Isolation and Characterization of Water-Soluble ...

Duquesne UniversityDuquesne Scholarship Collection

Electronic Theses and Dissertations

2005

Isolation and Characterization of Water-SolubleFluorescent Species from Human SerumAshraf Hamid Elamin

Follow this and additional works at: https://dsc.duq.edu/etd

This Immediate Access is brought to you for free and open access by Duquesne Scholarship Collection. It has been accepted for inclusion in ElectronicTheses and Dissertations by an authorized administrator of Duquesne Scholarship Collection. For more information, please [email protected].

Recommended CitationElamin, A. (2005). Isolation and Characterization of Water-Soluble Fluorescent Species from Human Serum (Doctoral dissertation,Duquesne University). Retrieved from https://dsc.duq.edu/etd/520

https://dsc.duq.edu?utm_source=dsc.duq.edu%2Fetd%2F520&utm_medium=PDF&utm_campaign=PDFCoverPages

https://dsc.duq.edu/etd?utm_source=dsc.duq.edu%2Fetd%2F520&utm_medium=PDF&utm_campaign=PDFCoverPages

https://dsc.duq.edu/etd?utm_source=dsc.duq.edu%2Fetd%2F520&utm_medium=PDF&utm_campaign=PDFCoverPages

https://dsc.duq.edu/etd/520?utm_source=dsc.duq.edu%2Fetd%2F520&utm_medium=PDF&utm_campaign=PDFCoverPages

mailto:[email protected]

Isolation and Characterization of Water-Soluble Fluorescent Species from Human Serum

A Dissertation Presented to the Department of Chemistry and Biochemistry at Duquesne University

In Partial Fulfillment of The Requirements for the Degree of Doctor of Philosophy

By

Ashraf H Elamin April 11, 2005

1

Abstract

Lipid peroxidation (LP) proceeds by a free radical chain reaction mechanism in

which molecular oxygen is incorporated into polyunsaturated fatty acids (PUFA)

to yield lipid hydroperoxides (LOOH). The ultimate end-products of LP are the

production of α, β unsaturated aldehydes, such as acrolein, 4-hydroxy-2-nonenal

(HNE) and malondialdehyde (MDA). These aldehydes are very reactive and

exhibit facile reactivity with various biomolecules, particularly proteins, resulting

in the formation of fluorescent adducts that could be useful biological marker in

diseases. These fluorescent adducts typically emit light between 400 and 600

nm when excited at wavelengths ranging from 300-400 nm. Lysine residues are

the major amino-acids that are predicated to react with aldehydes to yield

fluorescent adducts. LP has been implicated in several diseases such as

atherosclerosis, Parkinson, Alzheimer, and diabetes.

Water-soluble fluorescent species (WSF) are fluorescent species that exhibited

similar fluorescence characteristics for excitation and emission to those from

aldehyde-modified proteins and amino-acids analogues. WSF showed also

direct correlation with age, so it suggested that its components could serve as a

useful biomarker for in vivo LP.

We determined that WSF was composed of two main components, a high

molecular weight protein component and a low molecular weight non-protein

component. The non-protein component was unreactive towards proteins. We

were able to identify four major protein species in the protein component that

contributed to the fluorescence. These protein species were HSA, HST, Ig-G

and carbonic anhydrase, with HSA and HST being the major species contributing

to the overall fluorescence in the protein component. The fluorophore was

determined to be covalently bound to the protein component.

Using CNBr digestion on isolated HSA and HST, we were able to determine

multiple fluorophore binding sites within both proteins. We determined by MALDI

TOF/MS that there were indeed MW increases in HSA and HST after aldehyde

2

modification due to the fluorophores formation. In vitro studies on HSA

determined a change in the pI due to aldehyde modifications. In vitro functional

studies on HST suggested that there might be a loss of one the HST Fe binding

sites due to aldehyde modification.

All rights reserved © 2005

Ashraf H. Elamin

3

Acknowledgements

First I would like to thank Dr. David Seybert my dissertation advisor, for his help,

advice, guidance, and all that he has taught me during my graduate years. He

taught me from his valuable experiences how to be a good biochemist and even

more a good person, which I really appreciate and won’t be forgotten. I want to

acknowledge Dr. William Brown, Dr. Charles Dameron, Dr. Mitchell Johnson, and

Dr. David Wright for their guidance and I really learned a lot from their helpful

constructive comments and suggestions about my project and how to improve

my knowledge and skills preparing me for a productive future career. The

financial support of the Duquesne University Department of Chemistry and

Biochemistry for all my graduate years is also acknowledged. I want to thank

Anne Dunlap and Amy Pollack for their help, friendships, support and learning

from each other. I want to thank fellow graduate students throughout my

graduate years for their friendship, cooperation and support.

Special acknowledgment goes to my wife, Ayat, for her help, love and support

through the past years. Last acknowledgments go to my family (Hamid, Laila,

Sanaa, Wafaa, Naglaa and Sarah) and my in-laws for their continuous support to

get to where I am now, for without their encouragements I would not have make

this far.

4

Table of Contents 1.0 Background ......................................................................................................... 15

1.1 Lipid Peroxidation (LP) .................................................................................... 15 1.1.1 Significance of LP End-Products........................................................ 17 1.1.2 Role of reactive aldehydes in diseases................................................ 181.2 Biomolecules Modification by Aldehydes........................................................ 19 1.2.1 Protein Modification ........................................................................... 19 1.2.2 Aminophospholipid Modification....................................................... 23 1.2.3 DNA Modification. ............................................................................. 231.3 Water-soluble Fluorescent Species ................................................................... 24 1.4 Human Serum Albumin (HSA) ........................................................................ 27 1.4.1 Function .............................................................................................. 27 1.4.2 Structure.............................................................................................. 27 1.4.3 In vitro HSA Modification by LP End-Products ................................ 301.5 Human Serum Transferrin (HST) ..................................................................... 31 1.5.1 Function. ............................................................................................. 31 1.5.2 Structure.............................................................................................. 32 1.5.3 Iron Binding to HST ........................................................................... 32 1.5.4 Release of Iron. ................................................................................... 33 1.5.4.1 Release of Iron from N-Site. ........................................................ 33 1.5.4.2 Release of Iron from C-Site ......................................................... 341.6 Statement of Research....................................................................................... 35

2.0 Experimental Methods and Instrumentation................................................... 36 2.1 Extraction and characterizations of WSF from human serum.......................... 37 2.1.1 Extraction of WSF .............................................................................. 37 2.1.2 Attempted Chemical Reduction of WSF Extract................................ 372.2 Separation and Characterization of WSF’s Protein Component....................... 38 2.2.1 Gel Filtration Chromatography........................................................... 38 2.2.2 SDS-PAGE Gels ................................................................................. 38 2.2.3 Denaturation Experiments .................................................................. 38 2.2.4 Protein Concentration Determinations................................................ 39

2.2.5 Isolation of 67 kDa Protein Present in Peak B from the Sephadex G-75 Column.......................................................................................................... 392.2.6 Removal of HSA from other protein species present in peak A from the Sephadex G-75 Column.......................................................................... 40

2.2.7 Western Blot for Confirmation of 80 kDa Protein.............................. 40 2.2.8 Isolation of HST.................................................................................. 41 2.2.9 HSA and HST Aldehyde Modification............................................... 412.3 MALDI TOF/MS.............................................................................................. 42

2.3.1 Proteins Identifications and Confirmations from SDS-PAGE Gel Protein Plugs using MALDI-TOF/MS.......................................................... 42 2.3.2 In vitro Structural Studies on Aldehyde Modified HSA and HST using MALDI TOF/MS.......................................................................................... 43

2.4 Characterization Studies on HSA ..................................................................... 43 2.4.1 CNBr Digestion of Isolated HSA ....................................................... 43

5

2.4.2 Separation of Peptide Fragments using Gel Filtration Chromatography 45

2.4.3 Separation of Peptide Fragments from CNBr-digested of Isolated HSA using HPLC................................................................................................... 45

2.4.4 Isoelectric Focusing (IEF) Experiments on HSA ............................... 472.5 Structural studies on HST ................................................................................. 48 2.5.1 CNBr Digestion of Isolated HST and Peptide Fragments Separation 48

2.5.2 Reduction and Carboxyamidomethylation of Peak A of CNBr-digested Isolated HST from P-10 Column .................................................................. 502.5.3 Separation of CNBr-digested Isolated HST Peptide Fragments using HPLC ............................................................................................................ 50

2.6 In vitro functional Studies on HST ................................................................... 51 2.6.1 Protein Fluorescence Experiments...................................................... 51

2.6.2 Iron Uptake by Apo and Aldehyde Modified HST (Titration Experiments) ................................................................................................. 51 2.6.3 Iron Uptake by Apo and Aldehyde Modified HST (Saturation Experiments) ................................................................................................. 52

2.6.4..... Iron Kinetics of Release from Normal and Modified Diferric HST at pH 7.4.............................................................................................. 522.6.5 Iron Kinetics of Release from Normal and Aldehyde Modified Diferric HST at pH 5.6 ............................................................................................... 52

2.7 Characterizations of peak C of WSF (the non-protein component). ................ 53 2.7.1 Concentrating Peak C ......................................................................... 53

2.7.2 MW Determination of Peak C using Electrospray Ionization (ESI) MS 53

2.7.3 Peak C reactivity ................................................................................. 532.8 Serum Incubation Experiments......................................................................... 54 2.9 Human subjects’ study...................................................................................... 54

3.0 Results .................................................................................................................. 55 3.1 Components of WSF Extracted from Human Serum ....................................... 55 3.1.1 Isolation of WSF Extract from Human Serum ................................... 55 3.1.2 Reduction of WSF Extract with NaBH4............................................. 60

3.1.3 Binding that exists between fluorophore and protein component of WSF .............................................................................................................. 61

3.2 Characterization of the Protein Component of WSF ........................................ 63 3.2.1 Purification of the Major Fluorescent Species in Peak B from the Sephadex G-75 column................................................................................. 63

3.2.2 Protein identification of 67 kDa protein using MALDI TOF/MS ...... 653.2.3 Separation of proteins in peak A from the Sephadex G-75 column from HSA...................................................................................................... 68

3.2.4 Protein Identification and isolation of the 80 kDa protein from peak A* from the affinity column ....................................................................... 693.2.5 Protein Identifications of the 2 protein from DEAE column (peak A1). 75

3.3 Investigation of the nature of the non-protein WSF component (peak C from the Sephadex G-75 Column)......................................................................................... 79

6

3.3.1 Peak C Purification and Concentration............................................... 79 3.3.2 Molecular weight determination of peak C using ESI-MS................. 80 3.3.3 Peak C Reactivity................................................................................ 813.4 Characterization of Isolated Proteins ................................................................ 82 3.4.1 HSA..................................................................................................... 82

3.4.1.1 CNBr Digestion of Isolated HSA and Peptide Fragments Separation ................................................................................................. 82

3.4.1.2 In vitro characterization studies by MALDI TOF/MS ............... 87 3.4.1.3 In vitro Structural Characterizations using Isoelectric Focusing (IEF) Experiments..................................................................................... 91

3.4.2 HST............................................................................................................... 93 3.4.2.1 CNBr Digestion of Isolated HST Peptide Fragments Separation 93 3.4.2.2 In vitro characterizations studies using MALDI TOF/MS .......... 983.5 In Vitro Fluorescent Development Studies on HST. ...................................... 102 3.6 In Vitro Functional Studies on HST................................................................ 106 3.6.1 Iron Uptake ....................................................................................... 106 3.6.2 Fe Kinetics of Release from Diferric HST ....................................... 109 3.6.2.1 Fe Kinetics of Release from Commercial Diferric HST........... 109

3.6.2.2 Effect of Aldehyde Modification on Fe Kinetics of Release from Commercial Diferric HST........................................................................... 112

3.6.2.3 Saturation Experiment .............................................................. 116 3.6.2.4 Titration Experiment................................................................. 1243.7 Serum Incubation Experiments....................................................................... 131 3.8 Subject Study .................................................................................................. 134

4.0 Discussion........................................................................................................... 136 4.1 Separation and Characterization of WSF Extract Components...................... 136 4.1.1 Characterization of Peak C ............................................................... 137 4.1.2 Reduction Experiments on WSF....................................................... 138 4.1.3 Covalent Binding of Fluorophore ..................................................... 1394.2 Characterization of the Protein Component of WSF ...................................... 140 4.2.1 Isolation of Major Proteins Present in the WSF Protein Component1404.3 Characterizations Studies on HSA and HST .................................................. 144 4.3.1 Characterization of Isolated HSA ..................................................... 144 4.3.1.1 CNBr Digestion of Isolated HSA ............................................. 144 4.3.1.2 Isoelectric Focusing Experiments ............................................. 145 4.3.2 Characterizations of Isolated HST.................................................... 146 4.3.2.1 CNBr Digestion of Isolated HST.............................................. 1464.4 In Vitro MALDI Experiments on HSA and HST ........................................... 147 4.5 In Vitro Fluorescence Development Studies on HST..................................... 148 4.6 In Vitro Functional Studies on HST................................................................ 149 4.6.1 Iron Uptake by Apo HST.................................................................. 149 4.6.2 Fe Kinetics of Release from HST ..................................................... 150 4.6.2.1 Saturation and Titration Experiments ....................................... 1504.7 Serum Incubation Experiments....................................................................... 153 4.8 Subject Study .................................................................................................. 154 4.9 Is WSF a Reflection of In Vivo LP?................................................................ 155

7

5.0 References.......................................................................................................... 156

8

Table of Figures Fig 1 Model for the mechanism of initiation of LP [5].................................................... 16 Fig 2 Structures of some example of LP derived aldehydes [8] ....................................... 18 Fig 3 Scheme of proposed formation of Nα-acetyllysine-HNE fluorophore [24]............ 20 Fig 4 Fluorescence spectra of Nα-acetyllysine-HNE fluorophore showing excitation

maxima at 360 nm and emission maxima at 430 nm [24] ........................................ 21 Fig 5 Proposed structures of adducts formed from different aldehydes [8] ..................... 22 Fig 6 Structure of aminophospholipid adduct [27] ........................................................... 23 Fig 7 Structure of DNA adducts formed as a result of acrolein incubation [30] .............. 24 Fig 8 Fluorescence excitation and emission spectra of WSF pigments found in protein

fraction of mouse (A) and human (B) sera [31]........................................................ 25 Fig 9 Structures of HSA complexed with five different fatty acids. The protein secondary

structure is shown schematically. The A and B sub-domains are depicted in dark and light shades, respectively. Bound fatty acids are shown in space filling representations [35]................................................................................................... 29

Fig 10 Chemical structure of quanternary pyridinium salts [44]...................................... 30 Fig 11 CNBr fragments of HSA separated by tricine gel electrophoresis with the

assignment of MW to each respective band [66]...................................................... 44 Fig 12 Separation of CNBr fragments of HSA using reversed-phase C18 column HPLC.

Conditions are described in 2.3.3 [68] ...................................................................... 46 Fig 13 Separation of CNBr fragments of HST on Sephadex G-100. Elution profile

monitored at absorbance 280 nm [69] ...................................................................... 48 Fig 14 Separation of reduced and carboxyamidomethylated CN-A peptides on Sephadex

G-100. The elution profile monitored at absorbance 280 nm from G-100 column [69]............................................................................................................................ 49

Fig 15 Separation of reduced and carboxyamidomethylated CN-B peptides on Sephadex G-50. Elution profile monitored at absorbance 280 nm [69]................................... 49

Fig 16 Elution profile of WSF extract passed through the Sephadex G-75 column, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 nm and em 460 nm. Arrow indicates the void volume of the column........................................ 55

Fig 17 Top: Excitation spectrum of WSF extract from the Sephadex G-75 column. Bottom: Emission spectrum of peak WSF extract .................................................... 56

Fig 18 Top: Excitation spectrum of peak A from the Sephadex G-75 column. Bottom: Emission spectrum of peak A ................................................................................... 57

Fig 19 Top: Excitation spectrum of peak B from the Sephadex G-75 column. Bottom: Emission spectrum of peak B. .................................................................................. 58

Fig 20 Top: Excitation spectrum of peak C from the Sephadex G-75 column. Bottom: Emission spectrum of peak C ................................................................................... 59

Fig 21 The elution profile of control and NaBH4-reduced WSF passed through the Sephadex G-75 column monitored by fluorescence intensity at ex 350 nm and em 460 nm. Arrow indicates the void volume of column ............................................. 60

Fig 23 SDS-PAGE image showing WSF extract before passage through the Sephadex G-75, and peaks A and B eluted from the Sephadex G-75 column .............................. 63

Fig 24 The fluorescence elution profile of concentrate peak B through the Sephadex G-75 monitored at ex 350 nm and em 460 nm. Arrow shows the void volume of the column....................................................................................................................... 64

9

Fig 25 Emission spectrum of the isolated pure HSA. Insert: SDS-PAGE image of the pure isolated HAS..................................................................................................... 65

Fig 26 MALDI MS spectrum of trypsin digested 67 kDa band analyzed by 4700 Proteomics Analyzer TOF/MS as described in the experimental section................. 66

Fig 27 Peptide fragments mapping fingerprinting database search hit for the 67 kDa protein band .............................................................................................................. 67

Fig 28 The elution profile of concentrate peak A proteins passed through the Blue Sepharose affinity column with 1 M NaCl step gradient, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 and em 460 nm. Arrow indicates the beginning of the gradient .......................................................................................... 68

Fig 29 SDS image peak A before and after passage through the affinity column. Only the run-through peak is shown........................................................................................ 69

Fig 30 Western blot experiment confirming the presence of HST in peak A* from the affinity column. Standard (Std) contained HST ...................................................... 70

Fig 31 The elution profile of peak A* passed through the DEAE Sepharose column with 0-100 mM NaCl gradient, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 nm and em 460 nm. Arrow indicates the beginning of the gradient ..................................................................................................................... 71

Fig 32 Emission spectrum of the pure isolated HST. Insert: SDS-PAGE image of the pure isolated HST ..................................................................................................... 72

Fig 33 MALDI MS spectrum of 80 kDa band analyzed by 4700 Proteomics Analyzer TOF/MS as described in the experimental section ................................................... 73


Fig 35 SDS image of pass-through from the DEAE column, showing the 30 and 51 kDa proteins...................................................................................................................... 75

Fig 36 MALDI spectrum of trypsin-digested 51 kDa protein band analyzed by 4700 Proteomics Analyzer TOF/MS as described in the experimental section................. 76


Fig 38 MALDI MS spectrum of trypsin-digested 30 kDa protein band analyzed by 4700 Proteomics Analyzer TOF/MS as described in the experimental section................. 77


Fig 40 The fluorescence elution of WSF extract passed through the P-2 column, monitored at ex 350 nm and em 460 nm, as described in the experimental section. Arrow shows the void volume of column................................................................. 79

Fig 41 ESI-MS spectrum of peak C.................................................................................. 80 Fig 42 Incubated mixture of peak C and HSA and control HSA passed through the

Sephadex G-75 column. The fluorescence elution profile monitored at ex 350 nm and em 460 nm as described in the experimental section. Arrow shows the void volume of the column ............................................................................................... 81

Fig 43 SDS-PAGE image of CNBr digested HSA peptide fragments ............................ 82 Fig 44 The elution profile of CNBr fragments of isolated HSA passed through the

Ulrogel AcA gel filtration column, monitored by absorbance at 280 and fluorescence intensity at ex 350 and em 460 nm. Arrow shows the void volume of column....... 83

10

Fig 45 SDS-PAGE image of CNBr digested HSA peptide fragment and the 20 kDa peptide band of peak 1 from the Ulrogel AcA column............................................. 84

Fig 46 Elution of CNBr fragments of isolated HSA from reverse phase HPLC. Run conditions were described in the experimental section............................................. 85

Fig 47 The fluorescence and absorbance elution profiles of CNBr fragments of isolated HSA from reverse-phase HPLC. Run conditions were described in the experimental section ....................................................................................................................... 86

Fig 48 MALDI MS spectrum of commercial HSA .......................................................... 88 Fig 49 MALDI MS spectrum of 4-ONE modified HSA .................................................. 89 Fig 50 MALDI MS spectrum of hex modified HSA ........................................................ 90 Fig 51 IEF image of control HSA sample on a IPG strip of pH range 5-8...................... 92 Fig 52 IEF image of hex-modified HSA on a IPG strip with pH range5-8 .................. 92 Fig 53 IEF image of 4-ONE-modified HSA on a IPG strip with pH range 5-8 ............... 93 Fig 54 The elution profile of CNBr-digested isolated HST passed through the P-10

column, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 nm and em 460 nm. Arrow shows the void volume of the gel ...................................... 94

Fig 55 SDS-PAGE image of peak A of CNBr digested HST from the P-10 column....... 95 Fig 56 The elution profile of reduced and carboxyamidomethylated peak A of CNBr-

digested isolated HST through the P-10 column, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 nm and em 460 nm. Arrow shows the void volume of the column ............................................................................................... 96

Fig 57 The elution profile of CNBr fragments of isolated HST from reverse-phase HPLC. Run conditions were described in the experimental section..................................... 97

Fig 58 MALDI MS spectrum of commercial apo-HST.................................................... 99 Fig 59 MALDI MS spectrum of 4-ONE modified apo-HST.......................................... 100 Fig 60 MALDI MS spectrum of hex-modified apo-HST ............................................... 101 Fig 61 Fluorescence development time points plot of hex-modified apo and diferric HST

................................................................................................................................. 103 Fig 62 Fluorescence development time point plot of 4-ONE modified apo and diferric

HST......................................................................................................................... 104 Fig 63 SDS-PAGE image of hex and 4-ONE modified apo-HST versus unmodified apo-

HST......................................................................................................................... 105 Fig 64 Plot of Fe uptake titration experiment of control and 4-ONE-modified apo-HST

samples as described in the experimental section. Insert: Equivalence point determination using Origin software ...................................................................... 107

Fig 65 Plot of Fe uptake titration experiments of control and hex modified apo-HST sample as described in the experimental section. Insert: Equivalence point determination using Origin software ...................................................................... 108

Fig 66 Origin fit of kinetic of diferric-HST at pH 7.4 using Tiron as chelator and absorbance monitored at 480 nm ............................................................................ 110

Fig 67 Origin fit of kinetics of diferric-HST at pH 5.6 using EDTA as chelator and absorbance monitored at 470 nm ............................................................................ 111

Fig 68 Origin fit of Fe kinetic of release of control diferric-HST using Tiron as chelator and absorbance monitored at 470 nm at pH 7.4...................................................... 112

Fig 69 Origin fit of kinetic of hexenal modified diferric-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4 ............................................................ 113

11

Fig 70 Origin fit of kinetic of Fe release from control diferric HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6........................................ 114

Fig 71 Origin fit of kinetic of Fe release from hex-modified diferric HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6........................................ 115

Fig 72 Origin fit of Fe kinetics of release of control saturated apo-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4........................................ 117

Fig 73 Origin fit of Fe kinetics of release of hex modified saturated apo-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4 ................................... 118

Fig 74 Origin fit of Fe kinetic of control saturated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6 ............................................................ 119

Fig 75 Origin fit of Fe kinetics of release of hex-modified saturated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6........................ 120

Fig 76 Origin fit of Fe kinetics of release of 4-ONE modified saturated apo-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4 ......................... 121

Fig 77 Origin fit of Fe kinetic of 4-ONE modified saturated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6........................................ 123

Fig 78 Origin fit of Fe kinetics of release of control titrated apo-HST using Trion as chelator and absorbance monitored at 480 nm at pH 7.4........................................ 125

Fig 79 Origin fit of Fe kinetics of release of hex-modified titrated apo-HST run using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4 ......................... 126

Fig 80 Origin fit of Fe kinetics of release of control titrated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6........................................ 127

Fig 81 Origin fit of Fe kinetics of release of hex-modified titrated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6 ................................... 128

Fig 82 Origin fit of Fe kinetics of release run of 4-ONE modified titrated apo-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4 ......................... 129

Fig 83 Origin fit of Fe kinetics of release of 4-ONE modified titrated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6........................ 130

Fig 84 Fluorescence elution profile of hex modified WSF extract passed through the Sephadex G-75 column, monitored by fluorescence intensity at ex 350 nm and em 460 nm. Arrow shown the void volume of the column ......................................... 131

Fig 85 SDS-PAGE image of fractions of hex modified peaks A’ and B’ and control peak A, eluted from the Sephadex G-75 column ............................................................ 132

Fig 86 The elution profile of 4-ONE modified extract passed through the Sephadex G-75 column, monitored by fluorescence intensity at ex 350 nm and em 460 nm. Arrow shown void volume of the column.......................................................................... 133

Fig 87 Representative example of an Origin fit of one of the subjects Sephadex G-75 column fluorescence elution profile where peak areas of peaks A and B were calculated ................................................................................................................ 134

Fig 88 Comparison of the six subjects based on peak areas percentages of peaks A and B................................................................................................................................. 135

12

Table of Tables Table 1 MW and sequence positions of CNBr fragments of HSA [66] ........................... 43 Table 2 Fluorescence units calculated, as mentioned in experimental section, of NaBH4

reduced WSF before and after passage through the Sephadex G-75 column........... 61 Table 3 Distances calculated using MOE program of lysine pairs that are within close

proximity of HSA as described in the experimental section..................................... 84 Table 4 Summary of MALDI TOF/MS experiments of HSA and aldehyde modified

HSA (hex and 4-ONE).............................................................................................. 91 Table 5 Summary of MALDI TOF/MS experiments on control and aldehydes modified

apo-HST.................................................................................................................. 102 Table 6 Calculated rate constants compared with literature values of kinetics of iron

release of commercial diferric iron at pH 7.4 [61].................................................. 110 Table 7 Calculated rate constants compared with literature values of kinetics of iron

release of commercial diferric iron at pH 5.6 [61].................................................. 111 Table 8 Comparison of calculated rate constants from Origin fits of kinetics of control

and hex modified diferric HST at pH 7.4 ............................................................... 113 Table 9 Comparison of calculated rate constants from Origin fits between control and

hexenal modified HST at pH 5.6 ............................................................................ 115 Table 10 Comparison of calculated rate constants of control and hex modified saturated

apo-HST at pH 7.4 .................................................................................................. 118 Table 11 Comparison of calculated rate constants from control and hex-modified

saturated apo-HST samples kinetics at pH 5.6 ....................................................... 120 Table 12 Comparison of calculated rate constants of control and 4-ONE modified

saturated apo-HST at pH 7.4................................................................................... 122 Table 13 Comparison of calculated rate constants from control and 4-ONE modified

saturated apo-HST samples kinetics at pH 5.6 ....................................................... 123 Table 14 Comparison of calculated rate constants of kinetics from control and hex-

modified titrated apo-HST samples at pH 7.4 ........................................................ 126 Table 15 Comparison of calculated rate constants of control and hex-modified titrated

samples from kinetics at pH 5.6.............................................................................. 128 Table 16 Comparison of calculated rate constants of kinetics from control and 4-ONE

modified titrated apo-HST samples at pH 7.4 ........................................................ 129 Table 17 Comparison of calculated rate constants of control and hex-modified titrated

samples from kinetics at pH 5.6.............................................................................. 130

13

Abbreviations 4-ONE 4-hyroxy-2-nonenal

CNBr cyanogen bromide

DEAE diethylaminoethyl

DTT diththiothreitol

EDTA ethylenediaminetetracetic

Em fluorescence emission

ESI electrospray ionization

Hex E-2-hexenal

HEPES (N[2-hydroxyethyl]piperazine-N’-[2-ethanesulfonic acid])

HNE 4-hydroxy-2-nonenal

HPLC high performance liquid chromatography

HSA human serum albumin

HST human serum transferrin

IA iodoacetic acid

IEF isoelectric focusing

LDL Low density lipoproteins

LP lipid peroxidation

MALDI matrix assisted laser desorption ionization

MDA malondialdehyde

MES (2[N-Morpholino]ethanesulfonic acid)

MS mass spectrometry

PMF peptide mapping fingerprinting

PUFA polyunsaturated fatty acid

Tiron 4,5-dihydroxy-1,3-benzenedisulfonic acid

TOF time of flight

WSF water-soluble fluorescent species

14

1.0 Background

1.1 Lipid Peroxidation (LP)

LP serves as a marker of cellular oxidative stress and has long been recognized

to contribute to oxidative damage, as well as to inflammatory processes,

ischemia and reperfusion injury, and chronic diseases as atherosclerosis and

cancer [1, 2]. LP is a ubiquitous process that occurs in all organisms living under

aerobic conditions. While oxygen is essential to life, it can be potentially toxic to

living organisms due to its participation in the formation of free radicals. LP

proceeds by a free radical chain reaction mechanism in which molecular oxygen

is incorporated into polyunsaturated fatty acids (PUFA) to yield lipid

hydroperoxides (LOOH). Lipid hydroperoxides have been shown to produce

irreversible impairment of membrane fluidity and elasticity, which can lead to

rupture of the cell, and these changes are likely to be particularly significant in

long-lived and predominantly post-mitotic cells such as neurons. Most of these

processes generate superoxide anion radical (O2-•) and H2O [2, 3, 4].

