Dysregulation of the ERBB/EGFR signalling pathway causes multiple types of cancer. Accordingly, ADAM17, the primary shedding enzyme that releases and activates ERBB ligands, is tightly regulated. It has recently become clear that iRhom proteins, inactive members of the rhomboid-like superfamily, are regulatory cofactors for ADAM17. Here, we show that oncogenic KRAS mutants target the cytoplasmic domain of iRhom2 (also known as RHBDF2) to induce ADAM17-dependent shedding and the release of ERBB ligands. Activation of ERK1/2 by oncogenic KRAS induces the phosphorylation of iRhom2, recruitment of the phospho-binding 14-3-3 proteins, and consequent ADAM17-dependent shedding of ERBB ligands. In addition, cancer-associated mutations in iRhom2 act as sensitisers in this pathway by further increasing KRAS-induced shedding of ERBB ligands. This mechanism is conserved in lung cancer cells, where iRhom activity is required for tumour xenograft growth. In this context, the activity of oncogenic KRAS is modulated by the iRhom2-dependent release of ERBB ligands, thus placing the cytoplasmic domain of iRhom2 as a central component of a positive feedback loop in lung cancer cells.
The ERBB/EGFR signalling pathway is dysregulated in numerous cancers, especially of the lung, breast and ovary (Sigismund et al., 2018; Wang, 2017). In addition to oncogenic receptor mutations, tumorigenesis can be driven by excess ERBB ligand production (Arteaga and Engelman, 2014). ERBB family ligands are mostly synthesised as type I transmembrane domain proteins, and become active upon proteolytic cleavage and release (shedding) from the plasma membrane. Thus, shedding of ERBB ligands is a primary regulator of signalling that controls pathogenesis as well as cell proliferation, survival and differentiation.
This mode of regulation puts into the spotlight the enzymes responsible for shedding ligands. The metalloprotease ADAM17 is the most widespread sheddase of ERBB ligands, as well as controlling the release of many other growth factors, cytokines and other cell surface proteins (Zunke and Rose-John, 2017). Consistent with its potency, an intricate regulatory mechanism exists to control ADAM17, centred on iRhom1 and iRhom2 (also known as RHBDF1 and RHBDF2, respectively), which are rhomboid-like proteins that act as ADAM17 cofactors (Dulloo et al., 2019). iRhom proteins (iRhoms) are first required for the trafficking of ADAM17 from the endoplasmic reticulum (ER) to the Golgi, where the protease is matured by removal of its inhibitory pro-domain by the pro-protein convertase furin proteins (Adrain et al., 2012; Endres et al., 2003; McIlwain et al., 2012). iRhom2 has subsequently been shown to have an additional role at the plasma membrane – it regulates the activation of ADAM17 to release TNFα, the major type I inflammatory cytokine. This activation of ADAM17 by iRhom2 is triggered by phosphorylation of the cytoplasmic domain of iRhom2 mediated by ERK1 and ERK2 (ERK1/2; also known as MAPK3 and MAPK1, respectively) (Cavadas et al., 2017; Grieve et al., 2017).
Several lines of evidence have implicated iRhoms and ADAM17 in tumorigenesis, especially in lung, breast, cervical, oesophageal and colorectal cancers. iRhom2 and ADAM17 levels increase during cancer progression and correlate with lower survival rates (Agwae et al., 2020; Borrell-Pagès et al., 2003; Mochizuki and Okada, 2007; Walkiewicz et al., 2016; Xu et al., 2020; Zhou et al., 2006, 2014). The most direct link between the iRhom proteins and cancer are mutations in the cytoplasmic domain of iRhom2, which cause a rare familial syndrome, tylosis with oesophageal cancer (TOC), characterised by a very high lifetime risk of developing oesophageal cancer (Blaydon et al., 2012; Ellis et al., 2015; Mokoena et al., 2018; Qu et al., 2019; Saarinen et al., 2012). Increased activity of ADAM17 has been observed for iRhom2TOC mutations (Brooke et al., 2014; Maney et al., 2015) but, despite this strong genetic link, the precise mechanistic role of iRhoms in oncogenic signalling has been poorly explored.
ADAM17 is better characterised than iRhoms with respect to cancer, although until recently it too has not been the subject of the intense focus commensurate with its regulatory importance. For instance, oncogenic SRC triggers the ADAM17-dependent release of the ERBB ligand TGFα (Maretzky et al., 2011). It has also become clear that ADAM17 is important in cancers mediated by mutations in KRAS, which are the most frequent oncogenic mutations in human cancers, particularly in lung, colorectal and pancreatic tumours (Khan et al., 2019). Although oncogenic KRAS has long been considered to be constitutively active, and thus independent of upstream signals, a requirement for ADAM17 and ERBB1/EGFR in KRAS-induced pancreatic cancer has challenged this idea (Ardito et al., 2012; Navas et al., 2012). Indeed, ERBB signalling has now been shown to contribute to lung tumorigenesis by supporting activation of oncogenic KRAS (Kharbanda et al., 2020; Kruspig et al., 2018; Moll et al., 2018). In this context, it is also significant that KRAS-driven tumours express higher levels of ERBB ligands, in particular amphiregulin and TGFα (Ardito et al., 2012; Kruspig et al., 2018). However, as described above, ERBB ligands must be proteolytically shed to be active, and the regulation of shedding in cancer has been largely unknown. A recent advance has been the demonstration of a requirement for ADAM17 in KRAS-induced lung tumorigenesis (Saad et al., 2019). Using non-small-cell lung carcinoma (NSCLC) and patient-derived xenografts, as well as the KrasG12D mouse model, Saad et al. showed that depletion of ADAM17, or inhibition of its activity, suppressed lung tumour growth. They also found that oncogenic KRAS leads to increased activity of the p38 MAPKs and upregulated shedding of the ADAM17 substrate IL-6R.
