Type 1 diabetes is an autoimmune disorder. Although the pathogenesis of this disease is still under investigation, it is known that T-cell–mediated cytotoxicity and inflammatory cytokine damage cause dysfunction of insulin-secreting β cells in the pancreatic islets of Langerhans, resulting in β-cell destruction, insulin deficiency, hyperglycemia, and diabetes (1,2). Many studies have demonstrated that suppression of T-cell activation or reduction of inflammatory cytokine production could prevent or relieve this disease. One therapeutic approach is to use anti-inflammatory agents to delay or prevent the development of diabetes (3,4). Pancreatic islet transplantation is one of the treatments being developed to replace insulin-secreting β cells. Although new immunosuppressive regimens protect transplanted islets from immune rejection (5), autoimmune reactivity remains in type 1 diabetic recipients. Autoimmune reactive cells and cytokines destroy islets, contributing to transplantation failure in addition to allogeneic rejection (6).
The association between allogeneic rejection and autoimmune destruction after islet transplantation is not fully understood. However, T-cell activation and inflammatory cytokines are involved in both scenarios (6,7). Lisofylline (LSF) is a novel anti-inflammatory compound. LSF inhibits T-helper (Th) 1 cell differentiation and activation (8), reduces inflammatory cytokine production (9), and prevents autoimmune diabetes and multiple sclerosis in mice (3,10). Our studies show that LSF preserves insulin secretion in β cells exposed to inflammatory cytokines (3,11,12). In this study, we investigated whether LSF could prevent autoimmune destruction and diabetes recurrence after islet transplantation in the nonobese diabetic (NOD) mouse model. We demonstrated that LSF could protect transplanted islets from autoimmune damage in NOD recipients.
MATERIALS AND METHODS
Animals
NOD/LtJ (H2g7, as NOD) and NOD.CB17-Prkdc severe combined immunodeficiency (SCID)/J (H2g7, as NODSCID) mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and housed in a pathogen-free colony at the Center for Comparative Medicine, University of Virginia. The experimental animal protocol was fully approved by the Institutional Animal Care and Use Committee at the University of Virginia. In our NOD colony, 75% to 85% of female NOD mice become spontaneously diabetic by the age of 30 weeks, but only 30% to 40% of male NOD mice are diabetic at the same age. Female mice were screened for hyperglycemia weekly after the age of 12 weeks. Diabetes was considered present when blood glucose levels were higher than 250 mg/dL in three consecutive measurements. Diabetic mice were used as recipients 2 weeks after initial diagnosis, and were injected with 2 to 5 U/day of insulin (Humulin U Ultralente, Eli Lilly & Company, Indianapolis, IN) intramuscularly. Insulin injections in the recipient mice were withdrawn on the day of islet transplantation and thereafter.
Reagents
Lisofylline (LSF; 1-[5-R-hydroxyhexyl]-3,7-dimethylxanthine, or CT-1501R) was obtained from Cell Therapeutics (Seattle, WA). All other reagents were purchased commercially, including murine interleukin (IL)-1β, interferon (IFN)-γ, and tumor necrosis factor (TNF)-α (R & D Systems, Minneapolis, MN); collagenase P (Roche, Indianapolis, IN); anti-insulin antibody (H-86; Santa Cruz Biotechnology, Santa Cruz, CA) and murine cytokine enzyme-linked immunosorbent assay (ELISA) kits (BioSource International, Camarillo, CA); and Ficoll type-400 and D-glucose (Sigma Chemical Co., St. Louis, MO).
Islet Isolation, Treatment, and Functional Assay
Seven- to 8-week-old female NODscid mice were used as islet donors. Islets were isolated by means of collagenase P digestion and Ficoll type-400 discontinuous gradient separation as previously described (3). Freshly isolated islets were treated with 10 to 50 μM of LSF with or without a murine cytokine cocktail containing 5 ng/mL of IL-1β, 100 ng/mL of IFN-γ, and 10 ng/mL of TNF-α overnight as described (12). Static insulin secretion assay was performed as described (3).
