Structure-energy-based predictions and network modelling of RASopathy and cancer missense mutations

Abstract

The Ras/MAPK syndromes ('RASopathies') are a class of developmental disorders caused by germline mutations in 15 genes encoding proteins of the Ras/mitogen-activated protein kinase (MAPK) pathway frequently involved in cancer. Little is known about the molecular mechanisms underlying the differences in mutations of the same protein causing either cancer or RASopathies. Here, we shed light on 956 RASopathy and cancer missense mutations by combining protein network data with mutational analyses based on 3D structures. Using the protein design algorithm FoldX, we predict that most of the missense mutations with destabilising energies are in structural regions that control the activation of proteins, and only a few are predicted to compromise protein folding. We find a trend that energy changes are higher for cancer compared to RASopathy mutations. Through network modelling, we show that partly compensatory mutations in RASopathies result in only minor downstream pathway deregulation. In summary, we suggest that quantitative rather than qualitative network differences determine the phenotypic outcome of RASopathy compared to cancer mutations.

Figure 1. Genes affected in RASopathies

1. Network diagram showing affected genes in RASopathies. Proteins (PTPN11, SOS1 and SPRED) and protein groups (Ras, including NRAS, HRAS, KRAS and RIT1; GAP, including NF1 and RASA1; Raf, including RAF1 and BRAF; MEK, including MAP2K1 and MAP2K2) are displayed in white boxes and arranged in a network with their respective genes in grey. RASopathy diseases are indicated in blue.

2. Diseasome of RASopathies and cancer. Each node corresponds to a distinct disorder or cancer type. The size of the node corresponds to the total number of genes (among the 15 genes) that are involved in a particular disease. Abbreviations: NS, Noonan syndrome; NF1, neurofibromatosis type 1; CFC, cardiofaciocutaneous; LS, LEOPARD syndrome; HGF, hereditary gingvial fibromatosis; CM‐AVM, capillary malfunction‐arteriovenous malfunction; ALPS, autoimmune lymphoproliferative syndrome. Suffixes in NS (NS1, NS3, NS4, NS5, NS6, NS7, NS8 and NS‐like) and LS (LS1 to LS3) are different forms of the respective disease according to the classification in the OMIM database.

Figure 2. Distribution of somatic and germline mutations in 98 different structural domains and inter‐structural regions

For each gene, the number of different missense mutations were mapped on the respective domains and inter‐structural regions (called I1 to Ix, and ‘N’ = N‐terminal region; ‘C’ = C ‐terminal region). The colour of the bar diagrams represent mutations from different classes (see legend).

Figure 3. Classification of missense mutations and flow chart to analyse missense mutations based on structural information and FoldX energies

1. Flow chart of the analysis performed in this work.

2. Classification of missense mutations. Class 1a mutants can impair domain‐domain inhibitory interactions, as found in the phosphatase PTPN11 or the GEF Sos1. Class 1b mutants may release an inhibitory protein segment, thus resulting in activation, for example mutations in activation segment in kinases, such as BRAF. Class 2 mutants affect protein‐protein interactions and can result in a gain in signalling results from the loss of interactions with inhibitors or deactivating proteins. Examples are RAS mutations that prevent the down‐regulation by RASA1, or the binding of 14‐3‐3 proteins, which has been shown to interfere with Ras binding and inhibit Ras‐mediated plasma membrane recruitment of RAF1. Class 3 mutants destabilise a globular domain preventing folding. Class 4 mutants can affect protein localisation and/or half‐life. Class 5 mutants are neutral and are often localised on the protein surface.

Figure 4. Summary of the results from structure‐based energy calculations using FoldX

1. Structural coverage and the number of mutations predicted to be destabilising and non‐destabilising by FoldX.

2. Distribution of 427 missense mutations that map to a 3D structure into different classes.

3. Summary of the predicted effect for the 14 disease genes (proteins represented as pie diagrams). Proteins are arranged in a network similar as in Fig 1A. The grey numbers indicate the total number of different mutations represented in the pie diagrams.

Figure 5. FoldX energy changes for a gold set of RASopathy and cancer mutants

Figure 6. Schematic network model representation and model simulation results

1. Schematic representation of the network model. The 5 indicated rate constants represent the five experimentally measured rate constants for Ras WT and mutants. k_GEF is the GEF catalysed GDP dissociation, k_GAP is the GAP‐catalysed GTP hydrolysis, k_{GDP exch. intr.} is the intrinsic nucleotide dissociation, k_{GTP hydr. intr.} is the intrinsic GTP hydrolysis, and k_D is the affinity between Ras binding to the effector (EFF).

2. Results of the amount of Ras binding to effectors (Ras_EFF) in equilibrium for 12 network model simulations for using the respective 5 rate constants for Ras WT, Ras G12V (cancer) and RASopathy mutations (Ras K5N, V14I, Q22E, Q22R, P34L, P34R, T58I, G60R, E153V, F156L).