Impaired Crkl Expression Contributes to the Defective DNA Binding of Stat5b in Nonobese Diabetic Mice
http://www.100md.com
糖尿病学杂志 2006年第3期
1 Center for Biotechnology and Genomic Medicine, Medical College of Georgia, Augusta, Georgia
2 Molecular Reproduction Unit, Rajiv Gandhi Centre for Biotechnology, Kerala, India
3 Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida, Gainesville, Florida
4 Diabetes Research Center and Departments of Microbiology and Internal Medicine, University of Virginia, Charlottesville, Virginia
ECL, enhanced chemiluminescence; EMSA, electrophoretic mobility shift assay; GMCSF, granulocyte macrophage colonyeCstimulating factor; TCR, T-cell receptor
ABSTRACT
A point mutation in the Stat5b DNA binding domain in the nonobese diabetic (NOD) mouse was shown to have weaker DNA binding compared with the B6 Stat5b. Here, we assessed the binding ability of the mutant Stat5b in the B6 genetic background (B6.NOD-c11) and the wild-type Stat5b in the NOD background (NOD.Lc11). To our surprise, the binding ability of Stat5b is inconsistent with the presence or absence of the Stat5b mutation in these congenic mice but is correlated with the expression levels of the Crkl protein, which was coprecipitated by an anti-Stat5b antibody. Both the expression of Crkl and the Stat5b binding ability are the highest in B6.NOD-c11 and the lowest in NOD while intermediate in B6 and NOD.Lc11 mice. We demonstrated that the adapter molecule Crkl can bind Stat5b and that the Crkl protein is a Stat5b binding cofactor. More importantly, profection of Crkl recombinant protein significantly increased Stat5b binding ability and rescued the binding defect of the NOD mutant Stat5b, suggesting that Crkl is a key regulatory molecule for Stat5b binding. Therefore, the defective Crkl expression may contribute to the development of diabetes in the NOD mice by exacerbating the defective Stat5b binding ability.
Type 1 diabetes is an autoimmune disease resulting from the T-celleCmediated destruction of the insulin-producing -cells in the pancreas (1). Multiple environmental and genetic factors interact to precipitate the disease. The nonobese diabetic (NOD) mouse is the principle animal model for type 1 diabetes (2). Although the genetic factors responsible for the different disease stages have not yet been elucidated, genetic mapping studies have identified >20 diabetogenic intervals on different chromosomes, and these disease genes may control the development of the disease at the level of either initiation of insulitis and/or progression from insulitis to overt diabetes (3eC5).
In our previous study, we discovered a mutation in the Stat5b gene located on chromosome 11 within the Idd4 type 1 diabetes susceptibility interval in NOD mice. The unique mutation C1462A results in a leucine to methionine (L327M) substitution (6), which leads to a weaker DNA binding ability of Stat5b and consecutively an inapt transcriptional regulation Stat5b-regulated genes (IL-2R and Pim1). These studies established the Stat5b pathway as a key molecular defect in NOD mice. Since a number of cofactors may influence the DNA binding ability of Stat5b, we determined in this study the role of the cofactors in the pathogenesis of type 1 diabetes. Our study quickly zeroed in to the v-crk sarcoma virus CT10 oncogene homolog (avian)-like (Crkl) because it was coimmunoprecipitated by the anti-Stat5b antibody. Crkl is an SH2/SH3-containing 39-kDa protein of the Crk family, which function as adapter molecules. The Crk-activated signal transduction cascades include immune responses, the modulation of cell adhesion, and cell migration (7). Thrombopoietin and interleukin-2 induce association of Crk with Stat5 (8) and Crkl predominantly binds Stat5b (9). In response to interferon, Crkl associates with Stat5b and binds in vitro to the TTCTAGGAA palindromic element found in the promoters of a subset of interferon-stimulated genes (10). Association of Crkl with Stat5 in hematopoietic cells is stimulated by granulocyte-macrophage colony-stimulating factor or erythropoietin via the SH2 domain (11). We report here reduced expression levels of Crkl in NOD compared with B6 and two diabetes-resistant congenic strains. We also demonstrate that the DNA binding ability of Stat5b is modulated by Crkl and recombinant Crkl protein can rescue the binding ability of the mutant Stat5b protein. Our data suggest that the weaker expression of Crkl could compound the diminished DNA binding ability of the mutant Stat5b.
RESEARCH DESIGN AND METHODS
Ten-week-old female mice of three strains including C57B1/6J (B6) and NOD/LtJ (NOD), B6.NOD-c11 were obtained from our mouse colony at the University of Florida, while the NOD.Lc11 mice were obtained from Dr. Marcia McDuffie at the Diabetes Research Center, Departments of Microbiology and Internal Medicine, University of Virginia. Total RNA samples were prepared from spleen cells using standard methods. A subset of mice was induced with 0.5 e蘥 (l e蘥/ml) of granulocyte macrophage colonyeCstimulating factor (GMCSF) in PBS with 0.1% BSA through intraperitoneal injection. After 20 min of induction, the animals were killed and tissues were immediately snap frozen in liquid nitrogen and then stored at eC80°C. Sham treated controls were injected with 500 e蘬 0.1% BSA in 0.2 e蘭eCfiltered PBS (Gibco) and killed after 20 min as above. Nuclear and cytoplasmic extracts from spleens were prepared as described earlier (6) and protein concentration was measured using Bio-Rad protein assay.
SDS-PAGE, Western blotting, and immunobloting.
Cytosolic or nuclear proteins (10 e蘥) diluted 1:4 with Laemmli sample buffer were loaded onto a 10% running gel (pH 8.3) with a 4% stacking gel (pH 6.8) as detailed in our earlier paper (6). Gels were run on a MiniProtean III vertical electrophoresis system (BioRad) at 100 V for 3 h and the separated proteins transferred onto Sequiblot PVDF (0.2 e蘭; BioRad) in transfer buffer (25 mmol/l Tris, 190 mmol/l glycine [pH 8.2], 40% methanol [vol/vol]) using a Mini Transblot cell (BioRad) at 15 mA constant current for each membrane for 12eC16 h. The blots were developed by enhanced chemiluminescence (ECL) method per the manufacturer’s instructions. The membranes were blocked in 5% blocker for 60 min at room temperature with gentle shaking, washed five times in PBS-Tween 20 (PBS-T), and incubated for 1 h with optimal concentrations of primary antibodies. After incubation the membranes were extensively washed five times (5 min each) in PBS-T and then incubated for 60 min with horseradish peroxidaseeCconjugated secondary antibodies. Visualization was performed using an ECL kit per the manufacturer’s protocol (Amersham-Pharmacia). Antibodies used in immunoblotting studies were anti-Stat5b (sc-835) and anti-Crkl (sc-319) from Santa Cruz Biotechnology (Santa Cruz, CA) and anti-rabbit and anti-mouse from the ECL kit.
Amplification and direct DNA sequencing of coding region of murine Crkl gene.
Total RNA was extracted from splenocytes of NOD and B6 mice using Trizol kit (Sigma) according to instructions. Three micrograms of total RNA were converted to cDNA using reverse transcriptase (In Vitrogen) as described (6) and diluted 1:1 by TE buffer. Two microliters of diluted cDNA were used for PCR amplification of the entire coding region of the Crkl gene using forward primer 5'-ATGTCCTCCGCCAGGTTTC-3' and reverse primer 5'-AGCACAGCAATCACTCGTTG-3' with the mentioned conditions (94°C for 2 min for one cycle followed by 35 cycles at 94°C for 30 s, 64°C for 30 s, 72°C for 1 min 15 s, and 1 cycle for final extension at 72°C for 10 min.). The PCR products were loaded onto 1.5% agarose gel and then the DNA was recovered from the gel using QIAquick gel extraction kit (Qiagen). Purified PCR products were subjected to direct DNA sequencing using the above-mentioned primers on a Perkin Elmer 377 automated DNA sequencer.
Cloning, expression, and protein purification of murine Crkl.
