Files

Abstract

Poly(A)-binding protein (PABP, Pab1 in budding yeast), is a canonical stress granule marker that is consistently recruited to stress granules under a range of stresses. Pab1 autonomously condenses upon heat shock or starvation-induced acidification both in vivo and in vitro, and Pab1's condensation has been shown to be an adaptive stress response which promotes cellular fitness during stressed conditions. It was unclear how Pab1 senses thermal stress or acidification, transduces either stress and condenses. Particularly, Pab1's RRMs are necessary and sufficient for condensation whereas the IDR P-domain is not required, which is opposite to the general view of the field that IDR is essential for condensation. Pab1's condensation with an LCST also contrasts most condensing proteins with a UCST. The aim of this thesis is to investigate the molecular mechanism of Pab1's condensation, through a combination of different biophysical measurements. An outstanding issue of the field is that there is no readily available approach to probe the condensates with high structural resolution, due to that the usual set of high-resolution methods are not applicable to condensates. We address this problem by innovatively applying HDX-MS to Pab1 condensates to probe the hydrogen bonding in the condensates. In Chapter 2, I present HDX-MS data of both monomeric Pab1 and Pab1 condensates, which reveals that RRMs partially unfold upon condensation. RRMs exhibit different levels and patterns of unfolding with RRM3 remaining largely folded. HDX data of condensates exhibits high level of heterogeneity, which points to a structural diversity and interaction heterogeneity in the condensates. HDX-MS also indicates that pH- and Temperature-induced condensates are structurally similar. Our HDX data uncovers that partial unfolding underlies Pab1's condensation process, which supports the view that disorder and quinary interactions underlie the condensation of Pab1 as they do in other systems that condense, just that partial unfolding is an early step of the process. In Chapter 3, I show that RRMs participate differently in condensation. RRMs have different activation temperatures above which they partially unfold and participate in condensation. From this observation we porposed a "sequential activation" model. Co-demixing experiments indicate that the activation is required for an RRM to be engaged in condensation. We term the mechanism "thermodynamic specificity" that an activated RRM will strongly interact only with other activated RRMs rather than inactivated ones. In Chapter 4, I discuss other molecular factors of Pab1's condensation, including histidines' titration, nucleation effect, intramolecular inhibition, and linkers' flexibility. Importantly, we found that histidines partly confer Pab1's sensitivity to pH. Furthermore, SAXS studies show that Pab1's linkers are flexible, and the condensation is not auto-inhibited. FRET further confirms that Pab1 exhibits no significant large-scale conformational change upon condensation. Lastly, I present data from simulation for testing the mode of RRM interactions, which suggests non-specific interactions between RRMs. In Chapter 5, I summarize the findings and conclusions described in previous chapters and address some open questions and future directions. Overall, our multi-facet study investigates the molecular mechanism of Pab1's stress-triggered condensation and highlights the potential of HDX-MS as a novel high-resolution analysis tool for biomolecular condensates.

Details

Actions

PDF

from
to
Export
Download Full History