help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2003-1274
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
145/2/627    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, W.
Right arrow Articles by Delovitch, T. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, W.
Right arrow Articles by Delovitch, T. L.
Endocrinology Vol. 145, No. 2 627-638
Copyright © 2004 by The Endocrine Society

Insulin-Like Growth Factor (IGF)-I/IGF-Binding Protein-3 Complex: Therapeutic Efficacy and Mechanism of Protection against Type 1 Diabetes

Wei Chen, Konstantin V. Salojin, Qing-Sheng Mi, Marsha Grattan, T. Craig Meagher, Peter Zucker and Terry L. Delovitch

Autoimmunity/Diabetes Group (W.C., K.V.S., Q.-S.M., M.G., T.C.M., P.Z., T.L.D.), Robarts Research Institute, London, Ontario N6G 2V4, Canada; and the Department of Microbiology and Immunology (T.C.M., T.L.D.), University of Western Ontario, London, Ontario N6A 5C1, Canada

Address all correspondence and requests for reprints to: Dr. Terry L. Delovitch, Director, Autoimmunity/Diabetes Group, Robarts Research Institute, 1400 Western Road, London, Ontario N6G 2V4, Canada. E-mail: del{at}robarts.ca.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
IGF-I regulates islet ß-cell growth, survival, and metabolism and protects against type 1 diabetes (T1D). However, the therapeutic efficacy of free IGF-I may be limited by its biological half-life in vivo. We investigated whether prolongation of its half-life as an IGF-I/IGF binding protein (IGFBP)-3 complex affords increased protection against T1D and whether this occurs by influencing T cell function and/or islet ß-cell growth and survival. Administration of IGF-I either alone or as an IGF-I/IGFBP-3 complex reduced the severity of insulitis and delayed the onset of T1D in nonobese diabetic mice, but IGF-I/IGFBP-3 was significantly more effective. Protection from T1D elicited by IGF-I/IGFBP-3 was mediated by up-regulated CCL4 and down-regulated CCL3 gene expression in pancreatic draining lymph nodes, activation of the phosphatidylinositol 3-kinase and Akt/protein kinase B signaling pathway of ß-cells, reduced ß-cell apoptosis, and stimulation of ß-cell replication. Reduced ß-cell apoptosis resulted from elevated Bcl-2 and Bcl-XL activity and diminished caspase-9 activity, indicating a novel role for a mitochondrial-dependent pathway of ß-cell death. Thus, IGF-I/IGFBP-3 affords more efficient protection from insulitis, ß-cell destruction, and T1D than IGF-I, and this complex may represent an efficacious therapeutic treatment for the prevention of T1D.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
TYPE 1 DIABETES (T1D) is an autoimmune disorder resulting from the destruction of insulin-producing ß-cells of pancreatic islets (1). The onset of spontaneous T1D in humans and nonobese diabetic (NOD) mice is preceded by a progressive leukocyte infiltration into islets (insulitis), which persists for a relatively long period of time before eventual massive islet ß-cell destruction. Compelling evidence suggests that apoptosis is the principle mode of ß-cell death during the development of T1D (2, 3, 4). Cytotoxic cytokines secreted by infiltrating leukocytes, such as TNF-{alpha}, interferon-{gamma} (IFN-{gamma}), and IL-1ß, are major mediators of ß-cell apoptosis via multiple signaling pathways (5, 6, 7, 8, 9, 10, 11, 12). Traditionally, attempts to prevent ß-cell destruction and T1D have involved immunological strategies and ß-cell autoantigens to induce tolerance; however, it has become apparent that a number of approaches aimed at preserving ß-cell functions may also have therapeutic relevance in preventing the pathogenesis of T1D. Currently, an approach receiving much attention is the blockade of apoptosis signaling pathways that result in increased islet ß-cell survival and function (13, 14).

IGF-I, a single-chain peptide produced in many tissues, is abundant in the circulation as a predominant growth factor in regulating growth, survival, and metabolism (15). Recent studies show that IGF-I has potent antiapoptotic activity in many cell types. Previously, we demonstrated that local IGF-I secretion is inhibited in infiltrated islets and that exogenous IGF-I blocks the in vitro cytokine-induced apoptosis of islet cells from prediabetic NOD mice (16). IGF-I also protects islets against cytokine-mediated inhibition of insulin secretion, induction of nitric oxide synthesis, and cell apoptosis in vitro (17, 18, 19). In addition, human islets transfected with the IGF-I gene are resistant to IL-1ß-induced and Fas-mediated apoptosis (20). Furthermore, administration of human recombinant IGF-I to prediabetic NOD mice reduces the severity of insulitis and incidence of T1D in an adoptive transfer model of T1D (21). Nonetheless, the mechanism(s) by which IGF-I protects from T1D remain to be determined.

Treatment of humans with IGF-I has been shown to be of clinical value in the efficient control of many acute inflammatory conditions and improvement of metabolic conditions, including T1D (15, 22, 23). However, systemic administration of IGF-I may cause hypoglycemia because of a cross-reaction of IGF-I with insulin receptors. Furthermore, administration of uncoupled IGF-I has restricted therapeutic potential due to its short half-life, instability in the circulation, and several adverse effects (15). To overcome these limitations of IGF-I, a stable complex of IGF-I and IGF-I binding protein-3 (IGFBP-3) was synthesized. Systemic administration of the IGF-I/IGFBP-3 complex results in a 6-h longer systemic exposure to IGF-I due to its extended half-life in the circulation and slows the release of IGF-I from the vascular compartment that prevents the induction of hypoglycemia in animal models of T1D (24). Studies of IGF-I/IGFBP-3 in an experimental model of multiple sclerosis and human T1D (25, 26) demonstrated that administration of IGF-I/IGFBP-3 is not only safer but also has a higher therapeutic index than free IGF-I. Thus, IGF-I/IGFBP-3 may provide an improved mode of delivery of IGF-I that will facilitate in vivo functional studies and potential clinical uses.

In the present study, we compared the effects of IGF-I and IGF-I/IGFBP-3 treatments on the incidence of T1D in NOD mice. In particular, we explored the optimal dose and treatment regimen of IGF-I/IGFBP-3 for preventing T1D. Due to the expression of IGF-I receptors on many cell types, including lymphocyte subpopulations and islet ß-cells, we evaluated the effect of administration of IGF-I/IGFBP-3 on both systemic immune responses and islet ß-cells in vivo. Finally, we investigated signaling pathway intermediates and signaling pathways involved in protective mechanisms mediated by IGF-I by in vitro functional analyses of freshly isolated primary islet ß-cells and ß-cell lines. Recent clinical studies of IGF-I and IGF-I/IGFBP-3 have proven their beneficial effects on many diseases, including type 1 and type 2 diabetes (26, 27). Our findings provide new insight into the mechanisms of ß-cell apoptosis via the mitochondrial caspase-9-dependent pathway and the potential of IGF-I/IGFBP-3 as a prophylactic therapy in the prevention of autoimmune T1D in humans.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mice
NOD/Del and NOD.Scid mice were bred in a barrier facility at the Robarts Research Institute and were housed under specific pathogen-free conditions in the animal facility of University of Western Ontario for the duration of all experiments. In our colony of female NOD mice, the onset of insulitis and T1D occurs at 4–5 wk and 12 wk of age, respectively. The cumulative incidence of T1D is 75–85% by 35 wk of age. NOD.Scid mice were generously provided by Dr. L. Shultz (The Jackson Laboratory, Bar Harbor, ME). All procedures were performed under an approved protocol in compliance with guidelines set by the Animal Studies Committee of the University of Western Ontario.

Cell lines and reagents
The NIT-1 cell line was purchased from American Type Culture Collection (Manassas, VA), and cultured in Ham’s F-12K medium containing 10% fetal calf serum, 2 mM glutamine, and penicillin-streptomycin (Invitrogen, Inc., Burlington, Canada). Recombinant human IGF-I and IGF-I/IGFBP-3 complex (SomatoKine) used in this study were kindly provided by Genentech Inc. (San Francisco, CA) and Insmed Inc. (Glen Allen, VA), respectively. Recombinant human IGF-II was purchased from GroPep Inc. (Adelaide, Australia). Recombinant mouse TNF-{alpha} and IFN-{gamma} were purchased from BD Bioscience (Mississauga, Canada). Cyclophosphamide (CY), the phosphatidylinositol 3-kinase (PI3-K) inhibitor LY294002 [2-(4-morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloride], and 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) were purchased from Sigma Chemical Co. (St. Louis, MO).

