Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 6

Method Development in Quantitative and Structural Proteomics using Fourier Transform Ion Cyclotron Resonance Mass SpectrometryBYCHARLOTTE HAGMAN

ISSN 1651-6214ISBN 91-554-6134-4urn:nbn:se:uu:diva-4761

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2005

List of Papers

I Quantitative Analysis of Tryptic Protein Mixtures Using Elec-trospray Ionization Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. Hagman C, Ramström M, Håkansson P, Bergquist J. J. Proteome Res. 2004, 3, 587-594.

II Reproducibility of Tryptic Digestion Investigated by Quantitative Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. Hagman C, Ramström M, Jansson M, James P, Håkansson P,Bergquist J.J. Proteome Res. 2005, 4, in press.

III Depletion of High-Abundant Proteins in Body Fluids Prior to Liquid Chromatography Fourier Transform Ion Cyclotron Reso-nance Mass Spectrometry. Ramström M, Hagman C, Mitchell J.K, Derrick P.J, HåkanssonP, Bergquist J.J. Proteome Res. 2005, 4, in press.

IV A New Method for the Accurate Determination of the Isotopic State of Single Amide Hydrogens within Peptides using Mass Spectrometry. Buijs J, Håkansson K, Hagman C, Håkansson P, Oscarsson S. Rapid Commun. Mass Spectrom. 2000, 14, 1751-1756.

V Inter- and Intra-Molecular Migration of Peptide Amide Hydro-gens during Electrospray Ionization, Studied with FTICR Mass Spectrometry. Buijs J, Hagman C, Håkansson K, Richter J.H, Håkansson P, Oscarsson S. J. Am. Soc. Mass Spectrom. 2001, 12, 410-419.

VI Inter-Molecular Migration During Collisional Activation Moni-tored by Hydrogen/Deuterium Exchange FTICR Mass Spec-trometry. Hagman C, Håkansson P, Buijs J, Håkansson K.J. Am. Soc. Mass Spectrom. 2004, 15, 639-646.

VII A Site-Specific Investigation with Hydrogen/Deuterium Ex-change Using Collision-Induced Dissociation and Electron Cap-ture Dissociation. Hagman C, Tsybin Y, Håkansson P. In Manuscript

All published papers are reprinted with the permission from the journals in which they were originally published.

Author contribution:

I am responsible for the ideas behind paper VI and VII. The author and Mar-gareta Ramström are equally responsible for the development of the idea in Paper II. I wrote Papers I, II, VI, VII. I contributed to the writing in of Pa-pers III-V. I am responsible for the planning of Paper I, II, V-VII and for carrying out experiments and analysis in Paper I, V-VII. In Paper II most of the experi-ments and analysis was performed by me and in Paper III some of the ex-periments were performed by me. In Paper IV Jos Buijs and I are equally responsible for the analysis and experiments.

Related papers not included in this thesis:

A Novel Mass Spectrometric Approach to the Analysis of Hormonal Peptides in Extracts of Mouse Pancreatic Islets. Ramström M, Hagman C, Tsybin Y, Markides K, Håkansson P, Salehi A, Lundquist I, Håkanson R, Bergquist J. Eur. J. Biochem. 2003, 270, 3146-3152.

Conference presentations including material from the thesis:

1) Accurate Determination of the Isotopic State of Single Amide Hydrogens in Peptides. Buijs J, Håkansson K, Hagman C, Håkansson P, Oscarsson S. 48th ASMS, Long Beach, USA, June 2000.

2) Migration of Peptide Amide Hydrogens in the Gas-Phase. Håkansson K, Hagman C, Tsybin Y, Richter J, Håkansson P. 48th ASMS, Long Beach, USA, June 2000.

3) Structure Studies of Peptides in Solution and in Gas-Phase Using Hydro-gen/Deuterium Exchange Techniques in Combination with a 9.4 T Mass Spectrometer. Hagman C, Buijs J, Håkansson P. Analytical Days, Stock-holm, Sweden, June 2001.

4) Structural Information Obtained with Hydrogen/Deuterium Exchange. Hagman C, Buijs J, Håkansson K, Håkansson P. 6th European FTICR con-ference, Kerkrade, Netherlands, October 2001.

5) Hydrogen/Deuterium Exchange on Residue-Specific Level in Peptides Studied with Electron Capture Dissociation and a Fourier Transform Ion Cyclotron Resonance 9.4 T Mass Spectrometer. Hagman C, Budnik B, Tsybin Y, Håkansson P. 50th ASMS, Orlando, USA, June 2002.

6) Quantitative Analysis of Globally Labeled Tryptic Protein Mixtures Using Electrospray Ionisation Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. Hagman C, Ramström M, Håkansson P, Bergquist J. 51th

ASMS, Montreal, Canada, June 2003.

7) Quantitative Analysis of Tryptic Protein Mixtures Using Electrospray Ionization Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. Hagman C, Ramström M, Håkansson P, Bergquist J. Swedish ProteomeSociety Symposium, Lund, Sweden, December 2003.

8) Electrospray Ionization Fourier Transform Ion Cyclotron Resonance Mass Spectrometry for Global Quantitative Analysis of Multiply Labeled Protein Mixtures. Hagman C, Ramström M, Jansson M, James P, Håkansson P, Bergquist J. 52th ASMS, Nashville, USA June 2004.

Contents

1. Introduction 91.1 Protein Structure and Activity 101.2 Protein Expression and Complex Solutions 101.3 Proteomic Research 12

2. Techniques 132.1 Mass Spectrometry and Analysis of Complex Samples 132.2 Protein Identification 142.3 Relative Quantification 14 2.3.1 Quantitative Markers 16 2.3.2 Analysis of Quantitative Experiments 172.4 Structure Elucidation 182.5 Solution-Phase Hydrogen/Deuterium Exchange 18 2.5.1 Hydrogen/Deuterium Exchange Theory 19 2.5.2 High-Resolution Hydrogen/Deuterium Exchange 20 2.5.3 Experimental Conditions for Solution-Phase Exchange 202.6 Gas-Phase Hydrogen/Deuterium Exchange 21 2.6.1 Intra-Molecular Exchange 21 2.6.2 Inter-Molecular Exchange 21 2.6.3 Experimental Set-up for Gas-Phase Studies 22

3. Mass Spectrometric Methods 243.1 Fourier Transform Ion Cyclotron Resonance Mass Spectrometry 243.2 Electrospray Ionization 253.3 Collision-Induced Dissociation 263.4 Electron Capture Dissociation 283.5 Sample Introduction and Up-front Separation 29

4. Results and Discussion 314.1 Quantitative Analysis in Tryptic Protein Mixture 314.2 Quantitative Analysis in Body Fluids 334.3 Quantitative Evaluation of Depletion Kits 344.4 Hydrogen/Deuterium Exchange in Solution 354.5 Hydrogen/Deuterium Exchange in Gas-Phase 39

5. Conclusions 415.1 Quantitative Analysis 415.2 Structural Analysis 41

6. Future Aspects 427. Acknowledgments 438. Svensk sammanfattning 449. References 46

Abbreviations

2D Two-Dimensional BSA Bovine Serum Albumin CRM Charge Residue Model CE Capillary Electrophoresis CID Collision-Induced Dissociation DIGE Difference Gel Electrophoresis DNA Deoxyribonucleic acid D4 1-Nicotinoyloxy (D4)-Succinimide ester ECD Electron Capture Dissociation ESI Electrospray Ionization FAB Fast Atom Bombardment FFT Fast Fourier Transform FTICR Fourier Transform Ion Cyclotron Resonance GIST Global Internal Standard Technology H4 1-Nicotinoyloxy (H4)-Succinimide ester HDX Hydrogen/Deuterium Exchange HPLC High Performance Liquid Chromatography HSA Human Serum Albumin IRMPD Infra-Red Multi-Photon Dissociation IEM Ion Evaporation Model ICAT Isotope-Coded Affinity Tags LC Liquid Chromatography MCAT Mass-Coded Abundance Tags MALDI Matrix Assisted Laser Desorption/Ionization MS Mass Spectrometry NMR Nuclear Magnetic Resonance PA Proton Affinity PAGE Polyacrylamide Gel Electrophoresis PTMs Post-Translational Modifications QUEST Quantification Using Expressed Sequence Tags RF Radio Frequency RPLC Reversed Phase Liquid Chromatography RNA Ribonucleic Acid SMTA S-Methyl Thioacetimidate SMTP S-Methyl Thiopropionimidate SLD Soft Laser Desorption SORI Sustained Off-Resonance Irradiation

9

1. Introduction

Today, mass spectrometry is indispensable for identifying and characterizing proteins and peptides in life science research. In this thesis, methods have been developed for quantitative and structural analysis of proteins and pep-tides with Fourier transform ion cyclotron resonance, (FTICR),1-3 mass spec-trometry.

Quantitative measurements of proteins in complex solutions such as body fluids e.g. cerebrospinal fluid, (CSF), and plasma are performed with the intent to monitor up- or down-regulations as a consequence of e.g. disease state. These changes can be studied with the use of global labeling methods. Two sets of global markers with different properties were used for quantita-tive analysis in this thesis; S-Methyl Thioacetimidate, (SMTA), and S-Methyl Thiopropionimidate, (SMTP),4, 5 and the isotopic variants of 1-Nicotinoyloxy succinimide esters, the [H4]- and [D4]- markers.6

In Paper I, quantitative analysis of globally labeled tryptic protein mix-tures, using the SMTA- and SMTP-markers, is performed with reversed-phase High performance liquid chromatography,7, 8 (HPLC)-FTICR mass spectrometry. In Paper II the reproducibility of enzymatic digestion and de-rivatization, are investigated on a tryptic protein digest of CSF. Two sets of markers for global labeling are used; the SMTA- and SMTP-markers and the [H4]- and [D4]- markers. The lower limit for which it is possible to determine up- or down-regulations of proteins and peptides with HPLC-FTICR mass spectrometry is determined.

Proteins and peptides in body fluids are present in a wide dynamic range. Therefore, more abundant proteins can be depleted prior to analysis in order to increase the possibility for detecting low abundant proteins. In Paper III, two commercially available depletion kits are applied on plasma and CSF and evaluated with quantitative analysis using the SMTA- and SMTP-markers.

Hydrogen/deuterium exchange, (HDX), in conjugation with mass spec-trometry has been used for a decade for elucidating protein structures and dynamics.9-12 High-resolution information can be obtained by applying Col-lision-induced dissociation, (CID),13, 14 and determining the average isotopic incorporation in the intermediate residue.15, 16

In this thesis, site-specific information are obtained by comparing the iso-topic distributions of consecutive peptide fragment ions, see Paper IV-VII. In Paper IV, isotopic solution exchange processes of amide hydrogen at spe-cific residues in linear peptides are investigated with CID-FTICR mass spec-trometry. The exchange rates for amide hydrogens surrounded by side chains

10

of different sizes are determined and compared with theoretical exchange rates. In Paper V, residues where the isotopic solution-phase exchange proc-esses are difficult to monitor with CID-FTICR mass spectrometry are inves-tigated. In Paper VI, different fragment types are used to follow the isotopic solution-phase exchange process at specific residues. Gas-phase exchange reactions with D2O and CID-fragments are performed in the hexapole area and evaluated with respect to fragment type and size. In Paper VII, two complementary fragmentation methods, Electron capture dissociation, (ECD),17, 18 and CID, are used to monitor the isotopic solution-phase ex-change at a specific residue.

1.1 Protein Structure and Activity

The central dogma in biology states that the genetic information in a eu-karyotic cell is transcribed from Deoxyribonucleic acid, DNA, to Ribonu-cleic Acid, RNA, and from RNA into proteins through translation.19, 20 Dur-ing translation the amino acids are forming peptide bonds through a dehydra-tion reaction and are linked together to a polypeptide chain. Many proteins are large complexes which consist of several polypeptide chains.21

The native protein structure is not static. It can be seen as an equilibrium of un- and refolding-processes. The refolding occurs spontaneously and de-pends on the primary sequence.22 Protein folding of some proteins is assisted by molecular chaperones which prevent the single subunits to associate until the proper partners appear.23

Proteins are involved in most cell functions. They can be involved in regulatory24, 25, transport26 and signaling26 processes. Enzymes and receptors are also proteins that perform crucial biochemical tasks.27-29 The vast variety of protein functions can be explained through the many different three-dimensional forms proteins can adapt.

The protein activity can be altered through highly specific biochemical in-teractions with other molecules. These interactions involve protein binding sites, which are small patches located on the protein surface.30, 31 These sites can be involved in protein-protein32, protein-ligand and protein-antibody33

interactions.