Here, we report that iRhoms are essential for the oncogenic release of ERBB ligands mediated by KRAS-G12 mutants. Specifically, KRAS-induced shedding of ERBB ligands is triggered by the phosphorylation of the cytoplasmic domain of iRhom2, which allows the recruitment of the phospho-binding 14-3-3 proteins. Human cancer-associated mutations in the cytoplasmic domain of iRhom2 are sufficient to amplify this pathway, thus further establishing iRhom2 as an important component of oncogenic signalling. The pathological significance of this pathway was validated upon oncogenic KRAS expression in HEK239T cells and in the NSCLC cell line A549, which harbours an endogenous oncogenic KRASG12S mutation. Furthermore, loss of iRhom activity completely suppressed KRAS-driven tumour xenograft growth, demonstrating the requirement of iRhoms in a widely used model of lung cancer. Finally, we report that the cytoplasmic domain of iRhom2 is a hub for an ERBB-dependent positive feedback loop that maintains KRAS activity in lung cancer cells. Overall, our results demonstrate that iRhom2 plays a central role in oncogenic KRAS-induced signalling.
iRhoms are required for KRAS-driven shedding of ERBB ligands by ADAM17
Oncogenic KRAS induces the activation of ADAM17 (Ardito et al., 2012; Saad et al., 2019; Van Schaeybroeck et al., 2011), so we questioned whether iRhoms play a role in this process. First, to establish the effect of oncogenic KRAS in HEK293T cells, we expressed oncogenic KRASG12V. As expected, we observed a significant increase in the release of the ADAM17 substrate TGFα (Fig. 1A) especially compared to the effect of KRASS17N (Fig. 1A), a mutant with reduced GTPase activity (Farnsworth and Feig, 1991; Feig and Cooper, 1988). Using the inhibitors GI254023X and GW280264X, which respectively inhibit ADAM10, or both ADAM10 and ADAM17 (Hundhausen et al., 2003), we confirmed that TGFα shedding mediated by oncogenic KRAS was dependent on ADAM17 (Fig. 1B), which agrees with the reported ability of oncogenic KRAS to increase ADAM17-dependent shedding (Ardito et al., 2012; Saad et al., 2019; Van Schaeybroeck et al., 2011).
Although KRASG12V is one of the most well-studied oncogenic forms of KRAS (Muñoz-Maldonado et al., 2019), several other KRASG12X mutations are found in human cancers (Hobbs et al., 2016; Timar and Kashofer, 2020). We found that KRASG12S, KRASG12C and KRASG12D all caused elevated TGFα release (Fig. 1C). We also demonstrated that both isoforms of KRAS, 4A and 4B, induce shedding of TGFα (Fig. 1D). Overall, these results demonstrate the shared ability of KRAS oncogenic mutants to trigger growth factor release.
Having shown that oncogenic mutations in KRAS induce ADAM17-dependent shedding of TGFα, we next asked whether iRhoms are required for this activity. We found that KRAS-induced shedding of TGFα was completely blocked in HEK293T double-knockout (DKO) cells mutant for both iRhom1 and iRhom2 (Fig. 1E; Fig. S1). In single knockout lines, loss of iRhom1 had little effect, whereas iRhom2 KO showed a strong reduction in TGFα shedding (Fig. 1F), thereby demonstrating that iRhom2 is the primary mediator of KRAS-induced ADAM17-dependent shedding of TGFα.
KRAS-induced shedding depends on phosphorylation of the cytoplasmic domain of iRhom2 by the Raf/MEK/ERK pathway
To determine whether, as in inflammatory signalling (Cavadas et al., 2017; Grieve et al., 2017), iRhom2 phosphorylation participates in oncogenic ADAM17 signalling, we used a mutant version of iRhom2 (iRhom2site1-3) in which the three primary phosphorylation sites are changed to alanine residues (Grieve et al., 2017; Fig. S2A). We found that without iRhom2 phosphorylation at these three main sites, shedding of TGFα was significantly reduced (Fig. 2A). Importantly, this phosphorylation-deficient form of iRhom2 supported ADAM17 maturation as efficiently as wild-type iRhom2 (iRhom2WT; Fig. S2B), which aligns with our previous findings that iRhom2 phosphorylation is not needed for ADAM17 maturation (Grieve et al., 2017). These results reveal the role of iRhom2 phosphorylation in oncogenic signalling by KRAS.
In inflammatory signalling, iRhom2 phosphorylation is MAPK dependent (Cavadas et al., 2017; Grieve et al., 2017); it is also well established that oncogenic KRAS mutations act through the Raf/MEK/ERK MAPK pathway (Muñoz-Maldonado et al., 2019; Schubbert et al., 2007). We therefore asked whether the RAS/MAPK cascade also participates in iRhom2-dependent oncogenic signalling. TGFα shedding induced by KRASG12V was strongly inhibited by treating the cells with U1026 (Fig. 2B), a specific inhibitor of MEK1 and MEK2 (MEK1/2; also known as MAP2K1 and MAP2K2, respectively) (Favata et al., 1998), the kinases upstream of ERK1/2. We also found that oncogenic KRAS triggers the recruitment of 14-3-3 epsilon to iRhom2 and that, consistent with 14-3-3 proteins binding to phosphorylated residues (Fu et al., 2000), this recruitment was inhibited by treatment with U1026 (Fig. 2C; Fig. S2C). 14-3-3 recruitment to iRhom2 was associated with decreased binding between iRhom2 and ADAM17 (Fig. 2C). Although we have not investigated this phenomenon further, it agrees with our previous work on inflammatory signalling (Grieve et al., 2017) and suggests that the activation of ADAM17 by phosphorylated iRhom2 depends on an altered interaction between them. Since the recruitment of 14-3-3 to iRhom2 is sufficient for ADAM17 activation (Cavadas et al., 2017; Grieve et al., 2017), these results demonstrate that KRAS-induced shedding of ERBB ligands is mediated by ERK1/2-dependent phosphorylation of iRhom2.
