It was not until more than a century after the discovery of proteins that doubts began to emerge about a deceptively successful paradigm of protein mechanism: that a protein folds into a specific three-dimensional structure, which enables it to bind its complementary target, much like a lock and key. Yet, as anomalies in protein behaviour accumulated, this picture appeared incomplete. Some proteins seemed less like rigid locks and more like glass fresh from the furnace, still shifting and taking shape. Rather than settling into a single, static structure, these proteins can interconvert between a multitude of conformations within nanoseconds,, while transiently binding specific targets with high precision and specificity, but low affinity. These shape-shifting proteins, now known as intrinsically disordered proteins (IDPs), underpin a new paradigm in our understanding of proteins.
Answers from aggregating anomalies
The recognition of IDPs began with observations that were initially dismissed as mere anomalies. For example, different structural biology techniques yielded different structures for the same protein, certain atoms remained unobserved, structural determination produced missing data and unexpected trends were seen in biophysical properties (1). Many factors could plausibly explain these discrepancies, including human error, unmet experimental conditions, or coincidence arising from the ever-changing environment. Consequently, much of the experimental data were dismissed as outliers. It was difficult to cast suspicion on the protein itself, as doing so would have meant challenging a well-established scientific paradigm. Until around 2000, the accumulation of contradictory observations, alongside bioinformatics and protein characterisation studies, consolidated confidence in IDPs’ existence and pointed towards a fundamental shift in how proteins were viewed.
Up until the 2020s, experimental evidence supported that one in three eukaryotic proteins and 3133 proteins in total across the three kingdoms of life contain at least one long (more than 30 amino acids), functional, intrinsically disordered regions (IDR). Given the diversity of their functions and mechanisms, IDPs are further classified into eight functional categories, one of which is ‘unknown’, a reminder of the many questions that scientists have yet to answer (2).
Ordered disorder
We now know that IDPs are characterised by their disordered, ever-changing structures and ability to constantly shift among a dynamic ensemble of conformations. However, does this necessarily mean that IDPs are ‘unstructured’? The evidence suggests almost the opposite. Rather than simply existing in a state of pure randomness, IDPs are designed to be more intricate, much like an articulated doll with more segments or joints, where functions are distributed across many smaller, localised elements, allowing the whole structure to move flexibly and adapt.
Given their conformational variability, one might guess that IDPs are poorly conserved at the sequence level, meaning that their amino acid sequences vary substantially between proteins or across species. To a large extent, this is true. Nevertheless, hidden within these seemingly divergent sequences are conserved short linear motifs (SliMs), which are recurring sequences only a few amino acids long that are essential for controlling particular protein functions. A wide variety of SliMs exist within IDRs, often at conserved positions. To name a few, the FG motif, consisting of just two amino acids, phenylalanine and glycine, is repeatedly found in IDRs of the protein nucleoporin, where these motifs contribute to the selective trafficking of proteins through the nuclear pore complex. Likewise, the KR clusters are clusters enriched in lysine and arginine and function as nuclear localisation signals, directing proteins towards the nucleus on demand (3).
Interestingly, IDPs’ strategy parallels the broader and better-known process by which multicellular organisms become more complex, where functions become increasingly distributed among smaller, more specialised units, rather than being performed by the single, massive entity. One emerging view is also that these SliMs work cooperatively in a concerted way to create interactions, much like the division of labour in larger biological organisations (3).
In addition to their flexibility, another major benefit of this increased complexity is evolutionary efficiency. In general, IDPs tend to evolve faster than ordered proteins. These short, abundant SliMs occur within flexible disordered regions and can easily emerge through just one or two point mutations from nowhere. Consequently, instead of having to evolve the entire complex 3D protein structure, evolution simply has to generate a short linear motif within IDR with point mutations - a kind of evolutionary shortcut to acquiring new functions and growing complexity (4).
Controlled chaos
At high concentrations, IDPs can quickly become harmful, engaging in misinteractions with inappropriate target proteins. It easily creates chaos through their powerful ability to form transient but strong interactions with multiple proteins simultaneously. How, then, can we rely on disorder for critical life functions? Why can evolution afford this amount of chaos, and is it worth it to trade certainty for greater complexity?
The central answer lies in the word ‘control’. In general cases, cells exercise tight spatiotemporal control on IDPs, regulating them to ensure they exist in the right place, at the right concentration, and for the right duration. This regulation begins far before the protein is produced, starting with mRNA transcripts. Transcripts encoding IDPs exhibit higher decay rates, and are disproportionately targeted by small, non-coding RNAs that control translation. Once translated, IDPs are then subject to dynamic fine-tuning. Phosphorylation, for instance, is proposed to alter its net charge and binding affinity, modulating its stability and localisation on demand. On the other hand, IDPs undergo much more ubiquitination (attachment of an ubiquitin protein tag) compared to folded proteins, which renders them inherently sensitive to ubiquitin-proteasomal degradation pathways (5, 6). Evolution can therefore afford the apparent chaos of disorder because of the multiple checkpoints that tightly regulate the localisation, abundance and lifespan of IDPs. By doing so, evolution can trade for more efficient and complicated life processes more safely.
