ABSTRACT

THE BIOCHEMICAL STUDY OF AGE RELATED CHANGES IN HUMAN RETINAL PIGMENT EPITHELIUM AND BRUCHS MEMBRANE Laura S. Murdaugh, Ph.D. Department of Chemistry and Biochemistry Northern Illinois University, 2010 Elizabeth R. Gaillard, Director Age-related macular degeneration (AMD) is an ocular disease that causes severe visual loss and legal blindness in the elderly population. The pathophysiology of AMD is complex and may include genetic predispositions, accumulation of lipofuscin and drusen, local inflammation and neovascularization. Therefore, specific age-related changes in the retinal pigment epithelium (RPE) and Bruchs membrane have been investigated The accumulation of lipofuscin has been shown to precede the death of photoreceptor cells and the deterioration of the RPE. As a result, the determination of the photosensitive components of lipofuscin have been of major interest. One of these components, previously identified as a bis-retinoid pyridinium compound, is referred to as A2E. A2E has been characterized by mass spectrometry and is known to have a mass of 592 Da. The remaining chromaphores in RPE lipofuscin are structurally related to A2E as determined by their fragmentation pattern with losses of M+/- 190, 174 and/or 150 Da. Analysis of lipofuscin from various donors indicates that the

extracts consist of as many as fifteen of these hydrophobic components which are also observed to form spontaneously in vitro over extended periods of time. Previous studies have shown that numerous structural changes are induced in Bruchs membrane with age. These changes may have a harmful effect on Bruchs membrane, resulting in damage to RPE cells and the onset of AMD. Recent research has identified a commonly inherited variant of the complement factor H gene from different groups of AMD patients linking the genetics of the disease to inflammation. During inflammation there is activation of nitric oxide synthase and release of nitric oxide, which could lead to non-enzymatic nitration within extracellular deposits and/or intrinsic extracellular matrix (ECM) protein components of human Bruchs membrane. Two possible biomarkers for non-enzymatic nitration in aged human Bruchs membrane have been identified, which include 3-nitrotyrosine and nitrated A2E. The presence of nitrated A2E could not be detected in RPE extracts, suggesting that nitro-A2E may be a Bruchs membrane specific biomarker. The nonenzymatic glycation and nitration of the basement membrane protein laminin, as a model for aging Bruchs membrane, was also investigated. The results indicated that fragments containing lysine and arginine residues were preferentially modified in the glycated and irradiated samples. However, nitration of laminin fragments was not observed. Instead several of the fragments ending in a lysine residue appeared to bind to other fragments also ending in a lysine residue, indicating a polymerization-type reaction. This study provides evidence that glycation, nitration, and the presence of A2E may be involved in modifications to essential basement membrane proteins leading to deleterious changes within the RPE ECM environment.

NOTHERN ILLINOIS UNIVERSITY DEKALB, ILLINOIS

MAY 2010

THE BIOCHEMICAL STUDY OF AGE-RELATED CHANGES IN HUMAN RETINAL PIGMENT EPITHELIUM AND BRUCHS MEMBRANE

BY LAURA S. MURDAUGH 2010 Laura S. Murdaugh

A DISSERTATION SUBMITTED TO THE GRADUATE SCHOOL IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE DOCTOR OF PHILOSOPHY

DEPARTMENT OF CHEMISTRY AND BIOCHEMISTRY

Doctoral Director: Elizabeth R. Gaillard

UMI Number: 3404858

All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion.

UMI 3404858 Copyright 2010 by ProQuest LLC. All rights reserved. This edition of the work is protected against unauthorized copying under Title 17, United States Code.

ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, MI 48106-1346

ACKNOWLEDGEMENTS

I would like to sincerely thank several individuals who made this work possible. First, I would like to thank my advisor, Dr. Elizabeth Gaillard, for her guidance, support and hard work. Her advising and encouragement have been invaluable and have made me a better chemist. I would also like to thank the members of my dissertation committee, Dr. James Dillon, Dr. Victor Ryzhov, Dr. James Horn, and Dr. Linda Yasui, for their input and critically revising my dissertation. I want to give a special thank you to Dr. James Dillon for always taking the time to answer my questions and assist with technical problems and research. I would also like to thank my family, Linda and Gaylord Murdaugh, Ross and Judy Dill, and Christine and John Leal for their love, support, faith, and never-ending reassurance throughout the years. Finally, I would like to thank my husband, Adam Dill, for his love, patience, and understanding. His help and encouragement were essential to finishing this work.

TABLE OF CONTENTS

Page LIST OF TABLES..................................................................................... ..... vii LIST OF FIGURES.......................................................................................... ix Chapter 1. INTRODUCTION... The Visual System..... The Retinal Pigment Epithelium. .. 1 2 7

Bruchs Membrane.. .. 14 Lipofuscin.. 19 Oxidative Stress and the antioxidant Glutathione . 26 Inflammmation and AMD . 32 Advanced Glycation Endproducts and AMD 34 Dissertation Research .................................... 38 2. MATERIALS AND METHODS 40 Materials 40 Instrumentation .. 41 Methods . 48 Synthesis of A2E . 48

Chapter

iv Page Isolation of Lipofuscin ...... . 48 Auto-Oxidation of A2E............................. .. 55 Lipofuscin and A2E LC-MS Analysis 55 Determination of the Water-Octanol Partition Coefficient of A2E: Log P... 56 Cyclic Voltammetry ... 57 Reaction of A2E with Retinalaldehye .... 58 Separation of a Compound with m/z 920 from A2E RAL Reaction Mixture ... 58 Bruchs Membrane Preparation... 59 Preparation of Organic Soluble Materials from Bruchs Membrane... 60 Bruchs Membrane LC-MS Analysis.. 60 Acid Hydrolysis... 61 Bruchs Membrane LC-MS Analysis After Acid Hydrolysis and Standard Addition of 3Nitrotyrosine............................................ 62 Conditions of Tryptic Digests for Laminin Samples.63 Modifications with Glycolaldehyde to Laminin63 Modifications to Laminin with A2E..64 Modifications to Laminin with NaNO2. 64 LC-MS Analysis of Laminin Samples.65 Protein Prospector. 66

Chapter

v Page Bioworks Browser...... 67 SEQUEST... 67 Data analysis... 68

3.

THE COMPOSITIONAL STUDIES AND MOLECULAR MODIFICATIONS OF HUMAN RPE LIPOFSUCIN. Introduction... .. 69 69

Results... ... 72 Discussion 139 4. AGE-RELATED ACCUMULATION OF A2E AND NITROA2E IN HUMAN BRUCHS MEMBRANE. .146 Introduction....... 146 Results.. .151 Indentification of tyrosine nitration in Bruchs membrane ....151 Identification of nitro-A2E in Bruchs membrane ..154 Concentration of nitro-A2E in Bruchs membrane samples from different decades of life ....169 Discussion......176 5. MODIFICATIONS TO THE BASEMENT PROTEIN LAMININ AND A2E: A MODEL FOR AGING IN BRUCHS MEMBRANE..181

Chapter

vi Page Introduction. 181 Results. 185 Laminin modified with glycolaldehyde...185 Laminin modified with carboxymethyllysine........................................203 Laminin modified with A2E....209 Laminin modification with nitrite ...228 Discussion243

6.

CONCLUSIONS AND FUTURE WORK......248 Compositional studies of human retinal lipofuscin .....249 Accumulation of 3-nitrotyrosine and nitro-A2E in Bruchs membrane...........250 Modifications to laminin .........252

REFERENCES ........255

LIST OF TABLES Table 5.1 Page Laminin Control: Laminin fragments identified in the control sample including the observed m/z, associated charge, parent ions (MH+), and corresponding amino acid sequences .....189 Glycated Laminin Sample: Laminin fragments (without modifications) identified in the glycated laminin sample including the observed m/z, associated charge, the MH+, and corresponding amino acid sequence....196 Glycated Laminin: Most abundant laminin fragments modified with glycolaldehyde identified by LC-MS/MS including the observed m/z of the unmodified sequence, the associated charge, the observed m/z of the sequence after modification with glycolaldehyde, the observed intensity associated with the modified m/z, and the corresponding amino acid sequence with site of modification highlighted .....197 Glycated Laminin: Most abundant laminin fragments modified with CML identified by LC-MS/MS including the observed m/z of the unmodified sequence, the associated charge, the observed m/z of the sequence after modification with glycolaldehyde, the observed intensity associated with the modified m/z, and the corresponding amino acid sequence with site of modification highlighted .....205 Laminin fragments identified in A2E incubated laminin samples including the observed m/z of the laminin fragment, the associated charge, the MH+, and the corresponding amino acid sequence...216

5.2

5.3

5.4

5.5

viii Table 5.6 Page Laminin fragments modified with irradiated A2E including the observed m/z of the laminin fragment, the corresponding amino acid sequence with the site of modification highlighted, the associated charge, and the observed masses of laminin with modification A2E aldehydes.....217 Peptide fragments CSR, CSRAR, and CSRARK in the laminin control, glycated laminin, A2E laminin control, and irradiated A2E laminin samples including their corresponding observed m/z, associated charge, and retention time. The irradiated A2E sample also includes the mass of the corresponding A2E aldehyde modification.....229 Control Laminin Sample: Laminin fragments identified in the NaCl laminin sample including the observed m/z, associated charge, the MH+, and corresponding amino acid sequence...230 Nitrated Laminin Sample: Laminin fragments identified in the nitrated sample including the observed m/z, associated charge, the MH+, and corresponding amino acid sequence...231 Peptide fragments ARK, CSRARK, and QAASIK in the laminin control and nitrated laminin sample including their corresponding observed m/z, associated charge, and retention time ....233

5.7

5.8

5.9

5.10

LIST OF FIGURES Figure 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 1.12 1.13 2.1 2.2 2.3 2.4 Page Anatomy of the human eye...... 3 The Retina... 5 The Retinal Pigment Epithelium cell structure showing the relationship between the RPE cell and Bruchs Membrane.... 8 Formtion of phagolysosme and lipofuscin.... 11 The visual cycle. 12 The position and layers of Bruchs membrane.. 15 Transmission electron microscope image of drusen. 18 Transmission electron microscope image of lipofuscin granule... 21 Stucture of A2E and iso-A2E. ...22 Synthesis of A2E in vivo....................................................................... 23 Structures of A2E and oxidized A2E with corresponding aldehydes identified... 27 Structure of glutathione (GSH) and its dimer (GSSG).. 30 Maillard reaction35 Electrospray Ionization.. 42 Taylor Cone... 43 Quadrupole Ion Trap. 45 Electron multiplier. 46

x Figure 2.5 2.6 2.7 2.8 2.9 3.1 Page Chromatogram of the A2E reaction mixture using HPLC with PDA detection. A2E and iso-A2E are identified.. .49 The UV-Vis spectra of A2E and iso-A2E. 50 The mass spectrum of purified A2E.. 51 The MS/MS spectrum of purified A2E. 52 Isolation of Lipofuscin.. 54 The TIC from the Folch extract of lipofuscin granules (top) And the corresponding PDA chromatogram (bottom) are shown. The chromatogram consists of A2E, oxidized A2E, and a complex mixture of components..73 The mass spectrum of the Folch extract of human lipofuscin at time 62.93 mins. Group I, II, and III identify the related clusters of higher molecular weight compounds with mass to charge ratios of approximately 800, 1000, and 1400 respectively. Highlighted in red are the additions of 14 amu starting with m/z 847.9. 75 The mass spectrum of the Folch extract of human lipofuscin at time 86.26 mins Group II, and III identify the related clusters of higher molecular weight compounds with mass to charge ratios of approximately 1000 and 1400 respectively..... 76 The MS/MS scan for A2E identified in the Folch extract of lipofuscin granules. Peaks corresponding to the mass of 592 (red) with the loss of, 106 (m/z 486.5), 150 (m/z 442), 174 (m/z 418), and 190 (m/z 402) are identified... 77 The UV-visible spectrum of A2E.. 78 Characteristic cleavages for the fragmentation of A2E. 79 The MS/MS scan of peak with m/z 814 from lipofuscin sample. Peaks corresponding to the mass of 814 (red) with the loss of 106, 150, 174, and 190 are identified (blue) ............80 The UV-Vis absorption for the peak with m/z 81481

3.2

3.3

3.4

3.5 3.6 3.7

3.8

xi

Figure 3.9 3.10

Page Possible Structure of m/z 814 with cleavages identified...............83 The MS/MS scan for m/z 1081 located in lipofuscin. Peaks corresponding to the mass of 1081 (red) with the loss of 106 (m/z 975), 150 (m/z 931), 174 (m/z 907), and 190 (m/z 891) are identified (blue). ....84 The UV-Visible spectrum of m/z 1081 in lipofuscin. ...85 Possible structure of m/z 1081 with cleavages identified..86 The MS/MS results for the fragmentation of peak with m/z 1423 (red) in the lipofuscin sample. Peaks corresponding to the mass of 1423 with the loss of 174 (m/z 1249) and 190 (m/z 1233) are identified (blue) 87 Possible structure for m/z 1424 with cleavages identified 88 Calibration curve for Log P values of DDT, Triphenylamine, Phenanthrene, Benzophenone, and Cinnamic Acid to determine the Log P of A2E and higher molecular weight products .....90 Product from esterification reaction with A2E and R group. The R group being acetyl chloride, Hexanoyl chloride, or Cinnamoyl chloride ...91 The MS/MS of A2E acetyl ester (m/z 634) with the corresponding structure .....92 The MS/MS of the A2E hexanoyl ester (m/z = 690.5) with the corresponding structure ...93 CID of main fragment m/z 548 (red) with losses of 150 (m/z 398), 174 (m/z 374), and 190 (m/z 358)(blue) ..94 CID spectrum of species with m/z = 548 (red) with losses of 150 (m/z 398), 174 (m/z 374), and 190 (m/z 358) (blue) in full mass spectrum of human lipofuscin sample ...95

3.11 3.12 3.13

3.14 3.15

3.16

3.17 3.18 3.19 3.20

xii Figure 3.21 3.22 3.23 3.24 3.25 3.26 Page Rearrangement of esterification product yielding main fragment with m/z 548.......96 Possible structure and fragmentations of peak with m/z = 548.....97 MS/MS of Cinnamoyl chloride ester (m/z = 723).....98 Proposed product of Cinnamoyl chloride ester (m/z = 723) .......100 MacLafferty rearrangement in species with m/z = 574. ..101 The mass spectrum of A2E mixture at 93.52 minutes of chromatographic separation. Peaks found in lipofuscin mixture (Figures 3.2 and 3.3) are identified (blue) .....102 The MS/MS of m/z 859 in A2E. Peaks corresponding to mass 859 (red) with the loss of 150 (m/z 709), 174 (m/z 685), and 190 (m/z 669) are identified (blue) .....103 UV-visible spectrum of m/z 858 in aged A2E ....104 The MS/MS scan for m/z 1081 located in aged A2E. Peaks corresponding to the mass of 1081 (red) with the loss of 150 (m/z 931), 174 (m/z 907), and 190 (m/z 891) are identified (blue). The mass of A2E (m/z 592) and additional peaks corresponding to smaller molecular weight compounds (m/z 818 and 745) with similar losses identified in the same sample....105 The UV-Visible spectrum of m/z 1081 in aged A2E ..106 The MS/MS of m/z 859 in reaction mixture for A2E synthesis. Peaks corresponding to mass 859 (red) with the loss of 150 (m/z 709), 174 (m/z 685), and 190 (m/z 669) are identified (blue).....108 The MS/MS of m/z 920 in reaction mixture for A2E synthesis. Peaks corresponding to the mass of 920 (red) with the loss of 150 (m/z 771) and 190 (m/z 731) are identified (blue) ...109

3.27

3.28 3.29

3.30 3.31

3.32

xiii Figure 3.33 Page The MS/MS of 1189 in reaction mixture for A2E sythesis. Peaks corresponding to the mass of 1189 with the loss of 150 (m/z 1039), 174 (m/z 1015), and 190 (m/z 999) are identified .....110 The UV-Visible spectrum of m/z 859 in reaction mixture for A2E synthesis .......111 The UV-Visible spectrum of m/z 920 in reaction mixture for A2E synthesis ....112 The UV-Vis of m/z 1189 in reaction mixture for A2E sythesis .113 The full mass spectrum of the reaction between A2E and cinnamaldehyde...114 The full mass spectrum of the reaction between A2E and benzaldehyde ...115 The MS/MS spectrum of the higher molecular weight compound (m/z = 790) in A2E and Cinnamaldehyde reaction mixture using 40 % collision energy. Peaks corresponding to the mass of 790 (red) with the loss of 150 (m/z 640), 174 (m/z 617), and 190 (m/z 556) are identified (blue) .....116 The MS/MS spectrum of one of the higher molecular weight compounds in A2E benzaldehyde reaction mixture. Peaks corresponding to the mass of 794 (red) with the loss of 122 (m/z 672), 140 (m/z 654), 150 (m/z 644), and 190 (m/z 604) are identified (blue)..117 Structure and fragmentation of one of the higher molecular weight compounds from reaction of oxidized A2E and cinnmaldehyde .118 Possible structure and fragmentation of one of the higher molecular weight compounds from reaction of oxidized A2E and benzaldehyde ....119 The mass spectrum of A2E reacted with all-trans-retinal121

3.34 3.35 3.36 3.37 3.38 3.39

3.40

3.41

3.42

3.43

xiv Figure 3.44 Page The MS/MS spectrum of m/z 920 from A2E RAL reaction. Peaks corresponding to the mass of 920 (red) with the loss of 150 (m/z 771) and 190 (m/z 731) are identified (blue).. .122 The MS/MS spectrum of m/z 1189 from A2E RAL reaction. Peaks corresponding to the mass of 1189 (red) with the loss of 150 (m/z 1039), 174 (m/z 1015), and 190 (m/z 999) are identified (blue) ...123 Possible structure of m/z 920 with cleavages identified .....124 Possible structure of m/z 1189 125 The MS/MS of m/z 859 in A2E and all-trans-retinal reaction. Peaks corresponding to m/z 858 (red) with the loss of 150 (m/z 708), 174 (m/z 684), and 190 (m/z 668) are identified (blue).126 Possible structure for m/z 859 with cleavages identified 127 The chromatogram of the A2E RAL reaction mixture using HPLC and PDA detection. Compound with m/z 920 eluted at 35 min.. 128 The full mass spectrum if peak that eluted at 35 min in Figure 3.18...129 UV-Vis absorption for the peak with m/z 920 130 The MS/MS spectrum of m/z 920 ...131 The MS/MS spectrum of m/z 920 from Lipofuscin. Peaks corresponding to the mass of 920 (red) with the loss of 150 (m/z 771) and 190 (m/z 731) are identified (blue).... 132 The voltammogram of TEAP background.. 133 The voltammogram of ferrocene. 135 The voltammogram of benzaldehyde.. 136 The voltammogram of cinnamaldehyde.. 137 The voltammogram of all-trans retinal 138