ERK1/2 activation is not only induced by oncogenic KRAS but also by several other oncogenes (Liu et al., 2018; Roskoski, 2012; Wee and Wang, 2017), so we asked whether these other ERK1/2-activating oncogenes can similarly drive ADAM17 activity. HRASG12V, BRAFV600E and SRCY530F, all of which activated ERK1/2 (Fig. S2D), also induced elevated release of TGFα from HEK293T cells (Fig. 2D). This result is consistent with the increase in TGFα release seen previously with oncogenic SRC (Maretzky et al., 2011). ERK1/2-activating oncogenes KRASG12V and BRAFV600E also triggered the release of amphiregulin (Fig. S2E), another ADAM17-dependent ERBB ligand with a well-established role in oncogenesis (Busser et al., 2011; Xu et al., 2016). This contrasted with no increase of phosphorylated (p)ERK levels (Fig. S2D) (Beaver et al., 2013; Lauring et al., 2010) and a very small increase in amphiregulin release (Fig. S2E) caused by the oncogene AKT1E17K. These results suggest that the ability of ERK1/2-activating oncogenes to trigger the release of ERBB ligands depends on a common mechanism driven by phosphorylated iRhom2.
Cancer-associated mutations in iRhom2 potentiate KRAS-induced shedding of ERBB ligands
Our data demonstrate that iRhom2 phosphorylation participates in oncogenic signalling. The strongest and most direct evidence for the involvement of iRhom2 in human cancer is in the case of a rare inherited syndrome called tylosis with oesophageal cancer (TOC), which is caused by mutations in a small and highly conserved region within the cytoplasmic N-terminal domain of iRhom2 (Fig. 3A) (Blaydon et al., 2012). TOC is characterised by hyperkeratosis, oesophageal cancer, and at least in the case of one of the familial mutations, iRhom2D188N, by a susceptibility to other cancers (Saarinen et al., 2012). We therefore investigated whether the tylotic mutations affect oncogenic signalling through ADAM17. Replacing wild-type iRhom2 with tylotic iRhom2D188N caused a strong enhancement of KRAS-induced shedding of the ADAM17 substrate and the ERBB ligand amphiregulin (Fig. 3B). The shedding of EGF, which is triggered by ADAM10 rather than ADAM17 (Sahin et al., 2004), is not affected by iRhom2D188N (Fig. 3B), demonstrating the specificity of the oncogenic iRhom2 mutation for ADAM17. Strikingly, all analysed TOC mutations, including when combined, amplified KRAS-induced amphiregulin release (Fig. 3C) and none affected EGF shedding (Fig. S3A). Furthermore, none of the tylotic mutations altered ADAM17 maturation (Fig. S3B), consistent with our previous conclusion that the cytoplasmic tail of iRhom2 does not participate in the earlier iRhom2 function of promoting ER to Golgi trafficking of ADAM17 (Grieve et al., 2017). We conclude that TOC mutations are sufficient to potentiate KRAS-induced shedding of ADAM17 substrates in HEK293T cells, thereby establishing the direct effect of mutations in the N-terminus of iRhom2 in oncogene-driven signalling.
iRhoms are required for KRAS-driven tumorigenesis
Increased activation of ERBB1/EGFR as well as of the other ERBB receptors have widespread involvement in cancers (Hynes and MacDonald, 2009; Tebbutt et al., 2013; Wang, 2017) including, it has recently been established, in KRAS-induced lung tumorigenesis (Kharbanda et al., 2020; Kruspig et al., 2018; Moll et al., 2018). We therefore addressed the potential role of iRhoms in A549 cells, a widely used human lung adenocarcinoma cell model. These cells were selected because they are homozygous for KRASG12S, one of the mutations that we have shown drives TGFα release (Fig. 1C). Using CRISPR/Cas9, we knocked out both iRhom1 and iRhom2 in A549 cells to create three A549-DKO clonal cell lines, each of which, as expected, lacked ADAM17 maturation (Fig. S4A). The loss of iRhom activity, and thus of sheddase activity, was confirmed by the impairment shedding of the endogenous ERBB ligand amphiregulin (Fig. 4A; Fig. S4B). Unless otherwise specified, A549-DKO will from now on refer to A549 DKO clone 1. In addition, depletion of iRhoms induced a similar decrease in amphiregulin release in the KRAS mutant NSCLC cells lines NCI-H358 and NCI-H1792 (Fig. S4C), demonstrating the conserved function of iRhoms in promoting growth factor signalling in lung cancer cell lines. In support of this conclusion, A549-DKO cells also showed a decrease in cell proliferation (Fig. S4D).
