S T U D I E S O F N O N - C O V A L E N T M Y O G L O B I N I N T E R A C T I O N S B Y E L E C T R O S P R A Y I O N I Z A T I O N M A S S S P E C T R O M E T R Y by K E V I N J. M A R K B . S c , University of British Columbia, 2001 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Chemistry) T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A December 2006 © Kevin J. Mark, 2006 Abstract A c i d and base-induced unfolding of myoglobin in solution is monitored by electrospray ionization mass spectrometry. It is shown that acid-induced M b unfolding causes the appearance of different charge state distributions in positive and negative ion modes compared to base-induced unfolding, suggesting different protein conformations in solution. Coll ision cross sections of both apomyoglobin and holomyoglobin ions are measured at various orifice-skimmer voltage differences (AVOS) . The results show that at low A V O S , apomyglobin ions have greater collision cross sections than holomyoglobin ions, indicating that heme, when attached to the globin, helps maintain a more compact myoglobin structure. Coulomb effects in binding of heme in gas phase holomyoglobin ions are studied. Positive and negative ions are formed from solutions of F e + 2 (ferromyoglobin) and F e + 3 (ferrimyoglobin). The energy that must be added to the resulting holomyoglobin ions to cause heme loss is measured with a triple quadrupole M S / M S system. With negative ions, neutral heme is lost, regardless of the charge state of Fe in solution. It is likely that the F e + 3 is reduced to F e + 2 in the negative electrospray -process. Wi th positive ions, predominantly neutral heme loss is observed for ions formed from ferromyoglobin in solution, and positive heme loss for ions formed from ferrimyoglobin in solution. The energies required to induce neutral heme loss are similar for positive and negative ions. The energies required to induce charged heme loss from positive holomyoglobin ions are significantly less. Coulomb repulsion between the s charged heme and charged protein appears to lower the barrier for heme loss. These results are consistent with a simple model of a long range Coulomb repulsion and a short range attraction between the heme and protein. i i i Table of Contents Abstract n Table of Contents iv List of Tables v i i i List of Figures x List of Symbols and Abbreviations x i i i Acknowledgements xx Chapter 1: Introduction 1 1.1 Electrospray Ionization Mass Spectrometry 2 1.2 Studies of Non-Covalent Interactions in Myoglobin 5 1.2.1 Non-Covalent Interactions in Myoglobin 6 1.2.2 Application of Various Analytical Techniques to Study Myoglobin 7 1.2.3 Application of ESI -MS to Study Myoglobin 9 1.3 Studies of Myoglobin Conformations by Mass Spectrometry 11 1.3.1 Myoglobin Conformations in Solution 11 1.3.2 Myoglobin Conformations in the Gas Phase 14 1.4 Tandem Mass Spectrometry of Myoglobin 18 1.5 Coulomb Energy and The Dissociation of Heme From Myoglobin 19 1.6 Goals of This Work 22 iv 1.7 Outline of This Work 25 Chapter 2: Experimental Methods 27 2.1 Electrospray Ionization Mass Spectrometer 27 2.2 Electrospray Ionization 29 2.3 Interface Region '. 29 2.4 Quadrupole Potentials 30 2.5 Ion Motion in a Quadrupole 32 2.6 Instrument Operation 34 2.7 Ion Detection 35 2.8 Myoglobin Solutions and Reagents 37 Chapter 3: Myoglobin Spectra 38 3.1 Negative and Positive Ion Mass Spectra 38 3.2 Myoglobin Conformations in Solution Monitored in Negative Ion Mode 42 3.3 Summary 47 Chapter 4: Collision Cross Sections of Holomyoglobin Ions 49 4.1 Drag Model 49 4.2 Energy Loss Measurements 53 v 4.3 Interpreting the Data From Energy Loss Experiments 54 4.4 Collision Cross Sections of Holomyoglobin Ions from Ferri and Ferromyoglobin Solutions 56 4.5 Summary 58 Chapter 5: Collision Cross Sections of Apomyoglobin Ions 60 5.1 Mass Spectra of Apomyoglobin 61 5.2 Collision Cross Sections of Apomyoglobin Ions 63 5.3 Summary 68 Chapter 6: Coulomb Effects in Binding of Heme to Myoglobin 69 6.1 Tandem Mass Spectrometry of Holomyoglobin Ions 69 6.2 Collision Model 74 6.3 Neutral versus Charged Heme Loss 82 6.4 Comparison of AEint Values 83 6.5 Model Potential for Myoglobin 85 6.6 Summary 88 Chapter 7: Conclusions and Future Work 89 7.1 Conclusions 89 7.2 Future Work References List of Tables Table 2.1 Pressures in the different regions of the mass spectrometer 28 Table 2 .2 Voltages (V) used in positive ion mode M S / M S and collision cross section measurements. For negative ions, the polarity of these voltages was reversed 36 Table 4.1a) Coll is ion cross sections (A 2 ) of negative hMb ions produced from solutions of ferri and ferromyoglobin 56 Table 4.1b) Coll ision cross sections (A2) of positive hMb ions produced from solutions of ferri and ferromyoglobin 57 Table 5.1 Coll ision cross sections (A 2 ) of aMb ions formed in solution at A V O S of 30, 110 and 180 V 65 Table 5.2 Coll ision cross sections (A 2 ) of hMb ions at A V O S of 30 V and 110 V , and aMb ions formed in the gas phase by dissociating hMb ions, at A V O S of 110 and 180 V 6 6 Table 6.1 Dissociation voltages, pressures, collision cross sections and A E i n t values for different charge states of holomyoglobin: a) negative ions from ferriMb solution, b) negative ions from ferroMb solution, c) positive ions from ferriMb slution, and d) positive ions 78 from ferroMb solution Table 6.2 AEint values for a constant reaction time of 18 |is for a) negative v i i i ions from ferriMb solution, b) negative ions from ferroMb solution, c) positive ions from ferriMb solution, and d) positive ions from ferroMb solution List of Figures Figure 1.1 Schematic of ESI and a mass spectrometer interface region 3 Figure 1.2 Structure of heme [39] 6 Figure 2.1 Diagram of the E S I - M S system. ESI is the Electrospray ionization source, QO is the R F only quadrupole, SRO are short rods, Q0/Q1, Q1/Q2, and Q2/Q3 are ion lenses, Q I is the first mass analyzer, Q2 is an R F only quadrupole, Q3 is the second mass analyzer, E X I T is an aperture plate and C E M is the detector 27 Figure 2.2 Electrospray source used in M b experiments 29 Figure 2.3 Schematic of the quadrupole rods. The distance from the center to the rods is r 0 30 Figure 2.4 First stability region of ion motion in a quadrupole field based on the Mathieu equation. Different masses (mi, m 2 , and m3) lie on the operating lines labeled A and B 33 Figure 3.1 Negative ion ESI -MS spectra of a) 1 x 10"5 M ferrimyoglobin, and b) 5 x 10"6 M ferromyoglobin, in a solution of 50% methanol, at p H 7.1. The orifice-skimmer voltage difference was 170 V . Notation: -5h is -5 holomyoglobin (hMb) and -5a is -5 apomyoglobin (aMb) 39 Figure 3.2 Positive ion mode mass spectra of a) 1 x 10"5 M ferriMb, and b) 5 x 10"6 M ferroMb, in a solution of 50% M e O H , at p H 7.1. The A V O S was 170 V . 41 x Figure 3.3 Negative ion mass spectra of ferrimyoglobin ions formed from a solution containing 50% methanol, at pH: a) 3.2, and b) 10.1. A V O S was 170V 43 Figure 4.1 The stopping curves of +6 ferriMb ions at different cell pressures of Ar . The stopping curves from right to left correspond to added argon pressures of 0, 0.3, 0.6, 0.9, 1.2 and 1.5 mtorr, 54 Figure 4.2 E Cnnlm0 55 A plot of - I n — versus — for the +6 ferriMb ion E0 m, Figure 5.1 Mass spectra of apomyoglobin ions, formed from a solution containing 10% M e O H . The concentration of aMb was 1.2 x 10"5 M . The A V O S was a) 30 V , and b) 110 V 61 Figure 5.2 Mass spectra of apomyoglobin ions, formed from a solution containing 10% M e O H . The concentration of aMb was 1.2 x 10"5 M . The A V O S was 180 V -62 Figure 5.3 Coll is ion cross sections versus charge state of positive apomyoglobin ions formed in solution at A V O S (•) 30 V , (o) l 10 V , and ( T ) 1 8 0 V . . . 64 Figure 5.4 Coll ision cross sections vs. charge state for hMb and aMb ions at A V O S of 30 V 67 Figure 6.1 Tandem mass spectra of -7 holomyoglobin ions from a solution of ferrimyoglobin at a collision gas pressure of 1.7 mtorr. The collision energy E is a) 840 eV, b) 1085 eV, c) 1190 eV, and d) 1260 eV 70 xi Figure 6.2 Tandem mass spectra of +7 holomyoglobin ions formed from a solution of ferrimyoglobin, at a collision gas pressure of 1.7 mtorr. The collision energy E is a) 490 eV,.b) 630 eV, c) 770 eV, and d). 875 eV 71 Figure 6.3 Precursor and fragment ion intensities versus Q0-Q2 voltage difference for the -7 holomyoglobin ions formed from a solution of ferrimyoglobin 72 Figure 6.4 Dissociation voltage vs. pressure for negative ferroMb ions, for charges -4h to -7h 74 Figure 6.5 Calculated added internal energies, AEiM , versus reaction time for negative holomyoglobin ions formed from ferromyoglobin solution; -4 ( • ) ; - 5 ( o ) ; - 6 ( Y ) ; - 7 ( V ) 8 0 Figure 6.6 Model potentials for heme binding in neutral and +7 gas phase myoglobin, using Coulomb and Morse potentials 87 x i i List of Symbols and Abbreviations Symbol or Abbreviation Description aMb au,ax,ay A A P C I b B I R D c CD C D C E M CI CID C M C R M d D Peaks of apomyoglobin Sum of radii of the collision partners ApoMyoglobin Mathieu parameter Projection area Atmospheric Pressure Chemical Ionization Distance from the cell entrance Blackbody Infrared Radiation Dissociation Concentration Drag coefficient Circular Dichroism Channel Electron Multiplier Chemical Ionization Collisionally Induced Dissociation Center of mass Charged Residue Model Diameter Cross-sectional diameter Wel l depth (binding energy) Direct Current 3-(dimethylamino) propylamine Elemental charge Kinetic energy of an ion at the collision cell exit Kinetic energy of an ion at the collision cell entrance Stopping energy at one tenth of initial intensity Electric field Internal energies Electron Ionization Electrospray Ionization Aperture plate Ferrimyoglobin Ferromyoglobin Drag force Fast Atom Bombardment Fourier Transform Guandinium chloride Glycine Peaks of holomyoglobin Holomyoglobin Hydrogen/Deuterium His Histidine H P L C High Performance Liquid Chromatography H S P Heat shock protein IaMh Relative intensity of the fragment ion IhMh Relative intensity of the precursor ion ICP Inductively Coupled Plasma ICR Ion Cyclotron Resonance EEM Ion Evaporation Model IMS Ion Mobil i ty Spectrometry kB Boltzmann's constant K Ion mobility constant K0 Reduced ion mobility K(c) Equilibrium constant Kn Knudsen number / Length traveled in the collision cell ld Length of a drift tube lr Length of the collision cell over which reaction occurs Leu Leucine Lys Lysine m Constant mi , mi , iri3 Masses of ions m, Mass of an object xv m 2 Mass of the collision gas m, Mass of an ion mh Mass of a buffer gas ml z Mass to charge ratio M M -mx +m2 M A L D I Matrix Assisted Laser Desorption Ionization M b Myoglobin M e O H Methanol M S Mass Spectrometry M S / M S Tandem mass spectrometry M T B D l,3,4,6,7,8,-hexahydro-l-methyl-2H-pyrimido-[l,2-a] pyridine n Number gas density N Number of Collisions P Pressure qx, q2 Electrostatic charges qu,qx,qy Mathieu parameter Q Quadrupole QO Radio frequency only quadrupole Q l First mass analyzer Q2 Radio frequency only quadrupole/ collision cell Q3 Second mass analyzer xvi Q 0 / Q 1 . Q 1 / Q 2 , Q 2 / Q 3 r ro K D time R F s Ser SIMS SRO t T TI T O F u U U el-stat U H P Ion lenses Distance between heme and protein Distance from the center of a quadrupole to the rods Distance where the Morse potential has a minimum Distance between charges Reynolds number Reaction time Radio frequency Speed ratio Serine Secondary Ion Mass Spectrometry Short rods Time Drift time Temperature Thermal Ionization Time-of-Flight x or y D C voltage Electrostatic potential Ultra high purity Speed of an ion xvi i v 0 Velocity of an object relative to the gas ve Speed of ions entering the collision cell v, Speed of ions leaving the collision cell V R F voltage Vd Drift velocity V(r) Potential z Number of charges on the ion P Constant Ab Change in distance A E a Difference in activation energy AjE i n t Added internal energy A G Change in free energy of binding A V O S Orifice-skimmer voltage difference £ 0 Permittivity of vacuum heme + unfolded aMb This was confirmed by investigating the change in charge state distributions of hMb and aMb ions with the change in p H [66]. The gradual shift in charge states, with a change in pH from 8.5 to 2.5, indicated a non-cooperative myoglobin unfolding process which included multiple protein conformations. Conformations of myoglobin ions in solution have been studied by H / D exchange. Rates of exchange of backbone amide hydrogens were measured for both aMb and hMb ions [68]. It was found that the percentage of deuteration for aMb was four times higher than the percentage of deuteration for hMb, indicating a more compact conformation of hMb. Neutral and acidic p H values were used to investigate aMb conformations, by monitoring different charge state distributions and online H / D exchange [67]. 13 In the work presented in this thesis, conformational changes of aMb and h M b in solution are investigated with negative ion M S , through observations of differences in charge state distributions of ions produced from solutions at different p H values. 