1.2 Protein Expression and Complex Solutions

In the eukaryotic organism some proteins and peptides are expressed in all cells, whereas others are specific for a certain cell type.34 Often, these pro-teins and peptides are responsible for some characteristic processes. For

11

example, pancreatic beta-cells are producing insulin, which will make other cells able to store glucose. Red blood cells contain hemoglobin, which trans-ports oxygen in the body.

The most common regulatory step of the protein expression is the tran-scriptional control.35, 36 It is a dynamic process depending on factors such as cell cycle, stress and pharmacological treatment. Since the protein expres-sion is a dynamic process, changes in the protein and peptide patterns have been correlated to different disease states. These proteins and peptides can thus act as biomarkers.

Within the human body fluids, e.g. blood, saliva, urine, and CSF, num-bers of biomarkers are identified. Blood, (plasma and serum), are routinely analyzed in many clinical laboratories around the world.

In this thesis, CSF and plasma have been studied as potential sources for clinical markers. In plasma, the proteins can be divided into three groups. First, the abundant proteins which are secreted by the liver and intestines e.g. albumin and transthyretin. Second, the tissue leakage proteins, in this group important diagnostic markers are found. For example, cardiac troponines, creatine kinase or myoglobin are used in the diagnostics of myocardic infarc-tion. The third group is the signaling molecules e.g. cytokines, which are locally generated as a response to infection or inflammation. Cytokines are also important diagnostic markers.37

CSF is continuously secreted into the ventricles by the choroids plexus and circulates over the brain until it is absorbed by the arachnoid villi. It is estimated that the turnover rate is a replacement of the total volume, 150 mL, three to four times per day. Since the fluid is in contact with the brain some proteins and peptides as well as other low molecular species are continu-ously excreted, and therefore to a certain extent reflect the biochemical status of the brain.38 Biomarkers for neurological diseases such as Alz-heimer’s disease39, 40 and vascular dementia41 have been identified as well as for psychiatric disorders such as schizophrenia.42

One of the difficulties with the analysis of body fluids is the wide dy-namic range of proteins and peptide concentrations. For instance half of the total protein content in CSF is human serum albumin, (HSA),43 whereas other biologically and clinically more interesting proteins are present in much lower concentrations.

12

1.3 Proteomic Research

The systematic study of protein properties such as structure, function, inter-actions with other proteins and comparison between systems in health and disease is defined as Proteomics.44, 45

A crucial step for the mass spectrometric analysis of fragile bimolecules was the development of the soft ionization techniques; Soft laser desorption, (SLD),46 Matrix-assisted laser desorption, (MALDI),47 and Electrospray ionization, (ESI).48 These techniques, allow the transfer of large bio-molecules from solution into gas-phase without decomposition.

Since the biological samples often are very diverse, the sample complex-ity can be reduced by applying a separation step prior to mass analysis. For example in Capillary electrophoresis, (CE),49, 50 the separation is performed with respect to the analytes electrophoretic mobility. In Reversed-phase chromatography the different components are separated according to their hydrophobicity. 8, 51 These separations, desalts and pre-concentrates the sam-ple prior to analysis. After separation, the amount of proteins that are ionized simultaneously are reduced and therefore the ion suppression is less promi-nent.52

In eukaryotic cells, proteins are often modified after translation to in-crease the protein diversity and alter the activity. There are more than two hundred different post-translational modifications, (PTMs), described in the literature.53 The most common PTMs are; glycosylation,54 phosphorylation55,

56 and alkylation. By various mass spectrometric fragmentation techniques such as CID,57 ECD58 and Infra-red multi-photon dissociation, (IRMPD), the location of these PTMs can be determined.59, 60

The dynamic range of protein and peptide concentrations within complex samples can differ with ten orders of magnitude.37 In serum, albumin is pre-sent at 40 mg/mL and cytokines in pg/mL.61 To address this problem, sam-ples can be depleted of high abundant components prior to analysis.

The protein activity, can also be altered through up- and down-regulations of the protein expression. These changes, can be quantified by derivatizing samples corresponding to two different states with different quantitative markers. The markers, have a similar chemical reactivity but slightly differ-ent masses. After derivatization, the samples are mixed and analyzed. The signal intensity ratio between the light- and heavy-labeled peptides corre-sponds to the relative concentration ratio between the two different states.

Structural information can also be obtained by measuring the protein sol-vent accessibility by letting labile hydrogens equilibrate with isotopes in the surrounding solution. Information about isotopic exchange processes at spe-cific residues can be obtained by applying mass spectrometric fragmentation methods.

13

2. Techniques

2.1 Mass Spectrometry and Analysis of Complex Samples

Many different mass spectrometric techniques have been developed during the past 100 years. Here, only a summary of the dynamical development will be given. For a comprehensive description of the mass spectrometry history, the reader is referred to e.g. the book “Measuring Mass” by Michael Gross.62

The famous British scientist Sir J.J. Thomson, that discovered the elec-tron 1897, is also the father of mass spectrometry.63 His basic concept, for a mass spectrometer is still the same: an ion source followed by a mass disper-sive element and finally a detector. In his “parabola mass spectrograph” Thomson detected light ions like H+, H2

+, O+ and CO+.63, 64

The instrumentation was further developed by Thomson’s students Aston, that discovered many isotopes,65 and Dempster that developed an 180o mag-netic focusing spectrometer with an electron impact ion source.66

The first commercial mass spectrometer was installed 1943 in an oil com-pany and for many years the main users of mass spectrometry worked in the field of “petro-chemistry”. A turning point came 1974 when Ron Macfarlane showed that fission fragments from a radioactive source, 252-Cf, could be used to desorped and ionize thermally labile and non-volatile biomolecules like amino acids and peptides.67 Bovine insulin, which has a mass larger than 5000 Da, was ionized using plasma desorption.68

The development in the 80-thies was intense with the introduction of sev-eral new ionization techniques like Fast atom bombardment, (FAB),69

SLD,46 MALDI,47 and ESI.48, 70 These techniques, together with the time-of-flight technique for mass analysis, made it possible to study biomolecules with masses up to 105-106 Da.

This thesis, is focused on a rather modern mass spectrometric technology; i.e. Fourier Transform Ion Cyclotron Resonance, that was introduced in 1974 by Marshall and Comisarow.1 Some characteristic feature of the FTICR mass spectrometer are; ultra-high mass resolving power, mass accuracy in the sub ppm level, (even over a broad mass range), and sensitivity in the attomole region.2, 71

Mass spectrometry has become one of the most important techniques within proteomics during the last years. The most important parameter for protein identification is the mass measurement. Complex solutions, such as body fluids, contains numerous proteins in a broad concentration range. Dif-ferent masses, mi, can also occur in various charge states, zj, therefore the ratio, mi/zj, sometimes becomes almost identical. However, coupling a sepa-ration step prior to analysis reduces the amount of overlapping peaks as well as the ion suppression.

14

HPLC-FTICR mass spectrometry results in a further improved sensitivity, with ion trapping the detection limit is in the femtomole to attomole level.72,

73

Coupling a CE-separation step74, 75 to the FTICR results in even higher separation efficiency, shorter analysis time, and lower sample consumption compared to HPLC-FTICR mass spectrometry. The drawback with coupling the CE to the FTICR is the limited acquisition speed in the FTICR. There-fore, peaks are only present in few spectra which make it difficult to com-bine this approach with quantitative analysis. Therefore, HPLC-FTICR mass spectrometry is more suitable for quantitative determination within complex solutions.

2.2 Protein Identification

A complex sample which is enzymatically digested and analysed with HPLC FTICR mass spectrometry results in a mass chromatogram with the compo-nents separated according to their m/z-values and elution time. Another ap-proach is to use MS/MS after digestion and generate sequence tags.76 Any identification of proteins in complex solutions is dependent on a proper cali-bration of the mass chromatogram.

In this thesis, the identification was done by comparing the measured peptide masses with theoretical masses of the digested peptides.77 The accu-rate masses the calibration was performed with internal calibrants of known and abundant proteins. In order to compensate for the varying total charge during an entire HPLC-FTICR experiment the calibration peaks were evenly distributed over time in the mass chromatogram.

After calibration, an estimation of the relative mass measurement distribu-tion of a certain protein can be performed. The error distribution for a true match is narrower compared to the false matches.78

Another source of information that can be used to reduce the numbers of false matches is the distribution of the detected peptides in the protein se-quence.78

2.3 Relative Quantification

The traditional protein separation method in proteomics involves two-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis, (2D-PAGE).79, 80 On the gel, proteins are separated according to their isoelectric point and size. The drawback with this method, is the difficulties to analyze biomolecules with very high or low isoelectric points or extreme

15

molecular weight (i.e. >250 kDa and <10 kDa).81,82 Hydrophobic bio-molecules, e.g. membranproteins, are also difficult to analyze with 2D-PAGE. The sensitivity of 2D-PAGE is limited to femtomole levels depend-ant on the choice of staining.

Quantitative analysis of proteins can be performed e.g. with 2D fluores-cence difference gel electrophoesis, (2D-DIGE).83 Two different fluoropho-res are used in 2D-DIGE; the 1-(5-carboxypentyl)-1'-propylindocarbo-cyanine halide N-hydroxysuccinimidyl ester, (Cy3), and the 1-(5-carboxy-pentyl)-1'-methylindodicarbocyanine halide N-hydroxysuccinimidyl ester, (Cy5). They are structurally similar, but have different flourophores, (Cy3,

em=569 and Cy5, em= 645). Cy3 and Cy5 undergo a nucleophilic substitu-tion with the lysine -amino group of the sample. When the labeled samples are mixed and analyzed on the same 2D-gel, the color of the spot is corre-lated to the concentration ratio of the samples. The drawbacks with 2D-DIGE are that more than one protein might be present in the same spot, and thus complicating the quantitative analysis. The technique is also time con-suming, and the reproducibility of 2D-PAGE is hard to accomplish even for experienced users.

Several methods for relative quantification in liquid separation based pro-teomics have been presented during the last years. Quantitative analysis can be achieved by labeling two pools of proteins with different compounds which will result in a difference of the m/z- value between the two pools.

The most widely spread method for mass spectrometric quantitative analysis is Isotope-coded affinity tags (ICAT).84, 85 Using this method only enzymatic digested peptides containing cysteines will be isolated and further analyzed. This reduces the sample complexity. The drawback is that only sequences containing cysteine residues can be analyzed, and one out of seven proteins do not contain cysteine residues.61 The first generation of ICAT-markers was either deuterated (D8) or undeuterated (D0). The new generation of ICAT markers, contains 13C or 12C to reduce the effect of iso-topic shift in reversed phase separations, using C18 columns.86 Another ap-proach to global labeling is the Mass-coded abundance tag, (MCAT). The MCAT-markers reacts with the C-terminal lysine residue of the tryptic pep-tides.87 The lysine residue is converted through a guanidation reaction to a homoargine residue.88, 89 The drawback with this method is that the ioniza-tion efficiency of the native and derivated sample is different. Therefore, it is difficult to correlate the intensity ratio with the concentration ratio.

Yet another approach, is the Global internal standard technology, (GIST). The introduction of the labels takes place by using different isotopes, e.g. 16O and 18O, during the enzymatic digestion. The difficulty with this ap-proach, is that several isotopes might be incorporated during the tryptic di-gestion, and that back-exchange can occure.90, 91 The use of ESI, results in multiply charged peptide ions.48 Therefore, only mass spectrometers that provide high resolution can be used in these studies, since the mass/charge

16

difference between two multiply charged labeled samples can be difficult to distinguish at low resolution.

In studies on prokaryotic and simple eukaryotic species, the stabile iso-topes can be introduced in vivo during cultivation. This approach is not pos-sible to use for higher eukaryotic species. As an example, two cultures of Saccharomyces cerevisiae, a wild typ and mutant population, were grown on medium that contained the natural abundance of nitrogen, whereas the other medium was enriched with 15N. In that study, 42 abundant proteins were investigated. The quantitative analysis was initiated by separating the sample using HPLC and SDS-PAGE. Most of the investigated proteins, 39 out of 42, displayed a change in the intensity ratio that could be explained by the experimental error, approximately 10%, of the measurement. Three, of the investigated proteins were down-regulated with approximately 40% as a response to the mutation.92

2.3.1 Quantitative Markers

In this thesis, peptides obtained by enzymatic cleavage were globally labeled with two types of labels. The Quantification Using Expressed Sequence Tags, (QUEST),-markers are S-Methyl Thioacetimidate (SMTA) and S-Methyl Thiopropionimidate (SMTP).5 These markers, will react with the N-terminal amine group and the -amino group of the lysine residues.4 The labeling is regarded as global, since each enzymatic peptide contains at least one chemical group which is reactive. The amidation reaction, is efficient and fast.

a) b)

Figure 1. a) the SMTA-marker and b) the SMTP-marker.