15

Fig 1 Model for the mechanism of initiation of LP [5] Although a number of compounds initiate LP through direct production of

oxidizing radicals, further peroxidation occurs by secondary processes involving

enzyme activation in which unsaturated lipids serve as substrates.

The breakdown products of LP are also believed to contribute to the production

of lipofuscin, a structurally heterogeneous yellowish brown granular pigment that

accumulates with age. Lipofuscin is a fluorescent material that accumulates

within intracellular granules composed, in part, of protein and lipid in postmitotic

cells of a variety of organisms as they age. Although reports on the fluorescence

16

properties vary, this material typically emits light between 400 and 600 nm when

excited at wavelengths ranging from 300 to 400 nm. It has been proposed that

lipofuscin is generated upon reaction of LP products with various cellular

components [6, 7].

1.1.1 Significance of LP End-Products

Ultimate end-products of LP include α, β unsaturated aldehydes. Unlike reactive

free radicals, aldehydes are rather long-lived and can therefore, diffuse from site

of their origin (i.e. membranes) and reach and attack targets intracellularly or

extracellularly, which are distant from the initial free radical event. These

aldehydes once formed or activated induce diverse aspects of severe cellular

stress, including peroxide formation (oxidative stress), chromosomal aberrations,

sister chromatid exchanges, point mutation and cell killing. These aldehydes are

active and exhibit facile reactivity with various biomolecules including proteins,

DNA and phospholipids, generating stable products at the end of a series of LP

reactions. The protein-bound aldehyde is therefore considered to be a useful

biological marker in studies of degenerative diseases. The relative amount of

adducts varies with the type of reactive aldehyde, and is largely determined by

yield and reactivity of aldehydes and the stability of the adducts. Examples of

these aldehydes are malondialdehyde (MDA), acrolein, and 4-hydroxy-2-nonenal

(HNE) [8, 9, 10].

17

Fig 2 Structures of some example of LP derived aldehydes [8]

1.1.2 Role of reactive aldehydes in diseases

An important part of the pathogenesis of atherosclerosis has been implicated by

oxidative modification of low-density lipoproteins (LDL). Formation of LP

products bound to proteins in the vascular lesions, such as atherosclerotic

lesions, is a phenomenon common in most, if not all, types of vascular damage

associated with oxidative stress. The events in atherosclerosis are monocyte

migration from the blood stream, differentiation into macrophages in situ, uptake

of oxidized LDL by macrophage-scavenger receptors, transformation of lipid-

laden macrophage into foam cells, smooth muscle cell proliferation and

transformation into foam cells, and thus accumulation of foam cells leading to

fatty streaks and subsequent plaque formation. Normal LDL binds to receptors

on the surface of cells that require cholesterol and is internalized, releasing

cholesterol. The possibility that reactive aldehydes play a role in the

pathogenesis of atherosclerosis is suggested by the facts that (i) the level of

reactive aldehydes increases in plasma in relation of extensive aortic

atherosclerosis, (ii) high concentrations of aldehydes can be generated during

the oxidation of LDL phospholipids, and (iii) the structural and functional changes

18

associated with the in vivo oxidation of LDL can also be produced by direct

interaction of LDL with aldehydes [8,11-13].

HNE levels are elevated in diseased Alzeimer’s disease (AD) brain regions

compared to age-matched controls. In addition, free HNE is elevated in

cerebrospinal fluid from AD patients compared to AD patients. Protein adduction

of aldehydes such as HNE and acrolein creates new epitopes that can be

detected by appropriate antibodies [14].

The presence of the glyoxal-lysine adduct (CML) has been demonstrated in

various tissues from patients with diabetic nephropathy and chronic renal failure

and in atherosclerotic lesions of arterial walls. Quantification of the MG-derived

lysine-lysine dimer (MOLD) in human serum proteins revealed a significant

increase in diabetic samples compared with normal samples [16,17].

1.2 Biomolecules Modification by Aldehydes.

1.2.1 Protein Modification

Modification of proteins by α, β unsaturated aldehydes results in the formation of

fluorescent proteins. The adducts formed from the in vitro incubation of proteins

with aldehydes typically emit light between 400 and 600 nm when excited at

wavelengths ranging between 300 to 400 nm [18].

HNE is considered as the most toxic and reactive of the generated aldehydes. It

has a broad spectrum of biological toxicity including inhibition of DNA and protein

synthesis, inactivation of several enzymes, and modulation of expression of

several genes [19, 20]. It is an aliphatic α, β unsaturated aldehyde that alkylates

cellular nucleophilic groups located in proteins, nucleic acids or

aminophospholipids. HNE is a multifunctional alkylating agent. It possesses a

C2-C3 double bond in conjugation with a C1 aldehyde. This structure causes C3

to be a strong electrophilic center, a property that is enhanced by the electron

withdrawing C4 hydroxyl group. Michael addition at the C3 carbon can occur by

nucleophilic groups such as thiols, histidinyl side chains, the ε-amino group of

lysines, and other primary amines. The aldehyde functionality of HNE reacts with

19

the ε-amino group of lysine to form a Schiff base that can eventually lead to the

formation of a 2-pentylpyrrole through an imine intermediate. In vitro

experiments on small molecules and peptides had shown that 2 lysine residues

were needed to react with the aldehyde for fluorophore formation. Although it is

generally accepted that the fluorescence in HNE-treated proteins involves the

modifications of lysine residues, the structure of the fluorophore has so far not

been determined [20-25].

Acrolein is another type of reactive aldehyde produced from LP. It is a strong

electrophile and shows high reactivity with nuclphiles, such as the sulfhydryl

group of cysteine, imidazole group of histidine, and amino group of lysine. Its

high reactivity makes acrolein a dangerous substance for the living cell. Acrolein

undergoes nucleophilic addition at the double bond (C---3) to form a secondary

derivative with retention of the aldehyde group, resulting in the formation of the

Michael addition-type-acrolein-amino acid adducts. In vitro studies have

identified the Nε-(3-formyl-3,4-dehydropiperidino) lysine (FDP-lysine) [18,19].

Fig 3 Scheme of proposed formation of Nα-acetyllysine-HNE fluorophore [24]

20

Fig 4 Fluorescence spectra of Nα-acetyllysine-HNE fluorophore showing excitation maxima at 360 nm and emission maxima at 430 nm [24]

21

Fig 5 Proposed structures of adducts formed from different aldehydes [8]

22

Fig 6 Structure of aminophospholipid adduct [27]

1.2.2 Aminophospholipid Modification

In vitro studies have shown that HNE can react with aminophospholipids. A

study conducted by Guichardant et al., demonstrated that HNE can react with the

amino group of phosphatidylethanolamine (PE), yielding the same kind of

products found in amino acid treatment, namely a Michael adduct that represents

the main derivative and a Schiff-base which is partially converted to a pyrrole

derivative [27-29].

1.2.3 DNA Modification.

In vitro studies have been conducted to study the modification of DNA with

acrolein. It has been suggested that acrolein modification of DNA may contribute

to spontaneous mutagenesis, thereby playing a potential role in aging and

cancer. Upon incubation of acrolein with DNA, acrolein primarily reacts with 2-

deoxyguanosine and forms the 1,N-2-propano-2-deoxygyanosine adduct [30].

23

Fig 7 Structure of DNA adducts formed as a result of acrolein incubation [30]

1.3 Water-soluble Fluorescent Species

In 1985, Tsuchida et al. found that there were fluorescent substances in the

protein fraction of mouse and human sera. These fluorescent substances which

were called water-soluble fluorescent species (WSF), had fluorescence spectra

with excitation maxima at 335-340 nm and emission maxima at 435-440 nm in

mice, and excitation maxima were at 355-360 nm and emission maxima at 430-

435 nm, respectively for humans. Tsuchida et al. noted that these WSFs had

similar spectral characteristics to those reported with lipofuscin previously [31].

24

Fig 8 Fluorescence excitation and emission spectra of WSF pigments found in protein fraction of mouse (A) and human (B) sera [31] The aim of their study was to develop a reliable method to extract WSF from

serum, as they proposed WSF as a marker of in vivo LP.

Tsuchida et al. developed an assay for the extraction of WSF from sera of mice

and humans. In their studies, they determined that WSF correlated with the

induction of in vivo LP caused by carbon tetrachloride. Carbon tetrachloride is

considered a stimulative agent of LP in liver. Data showed that fluorescence

intensity significantly increased after the administration of carbon tetrachloride.

In addition, the spectral properties of WSF from test mice showed characteristics

similar to those of control mice. These data indicated that the fluorescence

responded to stimulation of in vivo LP, and it further supported the assumption

25

that WSF originally present in sera of control mice might be products resulting

from in vivo LP [31].

Tsuchida et al. also conducted a study to compare the fluorescence intensity of

WSF from sera of healthy subjects and subjects suffering from various diseases

such as diabetes, hypertension, hyperlipidemia etc. They noticed that the

fluorescence intensity of WSF from patients with diabetes or hypertension was

significantly higher than that from healthy subjects [31].

Seybert et al. conducted a study showing that WSF fluorescence appeared to

increase with age in humans even in normal population [32].

26

1.4 Human Serum Albumin (HSA)

Human serum albumin (HSA) is the most abundant and multifunctional protein in

serum. It is the major protein component of blood plasma, occurring at

concentration of ~ 0.6 mM. It can also be found in bodily tissues and secretions

[33].

1.4.1 Function

The primary function of HSA is to transport fatty acids, but because of its

versatile capacities and high concentration, it can assume a number of additional

functions. HSA carries around 0.1-2 mol of fatty acid per mol protein. HSA helps

to maintain colloidal osmotic blood pressure and contributes to bodily

detoxification by binding poisonous metabolites such as bilirubin [34].

Albumin carries bilirubin, the breakdown product of heme, to the liver for

conjugation with glucuronic acid and subsequent excretion. Albumin has one

very strong binding constant (~108 M-1) and two weaker binding constants for

bilirubin (~106 M-1). It has been observed clinically and in experimental systems

that when the bilirubin to HSA ratio exceeds 1, bilirubin binds to membranes [34].

HSA has been shown to be an important reservoir of the biological regulator

and neuromodulator, nitric oxide. In clinical situations, HSA affinity for an

amazing array of different drugs means that the protein can have a major effect

on pharmacokinetics. HSA is capable of binding an extraordinarily broad range

of drugs, and much of the clinical and pharmaceutical interest in the protein

derives from its effects on drug pharmacokinetics. Many studies have shown

that the presence of fatty acids has unpredictable effects on drug binding,

resulting in both co-operative and competitive interactions [35, 36].

1.4.2 Structure

HSA is a heart-shaped, monomeric protein of 585 amino acids with molecular

weight of 68, 500 daltons (Da). The protein is 67 % α-helical and entirely devoid

27

of β-sheet. HSA is organized into three homologous domains (labeled I-III) and

each domain is comprised of two sub-domains (A and B), which share common

structural elements. A total of 17 disulfide bridges contribute towards HSA’s

thermostability [37, 38].

Although it is now well established that HSA has multiple fatty acid binding sites

of varying affinities, the precise number of binding sites is not known. Although

the binding of fatty acids to HSA has been studied for over 40 years, still clear

understanding of these interactions is far from complete. Current consensus

seems to be that there are two or three high-affinity sites and at least three

further sites of lower affinity. The binding sites are distributed across the three

homologous domains. Bhattacharya et al. conducted a crystallographic study of

five HSA-fatty acid complexes formed using saturated medium-chain and long-

chain fatty acids (C10:0, C12:0, C14:0, C16:0, and C18:0). A total of seven

binding sites that are occupied by all medium-chain and long-chain fatty acids

have been identified, although medium-chain fatty acids are found to bind at

additional sites on the protein, yielding a total of 11 distinct binding locations.

The two principal drug-binding sites, in sub-domains IIA and IIIA, are observed to

be occupied by fatty acids and one of them (in IIIA) appears to coincide with a

high affinity long-chain fatty acid binding site [36, 39, 40].

In essence the three domains are each adapted differently for fatty acid binding

and no two binding sites are alike in detail, although each comprises a

hydrophobic pocket capped at one end with basic or polar side chains. The

architecture of fatty acid binding sites on HSA is quite distinct from that of

intracellular lipid-binding proteins which accommodate long-chain fatty acids

within a 10-strand anti-parallel β-barrel capped by a pair of short α-helices.

However, both types of protein bind fatty acids within hydrophobic cavities and

employ arginine or lysine residues, usually in combination with tyrosines or

serines, in order to hydrogen bond the carboxylate moiety of the bound lipid [41-

43].

28

Fig 9 Structures of HSA complexed with five different fatty acids. The protein secondary structure is shown schematically. The A and B sub-domains are depicted in dark and light shades, respectively. Bound fatty acids are shown in space filling representations [35]

29

1.4.3 In vitro HSA Modification by LP End-Products

Several in vitro studies had been conducted on HSA and bovine serum albumin

(BSA), where protein was incubated with α, β unsaturated aldehydes such as E-

2-octenal, HNE etc. Alaiz and Barragan conducted a study, where they

incubated BSA with E-2-octenal to determine changes induced in BSA. They

observed that the incubation produced fluorescent BSA with excitation maximum

at 350 nm and emission maximum at 440 nm. Structural analysis of fluorescent

adducts using lysine derivatives showed the formation of pyridinium derivatives

as stable end-products. They also performed amino acid analysis of modified

BSA to compare with unmodified BSA. They noticed that there was a 24.84%

loss of lysine residues due to modification. They also noticed high MW bands on

SDS-PAGE suggesting that E-2-octenal was capable of cross-linking the BSA

[44].

Fig 10 Chemical structure of quanternary pyridinium salts [44] Another study was conducted by Hidalgo and Zamora, where BSA was

incubated with 4,5-epoxy-2-E-heptenal (EH). 4,5-epoxy-2-alkenals are another

type of secondary products of LP. They also noticed after the incubation that

30

there was the production of fluorescent BSA with an excitation maximum at 350

nm and an emission maximum at 430 nm. Upon amino acid analysis of modified

BSA, there was a loss of lysine residues as compared with unmodified BSA. By

SDS-PAGE, they noticed that there was 5% increase in the percentage of protein

dimers [45].

1.5 Human Serum Transferrin (HST)

Human serum transferrin (HST) is a glycoprotein with MW of ~80 kDa found in

blood plasma. In normal healthy humans, the concentration of HST falls in the

range of 25-40 µM [46].

1.5.1 Function.

The main function of HST is the transport of iron in blood. HST has two high-

affinity iron-binding sites (K = 1020 M-1 at pH 7.4). The concentration of free Fe

(III) in the presence of HST is less than 10-12 M and at this concentration, iron is

unable to trigger damage via hydroxyl-radical formation. Iron is transported from

the intestine and the liver, via HST, to tissues that require iron for normal

metabolism. A large proportion of iron is directed to bone marrow, the site of

hemoglobin synthesis [47, 48].

Typically only 30% of the iron binding sites of HST are occupied. Thus, it retains

sufficient excess binding capacity to ensure that it is never saturated, except in

chronic iron overload disorders. This is largely to ensure the

compartmentalization and strict regulation of iron mobilization, which in the free

form is toxic. Among other roles, HST may also contribute to the defense against

infections by depriving microorganisms of iron, which is required for their growth

and reproduction [49].

31

1.5.2 Structure

HST is a single-chain, bilobal glycoprotein consisting of 679 amino acids with two

identical oligosaccharide moieties attached by N-glycosidic linkages to

asparagine side chains. The two lobes are homologous, termed N and C lobes,

each bearing a single iron-binding site. The two lobes share ~40% sequence

homology and are presumed to have arisen by a series of gene duplication and

fusion events. Each lobe is further divided into two domains connected by two

antiparallel β-strands that act as a hinge. In each lobe the iron binding site lies in

this interdomain [50].

HST is considered as one of the iron-tyrosinate protein family, where their

structural feature is the coordination of tyrosine to metal centers in proteins.

Structural data for many different transferrins show that each ferric ion is directly

coordinated to the side chains of two tyrosine residues, one histidine residue,

one aspartic residue, and two oxygens from the synergistic carbonate anion

which is anchored by an arginine residue. For the N-lobe the ligands are D63,

Y95, Y188 and H249. For the C-lobe the ligands are D392, Y426, Y517 and

H585. There is an absolute requirement for the presence of a synergistic anion

to obtain high-affinity iron binding. In the absence of a suitable synergistic anion,

the affinity of HST for iron is so low as not to compete successfully with hydrolytic

polymerization and non-specific binding effects. Iron binding leads to a large

conformational change in the HST structure in which the subdomains in each

lobe twist and come together to close the cleft over the metal [51-53].

1.5.3 Iron Binding to HST

Iron binding and release from HST occurs in a pH-dependent manner. Iron binds

to the protein in serum (pH ~7.4) and is delivered to the cells by receptor-

mediated endocytosis. The diferric HST-receptor complex is targeted to an

acidic medium (pH ~5.6) where iron is released. The receptor and apo-HST

return to the cell surface, and apo-HST is released to take more iron and repeat

the cycle. On average, one molecule of HST is believed to undergo 200 such

32

cycles before being degraded. HST ability to bind iron reversibly is easily

determined by spectral analysis, as iron binding leads to a visible absorbance

maximum around 465 nm, resulting in a change in the protein solution from

colorless to salmon pink [54, 55].

1.5.4 Release of Iron.

In vitro studies had shown that the two lobes of HST display different iron release

mechanisms. The two binding sites are not kinetically equivalent toward iron

release from HST. The C-site has the highest affinity for iron and retains it in

acidic media, whereas the N-site is considered more basic and does not retain

iron in acidic media. Therefore, the iron release from the N-site is faster than the

release from C-site. Experiments had also indicated that the release of iron from

the C-site is completely unaffected by whether the N-lobe is open or closed,

implying a lack of cooperativity between the lobes [56, 57].

1.5.4.1 Release of Iron from N-Site.

Extensive studies have established the significant role of a pair of lysines (K206

and K296) in the release of iron from the N-lobe. Structural studies of

recombinant HST N-lobe indicated that these lysine residues (which reside in

different domains) share a hydrogen bond when the two domains come together

as a result of iron binding. The distance between these two lysines is 3.1 Å. In

apo-HST, these two lysines are far apart (9.0 Å) as a result of cleft opening when

the two domains pivot away from each other. Mutagenesis studies had

demonstrated that mutation of K206 effectively locks the N-lobe into a closed

conformation at pH 7.4 and greatly reduces the rate of iron release at pH 5.6.

Studies showed that the two opposing lysines serve as an active trigger that

helps open the N-lobe. Protonation at low pH causes the two lysines to repel

each other resulting in the opening of the cleft and release of iron [58-61].

33

1.5.4.2 Release of Iron from C-Site

The C-lobe exhibits different iron release properties as compared to the N-lobe.

This is mainly due to the absence of an equivalent pair of lysines in the C-lobe.

Structural studies conducted by Dewan et al. suggested that a triad compromised

of a lysine (K534), arginine (R632), and aspartic acid (D634) might serve as the

pH sensitive motif in the C-lobe. Dewan et al. suggested that lowering the pH

would result in weakening or breaking of hydrogen bonds bridging the Arg and

Lys residues. The triad appears to serve as a trigger that helps to initiate iron

release rather than as a lock that prevents it. Each residue of this triad must play

an active role in the process of iron-release from the C-lobe, since mutation of

any of these residues to an alanine drastically slows chelator-mediated iron-

release [61-63].

34

1.6 Statement of Research

1. LP’s α, β unsaturated aldehyde end-products have been implicated in

numerous diseases such as atherosclerosis, diabetes and other disease

states. Modifications of proteins by aldehydes produce fluorescent

species. WSF extracted from human serum showed similar fluorescence

properties to those products detected with in vitro protein and amino-acid

derivatives modifications by aldehydes. The goal was to identify and

characterize the components of WSF (fluorophore binding to components

and fluorophore binding sites within the components), and to identify the

major species contributing to the fluorescence?

2. In vitro studies demonstrated that proteins when modified by aldehydes

can undergo conformational and functional changes as a result of

fluorophore formation accompanied with significant loss of lysine

residues. The goal was to determine by in vitro experiments on HSA and

HST, the structural and functional changes such as MW, pI and function

that may have occurred as a result of aldehyde modification.

35

2.0 Experimental Methods and Instrumentation

The following chemicals and reagents were obtained from Sigma: N-ethyl-

morpholine (99%), EDTA, Tiron, ammonium persulfate (>98%), sodium azide,

sodium citrate (tribasic), acrylamide/bis-acrylamide (30%), HSA (99%), apo-HST

(>98%), holo-HST (98%), BSA, idoacetic acid (99%), CNBr and iodoacetamide.

The following chemicals and reagents were purchased from Bio-Rad: Tris,

Coomassie brilliant blue R250, P-2 gel (fine), P-10, Tween, mercapoethanol and

TEMED.

The following chemicals and reagents were obtained from Fisher: potassium

chloride, potassium phosphate (ACS), potassium ferricyanide and ferrous

ammonium sulfate.

The following gel resins were from Pharmacia Chemicals: Sephadex G-75,

Sephadex G-100 and DEAE Sepharose.

Trans-2-hexenal (98%) was from Aldrich and trypsin was from Promega. All

other chemical and reagents used were of the highest commercial purity.

The instruments used were: HPLC was from Waters with the following

components: 600 pump, 470 scanning fluorescence detector and 2487 dual

absorbance detector. The ESI-MS was from Waters equipped with ZMD

quadrupole mass analyzer. The spectrofluorometer was from Photon

Technology International using Felix TM software for fluorescence intensity with

xenon arc lamp. UV-Vis spectra and readings were obtained with a Varian

CARY 3-E with temperature controller.

36

2.1 Extraction and characterizations of WSF from human serum

2.1.1 Extraction of WSF

The extraction of WSF from human serum was performed following the

procedure published by Tsuchida et al. [31]. To 100 µl of serum, 4 ml of

ethanol/ether mixture (3:1, v/v) was added. The mixture was vortexed vigorously

for 30 seconds, followed by centrifugation at 3000 rpm for 10 minutes. The

supernatant was discarded and the resulting pellet was mixed with 4 ml

ethanol/ether mixture. The mixture was vortexed again for 30 seconds followed

by centrifugation at 3000 rpm for 10 minutes. After discarding the supernatant,

the resulting pellet was dissolved in 200 µl DDH2O. Fluorescence intensity of the

WSF solution was measured with excitation at 350 nm and emission at 460 nm.

Slit widths of the spectrofluorometer of both the excitation and emission

monochromators were set at 5 nm (used in all experiments unless otherwise

specified). The fluorescence data were normalized using quinine (0.1 µg/ml) in

0.05 M H2SO4 as a standard to read 100 relative fluorescence units.

2.1.2 Attempted Chemical Reduction of WSF Extract

WSF was extracted from serum following procedure 2.1.1. WSF at a protein

concentration of 10 mg/ml was treated with 0.75 M NaBH4 for 24 h at room

temperature (RT). Reduced WSF (2 ml) extract was passed through a Sephadex

G-75 gel filtration column (2.2 cm x 93 cm). The equilibrating buffer components

were: 10 mM potassium monophosphate (KPi), 150 mM potassium chloride

(KCl) and 0.02% sodium azide (NaN3), pH 7.0. The buffer was prepared by

dissolving the appropriate amounts of buffer components in DDH2O, and pH was

adjusted to 7.0 by adding 1 M potassium hydroxide (KOH). All experiments

described that included the use of the Sephadex G-75 column, were performed

with the same equilibrating buffer unless otherwise specified. Four ml fractions

were collected and elution profiles were monitored by fluorescence (ex 350 nm,

em 460 nm). Fluorescence units of WSF were measured before applying to the

37

column. Collection and analysis of fractions from a control sample were the

same as the reduced WSF sample mentioned above.