1.3.2 Myog lob in Conformations i n the Gas Phase Determination of the size of protein ions in the gas phase may lead to better understanding of gas phase protein ion conformations. Initial experiments to evaluate the size of chemical compounds were performed by Mack [88]. Average cross section areas were determined for iodine, benzene, naphthalene, anthracene, toluene, aniline and benzidine by measuring their rates of evaporation. Measurement of protein cross sections is one mass spectrometric method to probe gas phase protein conformations. Cross sections have been measured by ion mobility and ion energy loss. Ion mobility spectrometry (IMS) was first applied as an analytical method to separate ions by Cohen and coworkers in 1970 [89]. In this technique, gas phase ions move through a drift tube under a constant electric field, while an inert buffer gas flows opposite to the drift direction of the ions of interest. The ion drift time can be recorded by a plasma chromatograph [89], or an ion detector following the mass analyzer [90]. Several mass analyzers have been used in combination with the drift tubes such as magnetic sectors [91], F T - I C R [92], ion traps [93], quadrupoles [94] and T O F instruments [95]. In general, ions with more folded structures have smaller cross 14 sectional areas, undergo fewer collisions with the buffer gas and therefore have larger ion mobilities. This is a unique feature, since ions with identical masses can be separated based on their sizes. The average drift velocity of an ion, Vd, is determined by the number of collisions with the buffer gas in the drift tube. The drift velocity is proportional to the electric field Eel: where K is the ion mobility constant. As ions pass through the drift tube with length ld , K can be calculated from where td is the drift time. To account for different temperatures and pressures, a reduced mobility K0 is calculated from: Vd=KE, el (1.1) K = (1.2) K0=K P 273.15 (1.3) 760 T 15 where P is the gas pressure in torr and T is the gas temperature in degrees Kelv in . Substitution of equation 1.2 into equation 1.3 gives the following expression for the reduced ion mobility constant: j ^ j L . v m (i.4) tdEa 760 T To obtain information about conformation, mobility can be converted to a collision integral: K = — [— (r——) ^ (—)] (1.5) 16 n fJkBT Q m:mh where n is the number gas density, e is the elemental charge, / /= —'• is the reduced mass, mi is the ion mass, mh is the buffer gas mass, T is the temperature, kB is Boltzmann's constant, and Q is the averaged collision integral [96, 97]. In the case of diffuse (inelastic) scattering of large ions like proteins, the collision integral is given by: £l=L22m2=l.22A (1.6) where a is the sum of the radii of the collision partners, and A is the projection area [98]. 16 Ion mobility measurements were used to distinguish different conformations of aMb ions in the gas phase [69]. The aMb charge states ranged from +4 to +22. The collision cross sections for the charges +8 to +22 of aMb ions were from 2800 to 3800 A 2 . High proton affinity bases ( D M A P A and M T B D ) were added in the interface region to "strip" protons from high charge states, to produce low charge state aMb ions. The collision cross sections for +4 aMb to +7 aMb ranged from 1500 to 1700 A 2 . It was concluded that "proton stripping" of higher charge state unfolded ions causes a spontaneous collapse into a partially folded conformation in the gas phase. The second method used to determine protein cross sections by M S is a measurement of the kinetic energy losses of ions [70, 71]. This technique has lower resolution than ion mobility [99]. Unlike IMS, energy loss experiments can be used to study weakly bound non-covalent complexes [59, 100]. Covey and Douglas [71] first reported the determination of collision cross sections by energy loss experiments with an ESI-triple quadrupole system. They found that an increase in charges on the ion leads to an increase in collision cross sections. The collision cross sections of proteins interpreted by ion mobility and by energy loss have been compared [70]. Douglas included an aerodynamic drag coefficient [101] that takes into account the thermal motion of the collision gas. Average center-of-mass scattering angle of 90° is assumed [71]. The drag coefficient model has been applied to calculate collision cross sections of gas phase ions of hMb and aMb [59, 72], cytochrome c [70], cytochrome c-cytochrome b$ complexes [100], D N A [99], and enzyme-inhibitor complexes [102]. 17 Collision cross sections of hMb and aMb ions have been measured at several orifice-skimmer voltage differences (AVOS) [59], with an ESI-triple quadrupole system. B y increasing the A V O S from 30 V to 180 V , the intensity of aMb ions in the mass spectrum increased, accompanied by a decrease in hMb ion intensity. Cross sections were measured for the +8 to +14 charge states of both aMb and hMb. For a given charge state, the increase in A V O S caused both aMb and hMb ions to unfold to give larger cross sections. 1.4 Tandem Mass Spectrometry of Myoglobin Tandem mass spectrometry is a process of multiple steps of mass analysis separated by ion fragmentation. It is used to provide structural information, by fragmenting an analyte ion and identifying the resulting fragments. Tandem mass spectrometry has been grouped into two methods, tandem in-time and tandem in-space [103]. In-time refers to mass analysers which can store, select, and fragment precursor ions such as quadrupole ion traps [104] and ion cyclotron resonance (ICR) traps [105] . In-space refers to at least two consecutive mass analyzers such as magnetic sectors [106], quadrupoles [107] and T O F [108]. In an ESI triple-quadrupole M S / M S system, an ion is selected in a first mass analyzing quadrupole. The ion then undergoes collisions with a neutral gas in a collision cell where it fragments. The product ions from the fragmentation are scanned by a second mass analyzing quadrupole. 18 One of the first tandem mass spectrometry investigations of myoglobin was done with an ESI-triple quadrupole system in 1994 [58]. Solutions of both ferri and ferromyoglobin were prepared for a comparative study. It was shown that fragmentation of positive ferriMb ions gives mostly charged heme loss, whereas fragmentation of positive ferroMb ions gives mostly neutral heme loss. Fragmentation of the +10 h M b ions, formed from a solution mixture of ferrimyoglobin and ferromyoglobin yielded +10 aMb and +9 aMb, as well as singly charged heme ions. Also in 1994, McLuckey and Ramsey used an ion trap to perform collisional activation of the +8 hMb ions in M S / M S experiments [57]. They observed a charged heme loss for positive ferriMb ions. Since then numerous M S / M S studies on myoglobin have been performed to investigate different aspects of its dissociation [60, 72, 75, 109]. 1.5 Coulomb Energy and the Dissociation of Heme From Myoglobin The Coulomb energy is the electrostatic potential energy between two charged bodies (for example heme and globin) and is given by: where e0 is the permittivity of vacuum, qx and q2 are the electrostatic charges and r is the distance between them. The Coulomb energy is positive for like charges, where the 19 force between the charges is repulsive. With opposite charges, the force is attractive and the Coulomb energy is negative. Coulomb effects can influence the dissociation of heme from gas phase myoglobin ions, because heme can leave the globin with an overall charge o f -1 , 0 or +1 [58, 109]. Simplistic arguments suggest that with ferrimyoglobin, the heme group wi l l have an overall charge of +1, and this has led some authors to speculate on the effects of this on the gas phase binding of heme to the multiply charged gas phase protein ions. Dissociation of +10 holomyoglobin ions, formed from a mixture of ferrimyoglobin and ferromyoglobin in the collision cell of a triple quadrupole system, was performed by Konishi and Feng [58]. The ratio of neutral to charged heme loss was independent of energy or collision gas target thickness, and it was concluded that ferrimyoglobin and ferromyoglobin had the same gas phase stability. McLuckey and Ramsey [57] speculated that loss of "Coulombic strain" would favour loss of charged heme from positive ferrimyoglobin ions particularly in higher charge states, and this was consistent with their observation of charged heme loss from +8 holomyoglobin ions in a trap M S / M S experiment. Schmidt and Karas [75] observed the preservation of negative holomyoglobin ions under "harsh" conditions (capillary 200°C, 20 V skimmer-octopole CID) , which led to loss of heme from positive holomyoglobin ions They concluded that negative ions were more stable because of Coulombic attraction between the positive heme and negative globin. 20 The overall charge on the heme group in solution is complicated by the presence of two propionic acid groups on the heme [37]. When iron is bound to a neutral free porphyrin, the porphyrin loses two protons and has a net charge of -2. If F e + 2 is bound, the overall heme group wi l l be neutral, and i f F e + 3 is bound, the heme wi l l have an overall charge of + 1 [110]. The propionic acid groups can also be charged, depending on solution pH. The p K a of these acid groups when heme is bound in the protein wi l l differ from the p K a of free heme ( p K a s o l u t l o n = 4.4) [111]. In cytochrome c the effects of binding the heme has been investigated. Wright et al. reported that the p K a values remain less than 6.5 for three cytochrome c variants [112]. Electrostatic calculations show the propionate groups have p K a values from about 4.0 to 6.5 in a variety of cytochromes [113]. The p K a values for the propionate groups of the heme in myoglobin have not been reported. Nevertheless, the groups are usually considered ionized at neutral p H [37]. The propionates are often described as participating in hydrogen bonds with the protein; one propionate shares a hydrogen with Lys45, and the second with Ser92 and Leu89 [39]. Thus, in solution depending on the description of the hydrogen bonding, the degree of ionization of the propionate groups and the Fe oxidation state, the heme can be considered to have an overall charge from -2 to +1. Measurement of the energies needed to dissociate the heme-globin complexes can give insight into Coulomb interactions within the complexes. Chen et al. [72] have introduced a collision model (described in detail in Chapter 6), to calculate the relative energies transferred to complexes to cause their dissociation in tandem mass spectrometry experiments. They applied this model to hMb and measured the relative energies 21 necessary to dissociate heme from myoglobin, for the charge states from +8 to +21. It was found that for lower and higher charge states, the calculated internal energies are similar. Another approach to investigate the Coulomb effect, in the process of heme loss from myoglobin, is the measurement of activation energies for heme loss by B I R D experiments. Gross and coworkers [60] studied the decomposition of h M b ions (charge states +9 to +12) in an ICR cell. They reported similar activation energies for the heme loss from the +9 to +12 hMb ions. These two types of experiments showed only small differences in heme binding energies between the low charge states, which therefore have similar Coulomb barriers for dissociation. 1.6 Goals of This Work Studies of non-covalent protein complexes by negative or positive ion E S I - M S can give insight into solution and gas phase protein conformations, their folding and unfolding mechanisms, energies required to dissociate these complexes, as well as the Coulomb effects on their dissociation. As discussed above, of the many non-covalent complexes, myoglobin has been extensively studied by various mass spectrometric techniques. In this study, collision cross sections and tandem mass spectrometry in positive and negative ion mode E S I - M S are used to investigate Coulomb effects on the dissociation of heme from myoglobin. Previously, acid and methanol-induced protein unfolding were studied by positive ion M S for lysozyme, cytochrome c, myoglobin and ubiquitin ions [64-66]. In addition, 22 comparison of unfolding for several proteins was investigated in positive and negative ion M S [76]. In this study, different M b conformations in solutions at low and high p H were investigated, by negative ion M S . The solution p H was changed, and different charge state distributions which indicate different protein conformations were observed. Here, conformations of gas phase negative myoglobin ions are probed by collision cross section measurements. Previously, Douglas and Collings [59] found that positive myoglobin ions can be activated in the orifice-skimmer region of an E S I triple-quadrupole mass spectrometer system. They dissociated hMb in the orifice-skimmer region to produce aMb ions and measured cross sections of the ions at three A V O S values. At 30 V , only hMb was observed. A t 110 V , both aMb and hMb were observed having similar cross sections. At 180 V , only aMb ions were observed, having cross sections similar to cross sections of aMb and hMb ions at 110 V . It was found that cross sections of hMb at 30 V were lower than cross sections of aMb and h M b at 110 V and 180 V , indicating more compact conformations of the protein with the heme bound at lower A V O S . Part of the work of this thesis describes a comparison of the