Search

KRAS mutations occur in roughly 15% of cancer patients, with KRAS G12D, KRAS G12V, and KRAS G12C substitutions constituting the majority of these cases. Cancer types also display puzzling preferences for certain RAS mutations; for instance, 49% of pancreatic ductal carcinomas (PDAC) are driven by KRAS G12D substitutions compared to 35% of colon adenocarcinomas (COAD) and 17% of lung adenocarcinomas (LUAD)1. In contrast, KRAS G12C mutations drive only 3% of PDAC and 10% of COAD, but 41% of LUAD cases1. Some of this discrepancy can be explained by mutagen exposure, as there is evidence that tobacco smoke carcinogens drive the predominance of KRAS G12C mutations in lung cancer. However, it is possible that inherent functional differences between the mutant proteins also contribute to such preferences across cancer types. These oncogenic substitutions do endow subtle and unique differences in biochemical properties, but it is unknown whether this translates to differences in cellular signaling potential. KRAS activates the MAPK and PI3K signaling pathways by binding directly to RAF1 or the p110 catalytic subunit of PI3K. These interactions are important signaling nodes that result in cellular proliferation, survival, movement, differentiation, and immune evasion, and therefore drive cancer when constitutively activated by mutant KRAS. Biochemical assessments of relative effector affinity between KRAS mutants lack crucial elements present in a cellular environment, such as GDP/GTP ratios, membrane interactions, subcellular localization, and engagement with other proteins. Here, we optimize a cell-based Bioluminescent Resonance Energy Transfer (BRET) system in HEK293T cells using NanoLuc- and mVenus-tagged proteins to quantify relative effector affinities between KRAS mutants. We assess the affinities of KRAS4b G12C, KRAS4b G12D, and KRAS4b WT for effectors p110α and RAF1(52-188), along with RAS binding deficient control mutants p110α T208D/K227A and RAF1(52-188)R89L, and show statistical differences in effector engagement between the wildtype and mutant proteins. We also demonstrate the efficacy of this optimized system as a drug screening tool by assessing compounds in clinical development for their ability to disrupt mutant KRAS-effector interactions within a cellular environment. 1Prior IA, Hood FE, Hartley JL. The frequency of Ras mutations in cancer. Cancer Research 2020. Citation Format: Megan Rigby, John Columbus, Vanessa Wall, Dominic Esposito, Thomas Turbyville, Dhirendra Simanshu, Dwight Nissley, Frank McCormick, Anna Maciag. Bioluminescence resonance energy transfer (BRET) as a tool for assessing mutant KRAS-effector affinity and drug efficacy [abstract]. In: Proceedings of the AACR Special Conference: Targeting RAS; 2023 Mar 5-8; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Res 2023;21(5_Suppl):Abstract nr B027.

The oncogene RAS is one of the most widely studied proteins in cancer biology, and mutant active RAS is a driver in many types of solid tumors and hematological malignancies. Yet the biological effects of different RAS mutations and the tissue-specific clinical implications are complex and nuanced. Here, we identified an internal tandem duplication (ITD) in the switch II domain of NRAS from a patient with extremely aggressive colorectal carcinoma. Results of whole-exome DNA sequencing of primary and metastatic tumors indicated that this mutation was present in all analyzed metastases and excluded the presence of any other clear oncogenic driver mutations. Biochemical analysis revealed increased interaction of the RAS ITD with Raf proto-oncogene Ser/Thr kinase (RAF), leading to increased phosphorylation of downstream MAPK/ERK kinase (MEK)/extracellular signal-regulated kinase (ERK). The ITD prevented interaction with neurofibromin 1 (NF1)-GTPase-activating protein (GAP), providing a mechanism for sustained activity of the RAS ITD protein. We present the first crystal structures of NRAS and KRAS ITD at 1.65-1.75 A resolution, respectively, providing insight into the physical interactions of this class of RAS variants with its regulatory and effector proteins. Our in-depth bedside-to-bench analysis uncovers the molecular mechanism underlying a case of highly aggressive colorectal cancer and illustrates the importance of robust biochemical and biophysical approaches in the implementation of individualized medicine.

