| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Genetic Medicine and Development (P.R., N.L.-M., E.H., J.-C.I., P.A.H.) and Cellular Physiology and Metabolism (F.C.), University Medical Center, 1211 Geneva, Switzerland; and Division of Endocrinology and Diabetes (J.A.E., M.B.-S., M.Y.D.), University Hospital of Zürich, 8091 Zürich, Switzerland
Address all correspondence and requests for reprints to: Dr. Pascale Ribaux, Department of Genetic Medicine and Development, University Medical Center, 1 Michel-Servet Street, 1211 Geneva-4, Switzerland. E-mail: Pascale.Ribaux{at}medecine.unige.ch.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B pathway inhibitor Bay 11–7082 it is demonstrated that CXCL1 expression and secretion are dependent on nuclear factor-B activation. IL-1 secreted by β-cells plated on 804G-ECM was found to be a key soluble mediator because treatment of cells with the IL-1 receptor antagonist significantly reduced both CXCL1 gene expression and secretion. It is concluded that ECM induces expression of cytokines including CXCL1 with amplification by IL-1 acting via a positive autocrine feedback loop. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, keratinocyte chemoattractant, or cytokine-induced neutrophil chemoattractant-1) and CXCL2 (Gro-β or macrophage inflammatory protein-2) belong to the CXC family and are classified into the ELR+ type, based on the presence of a triplet sequence (Glu-Leu-Arg) that imparts angiogenic function to this subset of chemokines (1, 4) whereas interferon-inducible protein-10 (IP10) (or CXCL10) is deprived of the ELR motif (3). The expression of many cytokines, including CXCL1 and IL-1β, depends on the transcription factor nuclear factor-B (NF-B) (5, 6, 7).
Although they were originally found to be secreted by immune cells in response to injury, cytokines can be expressed by many other cell types, including astrocytes (8), osteoclast precursors (9), adipocytes (10), and smooth muscle cells (11). Under specific circumstances, so far always shown to be cell detrimental, pancreatic β-cells can also produce cytokines. Indeed, IL-1β has been shown to be released by rat β-cells after exposure to double-stranded RNA and interferon-
(IFN) (12) and by human β-cells in response to high glucose concentration (13), although these data have been challenged (14). Moreover, treatment with IL-1β alone or in combination with IFN induces an increase in mRNA expression in rat primary β-cells for diverse chemokines, including IP10 (15, 16) and CXCL1 (15). Similarly, INS-1E cells treated with IL-1β and IFN produce chemokines such as IP10, CXCL1, and CXCL2 (17). In both type 1 and type 2 diabetes, cytokines seem to be implicated in β-cell death (18). In addition, expression and secretion of CXCL1 is increased in islets of high-fat-fed mice (19) and type 2 diabetic GK rats (Ehses, J. A., manuscript in preparation). But there is growing evidence for cytokine implication in nonapoptotic cellular processes such as angiogenesis (1, 4, 20), tumor growth (1, 5, 20), and development (21, 22).
In pancreatic islets, the extracellular matrix (ECM) surrounding β-cells is produced notably by local endothelial vascular cells (23), but the precise protein make-up of the ECM surrounding β-cells in their natural setting and the cellular origin of these components remain to be established. Using 804G-ECM (produced by rat bladder carcinoma cells) (24) as a model in vitro substratum, we have shown previously that it can activate the NF-B pathway and that this has positive effects on rat primary β-cell behavior via engagement of β1 integrins by laminin-5, a major component of this ECM (25, 26, 27, 28). Given the finding that activation of NF-B by fragments of fibronectin (another common ECM component) in other cellular models leads to induction of cytokine expression (29), we have now investigated whether cytokines could be expressed by rat β-cells in environmental conditions previously shown to improve their function.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B (C-20) from Santa Cruz Biotechnology (Santa Cruz, CA), polyclonal anti-Pdx1 (kind gift of Dr. C. V. Wright, Vanderbilt University, Nashville, TN), and monoclonal anti-CD68 (ED1) from Serotec (Düsseldorf, Germany). Secondary antibodies were Alexa Fluor 488 goat antimouse IgG (A11001) and Alexa Fluor 555 donkey antirabbit IgG (A31572) from Molecular Probes (Eugene, OR) and fluorescein isothiocyanate-conjugated goat antirabbit (Sigma). For blocking experiments, antibodies were rabbit antirat CXCL1 from Assay Designs (Ann Arbor, MI), hamster antirat CD29 (β1 integrin chain, Ha2/5) and control hamster IgM from Becton Dickinson Pharmingen (San Jose, CA), control mouse IgG from Sigma, and anti-3 chain of laminin-5 (CM6) kindly given by Dr. Vito Quaranta (Vanderbilt University). Biotin-16-uridine-5'-triphosphate used for the Oligo GEArray microarrays was from Roche (Mannheim, Germany).