Apoptosis Detection in Islets
Equal numbers (50 islets per well) of LSF-treated or cytokines plus LSF-treated islets were seeded onto a gelatin-coated 96-well plate. After 48-hr incubation, apoptosis was detected using an ApoPercentage Apoptosis Assay kit (BioColor Ltd., Belfast, Northern Ireland). After incubation with the apoptosis-detecting dye, apoptotic islets showed red-purple color that could be visualized and counted using a light microscope. After washing to eliminate unlabeled dye, a dye-releasing agent was added to the islets to release labeled dye into supernatant. A quantitation of apoptosis was assessed in the supernatant by a spectrometric analysis at 590 nm.
Islet Transplantation
Six hundred to 800 freshly isolated islets were implanted into subrenal capsules of the left kidney in each diabetic female NOD recipient. In the LSF treatment group, recipient mice started to receive daily intraperitoneal injections of LSF (50 mg/kg). In the saline treatment group, recipient mice received 100 μL of normal saline daily, intraperitoneally. Both treatments were started on the day of transplantation. By using an Accu-Chek blood glucose monitor (Roche Diagnostics, Indianapolis, IN), blood glucose levels of the recipient were checked daily for the first 10 days posttransplant and weekly thereafter.
Serum Cytokine Detection
Serum samples were collected from recipient mice at days 7, 21, and 28 after transplantation. Levels of cytokines were measured by ELISA kits according to the manufacturer’s protocols.
Nephrectomy, Histology, and Immunochemistry
The islet-engrafted kidney (left side) was removed from the recipient. The opposite kidney was kept intact to keep the recipient alive. At the end of the study, sera, pancreata, and islet-engrafted kidneys were collected. Histologic examination was performed on the pancreas and islet-engrafted kidney sections after a hematoxylin-eosin stain. Paraffin-embedded sections of pancreata and islet-engrafted kidneys were also used for insulin immunochemical staining. Pancreatic insulitis of all recipients was examined after termination of observation to confirm as evidence of diabetes.
Statistical Analysis
Data are presented as means±SEM. The statistical significance of the difference between groups was evaluated using analysis of variance, followed by Student-Newman-Keuls test for multiple comparisons. A value of P <0.05 was considered significant.
RESULTS
LSF Preserved Islet Insulin Secretion Function in the Presence of Inflammatory Cytokines
Inflammatory cytokines, such as IL-1β, IFN-γ, and TNF-α, cause dysfunction in murine pancreatic β cells in vitro (13). β-cell dysfunction can be demonstrated by reduced basal insulin secretion and responsiveness to glucose stimulation. In our study, freshly isolated islets were cultured with 10 to 50 μM of LSF with and without the cytokine cocktail overnight. Static insulin secretion assay showed that LSF enhanced β-cell basal insulin secretory capacity and its responsiveness to glucose stimulation in a dose-dependent fashion (Fig. 1A). In the presence of inflammatory cytokines (IL-1β, IFN-γ, and TNF-α), LSF significantly improved both basal and glucose-stimulated insulin secretion in isolated islets (Fig. 1B). These results indicate that LSF has protective effects against inflammatory cytokine-mediated damage in β cells.
;)
Figure 1: Static insulin secretion. Isolated NODscid islets were cultured with LSF 10 to 50 μM with or without the cytokine cocktail that contains IFN-γ 100 ng/mL, TNF-α 10 ng/mL, and IL-1β 5 ng/mL overnight. After washing to eliminate treatment agents, 50 islets per treatment group in duplicate were treated with 3 mM glucose-supplemented Krebs-Ringer bicarbonate (KRB) solution at 37°C for 30 min. Then, one set from each group was cultured with 28 mM glucose-KRB solution and the other set was cultured in 3 mM glucose-KRB solution for an additional 2 hr. Supernatant samples were collected for insulin measurement using an ELISA kit. A murine insulin standard was provided with the ELISA kit. (A) The dose effect of LSF in β-cell insulin secretion without cytokines. *P <0.05, **P <0.01 when compared with no LSF controls. (B) The effect of 50 μM of LSF treatment in protection of β-cell insulin secretion in the presence of an inflammatory cytokine cocktail. *P <0.01 in the pair comparison. All experiments were repeated three times and showed similar results.
LSF Reduced Islet Apoptosis in the Presence of Inflammatory Cytokines
Inflammatory cytokines intensify apoptosis in murine islets and cause β-cell dysfunction (13). Using optic density measurement of culture supernatant by a spectrophotometer, islet apoptoses can be compared quantitatively. We found a significant reduction of apoptosis in the LSF-treated islets after incubation with inflammatory cytokines (Fig. 2). The treatment with 25 to 50 μM of LSF inhibited inflammatory cytokine-mediated islet cell damage.