3.45

3.46 3.47 3.48

3.49 3.50

3.51 3.52 3.53 3.54

3.55 3.56 3.57 3.58 3.59

xv Figure 4.1 Page Selected Reaction Monitoring (SRM) chromatogram of 3-NT and acid hydrolysate of BM (SRM 227.1181.1).3-NT and acid hydrolysate of BM was analyzed by LC/MS as described in method. The SRM scan of BM acid hydrolysate has a peak with molecular mass 227 and fragment 181 and similar retention time (51 minutes) to 3-NT which indicates the presence of 3-NT in BM acid hydrolysate... 152 The tandem mass spectra of standard 3-nitrotyrosine and component with m/z 227.0 at RT 51mins in BM. The tandem mass spectrum of the component at RT 51mins from human BM extracted from 72-75 year old donors is very similar to the tandem mass spectrum of 3-NT. The inset gives the predicted fragmentation of 3-NT. 153 The zoom scan of BM with the standard addition of 3-nitrotyrosine (m/z 227). 155 The SRM scan of m/z 227 181 from the standard addition of 3-nitrotyrosine in BM samples from different age groups.. 156 Integration of area under SRM scan from Figure 4.4.. 157 Calibration curve for 3-nitrotyrosine... 158 The concentration of 3-nitrotyrosine in BM samples from ages of < 25, 40-60, and > 65 years.159 The selected ion chromatograms for synthetic A2E and nitro-A2E.. 160 The UV-Vis spectra for A2E (m/z 592.5) and for nitro-A2E (m/z 637.5). 162 Structures of A2E (m/z 592) and nitro-A2E (m/z 637) showing characteristic cleavage points and the resulting fragment molecular weights 163 The tandem mass spectrum of synthetic nitro-A2E induced dissociation to confirm the identification of nitro-A2E. 164

4.2

4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10

4.11

xvi Figure 4.12 Page The tandem mass spectrum of nitro-A2E isolated from 65 yrs and older BM. Box = mass same in synthetic nitro-A2E and nitro-A2E isolated from 65 yrs and older BM.166 Chromatogram of m/z 592.5 (A2E), m/z 637.5 (nitro A2E), m/z 653.4 (nitro A2E plus oxygen), and m/z 682.5 (A2E with 2 nitro substitutions).. 167 The selected ion chromatograms for A2E (m/z 592) and nitro-A2E (m/z 637) from RPE lipofuscin and BM extracts from human donor globes that were 65 yrs and older. Note that nitro A2E and A2E from the BM have similar concentrations, whereas no nitro-A2E could be detected from the RPE despite increasing the sensitivity of the detector...168 Integration of A2E in BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs)...170 The concentration of A2E in BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs)..171 The SRM scans of Nitro-A2E from BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs)..172 Integration of Nitro-A2E A2E in BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs) .....173 The concentration of Nitro-A2E in BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs) .....174 The concentration of A2E and Nitro-A2E in BM samples from <20, 40, 50, 60 70, and 80 decades of life and dry AMD...175 The amino acid sequence of laminin with B and Y ions identified.....186 The reaction scheme for glycation of lysine and arginine within the laminin fragment or with A2E and A2E derived aldehydes....187 The TIC for a typical enzymatically digested laminin control sample without modification is shown. The m/z ratios corresponding to fragments with amino acid sequences of CSRARK, AR, CSRARKQAASIKVAVSADR, and CSRAR are identified...188

4.13

4.14

4.15 4.16 4.17 4.18 4.19 4.20 5.1 5.2 5.3

xvii Figure 5.4 5.5 5.6 5.7 5.8 Page The MS/MS spectrum for digested laminin fragment, CSR (m/z 365),with B and Y ions identified ..190 The MS/MS spectrum for digested laminin fragment, ARK (m/z 374), with B and Y ions identified.....191 The MS/MS spectrum for digested laminin fragment, QAASIK (m/z 618), with B and Y ions identified......192 The MS/MS spectrum for digested laminin fragment, VAVSADR (m/z 718), with B and Y ions identified..193 The MS/MS spectrum for digested laminin fragment, CSRARKQAASIKVAVSADR (m/z 1009), with B and Y ions identified ..194 The MS/MS spectrum for digested laminin fragment, CSR (m/z 204), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue..198 The MS/MS spectrum for digested laminin fragment, CSRAR (m/z 634), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue....199 The MS/MS spectrum for digested laminin fragment, CSRARK (m/z 762), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue 200 The MS/MS spectrum for digested laminin fragment, QAASIK (m/z 659), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue....201 The ms/ms spectrum for digested laminin fragment, CSRARKQAASIKVAVSADR (m/z 1051), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue ...202

5.9

5.10

5.11

5.12

5.13

Figure 5.14

xviii Page The reaction scheme of glycolaldehyde with lysine producing carboxymethyl lysine (CML) and then the modification of primary amines in laminin by CML ....204 The MS/MS spectrum of CML located in the glycated laminin sample. The inset is the structure of CML (m/z 205) with characteristic cleavages identified ..206 The MS/MS spectrum for digested laminin fragment, ARK (m/z 543), modified by CML. The site of modification is highlighted in red and the B and Y ions are identified in blue....207 The proposed structure for CML modification of ARK fragment ..208 The MS/MS spectrum for digested laminin fragment, CSRARK (m/z 906), modified by CML. The site of modification is highlighted in red and the B and Y ions are identified in blue....210 The MS/MS spectrum for digested laminin fragment, QAASIK (m/z 803), modified by CML. The site of modification is highlighted in red and the B and Y ions are identified in blue....211 The proposed structure for CML modification of CSRARK Fragment......212 The proposed structure for CML modification of QAASIK Fragment......213 The cleavage positions and the molecular masses of corresponding aldehydes in A2E and oxidized A2E are shown .214 The MS/MS spectrum for digested laminin fragment, CSRAR (m/z 980), modified by A2E derived aldehyde with m/z 406. The site of modification is highlighted in red and the B and Y ions are identified in blue .218 The proposed structure for A2E derived aldehyde (m/z 406) modification of CSRAR fragment...220

5.15

5.16

5.17 5.18

5.19

5.20 5.21 5.22 5.23

5.24

xix Figure 5.25 Page The MS/MS spectrum for digested laminin fragment, CSRAR (m/z 956), modified by A2E derived aldehyde with m/z 382. The site of modification is highlighted in red and the B and Y ions are identified in blue 221 The proposed structure for A2E derived aldehyde (m/z 382) modification of CSRAR fragment...222 The MS/MS spectrum for digested laminin fragment, KQAASIK (m/z 1058), modified by A2E derived aldehyde with m/z 366. The site of modification is highlighted in red and the B and Y ions are identified in blue .223 The proposed structure for A2E derived aldehyde (m/z 366) modification of KQAASIK fragment ..224 The MS/MS spectrum for digested laminin fragment, KQAASIK (m/z 1058), modified by A2E derived aldehyde with m/z 382. The site of modification is highlighted in red and the B and Y ions are identified in blue .225 The proposed structure for A2E derived aldehyde (m/z 382) modification of KQAASIK fragment ..226 The HPLC total PDA trace of the laminin control, glycated laminin, A2E laminin control, and irradiated A2E laminin samples are shown respectively. Selected fragments are identified in each chromatogram .227 The MS/MS spectrum for digested laminin fragment QAASIKKRA (m/z 973). The B and Y ions are identified in blue ....234 The MS/MS spectrum for digested laminin fragment CSRARKKRARSC (m/z 711). The B and Y ions are identified in blue..235 The ms/ms spectrum for digested laminin fragment QAASIKKISAAQ (m/z 608). The B and Y ions are identified in blue....236 The MS/MS spectrum for digested laminin fragment ARKKRA (m/z 729). The B and Y ions are identified in blue ...237

5.26 5.27

5.28 5.29

5.30 5.31

5.32 5.33

5.34

5.35

xx Figure 5.36 5.37 5.38 5.39 5.40 Page The proposed structure for QAASIKKRA ..238 The proposed structure for CSRARKKRARSC..239 The proposed structure for QAASIKKISAAQ ...240 The proposed structure for ARKKRA.241 The HPLC total PDA trace of the laminin control and nitrated laminin samples are shown respectively. Selected fragments, ARK, CSRARK, AND QAASIK, are identified in each Chromatogram .242

CHAPTER 1 INTRODUCTION

Age-related macular degeneration (AMD) is the predominant cause of irreversible blindness in developed countries. Currently, 15 million Americans have been diagnosed with the disease, with an estimated 2 million new cases each year (Friedman, O'Colmain et al. 2004). Patients diagnosed with AMD lose their high resolution central vision. Initially, patients may exhibit mild symptoms of blurring and distortion but as the disease progresses a complete loss of their central vision generally occurs. AMD is characterized as two types: either atrophic (dry) or exudative (wet). Atrophic AMD involves the degeneration of the retinal pigment epithelium (RPE) and photoreceptor cells. This is the most common form, accounting for approximately 85 % of all cases. The more rapidly progressing form, exudative AMD, occurs in only a small percentage of patients. Choroidal neovascularization is a predominant symptom associated with exudative AMD. These new blood vessels created within the eye are generally weak and tend to break open, leading to bleeding within the retina and a sudden loss in central vision. Previously, literature has suggested that AMD is one disease with two forms and that a patient with the atrophic form will often develop the more severe exudative form (Brown, Brown et al. 2005). However, Hageman et al. recently proposed that AMD is actually two

2 separate and distinct diseases and, therefore, patient with atrophic AMD will not necessarily develop exudative AMD (Hageman, Luthert et al. 2001). Because of this ambiguity and the mutifactorial nature of AMD, the exact mechanism leading to the death of photoreceptor cells and the onset of AMD is still unknown. Recent research has suggested that age-related changes within the RPE and underlying Bruchs Membrane play a crucial role in the development of AMD (Dorey, Wu et al. 1989; Taylor, Munoz et al. 1990; Holz, Bellman et al. 2001). Therefore, age-related changes to the RPE and Bruchs Membrane and the mechanisms involved were investigated. Increasing our understanding of the biochemical and cellular changes occurring in the RPE and Bruchs Membrane may aid in the development of new therapies when diagnosing and treating patients with AMD.

The Visual System

The visual system is a series of complex interlinking processes that enable an organism to see and is part of the central nervous system. The modular arrangement of these processes includes lateralized, hierarchical, and parallel processing. Together these processes enable an organism to discriminate between colors, objects in motion, patterns, and dimensions. The basic anatomy of the eye, illustrated in Figure 1.1, is fundamental in understanding how the visual system works. As light enters the eye through the pupil, the cornea and lens focus the light onto the retina. Once the light (photons) reaches the retina, photoreceptor cells absorb the photons. The visual pigments convert these photons into an electrical

Figure 1.1 Anatomy of the human eye (McCarthy 2009)

4 signal that stimulates neurons in the retina. Retinal ganglion cell axons located in the optic nerve then transmit the visual image to the brain. The cornea is a transparent tissue of mainly collagen that protectively covers the iris, pupil, and anterior chamber (Oyster 1999). There are five layers that make up the cornea including the epithelium, Bowmans membrane, the stroma, Descemets membrane, and the endothelium. Together with the lens, the cornea refracts light and provides two-thirds of the eyes focusing power (Cassin and Solomon 1990; Goldstein 2007). However, this focus is fixed. Therefore, small adjustments to the eyes focus are controlled by the lens, where the curvature can be changed by the ciliary muscles. Located in the anterior segment, the lens also acts as a UV-filter containing compounds that absorb light from 295 to 400 nm (Dillon, Zheng et al. 1999; Gaillard, Zheng et al. 2000). Therefore, only light with wavelengths longer than 400 nm can reach the retina, protecting the retina from harmful UV-damage. The amount of light that enters the eye is controlled by the iris and pupil. The iris is a pigmented fibrovascular tissue located in front of the lens. At the center of the iris is a circular hole called the pupil. The iris regulates the size of the pupil, effectively changing the intensity of an image, the field of view, and the depth of focus. The retina is a complex seven-layered structure located in the back of the eye, as shown in Figure 1.2. Light initially enters through the ganglion cell layer and must travel through all cell types before reaching and activating the photoreceptor cells. The outer segments of both rods and cones contain the light-

Figure 1.2: The Retina (Molavi 1997)

6 sensing visual pigment and send signals to the cell bodies in the outer nuclear layer to axons. These axons in the outer plexiform layer connect with dendrites from bipolar and horizontal cells. The bipolar cells in the inner nuclear layer process the signal from the photoreceptor and horizontal cells and then send it to their axons. In the inner plexiform layer, the bipolar axons connect with ganglion cell dendrites and amacrine cells. The ganglion cells then send the signal with their axons from the ganglion cell layer through the optic fiber layer to the optic disc. An adult retina is on average 22 mm in diameter and .5 mm thick and contains approximately 7 million cone and 100 million rod photoreceptor cells. The outer segment of each photoreceptor cell contains an opsin-retinal complex known as the visual pigment. Each visual pigment contains the same chromophore of 11cis-retinal but the type of opsin differs between pigments. The visual pigment in rods, which are responsible for low light vision, is called rhodopsin. The remaining three pigments, responsible for color and bright light vision, are found in the different types of cone cells (Wald 1961; Brown and Wald 1964). These photoreceptor cells are unevenly distributed throughout the retina. Rods are located in the peripheral retina whereas cone cells are located almost exclusively in the fovea, which is responsible for high visual acuity. The area in and around the fovea that has a yellow pigmentation is called the macula (Kolb, Nelson et al. 2001). Damage to the macula can cause photoreceptor cells to die, which diminishes high visual acuity and leads to severe loss of vision. The 7th layer of the retina, the retinal pigment epithelium, is separated from the choroid by a thin layer of tissue known as Bruchs Membrane. The choroid

7 consists of four different layers, including: Hallers layer, Sattlers layer, the choroidal capillaries, and Bruchs Membrane. Each layer contains different size blood vessels that supply the eye with oxygen and nutrients. The choroidal capillaries contain the smallest blood vessels in the choroid. This structure controls the transport of oxygen, nutrients, and waste by passive diffusion (Olver 1990; Ramrattan, van der Schaft et al. 1994).

The Retinal Pigment Epithelium

The retinal pigment epithelium (RPE) is a pigmented single layer of cells that firmly attaches to the underlying choroid at the basal surface and weakly attaches to overlying photoreceptor cells at the apical surface (Cassin and Solomon 1990; Boyer, Poulsen et al. 2000). These smooth, hexagonally shaped cells, illustrated in Figure 1.3, are densely packed together with little extracellular space between each cell. Rivet-like structures, known as hemidesmosomes, connect the basal surface of the RPE to the basement membrane and may maintain the cohesive properties between the RPE and Bruchs Membrane (Miki, Bellhorn et al. 1975). The apical surface of the RPE cells contains microvilli, which are cellular membrane protrusions that increase surface area. These microvilli form a closefitting envelope around the ends of the photoreceptor cells (Schraermeyer and Heimann 1999). RPE cells also develop differently from most monolayer epithelial cells. Most epithelial cells develop adherin junctions several hours after cell-to-cell

Figure 1.3: The Retinal Pigment Epithelium cell structure showing the relationship between the RPE cell and Bruchs Membrane (Schraermeyer and Heimann 1999).