collision cross sections of aMb ions, formed from a myoglobin in solution and measured at A V O S of 30 V , with the collision cross sections of hMb ions measured at the same A V O S . At A V O S of 30 V , it is not possible to form aMb ions from holomyoglobin ions in the orifice-skimmer region. It is of interest to measure the cross sections of aMb at A V O S of 30 V , in order to compare them directly to those of hMb ions at that same A V O S . In this way, aMb ions do not gain excess energy that can cause them to unfold. The observed result contributes to a better understanding of the unfolding mechanism of myoglobin 23 during heme dissociation, and the role of heme binding in the stabilization of the myoglobin complex. Gas phase non-covalent complexes are free of the influence of solvent, and therefore it has been suggested that their dissociation is greatly influenced by Coulomb interactions between the charges on the gas phase ions [114]. Coulomb effects on the dissociation of non-covalent gas phase complexes have been reported for many protein multimers [115-121]. In many cases, it was found that the dissociation results in an asymmetric distribution of charges and masses between the fragments. These include homodimers of ecotin [115], cytochrome c and a-lactalbumin [117], tetramers of streptavidin [116], avidin, concavalin A and human hemoglobin [118, 120], pentamer of Shiga-like toxin I [119], and a 24-mer of heat-shock protein HSP16.5 [121]. The asymmetry of charge distributions in the fragment ions of all these complexes was found to depend on the precursor charge state, internal energy and conformation. It is of interest to investigate Coulomb effects in the dissociation of other types of gas phase non-covalent complexes, like protein small-molecule complexes. Myoglobin is particularly suited as a model system for this study, because the charge can be at least partially controlled by oxidizing or reducing the iron heme, which can therefore provide a more detailed study. Coulomb effects could in principle, either increase or decrease the barrier for heme loss, depending on the position of charges in the heme-protein complex and in the transition state for heme loss. The influence of Coulomb effects on the energies added to the ions to cause dissociation is discussed in the last chapter of this thesis. In principle, similar experiments could be performed with other protein small-molecule complexes, i f the 24 charge on the small molecule can be controlled in a systematic way. Measured collision cross sections, along with tandem M S experiments, are used to determine the energies necessary to dissociate positive and negative hMb ions. A simple model potential is proposed, to take into account the differences in Coulomb energies between neutral and charged heme loss from myoglobin. The M S / M S results for different oxidation states of the heme iron are of interest, since tandem mass spectrometry experiments on ferro and ferriMb ions show different heme loss pathways in positive ion mode M S [58, 109]. The possible differences in binding between heme F e 2 + and F e 3 + and the globin complicate the interpretation of the M S / M S experiments, but the simplest conclusion is that Coulomb repulsion lowers the barrier for heme loss. 1.7 Outline of This Work Chapter 2 details the basics of quadrupole theory, the ESI triple-quadrupole apparatus and other experimental details. Chapter 3 shows positive and negative ion mass spectra of ferri and ferromyoglobin in solution, at p H 7.1. A discussion is given of the conformation of ferrimyoglobin in solution, during acid and base-induced equilibrium unfolding, as monitored by charge states in negative ion E S I - M S . 25 Chapter 4 shows the results of collision cross section measurements for positive and negative hMb ions, produced from ferri and ferroMb in solution. The positive and negative ion cross sections are compared. Chapter 5 contains the results of collision cross section measurements of positive aMb ions formed upon removal of heme from hMb in solution, prior to E S I - M S . Cross sections of these ions, measured at different A V O S which can induce unfolding of ions, are compared with the cross sections of positive aMb ions formed by dissociating ferriMb ions in the orifice-skimmer region of the mass spectrometer. Removal of the heme prior to ESI -MS provides the possibility for a direct comparison of cross sections for aMb and hMb at a low A V O S , when aMb is not unfolded. Chapter 6 describes measurements of the added energies necessary to dissociate heme from positive and negative ions of hMb in the gas phase. Calculation of the internal energies for heme dissociation can be used to examine the effects of a neutral and charged heme loss, and whether any heme loss pathways are influenced by Coulomb effects. The collision cross section data from Chapter 4 are used, along with dissociation voltages obtained from tandem mass spectrometry experiments, to calculate these energies. The different internal energies needed to cause neutral and charged heme loss from myoglobin are rationalized by differences in Coulomb energy. A model for heme binding is described using Coulomb and Morse potentials. 26 Chapter 2 Experimental Methods This chapter includes an overview of the electrospray ionization triple quadrupole mass spectrometer operation. It also describes the materials and methods used to prepare myoglobin solutions. 2.1 Electrospray Ionization Mass Spectrometer collision gas ESI source,.' ill EXIT QO A SRO Qi • Q2 Q3 CEM zr curtain' plate • Q0/Q1 Q1/Q2 Q2/Q3 orifice skimmer Figure 2.1 Diagram of the ESI -MS system. ESI is the electrospray ionization source, QO is the R F only quadrupole, SRO are short rods, Q0/Q1, Q1/Q2, and Q2/Q3 are ion lenses, QI is the first mass analyzer, Q2 is an R F only quadrupole, Q3 is the second mass analyzer, E X I T is an aperture plate and C E M is the detector. 27 In the ESI-MS system used here [70], gas phase ions formed at atmospheric pressure by ESI pass through an orifice in a curtain plate, a N2 curtain gas, and a sampling orifice (0.25 mm diameter), followed by a skimmer (0.75 mm diameter orifice). Ions then enter an RF (radio frequency)-only quadrupole (Q0), a first quadrupole mass analyzer ( Q l ) , an RF-only quadrupole or collision cell (Q2), a second mass analyzer (Q3), followed by a channel electron multiplier (CEM) for ion detection. The low pressures of the system are achieved by a three stage differential pumping system. A mechanical pump is used for the interface region and one turbo molecular pump is used for each of the Q 0 chamber and the main chamber containing the quadrupoles. Table 2.1 gives the background pressures of the different regions in the MS system. Region Pressure (torr) Curtain-Orifice 760 Orifice-Skimmer 1.3 Q 0 Chamber 4 x 10"J Main Chamber 8 x 10"b Table 2.1 Pressures in the different regions of the mass spectrometer. 28 2.2 Electrospray Ionization epoxy glue solution F t fused silica fused silica capillary capillary stainless steel tube Figure 2.2 Electrospray source used in M b experiments. Figure 2.2 shows a schematic of the homemade electrospray source. The sprayer consists of a 3 cm long fused-silica capillary (Polymicro Technologies, Phoenix, A Z ) , with an inner diameter (i. d.) of 74 pm, and an outer diameter (o. d.) of 145 pm, connected to a 5 cm long stainless steel tube (Small Parts Inc., Miami Lakes, F L ) , with an i . d. of 0.02 cm, and an o. d. of 0.04 cm. A syringe pump (model 22, Harvard Apparatus, South Natick, M A ) infuses solutions into the source through a capillary, at a flow rate of 1 pL/min. High voltages (-2.0 to 4.5 k V ) are applied to the stainless steel tube. 2.3 Interface Region A s shown in Figure 1.1, charged droplets formed at atmospheric pressure travel towards the curtain plate inlet. Nitrogen (curtain gas) is passed through the region between the 29 curtain plate and the orifice. The curtain gas helps desolvate ions and prevents solvent from entering the first vacuum pumping stage. The flow of the curtain gas is important. Too high a flow may lead to a loss of analyte ions, while too low a flow could cause formation of analyte-solvent adducts. After passing through the orifice, ions are accelerated by a potential difference between the orifice and the skimmer. The curtain plate is held at 1000 V , the orifice at 290-320 V , and the skimmer at 120-150 V . Ions and gas pass through the skimmer and into the Q0 chamber. Operation of the quadrupoles is described below. 2.4 Quadrupole Potentials Figure 2.3 Schematic of the quadrupole rods. The distance from the center to the rods is e D C and R F voltages o-30 The ideal quadrupole mass filter consists of a set of four parallel hyperbolic rods with D C (direct current) and R F (radio frequency) potentials applied to the electrodes [122], as shown in Figure 2.3. Each pair of opposite rods is electrically connected, providing them with the same polarity. One pair of rods has an applied potential: O 0 =U-Vcos{ox) (2.1) where U is the D C voltage rod to ground, V is the zero to peak amplitude of the R F voltage between each rod and ground, co is the angular frequency of the R F voltage, t is time, and r 0 is the minimum distance from the center of the quadrupole to any of the rods. The other pair of electrodes has the opposite potential: - O 0 = U-Vcos{ax) (2.2) In two dimensions, the potential at any point (x, y) within the quadrupole is given by: f 2 2^ x - y V ro J O 0 (2.3) At the center of the quadrupole, y) = 0. 31 2.5 Ion Motion in a Quadrupole The motion of ions in the x -direction of a quadrupole field is determined by: ^ + - ^ y {U-V cos{ax))x = 0 dt miro where mi is the mass of an ion, and e is the elemental charge. Similarly in the direction: ^ Z _ ^ £ _ ( r 7 _ y c o s M ) ^ 0 at mirg The motion of ions in the x and y directions are independent. Let: __8eU_ y 2 2 4eV q* = -q, = — r -From equations (2.4) to (2.8), the equation of motion can be written as: ^ + ( a „ - 2 ^ c o s 2 ^ ) u = 0 (2.9) where u - x or y . Equation 2.9 is known as the Mathieu equation. The solutions of the Mathieu equation give the trajectories of the ions which can be "stable" or "unstable". Stable motion refers to the ion displacement in the x and y directions being less than r0, for all values of t. In this case, the oscillation of the ion through the quadrupole wi l l be confined within the rods, and the ion wi l l be transmitted. Conversely, unstable ion motion results in the ion striking the rods and not being transmitted through the quadrupole. For the set of different values of the Mathieu parameters, au and qu, a stability diagram like one presented in Figure 2.4, can be constructed [122, 123]. °U A 0.2 0.4 0.6 0.8 Figure 2.4 First stability region of ion motion in a quadrupole field based on the Mathieu equation.. Different masses (mi, m 2 , and 1113) lie on the operating lines labeled A and B . 33 Dividing equation 2.6 by equation 2.7 gives = —. Since this ratio is independent of V q m/z, masses of all the ions are located on the same line, called the "operating line" or "mass scan line". The — value is the slope of the operating line. A series of operating q 2U lines can be created by changing the ratio . One operating line corresponds to one resolution setting. In Figure 2.4 three ions: m i , m 2 , and m3 with different m/z ratios, lie on the operating line with mi < m 2 < m 3 . In the case of the operating line A , m 2 is stable in both the x and y directions, and therefore the ion m 2 is transmitted through the quadrupole. In the case of the operating line B , all three ions lie within the stability region and are transmitted. The operating line with the larger slope (line A ) results in higher resolution, because a shorter length of the operating line lies within the stability region. As a result, ions of mass m 2 are separated from ions with masses mi and m3, as both mi and m 3 lie in regions of instability. In the case of the operating line B , all three ions are transmitted with lower resolution. 2.6 Instrument Operation The triple quadrupole mass analyzer consists of four quadrupoles. After ions pass through the skimmer, they enter the QO quadrupole. Only R F voltage is applied between the rods of QO. Ions are transported to the first mass analyzer Q I , where ions with selected m/z ratios are transmitted. B y scanning the voltages on Q I , ions with different 34 m/z are transmitted through Q2 and Q3 to the detector to produce a Q l mass spectrum. While Q l is scanning, both Q2 and Q3 are in RF-only mode. Similarly, a Q3 mass spectrum can be generated by scanning the R F and D C voltages on Q3, to transmit different m/z ratios. In this case, Q l is operated in RF-only mode. In collision cross section experiments, Q l was operated as an ion guide in RF-only mode. Kinetic energies of ions leaving Q2 were determined by increasing the rod offset voltage of Q3, until the ion signal was decreased by several orders of magnitude for each pressure of argon. In M S / M S experiments with myoglobin, a product ion scan mode was used. Holomyoglobin ions that were mass selected in Q l , collided with argon in Q2 to cause dissociation. Fragment ions were mass analyzed in Q3. Coll ision energies of the hMb ions were determined by the difference between the QO and Q2 rod offset voltages. The Q3 rod offset was kept the same as the Q2 rod offset. The voltages used for tandem mass spectrometry and collision cross section experiments are listed in Table 2.2. When changing from positive to negative ion mode, all polarities were reversed. 