Islet isolation and culture
Islets were isolated from 4- to 6-wk-old male NOD mice as previously described (28). Distended pancreata were excised from anesthetized mice after intrabile duct collagenase digestion. Tissue was incubated in a 50-ml conical tube at 37 C for 10 min with agitation and digestion stopped with ice-cold Hanks’ basic salt solution. Islet tissue was filtered through a steel mesh (860 µm) and resuspended in dextran solution after washing twice with cold Hanks’ basic salt solution. The islet preparation was recovered after three-layer discontinuous dextran gradient centrifugation, and individual islets were hand picked to a purity of 95%. Purified islets were precultured in Ham’s F-12K medium at 37 C in 5% CO2 overnight before exposure to cytokines.

Treatment of mice with IGF-I/IGFBP-3 complex and free IGFs
Littermate-matched female NOD mice (2–8 wk old) were randomly divided into treatment groups that included recipients of either IGF-I, IGF-II (diluted in PBS + 1% NOD mouse serum), or IGF-I/IGFBP-3 as well as a control group that received vehicle (PBS + 1% NOD mouse serum). Treatment reagents were injected sc for 4 wk at the various doses, and the regimens are indicated in the figure legends.

Histological analysis of insulitis
The incidence and severity of insulitis and insulin immunostaining were analyzed using formalin-fixed (10%) and paraffin-embedded pancreatic tissues. After hematoxylin and eosin staining, insulitis scores were graded based on the degree of lymphocytic infiltration as follows: G0, normal; G1, periinsulitis (mononuclear cells surrounding islets and ducts but no infiltration of the islet architecture); G2, moderate insulitis (mononuclear cells infiltrating less than 50% of the islet architecture); and G3, severe insulitis (more than 50% of the islet tissue infiltrated by lymphocytes, accompanied by a reduction in insulin staining). Three randomly obtained and nonadjacent sections of each pancreas from coded slides were scored independently by two blind observers. Insulitis index (I) was calculated using the following formula: I = [(number of islets with score of G0 x 0) + (number of islets with score of G1 x 1) + (number of islets with score of G2 x 2) + (number islets with score of G3 x 3)]/3 x (total number of islets).

Assessment of diabetes
Mice were monitored for the development of T1D by measuring blood glucose levels (BGLs) either daily or weekly, dependent on the experimental design. Mice with a reading of BGL > 11.1 mmol/liter (200 mg/dl) on two consecutive occasions were diagnosed as being positive for T1D.

Detection of apoptosis in situ
For measurement of apoptosis, the in situ cell death detection kit (Roche Diagnostics, Laval, Canada) was used as suggested by the manufacturer. Briefly, a reaction of terminal deoxynucleotidyltransferase-mediated UTP end labeling (TUNEL) was carried out on paraffin-embedded islet tissue sections. The sections were further stained with antiinsulin antibody. The number of TUNEL/insulin double-positive cells per section were counted and are shown as a number of apoptotic nuclei per 100 islets. A minimum of 20 islets per mouse were analyzed, and six mice per group were used.

Assessment of ß-cell death by an MTT assay
Pancreatic islets (50/well) or NIT-1 cells (2 x 105/well) were precultured overnight in 96-well plates and then exposed to TNF-{alpha} + IFN-{gamma} with or without IGF-I for the indicated times. IGF-I was added to cultures 1 h before exposure to cytokines, after which medium was removed and MTT (0.5 mg/ml) added during an incubation for 2 h at 37 C in 5% CO2. The cells were then centrifuged, supernatants were removed, and acidic isopropanol (0.04N HCl) was added to dissolve MTT crystals. Absorbance of converted dye was measured at a wavelength of 575 nm with background subtraction at 655 nm.

Cytokine and chemokine assays
Pancreatic tissues and spleens were isolated and snap frozen in liquid nitrogen. Tissue samples were homogenized, sonicated in protease inhibitor buffered cocktail (Roche Diagnostics), and then filtered. Filtrates were analyzed for cytokine and chemokine levels by ELISA (BD Biosciences) as described (29), and the results were normalized relative to the amount of total pancreatic tissue protein.

Assay of caspase-3 activity
Caspase-3 activation/cleavage was determined by flow cytometry after intracellular staining of islet cells with a phycoerythrin-conjugated monoclonal antibody that recognizes only the processed/cleaved caspase-3 subunit (BD Biosciences), according to the protocol supplied by manufacturer.

Analysis of ß-cell replication and islet size
For measurement of ß-cell replication, mice were injected ip with 5-bromo-2-deoxyuridine (BrdU) (Sigma), 50 mg/kg body weight at the end of treatment. Six hours later, the pancreas was extracted and embedded in paraffin. Sections were stained for insulin and BrdU (mouse anti-BrdU monoclonal antibody, BD Biosciences). The number of insulin-stained BrdU-positive nuclei per total nuclei was determined. At least 1000 islet ß-cell nuclei were counted in each pancreatic section, and six mice per group were analyzed.

Islet size and area (µm2) was determined by calculation of sections using a digital camera mounted with a light microscope and analysis of each islet using Northern Eclipse Imagine analyzer. At least five sections (>20 islets) were measured per mouse.

cDNA microarray and RT-PCR analysis
Total RNA was extracted from tissue samples with the RNeasy Midi kit (QIAGEN, Mississauga, Canada). The quality of RNA was determined using Agilent Technologies 2100 Bioanalyzer (Caliper Technologies Corp., Mountain View, CA). Aliquots of RNA (3–4 µg) were used to analyze the apoptosis gene expression profile in islets by GEArray technology (Mouse Apoptosis GEArray Q series, catalog No. MM-002N-4, SuperArray Inc., Bethesda, MD). Relative amounts of mRNA transcripts were quantified using a Molecular Imager System and Molecular Analyst imaging software (Bio-Rad, Hercules, CA). Gene transcript levels were normalized to those of glyceraldehyde-3-phosphate dehydrogenase. A relative signal difference of >=2-fold in signal intensity was considered significant. Cytokine and chemokine gene expression was analyzed by RT-PCR, as described (29). To analyze CCL3 (MIP-1{alpha}) and CCL4 (MIP-1ß) chemokine gene expression, the following primers were used: MIP-1{alpha}, 5'-TTCTGCTGACAAGCTCACCCTCTGT-3' and 5'-GTCCCTCGATGTGGCTACTTGGCA-3'; and MIP-1ß, 5'-TGAAGCTTCTGGGCCCTGCAGTCC-3' and 5'-TCAGTGAGAAGCATCAGGGCTGAG-3'.

Immunoblotting
Islets or cells were lysed in ice-cold modified RIPA buffer (50 mM Tris-HCl, pH 8.0, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 5 mM EDTA) supplemented with a mixture of protease and phosphatase inhibitors (100 µM p-nitrophenyl guanidinobenzoate, 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 5 µg/ml pepstatin, 2 mM Na3VO4, and 10 mM NaF) (all obtained from Sigma). Protein concentrations in each sample were determined with a protein assay kit (R&D Systems, Mississauga,Canada). Equal amounts of sample protein were solubilized in 2x Laemmli sample buffer, resolved by SDS-PAGE 8–16% gradient gel (Novex, San Diego, CA) under reducing conditions, transferred to nitrocellulose membrane (Schleicher & Schuell, Keene, NH), and immunoblotted with antibodies. To avoid sample loading errors, ß-actin expression was determined in the blots to adjust and normalize the amount of sample loaded. Anti-Akt1 (goat polyclonal), anti-Bad (rabbit polyclonal), anti-Bax (rabbit polyclonal), anti-Bcl-2 (mouse monoclonal IgG1), and anti-Bcl-XL (rabbit polyclonal) antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA); anti-Akt, anti-caspase-9 and anti-caspase-12 polyclonal antibodies were purchased from Cell Signaling Technology, Inc. (Beverly, MA). Signal intensities were quantified using a Molecular Imager System and Molecular Analyst imaging software (Bio-Rad) and are presented as OD/mm2.

