Targeting Cellular Calcium Homeostasis to Prevent Cytokine-Mediated Beta Cell Death

Abstract

Pro-inflammatory cytokines are important mediators of islet inflammation, leading to beta cell death in type 1 diabetes. Although alterations in both endoplasmic reticulum (ER) and cytosolic free calcium levels are known to play a role in cytokine-mediated beta cell death, there are currently no treatments targeting cellular calcium homeostasis to combat type 1 diabetes. Here we show that modulation of cellular calcium homeostasis can mitigate cytokine- and ER stress-mediated beta cell death. The calcium modulating compounds, dantrolene and sitagliptin, both prevent cytokine and ER stress-induced activation of the pro-apoptotic calcium-dependent enzyme, calpain, and partly suppress beta cell death in INS1E cells and human primary islets. These agents are also able to restore cytokine-mediated suppression of functional ER calcium release. In addition, sitagliptin preserves function of the ER calcium pump, sarco-endoplasmic reticulum Ca²⁺-ATPase (SERCA), and decreases levels of the pro-apoptotic protein thioredoxin-interacting protein (TXNIP). Supporting the role of TXNIP in cytokine-mediated cell death, knock down of TXNIP in INS1-E cells prevents cytokine-mediated beta cell death. Our findings demonstrate that modulation of dynamic cellular calcium homeostasis and TXNIP suppression present viable pharmacologic targets to prevent cytokine-mediated beta cell loss in diabetes.

Figure 1

Dantrolene and Sitagliptin Protect INS-1E Cells From Cytokine and ER Stress Induced Cell Death. (a,b) INS-1E cells were pre-treated with 100 nM dantrolene, 10 μM pioglitazone, 10 μM verapamil, or 200 nM sitagliptin for 24 hours then stressed for 24 hours with cytokine cocktail (IL-1β and IFN-γ 50 ng/mL) or thapsigargin 10 nM. Apoptotic cell death was measured via caspase 3/7 activity assay. (c–e) The ER stress markers CHOP BiP and TXNIP were measured using quantitative RT-PCR. Data are expressed as mean ± SEM from at least three independent experiments. ^#p < 0.05 compared to cytokine or TG treated cells via unpaired t-test.

Figure 2

Dantrolene and Sitagliptin Restore Caffeine Mediated ER Calcium Release in Cytokine and ER Stressed Βeta Cells. (a) INS-1E cells were treated with 100 nM dantrolene, 200 nM sitagliptin or a combination of the two drugs for 24 hours followed by treatment with 10 nM thapsigargin for 24 hours or 6–24 hours of cytokine cocktail (IL-1β and IFN-γ 50 ng/mL) treatment. ER calcium levels were monitored in INS-1E cells stably expressing the D1ER probe via FACS analysis and presented as FRET/CFP. Fluorescence was monitored continuously for two minutes in INS-1E cells loaded with Fluo-4 AM imaged at baseline and after 5 mM caffeine stimulation and single cell calcium oscillation tracings were produced as depicted in panel b. Panel c shows representative single plane images at different times during the 2 minute acquisition after the addition of caffeine. Graphical representation of the relative change in fluorescence intensity averaged over 2 minutes of image acquisition, expressed as ΔF/F₀ (ΔF is the change in fluorescence intensity of the caffeine stimulated cells relative to the basal condition F₀) is displayed in the presence (panel d) and absence of calcium in the media (panel e). The average represents values over 2 minutes in 50 cells normalized for variations in the concentration of the dye in different cells. Data are expressed as mean ± SEM from at least three independent experiments. *p < 0.05 compared to control; ^#p < 0.05 compared to cytokine or TG treated cells via one way ANOVA test.

Figure 3

Sitagliptin Maintains SERCA Activity During Cytokine Stress. INS-1E cells were pre-treated with 100 nM dantrolene or 200 nM sitagliptin for 24 hours then stressed for 24 hours with cytokine cocktail (IL-1β and IFN-γ 50 ng/mL). (a) SERCA protein levels were measured by immunoblot. Full-length blot can be seen on-line (Supplementary Figure 1). (b) SERCA2b mRNA levels were measured by real-time PCR. (c) SERCA activity levels were monitored and expressed as nmol Ca/min/mg protein. Data are expressed as mean ± SEM from at least three independent experiments. *p < 0.05 compared to control; ^#p < 0.05 compared to cytokine or TG treated cells via one way ANOVA test.

