Introduction

Cytokines have been shown to have dramatic effects on pancreatic islets and insulin-secreting β-cell lines. Interleukin (IL)-1β is thought to be a key mediator of both impaired function and destruction of pancreatic β-cells during the development of autoimmune type 1 diabetes [1]. Furthermore, previous findings have established the role of IL-1β in β-cell failure in type 2 diabetes [2]. Indeed, exposure of cultured human islets to elevated glucose levels leads to β-cell production and release of IL-1β [2]. In turn, IL-1β feeds back on the β-cell to induce impaired function and apoptosis. Among the several signaling pathways activated by IL-1β in the β-cell is the activation of extracellular signal-regulated kinase (ERK). ERK constitutes a stress-activated member of the mitogen-activated protein kinase (MAPK) family and is involved in the transmission of stress and apoptotic signaling in many cells. ERK plays a critical role in cytokine-mediated apoptosis of β-cells, as suggested by the finding that ERK activation has been demonstrated to be required both for cytokine-induced expression of inducible nitric oxide (NO) synthase (iNOS) [3] and β-cell apoptosis [4]. Furthermore, glucose also stimulates activation of ERK in β-cells [5–10] and ERK activation was already shown to be required for double-stranded RNA- and virus-induced IL-1β expression [11]. Thus, strong evidence points toward ERK as a key regulator of β-cell function and apoptosis. However, the regulatory events in the activation of ERK by IL-1β remain largely unknown.

Increases in free intracellular Ca2+ concentration ([Ca2+]i) are critical for signal transduction in many cell types [6, 12]. Pancreatic β-cells are highly active in Ca2+ handling. Thus, β-cell uptake and metabolism of glucose lead to the closure of ATP-sensitive K+ channels, depolarization of the plasma membrane, and, subsequently, influx of Ca2+ through voltage-gated Ca2+ channels (VGCCs). The subsequent increase in cytosolic free Ca2+ concentration ([Ca2+]i) is an important determinant for insulin granule exocytosis. Ca2+ entry in response to glucose in insulin-secreting cells also leads to the activation of ERK [6, 12, 13], but not of c-Jun NH2-terminal kinase (JNK) or p38 MAP kinase to any significant degree [12, 14]. VGCCs also regulate β-cell apoptosis. Furthermore, cytokine-mediated β-cell death seems to be dependent on VGCC-mediated Ca2+ entry as suggested by the findings that blockers of L-type VGCCs can suppress β-cell apoptosis induced by IL-1β [15, 16] or IFNγ plus TNFα [17] and blockers of T-type VGCCs can repress β-cell apoptosis induced by a mixture of cytokines [18].

In the present study, we investigated if Ca2+ and ERK activation contributes to IL-1β-induced β-cell secretory dysfunction and apoptosis in rat islets. Our findings provide evidence for a potential role of Ca2+ and ERK activation in IL-1β-induced β-cell apoptosis signaling.

Materials and methods

Materials

Recombinant human IL-1β was obtained from R&D Systems (Minneapolis, MN). Mannitol was bought from Sigma (Shanghai, China). Reagents for SDS-polyacrylamide gel electrophoresis (SDS-PAGE) were all bought from Bio-Rad (Shanghai, China). Antibodies against ERK1/2 and phospho-ERK1/2 were obtained from Cell Signaling (Beverly, MA). Mibefradil was provided by Roche (Basel, Switzerland). Nimodipine, ionomycin, thapsigargin, W7, and bis-(o-aminophenoxy)- N,N,N′,N′-tetraacetic acid-acetoxymethyl (BAPTA-AM) were obtained from Calbiochem (San Diego, CA). Compound PD 098059 from New England Biolabs (Pickering, Ontario) is the specific inhibitor of MAPK/ERK kinase activation. Mibefradil were dissolved in water. Nimodipine and all other inhibitors were dissolved in polyethylene glycol or dimethylsulfoxide (DMSO).