Isolation of rat β-cells
Male Wistar rats of 150–200 g were used in all experiments. After collagenase perfusion, pancreases were extracted and islets of Langerhans were isolated by Ficoll gradient as described previously (25, 30). Rat islets were trypsinized and β-cells were collected by flow cytometry, based on their autofluorescence (30), with a cell sorter FACStar-Plus (Becton Dickinson, Sunnyvale, CA). Animals were treated according to protocols approved by the State Commissioner on Animal Care.
804G-ECM preparation
Rat bladder carcinoma 804G cells were grown in DMEM with 5.6 mM glucose and 10% fetal calf serum. At confluence, cells were washed, and medium deprived of serum was added and maintained for the next 3 d. Conditioned medium (referred to as 804G-ECM hereafter) was cleared by centrifugation (180 x g for 5 min), filtered (0.22 µm), and frozen at –20 C until use.
Coating of petri dishes
Petri dishes were coated with droplets of 60 µl of either poly-L-lysine (pLL) at 0.1 mg/ml or pure 804G-ECM. After an overnight incubation in a damp box at 37 C, dishes were washed three times for 5 min each with sterile water and air dried briefly before being plated with β-cells.
To block laminin-5, dishes coated with 804G-ECM were incubated with either CM6 or nonimmune mouse IgG antibodies at 0.1 mg/ml for 2 h at 37 C, whereas control dishes were incubated with PBS. Then dishes were washed three times for 5 min each with PBS and air dried.
β-Cell culture
Sorted β-cells were washed in culture medium (DMEM, 10% FCS, 11.2 mM glucose, 110 mg/liter Na pyruvate, 66 U/ml penicillin, 66 µg/ml streptomycin, and 50 mg/liter gentamycin) and allowed to recover overnight in suspension at 37 C in nonadherent 100-mm petri dishes. β-Cells were then collected, resuspended in culture medium to achieve a concentration of 600,000 cells/ml, and plated on coated petri dishes per droplet of 50 µl (except for chemotaxis assay, see below).
When indicated, before being plated, cells were preincubated in suspension with occasional shaking for 1 h at 37 C with either 4 µg/ml Ha2/5 antibody (or hamster IgM as a control) to check the role of β1 integrins, 5 µM Bay 11-7082 (or dimethylsulfoxide as a vehicle control) to evaluate the implication of the NF-B pathway or 1 µg/ml IL-1Ra to assess the role of IL-1.
Oligo GEArray experiments
Total RNA was extracted from sorted β-cells in suspension or after 4 h of culture on pLL or 804G-ECM, with the QIAshredder and RNeasy Mini kit (QIAGEN, Basel, Switzerland). Agarose gel electrophoresis was used to check the quality of the RNA. The UTP-biotinylated cRNA probe was synthesized with the TrueLabeling-AMP 2.0 Kit (SuperArray, Frederick, MD) and purified with the ArrayGrade cRNA Cleanup kit (SuperArray). Two different nylon membrane-based DNA microarrays were used (both from SuperArray): ORN-022 specific for rat chemokines and receptors and ORN-011 focused on rat inflammatory cytokines and receptors. The membranes were hybridized in the GEAhyb hybridization solution (SuperArray) with 3–4 µg cRNA probe and washed following the manufacturer’s instructions. Signals were detected with the chemiluminescent detection kit (SuperArray) and exposed to x-ray films.
