The presence of insoluble protein deposits in tissues and organs is a hallmark of many human pathologies. In addition, the formation of protein aggregates is considered one of the main bottlenecks to producing protein-based therapeutics. Thus, there is a high interest in rationalizing and predicting protein aggregation. For almost two decades, our laboratory has been working to provide solutions for these needs. We have traditionally combined the core tenets of both bioinformatics and wet lab biophysics to develop algorithms and databases to study protein aggregation and its functional implications. Here, we review the computational toolbox developed by our lab, including programs for identifying sequential or structural aggregation-prone regions at the individual protein and proteome levels, engineering protein solubility, finding and evaluating prion-like domains, studying disorder-to-order protein transitions, or categorizing non-conventional amyloid regions of polar nature, among others. In perspective, the succession of the tools we describe illustrates how our understanding of the protein aggregation phenomenon has evolved over the last fifteen years.

Intrinsically disordered proteins (IDPs) are essential players in the assembly of biomolecular condensates during liquid–liquid phase separation (LLPS). Disordered regions (IDRs) are significantly exposed to the solvent and, therefore, highly influenced by fluctuations in the microenvironment. Extrinsic factors, such as pH, modify the solubility and disorder state of IDPs, which in turn may impact the formation of liquid condensates. However, little attention has been paid to how the solution pH influences LLPS, despite knowing that this process is context-dependent. Here, we have conducted a large-scale in-silico analysis of pH-dependent solubility and disorder in IDRs known to be involved in LLPS (LLPS-DRs). We found that LLPS-DRs present maximum solubility around physiological pH, where LLPS often occurs, and identified significant differences in solubility and disorder between proteins that can phase-separate by themselves or those that require a partner. We also analyzed the effect of mutations in the resulting solubility profiles of LLPS-DRs and discussed how, as a general trend, LLPS-DRs display physicochemical properties that permit their LLPS at physiologically relevant pHs.

Proteome-wide analyses suggest that most globular proteins contain at least one amyloidogenic region, whereas these aggregation-prone segments are thought to be underrepresented in intrinsically disordered proteins (IDPs). In recent work, we reported that intrinsically disordered regions (IDRs) indeed sustain a significant amyloid load in the form of cryptic amyloidogenic regions (CARs). CARs are widespread in IDRs, but they are necessarily exposed to solvent, and thus they should be more polar and have a milder aggregation potential than conventional amyloid regions protected inside globular proteins. CARs are connected with IDPs function and, in particular, with the establishment of protein-protein interactions through their IDRs. However, their presence also appears associated with pathologies like cancer or Alzheimer’s disease. Given the relevance of CARs for both IDPs function and malfunction, we developed CARs-DB, a database containing precomputed predictions for all CARs present in the IDPs deposited in the DisProt database. This web tool allows for the fast and comprehensive exploration of previously unnoticed amyloidogenic regions embedded within IDRs sequences and might turn helpful in identifying disordered interacting regions. It contains >8,900 unique CARs identified in a total of 1711 IDRs. CARs-DB is freely available for users and can be accessed at http://carsdb.ppmclab.com. To validate CARs-DB, we demonstrate that two previously undescribed CARs selected from the database display full amyloidogenic potential. Overall, CARs-DB allows easy access to a previously unexplored amyloid sequence space.

No abstract