2.2 Separation and Characterization of WSF’s Protein Component

2.2.1 Gel Filtration Chromatography

A volume of 1.5 ml of WSF solution was passed through a G-75 column, and 4

ml fractions were collected. Elution profiles were monitored by absorbance (280

nm) and fluorescence (ex 350 nm and em 460 nm).

2.2.2 SDS-PAGE Gels

Samples were analyzed by sodium dodecyl sulfate-polyacrylamide gel

electrophoresis (SDS-PAGE) using 7-20% separating gels and 4-5% stacking

gels, as discrete percentages. The SDS-PAGE gels were prepared and run

following procedure by Laemmli et al. The high molecular weight standards that

were run with samples for 7%-10% discrete gels contained the following six

proteins with their molecular weights (MWs): myosin (200 kDa), β-glactosidase

(116 kDa), phosphorylase B (97.4 kDa), serum albumin (66 kDa), ovalbumin (45

kDa), and carbonic anhydrase (31 kDa). For SDS-PAGE gels of 20%

separating gels the standards contained the following proteins: phosphorylase b

(97.6 kDa), serum albumin (66 kDa), ovalbumin (45 kDa), carbonic anhydrase

(30 kDa), trypsin inhibitor (21.5 kDa), and lysozyme (14.4 kDa).

Estimates of MWs of unknown proteins were determined graphically from a

standard plot of mobility (mm) versus log MW of standard proteins. Kodak 3D

program was also used to calculate band intensities and areas for determination

of % purity of proteins as well as MW calculations.

2.2.3 Denaturation Experiments

Fractions from peaks A and B from G-75 were pooled together and concentrated

to 2 ml using ultrafiltration. The membrane used for ultrafiltration is PM10, with a

38

MW cutoff of 10 kDa. The pressure used was 40 psi and the membrane was

kept at 4°C during operation. A volume of 1 ml of a concentrated mixture of

peaks A and B was incubated with 1 ml of 8 M urea for 1 h at RT. A control

sample of just concentrated peaks A and B was left for 1 h at RT (same

experimental conditions as sample except for urea). Solution of 8 M urea was

prepared by dissolving the appropriate amount of urea in DDH2O. A volume of

1.5 ml of incubated mixtures of both control and sample were passed through

Sephadex G-75, and 4 ml fractions were collected. The elution profile was

monitored by absorbance (280 nm) and fluorescence intensity (ex 350 nm and

em 460 nm).

2.2.4 Protein Concentration Determinations

Protein concentrations present in peaks A and B from G-75 were determined by

Bradford assay. Coomassie Brilliant Blue G-250 dye was prepared by diluting 1

part dye concentrate with 4 parts DDH2O, followed by filtration through Whatman

#1 filter paper. A standard curve of absorbance at 595 nm versus protein

concentration (mg/ml) was determined using bovine serum albumin (BSA) as

protein standard. To determine concentration of unknown protein mixture, 20 µl

of protein sample were mixed with 1 ml dye. The protein-dye mixture was left to

sit for 10 minutes at RT. The solution’s absorbance was read at 595 nm.

Concentration was determined from the standard curve.

2.2.5 Isolation of 67 kDa Protein Present in Peak B from the Sephadex G-75 Column

The major band in peak B had a MW of 67 kDa as determined by SDS-PAGE.

To separate this band from the rest of the other protein bands in peak B, pooled

fractions of peak B from G-75 were concentrated using ultrafiltration. Two ml of

concentrated peak B at concentration of 10-14 mg/ml was passed through G-75

column. The elution profile was monitored by absorbance at 280 nm and

fluorescence (ex 350 nm, em 460 nm). Purity of the 67 kDa band across the

peak was monitored by SDS-PAGE (10% separating, 4% stacking).

39

2.2.6 Removal of HSA from other protein species present in peak A from the Sephadex G-75 Column

One of the major bands in peak A had a MW of 68 kDa and this was confirmed to

be HSA. To purify HSA, which is the major protein from the rest of protein

bands, pooled fractions of peak A were concentrated using ultrafiltration. Three

ml samples of concentrated peak A at protein concentration of 10-15 mg/ml were

passed through a Blue Sepharose CL-6B affinity column (1.2 cm x 9.3 cm).

Column was equilibrated with 0.05 M Tris/HCl, 0.1 M KCl, pH 7.0. A step

gradient was used to elute proteins from column. The high ionic strength buffer

was composed of 0.05 M Tris/HCl, 1.5 M KCl, pH 7.0. Buffers were made by

dissolving the appropriate amount of solid in water, and pH was adjusted by 1 M

HCl to the desired pH. Two ml fractions were collected and elution profile was

monitored by absorbance (280 nm) and fluorescence (ex 350 nm, em 460 nm).

The run-through peak and the eluting peaks were analyzed by SDS-PAGE (10%

separating, 4% stacking).

2.2.7 Western Blot for Confirmation of 80 kDa Protein

A western blot was conducted on run-through from the affinity column to confirm

presence of HST. SDS-PAGE gels (10% separating, 4% stacking) were

prepared and loaded with 8 µg of sample containing HST and prestained

standards (MWs 203, 137, 79, 42.3 and 31.6 kDa). The SDS-PAGE gel and the

nitrocellulose membrane (0.45 µm) were equilibrated in the transfer buffer for 20

minutes. Transfer buffer was composed of 25 mM Tris, 192 mM glycine, 20%

(v/v) methanol, pH 8.3. After assembling the sandwich containing the gel,

membrane and filter papers, transfer of proteins from gel to membrane was

performed electrophoretically for 1h at 100 V. After transfer, the membrane was

incubated with the blocking buffer for 20 minutes with gentle shaking. Blocking

buffer was composed of 20 mM Tris, 500 mM NaCl, 1% nonfat dry milk (w/v), pH

7.5. After the 20 minutes incubation, fresh 20 ml blocking buffer and 20 µl of

antitransferrin antibody were incubated with membrane for 1 h. After the 1 h

40

incubation, membrane was washed three times with 15 ml of blocking buffer.

Next 20 ml of fresh blocking buffer and 20 µl of horse-radish peroxidase (HRP)-

conjugated anti-goat IgG were incubated with membrane for 1 h. After

incubation membrane was washed twice with 15 ml buffer containing 20 mM Tris,

500 mM NaCl, 0.05% Tween (v/v), pH 7.5. Then membrane was washed once

with 10 ml buffer containing 20 mM Tris, 500 mM NaCl, pH 7.5. At the end, 10

ml of color reagent was added to the membrane and incubated until colors were

visible. The color reagent was prepared by mixing 10 ml 4-chloronapthol with 6

µl of 30% H2O2. All buffers used in this experiment were prepared by dissolving

the appropriate amount of solid in DDH2O and adjusting of pH by 1 M HCl.

2.2.8 Isolation of HST

Pooled fractions of run-through peak from the affinity column were concentrated

by ultrafiltration. Two ml samples at protein concentration of 2-6 mg/ml were

passed through a DEAE-Sepharose column (2cm x 27cm). Equilibrating buffer

was composed of 10 mM Tris/HCl, pH 7.4. Using a gradient mixer, a gradient of

0-100 mM NaCl was used to elute HST from column. Two ml fractions were

collected and the elution profile was monitored by absorbance (280 nm) and

fluorescence (ex 350 nm, em 460 nm). Purity of HST was assessed by SDS-

PAGE gel (10% separating, 4% stacking). A western blot was also performed to

confirm, presence to HST.

2.2.9 HSA and HST Aldehyde Modification

The aldehydes used for modification were hex and 4-ONE. An amount of 12 mg

of desired protein was dissolved in 10 mM KPi, pH 7.4. The concentrations of

aldehydes used were 4 mM for the 4-ONE and 15 mM for the hex. Then the

aldehydes were mixed with the proteins and incubated at 37°C. The time of

incubations was 6 h for HST and 48 h for HSA. The incubated mixtures were

then dialyzed versus DDH2O for 48 h to remove excess unreacted aldehydes.

Then the dialyzed mixtures were evaporated to dryness using a Speed Vac.

41

2.3 MALDI TOF/MS

2.3.1 Proteins Identifications and Confirmations from SDS-PAGE Gel Protein Plugs using MALDI-TOF/MS

Gel plugs from the desired protein band were cut from the SDS-PAGE gel. Gel

plugs were dehydrated for pre-trypsin digestion with the addition of 100 μl of 1:1

(v/v) methanol/ 50 mM NH4CO3 followed by shaking for 30 min at RT. The

supernatant was removed by pipetting followed by another addition of 100 μl of

1:1 (v/v) methanol / 50 mM NH4HCO3 followed by another shaking for 30 min at

RT. After supernatant removal, 50 μl of 100% acetonitrile (ACN) was added to

the gel plugs followed by Speed Vac drying for 15 min. The dried gel plugs were

then incubated with 10 μl trypsin for 4 h at 45°C. After the incubation, 7 μl of

supernatant was pipetted to a 0.65 ml tube. To the trypsin digested plugs, 60 μl

of 1% TFA/ 50% ACN was added followed by shaking for 30 min at RT to extract

proteins from gel plugs. Tubes containing extracted protein solutions were then

Speed Vac dried for 2 h at 45°C.

The dried pellets were dissolved in 3 μl 0.3% TFA/ 50% ACN and mixed with 3 μl

(1:1) of 10 mg/ml matrix dissolved in 800 μl 0.3% TFA/ 50% ACN. The matrix

used was α-cayno-4-hydroxycinnamic acid (CHCA). The sample/matrix mixtures

were spotted on a MALDI plate followed by MS analysis using the Applied

Biosystems 4700 Proteomics Analyzer. NCBlr human database search for

protein matches was performed using the Applied Biosystems GPS Explorer

software. The report generated would have the MW, PI, protein score, C.I. %

and accession number. The protein score is a relative indication of the quality of

the hit. It is dependent on a number of factors, and usually the higher the score

the higher the quality of the hit. The minimum protein score that is trusted to a

good protein hit is 60. The C.I.% (confidence index) represents the %

confidence of the program on the protein hit. Higher C.I.% is coupled with high

protein scores.

42

2.3.2 In vitro Structural Studies on Aldehyde Modified HSA and HST using MALDI TOF/MS

Studies were conducted on a Voyager-DE PRO for the MALDI analysis.

Aldehyde modified samples of HSA and HST was prepared as described in

2.2.9. Instrument conditions for sample analysis were as follows: linear and

positive mode, accelerating voltage = 25000 V, grid = 94%, guide wire = 0.1%,

delay time 500 nsec, matrix = sinapinic acid. Before all analyses, the instrument

was calibrated with commercial HSA and apo-HST.

2.4 Characterization Studies on HSA

2.4.1 CNBr Digestion of Isolated HSA

HSA was digested into seven peptide fragments using cyanogen bromide

(CNBr). CNBr is specific at cleaving at methionine residues within proteins.

Since HSA has six methionine residues within its amino acid sequence, seven

fragments were produced as a result by digestion with CNBr. Since there are 17

disulfide bonds, HSA was first reduced with dithiothreithol (DTT) followed by

reaction with iodoacetic acid (IA). Then the reduced and carboxymethylated

HSA was digested with CNBr [65, 66]. The molecular weights and locations in

protein sequence of the expected fragments are shown in Table1:

Fragment Partial Sequence M.W. (Da)

I 1-87 10 022

II 88-123 4334

III 124-298 20 704

IV 299-329 3357

V 330-446 13 902

VI 447-548 11 941

VII 549-585 4097

Table 1 MW and sequence positions of CNBr fragments of HSA [66]

43

Fig 11 CNBr fragments of HSA separated by tricine gel electrophoresis with the assignment of MW to each respective band [66] Isolated HSA was digested with CNBr to yield seven peptide fragments, MW

ranging from 3.4-20 kDa. CNBr digestion of HSA was carried out following a

procedure by Iadarola et al. HSA at a concentration of 10 mg/ml was reduced

under nitrogen for 3 h in a reduction medium containing 1 M Tris/HCl, 6 M

guanidine chloride (GuHCl), and a 10 fold molar excess of dithiothreithol (DTT)

over the disulfides bonds (SH) concentration, pH 8.0. Iodoacetic acid (IA) (2-fold

excess over total protein cysteines plus DTT concentrations) was added, and the

mixture was maintained in the dark for 1 h. Reaction was terminated by addition

of 6 µl of 2-mercapoethanol. Excess reagents were removed by dialysis versus

DDH O and the resulting protein was lyophilized. The lyophilized HSA pellet was

dissolved in 2 ml solution of CNBr solid dissolved in 70% formic acid. Amount of

CNBr mixed with HSA was a 200 molar excess of CNBr over the methionine

residues. The CNBr/HSA mixture was incubated for 24 h under nitrogen at RT in

dark. At the end of the 24 h incubation, the mixture was diluted 20-fold with

DDH O followed by lyophilization. The lyophilized CNBr digested HSA pellet was

dissolved in 2 ml 0.1% trifluoroacetic acid (TFA) for further analysis.

Fragmentation of HSA was monitored by SDS-PAGE gels (20% separating, 5%

stacking).

2

2

44

2.4.2 Separation of Peptide Fragments using Gel Filtration Chromatography

CNBr fragments of HSA were separated using Ulrogel AcA 202 column gel

filtration column (1.9 cm x 83 cm) with fractionation range of 1-20 kDa. The

column was equilibrated with 5% formic acid, 0.02% NaN3 mixture. One ml

samples at protein concentration of ~4 mg/ml were applied to the column. Two

ml fractions were collected and the elution profile was monitored by absorbance

(220 nm) and fluorescence (ex 350 nm, em 460 nm).

2.4.3 Separation of Peptide Fragments from CNBr-digested of Isolated HSA using HPLC

The literature showed the fragments could be separated and resolved using

HPLC. The column used in the separation was a Vydac C18 column. Two

solvents were used for the elution of fragments: A: 0.05% TFA, B: acetonitrile-2

proponal (2:1 v/v) containing 0.05% TFA. The HPLC C18 column was

equilibrated with 80% A and 20% B. The elution run was performed by running a

gradient of 40% A and 60% B for 30 minutes at room temperature (RT) with a

flow rate of 2 ml/min. The absorbance was monitored at 220 nm [67, 68].

45

CNBr digested HSA fragments were also separated by reversed-phase HPLC.

Separation of fragments was achieved by the use of a Varian C18 column (5 µm,

250 mm x 4.6 mm). Elution of fragments was achieved by a gradient of two

solvents: 0.05% TFA (solvent A), and acetonitrile-2 proponal (2:1, v/v) containing

0.08% TFA (solvent B). The column was equilibrated with 80% solvent A and

20% solvent B. For each run, 100 µl of peptide solution (corresponding to ~ 6

nmol of protein) was injected through column. The elution of peptide fragments

was achieved by the use of 30 min gradient to final conditions of 40% solvent A

and 60% solvent B at a flow rate of 0.5 ml/min [68]. All runs were performed at

RT and elution profile was monitored by absorbance (220 nm) and fluorescence

(ex 350 nm, em 460 nm).

Fig 12 Separation of CNBr fragments of HSA using reversed-phase C18 column HPLC. Conditions are described in 2.3.3 [68]

46

2.4.4 Isoelectric Focusing (IEF) Experiments on HSA

Two samples of HSA were prepared: one modified with hex and the other with 4-

ONE. A control sample of 5 mg/ml HSA dissolved in 20 mM KPi, pH 7.4 was also

prepared. Volumes of 100 μl of the three samples were resuspended in lysis

buffer. The lysis buffer was composed of 7 M urea and 2 M thiourea. The

precipitates of the three samples were resuspended in 175 μl of rehydration

buffer containing 25 mM DTT, 4% CHAPS and 0.2% ampholytes. The 1st

dimension was focusing in an 11 cm IPG strip (pH range 5-9) followed by the 2nd

dimension, which was a 12% SDS-PAGE gel. The stepwise run power

conditions for the IEF experiments were as follows: 100 V for 60 min, 250 V for

60 min, 500 V for 30 min (total 150 min).

47

2.5 Structural studies on HST

2.5.1 CNBr Digestion of Isolated HST and Peptide Fragments Separation

Apo-HST was digested with CNBr and peptide fragments were resolved using

the Sephadex G-100 gel filtration column equilibrated with 5% formic acid

separated the fragments produced. Three peaks were resolved and the elution

profile was monitored by absorbance intensity at 280 nm. The peaks were

assigned as CN-A, CN-B and CN-C, where CN-A contained four peptide

fragments, CN-B contained two peptide fragments and CN-C contained only one

peptide fragment [69].

Since HST has 19 disulfides bonds, CN-A and CN-B were reduced with

mercaptoethanol and followed by carboxyamidomethylation using

iodoacetamide. The reduced and carboxyamidomethylated peak CN-A was

passed through G-100 column [69, 70].

Fig 13 Separation of CNBr fragments of HST on Sephadex G-100. Elution profile monitored at absorbance 280 nm [69]

48

Fig 14 Separation of reduced and carboxyamidomethylated CN-A peptides on Sephadex G-100. The elution profile monitored at absorbance 280 nm from G-100 column [69]

Fig 15 Separation of reduced and carboxyamidomethylated CN-B peptides on Sephadex G-50. Elution profile monitored at absorbance 280 nm [69]

49

The four fragments were separated (fragments 1-4 from table 2). The reduced

and carboxyamidomethylated peak CN-B was passed through Sephadex G-50

gel filtration column equilibrated with 5% formic acid. Two peaks were resolved

corresponding to fragments 5 and 6 from table 2. The elution profile from G-50

column was monitored by absorbance intensity at 280 nm52.

Isolated HST at concentration of 10 mg/ml was incubated with 20 mg/ml CNBr

dissolved in 70% formic acid for 24 h in dark at 25°C. After 24 h incubation, the

mixture was lyophilized [69]. The lyophilized CNBr digest containing the peptide

fragments of M.W. ranging from 5-58 kDa, was dissolved in 5% formic acid

followed by elution through a P-10 column (1.5 cm x 83 cm), a gel filtration

column with fractionation range of 2-20 kDa. Three ml fractions were collected

and the elution profile was monitored by absorbance (at 280 nm) and

fluorescence (ex 350 nm, em 460 nm).

2.5.2 Reduction and Carboxyamidomethylation of Peak A of CNBr-digested Isolated HST from P-10 Column

Peak A was reduced and carboxyamidomethylated following the procedure of

Sutton et al. Lyophilized peak A at concentration of 10 mg/ml was dissolved in

0.2 M Tris/HCl, 8 M urea, pH 8.6, followed by addition of 1% (v/v) 2-

mercaptoethanol. The mixture was incubated at 25°C for 4 h, followed by

addition of 30 mg of iodoacetamide per ml and incubated for another 30 min.

Reaction was terminated by addition of excess 2-mercaptoethanol [69]. The

peptide fragments were applied through a P-10 column equilibrated with 5%

formic acid. Three ml fractions were collected and elution profile was monitored

by absorbance (at 280 nm) and fluorescence (ex 350 nm, em 460).

2.5.3 Separation of CNBr-digested Isolated HST Peptide Fragments using HPLC

CNBr digested HST fragments were also separated by reversed-phase HPLC.

Separation of fragments was achieved by the use of Varian C18 column (5 µm,

250 mm x 4.6 mm). Elution of fragments was achieved by a gradient of these

50

two solvents: 0.05% TFA (solvent A), and acetonitrile (solvent B). The column

was equilibrated with 100% solvent A. For each run, 100 µl of peptide solution

(corresponding to ~ 15 nmol of protein) was injected through column. Elution of

peptide fragments was achieved by the use of 35 min gradient to final conditions

of 40% solvent A and 60% solvent B at a flow rate of 0.5 ml/min. All runs were

performed at RT and elution profile was monitored by absorbance (280 nm) and

fluorescence (ex 350 nm, em 460 nm).

2.6 In vitro functional Studies on HST

2.6.1 Protein Fluorescence Experiments

Apo and diferric HST at concentrations 0.3 mM dissolved in 20 mM KPi, pH 7.5

buffer, were incubated with 15 mM hex or 4 mM 4-ONE dissolved in the same

buffer at 37°C. Control samples of both apo and holo HST were incubated at

37°C, without addition of hex or 4-ONE. Aliquots were taken of samples and

controls at 1 h intervals. Each aliquot consisted of 50 µl of sample or control

diluted to 2.5 ml with KPi buffer. Then the aliquots were read for fluorescence

units (ex 350 nm, em 460 nm), using quinine as standard. Aliquots were also

analyzed by SDS-PAGE gels (7.5% separating, 4% stacking).

2.6.2 Iron Uptake by Apo and Aldehyde Modified HST (Titration Experiments)

Fe(III)Cl3.6H2O was dissolved in 1.10 M trisodium citrate to give a final

concentration of 15 mg/ml. The solution (Fe (III)/citrate molar ratio, 0.05) was

diluted 50-fold with H2O and stored at 4°C. Apo-HST was dissolved to a

concentration of 10 mg/ml in 0.1 M Tris/HCl, 25 mM Na2CO3 pH 8.0. 10 µl

aliquots (each containing 10.8 nmol Fe) were added to apo-HST (5 mg; 62.5

nmol) dissolved in 1 ml 0.1 M Tris/HCl, 25 mM Na2CO3 in 2 ml cuvette. After

each addition, the reaction mixture was mixed with Pasteur pipette for 15

seconds, and the absorbance at 465 nm was measured after 1 min [55].

51

2.6.3 Iron Uptake by Apo and Aldehyde Modified HST (Saturation Experiments)

Apo-HST at concentration of 0.2 mM was dissolved in 1 ml 0.1 M HEPES, 25

mM NaHCO3 pH 7.5. The solution was mixed with 2 ml solution of 0.01 M

ferrous ammonium sulfate dissolved in 0.01 M HCl. After 24 h equilibration

period to ensure maximum binding, the percent saturation was assayed

spectrphotometrically, ε465 = 2500 M-1 cm-1/Fe. Measured saturation levels were

typically > 90% [54]. To produce hex and 4-ONE modified apo-HST, apo-HST at

concentration 0.3 mM was incubated with 15 mM hex or 4 mM 4-ONE for 7 h at

37°C. After incubation, the mixture was dialyzed vs H2O to ensure removal of

aldehydes. The dialyzed protein sample was dried using Speed Vac. The dried

pellet was dissolved in one of the two buffers used for the kinetics experiments

(mentioned below).

2.6.4 Iron Kinetics of Release from Normal and Modified Diferric HST at pH 7.4

Kinetics of iron release from HST samples at pH 7.4 was measured by

monitoring the absorbance change at 480 nm for release of Fe to the chelator

Tiron. A circulating water bath maintained the temperature at 25°C during the

experiment. Final concentration, of 6 µM protein and 12 mM Tiron in a total

volume of 2 ml in 0.1 M HEPES pH 7.4, were used for the kinetic assays [61].

The curve resulting from a plot of absorbance vs time was fitted to a double-

exponential function using Origin software.

2.6.5 Iron Kinetics of Release from Normal and Aldehyde Modified Diferric HST at pH 5.6

Kinetics of iron release from HST samples at pH 5.6 was measured by

monitoring the absorbance change at 470 nm for release of Fe from protein to

the chelator EDTA. Final concentration, of 6 µM protein and 4 mM EDTA in a

total volume of 2 ml in 0.1 M MES pH 5.6, were used in the kinetic runs (at

52

temperature 25°C) [61]. Since the rate of release from N-lobe was too rapid to

accurately measure under the conditions mentioned, the first two minutes of data

(N–lobe) were omitted from the analysis, and the remaining data were fit using a

single-exponential function using Origin software.

2.7 Characterizations of peak C of WSF (the non-protein component).

2.7.1 Concentrating Peak C

Pooled fractions of peak C from the Sephadex G-75 column were passed

through a P-2 gel filtration column (fractionation range 100-1800 Da). P-2

column (1.8 cm x 27.9 cm) was equilibrated with 100 mM N-ethyl-morpholine and

0.02% NaN3, pH 7.4. Three ml fractions were collected and the elution profile

was monitored by fluorescence (ex 350 nm, em 460 nm). Pooled fractions were

concentrated and evaporated to dryness using the Speed Vac instrument. The

resulting pellet was dissolved in DDH2O for further experiments.

2.7.2 MW Determination of Peak C using Electrospray Ionization (ESI) MS

Concentrated peak C was suspended in 100% ACN, HCl mixture and 2 ml was

injected into a Waters ESI/TOF MS for MW determinations. Instrument

conditions for the analysis were as follows: positive mode, cone voltage = 30 V,

ion energy = 0.2, source heater = 100, and desolvation temperature 100.

2.7.3 Peak C reactivity

To determine peak C reactivity towards proteins, 2 ml sample of pooled

concentrated fractions from P-2 column 100 mM N-ethyl-morpholine and 0.02%

NaN3, pH 7.4 were incubated with 5 mg/ml HSA (commercial) for 24 h at 37°C.

A control sample of 5 mg/ml HSA was also prepared and incubated for 24 h at

37°C. HSA was dissolved in 20 mM KPi, pH 7.4. Both control and experimental

samples of HSA and peak C were passed through G-75. Four ml fractions were

53

collected and elution profile was monitored by fluorescence (ex 350 nm, em 460

nm).

2.8 Serum Incubation Experiments

Two ml samples of whole serum were incubated with hex (15 mM) and 4-ONE

(4mM) for 24 h at 37°C. The incubated samples were dialyzed vs DDH2O to

remove excess aldehydes. Extraction was performed on dialyzed samples to

extract WSF following procedure mentioned in 1.1. The two samples of WSF

extracts dissolved in DDH2O of both aldehyde incubations were passed through

G-75 column. Four ml fractions were collected and elution profile was monitored

by fluorescence (ex 350 nm, em 460 nm). Fractions were also analyzed by SDS-

PAGE gels (10% separating, 4% stacking).

