Machine learning-driven multiscale modeling reveals lipid-dependent dynamics of RAS signaling proteins.

RAS is a signaling protein associated with the cell membrane that is mutated in up to 30% of human cancers. RAS signaling has been proposed to be regulated by dynamic heterogeneity of the cell membrane. Investigating such a mechanism requires near-atomistic detail at macroscopic temporal and spatial scales, which is not possible with conventional computational or experimental techniques. We demonstrate here a multiscale simulation infrastructure that uses machine learning to create a scale-bridging ensemble of over 100,000 simulations of active wild-type KRAS on a complex, asymmetric membrane. Initialized and validated with experimental data (including a new structure of active wild-type KRAS), these simulations represent a substantial advance in the ability to characterize RAS-membrane biology. We report distinctive patterns of local lipid composition that correlate with interfacially promiscuous RAS multimerization. These lipid fingerprints are coupled to RAS dynamics, predicted to influence effector binding, and therefore may be a mechanism for regulating cell signaling cascades.

Uncovering a membrane-distal conformation of KRAS available to recruit RAF to the plasma membrane.

The small GTPase KRAS is localized at the plasma membrane where it functions as a molecular switch, coupling extracellular growth factor stimulation to intracellular signaling networks. In this process, KRAS recruits effectors, such as RAF kinase, to the plasma membrane where they are activated by a series of complex molecular steps. Defining the membrane-bound state of KRAS is fundamental to understanding the activation of RAF kinase and in evaluating novel therapeutic opportunities for the inhibition of oncogenic KRAS-mediated signaling. We combined multiple biophysical measurements and computational methodologies to generate a consensus model for authentically processed, membrane-anchored KRAS. In contrast to the two membrane-proximal conformations previously reported, we identify a third significantly populated state using a combination of neutron reflectivity, fast photochemical oxidation of proteins (FPOP), and NMR. In this highly populated state, which we refer to as "membrane-distal" and estimate to comprise ∼90% of the ensemble, the G-domain does not directly contact the membrane but is tethered via its C-terminal hypervariable region and carboxymethylated farnesyl moiety, as shown by FPOP. Subsequent interaction of the RAF1 RAS binding domain with KRAS does not significantly change G-domain configurations on the membrane but affects their relative populations. Overall, our results are consistent with a directional fly-casting mechanism for KRAS, in which the membrane-distal state of the G-domain can effectively recruit RAF kinase from the cytoplasm for activation at the membrane.

Exploring CRD mobility during RAS/RAF engagement at the membrane.

During the activation of mitogen-activated protein kinase (MAPK) signaling, the RAS-binding domain (RBD) and cysteine-rich domain (CRD) of RAF bind to active RAS at the plasma membrane. The orientation of RAS at the membrane may be critical for formation of the RAS-RBDCRD complex and subsequent signaling. To explore how RAS membrane orientation relates to the protein dynamics within the RAS-RBDCRD complex, we perform multiscale coarse-grained and all-atom molecular dynamics (MD) simulations of KRAS4b bound to the RBD and CRD domains of RAF-1, both in solution and anchored to a model plasma membrane. Solution MD simulations describe dynamic KRAS4b-CRD conformations, suggesting that the CRD has sufficient flexibility in this environment to substantially change its binding interface with KRAS4b. In contrast, when the ternary complex is anchored to the membrane, the mobility of the CRD relative to KRAS4b is restricted, resulting in fewer distinct KRAS4b-CRD conformations. These simulations implicate membrane orientations of the ternary complex that are consistent with NMR measurements. While a crystal structure-like conformation is observed in both solution and membrane simulations, a particular intermolecular rearrangement of the ternary complex is observed only when it is anchored to the membrane. This configuration emerges when the CRD hydrophobic loops are inserted into the membrane and helices α3-5 of KRAS4b are solvent exposed. This membrane-specific configuration is stabilized by KRAS4b-CRD contacts that are not observed in the crystal structure. These results suggest modulatory interplay between the CRD and plasma membrane that correlate with RAS/RAF complex structure and dynamics, and potentially influence subsequent steps in the activation of MAPK signaling.

