Enhanced function of immuno-isolated islets in diabetes therapy by co-encapsulation with an anti-inflammatory drug

Abstract

Immuno-isolation of islets has the potential to enable the replacement of pancreatic function in diabetic patients. However, host response to the encapsulated islets frequently leads to fibrotic overgrowth with subsequent impairment of the transplanted grafts. Here, we identified and incorporated anti-inflammatory agents into islet-containing microcapsules to address this challenge. In vivo subcutaneous screening of 16 small molecule anti-inflammatory drugs was performed to identify promising compounds that could minimize the formation of fibrotic cell layers. Using parallel non-invasive fluorescent and bioluminescent imaging, we identified dexamethasone and curcumin as the most effective drugs in inhibiting the activities of inflammatory proteases and reactive oxygen species in the host response to subcutaneously injected biomaterials. Next, we demonstrated that co-encapsulating curcumin with pancreatic rat islets in alginate microcapsules reduced fibrotic overgrowth and improved glycemic control in a mouse model of chemically-induced type I diabetes. These results showed that localized administration of anti-inflammatory drug can improve the longevity of encapsulated islets and may facilitate the translation of this technology toward a long-term cure for type I diabetes.

Fig. 1. In vivo subcutaneous screening of anti-inflammatory drugs encapsulated in PLGA microparticles

SKH-1E mice were imaged three days after subcutaneous injection of microparticles. Data arranged in order of decreasing relative fluorescent or bioluminescent signals. Relative fluorescent or bioluminescent signal was calculated as a ratio of the signal from the drug-loaded microparticles to the signal from the control particles in the same mice. A) Activity of cathepsin enzymes was assessed using Prosense 680, a cathepsin-activated fluorescent imaging probe. B) Activity of reactive oxygen species was assessed using luminol which emits bioluminescence when oxidized by reactive oxygen species. Data: mean ± s.e.m (n=6 replicate injections). The dotted lines represent the position where the relative fluorescent or bioluminescent signal is equal to 1. The data points on the left of the lines (relative signal less than 1) correspond to drug formulations that were effective in decreasing the activity of the corresponding inflammation markers.

Fig. 2. Effects of selected drugs on the peak activities of cathepsin enzymes and ROS in the subcutaneous host response to PLGA microparticles

A) Injection pattern of the PLGA microparticles without ( ) and with drugs ( ) B–D) Bioluminescent images of representative mice on day 2 at the peak of ROS activity. E) Quantified bioluminescent signals on day 2. F–H) Fluorescent images of representative mice on day 9 at the peak of cathepsin activity. I) Quantified fluorescent signals on day 9. Data: mean ± s.e.m (n=15 replicate injections). *, **, *** denotes p<0.05, 0.01, 0.0001 respectively.

Fig. 3. Histology analysis of subcutaneously injected PLGA microparticles with and without drugs excised from SKH-1E mice at different time points

Scale bar represents 50um for all images. Yellow arrows indicate areas with minimal infiltration of immune cells. A–E) Samples with the control microparticles showed the typical time-course of the subcutaneous host response. F–J) Samples containing dexamethasone showed minimal infiltration of immune cells throughout the 28 day duration. K–O) Samples containing curcumin remained free of immune cells during the first two weeks (K–M) but cellular infiltration was observed at later time points (N–O). P–T) Samples containing ketoprofen showed the similar pattern of cellular infiltration as the control particles.

Fig. 4. Effects of hybrid drug-islet capsules on glycemic control of STZ-induced diabetic mice with a marginal islet mass transplanted into the intraperitoneal space

A–C) Phase contrast images of alginate microcapsules without any drug (A), with dexamethasone (E) and curcumin (F). D) Daily non-fasting blood glucose level of STZ-induced diabetic C57B6/J mice transplanted with control islet capsules (n=7), capsules containing dexamethasone (n=7) and curcumin (n=6) co-encapsulated with islets isolated from Sprague-Dawley rats. E) Fasting blood glucose level of the same groups of mice during the IPGTT on day 60. Data :mean ± s.e.m (n=6 or7).(*) and (#) represent p<0.01 and p<0.05 respectively.

Fig. 5. Characterization of fibrotic pericapsular overgrowth on microcapsules retrieved after transplantation into STZ-induced C57B6J diabetic mice

A) qPCR analysis of host (mouse) expression of immunological and fibrosis markers on alginate capsules and surrounding fat pad tissue retrieved one month post-transplantation. The markers were macrophage (Mϕ) marker CD68, B cell marker CD19, dendritic cell marker CD74, cytotoxic T cell marker CD8, inflammatory cytokines TNFα and TGFβ, and fibrosis-associated activated-fibroblast marker α-smooth muscle actin (α SMact) and collagen 1A1 (Col1a1). Data: mean ± s.e.m, (n=7). *, **, *** denotes p < 0.05, 0.01,0.001 respectively. B–D) Fluorescent images of DNA-stained control microcapsules (B) and microcapsules with dexamethasone (C) or curcumin (D) retrieved two month post-transplantation. E–F) Phase contrast images of the same control microcapsules (E) and microcapsules with dexamethasone (F) or curcumin (G). H–J) Histology H&E sections of retrieved control microcapsules (H) and microcapsules with dexamethasone (I) or curcumin (J).