Supplementary Figures 1-5, Tables 1-3

KRAS , the most frequently mutated oncogene in human cancer, produces two isoforms, KRAS4a and KRAS4b, through alternative splicing. These isoforms differ in exon 4, which encodes the final 15 residues of the G-domain and hypervariable regions (HVRs), vital for trafficking and membrane localization. While KRAS4b has been extensively studied, KRAS4a has been largely overlooked. Our multidisciplinary study compared the structural and functional characteristics of KRAS4a and KRAS4b, revealing distinct structural properties and thermal stability. Position 151 influences KRAS4a’s thermal stability, while position 153 affects binding to RAF1 CRD protein. Nuclear magnetic resonance analysis identified localized structural differences near sequence variations and provided a solution-state conformational ensemble. Notably, KRAS4a exhibits substantial transcript abundance in bile ducts, liver, and stomach, with transcript levels approaching KRAS4b in the colon and rectum. Functional disparities were observed in full-length KRAS variants, highlighting the impact of HVR variations on interaction with trafficking proteins and downstream effectors like RAF and PI3K within cells.

To dissect variant-function relationships in the KRAS oncoprotein, we performed deep mutational scanning (DMS) screens for both wild-type and KRASG12D mutant alleles. We defined the spectrum of oncogenic potential for nearly all possible KRAS variants, identifying several novel transforming alleles and elucidating a model to describe the frequency of KRAS mutations in human cancer as a function of transforming potential, mutational probability, and tissue-specific mutational signatures. Biochemical and structural analyses of variants identified in a KRASG12D second-site suppressor DMS screen revealed that attenuation of oncogenic KRAS can be mediated by protein instability and conformational rigidity, resulting in reduced binding affinity to effector proteins, such as RAF and PI3-kinases, or reduced SOS-mediated nucleotide exchange activity. These studies define the landscape of single amino acid alterations that modulate the function of KRAS, providing a resource for the clinical interpretation of KRAS variants and elucidating mechanisms of oncogenic KRAS inactivation for therapeutic exploitation.

The small GTPase Rat sarcoma virus proteins (RAS) are key regulators of cell growth and involved in 20-30% of cancers. RAS switches between its active state and inactive state via exchange of GTP (active) and GDP (inactive). Therefore, to study active protein, it needs to undergo nucleotide exchange to a non-hydrolysable GTP analog. Calf intestine alkaline phosphatase bound to agarose beads (CIP-agarose) is regularly used in a nucleotide exchange protocol to replace GDP with a non-hydrolysable analog. Due to pandemic supply problems and product shortages, we found the need for an alternative to this commercially available product. Here we describe how we generated a bacterial alkaline phosphatase (BAP) with an affinity tag bound to an agarose bead. This BAP completely exchanges the nucleotide in our samples, thereby demonstrating an alternative to the commercially available product using generally available laboratory equipment.

Abstract The diversity of chemical and structural attributes of proteins makes it inherently difficult to produce a wide range of proteins in a single recombinant protein production system. The nature of the target proteins themselves, along with cost, ease of use, and speed, are typically cited as major factors to consider in production. Despite a wide variety of alternative expression systems, most recombinant proteins for research and therapeutics are produced in a limited number of systems: Escherichia coli, yeast, insect cells, and the mammalian cell lines HEK293 and CHO. Recent interest in Vibrio natriegens as a new bacterial recombinant protein expression host is due in part to its short doubling time of ≤ 10 min but also stems from the promise of compatibility with techniques and genetic systems developed for E. coli. We successfully incorporated V. natriegens as an additional bacterial expression system for recombinant protein production and report improvements to published protocols as well as new protocols that expand the versatility of the system. While not all proteins benefit from production in V. natriegens, we successfully produced several proteins that were difficult or impossible to produce in E. coli. We also show that in some cases, the increased yield is due to higher levels of properly folded protein. Additionally, we were able to adapt our enhanced isotope incorporation methods for use with V. natriegens. Taken together, these observations and improvements allowed production of proteins for structural biology, biochemistry, assay development, and structure-based drug design in V. natriegens that were impossible and/or unaffordable to produce in E. coli.

