Synergistic Expression of the CXCL10 Gene in Response to IL-1β and IFN-γ Involves NF-κB, Phosphorylation of STAT1 at Tyr701, and Acetylation of Histones H3 and H4

Susan J. Burke, Matthew R. Goff, Danhong Lu, David Proud, Michael D. Karlstad and J. Jason Collier
J Immunol July 1, 2013, 191 (1) 323-336; DOI: https://doi.org/10.4049/jimmunol.1300344

Abstract

The CXCL10 gene encodes a peptide that chemoattracts a variety of leukocytes associated with type 1 and type 2 diabetes. The present study was undertaken to determine the molecular mechanisms required for expression of the CXCL10 gene in response to IL-1β and IFN-γ using rat islets and β cell lines. IL-1β induced the expression of the CXCL10 gene and promoter activity, whereas the combination of IL-1β plus IFN-γ was synergistic. Small interfering RNA–mediated suppression of NF-κB p65 markedly inhibited the ability of cytokines to induce the expression of the CXCL10 gene, whereas targeting STAT1 only diminished the synergy provided by IFN-γ. Furthermore, we found that a JAK1 inhibitor dose dependently reduced IFN-γ–controlled CXCL10 gene expression and promoter activity, concomitant with a decrease in STAT1 phosphorylation at Tyr701. We further discovered that, although the Tyr701 phosphorylation site is inducible (within 15 min of IFN-γ exposure), the Ser727 site within STAT1 is constitutively phosphorylated. Thus, we generated single-mutant STAT1 Y701F and double-mutant STAT1 Y701F/S727A adenoviruses. Using these recombinant adenoviruses, we determined that overexpression of either the single- or double-mutant STAT1 decreased the IFN-γ–mediated potentiation of CXCL10 gene expression, promoter activity, and secretion of protein. Moreover, the Ser727 phosphorylation was neither contingent on a functional Y701 site in β cells nor was it required for cytokine-mediated expression of the CXCL10 gene. We conclude that the synergism of IL-1β and IFN-γ to induce expression of the CXCL10 gene requires NF-κB, STAT1 phosphorylated at Tyr701, recruitment of coactivators, and acetylation of histones H3 and H4.

Introduction

Diabetes mellitus occurs when a reduction in functional pancreatic β cell mass is severe enough to produce hyperglycemia. This decrease in islet β cell function and mass occurs through organ-specific autoimmunity in type 1 diabetes (1, 2) and is strongly correlated with obesity-associated tissue dysfunction in type 2 diabetes (3, 4). A commonality between both of these endocrine diseases is the increase in pancreatic β cell production of soluble factors that, once secreted, operate as chemotactic cytokines (hereafter referred to as chemokines).

Chemokines are secreted peptides that recruit immune cells to the site of inflammation (5, 6). The chemokines are grouped according to placement of an N-terminal cysteine residue within the peptide, leading to assignment to the C, C-C, C-X-C, or C-X3-C families. Approximately 50 discrete chemokines have been identified, corresponding to ∼20 distinct chemokine receptors (reviewed in Refs. 6, 7). The expression of specific chemokine genes in the pancreatic β cell is increased in both rodents and humans prior to and during diabetes (8–11); therefore, it is possible that islet β cells are the very source of the soluble factors that initiate an influx of immune cells that contribute to their demise. The C-C and C-X-C chemokines are the largest families and contain some of the most well-studied chemotactic peptides, including CCL2 and CXCL10.

CXCL10 expression is elevated in pancreatic islets isolated from individuals with type 1 or type 2 diabetes (12, 13). Additionally, CXCL10 circulates at higher levels in diabetic patients than in nondiseased individuals (8, 14, 15). Transgenic expression of CXCL10 driven by pancreatic β cell–specific promoters revealed islets enriched with specific leukocytes (10, 16). An increase in the macrophage, T lymphocyte, and other immune cell populations results in an insulitis (islet inflammation) that leads to the release of proinflammatory cytokines, such as IL-1β and IFN-γ. Pancreatic β cell exposure to IL-1β and IFN-γ leads to accumulation of intracellular mediators of inflammation that suppress insulin secretion (17, 18), increased expression of genes that contribute to the inflammatory response (19–22), and, eventually, a decrease in functional β cell mass (23–25). Although IL-1β and IFN-γ clearly activate signaling pathways culminating in pancreatic β cell death and dysfunction, little is known about the mechanisms and associated protein factors that control gene expression of secreted proteins that potentially create a feed-forward amplification of this vicious inflammatory cycle.

IL-1β signals through the NF-κB pathway to regulate the expression of a variety of genes involved in inflammation. NF-κB is a dimer of proteins that includes RelA/p65, RelB, c-Rel, p50, and p52. p65 is held in the cytoplasm by a group of regulatory proteins, referred to as IκBs. Phosphorylation-induced degradation of IκBα by IκB kinases unmasks a nuclear-localization signal within p65, allowing it to be translocated into the nucleus (26). Nuclear p65 acts as either a homodimer or heterodimer with p50 to activate gene transcription. In addition, stimuli that activate NF-κB, such as IL-1β, are often potentiated by concurrent activation of the IFN-γ–signaling pathway (22, 27). However, mechanisms responsible for the IFN-γ–mediated potentiation of NF-κB–driven gene transcription in pancreatic β cells are not well understood.

IFN-γ signals through a cell surface receptor to activate STAT proteins to control transcriptional outputs (28). STAT1 is phosphorylated at Tyr701 and Ser727, and these sites are often responsible for maximal transcriptional activity in response to a specific stimulus. However, the requirement for STAT1 and associated phospho-acceptor sites in the cytokine-mediated regulation of the CXCL10 gene in pancreatic β cells has not been established. Therefore, we undertook the current study to investigate the signals, genomic response elements, and primary DNA-binding factors responsible for controlling expression of the CXCL10 gene by IL-1β and IFN-γ in insulin-producing cells. Several key discoveries were made: the CXCL10 gene is activated by IL-1β, a response that is synergistic with concomitant activation of the IFN-γ–signaling pathway; the proximal and distal κB sites within the CXCL10 gene promoter were required for the activation by both IL-1β and IFN-γ; the proximal and distal IFN-stimulated response elements (ISRE) within the CXCL10 promoter were required for signal integration by both IL-1β and IFN-γ; the Tyr701 site in STAT1 is inducible in pancreatic β cells, whereas the Ser727 site is constitutively phosphorylated; and maximal synthesis and secretion of CXCL10 are dependent upon IFN-γ–mediated phosphorylation of STAT1 at Tyr701 and increases in acetylation of histones H3 and H4.

Materials and Methods

Cell culture, islet isolation, and reagents

The 832/13 and INS-1E cell lines are derived from the parental INS-1 cell line (29). Culture and passage of the INS-1–derived 832/13 and INS-1E cell lines were described (30, 31). Islets were isolated from Wistar rats per a previous report (19). IL-1β was purchased from Thermo Fisher Scientific, IFN-γ was from Shenandoah Biotechnology, and JAK inhibitor was from EMD Millipore. The three copies of the consensus γ-activated sequence (3xGAS)-luciferase reporter plasmid were a gift from Dr. Thomas Spencer, Washington State University, Pullman, WA (32). Recombinant adenoviruses expressing β−galactosidase (βGAL) (33), wild-type STAT1 and Y701F STAT1 (20), IκBα superrepressor (34), and p65 wild-type (35) have been described. Silencer Select small interfering RNA (siRNA) oligonucleotides were purchased from Life Technologies.

Plasmid construction and site-directed mutagenesis

pGL3-luciferase vectors containing approximately −1 kb and −400 bp of the human CXCL10 promoter, in addition to NF-κB (distal and proximal) and proximal ISRE mutations in the −1 kb promoter, were described previously (36). Luciferase reporter vectors containing five copies of the distal κB element or five copies of the proximal κB element from the CXCL10 gene promoter were also described (37). The aforementioned CXCL10 −1 kb construct was used to generate mutations in the distal ISRE, whereas the CXCL10 −400 bp construct was used to generate a proximal ISRE mutation by site-directed mutagenesis using the QuikChange Site-Directed Mutagenesis kit, according to the manufacturer’s instructions (Agilent Technologies). The following primer pairs were used to incorporate these mutations: distal ISRE: (F) 5′−CATCGTAAATCAACCTGTTTCatTTCTGTCTCAAGCTACTTTC-3′ and (R) 5′−GAAAGTAGCTTGAGACAGAAatGAAACAGGTTGATTTACGATG-3′; proximal ISRE: (F) 5′−CAGTTCATGTTTTGGAcAGTGAcACCTAATTCAC-3′ and (R) 5′−GTGAATTAGGTgTCACTgTCCAAAACATGAACTG-3′.

Transfection and luciferase assay

Transient transfections of luciferase reporter constructs into 832/13 cells were achieved using TransIT-LT1 transfection reagent (Mirus Bio). siRNA oligonucleotides (siRNA ID for p65: s159517; siRNA ID for STAT1: s129044; negative control siRNA: catalog no. M4611) were transfected using DharmFECT Transfection Reagent 1 (Thermo Scientific). Concomitant transfection of plasmid and siRNA was performed using DharmaFECT Duo (Thermo Scientific). Whole-cell lysates were prepared, and luciferase activity was measured, as previously described (19).

Recombinant adenoviruses

pAC.CMV STAT1 (a gift from Dr. Guoxun Chen, University of Tennessee, Knoxville) was used as a template to generate the S727A and S727T mutations in STAT1. Recombinant adenovirus expressing STAT1 Y701F mutant (20) was described previously, and the pAC.CMV shuttle vector containing the Y701F mutation was used as a template to generate the Y701F/ S727A double mutation. Primer pairs used to incorporate mutations in STAT1 were as follows: S727A: (F) 5′−GAAAACCTGCTTCCCATGGCTCCAGAGGAGTTTGAC-3′ and (R) 5′−GTCAAACTCCTCTGGAGCCATGGGAAGCAGGTTTTC-3′; S727T: (F) 5′−GAAAACCTGCTTCCCATGACTCCAGAGGAGTTTGAC-3′ and (R) 5′−GTCAAACTCCTCTGGAGTCATGGGAAGCAGGTTTTC-3′. All mutations were generated using the QuikChange Site-Directed Mutagenesis kit, according to the manufacturer’s recommended protocol, and verified by dideoxy sequencing prior to construction of recombinant adenoviral vectors. Recombinant adenoviruses were generated by cotransfection of the pAC.CMV STAT1 shuttle vectors and pJM17 viral genome into HEK293 cells, as previously documented (38). All adenoviruses were confirmed to be insert positive and E1A negative, per prior recommendations (39).

