Ferreting Out Ferroptosis: Extending the Mechanism of Action of RSL3 to the Selenoproteome in Colorectal Cancer

Abstract

Although colorectal cancer (CRC) is the second leading cause of cancer-related mortality in the United States, there has been limited progress in recent years to identify new therapeutic options. However, cancer cells have been shown to be sensitive to ferroptosis, an iron-dependent lipid peroxide-induced form of cell death. In this issue of Cancer Research DeAngelo, Zhao, and colleagues aim to better understand the mechanisms underlying ferroptosis in CRC. However, using the ferroptosis-inducing small molecule, RSL3, they observed effects in CRC cells independent of RSL3’s presumed target, glutathione peroxidase 4 (GPx4). Investigating further, they find RSL3 inhibits multiple antioxidant proteins in the peroxidase and selenoprotein families to more broadly impact reactive oxygen species (ROS) and lipid peroxidation than previously assumed. Interestingly, the rare selenoprotein family, which includes GPx4, can be broadly reduced by inhibiting their specialized translational machinery. The authors next show that loss of AlkBH8, a tRNA methyltransferase responsible for modifying the selenocysteine-specific tRNA, broadly decreases selenoprotein activity and induces ferroptosis in CRC. This identifies the selenoproteome as a therapeutic target in CRC via induction of ROS, lipid peroxidation, and ferroptosis and adds to a growing body of literature on the potential utility of pro-oxidant mechanisms in cancer therapy.

Colorectal cancer (CRC) is the second leading cause of cancer-related mortality in the United States (1). Despite significant advancements in our understanding of CRC over the past years, therapeutic options for patients with advanced CRC have seen limited improvement. Notably, previous work from Dr. Yatrik Shah’s group has revealed that CRC cells exhibit an addiction to iron, identifying this as a promising vulnerability that could be therapeutically leveraged in CRC (2).

First described with the term “ferroptosis” by Dr. Brent Stockwell in 2012, this specialized form of regulated cell death is driven by the iron-dependent accumulation of membrane-associated lipid peroxides (3). While the precise molecular mechanisms of ferroptosis are complex and not yet fully elucidated, excess iron and reactive oxygen species (ROS) lead to the overwhelming accumulation of lipid hydroperoxides, irreparably damaging cell membranes and inducing cell death by osmolytic processes (4). Interestingly, many cancer cells are highly sensitive to cell death via ferroptosis, and due to this sensitivity ferroptosis induction has emerged as a promising strategy to improve anti-cancer therapies. This is particularly notable for cancers that accumulate high levels of iron, such as CRC.

Studies investigating the impact of ferroptosis in cancer cell death have been facilitated by a panel of small molecules known for their ability to activate ferroptosis, such as erastin, RSL3, and ML162 (3,5). RSL3 is a RAS-selective lethal (RSL) compound identified by the Stockwell group based on its synthetic lethality with oncogenic RAS. At the time, the mechanistic drivers of RSL3’s effects were known to be iron-dependent, but otherwise elusive, and it was the understanding of RSL-induced cell death which ultimately culminated in the description of ferroptosis in 2012 (3). Later, the selenium-containing antioxidant protein glutathione peroxidase 4 (GPx4) was identified as the molecular target of RSL3 through affinity-based chemo-proteomics and thereby emerged as a key regulator of ferroptosis (5). Originally known as phospholipid hydroperoxide glutathione peroxidase, the ability of GPx4 to target and reduce lipid-associated ROS was already well-established and made it an attractive candidate to mediate the accumulation of lipid peroxides that were observed with ferroptosis (6). This role for GPx4 is unique even among other members of the glutathione peroxidase (GPx) family, and its indispensable role in preventing ferroptosis and lipid peroxidation has been shown in numerous in vitro and in vivo studies (6).

However, as is common with cancer therapeutics, the mechanisms of small molecule inhibitors are often incompletely understood and can include off-target effects. Despite RSL3 being widely considered as a direct inhibitor of GPx4, recent studies have questioned this specificity. Although their ultimate findings and explanations differed, two previous studies have shown that RSL3 does not affect the activity of purified GPx4 protein (7,8). While one group postulated that this was due to the need of an adapter protein to mediate RSL3’s effect on GPx4, the Arnér group proposed that another selenium-containing antioxidant protein, thioredoxin reductase 1 (TXNRD1), may be the primary target of RSL3 instead of GPx4. However, since bona fide TXNRD1 inhibitors induce cell death that cannot be rescued with the ferroptosis inhibitor, ferrostatin-1, the mechanism by which RSL3 activates ferroptosis may not be limited to TXNRD1 inhibition (8). Taken together, these data suggest that RSL3-dependent ferroptosis may not be solely due to inhibition of GPx4, and its mechanism of action may involve a more complex network of proteins.

