Dampening oncogenic RAS signaling

See allHide authors and affiliations

Science  22 Mar 2019:
Vol. 363, Issue 6433, pp. 1280-1281
DOI: 10.1126/science.aav6703

The control of normal cellular homeostasis is a function of signaling by the RAS guanosine triphosphatases (GTPases) KRAS, NRAS, and HRAS and their downstream signaling proteins, such as the RAF kinases (BRAF and CRAF), mitogen-activated protein kinase (MAPK) kinase (MEK), and extracellular signal–regulated kinase (ERK), which together constitute the RAS-MAPK pathway (1). This pathway is dysregulated in human diseases, particularly cancer, in which mutations or other nongenetic events hyperactivate the pathway in more than 50% of cases (1). Activating mutations in RAS genes occur in more than 30% of all cancers and seemingly lock RAS into a constitutively active, GTP-bound state to signal autonomously without the need for upstream input (1). Therapeutic suppression of pathogenic RASMAPK signaling to maximize disease control in cancer patients remains an elusive goal. Multiple strategies targeting upstream [e.g., receptor tyrosine kinases (RTKs)] or downstream (e.g., MEK) RAS pathway proteins to limit RAS-GTP–mediated signaling in RAS-MAPK pathway–driven cancers are under evaluation (13). However, recent studies have extended our knowledge of how certain forms of oncogenic RAS and other oncogenic MAPK pathway proteins function not in an autonomous manner but instead semiautonomously, such that they still respond to upstream regulation (48). These findings reveal mechanisms by which RAS signaling is dysregulated in cancer and highlight newly identified therapeutic strategies with the potential to target oncogenic RAS-MAPK signaling.

RAS GTPases are evolutionarily conserved molecular switches with nucleotide [GTP and guanosine diphosphate (GDP)] binding capacity and intrinsic GTPase activity that regulate a variety of cellular processes, including cell proliferation, growth, and differentiation (1). When stimulated by upstream signals, such as RTK activation at cell membranes, RAS proteins bind GTP with the assistance of adaptor proteins and guanine nucleotide exchange factors (GEFs) such as growth factor receptor–bound protein 2 (GRB2), GRB2-associated binding protein (GAB), the tyrosine protein phosphatase SHP2, and SOS. When GTP-bound, RAS proteins undergo a conformational change that recruits downstream effector proteins, such as the RAF kinases, to cellular membranes to initiate signaling via the MAPK pathway, which in turn elicits transcriptional and posttranslational events that shape cellular phenotypes. These effects are down-regulated by the intrinsic conversion of GTP to GDP, which is accelerated by GTPase-activating proteins (GAPs), such as the tumor suppressor neurofibromatosis-1 (NF1). Oncogenic RAS mutations and oncogenic RAS-MAPK pathway alterations (e.g., mutant BRAF) have largely been considered to keep the RAS-MAPK pathway in a constitutively “on” state that is not sensitive to regulation by trans-acting factors (RTKs, adaptors, GEFs, and GAPs). This view is supported by studies showing that GTP-GDP cycling of RAS is often disabled through oncogenic mutation(s) that, in part, interfere with GAP-stimulated GTP hydrolysis (3).

This paradigm presents challenges for effective pharmacologic inhibition of the RAS-MAPK pathway. Direct interference of RAS binding to GTP is problematic because of the high affinity of RAS-GTP binding (3). There are emerging agents that target certain forms of oncogenic RAS directly (e.g., KRAS-Gly12Cys) (2), but these drugs may face the problem of cell signaling plasticity and pathway reactivation (often via RTK hyperactivation) that limits the clinical benefit of current RAS pathway–targeted drugs, such as RAF and MEK inhibitors (9). Inhibition of individual upstream RTKs [such as epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK)] to decrease RAS-GTP incurs the problem of RTK redundancy, wherein multiple RTKs can substitute for one another to reactivate RAS-MAPK signaling (10). One potential therapeutic strategy involves blocking protein-protein interactions that are critical connections between RTK and RAS activation, e.g., GRB2, SOS, or SHP2. However, inhibiting protein-protein interactions and protein phosphatase activity with the high specificity and selectivity required for clinical use remains a challenge.

With these challenges in mind, our understanding of how RAS signaling is corrupted in cancer remains incomplete. Cancer-associated RAS mutations commonly occur at certain codons, such as codons 12, 13, or 61, favoring different mutant amino acids at these sites in different cancer types. It remains unclear whether each type of cancer-associated RAS mutation in different RAS isoforms (KRAS, HRAS, and NRAS) dysregulates the GTP-GDP binding and hydrolysis state of the encoded protein uniformly. Similarly, it is unknown whether each type of mutation and RAS isoform affects interactions with or governance by regulatory proteins, or activation of various downstream signaling pathways, in an identical manner.

Several recent studies enhance our understanding of the layers of control of oncogenic RAS-MAPK signaling, with therapeutic implications (48, 11, 12). The collective work shows that oncogenic RAS-MAPK pathway output is tightly regulated, particularly by GRB2, SOS, and SHP2, which link RTK activation and RAS-GTP loading. Furthermore, downstream gating mechanisms have been identified that control how information flow from the MAPK pathway to its nuclear transcriptional effectors is interpreted to elicit gene expression and proliferative changes in cells (12). These regulatory mechanisms can be dysregulated in cancer, resulting in inappropriate MAPK pathway activation and cell proliferation. Therefore, focusing on GTP-GDP nucleotide cycling could be a promising approach to suppress signaling by certain forms of semiautonomous oncogenic RAS (e.g., KRAS-Gly12Cys and KRAS amplification), BRAF (class 3 mutants such as BRAF-Gly466Val, which do not cause direct hyperactivation of BRAF kinase activity and require RAS-GTP), and NF1 inactivation in cancer subtypes lacking clinically approved pathway-targeted drugs.