RNA isolation, synthesis of cDNA, and mRNA expression analysis

Total RNA isolation, cDNA synthesis, and gene-expression analysis via real-time RT-PCR were performed according to previous protocols (19). Primer sequences for RS9, p65, E1A, STAT1, and CXCL10 were generated using Primer3Plus software and are available upon request.

Protein isolation and immunoblotting

Whole-cell lysates were prepared using M-PER lysis solution supplemented with a protease/phosphatase inhibitor mixture (both from Thermo Fisher Scientific). Proteins were quantified by bicinchoninic acid assay (Thermo Fisher Scientific), separated on a 4–12% SDS-PAGE gel (Invitrogen), and transferred to polyvinylidene difluoride membranes. Membranes were blocked with Membrane Blocking Solution (Invitrogen) for 45 min prior to primary Ab incubation in 1% polyvinylpyrrolidone (24, 40). Membranes were stripped and reprobed with each Ab in the order shown in the respective figures (top to bottom). Abs used were from the following sources: tubulin, PO4-Y701 STAT1, PO4-S727 STAT1, and total STAT1 (Cell Signaling) and β-actin (Sigma-Aldrich).

Chromatin immunoprecipitation

832/13 cells were grown to confluence in 10-cm dishes, using one dish/treatment condition. After treatment of cells, media were aspirated, and formaldehyde was added at a final concentration of 1% (diluted in PBS) and incubated at room temperature for 10 min. Glycine was added to a final concentration of 125 mM for 5 min at room temperature. Cells were washed once with cold PBS and scraped into 1 ml prechilled PBS with 1× protease inhibitors (Halt Protease Inhibitor Cocktail; Thermo Scientific). Cell pellets were collected by centrifugation at 2000 rpm for 2 min at 4°C. Supernatants were discarded, and pellets were reconstituted in 0.5 ml SDS Lysis Buffer containing 1× protease inhibitors (1% SDS, 10 mM EDTA, 50 mM Tris-HCl [pH 8]). Lysates were incubated on ice for 15 min, followed by DNA sonication at 4°C using a Misonix Sonicator S-4000 (Thermo Scientific). The following sonication conditions generated fragments of DNA between 100 and 500 bp in length: Amplitude: 4, Process Time: 10 min, Pulse on time: 10 s, Pulse off time: 20 s. Sonicated fragments were centrifuged at 12,000 × g for 10 min at 4°C, and supernatants were transferred to a clean low-retention microcentrifuge tube. Preclearing was performed for 1 h at 4°C with rotation using 5 μl magnetic Protein G beads (Thermo Scientific). Once beads were removed, 100 μl sheared cross-linked DNA was diluted 10-fold in dilution buffer containing 1× protease inhibitors (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 167 mM NaCl, 17 mM Tris [pH 8]). Two micrograms of immunoprecipitating Abs was added and incubated overnight at 4°C with rotation. Immunoprecipitation of DNA–protein–Ab complexes was performed by incubation with 10 μl Protein A (for rabbit polyclonal Abs) or Protein G (for mouse monoclonal antisera) Dynabeads (Invitrogen) for 1 h at 4°C. The magnetic beads binding to the Ab–Ag–chromatin complexes were washed sequentially with 0.5 ml the following buffers: 1× Low-Salt Buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 150 mM NaCl, 20 mM Tris [pH 8]), 1× High-Salt Buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 500 mM NaCl, 20 mM Tris [pH 8]), 1× LiCl Buffer (0.25 M LiCl, 1% Nonidet P-40, 1 mM EDTA, 10 mM Tris [pH 8]), and 2× Tris-EDTA. Once the last wash buffer (Tris-EDTA) was completely removed, 100 μl 10% Chelex 100 resin in PBS was added, vortexed for 10 s, and heated to 100°C for 10 min. Following centrifugation at 17,000 × g for 1 min at 4°C, the supernatants were transferred to new tubes. A total of 120 μl nuclease-free water was added, vortexed, and recentrifuged. Supernatants were pooled. Inputs were processed as followed: 20 μl sonicated, precleared DNA was incubated overnight at 65°C with NaCl to a final concentration of 200 mM. Input DNA was then treated with RNase A for 30 min at 37°C and proteinase K for 1 h at 45°C, followed by purification using a QIAGEN Cleanup kit (QIAGEN). A total of 2.5 μl DNA was used as a template for downstream SYBR Green–based PCR reactions. Fold enrichment of promoter fragments was calculated using the δ-δ Ct method: 2(Ctexperimental immunoprecipitation – CtIgG control immunoprecipitation). All data are shown as the fold chromatin immunoprecipitation (ChIP) signal over the IgG background signal. Negative control regions corresponding to cAMP response element–binding regions of the NR4A2 gene were used to confirm transcription factor binding specificity. Abs were from the following sources: normal rabbit and mouse IgG were from Sigma Aldrich); p65, p50, CBP, p300, and STAT1 were from Santa Cruz Biotechnology; histone acetyl H3 K9, PO4-Ser2 RNA polymerase II, and PO4-Ser5 RNA polymerase II were from Abcam; and histone acetyl H4 K5, 8, 12, 16 and total RNA polymerase II were from Millipore. Primer sequences for amplifying the CXCL10 proximal and distal NF-κB and ISRE sites were generated using Primer3Plus software and are available upon request.

ELISA

Secretion of CXCL10 into the media was detected using the Rat IP-10 (CXCL10) Mini ELISA Development Kit (PeproTech), according to the suggested protocol. CXCL10 release into the media was normalized to total protein to account for any potential differences in cell number.

Statistical analysis

One-way ANOVA, followed by the Tukey post hoc correction, or Student t test was used to identify statistical significances at the confidence levels of 90, 95, and 99%, as reported within the figure legends.

Results

Expression of the CXCL10 gene is increased by IL-1β and IFN-γ in rat islets and β cell lines

CXCL10 levels are elevated in individuals with diabetes mellitus (41, 42), and expression of CXCL10 has been colocalized to the β cells of the pancreatic islets (13). Thus, we investigated the signals that increase the expression of the CXCL10 gene using rat islets and β cell lines. We found that 1 ng/ml IL-1β induced the CXCL10 gene by ∼47-fold in 832/13 rat insulinoma cells. After setting this 47-fold IL-1β response at 100%, we discovered that 100 U/ml IFN-γ potentiated the IL-1β response an additional 8-fold (Fig. 1A). A similar response was obtained in the INS-1E cell line (Fig. 1B). IL-1β and IFN-γ individually enhanced steady-state CXCL10 mRNA levels in isolated rat islets, and the combination of these cytokines synergistically augmented expression of the CXCL10 gene (Fig. 1C). To further examine the transcriptional responses, we next transfected 832/13 cells with a plasmid containing ∼1 kb of the CXCL10 proximal gene promoter (upstream of the transcriptional start site), which includes known κB and ISRE sequences, driving a luciferase reporter gene. We found that IL-1β increased promoter activity by 9-fold (set at 100% in Fig. 1D), whereas IFN-γ alone did not augment transcription over baseline. However, the combination of IL-1β plus IFN-γ induced a synergistic 27-fold cumulative response (shown as 3-fold potentiation of the IL-1β response; Fig. 1D). We further note that CXCL10 protein was secreted from 832/13 β cells in response to IL-1β and that this secretory response was further amplified by the addition of IFN-γ + IL-1β (Fig. 1E). Thus, the ability of IL-1β and IFN-γ to induce synthesis and secretion of CXCL10 is due, at least in part, to increased transcription of the CXCL10 gene by these proinflammatory cytokines.

FIGURE 1.
FIGURE 1.

Expression of the CXCL10 gene is increased by IL-1β and IFN-γ in rat islets and β cell lines. 832/13 (A) and INS-1E (B) insulinoma cells were stimulated with 1 ng/ml IL-1β, 100 U/ml IFN-γ, or IL-1β plus 100 U/ml IFN-γ for 6 h. (C) Rat islets were treated with 10 ng/ml IL-1β, 100 U/ml IFN-γ, or a combination of both cytokines for 6 h. (A–C) Total RNA was isolated, and CXCL10 mRNA levels were measured and normalized to those of the housekeeping gene, Ribosomal S9 (RS9). (D) 832/13 cells were transfected with −1 kb of the CXCL10 promoter upstream of the transcriptional start site fused to a luciferase reporter. At 24 h posttransfection, cells were stimulated for 4 h with 1 ng/ml IL-1β, 100 U/ml IFN-γ, or a combination of both cytokines. Relative promoter activity was measured and normalized to protein content via BCA assay. **p < 0.01 versus the indicated groups. (E) 832/13 cells were treated with 1 ng/ml IL-1β, 100 U/ml IFN-γ, or both cytokines for 3, 6, or 12 h. CXCL10 release into the media was measured by ELISA and was normalized to protein content via BCA assay. Data are mean ± SEM from three or four individual experiments, some of which were performed in duplicate. #p < 0.1, **p < 0.01 versus no treatment control (NT) at 3-h time point.

