Cell-Permeable Peptide Blocks TLR4 Signaling and Improves Islet Allograft Survival

Abstract

Toll-like receptor 4 (TLR4) activation in pancreatic β cells activates aberrant islet graft cellular pathways and contributes to immune rejection in allogeneic islet transplantation. As an approach to overcoming this problem, we determined the capacity of a 33-amino acid peptide consisting of a protein transduction domain (PTD) from the Hph-1 virus and a fragment of the intracellular domain of TLR4 from the C3H mice (PTD-dnTLR4) to block TLR4 signaling and improve allogeneic islet survival in vitro and after transplantation. The efficacy of PTD-dnTLR4 in blocking TLR4 signaling was assessed in the Raw264.7 macrophage line, in the islets, and the βTC3 cell line. In Raw264.7 cells, preculture with the peptide reduced LPS-induced NF-κB activation and production of proinflammatory cytokines (IL-1β, TNF-α, iNOS, and IL-6). In islets and β cells, preincubation with PTD-dnTLR4 suppressed LPS-induced TNF-α expression via inhibition of NF-κB activation and protected them from stress-induced cell death. In vivo, preincubation of BALB/c (H-2^d) islets with PTD-dnTLR4 resulted in significantly longer survival than control islets in a streptozotocin-induced diabetes model (two of seven grafts survived long term >100 days). PTD-dnTLR4-treated grafts exhibited reduced expression of TNF-α and iNOS and reduced macrophage infiltration posttransplant. The data indicate that PTD-dnTLR4 blocked TLR4 signaling in both macrophages and β cells, and prolonged allograft survival at least in part by suppressing inflammation and macrophage infiltration. This strategy for blocking TLR4 activity has potential utilization in the treatment of diseases where excessive TLR4 activation contributes to the pathologic cellular pathways such as islet transplantation.

Introduction

Islet cell transplantation has potential as a treatment for type 1 diabetes mellitus. Under optimal transplantation settings, 50–60% of islet cells undergo apoptosis, and at least two donors are required for one recipient to obtain normoglycemia ¹ . Transplanted islets are vulnerable in the first few days posttransplant, as apoptosis is induced by various stressors such as hypoxia, proinflammatory cytokines, and blood-mediated inflammatory response. Because apoptosis occurs before immune rejection, prevention of perioperative β-cell death is a promising strategy to enhance success of islet transplantation.

Toll-like receptors (TLRs) are critical mediators of innate immunity that can be activated by damage-associated molecular patterns (DAMPs). Different TLRs recognize different DAMPs. TLR4 recognizes lipopolysaccharide (LPS), a major component of the outer membrane of Gram-negative bacteria. TLR4 signaling can be mediated by myeloid differentiation primary response gene 88 (MyD88)-dependent and -independent pathways ² . Activation of the MyD88-dependent pathway causes phosphorylation of MAPKs, early activation of nuclear factor κ-light-chain-enhancer of activated B cells (NF-κB), and release of inflammatory cytokines including tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) ² . In the MyD88-independent pathway, TLR4 binds to TRIF, resulting in a delayed NF-κB activation and subsequent activation of interferon regulatory factor 3 (IRF3), which triggers secretion of interferon-β (INF-β), IL-6, and expression of inducible nitric oxide synthase (iNOS)^3,4.

TLR4 can also be activated by endogenous ligands including heat shock proteins such as hsp60, hsp70^5–8, extracellular matrix degradation products, high-mobility group box protein (HMGB1), β-defensin, surfactant protein A, and minimally modified low-density lipoprotein (MM-LDL). These endogenous ligands are neither similar to each other nor to microbial ligands^9–12, and most endogenous ligands reside inside cells. Their availability is tightly controlled under normal conditions, which avoids unnecessary simulation of cell surface TLR2 and TLR4. However, these factors can be released by tissue injury or damage and by inflammatory responses under stress, resulting in activation of TLR2/4 in neighboring cells. For example, high-mobility group box 1 (HMGB1) resides in the nucleus, and heat shock proteins (Hsps) normally reside in the cytosol or nucleus (Hsp70 and Hsp22), mitochondria (Hsp60 and Hsp70), or endoplasmic reticulum (gp96). These endogenous ligands can be released into the extracellular environment after cells undergo apoptosis or necrosis.