We next assayed the requirement for iRhoms in the growth of A549 spheroids, three-dimensional (3D) models of solid tumours (Gilazieva et al., 2020; Ham et al., 2016). Supporting the significance of the standard two-dimensional (2D) cell culture result (Fig. S4D), loss of iRhom1 and iRhom2 also significantly inhibited spheroid growth (Fig. 4B; Fig. S4E). Consistent with the implication that iRhom-induced release of ERBB ligands contributes to spheroid growth, inhibition of ADAM17 but not ADAM10 also inhibited growth (Fig. 4C; Fig. S4F).
These results prompted us to ask whether iRhoms also participate in tumorigenesis in vivo, using a xenograft model in which A549 cells are injected into immunodeficient mice. This xenograft model allows preclinical evaluation of the role of candidate target genes in tumour formation and maintenance (Gengenbacher et al., 2017). We established xenografts of A549 parental cells and A549-DKO cells, and found that loss of iRhoms had a profound effect, preventing all detectable tumour growth (Fig. 4D). Based on the results in A549 lung cancer cells in 2D cell culture, 3D spheroid growth and tumour xenograft, we conclude that iRhoms are required for oncogenic signalling and lung tumour growth.
iRhom2 phosphorylation regulates ADAM17-dependent release of ERBB ligand and tumour spheroid growth in lung cancer cells
Having established that iRhoms are required in a lung tumorigenesis model, we addressed the molecular mechanism that underlies the pro-tumorigenic function of iRhom2 in A549 cells, using our earlier work in HEK293T cells as a guide. First, we made a phosphomutant version of iRhom2 in which the three main phosphorylation sites (site1-3) as well as additional contributing sites were mutated to alanine residues (iRhom2pMUT, Fig. S2A). Although iRhom2pMUT efficiently rescued ADAM17 maturation (Fig. 5A) and thus the basal shedding activity, it significantly inhibited the release of endogenous amphiregulin from A549 cells (Fig. 5B), demonstrating that the phosphorylation-specific function of iRhom2 is required to trigger the full release of growth factors by ADAM17. Second, ERK1/2 kinases drive this mechanism, as evidenced by our observation that the inhibitor U1026 blocked amphiregulin release (Fig. S5). Third, phosphorylation of iRhom2 is required for 14-3-3 binding in A549 cells (Fig. 5C), indicating that the phosphorylated iRhom2/14-3-3/ADAM17 activation pathway controls shedding of the ERBB ligands in these lung cancer cells. Together with our results in HEK293T cells, these results support the conclusion that oncogenic KRAS drives ERBB signalling by inducing iRhom2 phosphorylation. We investigated the biological importance of phosphorylation in iRhom2-dependent shedding of ERBB ligand by performing the 3D spheroid assay, which showed that spheroid growth of reconstituted A549-DKO cells was significantly reduced in iRhom2pMUT-expressing cells, compared to what was seen in cells expressing iRhom2WT (Fig. 5D). Overall, our results highlight the pro-tumorigenic role of iRhom2-dependent shedding of ERBB ligands in lung cancer cells.
Cancer-associated mutations in iRhom2 increase RAS activity and drive a positive feedback loop in lung cancer cells
In HEK293T cells, the cancer causing tylotic iRhom2 mutant D188N enhanced amphiregulin release by oncogenic KRAS mutations (Fig. 3). The same experiment in A549 cells confirmed this result in the lung cancer cell line – compared to iRhom2WT, expressing iRhom2D188N in A549-DKO cells caused a more than two-fold increase in the release of endogenous amphiregulin in the presence of oncogenic KRAS (Fig. 6A,B), indicating that tylotic mutation sensitises iRhom2 to oncogenic signalling. Strikingly, iRhom2D188N also further increased spheroid growth compared to iRhom2WT (Fig. 6C), demonstrating that even in transformed A549 cells, the elevated release of ERBB ligand caused by the tylotic iRhom2 mutation is sufficient to further promote spheroid growth of lung cancer cells. Importantly, this shows that a single point mutation in the N-terminus of iRhom2 is sufficient to increase the growth of lung cancer cells in spheroids.
Our observation that iRhoms, and in particular tylotic iRhom2D188N, induce ERBB signalling, suggests the existence of a positive feedback loop, in which oncogenic KRAS, signalling through iRhom2, ADAM17 and amphiregulin, promotes ERBB activity and ultimately further KRAS activity. This possibility builds on recent results that show that oncogenic KRAS mutations are not fully constitutive: using an allele-specific inhibitor, it has been shown that the activity of a KRAS mutant is modulated by upstream ERBB signalling (Lito et al., 2016; Patricelli et al., 2016). To test this hypothesis, we assayed the activity of oncogenic KRAS in A549 cells by using the RAS-binding domain of Raf1 to pulldown active RASGTP. Compared to parental cells, RASGTP levels were reduced by the absence of iRhoms in A549-DKO as they were upon treatment with the pan-ERBB inhibitor afatinib (Fig. 6D), thus suggesting that iRhoms are required to maintain RAS activity by activating ERBB signalling. As tylotic iRhom2D188N triggers a strong increase in RASGTP (Fig. 6E), it further establishes the central role of iRhoms in controlling RAS activity. To definitively conclude whether this feedback loop acts through iRhom2-dependent shedding in the extracellular medium, we assessed the effect of conditioned medium from A549 cells on the ERBB1/EGFR reporter cell line A431. Conditioned medium from tylotic iRhom2D188N caused elevated activated ERK1/2 compared to iRhom2WT (Fig. 6F). We confirmed that iRhom2-driven activation of the Raf/MEK/ERK pathway depends on ERBB signalling by using afatinib (Fig. 6F). Together, these results support that iRhom-dependent shedding of ERBB ligands in the extracellular medium drives a positive feedback loop to maintain the activity of oncogenic KRAS in lung cancer cells.