Quantitative real-time RT-PCR
RT was performed with 300 ng total RNA with Superscript II (Invitrogen, Basel, Switzerland) in a volume of 20 µl. The quantitative detection of the PCR product was performed in the presence of SYBR Green I (Eurogentec, Seraing, Belgium) and fluorescein (Bio-Rad, Reinach, Switzerland) incorporated into the PCR buffer (qPCR core kit, Eurogentec) with the iCycler iQ System (Bio-Rad). The relative expression was normalized with the housekeeping gene EF1a. Oligonucleotide primers were as follows: rat CXCL1 forward 5'-AGA ACA TCC AGA GTT TGA AGG TGA T-3' and reverse 5'-GTG GCT ATG ACT TCG GTT TGG-3', rat CXCL2 forward 5'-TGG TTC AGA GGA TCG TCC AAA-3' and reverse 5'-CAG GAG CCC ATG TTC TTC CTT-3', rat IP10 forward 5'-TTC TTT GGC TCA CCG CTT TC-3' and reverse 5'-ATC CGG AAT CTG AGG CCA TC-3', rat IB forward 5'-TGC TGA GGC ACT TCT GAA AGC-3' and reverse 5'-TCC TCG AAA GTC TCG GAG CTC-3', rat CD68 forward 5'-CTT GCG CCA GTG ACC AAT C-3' and reverse 5'-GGA CCA GGC CAA TGA TGA GA-3', and rat EF1a forward 5'-AGC AAA AAT GAC CCA CCA ATG-3' and reverse 5'-ATC TGG CCT GGA TGG TTC AG-3'.
TaqMan gene expression assay
Total RNA of β-cells plated on pLL or 804G-ECM for 6 h was reverse transcribed using the Superscript II RNase H– reverse transcriptase kit (Invitrogen) according to the manufacturer’s instructions. RNA was primed with random hexamers (Microsynth, Balgach, Switzerland), and the reaction was carried out at 37 C for 2 h. Quantitative PCR of the cDNAs was performed using commercial TaqMan gene expression assays and the real-time PCR system 7000 of Applied Biosystems. The following assays were used: rat IL-1β Rn00580432_m1 and eukaryotic 18S rRNA Hs99999901_s1 (Applied Biosystems, Rotkreuz, Switzerland). Cycle threshold (Ct) values of cDNA samples were corrected for different amounts of input cDNA using 18S RNA as a reference. The data were analyzed using the 
Ct method and expressed as fold difference between the two culture conditions (pLL or 804G-ECM).
In situ hybridization
Sorted β-cells were cultured for 24 h on pLL or 804G-ECM before being treated or not with rat IL-1β for 6 h, trypsinized, and cytospun on Superfrost Plus slides at 800 rpm for 5 min (Cytospin 3; Shandon). Spots of cells were dried at room temperature for 30 min, and slides were frozen at –20 C until in situ hybridization was performed.
Digoxigenin-tagged RNA probes were generated from DNA templates. In vitro transcription was carried out in a cocktail of 1 mM rATP, rCTP, and rGTP, 0.65 mM rUTP, 0.35 mM digoxigenin-UTP (RNA labeling kit; Roche), 1 µl ribonuclease inhibitor (40 U/µl MBI, Promega, Wallisellen, Switzerland), 1 µg DNA template, and 0.5 µl RNA polymerase (50 U/liter T7 and 20 U/liter SP6; both from New England Biolabs, Hitchin, UK) in a final volume of 20 µl. The reaction mix was incubated at 37 C for at least 150 min followed by a 20-min incubation with a DNase I/MgCl2 mix to remove the DNA template [1.6 µl 300 mM MgCl2, 2 µl DNase I (10 U/liter; Roche), and 16.4 µl diethylpyrocarbonate (DEPC) water]. RNA was ethanol precipitated and redissolved in 22 µl DEPC water while vigorously shaking at room temperature. One microliter of the RNA was analyzed on a 1% agarose gel, and 1 µl was used to determine concentration in a photometer. Probes were diluted with DEPC-treated water to 150 ng/µl and stored as 20-µl aliquots at –80 C. Before use, riboprobes were diluted in hyb-mix (Ambion, Huntingdon, UK) to a final concentration of 30 ng/µl. Nonradioactive in situ hybridization was performed as described previously (31).
Detection of secreted cytokines
Sorted rat β-cells were cultured on pLL or 804G-ECM for various times in the presence or absence of different components (see figure legends for more details), and conditioned medium was collected and cleared by centrifugation (2200 x g for 10 min). A 25-µl volume of supernatant was then used to detect the presence of rat CXCL1, IL-1β, and IP10 by Luminex technology (RCYTO-80K; Linco Research, St. Charles, MO).