Innate immunity plays a critical role in destruction of islets and β cells during the peritransplant period. TLR4 and its associated molecules, myeloid differentiation factor (MD-2) and CD14, are expressed in pancreatic β cells ¹³ . Release of endogenous factors including the saturated fatty acid, palmitate, activates the TLR4/MyD88 pathway resulting in production of chemokines that recruit proinflammatory monocytes and macrophages to the islets leading to dysfunction of islets in type 1 diabetes (T1D) mouse models ¹⁴ . In addition to activating the innate immune response, TLRs are critical links between innate and adaptive immune systems. Activation of TLR4 in islet cells leads to allogeneic and xenogeneic graft rejection through induction of proinflammatory and procoagulant responses ¹⁵ . Blocking activation of TLRs, using TLR4-depleted islets or infecting islets with dominant negative TLR4 adenovirus, protects transplanted islets from cytokine-induced apoptosis and leads to long-term survival of islet allografts ¹⁶ . Therefore, availability of a clinically applicable approach to blocking TLR4 activation during the peritransplant period holds promise for increasing the efficacy of islet transplantation.

Protein transduction domain proteins (PTDs) can deliver a large variety of therapeutic proteins into various organs and cells, even across the blood–brain barrier (BBB), without affecting cell viability^17,18. They can transduce cells as rapidly as 15 min after transduction and with high transduction efficiency approaching 100%. PTDs are especially efficient at targeting the intracellular domain of a molecule, and fusion proteins composed of PTDs and various intracellular signaling mediators have elicited therapeutic effects in various disease models^17,19. PTD-based protein therapy can avoid side effects associated with the use of viral vectors and other toxic methods ²⁰ . In this study, we synthesized a 33-amino acid peptide containing the PTD domain of the Hph-1 virus and a fragment of the TLR4 intracellular domain of the C3H mice (PTD-dnTLR4). We determined the efficacy of PTD-dnTLR4 in blocking LPS-induced TLR4 signaling in vitro in Raw264.7, islets, and βTC3 cells and in vivo in pancreatic islets after transplantation.

Materials and Methods

Animals

Male BALB/C and C57BL/6 mice at 6–8 weeks of age were purchased from the Jackson Laboratory (Bar harbor, ME, USA). Mice were fed normal chow. All procedures and protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at the Medical University of South Carolina. All reagents were from Life Technologies unless otherwise specified.

PTD-dnTLR4 Peptide Synthesis

A peptide consisting of the PTD of Hph-1 virus together with a peptide corresponding to 21 amino acids of the TLR4 sequence of C3H/HeJ mice was commercially synthesized (Biomatik Corp) (Fig. 1A). The purity of the peptide was >95%. Peptide was dissolved in phosphate-buffered saline (PBS) (2 mg/ml), aliquoted, and stored at –80°C until use.

Figure 1.

PTD-dnTLR4 can enter Raw264.7 cells. (A) Peptide sequence of wtTLR4 (upper lane) and PTD-dnTLR4 (lower lane) used in this study. (B). Fluorescence microscopic analysis of Raw264.7 cells incubated with FITC-labeled PTD-dnTLR4 for 15 or 30 min. Magnification: 200×.

Peptide Labeling

Fluorescein-5-isothiocyanate (FITC, 3–4 mg; Life Technologies, Grand Island, NY, USA) was dissolved in 1 ml of 100 mM of sodium carbonate (pH 9.5) and was added dropwise to a peptide solution to a ratio of 20 μg of FITC per mg peptide. The mixture was incubated for 1 h at room temperature (RT) and then dialyzed against PBS in the dark. The dialysis solution was changed frequently until the absorbance of the solution reached an OD₄₈₀ of 0.

