Efficient T cell activation and effector responses require an antigenic peptide presented on the MHC complex to the TCR (signal 1), costimulatory molecule interactions between T cells and APCs (signal 2), and the synthesis of innate immune-derived proinflammatory cytokines and reactive oxygen species (signal 3). We previously demonstrated that the third signal dissipation impairs autoreactive T cell activation. In this study, we tested the hypothesis that encapsulation of Ag with an antioxidant-containing biomaterial would induce Ag-specific hyporesponsiveness. We cocultured bone marrow–derived dendritic cells with microcapsules composed of multilayer-assembled poly(N-vinylpyrrolidone) (PVPON) and the antioxidant tannic acid (TA). LPS-activated dendritic cells cocultured with (PVPON/TA) microcapsules displayed a decrease in TNF-α, IL-12p70, and CXCL10 synthesis. To study Ag-specific T cell responses, we incorporated chicken OVA into the (PVPON/TA) multilayers and stimulated OT-II splenocytes in a primary recall assay. Flow cytometric analysis demonstrated a significant inhibition of CD4 T cell activation markers, upregulation of CTLA-4 and PD-1, and blunted secretion of IL-2, IFN-γ, TNF-α, and CXCL10 by ELISA. To test microcapsule efficacy in vivo, we immunized OT-II mice with (PVPON/TA)-OVA microcapsules and performed an OVA recall assay. Immunization of OT-II mice with (PVPON/TA)-OVA microcapsules elicited a decrease in CD4 T cell differentiation and effector responses including IFN-γ, TNF-α, CCL3, and CCL5 by ELISA compared with OVA immunization alone. These data show that microcapsules composed of antioxidant and encapsulated Ags can effectively blunt innate immune-derived proinflammatory third signal synthesis necessary for Ag-specific effector T cell responses and present a prospective strategy for T cell–mediated autoimmunity.

Efficient T cell maturation and activation depends on three signals: binding of the TCR with processed antigenic peptides presented by MHC class I and MHC class II (MHC-II) complexes (signal 1), costimulatory molecule interactions between APCs and T cells (signal 2), and reactive oxygen species (ROS) synthesis and redox-dependent activation of proinflammatory cytokines (signal 3) (15). We previously demonstrated that dissipation of the ROS-dependent third signal impairs autoreactive T cell activation and diminishes CD4 and CD8 T cell differentiation and effector responses that result in pancreatic β cell destruction in NOD mice, as well as Ag-dependent activation in OT-II mice (1, 2, 59). One challenge in treating T cell–mediated diseases is avoiding disruption of innate and adaptive immune responses. Current attempts to employ global immunosuppressive agents can result in patient vulnerability to microbial and viral infections (10). For these reasons, development of Ag-specific approaches to avoid global immunosuppression while inhibiting aberrant T cell activation and restoring tolerance warrants further study.

Because of the integral role of T cells in numerous inflammatory- and autoimmune-mediated diseases, selectively suppressing T cell effector responses by using Ag-specific strategies is ideal and may limit the risk of off-target and systemic effects (10, 11). Cell-based therapies using ex vivo administration of Ag-specific autologous regulatory T cells (Tregs) are in clinical testing to examine the efficacy of these immunosuppressive cells to delay autoimmune diseases such as type 1 diabetes (T1D) (12, 13). Adoptive transfer of expanded Foxp3+ Tregs has shown promise in mice; however, technical limitations of human Tregs expansion ex vivo still need to be overcome (14). In addition to the use of Tregs, transfer of immunomodulatory dendritic cells (DCs) can restore Ag-specific peripheral tolerance and induce anergy in effector T cells (15, 16). As professional APCs, DCs are specialized to present numerous Ags to stimulate naive polyclonal T cells (17) and influence the maturation and effector responses of autoreactive T cells (18). Thus, DCs act as an essential bridge between the innate and adaptive arms of the immune system (19). In T1D, NOD macrophages and DCs are hyperactivated and hyperinflammatory, which contribute to the maturation of islet-reactive T cells (2022). Exposure of DCs to apoptotic cells can induce tolerance in CD4+ and CD8+ T cells (15, 23), and disruption of CD40/CD40L interaction in DCs can block clonal autoreactive CD4+ T cell expansion in NOD mice (16). Tarbell et al. (24) showed that DCs could induce the expansion of CD4+ CD25+ Tregs to prevent the adoptive transfer of T1D. In experimental autoimmune encephalomyelitis, a mouse model for multiple sclerosis, nanoparticle delivery of a T cell epitope of myelin oligodendrocyte glycoprotein to DCs can induce a tolerogenic DC phenotype, expand Foxp3+ Tregs, and ablate experimental autoimmune encephalomyelitis (25). Therefore, DCs represent an attractive target for intervention to prevent Ag-specific autoreactive T cell responses.

By exploiting the redox-dependent third signal necessary for naive T cell adaptive immune maturation, we investigated the feasibility of targeted Ag delivery to DCs accompanied by antioxidant treatment to inhibit the effector response of Ag-specific T cells. We recently reported that in vitro treatment of Ag-stimulated T cells with a layer-by-layer (LbL) coating of the antioxidant tannic acid (TA) assembled with the neutral polymer poly(N-vinylpyrrolidone) (PVPON) through hydrogen-bonding can dissipate ROS synthesis and inhibit the production of T cell–derived proinflammatory cytokines IFN-γ and TNF-α (26, 27). More recently, we demonstrated that (PVPON/TA) encapsulation of pancreatic islets can specifically inhibit T cell migration by diminishing the synthesis of proinflammatory CXCL10 and CCL5 chemokines (28).

In this study, we investigated whether incorporation of an Ag into the (PVPON/TA) microcapsule architecture could elicit Ag-specific hyporesponsiveness. We tested the hypothesis that encapsulation of Ags within these polymer microcapsules will abrogate DC-derived inflammatory third signal responses, and consequently, blunt Ag-specific effector T cell responses. By using OT-II mice, a CD4 TCR-transgenic mouse model that specifically recognizes aa 323–339 of chicken OVA (29), we show that administration of TA-containing microcapsules with encapsulated chicken OVA can attenuate CD4 T cell activation and effector cytokine production. Furthermore, these data provide evidence that encapsulation of autoantigens with (PVPON/TA) microcapsules may be efficacious in treating T cell–mediated autoimmune diseases.

NOD/ShiLtJ and OT-II mice (29) were bred and housed under pathogen-free conditions in the Research Support Building animal facility at the University of Alabama at Birmingham. OT-II mice were originally obtained as a kind gift from Dr. F. Lund at the University of Alabama at Birmingham. All mice were maintained on a light/dark (12/12 h) cycle at 23°C and received ad libitum standard laboratory chow and acidified water. Male NOD mice at 10–12 wk of age were used for bone marrow harvesting and DC differentiation. Male OT-II mice at 10–12 wk of age were used for all Ag-specific in vitro experiments, as well as all in vivo experiments. For immunization experiments, littermates were spread among experimental groups whenever possible. All animal studies were performed in accordance with the University of Alabama at Birmingham Institutional Animal Use and Care Committee in compliance with the laws of the United States of America.

Bone marrow hematopoietic stem cells were harvested from femurs and tibias of NOD mice as described (5, 9). Cells were plated onto tissue culture–treated petri dishes at a density of 106 cells per milliliter in DC media, consisting of RPMI 1640 supplemented with 10% heat-inactivated FBS, 10 mM HEPES buffer, 2 mM l-glutamine, 200 μM nonessential amino acids, 1 mM sodium pyruvate, 61.5 μM 2-ME, 100 μg/ml gentamicin, 12.5 mg/ml NG-monomethyl-l-arginine (L-NMMA), and 1 μg/ml indomethacin. The following day, nonadherent cells were collected and resuspended at a concentration of 1 × 105 cells per milliliter in DC media supplemented with GM-CSF (1000 U/ml; R&D Systems) and IL-4 (100 U/ml; R&D Systems). After 7 d of differentiation, 1 × 106 DCs were cultured with 107 microcapsules (a 1:10 ratio) per well in a 24-well tissue culture plate (CoStar) in the presence or absence of 1 μg/ml LPS from Escherichia coli (055:B5; Sigma-Aldrich). Supernatants were collected at 24, 48, and 72 h post-LPS stimulation for cytokine and chemokine analysis by ELISA, whereas cells were prepared for either flow cytometry or mRNA extraction at the same time points.

The redox-sensitive substrate luminol was used to detect superoxide synthesis (30). DCs were plated into polystyrene plates at 5 × 105 cells per well, either alone or cocultured with 5 × 106 microcapsules, and subsequently stimulated with 100 ng/ml PMA and 1 μg/ml ionomycin. Luminescence was generated by adding 200 μM luminol (Sigma-Aldrich) and 0.32 U/ml of HRP (Sigma-Aldrich). Luminescence was recorded with a Synergy 2 microplate reader (BioTek Instruments) with measurements taken every 1 min for 60 min.

