Probing protein dynamics and function under native and mildly denaturing conditions with hydrogen exchange and mass spectrometry

https://doi.org/10.1016/j.ijms.2004.09.021Get rights and content

Abstract

A combination of hydrogen exchange and mass spectrometry emerged in recent years as a powerful experimental tool capable of probing both structural and dynamic features of proteins. Although its concept is very simple, the interpretation of experimental data is not always straightforward, as a combination of chemical reactions (isotope exchange) and dynamic processes within protein molecules give rise to convoluted exchange patterns. This paper provides a historical background of this technique, candid assessment of its current state and limitations and a discussion of promising recent developments that can result in tremendous improvements and a dramatic expansion of the scope of its applications.

Introduction

Hydrogen–deuterium exchange (HDX) is a technique that became in recent decades one of the major experimental tools to probe both structural and dynamic features of proteins. The analytical value of HDX as a tool for probing macromolecular structure was recognized almost immediately after the discovery of deuterium [1] and the subsequent development of efficient methods of heavy water production [2]. Initial studies of the exchange reactions between small organic molecules and 2H2O carried out by Bonhoeffer and Klar [3] established that hydrogen atoms attached to carbon atoms (e.g., –CH3 groups) do not undergo facile exchange in solution, while the exchange rates for hetero-atoms (e.g., –OH groups) are generally high. Nevertheless, even such labile hydrogen atoms may exchange very slowly if they are not easily accessible by solvent, a situation that is rarely encountered among small organic molecules, but becomes increasingly common as the physical size of the molecule in question increases. Hvidt and Linderstrom-Lang [4], [5] later used HDX to measure the solvent accessibility of labile hydrogen atoms as a probe of polypeptide structure. While this study was the first attempt to use HDX to probe protein structure in solution, the early studies of hydrogen exchange reactions between water and biopolymers preceded this work by more than 15 years [6].

Replacement of a labile hydrogen atom with deuterium (or vice versa) changes two fundamental parameters, namely nuclear spin and mass. The former can be identified using NMR spectroscopy, while the latter can be measured using mass spectrometry. A change in mass also results in a significant alteration of a vibrational frequency, enabling the use of IR spectroscopy to determine the isotope content of a macromolecule. It is probably fair to say that the tremendous popularity enjoyed by HDX as an experimental technique is in large part due to the emergence and spectacular progress of high-resolution NMR [7], [8], [9], [10]. However, mass spectrometry (MS) is currently enjoying a dramatic surge in popularity in this field as well [11], [12], [13], [14]. Interestingly, the idea to use mass measurements as a means to monitor the extent of HDX within macromolecules actually precedes the use of NMR for the same purpose. In the mid-1950s, Burley et al. [15] estimated the extent of 2H incorporation into fibrous proteins by measuring the mass change of the protein sample as a result of HDX. In these experiments, quantitation of 2H content was carried out gravimetrically, i.e., by measuring the mass of an entire protein sample prior to and following the completion of the exchange reactions using a quartz spring. Although this idea was adopted by several other groups in the late 1950s–early 1960s [16], [17], [18], a lack of accuracy afforded by such measurements limited its use. This shortcoming could have been addressed by employing mass spectrometry to directly measure mass changes of macromolecules, however it took several decades before the dramatic technological advances in the field of mass spectrometry enabled desorption and ionization of intact macromolecular species. In the early 1980s McCloskey and co-workers [19] demonstrated that 2H incorporation into a peptide in solution can be accurately determined using fast atom bombardment (FAB) MS. Although initially this methodology was employed only as a means of providing an accurate count of labile hydrogen atoms within a peptide [19], its use has been later expanded to provide information on HDX kinetics in solution [20].

The advent of electrospray ionization (ESI) MS dramatically expanded the range of biopolymers for which the extent of 2H incorporation can be measured directly under a variety of conditions [21]. As a result of these developments, HDX MS has now become a powerful experimental tool for probing protein higher order structure. The number of applications of the HDX MS methodology to probe both architecture and dynamics of biomolecules continues to expand, as it offers several important advantages over HDX NMR. These include tolerance to paramagnetic ligands and co-factors as well as much more forgiving molecular weight limitations, and superior sensitivity, which often allows the experiments to be carried out at concentrations close to or even below the endogenous levels.