The derivatization, of a suitable chemical group, with a SMTA-marker will result in a mass shift of 41.030 Da. The SMTP-marker, contains an ad-

CH3

S

CCH3

CH3

S

CCH3 NH

CH3

S

CCH2 NH

CH3

CH3

S

CCH2

CH3

17

ditional CH2-group, and a covalent modification adds 55.042 Da to the pep-tide mass.

Peptides labeled with the SMTP-marker, are more hydrophobic compared to the corresponding SMTA-labeled peptide. Consequently, using reversed phase HPLC, the SMTA-labeled peptide will elute faster compared to the corresponding SMTP-labeled peptide. The other two markers, used in this thesis, are isotopic variants of 1-Nicotinoyloxy succinimide esters; the [H4]- and [D4]-marker. These markers, will by a succinylation reaction covalently modify the N-terminal amine group. At pH 5, the -amino group of the ly-sine residue are protonated, and therefore non-reactive. The derivatization of a tryptic peptide, with the [H4]- or [D4]-marker, results in a mass shifts of 106.029 Da or 110.034 Da, respectively.

When separating the labeled [H4]- and [D4]-peptides, with reversed phase chromatography, the difference in elution time will depend on isotope effects during separation. Therefore, the difference in elution time between a [H4]- and [D4]-labeled peptide pair will be less prominent compared to the corre-sponding SMTA- and SMTP-labeled peptide pair.

N

ON

O

HH

HH

ON

ON

O

HH

HH

O N O

ON

O

DD

DD

N O

ON

O

DD

DD

a) b)

Figure 2. The isotopic variants of 1-Nicotinoyloxy succinimide ester. In a) the [H4]-marker and in b) the [D4]-marker are depicted.

2.3.2 Analysis of Quantitative Experiments

In this thesis, the analysis of the mass chromatogram began with applying an algorithm that picked all the peaks above a certain threshold value. Another algorithm reduced the peak list into isotopic clusters. When performing quantitative analysis of a complex sample analyzed with HPLC-FTICR mass spectrometry, several thousands of peaks could be found in the mass chro-matogram.

18

Lists, of tryptic peptide masses of specific proteins were adapted to the covalent modifications with the QUEST- or [H4]- and [D4]- markers. These new peak lists were used to identify labeled peptides from the mass chroma-togram. Peptides corresponding to the same charge state and labeled with the same numbers of labels of either the light- or heavy-marker were matched. The intensity ratio between a matched peptide pair was used as a relative measurement of the concentration ratio between them.

2.4 Structure Elucidation

High-resolution information about the protein structure can be obtained by methods such as X-ray crystallography93 and multidimensional nuclear mag-netic resonance, (NMR), spectroscopy.94 NMR spectroscopy, allows the investigation of dynamics and intermolecular interactions of proteins in solu-tion and it has developed into an important technique for structure-based drug design.95, 96 The drawback for X-ray crystallography is that crystals suitable for analysis must be large, and thus requires a high initial protein concentration. Approximately 30%, of the total protein expression consists of membrane proteins which can not easily be crystallized. The membrane proteins are especially important for the life sciences, since the majority of all medical drugs interact through them.

Mass spectrometry in combination with Hydrogen/deuterium exchange, HDX, has been used for a decade to obtain information about protein fold-ing,10, 97 protein-ligand interactions,11 effects of point mutations98 and struc-tural changes upon adsorption on solid surfaces.99 In this thesis, mass spec-trometry and hydrogen/deuterium in conjugation with physical fragmenta-tion methods has been used to obtain site-specific information about isotopic exchange processes of amide hydrogens in peptides.

2.5 Solution-Phase Hydrogen/Deuterium Exchange

The theoretical equations, that describes the Hydrogen/Deuterium exchange, were written by Linderstrøm-Lang in the mid-1950s.100 During the 1960s, liquid scintillation technology was developed and the amount of isotopic exchange within a sample could be determined. Size exclusion chromatogra-phy was introduced as a useful method during this time period. In the 1980s, the proteins were enzymatic digested and separated by HPLC prior to analy-sis with liquid scintillation analysis. This improved the spatial information obtained with HDX.

19

In the same era the two-dimensional NMR spectroscopy was developed.94

The technique provides structural information from smaller proteins in solu-tion at Ångström resolution.

The first study combining solution-phase HDX with mass spectrometry was published in 1991. In that paper, Chait et al. described the conforma-tional changes induced in ubiqutine by changing the solvent composition.9During the last decade, mass spectrometry and HDX has proven to be an excellent analytical tool. Protein folding and unfolding has been studied withexchange rates ranging from milliseconds to days. Conformational changes of the catalytic domain, induced by phosphorylation has been studied.101 The maturing viral capside, was investigated and the dynamic expansion of the different viral segments were followed during that process. 102 The chaperon Gro-EL interaction with a vast variety of unfolded proteins in E. coli was probed and the results were in agreement with in vitro studies of the folding mechanism.103 In the HDX experiments the resolution is linked to the frag-mentation method.16, 104 Enzymatic cleavage will provide medium resolution since the incorporation in each peptide fragment is measured.

2.5.1 Hydrogen/Deuterium Exchange Theory

Hydrogens bound to O, N, and S are exchanging continuously with isotopes the surrounding solution and are therefore called labile hydrogens.

The exchange rate, kex, of the amide hydrogen in unstructured peptides is described in Equation 1. The reaction is catalyzed by hydroxide and hy-dronium ions in pH-dependent reactions and by water independently of pH.105 Reference values for the isotopic exchange at any amide hydrogen in a linear peptide were determined with polyalanine as a reference peptide.106

(1)WOHHex kOHkHkk

The isotopic exchange rate, as a function of pH, has its minimum at 2.7-2.9. A further reduction of the exchange rate, by a factor of ten, can be ob-tained by lowering the temperature from twenty degrees to zero degrees.

In the absence of folded structures, the HDX rate of the amide hydrogens is dependant on the shielding and the inductive effects of the closest neighboring side chains.106, 107 The shielding effect is larger for the side chains which have the amide hydrogen to the left (L-peptide), compared to side chains which have the amide hydrogen to the right (R-peptide).108 La-bile hydrogens situated on the side chains equilibrate instantly with the sur-rounding solution.

20

In folded structures, the labile hydrogens are shielded form the solution, and therefore exchange more slowly. The protein secondary structure, can be investigated by monitoring the exchange rate of amide hydrogens along the peptide backbone. 109 Conformational changes, are manifested as changes in the solvent accessibility, and results in changes of the isotopic exchange rate.

2.5.2 High Resolution Hydrogen/Deuterium Exchange

The hydrogen deuterium exchange of individual amide hydrogens can be monitored by combining HDX and mass spectrometric fragmentation meth-ods. Site-specific information from oxidized and reduced E.coli thioredoxin was obtained with HDX CID, and compared to results obtained with NMR. The study showed excellent correlation between the data obtained with the two methods.110

The advantage of using high resolution HDX mass spectrometry, is that large and more or less non-soluble proteins can be analyzed. For transmem-brane peptides, site-specific information on peptide-membrane interactions were determined using a- and y-type fragment ions.111 The fibrillary forma-tion within insulin, was investigated by Tito et al. using HDX.112 The study showed that the increased solvent accessibility of the C-terminal region, triggered the conversion from the soluble to the fibrillary form. The results were consistent with NMR studies.

2.5.3 Experimental Conditions for Solution-Phase Exchange

In this thesis, the isotopic solution-phase exchange was performed by let-ting the peptides totally exchange all labile hydrogens in D2O overnight. By mixing the deuterated peptide-solution with a spraying solution, containing H2O, the labile hydrogens will immediately start to equilibrate.

The , is the D/(D+H)-ratio in solution as shown in Equation 2. The kex, is the isotopic exchange rate of the amide hydrogen. The pD, is obtained by determining the distribution function for the amide hydrogen of interest, see Paper IV. If DHX only takes place in solution, the sequence specific ex-change of single labile amide hydrogens should obey the equation. Any de-viation from this equation is explained by gas-phase exchange.

)2()(1 texkD ep

21

2.6 Gas-Phase Hydrogen/Deuterium Exchange

The gas-phase proton affinity, (PA), kcal/mol of the protonated substrate and deuterated reagent are central parameters for gas-phase exchange reactions between simple ions. The general mechanism consists of three steps: the formation of the exchanging complex, the proton transfer, and the dissocia-tion of the exchanging complex.113

For the exchange to occur, the energy gained by forming the exchanging complex must exceed the threshold for internal proton transfer. The thresh-old mainly depends on the proton affinity difference between the two unpro-tonated species. It has been reported, that HDX can not occur in the gas phase between species that differs by 20 kcal/mol while other studies have observed isotopic exchange between species that differs by 50 kcal/mol.114,

115 An ion which easily donates a proton is more acidic and has a low PA-value. Whereas a higher PA-value, is associated with ions that have more basic properties.

2.6.1 Intra-Molecular Exchange

Intra-molecular gas-phase exchange reactions are often referred to as scram-bling. McLafferty et al. investigated cytochrome C activated with Sustained off-resonance irradiation, (SORI), and reported extensive scrambling.116

In several other studies, the scrambling was minimal or restricted to cer-tain positions. These studies show that the scrambling is depending on nature of the pressure, the collision activation energy115 and the gas-phase structure of the ion.117 SORI is a very slow heating process and therefore increases the possibility of intra-molecular exchange before dissociation.118

2.6.2 Inter-Molecular Exchange

Gas-phase exchange of labile hydrogens, requires the presence of two ex-changing protons and a basic non-protonated site. Ab initio calculations show that the exchange between the N-terminus and D2O, occurs through what is referred to as a relay mechanism, see Figure 3. The inter-molecular exchange between the N-terminal of the protonated peptide and D2O is initi-ated when a proton is shuttled from the site of protonation onto D2O. Simul-taneously the transfer of a deuteron from D2O to a less basic site on the pep-tide takes place.119, 120

22

N+H

H

H

CO

O

D

D

O

D

H

NH

H

C

O+

N

H

N

H

D

N+H

H

H

CO

O

D

D

O

D

H

NH

H

NH

H

C

O+

N

H

N

H

N

H

N

H

D

Figure 3. The inter-molecular exchange between the N-terminal and D2O is referred to as a relay mechanism.

Kaltashov et al. proposed a model for gas-phase exchange reactions be-tween peptide ions.117 The exchange process is initiated by a complex-formation between the reactant gas and a protonated site. The hydrogen ex-change occurs within that complex. Thereafter, the complex dissociates and intra-molecular exchange can occur. A high internal energy of the peptide ions will decrease the complex-formation rate, and will consequently lead to less incorporated isotopes in the gas-phase.115 The nature of the charge car-rier also effect the inter-molecular exchange. Cyclic peptide molecules cationized with sodium, were to less extent involved in inter-molecular gas-phase exchange compared to the protonated species.117

2.6.3 Experimental Set-up for Gas-Phase Studies

Gas-phase experiments were performed to monitor the gas-phase reactivity of the CID-fragment ions.121 In order to get comparable results, a defined amount of ions have to react with D2O-molecules for a defined time period. Since the ion source is continuously producing ions, a mechanical shutter was used to determine amount of ions participating in the gas-phase reac-tions, see Figure 4.

The mechanical shutter is of stainless steel, mounted in front of the capil-lary inlet, and held at the same potential as the capillary front. The shutter can be manipulated trough a pulse-width modulated servo into two different positions in which one of them allows the ions to enter the mass spectrome-ter.122 The system is controlled from the computer by the Bruker software.

23

Electrospray Capillary V Hexapole

Skimmer Gate

Electrospray Capillary V Hexapole

Skimmer Gate

Figure 4. The peptide ions were accumulated in a hexapole by applying a potential on the gate. The reaction time was fixed for all experiments. The pressure in the hexapole area was varied between 1·10-5 mbar and 5·10-5 mbar by leaking in gase-ous D2O. A shutter was mounted in front of the entrance to the mass spectrometer to control the amount of accumulated ions.

24

3. Mass Spectrometric Methods

3.1 Fourier Transform Ion Cyclotron Resonance Mass Spectrometry

In a Fourier transform ion cyclotron resonance, (FTICR),1 mass spectrome-ter, the ions to be studied are generated in an external ion source, e.g. using the electrospray ionization technique. The ions are then transferred and in-jected into a Penning trap inside a magnet with a strong and homogeneous magnetic field.123, 124

Due to the Lorenz force on the ions, they will rotate around the field lines with a cyclotron frequency, fc. The frequency is inversely proportional to the mass over charge, m/z, for the ion as shown in Equation 3.