2.7 Ion Detection A channel electron multiplier ( C E M ) was used as the ion detector. It is capable of detecting positive and negative ions. The C E M consists of a continuous dynode. When an ion strikes the surface of the dynode, secondary electrons are released. This process 35 continues as secondary electrons strike the dynode to further release more electrons, which create an electron cascade to produce a current. The total number of ions with different m/z ratios was recorded and presented as a mass spectrum. When changing from positive to negative ion detection, only the polarity was changed on the detector. M S / M S experiments Coll is ion cross section experiments ESI + 3500 +3500 Curtain +1000 +1000 Orifice + 320 +290 Skimmer +150 +120 QO +140 +110 Q0/Q1 +130 +105 Q i + 125 +100 Q1/Q2 +125 +95 Q2 Varied +100 Q2/Q3 Varied +100 Q3 Equal to Q2 Varied C E M -4800 -4800 Table 2.2 Voltages (V) used in positive ion mode M S / M S and collision cross section measurements. For negative ions, the polarity of these voltages was reversed. 36 2.8 Myoglobin Solutions and Reagents Horse heart myoglobin was purchased from Sigma (St. Louis, M O , U S A ) . Ferrimyoglobin (Fe + 3 ) was reduced to ferromyoglobin (Fe + 2 ) by mixing equal volumes of 1.6 x 10"3 M sodium L-ascorbate from Sigma (St. Louis, M O , U S A ) and 1 x 10"5 M myoglobin solutions. Reaction was allowed to proceed for 30 minutes [109]. The reduction of iron was verified by recording U V spectra (Beckmann D U 800). Apomyoglobin from equine skeletal muscle was purchased from Sigma (St. Louis, M O , U S A ) . The aMb was 1.2 x 10"5 M in 10% methanol. HPLC-grade methanol, acetic acid and ammonium hydroxide were purchased from Fisher Scientific (Nepean, O N , Canada). Argon (99.9999%, manufacturer's stated purity) and nitrogen (99.999%, manufacturer's stated purity, U H P grade) were from Praxair (Mississauga, O N , Canada). Solution p H was measured with an Accumet p H meter (Fisher Scientific, model 15, Arvada, CO) . 37 Chapter 3 Myoglobin Spectra The first part of this chapter shows positive and negative ion mass spectra of myoglobin ions from ferri and ferromyoglobin solutions at p H 7.1. The second part focuses on negative myoglobin ions from ferrimyoglobin solutions with different p H values, to induce unfolding. The ion charge states which can give an indication of protein conformations are monitored by E S I - M S . 3.1 Negative and Positive Ion Mass Spectra Mass spectra of negative M b ions, formed from ferri and ferromyoglobin solutions, are shown in Figures 3.1 a) and 3.1 b), respectively. The solutions contained 50% M e O H and the p H was measured to be 7.1. The orifice-skimmer voltage difference was 170 V . Under these experimental conditions, low charge states of holomyoglobin dominate the spectra. In the mass spectrum of ferriMb, shown in Figure 3.1 a), holomyoglobin ions have peaks at the following m/z ratios: 4393 (-4h), 3514 (-5h), 2928 (-6h), and 2510 (-7h). Corresponding to the hMb peaks, are lower intensity aMb peaks, at m/z ratios: 3389 38 CO C CD > CD CC 100 1000 5000 co 20 0 i 100 H - 1 0 - " cm" m. Figure 4.2 A plot of - In — - versus E m for the +6 ferriMb ion. are 120, 108, 96, 87 and 76 eV. The ratios — can be calculated accordingly. For the E° +6 hMb ion, the CD value is 2.5. Drag coefficients are determined from the graph of CD 55 versus s in [ 1 2 8 ] . The collision cross section of 1 1 6 7 A2 is determined from the slope of the plot - In — versus — - — — , shown in Figure 4 . 2 (R 2 = 0 . 9 9 9 5 , y intercept is EQ m, + 0 . 0 0 4 6 ) . 4.4 Collision Cross Sections of Holomyoglobin Ions from Ferri and Ferromyoglobin Solutions Collision cross sections measured for holomyoglobin ions from solutions of ferri and ferromyoglobin are shown in Table 4 . 1 . The cross sections are averages of three experiments performed on separate days. Errors are standard deviations. Charge Ferrimyoglobin Ferromyoglobin - 4 1 2 5 2 ± 9 6 1 2 7 1 ± 1 0 2 - 5 1 2 6 1 ± 7 4 1 2 7 0 ± 9 7 - 6 1 2 7 1 ± 2 1 1 2 5 9 ± 1 1 0 - 7 1 4 6 7 ± 5 4 1 4 1 6 ± 7 5 Table 4.1 a) Coll is ion cross sections (A 2 ) of negative hMb ions produced from solutions of ferri and ferromyoglobin. 5 6 Charge state Ferrimyoglobin Ferromyoglobin +5 1278 ± 1 4 9 1265 ± 5 7 +6 1235 ± 9 8 1264 ± 8 1 +7 1594 ± 5 4 1515 ± 1 0 1 +8 1777 ± 7 9 1791 ± 3 2 +9 1812 ± 2 0 Table 4.1 b) Coll ision cross sections ( A 2 ) of positive ions produced from solutions of ferri and ferromyoglobin. Typical uncertainties for the collision cross sections of hMb are about ± 5% with the largest ( ± 1 1 % ) for the +5 ferriMb ion. In general, the data show that an increase in the charge state of hMb ions leads to an increase in collision cross section. The increase in cross sections can be attributed to an increase in Coulomb repulsion as the number of charges on the ion increases. This trend is consistent with the previous findings for myoglobin and other proteins [59, 71]. A comparison between the cross sections from Table 4.1 and [59, 71] could not be done, as the charge states and A V O S used to measure M b cross sections were different from each other. These results show that for a given charge state, positive ions from solutions of ferri and ferroMb have similar cross sections. Also, for a given charge state, negative ions from 57 solutions of ferri and ferroMb have similar cross sections. These results indicate that these low charge state myoglobin ions have nearly identical sizes. It appears that cross sections of myoglobin do not depend greatly on the oxidation state of iron in heme and the ion polarity. As discussed later in detail, similar cross sections for ions formed from ferro and ferriMb solutions are a part of the evidence that, Fe + 3 is reduced to Fe + 2 in the negative electrospray process. Cross sections of positive and negative ions of cytochrome c, with charges ± 4 to ± 11, measured by ion mobility, have been compared by Hoaglund-Hyzer and coworkers [130]. It was found that for the low charge states (± 4 to ± 6), cross sections of positive and negative ions are similar. In addition, for higher charge states (± 9 to ± 11), cross sections are similar. However, the intermediate charge states, ± 7 and ± 8, differ by approximately 200 A 2. It was reasoned that the difference was attributed to different gas phase stabilities of positive and negative intermediate charge states of cytochrome c. The collision cross sections from Tables 4.1 a) and b) are used later in this thesis (Chapter 6), in the calculation of the internal energies needed to dissociate holomyoglobin. 4.5 Summary Energy loss experiments are performed with positive and negative myoglobin ions from ferri and ferroMb solutions. A drag coefficient model is used to calculate the cross 58 sections. The results show cross sections ranging from about 1250 A 2 to 1460 A 2 for the charge states -4 to -7 hMb. For the +5 to +9 hMb ions, the cross sections gradually increase from about 1260 A 2 to 1810 A 2 . For a given charge state, positive and negative ferri and ferroMb ions have comparable cross sections. 59 Chapter 5 Collision Cross Sections of Apomyoglobin Ions Collings and Douglas have studied dissociation of hMb in the orifice-skimmer region, to produce gas phase aMb ions [59]. It was reported that gas phase hMb ions have relatively compact structures that partially unfold when heme is dissociated from the globin. It was also shown that collisions in the orifice-skimmer region can significantly unfold hMb. They proposed the following unfolding mechanism: hMb —> unfolded hMb -> aMb At A V O S of 30 V , only peaks of hMb ions were present in the mass spectrum. A t high A V O S , only ions of aMb could be produced. A direct comparison of cross sections between hMb and aMb ions at the same low A V O S is needed, since at low A V O S aMb ions are still folded. This comparison is given in this chapter. Results of collision cross section measurements of aMb ions are given. These ions are formed from aMb in solution. Heme is removed from solution prior to ESI -MS. The cross sections are compared to collision cross sections of aMb formed in the gas phase by dissociating hMb ions in the interface region, by increasing the A V O S . 60 5.1 Mass Spectra of Apomyoglobin 120 100 80 2 60 0 > QJ CC 40 H 20 o3 CM +i a 03 ^ ^ 2 + + + + + 03 + 1 03 O + 1000 1500 2000 m/z CO c CD QJ > 0) CL 1000 1500 2000 m/z 03 r--+ 2500 2500 a) 03 + 3000 3000 Figure 5.1 Mass spectra of apomyoglobin ions, formed from a solution containing 10% M e O H . The concentration of aMb was 1.2 x 10"5 M . The A V O S was a) 30 V , and b) 110 V . Solution p H was 6.9. 61 120 1000 1500 2000 m/z 2500 3000 Figure 5.2 Mass spectra of apomyoglobin ions, formed from a solution containing 10% M e O H . The concentration of aMb was 1.2 x 10"5 M . The A V O S was 180 V . Solution p H was 6.9. Mass spectra of ions formed from a solution of apomyoglobin at different A V O S are shown in Figures 5.1 and 5.2. It can be seen that, at a A V O S of 30 V , the charge state distribution of apomyoglobin is wide, with charge states from +6 to +19. When the A V O S is increased to 110 V , the charge state distribution narrows to charge states from +6a to +15a. The intensities of the higher charge states (+12a and greater) decrease relative to the peak with the highest intensity (+10a). It can be seen that the noise level is higher in Figures 5.1 b) and 5.2, compared to Figure 5.1 a). It is possible that the energy given to the higher charge state ions causes them to fragment, which in turn increases the noise level. This is observed in the mass spectrum at A V O S of 110 V , as the noise level 62 increases between m/z 800 and 1800. A n increase of A V O S to 180 V further narrows the charge state distribution, with charge states from +6 to +14. Apomyoglobin ions further fragment at this A V O S , and the noise level increases. Mass spectra of apomyoglobin, in which heme is removed in solution, prior to E S I - M S , have been reported [67, 131]. In a solution containing 50% M e O H at p H 6.5, DePauw and coworkers [131] produced high charge states of apomyoglobin, up to +19a. They observed a decrease in ion intensity for higher charge states at high A V O S , similar to the results shown above. It was suggested that higher charge states preferentially fragment. Ion intensity versus cone voltage was plotted for the charge states +9, +11, +14, +17 and +19. It was shown that, for the +14a ion, decrease in the cone voltage from 160 V to 60 V (analogous to an increase in A V O S ) , caused the ion intensity of the +14a ion to increase from 10 7 up to a maximum of 10 9 (arbitrary units). A further decrease in cone Q R voltage to 20 V , resulted in an ion intensity drop from 10 to 10 . This trend was observed for the charge states +9, +11, +14, +17 and +19. 5.2 Collision Cross Sections of Apomyoglobin Ions Collision cross sections are measured for the +6a to the +14a ions, at A V O S of 30 V , 110 V and 180 V , and for the +15a and +16a ions, at A V O S of 30 V , and the results are shown in Figure 5.3. Typical standard deviations for these measurements are ± 5%. The results show that for a given charge state, collision cross sections of apomyoglobin ions formed from solution increase with an increase in A V O S . This indicates that when aMb 63 ions undergo more energetic collisions in the orifice-skimmer region at higher A V O S , the aMb ions unfold. Table 5.1 shows the collision cross sections of aMb ions formed from aMb in solution. 3 5 0 0 -\ O 1 5 0 0 -\ 4 6 8 10 12 14 16 18 Charge State Figure 5.3 Coll ision cross sections versus charge state of positive apomyoglobin ions formed in solution at A V O S (•) 30 V , (o)110 V , and (T)180 V . In addition, collision cross sections were measured for holomyoglobin ions from a 1 x 10"5 M solution, containing 10% M e O H (pH 6.9), at A V O S values of 30 V , 110 V , and 180 V . In this case, an increase in A V O S causes the formation of gas phase aMb ions from hMb ions by CID . Table 5.2 shows the cross sections of hMb and aMb ions 64 obtained at different A V O S values. Collings and Douglas have previously reported similar spectra and collision cross sections [59]. Charge A V O S = 30 V A V O S = 110 V A V O S = 180 V +6 1875+ 103 1889 ± 1 1 2 2 0 6 8 ± 1 2 1 +7 2005 ± 73 2 0 3 0 ± 1 0 5 2 1 0 2 ± 114 +8 2171 ± 5 5 2304 ± 1 0 7 2405 ± 90 +9 2275 ± 6 3 2497 ± 5 5 2655 ± 9 3 +10 2521 ± 7 0 2487 ± 45 2702 ± 8 0 + 11 2568 ± 8 7 2583 ± 80 2 9 5 1 ± 1 0 5 +12 2569 ± 8 6 2743 ± 105 3 0 0 2 ± 1 2 2 +13 2742 ± 109 2860 ± 5 0 3012 ± 6 9 +14 2 7 4 0 ± 110 2895 ± 1 2 0 3019 + 92 + 15 2825 ± 69 + 16 2833 ± 9 3 Table 5.1 Collision cross sections (A2) of aMb ions formed in solution at A V O S of 30, 110 and 180 V . 