Figure 2: Quantitative measurement of islet apoptosis. By using the ApoPercentage Apoptosis Assay kit, apoptotic islets were labeled with apoptosis dye. The concentration of the dye in supernatant was measured by an optic density reading (at 590 nm) with a spectrophotometer after addition of a dye-releasing agent. The studies were repeated three times and showed similar results. One of the representative studies is shown. P <0.05 between cytokine-treated and cytokine plus LSF-treated groups.
LSF’s effect on reduction of islet apoptosis was dose-dependent. We have tested a serial dosage of LSF from 10 to 50 μM and found that 25- to 50-μM concentrations were effective in our conditions. Two to 4 hrs of treatment seemed to be sufficient to observe LSF’s effect in vitro (data not shown).
LSF Prolonged Islet Graft Survival in Autoimmune Recipients
Syngeneic islet grafts are destroyed in autoimmune diabetic NOD recipients if a protective therapy is not provided. This phenomenon is known as autoimmune diabetes recurrence and is caused by the recipient’s autoimmunity destroying transplanted islets, even though both donor and recipient are genetically identical (14).
We have investigated whether LSF could prevent autoimmune diabetic recurrence in NOD recipients after islet transplantation. In the LSF treatment group, NOD mice received daily LSF injections beginning on the day of islet engraftment. This treatment was provided for only 3 weeks and was then completely withdrawn. Even with such a short period of treatment, LSF significantly prolonged islet graft survival as compared with the recipients that were treated with normal saline alone (average time of survival was 6 days in the saline group and >65 days in the LSF group) (Table 1). Daily blood glucose levels after grafting are shown in Figure 3.
Table 1Figure 3: Blood glucose measurement in the islet recipients. The islet recipients in both LSF and saline treatment groups were monitored for blood glucose levels before and after grafting. (A) Changes of blood glucose levels in saline-treated recipients individually. (B) Changes of blood glucose levels in LSF-treated recipients individually, including those mice that received nephrectomy (NT) at indicated dates after islet engraftment. Blood glucose levels higher than 200 mg/dL were considered as evidence of rejection after transplantation.
To identify the source of insulin, four of the LSF-treated recipients that maintained euglycemia for more than 30 days underwent nephrectomies to remove the islet-engrafted kidneys. Within 2 to 4 days after nephrectomy, all four mice returned to the diabetic state (average blood glucose level, 485 mg/dL). These results clearly indicated that insulin was released from the transplanted islets in the kidney. The longest islet graft survival reached more than 130 days, at which point they were terminated for assessments. Histology of these kidneys showed implanted islets in the subrenal capsules. Immunohistochemical staining with anti-insulin antibody demonstrated that insulin-positive cells in islet grafts could be detected even up to 130 days in LSF-treated recipients, suggesting the ability to produce insulin in the transplanted β cells (Fig. 4).
Figure 4: Islet graft histology. Islet-engrafted kidneys were collected from transplanted recipients and stained with hematoxylin-eosin (a, c, and e) or with an anti-insulin antibody (b, d, and f). (a and b) Obtained from an LSF-treated recipient 31 days after grafting; (c and d) an LSF-treated recipient 130 days posttransplant; and (e and f) a saline-treated recipient, at day 10 after islet engraftment. (arrows) Grafted islets. Similar observations were seen in the same treatment groups (magnification ×200–400).
LSF Treatment Reduced Serum Levels of Inflammatory Cytokines in Recipients
Previous studies have demonstrated the ability of LSF to reduce several inflammatory cytokines in infectious and autoimmune conditions (3,8–10). In this study, we have tested this effect of LSF in islet recipient mice in vivo. Daily LSF treatment in recipient mice was initiated on the day of transplant and continued for 3 weeks. As compared with normal saline-treated recipients, LSF treatment reduced serum levels of IFN-γ and TNF-α significantly. Reduction of the cytokines was seen beginning 1 week after treatment and lasted for 3 weeks. The rebound of IFN-γ and TNF-α production after LSF withdrawal was similar to the levels seen in young nondiabetic NOD mice but still below the average level of saline-treated NOD recipients (Fig. 5). However, we found no difference in the levels of IL-4 and IL-10 between LSF- and saline-treated groups (data not shown).
