Biopatterned CTLA4/Fc Matrices Facilitate Local Immunomodulation, Engraftment, and Glucose Homeostasis After Pancreatic Islet Transplantation

There is an immediate need for pancreatic islet transplantation (PIT) strategies that increase functional efficacy of small cell loads by optimizing engraftment and eliminating the need for chronic immunosuppression for graft survival. Costimulation blockade with cytotoxic T lymphocyte antigen 4 fusion protein (CTLA4-Ig) has been effective in downregulating allo- and autoimmune responses, leading to enhancement of islet allograft survival and inhibition of type 1 diabetes development (1–4). However, systemic administration of large doses of CTLA4-Ig not only has long-term side effects but also has failed to completely inhibit alloimmune destruction of islet β-cells and to induce tolerance (5). Site-specific immunomodulation may be a promising strategy to prevent acute rejection and minimize indiscriminate damage to essential cells and organs by systemic immunosuppression (6–8).

Our team has developed a cutting-edge biopatterning technology that can spatially control the distribution of physiologically relevant, pico-to-nanogram level doses of signaling molecules immobilized as “solid-phase” patterns to extracellular matrix (ECM) constructs via native binding affinities to direct cell behavior in vitro and tissue formation in vivo (9–11). Our a priori hypothesis is based on successful prior work where biopatterning of signaling molecules in solid phase significantly improved outcomes across multiple applications (12,13). Here, we fabricated immunoregulatory microenvironments using murine CTLA4/Fcγ2a heavy-chain chimeric fusion protein (CTLA4/Fc) constructs biopatterned within immunoneutral human-derived acellular dermal matrix (ADM). We investigated the effects of this CTLA4/Fc microenvironment on inhibiting alloimmune responses after PIT and promoting islet allograft survival without additional immunosuppression. Our work is the first to demonstrate that targeted, spatially controlled delivery of extremely small doses of solid-phase proregulatory agents, such as biopatterned CTLA4/Fc microenvironments, can promote engraftment, function, and long-term survival of islet allografts.

Uniform patterns of murine CTLA4/Fc bioink (200 μg/mL) were produced in our laboratory and printed onto 200-μm thick, 5 × 5 mm pieces of ADM (Synthes, West Chester, PA) using our inkjet-based biopatterning system (9,10,14). The printed bioink absorbs into the open-pore structure of the dermal surface and binds within the matrix to produce immobilized 3-D patterns that persist for sufficient periods of time needed to elicit the desired biological control (9). Binding retention of CTLA4/Fc was estimated by printing ¹²⁵I-CTLA4/Fc onto ADM using the method similar to other hormone bioinks (14,15), which demonstrated that biopatterned CTLA4/Fc persist stabilized release with 50% of bound remaining for 21 days under simulated in vitro conditions.

PIT was performed from donor DBA/2(H-2^d) mice to streptozotocin-induced diabetic C57BL/6(H-2^b) recipient mice as previously described (16). After prerinsing, ADM squares were cut into 1 × 5 mm strips and inserted below the renal capsule. Approximately 300 islet allografts were injected between strips, so that the biopatterned CTLA4/Fc could be in direct contact with islet allografts (Fig. 1). Other groups included single dose CTLA4/Fc (20 mg/kg) intraperitoneally (i.p.), CTLA4/Fc ADM implanted under the contralateral renal capsule, nonprinted ADM, and untreated recipients. Allograft function was assessed by monitoring blood glucose levels (BGL), and rejection was defined as BGL of >300 mg/dL on two consecutive measurements (17).

The production of IL-4, IFN-γ, IL-6, and IL-17 in serum samples obtained at 2 and 8 weeks posttransplantation were assessed by enzyme-linked immunosorbent assay (ELISA). Total RNA was extracted from draining lymph nodes or islet allografts. Levels of specific mRNA IL-4, IFN-γ, Gzmb, IL-17, and Foxp3 were quantified by real-time PCR.

Cells were harvested from lymph nodes, spleens, or peripheral blood of long-term survival recipients at day 150 after transplantation. The expression of cell surface and intracellular markers (CD3, CD4, CD25, and Foxp3) were analyzed by flow cytometric analysis. To test regulatory T-cell (Treg) suppression, flow-sorted CD4⁺CD25⁻ responder cells from naive C57BL/6 mice were cultured with CD4⁺CD25^hi T cells from long-term islet allograft survival recipients or age-matched naive mice in 96-well plates, using anti-CD3 and anti-CD28 monoclonal antibody (mAb) or irradiated splenocytes from donor (DBA/2) or the third party (C3H) as stimulators. Cultures were pulsed for the last 8 h of the incubation with 1 μCi/well ³H-thymidine, and ³H-thymidine incorporation was measured using standard scintillation procedures and expressed as percent inhibition of proliferation.

The kidney bearing the islet graft was removed from recipients and processed for hematoxylin-eosin (H&E) staining and immune-peroxidase histochemistry with anti-mouse Foxp3 mAb or anti-insulin mAb. Samples were evaluated in a blinded fashion at two different levels of sectioning.