Islet isolation and culture

Islets of Langerhans from Wistar Furth rats were isolated by handpicking after collagenase digestion of the pancreata [19]. Islets were precultured in 33-mm dishes for 5–7 days at 37°C in atmospheric humidified air in complete RPMI 1640 medium (Life Technologies, Inc.) containing 100 units/ml penicillin, 100 μg/ml streptomycin, and 10% FCS (Invitrogen). On the day of experimentation, around 250 rat islets were placed in six-well dishes in RPMI 1640 medium containing 11.1 mM glucose and supplemented with 5% fetal calf serum. All animal work was carried out according to national and international law and ethical standards.

Islet stimulation and lysis

At the time of experimentation (1 or 2 days after seeding of cells), the culture medium was removed and fresh medium (11.1 mM glucose) with or without drugs was added. Islets were incubated in the presence of drugs for 30 min to 1 h before exposure to IL-1β. When necessary, solvents (DMSO, polyethylene glycol, or ethanol) used for drugs were added to all conditions (1–25 μM final concentration). After IL-1β exposure, islets were washed in cold PBS, and lysed for 30 min on ice in RIPA buffer (Sigma). The solution was centrifuged at 15,000 × g for 5 min at 2–4°C. The supernatants (whole-cell lysates) were stored at −80°C until assayed.

TUNEL staining

A terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end-labeling (TUNEL) kit was used according to the manufacturer’s instructions (Boehringer Mannheim). Islet cells were fixed with paraformaldehyde, triton permeabilized, then incubated with a TUNEL reaction mixture containing fluorescein-dUTP and TdT to catalyze the attachment of fluoresceinated dUTP to the free 3′-OH ends of DNA strand breaks. A counterstain for cell nuclei [4′, 6′-diamino-2-phenylindole (DAPI)] was applied in mounting medium (Vector Laboratories). Cells were visualized using a Nikon 300 inverted fluorescence microscope with appropriate fluorescence filters. Percent apoptosis was calculated by dividing the number of green, TUNEL-staining cells by the total number of blue DAPI staining nuclei. At least five fields per sample containing a minimum of 100 cells were counted.

Caspase activity assay

Caspase-3 activity assay was performed according to the manufacturer’s instructions (CLONTECH Laboratories). Emissions were read using a fluorometer at 390 nm. Emissions of samples were compared with uninduced control cells to determine the fold increase in caspase-3 activity.

Detection of apoptosis by flow cytometry

Islet cells were cultured in the presence of cytokines or inhibitors for 24 h. A previously described method was used for quantification of apoptosis [20]. In brief, islet cells were incubated for 15 min with propidium iodide (10 mg/ml), rinsed in PBS, and subsequently trypsinized for 8 min at 37°C. By incubating the islets with propidium iodide prior to the trypsination procedure, trypsination artifacts are avoided. The dispersed islet cells were then washed and resuspended in culture medium containing fetal calf serum. The cells were analyzed in a Becton Dickinson FACSCalibur flow cytometer with respect to their forward scatter and FL3 fluorescence. Data were analyzed using the CellQuest software (Becton Dickinson Instruments).

IL-1β release

IL-1β release was evaluated in the culture medium collected before the termination of the experiment using a human IL-1β enzyme-linked immunosorbent assay (R&D Systems).

Immunoblotting

Equivalent amounts of cell extracts were mixed with SDS sample buffer, boiled for 5 min, and subjected to 10% SDS-PAGE. Proteins were electrically transferred to nitrocellulose filter membranes. Membranes were incubated for 1 h at room temperature in Tris-buffered saline solution with 0.1% Tween-20 (TBST) and 5% milk to block nonspecific binding sites. After washing in TBST, the membranes were incubated at 4°C overnight with anti-phosphospecific ERK1/2 (1:1,000) or total anti-ERK1/2 (1:1,000). Filter membranes were then washed in TBST and incubated for 1 h with horseradish peroxidase-conjugated secondary antibodies (1:20,000). Labeled proteins were visualized by chemiluminescence according to the manufacturer’s instructions (Amersham Biosciences).