Second harmonic generation detection of Ras conformational changes and discovery of a small molecule binder.

Second harmonic generation (SHG) is an emergent biophysical method that sensitively measures real-time conformational change of biomolecules in the presence of biological ligands and small molecules. This study describes the successful implementation of SHG as a primary screening platform to identify fragment ligands to oncogenic Kirsten rat sarcoma (KRas). KRas is the most frequently mutated driver of pancreatic, colon, and lung cancers; however, there are few well-characterized small molecule ligands due to a lack of deep binding pockets. Using SHG, we identified a fragment binder to KRasG12D and used 1H 15N transverse relaxation optimized spectroscopy (TROSY) heteronuclear single-quantum coherence (HSQC) NMR to characterize its binding site as a pocket adjacent to the switch 2 region. The unique sensitivity of SHG furthered our study by revealing distinct conformations induced by our hit fragment compared with 4,6-dichloro-2-methyl-3-aminoethyl-indole (DCAI), a Ras ligand previously described to bind the same pocket. This study highlights SHG as a high-throughput screening platform that reveals structural insights in addition to ligand binding.

Membrane-bound KRAS approximates an entropic ensemble of configurations.

KRAS4B is a membrane-anchored signaling protein and a primary target in cancer research. Predictions from molecular dynamics simulations that have previously shaped our mechanistic understanding of KRAS signaling disagree with recent experimental results from neutron reflectometry, NMR, and thermodynamic binding studies. To gain insight into these discrepancies, we compare this body of biophysical data to back-calculated experimental results from a series of molecular simulations that implement different subsets of molecular interactions. Our results show that KRAS4B approximates an entropic ensemble of configurations at model membranes containing 30% phosphatidylserine lipids, which is not significantly shaped by interactions between the globular G-domain of KRAS4B and the lipid membrane. These findings revise our understanding of KRAS signaling and promote a model in which the protein samples the accessible conformational space in a near-uniform manner while being available to bind to effector proteins.

Unveiling the Dynamics of KRAS4b on Lipid Model Membranes.

Small GTPase proteins are ubiquitous and responsible for regulating several processes related to cell growth and differentiation. Mutations that stabilize their active state can lead to uncontrolled cell proliferation and cancer. Although these proteins are well characterized at the cellular scale, the molecular mechanisms governing their functions are still poorly understood. In addition, there is limited information about the regulatory function of the cell membrane which supports their activity. Thus, we have studied the dynamics and conformations of the farnesylated KRAS4b in various membrane model systems, ranging from binary fluid mixtures to heterogeneous raft mimics. Our approach combines long time-scale coarse-grained (CG) simulations and Markov state models to dissect the membrane-supported dynamics of KRAS4b. Our simulations reveal that protein dynamics is mainly modulated by the presence of anionic lipids and to some extent by the nucleotide state (activation) of the protein. In addition, our results suggest that both the farnesyl and the polybasic hypervariable region (HVR) are responsible for its preferential partitioning within the liquid-disordered (Ld) domains in membranes, potentially enhancing the formation of membrane-driven signaling platforms.

Membrane lipids drive formation of KRAS4b-RAF1 RBDCRD nanoclusters on the membrane.