LPS Treatment

Raw264.7 cells or islets were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS). Cells were precultured with PTD-dnTLR4 or vehicle before challenge with ultrapure LPS EB (InvivoGen, San Diego, CA, USA). Ultrapure LPS EB is known to only induce TLR4 activation ²¹ .

Real-Time Quantitative PCR (RT-PCR) Analysis

RNA was extracted from cells and reverse transcribed into cDNA using an RT-PCR kit (Applied Biosystems, Carlsbad, CA, USA). Expression of TNF-α, IL-1β, IL-6, monocyte chemoattractant protein-1 (MCP-1), and iNOS mRNA were analyzed using commercially available primers from Applied Biosystems. PCR reactions were performed using the ABI 7700 sequence detection system (Perkin-Elmer, Applied Biosystems) as described previously ²² . Fold changes in gene expression normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression were plotted and compared between groups.

Islet Isolation

Islets were harvested from BALB/c mice by collagenase (type V, 0.6 mg/ml; Sigma Aldrich, St. Louis, MO, USA) digestion as described previously ²³ . Yield was assessed by light microscopic counting of dithizone (Sigma Aldrich)-stained islets. Islets were cultured in high-glucose DMEM with 10% FBS at 37°C and 5% CO₂.

Islet Transplantation

Hyperglycemia was induced in C57BL/6 (H-2^b) mice by a single intraperitoneal (IP) injection of streptozotocin (STZ; Sigma Aldrich) at 225 mg/kg. Mice with blood glucose levels >350 mg/dl on 2 consecutive days were used as recipients. BALB/C (H-2^d) islets [500–600 islet equivalents (IEQ)] were preincubated with PTD-dnTLR4 or vehicle and transplanted under the kidney capsule of each recipient. Blood glucose levels of recipients were measured twice per week and used an indirect indicator of islet function. Mice with blood glucose levels <200 mg/dl were considered normoglycemic. Grafts were considered rejected when two consecutive glucose levels were >300 mg/dl after a period of primary graft function as indicated by normoglycemia.

Apoptosis Assay

βTC3 cells were seeded in six-well plates at 1 × 10 ⁶ cells/well. Cells were incubated for 48 h in a hypoxia chamber (1% O₂, 37°C), and cell death was quantified by flow cytometry using propidium iodide (PI) staining.

Western Blot

Total cell lysates (30 μg) were separated by SDS-PAGE, transferred to nitrocellulose membranes, and incubated with goat anti-IκBα antibody (L-14, 1:500 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA), rabbit anti-cleaved-PRAP antibody, (c-PARP, 1:500 dilution; Cell Signaling Technology, Danvers, MA, USA), rabbit anti-Bim antibody (1:1,000; Cell Signaling Technology), rabbit anti-Bcl2 antibody (1:500 dilution; Cell Signaling Technology), or mouse anti-β-actin antibody (cytosol marker, 1:1,000 dilution; Sigma-Aldrich). Signals were visualized using an ECL detection kit (Amersham Pharmacia Biotech, Little Chalfont, UK). Relative expression of proteins was quantified using ImageJ software.

Immunohistochemistry

Immunohistochemistry was performed as described previously ¹⁶ . Kidneys carrying islet grafts were snap frozen in precooled 2-methylbutanolin liquid nitrogen. Sections of 5-μm size were fixed in cold acetone for 3 min, followed by immunohistochemistry with anti-F4/80 antibody (1:1,000; Serotec, Oxford, UK) to detect the presence of macrophages in the transplanted islets. The secondary antibody was biotinylated anti-rat antibody (1:1,000; Vector Laboratories, Burlingame, CA, USA). Slides were observed under light microscopy.

Antigen-Specific Tolerance Test

In recipients bearing long-term surviving grafts, the kidney carrying the islet graft was removed by nephrectomy, and blood glucose levels were measured. After confirmation of hyperglycemia, a second set of islets (500–600 IEQ), derived either from the original donor strain (BALB/C) or a third-party strain (DBA/1, H-2^q), were transplanted under the capsule of the lateral kidney. Rejection of the second graft was determined by blood glucose >350 mg/dl after a period of normoglycemia.

