Isolation and Characterization of Soluble Peptidoglycan from ...
Isolation and Characterization of Water-Soluble ...
Transcript of 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
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 34 Peptide fragments mapping fingerprinting database search hit for the 80 kDa protein band .............................................................................................................. 75
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 37 Peptide fragments mapping fingerprinting database search hit for the 51 kDa protein band .............................................................................................................. 77
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 39 Peptide fragments mapping fingerprinting database search hit for the 30 kDa protein band .............................................................................................................. 78
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
Rank Protein Name Accession No. Protein MW Protein Protein Protein
PI Score C. I. %
1 transferrin protein [Homo sapiens] gi|4557871 76999.6 6.81 160 100
Peptide Information
Calc. Mass Obsrv. Mass ± da ± ppm Start End Sequence Modification
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
4700 Reflector Spec #1 MC[BP = 1467.9, 1510]
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
Calc. Mass Obsrv. Mass ± da ± ppm Start End Sequence Modification
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
4700 Reflector Spec #1 MC[BP = 2327.3, 1224]
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
Rank Protein Name Accession No. Protein MW Protein Protein Protein
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
Calc. Mass Obsrv. Mass ± da ± ppm Start End Sequence Modification
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
experimental section.
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
ityVoyager Spec #1[BP = 1107.6, 19338]
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
ityVoyager Spec #1[BP = 1073.5, 33032]
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
ityVoyager Spec #1=>NF1.0[BP = 1067.0, 58682]
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
ityVoyager Spec #1=>NF1.0[BP = 1064.5, 56085]
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
Fe/apo-HST Molar Ratio
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
Fe/apo-HST Molar Ratio
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
Fe/apo-HST Molar Ratio
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
Fe/apo-HST Molar Ratio
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
Data: Data1_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 2.3965E-8R 2 = 0.99425 y0 0.00624 ±0.00002A1 0.00967 ±0.00006t1 31.35802 ±0.32213
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
Sample KN (x10-3) (min-1) KC (x10-3) (min-1)
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
Data: Data3_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 2.6455E-6R 2 = 0.93047 y0 0.10088 ±0.00019A1 0.02582 ±0.00051t1 39.55242 ±1.56531
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
Data: Data3_BModel: ExpDec1Equation: y = A1*exp(-x/t1) + y0Weighting: y No weighting Chi 2/DoF = 6.0123E-6R 2 = 0.97687 y0 0.0521 ±0.00019A1 0.08946 ±0.00121t1 20.58157 ±0.39958
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
Runs KC (x10-3) (min-1)
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
Sample KN (x10-3) (min-1) KC (x10-3) (min-1)
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
Runs KC (x10-3) (min-1)
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
Sample KC (x10-3) (min-1)
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
experimental section.
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
Sample KC (x10-3) (min-1)
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
3.7 Serum Incubation Experiments
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
4.7 Serum Incubation Experiments
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.
Biochemistry. 42, 3701-07.
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.
Biochemistry. 40, 1741-48.
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.
128. Lin, L., Mason, A., Woodworth, R., and Brandts, J. (1993) Calorimetric
Studies of the N-terminal Half-molecule of Transferrin and Mutant Forms
Modified Near the Fe3+–binding Site. Biochem. J. 293, 517-522.
168
129. Lin, L., Mason, A., Woodworth, R., and Brandts, J. (1993) Calorimetric
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