In the mid 1930s, James B. Sumner crystallized the first enzyme –urease from jack beans– and demonstrated that proteins could form crystals whilst keeping their active functional structure. It was difficult at the time to accept that biological molecules could crystallize in the same way as if they were just inanimate minerals. Some 20 years later, Max Perutz and John Kendrew solved the first crystal structures of proteins, hemoglobin and myoglobin, respectively. This marked the birth of, or, it might be better to say confirmed, a paradigm in which proteins are mainly formed of fixed, “ordered”, crystallizable structures, each of which has its own function. One structure for each function. The solution of thousands of these structures over the following fifty years seemed to confirm the paradigm to such an extent that it became something of a dogma: a single protein structure for a specific biological function. We could use the simile that the proteins we knew about at the time had been “tailor-made” for the function they were to perform. However, in the early 1990s, Nuclear Magnetic Resonance was developed, offering levels of resolution sufficient to determine the atomic structure of proteins in solution, with no need for crystallization. This resulted in researchers beginning to unravel the structure of proteins that were difficult to crystallize, detecting proteins that appeared to be disordered in the core of their cells. These proteins contradicted the traditional concept of a fixed and essentially ordered structure. In these proteins, the amino acid chain appears to be unfolded. So, how do they perform their function, if they do not have a specific structure? Beginning to understand this involves a complete paradigm shift: a heresy against the prevailing dogma. However, what we now call Intrinsically Disordered Proteins are not some rarity. In fact, they account for 20% of all proteins in the human body, for example.
The existence of these intrinsically disordered proteins is explained as they only fold when they are going to exercise their assigned function and, furthermore, being disordered, they can adapt to different functions, folding in different ways. The final structure they adopt varies depending on the nature of the molecule with which they are interacting, and the type of function being performed. To summarize, a single protein can perform more than one function. In other words, in this case we are dealing with plastic structures that are not crystallizable. This is a complete paradigm shift. Continuing with our simile, we can now describe these as ready-to-wear proteins, ready for use by different molecules for different purposes. This plasticity is an unexpected level of complexity in the way cells work, opening up multiple new avenues for understanding how the human body works. But this idea of plasticity is nothing new. It was actually proposed by Linus Pauling as long ago as 1940 to try to explain -although mistakenly in his case- the almost unbelievable versatility of the responses of the immune system in vertebrates. Furthermore, this plasticity is inherent to living beings, enabling them to adapt to multiple changing situations without the need for calculations: in reality, this is what engineers and scientists are trying to do when they design artificial intelligence systems, for example.
From this new perspective, analysis of the crystalline structures in the Protein Data Bank (the open-access international database containing the atomic coordinates of all proteins whose structure have been unraveled) reveals that there are thousands of these disordered structures, which are invisible to traditional X-ray diffraction mapping. They are also much more abundant in eukaryotes (organisms whose cells contain a nucleus) than in prokaryotes (no nucleus and, in principle, much less complex), in accordance with the school of thought that proposes that the complexity of living beings resides in the complexity and versatility of their regulating mechanisms, rather than in the number of molecules or genes they contain. And in effect, this disorder is usually associated with regulating functions. These intrinsically disordered proteins are usually involved in “switching on” and “switching off” genes and activating and deactivating enzymes. On the other hand, crystallizable proteins with well-folded structures are usually involved in catalytic processes, in the acceleration of chemical biological reactions.
Unfortunately, this structural disorder is also implicated in triggering illnesses. In summary, what we are talking about is a fine and delicate balance of molecular interactions. And a pathology can result when this balance is disturbed. The role of disordered proteins has been established in many illnesses, such as cystic fibrosis, diabetes, many forms of cancer and devastating neurodegenerative conditions, such as the sadly infamous Alzheimer’s disease. This condition develops with the formation of insoluble protein aggregates deposited in the brain, damaging our neurons. And the proteins we are discussing here are particularly inclined to this, precisely because their disordered structure makes their stability very delicate and means they have a propensity to interact with other proteins in a similar situation. This aggregation can be triggered by the slightest change in their environment. We could say that this aspect is the dark side of the plasticity we have been discussing.
As a result, intrinsically disordered proteins are now starting to be considered as potential targets for new drugs to help improve the symptoms of the illnesses they cause. And to conclude, once again we are looking at a fantastic example of how studying a problem in basic science can make a decisive contribution to our wellbeing and improve our quality of life.
Álvaro Martínez del Pozo
Professor , Universidad Complutense de Madrid
Babu MM, Kriwacki RW, Pappu RV (2012) Structural biology. Versatility from protein disorder. Science 337, 1460-1461.
Bernardó P, Pons M (2010) Proteínas invisibles. El Cultural/El Mundo (http://www.elcultural.com/revista/ciencia/Proteinas-invisibles/28201).
Choi UB, McCann JJ, Weninger KR, Bowen ME (2011) Beyond the random coil: Stochastic conformational switching in intrinsically disordered proteins. Structure 19, 566-576.
Hilser VJ (2013) Structural biology: Signalling from disordered proteins. Nature 498, 308-310.
Pons M (2011) Las proteínas que nos hacen humanos. SEBBM Divulgación: ciencia para todos/El rincón del profesor de ciencias/La ciencia al alcance de la mano (http://www.sebbm.es/archivos_tinymce/octubre2011_miquelpons.pdf).
Rangarajan N, Kulkarni P, Hannenhalli S (2015) Evolutionarily conserved network properties of intrinsically disordered proteins. PLoS One 10, e0126729.
Tompa P (2012) Intrinsically disordered proteins: a 10-year recap. Trends Biochem Sci. 37, 509-516.