Mutations in the small GTPases HRAS, NRAS, and splice variants KRAS4a and KRAS4b occur in roughly 20% of all cancers1. Each of these four RAS isoforms share a high degree of homology, yet play distinct biological roles; mouse models devoid of KRAS die mid-gestation, whereas HRAS and NRAS null mice survive2. The occurrence of RAS isoform mutations and their specific oncogenic codon substitutions are non-uniform across cancer types, with KRAS mutations in the G12 position frequently driving colorectal cancer while NRAS Q61 mutations are commonly present in skin cancer1. Several explanations have been proposed for this, including differential regulation at the level of transcription, mRNA stability, translation, and protein stability, along with distinctions in post-translational modification, localization, and tissue-specific availability of effectors3. Subtle differences in the biochemical properties of amino acid substitutions may also influence the dominant mutational forms across cancer types. However, the question of whether RAS isoforms and their specific codon substitutions preferentially engage with certain effectors within a cellular context remains unclear, despite considerable efforts. Here, we utilize bioluminescence resonance energy transfer (BRET) between mVenus- and NanoLuc-tagged recombinant proteins to derive quantitative measurements of effector affinity across RAS isoforms and common mutants. By transiently expressing a fixed concentration of NanoLuc-tagged effectors while titrating increasing amounts of mVenus-tagged RAS in a cellular system, we generated saturation curves that reflect the nature of binding between the protein pair in vivo to help untangle this longstanding question. We further validated interactions by immunoprecipitation, and determined construct expression and colocalization via western blot and confocal microscopy. 1Prior IA, Hood FE, Hartley JL. The frequency of Ras mutations in cancer. Cancer Research 20202Malumbres M, Barbacid M. RAS oncogenes: the first 30 years. Nature Reviews Cancer 2003;3:459-653Li S, Balmain A, Counter CM. A model for RAS mutation patterns in cancers: finding the sweet spot. Nature Reviews Cancer 2018;18:767-77 Citation Format: Megan Rigby, John Columbus, Vanessa Wall, Dominic Esposito, Thomas Turbyville. Interrogating RAS isoform and mutational specificity for effectors with bioluminescence resonance energy transfer (BRET) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2272.

Molecular testing of oncogenicRASmutations in colorectal cancer (CRC) have led to increased identification of mutations in patients with CRC.NRAS-mutated CRC has not been well characterized because it is less common (KRASmutations. Here, we report a novel 10 amino acid internal tandem duplication (ITD) inNRAS, which disrupts the switch II domain, in a patient with widely disseminated CRC. Hotspot next generation sequencing of a brain metastasis identified theNRASITD and aTP53missense mutation (p.P275F). Whole exome sequencing of the primary tumor and two metastatic lesions (lung and brain) confirmed that theNRASITD andTP53mutation were conserved between the primary tumor and both metastatic tumors, and identified an additional pathogenic mutation inCSMD1(a tumor suppressor gene). Structural biology and biochemical analyses demonstrated that theNRASITD prevented binding to GAP protein, leading to sustained RAS activation, increased interaction with RAF, and downstream MAPK activation. Additionally, we provide the first crystal structure of theRASITD. In conclusion, these studies indicate that theNRASITD was the probable primary driver mutation of this aggressive CRC. Identical or biologically similar ITDs inNRASandKRASmay be rare drivers of CRC and other aggressive malignancies.

KRAS4b is a major driver of cancers of the pancreas, lung, and colon, and until recently was considered undruggable. However, recent discoveries have revealed that molecules can target KRAS directly, and that our understanding of KRAS is incomplete. To exploit this new appreciation of KRAS druggability and biology, we have developed several protein-protein interaction (PPI) assays to target KRAS activity in live cells. Here we report on a campaign where we screened a 20,000-compound library against full length KRAS4b-G12D and RAF1 interactions in live HEK293 cells using the NanoBRET platform. From the list of screening hits, we identified several structurally related active compounds with dose-dependent, albeit weak, inhibition of the PPI signal. We expanded these active compounds to a more complete list of structural analogs which we tested in a full dose response. From this expanded group of compounds, we identified several with low µM IC50s. Follow up with these compounds in secondary assays revealed promising biological readouts, including inhibition of downstream pERK levels in G12D mutant PANC1 cells. In summary, by using well-designed live cell assays, we have identified promising leads, which we are pursuing with mechanism of action studies. Citation Format: John Columbus, Thomas J. Turbyville, Vanessa Wall, Dominic Esposito, David A. Barda, Sheng-Bin Peng. A live-cell, protein-protein interaction assay identified inhibitors of KRAS4b-G12D interaction with full-length RAF1 [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr LB-062.

