Decoding the Proteostasis Network in Resistance to KRAS Inhibitors
Xinxin Song,Zhuan Zhou,Kenneth D. Westover
DOI: https://doi.org/10.1016/j.xinn.2023.100526
2023-01-01
The Innovation
Abstract:The Kirsten rat sarcoma (KRAS) gene, which is one of the most frequently mutated genes in human cancers, has long captured the attention of oncologists. The recent approval of KRAS G12C inhibitors (KRASi), such as sotorasib and adagrasib, has signaled a promising turn in cancer therapy. However, emerging resistance to these drugs presents a formidable challenge.1Zhao Y. Murciano-Goroff Y.R. Xue J.Y. et al.Diverse alterations associated with resistance to KRAS(G12C) inhibition.Nature. 2021; 599: 679-683Crossref PubMed Scopus (127) Google Scholar Lv et al.’s2Lv X. Lu X. Cao J. et al.Modulation of the proteostasis network promotes tumor resistance to oncogenic KRAS inhibitors.Science. 2023; 381eabn4180Crossref Scopus (3) Google Scholar publication in Science suggests that the proteostasis network plays a role in mediating diverse KRASi resistance mechanisms. Their work shows that KRAS inhibition suppresses protein quality control in the majority of cancer cells. In the remaining cells, a subset can reactivate one of the endoplasmic reticulum (ER) stress sensors, specifically the inositol-requiring enzyme 1 alpha (IRE1α) branch of the unfolded protein response (UPR). This adaptation enables the cells to overcome their reliance on oncogenic KRAS, enhancing their resistance to KRASi (Figure 1). This insight underscores the complex interplay between oncogenic signaling and protein quality control in scenarios of drug resistance. Here, we review and discuss the major findings of this study in the following sections. Many compensatory mechanisms can cause KRASi resistance.1Zhao Y. Murciano-Goroff Y.R. Xue J.Y. et al.Diverse alterations associated with resistance to KRAS(G12C) inhibition.Nature. 2021; 599: 679-683Crossref PubMed Scopus (127) Google Scholar,3Awad M.M. Liu S. Rybkin I.I. et al.Acquired Resistance to KRAS(G12C) Inhibition in Cancer.N. Engl. J. Med. 2021; 384: 2382-2393Crossref PubMed Scopus (332) Google Scholar These include the activation of pathways such as epidermal growth factor (EGF), fibroblast growth factor (FGF), aurora kinase A (AURKA), and SOS1. Additionally, new mutations can emerge in genes such as KRAS, NRAS, BRAF, EGF receptor (EGFR), and FGF receptor 2 (FGFR2). Given this diverse array of resistance mechanisms, effectively countering KRASi is a daunting challenge. Aside from identifying these mechanisms in individual patients, another challenge is that simultaneously targeting these critical pathways can lead to harmful side effects, primarily due to the cumulative toxicity from multiple inhibitors. The notion that diverse mechanisms of resistance might converge on a singular, unifying pathway, such as the IRE1α-mediated reprogramming of proteostasis in resistant tumors, is attractive. Historically, RAS mutations have been associated with the initiation of a general UPR, resulting from ER stress during the oncogenic transformation of benign cells. Several studies have highlighted the significance of IRE1α in lung cancer.4Shimomura I. Watanabe N. Yamamoto T. et al.Selective targeting of KRAS-driven lung tumorigenesis via unresolved ER stress.JCI Insight. 2021; 6e137876Crossref Scopus (1) Google Scholar,5Crowley M.J.P. Bhinder B. Markowitz G.J. et al.Tumor-intrinsic IRE1α signaling controls protective immunity in lung cancer.Nat. Commun. 2023; 14: 120Crossref PubMed Scopus (5) Google Scholar Lv et al.2Lv X. Lu X. Cao J. et al.Modulation of the proteostasis network promotes tumor resistance to oncogenic KRAS inhibitors.Science. 