We have discovered that, by regulating ADAM17-dependent release of ERBB ligands, iRhoms are required for KRAS-driven tumorigenesis. Oncogenic mutants of KRAS induce ERK1/2-dependent phosphorylation of the cytoplasmic domain of iRhom2, triggering the recruitment of the phospho-binding 14-3-3 proteins, which in turn activate ADAM17 to shed ERBB ligands from the plasma membrane (Fig. 7). The relevance of this mechanism to human disease is demonstrated by our discovery that mutations in the cytoplasmic domain of iRhom2, known to be causative of the human cancer syndrome TOC, are sufficient to amplify this signalling pathway. The significance of iRhom2 to cancer pathogenesis is further reinforced by the result that loss of iRhom activity from A549 lung cancer cells completely blocks their ability to form tumours in a xenograft model.
iRhom2 is at the centre of a positive feedback loop to sustain oncogenic KRAS activity
As well as identifying iRhom2 as an essential player in KRAS-induced tumorigenesis, these results reveal the existence of a previously unidentified positive feedback loop that maintains RAS activity in lung cancer cells. In agreement with biochemical evidence proving that, contrary to prior belief, activated KRAS mutations are ‘hyperexcitable’ rather than constitutively locked in an active state (Lito et al., 2016; Patricelli et al., 2016), two recent studies have shown that oncogenic KRAS relies on upstream ERBB signalling to remain active, and thus to drive lung tumorigenesis (Kruspig et al., 2018; Moll et al., 2018). Our data establish that the cytoplasmic domain of iRhom2 is crucial in this mechanism; by being both downstream of oncogenic KRAS, and sufficient to increase ERBB-dependent RAS activation, the cytoplasmic domain of iRhom2 represents a central component of this newly uncovered positive feedback loop (Fig. 7). The existence of this feedback mechanism presupposes a sufficient pool of immature, plasma membrane-bound ERBB ligands that can be released in response to elevated iRhom2– ADAM17 activity to reinforce oncogenic KRAS activity. This requirement is supported by the recent observation that the expression of amphiregulin (and other ERBB ligands) is indeed elevated in KRAS-induced lung tumours (Kruspig et al., 2018). Overall, our results strengthen a now compelling body of evidence that overturns the earlier belief that oncogenic KRAS mutations are fully constitutive – instead, it is clear that KRAS-driven tumours are driven by signalling input to the activated KRAS oncoprotein. This opens a new potential strategy for therapeutic intervention.
iRhom2 induces activation of ADAM17 through a conserved mechanism in immune and cancer cells
Our results demonstrate that oncogenic and inflammatory signalling pathways share a conserved mechanism for the activation of the iRhom2–ADAM17 complex (this study and Cavadas et al., 2017; Grieve et al., 2017). One molecular aspect of the activation of ADAM17 by iRhom2 that we previously reported was that phosphorylation and 14-3-3 binding to iRhom2 causes some kind of conformational change in the complex between the two proteins, detected by weaker binding between them (Grieve et al., 2017). This partial uncoupling also occurred during KRAS-induced shedding (Fig. 2C), thus further demonstrating the conserved activation of the iRhom2–ADAM17 complex. Finally, the growing number of functional signalling complexes in which iRhom2 participates – iRhom2 and ADAM17 (Adrain et al., 2012; Cavadas et al., 2017; Grieve et al., 2017; McIlwain et al., 2012), iRhom2 and KRAS (Fig. 2C; Go et al., 2021), and iRhom2 and the previously described binding partner FRMD8 (Künzel et al., 2018; Oikonomidi et al., 2018) – strengthen the incentives to adopt mechanistic and structural approaches to understanding how iRhom2 controls ADAM17 signalling. For example, it would be interesting to investigate whether, in addition to 14-3-3, other iRhom2 interactors participate in KRAS-driven lung tumour formation.
Phosphorylation of the iRhom2–ADAM17 complex as an essential driver of KRAS-induced lung tumorigenesis
In work that complements these results, Saad et al. reported the requirement of ADAM17 in KRAS-induced lung tumorigenesis. They demonstrated that oncogenic KRAS induces the phosphorylation of ADAM17, as well as the release of its substrate IL-6R and an increase of ERK1/2 activation (Saad et al., 2019). Their work emphasises the need for better understanding of the potential role of the phosphorylation of ADAM17 on its catalytic activity (Düsterhöft et al., 2019; Zunke and Rose-John, 2017). Together with our work that demonstrates a clear molecular mechanism for the role of iRhom2 phosphorylation in the release of ERBB ligands, this establishes the wider significance of regulated shedding by ADAM17 as a mediator of oncogenic KRAS signalling. It will be interesting to explore the differences and possible crosstalk between the systems that lead, on one hand to shedding of soluble IL-6R triggered by phosphorylated ADAM17, and on the other hand, to ERBB ligand shedding induced by phosphorylated iRhom2.