Rat sarcoma (RAS) family of proteins function as GTP/GDP-dependent control switch to regulate Raf-MAPK effector pathway for cell proliferation and are frequently mutated in human cancer. RAS is anchored to the inner leaflet of the plasma membrane (PM) via C-terminal lipid-tether(s) and hypothesized to concentrate in distinct locations in membrane nano-domains to control the activation step. However, very little is known about the molecular dynamics of RAS on PM and the precise mechanism of activation. Here we characterize the molecular mobility of RAS variants by single-molecule tracking (SMT) method and estimate underlying mobility states. KRAS4b exhibit confined mobility with three diffusive states, a unique characteristic compared to all the other RAS isoforms (KRAS4a, NRAS, and HRAS). Although the major difference across RAS isoforms lies within the C-terminus HVR (hypervariable region) and its post-translational lipid modifications, the additional confinement for KRAS4b is contributed by the G-domain (globular domain) of the protein. Simulation of KRAS4b on membrane revealed a detailed atomistic mechanism that corroborates with the experimental results. Importantly, the oncogenic mutant of KRAS4b shows an altered mode of diffusion, implicating a plausible attribution to/from enhanced effector interaction. This study uncovers a novel underlying principle of KRAS4b functionality on living cell membrane with a potential for innovation in therapeutic strategies against its oncogenic variant. Citation Format: Debanjan Goswami, De Chen, John Columbus, Thomas Turbyville. Cooperative membrane interaction between G-domain and HVR defines unique diffusion behavior of KRAS4b [abstract]. In: Proceedings of the AACR Special Conference on Targeting RAS-Driven Cancers; 2018 Dec 9-12; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2020;18(5_Suppl):Abstract nr A09.

RAS proteins are GTP-dependent switches that control and regulate signaling pathways involved in cell fate and are frequently mutated in cancer. RAS association with the plasma membrane, or with certain endomembrane compartments, is a required step for its activity. Why precisely this is so remains an open question. One possibility is that RAS is merely required for recruitment of RAF and other effectors to the membrane where it can associate with key regulatory molecules that lead to signaling activation. However, four isoforms exist in humans (HRAS, NRAS, and two splice variants, KRAS4b and KRAS4a) and their differences lie within 22 amino acids in the C-terminal hypervariable region (HVR, aa 167-189). These differences in the domain responsible for membrane association result in the recruitment and organization of RAS into distinct membrane nanodomains, and it is thought that this results in differential signaling behavior from RAS isoforms. Isoform-specific RAS nanodomains are thought to both be highly dynamic, differentially composed of a variety of lipids and proteins, and supported by interactions with cytoskeletal structures. How these domains are regulated, and how they influence isoform-specific RAS signaling, remains unclear. Here, we report the molecular mobility of RAS variants in the plasma membrane of living cancer cells using single-molecule tracking methods. Detailed analysis of tracks revealed that KRAS4b molecules exhibit confined mobility with three diffusive states in the active plasma membrane of living cells. This diffusion characteristic was unique to KRAS4b and influenced by both the hypervariable region and globular domain of the protein, compared to all the other Ras isoforms. Importantly, the occupancy of each diffusive states was altered for the oncogenic mutant of KRAS4b, suggesting that the diffusive states we observe are related to signaling events. Our working hypothesis is that the HVR of KRAS4b may be directly involved in assembling the membrane lipid and protein nanodomain necessary for KRAS4b signaling activity. In addition, using two-color single-molecule tracking studies, we are beginning to characterize the interactions of KRAS4b with its major effector RAF. From these studies, we are learning about the kinetics of RAS/RAF interactions in the membrane of living cells. Understanding the underlying principle of KRAS4b functionality on cell membranes is useful for developing novel therapeutic strategies for targeting oncogenic KRAS4b. Citation Format: Debanjan Goswami, De Chen, John Columbus, Yue Yang, Felice Lightstone, Thomas Turbyville. KRAS4b’s unique diffusion behavior is defined by plasma membrane and effector interactions [abstract]. In: Proceedings of the AACR Special Conference on Targeting RAS-Driven Cancers; 2018 Dec 9-12; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2020;18(5_Suppl):Abstract nr IA07.

