Poly(2‐propylacrylic acid)/poly(lactic‐co‐glycolic acid) blend microparticles as a targeted antigen delivery system to direct either CD4+ or CD8+ T cell activation

Abstract Poly(lactic‐co‐glycolic acid) (PLGA) based microparticles (MPs) are widely investigated for their ability to load a range of molecules with high efficiency, including antigenic proteins, and release them in a controlled manner. Micron‐sized PLGA MPs are readily phagocytosed by antigen presenting cells, and localized to endosomes. Due to low pH and digestive enzymes, encapsulated protein cargo is largely degraded and processed in endosomes for MHC‐II loading and presentation to CD4+ T cells, with very little antigen delivered into the cytosol, limiting MHC‐I antigenic loading and presentation to CD8+ T cells. In this work, PLGA was blended with poly(2‐propylacrylic acid) (PPAA), a membrane destabilizing polymer, in order to incorporate an endosomal escape strategy into PLGA MPs as an easily fabricated platform with diverse loading capabilities, as a means to enable antigen presentation to CD8+ T cells. Ovalbumin (OVA)‐loaded MPs were fabricated using a water‐in‐oil double emulsion with a 0% (PLGA only), 3 and 10% PPAA composition. MPs were subsequently determined to have an average diameter of 1 µm, with high loading and a release profile characteristic of PLGA. Bone marrow derived dendritic cells (DCs) were then incubated with MPs in order to evaluate localization, processing, and presentation of ovalbumin. Endosomal escape of OVA was observed only in DC groups treated with PPAA/PLGA blends, which promoted high levels of activation of CD8+ OVA‐specific OT‐I T cells, compared to DCs treated with OVA‐loaded PLGA MPs which were unable activate CD8+ T cells. In contrast, DCs treated with OVA‐loaded PLGA MPs promoted OVA‐specific OT‐II CD4+ T cell activation, whereas PPAA incorporation into the MP blend did not permit CD4+ T cell activation. These studies demonstrate PLGA MP blends containing PPAA are able to provide an endosomal escape strategy for encapsulated protein antigen, enabling the targeted delivery of antigen for tunable presentation and activation of either CD4+ or CD8+ T cells.


| I N T R O D U C T I O N
Biomaterials have traditionally been developed to avoid provoking chronic aggravation of the immune system. 1 This complex network of cellular interactions and processes has evolved to provide protection against foreign invaders as well as homeostatic regulation of self to nonself. Strategies that aim to manipulate components of the immune system, particularly polymer-based vaccines, have gained attention in recent years. [1][2][3][4] Polymeric microparticles (MPs) offer a distinct advantage as their intrinsic immunomodulatory properties and chemical versatility fulfill desired properties in achieving long-term protection.
Generation of protective immunity following vaccination depends on antigen availability, delivery to antigen presenting cells (APCs) and further activation of innate immunity through the appropriate pathways. 5 Current vaccination strategies are adept at generating antibodies and CD4 1 T cell mediated immunity, but approaches to activate CD8 1 T cell pathways, necessary to combat intracellular pathogens such as malaria, HIV and cancer, are limited. Present strategies are based on live-attenuated or inactivated pathogens that induce robust humoral and cellular immunity. However, safety concerns associated with live vectors limits their use in infants, the elderly and the immunocompromised. 6 Conversely, recombinant antigens based on peptides hold high potential for CD8 1 T cell vaccine development but their dependency on adjuvants remains a major hurdle to clinical translation. 7 In this study, we develop a unique polymer blend for the fabrication of microparticles that allow for the tuning of CD4 1 and CD8 1 T cell responses through targeting of dendritic cells (DCs). We hypothesize incorporation of a small amount of Poly(2-propylacrylic acid) (PPAA) into a poly (lactic-co-glycolic acid) (PLGA) matrix, will lead to endosomal escape triggering cross-presentation pathway within DCs and further activation of CD8 1 T cells.
Dendritic cells are sentinels at the forefront of immunological responses as the most efficient APCs and regulators of adaptive immunity. Dendritic cells are constantly surveying the environment and are able to uptake, process and present antigen to naïve T cells and appropriately shape the resulting T and B cell responses. 8  have the ability to process exogenous proteins and present them in the context of MHC Class I. This function, known as cross-presentation, diversifies the ability of the immune system to generate immunity to various pathogens. 9 Microparticles fabricated of PLGA are one of the most widely used biocompatible, biodegradable polymeric carriers with multiple application in delivery of drugs, biomolecules and genes. PLGA systems are able to efficiently load hydrophilic and hydrophobic molecules and can easily be fabricated to specific micron size ranges, efficiently phagocytosed by DCs, and allowing targeted delivery of antigen to endosomal compartments. For sustained cytosolic delivery and activation of cross-presentation pathways an endosomal escape strategy must be employed. Several compounds have been incorporated into PLGA to allow blended formulation with improved stability, loading and delivery capacity. Poly (2-propylacrylic acid), has the ability to reversibly switch from a hydrophilic soluble conformation at physiological pH to a hydrophobic and membrane interactive state in response to acidic pH. Functionally, PPAA-containing systems trigger the disruption of and release from acidified endosomal compartments, effectively targeting cargo to cytosol. However, PPAA alone cannot readily support a matrix capable of encapsulating bioactive molecules. Therefore, in this study we incorporate PPAA into the PLGA matrix of MPs in order to provide cytosolic delivery of a model antigen, while maintaining robust properties of PLGA, thus accessing MHC-I loading and allowing the tuning of CD4 1 and CD8 1 T cell responses.