(3)2mzBfc

By measuring the ion cyclotron frequency in a homogeneous magnetic field, the m/z-ratio can be determined and thereby the mass. The frequency can be measured more accurately than any other experimental parameter and the cyclotron frequency is independent of the initial kinetic energy of the ions. Therefore, the FTICR technique provides higher mass resolving power than any other type of mass measurement.

The role of the magnetic field is to keep the ions in a circular orbit in a plane perpendicular to the magnetic field. In order to confine the ions inside the cell a trapping potential is applied, see Figure 5a. As a consequence of the applied electrical field, the ions undergo harmonic oscillations parallel to the z-axis. The ions will also move in a circular-like path in the xy-plane. This motion is refereed to as the magnetron motion.

To measure the cyclotron frequency, the ions are excited by applying a radio frequency, (RF), signal containing a broad band of frequencies to a pair of electrode plates in the cell.125 When a frequency equals the cyclotron frequency of a particular ion, they absorb energy and are coherently acceler-ated into larger orbits, see Figure 5b.126

After the RF-signal has been turned off, the image current produced by the ions are recorded through a second pair off electrodes, see Figure 5c.127

In contrast to other mass analyzers, the FTICR detection is non-destructive. Collisional damping mainly causes the decay of the image current. There-fore, the mass analysis takes place in ultra high vacuum.

25

+ +

a) b) c)

+ ++ +

a) b) c)

Figure 5. The principals of FTICR mass spectrometry, in a) Ion trapping for posi-tive ions b) RF-excitation and c) Ion detection.

The image current is amplified and sampled as a function of time. By ap-plying the Fast Fourier Transform, (FFT), technique,128 the time signal is converted to the corresponding frequency spectra and the m/z-values are determined from Equation 3.

Of all mass spectrometric techniques available today, FTICR gives the highest mass resolving power and mass accuracy in the ppm region over a broad mass range.2, 71 The accurate mass determination affect the reliability of protein identification, since it allows more precise settings using database identification, and thus reducing the risk of miss-matching.129

In this thesis, all measurements have been done using the Bruker Apex II FTICR mass spectrometer equipped with a passively shielded 9.4 Tesla su-perconducting high field magnet. The 9.4 T magnet provides a high mag-netic field and consequently improves FTICR-parameters such as; signal-to-noise level, dynamical range, mass selectivity and resolution.130

For further reading several reviews by Marshall are recommended.3, 131, 132

3.2 Electrospray Ionization

Electrospray ionization, (ESI), produces multiply charged ions at atmos-pheric pressure. The technique has had a tremendous impact on the devel-opment of mass spectrometry during the last years, since it allows the analy-sis of large, non-volatile biomolecules. 48, 70

In general, positive ionization electrospray can be used for the analysis of medium to high mass organic compounds, biological molecules such as pep-tides, proteins and glycoproteins.133

The working principal of ESI is that a potential difference of about 3-4 kV is applied between the electrospray needle and a counter electrode inside the mass spectrometer. The electrical field results in charge separation with the formation of a Taylor cone. When the droplet size reaches the Rayleigh

26

limit, droplets with excess charge will detach from the cone. These charged droplets experience the electrical field and are dragged towards the entrance of the mass spectrometer.

There are two models for the ion formation in the electrospray process.134

In the charge residue model, (CRM), the evaporation of solvent molecules leads to an increased charge density that will cause larger droplets to divide into smaller droplets until a bare ion is produced. There is a consensus that large ion molecules are formed according to the CRM.135, 136 In the ion evaporation model, (IEM), where the increased charge density leads to that the columbic repulsion exceeds the surface tension and eventually single gas-phase ions are produced. The hydrophobicity of the analyte causes its affinity for the droplet surface and results in increased chargeability.137, 138

Therefore, labeling the analyte with markers containing hydrophobic com-ponents, enhances the ionization.5

3.3 Collision-Induced Dissociation

Collision-induced dissociation, (CID),13, 139 can be applied in many mass spectrometers.140, 141 The main advantage of the method is its simplicity, in the high pressure area only one voltage has to be increased in order to obtain fragments. The potential difference between the nozzle and the skimmer will accelerate the ions into rest-gas molecules. When an ion with high transla-tional energy undergoes inelastic collisions, a part of the translation energy is converted to internal energy.

At dissociation, some energy is released as kinetic energy of the different fragment ions. A proper balance between the translation energy and the pres-sure results in extensive fragmentation.142 The fragmentation could be initi-ated with the migration of a proton to an amide nitrogen followed by a charge-directed fragmentation. This proposed fragmentation mechanism is called the mobile proton model.143

CID mainly leads to the dissociation of the peptide bond and the forma-tion of b- and y-fragments.144 The nomenclature,145 for the fragment ions of proteins and peptides is shown in Figure 6. In the Hydrogen/deuterium ex-change experiments, the site specific information was obtained by compar-ing consecutive peptide ion fragments, e.g. fragments that only differ with one amino acid residue in mass. By following the continuous ladder of these fragments the sequence of the peptide can be deduced.

27

N - C - C - N - C - C - N - C - C - N - C - C - N - C - C - N - C - C

O O O O O

OH

H

HR1 H R2 H R3 H H R5 H R6

b5b4

R4

a3 b3 c3

x3 y3 z3 y2 y1

H+

O

Figure 6. The a, b and c-fragments includes the N-terminal part of the peptide. The x, y and x-fragments contain the C-terminal.

The b2+-ion formation starts with a charge directed attack by the N-terminal carbonyl group to form a protonated oxalozone, see Figure 7a and b.146 If the internal energy is sufficiently high, the oxalozone ring structure opens up to form an acylic acylium ion, see Figure 7c. A part of the internal energy can be released by eliminating a CO-group, resulting in the a-ionformation, see Figure 7d. The a-ion has a higher internal energy compared to the ground state b-ion.

N C

C C

OO

CN

R

H

H

H

+

R

HH

N CC

OC O

CNH

XH+

R

R

H

H

H

c) Acyl formation d) The a-ion formation

H

H

R O R

O+ NH

H

R O

H H

N C

C C

OO

+

HH

N C

C C

O

+

H

N CC

OC O

CNH

R

H

H

H

N CC

OC O

CNH

H

R

H

H

H

a) A charge directed attack b) Oxalozone formation

c) Acyl formation d) The a-ion formation

H

H

R

N C C NH

H

H

H

H

HC

H

R

+N C C N C C

H H H

N C

C C

OO

CN

R

H

H

H

+

R

HH

N CC

OC O

CNH

XH+

R

R

H

H

H

c) Acyl formation d) The a-ion formation

H

H

R O R

O+O+ NH

H

R O

H H

N C

C C

OO

+

HH

N C

C C

O

+

H

N CC

OC O

CNH

R

H

H

H

N CC

OC O

CNH

H

R

H

H

H

a) A charge directed attack b) Oxalozone formation

c) Acyl formation d) The a-ion formation

H

H

R

N C C NH

H

H

H

H

HC

H

R

+C

H

R

+N C C N C C

H H H

Figure 7. An energetic b-ion in b) can decompose into an a-ion in d ) through an acylic acylium ion in c).

28

Ab initio calculations show, that larger b-ions, n = 3, 4, also are proto-nated oxalozone structures.147 In another study, the threshold energies for y1,a1 and b2-ion formation were investigated on a model peptide.148 The lowest threshold values were obtained for the b2-ions, whereas the y1-ion formation was slightly less energetic compared to the a1-ion formation.

3.4 Electron Capture Dissociation

Electron Capture Dissociation, (ECD), was developed by Zubarev et al. in 1998.17 In connection to the mass-analyzer cell, an electron gun with an indi-rectly heated cathode is mounted. The ions are stored inside the cell, and electrons are emitted during a short pulse from the electron gun. ECD has also been implemented in a radio frequency ion trap.149

The most used model for this dissociation, is the hot hydrogen atom model.150 In this model, it is suggested that the interaction between the ion and electron first results in a neutralization and production of either an odd-electron or a radical ion. The unstable odd-electron ion dissociates through an energetic, hot, hydrogen transfer to a backbone carbonyl group to form cand z·-ions, see Figure 8.

Stronger bonds can dissociate in presence of weaker bonds e.g. disulfide bonds.18, 151 Therefore, it has been postulated that ECD is a non-ergodic process.17 According to the principal of non-ergodicity, the dissociation takes place prior to energy randomization in the ion. The mechanism of dis-sociation as well as the postulated non-ergodic feature is under debate.152-155

OH

H+

N - C - C - N - C - C - N - C - C - N - C - C - N - C - C - N - C - C

O O O O OH

HR1 H R2 H R3 H H R5 H R6R4

c3 c4

z3 · z2•

O

OH

H+

N - C - C - N - C - C - N - C - C - N - C - C - N - C - C - N - C - C

O O O O OH

HR1 H R2 H R3 H H R5 H R6R4

c3 c4

z3 · z2

O

Figure 8. Pepetide fragments obtained with Electron Capture Dissociation.

29

Endogenous peptides from mouse pancreatic islets were characterized with HPLC-ECD FTICR mass spectrometry.151 For some peptides, the se-quence information obtained with ECD was complementary to what was obtained with CID. Analysis of low abundant peptides in a complex solu-tions, during a limited time period, requires further improvements of the ECD efficiency.

In contrast to IRMPD and CID, ECD preserves labile bonds and allows localization and characterization of PTMs. Modifications such as glycosyla-tions156 and phosphorylations157 have been investigated with ECD.

ECD takes place in the ultra-high vacuum region of the mass spectrome-ter, see Figure 9. Therefore, in comparison with CID, the probability of in-ter-molecular gas-phase exchange reactions is reduced.

Capillary Skimmer Hexapole Ion optics Electron gun

Ion Gate

~ -4 kV

Vdelta

103 10-3 10-6 10-9

Pressure (mbar)

Capillary Skimmer Hexapole Ion optics Electron gun

Ion Gate

~ -4 kV

Vdelta

103 10-3 10-6 10-9

Pressure (mbar)

Figure 9. The different pressure regimes in the FTICR mass spectrometer.

3.5 Sample Introduction and Up-front Separation

The FTICR mass spectrometer provides ultra-high resolution, high mass accuracy and sensitivity.2, 3 The sensitivity can further be improved by cou-pling a separation step prior to mass analysis. This is not trivial to achieve, since ions are continuously produced at atmospheric pressure in an external ion source and the detection in the FTICR takes place at ultra-high vacuum.

A solution to this problem, was to use differential pumping and an ion ac-cumulation step prior to injection to the mass analyzer. A further improve-ment was the use of micro-electro spray capillaries, with an inner diameter of 25-50 µm, resulting in flow rates of sub µL/min.72 This reduces the gas load, and preserves the ultra-high vacuum in the detection area of the FTICR mass spectrometer. Due to reduced space charge effects, improved ionization

30

efficiency, desalting and pre-concentration of the sample, 10-15 to 10-18 M detection limits have been reached with this approach.72, 73

HPLC FTICR mass spectrometry have been performed on tryptic protein digests of body fluids such as, CSF158, 159, plasma160 and amniotic fluid161.Analyzing a tryptic protein digest of CSF with HPLC FTICR mass spec-trometry results typically in 4000-7000 detected peptides.158 In a direct infu-sion experiment of CSF with ESI FTICR mass spectrometry, approximately 600 peptides were distinguished.162

31

4. Results and Discussion

4.1 Quantitative Analysis in Tryptic Protein Mixtures

The initial experiments in Paper I, was performed to investigate the correla-tion between the concentration ratio and intensity ratio with direct infusion ESI FTICR mass spectrometry. A trypic digest of bovine serum albumin, (BSA) was labeled with one of the two QUEST-markers; the S-Methyl Thioacetimidate, (SMTA)-marker, or the S-Methyl Thiopropionimidate, (SMTP)-marker. Through an amidination reaction, the primary amines at the N-terminal and the lysine residues are covalently modified by the QUEST-markers. Samples were prepared with a SMTA/SMTP concentration ratio of 1, 2 and 5. The identification was done with a MATLAB algorithm. The total intensities of peptides labeled with equal numbers of markers and con-taining the same numbers of charges were compared to each other. Peptides which were not fully labeled were also included in the analysis.

For the concentration ratio of 1, the intensity ratio was 0.97 ± 0.71, n = 21, the concentration ratio of 2 resulted in an intensity ratio of 3.38 ± 2.84, n = 15. The concentration ratio of 5 gave an intensity ratio of 4.56 ± 3.67, n = 6. For the concentration ratio of 1, the obtained median intensity of 0.97, was in excellent agreement with the theoretical value and obtained by comparing the total intensity of 21 identified labeled peptide pairs. The deviation for the other concentration ratios could be explained by the ion suppression due to the sample complexity and consequently less identified labeled peptide pairs.