Biochemical and biophysical assays using recombinant RAS require the protein to be in either the active or inactive state. Here we describe methods to exchange the nucleotide present in the purified RAS protein with either GDPβS, GppNHp, or GTP depending on the assay requirement. In addition, we also describe the HPLC method used to validate the exchange process and provide information on the efficiency of the nucleotide exchange.

Supplementary Materials and Methods

5' UTR structures of BONCAT identified proteins

Functional interaction of Ras signaling proteins with upstream, negative regulatory GTPase activating proteins (GAPs) represents a crucial step in cellular decision making related to growth and survival. Key components of the catalytic transition state for Ras deactivation by GAP-accelerated hydrolysis of Ras-bound guanosine triphosphate (GTP) are thought to include an arginine residue from the GAP (the arginine finger), a glutamine residue from Ras (Q61), and a water molecule that is likely coordinated by Q61 to engage in nucleophilic attack on GTP. Here, we use in-vitro fluorescence experiments to show that 0.1-100 mM concentrations of free arginine, imidazole, and other small nitrogenous molecule fail to accelerate GTP hydrolysis, even in the presence of the catalytic domain of a mutant GAP lacking its arginine finger (R1276A NF1). This result is surprising given that imidazole can chemically rescue enzyme activity in arginine-to-alanine mutant protein tyrosine kinases (PTKs) that share many active site components with Ras/GAP complexes. Complementary all-atom molecular dynamics (MD) simulations reveal that an arginine finger GAP mutant still functions to enhance Ras Q61-GTP interaction, though less extensively than wild-type GAP. This increased Q61-GTP proximity may promote more frequent fluctuations into configurations that enable GTP hydrolysis as a component of the mechanism by which GAPs accelerate Ras deactivation in the face of arginine finger mutations. The failure of small molecule analogs of arginine to chemically rescue catalytic deactivation of Ras is consistent with the idea that the influence of the GAP goes beyond the simple provision of its arginine finger. However, the failure of chemical rescue in the presence of R1276A NF1 suggests that the GAPs arginine finger is either unsusceptible to rescue due to exquisite positioning or that it is involved in complex multivalent interactions. Therefore, in the context of oncogenic Ras proteins with mutations at codons 12 or 13 that inhibit arginine finger penetration toward GTP, drug-based chemical rescue of GTP hydrolysis may have bifunctional chemical/geometric requirements that are more difficult to satisfy than those that result from arginine-to-alanine mutations in other enzymes for which chemical rescue has been demonstrated.

Legends for Supplementary Tables and Figures

Abstract RAS proteins are GTPases that regulate a wide range of cellular processes. RAS activity is dependent on its nucleotide-binding status, which is modulated by guanine nucleotide exchange factors (GEF) and GTPase-activating proteins (GAP). KRAS can be acetylated at lysine 104 (K104), and an acetylation-mimetic mutation of K104 to glutamine (K104Q) attenuates the in vitro–transforming capacity of oncogenic KRAS by interrupting GEF-induced nucleotide exchange. To assess the effect of this mutation in vivo, we used CRISPR-Cas9 to generate mouse models carrying the K104Q point mutation in wild-type and conditional KrasLSL-G12D alleles. Homozygous animals for K104Q were viable, fertile, and arose at the expected Mendelian frequency, indicating that K104Q is not a complete loss-of-function mutation. Consistent with our previous findings from in vitro studies, however, the oncogenic activity of KRASG12D was significantly attenuated by mutation at K104. Biochemical and structural analysis indicated that the G12D and K104Q mutations cooperate to suppress GEF-mediated nucleotide exchange, explaining the preferential effect of K104Q on oncogenic KRAS. Furthermore, K104 functioned in an allosteric network with M72, R73, and G75 on the α2 helix of the switch-II region. Intriguingly, point mutation of glycine 75 to alanine (G75A) also showed a strong negative regulatory effect on KRASG12D. These data demonstrate that lysine at position 104 is critical for the full oncogenic activity of mutant KRAS and suggest that modulating the sites in its allosteric network may provide a unique therapeutic approach in cancers expressing mutant KRAS. Significance: An allosteric network formed by interaction between lysine 104 and residues in the switch-II domain is required for KRAS oncogenicity, which could be exploited for developing inhibitors of the activated oncoprotein.