9 contact is made; however, fully developed RPE cells require weeks after confluence to develop mature junctions. In addition, RPE cells have been reported to express Ncadherin instead of E-cadherin (Lagunowich and Grunwald 1989; Davis, Bernstein et al. 1995; Huotari, Sormunen et al. 1995; Marrs, Andersson-Fisone et al. 1995; McKay, Irving et al. 1997), which is common in nonepithelial cells. However, Burke et al. reported that post confluent cultures of RPE cells contained both N- and E-cadherin (Burke, Cao et al. 1999). The expression of E-cadherin occurs in late confluency after N-cadherin is already present. Since E-cadherin is involved in desmosome assembly and Na/K ATPase polarity (Nelson, Shore et al. 1990; Marrs, Andersson-Fisone et al. 1995; Lewis, Wahl et al. 1997), the late expression of Ecadherin may be responsible for the absence of desmosomes in the RPE and the presence of RPE sodium pumps on the apical surface instead of the basolateral membrane (Burke, Cao et al. 1999). The RPE has numerous functions that are essential to maintaining the visual system. Photoreceptor cells are continuously renewed, synthesizing approximately 10 % of the outer segment each day (Young 1971a; Young 1971b). Because of the close proximity to photoreceptor cells, the RPE is responsible for the phagocytic uptake and break down of these shed photoreceptor outer segments. New discs are added to the base of the outer segment while the tips are engulfed and degraded by the RPE cells sending the waste to the choroidal capillaries. Approximately 3 billon discs can be engulfed by a single RPE cell in an average lifetime (Marshall 1987). This continual process of renewal and degradation of the photoreceptor cell outer segments is crucial in maintaining the viability of the photoreceptor cells because

10 they are continuously exposed to light and a relatively high oxygen concentration. The photoreceptor cells are susceptible to oxidative damage from reactive oxygen species (Winkler, Boulton et al. 1999). The RPE cell phagolysosmal system is highly efficient at digesting large quantities of the photoreceptor cells outer segment discs. These discs are visible in the RPE cytoplasm as membrane-bound vesicles known as phagosomes. Once inside the RPE, the phagosome can then fuse with a lysosome forming a phagolysosome, as seen in Figure 1.4. Under normal circumstances, hydrolytic digestion starts to break down the proteins, lipids, and polysaccharides within the phagolysosome. In young eyes, these molecules are generally reduced to 50 % of their original size in approximately 24 hrs. However, in older eyes, undigestible material from the phagolysosme, known as lipofuscin, starts to accumulate (FeeneyBurns and Eldred 1983; Feeney-Burns, Gao et al. 1988). As the eye ages, the number of photoreceptor and RPE cells disproportionately decreases, which results in a net increase in lipofuscin, melanosomes, and melanolipofuscin in the remaining cells. This overloads the RPE cells and decreases the cytoplasmic free space. The accumulation of lipofuscin, aging, and a variety of other factors changes the chemical composition within the RPE cells and increases oxidative stress. This can deleteriously affect their function and viability. RPE cells are also significantly involved in the visual cycle. Illustrated in Figure 1.5, the visual cycle involves the repeated movement of retinoids by the interphotoreceptor retinoid-binding protein (IRBP) between photoreceptor cells and

11

Figure 1.4: Formtion of phagolysosme and lipofuscin (Feeney-Burns and Eldred 1983)

12

Figure 1.5: The visual cycle (Cornwall 2009)

13 the RPE. Since free retinoids damage cells, the IRBP acts as a two-way carrier transporting retinoids through the interphotoreceptor matrix that separates the photoreceptor cells and the RPE (Pepperberg, Okajima et al. 1993). The visual cycle is initially activated when light is absorbed by rhodopsin, which causes the chromophore, 11-cis-retinal, to undergo photoisomerization to all-trans-retinal. Alltrans-retinal is released from rhodopsin and then reduced to all-trans retinol by alltrans-retinol dehydrogenase. The all-trans-retinol is then sent back to the RPE by IRBP to recharge. Once in the RPE, the all-trans-retinol is esterified by lecithin retinol acyltransferase and converted back to 11-cis-retinal by isomerohydrolase RPE65. Then, 11-cis-retinal is transported back to the photoreceptor cells by IRBP where it can bind with rhodopsin, restarting the visual cycle (Rando 2001). In addition to the phagocytic breakdown of photoreceptor cell outer segments and processing retinol in the visual cycle, the RPE has several other specialized functions including creating ion gradients within the interphotoreceptor matrix, uptake and storage of vitamin A for retinal synthesis, and building up the blood retinal barrier (Heller and Bok 1976; Bridges, Alvarez et al. 1982; Schraermeyer and Heimann 1999). The RPE also provides the only transport of oxygen, nutrients, and waste between the photoreceptor cells and choroid (Schraermeyer and Heimann 1999).

14 Bruchs Membrane

Bruchs Membrane is a thin extracellular membrane that is approximately 2-5 m thick depending on the age of the eye (Oyster 1999) and structurally separates the choroid from the RPE. Illustrated in Figure 1.6, the membrane is composed of five layers including the basal lamina of the RPE, the inner collagen layer, the elastin layer, the outer collagen layer, and the basement membrane of the choroidal capillaries. Although other small compounds are present, Bruchs Membrane is primarily composed of collagen, elastin, fibronectin, laminin, and heparin sulfate (Robey and Newsome 1983; Das, Frank et al. 1990). Bruchs membrane has a fundamental association with RPE cells that is mutually beneficial. Bruchs Membrane acts as a support for RPE cell attachment and survival. The RPE cells attach to the inner layer of Bruchs membrane, the basal lamina, which contains mostly laminin, Type IV collagen, and heparin sulfate. Therefore, a significant portion of the extracellular matrix (ECM) environment of RPE cells comes from Bruchs membrane. Components in Bruchs membrane ECM send signals that trigger cell differentiation, migration, and proliferation. Synergically with Bruchs membrane, the RPE maintains Bruchs membrane by synthesizing, regulating, and degrading ECM proteins. Changes within this ECM environment caused by oxidative stress, blue light damage, and lipofuscin or drusen accumulation can detrimentally affect the cell signals and disrupt cellular function

15

Figure 1.6: The diagram shows the position and layers of Bruchs Membrane (Anderson, Ozaki et al. 2001)

16 (Dorey, Wu et al. 1989; Winkler, Boulton et al. 1999; Sparrow, Nakanishi et al. 2000; Suter, Reme et al. 2000; Sparrow and Cai 2001; Liang and Godley 2003; Wang, Paik et al. 2005). In addition to providing a support and attachment site for RPE cells, Bruchs membrane also functions as a barrier that selectively filters nutrients between the RPE and choroidal capillaries (Hewitt, Nakazawa et al. 1989). Together with the closely packed RPE cells, Bruchs membrane forms the retinal blood barrier. This barrier is critical to maintaining the viability of the retina. Once the barrier is damaged, fluid can leak into the surrounding structures, causing damage to cells. Bruchs membrane also undergoes a variety of structural and compositional changes with aging. The thickness of Bruchs membrane increases with age primarily because of the accumulation of fibrous material within the inner collagen layer (Mishima 1978; Newsome, Huh et al. 1987). Collagens and their components within the lamina progressively start to cross-link, which decreases their solubility. The central elastin layer becomes less organized, undergoing fragmentation and calcification (Hogan 1972). In addition, previous studies have shown that the phospholipid, triglyceride, fatty acid, and free cholesterol content in Bruchs membrane exponentially increases with age. Curcio et.al have reported that there is a progressive accumulation of lipids and both esterified and unesterified cholesterol in Bruchs membrane (Curcio, Millican et al. 2001; Curcio, Presley et al. 2005). These cholesterol esters frequently oxidize generating cytotoxic oxysterols, which have proinflammatory properties (Fliesler 2002). The increase of these components

17 in Bruchs membrane may be related to the development of drusen (Coffey and Brownstein 1986). Drusen are deposits of extracellular material located between the basal lamina of the RPE and the inner collagen layer of Bruchs membrane (Figure 1.7). These deposits generally appear with age and are classified as either hard or soft with varying size, abundance, and shape. Hard drusen are generally small, round, and have well defined borders. Soft drusen, although larger than hard drusen, have varying size and shape, lack well defined borders, and generally have a pale yellow color (Davis, Gangnon et al. 2005; Ferris, Davis et al. 2005). The presence of soft drusen within the macula is a common indicator of AMD (Crabb, Miyagi et al. 2002). Although the exact composition of drusen is still unknown, several studies have reported the presence of lipids, carbohydrates, and proteins (Hageman, Luthert et al. 2001; Crabb, Miyagi et al. 2002). Newly formed drusen contain components that are similar to those found in aging Bruchs membrane, including membranebound vesicles that rupture, releasing granular and vesicular material. Neutral lipids found in drusen contain esterified and nonesterified cholesterol and account for approximately 3 % of the dry weight (Rudolf, Clark et al. 2008). The remainder is a mixture of proteins, carbohydrates, and cellular components of the RPE. The 129 proteins previously identified in drusen are a combination of blood, extracellular, and intracellular proteins (Rudolf, Clark et al. 2008; D'Souza, Jones et al. 2009). Recently, complement factors H and B gene sequence variants have been associated with an increased risk of developing AMD. Therefore, research focused specifically

18

Figure 1.7: Transmission electron microscope image of drusen (2002)

19 on drusen proteins involved in inflammation (immunoglobulins, factor X, C5, and C5b complex) and their relation to AMD has increased (Rudolf, Clark et al. 2008). In addition to lipids and proteins, current data indicates that RPE and photoreceptor cells are involved in the formation of drusen. Entire cells, cellular organelles and fragments, basal lamina from the RPE, lipofuscin, and melanin have all been found in drusen, suggesting a source of origin (Crabb, Miyagi et al. 2002). However, the exact relationship between RPE dysfunction and drusen formation is still unknown. Drusen may form as a result of RPE dysfunction initially caused by genetic and environmental factors, or the accumulation of drusen may actually cause RPE dysfunction by disrupting the exchange of nutrients and waste between the RPE and choriodal capillaries, leading to choroidal neovascularization (Sarks 1982; Sarks, Arnold et al. 1999; Chong, Keonin et al. 2005). The complex composition of drusen, including proteins, lipids, carbohydrates, and cellular components indicates that its pathogenesis may be a combined mechanism.

Lipofuscin

Lipofuscin is an autofluorescent heterogenous mixture present in the cytoplasm of postmitotic cells and is an ordinary morphological sign of aging (Feeney 1978; Kennedy, Rakoczy et al. 1995; Yin 1996). The composition of lipofuscin varies between different cell types. In the RPE, it results from the accumulation of undigestible material after phagocytosis of the photoreceptor cell outer segments (Feeney-Burns, Hilderbrand et al. 1984; Katz, Drea et al. 1986;

20 Boulton, McKechnie et al. 1989) and forms clusters of granules as seen in Figure 1.8. In an average lifetime, these granules can occupy approximately 20 % of the cytoplasmic free space (Haralampus-Grynaviski, Lamb et al. 2003). Although extensive research has been performed involving the mechanism of lipofuscin formation and its composition, the complex mixture is still relatively unknown. The mixture has both organic and water-soluble fractions that exhibit different fluorescence and UV-Visible absorption characteristics (Eldred and Katz 1988; Gaillard, Atherton et al. 1995). As the eye ages, the water-soluble portion increases (Rozanowska, Pawlak et al. 2004). Initially, Eldred and Katz were able to isolate 10 different fluorophores in human RPE extract (Eldred and Katz 1988). Two of these fluorophores were later identified as retinyl palmitate (Lamb, Zareba et al. 2001) and two isomers of a bis-retinoid pyridinium compound, A2E (2-[2,6dimethyl-8-(2,6,6-trimethyl-1-cyclohexen-1-yl)-1E, 3E,5E,7E-octatetraenyl]-1-(2hydroxyethyl)-4-[4-methyl-6-- (2,6,6-trimethyl-1-cyclohexen-1-yl)-1E,3E,5Ehexatrienyl]pyridinium) and iso-A2E, illustrated in Figure 1.9 (Eldred and Katz 1988; Sakai, Decatur et al. 1996). The chromophore was named A2E because the synthesis requires two equivalents of all-trans-retinal and one equivalent ethanolamine to produce the compound. Later, several other minor isomers of A2E were identified (Parish, Hashimoto et al. 1998). The mechanism for A2E synthesis in vivo is shown in Figure 1.10. Abundant in photoreceptor cells outer segments, all-trans-retinal reacts with phosphatidylethanolamine forming a Shiffs base known as NRPE. The NRPE reacts with a second molecule of all-trans-retinal, producing A2PE.

21

Figure 1.8: Transmission electron microscope image of lipofuscin granule (Haralampus-Grynaviski, Lamb et al. 2003)

22

Figure 1.9: Stucture of A2E and iso-A2E (Parish, Hashimoto et al. 1998)

23

Figure 1.10: Synthesis of A2E in vivo (Liu, Itagaki et al. 2000)

24 After the photoreceptor outer segment is phagocytized by the RPE, the phospholipid group is then hydrolyzed by the enzymes in the RPE lysosomes, forming A2E (Liu, Itagaki et al. 2000). The accumulation of lipofuscin has previously been suggested to cause retinal dysfunction by a variety of different mechanisms. Photochemically, lipofuscin can generate reactive oxygen species. Boultan et al. first reported that lipofuscin granules exposed to light generated significant amounts of superoxide ions (Boulton, Dontsov et al. 1993). The photogeneration of hydrogen peroxide and singlet oxygen was also observed in lipofuscin. Within the granules, free radicals such as superoxide ions are primarily responsible for the oxidation of lipids. Previous research reported that isolated granules exposed to visible light had a 30 % increase in lipid peroxidation when compared to controls (Rozanowska, JarvisEvans et al. 1995; Rozanowska, Wessels et al. 1998). In addition, lipofuscin accumulation in RPE cells has previously been studied in vitro and found to be photochemically toxic to RPE cells (Davies, Elliott et al. 2001). Lipofuscin is also capable of enzyme inactivation and causes a decrease in the phagocytic capacity of RPE cells (Sundelin, Wihlmark et al. 1998; Wassell, Davies et al. 1999). The accumulation of A2E, a major chromophore in lipofuscin, has a harmful effect on RPE cells. As seen in Figure 1.9, A2E structurally has two hydrophobic side chains connected by a positively charged pyridinium ring, making the molecule amphiphilic. This amphiphilic structure may disrupt the integrity of the RPE cell membranes. Sparrow et al. reported that when high concentrations of A2E are added to RPE cell membranes, the membranes become irregularly shaped and start to

25 bulge (Sparrow, Parish et al. 1999). Also, by acting as a detergent, A2E can disrupt the ATP-driven proton pumps, which causes the lysosomal pH to increase, preventing enzyme activity (Holz, Schutt et al. 1999; Bergmann, Schutt et al. 2004). In addition, the viability and function of RPE cells change with the accumulation of A2E. Under irradiation with blue light, A2E induces apoptosis in RPE cells and damages DNA (Sparrow, Nakanishi et al. 2000; Suter, Reme et al. 2000; Sparrow, Vollmer-Snarr et al. 2003). A2E was also reported to damage mitochondrial activity by inhibiting cytocrome c oxygenase (Suter, Reme et al. 2000; Shaban, Gazzotti et al. 2001), and cause membrane permeabilization inhibiting lysosomal function (Finnemann, Leung et al. 2002). Recently, the accumulation of A2E was suggested to cause specific phenotypic changes in the RPE, predisposing the retina to choroidal neovascularization (Iriyama, Fujiki et al. 2008). In addition to causing A2E-mediated damage, blue light causes oxidation of A2E. The oxidation of A2E results in the formation of three different types of compounds. The first group of compounds arises from the addition of oxygen to A2E, resulting in masses of 608 amu, 624 amu, and 640 amu. A total of nine oxygen additions have been observed (Ben-Shabat, Itagaki et al. 2002). These compounds were initially believed to form epoxides but later were shown to undergo rearrangement to the more stable furanoid oxides (Dillon, Wang et al. 2004). These products were identified in RPE cells that were fed A2E and in the organic soluble portion of human retinal lipofuscin. The second group of compounds results from the loss or addition of two hydrogen atoms from the oxidized A2E series described above. The third series of compounds were generated from cleavages along the

26 polyene chain in A2E and oxidized A2E, resulting in a series of lower molecular weight aldehydes (Figure 1.11) (Ben-Shabat, Itagaki et al. 2002; Avalle, Wang et al. 2004). These aldehydes are highly reactive and are more toxic to the cell than the oxidation products.

Oxidative Stress and the Antioxidant Glutathione

Oxidative stress results from an imbalance between the production of reactive oxygen species (ROS) and a systems ability to remove or repair the damage caused by these reactive intermediates. Normally, ROS such as superoxide anions, hydrogen peroxide, hydroxyl radicals, and singlet oxygen are produced at low levels within a cell and therefore, the damage they create can easily be repaired. Also, because of their involvement in redox signaling and their ability to kill pathogens, ROS can be extremely beneficial. However, high levels of oxidative stress can cause modifications to DNA, proteins, and lipids causing apoptosis or, in extreme cases, necrosis. Oxidative stress has been implicated in a variety of diseases including Alzheimers, Parkinsons and AMD. Because of the high oxygen consumption, the constant exposure to light, and the presence of polyunsaturated fatty acids, the retina is particularly susceptible to damage from oxidative stress (Beatty, Koh et al. 2000). Oxidative stress has been shown to cause damage to irradiated or hyperoxic tissues, suggesting that light irradiation and retinal damage are related. The high blood flow

27

HO

+
2 43 6 40

Corresponding aldehydes

472 432 406 366 A2E

2 47

HO N
+

Corresponding 488 aldehydes

448 422

382

Oxidized A2E

Figure 1.11: Structures of A2E and oxidized A2E with corresponding aldehydes identified (Wang, Keller et al. 2006)

28 through the choroidal capillaries and the phagocytosis of photoreceptor cells by the RPE generate a high concentration of hydrogen peroxide (Tate, Miceli et al. 1995). The accumulation of lipofuscin within the RPE also produces reactive oxygen species (Boulton, Dontsov et al. 1993; Gaillard, Atherton et al. 1995). Handa et al. reported that advanced glycation endproducts accumulate in Bruchs membrane with age, providing direct evidence that oxidative stress occurs in the vicinity of the RPE. In addition, RPE cells treated with hydrogen peroxide result in decreased expression of RPE cell markers (Alizadeh, Wada et al. 2001) and a disruption of RPE cell junctions and barrier integrity (Bailey, Kanuga et al. 2004). Several studies have suggested that antioxidants reduce the risk of developing or decrease the severity of AMD. Under normal conditions, enzymes maintain a reducing environment within the cell. Changes to this normal redox state can cause toxicity, resulting in damage to cellular components and disease. Previous studies have shown that decreased levels of the macular pigments lutein and zeaxanthin in the central retina result in an increased risk of developing AMD (Mozaffarieh, Sacu et al. 2003; Ahmed, Lott et al. 2005). A decrease in vitamin E was also shown to cause retinal degeneration (Hayes 1974). In addition, superoxide dismutase, catalase, and glutathione peroxidase were reported to have significantly lower levels in patients with AMD (Evereklioglu, Er et al. 2003). Glutathione (GSH) is a ubiquitous tripeptide that decreases in concentration with age, which has been associated with age-related chronic illnesses (Lang, Mills et al. 2000). Although the exact mechanism for the age-related decrease is unknown, a reduction in the enzyme activity of glutathione peroxidase, glutathione reductase,

29 and glutathione transferase are contributing factors (Sethna, Holleschau et al. 1982; Katakura, Kishida et al. 2004; Sastre, Martin et al. 2005). Involved in several critical cell processes, GSH is essential to maintain and regulate the redox status of a cell (Hammond, Lee et al. 2001; Schafer and Buettner 2001; Ballatori, Hammond et al. 2005). The ratio of GSH to glutathione dimer (GSSG), Figure 1.12, is an important indicator of the redox environment of a cell and changes to this ratio have been associated with cellular proliferation (Suthanthiran, Anderson et al. 1990), differentiation (Hansen, Carney et al. 2001), and apoptosis (Hammond, Lee et al. 2001). Within the cell, individual organelles have different redox requirements and therefore, different GSH:GSSG ratios. The endoplasmic reticulum has an oxidizing environment with a potential of -170 to -185 mV at pH 7.0, giving a ratio of GSH/GSSG of 1:1 to 3:1 (Hwang, Sinskey et al. 1992). The cytosol has a reducing environment with a potential of -290 mV at pH 7.0, giving a ratio of GSH/GSSG of 3,300:1 (Ostergaard, Tachibana et al. 2004). Isolated mitochondria have a redox potential from -250 to -280 mV at a pH 7.8, giving a GSH/GSSG ratio of 20:1 to 40:1 (Outten and Culotta 2004; Rebrin, Zicker et al. 2005; Shen, Dalton et al. 2005; Hu, Dong et al. 2008). Each cyclic voltammetry experiment used a standard hydrogen electrode. However, effectively measuring the GSH/GSSG ratio in mitochondria is difficult because separation of the matrix from the intermembrane space is nearly impossible and oxidiation of GSH often occurs during cell lysis and fractionation.