65 Charge hMb ions at A V O S = 30 V hMb ions at A V O S = 110 V aMb ions at A V O S = 110 V aMb ions at A V O S = 180 V +6 1 7 8 9 ± 1 1 2 1577 ± 6 0 1873 ± 9 4 +7 1472 ± 1 1 6 1847 ± 105 1708 ± 8 3 1986 ± 105 +8 1589 ± 9 3 2009 ± 107 2098 ± 7 1 2362 ± 82 +9 1823 ± 105 2298 ± 55 2198 ± 9 0 2501 ± 1 1 0 + 10 2022 ± 70 2291 ± 4 5 2388 ± 82 2601 ± 1 1 4 + 11 2275 ± 63 2531 ± 8 0 2399 ± 55 2703 ± 9 1 +12 2387 ± 93 2714 ± 105 2674 ± 84 2603 ± 90 +13 2432 ± 1 0 4 2839 ± 50 2780 ± 89 +14 2895 +120 Table 5.2 Coll ision cross sections (A2) of hMb ions at A V O S of 30 V and 110 V , and aMb ions formed in the gas phase by dissociating hMb ions, at A V O S of 110 and 180 V . A direct comparison between aMb ions (Table 5.1) formed from solution aMb, and hMb ions (Table 5.2), at the same A V O S of 30 V , shows that for all charge states, aMb ions have greater collision cross sections than hMb, by an average of about 400 A 2. This indicates that for the same A V O S , gas phase hMb ions have a more folded structures than aMb ions. Heme binding to the globin helps maintain the compact structures of myoglobin in the gas phase. Figure 5.4 shows a direct comparison of collision cross sections of hMb and aMb at A V O S of 30 V . 66 3500 3000 H 1000 H 1 1 1 1 1 1 1 4 6 8 10 12 14 16 18 Charge State Figure 5.4 Coll ision cross sections vs. charge state for hMb and aMb ions at A V O S of 30 V . N M R has been used to characterize the structure of apomyoglobin in solution [132]. These studies show aMb as being an unstable native, globule-like protein with some disorder at the heme binding site. The conformations and stability of aMb formed in solution at neutral pH, by ESI -MS and C D [67, 68]. It was found that aMb exchanges more hydrogen atoms in H / D exchange experiments than hMb, indicating a more open conformation of aMb. Small-angle x-ray scattering [133] and size exclusion chromatography [ 134] experiments showed that the cross section of M b in solution increases about 1.3 times after heme loss, and another 2-4 times after complete denaturation. The data presented here show that at A V O S of 30 V , hMb ions have cross 67 sections about 10-30% lower than aMb ions, for both low and high charge states, indicating more compact gas phase hMb ions. 5.3 Summary Collision cross sections are measured for hMb and aMb ions from hMb and aMb solutions, respectively, and for aMb ions formed by dissociation of gas phase hMb ions in the orifice-skimmer region of the mass spectrometer, causing native hMb to unfold, followed by heme loss. Mass spectra of aMb ions formed from a solution of aMb, are recorded, and collision cross sections are measured at A V O S of 30 V , 110 V and 180 V . The cross section data show that an increase in the A V O S causes the aMb ions to unfold moderately. Comparison of the cross sections of hMb and aMb ions at A V O S of 30 V provides evidence that heme, when attached to the globin, helps maintain a more compact myoglobin structure. This goes further than the findings by Collings and Douglas [59], since conformations of aMb and hMb are compared at a low A V O S , where both these forms of myoglobin are folded, as well as at high A V O S , at which both are unfolded. Thus, the conformations can be compared in all stages of the unfolding process. The charge state distribution and collision cross sections of hMb at A V O S of 30 V are found to be similar to those found by Collings and Douglas. Wi th the A V O S voltage that gives similar amounts of aMb and hMb in the spectra (110 V ) , cross sections are found to be similar to those of Collings and Douglas [59]. 68 Chapter 6 Coulomb Effects in Binding of Heme to Myoglobin This chapter describes tandem mass spectrometry experiments with hMb ions formed from solutions of ferri and ferroMb. A collision model is applied to calculate the added energy needed to induce dissociation of heme from myoglobin ions. A simple model potential (Coulomb and Morse) is used to account for the differences in the energies needed to induce dissociation of heme for different charge states of hMb. Most of the results in this chapter are taken from Mark and Douglas [124]. 6.1 Tandem Mass Spectrometry of Holomyoglobin Ions Holomyoglobin ions were formed from a solution containing 50% M e O H , at p H 7.1. Holomyoglobin ions were mass selected in QI and fragmented in Q2, with argon as a collision gas. Fragment ions were mass analyzed in Q3. For M S / M S of hMb ions, the Q3 rod offset was set equal to Q2. Holomyoglobin dissociates into aMb and heme. 69 2200 2400 2600 2800 2200 2400 2600 2800 m/z m/z Figure 6.1 M S / M S spectra of -7 holomyoglobin ions from a solution of ferrimyoglobin at a collision gas pressure of 1.7 mtorr. The collision energy E° was a) 840 eV, b) 1085 eV, c) 1190 eV, and d) 1260 eV. Figure 6.1 is a set of M S / M S spectra for dissociation of the -7 ferriMb ions, at a collision cell pressure of 1.7 mtorr. It can be seen that the major dissociation pathway of the -7 hMb ion is the loss of neutral heme, to form the -7 aMb ion. There is a small loss (10% of fragments) of negative heme to form -6 aMb. The major dissociation pathway, neutral heme loss, is observed for all charge states, from -4 to -7. Neutral heme loss as a major 70 dissociation pathway is also observed for negative and positive hMb ions, formed from a ferroMb solution. On the other hand, for positive ferriMb ions, the major dissociation pathway is the loss of a positively charged heme with a minor dissociation channel being neutral heme loss. Tandem mass spectra of +7 ferriMb ions are shown in Figure 6.2. 0 120 a) J3 + rt + E u = 490 eV rt so + c) + E° = 770 eV rt + jj VJUJ b) S 60 A 0 -2200 + E° = 630 eV rt + d) + E° = 875 eV rt + 2800 2200 m/z m/z Figure 6.2 Tandem mass spectra of +7 holomyoglobin ions formed from a solution of ferrimyoglobin, at a collision gas pressure of 1.7 mtorr. The collision energy E° is a) 490 eV, b) 630 eV, c) 770 eV, and d) 875 eV. 71 120 -20 -I 1 1 1 1 1 1— 60 80 100 120 140 160 180 Q 0 - Q 2 (V) Figure 6.3 Precursor and fragment ion intensities versus Q0-Q2 voltage difference for the -7 holomyoglobin ions formed from a solution of ferrimyoglobin. In M S / M S experiments, lab kinetic energies are systematically increased, to cause loss of precursor ions, and formation of fragment ions. A plot of fragment yields versus the rod offset difference between Q0 and Q2 is constructed to show the change in intensity of precursor and fragment ions during the tandem M S experiment. The precursor (hMb) and fragment (aMb) relative intensities are calculated as follows: W % ) = T - ^ 7 — 1 0 0 % (6.1) * hMb 'aMb 72 where IhMh is the relative intensity of the precursor ion and IaMh is the relative intensity of the fragment ion. Figure 6.3 shows that, for the -7 hMb ion, an increase in Q0-Q2 causes a decrease in the intensity of the -7 hMb ion, and an increase in the intensity of the -7 aMb ion. The intensity of -6 aMb, corresponding to a charged heme loss, remains less than 5% of the initial precursor intensity, and is always less than 12% of the neutral loss in all experiments with negative ions formed from ferrimyoglobin in solution. The sum of the intensities of precursor and fragment ions is normalized to 100%, at each of the collision energies used. The collision energy that creates a 50% loss of precursor ions is taken as the nominal energy, E°, required to induce dissociation. The Q0-Q2 rod offset voltage difference determining this energy is referred to as the dissociation voltage. Figure 6.4 shows dissociation voltages of negative ions formed from ferromyoglobin at five different collision cell pressures. A t higher pressures there are more collisions, and the required energy per collision is lower. Therefore, at higher pressures, ions require lower initial kinetic energies to induce dissociation. i 73 350 100 A 1 1 1 1 1 1 1 1 1 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 Pressure (mTorr) Figure 6.4 Dissociation voltage vs. pressure for negative ferroMb ions, for charges -4h to -7h. 6.2 Collision Model The maximum possible energy an ion can acquire in the collision cell is the center of mass energy of each collision, summed over all collisions. As shown, higher dissociation energies are needed at lower collision cell pressures. Also , different charge states of hMb have different collision cross sections, and therefore at a given cell pressure have a different number of collisions. A collision model has been developed [72, 100], which takes into consideration the effects of different collision cross sections, and different 74 kinetic energy losses as the ions move through the cell. A n ion at a distance b from the cell entrance has a lab kinetic energy given by: E = E°e m ' (6.3) The mean free path, A, of the ion is given by: X = — (6.4) n<7 In traveling a distance Ab through the collision cell, the number of collisions the ion experiences, N , is: • N = — - noAb (6.5) Let O be the average fraction of energy transferred to internal energy in a collision. This is taken as d> =1.0 [135]. The increase in internal energy of an ion, AEim , while traveling the distance Ab is: AEin=<£>^E°e m> noAb (6.6) M 75 where M = ra, + m 2 . Writing equation 6.6 in differential form and integrating over the length traveled, / , leads to: Ei, nodb (6.7) o ra, 1 l-e m. (6.8) M m2 C D For these experiments, CD is typically between 2.1 and 2.3. Table 6.1 a) shows the dissociation voltages and collision cell pressures used for M S / M S experiments, collision cross sections (Chapter 4), and the calculated added internal energies required to dissociate negative ions formed from ferriMb in solution. The calculated internal energy represents the change in internal energy from one state (hMb) to another (aMb). The absolute internal energy of hMb entering the collision cell is not known. Uncertainties in the cross sections and dissociation voltages are the standard deviations of three experiments. Uncertainties in Eint values are calculated from the combined uncertainties of the cross sections and dissociation voltages. The same experiments were repeated with negative ions, formed from ferromyoglobin in solution. Tandem mass spectra show predominantly loss of neutral heme, with less than 12% loss of charged heme. Dissociation voltages, collision cell pressures, cross sections and AE- values are shown in Table 6.1 b). Tables 6.1 c) and d) list dissociation 76 voltages, collision cell pressures, cross sections, and A£ i n t values for positive ions formed from solutions of ferri and ferroMb. a) Charge Dissociation voltage (V) P (mtorr) Cross section (A 2 ) A £ i n t (eV) -4 269 ± 2 2.15 1252 ± 9 6 273 ± 2 1 -5 279 ± 10 1.70 1261 ± 7 4 302 ± 20 -6 222 ± 3 1.70 1271 ± 2 1 290 ± 6 -7 165 ± 8 1.70 1467 ± 54 278 ± 17 average 285 ± 16 b) Charge Dissociation voltage (V) P (mtorr) Cross section (A 2 ) A £ i n t (eV) -4 257 ± 11 2.15 1 2 7 1 ± 1 0 2 263 ± 24 -5 248 ± 11 1.7 1270 ± 9 7 270 ± 24 -6 205 ± 7 1.7 1259 ± 110 266 ± 25 -7 160 ± 7 1.7 1416 ± 7 5 263 ± 17 average 265 ± 23 77 c) Charge Dissociation voltage (V) P (mtorr) Cross section (A 2 ) A £ i m (eV) +5 163 ± 3 1.7 1278 ± 1 4 9 178 ± 20 +6 137 ± 3 1.7 1235 ± 9 8 175 ± 14 +7 103 ± 5 1.7 1594 + 54 184 ± 11 +8 90 ± 0 1.7 1777 ± 79 197 ± 9 +9 75 ± 5 1.7 1812 + 20 187 ± 13 average 184 ± 13 d) Charge Dissociation voltage (V) P (mtorr) Cross section (A 2 ) A £ i n t (eV) +5 295 ± 5 1.7 1265 ± 5 7 320 ± 15 +6 246 ± 8 1.7 1264 ± 8 1 320 ± 23 +7 181 ± 3 1.7 1 5 1 5 ± 1 0 1 312 ± 2 1 +8 138 ± 6 1.7 1791 ± 3 2 303 ± 14 average 314 ± 18 Table 6.1 Dissociation voltages, pressures, collision cross sections and AEiBt values for different charge states of holomyoglobin: a) negative ions from a ferriMb solution, b) negative ions from a ferroMb solution, c) positive ions from a ferriMb solution, and d) positive ions from a ferroMb solution.. 78 For comparisons between the experiments, slight corrections are applied to these A E i n t values since in experiments at different pressures, and for ions with different collision cross sections, ions have slightly different times to react [100]. It was found by simulations, that under the conditions where there is a 50% loss of precursor and 50% formation of fragment ions, that most of the reaction occurs in the last 4 cm of the collision cell [100]. The ion speed in this region is calculated from equation 4.8 and the time to traverse this region is taken as the time available for reaction R!ime: R = • lime (6.9) 2E° m. -exp — CDnm2l<7 m. where lr = 4 cm is the length over which the reaction occurs. The calculated energies required for reaction, AZs i n t , added to the negative h M b ions, formed from ferromyoglobin solution for different reaction times are shown in Figure 6.5. The best comparison of these energies is done by using a constant reaction time of say 18 |is. The internal energies required to react in this time can be determined from this plot. This procedure is repeated for positive and negative ions from ferri and ferromyoglobin. The added internal energies required to induce reaction in 18 J0.S are shown in Table 6.2. 79 300 290 H 240 o " V ^ '-• 0... \ \ \ y-~~~ V 10 12 14 16 18 20 22 Reaction Time (JUS) Figure 6.5 Calculated added internal energies, A E i n t , versus reaction time for negative holomyoglobin ions formed from ferromyoglobin solution; -4 ( • ) ; -5 (o); -6 ( T ) ; -7 ( V ) . a) Charge A £ i n t (eV) -4 293 -5 293 -6 282 -7 280 average 287 ± 7 80 b) Charge A £ i n t (eV) -4 268 -5 257 -6 248 -7 247 average 255 ± 10 c) Charge A £ i n t (eV) +5 171 +6 174 +7 182 +8 205 average 183 ± 15 d) Charge A£, n t +5 306 +6 295 +7 291 +8 269 average 290 ± 15 Table 6.2 AEiM values for a constant reaction time of 18 ps for a) negative ions from ferriMb solution, b) negative ions from ferroMb solution, c) positive ions from ferriMb solution, and d) positive ions from ferroMb solution. 