For example, KRAS-Gly12Cys, which occurs in several cancer types including approximately 10 to 12% of non–small-cell lung cancer (NSCLC) cases, retains some intrinsic biochemical capacity to hydrolyze GTP to GDP and thus requires upstream regulation by RTK-SHP2-GRB2-SOS for optimal GTP loading. SHP2 inhibition can deprive the pathway of upstream RTK-mediated signaling flux and enhance therapeutic response to inhibition of upstream and downstream pathway components (e.g., RTKs such as ALK and effectors such as MEK) (48, 11, 12). SHP2 inhibition appears to be effective because, in part, it disrupts GRB2-SOS–mediated activation of RAS by RTKs such as EGFR. Other mechanisms by which SHP2 inhibition decreases RAS pathway signaling in these cancers likely exist and require further investigation. The emerging allosteric SHP2-targeted drugs, which regulate SHP2 by binding at a site in the protein other than the enzyme active site, provide a means to define the mechanistic basis for the role of SHP2 in promoting oncogenic RAS pathway signaling. Dampening the throttle by upstream pathway blockade of SHP2 appears to be a plausible way to target certain RTK-RAS pathway–driven cancers.

Regulating RAS pathway signaling in cancer

The conventional model of oncogenic RAS-MAPK pathway signaling in cancer suggests that mutations in the pathway render downstream signaling largely independent of regulation (autonomous). However, the emerging model of a semiautonomous state through which pathological RAS signaling remains under some control suggests a potential therapeutic opportunity to target upstream regulators, such as SHP2, SOS, and GRB2.


Instead of a static model of autonomous RAS-MAPK pathway activation, a more fluid semiautonomous model whereby controls of RAS pathway signaling kinetics and output can drive cancer reveals that targeting the GTPase cycle could be effective (see the figure). Regulatory proteins such as adaptors, GEFs, and GAPs, and protein phosphatases such as SHP2, which modulate RAS-MAPK pathway output in response to growth factor signaling, are good starting points. The emerging role of these proteins in promoting the oncogenic potential of semiautonomous RAS-MAPK pathway oncoproteins is consistent with the idea that such regulation may afford cancer cells the ability to titrate the amplitude and duration of RAS signaling to drive cancer, while also limiting excessive oncogene-induced stresses [e.g., replicative or proteostatic stress (13)] that can be a barrier to oncogenesis.

There are immediate therapeutic implications of the improved understanding of semiautonomous RAS pathway activation in cancer. SHP2 inhibition, either alone or in combination with inhibition of RTKs (EGFR or ALK), mutant KRAS (KRAS-Gly12Cys), or MEK, can suppress RAS-GTP levels and pathway output in certain RAS-mutant, BRAF-mutant, NF1-mutant, and RTK-driven tumors (48, 11, 12). SHP2 inhibition as monotherapy is in phase I trials. These trials will evaluate safety as a prelude to clinical efficacy studies and as a foundation for combination therapy strategies that are prompted by the new preclinical findings in NSCLC, melanoma, and other cancers. Targeting convergent signaling nodes, such as SHP2 or in the future GRB2 or SOS (which operate downstream of several RTKs), in combination regimens could mitigate the redundancy in RTK activation that occurs with blockade of individual RTKs, RAS, RAF, and MEK and limits drug response.

New strategies for drugging “undruggable” targets, such as oncogenic RAS, represent a key advance. Targeting SHP2 with an allosteric drug clears a path for drug development against other traditionally undruggable targets such as the oncoprotein MYC and the tumor suppressor PTEN. Another emerging area involves the development of allosteric small-molecule drugs that modulate the protein conformational states and dynamics of disease-relevant targets. Because they bind to a protein at a site that is not the active site, allosteric drugs provide complementary modulation of protein function against a target in concert with catalytic site–directed drugs [e.g., adenosine triphosphate (ATP)–competitive inhibitors] or drugs that covalently bind to their targets. Because of these distinct mechanisms of action, mutations in the protein target that cause resistance to one class of small-molecule drug (e.g., an ATP-competitive inhibitor) are less likely to induce resistance to another (e.g., an allosteric inhibitor). Thus, the development of more allosteric drugs for cancer could limit the potential for cross-resistance. Allosteric drugs could help control disease-driving targets that present challenges for catalytic modulation, or they could be used in combination regimens to blunt signaling plasticity, which can drive resistance to catalytic and covalent inhibitors.

There are several key questions and future directions. What are the precise molecular events that link RTK activation through SHP2, GRB2, and SOS to oncogenic RAS pathway output? Will these research advances allow the field to decrease the upstream signaling flux that underlies the full malignant potential of semiautonomous oncogenic RAS pathway alterations? What fraction of cancers harbor semiautonomous RAS-MAPK pathway oncogenic mutations? Will targeting the RAS GTP-GDP cycle, alone or in combination with other drugs targeting components of the RAS-MAPK pathway, be safely achieved in humans? Alternating drug dose and administration schedules could be tested in clinical trials to address this question. The improved mechanistic framework of semiautonomous oncogenic RAS pathway alterations highlights the potential of therapeutically targeting RAS GTP-GDP cycle regulation in cancer.

References and Notes

Acknowledgments: Supported by NIH grants NCIU01CA217882, NCIU54CA224081, NCIR01CA204302, NCIR01CA211052, and NCIR01CA169338 and by the Pew-Stewart Foundations. T.G.B. is a consultant to Novartis, AstraZeneca, Revolution Medicines, Takeda, and Array Biopharma and receives research funding from Novartis and Revolution Medicines. His spouse is an employee of Plexxikon Inc.
View Abstract

Stay Connected to Science

Navigate This Article