Statistical analysis
Statistical analyses were performed using the Student’s t test, Fisher’s exact test, or the Mann-Whitney sum of ranks test depending on the experimental design. Differences were considered statistically significant when P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
IGF-I/IGFBP-3 treatment reduces the severity of insulitis
To test whether administration of IGF-I/IGFBP-3 regulates the extent of lymphocyte infiltration into pancreatic islets, IGF-I/IGFBP-3 (1 mg/kg twice per day) was administered as described for IGF-I (21). To rule out possible effects of IGF-I on glucose metabolism, the BGLs on fasting and 1-h post injection of IGF-I/IGFBP-3 were determined after three daily injections. The fasting BGLs were represented as the mean value ± SD and were found to be 5.5 ± 0.5 vs. 5.7 ± 0.5 mmol/liter between treated and control groups, respectively. One hour after injection, the BGLs of treated and control mice were 6.3 ± 0.7 and 6.7 ± 1.0 mmol/liter, respectively. No clinical signs of hypoglycemia were observed in the treated mice during the treatment period, indicating that administration of IGF-I/IGFBP-3 does not significantly influence glucose metabolism. Next, we compared the efficacy of IGF-I/IGFBP-3 vs. IGF-I or IGF-II treatment in reducing the severity of insulitis. Groups of 4- to 5-wk-old female NOD mice were injected sc with either IGF-I/IGFBP-3 (2 mg/kg per day) or the equivalent amount of IGF-I or IGF-II (0.4 mg/kg) twice a day for 4 wk. Recipients of IGF-I/IGFBP-3 showed a significantly higher percentage of intact islets (G0 = 55%) with less severe infiltration (G3 = 14%, I = 0.291; Fig. 1AGo) compared with vehicle-treated control mice (G0 = 30%, G3 = 24%, I = 0.461; P < 0.001). Mice treated with IGF-I alone showed a similarly lower insulitis score compared with control mice (Fig. 1BGo; I = 0.221 vs. I = 0.432, respectively; P < 0.01). Moreover, the insulitis scores observed upon IGF-I/IGFBP-3 or IGF-I treatment did not differ significantly from each other. However, mice receiving IGF-II exhibited similar insulitis scores to those of control mice (data not shown). These data suggest that treatment with IGF-I/IGFBP-3 and IGF-I, but not IGF-II, efficiently blocks islet inflammation to an equivalent extent during the development of T1D in NOD mice.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 1. Severity of insulitis is reduced in IGF-I/IGFBP-3- and IGF-I-treated NOD mice. Female NOD mice (4–5 wk old; n = 6–8 per group) were injected sc with either IGFI/IGFBP-3 (1 mg/kg/d) (A), IGF-I (0.2 mg/kg, twice per day) (B), or vehicle (PBS) for 4 wk. Pancreata were examined histologically and scored for insulitis as described in Materials and Methods. *, P < 0.01; **, P < 0.001 vs. control mice treated with vehicle only (Fisher’s exact test). Results of one of two reproducible experiments are shown.

 
IGF-I/IGFBP-3 treatment prevents T1D in NOD mice
To determine whether administration of IGF-I/IGFBP-3 protects NOD mice from T1D and to identify an optimal regimen of IGF-I/IGFBP-3 treatment, female NOD mice (4–5 wk old) were treated sc with various doses of IGF-I/IGFBP-3 for 4 wk and then monitored for their incidence of T1D. Treatment with IGF-I/IGFBP-3 at a dose of 2 mg/kg/d, but neither 0.2 nor 1 mg/kg/d, significantly reduced the incidence of T1D by 50% compared with the control group (P < 0.05) (Fig. 2AGo). An increase in dose of IGF-I/IGFBP-3 to 4 mg/kg/d showed no further beneficial effect (25% incidence vs. 50% incidence in control mice at 24 wk of age; n = 8–10 mice per group). Surprisingly, 80% of mice that received IGF-I/IGFBP-3 at 1 mg/kg every other day displayed the lowest incidence of T1D (<=20%) up to 35 wk of age. The latter treatment protocol may represent an optimum therapeutic regimen, because use of this protocol beginning at either 2 or 8–10 wk of age did not further reduce the incidence of TID in NOD mice (our unpublished observations).



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 2. IGF-I/IGFBP-3 protects against T1D more efficiently than IGF-I in NOD mice. A, female NOD mice (4–5 wk old; n = 10 per group) received various doses of IGF-I/IGFBP-3 sc for 4 wk and remained untreated thereafter. B, Female NOD mice (4–5 wk old) were either treated with IGF-I [0.2 mg/kg/every other day (b.o.d.)] or vehicle for 4 wk or were left untreated. C, Female NOD mice (4 wk old) were treated sc with IGF-I/IGFBP-3 (2 mg/kg/d or 1 mg/kg/b.o.d) or vehicle for 4 wk and were then challenged with CY (300 mg/kg ip) at 10 wk of age. The cumulative incidence of T1D was monitored. *, P < 0.05; **, P < 0.01 vs. control mice treated with vehicle only (Mann-Whitney sum of ranks test). Results of one of two reproducible experiments are shown.

 
To explore whether a similar protocol of IGF-I treatment prevents the onset of T1D, we monitored the incidence of T1D in NOD mice either treated from 4 wk of age with IGF-I (0.4 mg/kg/d, twice a day) or vehicle or left untreated. Disease onset was delayed for only approximately 4 wk in IGF-I-treated compared with vehicle-treated or untreated mice, and the incidence of T1D was similar in all three treatment groups at 32 wk of age (P > 0.05) (Fig. 2BGo). Thus, IGF-I/IGFBP-3 appears to afford more efficient protection from T1D than free IGF-I.

Further evidence in support of the use of IGF-I/IGFBP-3 to protect against T1D was obtained from analyses of CY-induced T1D in NOD mice. Mice treated with IGF-I/IGFBP-3 at 2 mg/kg/d showed a 2-wk delay in disease onset compared with vehicle-treated control mice, but a high cumulative incidence of T1D (55 and 75%, respectively) was observed in both groups (Fig. 2CGo). In contrast, administration of IGF-I/IGFBP-3 at 1 mg/kg/d every other day reduced the incidence of T1D to 30% (P < 0.01), demonstrating that IGF-I/IGFBP-3 treatment can also prevent T1D in an accelerated and more aggressive model of the disease.

IGF-I/IGFBP-3 treatment may protect from T1D by providing an antiinflammatory environment in the pancreas rather than activating regulatory T cells
We have shown that IL-4-induced polarization of T helper-1 to T helper-2 cell responses is associated with the protection from T1D in NOD mice (30, 31). In addition, we reported that increased expression of the CCL3 chemokine in the pancreas is associated with progression to T1D, whereas increased CCL4 expression in the pancreas leads to protection from T1D in NOD mice (32). Because activated T cells express IGF-I receptors (33), it is possible that protection from T1D by IGF-I/IGFBP-3 is mediated by the modulation of activated T cell responses to islet ß-cells. To test this possibility, we initially examined the effects of IGF-I/IGFBP-3 treatment on the expression of IFN-{gamma}, IL-4, CCL3, and CCL4 in the spleen and pancreas of female NOD mice. After a 4-wk treatment with IGF-I/IGFBP-3, decreased CCL3 and increased CCL4 mRNA expression was observed in the pancreatic lymph nodes (PLNs) but not in the islets (Fig. 3AGo) or spleen (data not shown) of treated mice relative to that detected in vehicle-treated control mice. A modest reduction (15% decrease vs. control; P < 0.05) in CCL3 but not CCL4 protein expression was noted in the pancreas (Fig. 3BGo). Similarly, IGF-I/IGFBP-3 treatment significantly reduced the level of IFN-{gamma} protein expression in the pancreas (Fig. 3CGo) but did not affect the level of IL-4 expression (Fig. 3DGo). These findings are consistent with the notion that IGF-I/IGFBP-3 treatment may elicit a T helper-2 cell-like regulatory T cell environment in the pancreas that preserves ß-cell survival and function. However, our adoptive T cell transfer studies demonstrated that splenic T cells from IGF-I/IGFBP-3-treated mice do not transfer protection from T1D in NOD.Scid recipients (Fig. 3EGo), suggesting that regulatory T cells are not induced by our IGF-I/IGFBP-3 treatment protocol. Thus, the protection from T1D conferred by IGF-I/IGFBP-3 may be partially related to the induction of an antiinflammatory milieu rather than activation of regulatory T cells, implying that IGF-I/IGFBP-3 may have more direct effects on the survival and function of islet ß-cells than T cells.



