Periodic Reporting for period 4 - Agglomerates (Infinite Protein Self-Assembly in Health and Disease)
Reporting period: 2023-10-01 to 2024-02-29
Understanding how proteins respond to mutations is paramount to biology and disease. While protein stability and misfolding have been instrumental in rationalizing the impact of mutations, we recently discovered that an alternative route is also frequent, where mutations at the surface of symmetric proteins trigger novel self-interactions that lead to infinite self-assembly. This mechanism can be involved in disease, as in sickle-cell anemia, but may also serve in adaptation. Notably, it differs fundamentally from aggregation because misfolding does not drive it. Thus, we term it “agglomeration”. The ease with which agglomeration can occur, even by single point mutations, shifts the paradigm of how quickly new protein assemblies can emerge, both in health and disease. This prompts us to determine the basic principles of protein agglomeration and explore its implications in cell physiology and human disease.
We propose an interdisciplinary research program bridging atomic and cellular scales to explore agglomeration in three aims: (i) Map the landscape of protein agglomeration in response to mutation; (ii) Characterize how yeast physiology impacts agglomeration by changes in gene expression or cell state, and, conversely, how protein agglomerates impact yeast fitness. (iii) Analyze agglomeration in relation to human disease via two approaches. First, by predicting single nucleotide polymorphisms that trigger agglomeration, prioritizing them using knowledge from Aims 1 & 2, and characterizing them experimentally. Second, by providing a proof-of-concept that agglomeration can be exploited in drug design, whereby drugs induce its formation, like mutations can do. Overall, through this research, we aim to establish agglomeration as a paradigm for protein assembly, with implications for our understanding of evolution, physiology, and disease.
We propose an interdisciplinary research program bridging atomic and cellular scales to explore agglomeration in three aims: (i) Map the landscape of protein agglomeration in response to mutation; (ii) Characterize how yeast physiology impacts agglomeration by changes in gene expression or cell state, and, conversely, how protein agglomerates impact yeast fitness. (iii) Analyze agglomeration in relation to human disease via two approaches. First, by predicting single nucleotide polymorphisms that trigger agglomeration, prioritizing them using knowledge from Aims 1 & 2, and characterizing them experimentally. Second, by providing a proof-of-concept that agglomeration can be exploited in drug design, whereby drugs induce its formation, like mutations can do. Overall, through this research, we aim to establish agglomeration as a paradigm for protein assembly, with implications for our understanding of evolution, physiology, and disease.
We have made progress in several directions. Regarding Aim 1, we mapped the agglomeration landscape of two proteins following hundreds of random mutations targeting three surface positions. While surface mutations are classically viewed as benign, our analysis of several hundred mutants revealed they often trigger three main phenotypes in these proteins: nuclear localization, the formation of puncta, and fibers. Strikingly, more than 50% of random mutants induced one of these phenotypes in both complexes. Analyzing the mutant’s sequences showed that surface stickiness and net charge are two fundamental physicochemical properties associated with these changes. Additionally, for one of the proteins, more than 60% of mutants are self-assembled into fibers. Such a high frequency is explained by negative design: charged residues shielded this complex from self-interacting with copies of itself, and the sole removal of the charges induced its supramolecular self-assembly. A subsequent analysis of several other complexes targeted with alanine mutations suggested that such negative design is common. Together these results (Garcia-Seisdedos et al. PNAS, in press) highlight that minimal perturbations in protein surfaces’ physicochemical properties can frequently drive assembly and localization changes in a cellular context. Also in relation to Aim 1, we demonstrated that surface stickiness is selected against in disordered regions to minimize non-functional interactions (Dubreuil et al. J.Mol.Biol 2019). And finally, we came up with a more accurate stickiness scale (Villegas and Levy, bioRxiv. 2021), which we will use for the rest of this ERC project. Concerning Aim 2, the work is ongoing: we are analyzing co-localization between agglomerates and >100 protein-quality-control components by high-throughput confocal microscopy. We also started dissecting the impact that functional and non-functional agglomerates have on yeast fitness and cellular functions. Finally, concerning Aim 3, we are taking advantage of the recent structure-prediction revolution to maximize the coverage of the human proteome with symmetry information.
The creation of new protein-protein interactions is classically viewed as requiring new (positive design) elements. For example the addition of a hydrophobic residue that acts as a hot-spot for mediating an interaction. In the case of symmetric homo-oligomers, we found that this is not necessarily the case. Indeed, we saw that latent self-interacting interfaces may pre-exist at the surface and that the sole removal of a residue’s side chain may reveal a latent interface and allow binding. This finding adds to the textbook description of how protein interfaces can form. This part of the project is completed and our current work promises to be equally exciting. In particular, our knowledge of general principles by which agglomerates interact with cellular proteins and processes is inexistent. Thus, we anticipate this research to yield fundamental insights, beyond the state-of-the-art.