Periodic Reporting for period 1 - GPIDR (Understanding the genetic encoding of dynamic interactions underlying transcriptional regulation)
Reporting period: 2022-07-15 to 2024-07-14
Protein-protein interactions underlie a wealth of important biological functions. While many of these interactions occur between proteins or parts of proteins that have stable structures, a large portion of them occur in proteins that are interacting in a more dynamic way, often facilitated by so-called "intrinsic disorder". Indeed, one third of the eukaryotic proteome is made up of these intrinsically disordered regions that do not have a stable structure, many of which remain uncharacterized. Understanding how intrinsically disordered proteins engage in dynamic protein-protein interactions not only has important implications for fundamental biology, but also for understanding human disease and enabling drug discovery. In order to address this gap in knowledge, we focused our study on an interaction between a model protein domain (PDZ3) and a disordered peptide (CRIPT). When the domain and peptide interact, one half of the peptide becomes stable, while the other half remains dynamic, or "fuzzy". This feature of the system enabled us to study both modes of interaction in the same system, providing us with directly comparable results that we could use to understand how two aspects of protein binding, affinity (how strongly the protein binds) and specificity (how selectively the protein binds), are encoded for fuzzy and non-fuzzy binding modes.
In order to understand the encoding of fuzzy and non-fuzzy binding, we used a deep mutational scanning approach whereby we assayed various libraries of protein variants (totaling >400,000 variants across experiments) in PDZ3 and CRIPT using a protein complementation (PCA) based selection in budding yeast. We made almost every possible mutation and combinations thereof in the fuzzy and non-fuzzy parts of the CRIPT peptide to understand how protein affinity is encoded in these two binding modes, finding that while the dynamic part of CRIPT is more robust to mutation, it additively contributes to the binding affinity (strength of binding) in the interaction. We then made a library of protein variants in which we mutated PDZ3 in combination with CRIPT to understand how specificity is encoded in the interaction. Quantifying >200,000 energetic interactions between the PDZ domain and its ligand identified 20 major energetically coupled pairs of sites that control specificity. These are organized into six modules, with most mutations in each module reprogramming specificity for a single position in the ligand. We found that nine of the major energetic couplings controlling specificity are between structural contacts and 11 have an allosteric mechanism of action. Finally, we made a library of variants in which we mutated two residues at a time across the fuzzy and non-fuzzy parts of the CRIPT peptide. This enabled us to quantify whether and how the dynamic and stable parts of the peptide interact with each other. Remarkably, we found that the structured and fuzzy parts of the peptide communicate with each other through one central interaction, and show that this interaction has the potential to allow distal residues in the dynamic part of CRIPT to communicate with the structured residues in the binding pocket. In sum, our results quantify the binding specificities of >1,800 globular proteins to reveal how specificity is encoded and provide a direct comparison of the encoding of affinity and specificity in structured and dynamic molecular recognition.
We presented the first complete map of how binding affinity and specificity are encoded in a globular protein domain interacting with a disordered peptide through fuzzy and non-fuzzy binding modes. Our map reveals that specificity encoding in the domain is highly modular, with 20 distinct residues determining specificity for each position in the ligand by both direct and allosteric mechanisms. The more dynamic, ‘fuzzy’, tail of the peptide is more robust to mutation but can be used to additively tune affinity. These more dynamic residues make only small contributions to specificity through interactions with the domain, but they do contribute to specificity via interactions with the structured part of the peptide.