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 3. IGF-I/IGFBP-3 treatment protects from T1D by stimulating an antiinflammatory environment in the pancreas of NOD mice. A and B, RT-PCR analysis of CCL3 and CCL4 gene expression in NOD PLNs and islets. Total tissue RNA was pooled from six mice per group after IGF-I/IGFBP-3 treatment (2 mg/kg/d). PCRs were run on an ethidium bromide-stained 1.5% agarose gel and amplification products visualized and quantified using a molecular imager as described in Materials and Methods. Data are represented as mean ± SD of intensities of RT-PCR signals for chemokine mRNA, as determined by densitometry of cDNA followed by normalization to the glyceraldehyde-3-phosphate dehydrogenase signals in individual samples of three experiments. *, P < 0.05; **, P < 0.01 compared with controls. C–E, Chemokine and cytokine levels in the pancreas after treatment. Intrapancreatic concentrations of CCL3 and CCL4 (C), IFN-{gamma} (D), and IL-4 (E) were determined by ELISA. Data represent the mean concentrations ± SD in six mice per group. *, P < 0.05 vs. controls. F, Cumulative incidence of T1D in NOD.Scid recipient mice. Female NOD.Scid recipients were injected ip with diabetogenic spleen T cells (5 x 106) and an equal number of spleen T cells from either IGF-I/IGFBP-3- or vehicle-treated female NOD mice. Results of one of two reproducible experiments are shown. A Student’s t test was used to analyze statistical differences between the above data sets.

 
IGF-I/IGFBP-3 reduces ß-cell apoptosis in NOD mice
To assess whether IGF-I/IGFBP-3 regulates ß-cell survival in treated mice, an in situ TUNEL assay was used to evaluate the levels of ß-cell apoptosis in pancreatic tissue sections obtained from NOD mice treated with IGF-I/IGFBP-3 or vehicle control. Islets from mice that received IGF-I/IGFBP-3 displayed a 2.5-fold reduction (P < 0.02) in the number of intra-islet apoptotic nuclei relative to control mice (Fig. 4AGo). Positive staining for insulin identified the apoptotic nuclei to be localized in ß-cells. To further determine whether IGF-I/IGFBP-3 rescues ß-cells from apoptosis, primary islets were cultured in vitro with apoptosis-inducing cytokines and the effects of IGF-I on ß-cell death were monitored. Our preliminary studies indicated that IGF-I, in contrast to IGF-I/IGFBP-3, does not function efficiently in vivo as a complex when cultured with either islets or ß-cells. This result suggested that there is a lack of function of specific proteases that release free IGF-I from the complex; such proteases are abundant in serum and blood vessels (34). In addition, we found that IGFBP-3 alone or in combination with IGF-I did not alter the bioactivity of islet ß-cells. Thus, free IGF-I was used for the following in vitro studies to explore whether the protection from T1D is mediated by an IGF-I/IGFBP-3 complex. TNF-{alpha} and IFN-{gamma} were chosen for study, as these cytokines mediate insulitis and elicit ß-cell apoptosis (6, 35, 36, 37). In vitro exposure of primary islet ß-cells to TNF-{alpha} + IFN-{gamma} in the absence of IL-1ß for 48 h efficiently induced apoptosis in approximately 50% of the cells (Fig. 4BGo), as estimated by an MTT assay (38). Addition of IGF-I to these cultures reduced the percentage of apoptotic islet cells to approximately 25%, indicating their greater resistance to cytokine-induced cell death (P < 0.01). Next, we examined whether cytokine-induced cell death occurs through activation of caspase-3, a mediator and early indicator of apoptosis (33). After exposure to TNF-{alpha} + IFN-{gamma} for 24 h, 35% of islet cells stained positive for caspase-3 compared with 10% for control samples. Upon addition of IGF-I, the number of caspase-3-positive islet cells was reduced to approximately 20%. These data further support the idea that ß-cell death induced by TNF-{alpha} + IFN-{gamma} occurs via apoptosis and that IGF-I blocks this apoptosis and results in ß-cell survival. Thus, administration of IGF-I/IGFBP-3 may afford protection from T1D by preventing the apoptosis of ß-cells induced by cytokines expressed during disease development.



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 4. IGF-I inhibits islet ß-cell apoptosis. A, analysis of apoptosis in IGF-I/IGFBP-3-treated mice. The number of insulin/TUNEL double-positive nuclei (mean ± SD) was determined in the pancreas of IGF-I/IGFBP-3-treated (black bars) and vehicle-treated (white bars) mice (n = 6/group) before and after a 4-wk treatment (2 mg/kg/d, sc) as described in Materials and Methods. Results are expressed as the mean values ± SD; *, P < 0.01 vs. vehicle-treated mice (Student’s t test). B, IGF-I reduces cytokine-induced cell death of primary islets. Islets were cultured with TNF-{alpha} + IFN-{gamma} (10 ng/ml each) in the presence or absence of IGF-I (100 ng/ml) for 48 h. An MTT assay was used to assess cell viability in cultures. Results are expressed as mean value ± SD of the percentage of viable cells normalized to cells cultured in medium alone taken to be 100% viable. Data from three independent experiments are shown. *, P < 0.05 vs. cytokine treatment alone; **, P < 0.005 vs. medium control (Student’s t test). C, Caspase-3 activity in islet cells. Islets were treated as in B for 24 h. Caspase-3 activity in islet cells was determined by FACS analysis. Results are expressed as mean values ± SD from three independent experiments. *, P < 0.05 vs. cytokine treatment alone; **, P < 0.01 vs. medium control (Fisher’s exact test).

 
Regulation of Bcl-2 family members and caspase-9 activity by IGF-I contributes to inhibition of ß-cell apoptosis
Bcl-2 protein family members possesses both pro- and antiapoptotic functions in a mitochondrial-mediated pathway of ß-cell apoptosis (13, 39, 40). To elucidate the mechanism(s) of how IGF-I inhibits ß-cell apoptosis induced by cytokines, we examined the expression of Bcl-2 family members and caspase-9 activity in the NIT-1 ß-cell line and primary islets after exposure to TNF-{alpha} + IFN-{gamma} ± IGF-I in vitro. NIT-1 cells exposed to TNF-{alpha} + IFN-{gamma} express a 2-fold increase in proapoptotic Bax as revealed in immunoblots (Fig. 5AGo, lane 3), suggesting that Bcl-2 family members regulate islet ß-cell homeostasis. Addition of IGF-1 (100 ng/ml) significantly reduced the level of Bax expression (8-fold) and increased the level of Bcl-XL expression (2-fold) relative to that observed for NIT-1 cells exposed to TNF-{alpha} + IFN-{gamma} alone (Fig. 5AGo, lane 4). This stimulation of Bcl-XL expression and significant reduction of Bax expression was not observed when IGF-I was added at 1000 ng/ml (Fig. 5AGo, lane 5). Similar to Bax, Bad expression was normal and efficiently phosphorylated in NIT-1 cells cultured for 24 h in medium alone (Fig. 5BGo, lane 1). Exposure to IGF-I alone reduced the expression of Bad and maintained its level of phosphorylation (Fig. 5BGo, lane 2). In contrast, a normal level of Bad expression was maintained (relative to medium alone), but Bad phosphorylation was significantly diminished in cells exposed to TNF-{alpha} + IFN-{gamma} (Fig. 5BGo, lane 3). Addition of IGF-I to cultures containing TNF-{alpha} + IFN-{gamma} significantly reduced the level of Bad expression and augmented the level of Bad phosphorylation (Fig. 5BGo, lane 4).



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 5. Effects of IGF-I on Bcl-2 family and caspase-9 activity during ß-cell apoptosis induced by cytokines. A–C, NIT-1 cells (2 x 106 cells/per sample) (A and B) or islets (200 per sample) (C) were treated with IGF-I in the presence or absence of TNF-{alpha} + IFN-{gamma} (10 ng/ml each). After 24 h, cell lysates (50 µg of protein) were prepared, resolved by SDS-PAGE, and immunoblotted (IB) with antibodies against Bax, Bcl-XL, Bad, phospho-Bad (P-Bad), Bcl-2, ß-actin, and caspase-9. D, Caspase-9 activity was assayed in NIT-1 cells. Recombinant Bad (rBad) and recombinant phosphorylated-Bad (rP-Bad) were loaded as positive controls. The positions of migration of uncleaved (p49) and cleaved (p39) caspase-9 are indicated by arrows. IGF-I was used at either 100 ng/ml (lanes 2 and 4) or 1000 ng/ml (*, lane 5). The relative signal intensity (OD value) of each band was quantified using a Molecular Imager System and Molecular Analyst imaging software as described in Materials and Methods.