In another study, quantitative MALDI-TOF mass spectrometry was per-formed with SMTA- and SMTP-labeled tryptic protein digest of cytochrome c and hemoglobin.5 The average intensity ratios for smaller concentration ratios, (0.25, 0.5, 1, 2), were in good agreement with the theoretical value. The obtained intensity ratio corresponding to a concentration ratio of 4, de-viated with ~ 30% from the theoretical value.

In order to reduce the ion suppression and to further increase the sensi-tivity a HPLC-system was coupled to the FTICR mass spectrometer. The difference between the SMTA- and SMTP-marker is an additional CH2-group. Therefore, SMTP-labeled peptides are more hydrophobic compared to the corresponding SMTA-labeled peptide. In average the SMTA-labeled peptide eluted 70 seconds prior to the corresponding SMTP-labeled peptide, see Figure 10.

32

675 677 679 681 m/z84

89

94

99

points

675 677 679 681 m/z84

89

94

99

points

675 677 679 681 m/z20

21

22

23

min

m/z

min

SMTA-labeled peptide SMTP-labeled peptide

48

36

24

12

675 677 679 681 m/z84

89

94

99

points

675 677 679 681 m/z84

89

94

99

points

675 677 679 681 m/z20

21

22

23

min

m/z

min

675 677 679 681 m/z84

89

94

99

points

675 677 679 681 m/z84

89

94

99

points

675 677 679 681 m/z84

89

94

99

points

675 677 679 681 m/z84

89

94

99

points

675 677 679 681 m/z20

21

22

23

min

m/z

min

SMTA-labeled peptide SMTP-labeled peptide

48

36

24

12

Figure 10. A chromatogram with an identified SMTA- and SMTP-labeled peptide pair. The difference in elution time was approximately 70 seconds.

The identification of the numbers of different pairs could be set at several levels of mass accuracy. In Table 1, the number of assigned labeled pairs, median intensity ratio and range are included.

Concentration ratio Numbers of labels Assigned Intensity ratio SMTA/SMTP SMTA/SMTP SMTA SMTP matches Median Range

10 ppm1 190 190 113 1.14 ( 0.04-16.27)

2 197 188 111 2.35 ( 0.16-28.22)

5 207 129 64 8.59 ( 1.75-173.7)

5 ppm

1 134 166 84 0.95 ( 0.05-16.27)

2 125 140 70 2.14 ( 0.16-21.1)

5 130 89 40 7.18 ( 1.75-173.7)

Table 1. The median intensity ratio as well as the range of SMTA- and SMTP-labeled BSA digest was determined with HPLC-FTICR mass spectrometry.

33

The obtained median intensity rations are in good agreement with the 1:1 and 2:1 concentration ratios. Using the algorithm at 5 ppm, the obtained intensity ratios were 0.95 and 2.14. These results were obtained by analyzing the total intensity ratio of 84 and 70 identified SMTA-and SMTP-labeled peptide pairs, respectively. For the 5:1 concentration ratio, the intensity ratio of the SMTA-and SMTP-labeled peptides was overestimated with approxi-mately 40%. These results show that it may not be optimal to correctly quan-tify large up- and down regulations. Anyhow, it can clearly show that a large alteration in ratio has taken place.

4.2 Quantitative Analysis in Body Fluids

In quantitative analysis of complex samples several steps are applied prior to analysis. First, the biological material is enzymatically digested often with trypsine. Secondly, the two samples are labeled with different markers and subsequently mixed. All sample preparation is performed on the mixed sam-ple. Therefore no selective effect on the ratio should be expected.

In Paper II, the accuracy with which an up- or down-regulation in com-plex solutions can be measured with HPLC-FTICR mass spectrometry was determined. The reproducibility of tryptic digestion and the error introduced with derivatization was investigated with quantitative analysis. Two types of global markers were used in the quantitative analysis; the SMTA- and SMTP-marker and the [H4]- and [D4]-marker. Compared to the QUEST-markers the later are coeluting to a higher extent.

The result of the study showed, that comparing samples which originated from different digest events with quantitative analysis of abundant proteins using the QUEST- and [H4]- and [D4]-markers, the experimental error was not more than 30%. Up- or down-regulations over 30%, should thus be pos-sible to determine with HPLC-FTICR mass spectrometry.

As stated previously, the dynamic range in complex solutions is broad and therefore low abundant proteins are suppressed in the ionization process. An attempt was made to include lower abundant proteins in the analysis. In some experiments, a few labeled peptide pairs were identified originating from e.g. prostaglandin H2-isomerase, whereas in other experiments no la-beled peptide pairs could be assigned. To address this problem, the initial sample concentration could be increased. Another approach, would be to deplete the sample of abundant proteins prior to analysis.

Quantitative analysis with ICAT in complex mixtures has been used to elucidate specific signaling pathways. Totally 1331 proteins from Pseudo-monas areuinosa were identified. When limiting the access to magnesium, 145 proteins were differently expressed. The change in relative abundance of 72 of theses 145 proteins was 1.5- to 2.0-fold.163

34

4.3 Quantitative Evaluation of Depletion Kits

In Paper III, two commercially available depletion kits were evaluate with quantitative HPLC-FTICR mass spectrometry, using the SMTA- and SMTP-markers. The HSA-depletion kit and HSA/IgG-removal kit, were imple-mented on plasma and CSF. These kits should remove HSA and HSA/IgG, respectively from complex body fluids and other biological matrixes. The efficiency and reproducibility of the kits, are presented by the average inten-sities of some abundant components. The average intensity in depleted sam-ples, were normalized to average intensities in native plasma and CSF, see Figure 11.

a)

b)

Figure 11. The effect of depletion using two types of kits is monitored by quantitative analysis using the SMTA- and SMTP-markers. In a) Plasma and b) CSF. The me-dian values are given, the error bars show the range for three experiments.

35

In both plasma and CSF, the relative intensity of HSA is significantly de-creased after applying the HSA/IgG-removal kit. Whereas the average inten-sity for HSA in plasma after depletion with the HSA-depletion kit is close to unity. The HSA/IgG-removal kit, removes IgG efficiently in both plasma and CSF, since the average intensities for IgG are reduced, see Figure 11.

HSA constitutes ~ 50% of the total protein content in plasma. If it would be at least partly removed, lower abundant proteins would be easier to ana-lyze. Due to reduced ion suppression after depletion, an estimation of the numbers of identified proteins from the complex solutions was performed. These results show that significantly more protein could be detected after depletion, see Paper III. The concentration difference between the lowest abundant protein identified before and after applying the HSA/IgG-removal kit is approximately 10-fold, see Figure 12.

Before depletion and after depletionBefore depletion and after depletion

Figure 12. Without any depletion, the lowest abundant component in plasma identi-fied was Apolipoprotein A-I. The concentration range for the undepleted plasma sample is represented by the smaller light gray box. After depletion with HSA/IgG-removal kit, the larger darker gay box shows the concentration range of the identi-fied proteins. The lowest abundant protein was C1q complement. Modified picture from Anderson et al.37

In a study by Ramström et al. the CSF proteome was compared between healthy donors and patients with amyotrophic lateral sclerosis using HPLC FTICR mass spectrometry.159 It was possible to monitor differences in pro-tein patterns between the two groups. However, no biomarker could be iden-tified. Many clinically interesting proteins and peptides are present in low concentrations. Therefore, applying a depletion step prior to analysis should affect the outcome of these kind of studies.

36

4.4 Hydrogen/Deuterium Exchange in Solution

The HDX-process is correlated to the solvent accessibility of the labile am-ide hydrogen on the peptide backbone. For amide hydrogens on linear pep-tides, the isotopic exchange rate reflects the shielding effects of the two neighboring side chains. Amide hydrogens which are surrounded by large and bulky side chains have a relatively slow isotopic exchange rate since the solvent accessibility is reduced.

In Paper IV, hexa-glycine and hexa-tyrosine was used to demonstrate the role of side chain shielding on the isotopic exchange rate. The site-specific exchange at Gly5, the fifth residue in hexa-glycine, was determined to 0.63/min. The exchange rate at the Tyr4 residue in hexa-tyrosine was slower, 0.11/min, due to the large and bulky side chains surrounding the labile amide hydrogen, see Figure 13. The exchange rate was determine by using Equa-tion 2. Using the reference constants for the different side chains, as reported by Bay et al.,105 the exchange rate was calculated to 0.51/min for the Tyr4residue and 0.09/min for the Gly5 residue.

probability(D)

0

0.2

0.4

0.6

0.8

1

0 10 20 30 40 50time (min)

Figure 13. Due to differences in solvent accessibility, the site-specific exchange rate for the Gly5 residue in hexa-glycine is faster compared to the Tyr4 residue in hexa-tyrosine. The isotopic state of the amide hydrogen at the Tyr4 residue in hexa-tyrosine is represented by . The isotopic state of the amide hydrogen at the Gly5residue in hexa-glycine is shown by .

For alanine homopeptides, containing three to ten residues, the isotopic exchange rate was determined to 0.156 ± 0.005/min. This can be compared to the exchange rate of 0.159/min that was determined with NMR.105, 106 In Paper V, penta-phenyl alanine and penta-aspartic acid were investigated. For these peptides, the site-specific isotopic exchange process could not be monitored with CID HDX mass spectrometry. For alanine homopeptides,

37

containing more than three residues, the average deuterium content reflected the solution-phase exchange. The smaller alanine homopeptides, were de-pleted of deuterium during the mass spectrometric analysis. These observa-tions could be explained by different gas-phase structures and abilities to be involved in inter-molecular gas-phase exchange. In another study with gly-cine homopeptides, it was shown that the isotopic gas-phase exchange was strongly reduced with increasing peptide length. They suggested, that longer peptides fold into -sheet configurations with hydrogen bonds to solvate the charge site.120 Consequently, these peptides are less involved in isotopic exchange reactions in the gas-phase.

In Paper VI, site-specific investigations were performed on substance P and bombesin. The CID-fragmentation resulted in both a- and b-fragments. The isotopic state of the amide hydrogen at a specific residue could therefore be determined by two independent calculations. The result showed that the isotopic solution-phase exchange was more preserved during mass spectro-metric analysis using consecutive a-fragments compared to the b-fragments for the alanine residue in bombesin, see Figure 14. The specific features of the fragmentation mechanism might suggest a possible explanation. Peptide ions are accelerated into nebulising gas and solvent molecules during CID. The collisions lead to an accumulation of the internal energy. The peptide bond, which is the weakest bond, will dissociate when the internal energy reaches the threshold energy for dissociation. The subpopulation with high-est internal energy will form a-fragments. 146,148

probability(D)

00.2

0.40.6

0.81

0 10 20 30 40 50time (min)

CC

NC

CN

CC

NC

CN

CC

O

O

O

O

O

C

C

H2N O

H

C

C

C

N H2O

H

C

H

C

CH3

NH

b8 b9

a8 a9

(Ala)9

H

probability(D)

00.2

0.40.6

0.81

0 10 20 30 40 50time (min)

CC

NC

CN

CC

NC

CN

CC

O

O

O

O

O

C

C

H2N O

H

C

C

C

N H2O

H

C

H

C

CH3

NH

b8 b9

a8 a9

(Ala)9

H

Figure 14. The isotopic solution phase exchange for the alanine residue in bombesin was more preserved during mass spectrometric analysis using the a-fragments in the calculation, , instead of the b-fragments, .

38

A high internal energy, both before and after fragmentation, results in difficulties in forming the exchanging gas-phase complex.117 Therefore, the a-fragments should form unstable exchanging complex with other molecules in the gas-phase. Consequently, it is suggested that the solution-phase ex-change should be better preserved during mass spectrometric analysis using energetic fragments. Another explanation, is that the different types of ions adapt different gas-phase conformations and therefore are more or less sus-ceptible to gas-phase exchange.164, 165

In Paper VII, two complementary fragmentation methods, ECD and CID, were used in combination with HDX mass spectrometry. To calculate the isotopic state at the intermediate residue, the isotopic distributions of two consecutive peptide ions need to be intense. In some cases, it was possible to use consecutive c-fragments to determine the isotopic state at the amide hy-drogen position. The same exchange process was also was analyzed using b-fragments obtained with CID. In Figure 15, the exchange process at the phenylalanine residue, Phe8, in substance P was monitored as a function of time using consecutive c-fragments. More deuterium was preserved during the analysis compared to the corresponding CID experiments, see Paper VII, and the exchange process was in better agreement with Equation 2. Using ECD, the probability of gas-phase exchange after fragmentation is reduced since the fragmentation takes place inside the analyzer cell. The vacuum is approximately 10-9 mbar during ECD.

probability(D)

0

0.2

0.4

0.6

0.8

1

0 10 20 30 40 50time (min)

NC

CN

CC

NC

CN

O

O

OH

C

C

C

NH2O

H

H

H

C

C

c6

(Phe)8

c7probability(D)

0

0.2

0.4

0.6

0.8

1

0 10 20 30 40 50time (min)

NC

CN

CC

NC

CN

O

O

OH

C

C

C

NH2O

H

H

H

C

C

c6

(Phe)8

c7

NC

CN

CC

NC

CN

O

O

OH

C

C

C

NH2O

HH

H

H

C

C

c6

(Phe)8

c7

Figure 15. The isotopic exchange at the phenylalanine, Phe8 residue, was studied using consecutive fragments obtained with ECD.