Statistical Analyses

Islet graft survival was compared by Kaplan-Meier survival curves plotted using StatView software. Survival data are expressed as mean survival time ± standard deviation (MST ± SD). Significant differences were determined by Log-rank test; p < 0.05 was considered significant. Expression of cytokines was expressed as mean level ± standard deviation. Differences in cytokine expression were compared by Student's t-test with Bonferroni correction.

Results

PTD-dnTLR4 Rapidly Enters Cells and Blocks TLR4 Signaling

The PTD-dnTLR4 peptide consists of 33 amino acids that include the protein transfer domain of Hph-1 virus (YARVRRRGPRR) together with 22 amino acids identical to the amino acids 702 to 723 of TLR4 in the C3H/HeJ mice (Fig. 1A). To assess the penetration of the cell membrane by FITC-labeled peptide, we incubated Raw264.7 cells with 5 μM of FITC-labeled PTD-dnLR4 for 15 or 30 min at 37°C. Membrane-associated green fluorescence was observed at 15 min (Fig. 1B) and intracellular fluorescence was observed in treated cells at 30 min (Fig. 1B), demonstrating that PTD-dnTLR4 rapidly penetrated the cell membrane and entered cells.

To determine the effect of PTD-dnTLR4 on LPS-induced TLR4 signaling, we measured LPS-induced cytokine expression in Raw264.7 cells following treatment with peptide. The concentration of PTD-dnTLR4 used was 0.5 μM, and 5 μM was preincubated with cells for 30 min. Cells were then stimulated with LPS (100 ng/ml) for 1 to 4 h. Preincubation with the blocking peptide significantly suppressed LPS-induced IL-1β and TNF-α mRNA expression as early as 1 h after LPS stimulation (Fig. 2A and B) and suppressed iNOS and IL-6 within 4 h after LPS stimulation (Fig. 2C and D). Therefore, it seems PTD-dnTLR4 can effectively block LPS-induced TLR4 signaling in Raw264.7 cells.

Figure 2.

PTD-dnTLR4 suppresses LPS-induced cytokine expression in Raw264.7 cells. Cells were incubated with dnTLR4 at different concentrations. Expression of IL-1β, (A) TNF-α (B), iNOS (C), and IL-6 (D) mRNA in Raw264.7 cells after challenge with LPS (100 ng/ml) for 1 or 4 h. Data are representatives of at least three individual experiments. ∗p < 0.05, and ∗∗p < 0.01, Student's t-test.

Incubation of Islets/βTC3 Cells with PTD-dnTLR4 Protects Them From Hypoxia-Induced Cell Death Ex Vivo

We next assessed the effect of pretreatment with PTD-dnTLR4 on the islets harvested from BALB/c mice and determined whether the peptide can block LPS-induced TLR4 signaling. As we observed in Raw264.7 cells, treatment with LPS resulted in an increase in TNF-α expression in islets. Preincubation with PTD-TLR4 significantly inhibited LPS-induced TNF-α generation (Fig. 3A). We have previously shown that proinflammatory cytokines induced TNF-α expression via NF-κB activation/IκBα degradation, which was observed at 15 min and recovered at 30 min after treatment ²² . We measured IκBα expression in islets at different time points post-LPS treatment, and our data showed that IκBα degradation was inhibited in PTD-TLR4 preincubated cells at 15 min posttreatment (Fig. 3B and C), demonstrating that PTD-dnTLR4 blocks LPS-induced TLR4 activation in islets/β cells via inhibition of NF-κB activation.

Figure 3.

PTD-dnTLR4 inhibits LPS-induced TLR4 activation in isolated BALB/c islets. (A) Effect of pretreatment with PTD-dnTLR4 on LPS-induced TNF-α mRNA expression in islets. The gray bars indicate control islets (CTR) and the black bars indicate islet cells pretreated with PTD-dnTLR4 (5 μM, 30 min). Values are representative of three independent experiments. (B) Islets were incubated with PTD-dnTLR4 followed by LPS stimulation as indicated. IκBα degradation was assessed by Western blot 15 min after LPS treatment. (C) Histograms showing the relative quantity of IκBα (densities of IκBα divided by densities of β-actin of each sample) among groups (n = 4). Data are representatives of three individual experiments. ∗p < 0.05, ∗∗p < 0.01, Student's t-test.