30

Figure 1.12: Structure of glutathione (GSH) and its dimer (GSSG) (King 2009)

31 Within the eye, GSH is found in high levels throughout the lens, cornea, aqueous humor, and retina, protecting the eye from oxidative damage. The agerelated decrease of GSH levels has been associated with the development of cataracts, glaucoma, and AMD. Nuclear cataracts progressively deteriorate as the oxidation of methionine and cysteine residues and the loss of thiol groups in structural proteins increase. High GSH levels are essential to protect the thiol groups from ROS. Therefore, a decrease in GSH with an increase in GSSG causes an imbalance in the GSH/GSSG ratio and extensive protein cross-linking and insolubility occurs (David and Shearer 1984; Calvin, Medvedovsky et al. 1986). Glaucoma results from an increase in intraocular pressure with progressive loss of retinal ganglion cells by apoptosis. The production of ROS and a decrease in GSH levels causes the apoptosis of the ganglion cells (Maher 2005). In patients with exudative AMD, the total thiol and GSH concentration significantly decreases (Coral, Raman et al. 2006), possibly because of insufficient GSH synthesis and recycling in RPE cells (Sternberg, Davidson et al. 1993; Cohen, Olin et al. 1994). Reduced antioxidant properties in the RPE cells may increase the RPE cells susceptibility to oxidative stress. Compromised antioxidant defense systems are associated with age-related eye diseases and therefore supplements replacing these diminished antioxidants may be therapeutically beneficial.

32 Inflammation and AMD

Chronic inflammation has been implicated in age-related diseases including Alzheimers disease, heart disease, atherosclerosis, and AMD. Inflammation accelerates the production of free radicals, which can normally be controlled by antioxidants. However, when inflammation becomes chronic, the accelerated production of reactive oxygen species (ROS) and reactive nitrogen species (RNS) leads to increased concentrations of these species and associated tissue damage. Inflammatory mediators such as prostaglandins are a major source for the production of superoxide, hydroxyl radicals, and hydrogen peroxide (Chung, Kim et al. 2001). The activated phagocytes, neutrophils and macrophages, can also produce significant quantities of superoxide and hydrogen peroxide while simultaneously activating nitric oxide synthase to produce large increases in nitric oxide (NO) (Carreras, Pargament et al. 1994; Miller and MacFarlane 1995). Although considered relatively nonreactive, NO can form more toxic intermediates including nitrite (NO2-), peroxynitrite (ONOO -), nitrogen dioxide (NO2), and dinitrogen trioxide (N2O3). These highly reactive intermediates interact with macrophages, hepatocytes, thiols, and a variety of enzymes causing DNA damage, neurotoxicity, and apoptosis (Dawson 1995; Ignarro 1996). When superoxide is simultaneously released with NO, peroxynitrite is produced, which can cause the oxidation of proteins, lipid peroxidation, and tyrosine nitration. Previous research reported that 60 % of bovine RPE cells died approximately 6 hours after treatment with peroxynitrite. Also, using an anti nitrotyrosine antibody, protein modification within

33 RPE cells was detected after treatment with peroxynitrite (Behar-Cohen, Goureau et al. 1996). The nonenzymatic nitration of long-lived proteins has been notably associated with inflammation (Bailey, Paul et al. 1998; Paik, Dillon et al. 2001). Paik et al. reported the nonenzymatic nitration of extracellular matrix proteins by nitrite at physiological pH (Paik, Ramey et al. 1997; Paik, Dillon et al. 2001). In addition, the nonenzymatic nitration of extracellular matrix proteins was reported to detrimentally affect RPE function and viability (Wang, Paik et al. 2005) and reduce the RPE phagocytic capacity (Sun, Cai et al. 2007). Hageman and colleagues were the first group to suggest that inflammation and AMD were related based on the presence of immune response proteins in drusen that were isolated from the retinas of AMD patients (Hageman, Luthert et al. 2001). Immunohistochemical studies later identified a variety of ultrastructural components within drusen, including: immunoglobulins, components involved in the complement system (C5a, C3, and C5b-9), molecules involved in the acutephase response to inflammation (vitronectin, Amyloid P, and clusterin), and proteins that maintain and regulate the immune response (fibronectin, ubiquitin, and apolipoprotein E) (Hageman and Mullins 1999; Hageman, Mullins et al. 1999; Mullins and Hageman 1999; Rodrigues 2007). In addition, drusen contain glycoprotein-rich domains from dendritic cells. Later, Johnson et al. proposed that drusen formation starts after RPE degeneration initiates dendritic cells, which causes the release of regulatory proteins and activates the complement cascade, evoking an immune response (Hageman, Luthert et al. 2001; Johnson, Leitner et al. 2001).

34 Direct evidence supporting the relationship between inflammation and AMD involved four independent genetic studies. The genomes of AMD patients examined all had the same inherited variant, Y402H, on the same gene called complement factor H (CFH), which significantly increases a patients risk for developing AMD. CFH regulates inflammation and therefore the inherited variant may result in an overactive inflammatory process. Since the identification of CFH, AMD susceptibility variants have also been identified in complement component 2, complement factor B, and complement component 3. In addition, chemokine receptor 1, toll-like receptor 4, and the major histocompatibility complex class 1 genes have been suggested to play a role in the development of AMD.

Advanced Glycation Endproducts and AMD

Advanced glycation endproducts (AGEs) are a heterogeneous collection of modifications, mainly oxidative, that result from the spontaneous reaction of aldehydes and proteins through the Maillard reaction (Figure 1.13). The Maillard reaction, or nonenzymatic glycation, refers to chemical reactions involving primary or secondary amines and carbonyl compounds. In biological systems, the main source of amines comes from N-terminal amino groups, primary amino groups on free amino acids, and the -amino group on lysine residues within proteins. The primary sources for carbonyl compounds are from reducing sugars such as glucose, fructose, and lactose. Initially, there is a nucleophilic substitution between the carbonyl on a reducing sugar and amino group within a protein producing a Schiffs

35

Figure 1.13: Maillard reaction (Koldunov, Kononov et al.)

36 base. The Schiffs base undergoes spontaneous rearrangement to produce a relatively stable Amadori product. The Shiffs base and Amadori product can then further react through polymerization, cyclization, enolization, and oxidation to produce numerous AGEs. The rate of AGE formation during aging is greater than the rate predicted by first order kinetics. The rate of the reaction is dependent on the pKa of the amino group, the location of the amino group within the protein, the electrophilicity of the carbonyl carbon, and the ratio of the sugar cyclic to acyclic form (Bunn and Higgins 1981; Baynes, Watkins et al. 1989; Labuza and Baisier 1992; Naranjo, Malec et al. 1998). Therefore over time, there is a significant accumulation of AGEs on long-lived proteins. The accumulation of these irreversible AGE adducts depends on the lifetime of the modified protein, oxidative stress, redox status, and the availability of metal ions. Modification of proteins by AGEs often leads to protein cross-linking, pigmentation, and fluorescence (Thorpe and Baynes 2003). The presence of oxygen can also influence the specificity and rate of AGE formation. Previous studies have reported that the initial rate of glycation and selectivity of amino groups within a protein are reduced in the presence of oxygen, which was attributed to competitive parallel glycoxidation reactions with reducing sugars (Yeboah, Alli et al. 1999; Yeboah, Alli et al. 2000). These competitive reactions decrease the concentration of the reducing sugars available to react through glycation, resulting in a decreased rate of the initial reaction. However, when transition metal ions and oxygen are both present, oxidation of reducing sugars easily occurs. For example, aldoses form glyoxal and glucosone, which are

37 more reactive oxidation products. Since these products are more effective at glycating primary amines and the guanidine group on arginine residues, the rate of glycation reactions will increase as their concentration increases (Hayase, Yamamoto et al. 1996). The formation and presence of AGEs has been reported to accelerate agerelated changes and contribute to age-related diseases including; arthritis, cataracts, diabetic retinopathy, and AMD. Previous literature has reported that AGEs accumulate in human BM and basal deposits. Specifically, carboxymethyllysine (CML) was the first AGE identified in BM and drusen from AMD patients (Ishibashi, Murata et al. 1998). CML and pentosidine, a fluorescent cross-linking AGE, were reported to increase in BM with age (Handa, Verzijl et al. 1999; Glenn, Beattie et al. 2007). AGEs have also been detected in the RPE as free adducts or as AGE-modified proteins in lipofuscin granules (Schutt, Bergmann et al. 2003). RPE cells that were grown on AGE-modified substrate accumulated an increased quantity of lipofuscin, which is related to a decrease in lysosomal enzyme activity (Glenn, Mahaffy et al. 2009). AGE receptors such as RAGE, AGE-R1, and AGER3 have also been reported to increase in the RPE and photoreceptor cells or BM of AMD patients (Howes, Liu et al. 2004; Gu, Yuan et al. 2009). The presence of RAGE in vivo has been associated with chronic inflammation. The activation of RAGE and AGEs changes CD59, a major regulatory protein, and increases the inflammatory response (Cheng and Gao 2005). In addition, AGEs also occur at relatively high concentrations in the membranes associated with choroidal neovascularization (CNV) (Swamy-Mruthinti, Miriam et al. 2002). Elevated in BM

38 of AMD patients, the AGEs CML and carboxyethylpyrrole, promote neovascularization in vivo by stimulating vascular endothelial growth factor (VEGF) (Kobayashi, Nomura et al. 2007), which is related to wet AMD. Several AGEs have also been reported to produce the expression of pro-angiogenic growth factor in RPE in vitro (Zhou, Cai et al. 2005). Therefore, AGEs have been implicated in the pathology of several retinal diseases, suggesting their potential benefit as critical biomarkers in diagnosing a patients susceptibility to these diseases.

Dissertation Research

Vision loss associated with AMD is currently the predominant cause of irreversible blindness in developed countries. As a result of the increased number of documented cases and the severity of the disease, research focused specifically on the origin and progression of the disease is essential in order to treat patients effectively. The characteristic central vision loss associated with AMD is caused by photoreceptor cell death. However, the exact mechanism leading to the death of these cells and the onset of AMD is still unknown. The retina consists of several layers including the neural retina (neurons and photoreceptors), retinal pigment epithelium (RPE), and Bruchs membrane. Recent research has suggested that agerelated changes within the RPE and underlying Bruchs membrane may play a crucial role in the development of AMD. These changes include the accumulation of debris called lipofuscin and its major chromophore, A2E, in the RPE and the

39 development of lipid-like deposits on and in Bruchs membrane from the RPE. Therefore, this dissertation will focus on investigation of these age-related changes in the RPE and Bruchs Membrane. One of the major contributors to detrimentally affect RPE cell viability is the accumulation of lipofuscin. However, the origin of lipofuscin granules is still unknown. Therefore, the structures and reactivities of the higher molecular weight, more hydrophobic relatives of A2E within lipofuscin granules, were investigated to identify the compounds and suggest possible sources of formation. In addition to damage caused by lipofuscin, RPE cells are also affected by alteration to Bruchs membrane including the accumulation of debris, chemical modifications, and compounds involved in inflammation. This study also focuses on the formation of A2E and A2E-related compounds within Bruchs membrane and modifications to extracellular matrix proteins by nitrite, glycolaldehyde, and A2E as possible sources for age-related changes observed in patients with AMD. These results will increase the understanding of biochemical and cellular changes occurring in RPE cells and Bruchs membrane in relation to AMD.

Chapter 2 MATERIALS AND METHODS

Materials

All chemicals used were of the highest possible purity commercially available. All solvents used were HPLC grade and were purchased from Thermo Fisher Scientific Inc. (Waltham, MA). All-trans-retinal, tryptophan, dithiothreitol, ammonium bicarbonate, urea, iodoacetamide, glycolaldehyde, formic acid, acetic acid, sodium nitrite, ammonium acetate, sodium chloride, sulfanilamide, N-naphylethylenediamine, hydrochloric acid, dichloro-diphenyl trichloroethane, triphenylamine, phenanthrene, benzophenone, and cinnamic acid, ferrocene, benzaldehyde, cinnamaldehyde, 3-nitrotyrosine, and the Cys-laminin chain were purchased from Sigma Aldrich Co. (St. Louis, MO). Ethanolamine was purchased from ACROS Organics (Pittsburgh, PA). The sequencing grade modified trypsin was purchased from Promega Corp. (Madison, WI). Water was purified by using a Millipore Milli-Q Plus PUREpak 2 (18.2 M) water purification system.

41 Instrumentation

The UV-Visible absorption spectra were obtained from the Ocean Optics spectrophotometer (Dunedin, FL). The HPLC system consists of Hewlett Packard quaternary pump (Ti series, 1050, Hewlett Packard, France) with a diode array detector. The column used for separations was a C-18 reverse phase (RP), 250 10 mm, and C-12 RP, 150 4.60, 4 m size columns from Phenomenex (Torrance, CA). For mass spectrometric analysis, a Thermo Finnigan LCQ Advantage with Surveyor LC-pump (Thermo electron, San Jose, CA) was used. Steady-state irradiation was performed using a Philips special blue light (Oriel, Stratford, CT, model number 6292) in a quarter-inch acrylic glass irradiating chamber. Electrospray Ionization Mass Spectrometry (ESI-MS) was used in all proceeding studies to analyze human retinal lipofuscin, Bruchs membrane, and the modifications to laminin and A2E. Electrospray ionization is a powerful technique usually coupled to mass spectrometry, which creates ions from a solution containing the analytes of interest at atmospheric pressure (Figure 2.1). The sample is injected through a sample loop or syringe into capillary tubing or emitter, where a high voltage is applied (2-5 kV). The liquid sample then reaches the tip of the emitter where a Taylor cone is formed. At the center of the cone a jet of liquid sample is emitted, which ends in a fan-shaped plume (Figure 2.2). The initial sample generally has acid such as acetic acid or TFA added to it to increase the conductivity of the solution, which decreases the size of the droplets initially formed. The charged droplets then undergo further nebulization by interacting with an inert gas such as

42

Figure 2.1: Electrospray Ionization (Gates 2004)

43

Figure 2.2: Taylor Cone (New Objective 2004)

44 nitrogen. The charged particles then enter the first vacuum area of the mass spectrometer through an ion transfer tube. This tube is heated, which heats the counter flow nitrogen gas, increasing evaporation of the charged droplets. The evaporation continues until the particles become unstable, reaching their Rayleigh limit. At this critical limit, Coulombic explosion occurs, creating even smaller droplets. This process continues until the analyte ions of interest are released from the droplets. The charge on the droplets and subsequent ions formed depends on the voltage initially applied. The charged ions are then carried to the mass analyzer, which is a quadrupole ion trap (Figure 2.3) in the LCQ Advantage. The trap is made up of two hyperbolic endcap electrodes with a ring electrode in between. Within the quadrupole ion trap, constant direct current (DC) and radio frequencies (RF) and oscillating alternating current (AC) electric fields are used to trap the ions. Ions are then separated and sequentially ejected based on the stability of their trajectories in the oscillating field. The ions are then carried to the detector, which is a continuous dynode electron multiplier in the LCQ Advantage (Figure 2.4). After leaving the mass analyzer, ions strike the starting electrode with enough energy to cause secondary emission. The emitted electrons are then accelerated down the multiplier where the electrons can strike again, producing even more electrons and amplifying the signal. The secondary electrons are eventually collected at the end of the electron multiplier at a second electrode known as the anode. The signal can then be recorded and displayed. The LCQ Advantage has several scan types that were used to analyze human retinal lipofuscin, Bruchs membrane, and laminin, which include single-stage full

45

Figure 2.3: Quadrupole Ion Trap (Gates 2004)

46

Figure 2.4: Electron Multiplier (Kvech 2000)

47 scans, two-stage full scans, selected reaction monitoring, and zoom scans. The single-stage full scan type has one stage of mass analysis. The ions formed from ESI are stored in the mass analyzer. Then, these ions are sequentially scanned out of the mass analyzer to produce a full mass spectrum. The two-stage full scan type has two stages of mass analysis. In the first stage, the ions from ESI are stored in the mass analyzer. Then, the ion of a certain mass-to-charge ratio, also called the parent ion, is selected and all other ions are ejected from the mass analyzer. The precursor ion is excited, causing collisions with the background gas (helium) that is present in the mass analyzer. These collisions cause the parent ion to fragment, producing fragment ions. These daughter ions are stored in the mass analyzer and then are sequentially scanned out of the mass analyzer to produce a full product ion mass spectrum (MS/MS or MS2). Selected reaction monitoring (SRM) is a two-stage technique in which precursor ion and fragment ions are monitored. In the first stage of mass analysis, the ions formed from ESI are stored in the mass analyzer. The parent ion is selected and all other ions are ejected from the mass analyzer. Then, the parent ion is excited and collides with helium. The collisions of the parent ion cause fragmentation, generating the corresponding daughter ions. The parent and corresponding daughter ions of interest are then stored in the mass analyzer and all other ions are ejected. These selected ions are then sequentially scanned out of the mass analyzer, producing the SRM product ion spectrum. Finally, to confirm the molecular weight and charge state of certain compounds, higher resolution zoom scans were also performed.