The entire coding region of the murine Crkl region was PCR amplified with the same primers mentioned above, but the forward and reverse primer bear the SpeI and XhoI sequences at their 5' end for cloning purposes. The amplified PCR products were recovered from the agarose gel using QIAquick gel extraction kit (QIAgen) according to the manufacturer’s instructions. The purified PCR products were digested by appropriate restriction enzymes (New England Biolabs), then the full-length Crkl gene was ligated into the pTYB11 vector (IMPACT CN New England BioLabs) using the SpeI and XhoI sites and processed for transformation of BL21 DE3 host cells (Novagen). Screening of transfected clones was performed and all 18 clones were found to be positive for the insert gene. Direct DNA sequencing of the selected clones did not provide any proof of a mutation within the gene due to cloning procedures. The pTYB11 clones containing the entire murine Crkl coding sequence were induced by 1.0 mmol/l IPTG (Sigma) for 4 h at 30°C, and purification of the protein was carried out using chitin beads (IMPACT-CN system; New England Biolabs) as recommended by the manufacturer’s guidelines.
FarWestern analysis.
To determine direct interaction of Stat5b with Crkl, immunoprecipitated Stat5b from nuclear extracts of splenocytes of NOD and B6 mice, along with recombinant Aire and ADP proteins as negative controls, were run on native gels and transferred onto a PVDF membrane. The membrane was blocked in 5% blocker (Amersham-Pharmacia Biotech) in 1 x PBS-T for 1 h at room temperature. Membranes were then incubated in recombinant Crkl protein for 1 h followed by five washes with PBS-T each for 5 min. Then, membranes were incubated with anti-Crkl rabbit antibody conjugated with HRP for 1 h. The membranes were washed thoroughly and signals were detected using an ECL kit (Amersham-Pharmacia) according to the manufacturer’s instructions.
Profection.
Splenocytes were cultured and profection (protein transfection) was done using ProVectin reagent (Imgenex) per the method detailed below.
Splenocyte culture.
Spleen tissue was dissected, freed of adhering fat, and washed in sterile Hank’s balanced salt solution (Gibco). Mouse spleen was cut into several small pieces and ground between two sterilized frosted microscope glass slides. The cell suspension was collected in 15 ml Falcon tube and subjected to erythrocyte hemolysis using 1 ml RBC lysis buffer. The remaining debris was separated by gravity sedimentation by letting the tube stand up for about 10 min. The cell suspension was collected in another tube and centrifuged for 7 min at 2,000 rpm. The pellet was reconstituted in 1 ml Adoptive Immunotherapy MediaeCV (AIM-V) medium and the cell density was adjusted to 1 x 106 cell/ml. The splenocytes were cultured in supplemented, serum free, AIM-V at the concentration of 1.0 x 106 cells/ml. AIM-V was supplemented with 0.1 mmol/l modified Eagle’s medium (MEM) nonessential amino acids, 1.0 mmol/l sodium pyruvate, 3.0 x 10eC5 mmol/l 2-mercaptoethanol, 2.0 mmol/l L-glutamine, 10.0 mmol/l HEPES, and 100 units antibiotic-antimycotic solution (Gibco) at final concentration.
Protein transfection.
ProVectin reagent was prepared per the manufacturer’s instructions (Imgenex). The diluted recombinant Crkl in 25 e蘬 PBS (20 mmol/l Na phosphate, 150 mmol/l NaCl, pH 7.4) was used to hydrate the dried ProVectin reagent. AIM-V serum-free medium was added to make the volume 250 e蘬. This profection mix was added to splenocytes and cultured for 4 h in CO2 incubator in 96-well microtiter plates at 37°C. The splenocytes were induced for 15 min with GMCSF (1 e蘥/ml), and cells were collected in Falcon centrifuge tubes and centrifuged for 10 min/12,000 rpm. The medium was aspirated and cells washed with chilled PBS-T and stored at eC80°C. The cells are processed for cytosolic/nuclear extract preparation.
Electrophoretic mobility shift assays.
Electrophoretic mobility shift assays (EMSAs) of Stat5b were done exactly as described earlier (6). Stat5b consensus sequence (5'-TTTCTAGGAATT-3') was used for EMSA. Complementary oligos were used to make double-strand DNA, which was end-labeled using 170 e蘡i [-32P]ATP and T4 polynucleotide kinase. Equal amount of nuclear protein extract (5 e蘥) was incubated with double-strand oligonucleotides (consensus/mutant) at room temperature for 20 min in a binding buffer (13 mmol/l HEPES, pH 7.9, 80 mmol/l NaCl, 8% glycerol, 0.15 mmol/l EDTA, 1 mmol/l dithiothreitol) (12) and in the presence of poly dI-dC (2.5 e蘥/e蘬). For supershift assay, the protein extract was first incubated with 0.8 e蘥 of antibody for 20 min at room temperature. Oligonucleotide was then added and the reaction continued for another 30 min at room temperature. The anti-Stat5b (sc-835X) and anti-Crkl (sc-319) antibodies are rabbit polyclonal antibodies from Santa Cruz Biotechnology. DNA protein complexes were resolved on a 5% nondenaturing gel containing 5% glycerol and 0.5x Tris borate EDTA buffer (TBE). The gels were prerun in 0.5x TBE for 30 min at 150 V. A total of 2 e蘬 of loading mix (40% sucrose, 0.25% bromophenol blue, and 0.25% xylene cyanol) was added to each sample before loading. The gels were run at 250 mV for the first 15 min and then at constant 150 mV for about 4 h. The gels were exposed to X-ray films and then kept at eC80°C for at least 48 h.
Statistical analysis.
Western blots/EMSAs were digitized on a Chem-imager 4400 (Alpha Innotech). Band volumes were integrated using Phoretix 1-D image analysis software (version 5.20; Nonlinear Dynamics). The data were subjected to a two-tailed t test.
RESULTS
DNA binding ability of Stat5b in congenic mice.
We sequenced the Stat5b gene to confirm the Stat5b genotypes in two congenic mouse strains that are used in this study. The B6.NOD-c11 strain is a diabetes-resistant B6 mouse with a 42-cM Idd4 interval on chromosome 11 (extending from Csf2 to D11Mit42) that was derived from the NOD mouse and contains the NOD-derived mutant Stat5b. The NOD.Lc11 (formerly known as NOD.DR-3) strain is a diabetes-resistant NOD mouse which contains a 52-cM segment (25eC77 cM) spanning locus D11Mit87 to D11Mit42 from C57L/J mice and the C57L-derived wild-type Stat5b (Fig. 1A). The C57L and C57BL/6 were derived from common ancestors and appear to share the entire telomeric half of chromosome 11, which carries the Stat5a/5b/3 complex, by dense single nucleotide polymorphism analysis (13). The DNA binding abilities of B6.NOD-c11 and NOD.Lc11 Stat5b were analyzed using EMSA. The results were compared with those already obtained for the B6 and NOD mice. Protein extracts from the cytosolic and nuclear fractions were prepared from splenocytes with and without in vivo stimulation with GMCSF. Nuclear extracts were used in EMSA with the consensus DNA sequence recognized by Stat5 proteins (6). Both B6.NOD-c11 and NOD.Lc11 animals showed strong gel shift with the Stat5 consensus sequence in the samples stimulated with GMCSF as compared with a negligible shift in the unstimulated preparations (Fig. 1B). A supershift assay using a polyclonal anti-Stat5b antibody showed that the NOD.Lc11 Stat5b binding is similar to that of B6 (P = 0.4) but fivefold higher than NOD (P < 0.008) (Fig. 1C). These results are expected since the NOD.Lc11 mice carry the wild-type Stat5b gene. However, the Stat5b supershift is 7-fold (P < 10eC5) and surprisingly 8.5-fold higher (P < 10eC4) for B6.NOD-C11 than B6 and NOD, respectively (Fig. 1B and C). These results were unexpected because the B6.NOD-C11 mice have the NOD-derived mutant Stat5b, which is expected to bind DNA weakly. This observation led us to hypothesize that other factor(s) may modulate the DNA binding ability of Stat5b. Our data suggest that some cofactors should be located outside of the NOD-derived 42-cM chromosome 11 interval.
Western blot analysis of Crkl expression.