| Mice
The studies reported here conform to the animal welfare ACT and the

| Isolation and culture of murine bone marrow dendritic cells
Dendritic cells were derived from mouse bone marrow as previously described. 11 In brief, mice were euthanized by CO 2 asphyxiation followed by cervical dislocation. Tibias and femurs were harvested and marrow cells collected by flushing the shafts of the bones with a 25-g needle using wash media (RPMI medium containing 10% heat inactivated filter sterilized fetal bovine serum, and 1% penicillin-streptomycin) and mixed to obtain a single-cell suspension. The solution was then strained through a 70 mm cell strainer and centrifuged at 1,700 rpm for 5 min at 48C. Erythrocytes were then removed by resuspending in ACK lysis buffer for 5 min at room temperature and centrifuging at the conditions specified above. Resulting cells were cultured in DC media (complete DMEM/F12 medium containing 2.5 mM L-glutamine, 100 U/ml penicillin-streptomycin, 10% heat inactivated filter sterilized FBS, 1 mM sodium pyruvate, and 1x NEAA mixture and 20 ng/ml GM-CSF). Monocytes were selected by plastic adherence for 6 days at 378C with 95% relative humidity and 5% CO 2 . The culture was fed every 2 days by gently aspirating 50% medium and adding fresh medium. On day 6 of culture, the medium was discarded and non/ loosely adherent cells containing dendritic cells were collected. A single-cell suspension of cells was then plated immediately in round coverslips placed in 6-well plates at 5 x 10 5 cells/well for endosomal disruption study, or in round-bottom 96-well micro-titer plates (Corning, Corning, NY) at 2.5 x 10 4 cells/well for T cell proliferation assay.

| T cell proliferation assay by flow cytometry
On day 8 of culture, DCs were seeded with OVA-loaded PLGA-based microparticles at a 50:1 ratio (MPs: cells) and incubated for an additional 2 days. On day 10, CD4 1 and CD8 1 T cells were isolated from the spleen of OT-I and OT-II 8-week old female mice. Briefly, animals were euthanized by CO 2 asphyxiation followed by cervical dislocation.
Spleens were aseptically harvested and homogenized in wash media.
Cells were then passed through a 70 mm cell strainer and centrifuged at 1,700 rpm for 5 min at 48C. Red blood cells were eliminated using ACK lysis buffer followed by centrifugation to recover lymphocytes.

| Statistical analysis
Data reported were analyzed and charts were generated using Prism 5.0 (GraphPad, San Diego, CA). Statistics were done using a one-way ANOVA followed by Tukey's post hoc analysis to make pair-wise comparisons, with 95% confidence intervals. Two-way ANOVA was used to determine differences in T cell proliferation with a Bonferroni's post hoc analysis. Unless otherwise indicated, data represent the mean 6 SEM, with p < .05 considered statistically significant.  (Figure 3).

| PPAA-mediated endosomal disruption
The PPAA-mediated endosomal escape was examined by micros-  able to escape the endosome to some degree, a significant increase was observed with both the 3 and 10% blend. The capacity to disrupt endosomal membranes increased directly proportional to the percentage of PPAA in the particle formulation.

| Antigen-specific T cell proliferation study
DCs pretreated for 24 hr with OVA-loaded microparticle formulations were co-cultured with CFSE-labeled OVA-specific T cells for 5 days. As shown in Figure 5a

| DISCUSSION
This study has provided a novel approach to endosomal escape strategies, enabling the tuning of antigen delivery to either endosomal or cytosolic compartments to shape subsequent CD4 1