48 Methods

Synthesis of A2E

A2E for all reactions was prepared from all-trans-retinal and ethanolamine in acetic acid and ethanol as previously described by Parish et al. (Parish, Hashimoto et al. 1998). The mixture was stirred in the dark for three days at room temperature. After excess solvent was removed by drying under argon, the A2E was separated from the initial reaction mixture using a HP 1050 Ti HPLC and a C18 RP column. Using an isocratic gradient of MeOH:H2O (90:10) and a flow rate of 1.0 mL/min, the retention time of A2E was approximately 28 min monitored with a photodiode array detector at 430 nm, as shown in Figure 2.5. The concentration of the purified A2E was determined by measuring the absorbance at 439 nm using an Ocean Optics spectrometer, given an extinction coefficient of 36,900 L/molcm (Parish, Hashimoto et al. 1998). The absorption spectra for A2E and iso-A2E are displayed in Figure 2.6. After collection, the pure A2E fraction (Figure 2.7) was confirmed on the LCQ Advantage mass spectrometer using collision induced dissociation (CID) (Figure 2.8). The sample was then dried under argon and stored at -70 C for further analysis.

Isolation of Lipofuscin

Human RPE lipofuscin granules were extracted and isolated from donor

49

Figure 2.5: Chromatogram of the A2E reaction mixture using HPLC with PDA detection. A2E and iso-A2E are identified.

50

1000 800 600 mAU 400 200 0 200 250 300 350 400 nm 450

iso A2E

A2E

500

550

600

Figure 2.6: The UV-Vis spectra of A2E and iso-A2E

51

7.00E+008 6.00E+008 5.00E+008

592.5

Intensity (AU)

4.00E+008 3.00E+008 2.00E+008 1.00E+008 0.00E+000

200

300

400

500

600

700

800

900

m/z

Figure 2.7: The mass spectrum of purified A2E (m/z 592)

52

MS/MS 592
2.0x10 1.8x10 1.6x10 1.4x10
6

418.4

HO N

2 40

44

Intensity (AU)

1.2x10 1.0x10 8.0x10 6.0x10 4.0x10 2.0x10

46 8

15

442.4 402.4 376.4 364.4 352.4 486.5 468.4 536.5 592.6 576.5

0.0 300

400

500

600

m/z

Figure 2.8: The MS/MS spectrum of purified A2E (m/z 592)

124

418

190

4 17

53 globes (Midwest Eye Banks and Transplantation Centers, Chicago, IL) as previously described by Feeney-Burns (Feeney-Burns and Eldred 1983). Extraneous fat and muscle were removed from the periphery of the eyeballs to reduce contamination by other cells and to aid in dissection. The outside of the sclera was cut mid-coronal, approximately one centimeter posterior to the cornea, with a razor blade. An incision was made through the sclera and fine-tipped scissors were used to cut the eyeball following the initial path from the razor blade. After the eyeball was cut into two parts, the anterior portion of the eye was removed and discarded along with the vitreous humor. A cold (4 oC) phosphate (0.1 M) buffered sucrose solution (0.32 M) was pipetted into the eyecup so that the level was below the edge of the incision (to prevent contamination of the eyecup) and the neural retina was allowed to float up in the sucrose solution. The optic nerve was cut at the base where it enters the interior of the eye and the neural retina was removed. The eyecup was filled with approximately 1 mL of the sucrose solution and brushed gently with a camel hair paint brush to remove the RPE cells, which were placed into a 15 mL centrifuge tube. The eyecup was repeatedly brushed until all the cells appeared to be removed and the sucrose solution appeared clear. The solution was then pipetted out of the eyecup into the centrifuge tube, which was subsequently centrifuged for 5 min at 100X G using a Beckman J2-HS centrifuge and a JA-20 rotor to remove unbroken cells and melanin. The supernatant was then layered on a 3 step sucrose density gradient of 0.63, 1.37, and 2.25 M, which was then centrifuged for 15 min at 8,500X G. The interface between the top two layers was then removed, lyophilized, and frozen at -70 oC (Figure 2.9). To obtain the organic-

54

Figure 2.9: Isolation of Lipofuscin (Feeney-Burns and Eldred 1983)

55 soluble portion of lipofuscin, the samples were re-suspended and a Folch extraction was performed using a 1:1:1 ratio of CHCl3:CH3OH:H2O. The organic soluble portion was then dried under argon, resuspended in 1 mL MeOH, and analyzed using ESI-MS/MS with with simultaneous PDA detection.

Auto-Oxidation of A2E

After synthesis and purification, 2 mL aliquots of 15 M A2E was transferred to a 4 C refrigerator in the dark. To determine the products formed from the auto-oxidation of A2E over time, an aliquot (20 L) was removed from the sample at times 0, 30 and 60 days and analyzed on ESI-MS/MS with simultaneous PDA detection.

Lipofuscin and A2E LC-MS Analysis

All samples were analyzed on a Thermo Finnigan LCQ Advantage mass spectrometer. The mass spectrometer was set to positive ion mode with a capillary temperature of 200 C, source voltage of 4.0 kV, capillary voltage of 42 V, and a tube lens offset of 50 V. The mass-to-charge ratios were collected from 200 to 2000 and a normalized collision energy between 30-40 % was used for the MS/MS data. The lipofuscin and purified A2E samples were separated using the Surveyor LC system with a Synergi Max-RP C12 column. The flow rate was set to 0.2 mL/min

56 with a mobile phase of MeOH balanced with H2O (both containing 0.1 % formic acid) using a gradient of 80 % MeOH for 30 min and 80-100% MeOH for 90 min.

Determination of the Water-Octanol Partition Coefficient of A2E: Log P

A stock solution of purified A2E was determined to have a concentration of approximately 5 x 10-5 M by UV-Visible spectroscopy in methanol at 439 nm. A 0.5 mL aliquot of the purified A2E was added to 5.00 mL of octanol in a 30 mL separatory funnel. The separatory funnel was inverted several times and allowed to equilibrate for one hour. Next, the two layers were separated and collected for HPLC analysis. Triplicate injections of each layer were performed on an Hewlett Packard 1050 HPLC with a 100 L sample loop and a Phenomenex Synergi (4 m, 15 mm x 5 mm) column while monitoring the absorbance at 439 nm. The peak area of each injection was recorded and used for calculation of the partition coefficient. The theoretical Log P value was also calculated using the Sparc software program and was then compared to the experimental Log P value for A2E. To determine the approximate partition coefficients for the higher molecular weight compounds located in the lipofuscin samples, A2E, dichloro-diphenyl trichloroethane, triphenylamine, phenanthrene, benzophenone, and cinnamic acid were separated by HPLC using a flow rate of 1.0 mL/min on a C12 RP column and an isocratic gradient of 90:10 MeOH:H2O while monitoring the UV-Vis. The elution times and known partition coefficients were plotted to form a linear leastsquares calibration curve (Figure 3.15), which was then used to determine the

57 approximate partition coefficients for the higher molecular weight compounds in the lipofuscin sample based on extrapolated linear correlation to the compound elution time.

Cyclic Voltammetry

All voltammograms were obtained using a three electrode system comprising a platinum button working electrode, a platinum wire counter electrode, a silver wire as a quasi-reference electrode, a BioAnalytical Systems, Inc. CV27 Potentiostat, and an analog-to-digital converter equipped with computer data acquisition. The scan rate was set at 50 mV/s. Also, each scan was started at 0.0 V in forward (positive potential) direction and the gain was constant throughout the experiments. All half-cell potentials were reported with respect to the ferrocene/ferrocenium half-cell potential (vs. Ef(Fc+/Fc)) for reversible reactions. For non-reversible reactions, the simple peak potential was reported vs. Ef(Fc+/Fc). A background solution containing 0.1 M tetraethylammonium perchlorate (TEAP) in acetonitrile was prepared. All acetonitrile was dried using 5A molecular sieves prior to electrochemical analysis. This solution was then analyzed with the CV27 potentiostat to determine the working potential window. The potential window used for analysis was from -1.75 to 1.6 V. An internal standard containing solution was prepared containing 0.05 M ferrocene and 0.1M TEAP in acetonitrile. This internal standard containing solution was then analyzed using the above system with identical analysis parameters to identify the position of the ferrocene/ferrocenium

58 redox couple. Also, this solution was used as the solvent to prepare the analytecontaining solutions. To prepare these solutions, the appropriate masses of the analytes--benzaldehyde (0.05 M), cinnamaldehyde (0.05 M), and all-trans retinal (0.015 M)--respectively--were added to a 100 mL volumetric flask and diluted to volume using the internal standard containing solution.

Reaction of A2E with Retinaldehyde (RAL)

After synthesis and purification, a 5 mL aliquot of 50 M A2E was mixed with 100 M RAL and a catalytic amount of acetic acid. The mixture was then bubbled with argon and irradiated for 1 h. An aliquot was then removed and diluted with methanol and analyzed on ESI-MS/MS with simultaneous PDA detection.

Separation of a Compound with m/z 920 from A2E RAL Reaction Mixture

Once the A2E and RAL reaction was complete, a 200 L aliquot was injected onto a HP 1050 Ti HPLC using a C18 RP column. Using a flow rate of 1 mL/min, the compound with m/z 920 eluted at approximately 35 min with a gradient of 80:20 MeOH/H2O for 15 min followed by a linear increase to 100% MeOH over 10 min, which was then maintained for an additional 35 min. The peak that eluted at 35 min was collected and confirmed using direct injection on the LCQ Advantage mass spectrometer. This procedure was then repeated multiple times

59 until approximately 5 mg was collected. The sample was then dried and resuspended in deuterated chloroform for future analysis.

Bruchs Membrane Preparation

Donor globes were purchased from Chicago Eye Bank (Midwest Eye Banks and Transplantation Centers). BM tissues from different decades, including donors of 18, 40, 50, 60, 70, and 80 years of age, and patients diagnosed with dry AMD were used. The preparation of BM followed the method described by Karwatowski et al.(Karwatowski, Jeffries et al. 1995). The eye globe was opened by circumferential incision along the iris. The lens and vitreous humor were separated. The neuronal retina was removed and the BM and choroid complex was incubated in 0.01 vol% trypsin in 10 mM phosphate buffer solution (PBS, pH 7.4) for 10 min at 37 oC and subsequently rinsed in PBS. The tissue was then gently brushed to remove the RPE cells and choroidal tissue. According to Karwatowski et al. (Karwatowski, Jeffries et al. 1995), this treatment removed most of the debris from the basement membrane and left only Bruchs membrane and some choriodal capillaries, and only a small amount of collagen (4%) was released during this treatment. Once the RPE was removed, BM was gently cut out.

60 Preparation of Organic Soluble Materials from Bruch's Membrane

Isolated Bruchs membrane was cut into small pieces and placed in a homogenizer. An equal amount of CHCl3:CH3OH:H2O was added and gently homogenized to extract the organic soluble components. Glass wool was inserted into a Pasteur pipette and the homogenized Bruchs membrane sample was filtered through the pipette, separating the solid from the supernatant. The organic layer of the extract was separated from the water-soluble layer by decanting. The organic supernatant was centrifuged for 15 min at a speed of 5000 rpm. The supernatant from the centrifuged solution was collected and the excess solvent was evaporated under argon. Approximately 50 L of methanol was added to the dried extract and 20 L of the extract solution was injected and analyzed by liquid chromatography tandem mass spectrometry (LC-MS/MS).

Bruchs Membrane LC-MS Analysis

To investigate the synthesized nitro-A2E and organic solvent extracts of Bruch's membranes, samples were dissolved in methanol as described above and analyzed by LC-ESI-MS/MS. The conditions for mass spectrometry for the organic soluble extract of BM were: positive polarity, capillary temperature of 200 oC, source voltage of 4.5 kV, capillary voltage of 43 V, and tube lens offset of 50 V, m/z range: 200-1,000, and a normalized collision energy of 25%. The separation was carried out on a 1504.6 mm Synergi Max-RP C12 column using a linear

61 gradient of 85% to 96% methanol for 60 min and 96%-100% methanol for 10 min with a balance of water containing 0.1% trifluoro acetic acid (TFA) and a flow rate of 0.3 mL/min. For synthesized nitro A2E analysis, the separation was carried out using an isocratic mobile phase of 5% methanol for 10 min and linear gradient of 5100% methanol for 30 min balanced with water with 0.1% formic acid and a flow rate of 0.3 mL/min (monitored at 430 nm, 350 nm, and 250 nm). The compounds with m/z values of 592, 637, 653 and 682 were selected for subsequent MS/MS scans using normalized collision energy of 52%. These are the molecular weights of A2E, nitrated A2E, nitrated A2E plus one oxygen, and A2E with two sites of nitration. The mass spectrometer was set as source voltage 4 kV, capillary voltage 3.3 V, capillary temperature 200 oC, and tube lens voltage of 25 V.

Acid Hydrolysis

Bruchs membranes from different decades of life were pooled into three samples including < 25 yrs, 40-60 yrs, and >65 yrs. These samples were dissected and prepared as previously described and then hydrolyzed in 6 M HCl at 110 oC for 24 hours using homemade glass tubes with Teflon-lined screw caps. Before hydrolysis, deoxygenation of the samples was achieved by six freeze-pump-thaw cycles. After samples were placed in the tubes, air was removed by applying a vacuum for approximately 5 min. After hydrolysis, excess acid was evaporated using argon gas. The samples were then resuspended in 50L H2O and spiked with

62 50 L of 100 M 3-nitrotyrosine. The samples were then analyzed by LC-MS and the concentration was calculated using standard addition.

Bruchs Membrane LC-MS Analysis After Acid Hydrolysis and Standard Addition of 3-Nitrotyrosine

HPLC separation was performed using a Synergi Max-RP C12 column (150 4.6 mm). To analyze the acid hydrolysates of Bruch's membrane, the LC mobile phase was acetonitrile (ACN) balanced with H2O (both containing 0.1% TFA) with the following gradients: 1-10% ACN for 50 min, 10-60% ACN for 30 min, 60100% ACN for 20 min and a flow rate 0.2 mL/min. The conditions for mass spectrometry (Thermo Finnigan LCQ Advantage and Surveyor LC system, San Jose, CA) were: positive polarity, capillary temperature of 200 oC, source voltage of 4.5 kV, capillary voltage of 43 V, and tube lens offset of 50 V, m/z range: 200-1,000, normalized collision energy of 30% to investigate whether 3-nitrotyrosine (m/z of [MH]+ is 227.1) was present within the sample. The MS method contains one zoom scan (m/z 222.1-232.1), one MS2 scan with a parent mass of 227.1 and a selective reaction monitoring (SRM) scan with a parent mass of 227.1 and a fragment mass of 181.1 (corresponding to loss of a nitro group).

63 Conditions of Tryptic Digests for Laminin Samples

Enzymatic digests were performed on all laminin samples. Each protein was prepared to have a concentration of 1mg/mL in water. A solution of 8 M urea and 0.4 M ammonium bicarbonate (pH 7.5- 8.5) was prepared as the digestion buffer. An aliquot of 150 L of the protein was added to 200 L of the urea and ammonium bicarbonate solution. A 50 L sample of 50 mM dithiothreitol was added to the sample and then allowed to incubate at 50 C for 15 mins. After cooling to room temperature, a 50 L sample of 100 mM iodoacetamide was added to the protein sample and left to react in the dark for 15 mins. The sample was then diluted by adding 350 L of Milli-Q water. The Promega grade trypsin was suspended in 200 L of 50 mM acetic acid and a 25 L aliquot was removed and added to the diluted protein sample. The sample was then allowed to incubate at 37 C overnight or up to a total of 24 hrs.

Modifications with Glycolaldehyde to Laminin

The Cys-laminin chain, CSRARKQAASIKVAVSADR, was dissolved in 1 mL of Milli-Q water and divided into two equivalent samples. The first sample was digested with trypsin for 18 hrs and then dried under argon. Glycolaldehyde modified laminin was prepared by adding 150 L of 50 mM glycolaldehyde solution to 150 L of 0.1 mg/mL Cys-laminin chain. The concentration of glycolaldehyde was selected based on previous literature reports regarding the

64 glycation of proteins with glycolaldehyde (Nagai, Matsumoto et al. 2000; Nakajou, Horiuchi et al. 2005). The mixture was then incubated for 12 hrs at 37 C. All aliquouts were dialyzed using PBS (1mM KH2PO4, 10 mM Na2HPO4, 137 mM NaCl, 2.7 mM KCl, pH 7.4) to removed unreacted glycolaldehyde.

Modifications to Laminin with A2E

An additional sample of 1.0 mg/mL Cys-laminin chain was dissolved in 1 mL of Milli-Q water. The purified A2E (1 mL, 18 M) was then added to the laminin peptide in the dark. The mixture was divided into two equivalent aliquots. The first aliquot was kept in the dark at room temperature for 60 min. The second aliquot was irradiated through a quarter-inch piece of acrylic glass with a Phillips Special Blue bilirubin bulb for 60 min. The bilirubin bulb produces a narrow bandwidth of blue light approximately 420-480 nm that is used to treat hyperbilirubinemia (Sarici, Alpay et al. 1999).

Modifications to Laminin with NaNO2

A suspension of 1 mg of laminin in 200 mM NaNO2 or 200 mM NaCl dissolved in 10 mM phosphate buffer (pH 7.4) was prepared. Both samples were incubated in the dark for approximately 7 days at 37 C. The excess salt was then removed by dialysis against 10 mM phosphate buffer. The dialysis was stopped once the solution surrounding the dialysis tubing was Griess assay negative for

65 nitrite modification (Tsikas, Gutzki et al. 1997; Romitelli, Santini et al. 2007). Initially, sulfanilamide (2mg/ml) and 4 N HCl were added to 1 ml aliquot of the dialysis solution in a 1:1:1 ratio. Next, 1 ml of N-naphyl-ethylenediamine (NED) (1mg/ml) was added to the previous mixture. The nitrite in the solution reacts with the sulfanilamide in acid to form a diazonium salt. The salt then reacts with the NED, producing a stable azo compound, which has an intense purple color. The dialysis solution was determined to be Griess assay negative once the solution remained clear after the addition of these compounds. The proteins within the dialysis tubing were then removed and a tryptic digest was preformed followed by analysis with LC-MS.