A number of cofactors may modulate the Stat5b binding ability. Western blot analyses were used to evaluate one of the important candidate proteins: Crkl. The protein concentration for Western analyses was normalized by multiple housekeeping proteins. We also found similar levels for cytosolic and nuclear Stat5b in B6 NOD, B6.NOD-c11, and NOD.Lc11, confirming our previous findings in NOD and B6 (Fig. 2). In contrast, Crkl had a significantly higher expression in cytosol of spleen extracts from B6 than NOD (P < 10eC4; Fig. 2A). To our surprise, the B6.NOD-c11 mice had higher Crkl protein in the cytosol than both B6 and NOD (P < 10eC4), while the NOD.Lc11 had Crkl levels equivalent of B6 (Fig. 2). The higher Crkl levels are also found in the nuclear fractions of B6.NOD-C11 compared with B6 (P < 0.005) and NOD (P < 0.0002) (Fig. 2B).
In vitro interaction between Stat5b and Crkl.
In order to study Crkl and Stat5b interaction, we produced recombinant protein using the full-length B6 Crkl gene cloned into the pTYB11 vector (Fig. 3). The purified recombinant Crkl protein was confirmed by Western analysis with an anti-Crkl antibody and MALDI-TOF analysis of the protein (Fig. 3). FarWestern analysis using purified recombinant Crkl and immunoprecipitated Stat5b revealed that Crkl interacted with Stat5b from B6, NOD, and B6.NOD-c11 (Fig. 3F). As Crkl does not bind two control proteins, the interaction between Stat5b and Crkl appears to be highly specific (Fig. 3F). Our results also suggest that the mutant Stat5b from NOD has normal ability to bind Crkl (Fig. 3F).
In vivo interaction between Crkl and Stat5b.
To confirm the potential in vivo interaction between Crkl and Stat5b, we performed immunoprecipitation using anti-Crkl or anti-Stat5b antibodies. As shown in Fig. 4, the immunoprecipitated products using the anti-Stat5b antibody contain a visible Crkl band that was confirmed by Western blot with the anti-Crkl antibody. These data suggest that Crkl can bind Stat5b in vivo. In subsequent experiments we sought to determine the presence of Crkl in the Stat5b gel shift complex. EMSA was done using Stat5 consensus sequence and supershifts were performed using an anti-Crkl antibody. A strong shift band was observed for GMCSF-induced samples from all four mouse strains (B6 NOD, B6.NOD-c11, and NOD.Lc11) (Fig. 5). While there was no supershift band induced by the Crkl antibody, the shift band was completely abolished in the supershift assay with the anti-Crkl antibody when the B6 nuclear extracts were examined, suggesting the presence of Crkl in the Stat5b shifted complex (Fig. 5, lane 4). However, the shifted band from nuclear extracts of NOD mice could not be abolished by the Crkl antibody (Fig. 5, lane 7), suggesting the absence or reduced amount of Crkl in the Stat5b-DNA binding complex in the NOD extracts. The reduction of the Stat5b shift by the Crkl antibody is significantly different between B6 and NOD (P < 0.01), and this observation is consistent with the lower Crkl expression level in the nuclear extracts of NOD spleens. The B6.NOD-c11 showed heavy DNA binding, and the addition of Crkl antibody caused a decrease in the shifted band intensity (Fig. 5A). NOD.Lc11 had shift levels similar to B6 and the DNA/Stat5b complex was abolished by the Crkl antibody (Fig. 5A). The intensity difference in shift position with and without the Crkl antibody was quantitated to estimate the Crkl-sensitive component in the Stat5b-Crkl complex (Fig. 5B). The Crkl-sensitive component in Stat5b-Crkl complex is highest in B6 and NOD.Lc11. B6.NOD-c11 had moderate Crkl sensitive fraction and NOD had negligible Crkl-sensitive component in the Stat5b-Crkl complex (Fig. 5C). Thus, the ability of Stat5b to bind DNA is correlated with the amount of Crkl, suggesting that Crkl is an important factor for the DNA binding ability of Stat5b.
Rescue of Stat5b DNA binding ability by Crkl.
In order to prove that Crkl is an active partner in the transcriptional activation by Stat5b, we carried out profection studies with Crkl on splenocytes from B6 and NOD mice. Western blot analysis revealed similar Stat5b levels in cytosolic and nuclear compartments of splenocytes (Fig. 6A and B). The Crkl levels were higher in profected cytosolic and nuclear extracts from B6 and NOD splenocytes compared with nonprofected controls (Fig. 6A and B). Profected B6 splenocytes exhibited significant (P < 0.00003) increase in Stat5b DNA binding ability compared with nonprofected B6 samples (Fig. 6C). The nonprofected NOD splenocytes had weaker Stat5b DNA binding ability compared with nonprofected B6 (P < 0.005), corroborating our earlier results of a loss in the DNA binding ability for the mutant Stat5b in NOD. The Crkl-profected NOD splenocytes displayed an enhancement in Stat5b DNA binding compared with nonprofected NOD (P < 0.003) and reached a level comparable to wild-type Stat5b from B6 mice (Fig. 6C and D). Therefore, Crkl appears to be able to rescue the DNA binding deficiency of the mutant Stat5b in NOD mice, and our data strongly argue for a cooperative role for Crkl in Stat5b DNA binding.
DISCUSSION
This study provided confirmatory evidence that the Stat5b mutation in the NOD mice has weaker DNA binding affinity toward the Stat5b consensus sequence. More importantly, we provided evidence that NOD is deficient in the levels of the adapter molecule Crkl encoded by the Crkl gene on chromosome 16. Using NOD congenic mice that contain the chromosome 11 interval from B6 (NOD.Lc11) and the B6 congenic mice that contain a chromosome 11 interval from the NOD (B6.NOD-C11), we demonstrated that the DNA binding affinity of Stat5b is correlated with the Stat5b mutation and also the expression level of Crkl. Furthermore, the Crkl expression level is determined by the chromosome 11 interval as well as the rest of the genome.
In a series of biochemical studies, we demonstrated that Crkl can interact with Stat5b in vitro and in vivo. FarWestern analysis showed that Crkl can specifically bind Stat5b. Immunoprecipitation with the anti-Stat5b antibody can also immunoprecipitate the Crkl protein, suggesting that Crkl binds to Stat5b in vivo. More importantly, profection experiments with recombinant Crkl protein increases the DNA binding affinity of Stat5b in both the B6 and NOD mice. The Crkl profection can indeed rescue the Stat5b binding deficiency observed in the NOD mice, suggesting that Crkl is a critical cofactor for Stat5b function.
Crkl is a member of the Crk family with high homology in the SH2/SH3 domains and a similar overall structure. The Crk family of proteins are linked to multiple signaling pathways in different cell types. These proteins function as adaptor molecules, which serve to bridge two or more interacting proteins using their binding domains. Crk proteins are involved in the early steps of lymphocyte activation through their SH2-mediated transient interaction with signal-transducing molecules, such as Cbl, ZAP-70, Cas-L, and Stat5.
T-cells play a critical role in mediating the destruction of pancreatic islet -cells as transfer of diabetic T-cells into young NOD mice or NOD/scid mice can rapidly transfer disease (14,15). It has been suggested that type 1 diabetes can be inhibited at the level of development and/or activation of diabetogenic T-cells (16). Thymic and peripheral T-cells from NOD mice exhibit a proliferative hyporesponsiveness upon crosslinking of the T-cell receptor (TCR)/CD3 complex (17,18). This T-cell hyporesponsiveness is genetically inheritable and is thought to be mediated by the reduced secretion of both interleukin-2 and interleukin-4 (18eC20) as well as a defect in the protein kinase C/ras/mitogen-activated protein kinase T-cell activation signaling pathway (17). The patterns of unresponsiveness and lymphokine secretion seen in anti-TCR/CD3-activated NOD thymic T-cells has also been observed in activated NOD peripheral spleen T-cells (19).