Deletion and mutational analysis of the CXCL10 gene promoter reveals the importance of κB and ISRE sequences controlling gene transcription in response to cytokines

We next examined the transcriptional response to IL-1β and IFN-γ, alone and in combination, using either −1 kb or −400 bp constructs of the CXCL10 proximal gene promoter. The −1 kb genomic sequence containing two NF-κB response elements was 8-fold responsive to 1 ng/ml IL-1β (IL-1β response set at 100%; Fig. 2A). IFN-γ alone did not activate the promoter (Fig. 1D), but it did potentiate the response to IL-1β by an additional 3.3-fold (Fig. 2A). Mutating either the distal or proximal κB element led to an 89.4 and 67.8% decrease, respectively, in the ability of IL-1β to induce transcription of the promoter luciferase reporter gene. Additionally, the ability of IFN-γ to potentiate the response to IL-1β was also diminished by ∼50% (1.5- and 1.6-fold potentiation for distal and proximal mutations, respectively).

FIGURE 2.
FIGURE 2.

Cytokine-mediated activation of the CXCL10 promoter requires both NF-κB and ISRE response elements. (A) 832/13 cells were transfected with −1 kb of the CXCL10 promoter (−1 kb) or −1 kb with mutations in the distal or proximal NF-κB response elements. *wp < 0.05 versus −1 kb with IL-1β (w = white bar), **wp < 0.01 versus −1 kb with IL-1β, **bp < 0.01 versus −1 kb with IL-1β + IFN-γ (bblack bar). (B) 832/13 cells were transfected with either distal or proximal 5xNF-κB–luciferase plasmids overnight, followed by analysis of promoter activity. (C) 832/13 cells were transfected with the indicated plasmids; 24 h later they were treated for 4 h with increasing concentrations of IL-1β (0.1, 1, and 10 ng/ml). (D) 832/13 cells were transfected with −400 bp of the CXCL10 promoter (−400 bp) or −400 bp with a mutation in the ISRE response element. **p < 0.01 versus −400 bp with IL-1β + IFN-γ. (E) 832/13 cells were transfected with −1 kb or −1 kb with mutations in the proximal ISRE or distal ISRE response elements. **wp < 0.01 versus −1 kb with IL-1β, **bp < 0.01 versus −1 kb with IL-1β + IFN-γ. In (A), (D), and (E), at 24 h posttransfection, cells were incubated with 1 ng/ml IL-1β (white bars), 100 U/ml IFN-γ (gray bars), or a combination of both IL-1β and IFN-γ (black bars) for 4 h. Luciferase activity values are represented as mean ± SEM from three individual experiments.

Because of the differential magnitude of transcriptional effects observed after mutating the distal versus proximal κB response element on the cytokine response, we used multimerized response elements to individually examine transcriptional activities in both the basal (unstimulated) and inducible states. We discovered that the distal κB element drove almost a 2-fold greater transcriptional response in the basal state compared with the proximal κB element (Fig. 2B). In contrast, when each individual κB construct was exposed to a concentration range of IL-1β, the proximal κB element was revealed to have the most robust response to IL-1β (56.6-fold induction of the proximal κB element versus 26.2-fold induction of the distal κB element in response to 0.1 ng/ml IL-1β; Fig. 2C). This observation remained consistent, regardless of IL-1β concentration.

We next examined the cytokine responsiveness of a CXCL10 promoter–luciferase deletion construct, which contains 400 nucleotides upstream of the transcriptional start site but does not include the distal ISRE element. In these experiments, we found that the proximal ISRE site was largely responsible for the IFN-γ–mediated potentiation of the IL-1β response, because deletion of the 600 nucleotides upstream (containing the distal ISRE site) had minimal effects on the IL-1β responsiveness or the ability of IFN-γ to potentiate the IL-1β response (Fig. 2D).

To further investigate the observations made using the −400 kb promoter, we mutated the distal and proximal ISRE sites individually and in combination within the −1 kb gene promoter construct. Using these promoter constructs, we found that mutations in the proximal ISRE element eliminated both IL-1β– and IFN-γ–mediated gene activation (Fig. 2E). Moreover, although the distal site mutant also markedly impaired IL-1β activation (94% decrease relative to wild-type), the potentiation by IFN-γ was maintained (2.8-fold in wild-type versus 3.1-fold in distal ISRE mutant; Fig. 2E), albeit with a diminution in the magnitude of the response. Mutations in both the distal and proximal ISRE elements completely abolished transcription of the −1 kb CXCL10 gene promoter in response to IL-1β and IFN-γ (Fig. 2E). Therefore, despite both κB sites being intact, the ISRE elements contain nucleotide information required to confer overall signal integration at the CXCL10 gene promoter in response to IL-1β. Furthermore, although both ISRE sites clearly participate in regulation of the IL-1β response, the proximal site most likely is the key regulator of the IFN-γ–mediated potentiation.

p65/RelA and STAT1 are required for activation of the CXCL10 gene by cytokines

After determining the genomic response elements required for IL-1β– and IFN-γ–mediated activation of the CXCL10 gene (Fig. 2), we next examined which primary DNA-binding factors were necessary for gene activation. Because p65 and STAT1 were the two most likely candidates to activate κB and ISRE elements, respectively, we used two strategies for this approach: overexpression of the IκBα superrepressor (IκBαSR) and siRNA-mediated targeting of either the p65/RelA subunit of NF-κB or STAT1. The IκBαSR protein contains S32A/S36A mutations that prevent phosphorylation-induced degradation and, thus, promote cytoplasmic retention of p65 (34). Overexpression of the IκBαSR in 832/13 cells completely blocked the cytokine-mediated induction of the CXCL10 gene (Fig. 3A), which indicates involvement of NF-κB proteins and is consistent with the requirement of κB sites to confer transcriptional responsiveness by cytokines (Fig. 2A). In addition, IκBαSR diminished the cytokine-mediated secretion of CXCL10 (49% decrease relative to βGAL; Fig. 3B), confirming that NF-κB–mediated increases in transcription also augment protein production and release. To directly examine p65 subunit involvement, we next used siRNA-directed suppression and found that expression of the p65 transcript was decreased by 75.4% by sip65, whereas the STAT1 transcript was diminished by 91.3% with siSTAT1 duplex transfection (Fig. 3C). The reduction in mRNA corresponding to the p65 and STAT1 genes produced a decrease in total p65 and STAT1 protein (Fig. 3D). Using these specific siRNA approaches, we discovered that the IL-1β–stimulated expression of the CXCL10 gene was impaired by 74% (Fig. 3E, white bars); in contrast, the potentiation of the IL-1β response by IFN-γ was unaffected (5.7-fold versus 5.6-fold; Fig. 3E, black bars). Alternatively, depletion of STAT1 by siRNA duplex transfection decreased the IFN-γ–mediated potentiation by 28.7% (12.2-fold versus 8.7-fold potentiation, Fig. 3F, black bars), but it did not impair the CXCL10 response to IL-1β (Fig. 3F, white bars). Similarly, using −1 kb of the CXCL10 proximal gene promoter, we observed the strongest repression of promoter activity when p65 abundance was decreased by siRNA duplex transfection (77% decrease in IL-1β response while potentiation by IFN-γ was maintained; Fig. 3G). Again, targeting STAT1 with siRNA duplexes only impacted the IFN-γ–mediated potentiation of the IL-1β response (56% decrease relative to siScramble control; Fig. 3F, black bars).

FIGURE 3.
FIGURE 3.

p65/RelA and STAT1 are required for activation of the CXCL10 gene by cytokines. (A) 832/13 cells were transduced with recombinant adenoviruses expressing either βGAL or IκBα S32A/S36A (IκBαSR). At 24 h posttransduction, cells were stimulated for 6 h with either 1 ng/ml IL-1β or IL-1β plus 100 U/ml IFN-γ. *wp < 0.05 versus βGAL with IL-1β, *bp < 0.05 versus βGAL with IL-1β + IFN-γ. (B) 832/13 cells were transduced with the indicated adenoviruses. At 24 h posttransduction, cells were left untreated or were stimulated for 12 h with 1 ng/ml IL-1β plus 100 U/ml IFN-γ. CXCL10 release into the media was quantified by ELISA. **p < 0.01. (C and D) 832/13 cells were transfected with siRNA duplexes against either p65 or STAT1 with a negative control siScramble. (C) Expression of p65 and STAT1 was quantified by RT-PCR. (D) STAT1 and p65 abundance was measured via immunoblot with β-actin as a loading control. 832/13 cells were transfected with siRNA duplexes targeting either p65 (E) or STAT1 (F), with a negative scrambled sequence as a control for specificity (siScramble). At 48 h posttransfection, cells were stimulated with either 1 ng/ml IL-1β or IL-1β plus 100 U/ml IFN-γ for 6 h. In (E): *p < 0.05 versus siScramble with IL-1β. (F): *p < 0.05 versus siScramble with IL-1β + IFN-γ. In (A), (E), and (F), total RNA was isolated, and the mRNA level of CXCL10 was normalized to that of RS9. 832/13 cells were transfected concomitantly with CXCL10 −1 kb and siRNA duplexes targeting either p65 (G) or STAT1 (H). At 24 h posttransfection, cells were stimulated with IL-1β alone or in combination with IFN-γ for 4 h prior to measurement of luciferase activity. Values are mean ± SEM from three or four individual experiments. **wp < 0.01 versus siScramble with IL-1β, **bp < 0.01 versus siScramble with IL-1β + IFN-γ, **p < 0.01 versus siScramble with IL-1β + IFN-γ.

p65, STAT1, and RNA polymerase II are recruited to the CXCL10 gene promoter in a signal-specific manner

Using ChIP assays, we examined the occupancy of the key regulatory factors p65 and STAT1 at the CXCL10 gene promoter in response to IL-1β or IFN-γ. We detected p65 occupancy at the CXCL10 κB sites in the basal state (Fig. 4A, white bar), with a 130% increase in response to IL-1β (Fig. 4A, black bars).