2.9 Human subjects’ study

WSF were extracted from serum of six different individuals using procedure

mentioned in 1.1. The extracts from the six subjects were passed through G-75

column. For each subject, four ml fractions were collected and elution profile

was monitored by absorbance (280 nm) and fluorescence (ex 350 nm, em 460

nm). Peak areas were calculated by fitting peaks to nonlinear Gaussian fits

using Origin Software.

54

3.0 Results

3.1 Components of WSF Extracted from Human Serum

3.1.1 Isolation of WSF Extract from Human Serum

WSF was extracted from human serum according to the procedure by Tshcuida

et al. (as described in the experimental section, followed by passage of extract

through the Sephadex G-75 gel filtration column.

Three fluorescence peaks and two absorbance peaks were detected from the

Sephadex G-75 column elution profile. The peaks were labeled peaks A, B and

C, according to the fluorescence elution order.

0 50 100 150 200 250 300 350

400

600

800

1000

1200

1400

1600

Fluorescence Absorbance

Volume (ml)

Fluo

resc

ence

Inte

nsity

at 4

60 n

m

A

B

C

0.0

0.2

0.4

0.6

0.8

Absorbance at 280 nm

Fig 16 Elution profile of WSF extract passed through the Sephadex G-75 column, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 nm and em 460 nm. Arrow indicates the void volume of the column

55

Fig 17 Top: Excitation spectrum of WSF extract from the Sephadex G-75 column. Bottom: Emission spectrum of peak WSF extract

56

Fig 18 Top: Excitation spectrum of peak A from the Sephadex G-75 column. Bottom: Emission spectrum of peak A

57

Fig 19 Top: Excitation spectrum of peak B from the Sephadex G-75 column. Bottom: Emission spectrum of peak B.

58

Fig 20 Top: Excitation spectrum of peak C from the Sephadex G-75 column. Bottom: Emission spectrum of peak C Peaks A and B eluted in the excluded volume and high MW volumes of the

Sephadex G-75 column, while peak C eluted in the included volume and low MW

region of the Sephadex G-75 column. Peaks A and B exhibited strong 280 nm

absorbance content which was not detectable in peak C. This suggested that

WSF consisted of 2 major components: first component was the protein

59

component composed of peaks A and B and a second non-protein component

composed of peak C. Peak C showed no protein content using BioRad assay.

SDS-PAGE gel of peaks A and B showed multiple protein bands.

3.1.2 Reduction of WSF Extract with NaBH4

To detect if there would be an effect of reduction with NaBH4 on fluorescence

properties of WSF components, WSF extract was incubated with NaBH4 as

mentioned in experimental section. Reduced and control samples of WSF were

passed through a Sephadex G-75 column.

0 50 100 150 200 250 300 3500

1000

2000

3000

4000

5000

6000

Fluo

resc

ence

Inte

nsity

at E

mis

sion

460

nm

V o lum e (m l)

10 m g/m l W SF reduced w ith N aBH 4 C ontrol 10 m g/m l W SF

A B

Fig 21 The elution profile of control and NaBH -reduced WSF passed through the Sephadex G-75 column monitored by fluorescence intensity at ex 350 nm and em 460 nm. Arrow indicates the void volume of column

4

Comparisons between fluorescence units of reduced WSF extract before and

after elution through the Sephadex G-75 are illustrated in Table 2. This was

performed to track fluorescence units before and after passage through the

60

Sephadex G-75 column, and observe if fluorescence units would be the same or

different.

Passage through G-75 Fluorescence Units

Before 18.5

After 18.6

Table 2 Fluorescence units calculated, as mentioned in experimental section, of NaBH4 reduced WSF before and after passage through the Sephadex G-75 column The elution profile showed that there was no difference between control and

reduced WSF, suggesting that reduction had no effect of fluorescence properties

of peaks A, B and C. There was also no loss of fluorescence as reduced WSF

was passed through the Sephadex G-75 column.

3.1.3 Binding that exists between fluorophore and protein component of WSF

The aim was to determine if the fluorophore(s) was (were) covalently or

noncovalently bound to proteins present in the protein component of WSF.

Concentrated pooled peaks A and B fractions from the Sephadex G-75 column

were incubated with urea as described in the experimental section. Control and

urea-treated samples were passed through the Sephadex G-75 column.

61

0 5 0 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0 3 5 0

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0

1 2 0 0

1 4 0 0

1 6 0 0

1 8 0 0

2 0 0 0Fl

uore

scen

ce In

tens

ity a

t 460

nm

V o lu m e (m l)

P e a k s A a n d B w / u re a C o n tro l

A

B

Fig 22 The elution profile of concentrated peaks A and B sample incubated with 8 M urea for 1 h at 37°C and control samples eluted through the Sephadex G-75 column, monitored by fluorescence intensity at ex 350 and em 460 nm. Arrow indicates void volume of column

The elution profile from Fig 22 showed that control and urea-treated samples

displayed identical elution profiles from Sephadex G-75 column. The elution

profiles suggest that the fluorophore(s) are covalently bound to the proteins in the

protein component of WSF as all of the fluorescence came out in the excluded

volume of the Sephadex G-75 column.

62

3.2 Characterization of the Protein Component of WSF

3.2.1 Purification of the Major Fluorescent Species in Peak B from the Sephadex G-75 column

Concentrated samples of peak B fractions were composed of three major protein

bands, with MWs of 80, 67 and 51 kDa as determined by SDS-PAGE.

Fig 23 SDS-PAGE image showing WSF extract before passage through the Sephadex G-75, and peaks A and B eluted from the Sephadex G-75 column The major protein at 67 kDa suggested that it could be human serum albumin

(HSA). HSA is the most abundant protein in serum and has a MW of 68 kDa. To

purify the 67 kDa band from the other species, concentrated peak B was passed

through a second Sephadex G-75 column (Fig 24).

63

0 50 100 150 200 250

0

1000

2000

3000

4000

5000

6000

Fluo

resc

ence

Inte

nsity

at E

mis

sion

460

nm

Volum e (m l)

Fig 24 The fluorescence elution profile of concentrate peak B through the Sephadex G-75 monitored at ex 350 nm and em 460 nm. Arrow shows the void volume of the column Purity of 67 kDa protein was monitored by SDS-PAGE gel and >98% purity was

obtained as described in the experimental section.

64

Fig 25 Emission spectrum of the isolated pure HSA. Insert: SDS-PAGE image of the pure isolated HAS The emission spectrum show similar charactertics to those reported in the

literature, suggesting that similar fluorophores were present on isolated HSA [25].

3.2.2 Protein identification of 67 kDa protein using MALDI TOF/MS

In order to verify the identity of the 67 kDa protein, MALDI TOF/MS was used

coupled with peptide mapping fingerprinting (PMF) database search for protein

identification. The 68 kDa SDS-PAGE band gel plug was trypsin digested

followed by analysis using MALDI TOF/MS (Fig 26). The generated peptides

were searched against the protein database (Fig 27) as described in the

experimental section.

65

699.0 1401.8 2104.6 2807.4 3510.2 4213.0Mass (m/z)

406.7

0

10

20

30

40

50

60

70

80

90

100

% In

tens

ity

4700 Reflector Spec #1 MC[BP = 1467.9, 407]

67 kDa Band

1467

.894

2

2045

.168

9

1899

.063

6

1623

.848

416

39.9

863

804.

2894

960.

5756

1149

.662

1

1311

.783

3

1925

.024

5

2559

.293

224

83.1

567

2792

.468

3

1742

.962

0

1226

.625

9

726.

3108

1853

.960

0

2651

.317

9

1671

.835

3

1074

.576

7

880.

3965

2211

.193

8

1600

.791

415

11.9

091

2274

.099

1

2988

.469

7

1992

.048

1

2402

.211

7

1449

.460

1

3878

.949

7

Fig 26 MALDI MS spectrum of trypsin digested 67 kDa band analyzed by 4700 Proteomics Analyzer TOF/MS as described in the experimental section

66

Rank Protein Name Accession No. Protein MW Protein Protein Protein

PI Score C. I. %

Human Serum Albumin In A Complex With Myristic gi|4389275 65992.7 5.69 113 100

Acid And Tri-Iodobenzoic Acid

Peptide Information

Calc. Mass Obsrv. Mass ± da ± ppm Start End Sequence Modification

Seq. Seq.

927.4934 927.5151 0.0217 23 135 141 YLYEIAR

960.5624 960.5754 0.013 14 400 407 FQNALLVR

1074.5425 1074.579 0.0365 34 179 187 LDELRDEGK

1149.615 1149.6636 0.0486 42 39 48 LVNEVTEFAK

1226.6051 1226.623 0.0179 15 8 17 FKDLGEENFK

1311.7419 1311.7794 0.0375 29 335 345 HPDYSVVLLLR

1342.6346 1342.6931 0.0585 44 543 554 AVMDDFAAFVEK

1467.843 1467.8947 0.0517 35 334 345 RHPDYSVVLLLR

1600.7311 1600.7909 0.0598 37 387 399 QNCELFEQLGEYK

1623.7875 1623.8462 0.0587 36 321 333 DVFLGMFLYEYAR

1639.9376 1639.9863 0.0487 30 411 425 KVPQVSTPTLVEVSR

1742.894 1742.9626 0.0686 39 143 156 HPYFYAPELLFFAK

1853.9102 1853.952 0.0418 23 482 497 RPCFSALEVDETYVPK

1875.0156 1875.0869 0.0713 38 62 78 SLHTLFGDKLCTVATLR

1898.9951 1899.0638 0.0687 36 143 157 HPYFYAPELLFFAKR

2045.0952 2045.1669 0.0717 35 370 386 VFDEFKPLVEEPQNLIK

2650.2639 2650.3396 0.0757 29 112 133 LVRPEVDVMCTAFHDNEETFLK

Fig 27 Peptide fragments mapping fingerprinting database search hit for the 67 kDa protein band MALDI TOF/MS confirm that the 67 kDa protein was HSA as we suggested

previously. The high protein score coupled with a 100% C.I. % confirmed HSA.

This demonstrates that HSA was the major species in the WSF and the most

abundant protein.

67

3.2.3 Separation of proteins in peak A from the Sephadex G-75 column from HSA.

Concentrated pools of peak A fractions were composed of at least six protein

bands as determined by SDS-PAGE gel, with major bands with MWs of 80, 67,

51, and 30 kDa (Fig 23).

As the 67 kDa protein was identified to be HSA and it was the most abundant

protein, we wanted to observe if the fluorescence was present only in HSA or

there were other species that possessed the fluorescence. To separate the HSA

from the rest of the species, affinity chromatography was used. Blue Sepharose

CL-6B affinity column (with a high binding specificity for HSA) was used for the

separation as described in the experimental section.

0 1 0 2 0 3 0 4 0 5 0 6 0

0

2

A b s o rb a n c e F lu o re s c e n ce

V o lu m e (m l)

Abs

@ 2

80 n

m

0

2 0 0 0

4 0 0 0

6 0 0 0

8 0 0 0

1 0 0 0 0

1 2 0 0 0

1 4 0 0 0 Fluorescence Intensity @ E

mission 460 nm

Fig 28 The elution profile of concentrate peak A proteins passed through the Blue Sepharose affinity column with 1 M NaCl step gradient, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 and em 460 nm. Arrow indicates the beginning of the gradient

68

Effectiveness of HSA removal from other protein was monitored by SDS-PAGE

gels, which shown that >95% of HSA was removed and separated from the rest

of species (Fig 29).

Fig 29 SDS image peak A before and after passage through the affinity column. Only the run-through peak is shown The elution profile from Blue Sepharose affinity column also shown there was

fluorescence in the run-through peak (A*) suggesting that there were proteins

and species other than HSA that exhibit the anomalous fluorescence. The run-

through peak was composed of three major bands of 80, 51 and 30 kDa

respectively, as determined by SDS-PAGE gel.

3.2.4 Protein Identification and isolation of the 80 kDa protein from peak A* from the affinity column

A protein of 80 kDa MW suggested that it could be human serum transferrin

(HST). HST is one of the abundant proteins in serum and has a MW of 80 kDa.

69

To confirm the presence of HST, a western blot experiment was performed on

peak A* to detect if HST is the 80 kDA protein (Fig 30).

Fig 30 Western blot experiment confirming the presence of HST in peak A* from the affinity column. Standard (Std) contained HST The western blot confirmed that the 80 kDa protein corresponds to HST. The

next step was to isolate HST from the other two proteins mainly the 51 and 30

kDa proteins, and to observe whether the fluorescence is present in HST or

whether the other proteins contribute most of the fluorescence.

To isolate HST, a DEAE Sepharose ion-exchange column was used as

described in the experimental section [69] (Fig 31).

70

0 20 40 60 80 100

0.0

0.1

0.2

0.3

0.4

0.5

Abs @ 280 nm Fluorescence

Volume (ml)

Abs

@ 2

80 n

m

200

400

600

800

1000

1200

1400

1600

Fluorescence Intesity @ Em

iss 460 nm

Peak A1 HST

Fig 31 The elution profile of peak A* passed through the DEAE Sepharose column with 0-100 mM NaCl gradient, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 nm and em 460 nm. Arrow indicates the beginning of the gradient The elution profile showed that both the HST and the run-through (peak A1)

peaks contained fluorescence. Peak A1 contained 2 bands with MW of 55 and

30 kDa as determined by SDS-PAGE gel (Fig 35). The majority of the

fluorescence was present in HST as opposed to the other two proteins. This

suggested that HSA and HST were the two main and major species of the WSF

protein component. Purity of isolated HST was determined to be > 98% pure as

determined by SDS-PAGE gel.

71

The fluorescence emission spectrum of isolated HST showed similar

characteristics to that of isolated HSA suggesting the same fluorophore maybe

present in both proteins. This was significant as this suggests that WSF protein’s

component contained similar fluorophore as that of in vitro studies [25].

Fig 32 Emission spectrum of the pure isolated HST. Insert: SDS-PAGE image of the pure isolated HST

72

699.0 1361.8 2024.6 2687.4 3350.2 4013.0Mass (m/z)

3453

0

10

20

30

40

50

60

70

80

90

100

% In

tens

ity4700 Reflector Spec #1 MC[BP = 1586.8, 3454]

Isolated HST

1586

.756

7

1283

.555

2

842.

5099

1494

.713

1

871.

0183

2211

.106

2

855.

0409

1952

.930

1

1195

.528

8

2171

.086

7

2549

.303

7

1577

.637

7

1478

.720

2

1881

.854

1

1000

.488

5

893.

0013

1097

.994

3

1211

.525

312

76.6

123

2225

.122

1

1433

.576

7

1659

.761

7

839.

0706

1593

.778

915

14.6

587

2176

.064

9

942.

4503

1366

.681

9

2071

.914

6

1060

.011

6

1975

.968

418

85.8

684

1709

.939

1

2301

.169

7

1794

.813

2

2554

.314

0

2989

.368

9

2458

.270

823

85.0

217

2791

.389

2

3136

.577

6

Fig 33 MALDI MS spectrum of 80 kDa band analyzed by 4700 Proteomics Analyzer TOF/MS as described in the experimental section

73


PI Score C. I. %

1 transferrin protein [Homo sapiens] gi|4557871 76999.6 6.81 160 100

Peptide Information


Seq. Seq.

827.4046 827.4145 0.0099 12 565 571 NPDPWAK

964.5322 964.5309 -0.0013 -1 601 609 APNHAVVTR

997.4771 997.4808 0.0037 4 62 69 ASYLDCIR

1000.4985 1000.5045 0.006 6 669 676 YLGEEYVK

1125.6196 1125.5979 -0.0217 -19 613 621 EACVHKILR

1195.5524 1195.5582 0.0058 5 123 132 DSGFQMNQLR

1211.5472 1211.5605 0.0133 11 123 132 DSGFQMNQLR

1273.6534 1273.6676 0.0142 11 226 236 HSTIFENLANK

1276.632 1276.6624 0.0304 24 300 310 EFQLFSSPHGK

1283.5691 1283.5856 0.0165 13 531 541 EGYYGYTGAFR

1323.7167 1323.6871 -0.0296 -22 316 327 DSAHGFLKVPPR

1364.7241 1364.715 -0.0091 -7 542 553 CLVEKGDVAFVK

1415.7198 1415.7311 0.0113 8 47 60 SVIPSDGPSVACVK

1478.7347 1478.7516 0.0169 11 332 343 MYLGYEYVTAIR

1491.759 1491.7677 0.0087 6 298 310 SKEFQLFSSPHGK

1494.7296 1494.7467 0.0171 11 332 343 MYLGYEYVTAIR

1521.7366 1521.7572 0.0206 14 372 384 LKCDEWSVNSVGK

1529.7528 1529.796 0.0432 28 588 600 KPVEEYANCHLAR

1531.6879 1531.7059 0.018 12 684 696 CSTSSLLEACTFR

1565.7992 1565.8234 0.0242 15 647 659 DLLFRDDTVCLAK1

1577.6576 1577.6754 0.0178 11 495 508 FDEFFSEGCAPGSK

1586.7743 1586.7919 0.0176 11 588 600 KPVEEYANCHLAR

1593.8093 1593.8243 0.015 9 476 489 TAGWNIPMGLLYNK

1659.7828 1659.8014 0.0186 11 683 696 KCSTSSLLEACTFR

1705.7526 1705.762 0.0094 6 495 509 FDEFFSEGCAPGSKK

1817.8043 1817.8513 0.047 26 347 362 EGTCPEAPTDECKPVK

1881.8759 1881.8922 0.0163 9 237 251 ADRDQYELLCLDNTR

1952.9381 1952.9438 0.0057 3 572 587 NLNEKDYELLCLDGTR

2171.0952 2171.095 -0.0002 0 144 162 SAGWNIPIGLLYCDLPEPR

2201.0325 2201.0547 0.0222 10 344 362 NLREGTCPEAPTDECKPVK

2233.1248 2233.1138 -0.011 -5 216 236 DGAGDVAFVKHSTIFENLANK

74

2549.2927 2549.2673 -0.0254 -10 252 273 KPVDEYKDCHLAQVPSHTVVAR

Fig 34 Peptide fragments mapping fingerprinting database search hit for the 80 kDa protein band MALDI TOF/MS analysis further confirmed isolated HSA. A gel plug of HST was

trypsin digested and analyzed by 4700 MS as described in the experimental

section. Peptide mapping fingerprinting confirmed HST with a high protein score

and a 100% C.I.% HST.

51

30

Fig 35 SDS image of pass-through from the DEAE column, showing the 30 and 51 kDa proteins

3.2.5 Protein Identifications of the 2 protein from DEAE column (peak A1)

To gain knowledge about the identities of the 2 protein bands present in peak A1

from DEAE Sepharose column, a MALDI experiment was conducted again

coupled with peptide mapping fingerprinting database search. Gel plugs of the 2

proteins (51 and 30 kDa Fig 35) were trypsin digested and analyzed by MALDI

as described in the experimental section.

75

699.0 1401.8 2104.6 2807.4 3510.2 4213.0Mass (m/z)

1509

0

10

20

30

40

50

60

70

80

90

100

% In

tens

ity


51 kDa Band

1467

.852

1

1639

.947

0

1899

.002

3

2045

.099

6

1623

.801

1

927.

4941

1873

.952

3

1677

.810

5

1924

.950

6

804.

2793

2211

.105

7

1343

.645

8

2153

.038

6

2560

.190

9

1808

.000

9

1000

.623

6

1286

.678

1

1186

.647

9

2483

.079

3

909.

4245

1451

.322

6

1742

.881

6

1074

.562

3

1511

.850

515

77.6

835

1128

.705

9

2274

.038

6

2792

.383

1

2090

.108

2

726.

3082

2329

.190

7

1984

.829

6

2416

.086

9

1868

.011

5

2665

.232

426

13.2

942

3005

.346

9

2846

.279

1

2717

.254

9

Fig 36 MALDI spectrum of trypsin-digested 51 kDa protein band analyzed by 4700 Proteomics Analyzer TOF/MS as described in the experimental section Rank Protein Name Accession No. Protein MW Protein Protein Protein

PI Score C. I. %

IGG protein [Homo sapiens] gi|38382881 51491.6 7.51 78 99.836

Peptide Information


Seq. Seq.

838.5032 838.5125 0.0093 11 350 357 ALPAPIEK

1186.6466 1186.6469 0.0003 0 145 156 GPSVFPLAPSSK

1286.6738 1286.6774 0.0036 3 368 378 EPQVYTLPPSR

1671.8085 1671.7833 -0.0252 -15 312 324 TKPREEQYNSTYR

76

1677.8019 1677.8094 0.0075 4 298 311 FNWYVDGVEVHNAK

1808.0065 1808.0044 -0.0021 -1 325 340 VVSVLTVLHQDWLNGK

1872.9701 1872.9779 0.0078 4 368 383 EPQVYTLPPSRDELTK

1873.9218 1873.9512 0.0294 16 416 432 TTPPVLDSDGSFFLYSK

2544.1313 2544.1458 0.0145 6 394 415 GFYPSDIAVEWESNGQPENNYK

Fig 37 Peptide fragments mapping fingerprinting database search hit for the 51 kDa protein band

699.0 1401.8 2104.6 2807.4 3510.2 4213.0Mass (m/z)

1223

0

10

20

30

40

60

70

80

90

100

50

% In

tens

ity


30 kDa Band

2326

.267

622

53.1

536

2198

.212

6

2584

.134

8

1581

.825

1

2141

.985

6

2868

.431

9

1013

.475

3

1709

.932

3

2511

.138

7

2211

.105

7

2712

.223

1

842.

5101

1346

.719

4

1145

.581

9

2807

.310

5

2277

.134

5

2600

.128

4

2115

.868

4

1292

.568

6

1435

.721

1

1598

.820

4

1837

.986

2

3487

.559

8

994.

5864

2883

.505

9

1537

.793

3

2391

.999

8

2956

.431

9

Fig 38 MALDI MS spectrum of trypsin-digested 30 kDa protein band analyzed by 4700 Proteomics Analyzer TOF/MS as described in the experimental section

77


PI Score C. I. %

Carbonic Anhydrase Ii (Carbonate Dehydratase, Hca Ii,gi|1065006 29097.9 6.63 42 0

Ca2) (E.C.4.2.1.1) Mutant With Gln 92 Replac

Peptide Information


Seq. Seq.

1141.5286 1141.5642 0.0356 31 9 17 HNGPEHWHK

1346.7426 1346.7205 -0.0221 -16 76 88 AVLKGGPLDGTYR

1581.8171 1581.8246 0.0075 5 113 125 YAAELHLVHWNTK

1709.9121 1709.9331 0.021 12 112 125 KYAAELHLVHWNTK

2667.2949 2667.2065 -0.08 -33 89 111 LIEFHFHWGSLDGQGSEHTVDKK

Fig 39 Peptide fragments mapping fingerprinting database search hit for the 30 kDa protein band The peptide mapping fingerprinting database search the 51 kDa protein identified

Ig-G with a high protein score and C.I.%. For the 31 kDa protein, the database

identified carbonic anhydrase as the protein but with a low protein score and 0

C.I.%. This could be due to the fact that carbonic anhydrase was in low

abundance and amount that are major factors influencing the score. But the MW

agreed with what was expected according to SDS-PAGE gel and that is a valid

parameter for confirming proteins.

78

3.3 Investigation of the nature of the non-protein WSF component (peak C from the Sephadex G-75 Column)

3.3.1 Peak C Purification and Concentration

Peak C was designated as the non-protein component of WSF as it was negative

towards protein assays and absorbance spectrum did not show any

characteristic spectrum of protein and peptides (absorbance at 220 nm and 280

nm). Peak C was purified using P-2 gel filtration column (Fig 39) and

concentrated using Speed Vac as described in the experimental section.

0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 00

1 0 0 0

2 0 0 0

3 0 0 0

4 0 0 0

5 0 0 0

6 0 0 0

Fluo

resc

ence

Inte

nsity

at E

mis

sion

460

nm

V o lu m e (m l)

P e a k s A a n d B

P e a k C

Fig 40 The fluorescence elution of WSF extract passed through the P-2 column, monitored at ex 350 nm and em 460 nm, as described in the experimental section. Arrow shows the void volume of column

79

3.3.2 Molecular weight determination of peak C using ESI-MS

Fig 41 ESI-MS spectrum of peak C The molecular weight of peak C was determined to be 381.5 Da using ESI-MS as

described in the experimental section. The 267.5 Da peak corresponded to the

buffer components, mainly the moropholine buffer used for elution from the P-2

column.

80

3.3.3 Peak C Reactivity

The aim of this experiment was to determine if peak C was a reactive species

capable of reacting with proteins to generate a covalent fluorophore. Peak C

was incubated for 24 h with commercial HSA at 37°C. The incubated mixture

was passed through a Sephadex G-75 column as described in the experimental

section.

0 50 100 150 200 250 300 350200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

Fluo

resc

ence

Inte

nsity

@ E

mis

sion

460

nm

Volum e (m l)

5 m g/m l HSA + Peak C Control 5 m g/m l HSA

HSA

Peak C

Fig 42 Incubated mixture of peak C and HSA and control HSA passed through the Sephadex G-75 column. The fluorescence elution profile monitored at ex 350 nm and em 460 nm as described in the experimental section. Arrow shows the void volume of the column The elution profile from the Sephadex G-75 column showed that the incubated

mixture of peak C and HSA eluted as two peaks. The control HSA matched

perfectly with the HSA sample. The elution profile suggested that peak C was

81

not reactive with HSA, as fluorescence was detected in the included volume of

the Sephadex G-75 column, where it elutes normally for peak C.

3.4 Characterization of Isolated Proteins

3.4.1 HSA

3.4.1.1 CNBr Digestion of Isolated HSA and Peptide Fragments Separation

The aim of the experiment was to CNBr digest isolated HSA and determine if

there were a single or multiple fluorophore binding sites. This analysis, in

addition to identifying multiple sites, could also determine where approximately in

the amino acid sequence the fluorophore(s) are located. CNBr digestion of

isolated HSA produced the expected seven fragments as produced in the

literature Fig 43 [66].

Separation of CNBr fragments of isolated HSA was achieved by using Ulrogel

AcA gel filtration column as described in the experimental section.