Figure 5: Serum levels of cytokines in NOD mice. Serum samples were collected from nondiabetic NOD female mice at the age of 5 weeks or from islet transplant recipients at 1, 3, and 4 weeks after engraftment and treatment with saline or LSF, respectively. Four mice were used for each group. Murine TNF-α (A) and IFN-γ (B) were measured by ELISA kits with their standards. P <0.05 between saline- and LSF-treated recipients in IFN-γ and TNF-α levels.
DISCUSSION
Autoimmunity destroys grafted islets in type 1 diabetic recipients, even when allogeneic rejection is controlled by immunosuppressive treatment. The differences between allogeneic immune rejection and autoimmune-mediated destruction in islet transplantation are not fully understood. In a human study, it was clear that recurrent autoimmunity existed after islet engraftment even when allogeneic reactivity was controlled (6). This observation suggests that the islet-specific autoimmune destruction is present in type 1 diabetic recipients. However, the mechanism of this phenomenon needs to be defined.
Syngeneic islet grafts are destroyed in NOD recipients shortly after engraftment if there is no immunosuppressive therapy. Many studies reported success in prevention of autoimmune diabetic recurrence after islet transplantation (15–19). Each study took different approaches to protect islets from autoimmune destruction, such as administration of immunosuppressants, antibodies, recombinant proteins, virus-mediated gene therapy, or genetic manipulation. In this article we report a new approach using a short-term treatment with LSF, a novel anti-inflammatory agent to prevent autoimmune-mediated islet destruction in transplantation. LSF works by means of a separate mechanism that is associated with suppressing Th1 cell activation and reducing inflammatory cytokine production. LSF generally is an anti-inflammatory compound, not an immunosuppressant. Using LSF does not compromise host innate defenses (20–22), as other immunosuppressants may.
In this study, we have chosen autoimmune diabetic NOD mice to test the potential of suppressing islet autoimmune destruction in transplantation by LSF. LSF has been used in human clinical trials for bone marrow transplant patients and shown to be well tolerated in humans (23) and mice (3). When we used LSF to treat the NOD recipients, we saw a prolonged islet graft survival. Our in vitro study also suggested that LSF treatment reduced islet apoptosis, increased β-cell resistance to inflammatory cytokines, and preserved insulin secretory function (3,12,24). In a mechanism study using β-cell lines, we found that LSF promoted mitochondrial metabolism and increased β-cell intracellular adenosine triphosphate production, leading to reduction of β-cell death against multiple inflammatory cytokine effects, and enhancement of β-cell glucose responsiveness and insulin secretory ability (12). The unique combination of suppressed Th1 activity and improved β-cell function is the likely explanation for the effects seen with LSF.
Immune cellular activation and inflammatory cytokine production contribute to autoimmune diabetes recurrence. Studies have already demonstrated that LSF has several biologic effects. LSF suppresses Th1 T-cell differentiation and activation (8,10), reduces inflammatory cytokine production (such as IFN-γ and TNF-α) (9), decreases neutrophil migration and degranulation (25), and modifies macrophage migration (3) (Z.Y., unpublished observations). Splenocytes from LSF-treated NOD mice failed to induce diabetes in NODscid recipients, suggesting that LSF may cause changes in cellular function (3). Such changes in splenocyte function may result in immunoregulation to suppress diabetes development. In the study using the LSF-treated NOD splenocytes, we have found that signal transducer and activator of transcription (STAT)-4 activation was suppressed and IFN-γ production was reduced. Further analysis has shown that LSF’s protection was associated with suppression of STAT-4–mediated IL-12 signaling (24). Blockade of STAT-4 activation leads to interruption of IL-12 intracellular signaling and suppression of IL-12 activity (26). Interruption of IL-12 signaling reduces Th cell differentiation to Th1 type and decreases natural killer cell activity. Suppressed T-cell and natural killer cell activation reduce IFN-γ and other cytolytic cytokine production (27). Both T-cell activation and Th1 type cytokines could cause β-cell dysfunction. Therefore, LSF treatment may suppress T-cell functions by reducing STAT-4 activation, reducing inflammatory cytokine production, and providing a direct effect to protect grafted islets. Our experimental results strongly suggest that LSF could reduce islet apoptosis in the presence of inflammatory cytokines (3,12,24), but the direct association between protection of islets and inhibition of apoptosis by LSF needs to be confirmed in an in vivo model.