Section snippets

Protein structure and dynamics reflected in HDX MS patterns

The concept of HDX experiments is very simple, although the interpretation of experimental data is not always straightforward. Indeed, a combination of chemical reactions (isotope exchange) and dynamic processes within a protein molecule often give rise to convoluted exchange patterns. Labile hydrogen atoms exchange slowly if they are shielded from solvent (e.g., reside in a hydrophobic core of a protein) or involved in a hydrogen-bonding network. In order for the exchange to occur, such

Protein dynamics and function under native conditions: quantitative assessment of protein–ligand binding

The ability of HDX measurements to detect changes in the solvent exposure of polypeptide chains has been used in numerous studies of protein binding processes [30], [31], [32]. The basic premise of such analyses is that protein–protein (or, more generally, protein-large ligand) binding inevitably leads to solvent exclusion from the interface region, resulting in significant reduction of HDX rates for all amides located at the binding interface. In many cases, however, the ligand is too small to

Protein dynamics and function under mildly denaturing conditions: a glimpse at the mechanism

HDX MS measurements carried out under native conditions allow the thermodynamic parameters of the RA-CRABP I interaction to be quantitatively assessed, however, they do not provide direct information on the putative intermediate states of the protein and their involvement in ligand binding. Indeed, the exchange under these conditions proceeds via the EX2 mechanism and, therefore, is uncorrelated. As a result, HDX MS produces a picture of protein dynamics averaged across the entire population.

Understanding how proteins work: site-specific HDX MS measurements and protein ion fragmentation

While HDX MS unequivocally establishes the functional importance of partially unstructured protein states and even affords their detection, it provides little information on their structure. Indeed, site-specific assignment of 2H incorporation is a challenging task due to the reversible character of HDX. Still, there are conditions (pH 2.5–3, T = 0 °C) under which the exchange of backbone amide hydrogen atoms is relatively slow (Fig. 1). Coupling of HDX carried out under native conditions with

Conclusions

Functional importance of transient non-native protein structures can be seen in processes as diverse as recognition, signaling and transport. However, characterization and even distinct detection of such states is a challenging experimental task. Selected examples presented in this paper clearly demonstrate that HDX MS and HDX CAD MS carried out under native or mildly denaturing conditions often provide a unique opportunity to detect and characterize these elusive species in great detail.

Acknowledgements

I wish to thank my group members and colleagues who have been directly involved with various stages of the work presented in this paper: Hui Xiao, Andras Dobo, Stephen J. Eyles and Joshua Hoerner. I am also very grateful to David L. Smith (University of Nebraska), Max Deinzer (Oregon State University), Lars Konermann (University of Western Ontario) and John Engen (University of New Mexico) for very helpful and informative discussions we have had in the past several years. The work presented in

References (50)

  • A. Hvidt et al.

    Biochim. Biophys. Acta

    (1954)
  • A. Hvidt et al.

    Biochim. Biophys. Acta

    (1955)
  • T.M. Raschke et al.

    Curr. Opin. Biotechnol.

    (1998)
  • J.L. Morrison

    Biochim. Biophys. Acta

    (1961)
  • S.K. Sethi et al.

    Biochem. Biophys. Res. Commun.

    (1983)
  • A. Hvidt et al.

    Adv. Protein Chem.

    (1966)
  • S.W. Englander et al.

    Biophys. Chem.

    (2002)
  • C.B. Arrington et al.

    J. Mol. Biol.

    (1999)
  • C.B. Arrington et al.

    J. Mol. Biol.

    (2000)
  • J. Lanman et al.

    Curr. Opin. Struct. Biol.

    (2004)
  • H. Xiao et al.

    J. Am. Soc. Mass Spectrom.

    (2003)
  • X. Yan et al.

    Mol. Cell. Proteomics

    (2004)
  • G.J. Kleywegt et al.

    Structure

    (1994)
  • C.E. Dempsey

    Progr. Nucl. Magn. Res. Spectrosc.

    (2001)
  • H.C. Urey et al.

    Phys. Rev.

    (1932)
  • G.N. Lewis et al.

    J. Am. Chem. Soc.

    (1933)
  • K.F. Bonhoeffer et al.

    Naturwissenschaften

    (1934)
  • G. Champetier et al.

    B Soc. Chim.

    (1938)
  • S.W. Englander et al.

    Annu. Rev. Biophys. Biomol. Struct.

    (1992)
  • H.J. Dyson et al.

    Annu. Rev. Phys. Chem.

    (1996)
  • J.G. Kempf et al.

    Cell. Biochem. Biophys.

    (2003)
  • J.R. Engen et al.

    Anal. Chem.

    (2001)
  • I.A. Kaltashov et al.

    Mass Spectrom. Rev.

    (2002)
  • A.N. Hoofnagle et al.

    Annu. Rev. Biophys. Biomol. Struct.

    (2003)
  • L. Konermann et al.

    Mass Spectrom. Rev.

    (2003)
  • Cited by (0)

    View full text