2023; 381eabn4180Crossref Scopus (3) Google Scholar delineated a dynamic modulation of IRE1α activity in scenarios of KRAS activation, suppression, and resistance to KRASi. They show that oncogenic KRAS activation diminishes ER stress in KRAS-mutant cells, predominantly via the oncogenic kinase-mediated phosphorylation of IRE1α (Figure 1A). Upon KRAS suppression, a disruption in proteostasis emerges, evident through pronounced protein aggregation observed in doxycycline inducible KRAS G12D and KRAS G12C models (Figure 1B). Notably, KRASi-resistant cells show adaptability by navigating through proteotoxic stress and restoring proteostasis (Figure 1C). Proteostasis is a coordination of protein synthesis, folding, quality control, and degradation. The degradation aspect involves the ubiquitin-proteasome and autophagy-lysosome systems, both of which are essential for the selective disposal of aberrant proteins. When encountering disruptions in proteostasis, cancer cells use stress sensors to preserve protein quality control. Two main mechanisms are (1) the heat shock response, helmed by heat shock factor 1 (HSF1), which oversees cytosolic proteins, and (2) the UPR, which monitors transmembrane and secretory proteins within the ER. Any imbalance in ER homeostasis activates the UPR via ER membrane sensors, such as IRE1α, activating transcription factor 6 (ATF6), and protein kinase RNA-like ER kinase (PERK). Importantly, PERK, along with kinases such as general control non-derepressible 2 (GCN2), heme-regulated inhibitor (HRI), and protein kinase R (PKR), modulates eukaryotic translation initiation factor 2α (eIF2α), thereby affecting the integrated stress response (ISR). Lv et al.2Lv X. Lu X. Cao J. et al.Modulation of the proteostasis network promotes tumor resistance to oncogenic KRAS inhibitors.Science. 2023; 381eabn4180Crossref Scopus (3) Google Scholar found that oncogenic KRAS deactivation initially does not impart proteasomal, autophagy, or lysosomal activities. Instead, it hinders the ER, cytosolic protein quality control systems, and the ISR. However, in KRASi-resistant cells, only the IRE1α/X-box binding protein 1 (XBP1) pathway is reactivated. HSF1 and ISR are unaffected, indicating that oncogenic KRAS deactivation primarily affects the IRE1α/XBP1 pathway in KRASi-resistant cancers (Figure 1B). IRE1α is a substrate of the SEL1L/HRD1 ER-associated degradation (ERAD) complex, which is composed of the E3 ubiquitin ligase HRD1 (HMG-CoA reductase degradation 1 homolog) and its adapter protein, namely, suppressor/enhancer of lin-12-like (SEL1L). Mitogen-activated protein kinase (MAPK) promotes IRE1α protein stability by leading to complex disassociation and thus inhibiting SEL1L/HRD1-mediated IRE1α ubiquitination. Sotorasib, on the other hand, promoted the interaction of IRE1α with p97/valosin-containing protein (VCP) and nuclear protein localization protein 4 (NPL4), which deliver the ubiquitinated ERAD substrates to the proteasome for degradation (Figure 1B). During ER stress, IRE1α undergoes oligomerization and trans-autophosphorylation, thereby activating its ribonuclease (RNase) domain. This transformation leads to the conversion of unspliced XBP1 mRNA into spliced XBP1 (XBP1s), a potent transcriptional enhancer for UPR proteins. This mechanism is vital for proteostasis in KRASi-resistant cancer cells. The kinase activity of IRE1α is fundamental for its RNase activation. Neither the kinase-dead IRE1αK599A nor the RNase-deficient IRE1αK907A could rescue IRE1α depletion-induced phenotypes in KRASi-resistant cells, emphasizing IRE1α’s reliance on its intrinsic RNase activity. Another IRE1α variant, IRE1αD2M, which cannot sense the ER stress, was still functional in KRASi-resistant cells, indicating that IRE1α reactivation is ER stress independent. Experiments with the kinase-dead variant, IRE1αK599A, also enabled the identification of IRE1α phosphorylation sites by extracellular signal-regulated kinase (ERK) at specific sites, such as T973, S525, S529, and S549. Collectively, an array of receptor tyrosine kinases (RTKs), directing ERK and AKT activity in various KRASi-resistant tumors, consistently converge in regulating IRE1α (Figure 1C). Genetic interventions, such as IRE1α deficiency or XBP1 deletion, accentuate the efficacy of sotorasib, but are not tractable therapeutic strategies. ORIN1001, a promising new IRE1α RNase inhibitor, has exhibited safety in early-phase trials in breast cancer (NCT03950570). The extension of clinical trials to encompass KRAS G12C mutant cancers is warranted. A key observation of study of Lv et al.2Lv X. Lu X. Cao J. et al.Modulation of the proteostasis network promotes tumor resistance to oncogenic KRAS inhibitors.Science. 