Hypoxia is one of the major stressors that induces TLR4 activation in β cells and contributes to islet death posttransplantation ²² . To assess the capacity of PTD-dnTLR4 to protect β cells from hypoxia-induced cell death, βTC3 cells were preincubated with PTD-dnTLR4 (0.5 or 5 μM) for 30 min and then cultured under hypoxic conditions (1% O₂) for 48 h. Cell death was measured by propidium iodide (PI) staining and quantified by flow cytometry. Under hypoxic conditions 31.85 ± 1.91% of untreated cells underwent apoptosis. In contrast, preincubation with dnTLR4 significantly reduced cell death (20.45% ± 4.17% at 0.5 μM and 12.4 ± 2.82% at 5 μM) (Fig. 4A and B). We further measured expression of antiapoptotic gene Bcl2 and proapoptotic gene Bim, and cleaved PARP protein level in cells pretreated with PTD-dnTLR4 and exposed to hypoxia for 4 to 24 h. As is evident in Figure 4C, Bcl2 expression was increased at 4 and 8 h posthypoxia in cells preincubated with dnTLR4 compared to controls, Bim was suppressed at 8 and 12 h posthypoxia, and c-PARP was reduced at 4 h after hypoxia. These data suggest that blocking TLR4 signaling protects β cells from hypoxic stress-induced cell death via regulation of apoptotic-related protein expression.

Figure 4.

Preincubation with PTD-dnTLR4 protects βTC3 from hypoxia-induced cell death. (A) Apoptosis in βTC3 cells incubated with PTD-dnTLR4 (0.5 μM or 5 μM) and cultured under hypoxic conditions (1% O₂) for 48 h. Cells were stained with propidium iodide and analyzed by flow cytometry. (B) Percentages of apoptotic cells in each group shown in (A). Values are representative of three independent experiments; ∗p < 0.05 versus Control. (C) Expression of Bcl2, Bim, c-PRAP, and β-actin at protein levels measured in cells incubated with PTD-dnTLR4 (0.5 μM or 5 μM) and cultured under hypoxic conditions (1% O₂) for 4 to 24 h.

Preincubation with the PTD-dnTLR4 Before Transplant Prolongs Islet Allograft Survival in a Major MHC-Mismatched Allogeneic Islet Transplantation Model

To determine whether preincubation of islets with PTD-dnTLR4 improved survival and function of islets posttransplant, freshly isolated BALB/c islets were incubated with PTD-dnTLR4 (5 μM, 30 min) before being transplanted into C57BL/6 recipients that had been rendered diabetic by STZ. Grafts of untreated control islets were rejected in 20.3 ± 6.7 days, and their blood glucose levels remained >350 mg/dl until the end of experiments (n = 7) (Fig. 5A). Among grafts of islets incubated with PTD-dnTLR4 (n = 7), five were rejected in 30.6 ± 10.0 days, but two survived long term (>100 days, p = 0.004 vs. control, log-rank test) (Fig. 5A,), indicating that preculture with PTD-dnTLR4 before islet transplantation prolongs islet allograft survival.

Figure 5.

Preincubation of BALB/c islets with PTD-dnTLR4 prolongs islet allograft survival in diabetic C57BL/6 recipients. (A) Kaplan–Meier survival plot of grafted untreated BALB/c islets (gray circles, n = 7) and islets preincubated with PTD-dnTLR4 peptide (black circles, n = 7). (B) Changes of blood glucose levels of mice with long-term surviving allografts after removal of the first islet graft and transplantation of a second islet graft from either a BALB/c donor (solid line) or a DBA/1 donor (dotted line). Horizontal line indicates blood glucose of 200 mg/dl. (C) Expression of TNF-α, iNOS (D), and MCP-1 (E) mRNA in fresh islets, islet allografts preincubated with PTD-dnTLR4 (black bars), or vehicle (CTR, gray bars) at 1 day posttransplant. (F) Immunohistochemical analysis of macrophage infiltration in islet grafts 1 and 7 days posttransplantation. Kidney sections containing transplanted islets were stained with anti-F4/80 antibody. Magnification: ×200. ∗p < 0.05 versus CTR, Student's t-test with Bonferroni correction. Values are means of at least three grafts per group.