Prevention of autoimmunity in type 1 diabetes can be achieved by various mechanisms. Suppression of Th1-cell function and inflammatory cytokines or enhancement of Th2- cell activation and Th2 cytokine levels can protect NOD mice from autoimmune diabetes. The use of an IL-12 antagonist could drive pancreatic infiltrating cells to Th2 differentiation, preventing autoimmune diabetes (28). Several anti-inflammatory drugs, such as pentoxifylline and rolipram, could inhibit IL-12 and IFN-γ production to prevent diabetes in NOD mice (4). However, LSF, unlike these agents, does not lead to a deviation of T cells to the Th2 type, as demonstrated by no change in Th2 type cytokines (IL-4 and IL-10). LSF has no direct effect on IL-12 production and gene expression; rather, it blocks the signal transduction and transcription (by means of the STAT-4) pathway (3,8,29). Because the IFN-γ–to–IL-4 ratio is more reliable for evaluating therapeutic outcome in autoimmune disorders (30), reduction of IFN-γ by LSF decreases the IFN-γ–to–IL-4 ratio even when there is no increase in IL-4 production.
In autoimmune recipients, allogeneic and autoimmune reactions are mixed when allogeneic grafts are transplanted. To demonstrate the autoimmune response clearly, we have used major histocompatibility complex-matched NODscid mice as islet donors for diabetic NOD recipients. Therefore, allogeneic rejection and donor-recipient chimerism may not be a major issue here, because both T- and B-cell function are deficient in NODscid mice. This model may provide an experimental system with which to test autoimmune responses. In clinical practice, however, most transplantation is conducted in allogeneic combinations. Therefore, further studies of LSF’s effect in allogeneic protection are needed.
CONCLUSION
From both in vitro and in vivo studies, LSF’s protection in autoimmune diabetic recurrence is the result of prevention of islet death, enhancement of β-cell insulin secretion, suppression of immune T-cell activation, and reduction of inflammatory cytokines. Our study suggests that anti-inflammatory agents, such as LSF, may have the potential to be used as adjunctive agents in current regimens of immunosuppression for islet transplantation.
Acknowledgments.
The authors thank Marcia McDuffie, M.D., University of Virginia, for helpful discussion; Drs. M. E. Coon and J. W. Singer, Cell Therapeutics, for LSF supply; and Hongyu Fang and Runpei Wu for technical support.
REFERENCES
1. Tisch R, McDevitt H. Insulin dependent diabetes mellitus. Cell 1996; 85: 291.
2. Rabinovitch A. Immunoregulatory and cytokine imbalances in the pathogenesis of IDDM. Diabetes 1994; 43: 613.
3. Yang Z, Chen M, Wu R, et al. The anti-inflammatory compound lisofylline prevents type 1 diabetes in non-obese diabetic mice. Diabetologia 2002; 45: 1307.
4. Liang L, Beshay E, Prud’homme GJ. The phosphodiesterase inhibitors pentoxifylline and rolipram prevent diabetes in NOD mice. Diabetes 1998; 47: 570.
5. Shapiro AM, Lakey JR, Ryan EA, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 2000; 343: 230.
6. Stegall MD, Lafferty KJ, Kam I, et al. Evidence of recurrent autoimmunity in human allogeneic islet transplantation. Transplantation 1996; 61: 1272.
7. Santamaria P, Nakhleh RE, Sutherland DE, et al. Characterization of T lymphocytes infiltrating human pancreas allograft affected by isletitis and recurrent diabetes. Diabetes 1992; 41: 53.
8. Coon ME, Diegel M, Leshinsky N, et al. Selective pharmacologic inhibition of murine and human IL-12-dependent Th1 differentiation and IL-12 signaling. J Immunol 199; 163: 6567.
9. Rice GC, Rosen J, Weeks R, et al. CT-1501R selectively inhibits induced inflammatory monokines in human whole blood ex vivo. Shock 1994; 1: 254.
10. Bight JJ, Du C, Coon M, et al. Prevention of experimental allergic encephalomyelitis via inhibition of IL-12 signaling and IL-12-mediated Th1 differentiation: An effect of the novel anti-inflammatory drug lisofylline. J Immunol 1998; 161: 7015.