Fig 43 SDS-PAGE image of CNBr digested HSA peptide fragments

82

0 2 0 4 0 6 0 8 0 1 0 0

0 .0 0

0 .0 5

0 .1 0

0 .1 5

0 .2 0

0 .2 5

0 .3 0

A b s o rb a n c e F lu o re s c e n c e

V o lu m e (m l)

Abs

@ 2

20 n

m

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0

1 2 0 0

1 4 0 0

1 6 0 0 Fluorescence Intensity @ E

mission 460 nm

1

2

34

Fig 44 The elution profile of CNBr fragments of isolated HSA passed through the Ulrogel AcA gel filtration column, monitored by absorbance at 280 and fluorescence intensity at ex 350 and em 460 nm. Arrow shows the void volume of column Absorbance elution profile from Ulrogel AcA column of CNBr peptide fragments

of isolated HSA showed that there were four peaks, labeled peaks 1, 2, 3 and 4

Fig 44. The fluorescence elution profile detected three peaks correlating with

absorbance peaks 1, 3 and 4 Fig 44. The elution profile suggested that there

were, in fact multiple binding sites of fluorophore within HSA. Peak 1 had the

highest absorbance and fluorescence intensity and confirmed by SDS-PAGE to

be the 20 kDa peptide Fig 45. We conducted distance calculations and

determination of lysine residues using computational programs (MOE) Table 3 in

HSA which identified 4 lysine pairs on the 20 kDa peptide fragment with very

close proximity that could result on fluorophore formation and that explained the

high fluorescence intensity as compared to other peptide fragments peaks.

Peaks 3 and 4 had similar absorbance and fluorescence intensity and they both

suggested that those were composed of more than one peptide fragment.

83

Fig 45 SDS-PAGE image of CNBr digested HSA peptide fragment and the 20 kDa peptide band of peak 1 from the Ulrogel AcA column.

Residue 1 Residue 2 Distances (Å) Peptide Fragment

132 133 2.72 III (20 kDa) 132 155 2.95 III (20 kDa) 270 272 4.91 III (20 kDa) 277 282 5.72 III (20 kDa) 409 410 2.8 IV (3.4 kDa) 432 435 60.3 IV (3.4 kDa) 520 521 2.76 I (9.4 kDa) 530 532 4.42 I (9.4 kDa) 532 534 5.61 I (9.4 kDa) 553 556 5.30 VII (4.0 kDa) 537 541 6.56 VII (4.0 kDa)

Table 3 Distances calculated using MOE program of lysine pairs that are within close proximity of HSA as described in the experimental section.

84

In order to better identify the CNBr fragments of isolated HSA that contained the

fluorescence, reverse-phase HPLC was used. HPLC should offer better

separation and resolution than Ultrogel gel filtration column, plus it was coupled

with a fluorescence detector to detect simultaneously absorbance and

fluorescence. C18 column was used for the separation as described in the


Fig 46 Elution of CNBr fragments of isolated HSA from reverse phase HPLC. Run conditions were described in the experimental section

85

-0.1

0.4

0.9

1.4

1.9

2.4

0 10 20 30 40 50 60

Time (min)

Inte

nsity

(arb

itury

)Absorbance

Fluorescence

Fig 47 The fluorescence and absorbance elution profiles of CNBr fragments of isolated HSA from reverse-phase HPLC. Run conditions were described in the experimental section The elution profile from HPLC suggested also multiple binding sites of

fluorophore within HSA. The following peptides showed fluorescence: 20, 9.4,

4.0 and 3.4 kDa. HPLC elution profile also showed that there were multiple

fluorophore binding sites. The 20 kDa peptide fragment showed again the

highest fluorescence intensity as compared to other fragments.

The elution profiles from both theUlrogel AcA column and the HPLC reverse-

phase C18 column confirmed that there were multiple binding sites of fluorophore

within HSA.

86

3.4.1.2 In vitro characterization studies by MALDI TOF/MS

Experiments were conducted using MALDI TOF/MS conducted on commercial

HSA and in vitro aldehyde-modified HSA to detect any differences in MW

between modified versus unmodified protein. The aldehydes used for the

modification of HSA were 2-hexenal (hex) and 4-oxo-2-nonenal (4-ONE).

Pollack’s work in our lab showed that incubation of HSA with aldehyde at 37°C

resulted in the development of fluorescence over a period of 48 hours. So, if

there was fluorescence development over time due to fluorophore formation,

then a shift in MW of modified samples compared to control should occur and

MALDI TOF/MS should be able to detect it.

87

999.0 20799.6 40600.2 60400.8 80201.4 100002.0Mass (m/z)

1.0E

0

10

20

30

40

60

70

80

90

100

50

% In

tens

ityVoyager Spec #1[BP = 1062.9, 10006]

Commercial HSA

1062.96

1163.61

1300.22

1429.54

66363.031544.09 66263.66

66591.591714.3666066.162642.7066712.30

1970.34

2939.42 66818.1465820.002232.42

33181.043284.85 66927.7733278.68

65598.6733067.473446.81

65469.913948.14 33422.3567240.38

9039.084404.0465182.485121.9010736.46

Fig 48 MALDI MS spectrum of commercial HSA

88

999.0 20799.6 40600.2 60400.8 80201.4 100002.0Mass (m/z)

1.9E

0

10

20

30

40

60

70

80

90

100

50

% In

tens


4-ONE Modified HSA

1107.19

1224.96

1327.83

1427.34

1527.93

1716.22

1971.59

2085.39

69357.812646.1569922.37

2869.64 70202.803815.44 70663.3334709.554134.17

Fig 49 MALDI MS spectrum of 4-ONE modified HSA

89

999.0 20799.6 40600.2 60400.8 80201.4 100002.0Mass (m/z)

3.3E

0

10

20

30

40

60

70

80

90

100

50

% In

tens


Hexenal modified HSA

1073.99

1225.23

1340.13

1436.73

1534.552121.36

1652.321893.98

2333.682443.14

2839.86

3061.5669670.493342.82

4143.44 68747.074953.71 35193.86

Fig 50 MALDI MS spectrum of hex modified HSA

90

Sample Observed MW Difference Aldehyde Molecules Reacting

HSA (control) 66363 -

4-ONE modified HSA

69357 2994 19

Hex modified HSA

69670 3307 34

Table 4 Summary of MALDI TOF/MS experiments of HSA and aldehyde modified HSA (hex and 4-ONE) MALDI experiments confirmed the hypothesis that there was a MW increase in

HSA due to modification. Data had shown also that the number of hex molecules

reacting to form the fluorophore(s) was higher than 4-ONE molecules. This could

be attributed to the fact that the concentration of hex used for the modification

experiments (15 mM) was much higher than that of 4-ONE (4 mM), or it might

reflect different steric constraints on the number of sites modified.

3.4.1.3 In vitro Structural Characterizations using Isoelectric Focusing (IEF) Experiments

The aim of this experiment was to detect any changes in the pI of HSA (5.69)

due to aldehyde modification. Aldehydes used were hex and 4-ONE as

described in the experimental section. Experiments in sec 3.4.1.2 showed that

there was significant MW shift of aldehyde modified HSA versus a control

unmodified HSA. This meant that there was fluorophore formation and therefore,

presumptive loss of lysine residues due to fluorophore formation. If a significant

number of lysine residues were lost due to modification then a change in pI

should be observed with a shift towards the acidic side, concomitant with the loss

of basic amino acids residues.

91

Basic side Acidic side

Fig 51 IEF image of control HSA sample on a IPG strip of pH range 5-8 Basic side Acidic side

Fig 52 IEF image of hex-modified HSA on a IPG strip with pH range5-8

92

Basic side Acidic side

Fig 53 IEF image of 4-ONE-modified HSA on a IPG strip with pH range 5-8 IEF experimental results showed that there was a pI shift towards the acidic side

for hex and 4-ONE modified HSA samples as compared to control unmodified

HSA. 4-ONE sample showed more pI change than hex sample. HSA isoforms

were not that clear as the control sample as the isoforms are not clearly

separated suggesting that there might be some crosslinking occurring between

some isoforms.

3.4.2 HST

3.4.2.1 CNBr Digestion of Isolated HST Peptide Fragments Separation

The aim of the experiment was to CNBr digest isolated

HST and determine if there was one or multiple fluorophore binding sites.

Isolated HST was digested with CNBr as described in the experimental section.

Then CNBr digested isolated HST was passed through a P-10 gel filtration

column. The elution profile from P-10 column is shown in Fig 54.

93

0 20 40 60 80 100 120 140 160 180 200

0

2

Absorbance Fluorescence

Volume (ml)

Abs

@ 2

80 n

m

200

400

600

800

1000

1200

1400

1600

1800

2000Fluorescence Intensity @

460 nmA

BC

Fig 54 The elution profile of CNBr-digested isolated HST passed through the P-10 column, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 nm and em 460 nm. Arrow shows the void volume of the gel Three peaks were detected by absorbance at 280 nm, labeled peaks A, B and C

according to their elution order from column. The fluorescence elution profile

showed only one peak corresponding to absorbance peak A. Peak A consisted

of four peptide fragments of MW ranging from 7-27 kDa Fig 55. Peak B

consisted of two fragments of MWs 3 and 9 kDa. Peak C contained only 1

fragment with M.W. 5.8 kDa.

94

Fig 55 SDS-PAGE image of peak A of CNBr digested HST from the P-10 column

95

0 20 40 60 80 100 120 140 160 180 200

0

2

Absorbance Fluorescence

Volume (ml)

Abs

@ 2

80 n

m

200

400

600

800

1000

1200Fluorescence Intensity @

460 nm

Fig 56 The elution profile of reduced and carboxyamidomethylated peak A of CNBr-digested isolated HST through the P-10 column, monitored by absorbance at 280 nm and fluorescence intensity at ex 350 nm and em 460 nm. Arrow shows the void volume of the column To separate the four CNBr peptide fragments present in the fluorescent peak A,

the peptide mixture was reduced and carboxyamidomethylated, then passed

through a P-10 column as described in the experimental section.

The elution profile from P-10 column showed two absorbance peaks, labeled

peaks 1 and 2. Fluorescence elution profile showed also 2 peaks correlating with

absorbance peaks 1 and 2.

The absorbance and fluorescence elution profile from P-10 column suggested

multiple fluorophore binding sites within HST. The four CNBr were not resolved

using P-10 but the 2 peaks suggested that they each contained 2 peptide

fragments.

96

In order to achieve better separation of CNBr isolated HST peptide fragments,

reverse-phase HPLC was used. This was a pilot experiment as there is no

published procedure for separation of CNBr HST peptide fragments using HPLC.

The advantage was that a better separation, resolution and using less peptide

amounts could be attained using HPLC. A C18 column was used for the

separation as described in the experimental section.

Fig 57 The elution profile of CNBr fragments of isolated HST from reverse-phase HPLC. Run conditions were described in the experimental section We were only successful on separating peaks A and B of CNBr digested isolated

HST. Attempts to separate the reduced and carboxyamidomethylated peak A

peptide fragments were not successful.

97

3.4.2.2 In vitro characterizations studies using MALDI TOF/MS

Experiments were conducted using MALDI TOF/MS conducted on commercial

apo-HST and in vitro aldehyde-modified HST to detect any differences in MW

shifts between modified versus unmodified samples. The aldehydes used for the

modification of HSA were hex and 4-ONE. These experiments were conducted

in parallel to those performed on HSA where significant MW shifts were observed

due to modification. Shifts in MW in aldehyde modified samples compared to

control unmodified protein samples would corroborate that fluorophore formation

had resulted from covalent adducts.

98

10000.0 28000.4 46000.8 64001.2 82001.6 100002.0Mass (m/z)

963

0

10

20

30

40

60

70

80

90

100

50

% In

tens

ityVoyager Spec #1=>NF1.0[BP = 1061.8, 17255]

Commercial HST78703.65

78265.68

78129.01

79126.84

78013.46

77928.09

77715.77

77601.25

39337.90

79840.4839127.65

77106.1039650.61

76757.4138973.57

39810.04

76009.53

Fig 58 MALDI MS spectrum of commercial apo-HST

99

10000.0 28000.4 46000.8 64001.2 82001.6 100002.0Mass (m/z)

167

0

10

20

30

40

0

60

70

80

90

100

5

% In

tens


80690.3780949.9880420.01

81831.30

80087.61

82576.10

82802.1079588.70

79348.33

83299.38

83449.18

78605.05

4-ONE modified HST

Fig 59 MALDI MS spectrum of 4-ONE modified apo-HST

100

10000.0 28000.4 46000.8 64001.2 82001.6 100002.0Mass (m/z)

116

0

10

20

30

40

0

60

70

80

90

100

5

% In

tens


81732.3782440.04

82198.29

79645.05

Hexenal-modified HST

Fig 60 MALDI MS spectrum of hex-modified apo-HST

101

Sample Observed MW Difference Aldehyde molecules reacting

HST (control) 78302 -

4-ONE modified 80178 1876 12

Hex modified 81732 3430 35

Table 5 Summary of MALDI TOF/MS experiments on control and aldehydes modified apo-HST MALDI experiments showed that there were significant MW shifts of aldehyde

modified samples as compared to control unmodified apo-HST. This confirmed

that fluorophore formation led to shifts in MW of HST as it happened on HSA. As

in the case with HSA, the number of hex molecules reacting was higher again

than 4-ONE molecules as this could be due to the higher hex concentration used

for modification (15 mM) as apposed to 4-ONE concentration (4 mM), or it might

reflect different steric constraints on the number of sites modified.

3.5 In Vitro Fluorescent Development Studies on HST.

In vitro studies were conducted on commercial apo and diferric HST to determine

if there would be fluorescence development over time due to aldehyde

modification. Pollack’s work in our lab had demonstrated that there was

fluorescence development of HSA over time upon incubation with aldehydes.

This experiment was conducted to determine if similar pattern of fluorescence

development over time would be detected.

Apo and diferric HST were incubated for 24 h at 37°C with hex and 4-ONE, and

time points were taken and fluorescence values were normalized using quinine

as a standard as described in the experimental section. Both apo and diferric

forms were incubated with the aldehydes, to determine if either form would be

more susceptible to aldehyde modification.

102

0 5 10 15 20 252

4

6

8

10

12U

nits

of F

luor

esce

nce

Time (h)

Control Apo-HST Holo-HST

Fig 61 Fluorescence development time points plot of hex-modified apo and diferric HST

103

0 5 10 15 20 250

5

10

15

20

25

30

35

40

45

Nor

mal

ized

Flu

ores

cenc

e U

nits

Time (h)

Control Apo-HST Holo-HST

Fig 62 Fluorescence development time point plot of 4-ONE modified apo and diferric HST

104

Fig 63 SDS-PAGE image of hex and 4-ONE modified apo-HST versus unmodified apo-HST Results showed rapid increase in fluorescence over time in aldehyde modified

protein samples compared to control unmodified apo-HST, which displayed

essentially no change. Both aldehyde modified forms showed fluorescence

development with time, but the apo-form showed a more pronounced

fluorescence increase. 4-ONE modified forms of HST displayed higher

fluorescence development as compared to hex modified forms of HST. With

both hex and 4-ONE modified HST, the apo form showed higher fluorescence

development and increase compared to diferric HST.

SDS-PAGE showed formation of high MW species for both hex and 4-ONE

modified sample compared to the control sample, suggesting formation of

crosslinked species as a result of aldehyde modification.

105

3.6 In Vitro Functional Studies on HST

3.6.1 Iron Uptake

The aim of this experiment was to subject apo-HST to aldehyde modification and

determine if modification had an effect on the ability of apo-HST to bind Fe. This

was significant, as it would illustrate if aldehyde modification would have an effect

on the function of apo-HST, which is the binding and transport of Fe.

Iron uptake experiments were conducted by titration of aldehyde (hex and 4-

ONE) modified apo-HST with Fe (III)-citrate. The titration was monitored by

absorbance at 465 nm, and control and aldehyde (hex and 4-0NE) modified apo-

HST samples were analyzed as described in the experimental section.

106

0.0 0.5 1.0 1.5 2.0 2.5 3.0-0.005

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.050

0.055

Modified apo-HST w/ 4 mM 4-ONEA

bs @

465

nm

/mg

of a

po-H

ST

Fe/apo-HST Molar Ratio

Control

0.0 0.5 1.0 1.5 2.0 2.5-0.005

0.000

0.005

0.010

0.015

0.020

0.025

Fig 64 Plot of Fe uptake titration experiment of control and 4-ONE-modified apo-HST samples as described in the experimental section. Insert: Equivalence point determination using Origin software

1.38

Abs

@ 4

65 n

m/m

g ap

o-H

ST

per m

l


0.0 0.5 1.0 1.5 2.0 2.5 3.0

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Abs

@46

5/m

g ap

o-H

ST p

er m

l


Control

2.24

107

0.0 0.5 1.0 1.5 2.0 2.5 3.00.00

0.01

0.02

0.03

0.04

0.05

ControlAb

s @

465

/mg

apo-

HST

per

ml

Fe/aop-HST Molar Ratio

Modified apo-HST w/ 15 mM E-2-hexenal

0.0 0.5 1.0 1.5 2.0 2.5 3.0

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Fig 65 Plot of Fe uptake titration experiments of control and hex modified apo-HST sample as described in the experimental section. Insert: Equivalence point determination using Origin software Iron uptake experiments results showed that there were significant differences

between the control and the aldehyde modified samples. Both aldehyde

modified samples showed decrease to iron binding as compared to control

samples. Equivalence points of the control samples showed a molar ratio of

2.24, reflecting of the two binding sites of apo-HST. The hex and 4-ONE

modified apo-HST equivalence points showed molar ratios of 0.73 and 1.38

respectively. This suggested that there could be a loss of one of the binding

2.24

Abs

@46

5/m

g ap

o-H

ST p

er m

l


Control

0.0 0.5 1.0 1.5 2.00.008

0.010

0.012

0.014

0.016

0.018

0.020

0.022

0.024

0.73Abs

@ 4

65 n

m/m

g ap

o-H

ST

per m

l


Hex modified apo-HST

108

sites. The other possibly that there was a decrease of iron binding on both sites

due to aldehyde modification.

3.6.2 Fe Kinetics of Release from Diferric HST

To further investigate the apparent loss of binding sites of apo-HST due to

aldehyde modification, in vitro kinetic of Fe release experiments were conducted

measuring the rate of dissociation of Fe. The hypothesis was that if there was a

loss of a binding site, it might be the N-site as opposed to the C-site. This was

due to the fact that the N-site function depended on the action of two lysine

residues for the uptake and release of Fe from HST, as mentioned in sec 1, thus

making the N-site potentially more susceptible. Kinetics of Fe release were

conducted at pH 7.4 and 5.6, which were the pHs for iron uptake and release

respectively as described in the experimental section.

3.6.2.1 Fe Kinetics of Release from Commercial Diferric HST

Commercial diferric-HST was used to optimize the kinetic experimental

conditions and to observe if calculated rate constants were comparable with

literature rate constant values.

109

0 50 100 150 200 2500.00

0.01

0.02

0.03

0.04

0.05

Data: Data3_BModel: ExpDec2Equation: y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0Weighting: y No weighting Chi^2/DoF = 9.5476E-8R^2 = 0.99876 y0 0.05207 ±0.00061A1 -0.02466 ±0.00024t1 155.85045 ±11.15245A2 -0.01821 ±0.00065t2 23.83248 ±0.89164

Abs

@ 4

80 n

m

Time (min)

Fig 66 Origin fit of kinetic of diferric-HST at pH 7.4 using Tiron as chelator and absorbance monitored at 480 nm

Runs KN (x10-3) (min-1) KC (x10-3) (min-1)

Literature 32.3 ± 1.4 8.5 ± 0.6

Experimental 31 ± 7.8 4.7 ± 3.8

Table 6 Calculated rate constants compared with literature values of kinetics of iron release of commercial diferric iron at pH 7.4 [61]

110

0 50 100 150 2000.024

0.026

0.028

0.030

0.032

0.034

Data: Data3_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 3.2405E-8R 2 = 0.99113 y0 0.02511 ±0.00002A1 0.00812 ±0.00005t1 43.01664 ±0.60919

Abs

@ 4

70 n

m

Time (min)

Fig 67 Origin fit of kinetics of diferric-HST at pH 5.6 using EDTA as chelator and absorbance monitored at 470 nm

Runs KC (x10-3) (min-1)

Literature 24 ± 0.8

Experimental 23 ± 0.73

Table 7 Calculated rate constants compared with literature values of kinetics of iron release of commercial diferric iron at pH 5.6 [61]

111

The calculated rate constants for pH 7.4 and pH 5.6 were very comparable with

the literature rate constant values. This suggested that the kinetic experiments

were optimized and would yield reliable rate constants.

3.6.2.2 Effect of Aldehyde Modification on Fe Kinetics of Release from Commercial Diferric HST

Commercial diferric-HST was incubated with hex, and the kinetics of iron release

were determined on control and hex-modified samples at pH 7.4 and 5.6, as

described in the experimental section.

0 50 100 150 200 2500.00

0.01

0.02

0.03

0.04

0.05

Data: Data1_BModel: ExpDec2Equation: y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0Weighting: y No weighting Chi 2/DoF = 9.5476E-8R^2 = 0.99876 y0 0.05207 ±0.00061A1 -0.02466 ±0.00024t1 155.85045 ±11.15245A2 -0.01821 ±0.00065t2 23.83248 ±0.89164

Abs

@ 4

70 n

m

Time (min)

Fig 68 Origin fit of Fe kinetic of release of control diferric-HST using Tiron as chelator and absorbance monitored at 470 nm at pH 7.4

112

0 50 100 150 200 2500.020

0.025

0.030

0.035

0.040

0.045

0.050

0.055

Data: Data3_BModel: ExpDec2Equation: y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0Weighting: y No weighting Chi 2/DoF = 3.7124E-8R 2 = 0.99897 y0 0.05695 ±0.00064A1 -0.01106 ±0.00028t1 22.18078 ±0.72218A2 -0.0214±0.00036t2 210.61155 ±13.50067

Abs

@ 4

80 n

m

Time (min)

Fig 69 Origin fit of kinetic of hexenal modified diferric-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4

Sample KN (x10-3) (min-1) KC (x10-3) (min-1)

Control 35 ± 6.4 6.4 ± 3.4

Hex modified 45 ± 10.1 5.4 ± 2.3

Table 8 Comparison of calculated rate constants from Origin fits of kinetics of control and hex modified diferric HST at pH 7.4

113

0 50 100 150 200

0.006

0.008

0.010

0.012

0.014

0.016


Abs

@ 4

70 n

m

Time (min)

Fig 70 Origin fit of kinetic of Fe release from control diferric HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6

114

0 50 100 150 2000.010

0.011

0.012

0.013

0.014

0.015

0.016

0.017

0.018

0.019

Data: Data3_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 3.4317E-8R2 = 0.9901 y0 0.01115 ±0.00003A1 0.00763 ±0.00005t1 51.09792 ±0.83102A

bs @

470

nm

Time (min)

Fig 71 Origin fit of kinetic of Fe release from hex-modified diferric HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6

Sample KC (x10-3) (min-1)

Control 31 ± 1.1

Hex modified 21 ± 1.2

Table 9 Comparison of calculated rate constants from Origin fits between control and hexenal modified HST at pH 5.6

115

At pH 7.4, there was no significant change or minimal change in rate constants at

both The N and C-sites of the control and hex modified sample of commercial

diferric HST. At pH 5.6, there was a decrease in C-site rate constant hex

modified sample as compared to the control sample of diferric HST.

3.6.2.3 Saturation Experiment

The aim of this experiment was to modify apo-HST with aldehydes (hex and 4-

ONE) followed by saturation with Fe to determine the effect of modification on

uptake and release of Fe. If there was a loss of one of the binding sites of apo-

HST due to aldehyde modification, then uptake experiments followed by

saturation with Fe, would result in the saturation of only one binding site and this

should be reflected in the kinetics of Fe release experiments.

Apo-HST was incubated with aldehydes (hex and 4-ONE) followed by saturation

with ferrous ammonium sulfate as described in the experimental section. A

control sample was also prepared with only apo-HST and run through the same

experimental conditions as the modified sample. Fe release kinetics were

determined on both control and modified diferric HST samples.

116

0 50 100 150 200 2500.615

0.620

0.625

0.630

0.635

0.640

0.645

0.650

0.655

Data: Data3_BModel: ExpDec2Equation: y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0Weighting: y No weighting Chi 2/DoF = 1.5561E-7R2 = 0.99436 y0 0.67183 ±0.00236A1 -0.01705 ±0.00029t1 4.23417 ±0.12932A2 -0.03608 ±0.00228t2 440.46851 ±38.19385

Abs

@ 4

80 n

m

Time (min)

Fig 72 Origin fit of Fe kinetics of release of control saturated apo-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4

117

0 50 100 150 200 2500.09

0.10

0.11

0.12

0.13

0.14

0.15

Data: Data3_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 7.4536E-7R 2 = 0.99675 y0 0.21313 ±0.00389A1 -0.1196±0.00379t1 418.7506 ±17.95216

Abs

@ 4

80 n

m

Time (min)

Fig 73 Origin fit of Fe kinetics of release of hex modified saturated apo-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4


Control 120 ± 24 2.4 ± 0.23

Hex modified n/a 2.3 ± 0.53

Table 10 Comparison of calculated rate constants of control and hex modified saturated apo-HST at pH 7.4

118

For pH 7.4, there was a clear difference between the control and hex modified

diferric HST. The control had the clear biphasic kinetics and displaying the

release from the two binding sites. On the other hand, the hex modified

saturated diferric HST clearly displayed a monophasic kinetics suggesting that

there maybe a loss of a binding site. Calculations of rate constants showed that

the calculated rate constant was for the C-site rate constant.

0 50 100 150 2000.095

0.100

0.105

0.110

0.115

0.120

0.125

0.130

0.135

0.140


Abs

@ 4

70 n

m

Time (min)

Fig 74 Origin fit of Fe kinetic of control saturated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6

119

0 50 100 150 2000.04

0.06

0.08

0.10

0.12

0.14


Abs

@ 4

70 n

m

Time (min)

Fig 75 Origin fit of Fe kinetics of release of hex-modified saturated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6


Control 26 ± 5

Sample 46 ± 4.1

Table 11 Comparison of calculated rate constants from control and hex-modified saturated apo-HST samples kinetics at pH 5.6

120

Comparing rate constants between control and hex-modified saturated apo-HST

samples, the results showed there were minimal changes in the rate constant of

site C in pH 5.6.