To determine whether antigen-specific tolerance had been achieved in the two recipients carrying long-term surviving grafts, the original islet grafts were removed at 110 days posttransplant by nephrectomy. As expected, blood glucose levels of those mice increased from less than 150 mg/dl to 300 mg/dl after removal of the grafted kidney, indicating that normoglycemia had been maintained by the transplanted islets in these mice. We then placed a second subcapsular islet graft into the contralateral kidney. The second graft originated from either the original donor strain (BALB/c, H-2^d) or a third-party strain (DBA/1, H-2^q). Both grafts were rejected after a short term of primary function (Fig. 5B), suggesting that tolerance was not induced. In view of the fact that the number of mice tested was extremely limited (n = 1 per group), further experiments will be required.

Incubation with PTD-dnTLR4 Reduces Expression of Proinflammatory Genes and Inhibits Macrophage Infiltration Into Transplanted Islets

In a previous study, we have demonstrated less inflammation among islet grafts derived from TLR4-depleted donors than among grafts from wild-type donors. We speculated that the reduction in inflammation might contribute to long-term survival of islet grafts ²² . To assess the mechanism by which PTD-dnTLR4 prolongs islet allograft survival, we measured the expression of proinflammatory cytokines in transplanted islets at 1 day posttransplant. We found that the expression of proinflammatory cytokines TNF-α, iNOS, and MCP-1 was significantly increased in islet allografts compared to freshly isolated islets (Fig. 5C–E). In contrast, islets preincubated with PTD-dnTLR4 showed significantly reduced expression of TNF-α and iNOS. Expression levels of MCP-1 were also reduced in PTD-dnTLR4-treated islets, but the differences were not significant (Fig. 5E). We measured macrophage infiltration into islet grafts at 1 and 7 days posttransplant by immunohistochemical analysis using F4/80 antibody. Only a few F4/80⁺ cells were observed in grafts from either control or treated grafts 1 day posttransplant (Fig. 5F). However, at 7 days posttransplant, significantly fewer F4/80⁺ cells were observed in PTD-dnTLR4-treated islets than in controls, suggesting that treatment of islets with PTD-dnTLR4 prior to grafting reduced macrophage infiltration into the grafts. These results indicate that the PTD-dnTLR4 blocking peptide effectively inhibited TLR4 activation and reduced inflammation in transplanted islets, suggesting that PTD-dnTLR4 might be used as an additive therapy to enhance the efficacy of islet transplantation.

Discussion

Excessive TLR4 signaling contributes to the pathogenesis of numerous inflammatory diseases including sepsis, systemic sclerosis, ischemia reperfusion injury, and graft failure posttransplantation ²⁴ . Inhibiting or blocking TLR4 signaling using a clinically applicable approach may have therapeutic value. We previously showed that TLR4-depleted islet grafts (obtained from TLR4 knockout mice or by adenoviral infection) exhibited long-term survival in a mouse allogeneic islet transplantation model ²² . Here we approached TLR4 depletion using a synthetic peptide of 33 amino acids (PTD-dnTLR4). We demonstrated that this peptide rapidly entered cells and reduced LPS-induced TLR4 activation in macrophages and islets/β cells. Preincubation of islets with PTD-dnTLR4 for 30 min protected β cells from hypoxia-induced apoptosis and prolonged survival of islet allografts in a major MHC-mismatched islet transplantation model. The data suggested that the mechanism of protection was likely through suppression of expression of proinflammatory cytokines and inhibition of macrophage infiltration into the islet graft.

