TL;DR
This study reveals that Ras protein stability plays a crucial role in determining the activating effects of mutations, particularly for non-hotspot mutations. By comparing mutagenesis data from bacterial systems and mammalian cells, the researchers found that mutations destabilizing Ras can lead to increased signaling activity, but only up to a certain point, after which they can cause loss of function. This work highlights the importance of protein stability in understanding Ras-driven cancers and suggests that cancer hotspots are specifically selected for their ability to bypass regulatory mechanisms without compromising protein integrity.
What Problem Does This Paper Address?
Ras proteins are central players in cell signaling, and mutations in Ras are common drivers of cancer. While it's known that mutations at specific "hotspots" (Gly 12, Gly 13, and Gln 61) are frequently found in cancer, previous deep mutagenesis studies have identified many other mutations that can also activate Ras. This raises a question: why are only a few mutations so prevalent in cancer? This paper aims to bridge this gap by investigating how protein stability influences Ras activation and whether this can explain the observed mutational landscape in cancer. The authors hypothesize that the interplay between protein stability and the presence of regulatory proteins like GTPase-activating proteins (GAPs) might be key to understanding why certain mutations are more common in cancer than others.
Methods
The study employed a multi-pronged approach:
- Saturation Mutagenesis: This technique systematically introduces all possible amino acid substitutions at each position in a protein to map the "fitness landscape" – how each mutation affects protein function. This was performed in two main systems:
- Mammalian Ba/F3 cells: This system uses full-length H-Ras and assesses its ability to promote cell proliferation in the absence of growth factors.
- Bacterial two-hybrid assay: This system uses a truncated H-Ras or K-Ras construct and measures Ras signaling activity by coupling it to antibiotic resistance. Crucially, this assay was performed under different conditions, including the presence or absence of a GAP.
- Hydrogen-Deuterium Exchange (HDX) coupled with NMR: This technique probes protein dynamics and conformational stability by measuring how quickly backbone amide hydrogens exchange with deuterium in solution. Increased HDX rates suggest decreased stability.
- Pulse Proteolysis and Circular Dichroism (CD) Spectroscopy: These biophysical methods were used to directly measure the thermodynamic stability of different Ras constructs and mutants.
- Receiver Operating Characteristic (ROC) curve analysis: This statistical method was used to quantitatively compare the predictive power of different mutagenesis datasets (mammalian vs. bacterial) in identifying cancer-relevant mutations.
Datasets
The study utilized:
- COSMIC (Catalogue of Somatic Mutations in Cancer) database: This database was used to identify mutations found in human cancers.
- gnomAD (Genome Aggregation Database): This database was used to identify non-pathogenic variants for true negative controls.
- cBioPortal database: This database was used as an alternative source of oncogenic mutation data for validation.
The authors also generated extensive mutagenesis datasets from their bacterial and mammalian cell experiments.
Key Insights and Ideas
- Stability is Key for Non-Hotspot Mutations: The most striking finding is that mutations at less frequent sites (e.g., Val 14, Asp 119) can be activating or deleterious depending on the stability of the Ras construct. This suggests that these mutations' effects are not solely determined by their direct interaction with signaling partners but also by their impact on the protein's overall fold.
- Cancer Hotspots are "Stability-Independent": In contrast, mutations at the canonical cancer hotspots (Gly 12, Gly 13, Gln 61) appear to activate Ras regardless of construct stability. This implies they function through a different mechanism, likely by directly interfering with GAP binding or catalysis, rather than by destabilizing the protein.
- The "Sweet Spot" for Activation: The study proposes a model where there's an optimal range of stability for Ras activation. Mutations that slightly decrease stability can enhance signaling by promoting nucleotide exchange, but further destabilization leads to loss of function. This "stability threshold" explains why only a narrow window of mutations is observed in cancer.
- Bacterial Assays are Predictive, with Caveats: The bacterial two-hybrid assay, especially when including a GAP, proved to be a good predictor of Ras behavior in mammalian cells. However, extending the Ras construct in the bacterial system (longer constructs) improved the correlation, suggesting that the C-terminal helix plays a role in Ras stability and function that is better captured by longer constructs.
Key Results
- Mammalian vs. Bacterial Correlation: Saturation mutagenesis data from mammalian Ba/F3 cells showed a strong correlation with bacterial experiments where Ras was co-expressed with a GAP.
- Stability-Dependent Mutations: Mutations at sites like Val 14 and Asp 119 were activating in more stable (longer) Ras constructs but deleterious in less stable (shorter) constructs.
- HDX Data Supports Destabilization: Activating mutations at non-hotspot sites (H27G, L120A, Y157Q) led to increased HDX rates, consistent with structural destabilization.
- ROC AUC Scores: The mammalian Ba/F3 dataset and bacterial Ras+GAP datasets showed high AUC scores (up to 0.89) for predicting cancer mutations, indicating their predictive power.
- Conservation Patterns: Cancer hotspots (Gly 12, 13, Gln 61) are less conserved in the Ras superfamily compared to stability-dependent sites, suggesting they are less critical for maintaining the core Ras fold and more specialized for interacting with regulators.
Limitations and Future Work
- Bacterial System Context: While the bacterial system is powerful, it lacks the full cellular context of mammalian cells, such as membrane localization and the complete set of interacting proteins. The study acknowledges this by comparing bacterial results to mammalian data and using longer constructs to better mimic cellular conditions.
- GAP Identity in Ba/F3 Cells: The specific GAP responsible for surveillance in the Ba/F3 cells is not identified, which could be a target for future investigation.
- Chaperone Protection: The paper notes that unlike some kinases, Ras proteins do not appear to be protected by chaperones like HSP90 from the consequences of destabilizing mutations. Further research could explore the role of other cellular factors in modulating Ras stability and activation.
- Tissue-Specific Patterns: The authors suggest that differences in chaperone activity and other factors might explain tissue-specific patterns of activating mutations in Ras, an area ripe for further exploration.
Summary updated