81 6.3 Neutral versus Charged Heme Loss Loss of neutral heme is seen for negative ions from both ferromyoglobin and ferrimyoglobin in solution. Loss of positive heme from a negatively charged protein is not expected, because of the Coulomb attraction between the heme and protein. For example, from the cross section of 1271 A 2 for the -6 ion, a nominal ion "radius" of about 20 A can be calculated. If the -6 ion is comprised of -7 protein and +1 heme ions, the Coulomb binding energy at 20 A is 5.0 eV, considerably greater than the binding energy of heme in solution (1.1 eV) [39], or the activation energy for loss of positive ions from the positive protein (0.9 eV) [60]. Thus neutral heme loss might be expected even i f the heme is positively charged in the protein. The energies required to cause neutral heme loss from negative ions, formed from ferrimyoglobin and ferromyoglobin solutions, are similar (Table 6.1 a), and b)). This suggests the same reaction mechanism, loss of a neutral heme with neutral propionate groups and F e + 2 . The small loss of a negative heme likely arises from heme with F e + 2 and one ionized propionic acid group. For positive ions, charged heme loss dominates with ions formed from ferrimyoglobin, and neutral loss with ions formed from ferromyoglobin solution. The simplest interpretation is that heme in gas phase holomyoglobin ions has an overall charge +1 when formed from ferrimyoglobin, and is neutral when formed from ferromyoglobin. If the propionate groups are ionized in solution, they are likely re-protonated in the ESI process. The energy required to induce loss of neutral heme from positive ions is similar 82 to the energy required to induce neutral loss from the negative ions. It appears that the heme is overall neutral in these three cases. 6.4 Comparison of AEint Values The values of AEiM in Tables 6.1 and 6.2 are very similar, because the ions have similar cross sections and similar binding energies. Corrections for different reaction times, to Ais j n t calculated from equation 6.8, are quite small in this particular case. The values of A E i n t do not vary much over the limited range of charge states formed in these experiments. Tables 6.1 and 6.2 also show the A E i n t values averaged over charge states, and the average uncertainties. With positive or negative ions, the average energies required to induce loss of neutral heme are all about 280 eV. In contrast to this, the energy required to induce loss of charged heme from positive ions is substantially less, about 184 eV. Thus, it appears that Coulomb repulsion between the heme and positively charged globin lowers the barrier for heme loss. In the case of F e + 2 and F e + 3 , it is possible that the binding between heme and the protein in the gas phase complex changes. This could contribute to the difference in binding energy of the heme. In solution myoglobin, when Fe is in the oxidation state +2, the heme is slightly more strongly bound to the proximal histidine of the protein, than when Fe is in oxidation state +3, and this increases the overall binding energy of heme to the protein [136]. This difference in solution binding energy can be estimated in four different ways, as follows: 83 1) When C N " is bound to myoglobin, the Fe-His bond becomes stronger, analogous to F e + 2 myoglobin. Bunn and Jandl measured rates of heme exchange between different hemoglobins [137]. They found that ferrihemoglobin (37 ° C, p H 7.18) exchanged 76% of its hemes in 5 hours whereas cyanohemoglobin showed only 3.1-3.3% exchange. If this difference in rates (a factor of about 76/3 ~ 25) is attributed to differences in activation energies for heme loss (AE a ) , a difference of A E a = 0.086 eV is calculated. 13 1 2) The equilibrium constant for heme association with wi ld type myoglobin is 8x10 M " and for His93Gly myoglobin, where binding between the heme and proximal histidine is completely removed, 6 x l 0 9 M " 1 [138]. These equilibrium constants give A G values for binding of 0.807 and 0.567 eV respectively, and a difference of 0.24 eV or 22% of the activation energy for heme loss. 3) Hargrove and coworkers [139] compared rates of heme loss from wild type and many different mutant myoglobins. For the His93Gly mutant of sperm whale myoglobin (37 ° C, p H 5.0) the rate increases from 1.0 ± 0.5 hr"1 to 660 hr"1. If this change in rate is attributed to differences in activation energy for heme loss, a difference of A E a = 0.16 eV is calculated. This is about 15% of the activation energy for heme loss (1.1 eV). Both these experiments with His93Gly overestimate the difference between binding of Fe and F e + 3 heme. 84 4) Hargrove and coworkers [140] compared unfolding of aquometmyoglobin with cyanometmyoglobin by guanidinium chloride (GdmCl). The unfolding was modeled as an equilibrium between denatured and folded myoglobin with an equilibrium constant given by K{c) = ^exp(mc) where c is the G d m C l concentration and m = 4.1 M " 1 . For aquoferrimyoglobin K = 140 x 10"6 and for cyanoferrimyoglobin K = 6 x 10"6. This gives a difference in A G for binding heme of 0.0787 eV, small in comparison with the total binding energy of 1.1 eV. These four estimates, while approximate, show that the difference in heme binding between F e + 2 and F e + 3 is relatively small in comparison to the total binding energy, likely about 10% or less different. If these differences in binding energy apply to gas phase myoglobin ions, they seem too small to explain the ca. 30% difference in added energy, required for dissociation between neutral and charged heme loss. The simplest interpretation of these experiments is that Coulomb repulsion between the positive heme and positive protein lowers the barrier for heme loss. 6.5 Model Potential for Myoglobin The change in energy for charged and neutral heme loss is determined by the difference in Coulomb energy when heme is bound, and when heme is at the transition state for heme loss. The differences in energies required to dissociate charged heme from different charge states of holomyoglobin are small (Tables 6.1, 6.2), as are the differences in activation energies for loss of heme measured in B I R D experiments [60]. The differences 85 in Coulomb energy when heme is bound and when heme is at the transition state, must be small. This implies that the difference in spacing between the charged heme and charges on the protein, when the heme is bound and at the transition state, must also be relatively small. A simple-model potential for the heme binding that accounts for this and other properties of gas phase myoglobin is shown in Figure 6.6. The potential is given by: V(r) = [D, (1 - exp(-/?(r -re)f]-DE]+ q\a2 A7T£Qr (6.10) where the first term is a Morse potential with well depth DE, modified so V(r) —> 0 as r —> oo, and the second term describes the Coulomb energy between the heme and protein. The contributions of these two terms and their sum are shown in Figure 6.6. In equation 6.10, r is the distance between the heme and the protein, re is the distance between the heme and neutral protein when the neutral heme is bound, ft is a constant that determines the range of the Morse potential, qt is the heme charge, q2 is the protein charge and £ 0 is the permittivity of vacuum. Figure 6.6 shows this potential calculated with DE =1.1 eV, r e = 22 A , /? = 2.4 A " 1 , a charge of +6 on the protein and +1 on the heme (overall ion charge +7). The parameters of this potential are chosen as follows: (1) The value re = 22 A gives a cross sect ions; 2 of 1520 A 2 , like that of holomyoglobin +7 (Table 6.1); (2) The binding energy DE = 1.1 eV of heme in the absence of Coulomb repulsion is the activation energy to remove heme from solution myoglobin; (3) The value of P = 2.4 A " 1 was chosen because this gives a difference in activation energy for 86 heme loss from holomyoglobin +9 and holomyoglobin +12 ions of 0.1 eV, as found approximately in B I R D experiments [60]. Figure 6.6 shows that binding has a long range Coulomb repulsion, followed by a very short range attraction between the heme and protein. This behavior has been described by Rockwood and coworkers [141], for simple models of proteins as charges on a string, although it is now known that gas phase proteins [142], including myoglobin [72], in low charge states, contain considerable folded structure and are not "strings". 10 > -2 20 Coulomb •Morse+Coulomb Morse —i— 22 —i— 24 —i— 26 r(A) —i— 28 30 Figure 6 . 6 Model potentials for heme binding in neutral and +7 gas phase myoglobin, using Coulomb and Morse potentials. 87 6.6 Summary Myoglobin is particularly suited for studies of the Coulomb effects on binding because the charge can be at least partially controlled by oxidizing or reducing the heme. In principle, similar experiments could be performed with other protein small-molecule complexes, i f the charge on the small molecule could be controlled in a systematic way. +2 With gas phase myoglobin ions, the possible different bonding between heme Fe and F e + 3 and the protein complicate the interpretation of the M S / M S experiments, but the simplest conclusion is that Coulomb repulsion lowers the barrier for heme loss. This is consistent with a model potential that has long range Coulomb repulsion and short range binding. 88 Chapter 7 Conclusions and Future Work 7.1 Conclusions This thesis describes a study of the non-covalent heme-globin complex myoglobin. It examines the influence of Coulomb effects on the energy needed to dissociate the heme. This work is also an investigation of myoglobin conformations in solution and the gas phase, by monitoring charge state distributions and measuring collision cross sections with an electrospray ionization triple quadrupole system. A collision model is used to calculate the energy needed to dissociate the heme from myoglobin in tandem mass spectrometry experiments. The M S / M S results show that negative hMb ions from ferri and ferroMb solutions, and positive hMb ions from a ferroMb solution, give neutral heme loss as a major dissociation pathway and charged heme loss, as a minor dissociation pathway. For positive hMb ions formed from a solution of ferriMb, the charged heme loss is the major dissociation channel. Tandem mass spectrometry for a given charge state shows that a lower dissociation voltage is needed at higher A r collision gas pressures. Calculated internal energies show 89 that about 1.5 times more energy is needed to induce neutral heme loss, than to induce charged heme loss. The differences in energy can be attributed to Coulomb effects. Coulomb repulsion between the charged protein and charged heme lowers the barrier for heme loss. A model potential shows a short range attraction between heme and globin, and a long range Coulomb repulsion. The solution conformation change of negative myoglobin ions, formed from a ferriMb solution, is monitored by E S I - M S . At p H 7.1, low charge states (-4h to -7h) of holomyoglobin are observed in the ESI mass spectra. A t p H 3.2, a narrow charge state distribution of aMb ions, with low charge states, is formed. In comparison, in positive ion mode M S of myoglobin at a p H 4 [65], a broad charge state distribution of aMb ions with high charge states is observed. Coll ision cross sections of hMb and aMb ions, formed from a solution of hMb, are measured at different A V O S , and the results are similar to those reported by Collings and Douglas [59]. They found that the increase in A V O S activates h M b ions and causes them to unfold, followed by a loss of heme, to give aMb ions in the gas phase. In this study, apomyoglobin ions are formed from a solution of aMb. Apomyoglobin ions formed this way, have cross sections greater than cross sections of hMb ions at low A V O S . This implies that the heme helps maintain a more compact conformation of the gas phase myoglobin complex ions at low A V O S . 90 7.2 Future Work Protein conformations in solution have been monitored by E S I - M S , mainly focusing on positive protein ions. For a given protein, differences in charge state distributions between positive and negative protein ions have been observed [76]. For a more complete comparison between negative and positive M b ions, different M b charge state distributions at different solution p H values should be evaluated by E S I - M S . Furthermore, collision cross section measurements can be performed for ions formed from solutions at different p H values. As well , upon acid and base-induced denaturation protein unfolding in solution could be probed by measuring H / D exchange levels of negative M b ions. This may lead to better understanding of the differences between these two types of unfolding mechanisms at different p H values. The collision cross section results for aMb ions formed from aMb solution, could be supplemented by gas phase H / D exchange measurements. This might give an insight into the conformation differences between aMb and hMb ions when they are not unfolded. Measuring H / D exchange levels at different A V O S may give different exchange rates for gas phase aMb and hMb ions, which could be compared to the collision cross section data obtained in this study. It would be of interest to use the collision model to measure the energy needed for the dissociation of heme from complexes in which heme is bound covalently, like 91 cytochrome c, as well as from non-covalently bound protein-protein complexes, like hemoglobin, and to compare it to the energies required to dissociate myoglobin. Another way to explore the binding of heme would be to allow free heme to react with aMb in solution, prior to E S I - M S . Measuring the energy needed to dissociate heme from myoglobin in solution formed this way, could indicate the binding strength of the heme-globin complex as the heme attaches itself to aMb. Comparison to the binding in the gas phase can reveal i f differences exist. If they do, they could be attributed to the influence of solvent on heme binding. 