Engagement of TCR initiates signal transduction involving tyrosine phosphorylation of multiple effector molecules and the formation of multimolecular complexes at the receptor site. Our data suggest that the defect in NOD mice at the TCR-mediated signaling level could be due to the low levels of Crkl in NOD mice since Crk proteins are an established component in the plasma membrane scaffold of the T-cell signal transduction complex. The Crk adapter proteins are assumed to play a role in T-cell activation because of their induced association with tyrosine-phosphorylated -chain associated protein (ZAP-70) and Cbl and with the phosphatidylinositol 3-kinase regulatory subunit, p85, following engagement of the T-cell antigen receptor (21,22). Crk, ZAP-70, and p85 interact with tyrosine-phosphorylated Cbl, which serves as a major scaffold protein in activated T-cells. T-cell activationeCdependent conformational changes in Crk and/or p85 promote an initial direct or indirect low-affinity interaction between the two molecules, which is then stabilized by a secondary high-affinity interaction mediated by direct binding of the Crk-SH3(N) to the p85-PBP domain (22). The insulin receptor substrates play important roles in signal transduction emanating from the insulin and insulin-like growth factor-I receptors. Insulin receptor stably recruits full-length Crk by association with its SH2 domain in an autophosphorylation-dependent manner (23). Thus, an impaired Crk expression in NOD mice could lead to dysregulation of Crk-associated function in diabetogenic T-cells due to a defect in the signaling pathway of T-cell activation.
Autoreactive T-cells are normally deleted in the thymus through apoptosis, and defects in lymphocyte apoptosis may lead to autoimmune disorders. Thymocytes and peripheral lymphocytes in NOD are relatively resistant to various apoptosis signals (24,25). We have shown in this study that mice carrying substantial levels of Crkl are diabetes resistant while NOD mice with impaired Crkl expression are susceptible to diabetes. Thus, the reduced level of Crkl could contribute to the apoptosis resistance and the escape of diabetogenic T-cells from apoptosis observed in NOD. Finally, it is important to stress that type 1 diabetes is caused by a multitude of defects and that Crkl may lead to multiple functional defects, including the exacerbation of the weaker binding ability of the mutated Stat5b and defective T-cell signaling.
ACKNOWLEDGMENTS
This work was partially supported by a grant from the National Institute of Allergy and Infectious Disease (2P01 AI-42288) to J.X.S. The NOD.Lc11 strain was developed with support to M.M. from the Juvenile Diabetes Research Foundation, the American Diabetes Association, and the National Institute of Diabetes and Digestive and Kidney Diseases (R01-DK51112).
The authors thank Shiny Titus for assistance in graph generation.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Pipeleers D, Ling Z: Pancreatic beta cells in insulin-dependent diabetes. Diabetes Metab Rev 8:209eC227, 1992
Atkinson MA, Leiter EH: The NOD mouse model of type 1 diabetes: as good as it gets Nat Med 5:601eC604, 1999
Lyons PA, Wicker LS: Localising quantitative trait loci in the NOD mouse model of type 1 diabetes (Review). Curr Dir Autoimmun 1:208eC225, 1999
Wicker LS, Todd JA, Peterson LB: Genetic control of autoimmune diabetes in the NOD mouse. Annu Rev Immunol 13:179eC200, 1995
Rogner UC, Boitard C, Morin J, Melanitou E, Avner P: Three loci on mouse chromosome 6 influence onset and final incidence of type I diabetes in NOD.C3H congenic strains. Genomics 74:163eC171, 2001
Davoodi-Semiromi A, Laloraya M, Kumar GP, Purohit S, Jha RK, She JX: A mutant Stat5b with weaker DNA binding affinity defines a key defective pathway in nonobese diabetic mice. J Biol Chem 279:11553eC11561, 2004
Feller SM: Crk family adaptors-signalling complex formation and biological roles. Oncogene 20:6348eC6371, 2001
Oda A, Wakao H, Fujihara M, Ozaki K, Komatsu N, Tanaka S, Ikeda H, Miyajima A, Ikebuchi K: Thrombopoietin and interleukin-2 induce association of CRK with STAT5. Biochem Biophys Res Commun 278:299eC305, 2000
Schulze H, Ballmaier M, Welte K, Germeshausen M: Thrombopoietin induces the generation of distinct Stat1, Stat3, Stat5a and Stat5b homo- and heterodimeric complexes with different kinetics in human platelets. Exp Hematol 28:294eC304, 2000
Fish EN, Uddin S, Korkmaz M, Majchrzak B, Druker BJ, Platanias LC: Activation of a CrkL-stat5 signaling complex by type I interferons. J Biol Chem 274:571eC573, 1999
Ota J, Kimura F, Sato K, Wakimoto N, Nakamura Y, Nagata N, Suzu S, Yamada M, Shimamura S, Motoyoshi K: Association of CrkL with STAT5 in hematopoietic cells stimulated by granulocyte-macrophage colony-stimulating factor or erythropoietin. Biochem Biophys Res Commun 252:779eC786, 1998
Arreaza GA, Cameron MJ, Jaramillo A, Gill BM, Hardy D, Laupland KB, Rapoport MJ, Zucker P, Chakrabarti S, Chensue SW, Qin HY, Singh B, Delovitch TL: Neonatal activation of CD28 signaling overcomes T cell anergy and prevents autoimmune diabetes by an IL-4-dependent mechanism. J Clin Invest 100:2243eC2253, 1997
Petkov PM, Ding Y, Cassell MA, Zhang W, Wagner G, Sargent EE, Asquith S, Crew V, Johnson KA, Robinson P, Scott VE, Wiles MV: An efficient SNP system for mouse genome scanning and elucidating strain relationships. Genome Res 14:1806eC1811, 2004
Daniel D, Gill RG, Schloot N, Wegmann D: Epitope specificity, cytokine production profile and diabetogenic activity of insulin-specific T cell clones isolated from NOD mice. Eur J Immunol 25:1056eC1062, 1995
Crawford M, Daniel D, Wegmann D, Yang H, Gill RG: Autoimmune islet damage mediated by insulin-specific T cells. Transplant Proc 29:758eC759, 1997
McDuffie M: Derivation of diabetes-resistant congenic lines from the nonobese diabetic mouse. Clin Immunol 96:119eC130, 2000
Rapoport MJ, Lazarus AH, Jaramillo A, Speck E, Delovitch TL: Thymic T cell anergy in autoimmune nonobese diabetic mice is mediated by deficient T cell receptor regulation of the pathway of p21ras activation. J Exp Med 177:1221eC1226, 1993
Zipris D, Lazarus AH, Crow AR, Hadzija M, Delovitch TL: Defective thymic T cell activation by concanavalin A and anti-CD3 in autoimmune nonobese diabetic mice: evidence for thymic T cell anergy that correlates with the onset of insulitis. J Immunol 146:3763eC3771, 1991
Rapoport MJ, Jaramillo A, Zipris D, Lazarus AH, Serreze DV, Leiter EH, Cyopick P, Danska JS, Delovitch TL: Interleukin 4 reverses T cell proliferative unresponsiveness and prevents the onset of diabetes in nonobese diabetic mice. J Exp Med 178:87eC99, 1993
Zipris D, Crow AR, Delovitch TL: Altered thymic and peripheral T-lymphocyte repertoire preceding onset of diabetes in NOD mice. Diabetes 40:429eC435, 1991
Gelkop S, Isakov N: T cell activation stimulates the association of enzymatically active tyrosine-phosphorylated ZAP-70 with the Crk adapter proteins. J Biol Chem 274:21519eC21527, 1999
Gelkop S, Babichev Y, Isakov N: T cell activation induces direct binding of the Crk adapter protein to the regulatory subunit of phosphatidylinositol 3-kinase (p85) via a complex mechanism involving the Cbl protein. J Biol Chem 276:36174eC36182, 2001
Klammt J, Barnikol-Oettler A, Kiess W: Mutational analysis of the interaction between insulin receptor and IGF-I receptor with c-Crk and Crk-L in a yeast two-hybrid system. Biochem Biophys Res Commun 325:183eC190, 2004
Bergman ML, Cilio CM, Penha-Goncalves C, Lamhamedi-Cherradi SE, Lofgren A, Colucci F, Lejon K, Garchon HJ, Holmberg D: CTLA-4eC/eC mice display T cell-apoptosis resistance resembling that ascribed to autoimmune-prone non-obese diabetic (NOD) mice. J Autoimmun 16:105eC113, 2001
Colucci F, Bergman ML, Penha-Goncalves C, Cilio CM, Holmberg D: Apoptosis resistance of nonobese diabetic peripheral lymphocytes linked to the Idd5 diabetes susceptibility region. Proc Natl Acad Sci U S A 94:8670eC8674, 1997(Malini Laloraya, Abdoreza)
2 Molecular Reproduction Unit, Rajiv Gandhi Centre for Biotechnology, Kerala, India
3 Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida, Gainesville, Florida
4 Diabetes Research Center and Departments of Microbiology and Internal Medicine, University of Virginia, Charlottesville, Virginia
ECL, enhanced chemiluminescence; EMSA, electrophoretic mobility shift assay; GMCSF, granulocyte macrophage colonyeCstimulating factor; TCR, T-cell receptor
ABSTRACT
A point mutation in the Stat5b DNA binding domain in the nonobese diabetic (NOD) mouse was shown to have weaker DNA binding compared with the B6 Stat5b. Here, we assessed the binding ability of the mutant Stat5b in the B6 genetic background (B6.NOD-c11) and the wild-type Stat5b in the NOD background (NOD.Lc11). To our surprise, the binding ability of Stat5b is inconsistent with the presence or absence of the Stat5b mutation in these congenic mice but is correlated with the expression levels of the Crkl protein, which was coprecipitated by an anti-Stat5b antibody. Both the expression of Crkl and the Stat5b binding ability are the highest in B6.NOD-c11 and the lowest in NOD while intermediate in B6 and NOD.Lc11 mice. We demonstrated that the adapter molecule Crkl can bind Stat5b and that the Crkl protein is a Stat5b binding cofactor. More importantly, profection of Crkl recombinant protein significantly increased Stat5b binding ability and rescued the binding defect of the NOD mutant Stat5b, suggesting that Crkl is a key regulatory molecule for Stat5b binding. Therefore, the defective Crkl expression may contribute to the development of diabetes in the NOD mice by exacerbating the defective Stat5b binding ability.
