pH-responsive polymer–antigen vaccine bioconjugates

Abstract

Protein-based vaccines play an important role in controlling infectious disease but their full clinical impact has been limited by their inability to generate a coordinated cellular CD4+ and CD8+ immune response. Vaccines that better deliver antigens to the cytosolic MHC1 display system could in principle provide a better coordinated response. Here, controlled radical polymerization was employed to prepare a diblock copolymer containing an endosomal releasing segment based on poly(propylacrylic acid) (poly(PAA)) and a hydrophilic segment containing thiol-reactive disulfide moieties for antigen conjugation. Propylacrylic acid (PAA) was polymerized in the presence of a trithiocarbonate based RAFT chain transfer agent (CTA). The resultant poly(PAA) was then employed as a macroCTA in the copolymerization of pyridyl disulfide methacrylamide (PDSMA) (thiol-reactive monomer) and N,N-dimethylacrylamide (DMA) as a hydrophilic comonomer. Copolymer compositions and molecular weights were determined via¹H NMR spectroscopy and size exclusion chromatography. Native polyacrylamide gel electrophoresis (PAGE) showed a complete disappearance of the bands corresponding to free thiolated ovalbumin after conjugation to the polymer at pyridyl disulfide to thiol ratios as low as 2.5. The ability of the poly[(PAA)-b-(DMA)_co(PDSMA)]–ovalbumin conjugates to activate CTLs was evaluated in vivo, tumor protection using the EG7 tumor protection model. Tumors were visible in the PBS and free ovalbumin immunized mice by day 7 but were not visible in the polyPAA–ovalbumin conjugate immunized mice until day 18. The mice immunized with the polyPAA–ovalbumin conjugate had a survival rate at day 21 of 100% versus 20% for PBS and 40% for ovalbumin immunized mice.

---

Introduction

For protein-based vaccines, class I presentation of cytosolic antigens remains a considerable intracellular delivery challenge. Class I antigen presentation leads to the generation of CD8+ cytotoxic T-lymphocytes (CTLs) which promote immunorecognition and direct apoptotic induction in cancer cells.¹ To stimulate CD8 + T-cell responses, a number of strategies have been explored to enhance cytosolic delivery of vaccine antigens, including the use of viral vectors and bacterial toxins, however these approaches have raised safety concerns.² Synthetic carriers for protein antigens have been explored as a safer alternative but have generally been less effective. Liposomal systems,³ poly(D,L-lactic-co-glycolic acid) (PLGA) particulate carriers,⁴ as well as additional particulate carriers⁵ have all been investigated for antigen delivery. Synthetic carriers are a desirable alternative because they avoid safety concerns of viral vectors, can be produced on larger scales, and can be readily synthesized with well-defined chemical modifications and/or architectures. These studies have largely exploited particle-based delivery systems that may display more efficient uptake via phagocytic pathways and may also provide potential adjuvancy effects of the synthetic materials.⁶

However, little attention has been paid to other strategies for directing intracellular trafficking of antigens within antigen presenting cells (APCs) to achieve cytosolic delivery and enhanced major histocompatibility complex class 1 (MHC-1) antigen presentation. Previous approaches have employed pH-responsive elements such as nanoparticle carriers based on poly(γ-glutamic acid)–poly(L-phenylalanine ethyl ester),⁷ and acid labile particles and microgels that degrade at endosomal pH.⁸ Until recently, polymer–antigen molecular conjugates employing pH-responsive polymers designed to achieve endosomal escape had not been explored. Our group has extensively investigated poly(propylacrylic acid) (PPAA) for intracellular delivery biomacromolecules such as proteins, peptides, and siRNA.⁹PPAA contains carboxylic acid groups that become more protonated in the lower pH environment within endosomes, shifting the polymer to a membrane destabilizing state that promotes cytoplasmic delivery. We recently explored PPAA as a protein vaccine carrier by evaluating its efficacy in an in vivo mouse tumor protection model.¹⁰ Both the soluble conjugate and particulate PPAA-based formulations were tested, and both carriers were able to mediate CTL activation, antibody production, and significant tumor protection in vivo. These data shed new light on a potential strategy for controlling intracellular trafficking within APCs that may allow for the production of effective, yet more easily constructed bioconjugate vaccines.