Chemotaxis assay
Sorted β-cells were cultured on pLL or 804G-ECM at high density (90,000 cells/30 µl) to increase the chemokine concentration secreted in the supernatant, which was collected after 24 h. All the supernatants were then diluted two times in culture medium deprived of serum to achieve the minimal volume required to perform the migration assay and reduce serum-dependent background. When indicated, supernatant from β-cells plated on 804G-ECM was incubated with a 50-fold excess of rat anti-CXCL1 antibodies for 3 h at 4 C before the migration assay was performed. Rat granulocytes were isolated from total blood by Histopaque-1077 gradient (Sigma) and resuspended at 6.5 x 106 cells/ml in serum-free RPMI (Invitrogen). Chemotaxis assay was performed according to Falk et al. (32) in 48-microwell chemotaxis chambers (Neuro Probe, Gaithersburg, MD) using 5-µm pore-size polyvinylpyrrolidone-free polycarbonate membranes (Neuro Probe). The wells in the lower chamber were filled with 28 µl supernatant. Serum-deprived culture medium was used as a control for random unstimulated migration, and recombinant rat CXCL1 (1 ng/ml) and fMLP (250 nM) were used as positive controls. Each well in the upper chamber was filled with 50 µl rat granulocyte suspension. Each experimental condition was run in sestuplicate. The chambers were incubated for 1 h at 37 C in a 5% CO2 humidified atmosphere. The membranes were then removed, and the cells were fixed with 100% ethanol and stained with toluidine blue. Cells that had not migrated were removed from the upper surface of the membranes with filter paper. Migration was measured by densitometric analysis with Scion Image software (Scion, Frederick, MD).
Immunofluorescence
NF-B staining.
After 24 h of culture on pLL or 804G-ECM, monolayers of β-cells were treated with or without 1 µg/ml IL-1Ra for 1 h, followed by an incubation period of 20 min with or without 0.1 ng/ml rat recombinant IL-1β. Then β-cells were briefly washed with PBS, fixed for 20 min at room temperature with 4% paraformaldehyde, and washed again three times for 5 min each. After permeabilization (0.5% Triton X-100 in PBS for 5 min) and brief rinses (three times for 3 min each in PBS), blocking of nonspecific sites was performed with 0.5% BSA in PBS for 1 h. Primary and secondary antibodies (fluorescein isothiocyanate-conjugated goat antirabbit) were diluted at 1:100 in PBS plus 0.5% BSA and applied for 1.5 and 1 h, respectively.
Macrophage staining.
Cultures of rat macrophages and rat β-cells plated for 24 h on 804G-ECM were fixed, permeabilized, and blocked as described above. Primary antibodies were diluted in PBS plus 0.5% BSA at 1:100 for ED1 and 1:2000 for Pdx1 and applied for 1.5 h. Alexa Fluor secondary antibodies were used at 1:1000 in PBS plus 0.5% BSA for 1 h. Nuclei were labeled with Hoechst 33342 (Sigma), and the preparations were observed with an Axiocam fluorescence microscope (Zeiss, Oberkochen, Germany).
Data and statistical analysis
All data are presented as means ± SE for n independent experiments. Levels of significance between groups were assessed by Student’s t test for unpaired groups.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) and the other specific for ILs (Fig. 1B). Despite a common pattern of mRNA expression for the three conditions of cell culture, some differences were detectable. Compiling the data from the two arrays, four cytokines were shown to be overexpressed in β-cells plated on 804G-ECM (circled spots in Fig. 1): CXCL1, CXCL2, and IP10 with a strong increase and IL-1β with a much fainter signal. The experiment was repeated twice with reproducible data.
|
). A time-course experiment was performed on cDNA extracted at different time points from rat β-cells plated on pLL or 804G-ECM. The fold induction of CXCL1 (Fig. 2A), IP10 (Fig. 2B), and CXCL2 (Fig. 2C) was normalized to the starting condition (β-cells in suspension immediately before culture) and expressed as a percentage of the value obtained from cells on 804G-ECM at 4 h. Although the peak fold induction for each chemokine was variable, the graphs exhibit a very similar pattern of expression for the three chemokines with a maximum at 4 h followed by a slow decrease. Moreover, the overexpression of CXCL1 and IP10 on the 804G-ECM a the 2-h time point (22.2- and 2.9-fold, respectively) was similar to previous reproducible data obtained from two independent experiments of high-density microarrays from Affymetrix (Santa Clara, CA), performed at the same time point, showing an overexpression of 35.4-fold for CXCL1 and 3.7-fold for IP10 (Irminger, J.-C., unpublished data). In keeping with the low signal intensity on the Oligo GEArray, a TaqMan gene expression assay, which is more sensitive, was required to detect IL-1β. The 6-fold induction of IL-1β in β-cells plated on 804G-ECM compared with pLL was significant (Fig. 2D).
|
). The expression of IL-1β induced by the 804G-ECM was too low to be detectable, but after treatment with 2 ng/ml IL-1β, β-cells were positive for IL-1β (Fig. 2E). Insulin was used as a positive control and to check that cells expressing these cytokines were indeed β-cells.