PVPON (average m.w., Mw = 1,300,000 g/mol), TA (m.w. = 1700 g/mol), poly(ethylene imine) (PEI) (average m.w., Mw = 25,000 g/mol), hydrofluoric acid, and mono- and dibasic sodium phosphate were purchased from Thermo Fisher Scientific. Ultrapure water (Evoqua Water Technologies) with a resistivity of 18.2 MΩ cm was used for preparation of buffered solutions. Nonporous silica microspheres 4.0 ± 0.1 μm in diameter were purchased from Cospheric. Chicken egg white OVA was purchased from Sigma-Aldrich (A5503). Hollow hydrogen-bonded (PVPON/TA) microcapsules were prepared using the stepwise LbL adsorption of polymers on the sacrificial silica particles (26). First, PVPON was adsorbed onto the silica surfaces by shaking a suspension of the silica microspheres in 0.5 mg/ml PVPON (0.01 M phosphate buffer, pH 4.5) for 10 min. The particles were pelleted at 2500 rpm, and the supernatant was removed before resuspending the particles in 0.01 M phosphate buffer at pH 4.5 and repeating the rinse cycle three times. TA was then adsorbed at 0.5 mg/ml solution (0.01 M phosphate buffer at pH 4.5) onto the PVPON-coated silica particles by shaking the suspension for 10 min, followed by centrifugation and rinsing with 0.01 M phosphate buffer as stated above. This process was repeated until the desired number of (PVPON/TA) bilayers had been adsorbed, giving capsule architectures either with TA as the outer layer [(PVPON/TA)8 labeled as TA-capped] or as PVPON as the outer layer [(PVPON/TA)8(PVPON) labeled as PVPON-capped], respectively, where the subscript denotes the number of polymer bilayers. The sacrificial silica templates were dissolved by suspending the core-shell particles in aqueous 8% (w/w) hydrofluoric acid for 10 h and then extensively dialyzed against 0.01 M phosphate buffer at pH 7.4 for 50 h. The hollow hydrogen-bonded (PVPON/TA) microcapsules were counted using a hemocytometer with typical concentrations ranging from 1 to 4 × 108 capsules per microliter.

To prepare the hydrogen-bonded microcapsules with chicken OVA encapsulated within the capsule shell, the TA and PVPON layers were deposited from 0.5 mg/ml polymer solutions (0.01 M phosphate buffer, 0.1 M NaCl [pH 7.4]) as described above onto surfaces of PEI-coated silica particles (PEI, 1 mg/ml, aqueous solution) for 5 min per layer. Lyophilized chicken OVA solution (1 mg/ml; in 0.01 M phosphate buffer, 0.1 M NaCl [pH 7.4]) was allowed to adsorb onto the TA layer for 10 min, followed by centrifugation and buffer rinses as described above. The (TA/PVPON) bilayer was adsorbed onto the OVA layer as described for the OVA-free microcapsules, with the final shell architectures of (TA/PVPON)6(TA/OVA)(TA/PVPON) labeled as OVA.1 and (TA/PVPON)5(TA/OVA)2(TA/PVPON) labeled as OVA.2, respectively. As the OVA-free control, the (TA/PVPON)8 microcapsules labeled as PVPON-capped were synthesized. The sacrificial silica templates were removed as described for OVA-free microcapsules. The OVA-containing capsules were prepared at pH 7.5 in the presence of 0.1 M NaCl (0.01 M phosphate buffer) to avoid denaturing of the protein during the multilayer shell formation. Because no hydrogen bonds can be formed between the silica core surfaces and PVPON at pH 7.4, PEI was deposited first at the microparticle surfaces to facilitate interaction of TA with PEI through ionic pairing at pH 7.4 followed by hydrogen-bonding interactions between PVPON and TA assembled alternatively at this pH condition. These polymer assembly conditions also provided the thickness of the polymer shell similar to that assembled at low pH (pH 4.5) and low ionic strength (0.01 M phosphate buffer) for the OVA-free hydrogen-bonded capsules as we demonstrated earlier (31).

Confirmation of equivalent OVA encapsulation was accomplished by SDS-PAGE and Western blot analysis for each batch of microcapsules generated. Based on the UV/vis spectrophotometer readings of encapsulated chicken OVA, 10 μg of (PVPON/TA/OVA) microcapsules were resuspended in cell lysis buffer (no. 9803; Cell Signaling Technology) and denatured with SDS-PAGE sample buffer at 95°C for 5 min. The microcapsule samples along with known OVA concentrations were separated on 10% SDS-PAGE denaturing gel and transferred to nitrocellulose membranes as we described (5). OVA was probed with an anti-chicken OVA (ab1221; Abcam) polyclonal Ab followed by an anti-rabbit IRDye 800CW secondary Ab (LI-COR). Blots were visualized on an Odyssey CLx Imager with Image Studio v4.0 software to calculate densitometry.

Spleens were isolated from OT-II mice and homogenized to single-cell suspensions in splenocyte media consisting of DMEM supplemented with 10% heat-inactivated FBS, 10 mM HEPES buffer, 4 mM l-glutamine, 200 μM nonessential amino acids, 1 mM sodium pyruvate, 61.5 μM 2-ME, and 100 μg/ml gentamicin. Splenocytes were seeded at 5 × 105 cells per well in a 96-well round-bottom plate and stimulated with either 1 μM OVA323–339 peptide, 40 μg/ml (1 μM) chicken OVA, or OVA.1 or OVA.2 microcapsules constituting 40 μg/ml chicken OVA. Equal numbers of OVA-free PVPON-capped microcapsules lacking chicken OVA were added as controls to compare with OVA.1 and OVA.2 microcapsules. rOVA peptide OVA323–339 (ISQAVHAAHAEINEAGR) was synthesized by Mimotopes and lyophilized chicken OVA protein was purchased from Sigma-Aldrich (A5503). Supernatants were collected at 24, 48, 72, 96, and 120 h poststimulation for cytokine and chemokine analysis by ELISA. Cells were analyzed by flow cytometry at the same timepoints.

OVA.1 and OVA.2 microcapsules were washed and resuspended in sterile 1× PBS (pH 7.4; Life Technologies) at a concentration of 1.0 mg/ml OVA as determined above by Western blot analysis and emulsified with an equal volume of CFA to achieve a final OVA concentration of 500 μg/ml. Mice were injected s.c. at the base of the tail in a final volume of 200 μl (100 μg OVA). At 7 d postimmunization, the inguinal lymph nodes (LNs) and spleen were harvested and separately homogenized into single-cell suspensions. Cells were set aside for immunophenotyping by flow cytometry and seeded at 5 × 105 cells per well in a 96-well round-bottom plate to be used in an Ag-recall assay with 125 μg/ml (4 μM) chicken OVA or an equimolar amount of OVA.1 and OVA.2 microcapsules. Supernatants were collected at 48, 72, 96, and 120 h poststimulation for cytokine and chemokine analysis by ELISA. Cells were analyzed by flow cytometry at the same timepoints.

Detection of cytokines and chemokines in cell culture supernatants was detected by ELISA. IFN-γ and IL-2 levels were measured using Ab pairs purchased from BD Biosciences, whereas CCL3, CCL5, CXCL10, IL-12p70, and TNF-α were detected using DuoSet ELISA Kits purchased from R&D Systems as we described (6). ELISA plates were read on a Synergy2 microplate reader, and cytokine or chemokine production was analyzed and assessed via Gen5 v.1.10 software (BioTek Instruments).

Flow cytometry was performed on DCs, LN cells, and splenocytes to analyze the expression of surface activation markers and costimulatory molecules as we previously published (6). Cells were resuspended in FACS buffer (1% BSA in PBS) and nonspecific Fc binding was blocked with anti-mouse CD16/CD32 (BD Biosciences). DC-specific cell surface markers analyzed were CD11c (clone HL3; BD Biosciences), I-Ag7 (clone OX-6; BD Biosciences), CD40 (clone 3/23; BD Biosciences), CD80 (clone 16-10A1; BD Biosciences), and CD86 (clone PO3; BioLegend). T cell–specific cell surface markers consisted of CD4 (clone GK1.5; BD Biosciences), CD25 (clone PC61; BD Biosciences), CD44 (clone IM7; BD Biosciences), CD62L (clone MEL-14, BD Biosciences) CD69 (clone H1.2F3; BD Biosciences), PD-1 (clone 29F.1A12; BioLegend), CTLA-4 (clone MEL-14; eBioscience), and a fixable Live/Dead exclusion dye (Invitrogen). After staining, cells were washed twice with FACS buffer, then fixed with BD Perm/Wash buffer (BD Biosciences) for 15 min, followed by another wash with FACS buffer and resuspended in FACS buffer. Samples were acquired using an Attune NxT Flow Cytometer (Thermo Fisher Scientific) and analyzed with FlowJo version 10.4 software (BD Biosciences). A minimum of 300,000 events were collected per sample. Gating strategies for DC and CD4 T cells are provided in Supplemental Figs. 1 and 2, respectively.

Fluorescently labeled (PVPON/TA) microcapsules were generated as detailed above, with Alexa Fluor 536–labeled PVPON being deposited on the outermost bilayer (27). The Alexa Fluor 536–labeled PVPON was obtained by labeling amine-containing PVPON with Alexa Fluor 536 carboxylic acid succinimidyl ester (Invitrogen) as we reported earlier (27). Bone marrow–derived DCs (5 × 106 cells) from NOD mice were incubated for 24 and 48 h in the presence of Alexa Fluor 536–labeled (PVPON/TA) microcapsules (1 × 107 microcapsules) containing TA (TA-capped) or PVPON (PVPON-capped) on the outer layer. After incubation, cells were collected, washed in FACS buffer, and examined by flow cytometry for phagocytosis of fluorescently labeled TA-capped and PVPON-capped microcapsules.