In this issue of Cancer Research, DeAngelo, Zhao, et al. investigate the mechanism by which RSL3 activates ferroptosis in CRC (9). Using a panel of CRC cell lines, the authors determined that sensitivity to RSL3-induced ferroptosis did not correlate with the requirement for GPx4. Notably, some cell lines were sensitive to RSL3 but relatively insensitive to loss of GPX4, suggesting that RSL3 can induce ferroptosis through GPx4-independent mechanisms. To identify additional proteins that may be bound and inhibited by RSL3, the authors employed tandem mass tag labeling in combination with next-generation mass spectrometry. These biotinylation studies revealed that RSL3 indeed has broader binding and inhibitory capabilities than GPx4 and targeted multiple proteins with antioxidant capability, particularly those in the peroxidase and selenoprotein families. The authors also confirmed RSL3’s inhibition of TXNRD1, a selenoprotein family member crucial for cellular peroxidase activity, consistent with prior studies (8). Collectively, these findings argue that RSL3’s potent induction of oxidative stress, lipid peroxidation, and ferroptosis is due to simultaneous inhibition of multiple antioxidant molecules rather than inhibition of GPx4, alone. Importantly, this broader specificity must be considered for future experiments to fully interpret results downstream of treatment with RSL3 or its related compounds. Furthermore, the true targets and effects of RSL3 treatment may vary between different cell and tumor types, depending on the specific complement of antioxidant proteins present.

As noted, GPx4, TXNRD1, and many of their family members are “selenoproteins,” and collectively constitute a substantial portion of the cellular defenses against peroxides. Members of this rare protein family incorporate selenium into their primary structure as the modified animo acid, selenocysteine, with many serving antioxidant roles mediated by this selenocysteine residue (10). However, as selenocysteine is achieved by recoding of a “UGA” codon, selenoprotein synthesis is a highly regulated process that requires specialized translational machinery. Given that their results suggest RSL3 broadly inhibits the selenproteome to induce ferroptosis, DeAngelo, Zhao, and colleagues next hypothesized that targeting selenoprotein translation, in general, could have therapeutic effect in CRC cells. Here, they identify AlkBH8, a tRNA methyltransferase responsible for modifying the selenocysteine-specific tRNA, as a potential means to broadly impair selenoprotein synthesis. Indeed, loss of AlkBH8 was shown to induce cell death in CRC cell lines with minimal toxicity in normal intestinal cells. Together, this work suggests broadly inhibiting antioxidant selenoproteins or targeting their translational machinery may offer a novel therapeutic approach to improve cancer treatment by inducing widespread oxidative stress and ferroptosis.

In addition to identifying the selenoproteome as novel therapeutic targets for cancer therapy, this study also adds complexity to the role of selenium in human health. For years, selenium supplementation to promote activity of antioxidant selenoproteins was assumed to protect against cancer development. To this effect, multiple epidemiological studies from across the globe have indicated that lower levels of selenium are associated with increased cancer risk (9). However, large-scale clinical trials have failed to demonstrate a protective effect of selenium supplementation in American patients. While improvements in dosing strategies or patient selection may identify scenarios in which selenium supplementation is therapeutically beneficial, U.S. patients are also broadly selenium replete which likely negates the effect of further dietary supplementation. Interestingly, this study adds to a growing body of evidence that inhibition of selenium/selenoprotein activity and the use of pro-oxidant molecules may be more therapeutically relevant to cancer patients than selenium supplementation and promotion of antioxidant activity. This area of research is likely to be an exciting focus in the coming years, potentially providing novel targets to enhance therapeutic responses in CRC.

Figure:

In this issue, DeAngelo, Zhao, and colleagues determine that RSL3 induces oxidative stress and ferroptosis by broadly targeting antioxidant selenoprotein family members and not GPx4, alone. These studies also suggest that selenoproteins can inhibited by targeting their specialized translation machinery, which may provide novel therapeutics capable of inducing cancer cell death through ferroptosis. New findings from this study are denoted by red lines.