Type 1 diabetes is an autoimmune disease resulting from the T-celleCmediated destruction of the insulin-producing -cells in the pancreas (1). Multiple environmental and genetic factors interact to precipitate the disease. The nonobese diabetic (NOD) mouse is the principle animal model for type 1 diabetes (2). Although the genetic factors responsible for the different disease stages have not yet been elucidated, genetic mapping studies have identified >20 diabetogenic intervals on different chromosomes, and these disease genes may control the development of the disease at the level of either initiation of insulitis and/or progression from insulitis to overt diabetes (3eC5).
In our previous study, we discovered a mutation in the Stat5b gene located on chromosome 11 within the Idd4 type 1 diabetes susceptibility interval in NOD mice. The unique mutation C1462A results in a leucine to methionine (L327M) substitution (6), which leads to a weaker DNA binding ability of Stat5b and consecutively an inapt transcriptional regulation Stat5b-regulated genes (IL-2R and Pim1). These studies established the Stat5b pathway as a key molecular defect in NOD mice. Since a number of cofactors may influence the DNA binding ability of Stat5b, we determined in this study the role of the cofactors in the pathogenesis of type 1 diabetes. Our study quickly zeroed in to the v-crk sarcoma virus CT10 oncogene homolog (avian)-like (Crkl) because it was coimmunoprecipitated by the anti-Stat5b antibody. Crkl is an SH2/SH3-containing 39-kDa protein of the Crk family, which function as adapter molecules. The Crk-activated signal transduction cascades include immune responses, the modulation of cell adhesion, and cell migration (7). Thrombopoietin and interleukin-2 induce association of Crk with Stat5 (8) and Crkl predominantly binds Stat5b (9). In response to interferon, Crkl associates with Stat5b and binds in vitro to the TTCTAGGAA palindromic element found in the promoters of a subset of interferon-stimulated genes (10). Association of Crkl with Stat5 in hematopoietic cells is stimulated by granulocyte-macrophage colony-stimulating factor or erythropoietin via the SH2 domain (11). We report here reduced expression levels of Crkl in NOD compared with B6 and two diabetes-resistant congenic strains. We also demonstrate that the DNA binding ability of Stat5b is modulated by Crkl and recombinant Crkl protein can rescue the binding ability of the mutant Stat5b protein. Our data suggest that the weaker expression of Crkl could compound the diminished DNA binding ability of the mutant Stat5b.
RESEARCH DESIGN AND METHODS
Ten-week-old female mice of three strains including C57B1/6J (B6) and NOD/LtJ (NOD), B6.NOD-c11 were obtained from our mouse colony at the University of Florida, while the NOD.Lc11 mice were obtained from Dr. Marcia McDuffie at the Diabetes Research Center, Departments of Microbiology and Internal Medicine, University of Virginia. Total RNA samples were prepared from spleen cells using standard methods. A subset of mice was induced with 0.5 e蘥 (l e蘥/ml) of granulocyte macrophage colonyeCstimulating factor (GMCSF) in PBS with 0.1% BSA through intraperitoneal injection. After 20 min of induction, the animals were killed and tissues were immediately snap frozen in liquid nitrogen and then stored at eC80°C. Sham treated controls were injected with 500 e蘬 0.1% BSA in 0.2 e蘭eCfiltered PBS (Gibco) and killed after 20 min as above. Nuclear and cytoplasmic extracts from spleens were prepared as described earlier (6) and protein concentration was measured using Bio-Rad protein assay.
SDS-PAGE, Western blotting, and immunobloting.
Cytosolic or nuclear proteins (10 e蘥) diluted 1:4 with Laemmli sample buffer were loaded onto a 10% running gel (pH 8.3) with a 4% stacking gel (pH 6.8) as detailed in our earlier paper (6). Gels were run on a MiniProtean III vertical electrophoresis system (BioRad) at 100 V for 3 h and the separated proteins transferred onto Sequiblot PVDF (0.2 e蘭; BioRad) in transfer buffer (25 mmol/l Tris, 190 mmol/l glycine [pH 8.2], 40% methanol [vol/vol]) using a Mini Transblot cell (BioRad) at 15 mA constant current for each membrane for 12eC16 h. The blots were developed by enhanced chemiluminescence (ECL) method per the manufacturer’s instructions. The membranes were blocked in 5% blocker for 60 min at room temperature with gentle shaking, washed five times in PBS-Tween 20 (PBS-T), and incubated for 1 h with optimal concentrations of primary antibodies. After incubation the membranes were extensively washed five times (5 min each) in PBS-T and then incubated for 60 min with horseradish peroxidaseeCconjugated secondary antibodies. Visualization was performed using an ECL kit per the manufacturer’s protocol (Amersham-Pharmacia). Antibodies used in immunoblotting studies were anti-Stat5b (sc-835) and anti-Crkl (sc-319) from Santa Cruz Biotechnology (Santa Cruz, CA) and anti-rabbit and anti-mouse from the ECL kit.
Amplification and direct DNA sequencing of coding region of murine Crkl gene.
Total RNA was extracted from splenocytes of NOD and B6 mice using Trizol kit (Sigma) according to instructions. Three micrograms of total RNA were converted to cDNA using reverse transcriptase (In Vitrogen) as described (6) and diluted 1:1 by TE buffer. Two microliters of diluted cDNA were used for PCR amplification of the entire coding region of the Crkl gene using forward primer 5'-ATGTCCTCCGCCAGGTTTC-3' and reverse primer 5'-AGCACAGCAATCACTCGTTG-3' with the mentioned conditions (94°C for 2 min for one cycle followed by 35 cycles at 94°C for 30 s, 64°C for 30 s, 72°C for 1 min 15 s, and 1 cycle for final extension at 72°C for 10 min.). The PCR products were loaded onto 1.5% agarose gel and then the DNA was recovered from the gel using QIAquick gel extraction kit (Qiagen). Purified PCR products were subjected to direct DNA sequencing using the above-mentioned primers on a Perkin Elmer 377 automated DNA sequencer.
Cloning, expression, and protein purification of murine Crkl.
