Golgi Stress

[Temika]

The Golgi complex is a central organelle of the secretory pathway where sorting and processing of cargo occurs. While Golgi structure is important for the efficient processing of secretory cargo, the unusual organization suggests additional potential functions. The Golgi is disassembled after various cellular stresses, and we hypothesize that Golgi disassembly activates a stress signaling pathway. This pathway would function to correct the stress if possible, with irreparable stress resulting in apoptosis. Neurons appear to be particularly sensitive to Golgi stress; early disassembly of the organelle correlates with many neurodegenerative diseases. Here, Golgi stress and potential signaling pathways to the nucleus are reviewed.

Figure 1. Golgi structure in life, stress and death. (A) Golgi morphology in a typical mammalian cell, with the key structural players shown in the inset. For simplicity, individual golgins and GRASPs are not indicated. (B) Golgi stress due to cargo load or size, ionic imbalance, infection with intracellular pathogens, or perturbation of glycosylation or the cytoskeleton results in structural alterations that can signal to the nucleus to help repair the stress. (1) Dephosphorylation of TFE3 and tranlocation to the nucleus results in transcription of genes with a GASE, including some glycosyltransferases and trafficking components. (2) Activation of local caspase-2 cleaves select golgins, and fragments enter the nucleus to perform an unknown function. (3) Phosphorylation of GRASPs and golgins or their cleavage can result in a more complete disassembly of the Golgi, although the consequences for signaling to the nucleus are unknown. (C) With irreparable stress, the Golgi is completely disassembled as the cell undergoes apoptosis.

Copyright 2015 Machamer. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Cytotoxic activities of several Golgi-dispersing compounds including AMF-26/M-COPA, brefeldin A and golgicide A have previously been shown to induce autophagy or apoptosis. Here, we demonstrate that these Golgi disruptors also trigger ferroptosis, a non-apoptotic form of cell death characterized by iron-dependent oxidative degradation of lipids. Inhibitors of ferroptosis not only counteract cell death, but they also protect from Golgi dispersal and inhibition of protein secretion in response to several Golgi stress agents. Furthermore, the application of sublethal doses of ferroptosis-inducers such as erastin and sorafenib, low cystine growth conditions, or genetic knockdown of SLC7A11 and GPX4 all similarly protect cells from Golgi stress and lead to modulation of ACSL4, SLC7A5, SLC7A11 or GPX4 levels. Collectively, this study suggests a previously unrecognized function of the Golgi apparatus, which involves cellular redox control and prevents ferroptotic cell death.

Multiple reports have demonstrated the ability of brefeldin A (BFA) to induce apoptosis in various cancer cell lines independently of their p53 status15,16,17,18,19. Similar to BFA, golgicide A (GCA), and AMF-26 (also called M-COPA) are Golgi disruptors and reversible inhibitors of ARF1-GBF1 with a mode of action comparable to BFA20,21,22,23. However, a refined picture of the cell death programs triggered downstream of these Golgi stress-inducing compounds has not been elucidated. In addition, it is unknown whether BFA can activate alternative cell death mechanisms besides apoptosis and autophagy24.

Here, we find that in multiple human cell lines Golgi-dispersing agents including BFA, GCA, AMF-26 or AG1478/tyrphostin induce ferroptosis. Accumulation of lipid peroxides, a reduction in the intracellular glutathione pool and changes in expression levels of several ferroptosis signaling components are observed following Golgi stress. Furthermore, antioxidants, iron chelators, and reactive oxygen species (ROS) scavengers as well as overexpression of glutathione S-transferase alpha 1 (GSTA1), SLC7A11 and GPX4, or ACSL4 knockdown protect cells from Golgi stress-mediated cell death. Notably, BFA-induced Golgi dispersal, suppression of protein secretion, endoplasmic reticulum (ER) stress or DNA damage is prevented by ferroptosis inhibitor co-treatment suggesting that the control of lipid ROS formation is critical for secretory pathway homeostasis. On the other hand, overexpression of the Golgi-associated small GTPase ADP ribosylation factor 1 (ARF1) is sufficient to counteract BFA-induced lipid peroxide formation. Unexpectedly, similar to ferroptosis inhibitors, several ferroptosis inducers such as erastin or sorafenib, used at nontoxic concentrations unable to elicit discernable lipid peroxidation in cells, prevent Golgi stress-induced dispersal and lethality, which is dependent on the transsulfuration pathway. Further, shRNA-mediated knockdown of SLC7A11 or GPX4 results in enhanced viability upon BFA treatment, which might be caused by concomitant ACSL4 downregulation and by reduced autophagy levels in these cells.