The intracellular domain of TLR4 plays a pivotal role in receptor dimerization and activation of signaling and is critical component of TLR4 function^25,26. For example, because of a single mutation in this domain, which impedes signal transduction, C3H/HeJ mice are hyporesponsive to LPS. Our results using PTD-dnTLR4, which carries the C3H/HeJ mutation, support the critical role of this mutation in blocking the TLR4 signaling pathway. We demonstrated that both macrophages and β cells, which were preincubated with PTD-dnTLR4, exhibited reduced expression of proinflammatory cytokine expression after LPS stimulation. Based on cytokine expression, we infer that PTD-dnTLR4 was effective in blocking both Myd88-dependent cytokines (IL-1β and TNF-α) and TRIF pathway-dependent cytokines (IL-6) in both cell types.

Primary nonfunction and early islet destruction are major obstacles hindering success of islet transplantation. Graft function is frequently compromised because donor islet cells undergo apoptosis induced by various stressors mediated by the innate immune response after transplant. Thus, more islets are needed for each transplant, straining the already limited supply of donor islets. The traditional approach to improving graft function has centered on improving immunosuppression after transplantation; however, we have emphasized imparting “resistance” in donor islets to various insults via limiting innate immune responses and reducing stresses using various approaches including blocking activation of TLR4^16,22 and induction of protective gene expression in islet grafts^27,28.

Multiple stress factors contribute to islet dysfunction and death after transplantation. Among them, hypoxia plays a major role ²⁹ . β Cells are particularly sensitive to hypoxic conditions due to their high oxygen demand during insulin secretion ³⁰ . Inflammation is another cause of β-cell death posttransplant. Proinflammatory cytokines (including TNF-α, INF-γ, and IL-1β) produced by infiltrating cells of the recipient's innate immune response can cause early islet dysfunction and death^31–34. Proinflammatory cytokines can induce expression of chemokines (IP-10, MIP-1α, MIP-1β, MCP-1, and RANTES) in β cells and can contribute to the recruitment of recipient-derived inflammatory cells that infiltrate and participate in rejecting the islet graft^35–37. Both hypoxia and inflammation activate TLR4 in transplanted islets, especially in β cells, and might contribute to islet cell death^38,39. Moreover, TLR4 can be induced by various endogenous factors (e.g., HMGB1) released by dead cells leading to further islet cell death and graft dysfunction^40,41. By blocking TLR4 signaling, treatment with PTD-dnTLR4 reduced hypoxia and inflammation-induced β-cell apoptosis. As suggested by the danger hypothesis, without inflammation, the immune response will be weak; the graft might escape immune rejection and survive long term ⁴² .

Although only two of seven peptide-treated islet grafts survived long term in this study, graft survival was also significantly prolonged in recipients that ultimately rejected the islet graft. It seems that indefinite graft survival was achieved by graft accommodation without achieving antigen-specific tolerance, as shown in another islet transplantation study in which TLR4 was blocked by an anti-TLR4 antibody ⁴³ .

Whether TLR4 must be blocked in both the graft and recipient is uncertain. Conflicting studies have shown that to achieve graft survival TLR4 must be blocked only in the graft ⁴⁴ , only in the recipient ⁴⁵ , or in both ⁴³ . In addition, PTD-transduced islets when transplanted into recipients will again encounter lots of endogenous ligands (e.g., Hsp60) released during surgery. Therefore, if the PTD-dnTLR4 effect is transient, recipient treatment to further block TLR4 activation posttransplantation might bring additional benefits to transplanted islets. In future studies aimed at generating an optimal treatment protocol, we will further assess the contributions of treating only the graft, treating only the recipient, or treating both to improving graft survival.

Our data indicate that PTD-dnTLR4 effectively blocked activation of TLR4 in macrophages and in islets. PTD-dnTLR4 enhanced islet graft function with a concomitant improvement in long-term islet survival. Thus, treatment with PTD-dnTLR4 or a similar agent might be useful as a therapeutic approach for islet transplantation and other diseases in which blocking TLR4 signaling would be advantageous.