e15069 Background: Recent clinical guidelines for expansion of molecular testing of oncogenic RAS mutations in colorectal carcinoma have led to increased identification of mutations in this patient population. The clinical characteristics of NRAS-mutated colorectal carcinoma have not been well established because of the lower prevalence of these mutations (< 6%) as compared to oncogenic KRAS mutations. Here, we report the discovery of a novel internal tandem duplication (ITD) mutation of NRAS, which disrupts the Switch II domain in an index case of a patient with widely disseminated colorectal cancer. Methods: Hotspot next generation sequencing of a brain metastasis using a clinical solid tumor panel identified this novel NRAS ITD and a TP53 missense mutation (p.P275F). Whole exome sequencing of the primary tumor and two metastatic lesions (lung and brain) was performed to identify other genetic factors potentially driving the disease. Results: The above approach confirmed that the NRAS ITD and TP53 mutation were conserved between the primary tumor and both metastatic tumors and identified an additional pathogenic mutation in CSMD1 (a tumor suppressor gene). No other clearly pathogenic driver mutations were identified. Structural biology and biochemical analyses demonstrated that the inserted 15 amino acids prevented binding to GAP protein, leading to sustained RAS activation and increased interaction with RAF to stimulate downstream MAPK activation. Conclusions: These genomic and biochemical studies indicate that a novel NRAS ITD was the primary driver mutation of this aggressive colorectal carcinoma. Identical or similar ITDs in NRAS and KRAS may also drive other forms of colorectal cancer and other aggressive malignancies and represent a new subset of RAS-driven drug-resistant cancers.

All RAS isoforms, including the oncogenic variants, must localize to the plasma membrane (PM) to initiate downstream signaling. H-RAS, N-RAS and the KRAS splice variant KRAS4a all transit to the PM through the Golgi, and rely on cycles of reversible palmitoylation on one or two cysteines in the hyper variable region (HVR) to regulate exchange on and off the PM. In contrast, the KRAS4b HVR contains a series of positively charged lysine residues, which target KRAS4b to regions of the PM rich in negatively charged phospholipids. Though the details have not been fully worked out, evidence suggests that KRAS4b trafficking to and from the PM is mediated by chaperones such as PDE6δ and smgGDS. Since its path to the PM is distinct, and since membrane localization is a requirement for downstream signaling, perturbing the spatial organization of KRAS4b may uncover new molecular targets for KRAS-dependent cancers. To this end, we are developing a high content assay and image analysis pipeline to screen for molecules that disrupt the PM localization of KRAS4b. The assay is based on a doxycycline-inducible GFP-KRASG12V expressing HeLa cell line, and uses confocal microcopy to automatically image cells at sub-micron resolution in multi-well glass bottom plates. To specifically counterstain the PM compartment and nucleus, we use fluorescently labelled Concanavalin A (ConA) and Hoechst (a DNA specific dye) respectively. An image analysis pipeline was developed to segment and quantitate the levels of GFP-KRAS-G12V in the PM compartment. The pipeline includes a supervised machine learning algorithm to convert the grey-scale intensity values of the PM channel (ConA) to the probability that the ConA signal belongs to the PM. These probability values are subsequently used to segment the cell boundary and PM using a graph-cut method and a thresholding technique. We use the nuclear channel to eliminate any segmented boundaries that do not belong to a cell. Calculated Z’ factors using the mean values per well of membrane localized GFP-KRAS-G12V in doxycycline treated and untreated cells show acceptable values (0.7) for high content screening (HCS). Furthermore, as a secondary assay to evaluate hits from the primary HCS, we are developing fluorescence correlation spectroscopy (FCS) techniques to measure the temporal dynamics of GFP-KRAS-G12V membrane localization in live cells. In preliminary FCS experiments, we tested farnesyl thiosalicylic acid (FTS), a molecule reported to dissociate RAS molecules from the PM. We observed increased lateral diffusion (2.38±0.55 μm2/s) in drug treated cells (15-30 mins after treatment) compared to untreated cells (0.85±0.12 μm2/s) consistent with values previously reported using fluorescence recovery after photobleaching measurements (Niv H. et al. J. Biol. Chem. 1999; 274:1606-1613). Taken together, we demonstrate that our screening platform can be used to identify small molecule disruptors of oncogenic KRAS4b spatial and temporal localization in the PM of cells. Citation Format: Alla Brafman, Prabhakar Gudla, Kaustav Nandy, John Columbus, De Chen, Karen Worthy, Stephen Lockett, Thomas Turbyville. Targeting KRAS4b plasma membrane localization in cells. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4542. doi:10.1158/1538-7445.AM2015-4542