Ask1 Gene Deletion Blocks Maternal Diabetes–Induced Endoplasmic Reticulum Stress in the Developing Embryo by Disrupting the Unfolded Protein Response Signalosome
Article Figures & Tables
Figures
- Figure 1
Ask1 deletion diminishes maternal diabetes–induced UPR sensor activation. A: Blood glucose levels at E8.75, insulin levels at E8.75, and litter size (n = 5 means five dams per group). B: Levels of p-IRE1α. C: Levels of p-PERK. D: Levels of p-eIF2α. In B–D, representative lanes were from one embryo with bar graphs showing means ± SEM of three lanes of three embryos (n = 3), each from a separate dam. B–D were separate gels of different embryonic samples. Within each panel, the membrane was probed first for the phosphorylated protein and sequentially stripped for probing total protein, ASK1, and β-actin. This is the case for all figures. *Significant differences (P < 0.05) compared with the other three groups. DM, diabetic; ND, nondiabetic. B–D were separate gels and were reprobed for separate proteins in each panel.
- Figure 2
Ask1 deletion blocked maternal diabetes–induced ER stress markers and ER chaperone gene expression. Protein levels of BiP (A), calnexin (B), and CHOP (C). D: mRNA levels of ER chaperone genes (n = 5). A–C: Experiments were repeated three times using embryos of three different dams (n = 3) per group. *Significant differences compared with the other three groups. DM, diabetic; ND, nondiabetic. A–C were separate gels and were reprobed for separate proteins in each panel.
- Figure 3
Maternal diabetes enhances the formation of the IRE1α-ASK1-TRAF2 complex, and Ask1 deletion blocks mitochondrial translocation of Bax and Bak. A: Representative images of IP using an ASK1 antibody and an anti–p-ASK1 antibody. In the bar graphs, levels of p-IRE1α, TRAF2, and p-ASK1 were assessed in ASK1 or p-ASK1 immunoprecipitates and were normalized by respective levels in 10% input. Normal rabbit IgG was used as control. B: Protein levels of Bax and Bak in mitochondrial extracts were normalized by the mitochondrial marker prohibitin. Total Bax and Bak levels in whole tissue lysates were also determined. A and B: Embryos from one litter per group were used per IP run or mitochondria extraction, and three litters (n = 3) per group were used. *Significant differences compared with the other groups. DM, diabetic; ND, nondiabetic; WB, Western blotting. A and B were separate gels and were reprobed for separate proteins in each panel.
- Figure 4
ASK1 is required for the RNase activity of IRE1α. A: XBP1 mRNA splicing was detected in E8.75 embryos by reverse transcription and subsequent PCR. n = 2 mean two embryos from separate dams per group. Although the levels of cleaved XBP1 PCR product could be quantified, quantitative data might not be necessary because XBP1 cleavage was only detected in WT embryos of the diabetic group. B: Levels of miR-17, miR-125b, miR-96, and miR-34a. Experiments were repeated three times using embryos of three different dams (n = 3) per group. *Significant differences compared with the other three groups. DM, diabetic; ND, nondiabetic.
- Figure 5
Ask1 deletion blocks maternal diabetes–induced Txnip upregulation and caspase 2 cleavage. Levels of mRNA and protein levels of Txnip (A and B) and levels of mRNA and protein levels of caspase 2 (C and D). Cleaved caspase 2 was indicated in D. Experiments were repeated three times using embryos of three different dams (n = 3) per group. *Significant differences compared with the other three groups. DM, diabetic; ND, nondiabetic. B and D were separate gels and were reprobed for separate proteins in each panel.
- Figure 6
Maternal diabetes alters ASK1 and Txnip association with Trx in diabetic embryopathy. Representative images of IP using an ASK1 antibody. In the bar graphs, levels of p-IRE1α, TRAF2, and p-ASK1 were assessed in ASK1 immunoprecipitates and were normalized by respective levels in 10% input. Normal rabbit IgG was used as control. Embryos from one litter per group were used per IP run, and three litters (n = 3) per group were used. *Significant differences compared with the other group. DM, diabetic; ND, nondiabetic.
- Figure 7
ASK1 knockdown blocks tunicamycin-induced IRE1α activation, ER stress, and apoptosis in C17.2 neural stem cells. A: mRNA and protein levels of ASK1. B: Protein levels of p-IRE1α, IRE1α, BiP, CHOP, ASK1, and β-actin. C: miR expression. D: Txnip mRNA levels. E: XBP1 mRNA splicing. F: Levels of total and cleaved caspase 2 and 3. A–C, E, and F: Experiments were repeated three times (n = 3). G: Representative images of the TUNEL assay. Apoptotic cells were labeled as red and all cell nuclei were stained as blue. Experiments were repeated three times (n = 3), and the quantification of the data is shown in the bar graph. Tm, tunicamycin. *Significant difference compared with other groups (P < 0.01).
- Figure 8
ASK1 knockdown suppresses high glucose–induced IRE1α activation and ER stress. 4-PBA blocks high glucose–increased miR and Txnip expression in C17.2 neural stem cells. A: Protein levels of p-IRE1α, p-PERK, p-eIF2α, CHOP, ASK1, and β-actin. B: Levels of miRs. C: Levels of Txnip mRNA and protein. A–C: Experiments were repeated three times (n = 3). *Significant difference compared with other groups. D: A schematic diagram shows that ASK1 plays an essential role in the assembly of the IRE1α signalosome, which leads to prolonged UPR, severe ER stress, and apoptosis. Maternal diabetes activates ASK1 by oxidizing its inhibitor, Trx, and simultaneously removes the ER luminal inhibition of IRE1α. Maternal diabetes induces the formation of IRE1α-TRAF2-ASK1 complexes, which is indispensable for the proapoptotic kinase and the RNase activities of IRE1α. Ask1 deletion disrupts the IRE1α-TRAF2-ASK1 complexes and thus blocks the proapoptotic kinase and the RNase activities of IRE1α. Disruption of the IRE1α signalosome also inhibits PERK activation. Maternal diabetes increases Txnip expression, a downstream effector of the IRE1α RNase activity. Txnip sequesters thioredoxin from ASK1 and thus enhances ASK1 activation.
In this Issue