2023; 381eabn4180Crossref Scopus (3) Google Scholar was experimental support for a direct interaction between IRE1α and ORIN1001 at valine 918, supporting ORIN1001’s on-target effects. This is important given that structures of ORIN1001 bound to IRE1α have not been reported. Indeed, ORIN1001 augments responses in KRAS-mutant lung or pancreatic cancer models when combined with RAS pathway-directed approaches such as sotorasib or trametinib (Figure 1D). A number of important questions must be answered before UPR-targeted approaches can move forward. Foremost, the generalizability of these results is not known. The study of Lv et al.2Lv X. Lu X. Cao J. et al.Modulation of the proteostasis network promotes tumor resistance to oncogenic KRAS inhibitors.Science. 2023; 381eabn4180Crossref Scopus (3) Google Scholar was focused primarily on two pancreatic cancer cell lines to establish the role of UPR in mediating KRASi resistance. As validation, they evaluated five NSCLC PDX models, but sotorasib failed to effectively suppress MAPK signaling in 2 of 5 patient-derived xenograft models. Thus, proteostasis was largely unaffected in these two models. It will be critical to establish the frequency at which these mechanisms occur in models of various cancer types, and in patients, to know how widely this intervention might be applied. Second, in studies of clinical resistance to KRASi, a large proportion of patients developed secondary resistance mutations.3Awad M.M. Liu S. Rybkin I.I. et al.Acquired Resistance to KRAS(G12C) Inhibition in Cancer.N. Engl. J. Med. 2021; 384: 2382-2393Crossref PubMed Scopus (332) Google Scholar It is unknown if UPR plays a role in this scenario, although it seems unlikely. Finally, it is unknown if UPR mechanisms are particular to KRASi or if they may also apply when inhibiting other KRAS mutants, such as G12D, for which inhibitors such as MRTX1133 are currently in trials, or for pan-KRAS inhibitors, such as RMC-6236, also in clinical testing. Despite these potential limitations, this groundbreaking study suggests a hopeful trajectory: by targeting the IRE1α protein, proteostasis can be destabilized in resistant cells, thereby enhancing their susceptibility to oncogenic KRASi. Moving forward, it will be crucial to thoroughly investigate the extent to which KRAS-driven cancers rely on IRE1α-centric resistance mechanisms as they develop resistance to KRASi. Additionally, evaluating whether specific parameters, such as the phosphorylation of IRE1α at S549, S525, S529, and T973, can serve as reliable biomarkers for identifying and monitoring KRASi resistance in clinical samples will be important. Indeed, progress in precision medicine for KRAS-driven cancers will likely require an assessment of resistance pathways in individual patients and the formulation of responsive strategies. The discoveries in this study could herald the beginning of a transformative phase in the treatment of KRAS-driven cancers. We thank Dave Primm (Department of Surgery, University of Texas Southwestern Medical Center) for his critical reading of the manuscript. X.S. was supported by an American Cancer Society grant (IRG-21-142-16) and a National Institutes of Health grant (P30CA142543) to the Simmons Comprehensive Cancer Center and is supported by a startup fund by the Department of Surgery at the University of Texas Southwestern Medical Center. K.D.W. is supported by grants from the National Institutes of Health (R01CA244341) and the Cancer Prevention and Research Institute of Texas (RP220145). K.D.W. serves on advisory boards for Vibliome Therapeutics, Sanofi, and Reactive Biosciences and has equity in Vibliome Therapeutics.