FIGURE 4.
FIGURE 4.

p65, STAT1, and RNA polymerase II are recruited to the CXCL10 gene promoter in a signal-specific manner. (A–C) 832/13 cells were stimulated with either 1 ng/ml IL-1β or 100 U/ml IFN-γ for 0, 15, 30, and 60 min. ChIP assays were performed using Abs against p65 (A), p50 (B), or STAT1 (C) and normalized to assays performed using control IgG Ab. The recovered CXCL10 NF-κB and ISRE genomic regions were amplified by SYBR Green–based RT-PCR. (D) 832/13 cells were incubated with 1 ng/ml IL-1β for 15 min. ChIP assays were conducted using Abs against p65, p300, and CBP. (E–H) 832/13 cells were stimulated with 1 ng/ml IL-1β for 15, 30, or 60 min. ChIP assays were performed using Abs against acetylated histone H3 and H4 (E), total RNA polymerase II (F), polymerase II CTD-PO4-Ser5 (G), or polymerase II CTD-PO4-Ser2 (H). In (A–F), the recovered CXCL10 genomic regions targeted for amplification via real-time RT-PCR are shown above each individual graph. Data are mean ± SEM from three independent experiments. *p < 0.05 versus corresponding NT groups (A–D, F); #p < 0.1 versus the other corresponding NT groups, *p < 0.05 versus the intragroup comparison (G), **p < 0.01 versus corresponding NT groups. n.s., Not significant versus NT.

An increase in promoter-bound p50, the typical heterodimer partner for p65, was also detected at κB sites, and IL-1β stimulation produced a 109% increase in ChIP signal (Fig. 4B). Consistent with IFN-γ potentiating the effects of IL-1β, we detected a 6.86-fold increase in STAT1 ChIP signal in response to this cytokine within 15 min, which diminished to 2.49- and 1.94-fold at 30 and 60 min, respectively (Fig. 4C).

We next examined whether p300 or CBP, known coactivators of p65 (43, 44), were recruited to the CXCL10 gene promoter in response to IL-1β. There was a 181% increase in p300 at the κΒ sites within 15 min of IL-1β stimulation (Fig. 4D), consistent with the kinetics of p65 occupancy. CBP was present at high levels in the basal state, which matches the CXCL10 gene having a relatively high baseline expression level. In addition, IL-1β promoted an increase in CBP ChIP signal from 6.04-fold to 14.72-fold (144% increase; Fig. 4D). This accumulation of regulatory proteins at the CXCL10 gene promoter correlated with an approximate doubling in the acetylation of both histones H3 and H4 in response to IL-1β (Fig. 4E). Moreover, there was a 59.24-fold increase in the recruitment of total RNA polymerase II to the gene promoter (Fig. 4F). We also found a 3.1-fold increase in Ser5 phosphorylation of the carboxyl-terminal domain (CTD) of RNA polymerase II at the core promoter at 15 min, with further increases at 60 min (Fig. 4G) and a 3.71-, 4.96-, and 6-fold enhancement of Ser2 phosphorylation of the RNA polymerase II CTD on the coding region after 15, 30, and 60 min of IL-1β exposure, respectively (Fig. 4H). These changes are consistent with a gene undergoing transcriptional activation (45) and, to our knowledge, represent the first report of such alterations for the CXCL10 gene in response to proinflammatory cytokines in β cells.

Thus, we conclude that the synergistic increase in the expression of the CXCL10 gene by IL-1β and IFN-γ is through the docking of the key regulatory proteins NF-κB and STAT1 at the genomic response elements in the proximal promoter region, associated recruitment of the coactivators p300 and CBP, and alterations in the local chromatin environment (i.e., increases in acetylation of histones H3 and H4). These changes at the CXCL10 gene promoter likely assist in recruitment of RNA polymerase II and promote signal-induced modifications to the enzyme that allow transcriptional activation of the CXCL10 gene.

Pharmacological inhibition of JAK1 blocks the IFN-γ–mediated potentiation of IL-1β–stimulated CXCL10 gene expression

Because the maximal expression of the CXCL10 gene requires both IL-1β and IFN-γ, with IFN-γ augmenting the IL-1β response, we investigated whether phosphorylation of STAT1 was required for this process. First, we observed that IFN-γ induced an early (within 15 min) phosphorylation of STAT1 at Tyr701 (Fig. 5A). Interestingly, Ser727 was constitutively phosphorylated in the pancreatic β cell (Fig. 5A), with no attendant increases in response to IFN-γ. Similar results were obtained using INS-1E cells (data not shown) and in isolated rat islets (Fig. 5B). These results are in contrast to other tissues in which Ser727 is an inducible site and is dependent on prior phosphorylation of Tyr701 (46). Next, we found that titrating in increasing amounts of a JAK1 inhibitor led to impairments in phosphorylation of Tyr701 by IFN-γ (Fig. 5C). At the highest dose used (100 nM), we also noticed a decrease in Ser727 phospho-signal (Fig. 5C). The decrease in phospho-STAT1 at Tyr701 correlated with a concentration-dependent reduction in transcriptional activation of a synthetic luciferase reporter gene driven by 3xGAS (Fig. 5D). Furthermore, IFN-γ–mediated potentiation of the CXCL10 gene in the presence of the JAK1 inhibitor was also diminished by 59% at the highest concentration (Fig. 5E). We also observed similar results using the −1 kb proximal CXCL10 gene promoter, with 100 nM JAK1 inhibitor decreasing the IFN-γ potentiation by 42% (Fig. 5F). Thus, the ability of IFN-γ to augment IL-1β–mediated transcription requires JAK1 activity, which most likely signals through the inducible Tyr701 phosphorylation of STAT1.

FIGURE 5.
FIGURE 5.

Pharmacological inhibition of JAK1 blocks the IFN-γ–mediated potentiation of IL-1β–stimulated CXCL10 gene expression. 832/13 cells (A) and rat islets (B) were stimulated with 100 U/ml IFN-γ for the indicated times. Whole-cell lysates were extracted and blotted for phospho-STAT1 (Y701 and S727) and total STAT1 abundance. The abundance of tubulin (A) and β-actin (B) serves as loading control. (C) 832/13 cells were pretreated with increasing concentrations (0.1, 1, 10, or 100 nM) of a JAK inhibitor (JAKi) for 1 h, followed by 15 min of stimulation with 100 U/ml IFN-γ. Whole-cell lysates were extracted and analyzed for phospho-STAT1 (Y701 and S727) and total STAT1 abundance. Immunoblotting with β-actin is shown as a loading control. (D) 832/13 cells were transfected with a 3xGAS-luciferase construct; at 24 h posttransfection, cells were pretreated for 1 h with increasing concentrations of JAKi (1, 10, or 100 nM), followed by a 4-h stimulation with 100 U/ml IFN-γ. *p < 0.05 versus IFN-γ with DMSO. (E) 832/13 cells were pretreated for 1 h with increasing concentrations of JAKi (1, 10, or 100 nM), followed by a 6-h stimulation with IL-1β alone or IL-1β plus IFN-γ. Total RNA was isolated, and CXCL10 mRNA levels were quantified by real-time PCR. **p < 0.01 versus DMSO with IL-1β + IFN-γ. (F) 832/13 cells were transfected with CXCL10 −1 kb promoter-luciferase construct. At 24 h posttransfection, the cells were treated with 1, 10, or 100 nM JAKi for 1 h, followed by a 4-h stimulation with 1 ng/ml IL-1β or IL-1β plus 100 U/ml IFN-γ. In (D) and (F), promoter luciferase activity was measured. Data are mean ± SE from three independent experiments. *p < 0.05 versus DMSO with IL-1β + IFN-γ.

Removal of the Tyr701 phospho-acceptor site blunts the ability of IFN-γ to increase expression of the CXCL10 gene

Inhibition of signaling through JAK1 blocked the ability of IFN-γ to potentiate the IL-1β–stimulated expression of the CXCL10 gene (Fig. 5). Because Tyr701 is a major phospho-acceptor site activated in response to IFN-γ (Fig. 5A) (47), we generated a recombinant adenovirus that encodes STAT1 with a Y701F mutation. As expected, the mutant protein is not phosphorylated upon cellular exposure to IFN-γ (Fig. 6A). However, the constitutive phosphorylation of STAT1 at Ser727 in the pancreatic β cell was still present in the Y701F mutant STAT1 protein (Fig. 6A), indicating the phospho-acceptor sites are not dependent on each other for their individual activation. We note that the STAT1 Y701F has much less activity than does wild-type STAT1 on a synthetic 3xGAS-luciferase reporter gene, despite retaining constitutive Ser727 phosphorylation (Fig. 6B). Additionally, the IFN-γ–mediated potentiation of the CXCL10 gene was reduced by 39 and 51%, in the presence of increasing concentrations of STAT1 Y701F protein (Fig. 6C). Moreover, we discovered that, although wild-type STAT1 is capable of enhancing the IFN-γ–stimulated increase in promoter activity using the −1 kb CXCL10 luciferase reporter, STAT1 Y701F did not retain this ability (Fig. 6D). Further, cytokine-dependent secretion of CXCL10 protein was diminished by 43% in the presence of overexpressed STAT1 Y701F (Fig. 6E). Thus, phosphorylation of STAT1 at Tyr701 is required for maximal activation of the CXCL10 gene and release of protein in response to IL-1β + IFN-γ.

FIGURE 6.
FIGURE 6.

Removal of the Tyr701 phospho-acceptor site blunts the ability of IFN-γ to increase expression of the CXCL10 gene. (A) 832/13 cells were transduced with adenoviruses expressing βGAL, STAT1, or mutant STAT1 (Y701F). At 24 h posttransduction, cells were treated for 15 min with 100 U/ml IFN-γ. Whole-cell lysates were extracted and blotted for phospho-STAT1 (Y701 and S727) and total STAT1 abundance, with β-actin serving as a loading control. (B) 832/13 cells were transfected with a 3xGAS-luciferase construct; at 4 h posttransfection, adenoviruses expressing βGAL, STAT1, or STAT1 Y701F were transduced into cells. At 18 h postadenoviral delivery, cells were stimulated for 4 h with 100 U/ml IFN-γ. **p < 0.01 versus STAT1. (C) 832/13 cells were transduced with the indicated recombinant adenoviruses; at 24 h posttransduction, cells were stimulated with either 1 ng/ml IL-1β alone or IL-1β plus 100 U/ml IFN-γ for 6 h. Total RNA was isolated, and mRNA abundance of CXCL10 was monitored via real-time RT-PCR. **p < 0.01 versus βGAL with IL-1β + IFN-γ. (D) 832/13 cells were transfected with CXCL10 −1 kb plasmid, followed by transduction overnight with the indicated adenoviruses. At 18 h posttransduction, a 4-h stimulation with 1 ng/ml IL-1β alone or IL-1β plus 100 U/ml IFN-γ preceded measurement of promoter luciferase activity. **p < 0.01 versus STAT1 with IL-1β + IFN-γ. (E) 832/13 cells were treated with the labeled adenoviruses; at 24 h posttransduction, these cells were stimulated for 12 h with either 1 ng/ml IL-1β alone or IL-1β plus 100 U/ml IFN-γ. CXCL10 release into the media was quantified via ELISA and normalized to cellular protein content. Data are mean ± SE from three or four independent experiments. *p < 0.05 versus βGAL with IL-1β + IFN-γ.