Measurements of intracellular calcium concentration

For measurement of whole cell intracellular calcium ion concentration ([Ca2+]i), cells on coverslips were loaded with 3 mM fura-2/AM (a Ca2+-sensitive fluorescent dye) for 20 min at 37°C After incubation with fura-2/AM, inspection of cells by fluorescence microscopy demonstrated no vesicular compartmentalization of fura-2, suggesting that the dye loading method permitted measurement of cytosolic [Ca2+]i. Measurements were performed with the buffer 150 mM NaCl, 5 mM KCl, 10 mM d-glucose, 1 mM MgSO4, 1 mM Na2HPO4, and 20 mM HEPES at pH 7.4 with an osmolarity of 291 mosM. Estimates of whole cell [Ca2+]i was calculated according to the equation of Grynkiewicz [21] from the emitted fluorescence measurements using a Nikon Diaphot II inverted microscope optically interfaced to a Deltascan 4000, dual beam, epifluorescence spectrofluorimeter, and analysis system (Photon Technology Int.).

Insulin release and content

For determining the acute insulin release in response to glucose stimulation, islets were washed in Krebs–Ringer bicarbonate buffer (KRB-HEPES [pH 7.4]: 4.8 mM KCl, 134 mM NaCl, 5 mM NaHCO3, 1.2 mM MgSO4, 1.2 mM KH2PO4, 1 mM CaCl2, 0.5% BSA, and 10 mM HEPES) that contained 3.3 mM glucose and preincubated for 30 min in the same buffer. The Krebs-Ringer buffer was then discarded and replaced by fresh buffer that contained 3.3 mM glucose for 1 h for basal secretion, followed by an additional 1 h incubation in Krebs–Ringer buffer that contained 16.7 mM glucose. Supernatants were collected and frozen for insulin assays. Thereafter, islets were washed with PBS and extracted with 0.18 N HCl in 70% ethanol for 24 h at 4°C. The acid–ethanol extracts were collected for determination of insulin content. Insulin was determined by a human insulin radioimmunoassay kit (CIS Bio International).

Data analysis

Data were analyzed by ANOVA with correction for multiple comparisons or by paired or unpaired Student’s t-test, as indicated.

Results

Glucose- and IL-1β-induced islet cell apoptosis and insulin release require L-type Ca2+ influx and activation of ERK

We first confirmed glucose- and IL-1β-induced islet cells apoptosis by using TUNEL staining and caspase-3 activity assay analysis. Rat pancreatic islets were cultured for 4 days in RPMI 1640 medium containing 11.1 and 33.3 mM glucose, then treated for 3 h with 140 ng/ml IL-1β and 300 mM/l mannitol (which was reported to induce programmed cell death [22]), respectively. As shown in Fig. 1a, a higher concentration of glucose (33.3 mM) was more effective than 11.1 mM glucose on islet cells TUNEL staining (Fig. 1a). IL-1β increased TUNEL staining in both 11.1 and 33.3 mM glucose. The proportion of TUNEL-positive islet cells treated with IL-1β in 11.1 and 33.3 mM glucose was approximately 28% and 48%, respectively (Fig. 1a). The positive control mannitol produced an increase in TUNEL-stained cells that was approximately 40% of total cells (Fig. 1a). The response of islet cells to glucose and IL-1β was investigated to determine a role for glucose and IL-1β in apoptotic enzyme activation. Caspase-3 activity in islet cells was activated by treatment of the cells with 140 ng/ml IL-1β in both 11.1 and 33.3 mM glucose (Fig. 1b). The presence of 33.3 mM glucose significantly increased the IL-1β-induced caspase-3 activity compared with 11.1 mM glucose (Fig. 1b). 11.1 mM glucose alone did not significantly affect caspase-3 activity, which was consistent with the results of TUNEL staining. In addition, the positive control mannitol induced increase in islets cell caspase-3 activity. Above results indicated that both glucose and IL-1β induced islet cells apoptosis.

