Proinflammatory cytokines induce ER stress in many
Proinflammatory cytokines induce ER stress in many mammalian cell systems.32, 33, 34 Our results demonstrate that these cytokines also trigger ER stress in trophoblast JEG-3 cells. However, the severity of stress is likely low grade because only the PERK-EIF2A arm of the UPRER pathway was activated. Similar low-grade ER stress was observed in the trophoblast SynaptoRedTM C2 weight of the human placenta after pregnancy at high altitude and in the mouse placenta on hypoxic challenge. It has been suggested that the severity of ER stress induced by the proinflammatory cytokines can be cell type specific and also species specific. The mechanisms by which they activate UPR pathways in ER are unclear, but several studies suggest both direct activation and indirect activation mediated by nitric oxide or perturbation of calcium homeostasis.54, 55 A mixture of IL-1β, TNF-α, and IFN-γ induces splicing of XBP1, a downstream effector of the IRE1α arm of the UPRER pathway, and phosphorylation of Eif2A in mouse islet and MIN6 cells independent of nitric oxide production. Conversely, a combination of IL-1β and IFN-γ facilitates ER Ca2+ depletion mediated through inhibition of the sarcoplasmic reticulum Ca2+ ATPase (SERCA2B), as well as production of nitric oxide in pancreatic β cells. The activation of IRE1α by these cytokines is likely to be transient. The study by Brozzi et al found that IL-1β and IFN-γ gradually facilitate IRE1α activation, which peaks at 16 hours before decreasing. This finding may explain why only activation of PERK-EIF2A was observed in this study where the incubation time was limited to 24 hours. Finally, the IRE1α signaling pathway has been linked to cellular inflammatory response mechanisms through the JNK and NF-κB pathways, thereby possibly providing a positive feedback loop. Other stressors closely linked to the preeclampsia may also act through the same pathways, for example, high levels of maternal endothelin (ET)-1 or elevated plasma concentrations of homocysteine.57, 58, 59 We found that ET-1 down-regulates MMP-14 and -15 expression in first trimester trophoblast cells, and both ET-1 and homocysteine induce ER stress in trophoblast and other cell types.61, 62 Therefore, the finding of ER stress–regulated MMP2 expression may provide an additional mechanistic explanation for the actions of ET-1 and homocysteine in the development of pregnancy complications. These results elucidated the coexistence of both transcriptional and translational regulation of MMP-2 by ATF4 and EIF2A, respectively, under ER stress. This finding is supported by the siRNA-mediated knockdown of the ATF4 gene and GSK2606414-mediated suppression of phosphorylation of EIF2A in restoration of MMP2 transcript and protein levels, respectively (Figure 5, E and F). Interestingly, ATF4 nuclear localization was blocked by GSK2606414 treatment under ER stress. Although the mechanism behind this failure is unknown, recent studies have revealed roles for posttranslational modifications in determination of ATF4 protein stability, nuclear localization, and transcriptional activity. Nevertheless, without nuclear translocation, the inhibitory role of ATF4 in MMP2 transcription is minimal, thereby facilitating both transcription and translation of MMP-2 in the presence of GSK2606414. A combination of both transcriptional and translational regulation of MMP-2 ensures no cells invade into or towards an unfavorable environment. Finally, these results are consistent with the study by Lian et al, which indicate that decidual ER stress is increased in pregnancies complicated by FGR and preeclampsia via up-regulation of the PERK/EIF2A signaling mechanism. We recognize that use of placental bed samples collected after delivery at term was a limitation in this study. Ideally, placental bed samples from the first trimester are the most appropriate, but these are impossible to obtain for ethical and technical reasons. The use of trophoblast cell lines is also not ideal, but primary trophoblast cells demonstrate high levels of ER stress induced during the isolation procedure (H.W. Yung, unpublished data). This stress would mask the low-grade ER stress induced by the treatment with proinflammatory cytokines and confound the experiments.