92 References 1. Thomson, J. J., Rays of Positive Electricity and Their Application to Chemical Analysis, Longmans, Green and Co. Ltd. , London, 1913, p. p. 5. 2. Munson, M . S. B . and Field, F. H . , Chemical Ionization Mass Spectrometry I. General Introduction, J. Am. Chem. Soc, 1966, 88(12), p. 2621-2630. 3. Munson, B . , Chemical Ionization Mass Spectrometry - 10 Years Later, Anal. Chem., 1977, 49(9), p. A772-A778. 4. Horning, E . C , Horning, M . G. , Carroll, D . I., Dzidic , I., and Stil lwel, R. N . , New Picogram Detection System Based on a Mass Spectrometer with an External Ionization Source at Atmospheric Pressure, Anal. Chem., 1973, 45(6), p. 936-943. 5. Dempster, A . J., A New Method of Positive Ray Analysis, Phys. Rev. , 1918, 11(4), p. 316-325. 6. Becker, K . H . and Tarnovsky, V . , Electron-Impact Ionization of Atoms, Molecules, Ions and Transient Species, Plasma Sources Sci. Technol., 1995, 4, p. 307-315. 7. Barber, M . , Bordoli , R. S., Sedgwick, R. D. , and Tyler, A . N . , Fast Atom Bombardment of Solids as an Ion Source in Mass Spectrometry, Nature, 1981, 293(5830), p. 270-275. 93 8. Barber, M . , Bordoli , R. S., Elliott, G . J., Sedgwick, R. D. , and Tyler, A . N . , Fast Atom Bombardment Mass Spectrometry, Anal. Chem., 1982, 54(4), p. A645-A657. 9. Day, R. J., Unger, S. E . , and Cooks, R. G. , Molecular Secondary Ion Mass Spectrometry, Anal. Chem., 1980, 52(4), p. A557-A572. 10. Dole, M . , Mack, L . L . , and Hines, R. L . , Molecular Beams of Macroions, J. Chem. Phys., 1968, 49(5), p. 2240-2249. 11. Dole, M . , Hines, R. L . , Mack, L . L . , Mobley, R. C , Ferguson, L . D . , and Al ice , M . B . , Gas Phase Macroions, Macromolecules, 1968,1(1), p. 96 - 97. 12. Karas, M . and Hillenkamp, F., Laser Desorption Ionization of Proteins with Molecular Masses Exceeding 10,000 Daltons, Anal. Chem., 1988, 60(20), p. 2299-2301. 13. L i u , L . K . , Busch, K . L . , and Cooks, R. G. , Matrix-Assisted Secondary Ion Mass Spectra of Biological Compounds, Anal. Chem., 1981, 55(1), p. 109-113. 14. Houk, R. S., Fassel, V . A . , Flesch, G . D . , Svec, H . J., Gray, A . L . , and Taylor, C. E. , Inductively Coupled Argon Plasma as an Ion Source for Mass Spectrometric Determination of Trace Elements, Anal. Chem., 1980, 52(14), p. 2283-2289. 15. Taylor, H . E . , Inductively Coupled Plasma-Mass Spectrometry: Practices and Techniques, Academic Press, San Diego, C A , 2001, p. p. 55-88. 16. Nier, A . O., Ney, E . P., and Inghram, M . G . , A N u l l Method for the Comparison of Two Ion Currents in a Mass Spectrometer, Rev. Sci. Instrum., 1947,18(5), p. 294-297. 17. Paul, W . and Raether, M . Z . , Physik, 1955,140, p. 262-273. 94 18. Cameron, A . E . and Eggers Jr, A . E . D . F. , A n Ion "Velocitron", Rev. Sci. Instrum., 1948, 79(9), p. 605-607. 19. Wolff, M . M . and Stephens, W . E . , A Pulsed Mass Spectrometer with Time Dispersion, Rev. Sci. Instrum., 1953, 24(8), p. 616-617. 20. Katzenstein, H . S. and Friedland, S. S., New Time-of-Flight Mass Spectrometer, Rev. Sci. Instrum., 1955, 25(4), p. 324-327. 21. Chernusevich, I. V . , Ens, W. , and Standing, K . G . , Electrospray Ionization Mass Spectrometry, John Wiley and Sons, New York, N Y , 1997, p. p. 385-419. 22. Wiley, W . C. and McLaren , I. H . , Rev. Sci. Instrum., 1955(26), p. 1150-1157. 23. Comisarow, M . and Marshall, A . G . , Fourier-Transform Ion-Cyclotron Resonance Spectroscopy, Chem. Phys. Lett., 1974, 25(2), p. 282-283. 24. Paul, W . and Steinwedel, H . Z . , Naturforsch., 1953(8), p. 448-450. 25. Cole, R. B . , ed. Electrospray Ionization Mass Spectrometry: Fundamentals, Instrumentation and Applications, John Wiley & Sons, New York, N Y , 1997, p. p. 177-202. 26. Fenn, J. B . , Mann, M . , Meng, C. K . , Wong, S. F. , and Whitehouse, C . M . , Electrospray Ionization for Mass Spectrometry of Large Biomolecules, Science, 1989, 246(4926), p. 64-71. 27. Tanaka, K . , Waki , FL, Ido, Y . , Akita , S., Yoshida, Y . , and Yoshida, T., Protein and Polymer Analyses up to m/z 100,000 by Laser Ionization Time-of-Flight Mass Spectrometry, Rapid Commun. Mass Spectrom., 1988, 2, p. 151-153. 28. Cole, R. B . , Some Tenets Pertaining to Electrospray Ionization Mass Spectrometry, J. Mass Spectrom., 2000, 35(7), p. 763-772. 95 29. Iribarne, J. V . and Thomson, B . A . , Evaporation of Small Ions from Charged Droplets, J. Chem. Phys., 1976, 64(6), p. 2287-2294. 30. Thomson, B . A . and Iribarne, J. V . , Field-Induced Ion Evaporation from Liquid Surfaces at Atmospheric Pressure, J. Chem. Phys., 1979, 77(11), p. 4451-4463. 31. Schmelzeisenredeker, G . , Giessmann, U . , and Rollgen, F. W . , Field Effects in Molecular Ion Formation by the Thermospray Technique, Journal de Physique, 1984, 45(NC9), p. p. 297-302. 32. Covey, T. R., Bruins, A . P., and Henion, J. D. , Comparison of Thermospray and Ion Spray Mass Spectrometry in an Atmospheric Pressure Ion Source, Org. Mass Spectrom., 1988, 25(3), p. 178-186. 33. Al len , M . H . and Vestal, M . L . , Design and Performance of a Novel Electrospray Interface, J. Am. Soc. Mass Spectrom., 1992, 5(1), p. 18-26. 34. W i l m , M . S. and Mann, M . , Electrospray and Taylor Cone Theory, Dole's Beam of Macromolecules at Last, Int. J. Mass Spectrom. Ion Proc, 1994,136(2-3), p. 167-180. 35. W i l m , M . and Mann, M . , Analytical Properties of the Nanoelectrospray Ion Source, Anal. Chem., 1996, 68(1), p. 1-8. 36. Reedy, C. J. and Gibney, B . R., Heme Protein Assemblies, Chem. Rev., 2004, 104(2), p. 617-650. 37. Stryer, L . , Biochemistry, Freeman and Company, San Francisco, 1981, p. p. 44-51. 38. Budavari, S., ed. The Merck Index, 12 ed., Merck Research Laboratories, Whitehouse Station, N J , 1996, p. p. 4668-4669. 96 39. Hunter, C. L . , Mauk, A . G . , and Douglas, D . J., Dissociation of Heme from Myoglobin and Cytochrome bs: Comparison of Behavior in Solution and the Gas Phase, Biochemistry, 1997, 36(5), p. 1018-1025. 40. Hunter, C . L . , L loyd , E . , Eltis, L . D . , Rafferty, S. P., Lee, H . , Smith, M . , and Mauk, A . G . , Role of the Heme Propionates in the Interaction of Heme with Apomyoglobin and Apocytochrome bs, Biochemistry, 1997, 36(5), p. 1010-1017. 41. Kendrew, J. C , Bodo, G . Dintzis, H . M . , Parrish, R . G . , Wyckoff, H . , Phillips, D .C . , A Three-Dimensional Model of the Myoglobin Molecule Obtained by X -Ray Analysis, Nature, 1958,181, p. 662-666. 42. Bodo, G . , Dintzis, H . M . , Kendrew, J . C , Wickoff, H .W. , The Crystal Structure of Myoglobin V . A Low Resolution Three Dimensional Fourier Synthesis of Sperm-Whale Myoglobin Crystals, Proc. Royal Soc. London, 1959, 253, p. 70-102. 43. Takano, T., Structure of Myoglobin Refined at 2.0 A Resolution 1. Crystallographic Refinement of Metmyoglobin from Sperm Whale, J. Moi. Biol, 1977, 770(3), p. 537-568. 44. Takano, T., Structure of Myoglobin Refined at 2.0 A Resolution 2. Structure of Deoxymyoglobin from Sperm Whale, J. Moi. Biol, 1977, 770(3), p. 569-584. o 45. Phillips, S. E . V . , Structure and Refinement of Oxymyoglobin at 1.6 A Resolution, J. Moi. Biol, 1980, 742(4), p. 531-554. 46. Fasman, J. D. , Practical Handbook of Biochemistry and Molecular Biology, C R C Press, Boca Raton, Florida, 1989, p. p. 286-293. 47. Postnikova, G . B . and Yumakova, E . M . , Fluorescence Study of the Conformational Properties of Myoglobin Structure. pH-Dependent Changes in 97 Porphyrin and Tryptophan Fluorescence of the Complex of Sperm Whale Apomyoglobin with Protoporphyrin-LX - the Role of the Porphyrin Macrocycle and Iron in Formation of Native Myoglobin Structure, Eur. J. Biochem., 1991, 798(1), p. 241-246. 48. Nozawa, T., Yamamoto, T., and Hatano, M . , Infrared Magnetic Circular Dichroism of Myoglobin Derivatives, Biochim. Biophys. Acta, 1976, 427(1), p. 28-37. 49. Matsuoka, A . , Kobayashi, N . , and Shikama, K . , The Soret Magnetic Circular Dichroism of Ferric High-Spin Myoglobins - a Probe for the Distal Histidine Residue, Eur. J. Biochem., 1992, 270(1), p. 337-341. 50. Patel, D . J., Kampa, L . , Shulman, R. G . , Yamane, T., and Wyluda, B . J. , Proton Nuclear Magnetic Resonance Studies of Myoglobin in H2O, Proc. Natl. Acad. Sci. US, 1970, 67(3), p. 1109-1115. 51. Kreutzer, U. and Jue, T., ' H - N M R Signal of Arenicola marina Myoglobin in vivo as an Index of Tissue Oxygenation, Eur. J. Biochem., 1996, 255(3), p. 622-628. 52. Norris, C . L . and Peters, K . S., A Photoacoustic Calorimetry Study of Horse Carboxymyoglobin on the 10-nanosecond Time Scale, Biophys. J., 1993, 65(4), p. 1660-1665. 53. Jamin, M . , Antalik, M . , Loh , S. N . , Bolen, D . W. , and Baldwin, R. L . , The Unfolding Enthalpy of the p H 4 Molten Globule of Apomyoglobin Measured by Isothermal Titration Calorimetry, Protein Sci., 2000, 9(7), p. 1340-1346. 98 54. Ganem, B . , L i , Y . T., and Henion, J. D. , Observation of Noncovalent Enzyme Substrate and Enzyme Product Complexes by Ion-Spray Mass Spectrometry, J. Am. Chem. Soc., 1991,113(20), p. 7818-7819. 55. Ganem, B . , L i , Y . T., and Henion, J. D . , Detection of Noncovalent Receptor Ligand Complexes by Mass Spectrometry, J. Am. Chem. Soc, 1991, 7/3(16), p. 6294-6296. 56. Jaquinod, M . , Leize, E . , Potier, N . , Albrecht, A . M . , Shanzer, A . , and Vandorsselaer, A . , Characterization of Noncovalent Complexes by Electrospray Mass Spectrometry, Tetrahedron Lett., 1993, 34(11), p. 2111-211 A. 57. Mcluckey, S. A . and Ramsey, R. S., Gaseous Myoglobin Ions Stored at Greater Than 300 K , J. Am. Soc. Mass Spectrom., 1994, 5(4), p. 324-327. 58. Konishi, Y . and Feng, R., Conformational Stability of Heme-Proteins in Vacuo, Biochemistry, 1994, 53(32), p. 9706-9711. 59. Collings, B . A . and Douglas, D . J. , Conformation of Gas-Phase Myoglobin Ions, /. Am. Chem. Soc, 1996, 775(18), p. 4488-4489. 60. Gross, D . S., Zhao, Y . X . , and Will iams, E . R., Dissociation of Heme-Globin Complexes by Blackbody Infrared Radiative Dissociation: Molecular Specificity in the Gas Phase?, J. Am. Soc. Mass Spectrom., 1997, 5(5), p. 519-524. 61. Katta, V . and Chait, B . T., Observation of the Heme Globin Complex in Native Myoglobin by Electrospray Ionization Mass Spectrometry, J. Am. Chem. Soc, 1991, 773(22), p. 8534-8535. 99 62. Mao, D . M . , Ding, C. F. , and Douglas, D . J., Hydrogen/Deuterium Exchange of Myoglobin Ions in a Linear Quadrupole Ion Trap, Rapid Commun. Mass Spectrom., 2002, 76(20), p. 1941-1945. 63. Loo, J. A . , Giordani, A . B . , and Muenster, FL, Observation of Intact (Heme-Bound) Myoglobin by Electrospray Ionization on a Double-Focusing Mass Spectrometer, Rapid Commun. Mass Spectrom., 1993, 7(3), p. 186-189. 64. Konermann, L . , Rosell, F. I., Mauk, A . G . , and Douglas, D . J., Acid-Induced Denaturation of Myoglobin Studied by Time-Resolved Electrospray Ionization Mass Spectrometry, Biochemistry, 1997, 36(21), p. 6448-6454. 65. Babu, K . R. and Douglas, D . J., Methanol-Induced Conformations of Myoglobin at p H 4.0, Biochemistry, 2000, 39(47), p. 14702-14710. 66. Konermann, L . and Douglas, D . J., Equilibrium Unfolding of Proteins Monitored by Electrospray Ionization Mass Spectrometry: Distinguishing Two-State from Multi-State Transitions, Rapid Commun. Mass Spectrom., 1998, 72(8), p. 435-442. 67. Wang, F. and Tang, X . J., Conformational Heterogeneity and Stability of Apomyoglobin Studied by Hydrogen Deuterium Exchange and Electrospray Ionization Mass Spectrometry, Biochemistry, 1996, 35(13), p. 4069-4078. 68. Johnson, R. S. and Walsh, K . A . , Mass Spectrometric Measurement of Protein Amide Hydrogen Exchange Rates of Apo-Myoglobin and Holo-Myoglobin, Protein Sci., 1994, 3(12), p. 2411-2418. 100 69. Shelimov, K . B . and Jarrold, M . F. , Conformations, Unfolding, and Refolding of Apomyoglobin in Vacuum: A n Activation Barrier for Gas-Phase Protein Folding, J. Am. Chem. Soc., 1997, 779(13), p. 2987-2994. 70. Chen, Y . L . , Collings, B . A . , and Douglas, D . J., Coll is ion Cross Sections of Myoglobin and Cytochrome c Ions with Ne, A r , and Kr, J. Am. Soc. Mass . Spectrom., 1997, 8(1), p. 681-687. 71. Covey, T. and Douglas, D . J., Collision Cross Sections for Protein Ions, J. Am. Soc. Mass Spectrom., 1993, 4(8), p. 616-623. 72. Chen, Y . L . , Campbell, J. M . , Collings, B . A . , Konermann, L . , and Douglas, D . J., Stability of a Highly Charged Noncovalent Complex in the Gas Phase: Holomyoglobin, Rapid Commun. Mass Spectrom., 1998, 72(15), p. 1003-1010. 