Figure 4

Dantrolene and Sitagliptin Significantly Decrease Calpain Activation in INS-1E Cells. (a) INS-1E cells were pre-treated with 100 nM dantrolene or 200 nM sitagliptin for 24 hours then stressed for 6–48 hours with cytokine cocktail (IL-1β and IFN-γ 50 ng/mL) or 10 nM thapsigargin. (a) cytosolic free calcium activity was determined in INS-1E cells via Fura-2 assay and presented as A340/A380. (b) Calpain activation was measured via immunoblot for spectrin cleavage products and (c) quantitative densitometry was performed via ImageJ software (NIH). Full-length blot can be seen on-line (Supplementary Figure 1). (d) INS-1E cells were transfected with SiRNA against a scrambled construct (SiSC) or calpain 2 (SiCAPN2) and immunoblot was performed for calpain-2 expression. (e) SiSC and SiCAPN2 transfected INS-1E cells were then exposed to cytokines (IL-1β and IFN-γ 50 ng/mL) or 10 nM thapsigargin for 24 hours and cell death was measured via caspase 3/7 activity. (f) INS-1E cellspre-treated for 24 hours with varying doses of calpain inhibitor III. Cells werethen stressed with cytokines (IL-1β and IFN-γ 50 ng/mL) or 10 nM thapsigargin for 24 hours and cell death was monitored via caspase 3/7 activity. Data are expressed asmean ± SEM from at least three independent experiments. ^#p < 0.05 compared tocytokine or TG treated cells unpaired t-test. Full-length blot can be seen on-line (Supplementary Figure 1).

Figure 5

Knock down of TXNIP in INS-1E Cells Prevents Cytokine Induced Caspase 3/7 Activity. (a) Western blot showing that TXNIP expression is decreased in tet-shTXNIP cells. Fold over control data are pooled from 5 experiments. Full-length blot can be seen on-line (Supplementary Figure 1). (b) Stable Tet-shTXNIP INS1 cell lines were treated with DMSO or doxycycline for 48 hours followed by exposure to a cytokine cocktail of either 10 ng/mL (panel b) or 50 ng/mL (panel c) IL-1β and IFN-γ for 24 hours. Caspase 3/7 activity was measured and data was presented as fold over control for each condition. Statistical analysis was done via two-tailed t tests ^#P < 0.05, n = 3–8.

Figure 6

Dantrolene and Sitagliptin do not effect glucose stimulated insulin secretion (GSIS) in INS-1E cells or human islets. (a) INS-1E cells were treated with 100 nM dantrolene or 200 nM sitagliptin for 24 hours and GSIS assay was performed. Results are expressed as ng insulin secreted normalized for total protein. (b) Human islets from 3 separate donors were treated for 24 hours with 10 μM dantrolene or 200 nM sitagliptin and GSIS assay was performed. Results are expressed as ρg insulin secreted normalized to total DNA. Data are expressed as mean ± SEM. Statistical analysis to determine if drugs had an effect on insulin secretion was done via one-way ANOVA, n = 3–4.

Figure 7

The Effect of Dantrolene and Sitagliptin on Cytokine Mediated Cell Death In Human Islets. (a–h) Human islet cells from 7 different organ donors were pre-treated for 24 hours with dantrolene 10 μM or sitagliptin 200 nM then stressed with cytokines (IL-1β 50 u/mL and IFN-γ 1000 u/mL) or thapsigargin 2 μM for 48 hours. Data are presented as combined results (panel a) as well as individual results (panel b–h) for all donors. Detailed information obtained from islet distribution centers on patient age, body mass index (BMI), hemoglobin A1C (A1C), Glucose Stimulated Insulin Release Stimulation Index (SI) and cause of death are presented in panel i. Data are expressed as mean ± SEM for all figures. ^#p < 0.05 compared to cytokine or TG treated cells unpaired t-test.

Figure 8

Cytokines Lead to Altered Cellular Calcium Homeostasis, and Induction of Βeta Cell Death. In beta cells cytokines cause increased ER stress, depletion of ER calcium, prevention of functional ER calcium release and elevation of cytosolic free calcium levels. This chain of events ultimately culminates in cell death. This study has proved that drugs targeting modulation of ER and cytosolic free calcium levels can prevent the activation of cytokine and ER stresses mediated pro-apoptotic pathways and decrease beta cell death.