Highly metastatic osteosarcoma cells activate the mTOR pathway during metastatic progression

Ras proteins are GTPases that regulate a wide range of cellular processes. The activity of Ras is dependent on its nucleotide-binding status, which is modulated by guanine nucleotide exchange factors (GEFs) and GTPase-activating proteins (GAPs). Previously, we demonstrated that mutation of lysine 104 to glutamine (K104Q) attenuates the transforming capacity of oncogenic K-Ras by interrupting GEF induced nucleotide exchange. To assess the effect of this mutation in vivo , we used CRISPR/Cas9 to generate mouse models carrying the K104Q point mutation in wild-type and conditional K-RasLSL-G12D alleles. Consistent with our previous findings from in vitro studies, the oncogenic activity of K-RasG12D was significantly attenuated by mutation at K104 in vivo . These data demonstrate that lysine at position 104 is critical for the full oncogenic activity of mutant K-Ras and suggest that modification at K104, for example acetylation, may also regulate its activity. In addition, animals homozygous for K104Q were viable, fertile, and arose at Mendelian frequency, indicating that K104Q is not a complete loss of function mutation. Using biochemical and structural analysis, we found that the G12D and K104Q mutations cooperate to suppress GEF-mediated nucleotide exchange, explaining the preferential effect of K104Q on oncogenic K-Ras. Finally, we discovered an allosteric regulatory network consisting of K104 and residues including G75 on switch II (SWII) that is the key for regulating the stability of the α helix on SWII. In this allosteric network, K104-G75 interaction might be primary for keeping stabilization of SWII. Given the high frequency of KRAS mutations in human cancers, modulation of this network may provide a unique therapeutic approach.### Competing Interest StatementThe authors have declared no competing interest.

Background: The RAS genes, which are the prototype for the RAS superfamily of small GTPases, encode guanine nucleotide binding proteins that are frequently mutated in cancer. The superfamily has five major branches: RAS, RHO, RAB, RAN, and ARF. Its wild type members function as molecular switches that are active when GTP-bound and inactive when GDP-bound. Previously analyzed cancer-associated point mutants of RAS, such as those affecting codons 12, 13, and 61, and mutants of other superfamily members have followed this paradigm, as their steady-state level of bound GTP is higher than that of the wild type protein. However, it is possible that a less common cancer-associated mutant might be active when bound to GDP. Methods: We searched the Catalog of Somatic Mutations in Cancer (COSMIC) database and the TCGA database in the NCI Genomic Data Commons cancer-associated mutations in the RAS superfamily beyond the commonly mutated RAS codons, conducted biological and biochemical tissue culture studies and in vitro biochemical guanine nucleotide binding studies of a K-RAS mutant identified by this search, and complemented these findings with all-atom molecular dynamics simulations. Results: The database search identified a recurrent cancer-associated K-RAS mutant, M67L, as well as another cancer-associated mutant that affect the same codon in K-RAS and N-RAS and additional point mutants in the analogous codon in 10 other members of the RAS superfamily of GTPases. Detailed analysis of the K-RAS M67L mutant indicated that it has a gain-of-function biologically, but its low GTP level is similar to that of wild type K-RAS. Unlike the GDP-bound wild type, the GDP-bound M67L mutant interacts with and activates at least three RAS effectors: RALGDS, PI3K-p110a, and RAPGEF6. However, it interacts weakly with the prototypic RAS effector, RAF. All-atom molecular dynamics simulations of the mutant, conducted with and without RALGDS, provide structural insight into these characteristics of the GDP-bound mutant. Conclusion: Cancer-associated mutations that are active when GDP-bound occur in K-RAS and probably in other members of the RAS superfamily.