73. L i , Y . T., Hsieh, Y . L . , Henion, J. D . , and Ganem, B . , Studies on Heme-Binding in Myoglobin, Hemoglobin, and Cytochrome c by Ion-Spray Mass Spectrometry, J. Am. Soc. Mass Spectrom., 1993, 4(8), p. 631-637. 74. Sogbein, O. O., Simmons, D . A . , and Konermann, L . , Effects of p H on the Kinetic Reaction Mechanism of Myoglobin Unfolding Studied by Time-Resolved Electrospray Ionization Mass Spectrometry, J. Am. Soc. Mass Spectrom., 2000, 77(4), p. 312-319. 75. Schmidt, A . and Karas, M . , The Influence of Electrostatic Interactions on the Detection of Heme-Globin Complexes in ESI-MS, J. Am. Soc. Mass Spectrom., 2001, 72(10), p. 1092-1098. 101 76. Konermann, L . and Douglas, D . J., Unfolding of Proteins Monitored by Electrospray Ionization Mass Spectrometry: A Comparison of Positive and Negative Ion Modes, J. Am. Soc. Mass Spectrom., 1998, 9(12), p. 1248-1254. 77. Mirza , U . A . , Cohen, S. L . , and Chait, B . T., Heat-Induced Conformational Changes in Proteins Studied by Electrospray Ionization Mass Spectrometry, Anal. Chem., 1993, 65(1), p. 1-6. 78. Loo, J. A . , Edmonds, C . G . , Udseth, H . R., and Smith, R. D. , Effect of Reducing Disulfide-Containing Proteins on Electrospray Ionization Mass Spectra, Anal. Chem., 1990, 62(7), p. 693-698. 79. Konermann, L . and Douglas, D . J., Acid-Induced Unfolding of Cytochrome c at Different Methanol Concentrations: Electrospray Ionization Mass Spectrometry Specifically Monitors Changes in the Tertiary Structure, Biochemistry, 1997, 56(40), p. 12296-12302. 80. Chowdhury, S. K . , Katta, V . , and Chait, B . T., Probing Conformational Changes in Proteins by Mass Spectrometry, J. Am. Chem. Soc, 1990, 772(24), p. 9012-9013. 81. Veenstra, T. D. , Johnson, K . L . , Tomlinson, A . J., Naylor, S., and Kumar, R., Determination of Calcium-Binding Sites in Rat Brain Calbindin D - 2 8 K by Electrospray Ionization Mass Spectrometry, Biochemistry, 1997, 56(12), p. 3535-3542. 82. Nemirovskiy, O. V . , Ramanathan, R., and Gross, M . L . , Investigation of Calcium-Induced, Noncovalent Association of Calmodulin with Melit t in by Electrospray 102 Ionization Mass Spectrometry, J. Am. Soc. Mass Spectrom., 1997, 5(8), p. 809-812. 83. Veenstra, T. D . , Johnson, K . L . , Tomlinson, A . J., Craig, T. A . , Kumar, R., and Naylor, S., Zinc-Induced Conformational Changes in the D N A - B i n d i n g Domain of the Vitamin D Receptor Determined by Electrospray Ionization Mass Spectrometry, / . Am. Soc. Mass Spectrom., 1998, 9(1), p. 8-14. 84. Winston, R. L . and Fitzgerald, M . C , Mass Spectrometry as a Readout of Protein Structure and Function, Mass Spectrom. Rev., 1997,16(4), p. 165-179. 85. Cole, R. B . , ed. Electrospray Ionization Mass Spectrometry: Fundamentals Instrumentation and Applications, John Wiley & Sons, New York, N Y , 1997, p. p. 385-419. 86. Grandori, R., Matecko, I., Mayr, P., and Muller , N . , Probing Protein Stabilization by Glycerol Using Electrospray Mass Spectrometry, J. Mass Spectrom., 2001, 36(8), p. 918-922. 87. Konermann, L . , Collings, B . A . , and Douglas, D . J., Cytochrome c Folding Kinetics Studied by Time-Resolved Electrospray Ionization Mass Spectrometry, Biochemistry, 1997, 36(18), p. 5554-5559. 88. Mack, E . , Average Cross Sectional Areas of Molecules by Gaseous Diffusion Methods, J. Am. Chem. Soc, 1925, 47(10), p. 2468-2482. 89. Cohen, M . J. and Karasek, F. W. , Plasma Chromatography - a New Dimension for Gas Chromatography and Mass Spectrometry, J. Chromatogr. Sci., 1970, 5(6), p. 330-337. 103 90. Jarrold, M . F., Peptides and Proteins in the Vapor Phase, Annu. Rev. Phys. Chem., 2000, 57, p. 179-207. 91. Kemper, P. R. and Bowers, M . T., A Hybrid Double-Focusing Mass Spectrometer - High-Pressure Drift Reaction Cel l to Study Thermal Energy Reactions of Mass-Selected Ions, J. Am. Soc. Mass Spectrom., 1990, 7(3), p. 197-207. 92. Bluhm, B . K . , G i l l i g , K . J., and Russell, D . H . , Development of a Fourier-Transform Ion Cyclotron Resonance Mass Spectrometer-Ion Mobi l i ty Spectrometer, Rev. Sci. Instrum., 2000, 77(11), p. 4078-4086. 93. Creaser, C. S., Benyezzar, M . , Griffiths, J. R., and Stygall, J. W . , A Tandem Ion Trap/Ion Mobi l i ty Spectrometer, Anal. Chem., 2000, 72(13), p. 2724-2729. 94. Matz, L . M . , Asbury, G . R., and H i l l , H . H . , Two-Dimensional Separations with Electrospray Ionization Ambient Pressure High-Resolution Ion Mobi l i ty Spectrometry/Quadrupole Mass Spectrometry, Rapid Commun. Mass Spectrom., 2002, 76(7), p. 670-675. 95. Hoaglund, C. S., Valentine, S. J., Sporleder, C. R., Reil ly, J. P., and Clemmer, D . E . , Three-Dimensional Ion Mobil i ty T O F M S Analysis of Electrosprayed Biomolecules, Anal. Chem., 1998, 70(11), p. 2236-2242. 96. Collins, D . C. and Lee, M . L . , Developments in Ion Mobil i ty Spectrometry-Mass Spectrometry, Anal. Bioanal. Chem., 2002, 372(1), p. 66-73. 97. Mason, E . A . and McDanie l , E . W. , Transport Properties of Ions in Gases, Wiley & Sons, New York N Y , 1988, p. p. 212. 98. Shvartsburg, A . A . and Jarrold, M . F., A n Exact Hard Spheres Scattering Model for the Mobilities of Polyatomic Ions, Chem. Phys. Lett., 1996, 267, p. 86-91. 104 99. Moradian, A . , Scalf, M . , Westphall, M . S., Smith, L . M . , and Douglas, D . J., Coll ision Cross Sections of Gas Phase D N A Ions, Int. J. Mass Spectrom., 2002, 279(1), p. 161-170. 100. Mauk, M . R., Mauk, A . G. , Chen, Y . L . , and Douglas, D . J., Tandem Mass Spectrometry of Protein-Protein Complexes: Cytochrome c-Cytochrome bs, J. Am. Soc. Mass Spectrom., 2002, 73(1), p. 59-71. 101. Douglas, D . J., A n Aerodynamic Drag Model for Protein Ions, J. Am. Soc. Mass Spectrom., 1994, 5(1), p. 17-18. 102. Tesic, M . , W i c k i , J., Poon, D . , Withers, S. G . , and Douglas, D . J. , Gas Phase Noncovalent Protein Complexes that Retain Solution Binding Properties: Binding of Xylobiose Inhibitors to the (3-1,4 Exoglucanase From Cellulomonas fimi, J. Am. Soc. Mass Spectrom., 2006, p. in press. 103. Shukla, A . K . and Futrell, J. FL, Tandem Mass Spectrometry: Dissociation of Ions by Collisional Activation, J. Mass Spectrom., 2000, 35(9), p. 1069-1090. 104. March, R. E . , Practical Aspects of Ion Trap Mass Spectrometry, C R C Press, Boca Raton, 1995, p. p. 164-172. 105. Mclafferty, F. W. , Todd, P. J., Mcgilvery, D . C , and Baldwin, M . A . , Collisional Activation and Metastable Ion Characteristics .73. High-Resolution Tandem Mass Spectrometer ( M S - M S ) of Increased Sensitivity and Mass Range, J. Am. Chem. Soc, 1980, 702(10), p. 3360-3363. 106. Loo, J. A . , Ogorzalek Loo, R. R., and Andrews, P. C , Primary to Quaternary Protein Structure Determination with Electrospray Ionization and Magnetic Sector Mass Spectrometry, Org. Mass Spectrom., 1993, 25(12), p. 1640-1649. 105 107. Yost, R. A . and Enke, C. G . , Selected Ion Fragmentation with a Tandem Quadrupole Mass Spectrometer, 7. Am. Chem. Soc., 1978,100(1), p. 2274-2275. 108. Schey, K . , Cooks, R. G . , Grix , R., and Wollnik, H . , A Tandem Time-of-Flight Mass Spectrometer for Surface-Induced Dissociation, Int. J. Mass Spectrom. Ion Proc, 1987, 77(1), p. 49-61. 109. Chrisman, P. A . , Newton, K . A . , Reid, G . E . , Wells , J. M . , and McLuckey , S. A . , Loss of Charged Versus Neutral Heme from Gaseous Holomyoglobin Ions, Rapid Commun. Mass Spectrom., 2001, 75(23), p. 2334-2340. 110. Lippard, S. J., Principles of Bioinorganic Chemistry, University Science Books, M i l l Valley, C A , 1994, p. p. 219. 111. Wisz , M . S. and Hellinga, H . W. , A n Empirical Model for Electrostatic Interactions in Proteins Incorporating Multiple Geometry Dependent Dielectric Constants, Proteins:Structrure,Function,Genetics, 2003, 57(3), p. 360-377. 112. Wright, W . W. , Laberge, M . , and Vanderkooi, J. M . , Surface of Cytochrome c: Infrared Spectroscopy of Carboxyl Groups, Biochemistry, 1997, 36(48), p. 14724-14732. 113. Hauser, K . , Mao, J., and Gunner, M . R., p H Dependence of Heme Electrochemistry in Cytochromes Investigated by Multiconformation Continuum Electrostatic Calculations, Biopolymers, 2004, 74(1-2), p. 51-54. 114. Csiszar, S. and Thachuk, M . , Using Ellipsoids to Model Charge Distributions in Gas Phase Protein Complex Ion Dissociation, Can. I. Chem., 2004, 82(12), p. 1736-1744. 106 115. Felitsyn, N . , Kitova, E . N . , and Klassen, J. S., Thermal Dissociation of the Protein Homodimer Ecotin in the Gas Phase, J. Am. Soc. Mass Spectrom., 2002,75(12), p. 1432-1442. 116. Schwartz, B . L . , Bruce, J. E . , Anderson, G . A . , Hofstadler, S. A . , Rockwood, A . L . , Smith, R. D. , Chilkoti , A . , and Stayton, P. S., Dissociation of Tetrameric Ions of Noncovalent Streptavidin Complexes Formed by Electrospray Ionization, / . Am. Soc. Mass Spectrom., 1995, 6(6), p. 459-465. 117. Jurchen, J. C. and Will iams, E . R., Origin of Asymmetric Charge Partitioning in the Dissociation of Gas-Phase Protein Homodimers, J. Am. Chem. Soc, 2003, 725(9), p. 2817-2826. 118. Lightwahl, K . J., Schwartz, B . L . , and Smith, R. D . , Observation of the Noncovalent Quaternary Associations of Proteins by Electrospray Ionization Mass Spectrometry, J. Am. Chem. Soc, 1994, 776(12), p. 5271-5278. 119. Felitsyn, N . , Kitova, E . N . , and Klassen, J. S., Thermal Decomposition of a Gaseous Multiprotein Complex Studied by Blackbody Infrared Radiative Dissociation. Investigating the Origin of the Asymmetric Dissociation Behavior, Anal. Chem., 2001, 75(19), p. 4647-4661. 120. Versluis, C. and Heck, A . J. R., Gas-Phase Dissociation of Hemoglobin, Int. J. Mass Spectrom., 2001, 270(1-3), p. 637-649. 121. Benesch, J. L . and Robinson, C. V . , Mass Spectrometry of Macromolecular Assemblies: Preservation and Dissociation, Curr. Opin. Struc. Biol., 2006, 76(2), p. 245-251. 107 122. Dawson, P. H . , ed. Quadrupole Mass Spectrometry and its Applications, American Vacuum Society Classics, American Institute of Physics, Woodbury, N . Y . , 1995, p. p. 10-20. 123. March, R. E . and Hughes, R. J., Quadrupole Storage Mass Spectrometry 102, Chemical Analysis Series, John Wiley & Sons, New York, 1989, p. p. 46-47. 124. Mark, K . J. and Douglas, D . J., Coulomb effects in binding of heme in gas-phase ions of myoglobin, Rapid Commun. Mass Spectrom., 2006, 20(2), p. 111-117. 125. Loo, J. A . , Loo, R. R. O., Light, K . J., Edmonds, C. G . , and Smith, R. D. , Mult iply Charged Negative Ions by Electrospray Ionization of Polypeptides and Proteins, Anal. Chem., 1992, 64(1), p. 81-88. 126. Lee, V . W . S., Chen, Y . L . , and Konermann, L . , Reconstitution of A c i d -Denatured Holomyoglobin Studied by Time-Resolved Electrospray Ionization Mass Spectrometry, Anal. Chem., 1999, 77(19), p. 4154-4159. 127. Halliday, D. , Resnick, R., Fundamentals of Physics, John Wiley & Sons, New York, N Y , 1988, p. p. 205-217. 128. Stadler, J. R. and Zurich, V . J., Theoretical Aerodynamic Characteristics of Bodies in a Free-Molecule-Flow Field, National Advisory Committee for Aeronautics TN2423,1951, Moffett Field, C A . 129. Henderson, C. B . , Drag Coefficients of Spheres in Continuum and Rarefied Flows, AIAA J., 1976, 74(6), p. 707-708. 130. Hoaglund-Hyzer, C. S., Counterman, A . E . , and Clemmer, D . E . , Anhydrous Protein Ions, Chem. Rev., 1999, 99(10), p. 3037-3079. 108 131. Meunier, C , Jamin, M , and De Pauw, E . , On the Origin of the Abundance Distribution of Apomyoglobin Mult iply Charged Ions in Electrospray Mass Spectrometry, Rapid Commun. Mass Spectrom., 1998,12(5), p. 239-245. 132. Eliezer, D . and Wright, P. E . , Is Apomyoglobin a Molten Globule? Structural Characterization by N M R , J. Moi. Biol., 1996, 263(A), p. 531-538. 133. Nishi i , I., Kataoka, J., Tokunaga, F., and Goto, Y . , Co ld Denaturation of the Molten Globule States of Apomyoglobin and a Profile for Protein Folding, Biochemistry, 1994, 33, p. 4903-4904. 134. Goto, Y . , Calciano, L . , and Fink, A . , Acid-Induced Folding of Proteins, Proc. Natl. Acad. Sci. US, 1990, 87(2), p. 573-577. 135. Marzluff, E . M . , Campbell, S., Rodgers, M . T., and Beauchamp, J. L . , Collisional Activation of Large Molecules Is an Efficient Process, J. Am. Chem. Soc, 1994, 776(15), p. 6947-6948. 136. Hargrove, M . S., Singleton, E . W. , Quil l in , M . L . , Ortiz, L . A . , Phillips, G . N , Olson, J. S., and Mathews, A . J., His(64)(E7)-]Tyr Apomyoglobin as a Reagent for Measuring Rates of Hemin Dissociation, J. Biol. Chem., 1994, 269(6), p. 4207-4214. 137. Bunn, H . F. and Jandl, J. H . , Exchange of Heme among Hemoglobins and between Hemoglobin and Albumin, J. Biol. Chem., 1968, 243(3), p. 465-475. 138. Hargrove, M . S., Barrick, D . , and Olson, J. S., The Association Rate Constant for Heme Binding to Globin is Independent of Protein Structure, Biochemistry, 1996, 35(35), p. 11293-11299. 109 139. Hargrove, M . S., Wilkinson, A . J., and Olson, J. S., Structural Factors Governing Hemin Dissociation from Metmyoglobin, Biochemistry, 1996, 35(35), p. 11300-11309. 140. Hargrove, M . S. and Olson, J. S., The Stability of Holomyoglobin is Determined by Heme Affinity, Biochemistry, 1996, 35(35), p. 11310-11318. 141. Rockwood, A . L . , Busman, M . , and Smith, R. D . , Coulombic Effects in the Dissociation of Large Highly Charged Ions, Int. J. Mass Spectrom. Ion Proc, 1991,111, p. 103-129. 142. L i u , Y . S., Valentine, S. J., Counterman, A . E . , Hoaglund, C. S., and Clemmer, D . E. , Injected-Ion Mobil i ty Analysis of Biomolecules, Anal. Chem., 1997, 69(23), p. A728-A735. 110