Fig. 1
figure 1

Glucose- and IL-1β induced islet cell apoptosis. Rat islets were cultured on six-well dishes for 4 days in 11.1 and 33.3 mM glucose, then treated with IL-1β (140 ng/ml) or mannitol (300 mM/L) for 3 h. (a) TUNEL staining of islet cells. Percent apoptosis was determined by TUNEL staining assay and expressed as means ± SE of the percentage of apoptosis cell number. (b) Relative fluorescence for caspase-3 activity was determined using DEVD-AFC as a substrate (asp-glu-val-asp linked to the fluorochrome 7 amino 4 trifluoromethyl coumarin). Fluorescence value for cell control was assigned a value of one and other values calculated as relative fold increase/decrease to cell control. * P < 0.01 and ** P < 0.01 versus control islet cells at 11.1 and 33.3 mM glucose alone, respectively

Full size image

We next studied the functional role of glucose- and IL-1β-induced, calcium-mediated, ERK activation in rat islets; specifically, we investigated the ability of Ca2+ channel blocker nimodipine or ERK inhibitor PD98059 to protect β-cells from glucose- and IL-1β-induced apoptosis and impaired function. Exposure of rat islets to 11.1 and 33.3 mM glucose in the presence of 10 μM nimodipine as well as 10 μM PD98059 protected the β-cells from 11.1 and 33.3 mM glucose- and IL-1β-induced apoptosis. However, in 11.1 mM glucose, this protective effection is not statistically significant compared with cells in 33.3 mM glucose (Fig. 2). Furthermore, 10 μM nimodipine and 10 μM PD98059 induced a 3.6- and 4.3-fold increase of insulin content, respectively, in islets cultured at 33.3 mM glucose (P < 0.01). A total of 10 μM of each compound had no significant effect at 11.1 mM glucose (Fig. 2).

Fig. 2
figure 2

Glucose- and IL-1β-induced β-cell apoptosis and impaired function require L-type Ca2+ influx and ERK activation. Rat islets were cultured on six-well dishes for 4 days in 11.1 and 33.3 mM glucose. To some group of islets, 140 ng/ml IL-1β, 10 μM nimodipine, or 10 μM PD98059 were added. (a) The number of cells with intermediate apoptosis (propidium iodide uptake) was determined using flow cytometry. A total of 10,000 cells were counted in each group. Results are means ± SE of the percentage of apoptopsis cell number. (b) Stimulated insulin secretion during successive 1-h incubation at 3.3 (basal) and 16.7 (stimulated) mM glucose after the 4-day culture period. Results are means ± SE of the insulin secreted per islet relative to control incubations at 11.1 mM glucose alone. In each experiment, the data were collected from three plates per treatment. #  P < 0.01 versus control islets at 11.1 mM glucose alone; * P < 0.01 versus islets exposed to IL-1β alone; ** P < 0.01 versus nimodipine or/and PD98059-treated islets

Full size image

Glucose and IL-1β activate ERK in rat islets

In order to investigate whether ERK is activated by glucose and IL-1β, we first characterized the dose–response relationship between IL-1β exposure and ERK activation in cultured rat islets. By immunoblot analysis using antibodies against Thr202/Tyr204-phosphorylated ERK (P-ERK), IL-1β, in a dose-dependent manner, induced an increase in the phosphorylation of ERK (Fig. 3a, upper panel). Under these conditions, the level of total ERK protein was unchanged (Fig. 3a, lower panel). From this experiment, it was found that an IL-1β concentration of 140–180 ng/ml, giving roughly a half-maximal effect on ERK, would be appropriate for the remainder of the study. A time-course study of IL-1β-induced phosphorylation of ERK showed that ERK phosphorylation in response to IL-1β peaked at 40 min and persisted for at least 2 h (Fig. 3b). As seen in Fig. 3c, exposure of cultured rat islets to 33.3 mM glucose and 140 ng/ml IL-1β for 1 h enhanced the phosphorylation of ERK, whereas total ERK remained unchanged.