Abstract RAS oncoproteins are frequently mutated in pancreatic, colon and lung cancers and play a causal role in human carcinogenesis. Although it is a potentially good target for anticancer drugs development, it has been challenging to specifically target different forms of RAS using small molecules because of absence of evident target binding sites. Chemical reactivity of protein functional groups depends on the environment of a particular residue and reagent accessibility due to protein structure. We wanted to explore chemical reactivity of KRAS 4b functional groups using chemical reagents to identify novel sites for covalent and noncovalent drug development. NHS esters have been used for years for chemical modification of proteins. They are believed to selectively react with primary amino groups. However, there are reports in the literature that other nucleophilic groups such as sulfhydryl group of Cys can react with N-hydroxysuccinimide esters. It is possible that other functional groups may also react. Sulfo-NHS-acetate, sulfo-NHS-biotin and NHS-biotin were used for modification of K, HRAS proteins in GDP and GppNHp-loaded forms. To follow the extent of protein modification, MALDI-TOF MS was used. Generally, the reaction was performed until almost no unmodified protein was left. After modification the protein was digested by Glu-C and peptides were analyzed by MALDI-TOF MS and ES-MS/MS. A “modification index” (ratio of intensities of modified and unmodified peptide forms in MALDI-TOF MS spectra) was calculated to enable relative quantitation for all modified peptides. The comparison of the modification indexes for peptides derived from NHS-biotin modified KRAS and HRAS in GDP or GppNHp form suggested that GppNHp-loaded protein in general allows higher degree of modification. This suggests that there are inducible and expandable binding sites for small organic molecules that could be exploited for design of covalent and noncovalent drug leads. Acknowledgments: Supported by NCI contract HHSN261200800001E. Citation Format: Oleg Chertov, Xiaoying Ye, Timothy Waybright, Andrew G. Stephen. Probing amino acids residues chemical reactivity of KRAS 4b using N-hydroxysuccinimide esters [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 A02.

Protein prenylation is a key modification that is responsible for targeting proteins to intracellular membranes. KRAS4b, which is mutated in 22% of human cancers, is processed by farnesylation and carboxymethylation due to the presence of a 'CAAX' box motif at the C-terminus. An engineered baculovirus system was used to express farnesylated and carboxymethylated KRAS4b in insect cells and has been described previously. Here, we describe the detailed, practical purification and biochemical characterization of the protein. Specifically, affinity and ion exchange chromatography were used to purify the protein to homogeneity. Intact and native mass spectrometry was used to validate the correct modification of KRAS4b and to verify nucleotide binding. Finally, membrane association of farnesylated and carboxymethylated KRAS4b to liposomes was measured using surface plasmon resonance spectroscopy.