This study focuses on the design of a second generation polyPAA based antigen carrier that is based on a diblock design. The first generation carriers had relatively higher polydispersities and were based on a graft design where pyridyl disulfide (PDS) functionalized monomers were randomly dispersed throughout the polymer. This design potentially led to cross-linking and further carrier heterogeneity upon protein conjugation. Furthermore, an acrylate version of the PDS monomer was used for these polymer carriers, which is innately less stable. Here, we describe RAFT polymerization of diblock copolymers containing a prominent pH-responsive polyPAA 1^st block with a short 2^nd block containing a more stable pyridyl disulfide methacrylamide (PDSMA) monomer. This polymer design allows the endosomolytic component to freely interact with membranes. In addition, the second short block contains a small number of pyridyl disulfide groups that could reduce cross-linking. The synthesis and characterization of this new pH-responsive diblock copolymer are described along with its initial in vivo characterization using a murine ovalbumin (Ova) expressing tumor protection model. This model allows the characterization of prophylactic vaccine tumor protection, and additionally, it allows the quantitation of Ova-specific CD8 induction to test whether the endosomal-releasing polymers enhance this presentation pathway.

Results and discussion

Synthesis and characterization of poly[(PAA)-b-(DMA)_co(PDSMA)]

The successful generation of a potent and coordinated immune response remains a major hurdle in cancer immunotherapy.² Previously we have shown poly(PAA) to be highly effective as an endosomal-releasing agent. Here we prepared diblock copolymers containing a poly(PAA) block as the endosomal releasing agent and a second hydrophilic block for covalent conjugation of the antigen. As shown in Scheme 1a, the diblock copolymer was synthesized via RAFT using a trithiocarbonate based RAFT agent. The polymerization of sterically bulky α-alkyl substituted acrylic acid monomers is generally difficult and these reactions often exhibit slow polymerization rates and low monomer conversion.¹¹ For this reason the RAFT polymerization of poly(PAA) was conducted under bulk monomer conditions with a low initial CTA to initiator ratio ([CTA]₀/[I]₀ = 1).

Scheme 1 (a) Synthetic pathway for the preparation of poly[(PAA)-b-(DMA)_co(PDSMA)] viablock copolymerization of DMA and PDSMA from a poly(PAA) macroCTA. (b) Conjugation of thiol-functionalized ovalbumin to the diblock copolymerviathiol exchange with pyridyl disulfide moieties in the 2^nd block.

In order to maximize antigen conjugation to the polymeric scaffold while minimizing disruption of the intrinsic membrane disruptive properties of poly(PAA), the thiol-reactive pyridyl disulfide moieties were incorporated into a second spatially discrete block. This block was composed primarily of N,N-dimethylacrylamide (DMA) that was selected for its low steric bulk and high hydrophilicity. The poly(PAA) macroCTA was chain extended with DMA and PDSMA in DMSO at 70 °C to prepare the poly[(PAA)-b-(DMA)_co(PDSMA)] diblock copolymer. Molecular weights, comonomer compositions, and polydispersities for the macroCTA and the corresponding diblock copolymer are shown in Table 1. Absolute molecular weights for the poly(PAA) macroCTA and poly[(PAA)-b-(DMA)_co(PDSMA)] were determined to be 13.1 and 2.8 kDa respectively. The molecular weight of the second block was designed to be small (M_n < 5000 g mol⁻¹) in order to limit the number of protein antigens that may be conjugated per polymer. The PDIs for both materials were observed to be approximately 1.6, which while somewhat high is consistent with the RAFT polymerization of PAA and other sterically hindered monomers. ¹H NMR analysis of the diblock copolymer (Fig. 1) shows resonances at 7.21, 7.73, and 8.44 ppm corresponding to aromatic pyridyl disulfide residues. This suggests that the pyridyl disulfide groups, necessary for antigen conjugation, were not degraded during the course of the polymerization. By comparison of the resonances associated with the poly(PAA) carboxyl proton (11.5–12.5 ppm) to those associated with the pyridyl disulfide groups (5.8–6.4 ppm), it was possible to determine the ratio of PAA to PDSMA in the diblock copolymer (19 : 1). The methyl protons associated with dimethylacrylamide (2.8–3.2 ppm) slightly overlapped with the water peak (3.2–3.4 ppm) making it difficult to directly determine DMA incorporation via¹H NMR. As a result, the molecular weight of the second block as determined by SEC was used to back-calculate the percentage of dimethylacrylamide by using the polyPAA : PDSMA ratio, subtracting out the PDSMA content from the second block and calculating the DMA content. These calculations suggest that the PDSMA content in the diblock copolymer is 17 mol% which is in good agreement with the feed (20 mol%). TCEP reduction and subsequent UV analysis of the released pyridine 2-thione were also employed to provide a second method for determining the PDSMA content. These experiments indicate the presence of approximately 2.4 PDSMA residues per polymer which is consistent with the 1.9 PDSMA residues calculated via a combination of SEC and ¹H NMR.