An analogous series of kinetic experiments were repeated using 4-ONE

modification of apo-HST, followed by saturation and kinetics of Fe release

experiments as described in the experimental section.

0 50 100 150 200 2500.06

0.07

0.08

0.09

0.10

0.11

Data: Data1_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 5.5758E-7R 2 = 0.99521 y0 0.11789 ±0.00041A1 -0.04647 ±0.00034t1 142.11641 ±2.64662

Abs

@ 4

80 n

m

Time (min)

Fig 76 Origin fit of Fe kinetics of release of 4-ONE modified saturated apo-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4

121


Control 120 ± 24 2.4 ± 0.23

4-ONE modified n/a 7.1 ± 3.5

Table 12 Comparison of calculated rate constants of control and 4-ONE modified saturated apo-HST at pH 7.4 At pH 7.4, there was a difference between the control and 4-ONE modified

diferric HST kinetics. The control displayed biphasic kinetics while the 4-ONE

modified displayed close to monophasic kinetics. This suggested that there was

loss of Fe binding at the N-site due to 4-ONE modification. The calculated rate

constants were comparable to the rate constants of the C-site. The data was

consistent with data of hex modified kinetics.

122

0 50 100 150 200

0.0410

0.0412

0.0414

0.0416

0.0418

0.0420

0.0422

0.0424

0.0426

0.0428

Data: Data1_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 2.6251E-8R2 = 0.79379 y0 0.04012 ±0.00064A1 0.00233 ±0.00061t1 305.79339 ±114.86835

Abs

@ 4

70 n

m

Time (min)

Fig 77 Origin fit of Fe kinetic of 4-ONE modified saturated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6


Control 26 ± 5

Sample 3.2 ± 6

Table 13 Comparison of calculated rate constants from control and 4-ONE modified saturated apo-HST samples kinetics at pH 5.6

123

At pH 5.6, there was significant difference between rate constants of control and

4-ONE modified diferric sample of the C-site.

3.6.2.4 Titration Experiment

The aim of the experiment was to titrate Fe into aldehyde-modified (hex and 4-

ONE) apo-HST followed by kinetic of release experiments to observe differences

in calculated rate constants and to confirm results obtained from the saturation

experiments. These were experiments also to give insights if there was a loss of

a binding site due to aldehyde modification. Control and aldehyde-modified (hex

and 4-ONE) apo-HST samples were titrated with Fe as described in the

experimental section, and kinetics of release were determined on titrated control

and aldehyde modified diferric HST.

124

0 50 100 150 200 2500.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

0.11

Data: Data3_BModel: ExpDec2Equation: y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0Weighting: y No weighting Chi 2/DoF = 1.0087E-7R2 = 0.9992 y0 0.10295 ±0.00013A1 -0.02935 ±0.00014t1 97.7261 ±1.54836A2 -0.0352±0.00023t2 9.71277 ±0.12382

Abs

@ 4

80 n

m

Time (min)

Fig 78 Origin fit of Fe kinetics of release of control titrated apo-HST using Trion as chelator and absorbance monitored at 480 nm at pH 7.4

125

0 50 100 150 200 250

0.060

0.062

0.064

0.066

0.068

Data: Data16_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 3.4735E-8R2 = 0.99276 y0 0.15712 ±0.04212A1 -0.09705 ±0.0421t1 503.76994 ±13.44128

Abs

@ 4

80 n

m

Time (min)

Fig 79 Origin fit of Fe kinetics of release of hex-modified titrated apo-HST run using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4

Sample KN (x10-3)) (min-1) KC (x10-3) (min-1)

Control 101 ± 2.8 9.6 ± 0.57

Hex modified n/a 2.5 ± 0.61

Table 14 Comparison of calculated rate constants of kinetics from control and hex-modified titrated apo-HST samples at pH 7.4

126

At pH 7.4, there were differences between the control and hex modified diferric

HST kinetics. The control displayed the biphasic kinetics as compared to the

monophasic kinetics of the hex modified diferric sample. The calculated rate

constant was comparable to those of the control C-site rate constant. There was

a 3-4 fold decrease in rate constants of the hex modified as compared to the

control diferric HST.

0 50 100 150 200

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

0.018

0.020

Data: Data20_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 5.2459E-8R 2 = 0.9966 y0 0.0022 ±0.00005A1 0.00597 ±0.00006t1 44.99099 ±0.69558Abs

@ 4

70 n

m

Time (min)

Fig 80 Origin fit of Fe kinetics of release of control titrated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6

127

0 50 100 150 2000.010

0.011

0.012

0.013

0.014

0.015

0.016

0.017

0.018

0.019

Data: Data16_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 3.4317E-8R2 = 0.9901 y0 0.01115 ±0.00003A1 0.00763 ±0.00005t1 51.09792 ±0.83102A

bs @

470

nm

Time (min)

Fig 81 Origin fit of Fe kinetics of release of hex-modified titrated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6


Control 23 ± 0.71

Hex modified 26 ± 6

Table 15 Comparison of calculated rate constants of control and hex-modified titrated samples from kinetics at pH 5.6

128

Kinetics of Fe release at pH 5.6 showed comparable rate constants of the control

and the hex modified diferric HST.

Analogous experiments were conducted using 4-ONE as the modifying aldehyde

followed by Fe titration and kinetics of release experiments as described in the


0 50 100 150 200 2500.049

0.050

0.051

0.052

0.053

0.054

0.055

0.056

Data: Data1_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi^2/DoF = 4.044E-8R^2 = 0.97821 y0 0.05947 ±0.00037A1 -0.00885 ±0.00035t1 319.48151 ±19.16059

Abs

@ 4

80 n

m

Time (min)

Fig 82 Origin fit of Fe kinetics of release run of 4-ONE modified titrated apo-HST using Tiron as chelator and absorbance monitored at 480 nm at pH 7.4

Sample KN (x10-3)) (min-1) KC (x10-3) (min-1)

Control 101 ± 2.8 9.6 ± 0.57

4-ONE n/a 3.1 ± 0.89

Table 16 Comparison of calculated rate constants of kinetics from control and 4-ONE modified titrated apo-HST samples at pH 7.4 At pH 7.4, there was a difference between the control and 4-ONE modified

kinetics of diferric HST. The control displayed the biphasic kinetics while the 4-

129

ONE modified displayed essentially monophasic kinetics. Rate constant for the

C-site showed approximately a 3-fold decrease due to modification.

0 50 100 150 200

0.0130

0.0135

0.0140

0.0145

0.0150

0.0155

Data: Data1_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi^2/DoF = 3.0735E-8R^2 = 0.85902 y0 0.01306 ±0.00005A1 0.00175 ±0.00004t1 79.03107 ±5.92683

Abs

@ 4

70 n

m

Time (min)

Fig 83 Origin fit of Fe kinetics of release of 4-ONE modified titrated apo-HST using EDTA as chelator and absorbance monitored at 470 nm at pH 5.6


Control 23 ± 0.71

4-ONE modified 13 ± 9

Table 17 Comparison of calculated rate constants of control and hex-modified titrated samples from kinetics at pH 5.6 At pH 5.6, there was an approximate 2-fold decrease in the rate of C-site Fe

release for the 4-ONE modified HST.

130


The first aim of this experiment was to incubate whole serum with aldehydes (hex

and 4-ONE) to determine if there would be differential changes in fluorescence

intensities of the various WSF protein components. The second aim was to try to

detect if there would be evidence for protein-protein crosslinking. Serum was

incubated with aldehydes (hex and 4-ONE) for 24 h at 37°C, followed by

extraction of the WSF and elution of extract through Sephadex G-75 column as

described in the experimental section. A control sample containing only serum

was also prepared and eluted through the same experimental conditions as the

aldehyde-modified samples.

0 50 100 1 50 2 00

0

1 000

2 000

3 000

4 000

5 000

6 000

Flur

esce

nce

Inte

nsity

@ E

mis

s 46

0 nm

V o lu m e (m l)

H ex M od ified S erum E xtrac t C on tro l

Fig 84 Fluorescence elution profile of hex modified WSF extract passed through the Sephadex G-75 column, monitored by fluorescence intensity at ex 350 nm and em 460 nm. Arrow shown the void volume of the column

131

Fig 85 SDS-PAGE image of fractions of hex modified peaks A’ and B’ and control peak A, eluted from the Sephadex G-75 column The elution profile from the Sephadex G-75 column (Fig 81) showed almost a 3-

fold increase of fluorescence intensity of peaks A and B as compared to peaks A

and B of the control sample. The SDS-PAGE gel showed that there was

formation of higher MW species possibly trimers of HSA (67 kDa band), as the

high MW species showed a MW of about 190 kDa. Peak B did not show any

appearance of high MW species as compared to peak A.

132

0 50 100 150 200 250 300 350

0

10000

20000

30000

40000

50000

4-ONE Modified Serum Extract Control

Time (min)

Fluo

resc

ence

Inte

nsity

@ E

mis

sion

460

nm

200

400

600

800

1000

1200

1400

1600

1800

2000

Fluorescence Intensity @ 460 nm

Fig 86 The elution profile of 4-ONE modified extract passed through the Sephadex G-75 column, monitored by fluorescence intensity at ex 350 nm and em 460 nm. Arrow shown void volume of the column The 4-ONE elution profile from the Sephadex G-75 column showed almost a 24-

fold fluorescence increase of peak A, and 20-fold fluorescence increase of peak

B, as compared to the control unmodified peaks A and B sample. The 4-ONE

elution profile showed a 9-fold fluorescence increase for peak A and a 10-fold

133

fluorescence increase of peak B, as compared peaks A and B from the hex

elution profile.

3.7 Subject Study

The aim was to conduct a pilot study on serum samples from six different elderly

subjects to observe if there would be variations of the component peaks of WSF

extract between them. WSF’s extract of the six subjects were passed through

Sephadex G-75 column, and the elution profile was monitored by fluorescence

and absorbance, and Origin program was used for the calculations of peak areas

for peaks A and B as described in the experimental section.

0 50 100 150 200

200

400

600

800

1000

1200

1400

1600

1800

2000

Data: Data1_BModel: LorentzEquation: y = y0 + (2*A/PI)*(w/(4*(x-xc)^2 + w^2))Weighting: y No weighting Chi^2/DoF = 686.59207R^2 = 0.99596 y0 294.50831 ±7.3581xc1 117.19933 ±0.25819w1 12.99369 ±0.87744A1 12325.96315 ±686.89547xc2 94.59057 ±0.06798w2 7.40521 ±0.28227A2 19689.12853 ±491.87924

Fluo

resc

ence

Inte

sity

@ E

mis

sion

460

nm

Volume (ml)

Fig 87 Representative example of an Origin fit of one of the subjects Sephadex G-75 column fluorescence elution profile where peak areas of peaks A and B were calculated

134

Subject 1

Subject 2

Subject 3

Subject 4

Subject 5

Subject 6

0 10 20 30 40 50 60 70 80

Peak Areas %

Peak B Peak A

Fig 88 Comparison of the six subjects based on peak areas percentages of peaks A and B The bar graph of the results from the study shows significant variations between

the different individual serum samples in the composition of peaks A and B.

135

4.0 Discussion

4.1 Separation and Characterization of WSF Extract Components

WSF was extracted from human serum following procedure published by

Tsuchida et al [31]. SDS-PAGE gel showed that WSF extract had multiple

protein bands present with major bands at 80, 67, 51 and 30 kDa (Fig 23). The

presence of proteins in the WSF extract suggested that they might be

contributors to the WSF, consistent with the literature that LP α, β unsaturated

aldehydes end-products could react with proteins, resulting in the formation of

fluorescent adducts. These fluorophores with suggestions from the literature

could induce structural and functional changes on the proteins that they react

with [10, 18, 21].

Fluorescence excitation and emission spectra of extract showed an excitation

maximum at 350 nm and emission maximum at 460 nm, which agreed with

literature values [23, 24, 26] (Fig 17). The observed excitation and emission

wavelengths of extract were also in agreement of literature values of in vitro

experiments of proteins and small amino acid analogues incubated with α, β

unsaturated aldehydes such as HNE, hex, octenal, and 4-ONE. These

aldehydes are end-products from LP process, and they are very reactive towards

biomolecules such as proteins. Literature in vitro experiments showed that there

was development of fluorescence on proteins as a result of incubation of

proteins, due to modification of certain amino acids residues with aldehydes,

particularly lysine residues [23, 26]. Fluorescence developed due to formation of

adducts as a result of a reaction between aldehydes and amino acid residues of

proteins. This led us to believe that the WSF extracted from human serum was

in fact a byproduct or marker of in vivo LP.

The next question was, what were the different components or contributors to

WSF? To separate the different components of WSF, extract was eluted through

a Sephadex G-75 gel filtration column as described in the experimental section.

The fluorescence elution profile of WSF extract from Sephadex G-75 column

136

showed three major peaks, labeled peaks A, B and C, according to their elution

order (Fig 16). The absorbance elution profile at 280 nm identified two peaks

coinciding with fluorescent peaks A and B (Fig 16). Peaks A and B eluted in the

high molecular weight region of column while peak C eluted in the low molecular

weight (included) region of column. The fluorescent and absorbance elution

profiles suggested that WSF was made up of two main components: a high

molecular weight protein-containing component (peaks A and B) and a low

molecular weight non-protein component (peak C). Excitation and emission

fluorescent spectra of the WSF components were consistent with what was

reported in the literature (Figs 16-20) [23, 24, 26].

4.1.1 Characterization of Peak C

Peak C was determined to be non-protein species as it did not show any

characteristic absorbance at 280 and 220 (data not shown) nm, which are the

two main absorptions corresponding to the aromatic residues in proteins and to

peptides, respectively. Peak C was also negative towards Bio-Rad protein

assay. Peak C also eluted consistently through the different runs of WSF extract

through Sephadex G-75 column, suggesting that it was an actual component of

the WSF. This suggested that peak C was the non-protein, low molecular

component of WSF. Fluorescence spectra of peak C exhibited an excitation at

350 nm and emission at 460 nm, similar to the literature values for the

fluorophore formation and similar to the high molecular species, suggesting that

the same fluorophore was present in peak C species [23, 24, 26].

To gain a better understanding of the nature of peak C species, several

experiments were conducted. The first experiment was to try to determine the

MW of the peak C species as this might give an insight of the identity of the

species. The P-2 gel filtration column was used to collect fractions of peak C,

followed by Speed Vac of samples to concentrate the fractions (Fig 40). To

determine the mass of peak C species, ESI/MS was used. The ESI/MS

spectrum of peak C showed two major peaks at 267.5 Da and 381.5 Da (Fig 41). The 267.5 Da was consistent with the MW of morpholine buffer components

137

used for the elution through the P-2 column. This suggested that morpholine

buffer wasn’t completely removed by the Speed Vac prior to the ESI-MS

analysis.

On the other hand, peak 381.5 Da suggested that it was the MW of the peak C

species. The 381.5 Da peak appeared consistently on multiple analysis by ESI-

MS. The MW did not give us any insights of what it could be. One possibility

that it could be a degraded peptide fragment of one of the modified serum

proteins that might contain the fluorophore. The other possibility that it could be

a phosopholipid molecule modified by LP’s aldehyde, as literature showed that

such modification could occur. One other possibility that could be an amine-

containing compound that might be modified and contained the fluorophore.

To gain a better understanding about the potential reactivity of peak C species

towards biomolecules particulary proteins, concentrated peak C was incubated

with commercial HSA. Control HSA and HSA-peak C sample, were eluted

through the Sephadex G-75 column. The elution profile from Sephadex G-75

column showed that peak C species was not a reactive species towards proteins

(Fig 42). This was demonstrated as peak C eluted at its normal low MW region

and HSA eluted in the high MW region (as well as the control HSA sample) of the

Sephadex G-75 column. This demonstrates that peak C was an unreactive

species towards proteins. This raised the possibility it could be a cleaved

fragment of a modified aminophospholipid that contained the fluorophore. Or it

could be a degraded fragment that contained the fluorophore. Or it could be an

end-product fluorophore of a low molecular weight amine-containing serum

component.

4.1.2 Reduction Experiments on WSF

The literature suggested that formation of adducts within proteins due to

modification with aldehydes could proceed through Schiff-base or Michael

addition reactions or both [23-26]. Schiff-base reactions are reversible reactions,

so if WSF was formed from a Schiff-base reaction intermediate without further

138

reaction then it would be a reversible intermediate and its fluorescence should be

altered if reduced. To address this Schiff base question, WSF extract was

reduced with NaBH4 followed by eluting through the Sephadex G-75 column.

The fluorescence elution profile from the Sephadex G-75 column did not show

differences from control samples (Fig 21). Calculated normalized fluorescence

units of reduced sample showed no difference before and after passage through

Sephadex G-75 column (Table 2). This suggested no loss of fluorescence after

reduction and eluting through the Sephadex G-75 column.

The elution profile from the Sephadex G-75 column showed that there was no

effect of reduction on the fluorescent properties of WSF’s components,

consistent with the absence of a Schiff-base adduct. This finding is consistent

with a fluorophore, which may have passed through a Schiff-base intermediate

and undergone subsequent transformations to a stable compound or product.

4.1.3 Covalent Binding of Fluorophore

The next question addressed was whether the fluorophore was covalently or

non-covalently bound to protein species of the protein component of the WSF.

The concentrate of peaks A and B was denatured with urea, followed by eluting

through the Sephadex G-75 column.

There was no difference in the fluorescence elution profile of control and urea

treated samples, as both eluted in the high MW region of Sephadex G-75 column

(Fig 22). These results indicate that the fluorophore is in fact covalently bound to

the proteins. This is due to fact that protein component of WSF was denatured

with urea and if fluorophore was non-covalently bound to protein component then

fluorescence elution profile should have shown appearance of fluorescence in

the low MW region and not the high MW region of the Sephadex G-75 column.

This agreed with the literature based on low molecular weight model compounds

that the formed fluorophore(s) were covalently bound [23-26].

139

4.2 Characterization of the Protein Component of WSF

Peak A eluted from the Sephadex G-75 column showed multiple protein bands

with major bands at 80, 67, 51, and 30 kDa (Fig 23). Peak B eluted from the

Sephadex G-75 column showed three protein bands with major bands at 80, 68

and 51 kDa (Fig 23). Excitation and emission spectra of peaks A (Fig 18) and

B (Fig 19) were similar to published spectra for aldehyde modified lysine

derivatives [18, 21, 24, 31].

The next step was to determine if there was one or multiple species in the protein

component that contained the fluorophore. This would give insights if aldehyde

modifications of proteins were indiscriminate or specific toward one species.

4.2.1 Isolation of Major Proteins Present in the WSF Protein Component

The major protein in peak B had a MW of 67 kDa. The first predication was that

this 67 kDa protein could be human serum albumin (HSA). HSA is the major

protein in human serum comprising 60% of the total protein composition and it

has a MW of 68 kDa. The 67 kDa protein was isolated by elution through

Sephadex G-75 (Fig 24) and purity was monitored by SDS-PAGE gel. SDS-

PAGE gels of concentrates of 67 kDa fractions showed that the 67 kDa protein

was isolated and purified to >98% purity (Fig 25, insert).

The next step was the protein identification of 67 kDa band. This was performed

by tryptic digestion of the isolated 67 kDa gel plug. Tryptic fragments of digested

67 kDa protein gel plug were analyzed using MALDI TOF 4700 Analyzer MS (Fig

27). Protein was identified by running a peptide mapping fingerprinting database

search, which confirmed with 100% confidence interval that the 67 kDa band was

HSA (Fig 27). HSA was identified with a high protein score, which was a good

parameter of protein identification using MS. The MW matched of that

determined from SDS-PAGE gels. This confirmed our hypothesis that the 67

kDa protein was HSA. This also showed that HSA was the major and the most

abundant species present in the protein component of WSF.

140

Emission spectra of the isolated HSA showed significant similarities to spectra

published in the literature of in vitro modified proteins with aldehydes, showing an

excitation maximum at 350 nm and emission maximum at 460 nm (Fig 25). The

emission spectrum of isolated HSA suggested that we were detecting the same

fluorophore(s) that was reported in literature [23-26].

As we identified HSA as the major and most abundant species present in the

protein component of WSF, the next question was: are there other protein

species that contained fluorescence or was all the fluorescence residing on

HSA? HSA was also the most abundant protein in peak A. In order to identify

other species that may contain the fluorescence, the first step was to eliminate

contaminated HSA from the rest of the protein species. This was achieved by

the use of Blue Sepharose CL-6B affinity column. This column has a high

capacity for absorbing HSA from mixtures of serum proteins to >95% removal. A

step gradient elution was used for elution of proteins and HSA as described in

the experimental section. The absorbance elution profile showed 2 peaks

correlating with 2 fluorescence peaks (Fig 28). The first peak eluting was the

run-through peak (A*) that did not containing HSA. The second peak, which is

eluted at a high ionic strength, was the HSA. SDS-PAGE gel of peak A* showed

>98 % HSA removal (Fig 29). The fluorescence elution profile suggested that

there were other proteins apart from HSA that contained fluorescence, i.e.,

multiple proteins present in WSF contained fluorescent adducts within their

structures. It also showed that more than 50% of the fluorescence was present

in peak A*. This showed that there were multiple protein species that contained

the fluorescence. This also showed that the modification of protein species was

at least somewhat indiscriminate and was not specific towards one protein.

Peak A* contained three major proteins with MW 80, 51 and 30 kDA. The goal

was to isolated individual proteins from this mixture and monitor the fluorescence

to give insights into what specific proteins contained the fluorophore. The 80 kDa

protein suggested that it could be human serum transferrin (HST). HST is one of

the more abundant serum proteins and has a MW of 80 kDa. If the 80 kDa band

could be confirmed to be HST, it would make isolation from other proteins easier

141

as there was a published procedure for purification of HST from other serum

proteins [69].

To confirm the presence of HST in peak A* from Blue Sepharose affinity column,

a western blot experiment was conducted on concentrate of peak A*. Western

blot experiment was conducted by using anti-HST antibody as described in the

experimental section. Western blot experiment confirmed the presence of HST

in peak A* for the 80 kDa band (Fig 30).

To isolate HST from the other 2 protein species, DEAE Sepharose ion-exchange

chromatography was used. The literature had shown that this column is very

effective on purifying HST from other serum proteins using a NaCl gradient.

Concentrates of peak A* from Blue Sepharose affinity column were passed

through a DEAE Sepharose column as described in the experimental section.

The absorbance elution profile showed 2 peaks with correlating fluorescence

peaks (Fig 31). The run-through peak (peak A1) contained 2 protein bands (55

and 30 kDa) as determined by SDS-PAGE gel (Fig 35). HST was eluted in the

second peak with a gradient of NaCl. The fluorescence elution profile

demonstrated that fluorescence was present in HST and peak A1. The majority

of fluorescence (64%) was present with HST. SDS-PAGE had shown that >98 %

pure isolated HST was obtained from DEAE Sepharose column (Fig 32, insert). Emission spectra of isolated HST (Fig 32) had shown similarity with isolated HSA

emission spectrum, suggesting that the same fluorophore(s) may be present in

both isolated proteins. This was significant in that it showed that two of the most

abundant proteins in serum were aldehyde modified. Both proteins have

important biological functions that, if altered, could result in biochemical or

metabolic disorders.

HST was further confirmed using MALDI TOF/MS as described in the

experimental section. HST gel plugs were trypsin digested followed by analysis

by 4700 Proteomics Analyzer MS (Fig 33). Peptide mapping fingerprinting

database search confirmed with 100% C.I. that the protein was HST (Fig 34).

142

The protein score was high and the MW weight matched that of HST. This

agreed with our hypothesis that the 80 kDa protein was HST.

HST and HSA proteins were the major protein species of WSF protein

component that contained fluorescence. This also showed that aldehyde

targeted the most abundant proteins in the serum and caused modification that

could have an effect on function. In order to determine differences and make a

comparison between the 2 isolated proteins, fluorescence units of both isolated

proteins were calculated (as described in the experimental section) per mg of

protein. Isolated HSA had 23 fluorescence units/mg protein and isolated HST

had 12 fluorescence units/mg protein. This suggested that HSA had almost

double the fluorescence units per amount of protein, and this would suggest that

HSA had more fluorescent fluorophore within its structure than HST. Both

proteins had about 10% lysine amino acids within their amino acids sequences.

We performed studies to identify possible lysine residues within HSA, and we

were able to identify around 11 lysine pairs with close distances that could result

in a fluorophore formation. With HST, we were not able to perform such

calculations, as there is no crystal structure for HST. But this could suggest that

HSA had more lysine pairs that could react with aldehyde to form the

fluorophore(s) than HST, i.e., more fluorophores were present in HSA than HST.

To identify the 2 protein species present in peak A1 from the DEAE Sepharose

column, a MALDI TOF/MS experiment was conducted. Gel plugs of the 2 protein

bands were tryptic digested followed by analysis using MALDI TOF/MS. The

peptide mapping fingerprinting database search identified with 99% C.I. Ig-G

protein for the 51 kDa (Figs 36, 37). Ig-G is also one of the abundant proteins in

serum. The protein score was not as high as HSA and HST, and this could be

due to the fact that the amount and concentration was low. This is one of the

major factors that influence the protein score.

For the 30 kDa protein, the peptide mapping fingerprinting database searches

identified carbonic anhydrase as the protein (Figs 38, 39). The protein score

143

was very low and C.I. was 0%. This could be attributed to the fact that carbonic

anhydrase was in low abundance and this could influence the score and C.I. as

mentioned earlier. But the MW matched with the predicted MW determined from

SDS-PAGE gels.

Fluorescence was present in both Ig-G and carbonic anhydrase and future work

should focus on further separation of these proteins and determination if

fluorescence would reside on one of them of both of them. Ion-exchange

chromatography could be a good separation technique for the separation. But

Ig-G or carbonic anhydrase or both of them are going to be minor species in the

protein component of WSF as the bulk and the majority of fluorescence was

present in HSA and HST.

4.3 Characterizations Studies on HSA and HST

The aim of these experiments was to digest isolated proteins into large peptide

fragments and to detect fragments that contained fluorescence. This would give

us information if there were a single or multiple modification sites within the

isolated proteins. This would provide an insight for future work that should focus

on modified fragments and try to digest them into smaller fragments and could

determine exactly where the fluorophore (s) present in the amino acid sequence

of proteins. This subsequently would lead to determination of the chemical

structure of the fluorophore.