LC-MS Analysis of Laminin Samples

All samples were prepared in triplicate and then separated and analyzed on a ThermoFinnigan LCQ Advantage and Surveyor LC system using a Synergi Max-RP C12 column. The mass spectrometer was set to positive polarity, a capillary temperature of 200 C, source voltage of 4.0 kV, capillary voltage of 42 V, and a tube lens offset of 50 V. The mass-to-charge ratios were collected from 200 to 2000 and a normalized collision energy of 35 % was used for the tandem mass spectrometry data. The laminin control and glycolaldehyde modified laminin samples were separated using a flow rate 0.2 mL/min and a mobile phase of MeOH balanced with H2O (both containing 0.1 vol% formic acid) with the following gradient: 1-10% MeOH for 60 min, 10-70% MeOH for 60 min, and 70-100% for 60

66 min. The A2E and laminin samples were separated using the same flow rate and mobile phase but the gradient started with 40-70% MeOH for 60 min, 70-90% MeOH for 60 min, and 90-100% MeOH for 60 min. For each sample, a datadependent method was designed to acquire one full MS scan and three MS/MS scans for the three most abundant peaks in the full MS scan. The data generated from the mass spectrometer was then analyzed using information regarding enzyme digests, Protein Prospector, and SEQUEST software. The data from all control samples is provided as supplemental material.

Protein Prospector (http://prospector.ucsf.edu/prospector/mshome.htm)

Protein Prospector is a proteomics tool for searching sequence databases to compare data from mass spectrometry experiments. The MS-Digest option was used to obtain and compare the results from the enzymatic digest for laminin control. The minimum and maximum fragment masses were set to 100 and 4000 Da with a minimum fragment length of one amino acid. The enzyme was set to trypsin with a maximum number of missed cleavages set to 5 and multiple charges reported. The peptide fragment entered was CSRARKQAASIKVAVSADR with the instrument set to ESI-ION-TRAP.

67 Bioworks Browser

The Bioworks browser program 3.1 enables the analysis of raw data files generated in X-calibur. All chromatograms were analyzed using the pepmap and pepmatch software within the Bioworks browser software package. Pepmap was used to identify separated digest fragments of peptides resulting from enzyme digestion. Pepmap matches the acquired spectrum against the predicted digest fragment masses. The parameters were set to 5 % threshold, a scan width of 1, a mass tolerance of 1.5, and a maximum of 5 incomplete digest with no disulfide bonds. Pepmatch was used to predict the product ions of a peptide analyzed by CID. Pepmatch calculates the mass-to-charge ratio of each predicted fragment and matches those masses to peaks in the displayed mass spectrum. The parameters were set to display and compare generated B and Y ions, with a threshold of 2 %, and multiple charges. The identification of these fragments was then compared and confirmed using SEQUEST.

SEQUEST

SEQUEST is a proteomics tool which correlates tandem mass spectra data with amino acid sequences from protein databases. A protein database was generated for the laminin fragment and used as the reference. The parameters for a positive match on SEQUEST were set to a delta correlation (DelCn) of 0.1, a preliminary score (Sp) of 200, and the ion probability of 70 % coverage. The cross

68 correlation value (Xcorr) was set to the standard 1.9 for +1, 2.2 for +2, and 3.75 for +3 charged peptides. Modifcations from nitrite were identified based on the mass addition of 45 and corresponding absorption spectra from the PDA output. The modifications from glycolaldehyde were based on the mass additions of 42 and 102, which correspond to the addition of one or two molecule of glycolaldehyde with the loss of water. The modifications from A2E had to be confirmed by identifying the peaks that were not present in the control and analyzing the peaks in the MS/MS data set.

Data Analysis

A standard t-test was used for all statistical analysis with a p<0.05 indicating that the difference between groups was statistically significant. In addition, ANOVA one-way statistical analysis with a 95 % confidence level was performed on the Bruchs membrane samples from different decades.

CHAPTER 3 THE COMPOSITIONAL STUDIES AND MOLECULAR MODIFICATIONS OF HUMAN RPE LIPOFUSCIN

Introduction

As organisms age, many metabolically active post mitotic cells accumulate autofluorescent lysosomal storage bodies known as lipofuscin. Lipofuscin is a brown-yellow, electron-dense, aging pigment that is composed of a complex heterogeneous mixture of lipid-protein aggregates that form clusters of granules in the RPE. Within the human eye, these granules are believed to be formed from the indigestible material of phagocytized photoreceptor outer segments (Feeney-Burns and Eldred 1983; Boulton, McKechnie et al. 1989) and may account for up to 19 % of the cytoplasmic volume by the age of 80 (Feeney-Burns, Hilderbrand et al. 1984; Weiter, Delori et al. 1986; Davies, Elliott et al. 2001). Lipofuscin has also been shown to generate a series of reactive oxygen species (ROS), which include singlet oxygen, hydrogen peroxide, and superoxide anions (Boulton, Dontsov et al. 1993; Gaillard, Atherton et al. 1995; Rozanowska, Wessels et al. 1998; Wassell, Davies et al. 1999; Davies, Elliott et al. 2001). Considered photochemically toxic, lipofuscin was found to decrease phagocytic

70 capacity (Sundelin, Wihlmark et al. 1998). Previous studies have shown that both the isolated granules and the organic soluble extract of lipofuscin are extremely photoreactive (Gaillard, Atherton et al. 1995; Rozanowska, Wessels et al. 1998; Winkler, Boulton et al. 1999). One of the major fluorophores of lipofuscin, A2E, has been extensively studied since it was first isolated by Eldred et al. Structurally, A2E is a pyridinium bis-retinoid, which is synthesized using two moles of all-trans-retinal (RAL) and one mole of ethanolamine (Eldred and Katz 1988; Eldred and Lasky 1993; Parish, Hashimoto et al. 1998). After A2E could be chemically synthesized, research focused on the effect A2E had on cellular function. Previous literature reported that visible and UV radiation can cause lesions on the neural retina and RPE cells. The RPE cells are unusually susceptible to damage by wavelengths corresponding to the blue region of the visible spectrum, which is where A2E has the strongest absorbance. Previous research has reported that RPE cells fed A2E were severely damaged or killed after they were irradiated with blue light. Another possible source of light-induced A2E mediated damage is related to the photo-oxidation of A2E, which generates ROS, such as peroxide and superoxide radicals (Reszka, Eldred et al. 1995; Ragauskaite, Heckathorn et al. 2001). BenShabat et al. were the first to propose that the photo-oxidation of A2E results in higher molecular weight compounds that differ by 16 amu, resulting in multiple epoxide formation along the polyene chain (Ben-Shabat, Itagaki et al. 2002). However, because of the acidic environment within a lysosome, the allylic epoxides would be unstable and undergo rearrangement. Dillon et al. proposed that these

71 allylic epoxides of A2E would rearrange forming a furanoid oxide structure, which is relatively stable (Dillon, Wang et al. 2004). Furthermore, oxidative cleavage of side chains results in the formation of highly reactive aldehydes and ketones, which could readily react with cell constituents and cause irreversible damage (Wang, Keller et al. 2006). This explanation is based on the structural similarities between carotenoids and A2E. In addition, the amphiphilic structure of A2E may be responsible for detergent-like action in membrane disruption. The quaternary amine structure of A2E may also aid in the inhibition of lysosomal function by complexing to specific lysosomal enzymes (Eldred and Katz 1988). Even though the formation and composition of lipofuscin and the major fluorophore A2E have received notable attention, the origin of the granules and the identity of most of the compounds and the consequence of A2E accumulation within the granules are still unknown. One hypothesis suggested that A2E could exist in a free or esterified form. In the RPE, all-trans retinol, produced from the visual cycle, is converted to all-trans retinyl ester, which then self-aggregates into a retinosome (Imanishi, Gerke et al. 2004). This prevents hydrophobic interactions with cellular components that would disrupt normal cell function. Since A2E is extremely hydrophobic and accumulates within RPE lysosomes, A2E was suggested to undergo a similar esterification reaction (Mandal 2008). In addition to the esterification reactions, a second hypothesis involving the modification of A2E by A2E derived aldehydes was also suggested. Within the acidic lysosomal environment, A2E undergoes rearrangements and oxidation, generating aldehydes and ketones that are structurally similar to -carotene oxidation products

72 (Sommerburg, Langhans et al. 2003). These aldehydes are extremely reactive and in the presence of A2E may interact, forming higher molecular weight products. Therefore, in this study, lipofuscin was analyzed using a reversed-phase HPLC with an electrospray ionization mass spectrometer (ESI-MS) to investigate the hydrophobic compounds that elute later than A2E and that absorb radiation with wavelengths greater than 400 nm (Figure 3.1). The results indicate that a large quantity of the components of lipofuscin have mass spectra analogous to that of A2E, but with higher molecular weights as determined by their fragmentation pattern with losses of 190, 174 and/or 150 amu and the formation of fragments of ca 592 amu. The vast majority of the relatively hydrophobic components correspond to derivatized A2E with discrete molecular weights of 800-900 m/z, 970-1080 m/z and above 1200 m/z regions. These modified components increase the hydrophobicity of A2E and may explain the formation of lipofuscin granules in the RPE. The present study is part of a continuing effort to identify the molecular modifications to the structure of A2E (Dillon, Wang et al. 2004; Wang, Keller et al. 2006) and their mechanisms of formation.

Results

To study the composition of lipofuscin, samples were isolated from donor globes and analyzed on LC-MS as previously described in Chapter 2. The total ion chromatogram (TIC) and total absorbance from the Folch extract of lipofuscin granules are displayed in Figure 3.1. The chromatogram consists of A2E, oxidized

73

Figure 3.1: The TIC from the Folch extract of lipofuscin granules (top) and the corresponding PDA chromatogram (bottom) are shown. The chromatogram consists of A2E, oxidized A2E, and a complex mixture of components

74 A2E, and a complex mixture of components. Integration of the peak areas indicated that A2E comprised only approximately 5-10 % by volume relative to the complex mixture. We assumed that all molecules in the mixture have similar ionization efficiency and all of the instrumental parameters, flow rate and solvent composition remained constant. After further analysis of the mass spectral data of compounds that eluted from 50-100 mins, the mass spectrum revealed a series of closely related compounds that differed by a mass of 14 amu. Representative mass spectra of compounds that eluted at approximately 60 mins and 80 mins are displayed in Figures 3.2 and 3.3, respectively. There are three clusters of eluting masses, labeled I, II, and III, in the range of 800, 1000, and 1400 amu that exhibit the characteristic addition of 14 amu. To determine if these components were structurally related to A2E, the MS/MS data was analyzed. The MS/MS spectrum and the total absorbance of A2E are displayed in Figures 3.4 and 3.5, respectively. The fragmentation pattern

displayed characteristic losses of 150, 174, and 190 amu from the parent ion mass of 592 amu. These distinctive cleavages are illustrated in Figure 3.6. Once identified, these losses were compared to the MS/MS data of the components located within the complex mixture of the lipofuscin sample and all of the material that eluted between 50-80 mins and approximately 50 % of the material between 80-110 min had analogous spectra. Figures 3.7 and 3.8 display the MS/MS and absorbance spectrum of peak with m/z 814, which is representative of components that eluted at approximately 62 min in the TIC displayed in Figure 3.1. Once analyzed, Figure 3.7 displayed ions

75

2.0x10 1.8x10 1.6x10 1.4x10

862.8 874.9 860.8 847.9 876.9 878.8 971.0 831.0

1020.8 998.9 1022.9

Intensity (AU)

1.2x10 1.0x10 8.0x10 6.0x10 4.0x10

814.2

2.0x10

0.0 700 750 800 850 900 950 1000 1050 1100 1150 1200

m/z

Figure 3.2: The mass spectrum of the Folch extract of human lipofuscin eluted at time 62.93 mins. Groups I, II, and III identify the related clusters of higher molecular weight compounds with mass to charge ratios of approximately 800, 1000, and 1400, respectively. Highlighted in red are the additions of 14 amu starting with m/z 847.9.

76

1.4x10 1.2x10 1.0x10

6

948.9

927.0 1083.2 1455.0

Intensity (AU)

8.0x10 6.0x10 4.0x10 2.0x10

1081.1 1277.0 904.9 1472.0

1863.6

0.0 400 600 800 1000 1200 1400 1600 1800 2000

m/z

Figure 3.3: The mass spectrum of the Folch extract of human lipofuscin eluted at time 86.26 min. Groups II and III identify the related clusters of higher molecular weight compounds with mass to charge ratios of approximately 1000 and 1400, respectively.

77

418.4
4x10
6

Intensity (AU)

3x10

2x10

402.4 442.4 392.4 486.5 536.4 352.3 592.6

1x10

0 200 250 300 350 400 450 500 550 600

m/z

Figure 3.4: The MS/MS scan for A2E identified in the Folch extract of lipofuscin granules. Peaks corresponding to the m/z of 592 (red) with the loss of 106 (m/z 486.5), 150 (m/z 442), 174 (m/z 418), and 190 (m/z 402) are identified.

78

1.2x10

1.0x10

Intensity (AU)

8.0x10

6.0x10

4.0x10

2.0x10

0.0 200

250

300

350

400

450

500

550

600

Wavelength (nm)

Figure 3.5: The UV-visible absorbance spectrum of A2E

79

Figure 3.6: Characteristic cleavages for the fragmentation of A2E

80

558.5
2.0x10
6

624.6

Intensity (AU)

1.5x10

1.0x10

640.6
5.0x10
5

813.7

598.5 434.5 358.3 488.5532.5 384.4 450.4 508.5

663.6 722.6 798.7 708.6 758.7

0.0 400

6 6 3
600

1 8

800

m/z

Figure 3.7: The MS/MS scan of peak with m/z 814 from lipofuscin sample. Peaks corresponding to the mass of 814 (red) with the loss of 106, 150, 174, and 190 are identified (blue).

81

7x10 6x10 5x10

Intensity (AU)

4x10 3x10 2x10

1x10

0 300 400 500 600

Wavelength (nm)

Figure 3.8: The UV-Vis absorption for the peak with m/z 814

82 with masses of 663, 640, and 634, which correspond to losses of 150, 174, and 190 from the parent ion of 814. The proposed structure is displayed in Figure 3.9, which could form from the addition of one molecules of all-trans retinal to A2E with the loss of water and the ethanol group on A2E. Next the components that eluted with masses in the range of 1000 and 1400 amu were analyzed, and the MS/MS data was again compared to the fragmentation pattern of A2E. Figures 3.10 and 3.11 present the MS/MS and absorbance spectra for m/z 1081. The

fragmentation pattern for m/z 1081 displayed ions with masses of 931, 907 and 891, which correspond to the characteristic losses of 150, 174, and 190 from the parent ion. The proposed structure is displayed in Figure 3.12, which represents the addition of two molecules of all-trans retinal to A2E with the loss of water and ethanol group. Figure 3.13 presents the MS/MS spectrum of m/z 1424. The fragmentation pattern displayed losses of 174 and 190 displaying fragments with masses 1249 and 1233. The proposed structure is displayed in Figure 3.14, which represents the addition of one molecule of A2E aldehyde with m/z 472 and one molecule of A2E aldehyde with m/z 422 to A2E with the loss of the ethanol group on A2E and water. The spectra for m/z 1081 and 1423 also displayed fragmentation ions that were the same as the compounds, m/z 757, 803, and 814, located within group I of the lipofuscin samples. These data suggest that components in group II and III result from the polymerization of derivatives from group I. In addition to MS/MS data, the Log P of A2E was measured to determine the aggregative characteristics of A2E in aqueous environments. Using HPLC, the absorbance was measured and peak area was used to calculate the Log P of 7.3 +/-

83

Figure 3.9: Possible Structure of m/z 814 with cleavages identified

84

1.6x10 1.4x10 1.2x10

813 891

Intensity (AU)

1.0x10 8.0x10 6.0x10 4.0x10 2.0x10

825

592 487
663

799 767 729

907 865 931 975

1081

0.0 400 500 600 700 800 900 1000

m/z

Figure 3.10: The MS/MS scan for m/z 1081 located in lipofuscin. Peaks corresponding to the mass of 1081 (red) with the loss of 106 (m/z 975), 150 (m/z 931), 174 (m/z 907), and 190 (m/z 891) are identified (blue).

85

7x10

6x10

5x10

Intensity (AU)

4x10

3x10

2x10

1x10

0 300 400 500 600

Wavelength (nm)

Figure 3.11: The UV-Visible spectrum of m/z 1081 in lipofuscin

86

Figure 3.12: Possible structure of m/z 1081 with cleavages identified

87

2.0x10

757

1233

Relative Abundance (AU)

1.5x10

863 795

1423

1.0x10

1219
5.0x10
4

1393

592 566 620

1203 1057 931 1019 1135 1249

0.0 400 600 800 1000 1200 1400

m/z

Figure 3.13: The MS/MS results for the fragmentation of peak with m/z 1423 (red) in the lipofuscin sample. Peaks corresponding to the mass of 1423 with the loss of 174 (m/z 1249) and 190 (m/z 1233) are identified (blue).

88

Figure 3.14: Possible structure for m/z 1424 with cleavages identified

89 1, which was in agreement with computational values using the Sparc software program (Log P = 8.2 +/- 1). Using this value for A2E, a calibration curve of compounds with similar Log P values (Figure 3.15), and elution times from the TIC of lipofuscin, the approximate Log P values for the higher molecular weight components were calculated. Group I was determined to have an approximate Log P value of 8.3 +/- 0.5 followed by group II and III, which were approximately 9.2 +/0.7 and 10.2 +/- 1, respectively. To investigate the possibility that these modifications of A2E resulted from esterification, A2E was first treated with either acetyl chloride or hexanoyl chloride to synthesize the esters as indicated in Figure 3.16. The MS/MS obtained from esterification reaction of acetyl A2E is displayed in Figure 3.17 with the structure displayed as an inset. The MS/MS for the A2E hexanoyl ester is displayed in Figure 3.18 with the proposed structure displayed as an inset. The resulting spectra for A2E acetyl and hexanoyl esters gave a major fragment with m/z = 548. Further fragmentation of this major peak (m/z = 548) yielded fragments, with m/z = 358, 410, 374 (Figure 3.19), which was also located in the human lipofuscin sample (Figure 3.20). This fragment can readily be explained by the rearrangement depicted in Figure 3.21, giving the structure displayed in Figure 3.22. However, the parent ions from the A2E esterified products are not seen in the lipofuscin sample and, therefore, these products are not structurally related to those found in the lipofuscin mixture. A2E was further esterified with cinnamoyl chloride, which more closely structurally resembles A2E. Figure 3.23 displays the MS/MS data for the A2E

90

14 12 10 Log P 8 6 4 2 0 0 2 4 6 8 Log K 10 12 14

Figure 3.15: Calibration curve for Log P values of DDT, Triphenylamine, Phenanthrene, Benzophenone, and Cinnamic Acid to determine the Log P of A2E and higher molecular weight products.