Table 1 Molecular weights, polydispersities, and monomer compositions for the polyPAA macroCTA and the resultant diblock copolymer

Polymer	M n 1^st block^a/g mol^−1	M n 2^nd block^a/g mol^−1	PDI	Mol% DMA (feed)	Mol% PDSMA (feed)	Mol% DMA (exp.)^b	Mol% PDSMA (exp.)^b
a Absolute molecular weights determined by SEC Tosoh TSK-GEL columns (TSK-α3000, α3000, α4000) (Tosoh Bioscience, Montgomeryville, PA) connected in series to a MINIDAWN TREOS Multi-angle Light Scattering detector and an Optilab rEX RI detector (Wyatt, Santa Barbara, CA). HPLC-grade DMF containing 0.1 wt% LiBr was used as the mobile phase. b As determined by ^1H NMR spectroscopy (3 wt% DMSO-d6, Bruker AV300).
MacroCTA	13 100	—	1.69	—	—	—	—
Diblock	—	2800	1.63	80	20	83	17

---

Fig. 1 ¹H NMR of poly[(PAA)-b-(DMA)_co(PDSMA)] in DMSO-d₆ showing resonances associated with PAA carboxyl protons (a) as well as resonances associated with DMA residues (b) and PDSMA residues (c–e).

Conjugation of poly[(PAA)-b-(DMA)_co(PDSMA)] to ovalbumin

Conjugation of ovalbumin to the polymeric scaffold was achieved by first introducing the thiol functionality onto the protein. Lysine residues were reacted with a 15 molar excess of 2-thiolanimine hydrochloride “Traut's reagent”. The thiol-functionalized ovalbumin was then reacted with poly[(PAA)-b-(DMA)_co(PDSMA)] at a pyridyl disulfide to thiol molar ratio of 2.5 and 3.5. After allowing poly[(PAA)-b-(DMA)_co(PDSMA)] and thiol functionalized ovalbumin to react for 3 hours, the release of pyridine 2-thione was measured spectrophotometrically. These measurements suggest that both the 2.5 and 3.5 molar excess pyridyl disulfides were sufficient to quantitatively react with the thiolated protein.

Native polyacrylamide gel electrophoresis (PAGE) was also performed in order to further verify the successful conjugation as shown in Fig. 2. Upon conjugation to poly[(PAA)-b-(DMA)_co(PDSMA)] at pyridyl disulfide to thiol molar ratios of 2.5 and 3.5, an increase in the molecular weight of ovalbumin was observed in lanes 5 and 6 respectively, as compared to ovalbumin and thiolated ovalbumin in lanes 2 and 3 respectively. The complete disappearance of the bands corresponding to free ovalbumin is observed at both pyridyl disulfide to thiol ratios. These results suggest that all of the protein has been conjugated to the diblock copolymer. Poly[(PAA)-b-(DMA)_co(PDSMA)], lane 7, does not stain under these conditions and therefore does not interfere with localization of the ovalbumin.