The kaleidoscope of biological order created by IDPs
The tension between disorder and order also appears across a remarkable kaleidoscope of IDP functions. Some IDPs help misfolded proteins in refolding by transiently binding exposed regions and compensating for the unfavourable free energy increase with stabilisation of the intermediates. Others act as entropic chains, maintaining the spatial organisation of protein domains, or as molecular assemblers, engaging multiple binding partners simultaneously to promote formation of higher-order protein complexes, including ribosomes. Still others act as molecular scavengers, storing and neutralising molecules or compounds harmful in large amounts, such as reactive oxygen species and heavy metals. Taken together, these functions suggest that biological order can often arise through disorder. Is biological order even possible without the presence of disorder then? The multitude of functions dependent on IDPs’ flexibility may have shown us that an ever-changing kaleidoscope of uncertainties and possibilities is sometimes what really underpins the appearance of tightly regulated biochemical processes.
The D² concept - Disorder in disorder
Disorder is, however, still disorder, and its flexibility makes it hard for cellular mechanisms to 100% guarantee that it is under control. One prominent risk of IDPs is their propensity to form amyloid fibres, structures notorious for their associations with neurodegenerative diseases such as Alzheimer’s, Parkinson’s and Huntington’s disease. While folded proteins have to undergo thermodynamically unfavourable unfolding or partial-unfolding before transitioning into amyloid states, IDPs already access aggregation-prone conformations. Underlying their delicate balance between flexibility and functionality is the capacity to self-assemble into β-sheet-rich structures, which can progressively aggregate into insoluble, elongated amyloid fibres through various mechanisms. As amyloid-like structures are highly thermodynamically favourable, once it has progressed, it is difficult for IDPs to get out of it (7, 8).
The consequences of loss of control extend beyond amyloid fibril formation because of the ubiquity of IDPs in signalling and regulatory networks. The intricate nature of biochemical processes naturally makes a single fault significant. Dysregulation of IDPs can disrupt and redirect biochemical pathways. For instance, prolonged cellular stress can result in an IDP-mediated protective pathway being altered into one that triggers programmed cell death. Cancer cells can likewise exploit dysregulated IDP-mediated pathways to tolerate elevated levels of reactive oxygen species and facilitate metastasis (8).
Evolution also doesn’t only benefit humans. By exploiting the flexibility of IDP, bacterial pathogens can mimic host proteins with surprising precision, hijacking the host’s own biological machinery for their benefit. In some cases, a bacterium needs little more than a tiny IDP with a few strategically positioned amino acids, to seize control of a host enzyme and rewrite its function. Like a puppeteer pulling the strings of a marionette, the pathogen redirects the enzyme’s activity to serve its own interests, helping it survive, thrive and persist within the host (9).
Evolution neither guarantees flawless solutions nor strives towards a predetermined destination. Biological systems, too, have no fixed formula of how much chaos and control to introduce, only a ceaseless process of balancing chaos and control. Nevertheless, uncertainty never negates the need for change, and nature never stops bringing uncertainty to life. Chaos is brought to the world by evolution, just like courage is brought, like the first flame flickers to life on a barren earth and the first words stumble tentatively from the mouth. None of them are perfectly safe in itself. It is through balanced use that each finds its right place to be.
References
Uversky VN, Kulkarni P. Intrinsically disordered proteins: Chronology of a discovery. Biophysical Chemistry. 2021;279:106694. doi:10.1016/j.bpc.2021.106694
Trivedi R, Nagarajaram HA. Intrinsically Disordered Proteins: An Overview. Int J Mol Sci. 2022;23(22):14050. doi:10.3390/ijms232214050
Nguyen Ba AN, Yeh BJ, van Dyk D, et al. Proteome-Wide Discovery of Evolutionary Conserved Sequences in Disordered Regions. Science Signaling. 2012;5(215):rs1-rs1. doi:10.1126/scisignal.2002515
Pajkos M, Dosztányi Z. Chapter Two - Functions of intrinsically disordered proteins through evolutionary lenses. In: Uversky VN, ed. Dancing Protein Clouds: Intrinsically Disordered Proteins in the Norm and Pathology, Part C. Vol 183. Progress in Molecular Biology and Translational Science. Academic Press; 2021:45-74. doi:10.1016/bs.pmbts.2021.06.017
Babu MM, van der Lee R, de Groot NS, Gsponer J. Intrinsically disordered proteins: regulation and disease. Current Opinion in Structural Biology. 2011;21(3):432-440. doi:10.1016/j.sbi.2011.03.011
Edwards YJ, Lobley AE, Pentony MM, Jones DT. Insights into the regulation of intrinsically disordered proteins in the human proteome by analyzing sequence and gene expression data. Genome Biol. 2009;10(5):R50. doi:10.1186/gb-2009-10-5-r50
Uversky VN, Oldfield CJ, Dunker AK. Intrinsically disordered proteins in human diseases: introducing the D2 concept. Annu Rev Biophys. 2008;37:215-246. doi:10.1146/annurev.biophys.37.032807.125924
Dormann D, Lemke EA. Adding intrinsically disordered proteins to biological ageing clocks. Nat Cell Biol. 2024;26(6):851-858. doi:10.1038/s41556-024-01423-w
Forni D, Cagliani R, Sironi M. Unstructured but dangerous—the emerging roles of intrinsically disordered protein regions in bacterial pathogenesis. FEBS Letters. 2026;600(18):2695-2715. doi:10.1002/1873-3468.70368
Back to
Issue 11