91

Figure 3.16: Product from esterification reaction with A2E and R group. The R group is acetyl chloride, Hexanoyl chloride, or Cinnamoyl chloride (Mandal 2008).

92

5x10

548.4

634.4

4x10

Intensity (AU)

3x10

CH 3

2x10

1x10

358.3
0 100 200 300 400 500 600 700

m/z

Figure 3.17: The MS/MS of A2E acetyl ester (m/z 634) with the corresponding structure (Mandal 2008)

93

3.0x10

548.4
2.5x10
5

Intensity (AU)

2.0x10

1.5x10

1.0x10

5.0x10

358.3
300 400 500

574.5
600 700

0.0 200

m/z

Figure 3.18: The MS/MS of the A2E hexanoyl ester (m/z = 690.5) with the corresponding structure (Mandal 2008)

94

1.0x10

358.3
8.0x10
4

548

Intensity (AU)

6.0x10

4.0x10

2.0x10

241.2 198.2

374.3410.2

398.2

506.4

0.0 200 300 400 500

m/z

Figure 3.19: CID of main fragment m/z 548 (red) with losses of 150 (m/z 398), 174 (m/z 374), and 190 (m/z 358)(blue) (Mandal 2008)

95

2.0x10

358.3

Intensity (AU)

1.5x10

1.0x10

410.4 5.0x10
4

548.6

241.2

398.4

374.2
348.4 442.5 400 492.4 500

0.0 200

300

m/z

Figure 3.20: CID spectrum of species with m/z = 548 (red) with losses of 150 (m/z 398), 174 (m/z 374), and 190 (m/z 358) (blue) in full mass spectrum of human lipofuscin sample

96

H N O

NH

m/z 548

Figure 3.21: Rearrangement of esterification product yielding main fragment with m/z 548 (Mandal 2008)

97

412

372 346

242

358

Figure 3.22: Possible structure and fragmentations of peak with m/z = 548

98

1.2x10

575.5
1.0x10
4

Intensity (AU)

8.0x10

6.0x10

4.0x10

2.0x10

533.3 549.2 359.3 483.2

617.4 723.4

0.0 400 500 600 700

m/z

3.23: MS/MS of Cinnamoyl chloride ester (m/z = 723) (Mandal 2008)

99 cinnamoyl ester with the proposed structure displayed in Figure 3.24. The spectrum is relatively simple, giving one major fragment (m/z 575), which was interpreted as resulting from a McLafferty rearrangement (Figure 3.25) (Mandal 2008). Again, these results were also not consistent with the compounds found in lipofuscin. A second hypothesis that may account for the hydrophobic mixture in lipofuscin involved reactions with A2E-derived aldehydes (Figure 2.4). It was proposed that, once formed, these aldehydes could then react with other A2E molecules, forming higher molecular weight species. To investigate this hypothesis, A2E reaction mixture was allowed to incubate at 4 C for 60 days. This sample led to a complex mixture, which included many of the compounds found in vivo. The aged A2E samples kept at 4 C for 0, 30 and 60 days were then analyzed to determine if similar products were formed from pure A2E over time. The TIC for the aged A2E sample after 60 days displayed peaks similar to the lipofuscin sample, including ions with m/z of 859 and 1081 (Figure 3.26). The MS/MS for 859 and corresponding absorbance spectrum are displayed in Figures 3.27 and 3.28, respectively. The MS/MS for 1081 and corresponding absorbance spectrum are displayed in Figures 3.29 and 3.30, respectively. Both figures displayed peaks corresponding to losses of 150, 174, and 190 from the parent ion. In addition, the absorption spectra show two peaks with maxima at 330 and 500 nm. However, the intensity of the ion generated for the 1081 peak was smaller than the intensity of the ions generated for the 859 peak. Also, after 60 days, the more hydrophobic higher

100

Figure 3.24: Proposed product of Cinnamoyl chloride ester (m/z = 723)(Mandal 2008)

101

O H O

Figure 3.25: MacLafferty rearrangement in species with m/z = 574 (Mandal 2008)

102

4.5x10 4.0x10 3.5x10 3.0x10

1188.1

1236.1
6

Intensity (AU)

2.5x10 2.0x10 1.5x10 1.0x10 5.0x10

831.8 813.8 859.4

1203.1

784.3
6

1081.7 839.9 887.9 907.1

981.4 1070.1
1032.2 1113.1 13 1
1200

0.0 800 1000

m/z

Figure 3.26: The mass spectrum of A2E fraction that eluted at 93.52 minutes of chromatographic separation. Peaks found in lipofuscin mixture (Figures 3.2 and 3.3) are identified (blue).

103

669
4x10
5

Intensity (AU)

3x10

2x10

643 685

859

1x10

709 415 603 493 531 630

600

721 753
800

0 400

m/z

Figure 3.27: The MS/MS of m/z 859 in A2E. Peaks corresponding to mass 859 (red) with the loss of 150 (m/z 709), 174 (m/z 685), and 190 (m/z 669) are identified (blue).

104

m/z 859
1.2x10
4

1.0x10

Intensity (AU)

8.0x10

6.0x10

4.0x10

2.0x10

0.0 300 400 500 600

wavelength (nm)

Figure 3:28: UV-visible absorption spectrum of m/z 858 in aged A2E

105

1.0x10

1008.7

8.0x10

Intensity (AU)

6.0x10

1051.7

4.0x10

754.5 663.3 818.6 689.6

891.6 907.6

2.0x10

642.2
591.8

931.6
1081

0.0 400 500 600 700 800 900 1000 1100

m/z

Figure 3.29 The MS/MS scan for m/z 1081 located in aged A2E. Peaks corresponding to the mass of 1081 (red) with the loss of 150 (m/z 931), 174 (m/z 907), and 190 (m/z 891) are identified (blue). The mass of A2E (m/z 592) and additional peaks corresponding to smaller molecular weight compounds (m/z 818 and 745) with similar losses identified in the same sample.

106

7x10

6x10

5x10

Intensity (AU)

4x10

3x10

2x10

1x10

0 300 400 500 600

Wavelength (nm)

Figure 3.30: The UV-Visible absorption spectrum of m/z 1081 in aged A2E

107 molecular weight compounds located within the lipofuscin sample were either absent from the aged A2E sample or were not abundant enough for adequate identification. The MS/MS of three of the major peaks with m/z = 859, m/z = 920, and m/z = 1188 are displayed in Figures 3.31, 3.32, and 3.33 with characteristic losses of 150, 174 and 190 identified. The corresponding absorbance spectra for each compound are displayed in Figures 3.34, 3.35, and 3.36, respectively. Based on these spectra, the compounds with m/z 920 and 859 are clearly related. To investigate the specific mechanism of formation of the higher molecular weight compounds formed from the reaction of A2E with aldehydes, A2E was reacted with specific aldehydes, either cinnamaldehyde or benzaldehyde, for approximately 12 h in the dark. The resulting full mass spectra from cinnamaldehyde and benzaldehyde showed completely oxidized A2E and peaks with the oxidized A2E and attached aldehydes (Figures 3.37 and 3.38). As in human lipofuscin, the reactions with A2E appear as a series of discrete groups. For both the cinnamaldehyde and benzaldehyde reactions, group I is A2E and its oxidation products, group II is group I plus the addition of one aldehyde, and group III is the addition of a second aldehyde. The fragmentation patterns of one of the higher molecular weight compounds in A2E cinnamaldehyde and benzaldehyde reactions are displayed in Figure 3.39 and Figure 3.40 with corresponding proposed structures in Figures 3.41 and 3.42. The fragmentation of major peaks showed similar patterns with losses of 190, 174, and 150, which were also observed in oxidized A2E. The loss of 148 in

108

859.6

6.0x10

668.5

Intensity (AU)

4.0x10

642.5

2.0x10

684.6 708.6 414.4 478.4 530.3 454.4 602.5 658.6 5 4 5 828.6

800

0.0 400

600

m/z

Figure 3.31: The MS/MS of m/z 859 in reaction mixture for A2E synthesis. Peaks corresponding to mass 859 (red) with the loss of 150 (m/z 709), 174 (m/z 685), and 190 (m/z 669) are identified (blue).

109

859.6
4x10
5

3x10

Intensity (AU)

2x10

1x10

669.6 793.6 708.2 565.5 684.5 730.6 605.5 657.5 770.5

600 800

920.6 890.6

m/z

Figure 3.32: The MS/MS of m/z 920 in reaction mixture for A2E synthesis. Peaks corresponding to the mass of 920 (red) with the loss of 150 (m/z 771) and 190 (m/z 731) are identified (blue).

110

1127.8
1.2x10
6

1.0x10

Intensity (AU)

8.0x10

6.0x10

4.0x10

749.8
2.0x10
5

873.8 937.8 925.7

1039.7 1014.8 998.8 1082.8

709.8

770.7810.7849.7
800

0.0 1000

m/z

Figure 3.33: The MS/MS of 1189 in reaction mixture for A2E sythesis. Peaks corresponding to the mass of 1189 with the loss of 150 (m/z 1039), 174 (m/z 1015), and 190 (m/z 999) are identified.

111

1.4x10

1.2x10

1.0x10

Intensity (AU)

8.0x10

6.0x10

4.0x10

2.0x10

0.0 250

300

350

400

450

500

550

600

wavelength (nm)

Figure 3.34: The UV-Visible absorption spectrum of m/z 859 in reaction mixture for A2E synthesis

112

M+ 920
6x10
5

5x10

Intensity (AU)

4x10

3x10

2x10

1x10

200

300

400

500

Wavelength (nm)

Figure 3.35: The UV-Visible absorption spectrum of m/z 920 in reaction mixture for A2E synthesis

113

M+ 1189
3.5x10
5

3.0x10

Intensity (AU)

2.5x10

2.0x10

1.5x10

1.0x10

5.0x10

300

400

500

Wavelength (nm)

Figure 3.36: The UV-Visible absorption spectrum of m/z 1188 in reaction mixture for A2E sythesis

114

Ox A2E
3.0x10
5

641

2.5x10

Intensity (AU)

2.0x10

625 642.1

Ox A2E + CAL

788.9
1.5x10
5

1.0x10

551.2

657 772.9 686.9 708

805
Ox A2E + 2CAL

5.0x10

836.7

952.5
1000

0.0 400

600

800

m/z

Figure 3.37: The full mass spectrum of the reaction between A2E and cinnamaldehyde (Mandal 2008)

115

3.5x10 3.0x10 2.5x10

Ox A2E + BAL

Ox A2E + 2BAL

794.8
6

916.7 932.7

810.8

Intensity (AU)

2.0x10 1.5x10 1.0x10

Ox A2E

826.8

557.4

672.9 750.7

842.7

964.6

5.0x10

617.3

0.0 500 600 700 800 900 1000

m/z

Figure 3.38: The full mass spectrum of the reaction between A2E and benzaldehyde (Mandal 2008)

116

8000

640.4

556.2
6000

Intensity (AU)

4000

2000

422.2 346 374

0 400

512.3 540 454.5

500

617 684.3
600 700

762.4 791
800

m/z

Figure 3.39: The MS/MS spectrum of the higher molecular weight compound (m/z = 790) in A2E and Cinnamaldehyde reaction mixture using 40 % collision energy. Peaks corresponding to the mass of 790 (red) with the loss of 150 (m/z 640), 174 (m/z 617), and 190 (m/z 556) are identified (blue) (Mandal 2008).

117

70000 60000 50000

672.4

Intensity (AU)

40000 30000 20000 10000 0 450

766.4 794 490 468 508

500

574.3 604 561

550 600

654.5 644
650

722.5
700 750 800

m/z

Figure 3.40 The MS/MS spectrum of one of the higher molecular weight compounds in A2E benzaldehyde reaction mixture. Peaks corresponding to the mass of 794 (red) with the loss of 122 (m/z 672), 140 (m/z 654), 150 (m/z 644), and 190 (m/z 604) are identified (blue) (Mandal 2008).

118

Figure 3.41 Possible structure and fragmentation of one of the higher molecular weight compounds from reaction of oxidized A2E and cinnmaldehyde (Mandal 2008)

119

Figure 3.42: Possible structure and fragmentation of one of the higher molecular weight compounds from reaction of oxidized A2E and benzaldehyde (Mandal 2008).

120 the A2E cinnamaldehyde spectrum could be attributed to cinnamic acid and the loss of 122 in the A2E benzaldehyde spectrum could be due to the loss of the benzoic acid moiety. These fragments signify that the side chains of A2E are intact and the modifications are occurring at the ends of the polyene chain (Mandal 2008). The photolysis of all-trans-retinal in the presence of A2E was also performed. The full mass spectrum displayed in Figure 3.43 indicates the formation of two main products with m/z = 920 and 1188 after one hour of irradiation. The fragmentation pattern of m/z = 920 and 1188 are displayed in Figures 3.44 and 3.45 with the proposed structures displayed in Figures 3.46 and 3.47, respectively. The same characteristic losses of 150, 174, and 190 and a major fragment with m/z = 858 (Figures 3.48 and 3.49) are identified. This reaction was also performed without irradiation; however, the formation of products with m/z = 920 and 1188 was much slower, appearing after 18 hrs. Once the A2E and RAL reaction was complete, the mixture was injected onto an HPLC to separate and collect the compound with m/z = 920 (Figure 3.50). The compound that eluted at 35 mins was directly injected into the mass spectrometer (Figure 3.51) and confirmed by UV-Vis (Figure 3.52) and MS/MS (Figure 3.53) to be the same compound identified in the A2E reaction mixture (Figures 3.33 and 3.34) and lipofuscin samples (Figure 3.54). To further investigate the chemical reactions involved in producing the higher molecular weight products found in lipofuscin, cyclic voltammetry was performed on benzaldehyde, cinnamaldehyde and all-trans retinal. Initially, a background was taken of the working solution (Figure 3.55). This solution contained an electrolyte (0.1 M TEAP) added to 100 ml of anhydrous acetonitrile to

121

A2E + RAL rxn

592.9
8.0x10
6

Intensity (AU)

6.0x10

4.0x10

920.1

2.0x10

1188.2 727.5
858.9

0.0 400 600 800 1000 1200 1400

m/z

Figure 3.43: The mass spectrum of A2E reacted with all-trans-retinal

122

859.7
1.8x10 1.6x10 1.4x10
4

Intensity (AU)

1.2x10 1.0x10 8.0x10 6.0x10 4.0x10 2.0x10

920.8

793.6 669.6

771.8 565.4
600

730.6

833.7
800 1000

0.0 400

m/z

Figure 3.44: The MS/MS spectrum of m/z 920 from A2E RAL reaction. Peaks corresponding to the mass of 920 (red) with the loss of 150 (m/z 771) and 190 (m/z 731) are identified (blue).

123

1129.0
3

4x10

1127.9

3x10

Intensity (AU)

2x10

873.7 749.6 861.6 937.8

998.8 1038.8 1115.8 1014.8 1082.9

1159

1x10

1188
0 800 1000 1200

m/z

Figure 3.45: The MS/MS spectrum of m/z 1188 from A2E RAL reaction. Peaks corresponding to the mass of 1188 (red) with the loss of 150 (m/z 1038), 174 (m/z 1014), and 190 (m/z 998) are identified (blue).

124

Figure 3.46: Possible structure of m/z 920 with cleavages identified

125

Figure 3.47: Possible structure and fragmentation pattern of m/z 1188

126

2.0x10 1.8x10 1.6x10 1.4x10

668.6

858.6

Intensity (AU)

1.2x10 1.0x10 8.0x10 6.0x10 4.0x10

654.5
684.4

440.3 466.3
532.2 602.4

708.5

2.0x10

0.0 400

600

800

m/z

Figure 3.48: The MS/MS of m/z 858 in A2E and all-trans-retinal reaction. Peaks corresponding to m/z 858 (red) with the loss of 150 (m/z 708), 174 (m/z 684), and 190 (m/z 668) are identified (blue).

127

Figure 3.49: Possible structure for m/z 858 with cleavages identified

128

mAU

2000

920
1500

1000

500

10

20

30

40

50

min

Figure 3.50: The chromatogram of the A2E RAL reaction mixture using HPLC and PDA detection. Compound with m/z 920 eluted at 35 min.

129

920.97
6x10
7

5x10

Intensity (AU)

4x10

3x10

2x10

1x10

0 400 600 800 1000 1200 1400 1600 1800 2000

m/z

Figure 3.51: The full mass spectrum of peak that eluted at 35 min. in Figure 3.18

130

6x10

5x10

Relative Absorbance (AU)

4x10

3x10

2x10

1x10

0 200

300

400

500

Wavelength (nm)

Figure 3.52: UV-Vis absorption for the peak with m/z 920

131

3.5x10 3.0x10 2.5x10

859.8

Intensity (AU)

2.0x10 1.5x10 1.0x10

669.9 440.9 565.8 502.8 606

600

5.0x10

708.9 685.8

793.8 920.5

0.0 800

m/z

Figure 3.53: The MS/MS spectrum of m/z 920

132

858.5
1.2x10
5

1.0x10

668.9
4

Intensity (AU)

8.0x10

920.4

6.0x10

4.0x10

473.4 459.3 572.9

730.2 770.6

2.0x10

0.0 400 600 800 1000

m/z

Figure 3.54: The MS/MS spectrum of m/z 920 from Lipofuscin. Peaks corresponding to the mass of 920 (red) with the loss of 150 (m/z 771) and 190 (m/z 731) are identified (blue).