Fig. 2 Native polyacrylamide gel electrophoresis (PAGE) lane (1) ladder, (2) ovalbumin, (3) thiolated ovalbumin, (4) ladder, (5) poly[(PAA)-b-(DMA)_co(PDSMA)]–ovalbumin conjugate ([polymer]₀/[protein]₀ = 2.5 : 1), (6) poly[(PAA)-b-(DMA)_co(PDSMA)]–ovalbumin conjugate ([polymer]₀/[protein]₀ = 3.5 : 1) and (7)poly[(PAA)-b-(DMA)_co(PDSMA)] the same concentration as in lane 5.

pH-responsive membrane destabilizing activity of poly[(PAA)-b-(DMA)_co(PDSMA)] and ovalbumin conjugates

Both polymer and polymer–ovalbumin conjugates were evaluated for their ability to induce red blood cell hemolysis at pH values relevant to the endosomal/lysosomal trafficking pathway. Three different pH conditions were used to mimic the extracellular environment (pH = 7.4) as well as the early (pH =6.6), and late endosome (pH = 5.8) (Fig. 3). As expected free ovalbumin did not show any significant red blood cell lysis under any of the pH conditions tested. Both the poly[(PAA)-b-(DMA)_co(PDSMA)] and the polymer–ovalbumin conjugates exhibit low red blood cell lysis at physiologic pH. An increase in the membrane destabilizing potential of both the diblock copolymer as well as the corresponding polymer–ovalbumin conjugates is observed at a pH of 6.6. Interestingly, in both cases the membrane disruptive properties are less than those typically observed for free PAA. This result suggests that the incorporation of a 2^nd hydrophilic block has an effect on the endosomalytic properties of the PAA segment. At pH 5.8, however, dramatic increase in the membrane disruptive properties is observed for both the diblock copolymer and the polymer–protein conjugate. In both cases nearly quantitative red blood cell lysis is observed at a polymer concentration of 20 μg mL⁻¹ which is consistent with hemolysis results for free poly(PAA) at these concentrations and pH values. These results taken together suggest that the incorporation of a hydrophilic segment does influence the magnitude of red blood cell lysis at intermediate pH values but that the conjugation of a large protein does not diminish the membrane disruptive potential of these materials at lower pH values.

Fig. 3 Red blood cell hemolysis as a function of pH for thiolated ovalbumin (22.15 μg mL⁻¹), poly[(PAA)-b-(DMA)_co(PDSMA)] diblock copolymer (20 μg mL⁻¹), and ovalbumin–poly[(PAA)-b-(DMA)_co(PDSMA)] conjugates (20 μg mL⁻¹). Data represent a single experiment in quadruplicate ± standard deviation and are normalized to a positive control, 1% v/v Triton X-100.

Tumor protection assay

To evaluate the ability of the poly[(PAA)-b-(DMA)_co(PDSMA)]–ovalbumin conjugates to activate CTLs in vivo, a preliminary tumor protection experiment was performed using the EG7 tumor model. The experiment involved three groups of animals as follows: mice immunized with PBS, mice immunized with free ovalbumin, and mice immunized with an equal amount of ovalbumin conjugated to poly[(PAA)-b-(DMA)_co(PDSMA)]. Each mouse received a subcutaneous vaccine injection in the right flank seven days prior to EG7.OVA tumor cell injections in the left flank (Fig. 4). Tumors were visible in the PBS and free ovalbumin immunized mice by day 7 but were not visible in the polyPAA–ovalbumin conjugate immunized mice until day 18. The mice immunized with the polyPAA–ovalbumin conjugate had a survival rate at day 21 of 100% versus 20% for PBS and 40% for ovalbumin immunized mice. Despite common administration in this model, no boosters were utilized in this study.¹²

Fig. 4 Kaplan–Meier survival plot. Mice were immunized with PBS (diamonds), free ovalbumin (boxes), and poly[(PAA)-b-(DMA)_co(PDSMA)]–ovalbumin conjugates (triangles) seven days prior to EG.7-OVA tumor cell injections (administered on day 1). Mice were eliminated from the study and euthanized when tumors reached 1250 mm³.