39

4.5 Hydrogen/Deuterium Exchange in Gas-Phase

In Paper VI, the gas-phase exchange properties of CID fragment ions were investigated by trapping fragment ions in the hexapole area of the FTICR mass spectrometer. A mechanical shutter was used to control the amount of ions which was allowed to enter the hexapole.122 A peptide, penta-phenylalanine, which has lost all the isotopic information during analysis in the solution experiments, was used to investigate the gas-phase reactivity of peptide CID-fragments. The fragmentation process resulted in a-, b- and y-ion formation. The pressure in the accumulation hexapole was increased by leaking in D2O prior to the experiment. The different CID-fragments dis-played different gas-phase reactivity. A plot of the relative mass incorpora-tion in the fragment ion as a function of the pressure in the hexapole area can be seen in Figure 16. By externally performing the gas-phase reactions, the pressure around the analyzer cell is still ultra-high during the entire experi-ment.

0

0.002

0.004

0.006

0 10 20 30 40 50 60pressure (ubar)

0

0.0005

0.001

0.0015

0 10 20 30 40 50 60pressure (ubar)

Mdelta/MaverageMdelta/Maverage

0

0.002

0.004

0.006

0 10 20 30 40 50 60pressure (ubar)

0

0.0005

0.001

0.0015

0 10 20 30 40 50 60pressure (ubar)

Mdelta/MaverageMdelta/Maverage

a) b)

c)

0

0.002

0.004

0.006

0 10 20 30 40 50 60pressure (ubar)

Mdelta/Maverage

0

0.002

0.004

0.006

0 10 20 30 40 50 60pressure (ubar)

Mdelta/Maverage

Figure 16. The relative incorporation of deuterium into different CID fragments as a function of the pressure in the hexapole area. The relative deuterium incorporation into the a4 fragment can be seen in a), the incorporation into the b-fragments in b), = b4, = b3, = b2, and the incorporation into the y-fragments, = y4, = y3,= y2, in c).

40

The inter-molecular gas-phase exchange depends on the gas-phase struc-ture,165 and the internal energy of the exchanging components. It has been shown that the internal energy for a- , b-, and y-fragments are as follows: a > y > b.148

The gas-phase experiments performed on penta-phenyl alanine are consis-tent with these results since the more energetic a-fragment is less reactive in the gas-phase exchange compared to the b-fragment.146, 147 Within a frag-ment type, the different gas-phase exchange profiles could be related to the individual gas-phase exchange structures. Larger fragments can more easily adopt compact gas-phase structures and are consequently less accessible for gas-phase exchange.120

41

5. Conclusions

5.1 Quantitative Analysis

Complex samples were investigated with quantitative analysis using global labeling in conjugation with HPLC FTICR mass spectrometry. In a tryptic digest of BSA the median intensity ratio between two pools of SMTA- and SMTP-labeled peptides was determined. For smaller intensity ratios, (1:1, 2:1), the experimental and theoretical values were in good agreement. For the larger concentration ratio 5:1, it could be established that an alteration has taken place but it could not be accurately quantified. To estimate the variation introduced with tryptic digestion and derivatization with global markers, quantitative analysis was performed on abundant proteins in CSF. Although, labeling peptides with the [H4]- and [D4]-markers were coeluting to a higher extent, compared to the corresponding SMTA- and SMTP-labeled peptide pair, the introduced error was the same. The results estimates the lower limit for the up- or down-regulations that can be detected with quantitative HPLC FTICR mass spectrometry to 30%. Two depletion kits, the HSA/IgG-removal kit and the HSA-depletion kit, were evaluated with quantitative analysis using the SMTA- and SMTP-markers. After depletion, more low abundant proteins were identified. This is of great importance, when searching for biomarkers in complex solutions such as body fluids.

5.2 Structural Analysis

The isotopic solution-phase exchange at specific residues could be moni-tored using CID FTICR mass spectrometry. The exchange rates of amide hydrogens at specific residues in homopeptides of glycine and tyrosine were in good agreement with theoretical values and reflected the different solvent accessibilities due to side-chain shielding. At some residues, the isotopic solution exchange could not be monitored with mass spectrometry. Gas-phase studies showed that consecutive penta-phenylalanine CID-fragments had different gas-phase exchange properties. Therefore, the information about the solution-phase exchange at these residues was lost during analysis. It is crucial to be able to determine whether the absence of an exchange pro-file can be explained by protection in the solution-phase or to gas-phase ex-change during analysis. With an in-cell fragmentation method the solution phase exchange at a specific residue, which was lost with CID, could be monitored. HDX mass spectrometry is an important complement other methods in structural biology since it has several advantages such as speed, higher sensitivity and the ability to analyze larger molecules.

42

6. Future Aspects

Quantitative analysis of biomarkers in body fluids is central when relating the protein expression to disease. To understand the complexity of a disease state one must understand the dynamics of cellular activation, growth, and apoptosis. The identification of crucial proteins in proteins pathways gives the possibility to design medical strategies to slow down or stop the disease progress.

In order to succeed with this task, some issues have to be solved. Publicly accessible databases have to be constructed to identify the components in complex samples. To address the abundance problem, techniques to deplete samples of highly abundant and clinically non-interesting proteins have to be further developed. Better quality of the clinical samples to standardize clini-cal measurement and to correlate the result with biological activity is also important. In comparison with a biopsy, CSF and plasma are body fluids which are relatively easy to withdraw from the patient. Therefore, discover-ing biomarkers in these body fluids is of great clinical interest.

The combination of HDX and mass spectrometry is emerging as a new technique in structural biology since it has several advantages compared to X-ray analysis and NMR. Mass spectrometry is not size limited, the sensitiv-ity is higher, and the analysis is faster. HDX mass spectrometry can be used for studying structural changes, protein dynamics, large macromoleular com-plexes, and multi-subunit proteins. By using specific CID-fragment types, or modifying the charge carrier, the gas-phase exchange is reduced and the solution-phase exchange is better preserved during the mass spectrometric analysis. Physical fragmentation at ultra-high vacuum is a specific feature of the FTICR mass spectrometer. This reduces the possibility of fragment ions to be involved in inter-molecular gas-phase exchange and is therefore bene-ficial in high-resolution HDX mass spectrometry.

The global view of proteomics, envisions the possibility to describe the total flow of metabolic compounds in living organisms. In order to under-stand the life processes at a detailed level, mass spectrometric methods for both quantitative and structural elucidation are required. The work in this thesis clearly shows the important role of FTICR mass spectrometry in these applications.

43

7. Acknowledgments

First of all, I would like to thank my supervisor Per Håkansson for guiding me in the course of my doctorate studies with great scientific knowledge and human gentleness. Although your office is next to mine, you have let me develop rather independently during these years. It never felt intimidating since I knew you would support me whenever I would encounter problems. I am most grateful for the time we have been working together.

Jonas Bergquist, my assistant supervisor for constructive criticism and sharing your knowledge and enthusiasm for science, especially the quantita-tive and clinical aspects proteomics.

Jos Buijs, my second assistant supervisor, for introducing me to the HDX method during my masters project.

Kristina Håkansson, for giving me useful instructions concerning writing and taking care of me during my first days with the FTICR.

Johan Kjellberg, our excellent research engineer, for always being helpful and finding creative solutions to all problems.

My room mate, Margareta Ramström, for all interesting discussions and shared laughs. I can not even imagine my time in the Ion Physics corridor without you!

Another thing I really enjoyed during these years, is the international at-mosphere in the Ion Physics group. During our coffee breaks you can get reports from all around the world. I would like to thank present and past members of the Ion Physics group: Vassili, Edvard, Susanna, Fernando, Youri, Magnus, Christian, Srimali, Nadia, Jan, Eric, Laurent, Nadezhda, Jennifer, Ceasar, Gamini, Ari, Bogdan, Chris, Mikhail, Michael, Igor, Oleg, Frank, Roman, Anne, Aleneka, Yanwen, Jens, Fredrik, Sincat, Anders, Paula, Marit, Maud, Tomas, Raad, Karin, Gösta, Yngve, Inge, Rogerio, Bengt, Jonas and Göran.

All my wonderful friends for sharing great moments, interesting discus-sions, making me relax and focus on other things than science.

Finally I would like to thank my family: Mamma, Pappa, my brother Jan,Iliana and little Oscar for always supporting and encouraging me!

44

8. Svensk sammanfattning

Arbetet i den här avhandlingen visar betydelsen av FTICR-tekniken vid utvecklandet av kvantitativa och strukturbiologiska metoder inom proteomi-ken. Termen proteomik innefattar studier av proteinstrukturer, proteininteraktioner samt jämförelsen av protein- och peptiduttryck mellan olika tillstånd. Varje cell innehåller tiotusentals olika proteiner i varierande koncentration. Vissa proteiner uttrycks i samband med vissa sjukdoms-tillstånd. Dessa proteiner kan då ibland användas som biomarkörer. I den här avhandlingen, har metoder för att detektera biomarkörer i cerebro-spinal vätska och plasma utvecklats. Det är relativt enkelt för en patient att lämna dessa prover, tillskillnad från biopsier, och därför har dessa metoder större klinisk relevans vid explorativa studier.

För att detektera och kvantifiera förändringar i proteinuttryck använder man sig av olika markörer som kovalent binder till specifika kemiska grup-per hos proteiner och peptider. En inmärkning resulterar i en karakteristik ökning av massan som detekteras med Fouriertransform-joncyklotronresonans, FTICR, masspektrometern. I Paper I, har ett tryptiskt digerat av bovint serum albumin, märkts in med två kvantitativa markörer; SMTA och SMTP. De SMTA- och SMTP-inmärkta provet blandas i olika kvotförhållanden. Genom att koppla ett vätskeseparations system, HPLC, till FTICR-masspektrometern så ökar känsligheten ytterligare i analysen. Den kvantitativa analysen sker genom att man korrelerar intensitetskvoten med koncentrationskvoten. För koncentrationskvoter mellan 1 och 2 var intensi-tetskvoten i mycket god överensstämmelse. I Paper II, undersöktes vilken den lägsta upp- eller ned-reglering av vanligt förekommande proteiner man kan detektera med kvantitativ HPLC-FTICR masspektrometri. De två set av kvantitativa markörerna som användes i denna studie var SMTA och SMTP samt [H4] och [D4]. Båda markörer gav ett liknande resultat för den lägsta gräns, i vårt fall ungefär 30%, för vilken man kan studera upp- och ned-regleringar av proteiner med HPLC-FTICR masspektrometri. Protein-koncentrationerna kan skilja sig mycket åt i kroppsvätskor och därför är det svårare att detektera proteiner som är närvarande i låg koncentration jämfört med proteiner av hög koncentration. Det kan vara så att vissa biomarkörer inte kan detekteras eftersom andra proteiner finns i mycket högre koncentra-tion. Därför försöker man selektera bort vissa utav de vanligaste proteinerna ur kroppsvätskan innan analysen. I Paper III, har två kommersiella selek-tionskit evaluerats med hjälp av kvantitativ analys. Det ena kittet var anti-kroppsbaserat, HSA/IgG-kitet, och det andra, HSA- kittet, baserat på en för oss okänd mekanism. Mängden protein som fanns kvar efter selektionen med de båda kitten bestämdes med hjälp av kvantitativ HPLC-FTICR analys. Fler proteiner av lägre koncentration kunde identifi-eras efter det att de mest van-ligt förekommande proteiner selekterats bort. Den här tillämpningen ökar

45

möjligheten att identifiera biomarkörer kroppsvätskor med HPLC-FTICR masspektrometri.

I den här avhandlingen har även metoder för att studera proteinstrukturer utvecklats. Många biokemiska reaktioner induceras genom att bio-molekylerna känner igen varandra, därför är formen hos biomolekyler för-knippad med funktionen. Proteiner i lösning kan jämvikta med isotoper i den omgivande lösningen. Det är vissa väteatomer t.ex. amidväten på protein-kedjan som kan bytas ut mot den tyngre isotopen deuterium. Då detta sker i lösning, kan man dra slutsatser om proteinstrukturen. De områden på protei-net som har inkorporerat många isotoper på befinner sig på ytan av proteinet medan de inre delarna av proteinet är mer svårtillgängliga och har därför inkorporerat färre isotoper. Väte/deuterium utbytet, HDX, hos proteiner i lösning brukar kombineras med tryptisk digerering och resulterar i en upp-lösning på 10-15 aminosyror. Denna teknik har använts för att studera prote-in-protein interaktionen, proteindynamik och strukturför-ändringar inducera-de av punktmutationer.