The iRhom2–ADAM17 complex in other cancer types
Oncogenic KRAS is a driver of multiple cancers in addition to lung adenocarcinoma, so our work raises the question of whether iRhom2 also has a role in these other cancers. Pancreatic adenocarcinoma, the seventh leading cause of cancer-related deaths worldwide, is considered the most KRAS-addicted cancer (Bray et al., 2018; Rawla et al., 2019b; Waters and Der, 2018). Strikingly, ADAM17 and ERBB1/EGFR are both required to maintain high RAS activity in a KrasG12D mouse model of pancreatic ductal carcinoma (Ardito et al., 2012; Navas et al., 2012). In the light of the results we report here, it will be interesting to investigate whether iRhom2 plays a similar role in supporting a positive feedback loop in this particularly aggressive oncogenic context. In support of this possibility, the cytoplasmic domain of iRhom2 has been found to be phosphorylated in the presence of KRASG12D in pancreatic cancer cells (Tape et al., 2016). Another case where there is now a strong incentive to explore the possible involvement of iRhom2 is colorectal cancer, the second leading cause of cancer-related deaths worldwide (Bray et al., 2018; Rawla et al., 2019a), which can also be driven by oncogenic KRAS mutations (Cicenas et al., 2017). Using patient-derived organoids and xenografts, it has recently been demonstrated that ERBB signalling promotes tumorigenesis by maintaining ERK activity in colorectal tumours (Ponsioen et al., 2021). Although ADAM17 has been shown to be required for colorectal tumour growth (Schmidt et al., 2018), the possible contribution of iRhom2 phosphorylation in colorectal tumorigenesis is currently unexplored. Finally, further work will be required to characterise the molecular mechanism by which TOC mutations trigger activation of ADAM17. For example, we speculate that TOC mutations might finetune the active iRhom2–ADAM17 complex, or that they affect the recruitment of the cytosolic effectors of iRhom2.
In summary, we have shown that by driving ADAM17-dependent ERBB signalling, iRhoms are essential components in KRAS-driven tumorigenesis. On a mechanistic level, we report the existence of a KRAS/iRhom2/ERBB positive feedback loop that maintains oncogenic KRAS activity and might explain the potency of KRAS-induced cancers. Finally, by establishing the role of iRhom2 in oncogenic activation of ADAM17, our results provide new routes to explore future therapeutic opportunities.
MATERIALS AND METHODS
iRhom2, KRAS4A, KRAS4B SRC, BRAF and AKT1 constructs were amplified by PCR from iRhom2 cDNA (Künzel et al., 2018), KRAS4A cDNA (Grieve and Rabouille, 2014), KRAS4B cDNA (a kind gift from Julian Downward, Francis Crick Institute, London), SRC cDNA (antibodies-online), BRAF cDNA (antibodies-online) and AKT1 cDNA (antibodies-online). They were mutated using the QuikChange Multi Site-Directed Mutagenesis Kit (Agilent Technologies, 200515) and subcloned using In-Fusion HD Cloning Kit (Takara Bio, 639649) according to the manufacturer's instructions. For all constructs, single colonies were picked and extracted DNA was verified by Sanger sequencing (Source Bioscience, Oxford, UK). The list of the plasmids used in the study is available in Table S1.
Cell culture and transfections
Human embryonic kidney (HEK) 293T cells and human non-small-cell lung cancer (NSCLC) A549 cells were cultured in DMEM (Sigma-Aldrich) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich) and 2 mM L-glutamine (Gibco) at 37°C with 5% CO2. Human carcinoma A431 cells were cultured in EMEM (Lonza) supplemented with 10% FBS and 2 mM L-glutamine (Gibco). FuGENE HD (Promega) was used for transient DNA transfection in HEK293T cells, with a ratio of 1 μg DNA and 4 μl transfection reagent diluted in OptiMEM (Gibco). Lipofectamine 2000 (Thermo Fisher Scientific) was used for transient DNA transfection of A549 cells, with a ratio of 0.3 μg DNA and 1 μl transfection reagent. The knockdown experiments in NSCLC cell lines were performed with 60 nM siRNA using Lipofectamine RNAiMAX (Thermo Fisher Scientific) following the manufacturer's instructions. The list of the siRNAs used in the study is available in Table S2. The HEK DKO cells stably expressing pLVX-TetOne-zeo constructs were stimulated with 100 ng/ml doxycycline (MP Biomedicals, 195044).
CRISPR/Cas9 genome editing in A549 and HEK293T cells
CRISPR/Cas9-mediated single knockout of human RHBDF1 (encoding iRhom1) or RHBDF2 (encoding iRhom2) in HEK293T was performed as described previously (Künzel et al., 2018), and the list of the primers used in the study is available in Table S3. In brief, the plasmids co-expressing Cas9 nickase (Cas9n) and the gRNA targeting RHBDF1 or RHBDF2 were transfected into HEK293T cells. Upon puromycin selection and isolation of single colonies, the loss of RHBDF1 or RHBDF2 was analysed by PCR.
For CRISPR/Cas9-mediated double knockout of human RHBDF1 and RHBDF2 in A549 cells, 4 μg of plasmids co-expressing Cas9n and the gRNA were transfected using the Neon Transfection System (Invitrogen) according to the manufacturer's instructions. The following electroporation settings were used: 1230 volts, 30 s pulse width, 2 pulses number and 8×106 cells/ml. Antibiotic selection was performed using 0.5 μg/ml puromycin for 48 h, before selecting single colonies to establish clonal cell lines, and analysing loss of RHBDF1 and RHBDF2 by PCR.
Lentiviral transduction of cell lines
A549 or HEK293T DKO cells stably expressing iRhom2 constructs were generated by lentiviral transduction using the pLVX-TetOne or pHRSIN constructs as previously described (Adrain et al., 2012). Cells were selected by adding 50 μg/ml zeocin or 10 μg/ml Blasticidin S HCl. The list of the cell lines used in the study is available in Table S4.