Experimental details

Materials

All reagents were purchased from Sigma-Aldrich and Wako Chemicals and used without further purification unless specified otherwise. The trithiocarbonate CTA ethyl cyanovaleric trithiocarbonate (ECT) was synthesized as previously described. N,N-Dimethylacrylamide was distilled under reduced pressure. Propylacrylic acid (PAA) was synthesized as previously reported.^13,142,2-Azobisisobutyronitrile (AIBN) was recrystallized from methanol.

Synthesis of pyridyl disulfide methacrylate

Aldrithiol-2 (5 g, 22.59 mmol) was dissolved in 40 mL of methanol and 1.8 mL of AcOH. The solution was added to a solution of 2-aminoethanethiol·HCl (1.28 g, 11.30 mmol) in 20 mL methanol for 30 minutes. The reaction was stirred under N₂ for 28 h at room temperature. After evaporation of solvents, the residual oil was washed twice with 40 mL of diethyl ether. The crude compound was dissolved in 10 mL of methanol and the product was precipitated twice with 50 mL of diethyl ether to yield pyridine dithioethylamine a white solid. Yield 95%.

Pyridine dithioethylamine (6.7 g, 30.07 mmol) and triethylamine (4.23 mL, 30.37 mmol) were dissolved in DMF (25 mL) and TEA (4.5 mL, 35 mmol) and methacryloyl chloride (3.33 mL, 33.08 mmol) was added slowly via syringe at 0 °C. The reaction mixture was then stirred for 2 hours at room temperature. After this time the reaction was quenched with saturated NaHCO₃ (100 mL) and extracted with ethyl acetate (150 mL × 3). The combined organic layer was further washed by 10% HCl (100 mL, 1 time) and pure water (100 mL, 2 times) and dried by MgSO₄. The resultant viscous oil was then purified by column chromatography (ethyl acetate/hexanes 1/10 to 2/1). ¹H NMR (CDCl₃): δ 8.45 (NCH), 7.61 (SCCHCH), 7.11 (NCHCH), 5.80 (CCH₂), 5.36 (CCH₂), 3.60 (SCH₂), 2.94 (SCH₂CH₂), CCH₃ (2.04).

Synthesis of poly(propylacrylic acid) macro chain transfer agents

The RAFT polymerization of PAA was conducted at 60 °C for 48 hours under a nitrogen atmosphere in septa-sealed vials. The initial CTA to initiator and monomer to CTA ratios ([M]₀/[CTA]₀) were 1 : 1 and 150 : 1 respectively. Briefly, PAA monomer (2 g, 17.5 mmol) was added to ECT (30.8 mg, 0.12 mmol) and AIBN (19.2 mg, 0.12 mmol). The polymerization solution was then transferred to a septa-sealed vial and purged with nitrogen for 30 minutes. After this time the vial was transferred to a preheated oil bath at 60 °C and allowed to react for 48 hours. The polymer was then isolated by precipitation as a solution in DMF into a 100× excess of diethyl ether (×5) (M_n = 13 100 g mol⁻¹, PDI = 1.69, conversion = 71%).

Block copolymerization of DMA and PDSMA from poly(PAA) macroCTA

To a solution of the poly(PAA) macroCTA (13 100 g mol⁻¹, PDI = 1.69) (0.4 g, 0.0305 mmol) in DMSO (1.5156 g) were added DMA (0.4535 g, 4.58 mmol), PDSMA (0.2925 g, 1.15 mmol), and AIBN (0.5 mg, 0.00305 mmol). The polymerization solution was then transferred to a septa-sealed vial and purged with nitrogen for 30 minutes. After this time the polymerization vial was transferred into a preheated oil bath at 70 °C and allowed to polymerize for 6 hours. The polymer was isolated by repeated precipitation in ether from dimethylsulfoxide (DMSO). The diblock copolymer was further purified by a PD-10 desalting column (Sephadex™ G-25, MWCO 5 kDa) and isolated by lyophilization. NMR and GPC analyses were then performed on the precipitated polymer (M_n = 15 900 g mol⁻¹, PDI = 1.63). TCEP reduction was then performed on the lyophilized polymer as a second method of evaluating the number of pyridyl disulfide groups per polymer.