Stimulated DCs and splenocytes were collected, washed with PBS nd stored in RNAlater (Ambion) at −20°C according to the manufacturer’s protocols. RNA was collected using RNeasy plus (Qiagen) and cDNA was generated using Superscript RT III (Life Technologies) and amplified on a Roche LightCycler 480 Quantitative Real-Time PCR thermocycler. TaqMan (Applied Biosystems) gene expression probes used included Tnf (Mm00443258_m1), Il12a (Mm00434165_m1), Il12b (Mm00434174_m1), Cxcl10 (Mm00445235_m1), and Gapdh (Mm99999915_g1). The relative mRNA levels were calculated using the 2−ΔΔCycle threshold method and Gapdh was used as a housekeeping control gene for normalization. The unstimulated samples were used as calibrator controls and set as one.

Data were analyzed using GraphPad Prism Version 8.0 statistical software. Differences between means of each experimental group were compared by one-way ANOVA with either Dunnett or Tukey multiple comparisons tests, with p < 0.05 considered significant. All experiments were performed independently at least three separate times with data obtained in a minimum of triplicate wells in each in vitro experiment.

The innate immune-derived proinflammatory third signal consisting of ROS synthesis and redox-dependent induction of proinflammatory cytokines is required to mature adaptive immune effector responses. We hypothesized that we could exploit this requirement by encapsulating Ags in conjunction with an antioxidant to dissipate ROS synthesis and proinflammatory third signal generation. This would curtail innate immune responses and inhibit Ag-specific T cell responses. To test this hypothesis, we generated (PVPON/TA) hydrogen-bonded microcapsules as described in (26, 31) and represented schematically in (Fig. 1A). We have shown extensively in previous studies that (PVPON/TA) multilayers can decrease cellular ROS generation (2628). The LbL assembly of the capsule architecture allows us to easily test the efficacy of TA either exposed to the microenvironment on the outer layer or shielded by an additional layer of PVPON. Microcapsules with an outer layer of TA were more effective in blunting the secretion of proinflammatory cytokines, chemokines, and ROS by CD4 T cells (26) and macrophages (28) than microcapsules with an outer layer of PVPON. In the current study, we wanted to assess if the microcapsule outer layer containing TA or PVPON had an effect on DC function and their ability to stimulate Ag-specific T cells. Microcapsules consisting of eight bilayers of (PVPON/TA) that have the outer layer of TA are named as TA-capped, whereas microcapsules consisting of eight bilayers plus an additional layer of PVPON on the outer layer are named PVPON-capped.

FIGURE 1.

Generation of (PVPON/TA) microcapsules and coculture with Alexa Fluor 536–labeled (PVPON/TA) microcapsules with DCs.

Schematic for multilayer assembly of PVPON and TA on sacrificial spherical silica particles to obtain hollow capsules. Hydrogen-bonded multilayer (PVPON/TA) capsules were prepared via the hydrogen-bonding LbL technique by coating monodisperse silica particles 4 μm in diameter with (PVPON/TA)n nonionic multilayers where “n” denotes the number of bilayers followed by sacrificial core dissolution (A). Flow cytometry analysis of CD11c+ NOD bone marrow–derived DCs treated with Alexa Fluor 536–labeled TA-capped or PVPON-capped (PVPON/TA) microcapsules for 24 and 48 h (B). Data are representative of two independent experiments.

FIGURE 1.

Generation of (PVPON/TA) microcapsules and coculture with Alexa Fluor 536–labeled (PVPON/TA) microcapsules with DCs.

Schematic for multilayer assembly of PVPON and TA on sacrificial spherical silica particles to obtain hollow capsules. Hydrogen-bonded multilayer (PVPON/TA) capsules were prepared via the hydrogen-bonding LbL technique by coating monodisperse silica particles 4 μm in diameter with (PVPON/TA)n nonionic multilayers where “n” denotes the number of bilayers followed by sacrificial core dissolution (A). Flow cytometry analysis of CD11c+ NOD bone marrow–derived DCs treated with Alexa Fluor 536–labeled TA-capped or PVPON-capped (PVPON/TA) microcapsules for 24 and 48 h (B). Data are representative of two independent experiments.

Close modal

To examine if (PVPON/TA) microcapsules were phagocytosed by DCs, we labeled (PVPON/TA) microcapsules with Alexa Fluor 536 and cocultured with bone marrow–derived DCs for 24 and 48 h (Fig. 1B). Flow cytometric analysis showed that CD11c-expressing DCs readily phagocytosed the Alexa Fluor 536–labeled (PVPON/TA) microcapsules either TA-capped or PVPON-capped (Fig. 1B) in contrast to untreated DCs. There was a slight increase in the frequency of Alexa Fluor 536–labeled PVPON-capped (PVPON/TA) microcapsules phagocytosed by DCs compared with TA-capped microcapsules at 24 and 48 h postincubation.

To determine if phagocytosed (PVPON/TA) microcapsules are immunomodulatory and can decrease MHC-II, costimulatory molecule expression (signal 2), and proinflammatory cytokine (signal 3) synthesis, LPS-stimulated bone marrow–derived DCs from NOD mice were cocultured with TA-capped or PVPON-capped (PVPON/TA) microcapsules for 24 h. Because DCs from NOD mice display an inherent exacerbated proinflammatory response (20, 21), we reasoned that this mouse model would be ideal to assess the anti-inflammatory effects of (PVPON/TA) microcapsules on innate immune responses. Flow cytometry analysis showed that the encapsulation material in the presence or absence of LPS exerted no negative effect on DC viability at 24 h (Fig. 2A). We next investigated the expression of the DC lineage marker CD11c (Fig. 2B), the MHC-II molecule I-Ag7 (Fig. 2C), and the costimulatory molecules CD40 (Fig. 2D), CD80 (Fig. 2E), and CD86 (Fig. 2F). Although LPS stimulation activated DCs and upregulated costimulatory molecules, coculture with (PVPON/TA) microcapsules did not inhibit costimulatory molecule expression. Importantly, we also showed that (PVPON/TA) microcapsules alone did not elicit an increase in DC activation, further corroborating our previous data that (PVPON/TA) microcapsules are nonimmunogenic and nontoxic (26).

FIGURE 2.

(PVPON/TA) microcapsules did not elicit changes in costimulatory molecule expression in LPS-stimulated DCs.

NOD bone marrow–derived DCs were cocultured with TA-capped and PVPON-capped microcapsules either under unstimulated or LPS-stimulated conditions for 24 h. Cell viability (A) was assessed by flow cytometry using a LIVE/DEAD Fixable Near-IR Dead Cell Stain Kit (Invitrogen). DCs were assessed for the expression of CD11c (B), MHC-II I-Ag7 (C), CD40 (D), CD80 (E), or CD86 (F). TA-capped denotes an outer layer of TA, whereas PVPON-capped has an outer layer of PVPON on the surface. Data are representative of three independent experiments.

FIGURE 2.

(PVPON/TA) microcapsules did not elicit changes in costimulatory molecule expression in LPS-stimulated DCs.

NOD bone marrow–derived DCs were cocultured with TA-capped and PVPON-capped microcapsules either under unstimulated or LPS-stimulated conditions for 24 h. Cell viability (A) was assessed by flow cytometry using a LIVE/DEAD Fixable Near-IR Dead Cell Stain Kit (Invitrogen). DCs were assessed for the expression of CD11c (B), MHC-II I-Ag7 (C), CD40 (D), CD80 (E), or CD86 (F). TA-capped denotes an outer layer of TA, whereas PVPON-capped has an outer layer of PVPON on the surface. Data are representative of three independent experiments.

Close modal

Because (PVPON/TA) microcapsules did not decrease costimulatory molecule expression of LPS-stimulated DCs, we wanted to assess if the synthesis of the proinflammatory third signal necessary for T cell adaptive immune maturation was altered (1). LPS-stimulated bone marrow–derived DCs were cocultured with TA-capped or PVPON-capped (PVPON/TA) microcapsules for 24 h. Importantly, we observed no significant increases in proinflammatory mRNA accumulation when DCs were treated with (PVPON/TA) microcapsules alone (Fig. 3), consistent with our previous findings that (PVPON/TA) is nonimmunogenic (27). Conversely, in LPS-stimulated DCs, we observed a downregulation of Tnf by ∼1.7-fold in TA-capped (p < 0.001) and by 1.6-fold in PVPON-capped (p < 0.001) microcapsules compared with LPS stimulation alone (Fig. 3A). Similarly, Cxcl10 mRNA accumulation was reduced 2-fold (p < 0.01) by PVPON-capped microcapsules (Fig. 3B). We also analyzed mRNA accumulation of Il12a and Il12b, which respectively encode for the two subunits of the heterodimer cytokine IL-12p70. We observed a 1.5-fold decrease (p < 0.01) of Il12a by TA-capped microcapsules and a 2.1-fold decrease (p < 0.001) by PVPON-capped microcapsules (Fig. 3C). Similarly, in Il12b, we observed a 2.7-fold decrease (p < 0.0001) of Il12b by TA-capped microcapsules and a 2.1-fold decrease (p < 0.0001) by PVPON-capped microcapsules (Fig. 3D) following LPS stimulation. With the exception of Cxcl10 transcript, which was not decreased by TA-capped microcapsules, there was a demonstrable reduction of cytokine and chemokine mRNA accumulation in both capsule groups.