I Paper IV-VII, har vi genom att kombinera masspektrometriska fragmen-teringsmetoder som kollisions-inducerad-dissociation, CID, och jon-elektron-infångningsdissociation, ECD, med HDX-tekniken, möjligheten att studera isotop utbyten på aminosyrainvå. Den höga upplösningen gör det möjligt att i detalj studera protein strukturen aminosyra för aminosyra. I Pa-per IV, studerades specifika utbytes processer i peptider med olika stora sidokedjor. HDX-tekniken är så känslig att vi kan mäta skillnader i utbytes-hastighet hos amidväten som är omgivna av stora, respektive små sidoked-jor. De utbyteshastigheter som vi erhöll överensstämde mycket bra med de teoretiska värdena. I Paper V, undersöktes aminosyror som är problematiska att studera med CID HDX-teknik eftersom deras isotop innehåll förändras under masspektrometrisk analys. I Paper VI, undersöktes CID-fragmentens reaktivitet i gasfas. Det visade sig att vissa fragment typer var mindre reakti-va i gasfas och därför också bättre lämpade att använda för högupplösande studier av isotoputbyten i lösning. En möjlig förklaring till att de olika frag-menten är olika reaktiva i gasfas skulle kunna vara att de har olika inre ener-gi. Om ett fragment är mycket energirikt rör det sig så mycket att utbytes-komplex i gasfasen blir instabilt. Därför är energetiska fragment bättre läm-pade för högupplösta HDX-studier. Fragmentstorleken påverkade också gasfasreaktiviteten, små fragment var mer reaktiva i gasfas jämfört med stör-re fragment. I Paper VII används två fragmenterings-tekniker, ECD och CID, i kombination med HDX-tekninken. Det visade sig att isotoputbytet vid en specifik aminosyra i en peptid som inte kunde följas med CID var möjlig att följa med ECD. En förklaring är att eftersom ECD sker vid ett tryck av 10-9 mbar och CID vid 10-4 mbar är sannolikheten högre för att gasfasutbyten ska ske vid CID.

46

9. References

(1) Comisarow, M. B.; Marshall, A. G. Chem. Phys. Lett. 1974, 25, 282-283.

(2) Amster, I. J. J. Mass. Spectrom. 1996, 31, 1325-1337. (3) Marshall, A. G.; Hendrickson, C. L.; Jackson, G. S. Mass Spectrom.

Rev. 1998, 17, 1-35. (4) Thumm, M.; Hoenes, J.; Pfleiderer, G. Biochim. Biophys. Acta 1987,

2, 263-267. (5) Beardsley, R. L.; Reilly, J. P. J Proteome Res 2003, 2, 15-21. (6) Munchbach, M.; Quandroni, M.; Miotti, G.; James, P. Anal. Chem.

2000, 72, 4047-4057. (7) Geng, X.; Regnier, F. E. J Chromatogr. 1984, 296, 15-30. (8) Stacey, C. C. Rapid Commun Mass Spectrom. 1994, 8, 513-515. (9) Katta, V.; Chait, B. T. Rapid Commun. Mass Spectrom. 1991, 5,

214-217. (10) Miranker, A.; Robinson, C. V.; Aplin, R. T.; C.M., D. Science 1993,

262, 896-900. (11) Mandell, J. G.; Falick, A. M.; Komives, E. A. Proc Natl Acad Sci

1998, 95, 14705-14710. (12) Buijs, J.; Ramstrom, M.; Danfelter, M.; Larsericsdotter, H.; Hakans-

son, P.; Oscarsson, S. J. Colloid Interface Sci. 2003, 263, 441-448. (13) Haddon, W. F.; McLafferty, F. W. J. Am. Chem. Soc. 1968, 90, 4745

- 4746. (14) Laskin, J.; Futrell, J. H. Mass Spectrom Rev. 2003, 22, 158-181. (15) Anderegg, R. J.; Wagner, D. S.; Stevenson, C. L.; Borchard, R. T. J.

Am. Soc. Mass Spectrom. 1994, 5, 425-433. (16) Akashi, S.; Naito, Y.; Takio, K. Anal. Chem. 1999, 71, 4974-4980. (17) Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem.

Soc. 1998, 120, 3265-3266. (18) Zubarev, R. A.; Kruger, N. A.; Fridriksson, E. K.; Lewis, M. A.;

Horn, D. M.; Carpenter, B. K.; McLafferty, F. W. J. Am. Chem. Soc.1999, 121, 2857 - 2862.

(19) Watson, J. D.; Crick, F. H. C. Nature 1953, 171, 737-738. (20) Crick, F. H. C. Nature 1970, 227, 561-563. (21) Perutz, M. F. W. H. Freeman and Company 1991.(22) Anfinsen, C. B. Science 1973, 181, 223-230. (23) Hartl, F. U.; Hlodan, R.; Langer, T. Trends Biochem Sci. 1994, 19,

20-25.(24) Muller, C. W. Curr. Opin. Struct. Biol. 2001, 11, 26-32. (25) Harrison, S. C. Nature 1991, 353, 715-719. (26) Pawson, T.; Scott, J. D. Science 1997, 278, 2075-2080. (27) Fantl, W. J.; Johnson, D. E.; Williams, L. T. Annu. Rev. Biochem.

1993, 62, 453-481. (28) Leinhard, G. E. Science 1973, 180, 149-154. (29) Hilvert, D. Annu. Rev. Biochem. 2000, 69, 751-794.

47

(30) Mathews, C. K.; van Holde, K. E.; Ahern, K.-G. Biochemistry 3rd edn 2000, San Francisco Benjamin Cummings.

(31) Branden, C.; Tooze, J. Introduction to Protein Structure; Garland Publishing, Inc: New York, 1991.

(32) Jones, S.; Thornton, J. M. Proc. Natl Acad. Sci. USA 1996, 93.(33) Rini, J.; Schultze-Gahmen, U.; Wilson, I. A. Nature 1992, 255, 959-

965.(34) Alberts, B.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P.

Molecular Biology of the Cell 4th ed. 2002, New York: Garland Publishing.

(35) Kornberg, R. D. Trends Cell Biol. 1999, 9, M46-49. (36) Velculescu, V. E.; Zhang, L.; Zhou, W.; Vogelstein, J.; Basrai, M.

A.; D.E.Jr., B.; Hieter, P.; Vogelstein, B.; Kinzler, K. W. Cell 1997,88, 243-251.

(37) Anderson, N. L.; Anderson, N. G. Mol Cell Proteomics 2002, 1,845-867.

(38) Johanson, C. E. Ventricles and cerebrospinal fluid; J. B. Lippincott,: Philadelphia, 1995.

(39) Sjogren, M.; Blomberg, M.; Jonsson, M.; Wahlund, L. O.; Edman, A.; Lind, K.; Rosengren, L.; Blennow, K.; Wallin, A. J Neurosci Res. 2001, 66, 510-516.

(40) Mitchell, A.; Brindle, N. Int. J. Geriatr Psychiatry 2003, 18, 407-411.

(41) Sjogren, M.; Davidson, P.; Gottfries, J.; Vanderstichele, H.; Edman, A.; Vanmechelen, E.; Wallin, A.; Blennow, K. Dement. Geriatr. Cogn. Disord. 2001, 12, 257-264.

(42) Blennow, K.; Davidson, P.; Gottfries, C. G.; Ekman, R.; Heilig, M. Lancet 1996, 348, 692- 693.

(43) Seyfert, S.; Kunzmann, V.; Schwertfeger, N.; Koch, H. C.; Faul-stich, A. J Neurol. 2002, 249, 1021-1026.

(44) Wilkins, M. R.; Sanchez, J. C.; Gooley, A. A.; Appel, R. D.; HumpherySmith, I.; Hochstrasser, D. F.; Williams, K. L. Biotech.Gen. Eng. Rev 1995, 13, 19-50.

(45) Patterson, S. D.; Aebersold, R. H. Nat Genet. 2003, 33, Suppl:311-323.

(46) Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yoshida, Y.; Yoshida, T. Rapid Commun. Mass Spectrom. 1988, 2, 151.

(47) Karas, M.; Hillenkamp, F. Anal Chem. 1988, 60, 2299-2301. (48) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C.

M. Science 1989, 246, 64-71. (49) Olivares, J. A.; Nguyen, N. T.; Yonker, C. R.; Smith, D. L. Analyti-

cal Chemistry 1987, 59, 1230-1232. (50) Schmitt-Kopplin, P.; Frommberger, M. Electrophoresis 2003, 24,

3837-3867. (51) Tomer, K. B. Chem Rev. 2001, 101, 297-328. (52) Annesley, T. M. Clin Chem. 2003, 49, 1041-1044.

48

(53) Krishna, R.; Wolf, F. Protein Structure - A Practical Approach edn 2; Oxford University Press: New York, 1997.

(54) Varki, A. Glycobiology 1993, 3, 97-130. (55) Krebs, E. G.; Beavo, J. A. Annual Review of Biochemistry 1979, 48,

923-959. (56) Hunter, T. Cell 1995, 83, 1-4. (57) Carr, S. A.; Biemann, K. Methods Enzymol 1984, 106, 29-58. (58) Cooper, H. J.; Hakansson, K.; Marshall, A. G. Mass Spectrom Rev.

2004.(59) Veenstra, T. D. Adv Protein Chem. 2003, 65, 161-194. (60) Hakansson, K.; Chalmers, M. J.; Quinn, J. P.; McFarland, M. A.;

Hendrickson, C. L.; Marshall, A. G. Anal Chem. 2003, 75, 3256-3262.

(61) Rabilloud, T. Proteomics 2002, 2, 3-10. (62) Gross, M., Ed. Measuring Mass; American Society for Mass Spec-

trometry, 2002. (63) Thomson, J. J. Philosophical Magazine 1899, 48, 547-567. (64) Thomson, J. J. The London, Edinburgh, and Dublin Philosophical

Magazine and Journal of Science 1907, XLVII. (65) Aston, F. W. Nature 1920, 105.(66) Dempster, A. J. Phys. Rev. 1918, 11, 316-324. (67) Macfarlane, R. D.; Torgerson, D. F. Science 1976, 191, 920-925. (68) Hakansson, P.; Kamensky, I.; Sundqvist, B.; Fohlman, J.; Peterson,

P.; McNeal, C. J.; Macfarlane, R. D. Journal of the American Chemical Society 1982, 104, 2948-2949.

(69) Barber, M.; Bordoli, R. S.; Sedgwick, R. D.; Tyler, A. N. Nature1981, 293, 270-275.

(70) Yamashita, M.; Fenn, J. B. J. Phys. Chem. 1984, 88, 4451 - 4459. (71) Marshall, A. G.; Hendrickson, C. L.; Jackson, G. S. Encyclopaedia

of Analytical Chemistry; Wiley: Chichester, 2000. (72) Emmett, M. R.; White, F. M.; Hendrickson, C. L.; Shi, S. D.; Mar-

shall, A. G. J Am Soc Mass Spectrom. 1998, 9, 333-340. (73) Martin, S. E.; Shabanowitz, J.; Hunt, D. F.; Marto, J. A. Anal Chem.

2000, 72, 4266-4274. (74) Olefirowicz, T. M.; Ewing, A. G. Anal Chem. 1990, 62, 1872-1876. (75) Bergquist, J.; Tarkowski, A.; Ekman, R.; Ewing, A. Proc Natl Acad

Sci U S A. 1994, 91, 12912-12916. (76) Mann, M.; Wilm, M. Anal Chem. 1994, 66, 4390-4399. (77) Mann, M.; Hojrup, P.; Roepstorff, P. Biol Mass Spectrom. 1993, 22,

338-345. (78) Palmblad, M.; Bergquist, J. In Emerging technologies in protein

genomic material analysis; Marko-Varga, G., Oroszlan, P., Eds.; El-sevier Science B.V.: Amsterdam, 2003; Vol. J. of Chromatography Library, pp 199-240.

(79) Klose, J. Humangenetik 1975, 26, 231-243. (80) O'Farrell, P. H. J. Biol. Chem. 1975, 250, 4007-4021.

49

(81) Bae, S. H.; Harris, A. G.; Hains, P. G.; Chen, H.; Garfin, D. E.; Ha-zell, S. L.; Paik, Y. K.; Walsh, B. J.; Cordwell, S. J. Proteomics2003, 3, 569-579.

