UC Merced

Proteins are the work horses of the cell that perform the vast majority of functions essential for life. The mechanism by which proteins fold to their functional native state has been a subject of extensive research for more than 50 years now. Downhill folders are the class of proteins whose folding reaction is heterogeneous, non-cooperative, and happens without encountering a significant free energy barrier, resulting in ultrafast kinetics. The single ensemble of conformations of a global downhill folding protein moves gradually from highly disordered to the unique native structure when thermodynamic parameters that affect the protein’s stability are changed (one-state folding). The gradual morphing of a one-state downhill folding protein structure in response to thermodynamic bias is referred to as a conformational rheostat. When such a conformational rheostat is coupled to binding an analyte, it can result in an ultrafast, broad band, and single-molecule analog biosensors. This thesis explores conformational rheostats as the mechanism behind the folding upon binding behavior of intrinsically disordered proteins and as broadband transducers towards engineering high-performance biosensors.The second chapter of this thesis describes a new methodology that we have developed to study the conformational landscape of intrinsically partially disordered proteins (IPDP). This methodology is inspired by the LEGO game, where the sequence of an IPDP is deconstructed into its local structural elements and their possible combinations based on the 3D structure the IPDP acquires upon binding its partners. The local structural elements are hence analogous to LEGO building blocks, and their combinations report on the interactions among them, like the complementary indentations of LEGO pieces. In particular, we chose the IPDP NCBD as model IPDP to develop the proof of concept for the method. Our results showed that even though the NCBD is highly flexible and apparently disordered, there are strong local signals and different sets of long-range transient interactions. These sets of interactions stabilize the overall fold and compete with one another hence resulting in a dynamic ensemble. The methodology developed in Chapter 2 is expected to be extremely useful in characterizing the incipient cooperativity of virtually any IPDP in their unbound form, a capability that is currently unavailable. Chapters 3 and 4 of this thesis deal with the design of a pH biosensor using the downhill folding protein gpW as a scaffold and unfolding coupled to ionization as a transducer. In chapter 3, a methodology for engineering conformational pH transducing into pH insensitive proteins using a histidine grafting approach was developed. The methodology was applied to the protein gpW to demonstrate an engineered, tunable broadband pH transducer based on the conformational rheostat mechanism. Chapter 4 explores general strategies for introducing fluorescence readouts capable of converting the gradual conformational changes of the rheostatic pH transducer into broadband fluorescence-based pH biosensors. Strategies that exploit the Förster Resonance Energy Transfer (FRET) and Photo Induced Electron Transfer (PET) mechanisms were explored as potential means to convert changes in conformation into suitable fluorescence signals were explored and characterized. We discovered that FRET signals using fluorophores in the visible (required for high-sensitivity biosensing) are insensitive to the localized conformational changes associated with conformational rheostats in native-like conditions. In contrast, the very short-range distance dependence of PET (< 1 nm) enabled efficient, high signal-to-noise broadband pH sensing, thus emerging as an extremely useful strategy for implementing rheostatic fluorescence biosensing.