The oncogene RAS, extensively studied for decades, presents persistent gaps in understanding, hindering the development of effective therapeutic strategies due to a lack of precise details on how RAS initiates MAPK signaling with RAF effector proteins at the plasma membrane. Recent advances in X-ray crystallography, cryo-EM, and super-resolution fluorescence microscopy offer structural and spatial insights, yet the molecular mechanisms involving protein-protein and protein-lipid interactions in RAS-mediated signaling require further characterization. This study utilizes single-molecule experimental techniques, nuclear magnetic resonance spectroscopy, and the computational Machine-Learned Modeling Infrastructure (MuMMI) to examine KRAS4b and RAF1 on a biologically relevant lipid bilayer. MuMMI captures long-timescale events while preserving detailed atomic descriptions, providing testable models for experimental validation. Both in vitro and computational studies reveal that RBDCRD binding alters KRAS lateral diffusion on the lipid bilayer, increasing cluster size and decreasing diffusion. RAS and membrane binding cause hydrophobic residues in the CRD region to penetrate the bilayer, stabilizing complexes through ß-strand elongation. These cooperative interactions among lipids, KRAS4b, and RAF1 are proposed as essential for forming nanoclusters, potentially a critical step in MAP kinase signal activation.

Natural Product Graveoline Modulates Kirsten Rat Sarcoma Viral Oncogene Homologue (KRAS) Membrane Association: Insights from Advanced Spectroscopic Studies.

The KRAS gene plays a pivotal role in numerous cancers by encoding a GTPase that upon association with the plasma membrane activates the MAPK pathway, promoting cellular proliferation. In our study, we investigated small molecules that disrupt KRAS's membrane interaction, hypothesizing that such disruption could in turn inhibit mutant RAS signaling. Native mass spectrometry screening of KRAS-FMe identified compounds with a preference for interacting with the hypervariable region (HVR), and surface plasmon resonance (SPR) further refined our selection to graveoline as a compound exhibiting preferential HVR binding. Subsequent nuclear magnetic resonance (NMR) analysis showed that graveoline's interaction with KRAS depends on C-terminal O-methylation. Moreover, our findings revealed multiple interaction sites, suggesting weak engagement with the KRAS G domain. Using nanodiscs as a membrane mimetic, further characterization through NMR and Förster resonance energy transfer (FRET) studies demonstrated graveoline's ability to perturb KRAS membrane interaction in a biochemical setting. Our biophysical approach sheds light on the intricate molecular mechanisms underlying KRAS-ligand interactions, providing valuable insights into understanding the KRAS-associated pathophysiology. These findings contribute to the translational aspect of our study, offering potential avenues for further research targeting KRAS membrane association with the potential to lead to a new class of RAS therapeutics.

Direct Modulators of K-Ras-Membrane Interactions.

Protein-membrane interactions (PMIs) are ubiquitous in cellular signaling. Initial steps of signal transduction cascades often rely on transient and dynamic interactions with the inner plasma membrane leaflet to populate and regulate signaling hotspots. Methods to target and modulate these interactions could yield attractive tool compounds and drug candidates. Here, we demonstrate that the conjugation of a medium-chain lipid tail to the covalent K-Ras(G12C) binder MRTX849 at a solvent-exposed site enables such direct modulation of PMIs. The conjugated lipid tail interacts with the tethered membrane and changes the relative membrane orientation and conformation of K-Ras(G12C), as shown by molecular dynamics (MD) simulation-supported NMR studies. In cells, this PMI modulation restricts the lateral mobility of K-Ras(G12C) and disrupts nanoclusters. The described strategy could be broadly applicable to selectively modulate transient PMIs.

Machine Learning-Driven Multiscale Modeling: Bridging the Scales with a Next-Generation Simulation Infrastructure.

Interdependence across time and length scales is common in biology, where atomic interactions can impact larger-scale phenomenon. Such dependence is especially true for a well-known cancer signaling pathway, where the membrane-bound RAS protein binds an effector protein called RAF. To capture the driving forces that bring RAS and RAF (represented as two domains, RBD and CRD) together on the plasma membrane, simulations with the ability to calculate atomic detail while having long time and large length- scales are needed. The Multiscale Machine-Learned Modeling Infrastructure (MuMMI) is able to resolve RAS/RAF protein-membrane interactions that identify specific lipid-protein fingerprints that enhance protein orientations viable for effector binding. MuMMI is a fully automated, ensemble-based multiscale approach connecting three resolution scales: (1) the coarsest scale is a continuum model able to simulate milliseconds of time for a 1 µm2 membrane, (2) the middle scale is a coarse-grained (CG) Martini bead model to explore protein-lipid interactions, and (3) the finest scale is an all-atom (AA) model capturing specific interactions between lipids and proteins. MuMMI dynamically couples adjacent scales in a pairwise manner using machine learning (ML). The dynamic coupling allows for better sampling of the refined scale from the adjacent coarse scale (forward) and on-the-fly feedback to improve the fidelity of the coarser scale from the adjacent refined scale (backward). MuMMI operates efficiently at any scale, from a few compute nodes to the largest supercomputers in the world, and is generalizable to simulate different systems. As computing resources continue to increase and multiscale methods continue to advance, fully automated multiscale simulations (like MuMMI) will be commonly used to address complex science questions.

Asynchronous Reciprocal Coupling of Martini 2.2 Coarse-Grained and CHARMM36 All-Atom Simulations in an Automated Multiscale Framework.

The appeal of multiscale modeling approaches is predicated on the promise of combinatorial synergy. However, this promise can only be realized when distinct scales are combined with reciprocal consistency. Here, we consider multiscale molecular dynamics (MD) simulations that combine the accuracy and macromolecular flexibility accessible to fixed-charge all-atom (AA) representations with the sampling speed accessible to reductive, coarse-grained (CG) representations. AA-to-CG conversions are relatively straightforward because deterministic routines with unique outcomes are achievable. Conversely, CG-to-AA conversions have many solutions due to a surge in the number of degrees of freedom. While automated tools for biomolecular CG-to-AA transformation exist, we find that one popular option, called Backward, is prone to stochastic failure and the AA models that it does generate frequently have compromised protein structure and incorrect stereochemistry. Although these shortcomings can likely be circumvented by human intervention in isolated instances, automated multiscale coupling requires reliable and robust scale conversion. Here, we detail an extension to Multiscale Machine-learned Modeling Infrastructure (MuMMI), including an improved CG-to-AA conversion tool called sinceCG. This tool is reliable (∼98% weakly correlated repeat success rate), automatable (no unrecoverable hangs), and yields AA models that generally preserve protein secondary structure and maintain correct stereochemistry. We describe how the MuMMI framework identifies CG system configurations of interest, converts them to AA representations, and simulates them at the AA scale while on-the-fly analyses provide feedback to update CG parameters. Application to systems containing the peripheral membrane protein RAS and proximal components of RAF kinase on complex eight-component lipid bilayers with ∼1.5 million atoms is discussed in the context of MuMMI.

Metabolic profiling for the detection of bladder cancer.

The development and progression of many human diseases often result in changes in gene expression and protein and metabolite concentrations. Changes at the protein and metabolite level often are detectable in biological fluids and tissues before the appearance of clinical symptoms, rendering them useful diagnostic and prognostic biomarkers. As with many conditions, the discovery of a sensitive and specific urinary biomarker for bladder cancer would save lives and reduce the suffering due to this condition. A number of potential urinary protein biomarkers for bladder cancer have been identified, but they lack the sensitivity and specificity required to replace cystoscopy and histopathology. We discuss the use of mass spectrometry and nuclear magnetic resonance spectroscopy for the detection of metabolites in biological samples, comment on their advantages and limitations, and discuss recently published work in urine metabolic profiling for bladder cancer detection.

RAS Nanoclusters: Dynamic Signaling Platforms Amenable to Therapeutic Intervention.

RAS proteins are mutated in approximately 20% of all cancers and are generally associated with poor clinical outcomes. RAS proteins are localized to the plasma membrane and function as molecular switches, turned on by partners that receive extracellular mitogenic signals. In the on-state, they activate intracellular signal transduction cascades. Membrane-bound RAS molecules segregate into multimers, known as nanoclusters. These nanoclusters, held together through weak protein-protein and protein-lipid associations, are highly dynamic and respond to cellular input signals and fluctuations in the local lipid environment. Disruption of RAS nanoclusters results in downregulation of RAS-mediated mitogenic signaling. In this review, we discuss the propensity of RAS proteins to display clustering behavior and the interfaces that are associated with these assemblies. Strategies to therapeutically disrupt nanocluster formation or the stabilization of signaling incompetent RAS complexes are discussed.

Schweinfurthins I and J from Macaranga schweinfurthii.

Demand for the experimental antineoplastic agent schweinfurthin A, for developmental testing, prompted a re-collection of leaf material of Macaranga schweinfurthii from the original collection site in Cameroon. During chromatographic purification of the organic solvent extract, analytical UPLC-PDA-TOFMS of stilbene-enriched fractions revealed the presence of six known schweinfurthins and two previously unknown stilbenes. The structures of these new compounds, schweinfurthins I and J (1 and 2), were elucidated by 1D- and 2D-NMR techniques.

Analytical and statistical approaches to metabolomics research.

Metabolomics, the global profiling of metabolites in different living systems, has experienced a rekindling of interest partially due to the improved detection capabilities of the instrumental techniques currently being used in this area of biomedical research. The analytical methods of choice for the analysis of metabolites in search of disease biomarkers in biological specimens, and for the study of various low molecular weight metabolic pathways include NMR spectroscopy, GC/MS, CE/MS, and HPLC/MS. Global metabolite analysis and profiling of two different sets of data results in a plethora of data that is difficult to manage or interpret manually because of their subtle differences. Multivariate statistical methods and pattern-recognition programs were developed to handle the acquired data and to search for the discriminating features between data acquired from two sample sets, healthy and diseased. Metabolomics have been used in toxicology, plant physiology, and biomedical research. In this paper, we discuss various aspects of metabolomic research including sample collection, handling, storage, requirements for sample analysis, peak alignment, data interpretation using statistical approaches, metabolite identification, and finally recommendations for successful analysis.

Biophysical studies of DNA modified with conformationally constrained nucleotides: comparison of 2'-exo (north) and 3'-exo (south) 'locked' templates.

The biophysical properties of oligodeoxyribonucleotides (ODNs) selectively modified with conformationally 'locked' bicyclo[3.1.0]hexane pseudosugars (Maier,M.A., Choi,Y., Gaus,H., Barchi,J.J. Jr, Marquez,V.E., Manoharan,M. (2004) Synthesis and characterization of oligonucleotides containing conformationally constrained bicyclo[3.1.0]hexane pseudosugar analogs Nucleic Acids Res., 32, 3642-3650) have been studied by various techniques. Six separate synthetic ODNs based on the Dickerson Drew dodecamer sequence (CGCGAAT*T*CGCG) were examined where each one (or both) of the thymidines (T*) were substituted with a bicyclic pseudosugar locked in either a North (2'-exo) or South (3'-exo) ring pucker. Circular dichroism spectroscopy, differential scanning calorimetry and (1)H NMR spectroscopy were used to examine the duplex stability and conformational properties of the ODNs. Replacement of one or both thymidines with North-locked sugars (RNA-like) into the dodecamer did not greatly affect duplex formation or melt temperatures but distinct differences in thermodynamic parameters were observed. In contrast, incorporation of South-locked sugar derivatives that were predicted to stabilize this standard B-DNA, had the unexpected effect of causing a conformational equilibrium between different duplex forms at specific strand and salt concentrations. Our data and those of others suggest that although DNA can tolerate modifications with RNA-like (North) nucleotides, a more complicated spectrum of changes emerges with modifications restricted to South (DNA-like) puckers.

Molecular recognition of RAS/RAF complex at the membrane: Role of RAF cysteine-rich domain.