Coupling of thiolated ovalbumin with poly[(PAA)-b-(DMA)_co(PDSMA)] viadisulfide exchange

A 15× molar excess of Traut's reagent was added to ovalbumin in 0.1 M phosphate buffer, pH 7.8, 0.15 M NaCl, 5 mM EDTA. The reaction was stirred at room temperature for 1 hour and purified using a PD-10 desalting column (Sephadex G-25, MWCO 5 kDa) according to the manufacturer's instructions. The degree of thiol modification with Traut's reagent was determined using Ellman's reagent. Briefly, 5,5′-dithio-bis-(2-nitrobenzoic acid) (Ellman's reagent) is reacted with sulfhydryls resulting in the formation of 2-nitro-5-thiobenzoic acid (ε = 14 150 M⁻¹ cm⁻¹ at 412 nm).

Thiol-functionalized ovalbumin and poly[(PAA)-b-(DMA)_co(PDSMA)] were coupled viadisulfide exchange reaction with polymeric pyridyl disulfide moieties. Poly[(PAA)-b-(DMA)_co(PDSMA)] was added to thiol functionalized ovalbumin at either a 2.5 or 3.5 times molar excess and allowed to react at room temperature for 3 hours in 0.1 M phosphate buffer, pH 7.8, 0.15 M NaCl, 5 mM EDTA. The degree of conjugation was determined by monitoring the release of pyridine 2-thione spectrophotometrically at 343 nm using an extinction coefficient of 8080 M⁻¹ cm⁻¹ for pyridine 2-thione. The degree of protein conjugation to the polymer was verified using native polyacrylamide gel electrophoresis (PAGE). A 4–20% Tris–HCl gel along with 1× running buffer (30.3 g Tris base, 144 g glycine, 10 g SDS in 1 L ddH₂O diluted 10×), 5× load buffer (0.25 M Tris, pH 7, 0.1% bromophenol blue, 50% glycerol), and Kaleidoscope Prestained Standards (BIO-RAD Catalog # 161-6324) were used for this assay. The gel was run for 1 hour at 125 Volts at room temperature and subsequently stained for 4 hours in GelCode blue and destained overnight with ddH₂O.

pH-dependent membrane disruption of carriers and ovalbumin conjugates

Hemolysis¹⁵ was used to determine the potential endosomolytic activity of both the poly[(PAA)-b-(DMA)_co(PDSMA)] and the corresponding ovalbumin conjugate at pH values that mimic endosomal trafficking (extracellular pH = 7.4, early endosome pH = 6.6, and late endosome pH = 5.8). Briefly, whole human blood was collected in vacutainers containing EDTA. Blood was centrifuged, plasma aspirated, and washed three times in 150 mM NaCl to isolate the red blood cells (RBCs). The RBCs were then resuspended in phosphate buffer (PB) at pH 7.4, pH 6.6, or pH 5.8. Poly[(PAA)-b-(DMA)_co(PDSMA)] (20 μg mL⁻¹) or the corresponding ovalbumin conjugates were then incubated with the RBC at the three pH values for 1 h at 37 °C. Intact RBCs were then centrifuged and the hemoglobin released into supernatant was measured by absorbance at 541 nm as an indication membrane disruption. Untreated cells and cells treated with Triton X served as the negative and positive controls respectively.

EG.7-OVA cell culture

EG.7 OVA cells were maintained between 1 × 10⁵ and 1 × 10⁶cells mL⁻¹ in a RPMI 1640 medium containing 2 mM L-glutamine, 1.5 g L⁻¹sodium bicarbonate, 4.5 g L⁻¹glucose, 10 mM HEPES and 1.0 mM sodium pyruvate and supplemented with 0.05 mM 2-mercaptoethanol and 0.4 mg mL⁻¹ G418 and 10% fetal bovine serum renewing fresh medium every 2 to 3 days.