Fig. 3
figure 3

IL-1β and glucose activate ERK in cultured rat islets. (a) Islets were exposed to the indicated concentrations of IL-1β for 1 h. After lysis of Islets, the activation state of JNK was examined by immunoblot analysis using antibodies to Thr202/Tyr204-phosphorylated ERK (P-ERK) and total ERK (ERK). (b) Islets were exposed to IL-1β (140 ng/ml) for the indicated time periods. The activation state of ERK was examined by immunoblot analysis. (c) Rat islets were cultured in 11.1 and 33.3 mM glucose for 4 days, then treated with or without IL-1 (140 ng/ml) for 1 h. Islet extracts were analyzed by immunoblot for P-ERK and total ERK. Each Blot shown is representative of three experiments. The proportions below each blot indicate the blot density compared with the maximum blot density in each group. IB, immunoblot

Full size image

Calcium channel blockade reduces IL-1β-induced ERK activation

In order to determine whether Ca2+ plays a role in ERK activation by IL-1β, we first examined whether the L- or T-type plasma membrane VGCCs Ca2+ channel blackade affects IL-1β-stimulated ERK activity in cultured rat islets. Nimodipine was used to block L-type channels and mibefradil to block T-type channels. The IL-1β-induced increase in activation of ERK by a 1-h exposure to IL-1β in culture medium containing 11.1 or 33.3 mM glucose was significantly inhibited by each Ca2+ channel blockade (Fig. 4a and b, upper panels). It was found that nimodipine and mibefradil were equally effective in inhibiting IL-1β-stimulated activation of ERK, suggesting that Ca2+ entering through both L and T-type channels contribute to IL-1β activation of ERK.

Fig. 4
figure 4

Nimodipine and mibefradil inhibit IL-1β-induced activation of ERK. Rat islets were incubated in the presence of 5 mM glucose overnight. Medium was then changed to a new one containing either 11.1 or 33.3 mM glucose in the presence or absence of IL-1β with or without 10 μM nimodipine (a) or 1 μM mibefradil (b). After 1 h stimulation, the activation state of ERK in whole-cell lysates was examined by immunoblot analysis. The proportions below each blot indicate the blot density compared with the maximum blot density in each group. Each experiment was repeated at least three times. IB, immunoblot

Full size image

Requirement of calcium for IL-1β stimulation of ERK activation

Since one effect of IL-1 stimulation, namely an increase in intracellular calcium, was happened in almost all the cells, we asked if this calcium response could be the restricting factor leading to the differential effects of IL-1 on ERK activation. In order to more directly examine the effect of [Ca2+]i on IL-1β activation of ERK, the effect of Ca2+ ionophore ionomycin was determined first. Preliminary dose–response experiments were conducted with ionomycin to determine the optimal dose for stimulation of Ca2+ flux, and this was found to be 2 μM in cultured rat islets. Ionomycin-induced ERK activation at a similar or less level than that of IL-1β alone, but the combination of ionomycin and IL-1β was able to strongly stimulate ERK activation (Fig. 5a). The increase in ERK activation was quantified to be about 300% (P < 0.01) above the level induced by IL-1β alone in culture medium containing 33.3 mM glucose. In order to determine the correlation between ionomycin-induced increases in [Ca2+]i and IL-1β-induced ERK activation, we measured [Ca2+]i after treatment with IL-1β in the presence or absence of ionomycin. Exposure to both IL-1β or ionomycin alone for 20–30 min caused a significant change in [Ca2+]i (Fig. 5b) compared with distilled water-treated islets (data not shown). Furthermore, combined with IL-1β, ionomycin caused a much higher rise in [Ca2+]i than that of IL-1β alone (Fig. 5b). Therefore, the ionomycin-induced increases in [Ca2+]i directly correlated with the augmenting effect of it on IL-1β-induced ERK activation.

