained by multimeric presentation of TLR-7/8 agonists to macrophages.^{25, 57} Since TLR-7/8 are located in the endosomal compartment of cells, it is also possible that PCPP can facilitate uptake of its cargo into endosomes, similar to the previously reported ability of its copolymers,⁵⁸ thereby increasing availability of the agonist to receptors.

PCPP Links RSQD with Model Vaccine Antigen Through Spontaneous Self-Assembly in Aqueous Solutions.

As mentioned above, one of the main challenges faced in the development of RSQD as vaccine adjuvant is that the compound quickly dissociates from the antigen once the formulation is delivered in vivo.¹⁵ Polyphosphazene adjuvants are known to spontaneously self-assemble with protein antigens thereby acting as vaccine delivery vehicles.^{32, 54, 59} Therefore, we investigated the ability of PCPP to act as an assembly scaffold for a concomitant delivery of RSQD and vaccine antigen. In this study we used hepatitis C virus (HCV) envelope glycoprotein E2 as the model antigen to evaluate PCPP-R and RSQD for their immunostimulatory potential.⁶⁰ This antigen has been already shown to non-covalently bind to PCPP.³² Moreover, TLR-7/8 dependent activation may be particularly beneficial for inducing immune responses to infections caused by RNA viruses, such as HCV.^{4, 61}

The analysis of PCPP-R and E2 protein formulation was conducted in PBS (pH 7.4) using AF4 method and DLS (Figures 3D and S2). Fractograms of E2, PCPP, PCPP-R and PCPP-E2 are shown for comparative purposes. As seen in the figure, ternary PCPP-R – E2 formulation displays (a) complete disappearance of peak characteristic for E2 protein (at approximately 9 min) and (b) dramatic increase in the macromolecular peak area (approximately 18 min) compared to PCPP-R formulation. This unambiguously proves efficient binding of E2 protein to PCPP-R. Furthermore, PCPP-R - E2 fractogram shows larger peak area than that for binary PCPP – E2 formulation, which demonstrates retention of RSQD in the ternary PCPP-R – E2 system.

PCPP-R Shows In Vivo Synergy Between PCPP and RSQD in Inducing Higher Neutralization Titers Against HCV E2 Antigen, Promoting Balanced Th1/Th2 Immunity, and Generating Antigen Specific Memory Response.

In vivo evaluation of PCPP-R was performed in BALB/c mice immunized with soluble E2 wild-type protein (sE2-WT) and its formulations with PCPP-R, PCPP, RSQD, and Alhydrogel (Alum). Immunization schedule included prime and two boosts. Figure 4A shows that serum neutralization titers against the homologous HCV isolate (H77) were significantly higher for animals immunized with PCPP-R adjuvanted E2 as compared with titers induced with Alum (2.5-fold), PCPP (2.7-fold), and RSQD (65-fold). Similar results were obtained for total antibody titers in which IgG titers induced by PCPP-R formulated E2 were approximately 10-fold higher than those observed for RSQD although there was not a significant difference than those observed for Alum and PCPP (Figure 4B).

Figure 4.

Immune responses in vivo. Antibody responses induced by E2 formulations: (A) serum neutralization ID₅₀ values against homologous HCV isolate (H77); (B) serum IgG; (C) IgG1; (D) IgG2a EC₅₀ values and (E) IgG2a/IgG1 ratio (BALB/c mice, day 42 post immunization, individual values and the geometric mean for the distribution); (F) FACS plots of cell mediated responses to E2 peptides as assessed ex-vivo in splenocytes derived from immunized BALB/c mice (TNF-α single (Y axis), IFN-γ single (X axis) and TNF-α/IFN-γ double producing cells (top right quadrant). Positive cell frequencies are shown in quadrants. Significance comparisons were calculated with Kruskal-Wallis one-way analysis of variance (ANOVA), * p≤0.05, ** p≤0.01, *** p≤0.001.

Directed alteration of the immune response is another desirable objective of vaccine adjuvants. Effective vaccines against pathogens require both humoral and T cell responses to engender protective antibody and/or cellular immunity, along with robust long-term memory responses.⁶² To that end, achieving functionally appropriate type of the immunity, such as T helper 1 (Th1) mediated immunity, is highly desirable. Since IgG2a and IgG1 isotypes are commonly used as surrogate markers of Th1 and Th2 responses in mice, respectively,^63–65 these subclasses were also monitored (Figure 4C and 4D). PCPP-R formulations showed similar levels of IgG1 titers to those induced by PCPP and Alum (Figure 4C) as both of these adjuvants are known to be potent antibody producers.^{64, 66, 67} However IgG2a titers (associated with Th1 immunity) induced by antigen adjuvanted with PCPP-R was significantly higher than Alum (8.0-fold) and PCPP (8.3-fold), and about 4.3-fold higher than RSQD (Figure 4D). A significant reduction in performance of RSQD in vivo is probably associated with previously mentioned rapid clearance and dissociation of this TLR agonist upon systemic administration.¹⁵ Isotope profiles (IgG2a/IgG1 ratios), which are shown in Figure 4E, suggest that in contrast to “unmodified” PCPP or Alum, PCPP-R is capable of shifting the immune response towards Th1 immunity. The IgG2a/IgG1 ratios for animal groups immunized with antigen alone or antigen adjuvanted with RSQD were lower and are not shown due to the very low levels of IgG isotopes observed for these groups.