FIGURE 3.

(PVPON/TA) microcapsules inhibit proinflammatory cytokine and chemokine synthesis in LPS-stimulated DCs.

Bone marrow–derived DCs (106 cells per well) from NOD mice were cocultured with the microcapsules (107 cells per well) in the presence or absence of 1.0 μg/ml of LPS for 24 h quantitative real-time PCR was performed to assess the expression of the proinflammatory genes Tnf (A), Cxcl10 (B), Il12a (C), and Il12b (D). Supernatants were assayed for TNF-α (48 h) (E), CXCL10 (24 h) (F), and IL-12p70 (48 h) (G) by ELISA. TA-capped denotes an outer layer of TA, whereas PVPON-capped microcapsules have an outer layer of PVPON on the surface. Data shown is representative of at least three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 3.

(PVPON/TA) microcapsules inhibit proinflammatory cytokine and chemokine synthesis in LPS-stimulated DCs.

Bone marrow–derived DCs (106 cells per well) from NOD mice were cocultured with the microcapsules (107 cells per well) in the presence or absence of 1.0 μg/ml of LPS for 24 h quantitative real-time PCR was performed to assess the expression of the proinflammatory genes Tnf (A), Cxcl10 (B), Il12a (C), and Il12b (D). Supernatants were assayed for TNF-α (48 h) (E), CXCL10 (24 h) (F), and IL-12p70 (48 h) (G) by ELISA. TA-capped denotes an outer layer of TA, whereas PVPON-capped microcapsules have an outer layer of PVPON on the surface. Data shown is representative of at least three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Close modal

Downregulation of proinflammatory cytokine and chemokine synthesis was confirmed at the protein level by ELISA. Unstimulated DCs did not secrete detectable levels of TNF-α, IL-12p70, or CXCL10, but following LPS stimulation at 24 h, the levels of secreted TNF-α by DC were reduced 1.8-fold by TA-capped microcapsules (p < 0.001) and 1.4-fold by PVPON-capped microcapsules (p < 0.01) (Fig. 3E). We also observed a downregulation of CXCL10 by all microcapsule groups compared with the noncapsule group. Secretion of CXCL10 was curtailed 1.5-fold (p < 0.01) by TA-capped microcapsules and 1.3-fold using PVPON-capped microcapsules (p < 0.01) (Fig. 3F). Finally, we observed a 1.8-fold decrease (p < 0.01) by TA-capped microcapsules and a 1.5-fold decrease (p < 0.05) of secreted IL-12p70 by PVPON-capped microcapsules poststimulation with LPS (Fig. 3G). Although (PVPON/TA) microcapsules did not impede the ability of LPS-stimulated DC to express costimulatory molecules (Fig. 2), we did observe a decrease in proinflammatory cytokine and chemokine synthesis (Fig. 3), essential third signals involved in T cell maturation (1).

Because our (PVPON/TA) microcapsules can be tailored to incorporate peptides or nominal Ags, we next investigated the ability of the microcapsules to abrogate Ag-specific T cell responses against chicken OVA. Although we observed similar decreases in proinflammatory cytokine and chemokine reductions with TA-capped and PVPON-capped microcapsules when cotreated with LPS-stimulated DCs, we determined that microcapsules composed of eight PVPON/TA bilayers and a PVPON-capped outer shell would maximize the delivery of available TA in vivo while not compromising capsule integrity due to PVPON hydrophilicity, which can prevent protein adsorption (32, 33). The outer PVPON layer may shield the TA molecules from extracellular redox reactions before phagocytosis by DCs. To investigate the ability of encapsulated Ags to elicit Ag-specific hyporesponsiveness, we synthesized (PVPON/TA) microcapsules with chicken OVA incorporated into the seventh bilayer, termed OVA.1, with the following structure: (TA/PVPON)6(TA/OVA)(TA/PVPON). We also synthesized (PVPON/TA) microcapsules with two layers of OVA termed OVA.2 and containing the following structure (TA/PVPON)5(TA/OVA)2(TA/PVPON). The quantitation of OVA concentration was initially determined by UV/vis spectroscopy (31) and confirmed on each freshly synthesized microcapsule batch by Western blot analysis before use in any downstream assays. OVA was efficiently encapsulated within both groups of microcapsules, with OVA.2 containing an increase in OVA compared with OVA.1 (Fig. 4A). Based on the Western blot results, the amount of OVA within OVA.1 and OVA.2 was normalized to ensure that equivalent concentrations of OVA (40 μg/ml) were used to properly stimulate OT-II splenocytes in a primary recall assay.

FIGURE 4.

OVA encapsulated with (PVPON/TA) microcapsules inhibit CD4 T cell activation.

To generate microcapsules containing OVA within the shell, PVPON was substituted for OVA at bilayer 7 (OVA.1) and bilayers 6 and 7 (OVA.2) during the LbL assembly to achieve the desired OVA concentration as shown in Fig. 1. Western blot analysis shows detectable OVA protein (42 kDa) in (PVPON/TA/OVA) OVA.1 and OVA.2 microcapsules. Densitometry analysis shows relative protein containing in microcapsules compared with 10 μg of chicken OVA–loading control (A). OT-II bone marrow–derived DCs were cocultured with (PVPON/TA) or (PVPON/TA/OVA) microcapsules, and stimulated with PMA/I (100 ng/ml PMA and 1 μg/ml ionomycin) to induce luminol oxidation (B). The (PVPON/TA) microcapsules containing either (PVPON/TA) alone without OVA (PVPON-capped), one layer (OVA.1), or two layers (OVA.2) of OVA encapsulated with (PVPON/TA) were cocultured with OT-II splenocytes. Flow cytometric analysis of DCs were assessed by MHC-II/I-Ab (72 h) (C), CD80 (72 h) (D), and CD86 (72 h) (E). CD4 T cells were analyzed by CD25 (72 h) (F), CD69 (72 h) (G), CD44+ CD62L effector/effector memory T cells (72 h) (H), CD44+ CD62L+ central memory T cells (72 h) (I), CTLA-4 (48 h) (J), and PD-1 (72 h) (K). Data shown represent three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. NS, not significant.

FIGURE 4.

OVA encapsulated with (PVPON/TA) microcapsules inhibit CD4 T cell activation.

To generate microcapsules containing OVA within the shell, PVPON was substituted for OVA at bilayer 7 (OVA.1) and bilayers 6 and 7 (OVA.2) during the LbL assembly to achieve the desired OVA concentration as shown in Fig. 1. Western blot analysis shows detectable OVA protein (42 kDa) in (PVPON/TA/OVA) OVA.1 and OVA.2 microcapsules. Densitometry analysis shows relative protein containing in microcapsules compared with 10 μg of chicken OVA–loading control (A). OT-II bone marrow–derived DCs were cocultured with (PVPON/TA) or (PVPON/TA/OVA) microcapsules, and stimulated with PMA/I (100 ng/ml PMA and 1 μg/ml ionomycin) to induce luminol oxidation (B). The (PVPON/TA) microcapsules containing either (PVPON/TA) alone without OVA (PVPON-capped), one layer (OVA.1), or two layers (OVA.2) of OVA encapsulated with (PVPON/TA) were cocultured with OT-II splenocytes. Flow cytometric analysis of DCs were assessed by MHC-II/I-Ab (72 h) (C), CD80 (72 h) (D), and CD86 (72 h) (E). CD4 T cells were analyzed by CD25 (72 h) (F), CD69 (72 h) (G), CD44+ CD62L effector/effector memory T cells (72 h) (H), CD44+ CD62L+ central memory T cells (72 h) (I), CTLA-4 (48 h) (J), and PD-1 (72 h) (K). Data shown represent three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. NS, not significant.

Close modal

We next determined the efficacy of the (PVPON/TA) microcapsules to scavenge ROS synthesis by measuring the oxidation of luminol by chemiluminescence. Bone marrow–derived DCs from OT-II mice were cocultured with OVA-free PVPON-capped, OVA.1, or OVA.2 OVA-containing microcapsules and stimulated with PMA/I to induce a respiratory burst (Fig. 4B). We observed a significant (p < 0.0001) decrease in oxidation for all microcapsule groups. PVPON-capped capsules elicited the greatest decrease (3.25-fold), followed by OVA.1 (2.75-fold) and OVA.2 (2.65-fold), whereas the differences between microcapsule groups was NS. These data show that the (PVPON/TA) microcapsules with or without OVA can successfully dissipate H2O2 synthesis in PMA/I–stimulated DCs.