Fig. 5
figure 5

Effects of ionomycin, thapsigargin, BAPTA/AM, and W7 on IL-1β-induced activation of ERK. (a) Rat islets were incubated in the presence of 5 mM glucose overnight. Medium was then changed to a new one containing either 11.1 or 33.3 mM glucose in the presence or absence of IL-1β (140 ng/ml) with or without 2 μM ionomycin for a 1-h stimulation. The activation state of ERK in whole-cell lysates was examined by immunoblot analysis. (b) Rat islets were incubated as described in A and stimulated with 2 μM ionomycin and IL-1β alone and combined. [Ca2+]i was measured by fura-2 method. Data are expressed as means ± SE of n = 10 for each condition. * P < 0.01 versus IL-1β alone. (b, d, e), Rat islets were incubated in the presence of 5 mM glucose overnight. Medium was then changed to a new one containing either 11.1 or 33.3 mM glucose in the presence or absence of IL-1β (140 ng/ml) with or without 1 μM thapsigargin, 3 μM BAPTA/AM, or 25 μM W7, respectively. After 1-h stimulation, the activation state of ERK in whole-cell lysates was examined by immunoblot analysis. The proportions below each blot indicate the blot density compared with the maximum blot density in each group. Blots shown are representative of three experiments. IB, immunoblot

Full size image

Because entry of Ca2+ via VGCCs into β-cells is accompanied by Ca2+ release from the ER (13), we next looked at whether Ca2+ release from the ER modulates ERK activation. Thapsigargin (1 μM), an inhibitor of sarco-(endo)plasmic reticulum Ca2+ ATPase leading to impairment of ER Ca2+ uptake and thus an increase in [Ca2+]i, in itself stimulated ERK activation in cultured rat islets (Fig. 5c). Also, the combination of IL-1β and thapsigargin resulted in higher ERK activation as compared with each treatment alone. Notably, amplitudes of IL-1β-induced ERK activation in culture medium containing 33.3 mM glucose are always at a higher level than those of 11.1 mM glucose, suggesting that glucose alone is necessary but not sufficient for ERK activation. Taken together, these observations indicate that most stimuli that increase [Ca2+]i cause higher ERK activation in response to IL-1β.

In the next series of experiments, we asked if various inhibitors of intracellular Ca2+ influx effected IL-1β activation of ERK in rat islets. As shown in Fig. 5d, a chelator of intracellular free Ca2+ (BAPTA-AM) reduced IL-1β-stimulated ERK activation in rat islets. Similarly, an inhibitor of calmodulin (W7) diminished IL-1β-induced phosphorylation of ERK (Fig. 5e).

Requirement of calcium and ERK activation for glucose-induced IL-1β release

Previous studies have shown that in vitro exposure of human islets to high glucose levels resulted in increased production and release of IL-1β [23]. We first determined whether rat islets exposure to glucose in vitro induce IL-1β release, and the time and dose–response relationship with glucose. Rat islets were exposed to elevated glucose concentrations (5.5, 11.1, and 33.3 mM) for 12, 24, 48, and 72 h. Measurement of IL-1β released in the culture medium revealed a 80% increase in islets cultured at 33.3 mM for 48 h incubation compared with islets in 5.5 mM glucose (Fig. 5a). The time-course effect of 33.3 mM glucose on IL-1β secretion became significant only after 24 h of exposure to high glucose, and persisted after 72 h. The present results suggest that high concentrations of glucose induce IL-1β production and secretion in rat islets and implicate locally produced IL-1β is an important mediator of glucotoxicity to islet cells.

Next, the role of Ca2+ channels and ERK in glucose-induced IL-1β release was tested. IL-1β secretion to the culture medium after a 48-h culture period at 11.1 and 33.3 mM were measured. 33.3 mM glucose increased 3.2-fold of IL-1β release compared with 11.1 mM glucose and was prevented by 10 μM nimodipine, 1 μM mibefradil, and 10 μM PD98059 (Fig. 6b), although this prevention reached statistical significance only at 33.3 mM glucose.