Conservative or loss-of-function mutations at the Ser727 phospho-acceptor site do not block IFN-γ–mediated increases in expression of the CXCL10 gene

The serine phosphorylation of STAT1 is required for the expression of many genes in a variety of tissues (48). Because the Ser727 site in STAT1 is constitutively phosphorylated in the pancreatic β cell and does not require prior phosphorylation at Tyr701, we generated recombinant adenoviruses that express either STAT1 with S727A (loss-of-function) or S727T (conservative) mutations. In Fig. 7A, it is clear that phosphorylation of STAT1 at Ser727 was diminished, albeit with different magnitudes, in these mutant proteins. The removal of the phospho-acceptor site (S727A) reduced the ability of STAT1 to support transcription from a 3xGAS-luciferase reporter gene (Fig. 7B). In contrast, the conservative change of serine to threonine, which retains the hydroxyl group that serves as a phospho-acceptor site, displayed activity similar to the wild-type STAT1 on a 3xGAS reporter gene (Fig. 7C). Despite the STAT1 S727A mutant having decreased activity on a synthetic STAT1-regulated reporter gene, this mutant STAT1 did not display any loss in activity when we examined cytokine-mediated CXCL10 gene expression (Fig. 7D). Moreover, the S727T mutant was also perfectly capable of supporting a response to the CXCL10 gene in response to IL-1β and IFN-γ (Fig. 7E). Thus, despite phosphorylation of Ser727 being required for expression of cytokine-responsive genes in other tissues (49, 50) and for activation of a consensus multimerized γ-activated sequence element in the pancreatic β cell (Fig. 7B), the CXCL10 gene does not require phosphorylation at Ser727 within STAT1 to support its expression.

FIGURE 7.
FIGURE 7.

Amino acid substitutions at the Ser727 phospho-acceptor site do not block IFN-γ–mediated increases in expression of the CXCL10 gene. (A) 832/13 cells were transduced with adenoviruses encoding βGAL, STAT1, or two increasing concentrations of STAT1 with either S727A or S727T mutations. Whole-cell lysates were immunoblotted for phospho-STAT1 (Y701 and S727) and total STAT1 abundance, with β-actin serving as a loading control (left panel). Densitometric analysis of the immunoblot displays the relative ratio of phosphorylated STAT1 (S727) to total STAT1 (right panel). Note the decrease in phospho-STAT1 in the mutants. 832/13 cells were transfected with a 3xGAS-luciferase construct; at 4 h posttransfection, cells were transduced with adenoviruses expressing βGAL, STAT1, or two increasing concentrations of either STAT1 S727A (B) or STAT1 S727T (C). Following overnight culture with the indicted adenoviruses, the cells were incubated for 4 h with 100 U/ml IFN-γ. Promoter activity was quantified and normalized to protein content via BCA assay. (D and E) 832/13 cells were transduced with the indicated recombinant adenoviruses; at 24 h posttransduction, cells were treated for 6 h with 1 ng/ml IL-1β alone or IL-1β plus 100 U/ml IFN-γ. Total RNA was extracted, and mRNA abundance of CXCL10 was quantified via real-time RT-PCR. The concentrations of adenoviruses used in (B)–(E) are identical to those shown in (A). Data shown are mean ± SEM from three independent experiments. **p < 0.01 versus STAT1.

To test the hypothesis that Ser727 synergizes with Tyr701 upon the latter’s inducible phosphorylation to activate transcription, we next examined how a double-mutant (DM) Y701F/S727A mutant impacted expression of the CXCL10 gene. The abundance of wild-type and DM STAT1 are shown in Fig. 8A, with the clear inability to augment phosphorylation at the mutated sites in response to cytokine stimulus. The decrease in phosphorylation at both of the two key regulatory sites in STAT1 resulted in impairments in the ability to activate the 3xGAS-luciferase reporter gene (Fig. 8B). Moreover, expression of the CXCL10 gene was dose dependently diminished by 53.6, 64.6, and 74.7% with increasing concentrations of STAT1 DM protein (Fig. 8C). In the isolated rat islet, the STAT1 DM displayed comparable activity, with a 54% decrease in expression of the CXCL10 gene in response to IL-1β + IFN-γ (Fig. 8D). In addition, promoter activity was reduced by 60.8% and secretion of CXCL10 protein was reduced by 30.1% with overexpression of the DM STAT1 protein (Fig. 8E, 8F). These data are incredibly similar to those seen with the single Y701F mutation and, therefore, support the lack of necessity for Ser727 phosphorylation to drive expression of the CXCL10 gene in pancreatic β cells. Thus, we conclude that maximal induction of the CXCL10 gene by IL-1β and IFN-γ in pancreatic β cells requires a functional phospho-acceptor site at Tyr701 within the STAT1 protein to communicate signals induced by IFN-γ.

FIGURE 8.
FIGURE 8.

A DM STAT1 at Y701F/S727A reveals that the Tyr701 site predominantly controls the response to IFN-γ. (A) 832/13 cells were exposed for 24 h to adenoviruses expressing βGAL, STAT1, or STAT1 DMs (mutations Y701F and S727A are both present). Following 15 min of stimulation with 100 U/ml IFN-γ, whole-cell lysates were extracted and blotted for phospho-STAT1 (Y701 and S727) and total STAT1 abundance, with β-actin as a loading control. (B) 832/13 cells were transfected with a 3xGAS-luciferase construct; at 4 h posttransfection, cells were exposed to adenoviruses expressing βGAL, STAT1, or STAT1 Y701F/S727A. After overnight transduction with the indicated adenoviruses, cells were stimulated for 4 h with 100 U/ml IFN-γ. *p < 0.05 versus STAT1. 832/13 cells (C) and rat islets (D) were treated with the indicated adenoviruses; at 24 h posttransduction, cells were stimulated for 6 h with either 1 ng/ml IL-1β alone or IL-1β plus 100 U/ml IFN-γ (C) or 10 ng/ml IL-1β plus 100 U/ml IFN-γ (D). Total RNA was isolated and CXCL10 transcript levels were analyzed. **p < 0.01 versus βGAL with IL-1β + IFN-γ. (E) 832/13 cells were transfected with CXCL10 −1 kb plasmid, followed by an 18-h exposure to the indicated adenoviruses and a subsequent 4-h incubation with 1 ng/ml IL-1β alone or IL-1β plus 100 U/ml IFN-γ. **p < 0.01 versus STAT1 with IL-1β + IFN-γ. In (B) and (E), promoter activity was quantified and normalized to protein abundance via BCA assay. (F) 832/13 cells were treated with the indicated adenoviruses; at 24 h posttransduction, cells were stimulated for 12 h with either 1 ng/ml IL-1β alone or IL-1β plus 100 U/ml IFN-γ. CXCL10 release into the media was quantified via ELISA and normalized to total cellular protein content. Data are mean ± SEM from three or four independent experiments. *p < 0.05 versus βGAL with IL-1β + IFN-γ. (G) IL-1β + IFN-γ synergistically increase transcription of the CXCL10 gene, thereby augmenting release of CXCL10 protein. A major part of this response includes constitutive phosphorylation of STAT1 at Ser727 and inducible phosphorylation (by IFN-γ) at Tyr701. Secreted CXCL10 promotes chemotaxis of immune cells via activation of the CXCR3 receptor and/or activates β cell TLR4 receptors in an autocrine/paracrine fashion. Dashed arrow indicates putative pathway activation.

Discussion

CXCL10 peptide levels are elevated in the blood of individuals with type 1 diabetes mellitus (8, 11, 12, 14) and are associated with rejection of transplanted islets (51, 52). Synthesis and secretion of CXCL10 from the pancreatic β cell occur in both type 1 and type 2 diabetes mellitus (13), and forced expression from pancreatic β cells of transgenic mice accelerates immune-mediated diabetes (10). Thus, a thorough understanding of the signals and factors required to enhance expression of the CXCL10 gene could be beneficial for developing strategies to diminish recruitment of proinflammatory stimuli–induced leukocyte populations. In the current study, we demonstrated a strong synergy between IL-1β and IFN-γ to induce the expression of the CXCL10 gene. To our knowledge, these data are the first to show the molecular regulation of the CXCL10 gene in response to proinflammatory cytokines in pancreatic β cells.

We specifically note the involvement of what appears to be a cooperative set of κB and ISREs. Normally, κB sites respond via NF-κB–activating stimuli (e.g., IL-1β, TNF-α), whereas γ-activated sequence/ISRE elements respond to type I and/or type II IFNs. In the case of IL-1β and IFN-γ signaling to the CXCL10 gene in pancreatic β cells, the κB and ISRE sites appear to be operating concurrently to fine tune the response of each signal in a cooperative manner (Figs. 1, 2). This type of signal integration at the genomic level offers the β cell the ability to rapidly increase or decrease the transcriptional responses at key promoters, such as CXCL10 and, thus, potentially control inflammation within cells of the pancreatic islets. Our observations regarding major response elements controlling expression of the CXCL10 gene are in agreement with those obtained using macrophage (53) and fibroblast (54) cell lines in response to IFN-γ paired with either LPS or TNF-α, respectively. This potentially indicates a general responsiveness of the CXCL10 gene promoter to proinflammatory stimuli, albeit with diverse regulatory mechanisms that may depend on the specific cell type. Notably, however, IFN-γ alone fails to activate transcription of the CXCL10 gene in two different rat pancreatic β cell lines, yet it potently potentiates the response provided by IL-1β (Figs. 1, 2). Thus, the IFN-γ–signaling pathway is intact, although the CXCL10 gene is not firing in response to this cytokine alone (Fig. 1), despite robust STAT1 phosphorylation (Fig. 5A, 5B).