4.3.1 Characterization of Isolated HSA

4.3.1.1 CNBr Digestion of Isolated HSA

CNBr digestion of isolated HSA produced the expected seven fragments (Fig 43) [66]. The Ulrogel AcA gel filtration column was used for the separation of the

CNBr peptide fragments. The absorbance elution profile had shown four peaks,

labeled peaks 1, 2, 3 and 4 according to elution order (Fig 44). The fluorescence

elution profile showed three peaks correlating with peaks 1, 3 and 4 absorbance

peaks. The fluorescence elution profile suggested there were multiple binding

144

sites of fluorophore within isolated HSA. Peak 1 from the elution profile had the

largest fluorescence intensity, and confirmed by SDS-PAGE that it was the 20

kDa peptide fragments (Fig 45). This agreed with our lysine residues study

where 4 out of the 11 lysine pairs were identified to be present in the 20 kDa

peptide fragment. Peaks 3 and 4 suggested that there were composed of

multiple peptide fragments bound together.

To gain more understanding and knowledge of what peptide fragments were

modified and contained the fluorophore (s), we used C18 reverse-phase HPLC.

The literature showed that HSA CNBr peptide fragments could be resolved by

HPLC using a C18 column, showing that five of the seven peptide fragments

could be completely resolved [67, 68]. A fluorescence detector was added to the

HPLC system so as to be able to monitor absorbance and fluorescence at the

same time. This would give accurate determination of what peptide fragments

contained the fluorescence intensity. The elution profile was comparable to one

published in the literature (Fig 46) [67, 68]. The fluorescence elution profile

showed multiple fragments containing fluorescence, suggesting also there were

multiple binding sites of fluorophore (Fig 47). The 20 kDa peptide fragment

showed also the highest fluorescence intensity.

Combining both elution profiles from the AcA column and HPLC, we predict that

there are at least 4 binding sites within HSA. The peptides fragments that

seemed to contain the fluorophore were the 20, 9.5, 4.0 and 3.4 kDa peptide

fragments. From our lysine pairs study, it showed that the 20 had 4 lysine pairs,

the 9.5 had 3 lysine pairs, the 4.0 had 2 lysine pairs, and 3.4 had 2 lysine pairs

(Table 3). Future work should focus on isolating these peptide fragments and

perform MALDI TOF/MS experiments coupled with trypsin digestion, to exactly

locate the position of the fluorophore in the amino acid sequence.

4.3.1.2 Isoelectric Focusing Experiments

The main aim of these experiments was to detect if shifts in pI shift due to in vitro

aldehyde modification of HSA. Our hypothesis was that if there was significant

145

lysine residues modification by aldehydes, then loss of those lysine residues

should have an effect on pI shifting HSA’s pI towards the acidic side. The IEF of

control HSA displayed the expected five isoforms of HSA (Fig 51). The IEF of

hex (Fig 52) and 4-ONE (Fig 53) modified HSA didn’t display clearly the isoforms

of HSA. But there were significant pI shift for both modified samples, but 4-ONE

sample showed more shift towards the acidic side. This could be true for 4-ONE

as it is predicated to be a more reactive aldehyde than hex. Approximation of the

pI for hex modified HSA was 5.50 and for 4-ONE was 5.32 as compared to HSA

pI of 5.62. The reason of not clearly visualizing the isoforms for both modified

samples could be that the aldehyde modification resulted in crosslinking of some

isoforms together. Future experiments should focus on performing MALDI

experiments on these isoforms and determine changes in MW as compared to a

control sample.

4.3.2 Characterizations of Isolated HST

4.3.2.1 CNBr Digestion of Isolated HST

Isolated HST was digested with CNBr producing the expected three fragments

[69]. One fragment had four peptide fragments bound together by disulfide

bonds with MWs ranging from 7-27 kDa. The second fragment had two peptides

bound together by disulfide bonds with MW 3 and 9 kDa. The third fragment

contained one peptide of MW 5.8 kDa.

Separations of peptide fragments were achieved by elution through the P-10

filtration column. The absorbance elution profile detected 3 peaks corresponding

to the three fragments described above, labeled peaks A, B and C according to

order of elution (Fig 54). The fluorescence elution profile showed only one peak

correlating with absorbance peak A. This suggested that there could be one or

multiple binding sites of fluorophore, as peak A was composed of 4 peptide

fragments joined by disulfide bonds. SDS-PAGE image of peak A showed the

expected four peptide fragments (Fig 55).

146

In order to determine if the fluorescence resided on one or multiple peptide

fragments in peak A, we needed to reduce the disulfide bonds and separated the

peptide fragments. Concentrated fractions of peak A from the P-10 column were

reduced and carboxyamidomethylated and then eluted through the P-10 column.

The absorbance elution profile showed three peaks, labeled peaks 1, and 2

according to the elution order (Fig 56). Peak 1 was composed of peptide

fragments 27 and 15 kDa and peak 2 was composed of peptide fragments 8 and

7 kDa. The fluorescence elution profile showed two peaks corresponding with

the two absorbance peaks (Fig 56). The fluorescence elution profile showed that

there were multiple binding sites of fluorophore within isolated HST. The levels

of fluorescence intensities were comparable in the 2 peaks. Future work should

focus on separating these fragments through the use of another gel filtration

column or ion-exchange column, and determine what peptide fragments

contained the fluorescence. This could lay the work for future MALDI

experiments to elucidate the exact position and MW of the fluorophore.

4.4 In Vitro MALDI Experiments on HSA and HST

Pollack’s work in our lab had demonstrated that HSA incubation in vitro with

aldehyde resulted in fluorescence development over the time of the incubation

[71]. The fluorescence reached its maximum after 48 h, where it reached a

plateau. If there was fluorophore formation resulting from covalent adducts,

then we should detect a significant shift in MW using MALDI TOF/MS. HSA

(Figs 48-50) and apo-HST (Figs 58-60) were incubated with aldehydes and were

analyzed by MALDI TOF/MS. For both proteins, significant MW increases were

detected with both aldehydes. These confirmed our hypothesis that aldehyde

modification resulted in an increase in MW due to fluorophore formation.

We estimated the number of aldehyde molecules that could be reacting based on

the MW difference between the control and modified sample. With 4-ONE

modified samples of HSA and HST, there were about comparable number of

molecules reacting (19 for HSA and 12 for HST). As for the hex modified

samples of HSA and HST, identical numbers of molecules were calculated (34

147

for HSA and 35 for HST). These results were not what we expect based on the

fact that 4-ONE was apparently more reactive as compared to hex. One

explanation could be that 4-ONE and hex generate different types of

fluorophores with different quantum yields. One useful future experiment is to

use HNE and 2-octenal and compare all the data and determine if a pattern

develops or not.

4.5 In Vitro Fluorescence Development Studies on HST

In vitro experiments were conducted on apo and diferric forms HST incubated

with aldehydes such as hex and 4-ONE. The main aim was to determine if there

will be protein fluorescence development with time as result of aldehyde

modification. This was based on work done by Pollack in our lab where she

determined that there was fluorescence development over time in HSA due to

aldehyde modification [71]. Fluorescence units time points of hex (Fig 61) and 4-

ONE (Fig 62) modified HST showed an increase in fluorescence in both apo and

diferric forms of HST. There was a larger fluorescence increase in the apo form

as compared with the diferric form with both aldehydes. This could be due to the

diferric form being in a more closed structural conformation and lysine residues

may not be surface accessible to react with aldehydes. For the apo form, it is

more of an open structural conformation of the protein, where the lysine residues

could be more accessible and can react readily with aldehydes. 4-ONE showed

a 3-fold fluoresence increase for both apo and diferric forms compared to the hex

sample.

The second aim of these experiments was to detect if there will be dimers or

trimers formation with time as a result of incubation of apo HST with aldehydes.

Samples were taken at different time points and were run on SDS-PAGE gel to

detect changes. SDS-PAGE gels had showed that there was indeed dimer and

possibly trimer formations of crosslinked HST as a result of incubation of apo

HST with both hex and 4-ONE aldehydes. This was significant because it

demonstrated that protein crosslinking (inter or intra) existed as a result of

148

aldehyde modification that resulted in the formation of dimers and possibly

trimers.

4.6 In Vitro Functional Studies on HST

4.6.1 Iron Uptake by Apo HST

The aim of these experiments was to address the following question: does

aldehyde modification of apo HST have an effect on the affinity or binding

capacity for iron? Apo-HST was aldehyde-modified (hex and 4-ONE), followed

by Fe titration, monitoring absorbance at 465 nm as described in the

experimental section. Absorbance changes at 465 nm were used to measure Fe

binding by apo-HST. Both hex and 4-ONE iron uptake plots demonstrated

almost a 50% decrease in absorbance per mg protein compared to the control

unmodified sample. The control unmodified apo-HST showed an equivalence

point of about 2, illustrating the two Fe binding sites of apo-HST. For the hex

(Fig 64) and 4-ONE (Fig 65) modified apo-HST, there was a decrease of the

equivalence point to 0.73 and 1.13 respectively. This suggested that there might

be a loss of an Fe binding site due to aldehyde modification or the binding

capacity of apo-HST was decrease by 50% for both binding sites. This led to the

next question that we needed to address: which of the two sites has lost it

binding as a result of aldehyde modification, or are both sites affected equally in

loss of their binding capabilities?

If there was a discriminating loss of one of the two sites, we postulated it might

be the N-site. This hypothesis was based on fact that the N-site functions by

what is termed as the “dilysine trigger” (as mentioned previously in Chapter 1)

[58, 59]. Two lysines within a close proximity and are responsible for the binding

of an iron. If one or both of these lysines became modified with aldehydes, then

apo-HST would be expected to lose Fe binding capability at the N-site. This

would subsequently affect the kinetics of release of iron from HST as the release

of Fe would be mainly attributed to the C-site. Experimentally, in order to prove

149

the N-site hypothesis, we had to perform experiments monitoring kinetics of iron

release from aldehyde modified versus unmodified diferric HST.

4.6.2 Fe Kinetics of Release from HST

Experiments were conducted to measure the Fe kinetics of release from HST as

described in the experimental section. Commercial diferric-HST experiments

optimized the experimental conditions and the calculated rate constants of both

sites were comparable with literature values [61]. Then commercial diferric-HST

was modified with hex to observe changes in rate constants due to aldehyde

modification. At pH 7.4, there were no changes in the rate constants between

the control and hex modified diferric HST for both the N-and the C-sites. This

could be due to the fact that the diferric HST was already in the closed

conformation and both irons were bound, so modification didn’t have an effect as

the susceptible (lysines) residues were buried or had diminished nucleophilicity

due to binding to other amino acid residues. At pH 5.6, there was a decrease in

the rate constants of the C-site compared to the control sample. This suggested

that aldehyde modification had an effect on kinetics of release of Fe from C-site.

This agreed with the fluorescence development experiments where we detected

fluorescence development over time of diferric HST when incubated with the

aldehydes. It showed that there was fluorophore formation and this could have

an effect on the kinetics of release in this experiment.

4.6.2.1 Saturation and Titration Experiments

To prove our previous hypothesis that the N-site binding capability was lost due

to aldehyde modification, saturation and titration experiments were performed on

aldehyde-modified apo-HST followed by monitoring of Fe kinetics of release.

Apo-HST was modified with hex and 4-ONE followed by either saturation or

titration experiments. Both the saturation and the titration experiments were

conducted on control unmodified and aldehyde-modified apo-HST, followed by

measurements of the kinetics of Fe release to calculate the rate constants of

both binding sites. Two different methods of binding iron to control and

150

aldehyde-modified apo-HST were used so as to obtain more reliable data as both

methods should yield similar or comparable results.

For the data analysis, our main focus would be on pH 7.4 kinetics and rate

constants as at pH 5.6 the N-site rate constant cannot be calculated accurately

as it is too fast to measure with the available instrument as mentioned previously.

Plots of control saturated and titrated apo-HST displayed biphasic kinetics due to

the release of Fe from both sites. If our hypothesis was correct or valid, then for

the aldehyde-modified saturated or titrated apo-HST we should observe

predominantly monophasic kinetics due to the loss of one of the binding sites.

Data and plots of kinetics runs of aldehyde-modified (hex and 4-ONE) saturated

and titrated apo-HST at pH 7.4 had shown different kinetic patterns as compared

to control. There was the disappearance of the biphasic kinetics as detected with

the control and the appearance of monophasic kinetics. The calculated rate

constants were comparable to the values of the C-site calculated for the control

unmodified sample. There were decreases in the rate constants detected for the

C-site in the titration experiments data for both the hex and 4-ONE, as opposed

to the saturation experiments where the rate constants for the control and

aldehyde modified samples were comparable.

On the other hand, data from both the saturation and the titration experiments

suggested that there was the loss of the N-site due to aldehyde modification at

pH 7.4. These data suggested that our hypothesis was correct and valid that the

binding site lost to aldehyde modification was the N-site. This also suggested

that the 2 lysine residues that act as the “dilysine trigger” for Fe release and

binding were somehow being aldehyde modified and this resulted on the loss of

the binding capability at the N-site.

At pH 5.6, there was a drastic decrease in the rate constants for the C-site

between the control unmodified and the 4-ONE modified HST for the saturation

experiments. The kinetic time course also was very noisy and this could have an

effect on the rate constants calculations, and it was reproducible on all kinetic

measurements. But it could also mean that 4-ONE modification had an effect on

151

the kinetics of release from the C-site. This was not noticeable with the hex

modified sample as the rate constants for the control and hex modified samples

were comparable. This was also detected in the titration experiments where the

hex modified samples didn’t show decrease in the rate constants for the C-site,

but the 4-ONE modified samples showed decrease. Comparing both the

saturation and the titration experiment data for the 4-ONE, it suggested that 4-

ONE had also an effect on the kinetics of Fe release from the C-site at pH 5.6.

The saturation and the titration experiments results were comparable on showing

the loss of N-site binding site at pH 7.4. This could be significant as this may

cause the development of chronic iron disorder as a result of the loss of the N-

site binding capability. There were differences concerning the C-site rate

constants as the titration experiments showed that there were also decreases in

the rate constants of the C-site at both pHs. This could also be valid as 4-ONE is

a very reactive aldehyde and could also induced modifications in the C-site that

could affect the kinetics of Fe release. A future experiment to try to better

understand the extent of modification of C-site, would be to use HNE and octenal

as modifying aldehydes and determine from the data obtained from all these four

aldehydes experiments the effect of modification of the C-site. On the other

hand, to reinforce our hypothesis there was a loss of N-site Fe binding

capabilities due to aldehyde modification, future experiments should focus on the

use of stopped flow experiments to be used at pH 5.6, to corroborate loss of N-

site binding capability at pH 5.6.

152


The main aim of these experiments was to modify whole serum with aldehydes

and determine changes if any to the fluorescence properties of peaks A and B,

and if there would be dimers or trimers formations as a result of aldehyde

modification. The elution profiles showed a large increase in fluorescence

intensities of peaks A and B as compared with control with both aldehydes (hex

and 4-ONE), with 4-ONE showing larger increase in fluorescence intensity than

hex. 4-ONE modified sample showed a 24-fold fluorescence increase and the

hex modified sample showed a 3-fold fluorescence increase, of peaks A and B

compared to their respective control samples. This suggests that modifying

whole serum may result in a large number of random cross-linking reactions,

resulting in an increase in the number of fluorophore generated.

An SDS-PAGE gel (Fig 85) of hex incubated serum had shown that there was

indeed formation of dimers and possibly trimers of HSA due to either intra or inter

crosslinking of proteins as a result of aldehyde modification. The cross links

were detected only in peak A and was absent from peak B. Future experiments

should focus on the identifications of the proteins that were crosslinking and this

could be done by MALDI TOF experiments where gel plugs of the crosslinks

could be analyzed to identify the proteins involved in crosslinking.

153

4.8 Subject Study

This was a pilot study where the main aim of the experiment was to observe if

there are variations between different elderly subjects on peaks A and B

fluorescence properties. This could give insights to if these WSFs were the

same in all subjects or if there were major differences in component peaks. The

study was conducted on serum samples from six different aging subjects. The

elution profiles from the Sephadex G-75 were fitted using Origin software where

peaks areas percentages of peaks A and B were calculated, as described in the

experimental section. Data showed that there were indeed significant variations

between subjects on peaks A and B fluorescence composition (Fig 88). For

example, subject 1 had 60% peak A and 40% peak B, while subject 6 had 25%

peak A and 75% peak B. This showed that within aged subjects there were

major differences between peaks A and B. Future experiments may include

performing a study of younger subjects and compare them with old age subjects,

and determine if age correlates with peaks A and B fluorescence properties.

Another experiment may include subjects at the different disease stages as

compared to control samples. This also raises the question of the origin of these

pattern differences and whether changes in A:B ratio reflect or correlate with

specific disease states of oxidative stress status.

154

4.9 Is WSF a Reflection of In Vivo LP?

Experiments conducted on this dissertation agreed with the conclusion that WSF

was a reflection of in vivo LP. The WSF components showed similar

fluorescence properties as in vitro reactions fluorescence properties conducted

on proteins and amino acid analogues. The fluorescence properties determined

on isolated HSA and HST agreed with the literature fluorescence properties of in

vitro aldehyde modified proteins. We did identify at least two major and

abundant proteins that are present in high protein amounts and are more

susceptible to aldehyde modification than other less abundant proteins. This

could be significant because WSF presence in these major serum proteins could

be used as a biomarker for LP associated diseases.

All rights reserved © 2005

Ashraf H. Elamin

155

5.0 References

1. Wickens P. A. (2001). Ageing and the free radical theory. Respiration

Physiology. 128, 379-391.

2. Picklo, MJ, Montine, TJ, Amarnath V, and Neely, MD. (2002). Carbonyl

toxicology and Alzheimer’s disease. Toxicology and Applied Pharmacology. 184

(3), 187-197.

3. Oberley, D. T., Toyokuni, S., and Szweda, I. L. (1999). Localization of

Hydoxynonenal Protein Adducts in Normal Human Kidney and Selected Human

Kidney Cancers. Free Radical Biology & Medicine. 27 (5/6), 695-703.

4. Huang, X., Atwood, C., Hartshorn, M. et al. (1999). The Aβ Peptide of

Alzheimer’s Disease Directly Produces Hydrogen Peroxide through Metal Ion

Reduction. Biochemistry. 38 (24), 7609-7616.

5. Gutteridge, J.M.C. and Halliwell, B. (1990). The measurement and

mechanism of lipid peroxidation in biological systems. TIBS. 15, 129-135.

6. Sevanian, A. and Ursini, F. (2000). Lipid Peroxidation in Membranes and

Low-Density Lipoproteins: Similarities and Differences. Free Radical Biology &

Medicine. 29 (3/4), 306-311.

7. Calingasan, N., Uchida, K., and Gibson, G. (1990). Protein-bound

acrolein: A novel marker for oxidative stress in Alzheimer’s disease. J

Neruochem. 72, 751-756.

8. Uchida, K. (2000). Role of Reactive Aldehyde in Cardiovascular Diseases.

Free Radical Biology & Medicine. 28 (12), 1685-1696.

9. Kato, Y., Mori, Y., Makino, Y. et al. (1999). Formation of Nε-

(Hexanonyl)lysine in Protein Exposed to Lipid Hydroperoxide. The Journal of

Biological Chemistry. 274 (29), 20406-20414.

156

10. Xu, G. and Sayre, M. L. (1999). Structural Elucidation of a 2:2 4-

Ketoaldehyde-Amine Adduct as a Model for Lysine-Directed Cross-Linking of

Proteins by 4-Ketoaldehydes. Chem. Res. Toxicol., 12, 862-868.

11. Keidar, S., Rosenblat, M., Fuhrman, B. et al. (1992). Involvement of the

macrophage LDL receptor-binding domains in the uptake of oxidized LDL. J Lipid

Res. 12, 484-493.

12. MacNamara, J.R., Small, D.M., Li, Z., et al. (1996). Difference in LDL

subspecies involves alterations in lipid composition and conformational changes

in apolipoprotein B. J Lipid Res. 37, 1924-1935.

13. Chi Chen, G., Liu, W., Allaart, J., et al. (1994). Conformation differencesin

human apolipoprotein B-100 among subspecies of low density lipoproteins (LDL).

J Biol Chem. 269, 29121-29128.

14. Mark, R.J., Lovell, M.A., Uchida, K., et al. (1997). A role for 4-

hydroxynonenal, an aldehydic product of lipid peroxidation, in disruption of ion

homeostasis and neuronal death induced by amyloid beta-peptide. J Neurochem.

68, 255-264.

15. Markesbery, W., and Lovell, M. (1998). Four-hydroxynonenal, a product of

lipid peroxidation, is increased in the brain in Alzheimer’s disease. Neurobiol

Aging. 19, 33-36.

16. Mitchell, D., and Peterson, D. (1987). The oxidation of unsaturated

aldehydic products in lipid peroxidation by rat liver aldehyde dehydrogenases.

Toxicol. Appl. Pharmacol. 87, 403-410.

17. Montine, K.S., Kim, P.J., Olson, S.J., et al. (1997). 4-hydroxy-2-nonenal

pyrrole adducts in human neurodegenerative disease. J. Neuropathol. Exp.

Neurol. 56, 866-871.

18. Hidalgo, J. F., Alaiz, M., and Zamora, R. (2001). Pyrrolizaiton and

Antioxidant Function of Proteins Following Oxidative Stress. Chem. Res. Toxicol.

14, 582-588.

157

19. Friguet, B., Stadtman, R. E., and Szweda, I. L. (1994). Modificaiton of

Glucose-6-phosphate Dehydrgonanse by 4-Hydroxy-2nonenal. The Journal of

Biological Chemistry. 269 (34), 21639-21643.

20. Lightning, L., Jones, J., Friedberg, T. et al. (2000). Mechanism-Based

Inactivation of Cytochrome. Biochemistry. 39, 4276-4287.

21. Korotchkina, L., Yang, H., Tirosh, O., Packer, L., and Patel, M. (2001).

Protection by Thiols of the Mitochondrial Complexes from 4-Hydroxy-2-Nonenal.

Free Radical Biology & Medicine. 30 (9), 992-999.

22. Xu, G., Liu, Y., and Sayre, L. (1999). Independent Synthesis, Solution

Behavior, and Studies on the Mechanism of Formation of a Primary Amine-

Derived Fluorophore Representing Cross-linking of Proteins by (E)-4-Hydroxy-2-

noneal. J. Org. Chem. 64, 5732-5745.

23. Doorn, J. and Petersen, D. (2002). Covalent Modification of Amino Acid

Nucleophiles by the Lipid Peroxidation Products 4-Hydroxy-2-nonenal and 4-

Oxo-2-noneanl. Chem. Res. Toxicol. Revised.

24. Tsai, I., Szweda, P., Vinogradova, O., and Szweda, L. (1998). Structural

characterization and immunochemical detection of a fluorophore derived from 4-

hydroxy-2-nonenal and lysine. Proc. Natl. Acad. Sci. 7975-7980.

25. Itakura, K., Osawa, T., and Uchida, K. (1998). Structural of a Fluorescent

Compound Formed from 4-Hydroxy-2-nonenal and Nα-Hippuryllysine: A Model

for Fluorophores Derived from Protein Modifications by Lipid Peroxidation. J.

Org. Chem. 63, 185-187.

26. Uchida, K., Kanematsu, M., Sakai, K., et al. (1998). Protein-bound

acrolein: Potential markers for oxidative stress. Biochemistry. 95 (9), 4882-4887.

27. Guichardant, M., Taibi-Tronche, P., Fay, LB, and Lagarde, M. (1998).

Covalent modifications of aminophospholipids by 4-hydroxynonenal. Free Radic

Biol Med. 25 (9), 1049-1056.

158

28. Benedetti, A., Comporti, M., and Esterbauer, H. (1980). 4-hydroxynonenal

as a cytotoxic product originating from the peroxidation of liver microsomal lipids.

Biochim. Biophys. Acta. 620, 281-296.

29. Porter, N.A., Caldwell, S.E., and Mills, K.A. (1995). Mechanisms of free

radical oxidation of unsaturated lipids. Lipids. 30, 277-290.

30. Kawal, Y., Furuhata, A., Toyokuni, S., Yasyaki, A., and Uchida, K. (2003).

Formation of acrolein-derived 2’-deoxyadenosine adduct in an iron-induced

carcinogenesis model. Journal of Biol Chem. Manuscript M309057200.

31. Tsuchida, M., Miura, T., Mizutani, K., and Aibara, K. (1985). Fluorescent

substances in mouse and human sera as a parameter of in vivo lipid

peroxidation. Biochemica et Biohysica Acta. 834, 196-204.

32. Seybert, D.W. Unpublished data.

33. Curry, S., Brick, P., and franks, N. (1999). Fatty acid binding to human

serum albumin: new insights from crystallographic studies. Biochemica et

Biohysica Acta. 1441, 131-140.

34. Ashbrook, J.D., Spector, A.A., and Fletcher, J.E. (1972). Medium chain

fatty acid binding to human plasma albumin. J. Biol. Chem. 247, 7030-7042.

35. Bhattacharya, A., Grüne, T., and Curry, S. (2000). Crystallographic

Analysis Reveals Common Modes of Binding of Medium and Long-chain Fatty

Acids to Human Serum Albumin. J. Mol. Bol. 303, 721-732.

36. Eckenhoff, R. G. (1996). Amino acid resolution of halothane binding sites

in serum albumin. J Biol Chem. 271, 15521-26.

37. Berde, C.B., Hudson, B.S., Simoni, R.D., et al. (1979). Human serum

albumin. Spectroscopic studies of binding and proximity relationships for fatty

acids and bilirubin. J. Biol. Chem. 254, 391-400.

38. Brown, J.R., and Shockley, P. (1982). Serum albumin: structure and

characterization of its ligand binding sites. In Lipid-Protein Interactions (Jost, P.

and Griffith, O.H., eds), vol. 1, pp. 26-68, Wiley and Sons, New York.

159

39. Elmadhoun, B. M., Wang, G.Q., Templeton, J.F. et al. (1998). Binding [3H]

palmitate to BSA. Amer Journal of Physiology. 275, G638-44.