Activation of RAF kinase involves the association of its RAS-binding domain (RBD) and cysteine-rich domain (CRD) with membrane-anchored RAS. However, the overall architecture of the RAS/RBD/CRD ternary complex and the orientations of its constituent domains at the membrane remain unclear. Here, we have combined all-atom and coarse-grained molecular dynamics (MD) simulations with experimental data to construct and validate a model of membrane-anchored CRD, and used this as a basis to explore models of membrane-anchored RAS/RBD/CRD complex. First, simulations of the CRD revealed that it anchors to the membrane via insertion of its two hydrophobic loops, which is consistent with our NMR measurements of CRD bound to nanodiscs. Simulations of the CRD in the context of membrane-anchored RAS/RBD then show how CRD association with either RAS or RBD could play an unexpected role in guiding the membrane orientations of RAS/RBD. This finding has implications for the formation of RAS-RAS dimers, as different membrane orientations of RAS expose distinct putative dimerization interfaces.

A 4'-C-ethynyl-2',3'-dideoxynucleoside analogue highlights the role of the 3'-OH in anti-HIV active 4'-C-ethynyl-2'-deoxy nucleosides.

4'-C-ethynyl-2'-deoxynucleosides belong to a novel class of nucleoside analogues endowed with potent activity against a wide spectrum of HIV viruses, including a variety of resistant clones. Although favorable selectivity indices were reported for several of these analogues, some concern still exists regarding the 3'-OH group and its role in cellular toxicity. To address this problem, we removed the 3'-OH group from 4'-C-ethynyl-2'-deoxycytidine (1a). This compound was chosen because of its combined high potency and low selectivity index. The removal of the 3'-OH was not straightforward; it required a different synthetic approach from the one used to synthesize the parent compound. Starting with glycidyl-4-methoxyphenyl ether, the target 4'-C-ethynyl-2',3'-dideoxycytidine analogue (rac-1h) was obtained after 13 steps. In a cellular assay, rac-1h was completely inactive (0.001-10 microM) against HIV(LAI), demonstrating the critical importance of the 3'-OH for antiviral activity. To determine whether the role of the 3'-OH was essential for the phosphorylation of the compound by cellular kinases or for inhibition of DNA polymerization, we synthesized and tested the 5'-triphosphate (rac-1h-TP) for its ability to inhibit HIV reverse transcriptase (RT). rac-1h-TP was slightly more potent than AZT-5'-triphosphate against wild-type HIV RT, suggesting that the role of the 3'-OH is crucial only for the activation of the drug by cellular kinases. The lipase-catalyzed resolution of rac-1h into ent-1h (beta-D-dideoxyribo) and ent-14 (beta-L-dideoxyribo) and the synthesis of the corresponding 5'-triphosphates established the stereochemical assignment based on HIV RT's preference for the beta-D-enantiomer, which was confirmed by assaying against the M184V variant, an RT mutant with a marked preference for incorporating nucleosides in the D-configuration.

The use of urine proteomic and metabonomic patterns for the diagnosis of interstitial cystitis and bacterial cystitis.

The advent of systems biology approaches that have stemmed from the sequencing of the human genome has led to the search for new methods to diagnose diseases. While much effort has been focused on the identification of disease-specific biomarkers, recent efforts are underway toward the use of proteomic and metabonomic patterns to indicate disease. We have developed and contrasted the use of both proteomic and metabonomic patterns in urine for the detection of interstitial cystitis (IC). The methodology relies on advanced bioinformatics to scrutinize information contained within mass spectrometry (MS) and high-resolution proton nuclear magnetic resonance (1H-NMR) spectral patterns to distinguish IC-affected from non-affected individuals as well as those suffering from bacterial cystitis (BC). We have applied a novel pattern recognition tool that employs an unsupervised system (self-organizing-type cluster mapping) as a fitness test for a supervised system (a genetic algorithm). With this approach, a training set comprised of mass spectra and 1H-NMR spectra from urine derived from either unaffected individuals or patients with IC is employed so that the most fit combination of relative, normalized intensity features defined at precise m/z or chemical shift values plotted in n-space can reliably distinguish the cohorts used in training. Using this bioinformatic approach, we were able to discriminate spectral patterns associated with IC-affected, BC-affected, and unaffected patients with a success rate of approximately 84%.