Many other chemokines, in addition to CXCL10, are expressed in islet β cells, and their expression can be increased considerably by exposure to IL-1β or combinations of inflammatory cytokines (55). However, very few studies have delineated the primary factors and associated molecular mechanisms required for the transcriptional output at these individual gene promoters; the current study provides such information regarding the CXCL10 gene. Additional examples of genes other than chemokines whose expression is also strongly controlled by proinflammatory cytokines in pancreatic β cells include those encoding inducible NO synthase (22, 56) and PG-endoperoxide synthase 2 (aka cyclooxygenase-2) (20, 57). Delineating precise complexes required for transcription of genes controlling inflammatory processes, such as those that encode chemokines and other proteins involved in inflammatory responses, will be required to understand how inflammation is managed at the molecular level.

Along these lines, we demonstrated previously that the CCL2 gene is highly responsive to IL-1β but is not potentiated by IFN-γ in pancreatic β cells (19), despite IFN-γ responsiveness in other tissues (58). The distinction in signal integration at given gene promoters, such as seen with CCL2 versus CXCL10, may be one way to help orchestrate a cautious recruitment of quantities and types of immune cells by carefully titrating amounts of chemotactic proteins made and secreted. The synergistic nature of IFN-γ amplifying the response to IL-1β to modulate CXCL10, but not CCL2, expression fits with this interpretation. In addition, TNF-α and IFN-γ synergize to induce the expression of CXCL10 gene in human airway smooth muscle cells (59), indicating a global importance for understanding the regulation of this gene and its role in a broad spectrum of inflammatory processes. However, no increase in acetylation of histone H3 was observed in airway smooth muscle cells in response to TNF-α or IFN-γ, indicating tissue-specific and/or signal-specific regulatory strategies to control gene transcription.

As part of our molecular analysis, we discovered a tissue-specific phosphorylation pattern of STAT1 in the β cell that is distinct from other cell types. The pancreatic β cell constitutively phosphorylates STAT1 at Ser727, regardless of whether the Tyr701 site is intact. This is in contrast to other tissues, which were shown to require Tyr701 phosphorylation as a prerequisite for Ser727 phosphorylation (46, 49). Thus, the phosphorylation patterns of STAT1 could partially help to explain the potentiation of the IL-1β response by IFN-γ, with activation of the inducible Tyr701 site assisting in promotion of a multiregulatory complex that enhances the transactivation capacity of primary DNA-binding factors and/or recruitment of RNA polymerase II. Interestingly, in the context of the CXCL10 gene promoter, whether Ser727 is phosphorylated or not does not appear to impact transcription of this gene in pancreatic β cells (Fig. 7D).

Moreover, the S727T STAT1 also displayed no loss in function at either a 3xGAS reporter gene (Fig. 7C) or in the context of the CXCL10 gene promoter (Fig. 7E), which could indicate that the kinase responsible for constitutive activation of the Ser727 site within STAT1 is a Ser/Thr kinase. We further note that we could not detect a direct association of p65 with STAT1 via coimmunoprecipitation (S.J. Burke, M.R. Goff, & J.J. Collier, unpublished observations) as an explanation for synergy between IL-1β and IFN-γ, even though both proteins are clearly required for maximal expression of the CXCL10 gene (Figs. 1–3), and each individual protein is detected at the relevant cognate DNA binding sites using ChIP assays (Fig. 4). Taken together, the binding data and differential STAT1 phosphorylation patterns may offer insights into the control of the CXCL10 gene by inflammatory signals. Phosphorylation at Ser727 within STAT1 is dispensable for full expression of the CXCL10 gene; however, potentiation of the IL-1β response by IFN-γ is contingent on both JAK1 activity (Fig. 5) and tyrosine phosphorylation at the 701 position in STAT1 (Fig. 6).

Soluble secreted CXCL10 is a strong chemoattractant for T lymphocytes (60, 61); autoimmune destruction of islet β cells leading to type 1 diabetes is due, in large part, to a massive infiltration of T lymphocytes into the pancreatic islets (62, 63). Thus, enhanced secretion of CXCL10 directly from β cells exposed to proinflammatory signals, such as IL-1β and IFN-γ, may be a key contributing factor to diabetes development (64). Moreover, the increase in secreted CXCL10 protein driven by the combination of IL-1β and IFN-γ may also directly participate in β cell death and dysfunction by signaling through TLR4 on the β cell surface (Fig. 8F) (13). This fits with our own results that β cells do not express appreciable amounts of the CXCR3 protein (data not shown), which normally recognizes CXCL10 and other chemokines (e.g., CXCL9 and CXCL11) as ligands (65).

It is plausible that resident macrophages, which provide enough IL-1β to alter gene-expression patterns in the pancreatic β cells (22), in concert with infiltrating T cells combine to target insulin-positive cells for autoimmune-mediated destruction. This is fitting with extant studies demonstrating that T cell clones isolated from NOD mice immunized with insulin B-chain peptide fragments accelerated diabetes in neonatal mice, whereas the GAD65 (another autoantigen) T cell clone did not display this phenotype (66). A key distinction between these two T cell populations is that the clone immunized against the insulin B-chain expressed high levels of CXCL10 (66), whereas the other clones tested did not. Thus, synthesis and secretion of CXCL10 from specific T cell populations, in addition to that secreted from pancreatic β cells, would keep infiltrating T cells from leaving the islets and likely exacerbate accumulation of these and other immune cells. Moreover, CCL2 release from pancreatic β cells in response to IL-1β would promote recruitment of additional monocytes and macrophages into the islets (9, 19), in addition to contributing to T lymphocyte movement toward islet β cells (67). Taken together, these observations may help to explain the transition from peri-insulitis to invasive insulitis observed in NOD mice (68).

Therefore, attempts to design novel therapeutics for treatment of diabetes mellitus will benefit from a thorough understanding of how the β cell integrates signals, such as those elicited by cell surface receptors recognizing cytokine and chemokine ligands, to coordinately control transcriptional pathways regulating key soluble secreted factors linked to immune cell infiltration. Our present study has provided new knowledge regarding the phosphorylation state of STAT1 within pancreatic β cells, in addition to providing critical insights into the regulation of the CXCL10 gene by NF-κB and STAT1 in response to IL-1β and IFN-γ. It is possible that if a small molecule were designed to be able to selectively reduce CXCL10 signaling, this strategy would likely dampen the ability of the β cell to participate in its own demise through chemotactic and immune cell–modulatory activities. Interference with CXCL10 synthesis, secretion, or receptor interaction may also effectively reduce signaling through innate immune receptors, such as TLR4 (13). This approach may also be more selective than glucocorticoid-based therapies, which broadly target inflammation and have a host of undesirable side effects. A further understanding of proinflammatory cytokine–mediated control of chemokine gene networks will be required to facilitate improvement of therapeutic approaches to treat diseases associated with pathological inflammation.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Drs. Guoxun Chen, Naima Moustaid-Moussa, Christopher Newgard, Thomas Spencer, and Jay Whelan for reagents. We also thank Emily Lazek for technical assistance.

Footnotes

  • Abbreviations used in this article:

    ChIP
    chromatin immunoprecipitation
    CTD
    carboxyl-terminal domain
    DM
    double mutant
    βGAL
    β-galactosidase
    IκBαSR
    IκBα superrepressor
    ISRE
    IFN-stimulated response element
    siRNA
    small interfering RNA
    3xGAS
    three copies of the consensus γ-activated sequence.

  • Received February 4, 2013.
  • Accepted April 25, 2013.