The entire coding region of the murine Crkl region was PCR amplified with the same primers mentioned above, but the forward and reverse primer bear the SpeI and XhoI sequences at their 5' end for cloning purposes. The amplified PCR products were recovered from the agarose gel using QIAquick gel extraction kit (QIAgen) according to the manufacturer’s instructions. The purified PCR products were digested by appropriate restriction enzymes (New England Biolabs), then the full-length Crkl gene was ligated into the pTYB11 vector (IMPACT CN New England BioLabs) using the SpeI and XhoI sites and processed for transformation of BL21 DE3 host cells (Novagen). Screening of transfected clones was performed and all 18 clones were found to be positive for the insert gene. Direct DNA sequencing of the selected clones did not provide any proof of a mutation within the gene due to cloning procedures. The pTYB11 clones containing the entire murine Crkl coding sequence were induced by 1.0 mmol/l IPTG (Sigma) for 4 h at 30°C, and purification of the protein was carried out using chitin beads (IMPACT-CN system; New England Biolabs) as recommended by the manufacturer’s guidelines.
FarWestern analysis.
To determine direct interaction of Stat5b with Crkl, immunoprecipitated Stat5b from nuclear extracts of splenocytes of NOD and B6 mice, along with recombinant Aire and ADP proteins as negative controls, were run on native gels and transferred onto a PVDF membrane. The membrane was blocked in 5% blocker (Amersham-Pharmacia Biotech) in 1 x PBS-T for 1 h at room temperature. Membranes were then incubated in recombinant Crkl protein for 1 h followed by five washes with PBS-T each for 5 min. Then, membranes were incubated with anti-Crkl rabbit antibody conjugated with HRP for 1 h. The membranes were washed thoroughly and signals were detected using an ECL kit (Amersham-Pharmacia) according to the manufacturer’s instructions.
Profection.
Splenocytes were cultured and profection (protein transfection) was done using ProVectin reagent (Imgenex) per the method detailed below.
Splenocyte culture.
Spleen tissue was dissected, freed of adhering fat, and washed in sterile Hank’s balanced salt solution (Gibco). Mouse spleen was cut into several small pieces and ground between two sterilized frosted microscope glass slides. The cell suspension was collected in 15 ml Falcon tube and subjected to erythrocyte hemolysis using 1 ml RBC lysis buffer. The remaining debris was separated by gravity sedimentation by letting the tube stand up for about 10 min. The cell suspension was collected in another tube and centrifuged for 7 min at 2,000 rpm. The pellet was reconstituted in 1 ml Adoptive Immunotherapy MediaeCV (AIM-V) medium and the cell density was adjusted to 1 x 106 cell/ml. The splenocytes were cultured in supplemented, serum free, AIM-V at the concentration of 1.0 x 106 cells/ml. AIM-V was supplemented with 0.1 mmol/l modified Eagle’s medium (MEM) nonessential amino acids, 1.0 mmol/l sodium pyruvate, 3.0 x 10eC5 mmol/l 2-mercaptoethanol, 2.0 mmol/l L-glutamine, 10.0 mmol/l HEPES, and 100 units antibiotic-antimycotic solution (Gibco) at final concentration.
Protein transfection.
ProVectin reagent was prepared per the manufacturer’s instructions (Imgenex). The diluted recombinant Crkl in 25 e蘬 PBS (20 mmol/l Na phosphate, 150 mmol/l NaCl, pH 7.4) was used to hydrate the dried ProVectin reagent. AIM-V serum-free medium was added to make the volume 250 e蘬. This profection mix was added to splenocytes and cultured for 4 h in CO2 incubator in 96-well microtiter plates at 37°C. The splenocytes were induced for 15 min with GMCSF (1 e蘥/ml), and cells were collected in Falcon centrifuge tubes and centrifuged for 10 min/12,000 rpm. The medium was aspirated and cells washed with chilled PBS-T and stored at eC80°C. The cells are processed for cytosolic/nuclear extract preparation.
Electrophoretic mobility shift assays.
Electrophoretic mobility shift assays (EMSAs) of Stat5b were done exactly as described earlier (6). Stat5b consensus sequence (5'-TTTCTAGGAATT-3') was used for EMSA. Complementary oligos were used to make double-strand DNA, which was end-labeled using 170 e蘡i [-32P]ATP and T4 polynucleotide kinase. Equal amount of nuclear protein extract (5 e蘥) was incubated with double-strand oligonucleotides (consensus/mutant) at room temperature for 20 min in a binding buffer (13 mmol/l HEPES, pH 7.9, 80 mmol/l NaCl, 8% glycerol, 0.15 mmol/l EDTA, 1 mmol/l dithiothreitol) (12) and in the presence of poly dI-dC (2.5 e蘥/e蘬). For supershift assay, the protein extract was first incubated with 0.8 e蘥 of antibody for 20 min at room temperature. Oligonucleotide was then added and the reaction continued for another 30 min at room temperature. The anti-Stat5b (sc-835X) and anti-Crkl (sc-319) antibodies are rabbit polyclonal antibodies from Santa Cruz Biotechnology. DNA protein complexes were resolved on a 5% nondenaturing gel containing 5% glycerol and 0.5x Tris borate EDTA buffer (TBE). The gels were prerun in 0.5x TBE for 30 min at 150 V. A total of 2 e蘬 of loading mix (40% sucrose, 0.25% bromophenol blue, and 0.25% xylene cyanol) was added to each sample before loading. The gels were run at 250 mV for the first 15 min and then at constant 150 mV for about 4 h. The gels were exposed to X-ray films and then kept at eC80°C for at least 48 h.
Statistical analysis.
Western blots/EMSAs were digitized on a Chem-imager 4400 (Alpha Innotech). Band volumes were integrated using Phoretix 1-D image analysis software (version 5.20; Nonlinear Dynamics). The data were subjected to a two-tailed t test.
RESULTS
DNA binding ability of Stat5b in congenic mice.
We sequenced the Stat5b gene to confirm the Stat5b genotypes in two congenic mouse strains that are used in this study. The B6.NOD-c11 strain is a diabetes-resistant B6 mouse with a 42-cM Idd4 interval on chromosome 11 (extending from Csf2 to D11Mit42) that was derived from the NOD mouse and contains the NOD-derived mutant Stat5b. The NOD.Lc11 (formerly known as NOD.DR-3) strain is a diabetes-resistant NOD mouse which contains a 52-cM segment (25eC77 cM) spanning locus D11Mit87 to D11Mit42 from C57L/J mice and the C57L-derived wild-type Stat5b (Fig. 1A). The C57L and C57BL/6 were derived from common ancestors and appear to share the entire telomeric half of chromosome 11, which carries the Stat5a/5b/3 complex, by dense single nucleotide polymorphism analysis (13). The DNA binding abilities of B6.NOD-c11 and NOD.Lc11 Stat5b were analyzed using EMSA. The results were compared with those already obtained for the B6 and NOD mice. Protein extracts from the cytosolic and nuclear fractions were prepared from splenocytes with and without in vivo stimulation with GMCSF. Nuclear extracts were used in EMSA with the consensus DNA sequence recognized by Stat5 proteins (6). Both B6.NOD-c11 and NOD.Lc11 animals showed strong gel shift with the Stat5 consensus sequence in the samples stimulated with GMCSF as compared with a negligible shift in the unstimulated preparations (Fig. 1B). A supershift assay using a polyclonal anti-Stat5b antibody showed that the NOD.Lc11 Stat5b binding is similar to that of B6 (P = 0.4) but fivefold higher than NOD (P < 0.008) (Fig. 1C). These results are expected since the NOD.Lc11 mice carry the wild-type Stat5b gene. However, the Stat5b supershift is 7-fold (P < 10eC5) and surprisingly 8.5-fold higher (P < 10eC4) for B6.NOD-C11 than B6 and NOD, respectively (Fig. 1B and C). These results were unexpected because the B6.NOD-C11 mice have the NOD-derived mutant Stat5b, which is expected to bind DNA weakly. This observation led us to hypothesize that other factor(s) may modulate the DNA binding ability of Stat5b. Our data suggest that some cofactors should be located outside of the NOD-derived 42-cM chromosome 11 interval.
Western blot analysis of Crkl expression.