In vivo tumor protection assay

Female C57BL/6 mice 6–8 weeks old were anesthetized with isoflurane and injected subcutaneously with 150 μL of PBS, ovalbumin, or poly[(PAA)-b-(DMA)_co(PDSMA)] conjugated to ovalbumin, with an equivalent of 100 μg ovalbumin injected in each mouse (n ≥ 5). Seven days after immunization, ovalbumin tumors were established by subcutaneously injecting 1 × 10⁶EG.7-OVA tumor cells in 100 μL saline into the left flank of each mouse. After injection, tumor growth was monitored every two to three days by measuring two perpendicular axes using digital calipers. Tumor volume was calculated using the ellipsoidal equation V = 0.5 × length × width².¹⁰ Once the tumor reached 1250 mm³, mice were eliminated from the study by euthanasia in accordance with the University of Washington IACUC protocols.

Instrumentation

The polymers in this study were characterized by gel permeation chromatography using three Tosoh TSK-GEL columns (TSK-α3000, α3000, α4000) (Tosoh Bioscience, Montgomeryville, PA) connected in series to a MiniDAWN TREOS multi-angle light scattering detector and an Optilab rEX RI detector (Wyatt, Santa Barbara, CA). The mobile phase was HPLC-grade DMF containing 0.1 wt% LiBr. For hemolysis and LDH cytotoxicity assays a fluorescence microplate reader (Tecan Safire2, Switzerland) was used. NMR and mass spectroscopy analyses were performed on a Bruker AVance AV300 (300 MHz) spectrometer and a Bruker Esquire Liquid Chromatograph Ion Trap Mass Spectrometer, respectively.

Conclusions

RAFT polymerization was used to polymerize pH-responsive, diblock copolymers containing approximately 2 pyridyl disulfide groups per polymer. These polymers were successfully coupled to thiolated ovalbumin viadisulfide exchange to prepare well-defined polymer–protein conjugates at [polymer]₀/[protein]₀ ratios as low as 2.5 : 1. The diblock copolymer alone and the ovalbumin conjugate exhibited nearly 100% red blood cell lysis at pH 5.8 at 20 μg mL⁻¹. A preliminary tumor protection experiment was performed using the EG7 ovalbumin mouse tumor model. Tumors were visible in the PBS and free ovalbumin immunized mice by day 7 but were not visible with the polyPAA conjugate vaccine until day 18. Initial animal studies demonstrated that well-defined polyPAA–ovalbumin conjugates can stimulate the immune system against an injected protein resulting in the enhanced rejection of EG7 tumor cellsin vivo. Further design optimizations will seek to improve these results by taking advantage of the synthetic flexibility of this system through incorporation of APC targeting and immunostimulatory agents.

Acknowledgements

This work was funded by the National Institutes of Health (R01EB002991), a National Science Foundation Graduate Fellowship to EC, and Center for the Intracellular Delivery of Biologics which is supported by the Washington State Life Science Discovery Fund (grant number 2496490).