Abstract While the Ras GTPases are frequently mutated, especially at codons 12, 13, or 61, the structurally related Rho GTPases, although they are implicated in tumor invasion and metastasis, are rarely mutated in solid tumors, except for diffuse gastric cancer. The vast majority of cancer-associated RHO mutations are not located in codons 14, 15, or 63, which correspond to the 3 canonical mutant Ras codons. Here we have evaluated the biological and biochemical phenotypes of several cancer-associated RhoA mutants, which indicated that some RhoA mutants have unusual phenotypes and therefore prompted us to generalize our findings to other small GTPases by constructing and analyzing the analogous mutants in K-Ras. We constructed 8 cancer-associated RhoA mutants from the TCGA database, and found that each was gain-of-function biologically. Using the Rhotekin binding assay to infer RhoGTP levels, 5 mutants (Y42C, L57V, D59G, A61D, D76G), had higher binding than WT RhoA. However, 3 mutants (G62E. L69P and L69R) did not bind Rhotekin, but did bind strongly to Citron, a Rho-dependent kinase implicated in cytokinesis and cell cycle regulation. Because the three mutants were highly transforming, we directly determined GTP/GDP binding levels, the “gold-standard” assay, for all mutants, which indicated G62E had high GTP-binding and low GDP-binding, in contrast to its negative binding to Rhotekin, while L69P and L69R had high GDP-binding and low GTP-binding similar to WT. Purified GDP-bound L69P and L69R formed a complex with Citron more efficiently than did WT GTP-bound or GDP-bound RhoA. Endogenous Citron and CDK1 contribute to anchorage-independent growth, as siRNA knockdown of either gene in lines transformed by G62E, L69R or L69P inhibited this phenotype. The cancer-associated K-Ras mutants G60E and M67L (which correspond to RhoA G62E and L69R) and also K-Ras M67R displayed gain-of-function biologically, and the GTP/GDP-binding phenotype of each was analogous to the respective RhoA mutant: high GTP-binding for G60E, and low Raf-RBD binding and high GDP-binding for M67L and M67R. The low GTP K-Ras mutants bound more strongly to the RBD domain of Ras effectors other than Raf, resulting in activation of RAP and AKT in transfected cells. Our microarray data revealed that these GDP-bound mutant K-Ras or RhoA can up-regulate cell-cycle progression genes and down-regulate tumor suppressor genes compared to WT, which may account for their transforming phenotype. We conclude that some cancer-associated RhoA mutants are transforming when GDP-bound, a phenotype that has not been described for any small GTPase. The equivalent K-Ras mutants display a similar phenotype, which implies the findings should be extrapolatable to many small GTPases. Such mutants can be strongly transforming, although they only bind a subset of preferred downstream effectors and activate their respective pathways. In addition, the widely used Rhotekin binding assay to infer RhoGTP levels may give misleading results for some mutants. Citation Format: Xiaolan Qian, Alex Papageorge, Dunrui Wang, Brajendra Tripathi, Yucheng Gao, Marian Durkin, Beatriz Sanchez-Solana, Dominic Esposito, Timothy Waybright, Andrew Stephen, Douglas R. Lowy. Identification of cancer-associated RhoA mutants and analogous K-Ras mutants that induce gain-of function despite low GTP binding [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 2586.

RAS proteins are responsible for a third of all cancer cases and are involved in both early onset and progression of the deadliest cancers, including lung, colon, and pancreatic cancers. RAS proteins act as binary molecular switches that cycle between active GTP-bound and inactive GDP-bound states. The conversion from inactive to the active form is stimulated by guanine nucleotide-exchange factors (GEFs). Conversion back to the inactive form is mediated by GTPase-activating proteins (GAPs). In the GTP-bound state, RAS proteins interact with a variety of effector proteins such as RAF Kinase, PI 3-Kinase, and RalGDS, leading to activation of several signaling cascades within the cell. Among the three RAS isoforms - HRAS, KRAS, and NRAS, KRAS is the most commonly mutated gene. 98% of oncogenic RAS mutations are found at amino acid positions G12, G13, and Q61, which impair intrinsic and GAP-mediated GTPase function resulting in accumulation of constitutively GTP-bound RAS in cells. To gain insights into the effect of oncogenic mutations on overall structure and GTP hydrolysis, we solved high-resolution crystal structures of wild-type and six oncogenic mutants (G12C, G12D, G12V, G13D, Q61L, and Q61R) of KRAS4b in complex with GMPPNP (a non-hydrolysable GTP analog) and magnesium. Structural comparison of mutants vs. wild-type KRA4b in complex with GMPPNP/Mg2+ and KRAS4b mutants vs. WT-HRAS bound to RASA1-GAP provides a rationale for impaired intrinsic and GAP-mediated GTP hydrolysis in the KRAS mutants. Interestingly, we noticed variations in the conformation of the switch I region which contacts both nucleotide and many downstream effector proteins. In the structure of three KRAS mutants, the switch I region adopts a partially open conformation known as the state I. Importantly, the state I structures reveal new pockets that could be exploited for structure-based drug design. Recently we have also solved the structures of GMPPNP-bound wild-type and Q61 mutants of KRAS in complex with RasGAPs. Structural analysis of KRAS Q61 mutants bound to RasGAPs provides a rationale for impaired GAP-mediated GTP hydrolysis and new insights for designing small molecules that could potentially restore GAP-mediated GTPase activity in these mutants. Citation Format: Timothy Tran, Srisathiyanarayanan Dharmaiah, Wupeng Yan, Albert Chan, Timothy Waybright, William Gillette, Dominic Esposito, Andrew Stephen, Dwight Nissley, Frank McCormick, Dhirendra K. Simanshu. Structures of oncogenic KRAS mutants in the active state and complexed with RasGAP provide new insights into targeting RAS-driven cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 960.