40. Wilton, D.C. (1990). The fatty acid analogue 11-(dansylamino)undecanoic

acid is a fluorescent probe for the bilirubin-binding sites of albumin. Biochem

Journal. 270, 163-66.

41. Curry, S., Mandelkow, H., Brick, P., et al. (1998). Crystal structure of

human serum albumin complexed with fatty acids reveals an asymmetric

distribution of binding sites. Nature Struct. Biol. 5, 827-835.

42. Hamilton, J.A., Era, S., Reed, R.G., et al. (1991). Locations of the three

primary binding sites for long-chain fatty acids on bovine serum albumin. Proc.

Natl. Acad. Sci. USA. 88, 2051-2054.

43. Koh, S.W. and Means, G.E. (1979). Characterization of a small apolar

anion binding site of human serum albumin. Arch. Biochem. Biophys. 192, 73-79.

44. Alaiz, M., Barragan, S. (1995). Changes induced in bovine serum albumin

following interactions with the lipid peroxidation product E-2-octenal. Chem

Physics Lipids. 77, 217-23.

45. Hidalgo, F.J., and Zamora, R. (2000). Modification of Bovine Serum

Albumin Structure following Reaction with 4,5(E)-Epoxy-2(E)-heptenal. Chem

Res Toxicol. 13, 501-8.

46. MacGillivray, R.T., Mendez, E., Shewale, J.G., et al. (1983) The primary

structure of human serum transferrin. J Biol Chem. 258(6), 3543-53.

47. Li, Y., Harris, W.R., Maxwell, A., et al. (1998). Kinetics Studies on the

Removal of Iron and Aluminum from Recombinant and Site-Directed Mutant N-

Lobe Half Transferrins. Biochemistry. 37, 14157-66.

48. He, Q., Mason, A.B., and Woodworth, R.C. (1997). Iron release from

recombinant N-lobe and single point Asp63 mutants of human transferrin by

EDTA. Biochem J. 328, 439-45.

160

49. Baldwin, D.A., Deolinda M.R., Rainer M.A. (1982). The Effect of pH on the

Kinetics of Iron Release From Human Transferrin. Biochim Biophysi Acta. 719,

140-46.

50. Bailey, S., Evans, R.W., Garratt, R.C., et al. (1988). Molecular Structure of

Serum Transferrin. J Biol Chem. 27, 5804-12.

51. Pyrz, J.W., Roe, A.L., Stern, L.J., et al. (1985) Model Studies of Iron-

Tyrosinate Proteins. J Am Chem Soc. 107, 614-20.

52. Nurizzo, D., Baker, H.M., He, Q., et al. (2000). Crystal Structures and Iron

Release of Mutants (K206A and K296A) That Abolish the Dilysine Ineraction in

the N-lobe of Human Transferrin. Biochemistry. 40, 1616-23.

53. Chahine, J.M., and Fain, D. (1994). The mechanism of iron release from

transferrin. Europ J Biochemistry. 223(2), 581-7.

54. Berry, L.R., and Hatton, M.W. (1986). A comparison of two procedures

used for complexing Fe(III) with human apotransferrin. Biochem Cell Biol. 64(9),

936-45.

55. Hatton, M.W., and Berry, L.R. (1987). A comparison of two procedures

used for complexing Fe(III) with human apotransferrin. II. 65(4), 271-9.

56. Baldwin, D.A. (1979). The Kinetics of Iron Release from Human Serum

Transferrin by EDTA. Biochim Biophysic Acta. 623, 183-98.

57. Marques, F.M., Watson, D.L., and Egan, T.J. (1990). Kinetics of Iron

Removal from Human Serum Monoferric Transferrins by Citrate. Inorg Chem. 30,

3758-62.

58. He, Q., Mason, A.B., Tam, B.M., et al. (1999). Dual Role of Lys206-

Lys296 Interaction in Human Serum Transferrin N-Lobe. Biochemistry. 38, 9704-

11.

59. Dewan, J.C., Mikami, B., Hirose, M., et al. (1993). Structural Evidence for

a pH-Sensitive Dilysine Trigger in the Hen Ovotransferrin N-Lobe. Biochemistry.

32, 11963-68.

161

60. Adams, T.E., Mason, A.B., He, Q., et al. (2003) The position of Arginine

124 Controls the Rate of Iron Release from the N-Lobe of Human Serum

Transferrin. J Biol Chem. 278(8), 6027-33.

61. Mason, A.B., He, Q., Halbrooks, P.J., et al. (2002). Differential Effect of a

His Tag at the N-and C- Termini of Recombinant Human Serum Transferrin.

Biochemistry. 41, 9448-54.

62. Liu, R., Guan, J., Zak, O., et al. (2003). Structural Reorganization of the

Transferrin C-Lobe and Transferrin Receptor upon Complex Formation.

Biochemistry. Accelerated Publications.

63. Halbrooks, P.J., He, Q., Briggs, S., et al. (2003). Investigation of the

Mechanism of Iron Release from the C-Lobe of Human Serum Transferrin.


64. Laemmli, U.K. (1970). Nature. 227, 80.

65. McMenamy, R.H., Dintzis, H.M., and Watson, F. (1971). Cyanogen

Bromide Fragments of Human Serum Albumin. J Biol Chem. 246, 4744-50.

66. Chuang,V.T., Kuniyasu, A., Nakayama, H. et al. (1999). Helix 6 of

subdomain III A of human serum albumin is the region primarily photolabeled by

ketoprofen. Biochim Biophys Acta. 1434, 18-30.

67. Campagnini, A., Foti, S., Salett, R. et al. (1998). Characterization of

Cyanogen Bromide Fragments of Reduced Human Serum Albumin by MALDI

MS. J Mass Spectrometry. 33, 673-76.

68. Iadarola, P., Ferri, G., Galliano, M., et al. (1984). Separation of cyanogen

bromide fragments from normal and abnormal human serum albumin by

reversed-phase HPLC. J Chromatogr. 298(2), 336-44.

69. Sutton, M.R., and Brew, K. (1973). Purification and Characterization of the

Seven Cyanogen Bromide Fragments of Human Serum Transferrin. Biochem J.

139, 163-8.

162

70. MacGillivray, R.T., Mendez, E., Shewal, J.G., et al. (1983). Primary

Structure of Human Serum Transferrin. J Biol Chem. 258(6), 3543-53.

71. Pollack, A., Seybert, D.W. Unpublished data.

72. Harrap, K.R., Jackson, R.C., Riches, P.G., et al. (1973). Biochim. Biophys.

Acta. 310, 104.

73. Modig, H.G. (1973). Biochem. Pharmacol. 22, 1623.

74. Lund, M., Bjerrum, O.J., and Bjerrum, M.J. (1999). Structural

heterogeneity of the binding sites of HSA for phenyl-groups and medium-chain

fatty acids. Eurp J Biochem. 260, 470-76.

75. Dasgupta, A., and Zdunek, T. (1992). In Vitro Lipid Peroxidation of Human

Serum Catalyzed by Curpric Ion. Life Sciences. 50, 875-82.

76. Lee, C., Yim, M.B., Chock, P.B., et al. (1998). Oxidation-Reduction

Properties of Methyglyoxa-modified Protein in Relation to Free Radical

Generation. J Biol Chem. 273(39), 25272-78.

77. Southwell, J.L., Yeargans, G.S., Kowalewski, C., et al. (2002). Acrolein

modifies and inhibits cytosolic aspartate aminotransferase. J Enzyme Inhib Med

Chem. 17, 19-23.

78. Glomb, M.A., and Monnier, V.M. (1995). Mechanism of Protein

Modification by Glyoxal and Glycolaldehyde. J Biol Chem. 270(17), 10017-26.

79. Belguendouz, L., Fremont, L. and Gozzelino, M.T. (1998). Interaction of

transresveratrol with plasma lipoproteins. Biochem Pharmacol. 55(6), 811-6.

80. Hotta, H., Sakamoto, H., Nagano, S., et al. (2001). Unusually large

numbers of electrons for the oxidation of polyphenolic antioxidants. Biochim

Biophysi Acta. 1526, 159-67.

81. Surgano, M., Sawada, S., Tsuchida, K., et al. (2000). Low density

lipoproteins develop resistance to oxidative modification. J Lipid Res. 41, 126-33.

82. Hofstee, B.H. (1973). Hydrophobic Affinity Chromatography of Proteins.

Anal Biochem. 52, 430-48.

163

83. Hartley, D.P., Kroll, D.J., and Peterson, D.R. (1997). Prooxidant-Initiated

Lipid Peroxidation in Isolated Rat Hepatocytes. Chem Res Toxicol. 10, 895-905.

84. Refsgaard, H.H., Tsai, L., and Stadtman, E.R. (2000). Modifications of

proteins by polyunsaturated fatty acid peroxidation products. Proc Natl Acad Sci

USA. 97(2), 611-6.

85. Klein, E.J., Rivera, S.L., and Porter, J.E. (2000). Optimization of ion-

exchange protein separations using a vector quantizing neutral network.

Biotechnol Prog. 16(3), 506-12.

86. Uchida, K., Khor, O.T., Oya, T., et al. (1997). Protein modification by a

Maillard reaction intermediate methylglyoxal. FEBS Letters. 410, 313-18.

87. Gree, R., Tourbah, H., and Carrie, R. (1986). Fumaraldehyde

monodimethyl acetal. Tetrahedron Letters. 27(41), 4983-86.

88. Hazen, S.L., Gaut, J.P., Hsu, F.F., et al. (1997)

Hydroxyphenylacetaldehyde Covalently Modifies ε-Amino Groups of Protein

Lysine Residues. 272(27), 16990-98.

89. Duda, D., Tu, C., Qian, M., et al. (2000). Structural and kinetic Analysis of

the Chemical Rescue of the Proton Transfer Function of Carbonic Anhydrase II.


90. Hoff, H.F., Whitaker, T.E., and O’Neil, J. (1992) Oxidation of Low Density

Lipoprotein Leads to Particle Aggregation and Altered Marcrophage Recognition.

J Biol Chem. 267, 602-9.

91. Steinbrecher, U.P. (1986) Oxidation of Human Low Density Lipoprotein

Results in Derivatization of Lysine Residues of Apolipoprotein B by Lipid

Peroxide Decomposition Products. J Biol Chem. 270, 3603-8.

92. Purcell, M., Neault, JF., and Tajmir-Riahi, HA. (2000) Interaction of taxol

with human serum albumin. Biochim Biohys Acta. 1478 (1), 61-68.

164

93. Watanabe, S. and Sato, T. (1996) Effects of Free Fatty Acids on the

Binding of Bovine and Human Serum Albumin with Steroid Hormones. Biochim

Biohys Acta. 1289 (3), 385-395.

94. Yong, PM. and Wheat TE. (1990) Optimization of High-Performance

Liquid Chromatographic Peptide Separations with Alternative Mobile and

Stationary Phases. J Chromatogr. 512, 273-281.

95. Mant, CT. and Hodges RS. (1985) General Method for the Separation of

Cyanogen Bromide Digests of Proteins by High-Performance Liquid

Chromatography. Rabbbit Skeletal Troponon I. J Chromatogr. 326, 349-356.

96. Berde, CB., Hudson, BS., Simoni, RD. and Sklar, LA. (1979) Human

Serum Albumin. Spectroscopic Studies of Binding and Proximity Relationships

for Fatty Acids and Bilirubin. J Biol Chem. 254 (2), 391-400.

97. Minchiotti, L., Kragh-Hansen, U., Nielsen, H. et al. (1999) Structural

Characterization, Stability and Fatty Acid-Binding Properties of Two French

Genetic Variants of Human Serum Albumin. Biochimica et Biophysica Acta.

1431, 223-231.

98. Iadarola, P., Zapponi MC., Stoppini M. et al. (1988) High-performance

Liquid Chromatography of Complex Mixtures of Cyanogen Bromide-produced

Peptides from Different Proteins. J Chromatogr. 443, 317-328.

99. Kragh-Hansen, U., Pedersen, AO., Galliano, M. et al. (1996) High-affinity

Binding of Laurate to Naturally Occuring Mutants of Human Serum Albumin and

Proalbumin. Biochem J. 320 (Pt 3), 911-916.

100. Luiken, J., Van der Zee, R., and Welling, GW. (1984) Structure and

Activity of Proteins after Reversed-phase High-performance Liquid

Chromatography. J Chromatogr. 284 (2), 482-486.

101. Ronchi, S., Minchiotti, L., Curti, B. et al. (1976) Isolation, Characterization

and Partial Sequence of Cyanogen Bromide Fragments and Thiol Peptides from

Pig Kidney D-Amino-Acid Oxidase. Biochimica et Biophysica Acta. 427, 634-643.

165

102. Lozier, J., Takahashi, N., and Putnam, FW. (1983) Purification of

Cyanogen Bromide Fragments from beta-2-glycogprotein I by High-performance

Liquid Chromatography. J Chromatogr. 266, 545-554.

103. Van der Rest, M., Bennett, HP., Solomon, S., and Glorieux, FH.

Separation of Collagen Cyanogen Bromide-derived Peptides by Reverse-phase

High-performance Liquid Chromatography. Biochem J. 191 (1), 253-256.

104. Brodersen, R., Andersen, S., Vroum, H., Nielsen, SU., and Pedersen, AO.

(1990) Multiple Fatty Acid Binding to Albumin in Human Blood Plasma. Eur J.

Biochem. 189 (2), 343-349.

105. Geisow, M. and Beaven, G. (1977) Large Fragments of Human Serum

Albumin. Biochem J. 161, 619-625.

106. Travis, J., Bowen, J., Tewksbury, D., Johnson, D., and Pannell, R. (1976)

Isolation of Albumin from Whole Human Plasma and Fractionation of Albumin-

Depleted Plasma. Biochem. J. 157, 301-306.

107. Ledden, DJ., Feldhoff, RC., and Chan, SK. Characterization of Fragments

of Human Albumin Purified by Cibacron Blue F3GA Affinity Chromatography.

Biochem J. 205 (2), 331-337.

108. Jacobsen, C. Lysine Residue 240 of Human Serum Albumin is Involved in

High-Affinity Binding of Bilirubin. Biochem J. 171, 453-459.

109. Westwood, M. and Thornalley, P. Molecular Characteristics of

Methylglyoxal-Modified Bovine and Human Serum Albumins. Comparison with

Glucose-Derived Advanced Glycation Endproducts Modified Serum Albumins. J

Protein Chemistry. 14 (5), 359-372.

110. Morrison, JR., Fidge, NH., and Grego, B. (1990) Studies on the Formation,

Separation, and Characterization of Cyanogen Bromide Fragments of Human AI

Apolipoprtein. Ana Biochem. 186 (1), 145-152.

111. Reed, R., Feldhoff, R., and Peters, T. Jr. (1976) Fragments of Bovine

Serum Albumin Produced by Limited Proteolysis: Complementary Behavior of

Two Large Fragments. Biochemistry, 15 (24), 5394-5398.

166

112. Kragh-Hansen, U., Donaldson, D., and Jensen, P. (2001) The Glycan

Structure of Albumin Redhill, a Glycosylated Variant of Human Serum Albumin.

1550 (Issue 1), 20-26.

113. Krull, RE., Chen, P., Novak, J., Kirk, M., Barnes, S., Baker, J., Krishna,

NR. and Caufield, PW. (2000) Biochemical Structural Analysis of the Iantiobiotic

Mutacin II. J Biol Chem., 275 (21), 15845-15850.

114. Iadarola, P., Zapponi, MC., Minchiotti, L., Meloni, ML., Galliano, M, and

Ferri, G. (1990) Separation of Fragments from Human Serum Albumin and its

Charged Variants by Reversed-phase and Cation-exchange High-performance

Liquid Chromatography. J Chromatogr., 512, 165-76.

115. Moncada, C. and Israel, Y. (1999) Generation of Acetate and Production

of Ethyl-Lysine in the Reaction of Acetaldehyde Plus Serum Albumin. Alcohol, 17

(1), 87-91.

116. Kato, Y. and Toshihiko, O. (1998) Detection of Oxidized Phospholipid-

protein Adducts Using Anti-15-Hydroperoxyeicostetranenoic Acid-Modified

Protein Antibody: Contribution of Esterified Fatty Acid-Protein Adduct to

Oxidative Modification of LDL. Arch Biochem Biophys. 351 (1), 106-114.

117. Lin, L., Mason, A., Woodworth, R., amd Brandts, J. (1994) Calorimetric

Studies of Serum Transferrin and Ovotransferrin. Estimate of Domain

Interactions, and Study of the Kinetic Complexities of Ferric Ion Binding.

Biochemistry, 33 (7), 1881-1886.

118. Foley, A. and Bates, G. (1988) The Influence of Inorganic Anions on the

Formation and Stability of Fe3+ -Transferrin-anion Complexes. Biochimica et

Biophysica Acta. 965, 154-162.

119. Chahine, J. and Fain, D. (1993) The Mechanism of Iron Transferrin

Interactions. Uptake of the Iron Nitrilotriactetic Acid Complex. J. Chem. Soc.

Dalton Trans.

167

120. Schlabach, M. and Bates, G. (1975) The Synergistic Binding of Anions

and Fe3+ by Transferrin. Implications for the Interlocking Sites Hypothesis. The

Journal of Biological Chemistry. 250 (6), 2182-2188.

121. Lin, L., Mason, A., Woodworth, R., and Brandts, J. (1991) Calorimetric

Studies of the Binding of Ferric Ions to Ovotransferrin and Interactions between

Binding Sites. Biochemistry. 30, 11660-11669.

122. Hadden, J., Bloemendal, M., Haris, P., Srai, S., and Chapman, D. (1994)

Fourier Transform Infrared Spectroscopy and Differential Scanning Calorimetry

of Transferrins: Human Serum Transferrin, Rabbit Serum Transferrin and Human

Lactroferrin. Biochimica et Biophysica Acta, 1205, 59-67.

123. Thompson, C., Grady, J., and Chasteen, D. (1986) The Influence of

Uncoordinated Histidines on Iron Release from Transferrin. A Chemical

Modification Study. The J of Bio. Chem. 261 (28), 13128-13134.

124. Kilar, F. and Simon, I. (1985) The Effect of Iron Binding on the

Conformation of Transferrin. A Small Angel X-Ray Scattering Study. Biophys. J.

48, 799-802.

125. Aisen, P., Leibman, A., and Zweier, J. (1978) Stoichiometric and Site

Characteristics of the Binding of Iron to Human Transferrin. The J of Bio. Chem.

253(6), 1930-1937.

126. MacGillivray, RT., Mendez, E., Sinha, SK., Sutton, MR., Lineback-Zins, J.,

and Brew, K. (1982) The Complete Amino Acid Sequence of Human Serum

Transferrin. Proc. Natl. Acad. Sci. USA, 79 (8), 2504-2508.

127. Mecklenburg, S., Donohoe, R., and Olah, G. (1997) Tertiary Structural

Changes and Iron Release From Human Serum Transferrin. J. of Mol. Bio. 270

(5), 739-750.


Studies of the N-terminal Half-molecule of Transferrin and Mutant Forms

Modified Near the Fe3+–binding Site. Biochem. J. 293, 517-522.

168


Studies of the Binding of Ferric Ions to Human Serum Transferrin. Biochem. J.

32, 9398-9406.

130. Esterbauer, H., Schaur, R., and Zollner, H. (1991) Chemistry and

Biochemistry of 4-Hydroxynonenal, Malonaldehyde and Related Aldheydes. Free

Radical Biology & Medicine. 11, 81-128.

131. Ichihashi, K., Osawa, T., Toyokuni, S., and Uchida, K. (2001) Endogenous

Formation of Protein Adducts with Carcinogenic Aldehydes. JBC Papers in

Press. Manuscript M101947200.

132. Uchida, K., Kanemastu, M., Morimitsu, Y., Osawa, T., Noguchi, N., and

Niki, E. (1998) Arcolein Is a Product of Lipid Peroxidation. Formation of Free

Acrolein and Its Conjugate with Lysine Residues in Oxidized Low Denisty

Liproproteins. The J. of Biol. Chem., 273 (26), 16058-16066.

133. Alaiz, M. and Barragan, S. (1995) Reaction of a Lysyl Residue Analogue

with E-2-octenal. Chem. And Phys. Of Lipids. 75, 43-49.

134. Bennaars-Eiden, A., Higgins, L., Kapphahn, RJ., Ferringon, DA., and

Bernlohr, DA. (2002) Covalent Modification of Epithelial Fatty Acid-binding

Protein by 4-hydroxynonenal In Vitro and In Vivo. Evidence for a Role in

Antioxidant Biology. J. Biol. Chem. 277 (52), 50693-50702.

135. Rohn, TT., Nelson, LK., Waeg, G., and Quinn, MT. (1998) U-101033 (2,4-

diaminopyrrolopyrimidine), a Potent Inhibitor of Membrane Lipid Peroxidation as

Assessed by the Production of 4-Hydroxynonenal, Malondialdehyde, and 4-

Hydroxynonenal—protein Adducts. Biochem. Parmacol. 56 (10), 1371-1379.

136. Schneider, C., Tallman, K., Porter, N., and Brash, A. (2001) Two Distinct

Pathways of Formation of 4-Hydroxy-nonenal: Mechanisms of Non-enzymatic

Transformation of the 9- and 13-Hydroperoxides of Linoleic Acid to 4-

Hydroxyalkenals. JBC Papers in Press. Manuscript M101821200.

137. Requena, JR., Fu, MX., Ahmed, MU., Jenkins, AJ., Lyons, TJ., Baynes,

JW., and Thorpe, SR. (1997) Quantification of Malondialdehyde and 4-

169

Hydroxynonenal Adducts to Lysine Residues in Native and Oxidized Human

Low-density Lipoprotein. Biochem J. 322 (Pt 1), 317-325.

138. Pietrzak, E., Shanley, B., and Kroon, P. (1995) Antibodies Made Against a

Formaldehyde-Protein Adducts Cross React with an Acetaldehyde-Protein

Adduct. Implication for the Origin of Antibodies in Human Serum which

Recognize Acetaldehyde-Protein Adducts. Alcohol & Alcoholism. 30 (3), 373-

378.

139. Aldini, G., Carini, M., Beretta, G., Bradamante, S., and Facino, RM. (2002)

Carnosine is a Quencher of 4-Hydroxy-Nonenal: Through what Mechanism of

Reaction? Biochem Biophys Res Commun. 298 (5), 699-706.

140. Esterbauer, H., Jurgens, G., Quehenberger, O., and Koller, E. (1987)

Autoxidation of Human Low Density Lipoprotein: Loss of Polyunsaturated Fatty

Acids and Vitamin E and Generation of Aldehydes. J of Lip. Res. 28, 495-509.

141. Kikugawa, K., Machida, Y., Kida, M., and Kurechi, T. (1981) Studies on

Peroxidized Lipids. III. Fluorescent Pigments Derived from the Reaction of

Malonaldehyde and Amino Acids. Chem. Pharm. Bull. 29 (10), 3003-3011.

142. Houglum, K., Filip, M., Witztum, J., and Chojkier, M. (1991)

Malondialdehyde and 4-Hydroxynonenal Protein Adducts in Plasma and Liver of

Rats with Iron Overload. J. of Clin. Invest. 86, 1991-1998.

143. Uchida, K., Toyokuni, S., Nishikawa, K., Kawakishi, S., Oda, H., Hiai, H.,

and Stadtman, ER. (1994) Michael Addition-Type 4-Hydroxy-2-Nonenal Adducts

in Modified Low-Density Lipoproteins: Markers for Atherosclerosis. Biochem. 33

(41), 12487-12494.

144. Uchida, K., Sakai, K., Itakura, K., Osawa, T., and Toyokuni, S. (1997)

Protein Modification by Lipid Peroxidation Products: Formation of

Malondialdehyde-Derived Nε-(2-Propenal) Lysine in Proteins. Arch. Of Biochem.

And Biophys. 346 (1), 45-52.

145. Cohnn, JA., Tsai, L., Friguet, B., and Szweda, LI. (1996) Chemical

Characterization of Protein-4-Hydroxy-2-Nonenal Cross-Link: Immunochemical

170

Detection in Mitochondria Exposed to Oxidative Stress. Arch. Biochem. Biophys.

328 (1), 158-164.

146. Xu, G., Liu, Y., Kansal, M., and Sayre, L. (1999) Rapid Cross-Linking of

Protein by 4-Ketoaldehydes and 4-Hydroxy-2-Alkenals Does Not Arise from the

Lysine-Derived Monoalkylpyrroles. Chem. Res. Toxicol. 12, 855-861.

147. Burcham, P.C., and Kuhan, Y.T. (1996). Introduction of carbonyl groups

into protein by the lipid peroxidation product, malondialdehyde. Biochem.

Biophys. Acta. 220, 996-1001.

148. Chio, K.S., and Tappel, A.L. (1972). Inactivation of ribonuclease and other

enzymes by peroxidizing lipids by malondialdehyde. Biochemistry. 8, 2827-2832.

149. Furuhata, A., Nakamura, M., Osawa, T., et al. (2002). Thiolation of

protein-bound carcinogenic aldehyde: An electrophile acrolein-lysine adduct that

covalently bound to thiols. J. Biol. Chem. 276, 12076-12083.

150. Meister, G.E., and Butler, A. (1996). Inactivation of vanadium

bromoperoxidase: formation of 2-oxohistidine. Biochemistry. 35, 11805-11811.

151. Uchida, K., and Kawakishi, S. (1993). 2-Oxo-histidine as a novel biological

marker for oxidatively modified proteins. FEBS Lett. 332, 208-210.

152. Rivett, A.J., and Levine, R.L. (1990). Metal-catalyzed oxidation of

Escherichia coli glutamine synthetase: correlation of structural an functional

changes. Arch. Biochem. Biophys. 278, 26-34.

153. Zhao, F., Ghezzo-Schoneich, E., Aced, G.I., et al. (1997). Metal-catalyzed

oxidation of histidine in human growth hormone: mechanism, isotope effects, and

inhibition by a mild denaturing alcohol. J. Biol. Chem. 72, 3019-9029.

171

Isolation and Characterization of Water-Soluble ...

Documents

Transcript of Isolation and Characterization of Water-Soluble ...