Cells were washed three times with ice-cold PBS before lysis in Triton X-100 lysis buffer (1% Triton X-100, 150 mM NaCl, 50 mM Tris-HCl pH 7.5) supplemented with EDTA-free complete protease inhibitor mix (Roche, 11873580001), 10 mM 1,10-phenanthroline (Sigma-Aldrich, 131377-5G) and PhosSTOP (Roche, 04906837001). Pre-washed anti-HA magnetic beads (Thermo Fisher Scientific, 88837) were added to the lysates cleared from cell debris by centrifugation at 21,000 g at 4°C for 15 min and incubated for at least 2 h on a rotor at 4°C. Beads were washed five times with Triton X-100 lysis buffer and eluted with a 10-min incubation at 65°C in 2× SDS sample buffer (0.25 M Tris-HCl pH 6.8, 10% SDS, 50% glycerol, 0.02% bromophenol blue) supplemented with 200 mM DTT.
Concanavalin A enrichment
Cell lysates were incubated with 30 μl concanavalin A–Sepharose (Sigma-Aldrich, C9017-25ML) at 4°C for 2 h on a rotor. Beads were pelleted at 1500 g for 2 min at 4°C and washed five times with Triton X-100 lysis buffer. Glycoroteins were eluted with 2× LDS buffer (Invitrogen) supplemented with 25% sucrose and 50 mM DTT for 10 min at 65°C.
To detect active RAS in A549 cells, RAS-GTP pulldown was performed according to the manufacturer's instructions using the Active Ras Detection Kit (Cell Signaling Technology, #8821). In brief, one confluent 10 cm dish of cells was rinsed with ice-cold PBS and lysed in 0.5 ml ice-cold lysis buffer supplemented with 1 mM PMSF. Cell lysates were cleared by centrifugation (16,000 g for 15 min at 4°C) and protein concentration was determined by Bradford assay. Cleared lysates were added to the pre-washed spin cup which contains 100 μl of the 50% resin slurry and 80 μg of GST-Raf1-RBD and incubated at 4°C for 1 h on a rotor. The resin was washed three times with Wash Buffer and the proteins bound to the resin were eluted with 50 μl of the sample buffer supplemented with 200 mM DTT. Samples were denatured at 95°C for 5 min and were subjected to western blot analysis.
SDS-PAGE and western blotting
Cells were washed with ice-cold PBS before lysis in Triton X-100 lysis buffer supplemented with EDTA-free complete protease inhibitor mix (Roche, 11873580001), 10 mM 1,10-phenanthroline (Sigma-Aldrich, 131377-5G) and, when blotting for phosphoproteins, with PhosSTOP (Roche, 04906837001). Cell lysates were denatured at 65°C for 10 min in sample buffer supplemented with 100 mM DTT. Samples were run in 4–12% Bis-Tris NuPAGE gradient gels (Invitrogen) and MOPS running buffer (50 mM MOPS, 50 mM Tris, 0.1% SDS, 1 mM EDTA, pH 7.7), or in Novex 8-16% Tris-Glycine Mini Gels with WedgeWell format (Thermo Fisher Scientific) and Tris-glycine running buffer (25 mM Tris base, 192 mM glycine, 0.1% SDS, pH 8.3). Proteins were then transferred to a methanol-activated polyvinylidene difluoride (PVDF) membrane (Millipore) in Bis-Tris or Tris-Glycine transfer buffer. 5% milk in PBST (0.1% Tween 20) or TBST (0.05% Tween 20) was used for blocking and antibody incubation, and PBST or TBST was used for washing. The membranes were incubated with secondary antibodies at the room temperature for 1 h. Blots were quantified using ImageJ. The list of the antibodies used in this study is available in Table S5, and images of uncropped western blots are shown in Fig. S6.
HEK293T cell lines were seeded in poly-L-lysine (PLL, Sigma-Aldrich) coated 24-well plates in triplicate 24 hours before transfection. 50 ng alkaline phosphatase (AP)-conjugated substrates were transfected with FuGENE HD (Promega, E2312). In KRAS-related experiments, 100 ng control plasmids or KRAS plasmids were transfected together with AP substrates. At 24 h after transfection, cells were washed twice with PBS and incubated for 18 h in 300 μl Phenol Red-free OptiMEM (Gibco, 11058-021) supplemented with 1 μM GW280264X (GW; Generon, AOB3632-5) or GI254023X (GI; Sigma, SML0789-5MG) when indicated. For the kinase inhibition assay, 300 μl Phenol Red-free OptiMEM were supplemented with 10 μM U0126 (Abcam, ab120241-5mg). After 3 h incubation, supernatants were collected and cells were lysed in 300 μl Triton X-100 lysis buffer supplemented with EDTA-free protease inhibitor mix (Roche). 100 μl supernatant and 100 μl diluted cell lysates were independently incubated with 100 μl AP substrate p-nitrophenyl phosphate (PNPP; Thermo Fisher Scientific, 37620) at room temperature and the absorbance was measured at 405 nm by a plate reader (SpectraMax M3, Molecular Devices). The percentage of substrate release was calculated by dividing the signal from the supernatant by the total signal (supernatant and cell lysate).
Tumour spheroids were generated as previously described in Molina-Arcas et al. (2019). In brief, 2500 cells were resuspended in culture medium supplemented with 2.5% growth-factor reduced Matrigel (Scientific Laboratory Supplies, #356231) and placed in a 96-well round-bottom ultra-low attachment plate (Corning, #7007). Formation of the spheroids was initiated by centrifugation at 300 g for 4 min. After 13 days, tumour spheroids were imaged using a stereoscopic microscope (Leica DFC310 FX), and cell viability was measured using the CellTiter-Glo Cell viability assay (Promega, #G9681) according to the manufacturer's instructions.
