Roughly one third of all proteins are transmembrane and secretory proteins that are modified, folded and assembled in the endoplasmic reticulum (ER). When the ER-folding machinery becomes overwhelmed, misfolded polypeptides accumulate leading to ER stress, with toxic consequences to cells. To adjust protein-folding capacity of the ER according to the need of the cell, a collection of ER-resident sensor/transducer proteins constantly monitor the protein folding status in the ER lumen and take corrective actions. Maintaining homeostasis between protein folding capacity and protein folding demand is crucial to ensure high fidelity in protein folding. The collection of conserved signaling pathways in charge of overseeing proper ER function is known as the unfolded protein response (UPR).
Since its discovery in our lab some 20 years ago, our knowledge of the UPR has vastly expanded; growing from insights obtained from the budding yeast Saccharomyces cerevisiae, where a single signal transduction pathway is in charge of responding to ER stress. The ER stress sensor/transducer, Ire1, is a kinase/endoribonuclease that lies within the ER membrane and responds to protein folding perturbations. Upon accumulation of unfolded proteins, Ire1 binds to them and, in response, oligomerizes in the plane of the ER membrane. The oligomerization reaction leads to Ire1’s trans-autophosphorylation and allosteric activation of its endoribonuclease activity. In S. cerevisiae, the transcript encoding the transcription factor Hac1 is the sole target of the Ire1’s nuclease activity. In an unconventional, cytoplasmic splicing reaction, Ire1 removes an intron from HAC1 mRNA, which is repaired by tRNA ligase, allowing the production of the Hac1 transcription factor that executes a transcriptional response to augment the ER’s capacity.
As unique as this mechanism is, it appears to be not the only way for dealing with ER stress. Our group has uncovered an alternative, pre-emptive mechanism in the fission yeast Schizosaccharomyces pombe. Rather than relying on the transcriptional regulation of several hundred genes as is the case in S. cerevisiae, S. pombe alleviates ER stress by cleaving mRNAs encoding ER client proteins, thereby lowering the load of proteins entering the organelle. This mechanism, initially discovered in Drosophila and known as regulated IRE-dependent degradation (RIDD), is the sole way employed by S. pombe to deal with ER stress. Interestingly, the mRNA encoding the most abundant ER chaperone Bip1 is also subject to this regulation in S. pombe. Paradoxically, cleavage of the Bip1 mRNA by Ire1 does not lead to its degradation but to its stabilization and a transient updraft in the levels of Bip1, which in turn help to alleviate ER stress. Thus, evolution has endowed these two different yeast species with distinct mechanisms to deal with ER stress. We are currently leveraging these two extremes to decipher the mechanistic differences in Ire1 that allow it to operate in the splicing or RIDD modes.
With the advent of multicellularity, higher eukaryotes tailored their UPR, entwining the complexities found in both S. cerevisiae and S. pombe. For instance, mammalian cells exhibit both RIDD and the unconventional splicing of the mRNA encoding XBP1 — the functional homolog of Hac1. Mammalian cells also possess two additional ER stress sensors, the kinase PERK, and the membrane-tethered transcription factor ATF6. Like RIDD, PERK activation results in a pre-emptive defense mechanism against ER stress. Active PERK phosphorylates the (alpha)-subunit of eukaryotic translation initiation factor 2, leading to a global attenuation of protein synthesis. This allows the cell to cope with upsurges in protein folding defects by down-tuning protein synthesis. On the other hand, ATF6, akin to Hac1 and XBP1, initiates a vast and comprehensive transcriptional program aimed at augmenting the folding capacity of the ER. Upon detection of ER stress, ATF6 travels in transport vesicles to the Golgi apparatus, where it undergoes regulated intramembrane proteolysis. This liberates its cytosolic transactivation domain, which then moves into the nucleus to activate its UPR target genes. Because of the comprehensive nature of the UPR, the fundamental response of the cells is to try to reinstate homeostasis by correcting the folding defects. However, if the stress persists, the cell commits to apoptosis, thereby ensuring that damaged cells are removed for the benefit of the organism.
By continuously expanding the depth of our knowledge of the UPR, we seek to understand how the UPR operates in normal cells and how it may be rewired in disease. For example, cancer cells abuse the UPR for selfish survival, exploiting the cytoprotective functions of the UPR while evading its apoptotic output. In the past years we have learned that genetic redundancies within the UPR accommodate for defects in downstream signaling and, by applying deep sequencing methods, we have started to discover the genetic vulnerabilities within the UPR of cancer cells that could be exploited in developing new therapeutics. We are beginning to understand how this fundamental signaling pathway can be co-opted by tumor cells.
Outstanding questions include the detailed understanding of the structural mechanisms governing Ire1, PERK and ATF6 activation upon sensing of unfolded proteins, which remain active areas of investigation in the lab. In addition, we are working towards a fundamental structural and dynamic understanding of the interaction of Ire1 with its cognate RNA substrates. While we know some of the sequence/structure determinants of RNA recognition in Ire1’s splicing substrates, this knowledge still evades us for RIDD substrates. Additionally, the events leading to ATF6 activation and trafficking remain unanswered. Finally, we do not understand the interplay between ER stress sensors, nor do we know much regarding interactions with other cellular pathways.