11. Bleich D, Chen S, Bursten SL, et al. Lisofylline, an inhibitor of unsaturated phosphatidic acid generation, ameliorates interleukin-1β–induced dysfunction in cultured rat islets. Endocrinology 1996; 137: 4871.
12. Chen M, Yang Z, Wu R, et al. Lisofylline, a novel antiinflammatory agent, protects pancreatic β-cells from proinflammatory cytokine damage by promoting mitochondrial metabolism. Endocrinology 2002; 143: 2341.
13. Rabinovitch A. Roles of cytokines in IDDM pathogenesis and islet β-cell destruction. Diabetes Rev 1993; 1: 215.
14. Sibley RK, Sutherland DER, Goetz F, et al. Recurrent diabetes mellitus in the pancreas iso- and allograft. Lab Invest 1985; 53: 132.
15. Suarez-Pinzon WL, Marcoux Y, Ghahary A, et al. Gene transfection and expression of transforming growth factor-beta 1 in nonobese diabetic mouse islets protects beta-cells in syngeneic islet grafts from autoimmune destruction. Cell Transplant 2002; 11: 519.
16. Casteels K, Waer M, Laureys J, et al. Prevention of autoimmune destruction of syngeneic islet grafts in spontaneously diabetic nonobese diabetic mice by a combination of a vitamin D3 analog and cyclosporine. Transplantation 1998; 65: 1225.
17. Shapiro AM, Suarez-Pinzon WL, Power R, et al. Combination therapy with low dose sirolimus and tacrolimus is synergistic in preventing spontaneous and recurrent autoimmune diabetes in non-obese diabetic mice. Diabetologia 2002; 45: 224.
18. Uchikoshi F, Yang Z, Rostami S, et al. Prevention of autoimmune recurrence and rejection by adenovirus mediated CTLA4Ig gene transfer to the pancreatic graft in BB rat. Diabetes 1999; 48: 652.
19. Panerai AE, Nicoletti F, Sacedote P, et al. MFP14, a multifunctional emerging protein with immunomodulation properties, prevents spontaneous and recurrence autoimmune diabetes in NOD mice. Diabetologia 2001; 44: 839.
20. Rice GC, Brown PA, Nelson RJ, et al. Protection from endotoxic shock in mice by pharmacologic inhibitor of phosphatidic acid. Proc Natl Acad Sci USA 1994; 91: 3857.
21. Cicalese L, Yacoub W, Rogers J, et al. Translocation of bacteria from the gastrointestinal tract: Protection afforded by lisofylline. Transplant Proc 1999; 31: 575.
22. Bursten S, Weeks R, West J, et al. Potential role for phosphatidic acid in mediating the inflammatory responses to TNF alpha and IL-1 beta. Circ Shock 1994; 44: 14.
23. List AF, Maziarz R, Stiff P, et al. A randomized placebo-controlled trial of lisofylline in HLA-identical, sibling-donor, allogeneic bone marrow transplant recipients. Bone Marrow Transplant 2000; 25: 283.
24. Yang Z, Chen M, Fialkow LB, et al. The novel anti-inflammatory compound, lisofylline, prevents diabetes in multiple-low dose streptozocin-treated mice. Pancreas 2003; 26: E99.
25. Oka Y, Hasegawa N, Nakayama M, et al. Selective downregulation of neutrophils by a phosphatidic acid generation inhibitor in a porcine sepsis model. J Surg Res 1999; 81: 147.
26. Kaplan MH, Sun YL, Hoey T, et al. Impaired IL-12 responses and enhanced development of Th2 cells in STAT4-deficient mice. Nature 1996; 382: 174.
27. Thierfelder WE, van Deursen JM, Yamamoto K, et al. Requirement for STAT4 in interleukin-12-mediated responses of natural killer and T cells. Nature 1996; 382: 171.
28. Trembleau S, Penna G, Gregori S, et al. Deviation of pancreas-infiltrating cells to Th2 by interleukin-12 antagonist administration inhibits autoimmune diabetes. Eur J Immunol 1997; 27: 2230.
29. Yang Z, Chen M, Fialkow LB, et al. Inhibition of STAT4 activation by lisofylline is associated with the protection of autoimmune diabetes. Diabetes Metab Res Rev 2002; 18: 31.
30. Ghoreschi K, Thomas P, Breit S, et al. Interleukin-4 therapy of psoriasis induces Th2 responses and improves human autoimmune disease. Nat Med 2003; 9: 40.