Cytokine expression by rat β-cells is not due to contamination by macrophages
Because the purity of the β-cell population is not absolute (
95% insulin-positive cells, with the remainder largely -cells based on glucagon staining) (30), it was possible that there was contamination by macrophages, which are a major source of cytokines. Double immunostaining with antibodies specific for β-cells (Pdx1) and macrophages (ED1), respectively, was performed. Purified rat macrophages were used to check the efficacy of the ED1 marker. As expected, these cells were negative for Pdx1 staining (Fig. 3A4) and positive for the ED1 marker (Fig. 3A5). On the other hand, no macrophages were visible in the β-cell population (Fig. 3A2) where nearly all cells were positive for nuclear Pdx1 staining (Fig. 3A1). The absence of macrophages was also confirmed by quantitative real-time RT-PCR (Fig. 3B). The expression of the macrophage-specific membrane receptor CD68 was compared in rat macrophages and in three independent FACS-sorted β-cell populations. As expected, CD68 was very strongly expressed in macrophages, whereas it was absent from the β-cell populations as well as from the insulinoma cell line INS-1E that was used as a negative control (Fig. 3B).
|
), suggesting that the concentration of IP10 and IL-1β was too low (beneath the level of detection of 3.78 and 2.32 pg/ml, respectively) to allow for detection (data not shown). A control experiment performed without cells indicated that the 804G-ECM itself released some CXCL1, but this amount was constant with time (80 pg/ml, symbolized by a dashed line in Fig. 4A). By contrast, in conditioned media from β-cells plated on 804G-ECM, the concentration of CXCL1 increased with time, and the differences vs. pLL were highly significant (P < 0.001) at each time point (Fig. 4A).
|
). Supernatant from β-cells on pLL did not attract granulocytes compared with control culture medium, whereas supernatant from β-cells on 804G-ECM significantly increased their migration (P < 0.001) to the same extent as recombinant rat CXCL1 at 1 ng/ml. Neutralization of CXCL1 in supernatant from β-cells on 804G-ECM with specific antibodies almost completely reversed the migration (Fig. 4B).
Laminin-5 is involved in CXCL1 expression
Laminin-5 is a major component of 804G-ECM, and we previously showed that its interaction with β1 integrins has positive effects on rat β-cell behavior (25, 26, 27, 28). To check whether it was also involved in CXCL1 expression, we used specific blocking antibodies against either laminin-5 (Fig. 5A) or β1 integrins (Fig. 5B). Results show that preventing the ligation of laminin-5 with β1 integrins by blocking either the ligand or the receptor with the relevant antibodies leads to a 40–50% significant (P < 0.05) reduction of CXCL1 expression on 804G-ECM.
|
B and IL-1 are key mediators of CXCL1 expressionB dependent, and we have shown previously that the 804G-ECM induces the activation of NF-B in rat β-cells (26). To check for involvement of NF-B in CXCL1 expression, β-cells were preincubated for 1 h with 5 µM Bay 11-7082, an inhibitor of NF-B activation used in our earlier study (26), and then plated on pLL or 804G-ECM for 10 h in continued presence of the inhibitor. Real-time quantitative RT-PCR was performed to assess the level of induction of CXCL1 (Fig. 6A) and IB (Fig. 6B), a well-characterized NF-B target gene. The release of CXCL1 was measured in the conditioned media (Fig. 6C). In the presence of Bay 11-7082, the reduction of the level of expression of IB was highly significant (P < 0.001) in β-cells cultured on 804G-ECM (Fig. 6B). In this same condition, the induction of CXCL1 was abolished at the mRNA level (Fig. 6A) as well as at the protein level (Fig. 6C).