A number of cofactors may modulate the Stat5b binding ability. Western blot analyses were used to evaluate one of the important candidate proteins: Crkl. The protein concentration for Western analyses was normalized by multiple housekeeping proteins. We also found similar levels for cytosolic and nuclear Stat5b in B6 NOD, B6.NOD-c11, and NOD.Lc11, confirming our previous findings in NOD and B6 (Fig. 2). In contrast, Crkl had a significantly higher expression in cytosol of spleen extracts from B6 than NOD (P < 10eC4; Fig. 2A). To our surprise, the B6.NOD-c11 mice had higher Crkl protein in the cytosol than both B6 and NOD (P < 10eC4), while the NOD.Lc11 had Crkl levels equivalent of B6 (Fig. 2). The higher Crkl levels are also found in the nuclear fractions of B6.NOD-C11 compared with B6 (P < 0.005) and NOD (P < 0.0002) (Fig. 2B).
In vitro interaction between Stat5b and Crkl.
In order to study Crkl and Stat5b interaction, we produced recombinant protein using the full-length B6 Crkl gene cloned into the pTYB11 vector (Fig. 3). The purified recombinant Crkl protein was confirmed by Western analysis with an anti-Crkl antibody and MALDI-TOF analysis of the protein (Fig. 3). FarWestern analysis using purified recombinant Crkl and immunoprecipitated Stat5b revealed that Crkl interacted with Stat5b from B6, NOD, and B6.NOD-c11 (Fig. 3F). As Crkl does not bind two control proteins, the interaction between Stat5b and Crkl appears to be highly specific (Fig. 3F). Our results also suggest that the mutant Stat5b from NOD has normal ability to bind Crkl (Fig. 3F).
In vivo interaction between Crkl and Stat5b.
To confirm the potential in vivo interaction between Crkl and Stat5b, we performed immunoprecipitation using anti-Crkl or anti-Stat5b antibodies. As shown in Fig. 4, the immunoprecipitated products using the anti-Stat5b antibody contain a visible Crkl band that was confirmed by Western blot with the anti-Crkl antibody. These data suggest that Crkl can bind Stat5b in vivo. In subsequent experiments we sought to determine the presence of Crkl in the Stat5b gel shift complex. EMSA was done using Stat5 consensus sequence and supershifts were performed using an anti-Crkl antibody. A strong shift band was observed for GMCSF-induced samples from all four mouse strains (B6 NOD, B6.NOD-c11, and NOD.Lc11) (Fig. 5). While there was no supershift band induced by the Crkl antibody, the shift band was completely abolished in the supershift assay with the anti-Crkl antibody when the B6 nuclear extracts were examined, suggesting the presence of Crkl in the Stat5b shifted complex (Fig. 5, lane 4). However, the shifted band from nuclear extracts of NOD mice could not be abolished by the Crkl antibody (Fig. 5, lane 7), suggesting the absence or reduced amount of Crkl in the Stat5b-DNA binding complex in the NOD extracts. The reduction of the Stat5b shift by the Crkl antibody is significantly different between B6 and NOD (P < 0.01), and this observation is consistent with the lower Crkl expression level in the nuclear extracts of NOD spleens. The B6.NOD-c11 showed heavy DNA binding, and the addition of Crkl antibody caused a decrease in the shifted band intensity (Fig. 5A). NOD.Lc11 had shift levels similar to B6 and the DNA/Stat5b complex was abolished by the Crkl antibody (Fig. 5A). The intensity difference in shift position with and without the Crkl antibody was quantitated to estimate the Crkl-sensitive component in the Stat5b-Crkl complex (Fig. 5B). The Crkl-sensitive component in Stat5b-Crkl complex is highest in B6 and NOD.Lc11. B6.NOD-c11 had moderate Crkl sensitive fraction and NOD had negligible Crkl-sensitive component in the Stat5b-Crkl complex (Fig. 5C). Thus, the ability of Stat5b to bind DNA is correlated with the amount of Crkl, suggesting that Crkl is an important factor for the DNA binding ability of Stat5b.
Rescue of Stat5b DNA binding ability by Crkl.
In order to prove that Crkl is an active partner in the transcriptional activation by Stat5b, we carried out profection studies with Crkl on splenocytes from B6 and NOD mice. Western blot analysis revealed similar Stat5b levels in cytosolic and nuclear compartments of splenocytes (Fig. 6A and B). The Crkl levels were higher in profected cytosolic and nuclear extracts from B6 and NOD splenocytes compared with nonprofected controls (Fig. 6A and B). Profected B6 splenocytes exhibited significant (P < 0.00003) increase in Stat5b DNA binding ability compared with nonprofected B6 samples (Fig. 6C). The nonprofected NOD splenocytes had weaker Stat5b DNA binding ability compared with nonprofected B6 (P < 0.005), corroborating our earlier results of a loss in the DNA binding ability for the mutant Stat5b in NOD. The Crkl-profected NOD splenocytes displayed an enhancement in Stat5b DNA binding compared with nonprofected NOD (P < 0.003) and reached a level comparable to wild-type Stat5b from B6 mice (Fig. 6C and D). Therefore, Crkl appears to be able to rescue the DNA binding deficiency of the mutant Stat5b in NOD mice, and our data strongly argue for a cooperative role for Crkl in Stat5b DNA binding.
DISCUSSION
This study provided confirmatory evidence that the Stat5b mutation in the NOD mice has weaker DNA binding affinity toward the Stat5b consensus sequence. More importantly, we provided evidence that NOD is deficient in the levels of the adapter molecule Crkl encoded by the Crkl gene on chromosome 16. Using NOD congenic mice that contain the chromosome 11 interval from B6 (NOD.Lc11) and the B6 congenic mice that contain a chromosome 11 interval from the NOD (B6.NOD-C11), we demonstrated that the DNA binding affinity of Stat5b is correlated with the Stat5b mutation and also the expression level of Crkl. Furthermore, the Crkl expression level is determined by the chromosome 11 interval as well as the rest of the genome.
In a series of biochemical studies, we demonstrated that Crkl can interact with Stat5b in vitro and in vivo. FarWestern analysis showed that Crkl can specifically bind Stat5b. Immunoprecipitation with the anti-Stat5b antibody can also immunoprecipitate the Crkl protein, suggesting that Crkl binds to Stat5b in vivo. More importantly, profection experiments with recombinant Crkl protein increases the DNA binding affinity of Stat5b in both the B6 and NOD mice. The Crkl profection can indeed rescue the Stat5b binding deficiency observed in the NOD mice, suggesting that Crkl is a critical cofactor for Stat5b function.
Crkl is a member of the Crk family with high homology in the SH2/SH3 domains and a similar overall structure. The Crk family of proteins are linked to multiple signaling pathways in different cell types. These proteins function as adaptor molecules, which serve to bridge two or more interacting proteins using their binding domains. Crk proteins are involved in the early steps of lymphocyte activation through their SH2-mediated transient interaction with signal-transducing molecules, such as Cbl, ZAP-70, Cas-L, and Stat5.
T-cells play a critical role in mediating the destruction of pancreatic islet -cells as transfer of diabetic T-cells into young NOD mice or NOD/scid mice can rapidly transfer disease (14,15). It has been suggested that type 1 diabetes can be inhibited at the level of development and/or activation of diabetogenic T-cells (16). Thymic and peripheral T-cells from NOD mice exhibit a proliferative hyporesponsiveness upon crosslinking of the T-cell receptor (TCR)/CD3 complex (17,18). This T-cell hyporesponsiveness is genetically inheritable and is thought to be mediated by the reduced secretion of both interleukin-2 and interleukin-4 (18eC20) as well as a defect in the protein kinase C/ras/mitogen-activated protein kinase T-cell activation signaling pathway (17). The patterns of unresponsiveness and lymphokine secretion seen in anti-TCR/CD3-activated NOD thymic T-cells has also been observed in activated NOD peripheral spleen T-cells (19).
