Probing noncovalent structural features of proteins by mass spectrometry

The temperature dependence of the unimolecular kinetics for dissociation of the heme group from holo-myoglobin (Mb) and holo-hemoglobin α-chain (Hb-α) was investigated with blackbody infrared radiative dissociation (BIRD). The rate constant for dissociation of the 9 + charge state of Mb formed by electrospray ionization from a "pseudo-native" solution is 60% lower than that of Hb-α at each of the temperatures investigated. In solutions of pH 5.5-8.0, the thermal dissociation rate for Mb is also lower than that of HB-α (Hargrove, M. S. et al. J. Biol. Chem. 1994, 269, 4207-4214). Thus, Mb is thermally more stable with respect to heme loss than Hb-α both in the gas phase and in solution. The Arrhenius activation parameters for both dissociation processes are indistinguishable within the current experimental error (activation energy 0.9 eV and pre-exponential factor of 10 8-10 s −1 ). The 9+ to 12+ charge states of Mb have similar Arrhenius parameters when these ions are formed from pseudo-native solutions. In contrast, the activation energies and pre-exponential factors decrease from 0.8 to 0.3 eV and 10 7 to 10 2 s −1 , respectively, for the 9 + to 12 + charge states formed from acidified solutions in which at least 50% of the secondary structure is lost. These results demonstrate that gas-phase Mb ions retain clear memory of the composition of the solution from which they are formed and that these differences can be probed by BIRD.Noncovalent interactions play a critical role in the function of biomolecules in solution [1]. Under gentle conditions, both electrospray ionization [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] and matrix-assisted laser desorption [17][18][19][20] can produce intact gas-phase ions of a variety of noncovalent biomolecule complexes. Recent studies indicate that information about the relative binding affinities of noncovalent complexes in solution can be obtained from the relative abundances of the corresponding ions observed in a mass spectrum [12][13][14][15]21]. Careful control experiments are critical [22], because other factors, such as ion source conditions and gas-phase chemistry, can influence these abundances. An important question is what structural changes occur when noncovalent complexes enter the gas phase, that is, are specific interactions retained in the absence of solvent, and if so, do their bindings strengths reflect those in solution? Recent studies, which probe the gas-phase conformation of ions by using H/D exchange [23,24], collisional cross section [5,25] and ion mobility [26,27], proton transfer reactivity [28,29], and high-energy surface-impact collisions [6], show that many proteins can have a highly compact conformation in the gas phase, although the exact nature of these conformations is not known. These results suggest the possibility that ions can retain some conformational attributes of their solution structure.Noncovalent heme-binding proteins such as myoglobin (Mb) and hemoglobin (Hb) have been extensively studied in solution [30][31][32][...

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

Dissociation of heme-globin complexes by blackbody infrared radiative dissociation: Molecular specificity in the gas phase?

Gross

Zhao

Williams

1997

“…The contribution of the specific complex to the total signal intensity is found to increase with the diacid chain length, which is in agreement with solution behavior. (J Am Soc Mass Spectrom 2002, 13, 946 -953) © 2002 American Society for Mass Spectrometry F or the last ten years, electrospray mass spectrometry (ES-MS) has been used to investigate noncovalent complexes of different classes of compounds [1][2][3][4][5][6]. Electrospray mass spectrometry can be used basically in two different ways, each giving access to a different kind of information.…”

mentioning

confidence: 99%

“…(1) Full-scan simple MS spectra can be recorded with soft conditions to detect the complexes that are present in the infused solution. (2) Once the complexes of interest are isolated in the gas phase, their intrinsic properties (without any influence of the solvent) can be studied. In the present paper, we will focus on the former aspect, namely whether electrospray mass spectra give a faithful image of the complexes that are present in the solution.…”

mentioning

confidence: 99%

On the specificity of cyclodextrin complexes detected by electrospray mass spectrometry

Gabelica

Galić

Pauw

2002

122

116

␣-cyclodextrin complexes with linear ␣, -dicarboxylic acids were investigated by electrospray mass spectrometry. These hydrophobic complexes are known to have an equilibrium binding constant that increases with the diacid chain length. However, the electrospray mass spectrometry (ES-MS) spectra showed that the relative intensity of the complex did not vary significantly with chain length. This contradiction is caused by a contribution of nonspecific adducts to the signal of the complex in ES-MS. In order to estimate the contribution of nonspecific adducts to the total intensity of the complexes with ␣-cyclodextrin, the comparison was made between ␣-cyclodextrin and maltohexaose, the latter being incapable of making inclusion complexes in solution. The signal observed for complexes between diacids and maltohexaose can only result from nonspecific electrostatic aggregation, and is found to be more favorable with the shorter diacids. This is also supported by MS/MS experiments. A procedure is described which allows estimation of the contribution of the nonspecific complex in the spectra of the complexes with ␣-cyclodextrin by using the relative intensity of the complex with maltohexaose. The contribution of the specific complex to the total signal intensity is found to increase with the diacid chain length, which is in agreement with solution behavior. (J Am Soc Mass Spectrom 2002, 13, 946 -953)

“…These considerations imply that the presence or absence of noncovalent complex ions in an ESI mass spectrum does not necessarily allow conclusions to be drawn regarding specific solution phase interactions. Control experiments are often required to rule out false-positive or false-negative scenarios [24,38,39]. It appears that these issues are at least partly responsible for the limited acceptance of ESI-MS as a general tool for "mix and measure" experiments in the context of high-throughput screening tests [3].…”

mentioning

confidence: 99%

Diffusion measurements by electrospray mass spectrometry for studying solution-phase noncovalent interactions

Clark

Konermann

2003

This study describes a novel approach for monitoring noncovalent interactions in solution by electrospray mass spectrometry (ESI-MS). The technique is based on measurements of analyte diffusion in solution. Diffusion coefficients of a target macromolecule and a potential low molecular weight binding partner are determined by measuring the spread of an initially sharp boundary between two solutions of different concentration in a laminar flow tube (Taylor dispersion), as described in Rapid Commun. Mass Spectrom. 2002Spectrom. , 16, 1454Spectrom. -1462. In the absence of noncovalent interactions, the measured ESI-MS dispersion profiles are expected to show a gradual transition for the macromolecule and a steep transition for the low molecular weight compound. However, if the two analytes form a noncovalent complex in solution the dispersion profiles of the two species will be very similar, since the translational diffusion of the small compound is determined by the slow Brownian motion of the macromolecule. In contrast to conventional ESI-MS-based techniques for studying noncovalent complexes, this approach does not rely on the preservation of solution-phase interactions in the gas phase. On the contrary, "harsh" conditions at the ion source are required to disrupt any potential gasphase interactions between the two species, such that their dispersion profiles can be monitored separately. The viability of this technique is demonstrated in studies on noncovalent heme-protein interactions in myoglobin. Tight noncovalent binding is observed in solutions of pH 10, both in the absence and in the presence of 30% acetonitrile. In contrast, a significant disruption of the noncovalent interactions is seen at an acetonitrile content of 50%. Under these conditions, the diffusion coefficient of heme in the presence of myoglobin is only slightly lower than that of heme in a protein-free solution. A breakdown of the noncovalent interactions is also observed in aqueous solution of pH 2.4, where myoglobin is known to adopt an acid-unfolded conformation. (J Am Soc Mass Spectrom 2003, 14, 430 -441)