 
We next examined whether similar mechanisms mediate the apoptosis of primary islet ß-cells. NOD mouse islet cells were cultured for 48 h under the same conditions used for NIT-1 cells. IGF-I-stimulated islet cells expressed 4-fold more Bcl-2 and 4-fold more Bcl-XL (Fig. 5CGo, lane 2) compared with control islet cells (Fig. 5CGo, lane 1). Similar increases in Bcl-2 and Bcl-XL expression were detected for islet cells stimulated in the presence of TNF-{alpha} + IFN-{gamma} (Fig. 5CGo, lane 3), suggesting that antiapoptotic molecules may be expressed as a compensatory measure in islets cells stressed under these conditions. It is noteworthy that IGF-I stimulation further enhanced both Bcl-2 (2-fold) and Bcl-XL (2.3-fold) expression in islet cells exposed to only TNF-{alpha} + IFN-{gamma} (Fig. 5CGo, lane 4). These results emphasize the potential importance of the mitochondrial-dependent pathway of caspase activation in TNF-{alpha} + IFN-{gamma}-induced apoptosis of ß-cells and suggest that IGF-I may promote islet ß-cell survival by regulating the levels of Bcl-2 family members and down-regulating mitochondrial-dependent apoptosis induced by cytokines.

Caspase-9 and -12 have been proposed to influence islet ß-cell apoptosis (41, 42, 43). Here, we analyzed whether IGF-I has an effect on these early apoptosis mediators whose accumulation is triggered by mitochondrial damage or stressed endoplasmic reticulum. After exposure to TNF-{alpha} + IFN-{gamma} for 6 h, caspase-9 was significantly cleaved in NIT-1 cells (Fig. 5DGo, lane 3), suggesting that activation of caspase-9 participates in cytokine-induced ß-cell death. Addition of IGF-I efficiently blocked the cleavage of caspase-9 activated by TNF-{alpha} + IFN-{gamma}. On the other hand, caspase-12 was not detectable in NIT-1 cells stimulated under the above-described conditions (our unpublished data), indicating that caspase-12 does not participate in TNF-{alpha} + IFN-{gamma}-induced ß-cell apoptosis. Our additional studies demonstrated that ß-cell death induced by TNF-{alpha} + IFN-{gamma} cannot be fully blocked by inhibitors of caspase-3, -8, and -9 (our unpublished data), suggesting that initiator caspases, such as caspase-2, may also regulate cytokine-induced ß-cell apoptosis (44).

IGF-I-mediated ß-cell protection against apoptosis depends on activation of the PI3-K/Akt-1 signaling pathway
The PI3-K/Akt-1 signaling pathway is a key pathway for ß-cell metabolism, proliferation and survival (45, 46, 47). To determine whether activation of this pathway contributes to IGF-I-mediated resistance to ß-cell apoptosis induced by inflammatory cytokines, we first examined the role of PI3-K activity in NIT-1 cell apoptosis by evaluating the effects of PI3-K blockade on ß-cell death. A high concentration (50 µM) of LY294002, an inhibitor of PI3-K, reduced NIT-1 cell viability by approximately 45% compared with that observed for cells cultured with medium alone, indicating that PI3-K controls the normal survival and/or proliferation of NIT-1 cells (Fig. 6AGo). As expected, TNF-{alpha} + IFN-{gamma} also induced NIT-1 cell death to a similar extent. However, NIT-1 cells are much more susceptible to TNF-{alpha} + IFN-{gamma}-induced apoptosis in the presence of 50 µM LY 294002, as cell viability was significantly decreased from approximately 54 to 11%. IGF-I-mediated rescue from apoptosis was also partially inhibited by LY294002 in a dose-dependent manner (5–20 µM). Given that one of the most important downstream effectors of PI3-K is Akt-1, which exerts its antiapoptosis function via the phosphorylation-dependent inactivation of Bad (48), we assayed Akt-1 activity in NIT-1 cells after exposure to IGF-I alone or TNF-{alpha} + IFN-{gamma} ± IGF-I. Whereas IGF-I activated the phosphorylation of Akt-1 in NIT-1cells (Fig. 6BGo, lane 2), this phosphorylation was not changed upon exposure to TNF-{alpha} + IFN-{gamma} compared with the control medium alone (Fig. 6BGo, lane 4). However, inclusion of IGF-I in the latter cultures partially enhanced the level of Akt-1 phosphorylation (Fig. 6BGo, lane 5) to that observed upon stimulation by cytotoxic cytokines (Fig. 6BGo, lane 4). The presence of LY294002 (20 µM) blocked the increase in Akt-1 phosphorylation induced by IGF-I (Fig. 6BGo, lane 6). Immunoblotting analyses of total Akt indicated that equivalent levels of Akt were present in NIT-1 cells under the various incubation conditions. Taken together, our data indicate that the stimulation of ß-cell survival by IGF-I in a proinflammatory and proapoptotic milieu depends on the PI3-K/Akt-1 signaling pathway.



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 6. Activation of Akt-1 contributes to IGF-I-mediated inhibition of ß-cell death. A, blockade of PI3-K by the LY294002 inhibitor reduces IGF-I-mediated inhibition of ß-cell death. NIT-1 cells (2 x 105/well in a 96-well plate) were treated under the conditions shown. LY294002 was added to cultures 1 h before exposure to cytokines and IGF-I. After 48 h, cell viability was assessed by the MTT assay, and the viability of cells cultured in medium was normalized to 100%. Results are shown as mean ± SD and are representative of three independent experiments. B, Effects of IGF-I and PI3-K for 1 h under various conditions, as shown. Cell lysates were prepared, resolved by SDS-PAGE (25 µg/lane), and immunoblotted with anti-phospho-Akt-1 and anti-Akt antibodies.

 
Increased ß-cell replication in IGF-I/IGFBP-3-treated mice
Because of its mitogenic capacity, IGF-I can regulate ß-cell proliferation in vitro and in vivo (49, 50, 51). It is possible that long-term treatment with IGF-I may prevent T1D in NOD mice by enhancing ß-cell proliferation. To test this possibility, we assayed ß-cell proliferation by BrdU incorporation in response to IGF-I and also measured the size of islets after IGF-I/IGFBP-3 treatment. ß-cell proliferation was significantly enhanced in mice treated with IGF-1/IGFBP-3 compared with vehicle-treated mice (Fig. 7AGo). These newly proliferating ß-cells were localized predominantly near the infiltrating zone of islets (our unpublished data). No increase in ß-cell mass was observed in mice treated with IGF-I/IGFBP-3. In addition, the distribution of islet size was not altered in mice treated with IGF-I/IGFBP-3 compared with vehicle-treated control mice, although a high number of small-size islets (<100 µm2) were observed in IGF-I/IGFBP-3-treated mice (Fig. 7BGo). Thus, IGF-I may protect from the development of T1D by stimulating ß-cell proliferation.



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 7. Effects of IGF-I/IGFBP-3 on ß-cell replication and islet size in NOD mice. NOD mice (4 wk old; six mice per group) were treated for 4 wk with either IGF-I/IGFBP-3 (1 mg/kg, every other day.) or vehicle. A, Analysis of ß-cell replication. The number of insulin-stained BrdU-positive nuclei was determined. Results are shown as mean values ± SD of percentage of BrdU-positive ß-cells per 6 h. *, P < 0.05 vs. vehicle (PBS)-treated mice (Student’s t test). B, Size distribution of islets in IGF-I/IGFBP-3-treated (1 mg/kg, every other day) NOD mice. Islet size was measured as described in Materials and Methods.

 
IGF-I suppresses the expression of several genes that encode proteins in the apoptotic signaling pathways of islet cells
To further understand how IGF-I prevents the apoptosis of islet ß-cells, we used cDNA microarrays to analyze the expression profiles of several genes that encode proteins in NOD islet cell apoptotic signaling pathways triggered by TNF-{alpha} + IFN-{gamma}. After exposure to these cytokines for 24 h, 12 islet gene transcripts were detected on a cDNA template filter consisting of 96 apoptosis-related genes (Table 1Go). These transcripts include those of the DAP kinase (2), DFFA, DR3, DR6, Myd88, TNF-ß, and Trail proapoptotic genes. Other proapoptotic genes, such as Flash and TNFRSF11A, exhibited lower levels of transcription. On the other hand, the expression of some antiapoptotic genes, including IAP-1 and NIP-3, was also elevated, which may indicate that islet cells up-regulate their self-antiapoptosis machinery to rescue themselves from cell death induced by cytokines. In comparison with gene profiles of islets exposed to cytokines alone, islets exposed to IGF-I did not yield the activation of some proapoptotic genes such as DAP kinase (2), DFFA, and Flash. Surprisingly, IGF-I significantly down-regulated the expression of a group of newly defined proapoptotic genes, such as DR3, DR6, Myd88, TNF-ß, TNFRSF11A, and Trail. The expression of some proapoptotic genes was also suppressed in islets cultured with IGF-I due to hypoxia resulting from the islet isolation procedures (data not shown). These results implicate the complexity of genes involved in the regulation of ß-cell apoptosis pathways activated by TNF-{alpha} + IFN-{gamma} and reinforce the importance of IGF-I as a potent antiapoptotic agent for ß-cells.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Effects of IGF-I on gene expression of apoptotic pathways activated by cytokines in pancreatic islets of NOD mice

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study, we examined the effects of IGF-I/IGFBP-3 on both spontaneous and CY-induced T1D in NOD mice. We demonstrate for the first time that administration of IGF-I/IGFBP-3 to NOD mice starting at 4 wk of age not only diminished the severity of insulitis but also significantly reduced the incidence of spontaneous and CY-induced T1D at 35 wk of age in a dose-dependent manner. Note that the content of IGF-I in the IGF-I/IGFBP-3 complex used here was one fifth (i.e. 0.2 mg/kg) of that used in a previous investigation (25). Furthermore, free IGF-I had only a transient effect on reducing insulitis but did not afford any significant protection against the development of T1D. Our observations demonstrate that the bioactivity of IGF-I present in the form of an IGF-I/IGFBP-3 complex is considerably enhanced in vivo compared with that of free IGF-I and provide further support for the potential therapeutic application of IGF-I/IGFBP-3 treatment in the prevention of human T1D. Nonetheless, it is important to bear in mind that the therapeutic efficacy of IGF-I/IGFBF-3 treatment can also be influenced by the dose and tissue sensitivity of this complex as well as the disease status of prospective recipients. An example of this was reported in a study of experimental allergic encephalomyelitis, in which a high dose of the IGF-I/IGFBP-3 complex administered after disease onset increased the severity of disease (25).

In addition to its role in IGF-I transport, IGFBP-3 can also regulate cell survival and apoptosis (34). Upon treatment with IGF-I/IGFBP-3, the protective effect in T1D mediated by IGF-I may be influenced in part by IGFBP-3 bioactivity. Interestingly, at the outset of this study, the effect of IGFBP-3 on islet cell survival and apoptosis was examined in vitro. Our data suggest that IGFBP-3 alone or in combination with IGF-I (10–1000 ng/ml) neither induces islet cell proliferation nor enhances cell survival under conditions of cytokine-induced islet ß-cell cytotoxicity (our unpublished observations). Rather, we found that IGFBP-3 blocks the antiapoptotic function of IGF-I. Our in vivo data support those obtained in previous animal (IGFBP-3 transgenic mice) and clinical (hyperglycemia and reduced insulin sensitivity) studies of the IGF/IGFBP-3 complex (26, 52, 53). Thus, it is unlikely that IGFBP-3 functions independently of IGF-I independent when present in the form of an IGF-I/IGFBP-3 complex.

To evaluate the effects of IGF-I on immune responses against islet cells, we determined the cytokine and chemokine profiles in the spleen and pancreas of IGF-I/IGFBP-3-treated mice and monitored their splenic regulatory T cell activity. These mice did not exhibit profound changes in their levels of IFN-{gamma}, IL-4, CCL3, and CCL4 in the pancreas, although there was an observed trend toward reduced IFN-{gamma} and CCL3 compared with vehicle-treated mice. However, reduced CCL3 and increased CCL4 mRNA expression was observed in the PLNs of IGF-I/IGFBP-3-treated mice. In the regulation of the CCL3 and CCL4 chemokines by IGF-I/IGFBP3, changes observed in the chemokine protein level in pancreatic tissues seem less significant than the changes detected in mRNA levels assayed by RT-PCR. This discrepancy may be due to several reasons. First, due to the relatively short half-life of chemokines at the site of inflammation, the high content of proteases, and excellent blood circulation in the pancreas, CCL3 and CCL4 may be quickly degraded or diluted during the euthanasia of mice and subsequent sample preparation. Thus, the difference in protein levels between the two groups may be artificially reduced. Second, at 8–10 wk of age, the distribution of chemokine-producing T cells and other leukocytes may differ between PLNs and islets. IGF-I may also have more functional activity in the PLNs relative to islets in treated mice. The latter point is supported by our RT-PCR data. However, when we measured chemokine protein levels in whole pancreatic tissue (includes a few small lymph nodes), the difference between the two groups of PLN samples was reduced possibly due to dilution by whole tissue-derived proteins. Considering the protein and mRNA levels of chemokines, severity of insulitis, and our laboratories’ previous reports on these chemokines during insulitis (32, 54, 55), we believe that the changes observed in CCL3 and CCL4 levels are associated with IGF-I-mediated protection from diabetes possibly through the inhibition of migration of diabetogenic T cells to islets. In fact, our ongoing study of CCL4 activity indicates that the ability of this chemokine to mediate protection from T1D involves the reduced recruitment of CD8+ T cells to islets during the development of T1D (our unpublished observations).

Recently, ß-cell apoptosis has been identified as a key mechanism of islet destruction in the pathogenesis of T1D. Here we examined the ability of IGF-I/IGFBP-3 treatment to counteract ß-cell apoptosis in NOD mice. We found that mice treated with IGF-I/IGFBP-3 had fewer apoptotic islet ß-cells than vehicle-treated mice. These data provide direct evidence that the administration of IGF-I in vivo in the form of an IGF-I/IGFBP-3 complex effectively enhances ß-cell resistance to cytokine-induced cytotoxicity. Although local expression of IGF-I in ß-cells prevents ß-cell apoptosis induced by multi-low-dose streptozotocin treatment in transgenic C57BL/6-SJL mice and human islets (20, 50), our study is the first to demonstrate that IGF-I administered systemically inhibits ß-cell apoptosis in NOD mice that spontaneously develop T1D. We further observed that this IGF-I-induced blockade of ß-cell apoptosis is mediated by the up-regulation of antiapoptotic Bcl-2 and Bcl-XL and down-regulation of Bad and Bax in ß-cells. Moreover, we showed that IGF-I inhibits the caspase-9-mediated mitochondrial-dependent pathway of ß-cell apoptosis. These observations not only emphasize the importance of the mitochondrial-dependent death pathway of ß-cells activated by cytokines but also reveal possible antiapoptotic mechanisms mediated by IGF-I as a survival signal in preventing ß-cell apoptosis.

The PI3-K and Akt-1 signaling pathway regulates ß-cell growth and metabolism (19, 45, 46, 56, 57). Previously, we showed that IGF-I is not detectable in lymphocyte-infiltrated pancreatic islets (16), suggesting that cytokines may diminish IGF-I-activated survival signals and render islet ß-cells more susceptible to cytokine-induced apoptosis. In support of this notion, we found that the LY294002 inhibitor of PI3-K abrogated the protective effect of IGF-I on NIT-1 cells. In addition, enhanced NIT cell death was detected in the presence of this inhibitor. Immunoblots of NIT-1 cell extracts showed that changes in Akt-1 activity parallel those in cell viability. Thus, our data further demonstrates that cytotoxic cytokines can inhibit survival signaling pathways in ß-cells, and emphasizes the importance of the PI3-K/Akt-1 signaling cascade as a key pathway of IGF-I-mediated protection from apoptosis in ß-cells. Although it remains to be determined how IGF-I counteracts cytokine-mediated death signaling in ß-cells, members of the suppressor of cytokine signaling (SOCS) protein family may be involved as SOCS family members are known to regulate IGF-I signaling pathways (58, 59). Whether a SOCS protein(s) indeed blocks cytokine-induced ß-cell apoptosis by activating IGF-I signaling pathways requires additional experimentation.