Cell proliferation assay
To assay cell proliferation in a 2D adherent format, 1000 cells were seeded in standard 96-well tissue culture plate. After 5 days, cell viability was measured using the CellTiter-Glo Cell viability assay (Promega, #G9681) according to the manufacturer's instructions, as previously described in Patricelli et al. (2016).
A431 ERBB1/EGFR activation assay
1.5×106 A549 or 3×106 A431 cells were seeded in a 10 cm tissue culture dish. After 3 days, A431 cells were washed once with PBS, and serum-starved in 10 ml of OptiMEM supplemented with 1 μM afatinib (Stratech, A8247-APE) when indicated, and with 2 μM GW and to prevent growth factor release from A431 cells. The following day, the medium of A431 cells was renewed with OptiMEM supplemented with the same inhibitors, while A549 cells were washed once with PBS before adding 5 ml OptiMEM constituting the conditioned medium. After 4 h of collection, A431 cells were incubated with the conditioned medium for 3 min before being placed on ice and lysed as described in the SDS-PAGE and western blotting section.
80,000 A549 cells were seeded in triplicate per well of a 24-well plate. To study the loss of shedding in A549-DKO and A549-DKO-iRhom2pMUT cells, the medium was replaced the following day with 350 μl of full medium (DMEM supplemented with FBS and L-glutamine) and collected after 18 h of incubation. To determine the increased shedding in A549-DKO-iRhom2D188N, a 4-h collection was performed 48 h after seeding the cells. Similarly, a 4-h collection in 350 μl of full medium supplemented with 10 μM U0126 was performed to determine the contribution of ERK1/2. In all cases, the concentration of amphiregulin in the supernatant was determined using the Human Amphiregulin Quantikine ELISA Kit (R&D Systems, DAR00) according to the manufacturer's instructions. In parallel, the cells were lysed in Triton X-100 lysis buffer and the total protein concentration was measured using the BCA Assay (Life Technologies). The substrate release was determined by normalising the amphiregulin concentration by the total protein concentration.
106 Ctrl and iRhom1/2 double-knockout (DKO) A549 cells were resuspended in Matrigel:PBS (50:50 v/v) before being subcutaneously injected in one flank of 12 (n=6 mice per cell line) 6-week-old female immunodeficient NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice (Charles River UK Ltd, Margate, Kent, UK). Xenograft growth was monitored with a calliper twice weekly; tumour volume was determined using the following formula: (length×width2)/2. At the end of the experiment, tumours were collected and photographed. Animal experiments were performed under the Home Office Project Licence PPL30/3395 (licence holder, A.J.R.).
Statistics were performed in R (version 4.1.2). The required assumptions for one-way ANOVA were tested for each statistical test. Bartlett's test was used to evaluate the variance homoscedasticity and Shapiro–Wilk test was used to verify residual normality. A log10 transformation was applied when heteroscedasticity or non-normal distribution of residuals was detected. Pairwise comparison was performed using a post-hoc Tukey's test.
We are indebted to Ervin Fodor (Dunn School, Oxford, UK) for providing A549 cells, and to Sally Cowley (Dunn School, Oxford, UK) for the help in generating the CRISPR/Cas9 knockout A549 cell lines. The statistical analysis was performed by Nicolas Guex and Maxime Jan (Bioinformatics Competence Center, University of Lausanne, Switzerland). We thank the whole Freeman and Ryan groups for contributions during the course of the project and for feedback on the manuscript. We are thankful to Miriam Molina-Arcas and Julian Downward (Francis Crick Institute, London, UK) for the exchange of ideas and for providing the plasmid to express KRAS4B, and to Michael van de Weijer (Dunn School, Oxford, UK) for providing the pLVX-TetOne-zeo plasmid.
Conceptualization: B.S., F.L., A.G.G., M.F.; Methodology: B.S., F.L., S.M.S., A.J.R.; Validation: B.S., F.L., S.M.S.; Formal analysis: B.S., F.L., S.M.S.; Investigation: B.S., F.L., S.M.S., A.G.G.; Writing - original draft preparation: B.S., F.L., M.F.; Writing - review & editing: B.S., F.L., S.M.S., A.G.G., A.J.R., M.F.; Supervision: A.J.R., M.F.; Funding acquisition: A.J.R., M.F.
The Freeman group was supported by a Wellcome Trust Investigator Awards (101035/Z/13/Z and 220887/Z/20/Z). This project was also supported by a Cancer Research UK Development Fund (CRUKDF 0920-MF). B.S. was funded by a Medical Research Council scholarship (1643127). F.L. was supported by the China Scholarship Council-University of Oxford Scholarship (201806240008). A.G.G. received funding from H2020 Marie Skłodowska-Curie Actions fellowship (659166), and A.G.G. and M.F. were funded by the Biotechnology and Biological Sciences Research Council Research grant 440 (BB/RO16771/1). S.M.S. and A.J.R. were funded by the Medical Research Council Programme Grant (MC_UU_00001/6). Open access funding provided by a Read and Publish agreement with the University of Oxford. Deposited in PMC for immediate release.
Peer review history
The peer review history is available online at https://journals.biologists.com/jcs/lookup/doi/10.1242/jcs.259949.reviewer-comments.pdf.
M.F. is a director of the Company of Biologists but was not included in any aspect of the editorial handling of this article or peer review process. The authors declare no financial interests.