Data are expressed as mean ± SD. Comparisons between two samples were performed using the Student t test. Multiple groups were compared by one-way ANOVA with Tukey multiple comparisons test. Allograft survival was analyzed using a log-rank test. Differences are considered significant if P < 0.05.

In the MHC-mismatched DBA/2 to C57BL/6 mice PIT model, treatment with local biopatterned CTLA4/Fc ADM significantly prolonged islet allograft survival with median survival time (MST) of 71 days (n = 10; P < 0.001 vs. untreated group; P < 0.05 vs. CTLA4/Fc i.p. group). All those recipients maintained normoglycemia for >28 days, 40% of which demonstrated long-term (>150 days) euglycemia. Contralaterally implanted CTLA4/Fc ADM did not induce engraftment of islet allografts, indicating that the biopatterned CTLA4/Fc functioned at the transplant site but not remotely (Fig. 2A). Furthermore, biopatterned CTLA4/Fc maintains persistence of detectable low serum levels longer than systemically delivered CTLA4/Fc as detected by ELISA over time after PIT (Fig. 2B).

The levels of IL-4 increased and IFN-γ and IL-6 decreased in the local biopatterned CTLA4/Fc treatment group compared with the nonprinted ADM group at 2 weeks posttransplantation. After 8 weeks, the alteration of IL-4 and IL-6 remained in biopatterned CTLA4/Fc–treated recipients, when a significant decrease of IL-17 was detected (Fig. 3A). Additionally, there was marked upregulation of IL-4 and Foxp3 mRNA and downregulation of IFN-γ and Gzmb mRNA expression in both islet allografts and draining lymph nodes of biopatterned CTLA4/Fc–treated recipients at 2 weeks posttransplantation (Fig. 3B). T cells from biopatterned CTLA4/Fc–treated, long-term surviving recipients demonstrated an increase of CD4⁺Foxp3⁺ Treg in peripheral blood (6.0 ± 0.3%), spleen (8.8 ± 0.5%), and lymph nodes (9.5 ± 0.4%) compared with the naive mice (Fig. 3C). Furthermore, CD4⁺CD25^hi Treg isolated from these long-term surviving recipients significantly inhibited the proliferation of CD4⁺CD25⁻ T cells activated by the same donor antigen in a mixed lymphocyte reaction (71.2 ± 2.7% vs. 31.6 ± 2.1% inhibition; P < 0.01) but could not suppress third-party antigen-induced alloreactivity. These Treg also exhibited more potent suppressive ability than Treg from naive mice (67.9 ± 3.1% vs. 24.7 ± 4.0% inhibition; P < 0.01) under nonspecific stimulation (Fig. 3D).

A decreased inflammatory cell infiltration and minimal damage of islet allografts were observed by H&E at day 14 posttransplantation in the biopatterned CTLA4/Fc–treated recipients compared with the nonprinted ADM–treated recipients (Fig. 4A). The low degree of inflammatory infiltration and significant preservation of islet allografts were further characterized in the biopatterned CTLA4/Fc–treated long-term surviving recipients at 150 days posttransplantation by H&E and immunohistochemical analysis. The strong insulin staining demonstrated the potent function of allogeneic islets with insulin-producing cells. An increase of Foxp3-expressing cells suggests that local immune regulation is dominant in the protection of islet allografts from alloimmune destruction (Fig. 4B).

Local or spatial control of the transplant microenvironment can modulate inflammation, immune cell infiltration, and activation associated alloimmune responses (18). Prior work has shown that T-cell–mediated cytotoxicity may be mitigated through local delivery of CTLA4-Ig (19). In this study, we sought to improve the efficiency of local immune regulation by targeting delivery of CTLA4/Fc to the transplant microenvironment of engrafted β-cells. Our goal was to eliminate systemic nonspecific immunosuppression to minimize islet toxicity, thereby reducing the islet cell load needed for PIT and supporting long-term engraftment, survival, and endocrine function.

We printed bioinks of CTLA4/Fc on a biodegradable ADM with defined surface concentrations and characterized spatial organization to create a tissue-engineered microenvironment using our established inkjet biopatterning technology, which is scalable and readily adaptable to other signaling molecules and/or ECMs (12,14). This technique allows for the solid-phase delivery of therapeutic proteins in extremely low dosages, with increased bioactivity relative to liquid-phase delivery. Our results show that a biopatterned CTLA4/Fc microenvironment implanted along with a low number of islet cells under the renal capsule achieved significantly prolonged survival, whereas a single i.p. injection of CTLA4/Fc or contralaterally implanted CTLA4/Fc ADM was less effective, suggesting that the immunoregulatory effect is exerted locally by CTLA4/Fc at the site of islet transplantation. Furthermore, prolonged low levels of serum CTLA4/Fc were observed following implantation of biopatterned CTLA4/Fc, indicating that high levels of CTLA4/Fc in the systemic circulation were not necessary. This would suggest that a persistent presence of solid-phase CTLA4/Fc might be optimal for immunoregulation.