References

  1. ↵
    1. Bluestone J. A.,
    2. K. Herold,
    3. G. Eisenbarth
    . 2010. Genetics, pathogenesis and clinical interventions in type 1 diabetes. Nature 464: 1293–1300.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Eisenbarth G. S.
    1986. Type I diabetes mellitus. A chronic autoimmune disease. N. Engl. J. Med. 314: 1360–1368.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Doria A.,
    2. M. E. Patti,
    3. C. R. Kahn
    . 2008. The emerging genetic architecture of type 2 diabetes. Cell Metab. 8: 186–200.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Unger R. H.
    2002. Lipotoxic diseases. Annu. Rev. Med. 53: 319–336.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Baggiolini M.
    1998. Chemokines and leukocyte traffic. Nature 392: 565–568.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Rotondi M.,
    2. L. Chiovato,
    3. S. Romagnani,
    4. M. Serio,
    5. P. Romagnani
    . 2007. Role of chemokines in endocrine autoimmune diseases. Endocr. Rev. 28: 492–520.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Charo I. F.,
    2. R. M. Ransohoff
    . 2006. The many roles of chemokines and chemokine receptors in inflammation. N. Engl. J. Med. 354: 610–621.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Shigihara T.,
    2. Y. Oikawa,
    3. Y. Kanazawa,
    4. Y. Okubo,
    5. S. Narumi,
    6. T. Saruta,
    7. A. Shimada
    . 2006. Significance of serum CXCL10/IP-10 level in type 1 diabetes. J. Autoimmun. 26: 66–71.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Sarkar S. A.,
    2. C. E. Lee,
    3. F. Victorino,
    4. T. T. Nguyen,
    5. J. A. Walters,
    6. A. Burrack,
    7. J. Eberlein,
    8. S. K. Hildemann,
    9. D. Homann
    . 2012. Expression and regulation of chemokines in murine and human type 1 diabetes. Diabetes 61: 436–446.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Rhode A.,
    2. M. E. Pauza,
    3. A. M. Barral,
    4. E. Rodrigo,
    5. M. B. Oldstone,
    6. M. G. von Herrath,
    7. U. Christen
    . 2005. Islet-specific expression of CXCL10 causes spontaneous islet infiltration and accelerates diabetes development. J. Immunol. 175: 3516–3524.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Nicoletti F.,
    2. I. Conget,
    3. M. Di Mauro,
    4. R. Di Marco,
    5. M. C. Mazzarino,
    6. K. Bendtzen,
    7. A. Messina,
    8. R. Gomis
    . 2002. Serum concentrations of the interferon-gamma-inducible chemokine IP-10/CXCL10 are augmented in both newly diagnosed Type I diabetes mellitus patients and subjects at risk of developing the disease. Diabetologia 45: 1107–1110.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Uno S.,
    2. A. Imagawa,
    3. K. Saisho,
    4. K. Okita,
    5. H. Iwahashi,
    6. T. Hanafusa,
    7. I. Shimomura
    . 2010. Expression of chemokines, CXC chemokine ligand 10 (CXCL10) and CXCR3 in the inflamed islets of patients with recent-onset autoimmune type 1 diabetes. Endocr. J. 57: 991–996.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Schulthess F. T.,
    2. F. Paroni,
    3. N. S. Sauter,
    4. L. Shu,
    5. P. Ribaux,
    6. L. Haataja,
    7. R. M. Strieter,
    8. J. Oberholzer,
    9. C. C. King,
    10. K. Maedler
    . 2009. CXCL10 impairs beta cell function and viability in diabetes through TLR4 signaling. Cell Metab. 9: 125–139.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Roep B. O.,
    2. F. S. Kleijwegt,
    3. A. G. van Halteren,
    4. V. Bonato,
    5. U. Boggi,
    6. F. Vendrame,
    7. P. Marchetti,
    8. F. Dotta
    . 2010. Islet inflammation and CXCL10 in recent-onset type 1 diabetes. Clin. Exp. Immunol. 159: 338–343.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Berg A. K.,
    2. T. Tuvemo,
    3. G. Frisk
    . 2010. Enterovirus markers and serum CXCL10 in children with type 1 diabetes. J. Med. Virol. 82: 1594–1599.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Kriegel M. A.,
    2. C. Rathinam,
    3. R. A. Flavell
    . 2012. Pancreatic islet expression of chemokine CCL2 suppresses autoimmune diabetes via tolerogenic CD11c+ CD11b+ dendritic cells. Proc. Natl. Acad. Sci. USA 109: 3457–3462.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Corbett J. A.,
    2. M. A. Sweetland,
    3. J. L. Wang,
    4. J. R. Lancaster Jr..,
    5. M. L. McDaniel
    . 1993. Nitric oxide mediates cytokine-induced inhibition of insulin secretion by human islets of Langerhans. Proc. Natl. Acad. Sci. USA 90: 1731–1735.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Corbett J. A.,
    2. J. L. Wang,
    3. M. A. Sweetland,
    4. J. R. Lancaster Jr..,
    5. M. L. McDaniel
    . 1992. Interleukin 1 beta induces the formation of nitric oxide by beta-cells purified from rodent islets of Langerhans. Evidence for the beta-cell as a source and site of action of nitric oxide. J. Clin. Invest. 90: 2384–2391.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Burke S. J.,
    2. M. R. Goff,
    3. B. L. Updegraff,
    4. D. Lu,
    5. P. L. Brown,
    6. S. C. Minkin Jr..,
    7. J. P. Biggerstaff,
    8. L. Zhao,
    9. M. D. Karlstad,
    10. J. J. Collier
    . 2012. Regulation of the CCL2 gene in pancreatic β-cells by IL-1β and glucocorticoids: role of MKP-1. PLoS ONE 7: e46986.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Burke S. J.,
    2. J. J. Collier
    . 2011. The gene encoding cyclooxygenase-2 is regulated by IL-1β and prostaglandins in 832/13 rat insulinoma cells. Cell. Immunol. 271: 379–384.
    OpenUrlCrossRefPubMed
    1. Heitmeier M. R.,
    2. A. L. Scarim,
    3. J. A. Corbett
    . 1999. Prolonged STAT1 activation is associated with interferon-gamma priming for interleukin-1-induced inducible nitric-oxide synthase expression by islets of Langerhans. J. Biol. Chem. 274: 29266–29273.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Heitmeier M. R.,
    2. A. L. Scarim,
    3. J. A. Corbett
    . 1997. Interferon-gamma increases the sensitivity of islets of Langerhans for inducible nitric-oxide synthase expression induced by interleukin 1. J. Biol. Chem. 272: 13697–13704.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Steer S. A.,
    2. A. L. Scarim,
    3. K. T. Chambers,
    4. J. A. Corbett
    . 2006. Interleukin-1 stimulates beta-cell necrosis and release of the immunological adjuvant HMGB1. PLoS Med. 3: e17.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Collier J. J.,
    2. P. T. Fueger,
    3. H. E. Hohmeier,
    4. C. B. Newgard
    . 2006. Pro- and antiapoptotic proteins regulate apoptosis but do not protect against cytokine-mediated cytotoxicity in rat islets and beta-cell lines. Diabetes 55: 1398–1406.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Gepts W.
    1965. Pathologic anatomy of the pancreas in juvenile diabetes mellitus. Diabetes 14: 619–633.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Hayden M. S.,
    2. S. Ghosh
    . 2012. NF-κB, the first quarter-century: remarkable progress and outstanding questions. Genes Dev. 26: 203–234.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Christmann B. S.,
    2. J. M. Moran,
    3. J. A. McGraw,
    4. R. M. Buller,
    5. J. A. Corbett
    . 2011. Ccr5 regulates inflammatory gene expression in response to encephalomyocarditis virus infection. Am. J. Pathol. 179: 2941–2951.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Levy D. E.,
    2. J. E. Darnell Jr..
    2002. Stats: transcriptional control and biological impact. Nat. Rev. Mol. Cell Biol. 3: 651–662.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Asfari M.,
    2. D. Janjic,
    3. P. Meda,
    4. G. Li,
    5. P. A. Halban,
    6. C. B. Wollheim
    . 1992. Establishment of 2-mercaptoethanol-dependent differentiated insulin-secreting cell lines. Endocrinology 130: 167–178.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Hohmeier H. E.,
    2. H. Mulder,
    3. G. Chen,
    4. R. Henkel-Rieger,
    5. M. Prentki,
    6. C. B. Newgard
    . 2000. Isolation of INS-1-derived cell lines with robust ATP-sensitive K+ channel-dependent and -independent glucose-stimulated insulin secretion. Diabetes 49: 424–430.
    OpenUrlAbstract
  31. ↵
    1. Janjic D.,
    2. P. Maechler,
    3. N. Sekine,
    4. C. Bartley,
    5. A. S. Annen,
    6. C. B. Wolheim
    . 1999. Free radical modulation of insulin release in INS-1 cells exposed to alloxan. Biochem. Pharmacol. 57: 639–648.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Stewart M. D.,
    2. G. A. Johnson,
    3. C. A. Vyhlidal,
    4. R. C. Burghardt,
    5. S. H. Safe,
    6. L. Y. Yu-Lee,
    7. F. W. Bazer,
    8. T. E. Spencer
    . 2001. Interferon-tau activates multiple signal transducer and activator of transcription proteins and has complex effects on interferon-responsive gene transcription in ovine endometrial epithelial cells. [Published erratum appears in 2001 Endocrinology 151: 2400.] Endocrinology 142: 98–107.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Herz J.,
    2. R. D. Gerard
    . 1993. Adenovirus-mediated transfer of low density lipoprotein receptor gene acutely accelerates cholesterol clearance in normal mice. Proc. Natl. Acad. Sci. USA 90: 2812–2816.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Jobin C.,
    2. A. Panja,
    3. C. Hellerbrand,
    4. Y. Iimuro,
    5. J. Didonato,
    6. D. A. Brenner,
    7. R. B. Sartor
    . 1998. Inhibition of proinflammatory molecule production by adenovirus-mediated expression of a nuclear factor kappaB super-repressor in human intestinal epithelial cells. J. Immunol. 160: 410–418.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Prasad R. C.,
    2. X. L. Wang,
    3. B. K. Law,
    4. B. Davis,
    5. G. Green,
    6. B. Boone,
    7. L. Sims,
    8. M. Law