To determine if OVA.1 and OVA.2 microcapsules could elicit Ag-specific hyporesponsiveness, we isolated splenocytes from OT-II mice, performed a primary recall assay with OVA323–339, whole-chicken OVA, PVPON-capped microcapsules (containing no Ag), OVA.1 microcapsules, or OVA.2 microcapsules, and immunophenotyped cells by flow cytometry. To determine the effect of the microcapsules on Ag presentation, we first investigated the status of CD11c+ DCs. We observed significant ∼2-fold reductions in MHC-II expression on CD11c+ DCs with OVA.1 (p < 0.01) and OVA.2 (p < 0.05) microcapsule treatment compared with OVA protein, approaching levels of unstimulated cells at 48 h (Fig. 4C). Although no differences in the levels of the costimulatory molecule CD40 were detected (data not shown), we observed an increase in CD80 expression in CD11c+ cells cocultured with PVPON-capped (p < 0.01, 2.2-fold), OVA.1 (p < 0.01, 2.0-fold), and OVA.2 (NS, 1.65-fold) microcapsules (Fig. 4D). Interestingly, we did not observe an increase in CD86 expression in CD11c+ cells (Fig. 4E), but PVPON-capped microcapsules elicited a decrease (p < 0.05, 1.9-fold). We next determined the activation status of CD4 T cells and showed that the expression of the T cell activation marker CD25 was inhibited (p < 0.0001) (Fig. 4F) by both OVA.1 (2.0-fold) and OVA.2 (1.9-fold) microcapsules, and CD69 (p < 0.01) (Fig. 4G) was also decreased 1.8-fold by OVA.1 and OVA.2 microcapsules. We also investigated the effects of OVA.1 and OVA.2 on the differentiation of CD4 T cells to effector/effector memory (CD44+ CD62L) and central memory (CD44+ CD62L+) populations. No significant changes in either effector/effector memory (Fig. 4H) or naive (data not shown) populations were detected, but we observed a significant decrease in the central memory population by OVA.1 (p < 0.0001) and OVA.2 (p < 0.01) microcapsules compared with OVA alone (Fig. 4I). Next, we wanted to investigate if the expression of immune inhibitory receptors was induced with our TA-containing microcapsules to partly explain the decreased Ag-specific immune response. We observed that CD4+ T cells upregulated CTLA-4 when cocultured with OVA.1 (3-fold, p < 0.01) or OVA-free PVPON-capped (2.6-fiold, p < 0.05) microcapsules (Fig. 4J). Finally, we observed an upregulation of the immune checkpoint marker PD-1 by OVA.1 (1.9-fold, p < 0.0001), OVA.2 (1.4-fold, p < 0.05), and PVPON-capped (1.8-fold, p < 0.001) microcapsules compared with OVA (Fig. 4K). Taken together, these data show that the microcapsules inhibit the activation of Ag-specific CD4 T cells and can induce CTLA-4 and PD-1 inhibitory receptor expression.

To determine if OVA.1 and OVA.2 microcapsules could diminish Th1 proinflammatory effector responses, supernatants from the primary recall assays were analyzed by ELISA for IL-2, IFN-γ, TNF-α, and the chemokine CXCL10. Stimulation of OT-II splenocytes with OVA323–339 peptide served as a positive control to indicate that our CD4+ T cells were Ag-specific. As expected, stimulation with OVA323–339 led to an increase in cytokine synthesis. Importantly, treatment with PVPON-capped microcapsules, which contain no OVA protein, elicited no response similarly to no Ag negative controls. Treatment with equivalent concentrations of OVA, OVA.1, and OVA.2 elicited a cytokine response from OT-II splenocytes. OVA323–339– and OVA-stimulated splenocytes secreted IL-2 at 72 h poststimulation, but IL-2 secretion was barely detectable by cells stimulated with either OVA.1 or OVA.2 microcapsules (p < 0.05) (Fig. 5A). By 72 h, IFN-γ synthesis was significantly dampened by both OVA.1 (p < 0.01, 3.1-fold) and OVA.2 (p < 0.05, 2.0-fold) microcapsules (Fig. 5B). The proinflammatory chemokine CXCL10 (Fig. 5C) was suppressed to nearly undetectable levels in both OVA.1 (p < 0.001, not detected) and OVA.2 (p < 0.001, 17.5-fold) microcapsules at 48 h. Finally, TNF-α was significantly suppressed (p < 0.01, 2.0-fold) by OVA.1 microcapsules, but not by OVA.2 microcapsules at 72 h (Fig. 5D).

FIGURE 5.

OVA encapsulated with (PVPON/TA) microcapsules inhibit Th1 effector responses.

Cell culture supernatants from OT-II splenoctye-microcapsule cocultures with ova peptide, OVA, PVPON-capped, OVA.1, and OVA.2 were collected and the synthesis of IL-2 (A), IFN-γ (B), CXCL10 (C), and TNF-α (D) was analyzed by ELISA. Data are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 ****p < 0.0001.

FIGURE 5.

OVA encapsulated with (PVPON/TA) microcapsules inhibit Th1 effector responses.

Cell culture supernatants from OT-II splenoctye-microcapsule cocultures with ova peptide, OVA, PVPON-capped, OVA.1, and OVA.2 were collected and the synthesis of IL-2 (A), IFN-γ (B), CXCL10 (C), and TNF-α (D) was analyzed by ELISA. Data are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 ****p < 0.0001.

Close modal

To determine if in vivo immunization with OVA.1 and OVA.2 could suppress Ag-specific immune responses, OT-II mice were s.c. injected with 100 μg of OVA or an equimolar amount of OVA.1 and OVA.2 microcapsules in the presence of CFA. At 7 d postimmunization, the draining inguinal LNs and spleens were harvested, homogenized to single-cell suspensions, immunophenotyped by flow cytometry, and restimulated with OVA in an Ag-recall assay. Initial immunophenotyping of splenocytes did not show any significant differences in cell viability (Fig. 6A), frequencies of CD4+ T cells (Fig. 6B), CD25+ CD4+ T cells (Fig. 6C), CD69+ CD4+ T cells (Fig. 6D), and MHC-II+ CD11c+ DCs (Fig. 6E) between OVA-, OVA.1-, and OVA.2-immunized groups. However, it should also be noted that both OVA, OVA.1, and OVA.2 elicited a response above PVPON-capped microcapsules containing no Ag. We also observed that OVA.1-immunized mice exhibited a 1.7-fold (NS) decrease in central memory (CD4+ CD44+ CD62L+) T cells (Fig. 6F) and a 1.5-fold decrease (p < 0.05) in effector/effector memory (CD4+ CD44+ CD62L) T cells (Fig. 6G) compared with OVA alone. Phenotyping of cells isolated from inguinal LN showed similar trends for cell viability, overall CD4+ populations, CD25 and CD69 activation status, MHC-II, and memory/effector status, but these were not statistically significant (data not shown). To confirm that OVA.1- and OVA.2-immunized mice can induce Ag-specific hyporesponsiveness, we performed an Ag-recall assay with LN cells from immunized mice by restimulating with 125 μg/ml of OVA. In agreement with the immunophenotyping results (Fig. 6), effector cytokine secretion was diminished from OVA-stimulated LN cells from mice immunized with OVA.1 and OVA.2 microcapsules versus OVA protein. We observed ∼25-fold decrease in IFN-γ (p < 0.0001) by both OVA.1 and OVA.2 (Fig. 7A), as well as a 4.4-fold decrease in TNF-α from OVA.1 microcapsules (Fig. 7B). We also observed a 1.6-fold (p < 0.05) decrease in the chemokine CCL3 (MIP-1α) by OVA.2 (Fig. 7C) and a significant decrease in CCL5 (RANTES) by both OVA.1 (2.0-fold, p < 0.01) and OVA.2 (1.7-fold, p < 0.05) (Fig. 7D). Because CCL3 is important for DC trafficking to LN and CCL5 can coordinate effector and memory T cell migration to peripheral LN (34), these data show that OVA.1 and OVA.2 microcapsules can decrease proinflammatory T cell effector responses and may be beneficial in eliciting Ag-specific hyporesponsiveness in the treatment of T cell–mediated inflammatory diseases.

FIGURE 6.

Immunization with OVA.1 and OVA.2 microcapsules decreases the number of effector and central memory T cells in vivo.

Splenocytes from immunized OT-II mice were isolated 7 d postimmunization and analyzed by flow cytometry. Flow cytometric analysis of cell viability (A), frequency of CD4 T cells (B), and activation status of CD4+ T cells by CD25 (C), CD69 (D), CD44+ CD62L+ central memory T cells (F), and CD44+ CD62L effector/effector memory T cells (G). Frequency of CD11c+ cells expressing I-Ab (E). Data shown represent three independent experiments. *p < 0.05.

FIGURE 6.

Immunization with OVA.1 and OVA.2 microcapsules decreases the number of effector and central memory T cells in vivo.

Splenocytes from immunized OT-II mice were isolated 7 d postimmunization and analyzed by flow cytometry. Flow cytometric analysis of cell viability (A), frequency of CD4 T cells (B), and activation status of CD4+ T cells by CD25 (C), CD69 (D), CD44+ CD62L+ central memory T cells (F), and CD44+ CD62L effector/effector memory T cells (G). Frequency of CD11c+ cells expressing I-Ab (E). Data shown represent three independent experiments. *p < 0.05.

Close modal
FIGURE 7.

Immunization with OVA.1 and OVA.2 microcapsules suppresses effector cytokine and chemokine production from LN cells.

OT-II LN from OVA-immunized mice were either unstimulated or stimulated with 125 μg/ml OVA. Cell culture supernatants were collected to detect cytokine synthesis by ELISA for IFN-γ [48 h, (A)], TNF-α [48 h, (B)], CCL3 [48 h, (C)], and CCL5 [48 h, (D)] following OVA stimulation. Data represent four individual experiments. *p < 0.05, **p < 0.01, ****p < 0.0001.

FIGURE 7.

Immunization with OVA.1 and OVA.2 microcapsules suppresses effector cytokine and chemokine production from LN cells.