ADM, as a native biological scaffold maintaining the original dermal ECM architecture and low immunogenicity, was determined to be a suitable surrogate material to sequester biologically active components (9). It has also been shown to support cell growth in vitro and to support neovascularization following implantation in vivo (20). Therefore, biopatterning technology combined with the features of ADM enables prolonged retention and an extended bioavailability of biopatterned CTLA4/Fc, thus allowing it to be sequestered at nanogram levels. This establishes a microenvironment that supports transplanted islet β-cells and exerts local immunomodulation by CTLA4/Fc to promote islet allograft engraftment and long-term function.

Type 1 T-helper cells (Th1) and type 2 T-helper cells (Th2) are the principal regulators of alloimmunity in transplantation. The induction of Th2 cytokines can inhibit Th1-mediated rejection responses, thereby promoting allograft acceptance (21). The cytokine profiles from biopatterned CTLA4/Fc–treated recipients revealed increased Th2 cytokine (IL-4) and decreased Th1 cytokines (IFN-γ) production in serum, with changes being more significant in the late posttransplantation period (8 weeks). In addition, the corresponding upregulation of IL-4 gene and downregulation of IFN-γ gene expression were detected in both islet allografts and draining lymph nodes. This finding suggests that biopatterned CTLA4/Fc delivered locally acts efficiently to modulate Th1/Th2 cytokines and to promote an immune deviation toward Th2 responses. Furthermore, histological examination showed decreased lymphocyte infiltration and long-term preserved islet allograft with increased Foxp3 staining in the biopatterned CTLA4/Fc–treated recipients, indicating the protective effect of biopatterned CTLA4/Fc in situ against alloreactive effector T-cell responses.

Active immune suppression by Treg is a prominent periphery tolerance mechanism. Nevertheless, inflammation of local tissue during transplantation not only limits Treg suppression but also promotes proinflammatory Th17 responses (22,23). Our data demonstrated that localized biopatterned CTLA4/Fc treatment reduced serum levels of IL-6 and IL-17 and decreased Gzmb gene and enhanced Foxp3 gene expression in both islet allografts and draining lymph nodes, accompanied by increased Treg cells. These findings further support the multifaceted immunomodulation conferred by localized biopatterned CTLA4/Fc. They may provide microenvironments that favor Treg generation in the hostile environment of an alloimmune response through suppressing cytotoxic T lymphocytes and inhibiting IL-6–mediated Th17 responses. In this study, the expansion of Treg and maintenance of their suppressive potency on effector T-cell proliferation against donor antigen in a mixed lymphocyte reaction may at least in part underlie the long-term allograft survival induced by the local biopatterned CTLA4/Fc therapy. A continuing presence of CTLA4/Fc in the islet allograft could especially benefit the recruitment of Treg to the allograft and ensure Treg effective control of antidonor reactivity. These factors encourage the further investigation of local immunoregulation for the induction of transplantation tolerance.

In conclusion, for the first time, we demonstrated that biopatterning of native matrices, such as ADM, with immunoregulatory agents, such as CTLA4/Fc, facilitates highly localized delivery of these agents without diminution of efficacy. Our study confirms that this technology promotes long-term engraftment of minimal loads of allogeneic islet cells and leads to sustained glucose homeostasis. The CTLA4/Fc microenvironment promoted expansion of Treg and shifted cytokine production and gene expression from proinflammatory to regulatory profiles. Our findings suggest that the targeted delivery of biopatterned CTLA4/Fc microenvironments could become a safe and effective strategy for minimizing systemic immunosuppression and promoting islet allograft survival in PIT.

Acknowledgments. The authors thank Kia M. Washington and Maxine Miller (Department of Plastic Surgery, University of Pittsburgh School of Medicine) for providing input and advice to our work.

Funding. This study was supported by a JDRF Innovative Grant (5-2012-308 to M.G.S. [principal investigator], X.X.Z., V.S.G., and P.G.C. [co-principal investigator]) and in part by the Plastic Surgery Foundation and Musculoskeletal Transplant Foundation Dermal Tissue Research Grant (349234 to W.Z. [principal investigator]).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. W.Z., V.S.G., P.G.C., L.E.W., and X.X.Z. generated the original study conception. W.Z., V.S.G., P.G.C., L.E.W., X.X.Z., and M.G.S. designed experiments, analyzed data, and interpreted results. W.Z., V.S.G., and P.G.C. wrote the manuscript and prepared the figures. W.Z., V.S.G., P.G.C., L.E.W., X.X.Z., and M.G.S. made critical scientific input and revisions for intellectual content. W.Z., Y.L., Y.Y., and C.K. performed animal experiments, immunological assays, and data acquisition. W.Z. and X.X.Z. constructed the CTLA4/Fc fusion protein. P.G.C. and L.E.W. made the biopatterned constructs. M.G.S. coordinated and directed the project. M.G.S. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.