    . 2009. Identification of genes, including the gene encoding p27Kip1, regulated by serine 276 phosphorylation of the p65 subunit of NF-kappaB. Cancer Lett. 275: 139–149.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Spurrell J. C.,
    2. S. Wiehler,
    3. R. S. Zaheer,
    4. S. P. Sanders,
    5. D. Proud
    . 2005. Human airway epithelial cells produce IP-10 (CXCL10) in vitro and in vivo upon rhinovirus infection. Am. J. Physiol. Lung Cell. Mol. Physiol. 289: L85–L95.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Koetzler, R., R. S. Zaheer, S. Wiehler, N. S. Holden, M. A. Giembycz, and D. Proud. 2009. Nitric oxide inhibits human rhinovirus-induced transcriptional activation of CXCL10 in airway epithelial cells. J. Allergy Clin. Immunol. 123: 201–208.e9.
  38. ↵
    Becker, T. C., R. J. Noel, W. S. Coats, A. M. Gomez-Foix, T. Alam, R. D. Gerard, and C. B. Newgard. 1994. Use of recombinant adenovirus for metabolic engineering of mammalian cells. Methods Cell. Biol. 43 Pt A: 161–189.
  39. ↵
    1. Lavine J. A.,
    2. P. W. Raess,
    3. D. B. Davis,
    4. M. E. Rabaglia,
    5. B. K. Presley,
    6. M. P. Keller,
    7. M. C. Beinfeld,
    8. A. S. Kopin,
    9. C. B. Newgard,
    10. A. D. Attie
    . 2010. Contamination with E1A-positive wild-type adenovirus accounts for species-specific stimulation of islet cell proliferation by CCK: a cautionary note. Mol. Endocrinol. 24: 464–467.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Haycock J. W.
    1993. Polyvinylpyrrolidone as a blocking agent in immunochemical studies. Anal. Biochem. 208: 397–399.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Herder C.,
    2. J. Baumert,
    3. B. Thorand,
    4. W. Koenig,
    5. W. de Jager,
    6. C. Meisinger,
    7. T. Illig,
    8. S. Martin,
    9. H. Kolb
    . 2006. Chemokines as risk factors for type 2 diabetes: results from the MONICA/KORA Augsburg study, 1984-2002. Diabetologia 49: 921–929.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Shimada A.,
    2. J. Morimoto,
    3. K. Kodama,
    4. R. Suzuki,
    5. Y. Oikawa,
    6. O. Funae,
    7. A. Kasuga,
    8. T. Saruta,
    9. S. Narumi
    . 2001. Elevated serum IP-10 levels observed in type 1 diabetes. Diabetes Care 24: 510–515.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Zhong H.,
    2. M. J. May,
    3. E. Jimi,
    4. S. Ghosh
    . 2002. The phosphorylation status of nuclear NF-kappa B determines its association with CBP/p300 or HDAC-1. Mol. Cell 9: 625–636.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Zhong H.,
    2. R. E. Voll,
    3. S. Ghosh
    . 1998. Phosphorylation of NF-kappa B p65 by PKA stimulates transcriptional activity by promoting a novel bivalent interaction with the coactivator CBP/p300. Mol. Cell 1: 661–671.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Ahn S. H.,
    2. M. Kim,
    3. S. Buratowski
    . 2004. Phosphorylation of serine 2 within the RNA polymerase II C-terminal domain couples transcription and 3′ end processing. Mol. Cell 13: 67–76.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Sadzak I.,
    2. M. Schiff,
    3. I. Gattermeier,
    4. R. Glinitzer,
    5. I. Sauer,
    6. A. Saalmüller,
    7. E. Yang,
    8. B. Schaljo,
    9. P. Kovarik
    . 2008. Recruitment of Stat1 to chromatin is required for interferon-induced serine phosphorylation of Stat1 transactivation domain. Proc. Natl. Acad. Sci. USA 105: 8944–8949.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    1. Quelle F. W.,
    2. W. Thierfelder,
    3. B. A. Witthuhn,
    4. B. Tang,
    5. S. Cohen,
    6. J. N. Ihle
    . 1995. Phosphorylation and activation of the DNA binding activity of purified Stat1 by the Janus protein-tyrosine kinases and the epidermal growth factor receptor. J. Biol. Chem. 270: 20775–20780.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Decker T.,
    2. P. Kovarik
    . 2000. Serine phosphorylation of STATs. Oncogene 19: 2628–2637.
    OpenUrlCrossRefPubMed
  49. ↵
    1. Kovarik P.,
    2. D. Stoiber,
    3. P. A. Eyers,
    4. R. Menghini,
    5. A. Neininger,
    6. M. Gaestel,
    7. P. Cohen,
    8. T. Decker
    . 1999. Stress-induced phosphorylation of STAT1 at Ser727 requires p38 mitogen-activated protein kinase whereas IFN-gamma uses a different signaling pathway. Proc. Natl. Acad. Sci. USA 96: 13956–13961.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Goh K. C.,
    2. S. J. Haque,
    3. B. R. Williams
    . 1999. p38 MAP kinase is required for STAT1 serine phosphorylation and transcriptional activation induced by interferons. EMBO J. 18: 5601–5608.
    OpenUrlAbstract
  51. ↵
    1. Abdi R.,
    2. T. K. Means,
    3. A. D. Luster
    . 2003. Chemokines in islet allograft rejection. Diabetes Metab. Res. Rev. 19: 186–190.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Romagnani P.,
    2. C. Crescioli
    . 2012. CXCL10: a candidate biomarker in transplantation. Clin. Chim. Acta 413: 1364–1373.
    OpenUrlCrossRefPubMed
  53. ↵
    1. Ohmori Y.,
    2. T. A. Hamilton
    . 1993. Cooperative interaction between interferon (IFN) stimulus response element and kappa B sequence motifs controls IFN gamma- and lipopolysaccharide-stimulated transcription from the murine IP-10 promoter. J. Biol. Chem. 268: 6677–6688.
    OpenUrlAbstract/FREE Full Text
  54. ↵
    1. Ohmori Y.,
    2. T. A. Hamilton
    . 1995. The interferon-stimulated response element and a kappa B site mediate synergistic induction of murine IP-10 gene transcription by IFN-gamma and TNF-alpha. J. Immunol. 154: 5235–5244.
    OpenUrlAbstract
  55. ↵
    Sarkar, S. A., C. E. Lee, F. Victorino, T. T. Nguyen, J. A. Walters, A. Burrack, J. Eberlein, S. K. Hildemann, and D. Homann. 2012. Expression and regulation of chemokines in murine and human type 1 diabetes. Diabetes 61: 436–446.
  56. ↵
    1. Darville M. I.,
    2. D. L. Eizirik
    . 1998. Regulation by cytokines of the inducible nitric oxide synthase promoter in insulin-producing cells. Diabetologia 41: 1101–1108.
    OpenUrlCrossRefPubMed
  57. ↵
    1. Corbett J. A.,
    2. G. Kwon,
    3. J. Turk,
    4. M. L. McDaniel
    . 1993. IL-1 beta induces the coexpression of both nitric oxide synthase and cyclooxygenase by islets of Langerhans: activation of cyclooxygenase by nitric oxide. Biochemistry 32: 13767–13770.
    OpenUrlCrossRefPubMed
  58. ↵
    1. Zhou Z. H.,
    2. P. Chaturvedi,
    3. Y. L. Han,
    4. S. Aras,
    5. Y. S. Li,
    6. P. E. Kolattukudy,
    7. D. Ping,
    8. J. M. Boss,
    9. R. M. Ransohoff
    . 1998. IFN-gamma induction of the human monocyte chemoattractant protein (hMCP)-1 gene in astrocytoma cells: functional interaction between an IFN-gamma-activated site and a GC-rich element. J. Immunol. 160: 3908–3916.
    OpenUrlAbstract/FREE Full Text
  59. ↵
    1. Clarke D. L.,
    2. R. L. Clifford,
    3. S. Jindarat,
    4. D. Proud,
    5. L. Pang,
    6. M. Belvisi,
    7. A. J. Knox
    . 2010. TNFα and IFNγ synergistically enhance transcriptional activation of CXCL10 in human airway smooth muscle cells via STAT-1, NF-κB, and the transcriptional coactivator CREB-binding protein. J. Biol. Chem. 285: 29101–29110.
    OpenUrlAbstract/FREE Full Text
  60. ↵
    1. Groom J. R.,
    2. A. D. Luster
    . 2011. CXCR3 ligands: redundant, collaborative and antagonistic functions. Immunol. Cell Biol. 89: 207–215.
    OpenUrlCrossRefPubMed
  61. ↵
    1. Bromley S. K.,
    2. T. R. Mempel,
    3. A. D. Luster
    . 2008. Orchestrating the orchestrators: chemokines in control of T cell traffic. Nat. Immunol. 9: 970–980.
    OpenUrlCrossRefPubMed
  62. ↵
    1. Bulek A. M.,
    2. D. K. Cole,
    3. A. Skowera,
    4. G. Dolton,
    5. S. Gras,
    6. F. Madura,
    7. A. Fuller,
    8. J. J. Miles,
    9. E. Gostick,
    10. D. A. Price,
    11. et al
    . 2012. Structural basis for the killing of human beta cells by CD8(+) T cells in type 1 diabetes. Nat. Immunol. 13: 283–289.
    OpenUrlCrossRefPubMed
  63. ↵
    1. Roep B. O.,
    2. M. Peakman
    . 2011. Diabetogenic T lymphocytes in human Type 1 diabetes. Curr. Opin. Immunol. 23: 746–753.
    OpenUrlCrossRefPubMed
  64. ↵
    1. Frigerio S.,
    2. T. Junt,
    3. B. Lu,
    4. C. Gerard,
    5. U. Zumsteg,
    6. G. A. Holländer,
    7. L. Piali
    . 2002. Beta cells are responsible for CXCR3-mediated T-cell infiltration in insulitis. Nat. Med. 8: 1414–1420.
    OpenUrlCrossRefPubMed
  65. ↵
    1. Clark-Lewis I.,
    2. I. Mattioli,
    3. J. H. Gong,
    4. P. Loetscher
    . 2003. Structure-function relationship between the human chemokine receptor CXCR3 and its ligands. J. Biol. Chem. 278: 289–295.
    OpenUrlAbstract/FREE Full Text
  66. ↵
    1. Ejrnaes M.,
    2. N. Videbaek,
    3. U. Christen,
    4. A. Cooke,
    5. B. K. Michelsen,
    6. M. von Herrath
    . 2005. Different diabetogenic potential of autoaggressive CD8+ clones associated with IFN-gamma-inducible protein 10 (CXC chemokine ligand 10) production but not cytokine expression, cytolytic activity, or homing characteristics. J. Immunol. 174: 2746–2755.
    OpenUrlAbstract/FREE Full Text
  67. ↵
    1. Carr M. W.,
    2. S. J. Roth,
    3. E. Luther,
    4. S. S. Rose,
    5. T. A. Springer
    . 1994. Monocyte chemoattractant protein 1 acts as a T-lymphocyte chemoattractant. Proc. Natl. Acad. Sci. USA 91: 3652–3656.
    OpenUrlAbstract/FREE Full Text
  68. ↵
    1. O’Reilly L. A.,
    2. P. R. Hutchings,
    3. P. R. Crocker,
    4. E. Simpson,
    5. T. Lund,
    6. D. Kioussis,
    7. F. Takei,
    8. J. Baird,
    9. A. Cooke
    . 1991. Characterization of pancreatic islet cell infiltrates in NOD mice: effect of cell transfer and transgene expression. Eur. J. Immunol. 21: 1171–1180.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section