|
B in β-cells exposed to exogenous IL-1β. In untreated cells (Fig. 7A, 1 and 5) as well as in cells exposed to IL-1Ra (1 µg/ml) for 1 h (Fig. 7A, 3 and 7), NF-B had a cytosolic localization, whereas after exposure to IL-1β (0.1 ng/ml) for 20 min, all nuclei stained for NF-B (Fig. 7A, 2 and 6). The pretreatment of β-cells with IL-1Ra completely inhibited the IL-1β-induced nuclear translocation of NF-B (Fig. 7A, 4 and 8).
|
B expression (Fig. 7C) that was highly significant (P < 0.001) on 804G-ECM compared with untreated cells. There was also a highly significant (P < 0.001) inhibition by IL-1Ra of 804G-ECM induction of CXCL1 gene expression (Fig. 7B) and secretion (Fig. 7D), but this inhibition was only partial with a significant increase of both expression (Fig. 7B) and secretion (Fig. 7D) for 804G-ECM vs. pLL in the presence of this receptor antagonist. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although the FACS-sorted β-cell population is highly pure, there are still less than 5% of non-β-cells remaining. So, one may argue that the cytokine mRNAs detected in this study derive from possible macrophage contamination. This was excluded because the specific macrophage marker CD68, which was efficient in labeling rat macrophages, was not detected at the mRNA and protein level in the β-cell population. In addition, the few non-β-cells present in the FACS-sorted β-cell culture did not exhibit the typical irregular nuclear shape of macrophages but rather had small round nuclei. Moreover, the overnight culture in suspension of freshly FACS-sorted β-cells before all experiments is in itself enough to eliminate any undesired macrophage, because the latter adhere very rapidly to the dishes used (Armanet, M., personal communication).
The time-course experiments show that the three chemokines CXCL1, CXCL2, and IP10 have a similar pattern of expression with a maximal induction after 4 h of culture on 804G-ECM. The level of expression of IL-1β induced by 804G-ECM in β-cells was lower than these. Nevertheless, it was clearly detected by the highly sensitive TaqMan assay, and treatment with IL-1β at 2 ng/ml induced an IL-1β expression visible by in situ hybridization, proving that sorted β-cells can produce IL-1β. The failure to measure IL-1β in the conditioned medium supports the common finding that the picomolar range of secreted IL-1 may easily escape detection by routine ELISAs (33). Nonetheless, we can affirm that IL-1 acts in an autocrine manner to enhance the expression of CXCL1 because treatment with IL-1Ra reduced significantly this action. This is in agreement with recent data reporting that release of chemokine KC (mouse homolog of rat CXCL1) from mouse islets exposed to a type 2 diabetic milieu is blunted by treatment with IL-1Ra (19). This also confirms previous results showing the implication of IL-1β in the expression of CXCL1 in rat β-cells (15) and suggests a role for IL-1 as a mediator of the beneficial effects induced by the 804G-ECM in rat β-cells. In keeping with this hypothesis, treatment of human β-cells cultured for 4 d on ECM with low doses of IL-1β (10–20 pg/ml) induces an increase in replication and a decrease in apoptosis (34). It was also shown many years ago that low levels of this cytokine improve insulin release from rat islets (35). The impact of cytokines secreted by islet cells appears to be strongly influenced by the amplitude and kinetics of expression as well as by the prevailing biological context. IL-1β has thus been shown to exert either beneficial (34) or detrimental (18) effects on β-cell function and survival. We postulate a similar situation for islet cell secretion of CXCL1 or its homologs, with possible beneficial effects when triggered by, e.g. ECM, and detrimental ones in an individual with type 2 diabetes.
As reported previously by our group, laminin-5 contained in 804G-ECM activates β1 integrins (25). We now show that preventing the interaction of β1 integrins and laminin-5 with blocking antibodies to either partner of this ligation results in a 40–50% inhibition of CXCL1 expression. This incomplete inhibition could be attributed to the indirect aspect of this method, although we do not exclude the involvement of other 804G-ECM components.
This matrix also stimulates the NF-B pathway in rat β-cells (26). The present data demonstrate that the 804G-ECM-induced expression of CXCL1 by these cells is dependent on NF-B activation. This would be in agreement with the recent finding that fibronectin fragments can activate NF-B via integrins 5β1 in human chondrocytes, inducing the production of diverse cytokines and notably CXCL1 and IL-1β (29). However, we cannot draw a strong parallel between β-cells and chondrocytes because leaving aside their different nature, the ECM surrounding these cells most probably differs in its composition. Nevertheless, the secreted IL-1β has an autocrine action in human chondrocytes (29), similar to what we see in rat β-cells.