References

1. T. Boon, J. C. Cerottini, B. Vandeneynde, P. Vanderbruggen and A. Vanpel, Annu. Rev. Immunol., 1994, 12, 337; H. D. Engers, T. Lahaye, G. D. Sorenson, A. L. Glasebrook, C. Horvath and K. T. Brunner, J. Immunol., 1984, 133, 1664.
2. M. C. Bonnet, J. Tartaglia, F. Verdier, P. Kourilsky, A. Lindberg, M. Klein and P. Moingeon, Immunol. Lett., 2000, 74, 11; R. Harrop, J. John and M. W. Carroll, Adv. Drug Delivery Rev., 2006, 58, 931; T. J. Goletz, K. R. Klimpel, S. H. Leppla, J. M. Keith and J. A. Berzofsky, Hum. Immunol., 1997, 54, 129; C. A. Shaw and M. N. Starnbach, Infect. Immun., 2006, 74, 1001.
3. J. G. Altin and C. R. Parish, Methods, 2006, 40, 39; G. Chikh and M. P. Schutze-Redelmeir, Biosci. Rep., 2002, 22, 339.
4. K. J. Maloy, A. M. Donachie, D. T. Ohagan and A. M. Mowat, Immunology, 1994, 81, 661; N. L. Bennewitz and J. E. Babensee, Biomaterials, 2005, 26, 2991; J. Panyam, W. Z. Zhou, S. Prabha, S. K. Sahoo and V. Labhasetwar, FASEB J., 2002, 16, 1217; K. D. Newman, P. Elamanchili, G. S. Kwon and J. Samuel, J. Biomed. Mater. Res., 2002, 60, 480.
5. Y. J. Kwon, S. M. Standley, S. L. Goh and J. M. Frechet, J. Controlled Release, 2005, 105, 199; S. M. Standley, Y. J. Kwon, N. Murthy, J. Kunisawa, N. Shastri, S. J. Guillaudeu, L. Lau and J. M. J. Frechet, Bioconjugate Chem., 2004, 15, 1281; E. M. Bachelder, T. T. Beaudette, K. E. Broaders, S. E. Paramonov, J. Dashe and J. M. J. Frechet, Mol. Pharmaceutics, 2008, 5, 876.
6. C. V. Harding and R. Song, J. Immunol., 1994, 153, 4925; M. Kovacsovics-Bankowski, K. Clark, B. Benacerraf and K. L. Rock, Proc. Natl. Acad. Sci. U. S. A., 1993, 90, 4942; L. Thiele, H. P. Merkle and E. Walter, Pharm. Res., 2003, 20, 221.
7. T. Yoshikawa, N. Okada, A. Oda, K. Matsuo, Y. Mukai, Y. Yoshioka, T. Akagi, M. Akashi and S. Nakagawa, Biochem. Biophys. Res. Commun., 2008, 366, 408; T. Yoshikawa, N. Okada, A. Oda, K. Matsuo, H. Kayamuro, Y. Ishii, T. Yoshinaga, T. Akagi, M. Akashi and S. Nakagawa, Vaccine, 2008, 26, 1303.
8. N. Murthy, M. C. Xu, S. Schuck, J. Kunisawa, N. Shastri and J. M. J. Frechet, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 4995; E. M. Bachelder, T. T. Beaudette, K. E. Broaders, S. E. Paramonov, J. Dashe and J. M. J. Frechet, Mol. Pharmaceutics, 2008, 5, 876.
9. N. Murthy, J. R. Robichaud, D. A. Tirrell, P. S. Stayton and A. S. Hoffman, J. Controlled Release, 1999, 61, 137; V. Bulmus, M. Woodward, L. Lin, N. Murthy, P. S. Stayton and A. S. Hoffman, J. Controlled Release, 2003, 93, 105; R. A. Jones, C. Y. Cheung, F. E. Black, J. K. Zia, P. S. Stayton, A. S. Hoffman and M. R. Wilson, Biochem. J., 2003, 372, 65.
10. S. Foster, C. L. Duvall, E. F. Crownover, A. S. Hoffman and P. S. Stayton, Bioconjugate Chem., 2010, 21, 2205.
11. G. W. Hastings, J. Chem. Soc. D, 1969, 1039.
12. S. T. Reddy, A. J. van der Vlies, E. Simeoni, V. Angeli, G. J. Randolph, C. P. O'Neill, L. K. Lee, M. A. Swartz and J. A. Hubbell, Nat. Biotechnol., 2007, 25, 1159.
13. M. Ferrito and D. A. Tirrell, Macromol. Synth., 1992, 11, 59–62.
14. A. J. Convertine, D. S. W. Benoit, C. L. Duvall, A. S. Hoffman and P. S. Stayton, J. Controlled Release, 2009, 133, 221.
15. C. A. Lackey, N. Murthy, O. W. Press, D. A. Tirrell, A. S. Hoffman and P. S. Stayton, Bioconjugate Chem., 1999, 10, 401.

---

Footnote

† Co-first authors.

---