(82) Candiano, G.; Musante, L.; Bruschi, M.; Ghiggeri, G. M.; Herbert, B.; Antonucci, F.; Righetti, P. G. Electrophoresis 2002, 23, 292-297.

(83) Unlu, M.; Morgan, M. E.; Minden, J. S. Electrophoresis 1997, 11,2071-2077.

(84) Gygi, S. P.; Rist, B.; Gerber, S. A.; Turecek, F.; Gelb, M. H.; Aeber-sold, R. Nat. Biotechnol. 1999, 17, 994-999.

(85) Han, D. K.; Eng, J.; Zhou, H.; Aebersold, R. Nat. Biotechnol. 2001,19, 946-951.

(86) Zhang, R.; Regnier, F. E. J Proteome Res 2002, 1, 139-147. (87) Cagney, G.; Emili, A. Nat Biotechnol 2002, 20, 163-170. (88) Greenstein, J. P. J Organic Chem 1938, 2, 480 - 483. (89) Hale, J. E.; Butler, J. P.; Knierman, M. D.; Becker, G. W. Anal Bio-

chem. 2000, 287, 110-117. (90) Mirgorodskaya, O. A.; Kozmin, Y. P.; Titov, M. I.; Korner, R.;

Sonksen, C. P.; Roepstorff, P. Rapid Commun Mass Spectrom. 2000,14, 1226-1232.

(91) Yao, X.; Freas, A.; Ramirez, J.; Demirev, P. A.; Fenselau, C. AnalChem. 2001, 73, 2836-2842.

(92) Oda, Y.; Huang, K.; Cross, F. R.; Cowburn, D.; Chait, B. T. ProcNatl Acad Sci U S A 1999, 96, 6591-6596.

(93) Perkins, S. J. Biochem. Sci. 1988, 254, 313-327. (94) Wagner, G.; Wuthrich, K. J. Mol. Biol. 1982, 160, 343-361. (95) Medek, A.; Hajduk, P. J.; Mack, J.; Fesik, S. W. J. Am. Chem. Soc.

2000, 122, 1241-1242. (96) Pellecchia, M. J. Biomol. NMR. 2002, 22, 165-173. (97) Roder, H.; Elove, G. A.; Englander, S. W. Nature 1988, 335, 700-

704.(98) Johnson, R. S. J. Am. Soc. Mass Spectrom. 1996, 7, 515-521. (99) Buijs, J.; Costa Vera, C.; Ayala, E.; Steensma, E.; Hakansson, P.;

Oscarsson, S. Anal Chem 1999, 71, 3219-3225. (100) Hvidt, A.; Linderstrom-Lang, K. Biochim Biophys Acta. 1954, 14,

574-575. (101) Hughes, C. A.; Mandell, J. G.; Anand, G. S.; Stock, A. M.; Ko-

mives, E. A. J Mol Biol. 2001, 307, 967-976. (102) Tuma, R.; Coward, L. U.; Kirk, M. C.; Barnes, S.; Prevelige, P. E. J.

J Mol Biol. 2001, 3, 389-396. (103) Robinson, C. V.; Gross, M.; Eyles, S. J.; Ewbank, J. J.; Mayhew,

M.; Hartl, F. U.; Dobson, C. M.; Radford, S. E. Nature 1994, 372,646-651.

(104) Smith, D. L.; Deng, Y.; Zhang, Z. J Mass Spectrom 1997, 32, 135-146.

(105) Bai, Y.; Milne, J. S.; Mayne, L.; Englander, S. W. Proteins 1993,17, 75-86.

50

(106) Connelly, G. P.; Bai, Y.; Jeng, M. F.; Englander, S. W. Proteins1993, 17, 87-92.

(107) Molday, R. S.; Englander, S. W.; Kallen, R. G. Biochemistry 1972,11, 150-158.

(108) Englander, S. W.; Sosnick, T. R.; Englander, J. J.; Mayne, L. CurrOpin Struct Biol 1996, 6, 18-23.

(109) Li, R.; Woodward, C. J. Am. Soc. Mass Spectrom. 1999, 8, 672-674. (110) Kim, M. Y.; Maier, C. S.; Reed, D. J.; Ho, P. S.; Deinzer, M. L.

Biochemistry 2001, 40, 14413-14421. (111) Demmers, J. A.; van Duijn, E.; Haverkamp, J.; Greathouse, D. V.;

Koeppe, R. E. n.; Heck, A. J.; Killian, J. A. J Biol Chem 2001, 276,34501-34508.

(112) Tito, P.; Nettleton, E. J.; Robinson, C. V. J. Mol. Biol. 2000, 303,267-278.

(113) Lias, S. G.; Lieberman, J. F.; Levin, R. D. J. Phys. Chem. Ref. Data1984, 13, 695-808.

(114) Ausloos, P.; Lias, S. G. J Am. Chem. Soc. 1981, 103, 3641-3645. (115) Cheng, X.; Fenselau, C. Int. J. Mass Spectrom. Ion Proc. 1992, 122,

109-119. (116) McLafferty, F. W.; Guan, Z.; Haupts, U.; Wood, T. D.; Kelleher, N.

L. J. Am. Chem. Soc. 1998, 120, 4732-4740. (117) Kaltashov, I. A.; Doroshenko, V. M.; Cotter, R. J. Proteins 1997, 28,

53-58.(118) Kaltashov, I. A.; Eyles, S. J. Mass Spectrom Rev. 2002, 21, 37-71. (119) Wyttenbach, T.; T., B. M. J. Am. Soc. Mass Spectrom. 1999, 10, 9-

14.(120) Campbell, S.; Rodgers, M. T.; Marzluff, E. M.; Beauchamp, J. L. J.

Am. Chem. Soc. 1995, 117, 12840-12854. (121) Hofstadler, S. A.; Sannes-Lowery, K. A.; Griffey, R. H. Rapid

Commun. Mass Spectrom. 1999, 13, 1971-1979. (122) Axelsson, J.; Hakansson, K.; Palmblad, M.; Hakansson, P. Rapid

Commun. Mass Spectrom. 1999, 13, 1550. (123) Guan, S.; Marshall, A. G. Int. J. Mass Spectrom. 1995, 146/147,

261-296. (124) Caravatti, P.; Allemann, M. Org. Mass Spectrom. 1991, 26, 514-

518.(125) Comisarow, M. B. Adv. Mass Spectrom. 1978, 7, 1042-1046. (126) Wang, M.; Marshall, A. G. Int. J. Mass Spectrom. 1990, 100, 324-

346.(127) Comisarow, M. B. Int. J. Mass Spectrom. Ion Phys. 1981, 37, 251-

257.(128) Cooley, J. W.; Tukey, J. W. Math. Comp. 1965, 19, 297-301. (129) Fenyo, D.; Qin, J.; Chait, B. T. Electrophoresis 1998, 19, 998-1005. (130) Guan, S.; Marshall, A. G. Rapid Commun. Mass Spectrom. 1996, 10,

1819-1823. (131) Marshall, A. G. Int. J. Mass Spectrom. 2000, 200, 331-356.

51

(132) Marshall, A. G.; Hendrickson, C. L. Int. J. Mass Spectrom. 2002,215, 59-75.

(133) Reinhold, V. N.; Reinhold, B. B.; Costello, C. E. Anal Chem. 1995,67, 1772-1784.

(134) Cole, R. B. J Mass Spectrom 2000, 35, 763-772. (135) Fernandez de la Mor, J. Anal. Chim. Acta 2000, 406, 93-104. (136) Kebarle, P. J Mass Spectrom 2000, 35, 804-817. (137) Iribarne, J. V.; Dziedzic, P. J.; Thomson, B. A. Int. J. Mass Spec-

trom. Ion Physics 1983, 50, 331-347. (138) Cech, N. B.; Krone, J. R.; Enke, C. G. Anal Chem. 2001, 73, 208-

213.(139) Aston, F. W. Proc. Cambridge Philos. Soc. 1919, 19, 317. (140) Yost, R. A.; Enke, C. G. J. Am. Chem. Soc. 1978, 100, 2274 - 2275. (141) McIver, R. T.; Hunter, R. L.; Bowers, W. D. Int. J. Mass Spectrom.

Ion Processes 1985, 64, 67-77. (142) Biemann, K.; Scoble, H. A. Science. 1987, 237, 992-998. (143) Dongre, A. R.; Jones, J. L.; Somogyi, A.; Wysocki, V. H. J. Am.

Chem. Soc. 1996, 118, 8365 -8374. (144) Kenney, P. T.; Nomoto, K.; Orlando, R. Rapid Commun. Mass Spec-

trom. 1992, 6, 95-97. (145) Roepstorff, P.; Fohlman, J. Biomed Mass Spectrom 1984, 11, 601. (146) Yalcin, T.; Khouw, C.; Csizmadia, I. G.; Peterson, M. R.; Harrison,

A. G. J. Am. Soc. Mass Spectrom. 1995, 6, 1165-1174. (147) Yalcin, T.; Csizmadia, I. G.; Peterson, M. R.; Harrison, A. G. J. Am.

Soc. Mass Spectrom. 1996, 7, 233-242. (148) Klassen, J. S.; Kebarle, P. J. Am. Chem. Soc. 1997, 119, 6552 -

6563. (149) Baba, T.; Hashimoto, Y.; Hasegawa, H.; Hirabayashi, A.; Waki, I.

Anal Chem. 2004, 76, 4263-4266. (150) Zubarev, R. A.; Horn, D. M.; Fridriksson, E. K.; Kelleher, N. L.;

Kruger, N. A.; Lewis, M. A.; Carpenter, B. K.; McLafferty, F. W. Anal. Chem. 2000, 72, 563-573.

(151) Ramstrom, M.; Hagman, C.; Tsybin, Y. O.; Markides, K. E.; Hakansson, P.; Salehi, A.; Lundquist, I.; Hakanson, R.; Bergquist, J. Eur J Biochem. 2003, 270, 3146-3152.

(152) Zubarev, R. A. Mass Spectrom Rev. 2003, 22, 57-77. (153) Leymarie, N.; Costello, C. E.; O'Connor, P. B. J Am Chem Soc.

2003, 125, 8949-8958. (154) Turecek, F. J Am Chem Soc. 2003, 125, 5954-5963. (155) Breuker, K.; Oh, H.; Lin, C.; Carpenter, B. K.; McLafferty, F. W.

Proc Natl Acad Sci U S A 2004, 101, 14011-14016. (156) Hakansson, K.; Cooper, H. J.; Emmett, M. R.; Costello, C.; Mar-

shall, A. G.; Nilsson, C. L. Anal Chem 2001, 73, 4530-4536. (157) Stensballe, A.; Jensen, O. N.; Olsen, J. V.; Haselmann, K. F.;

Zubarev, R. A. Rapid Commun Mass Spectrom. 2000, 14, 1793-1800.

52

(158) Ramstrom, M.; Palmblad, M.; Markides, K. E.; Hakansson, P.; Ber-gquist, J. Proteomics 2003, 3, 184-190.

(159) Ramstrom, M.; Ivonin, I.; Johansson, A.; Askmark, H.; Markides, K. E.; Zubarev, R.; Hakansson, P.; Aquilonius, S. M.; Bergquist, J. Proteomics 2004, 4, 4010-4018.

(160) Wu, S.-L.; Choudhary, G.; Ramstrom, M.; Bergquist, J.; Hancock, W. S. J. Proteome Res. 2003, 2, 383-393.

(161) Nilsson, S.; Ramstrom, M.; Palmblad, M.; Axelsson, O.; Bergquist, J. J Proteome Res. 2004, 3, 884-889.

(162) Bergquist, J.; Palmblad, M.; Wetterhall, M.; Hakansson, P.; Markides, K. E. Mass Spectrom Rev. 2002, 21, 2-15.

(163) Guina, T.; Purvine, S. O.; Yi, E. C.; Eng, J.; Goodlett, D. R.; Aeber-sold, R.; Miller, S. I. Proc Natl Acad Sci U S A. 2003, 100, 2771-2776.

(164) Freitas, M. A.; Hendrickson, C. L.; Emmett, M. R.; Marshall, A. G.J. Am. Soc. Mass Spectrom. 1998, 9, 1012-1019.

(165) Freitas, M. A.; Marshall, A. G. Int. J. Mass Spectrom. 1999,182/183, 221-231.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 6

Editor: The Dean of the Faculty of Science and Technology

A doctoral dissertation from the Faculty of Science and Technology, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology. (Prior to January, 2005, the series was published under the title "Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology".)

Distribution: publications.uu.seurn:nbn:se:uu:diva-4761

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2005