In this study, only the chemokine CXCL1 was secreted in sufficient amounts by rat β-cells plated on 804G-ECM to be quantified by Luminex technology, whereas IP10 and IL-1β were not, and CXCL2 was not included in the kit used. Similarly, the production and release of a substantial amount of chemokine KC and IL-8 (respective mouse and human functional homologs of rat CXCL1) were observed in mouse and human islets exposed to a type 2 diabetic milieu and also in mouse islets from hig- fat-fed mice (19). The release of some CXCL1 by the 804G-ECM itself was somewhat unexpected, but this does not put in question our results, because this amount is stable with time (80 pg/ml), whereas it increases in the presence of β-cells. Indeed, the concentration of CXCL1 secreted by 30,000 rat β-cells in 50 µl was about 800 pg/ml after 24 h of culture, thereby exceeding the background amounts released from 804G-ECM by an order of magnitude. Moreover, the use of the NF-B pathway inhibitor Bay 11-7082 completely prevented the rise of CXCL1 secretion in the presence of β-cells on 804G-ECM, proving its cellular origin.
CXCL1 is known to attract neutrophils and other polymorphonuclear cells, and the present migration assay proves that secreted CXCL1 induced by 804G-ECM is functional. Indeed, granulocyte migration was identical after exposure to 1 ng/ml rat recombinant CXCL1 or supernatant from β-cells plated on 804G-ECM. Moreover, blocking CXCL1 with specific antibodies almost completely reversed this effect. This correlates with recent finding showing that neutrophil migration was induced by conditioned medium from human islets exposed to a type 2 diabetic milieu and was abolished by IL-8 neutralization (19).
Purified rat CXCL1 protein used at 10 or 100 ng/ml does not seem to have any impact on proliferation, apoptosis, and insulin secretion of rat β-cells plated on pLL or 804G-ECM (data not shown). Moreover, it does not induce the expression and the secretion of CXCL1 in rat β-cells cultured on pLL and does not increase CXCL1 expression on 804G-ECM (data not shown). This suggests that CXCL1 does not have an autocrine action in rat β-cells. Similar conclusions were drawn in mouse islets, based on mild effects of KC (mouse homolog of rat CXCL1) on β-cell apoptosis and insulin secretion (19). The oligo-array experiments tend to confirm this argument seeing that among the two known receptors (CXCR1 and CXCR2) for CXCL1, CXCR1 alone is expressed in rat β-cells (data not shown). Based on the results of other groups, it seems that only the ligation of CXCL1 to CXCR2 induces biological effects (36, 37, 38). We suggest that CXCL1 secreted by rat β-cells acts more probably in a paracrine manner in situ in the natural setting of β-cells within islets. A possible target would be the microvascular endothelial cell, which expresses CXCR2 (37).
In conclusion, the present study demonstrates that rat β-cells express cytokines CXCL1, CXCL2, IP10, and IL-1β in response to ECM under conditions previously shown to improve function and survival. The production of CXCL1 is an NF-B-dependent event that is enhanced by IL-1 in an autocrine fashion (Fig. 8). It follows that inducing secretion of appropriate levels of both CXCL1 and IL-1 may be required for normal β-cell function via the natural vicinity of extracellular proteins in vivo and of benefit when transplanting islets or in any future attempt to regenerate the endocrine pancreas in individuals with diabetes.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
This work was supported by Grants 3200B0-101902 (to P.A.H.) and PP00B-68874/1 (to M.Y.D.) from the Swiss National Science Fund, Grant 7-2005-1158 (to P.A.H.) and a Fellowship (to J.A.E.) from the Juvenile Diabetes Research Foundation, a research grant from the European Foundation for the Study of Diabetes (EFSD)-Merck Sharp & Dohme (MSD) partnership program (to M.Y.D.), and the University Research Priority Program "Integrative Human Physiology" at the University of Zürich (to M.Y.D. and J.A.E.).
Disclosure Statement: P.R., J.A.E., N.L.-M., F.C., M.B.-S., E.H., J.-C.I., and M.Y.D. have nothing to declare. P.A.H. consults for Merck/MSD-Switzerland, Amylin-Lilly, and GlaxoSmithKline. P.A.H. received lecture fees from Merck/MSD-Switzerland.