We observed that NOD mice receiving IGF-I/IGFBP-3 had augmented ß-cell proliferation rather than increased ß-cell mass and islet size. In transgenic mouse studies, overexpression of IGF-I and the constitutive expression of Akt-1 in ß-cells results in increased ß-cell mass, hyperplasia of islets, and resistance to streptozotocin-induced diabetes (50, 56). In these transgenic models, mitogenic pathways are not continuously suppressed by cytotoxic cytokines, and ß-cells are destroyed relatively rapidly. These factors may offer the residual surviving islet ß-cells an opportunity for regeneration. However, during the pathogenesis of T1D in NOD mice, islet ß-cell death is a chronic event and the mitogenic activities initiated by the autocrine/paracrine form of IGF-I may be suppressed by cytokines secreted by islet-infiltrating lymphocytes. Moreover, considering the numerous stresses islets face during the process of inflammation, ß-cell survival might be a physiologically more important process than proliferation. Therefore, it is conceivable that IGF-I/IGFBP-3 treatment may provide only a limited mitogenic signal for ß-cell replication in NOD mice. Regardless, we believe that enhancement of ß-cell replication partially contributes to reduced susceptibility of T1D in NOD mice after IGF-I/IGFBP-3 treatment.

Various death receptor activation pathways, including those mediated by Fas/FasL and TNF receptors, are stimulated in islet ß-cells (13, 40, 60). However, it remains unknown how different death pathways interact with each other to regulate cell death and whether IGF-I is able to suppress these amplified death signals, because ß-cells are exposed to multiple cytokines during the pathogenesis of T1D. Our results show that many TNF receptor superfamily genes (DR3, DR6, TNF-ß, and Trail) are activated by TNF-{alpha} + IFN-{gamma} stimulation in islets, suggesting an amplification of death signals resulting from the expression of more than one surface death receptor. Moreover, we found that IGF-I down-regulated several of these death receptor gene transcripts. These findings further illustrate that IGF-I efficiently prevents ß-cell apoptosis, especially when induced by inflammatory stress.

In summary, the present study demonstrates the efficacy of IGF-I/IGFBP-3 treatment in protection from T1D in NOD mice. Because treatment of IGF-I/IGFBP-3 has proven to inhibit acute inflammation, protect against tissue injury, and offer improvement of metabolism in clinical studies (61), our study may provide a new approach for prevention of T1D in humans.


   Acknowledgments
 
We thank all the members of our laboratory for their advice, critique, and encouragement.


   Footnotes
 
This work was supported by grants (to T.L.D.) from the Canadian Institutes of Health Research (MT-5729) and the Juvenile Diabetes Research Foundation (JDRF). W.C. was supported as a postdoctoral fellow by the JDRF (152–03-1999). T.L.D. is the Sheldon H. Weinstein Scientist in Diabetes at the University of Western Ontario.

Abbreviations: BGL, Blood glucose level; BrdU, 5-bromo-2-deoxyuridine; CY, cyclophosphamide; IFN, interferon; IGFBP, IGF binding protein; MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; NOD, nonobese diabetic; PI3-K, phosphatidylinositol 3-kinase; PLN, pancreatic lymph node; SOCS, suppressor of cytokine signaling; T1D, type 1 diabetes; TUNEL, terminal deoxynucleotidyltransferase-mediated UTP end labeling.

Received September 23, 2003.