Engagement of TCR initiates signal transduction involving tyrosine phosphorylation of multiple effector molecules and the formation of multimolecular complexes at the receptor site. Our data suggest that the defect in NOD mice at the TCR-mediated signaling level could be due to the low levels of Crkl in NOD mice since Crk proteins are an established component in the plasma membrane scaffold of the T-cell signal transduction complex. The Crk adapter proteins are assumed to play a role in T-cell activation because of their induced association with tyrosine-phosphorylated -chain associated protein (ZAP-70) and Cbl and with the phosphatidylinositol 3-kinase regulatory subunit, p85, following engagement of the T-cell antigen receptor (21,22). Crk, ZAP-70, and p85 interact with tyrosine-phosphorylated Cbl, which serves as a major scaffold protein in activated T-cells. T-cell activationeCdependent conformational changes in Crk and/or p85 promote an initial direct or indirect low-affinity interaction between the two molecules, which is then stabilized by a secondary high-affinity interaction mediated by direct binding of the Crk-SH3(N) to the p85-PBP domain (22). The insulin receptor substrates play important roles in signal transduction emanating from the insulin and insulin-like growth factor-I receptors. Insulin receptor stably recruits full-length Crk by association with its SH2 domain in an autophosphorylation-dependent manner (23). Thus, an impaired Crk expression in NOD mice could lead to dysregulation of Crk-associated function in diabetogenic T-cells due to a defect in the signaling pathway of T-cell activation.
Autoreactive T-cells are normally deleted in the thymus through apoptosis, and defects in lymphocyte apoptosis may lead to autoimmune disorders. Thymocytes and peripheral lymphocytes in NOD are relatively resistant to various apoptosis signals (24,25). We have shown in this study that mice carrying substantial levels of Crkl are diabetes resistant while NOD mice with impaired Crkl expression are susceptible to diabetes. Thus, the reduced level of Crkl could contribute to the apoptosis resistance and the escape of diabetogenic T-cells from apoptosis observed in NOD. Finally, it is important to stress that type 1 diabetes is caused by a multitude of defects and that Crkl may lead to multiple functional defects, including the exacerbation of the weaker binding ability of the mutated Stat5b and defective T-cell signaling.
ACKNOWLEDGMENTS
This work was partially supported by a grant from the National Institute of Allergy and Infectious Disease (2P01 AI-42288) to J.X.S. The NOD.Lc11 strain was developed with support to M.M. from the Juvenile Diabetes Research Foundation, the American Diabetes Association, and the National Institute of Diabetes and Digestive and Kidney Diseases (R01-DK51112).
The authors thank Shiny Titus for assistance in graph generation.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Pipeleers D, Ling Z: Pancreatic beta cells in insulin-dependent diabetes. Diabetes Metab Rev 8:209eC227, 1992
Atkinson MA, Leiter EH: The NOD mouse model of type 1 diabetes: as good as it gets Nat Med 5:601eC604, 1999
Lyons PA, Wicker LS: Localising quantitative trait loci in the NOD mouse model of type 1 diabetes (Review). Curr Dir Autoimmun 1:208eC225, 1999
Wicker LS, Todd JA, Peterson LB: Genetic control of autoimmune diabetes in the NOD mouse. Annu Rev Immunol 13:179eC200, 1995
Rogner UC, Boitard C, Morin J, Melanitou E, Avner P: Three loci on mouse chromosome 6 influence onset and final incidence of type I diabetes in NOD.C3H congenic strains. Genomics 74:163eC171, 2001
Davoodi-Semiromi A, Laloraya M, Kumar GP, Purohit S, Jha RK, She JX: A mutant Stat5b with weaker DNA binding affinity defines a key defective pathway in nonobese diabetic mice. J Biol Chem 279:11553eC11561, 2004
Feller SM: Crk family adaptors-signalling complex formation and biological roles. Oncogene 20:6348eC6371, 2001
Oda A, Wakao H, Fujihara M, Ozaki K, Komatsu N, Tanaka S, Ikeda H, Miyajima A, Ikebuchi K: Thrombopoietin and interleukin-2 induce association of CRK with STAT5. Biochem Biophys Res Commun 278:299eC305, 2000
Schulze H, Ballmaier M, Welte K, Germeshausen M: Thrombopoietin induces the generation of distinct Stat1, Stat3, Stat5a and Stat5b homo- and heterodimeric complexes with different kinetics in human platelets. Exp Hematol 28:294eC304, 2000
Fish EN, Uddin S, Korkmaz M, Majchrzak B, Druker BJ, Platanias LC: Activation of a CrkL-stat5 signaling complex by type I interferons. J Biol Chem 274:571eC573, 1999
Ota J, Kimura F, Sato K, Wakimoto N, Nakamura Y, Nagata N, Suzu S, Yamada M, Shimamura S, Motoyoshi K: Association of CrkL with STAT5 in hematopoietic cells stimulated by granulocyte-macrophage colony-stimulating factor or erythropoietin. Biochem Biophys Res Commun 252:779eC786, 1998
Arreaza GA, Cameron MJ, Jaramillo A, Gill BM, Hardy D, Laupland KB, Rapoport MJ, Zucker P, Chakrabarti S, Chensue SW, Qin HY, Singh B, Delovitch TL: Neonatal activation of CD28 signaling overcomes T cell anergy and prevents autoimmune diabetes by an IL-4-dependent mechanism. J Clin Invest 100:2243eC2253, 1997
Petkov PM, Ding Y, Cassell MA, Zhang W, Wagner G, Sargent EE, Asquith S, Crew V, Johnson KA, Robinson P, Scott VE, Wiles MV: An efficient SNP system for mouse genome scanning and elucidating strain relationships. Genome Res 14:1806eC1811, 2004
Daniel D, Gill RG, Schloot N, Wegmann D: Epitope specificity, cytokine production profile and diabetogenic activity of insulin-specific T cell clones isolated from NOD mice. Eur J Immunol 25:1056eC1062, 1995
Crawford M, Daniel D, Wegmann D, Yang H, Gill RG: Autoimmune islet damage mediated by insulin-specific T cells. Transplant Proc 29:758eC759, 1997
McDuffie M: Derivation of diabetes-resistant congenic lines from the nonobese diabetic mouse. Clin Immunol 96:119eC130, 2000
Rapoport MJ, Lazarus AH, Jaramillo A, Speck E, Delovitch TL: Thymic T cell anergy in autoimmune nonobese diabetic mice is mediated by deficient T cell receptor regulation of the pathway of p21ras activation. J Exp Med 177:1221eC1226, 1993
Zipris D, Lazarus AH, Crow AR, Hadzija M, Delovitch TL: Defective thymic T cell activation by concanavalin A and anti-CD3 in autoimmune nonobese diabetic mice: evidence for thymic T cell anergy that correlates with the onset of insulitis. J Immunol 146:3763eC3771, 1991
Rapoport MJ, Jaramillo A, Zipris D, Lazarus AH, Serreze DV, Leiter EH, Cyopick P, Danska JS, Delovitch TL: Interleukin 4 reverses T cell proliferative unresponsiveness and prevents the onset of diabetes in nonobese diabetic mice. J Exp Med 178:87eC99, 1993
Zipris D, Crow AR, Delovitch TL: Altered thymic and peripheral T-lymphocyte repertoire preceding onset of diabetes in NOD mice. Diabetes 40:429eC435, 1991
Gelkop S, Isakov N: T cell activation stimulates the association of enzymatically active tyrosine-phosphorylated ZAP-70 with the Crk adapter proteins. J Biol Chem 274:21519eC21527, 1999
Gelkop S, Babichev Y, Isakov N: T cell activation induces direct binding of the Crk adapter protein to the regulatory subunit of phosphatidylinositol 3-kinase (p85) via a complex mechanism involving the Cbl protein. J Biol Chem 276:36174eC36182, 2001
Klammt J, Barnikol-Oettler A, Kiess W: Mutational analysis of the interaction between insulin receptor and IGF-I receptor with c-Crk and Crk-L in a yeast two-hybrid system. Biochem Biophys Res Commun 325:183eC190, 2004
Bergman ML, Cilio CM, Penha-Goncalves C, Lamhamedi-Cherradi SE, Lofgren A, Colucci F, Lejon K, Garchon HJ, Holmberg D: CTLA-4eC/eC mice display T cell-apoptosis resistance resembling that ascribed to autoimmune-prone non-obese diabetic (NOD) mice. J Autoimmun 16:105eC113, 2001
Colucci F, Bergman ML, Penha-Goncalves C, Cilio CM, Holmberg D: Apoptosis resistance of nonobese diabetic peripheral lymphocytes linked to the Idd5 diabetes susceptibility region. Proc Natl Acad Sci U S A 94:8670eC8674, 1997(Malini Laloraya, Abdoreza)