First Published Online August 16, 2007
Abbreviations: CXCL1, C-X-C motif ligand 1; DEPC, diethylpyrocarbonate; ECM, extracellular matrix; fMLP, formyl-Met-Leu-Phe; IFN, interferon-; IL-1Ra, IL-1 receptor antagonist; IP10, interferon-inducible protein-10; NF-B, nuclear factor-B; pLL, poly-L-lysine.
Received March 9, 2007.
Accepted for publication August 7, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B-dependent genes in primary rat pancreatic β-cells. J Biol Chem 276:48879–48886B through its type I but not its type II receptor in lymphocytes. J Biol Chem 267:15836–15841 regulation in airway smooth muscle by IL-1β and TNF-: role of NF-B and MAP kinases. Am J Physiol Lung Cell Mol Physiol 291:L66–L74 induce the expression of diverse chemokines and IL-15 in human and rat pancreatic islet cells, and in islets from pre-diabetic NOD mice. Diabetologia 46:255–266[Medline]B by extracellular matrix is involved in spreading and glucose-stimulated insulin secretion of pancreatic beta cells. J Biol Chem 280:30630–306376β1 integrin. Diabetes 49:233–243[Abstract]B mediates the stimulation of cytokine and chemokine expression by human articular chondrocytes in response to fibronectin fragments. J Immunol 174:5781–5788, GROβ, GRO, neutrophil-activating peptide-2, and epithelial cell-derived neutrophil-activating peptide-78 are potent agonists for the type B, but not the type A, human interleukin-8 receptor. J Biol Chem 271:20545–20550This article has been cited by other articles:
 |
|
 |
|
  M. Boni-Schnetzler, S. Boller, S. Debray, K. Bouzakri, D. T. Meier, R. Prazak, J. Kerr-Conte, F. Pattou, J. A. Ehses, F. C. Schuit, et al. Free Fatty Acids Induce a Proinflammatory Response in Islets via the Abundantly Expressed Interleukin-1 Receptor I Endocrinology, December 1, 2009; 150(12): 5218 - 5229. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bouzakri, P. Ribaux, and P. A. Halban Silencing Mitogen-activated Protein 4 Kinase 4 (MAP4K4) Protects Beta Cells from Tumor Necrosis Factor-{alpha}-induced Decrease of IRS-2 and Inhibition of Glucose-stimulated Insulin Secretion J. Biol. Chem., October 9, 2009; 284(41): 27892 - 27898. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Parnaud, E. Hammar, P. Ribaux, M. Y. Donath, T. Berney, and P. A. Halban Signaling Pathways Implicated in the Stimulation of {beta}-Cell Proliferation by Extracellular Matrix Mol. Endocrinol., August 1, 2009; 23(8): 1264 - 1271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Boni-Schnetzler, J. Thorne, G. Parnaud, L. Marselli, J. A. Ehses, J. Kerr-Conte, F. Pattou, P. A. Halban, G. C. Weir, and M. Y. Donath Increased Interleukin (IL)-1{beta} Messenger Ribonucleic Acid Expression in {beta}-Cells of Individuals with Type 2 Diabetes and Regulation of IL-1{beta} in Human Islets by Glucose and Autostimulation J. Clin. Endocrinol. Metab., October 1, 2008; 93(10): 4065 - 4074. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bouzakri, P. Ribaux, A. Tomas, G. Parnaud, K. Rickenbach, and P. A. Halban Rab GTPase-Activating Protein AS160 Is a Major Downstream Effector of Protein Kinase B/Akt Signaling in Pancreatic {beta}-Cells Diabetes, May 1, 2008; 57(5): 1195 - 1204. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) or 804G-ECM (
) for 2, 4, 10, or 24 h was used for quantitative real-time RT-PCR to detect the pattern of expression of CXCL1 (A), IP10 (B), and CXCL2 (C). On each graph, the level of the chemokine mRNA expression is normalized to the control condition (β-cells in suspension immediately before culture) and expressed as a percentage of the value obtained from cells on 804G-ECM at 4 h. n = 3; *, P < 0.05; **, P < 0.001 at respective time point. D, Expression of IL-1β was assessed by real-time TaqMan PCR on cDNA from rat β-cells plated on pLL or 804G-ECM for 6 h. n = 3; *, P < 0.05. E, In situ hybridization was performed on β-cells plated on 804G-ECM for 6 h with or without IL-1β (0.1 ng/ml for CXCL1 and 2 ng/ml for IL-1β). Sense probes were used as control (CTRL). Bar, 10 µm