Accepted for publication November 3, 2003.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Delovitch TL, Singh B 1997 The nonobese diabetic mouse as a model of autoimmune diabetes: immune dysregulation gets the NOD. Immunity 7:727–738[CrossRef][Medline]
  2. Augstein P, Elefanty AG, Allison J, Harrison LC 1998 Apoptosis and ß-cell destruction in pancreatic islets of NOD mice with spontaneous and cyclophosphamide-accelerated diabetes. Diabetologia 41:1381–1388[CrossRef][Medline]
  3. Eizirik DL, Darville MI 2001 ß-Cell apoptosis and defense mechanisms: lessons from type 1 diabetes. Diabetes 50 (Suppl 1):S64–S69
  4. O’Brien BA, Harmon BV, Cameron DP, Allan DJ 1997 Apoptosis is the mode of ß-cell death responsible for the development of IDDM in the nonobese diabetic (NOD) mouse. Diabetes 46:750–757[Abstract]
  5. Thomas HE, Kay TW 2000 ß-Cell destruction in the development of autoimmune diabetes in the non-obese diabetic (NOD) mouse. Diabetes Metab Res Rev 16:251–261[CrossRef][Medline]
  6. Suk K, Kim S, Kim YH, Kim KA, Chang I, Yagita H, Shong M, Lee MS 2001 IFN-{gamma}/TNF-{alpha} synergism as the final effector in autoimmune diabetes: a key role for STAT1/IFN regulatory factor-1 pathway in pancreatic ß-cell death. J Immunol 166:4481–4489[Abstract/Free Full Text]
  7. Kim KA, Kim S, Chang I, Kim GS, Min YK, Lee MK, Kim KW, Lee MS 2002 IFN{gamma}/TNF{alpha} synergism in MHC class II induction: effect of nicotinamide on MHC class II expression but not on islet-cell apoptosis. Diabetologia 45:385–393[CrossRef][Medline]
  8. Kagi D, Ho A, Odermatt B, Zakarian A, Ohashi PS, Mak TW 1999 TNF receptor 1-dependent ß-cell toxicity as an effector pathway in autoimmune diabetes. J Immunol 162:4598–4605[Abstract/Free Full Text]
  9. Rabinovitch A, Suarez-Pinzon WL, Shi Y, Morgan AR, Bleackley RC 1994 DNA fragmentation is an early event in cytokine-induced islet ß-cell destruction. Diabetologia 37:733–738[Medline]
  10. Eizirik DL, Mandrup-Poulsen T 2001 A choice of death: the signal-transduction of immune-mediated ß-cell apoptosis. Diabetologia 44:2115–2133[CrossRef][Medline]
  11. Stephens LA, Thomas HE, Ming L, Grell M, Darwiche R, Volodin L, Kay TW 1999 Tumor necrosis factor-{alpha}-activated cell death pathways in NIT-1 insulinoma cells and primary pancreatic ß-cells. Endocrinology 140:3219–3227[Abstract/Free Full Text]
  12. Thomas HE, Darwiche R, Corbett JA, Kay TW 1999 Evidence that ß-cell death in the nonobese diabetic mouse is Fas independent. J Immunol 163:1562–1569[Abstract/Free Full Text]
  13. Rabinovitch A, Suarez-Pinzon W, Strynadka K, Ju Q, Edelstein D, Brownlee M, Korbutt GS, Rajotte RV 1999 Transfection of human pancreatic islets with an anti-apoptotic gene (bcl-2) protects ß-cells from cytokine-induced destruction. Diabetes 48:1223–1229[Abstract]
  14. Cottet S, Dupraz P, Hamburger F, Dolci W, Jaquet M, Thorens B 2002 cFLIP protein prevents tumor necrosis factor-{alpha}-mediated induction of caspase-8-dependent apoptosis in insulin-secreting ßTc-Tet cells. Diabetes 51:1805–1814[Abstract/Free Full Text]
  15. Le Roith D 1997 Seminars in medicine of the Beth Israel Deaconess Medical Center. Insulin-like growth factors. N Engl J Med 336:633–640[Free Full Text]
  16. Hill DJ, Petrik J, Arany E, McDonald TJ, Delovitch TL 1999 Insulin-like growth factors prevent cytokine-mediated cell death in isolated islets of Langerhans from pre-diabetic non-obese diabetic mice. J Endocrinol 161:153–165[Abstract]
  17. Harrison M, Dunger AM, Berg S, Mabley J, John N, Green MH, Green IC 1998 Growth factor protection against cytokine-induced apoptosis in neonatal rat islets of Langerhans: role of Fas. FEBS Lett 435:207–210[CrossRef][Medline]
  18. Mabley JG, Belin V, John N, Green IC 1997 Insulin-like growth factor I reverses interleukin-1ß inhibition of insulin secretion, induction of nitric oxide synthase and cytokine-mediated apoptosis in rat islets of Langerhans. FEBS Lett 417:235–238[CrossRef][Medline]
  19. Castrillo A, Bodelon OG, Bosca L 2000 Inhibitory effect of IGF-I on type 2 nitric oxide synthase expression in Ins-1 cells and protection against activation-dependent apoptosis: involvement of phosphatidylinositol 3-kinase. Diabetes 49:209–217[Abstract]
  20. Giannoukakis N, Mi Z, Rudert WA, Gambotto A, Trucco M, Robbins P 2000 Prevention of ß-cell dysfunction and apoptosis activation in human islets by adenoviral gene transfer of the insulin-like growth factor I. Gene Ther 7:2015–2022[CrossRef][Medline]
  21. Bergerot I, Fabien N, Maguer V, Thivolet C 1995 Insulin-like growth factor-1 (IGF-1) protects NOD mice from insulitis and diabetes. Clin Exp Immunol 102:335–340[Medline]
  22. Dunger DB, Cheetham TD, Crowne EC 1995 Insulin-like growth factors (IGFs) and IGF-I treatment in the adolescent with insulin-dependent diabetes mellitus. Metabolism 44:119–123[Medline]
  23. Bach MA, Chin E, Bondy CA 1994 The effects of subcutaneous insulin-like growth factor-I infusion in insulin-dependent diabetes mellitus. J Clin Endocrinol Metab 79:1040–1045[Abstract]
  24. Adams S, Moore J, Chu S, Bagi C, DeLeon L, Liu C, Schmidt D, Sommer A 1995 Pharmacokinetics and bioavailability of rhIGF-I/IGFBP-3 in the rat and monkey. Prog Growth Factor Res 6:347–356[CrossRef][Medline]
  25. Lovett-Racke AE, Bittner P, Cross AH, Carlino JA, Racke MK 1998 Regulation of experimental autoimmune encephalomyelitis with insulin-like growth factor (IGF-1) and IGF-1/IGF-binding protein-3 complex (IGF-1/IGFBP3). J Clin Invest 101:1797–1804[Medline]
  26. Clemmons DR, Moses AC, McKay MJ, Sommer A, Rosen DM, Ruckle J 2000 The combination of insulin-like growth factor I and insulin-like growth factor-binding protein-3 reduces insulin requirements in insulin-dependent type 1 diabetes: evidence for in vivo biological activity. J Clin Endocrinol Metab 85:1518–1524[Abstract/Free Full Text]
  27. Herndon DN, Ramzy PI, DebRoy MA, Zheng M, Ferrando AA, Chinkes DL, Barret JP, Wolfe RR, Wolf SE 1999 Muscle protein catabolism after severe burn: effects of IGF-1/IGFBP-3 treatment. Ann Surg 229:713–720; discussion 720–722[CrossRef][Medline]
  28. Guan J, Behme MT, Zucker P, Atkison P, Hramiak I, Zhong R, Dupre J 1998 Glucose turnover and insulin sensitivity in rats with pancreatic islet transplants. Diabetes 47:1020–1026[Abstract]
  29. Chen W, Bergerot I, Elliott JF, Harrison LC, Abiru N, Eisenbarth GS, Delovitch TL 2001 Evidence that a peptide spanning the B-C junction of proinsulin is an early autoantigen epitope in the pathogenesis of type 1 diabetes. J Immunol 167:4926–4935[Abstract/Free Full Text]
  30. Rapoport MJ, Jaramillo A, Zipris D, Lazarus AH, Serreze DV, Leiter EH, Cyopick P, Danska JS, Delovitch TL 1993 Interleukin 4 reverses T cell proliferative unresponsiveness and prevents the onset of diabetes in nonobese diabetic mice. J Exp Med 178:87–99[Abstract/Free Full Text]
  31. Cameron MJ, Arreaza GA, Zucker P, Chensue SW, Strieter RM, Chakrabarti S, Delovitch TL 1997 IL-4 prevents insulitis and insulin-dependent diabetes mellitus in nonobese diabetic mice by potentiation of regulatory T helper-2 cell function. J Immunol 159:4686–4692[Abstract]
  32. Cameron MJ, Arreaza GA, Grattan M, Meagher C, Sharif S, Burdick MD, Strieter RM, Cook DN, Delovitch TL 2000 Differential expression of CC chemokines and the CCR5 receptor in the pancreas is associated with progression to type I diabetes. J Immunol 165:1102–1110[Abstract/Free Full Text]
  33. Cattan P, Berney T, Schena S, Molano RD, Pileggi A, Vizzardelli C, Ricordi C, Inverardi L 2001 Early assessment of apoptosis in isolated islets of Langerhans. Transplant Proc 33:264–265[CrossRef][Medline]
  34. Firth SM, Baxter RC 2002 Cellular actions of the insulin-like growth factor binding proteins. Endocr Rev 23:824–854[Abstract/Free Full Text]
  35. Liu D, Pavlovic D, Chen MC, Flodstrom M, Sandler S, Eizirik DL 2000 Cytokines induce apoptosis in ß-cells isolated from mice lacking the inducible isoform of nitric oxide synthase (iNOS-/-). Diabetes 49:1116–1122[Abstract]
  36. Green EA, Wong FS, Eshima K, Mora C, Flavell RA 2000 Neonatal tumor necrosis factor-{alpha} promotes diabetes in nonobese diabetic mice by CD154-independent antigen presentation to CD8(+) T cells. J Exp Med 191:225–238[Abstract/Free Full Text]
  37. Pakala SV, Chivetta M, Kelly CB, Katz JD 1999 In autoimmune diabetes the transition from benign to pernicious insulitis requires an islet cell response to tumor necrosis factor-{alpha}. J Exp Med 189:1053–1062[Abstract/Free Full Text]
  38. Kim YH, Kim S, Kim KA, Yagita H, Kayagaki N, Kim KW, Lee MS 1999 Apoptosis of pancreatic ß-cells detected in accelerated diabetes of NOD mice: no role of Fas-Fas ligand interaction in autoimmune diabetes. Eur J Immunol 29:455–465[CrossRef][Medline]
  39. Hanke J 2001 Apoptosis in cultured rat islets of Langerhans and occurrence of Bcl-2, Bak, Bax, Fas and Fas ligand. Cells Tissues Organs 169:113–124[CrossRef][Medline]
  40. Allison J, Thomas H, Beck D, Brady JL, Lew AM, Elefanty A, Kosaka H, Kay TW, Huang DC, Strasser A 2000 Transgenic overexpression of human Bcl-2 in islet ß-cells inhibits apoptosis but does not prevent autoimmune destruction. Int Immunol 12:9–17[Abstract/Free Full Text]
  41. Araki E, Oyadomari S, Mori M 2003 Endoplasmic reticulum stress and diabetes mellitus. Intern Med 42:7–14[Medline]
  42. Oyadomari S, Araki E, Mori M 2002 Endoplasmic reticulum stress-mediated apoptosis in pancreatic ß-cells. Apoptosis 7:335–345[CrossRef][Medline]
  43. Mathis D, Vence L, Benoist C 2001 ß-Cell death during progression to diabetes. Nature 414:792–798[CrossRef][Medline]
  44. Huo J, Luo RH, Metz SA, Li G 2002 Activation of caspase-2 mediates the apoptosis induced by GTP-depletion in insulin-secreting (HIT-T15) cells. Endocrinology 143:1695–1704[Abstract/Free Full Text]
  45. Liu W, Chin-Chance C, Lee EJ, Lowe Jr WL 2002 Activation of phosphatidylinositol 3-kinase contributes to insulin-like growth factor I-mediated inhibition of pancreatic ß-cell death. Endocrinology 143:3802–3812[Abstract/Free Full Text]
  46. Tuttle RL, Gill NS, Pugh W, Lee JP, Koeberlein B, Furth EE, Polonsky KS, Naji A, Birnbaum MJ 2001 Regulation of pancreatic ß-cell growth and survival by the serine/threonine protein kinase Akt1/PKB{alpha}. Nat Med 7:1133–1137[CrossRef][Medline]
  47. Srinivasan S, Bernal-Mizrachi E, Ohsugi M, Permutt MA 2002 Glucose promotes pancreatic islet ß-cell survival through a PI 3-kinase/Akt-signaling pathway. Am J Physiol Endocrinol Metab 283:E784–E793
  48. Brazil DP, Park J, Hemmings BA 2002 PKB binding proteins. Getting in on the Akt. Cell 111:293–303[CrossRef][Medline]
  49. Smith FE, Rosen KM, Villa-Komaroff L, Weir GC, Bonner-Weir S 1991 Enhanced insulin-like growth factor I gene expression in regenerating rat pancreas. Proc Natl Acad Sci USA 88:6152–6156[Abstract/Free Full Text]
  50. George M, Ayuso E, Casellas A, Costa C, Devedjian JC, Bosch F 2002 ß-Cell expression of IGF-I leads to recovery from type 1 diabetes. J Clin Invest 109:1153–1163[CrossRef][Medline]
  51. Hayakawa H, Kawarada Y, Mizumoto R, Hibasami H, Tanaka M, Nakashima K 1996 Induction and involvement of endogenous IGF-I in pancreas regeneration after partial pancreatectomy in the dog. J Endocrinol 149:259–267[Abstract]
  52. Silha JV, Murphy LJ 2002 Insights from insulin-like growth factor binding protein transgenic mice. Endocrinology 143:3711–3714