Fig. 6
figure 6

(a) Glucose induces IL-1β release in rat islet cells. Secretion of IL-1β from rat islets cultured in 5.5, 11.1, or 33.3 mM glucose for 12, 24, 48, and 72 h. Data are represented as mean ± SE of five experiments. * P < 0.01 versus islets cultured in 5.5 mM glucose. (b) Glucose-induced IL-1β release requires L- or V-type Ca2+ influx and ERK activation. Secretion of rat islets cultured for 48 h in 11.1 or 33.3 mM glucose alone or in the presence of 10 μM nimodipine, 1 μM mibefradil, or 10 μM PD98059. Each bar represents the means ± SE of three separate experiments # P < 0.01 versus control islets at 11.1 mM glucose alone; * P < 0.01 versus nimodipine, mibefradil or PD98059-treated islets

Full size image

Discussion

IL-1β and glucose may be involved in the pathogenesis of type 1 and 2 diabetes by causing cellular dysfunction and apoptosis of pancreatic β-cells. IL-1β activates ERK in β-cells and signaling via this pathway is critically involved in mediating IL-1β-induced apoptosis [11]. However, the cellular regulatory mechanisms contributing to IL-1β-induced ERK activation in β-cells are not well understood.

Direct as well as indirect evidence has pointed toward Ca2+ as an important determinant of β-cell apoptosis. In the present study, we found that both glucose- and IL-1β-induced apoptosis and impaired function were dependent on Ca2+ influx and ERK activation. The similarity of the effects induced by the nutrient glucose and the cytokine IL-1β is explained by the ability of glucose to induce β-cell production of IL-1β. Surprisingly, the L- and V-type Ca2+ channel blocker nimodipine, mibefradil, and the ERK inhibitor PD98059 also prevented glucose-induced IL-1β release, suggesting that glucose-induced Ca2+ influx is critical for the stimulus-secretion coupling. It follows that the pathway that leads to islet production of IL-1β is also used by IL-1β itself to induce its toxic effects. That glucose-induced β-cell release of IL-1β is dependent on ERK is in line with another report [24], showing that ERK activation is required for virus-induced IL-1β expression in macrophages.

We next detected phosphorylation of ERK by glucose and IL-1β. It was found that in rat islets exposed to IL-1β, ERK activation can be stimulated in a short incubation time and is still detectable after 24 h [25]. Moreover, PD98059 inhibited ERK phosphorylation not only in the short-time experiment but also apoptosis and impaired function in 4-day incubations. Therefore, ERK is also a mediator of the long-term effects of glucose and IL-1β. Furthermore, we show that ERK mediates the deleterious effects of IL-1β on β-cell function and is required for IL-1β-induced apoptosis. Thus, ERK may be a key regulator of β-cell function and apoptosis.

Furthermore, we investigated the role of Ca2+ in IL-1β activation of ERK. The results showed that IL-1β-induced detectable net increases in [Ca2+]i in rat islets. By using Ca2+ channel blockers, Ca2+ ionophore, and agents that increase or decrease [Ca2+]i, we provide evidence to suggest a role for Ca2+ in controlling IL-1β activation of ERK in rat islets. The molecular mechanism may involve activation of CaMKs by increases in [Ca2+]i. Also, the findings obtained using the sarco(endo)plasmic reticulum Ca2+ ATPase blocker thapsigargin suggest that Ca2+ release from internal stores posses the capability to activate ERK and augment IL-1β activation of ERK, since thapsigargin is a known activator of ER stress.

In conclusion, this study suggests that in rat islets, glucose- and IL-1β-induced β-cell secretory dysfunction and apoptosis are Ca2+ influx and ERK dependent. Even though the present findings were mainly obtained by a pharmacological approach, these findings should be confirmed by molecular approaches. Further detailed analysis of targets and regulators of Ca2+ and ERK signaling in the β-cell should reveal novel therapeutic options for the management and treatment of diabetes.