OT-II LN from OVA-immunized mice were either unstimulated or stimulated with 125 μg/ml OVA. Cell culture supernatants were collected to detect cytokine synthesis by ELISA for IFN-γ [48 h, (A)], TNF-α [48 h, (B)], CCL3 [48 h, (C)], and CCL5 [48 h, (D)] following OVA stimulation. Data represent four individual experiments. *p < 0.05, **p < 0.01, ****p < 0.0001.

Close modal

Effective treatment of T cell–mediated inflammatory diseases continues to present a major clinical hurdle. Use of nonsteroidal anti-inflammatory drugs, corticosteroids, and other anti-inflammatory drugs only mask symptoms while not treating the underlying pathological condition. Global immunosuppression is unattractive as it leaves the patient vulnerable to microbial infections. Ag-based strategies may be beneficial in decreasing aberrant T cell responses. We and others have previously shown that disruption of the innate immune-mediated proinflammatory third signal results in CD4 and CD8 T cell hyporesponsiveness and a decrease in Ag-specific effector function (1, 6, 35). Therefore, disruption of this signal represents an attractive strategy to suppress T cell–mediated autoimmune diseases including T1D.

The goal of this study was to investigate an Ag-specific approach to abrogating T cell activation. We tested the hypothesis that encapsulation of Ags within an antioxidant-containing polymer microcapsule would blunt the activation of DCs, and also inhibit Ag-specific T cell responses by perturbing the proinflammatory third signal of T cell activation. Our previous data showed that (PVPON/TA) microcapsules can inhibit ROS generation by PMA-stimulated macrophages in vitro (2628). In this study we showed that (PVPON/TA) microcapsules can be phagocytosed by DCs and cotreatment with LPS-activated NOD bone marrow–derived DCs did not affect MHC-II, CD40, CD80, or CD86 surface expression. However, when we examined the generation of the proinflammatory cytokines TNF-α and IL-12p70, as well as the chemokine CXCL10, we observed a robust decrease in these proinflammatory molecules. Treatment with (PVPON/TA) microcapsules suggests that the innate immune response of DCs is altered and may elicit Ag-specific hyporesponsiveness when presenting antigenic peptides to T cells. Importantly, DC-derived IL-12p70 is necessary for the differentiation of naive T cells to Th1 effector T cells, which is the effector response that mediates most autoimmune diseases including T1D (36). These findings also corroborate our previous observations that the production of proinflammatory cytokines by activated NOD macrophages was decreased when cocultured with (PVPON/TA) microcapsules (28).

In the current study, eight (PVPON/TA) bilayers with TA on the outer layer (TA-capped) or with an additional PVPON outer layer (PVPON-capped) was determined to be equally effective at inhibiting innate immune activation of LPS-stimulated DCs. Interestingly, this is contrary to our previous observations, in which microcapsules with exposed surface TA were more efficacious in dissipating free radical synthesis and IFN-γ by macrophages and splenic T cells, respectively (26, 28). The enhanced anti-inflammatory response with TA-capped and PVPON-capped microcapsules may be due to a 2-fold increase of TA being delivered compared with the microcapsules used in our previous studies. The additional PVPON outer layer in PVPON-capped microcapsules may help facilitate uptake and decrease endosomal degradation, as opposed to exposed TA being available to sequester extracellular ROS.

Based on our promising results with DCs, and owing to their specialized role as APCs to link both the innate and adaptive arms of the immune system, we incorporated OVA into the microcapsule bilayers to determine if Ag-specific immune responses with the CD4 T cell–transgenic OT-II mouse model could be inhibited. We found that when cocultured with OT-II splenocytes, OVA.1 and OVA.2 microcapsules were able to blunt the activation of CD4 T cells. This was evident by the decrease in expression of CD25 and CD69 T cell activation markers, decreases in central memory T cells, and an upregulation of the immune checkpoint markers CTLA-4 and PD-1. The effector function of CD4 T cells was also mitigated by OVA.1 and OVA.2 microcapsules, as observed by the inhibition of Th1 effector cytokines IL-2, IFN-γ, and TNF-α and the chemokine CXCL10. Although these in vitro data support our hypothesis, we wanted to define the in vivo efficacy of OVA.1 and OVA.2 microcapsules to suppress Ag-specific immune responses. We showed that CD4 T cells from mice immunized with OVA.1 and OVA.2 microcapsules had a lower proportion of CD44+ CD62L effector/effector memory T cells by flow cytometry. Most importantly, when subjected to an Ag-recall assay, CD4 T cells from the draining LNs secreted less IFN-γ, TNF-α, CCL3, and CCL5 following immunization with OVA.1 and OVA.2 microcapsules. IFN-γ and TNF-α are involved in pancreatic β cell destruction in T1D (37, 38), whereas CCL3 and CCL5 are critical chemokines necessary for coordinating DC trafficking to peripheral LN and engagement with CD4 T cells (39, 40). These results provide evidence that encapsulation of autoantigens/peptides with (PVPON/TA) shells may be efficacious in suppressing autoreactive T cell responses and delaying autoimmunity.

We and others have previously shown the importance of ROS synthesis in the coordination of adaptive immune responses (1, 4, 5, 7, 41). In the absence of ROS, innate-derived immune signals are blunted resulting in abrogated maturation of adaptive immune effector response and diminished memory function (7). Furthermore, decreasing NADPH oxidase (NOX)–derived superoxide in autoimmune-prone NOD mice can delay the development of autoimmunity by inhibiting both innate (4, 42) and adaptive (7, 9) immune effector responses. Our prior studies highlighted the ability of (PVPON/TA) microcapsules and islets encapsulated in (PVPON/TA) coatings to ablate local generation of ROS species and inhibit autoreactive T cell responses (2628). Coupled with the knowledge that TNF-α is redox-regulated (43) and intracellular ROS generation is elevated during Ag recognition (44), delivery of (PVPON/TA/Ag) microcapsules can effectively ablate the proinflammatory third signal necessary for T cell activation. We observed that OVA.1 was more effective than OVA.2 to inhibit TNF-α synthesis in an OT-II primary recall assay and in an Ag-recall assay with LN cells from immunized OT-II mice. Although the amount of OVA and TA was normalized for all T cell assays, the configuration of OVA.2 microcapsules containing one layer of TA sandwiched between two layers of OVA may affect the ability of TA to effectively decrease ROS synthesis in contrast to OVA.1 microcapsules. In addition, the ability of OVA.2 to inhibit redox-sensitive signaling pathways involved in TNF-α synthesis including NF-κB and MAPK may not be as effective as OVA.1 treatment (45). Future studies will increase the concentration of TA within the microcapsules to determine if an increase in antioxidant defenses can further abrogate Ag-specific T cell responses. Alternatively, we recently demonstrated that (PVPON/TA) coatings conjugated with a manganese metalloporphyrin antioxidant were effective in decreasing ROS synthesis and display a broader range of antioxidant activity than TA (46). We can also conjugate antigenic peptides to manganese metalloporphyrin (PVPON/TA) coatings and determine if these new capsules are more effective in eliciting Ag-specific hyporesponsiveness.

We observed increased expression of the T cell checkpoint receptors CTLA-4 and PD-1 by splenic CD4 T cells following coculture with OVA-free PVPON-capped and OVA.1 microcapsules. Although the differentiation of Tregs was not enhanced with OVA.1 or OVA.2 treatment (data not shown), our results provide further support that an increase in polyphenolic antioxidants can increase CTLA-4 expression (47). Interestingly, we observed a decrease in MHC-II+, no change with CD86+, and an increase in CD80+ expression in splenic DCs following microcapsule coculture. This would seem feasible, given that CD80 has higher affinity and avidity for CTLA-4 than CD86, and that CD80 and CTLA-4 interactions are preferential in regulating self-tolerance than CD86 and CTLA-4 interactions (48, 49). It is plausible that antioxidants are not only efficient in decreasing proinflammatory cues for T cell activation, but may also induce the expression of inhibitory receptors such as CTLA-4. In support of this hypothesis, previous results from Delmastro et al. (50) demonstrated that treatment of NOD mice with a catalytic antioxidant was effective in delaying spontaneous T1D progression. One mechanism of protection mediated by catalytic antioxidant treatment was an increase in surface expression of LAG-3, an inhibitory receptor that interacts with MHC-II to decrease inflammatory responses from CD4 T cells (51). Treatment of BDC-2.5 splenocytes with the catalytic antioxidant decreased Th1 cytokine responses and increased LAG-3 expression partly due to redox-regulation of the metalloprotease TNF-α–converting enzyme (TACE) involved in regulating LAG-3. Future studies are warranted to further understand how changes in redox status and treatment with TA can influence the expression of inhibitory receptors such as CTLA-4.

In conclusion, this study shows that combined delivery of a nominal Ag with TA, a broad acting antioxidant, can blunt DC-derived proinflammatory responses and inhibit adaptive immune effector Th1 responses from Ag-specific CD4 T cells. Methods to tolerize DCs and/or promote an anti-inflammatory DC response represent an attractive strategy to inhibit T cell–mediated inflammatory diseases and autoreactive T cell responses (52, 53). Our future studies will encapsulate T1D autoantigens such as insulin (InsB:9–23, InsB:15–23, and InsA:14–20), hybrid insulin peptides (54, 55), glutamic acid decarboxylase-65 (GAD65) peptides, and islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP) peptides. By conjugating with antigenic peptides rather than full-length proteins, we can bypass Ag processing, target autoreactive T cells, and determine if Ag-specific hyporesponsiveness can be generated. Reprogramming DCs to decrease proinflammatory effector responses of autoreactive T cells with (PVPON/TA)-encapsulated Ags, may be a novel therapy to delay pancreatic β cell destruction in T1D, as well as aberrant T cell activation in other disease contexts.

