Abstract
Thapsigargin (Tg) is a naturally occurring plant compound that potently blocks the endoplasmic reticulum (ER) Ca2+ pump SERCA (sarco/endoplasmic reticulum Ca2+ ATPase). This leads to ER Ca2+ depletion, unresolved ER stress, unfolded protein response (UPR) activation, and, eventually, cell death. Understanding the biological mechanisms of how Tg affects cellular functions is important because many diseases are associated with decreased ER Ca2+ levels, UPR, and cell death. Moreover, Tg may find use in targeted cancer therapy, in the form of Tg analog-based prodrugs.
In this study, we have analyzed the effects of Tg and various Tg analogs with respect to their abilities to inhibit SERCA, deplete ER Ca2+ stores, induce the UPR, inhibit cell proliferation, and initiate cell death (paper 1). Moreover, we have explored the detailed relationship between the degree of ER Ca2+ depletion versus effects on various cellular functions (paper 2). Finally, we have delineated central molecular mechanisms and the individual roles of the major UPR components in Tg- and Tg analog-induced cell death (paper 3).
In the first paper, we found that various Tg analogs bind and inhibit SERCA with different affinity and kinetics. Overall, this translated into corresponding differences in cytotoxicity. We could detect a correlation of ER Ca2+ depleting concentrations of Tg or Tg analogs and their ability to induce sustained UPR and cell death, while an increase in cytosolic Ca2+ levels via SOCE (store-operated Ca2+ entry) was not crucial for cell death induction. Comparison of in vitro and in vivo data indicated that SERCA binding kinetics are not decisive for long-term ER Ca2+ depletion and cytotoxicity, and that incomplete inhibition of SERCA can be sufficient to drain ER Ca2+ stores, with detrimental consequences. In the second paper, we investigated in detail the correlation between the extent of ER Ca2+ depletion versus UPR induction and other cellular effects. Astonishingly, Tg/Tg analog titration experiments showed that relatively low drug concentrations could deplete the ER Ca2+ stores to virtually undetectable levels without inducing UPR, cell death or inhibiting autophagy. Only at substantially higher drug concentrations could we observe those effects. This indicates that ER Ca2+ depletion-induced effects on UPR, autophagy and cell death require either extreme global ER Ca2+ depletion or, alternatively, that Ca2+ depletion of an ER area more resistant to Ca2+ depletion than the bulk of the ER, is the key trigger. In the third paper, we focused on the death-inducing signaling pathways of Tg and two therapeutically relevant Tg analogs. In two different cancer cell lines we found cell death induction to depend on death receptor 5 (DR5), caspase-8 and a non-autophagic function of the autophagy-related protein LC3B. The ER stress sensor PERK and its downstream targets ATF4 and CHOP were essential for cell death induction, yet appeared to act in an uncoupled manner; while ATF4 and CHOP were independently required for DR5 and LC3B upregulation, PERK did not regulate DR5 and LC3B. Also surprisingly, we identified a previously unrecognized dependency of Tg-induced cell death on IRE1-mediated JNK activation via XBP1, which occurred in a cell type-specific manner. Finally, and importantly, we found that the therapeutically relevant Tg analogs induced cell death with the same molecular dependencies as observed with Tg. Together, our results have clarified important aspects related to the biological effects of ER Ca2+ depletion, and have revealed central mechanisms of UPR-induced cell death triggered by Tg and therapeutically relevant Tg analogs. This increased knowledge opens up new questions in the field of ER stress biology and can aid future developments of novel therapeutic strategies towards various human diseases.