The ability to increase generation of memory T and B cells (antigen specific cells that are long-lived and can be rapidly converted into effector cells upon infection) is a highly desirable feature of optimal vaccine adjuvants.^{68, 69} Antiviral CD4 T cells play a vital role in the maintenance of functional CD8 responses, as well as formation of memory B cells.⁷⁰ Their function is typically evaluated by monitoring production of cytokines, such as gamma interferon (IFN-γ) and tumor necrosis factor alpha (TNF-α), with multiple-cytokine producing cells suggested to be superior.⁷⁰ The cell-mediated memory recall response with E2 was evaluated by ex vivo re-stimulation of splenocytes with E2 peptides to identify antigen-specific CD4 T cells. Splenocytes were stained for extracellular surface markers and intracellular cytokines to identify antigen-specific effector memory CD4 T cells by fluorescent activated cell sorting (FACS). Density plots of CD4 T cell populations displaying the frequency of TNF-α and IFN-γ single producing (SP) and TNF-α/IFN-γ double producing (DP) cells in unstimulated and E2 peptide stimulated conditions are shown in Figures 4F and S3. As shown in the figure, PCPP-R adjuvanted formulation displayed a higher frequency of IFN-γ single producers and IFN-γ/TNF-α double producers as compared to all other formulations including those adjuvanted by PCPP and RSQD. These results suggest generation of antigen-specific memory response in mice immunized with PCPP-R adjuvanted antigen. It also correlates with data on IgG isotope profiles confirming that a new adjuvant favors Th1 responses. Further studies are warrantied, which can include promising next generation polyphosphazene adjuvant, PCEP, as it showed greater potency and higher quality immune responses than PCPP in previous studies.^{31, 71, 72} Overall, the dramatically different performances observed for PCPP-R and its components in vivo unambiguously proves the viability of the molecular construct and ion paired polyelectrolyte-drug approach for parenteral applications.

CONCLUSIONS

Solution interactions between RSQD, a small molecule cationic TLR-7/8 agonist, and negatively charged polyphosphazene polyelectrolyte, PCPP, a vaccine delivery vehicle, were studied by DLS, turbidimetry, AF4, 1H NMR, and fluorescence spectroscopy and revealed formation of intermolecular complex, PCPP-R. We attempted to exploit a chemical synergy between these molecules with a goal of addressing some of the major challenges faced by RSQD in immunoadjuvant applications, i.e., high in vivo clearance rate and lack of association with vaccine antigen. With this objective, RSQD was converted into a multimeric form by supramolecular assembly with PCPP and its in vitro stability was assessed at near physiological conditions (PBS, pH 7.4). Despite a relative weakness of electrostatic interactions, the supramolecular construct of PCPP with RSQD was able to maintain at least 250 molecules of RSQD per polyphosphazene chain in the release experiments, emphasizing the importance of hydrophobic interactions in such systems. Moreover, PCPP-R displayed improved hemocompatibility and higher immune stimulating potency in cellular assays. The latter was possibly attributed to either multimeric presentation of TLR agonist in PCPP-R coupled with the improved delivery of antigen to endosomes that can be facilitated by PCPP carrier. The PCPP-R construct is amenable to co-formulation with antigens in a ternary complex as shown by using HCV E2 envelope glycoprotein as a model antigen combined with PCPP-R, thereby linking three components PCPP, RSQD, and E2 protein into a single supramolecular assembly as confirmed by AF4 analysis. Finally, in vivo performance of PCPP-R was evaluated in E2 immunization experiments in mice. In contrast to its molecular subunits, PCPP and RSQD, PCPP-R demonstrated higher HCV neutralization titers and a balanced Th1/Th2 response, and has the potential to engender antigen-specific memory responses. The above results clearly demonstrate viability of the approach, which is based on water-soluble constructs of a charged small molecule drug with a biodegradable polyelectrolyte, for pharmaceutical applications requiring parenteral administration. The proposed methodology avoids the need for expensive and laborious step of covalently linking the drug to a pharmaceutical carrier (prodrug) or encapsulation in particulate systems and can be potentially applied to a large number of pharmaceutical agents.

ABBREVIATIONS

RSQD or R848

resiquimod

TLR

Toll-like receptor

PCPP

poly[di(carboxylatophenoxy)phosphazene

PCPP-R

complex of PCPP and RSQD

HCV

Hepatitis C virus

HCVpp

HCV pseudoparticles

AF4

asymmetric flow field flow fractionation

DLS

dynamic lights scattering

NMR

nuclear magnetic resonance

IgG

immunoglobulin G

IFN-γ

interferon gamma

TNF-α

tumor necrosis factor alpha

IL

Interleukin

CD4, CD8, CD44

cell-surface glycoproteins 4, 8 and 44

PRR

pattern recognition receptors

GLA

glucopyranosyl lipid

PK/PD

pharmacokinetic-pharmacodynamic

PEC

polyelectrolyte complex

E2

envelope protein vaccine antigen

RAW-Blue cells

mouse macrophage reporter cells

PBS

phosphate buffered saline

ID50

50% inhibitory dilution

ANOVA

analysis of variance

ELISA

enzyme-linked immunosorbent assay

DMEM

Dulbecco’s modified eagle medium

FACS

fluorescence-activated cell sorting

SEAP

Secreted Alkaline Phosphatase

NF-kB

nuclear factor kappa-light-chain-enhancer of activated B cells

Th1 and Th2

T helper 1 and 2

SP

single producing

DP

double producing