Resolving Conformational and Rotameric Exchange in

Authors: Michael D Bridges Kálmán Hideg Wayne L Hubbell

Publish Date: 2009/11/14

Volume: 37, Issue: 1-4, Pages: 363-390

PDF Link

Abstract

The function of many proteins involves equilibria between conformational substates and to elucidate mechanisms of function it is essential to have experimental tools to detect the presence of conformational substates and to determine the time scale of exchange between them Sitedirected spin labeling SDSL has the potential to serve this purpose In proteins containing a nitroxide side chain R1 multicomponent electron paramagnetic resonance EPR spectra can arise either from equilibria involving different conformational substates or rotamers of R1 To employ SDSL to uniquely identify conformational equilibria it is thus essential to distinguish between these origins of multicomponent spectra Here we show that this is possible based on the time scale for exchange of the nitroxide between distinct environments that give rise to multicomponent EPR spectra rotamer exchange for R1 lies in the ≈01–1 μs range while conformational exchange is at least an order of magnitude slower The time scales of exchange events are determined by saturation recovery EPR and in favorable cases the exchange rate constants between substates with lifetimes of approximately 1–70 μs can be estimated by the approachHierarchy of protein dynamics A protein in a given state resides in a global free energy minimum in multidimensional conformational space here represented by a single conformational coordinate top panel Within the global conformation of a given state there exist taxonomic substates in equilibrium with exchange lifetimes on the μs–ms time scale center panel Within each taxonomic substate there exist statistical substates of very short lifetimes on the ps–ns time scale bottom panel Transitions between taxonomic substates correspond to conformational exchange while transitions between statistical substates correspond to backbone fluctuations about the average structure of the substateThere is little doubt now that the flexibility of proteins on the μs–ms time scale in many cases is related to function An example is provided by the exchange between the open and closed conformational substates of molecular gates in protein–ligand systems 11 that coordinate ligand binding and dissociation reactions In other cases conformational substates may correspond to specific intermediates exhibited during enzyme catalysis 12 13 generate the diversity of ligand recognition in antibodies 14 15 or result in promiscuity in protein–protein recognition in signal transduction 16 17 Of particular interest is a recent NMR study that showed the conformational substates explored by ubiquitin in solution to constitute the complete manifold of substates involved in promiscuous binding interactions 18Flexibility on the psns time scale due to transitions between statistical substates plays a role in protein–ligand and protein–protein interactions 19 20 21 and provides for flexible regions that enable the larger scale conformational exchange events 22 A remarkable recent discovery is that a substantial number of proteins exist in a completely unstructured state “intrinsically disordered proteins” 23 24 characterized by high backbone flexibility such proteins adopt a particular conformation only upon binding to a structured partner 25 The intrinsically disordered nature allows for rapid searchandbind kinetics as well as the potential for recognition of multiple partners 26To explore molecular mechanisms of protein function it is apparent that experimental techniques capable of monitoring dynamic protein modes over a wide range of time scales are required NMR has led the way in solutionphase studies of relatively small proteins but equivalent measurements on larger soluble proteins membranebound proteins and transient complexes remain challenging and so new methodologies would be welcome to both extend and complement NMRSpin labeling of a cysteine sidechain in a protein A cysteine residue can be modified by the methanthiosulfonate reagent 1oxyl2255tetramethylpyrroline3methyl methanethiosulfonate HO225 or its variants HO1943 HO2101 HO1944 reagent yielding the R1 a R1b b R1f c or RX d sidechains respectively In the case of the RX sidechain two cysteine sites in close proximity are crosslinked by this reagentStructural origin of complex spectra of R1 a Hypothetical example of conformational exchange between two states “A” and “B” related to each other by the outward rotation of a domain In “B” the nitroxide is buried and has little motion relative to protein giving rise to a characteristic broad line shape magenta trace When the domain rotates out motional constraints on the nitroxide are relieved giving rise to a mobile state “A” with a relatively sharp line shape cyan trace In an equilibrium mixture with slow exchange between states the spectrum is a sum of the two components in proportion to their population green trace b Models of two R1 rotamers in which the nitroxide makes immobilizing contacts with the protein in one configuration while the nitroxide retains high mobility in the other This situation will also give rise to a complex spectrum similar to that shown due to the simultaneous presence of both rotamers Rotamer model is based on 115R1 in T4 Lysozyme Protein Data Bank 2OU9Time scale for EPR exchange spectroscopy and some exchange events in proteins Together various EPR detection methods span a wide time range encompassing part or all of the events of interest in proteins For short lifetimes on the T 2 time scale exchange leads to spectral averaging illustrated by the merging of resolved peaks in the CW EPR spectrum arrows lower right spectrum Rotamer lifetimes for some native side chains fall in this time domain 74 Pulsed saturation recovery and relaxation spectroscopy T or Pjump cover the range of lifetimes expected for protein conformational exchange inferred from NMR 9

Keywords: