Sequence Determinants of the Conformational Properties of an Intrinsically Disordered Protein Prior to and upon Multisite Phosphorylation
J. Am. Chem. Soc. 138, 15323 - 15335
Martin, E.W.,* Holehouse, A.S.* Grace, C.R., Hughes, A., Pappu, R.V., and Mittag, T. (* - co-first authors)
People often think that all 'unstructured' or 'disordered' proteins and protein regions behave in the same way - as unbiased random-coil like ensembles. In fact, disordered regions show a strong conformational preferences in a sequence dependent manner. Our lab (and others) have done a lot of work to try and elucidate and understand this sequence-to-ensemble mapping so given only a protein sequence we can make a reasonable educated guess as to how that region might behave in solution. The yeast transcription factor Ash1 contains a large C-terminal intrinsically disordered region (IDR) that undergoes extensive phosphorylation. Given many IDPs undergo phosphorylation, and phosphorylation provides a mechanism to dynamically change an IDRs primary sequence, we and Tanja's group at St. Jude were interested to determine how this region may change upon phosphorylation.
To make this questions tractable, we sought to determine how an 85 residue region within the Ash1 IDR (Ash1425-500) changed upon phosphorylation at ten naturally occuring sites. Based on its sequence composition, we expected Ash1-425-500 to form a relatively compact ensemble. Upon analysis of unphosphorylated Ash1 by all-atom simulation, small angle X-ray scattering (SAXS), and NMR, we discovered that Ash1-425-500 is actually highly expanded. Moreover, upon ten-fold phosphorylation we almost no change in global dimensions were observed. After extensive further experiments and simulations, we determined two important insights based on these puzzling results.
Firstly, the conformational behaviour of IDRs is strongly influenced by the presence and patterning of proline residues. It had been previously shown that proline had a tendancy to make IDRs more expanded, but in this work we showed through extensive sequence design that the patterning of proline and charged residues relative to other residue types provides a formal order parameter that predicts the degree of expansion. In short, it's not enough to just have a load of proline residues, it matters where in the sequence they are relative to everything else. This order parameter (Omega) is implemented in the localCIDER sequence analysis software.
Secondly, while we didn't observe a change in the global dimensions of Ash1-425-450 upon 10-fold phosphorylation, simulations, and extensive 1D and 2D NMR experiments revealed that there were substantial local changes. Phosphorylation leads to the simultenaous local expansion of some regions and local compaction of others, leading to mutually compensatory changes across the sequence. The upshot is that while the global dimensions are unchanged, locally we see well defined changes to conformational preferences. This is an important result, because previous work from many different labs examining shorter (15-30 residue) IDRs had shown apparently contradictory results, where phosphorylation leads to collapse or expansion. We showed that phosphorylation can induce both compaction and expansion simultaneously within the same disordered protein, suggesting that overall sequence context has an important role in determining how IDR phosphorylation leads to functional changes.
This work started as a discussion at the 2015 Biophysical Society meeting and ended in a substantial body of work (be sure to check out the 46 pages of supplementary information!). Working with Tanja and Erik has been an absolute pleasure, and we are continuing to collaborate on further work extending this system and examining others.