We next analyzed by immunofluorescence microscopy Golgi morphology of cells exposed to BFA, GSH, Fer-1 or to the combination of BFA with either GSH or Fer-1. Immunofluorescence staining for the cis-Golgi marker GM130 revealed that BFA-induced Golgi dispersal was substantially diminished upon simultaneous GSH exposure leading to complete restoration of the measured Golgi area to baseline levels of untreated cells (Fig. 3a). Akin to the results obtained with GSH, a striking rescue of Golgi scattering in cells co-treated with BFA and Fer-1 was detected compared to BFA-only treatment (Fig. 3b). Furthermore, co-treatment of BFA with GSH and Fer-1, respectively, improved secretion of a reporter protein (Gaussia luciferase, Gluc) relative to BFA-only treatment (Fig. 3c, d). Interestingly, Fer-1 by itself appeared to promote protein secretion (Fig. 3d). Together, these data not only demonstrate a key role for ferroptosis in governing Golgi stress-triggered cell death, but also suggest that reduced accumulation of lipid peroxides rectifies Golgi dispersal as well as protein secretion in response to AMF-26, BFA or GCA.

Overexpression of SLC7A11 or GPX4-cyto was previously reported to confer protection to erastin-induced ferroptosis6,27. In line with this and our own results from above, we found that stable overexpression of epitope-tagged SLC7A11 (Supplementary Fig. 3a) or GPX4-mito (Supplementary Fig. 3b) made HeLa cells more resistant to Golgi stressors. Moreover, treatment of HeLa cells with BFA impaired in vitro colony formation, an effect which was attenuated when GPX4 was overexpressed (Supplementary Fig. 3c). Recently, ACSL4 was shown to be essential for ferroptotic cell death as its knockdown provided marked resistance to ferroptosis-inducing agents13,14,42. Recapitulating these effects, we observed that stable ACSL4 HeLa knockdown cell lines became resistant to BFA treatment (Fig. 4a). Conversely, ACSL4 overexpression sensitized HeLa cells to BFA treatment (Supplementary Fig. 3d). In addition to the GSH-glutaredoxin antioxidant pathway, the cellular redox environment is also controlled by the thioredoxin system, the latter one including thioredoxin reductase 1 (TXNRD1). Crosstalk and functional compensation has been observed between these two systems43,44,45. Lentiviral-mediated knockdown of TXNRD1-sensitized cells to cell death in the presence of BFA (Supplementary Fig. 3e). Similarly, co-incubation of HeLa cells with the thioredoxin inhibitor Auranofin and BFA caused a further decrease in viability relative to single BFA treatment (Supplementary Fig. 3f). This suggests that redox balance in response to Golgi stress-inducing compounds is regulated by both antioxidant systems. Despite the observation by Tang and colleagues that activation of the KEAP1-NRF2 pathway protects hepatocellular carcinoma cells against ferroptosis46, in our hands knockdown of NRF2 in HeLa cells did not show obvious alterations in their sensitivity towards BFA relative to control knockdown cells (Supplementary Fig. 3g). Intriguingly, lentiviral hairpin-mediated SLC7A11 or GPX4 depletion markedly increased cellular resistance to BFA (Fig. 4b, c). This result was unexpected, because SLC7A11 or GPX4 gain-of-function prevent BFA-induced cell death (Supplementary Fig. 3a, 3b), Gpx4 deficiency by itself was shown to entail ferroptosis31, and RNAi-mediated GPX4 or SLC7A11 knockdown sensitized to RSL3- or erastin-induced ferroptosis6,27, and hence a similar sensitization phenotype could have been expected in the presence of BFA. To confirm the previously described ferroptosis sensitization effects, we treated the SLC7A11 knockdown and control cells concurrently with either BFA or erastin. We observed the anticipated erastin sensitivity phenotype suggesting that BFA-resistance upon SLC7A11 depletion in our system is not due to hairpin-off target effects (Fig. 4b). Strikingly, when lysates of SLC7A11- or GPX4-depleted HeLa cells were analyzed, which prior to lysis were left untreated or treated with BFA or GCA, we found ACSL4 expression to be concomitantly reduced in SLC7A11 or GPX4 knockdown cells relative to shRFP control cells (Fig. 4d, e). Similar to BFA/GCA treatment, it was previously reported that erastin suppresses GPX4 protein expression47. Of note, GPX4 expression was slightly enhanced in SLC7A11 knockdown cells and became to a lesser extent downregulated in response to BFA or GCA relative to control cells (Fig. 4d). On the other hand, SLC7A11 expression was induced following GPX4 downregulation (Fig. 4e). These effects are likely to contribute to enhanced BFA/GCA resistance in the respective knockdown cells. Since loss of ACSL4 renders cells resistant to BFA-induced cell death (Fig. 4a), it is conceivable that the observed ACSL4 downregulation as a result of shRNA-induced SLC7A11 or GPX4 knockdown is epistatic to loss of SLC7A11 or GPX4. In other words, in SLC7A11- or GPX4-depleted HeLa cells BFA-induced ferroptosis progression is attenuated due to a concomitant reduction of ACSL4 levels. In a similar vein, Doll et al.13 demonstrated that lipid oxidation caused by GPX4 inhibition requires ACSL4.

c80f0f1006