We thank Jessie Barra, Samuel Blum, Katie Heath, and Dr. Jared Taylor for careful reading of this manuscript.

This work was supported by a National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases R01 award (DK099550) (to H.M.T.), a Juvenile Diabetes Research Foundation (JDRF) research award (1-SRA-2015-42-A-N) (to H.M.T.), a JDRF postdoctoral fellowship award (3-PDF-2018-592-A-N) (to J.M.F.), and National Science Foundation Division of Materials Research Award 1608728 (to E.K.).

Abbreviations used in this article:

DC

dendritic cell

LbL

layer-by-layer

LN

lymph node

MHC-II

MHC class II

PEI

poly(ethylene imine)

PVPON

poly(N-vinylpyrrolidone)

ROS

reactive oxygen species

TA

tannic acid

T1D

type 1 diabetes

Treg

regulatory T cell.

The online version of this article contains supplemental material.

1
Tse
H. M.
,
M. J.
Milton
,
S.
Schreiner
,
J. L.
Profozich
,
M.
Trucco
,
J. D.
Piganelli
.
2007
.
Disruption of innate-mediated proinflammatory cytokine and reactive oxygen species third signal leads to antigen-specific hyporesponsiveness.
J. Immunol.
178
:
908
917
.
2
Curtsinger
J. M.
,
M. F.
Mescher
.
2010
.
Inflammatory cytokines as a third signal for T cell activation.
Curr. Opin. Immunol.
22
:
333
340
.
3
Sklavos
M. M.
,
H. M.
Tse
,
J. D.
Piganelli
.
2008
.
Redox modulation inhibits CD8 T cell effector function.
Free Radic. Biol. Med.
45
:
1477
1486
.
4
Thayer
T. C.
,
M.
Delano
,
C.
Liu
,
J.
Chen
,
L. E.
Padgett
,
H. M.
Tse
,
M.
Annamali
,
J. D.
Piganelli
,
L. L.
Moldawer
,
C. E.
Mathews
.
2011
.
Superoxide production by macrophages and T cells is critical for the induction of autoreactivity and type 1 diabetes.
Diabetes
60
:
2144
2151
.
5
Tse
H. M.
,
M. J.
Milton
,
J. D.
Piganelli
.
2004
.
Mechanistic analysis of the immunomodulatory effects of a catalytic antioxidant on antigen-presenting cells: implication for their use in targeting oxidation-reduction reactions in innate immunity.
Free Radic. Biol. Med.
36
:
233
247
.
6
Padgett
L. E.
,
H. M.
Tse
.
2016
.
NADPH oxidase-derived superoxide provides a third signal for CD4 T cell effector responses.
J. Immunol.
197
:
1733
1742
.
7
Tse
H. M.
,
T. C.
Thayer
,
C.
Steele
,
C. M.
Cuda
,
L.
Morel
,
J. D.
Piganelli
,
C. E.
Mathews
.
2010
.
NADPH oxidase deficiency regulates Th lineage commitment and modulates autoimmunity.
J. Immunol.
185
:
5247
5258
.
8
Delmastro-Greenwood
M. M.
,
H. M.
Tse
,
J. D.
Piganelli
.
2014
.
Effects of metalloporphyrins on reducing inflammation and autoimmunity.
Antioxid. Redox Signal.
20
:
2465
2477
.
9
Liu
C.
,
R. L.
Whitener
,
A.
Lin
,
Y.
Xu
,
J.
Chen
,
A.
Savinov
,
J. W.
Leiding
,
M. A.
Wallet
,
C. E.
Mathews
.
2019
.
Neutrophil cytosolic factor 1 in dendritic cells promotes autoreactive CD8+ T cell activation via cross-presentation in type 1 diabetes.
Front. Immunol.
10
:
952
.
10
Luo
X.
,
K. C.
Herold
,
S. D.
Miller
.
2010
.
Immunotherapy of type 1 diabetes: where are we and where should we be going?
Immunity
32
:
488
499
.
11
Herold
K. C.
,
D. A.
Vignali
,
A.
Cooke
,
J. A.
Bluestone
.
2013
.
Type 1 diabetes: translating mechanistic observations into effective clinical outcomes.
Nat. Rev. Immunol.
13
:
243
256
.
12
Battaglia
M.
,
M. G.
Roncarolo
.
2011
.
Immune intervention with T regulatory cells: past lessons and future perspectives for type 1 diabetes.
Semin. Immunol.
23
:
182
194
.
13
Bluestone
J. A.
,
J. H.
Buckner
,
M.
Fitch
,
S. E.
Gitelman
,
S.
Gupta
,
M. K.
Hellerstein
,
K. C.
Herold
,
A.
Lares
,
M. R.
Lee
,
K.
Li
, et al
.
2015
.
Type 1 diabetes immunotherapy using polyclonal regulatory T cells.
Sci. Transl. Med.
7
: 315ra189.
14
Kroger
C. J.
,
M.
Clark
,
Q.
Ke
,
R. M.
Tisch
.
2018
.
Therapies to suppress β cell autoimmunity in type 1 diabetes.
Front. Immunol.
9
:
1891
.
15
Liu
K.
,
T.
Iyoda
,
M.
Saternus
,
Y.
Kimura
,
K.
Inaba
,
R. M.
Steinman
.
2002
.
Immune tolerance after delivery of dying cells to dendritic cells in situ.
J. Exp. Med.
196
:
1091
1097
.
16
Price
J. D.
,
N. M.
Beauchamp
,
G.
Rahir
,
Y.
Zhao
,
C. C.
Rieger
,
A. W.
Lau-Kilby
,
K. V.
Tarbell
.
2014
.
CD8+ dendritic cell-mediated tolerance of autoreactive CD4+ T cells is deficient in NOD mice and can be corrected by blocking CD40L.
J. Leukoc. Biol.
95
:
325
336
.
17
Mellman
I.
,
R. M.
Steinman
.
2001
.
Dendritic cells: specialized and regulated antigen processing machines.
Cell
106
:
255
258
.
18
Mehling
A.
,
S.
Beissert
.
2003
.
Dendritic cells under investigation in autoimmune disease.
Crit. Rev. Biochem. Mol. Biol.
38
:
1
21
.
19
Tisch
R.
,
B.
Wang
.
2009
.
Role of plasmacytoid dendritic cells in type 1 diabetes: friend or foe?
Diabetes
58
:
12
13
.
20
Poligone
B.
,
D. J.
Weaver
Jr.
,
P.
Sen
,
A. S.
Baldwin
Jr
,
R.
Tisch
.
2002
.
Elevated NF-kappaB activation in nonobese diabetic mouse dendritic cells results in enhanced APC function.
J. Immunol.
168
:
188
196
.
21
Sen
P.
,
S.
Bhattacharyya
,
M.
Wallet
,
C. P.
Wong
,
B.
Poligone
,
M.
Sen
,
A. S.
Baldwin
Jr
,
R.
Tisch
.
2003
.
NF-kappa B hyperactivation has differential effects on the APC function of nonobese diabetic mouse macrophages.
J. Immunol.
170
:
1770
1780
.
22
Piganelli
J. D.
,
T.
Martin
,
K.
Haskins
.
1998
.
Splenic macrophages from the NOD mouse are defective in the ability to present antigen.
Diabetes
47
:
1212
1218
.
23
Iyoda
T.
,
S.
Shimoyama
,
K.
Liu
,
Y.
Omatsu
,
Y.
Akiyama
,
Y.
Maeda
,
K.
Takahara
,
R. M.
Steinman
,
K.
Inaba
.
2002
.
The CD8+ dendritic cell subset selectively endocytoses dying cells in culture and in vivo.
J. Exp. Med.
195
:
1289
1302
.
24
Tarbell
K. V.
,
S.
Yamazaki
,
K.
Olson
,
P.
Toy
,
R. M.
Steinman
.
2004
.
CD25+ CD4+ T cells, expanded with dendritic cells presenting a single autoantigenic peptide, suppress autoimmune diabetes.
J. Exp. Med.
199
:
1467
1477
.
25
Yeste
A.
,
M.
Nadeau
,
E. J.
Burns
,
H. L.
Weiner
,
F. J.
Quintana
.
2012
.
Nanoparticle-mediated codelivery of myelin antigen and a tolerogenic small molecule suppresses experimental autoimmune encephalomyelitis.
Proc. Natl. Acad. Sci. USA
109
:
11270
11275
.
26
Kozlovskaya
V.
,
B.
Xue
,
W.
Lei
,
L. E.
Padgett
,
H. M.
Tse
,
E.
Kharlampieva
.
2015
.
Hydrogen-bonded multilayers of tannic acid as mediators of T-cell immunity.
Adv. Healthc. Mater.
4
:
686
694
.
27
Kozlovskaya
V.
,
O.
Zavgorodnya
,
Y.
Chen
,
K.
Ellis
,
H. M.
Tse
,
W.
Cui
,
J. A.
Thompson
,
E.
Kharlampieva
.
2012
.
Ultrathin polymeric coatings based on hydrogen-bonded polyphenol for protection of pancreatic islet cells.