Neurofibromin is a tumor suppressor encoded by the NF1 gene, which is mutated in Rasopathy disease neurofibromatosis type I. Defects in NF1 lead to aberrant signaling through the RAS–mitogen-activated protein kinase pathway due to disruption of the neurofibromin GTPase-activating function on RAS family small GTPases. Very little is known about the function of most of the neurofibromin protein; to date, biochemical and structural data exist only for its GAP domain and a region containing a Sec-PH motif. To better understand the role of this large protein, here we carried out a series of biochemical and biophysical experiments, including size-exclusion chromatography–multiangle light scattering (SEC-MALS), small-angle X-ray and neutron scattering, and analytical ultracentrifugation, indicating that full-length neurofibromin forms a high-affinity dimer. We observed that neurofibromin dimerization also occurs in human cells and likely has biological and clinical implications. Analysis of purified full-length and truncated neurofibromin variants by negative-stain EM revealed the overall architecture of the dimer and predicted the potential interactions that contribute to the dimer interface. We could reconstitute structures resembling high-affinity full-length dimers by mixing N- and C-terminal protein domains in vitro. The reconstituted neurofibromin was capable of GTPase activation in vitro, and co-expression of the two domains in human cells effectively recapitulated the activity of full-length neurofibromin. Taken together, these results suggest how neurofibromin dimers might form and be stabilized within the cell.

Although post-translational modification of the C-terminus of RAS has been studied extensively, little is known about N-terminal processing. Mass spectrometric characterization of KRAS expressed in mammalian cells showed cleavage of the initiator methionine (iMet) and N-acetylation of the nascent N-terminus. Interestingly, structural studies on GDP-and GMPPNP-bound KRAS lacking the iMet and N-acetylation resulted in Mg2+-free structures of KRAS with flexible N-termini. In the Mg2+-free KRAS-GDP structure, the flexible N-terminus causes conformational changes in the interswitch region resulting in a fully open conformation of switch I. In the Mg2+-free KRAS-GMPPNP structure, the flexible N-terminus causes conformational changes around residue A59 resulting in the loss of Mg2+ and switch I in the inactive state 1 conformation. Structural studies on N-acetylated KRAS-GDP lacking the iMet revealed the presence of Mg2+ and a conformation of switch regions also observed in the structure of GDP-bound unprocessed KRAS with the iMet. In the absence of the iMet, the N-acetyl group interacts with the central beta-sheet and stabilizes the N-terminus and the switch regions. These results suggest there is crosstalk between the N-terminus and the Mg2+ binding site, and that N-acetylation plays an important role by stabilizing the N-terminus of RAS upon excision of the iMet.