Adv. Funct. Mater.
22
:
3389
3398
.
28
Pham-Hua
D.
,
L. E.
Padgett
,
B.
Xue
,
B.
Anderson
,
M.
Zeiger
,
J. M.
Barra
,
M.
Bethea
,
C. S.
Hunter
,
V.
Kozlovskaya
,
E.
Kharlampieva
,
H. M.
Tse
.
2017
.
Islet encapsulation with polyphenol coatings decreases pro-inflammatory chemokine synthesis and T cell trafficking.
Biomaterials
128
:
19
32
.
29
Barnden
M. J.
,
J.
Allison
,
W. R.
Heath
,
F. R.
Carbone
.
1998
.
Defective TCR expression in transgenic mice constructed using cDNA-based alpha- and beta-chain genes under the control of heterologous regulatory elements.
Immunol. Cell Biol.
76
:
34
40
.
30
Johansson
A.
,
C.
Dahlgren
.
1989
.
Characterization of the luminol-amplified light-generating reaction induced in human monocytes.
J. Leukoc. Biol.
45
:
444
451
.
31
Liu
F.
,
V.
Kozlovskaya
,
O.
Zavgorodnya
,
C.
Martinez-Lopez
,
S.
Catledge
,
E.
Kharlampieva
.
2014
.
Encapsulation of anticancer drug by hydrogen-bonded multilayers of tannic acid.
Soft Matter
10
:
9237
9247
.
32
Gaucher
G.
,
K.
Asahina
,
J.
Wang
,
J. C.
Leroux
.
2009
.
Effect of poly(N-vinyl-pyrrolidone)-block-poly(D,L-lactide) as coating agent on the opsonization, phagocytosis, and pharmacokinetics of biodegradable nanoparticles.
Biomacromolecules
10
:
408
416
.
33
Andersen
T. E.
,
Y.
Palarasah
,
M. O.
Skjødt
,
R.
Ogaki
,
M.
Benter
,
M.
Alei
,
H. J.
Kolmos
,
C.
Koch
,
P.
Kingshott
.
2011
.
Decreased material-activation of the complement system using low-energy plasma polymerized poly(vinyl pyrrolidone) coatings.
Biomaterials
32
:
4481
4488
.
34
Schaller
T. H.
,
K. A.
Batich
,
C. M.
Suryadevara
,
R.
Desai
,
J. H.
Sampson
.
2017
.
Chemokines as adjuvants for immunotherapy: implications for immune activation with CCL3.
Expert Rev. Clin. Immunol.
13
:
1049
1060
.
35
Curtsinger
J. M.
,
C. M.
Johnson
,
M. F.
Mescher
.
2003
.
CD8 T cell clonal expansion and development of effector function require prolonged exposure to antigen, costimulation, and signal 3 cytokine.
J. Immunol.
171
:
5165
5171
.
36
Bending
D.
,
H.
De la Peña
,
M.
Veldhoen
,
J. M.
Phillips
,
C.
Uyttenhove
,
B.
Stockinger
,
A.
Cooke
.
2009
.
Highly purified Th17 cells from BDC2.5NOD mice convert into Th1-like cells in NOD/SCID recipient mice.
J. Clin. Invest.
119
:
565
572
.
37
Calderon
B.
,
A.
Suri
,
X. O.
Pan
,
J. C.
Mills
,
E. R.
Unanue
.
2008
.
IFN-gamma-dependent regulatory circuits in immune inflammation highlighted in diabetes.
J. Immunol.
181
:
6964
6974
.
38
von Herrath
M. G.
,
M. B.
Oldstone
.
1997
.
Interferon-gamma is essential for destruction of beta cells and development of insulin-dependent diabetes mellitus.
J. Exp. Med.
185
:
531
539
.
39
Castellino
F.
,
A. Y.
Huang
,
G.
Altan-Bonnet
,
S.
Stoll
,
C.
Scheinecker
,
R. N.
Germain
.
2006
.
Chemokines enhance immunity by guiding naive CD8+ T cells to sites of CD4+ T cell-dendritic cell interaction.
Nature
440
:
890
895
.
40
Castellino
F.
,
R. N.
Germain
.
2007
.
Chemokine-guided CD4+ T cell help enhances generation of IL-6RalphahighIL-7Ralpha high prememory CD8+ T cells.
J. Immunol.
178
:
778
787
.
41
Piganelli
J. D.
,
S. C.
Flores
,
C.
Cruz
,
J.
Koepp
,
I.
Batinic-Haberle
,
J.
Crapo
,
B.
Day
,
R.
Kachadourian
,
R.
Young
,
B.
Bradley
,
K.
Haskins
.
2002
.
A metalloporphyrin-based superoxide dismutase mimic inhibits adoptive transfer of autoimmune diabetes by a diabetogenic T-cell clone.
Diabetes
51
:
347
355
.
42
Padgett
L. E.
,
A. R.
Burg
,
W.
Lei
,
H. M.
Tse
.
2015
.
Loss of NADPH oxidase-derived superoxide skews macrophage phenotypes to delay type 1 diabetes.
Diabetes
64
:
937
946
.
43
Han
D.
,
M. D.
Ybanez
,
S.
Ahmadi
,
K.
Yeh
,
N.
Kaplowitz
.
2009
.
Redox regulation of tumor necrosis factor signaling.
Antioxid. Redox Signal.
11
:
2245
2263
.
44
Matsue
H.
,
D.
Edelbaum
,
D.
Shalhevet
,
N.
Mizumoto
,
C.
Yang
,
M. E.
Mummert
,
J.
Oeda
,
H.
Masayasu
,
A.
Takashima
.
2003
.
Generation and function of reactive oxygen species in dendritic cells during antigen presentation.
J. Immunol.
171
:
3010
3018
.
45
Burg
A. R.
,
H. M.
Tse
.
2018
.
Redox-sensitive innate immune pathways during macrophage activation in type 1 diabetes.
Antioxid. Redox Signal.
29
:
1373
1398
.
46
Alford
A.
,
V.
Kozlovskaya
,
B.
Xue
,
N.
Gupta
,
W.
Higgins
,
D.
Pham-Hua
,
L.
He
,
V. S.
Urban
,
H. M.
Tse
,
E.
Kharlampieva
.
2018
.
Manganoporphyrin-polyphenol multilayer capsules as radical and reactive oxygen species (ROS) scavengers.
Chem. Mater.
30
:
344
357
.
47
Sharma
S.
,
K.
Chopra
,
S. K.
Kulkarni
,
J. N.
Agrewala
.
2007
.
Resveratrol and curcumin suppress immune response through CD28/CTLA-4 and CD80 co-stimulatory pathway.
Clin. Exp. Immunol.
147
:
155
163
.
48
Sansom
D. M.
,
C. N.
Manzotti
,
Y.
Zheng
.
2003
.
What’s the difference between CD80 and CD86?
Trends Immunol.
24
:
314
319
.
49
Rowshanravan
B.
,
N.
Halliday
,
D. M.
Sansom
.
2018
.
CTLA-4: a moving target in immunotherapy.
Blood
131
:
58
67
.
50
Delmastro
M. M.
,
A. J.
Styche
,
M. M.
Trucco
,
C. J.
Workman
,
D. A.
Vignali
,
J. D.
Piganelli
.
2012
.
Modulation of redox balance leaves murine diabetogenic TH1 T cells “LAG-3-ing” behind.
Diabetes
61
:
1760
1768
.
51
Huard
B.
,
P.
Prigent
,
M.
Tournier
,
D.
Bruniquel
,
F.
Triebel
.
1995
.
CD4/major histocompatibility complex class II interaction analyzed with CD4- and lymphocyte activation gene-3 (LAG-3)-Ig fusion proteins.
Eur. J. Immunol.
25
:
2718
2721
.
52
Tarbell
K. V.
,
S.
Yamazaki
,
R. M.
Steinman
.
2006
.
The interactions of dendritic cells with antigen-specific, regulatory T cells that suppress autoimmunity.
Semin. Immunol.
18
:
93
102
.
53
Creusot
R. J.
,
N.
Giannoukakis
,
M.
Trucco
,
M. J.
Clare-Salzler
,
C. G.
Fathman
.
2014
.
It’s time to bring dendritic cell therapy to type 1 diabetes.
Diabetes
63
:
20
30
.
54
Delong
T.
,
T. A.
Wiles
,
R. L.
Baker
,
B.
Bradley
,
G.
Barbour
,
R.
Reisdorph
,
M.
Armstrong
,
R. L.
Powell
,
N.
Reisdorph
,
N.
Kumar
, et al
.
2016
.
Pathogenic CD4 T cells in type 1 diabetes recognize epitopes formed by peptide fusion.
Science
351
:
711
714
.
55
Wiles
T. A.
,
T.
Delong
,
R. L.
Baker
,
B.
Bradley
,
G.
Barbour
,
R. L.
Powell
,
N.
Reisdorph
,
K.
Haskins
.
2017
.
An insulin-IAPP hybrid peptide is an endogenous antigen for CD4 T cells in the non-obese diabetic mouse.
J. Autoimmun.
78
:
11
18
.

The authors have no financial conflicts of interest.

This article is distributed under the terms of the CC BY-NC-ND 4.0 Unported license.

Supplementary data