133

ACN with 0.1 M TEAP Background 50 mV/s Scan

10

0 -2 -1.5 -1 -0.5 -10 Current (uA) 0 0.5 1 1.5 2

-20

-30

-40

-50 Applied Potential (V)

Figure 3.55: The voltammogram of TEAP background

ensure sufficient conductivity. The redox couple ferrocenium/ ferrocene (Fc /Fc) (0.05 M) was added to each of the sample solutions to serve as an internal standard, including the initial working solution (Figure 3.56), the benzaldehyde (Figure 3.57), the cinnamaldehyde (Figure 3.58), and the all-trans retinal (Figure 3.59). Benzaldehyde appears to undergo two irreversible reductions at approximately 0.958 and -1.817 V vs. Ef(Fc+/Fc). These reductions could be irreversible as a result of kinetic considerations, the reduction being much more highly favored than the oxidation, or because of subsequent chemical processes, such as decomposition of the reduction product prior to the oxidation. Cinnamaldehyde appears to undergo a reversible redox reaction at -1.920 V vs. Ef(Fc+/Fc). However, from the disproportionate intensity of the redox peaks (the reduction peak being much more intense than the oxidation peak) it appears that the reduction is a more highly favored reaction. Alternatively, the reduction product could also be more stable and simply take more time to decompose, leaving a smaller amount of cinnamaldehyde reduction product to be subsequently oxidized. All-trans retinal appears to undergo an irreversible oxidation at 0.721 V and two irreversible reduction at -1.830 V and -1.641 V vs. Ef(Fc+/Fc). However, further analysis of these compounds is still needed to confirm the potentials. In addition, to determine the redox environment within the lipofsucin granules, cyclic voltammetry should be performed on A2E and the higher molecular weight products.

134

135

ACN with 0.1 M TEAP + 0.05 M Ferrocene 50 mV/s Scan Run 01

60

40

20 Current (uA)

0 -2 -1.5 -1 -0.5 -20 0 0.5 1 1.5 2

-40

-60 Applied Potential (V)

Figure 3.56: The voltammogram of ferrocene

136

ACN with 0.1 M TEAP + 0.05 M Benzaldehyde 50 mV/s Scan

60 0.6231, 40.879 40 20 Current (uA) 0 -2 -1.5 -1 -1.2883, -26.327 -0.5 0 0.5 0.5024, -21.448 1 1.5 2 -20 -0.3948, -17.296 -40 -1.2542, -50.208 -60 -80 Applied Potential (V)

Figure 3.57: The voltammogram of benzaldehyde

137

ACN with 0.1 M TEAP + 0.05 M Cinnamaldehyde 50 mV/s Scan

60 40 20 -1.1891, 3.912 Current (uA) 0 -2 -1.5 -1 -0.5 -20 0.5252, -20.372 -40 -60 -1.4484, -69.931 -80 Applied Potential (V) 0 0.5 1 1.5 2 0.6763, 36.122

Figure 3.58: The voltammogram of cinnamaldehyde

138

ACN with 0.1 M TEAP + 0.015 M All-Trans 50 mV/s Scan

120 100 80 60 Current (uA) 40 20 0 -1.076, -2.622 -1 -0.5 -1.0556, -25.102 0 -20 -40 -60 -1.2307, -62.862 -80 Applied Potential (V) 0 0.5 0.5333, -17.211 1 1.5 2 0.665, 36.012 1.3153, 86.879

-2

-1.5

Figure 3.59: The voltammogram of all-trans retinal (blue) and control (red)

139 Discussion

In this chapter, the composition of lipofuscin and the individual components of lipofuscin have been investigated. The results reported support the hypothesis that the reason A2E is being sequestered within lipofuscin granules is to minimize damage to the RPE, and that the higher molecular weight products originate from the reaction of A2E derived aldehydes with other molecules of A2E present in the lipofuscin mixture, not esterification reactions. Previously, research has suggested that retinal lipofuscin is extremely phototoxic. However, when RPE cells were fed lipofuscin granules, the RPE cells did not show an appreciable amount of damage (Ligget 2007), suggesting that the individual compounds within the lipofuscin granules are responsible for observed phototoxicity and not the granules themselves. A2E has been reported to cause damage to RPE cells by photochemically initiating free radical reactions and acting as a detergent by disrupting cell membranes. Sparrow et al. reported that A2E mediates blue light-induced apoptosis and that blue light damages DNA in A2Eladen RPE cell solutions (Sparrow, Parish et al. 1999; Sparrow and Cai 2001). However, these cells were fed free A2E and not lipofuscin granules. Boulton et al. later reported that the physiological concentration of A2E in lipofuscin granules was too small to account for the blue light-induced phototoxicity observed when RPE cells are fed lipofuscin granules (Davies, Elliott et al. 2001). Nevertheless, the cells were still damaged after irradiation with blue light, indicating that the major phototoxic component was not A2E and still is unidentified. This is supported by

140 research that discovered that A2E was excited by energy transfer within the lipofuscin granules and consequently could not be the dominant blue-absorbing chromophore (Haralampus-Grynaviski, Lamb et al. 2003). The comparison of fluorescence spectra of lipofuscin and A2E were similar, but variations in lipofuscin spectra suggest that multiple components contribute to the absorbance and fluorescence of lipofuscin. Therefore, the biological activity of the lipofuscin granules and the damaging effects associated with the granules, A2E, and other components of lipofuscin are still controversial subjects. Numerous compounds including A2E, oxidized A2E, and a complex mixture of hydrophobic components have been identified in Lipofuscin (Figure 3.1). This complex mixture of higher molecular weight compounds accounts for a large portion of the lipofuscin sample. Analysis of the corresponding spectra revealed a series of closely related compounds that differed by 14 amu, which most likely result from the addition of methylene groups. Eluting between 50 and 110 mins with 100 % methanol, these components were also determined to be relatively hydrophobic. The collision-induced dissociation (CID) and corresponding absorbance spectra for several of these higher molecular weight compounds were analyzed and then compared to the fragmentation pattern and absorbance spectra of A2E within the lipofuscin sample. Upon CID, the parent ion of A2E exhibits characteristic losses of 150, 174, and 190. These characteristic losses and the parent ion mass were observed in components of the complex mixture. All of the components that eluted between 50-80 min and approximately 50 % of the material that eluted from 80-110 min had analogous spectra to A2E, suggesting that these

141 higher molecular weight products are derivatized A2E. In addition, numerous spectra located within group II and III in the lipofuscin samples displayed ions consistent with compounds present in group I. For example, Figure 3.2 displays ions with m/z 814 and 863, which are also present in mass spectra of fragments found in group II (m/z 1081) and III (m/z 1424) of the lipofuscin sample (Figure 3.10 and 3.13), suggesting that these higher molecular weight products are the result of a polymerization reaction. Since hydrophobic substances like all-trans retinol are stored in the RPE as esters, esterification of A2E was previously investigated to identify the higher molecular weight products in lipofuscin. Initially, esterification of A2E was performed with acetyl chloride and hexanoyl chloride. Both reactions were used as model systems to represent the possible short and long chain fatty acids that exist in the retina (RPE) that could derivatize A2E to form esters. The fragmentation pattern for both esters could not be found in the human lipofuscin extract. However, the major fragment, m/z = 548, was identified in the full mass spectrum of lipofuscin and had a similar fragmentation pattern to the synthesized A2E ester, suggesting a similar structure. The predicted structure for species with m/z = 548 was A2E with the loss of the ethanol group. The loss of the ethanol group was also later seen in compounds with m/z 814 (Figure 3.9) and 1081 (Figure 3.12). Since both esterification products displayed the same fragmentation pattern, A2E was treated with structurally similar cinnamoyl chloride. The fragmentation pattern of the A2E cinnamoyl chloride ester differed from the acetyl and hexanoyl esters, the major peak had m/z = 575, which was attributed to a McLafferty rearrangement. This

142 mechanism could not be traced in human RPE samples, indicating that A2E is not stored as esters (Mandal 2008). To investigate the relationship between these higher molecular weight compounds and A2E, samples of pure A2E were aged for 60 days. The TIC of the aged A2E samples displayed similar clusters of peaks that were located in the lipofuscin sample (Figure 3.26). The CID of these peaks displayed in Figures 3.27 and 3.29 were also similar to the CID of the peaks located within lipofuscin displayed in Figures 3.7, 3.10, and 3.13. These figures show ions corresponding to, in most cases, A2E and losses of 150, 174, and 190 from the parent ion, which is also observed in the CID of A2E. However in the aged A2E, the peak with m/z = 859 had a greater abundance than peak with m/z = 1081. These data suggest that as A2E ages, the compound forms higher molecular weight derivatives and that these derivatives increase in abundance with age. The cluster of peaks with m/z of approximately 1400 in the lipofuscin sample was not observed in the synthetically aged A2E, which was attributed to the sample not being aged long enough. Ions corresponding to compounds located within group I were also observed in the MS/MS spectra from compounds in group II within the aged A2E sample, which supports a polymerization reaction. In addition, the reaction mixture of ethanolamine and all-trans retinal that produces A2E generated higher molecular weight products found in the lipofuscin sample. Since the esters previously synthesized were not consistent with the compounds found in lipofuscin, a second hypothesis involving the reaction of A2E with aldehydes was tested. The autooxidation of A2E in the presence of cinnamaldehyde and benzaldehyde yielded a

143 series of compounds including oxidation and addition products. The fragmentation patterns and characteristic losses of these products were similar to those found in oxidized A2E (Mandal 2008). The photolysis of retinal in the presence of A2E also generated compounds with the same characteristic fragmentation patterns with losses of 150, 174, and 190 found in aged A2E, the A2E reaction mixture, and human retinal lipofuscin. The absorption spectra show two peaks with maxima at 330 and 500 nm, which is in agreement with previously reported compounds located in photoreceptor cell out segments (Bui, Han et al. 2006). The compound with m/z 920 was suggested to be A2E plus the addition of one molecule of all-trans-retinal and one molecule of CH2COOH with the loss of oxygen (Figure 3.46). The compound with m/z 1188 was suggested to be 2 molecules of all-trans-retinal and one molecule of CH2COOH with the loss of two oxygens (Figure 3.47). The MS/MS of 858 was consistent with the addition of one molecule all-trans-retinal to A2E with the loss of water (Figure 3.49). The spectroscopic characteristics and fragmentation patterns associated with these compounds supports the hypothesis that A2E is reacting with aldehydes such as alltrans-retinal (Figures 3.46, 3.47, and 3.49), A2E-derived aldehydes (Figure 3.14), cinnamaldehyde and benzaldehyde (Figures 3.41 and 3.42) to form the higher molecular weight compounds found in lipofuscin. Also, the previously described cyclic voltammetric experiments support these observations. All three aldehydes--benzaldehyde, cinnamaldehyde, and alltrans-retinal--undergo irreversible reductions, as previously described. The fact that these aldehydes undergo irreversible reductions suggests that the reduction products

144 are highly reactive and that the rate of their disappearancepotentially through subsequent reactionsis quite fast. Previous studies have shown that electrochemical electron transfer to benzaldehyde produces a radical anion that can dimerize with an identical radical anion or the parent molecule, forming a higher molecular weight product (Armstrong, Quinn et al. 1974; Yeh 1977; Fawcett and Lasia 1981). This same type of polymerization reaction has been shown with many aromatic aldehydes and, in cases with aldehydes similar to benzaldehyde, have been shown to undergo further polymerization initiated by radical anion formation (Yeh, Liu et al. 2004). In addition, Simon et al. reported that the surface of lipofuscin granules contained small distinctive areas that were separated by thin layers, indicating that lipofuscin is an aggregated material (Haralampus-Grynaviski, Lamb et al. 2003). One of the major fluorescent components in the hydrophobic fraction of lipofuscin, A2E, was determined to have a log P of approximately 7.3, indicating that A2E is lipophilic and in aqueous solution will aggregate, minimizing contact with water or other polar substances. However, contrary to previous literature, A2E is not the dominant blue-absorbing chromophore or yellow-emitting fluorophore in lipofuscin. A2E becomes electronically excited mainly by energy transfer (HaralampusGrynaviski, Lamb et al. 2003). The lack of fluorescence suggests that A2E may self quench as it aggregates, forming higher molecular weight products (Ragauskaite, Heckathorn et al. 2001). This supports our results indicating that the majority of components in the hydrophobic portion of RPE lipofuscin granules consist of derivatized A2E generating a series of relatively hydrophobic compounds from

145 auto-oxidation. The higher molecular weight compounds identified result from an Aldol type condensation. These A2E modifying reactions assist in selfaggregation to form hard granules, which sequester A2E, diminishing its destructive ability.

CHAPTER 4 AGE-RELATED ACCUMULATION OF 3-NITROTYROSINE AND NITRO-A2E IN HUMAN BRUCHS MEMBRANE

Introduction

Recently four independent research groups used different methods to screen the genomes from different groups of AMD patients. All four studies discovered a commonly inherited variant (Y402H) of the complement factor H (CFH) gene that significantly increases the risk of AMD (Edwards, Ritter et al. 2005; Hageman, Anderson et al. 2005; Haines, Hauser et al. 2005; Klein, Zeiss et al. 2005). This finding links genetics and inflammation. Before this finding, the study of the components of drusen had provided compelling evidence that inflammatory and immune-mediated events participate in the development of drusen and progression of AMD. Protein components of drusen include immunoglobulins, components of the complement pathway (e.g., C5 and C5b-9), molecules involved in the acutephase response to inflammation (e.g., Amyloid P component), and proteins that modulate the immune response (e.g., vitronectin, clusterin, and apolipoprotein E) (Hageman and Mullins 1999; Hageman, Mullins et al. 1999; Johnson, Ozaki et al. 2000; Mullins, Russell et al. 2000). The finding that macrophages are important in choroidal neovascularization (CNV) also supports the involvement of inflammation

147 in AMD (Grossniklaus, Ling et al. 2002). Recent research provided further evidence that inflammation is involved in the development of AMD (Chen, Forrester et al. 2007; Laine, Jarva et al. 2007; Schaumberg, Christen et al. 2007; Skerka, Lauer et al. 2007) and the link between inflammation, drusen and oxidative stress (Wu, Lauer et al. 2007; Hollyfield, Bonilha et al. 2008; Wang, Ohno-Matsui et al. 2008). During inflammation, large fluxes of nitric oxide (NO) are released through the activation of inducible nitric oxide synthase (Marletta, Yoon et al. 1988; Carreras, Pargament et al. 1994). Nitrite concentration is reported to be nearly doubled in the diabetic retina (El-Remessy, Behzadian et al. 2003). Cigarette smoking, which has been strongly associated with the development of AMD (Solberg, Rosner et al. 1998), is also an important chronic contributor to human NO exposure (Council 1986; Borland and Higenbottam 1987). Patients with AMD have significantly higher plasma NO levels than control subjects (Evereklioglu, Er et al. 2003). NO itself is a relatively unreactive radical; however, it is able to form other reactive intermediates including nitrite (NO2-), peroxynitrite (ONOO-), NO2, and N2O3, etc that can modify proteins, lipids and other compounds. Nitrite is one of the major NO metabolic products and has been used as a marker of NO production (Farrell, Blake et al. 1992; Gaston, Reilly et al. 1993). In addition, nonenzymatic nitration of long-lived proteins such as extracellular matrix proteins is a well known pathway that has been associated with inflammation (Bailey, Paul et al. 1998; Paik, Dillon et al. 2001). The extracellular matrix proteins such as collagen and elastin have been reported to be nonenzymatically modified by nitrite at physiological pH (Paik, Ramey et al. 1997; Paik, Dillon et al. 2001). It has been reported that nitrite-

148 modification of basement membrane-like extracellular matrix proteins can impart deleterious effects on adjacent epithelial cell function and viability (Wang, Paik et al. 2005) and impair phagocytic capacity (Sun, Cai et al. 2007). Bruchs membrane lies between the choroidal capillary bed and retinal pigment epithelial (RPE) cells. The exchange of various materials between the underlying choriocapillaris and overlying RPE occurs through Bruchs membrane (Lyda, Eriksen et al. 1957; Sellner 1986). Bruchs membrane is permeable to macromolecules up to 300kD in size in healthy eyes, but there are numerous examples of pathological processes in which larger macromolecules or even cells, including macrophages and leukocytes, can traverse Bruchs membrane in the diseased eye (Crane and Liversidge 2008). In addition to Bruchs membrane, trafficking of material from the RPE to the choriocapillaris is limited in the healthy eye by tight junctions between adjacent cells of the RPE monolayer. This outer blood-retinal barrier is part of the specialized ocular microenvironment that confers protection or immune privilege to mitigate the effect of deleterious immune responses (Streilein 2003). Nevertheless, this barrier is altered in pathological circumstances, and breakdown of the outer blood retinal barrier, including macrophage and leukocyte infiltration of the retina, are implicated in many diseases including AMD (Jha, Bora et al. 2007). Several investigators have suggested that age-related damage to Bruchs membrane allows for the accumulation of abnormal extracellular deposits, called drusen, between the basal lamina of the RPE and the inner collagen layer of Bruchs membrane (Newsome, Huh et al. 1987; Pauleikhoff, Barondes et al. 1990; Mullins, Russell et al. 2000; Crabb, Miyagi et al. 2002). The

149 accumulation of drusen is thought to elicit a local inflammatory response (Anderson, Mullins et al. 2002; Yasukawa, Wiedemann et al. 2007; Hollyfield, Bonilha et al. 2008). Recently, research has shown that age-related changes in human Bruchs membrane can exert significant deleterious effects on RPE function that are independent of cell aging, including impairing the ability of cultured RPE to phagocytize photoreceptor outer segments (Sun, Cai et al. 2007). A similar effect on RPE function is observed after nonenzymatic nitration of RPE basement membrane in tissue culture (Wang, Paik et al. 2005). We hypothesize that inflammation will produce reactive nitrogen species that will modify intrinsic extracellular matrix proteins and/or extrinsic deposits accumulated in Bruch's membrane. Surprisingly, there have been no studies that have reported nitrite modification occurring in intrinsic Bruchs membrane proteins or extrinsic deposits, although tyrosine nitration has been shown to occur in photoreceptor cells (Miyagi, Sakaguchi et al. 2002). However, previous studies have demonstrated that numerous structural and molecular alterations occur within human Bruchs membrane as a function of age. These changes, which disrupt the normal molecular architecture of Bruchs membrane, include: (1) structural changes in the main collagen framework, including cross-linking and deposition of long-spaced collagen (Yamamoto and Yamashita 1989), qualitative and quantitative changes in the native extracellular matrix molecules (Pauleikhoff, Wojteki et al. 2000), deposition of abnormal extrinsic molecules including fluorescent products that accumulate in drusen (Ruberti, Curcio et al. 2003), macromolecular changes in the structure of Bruchs

150 membrane, such as calcification, cracks or loss of inner layers due to inadequate basal membrane regeneration as in geographic atrophy (Feeney-Burns and Ellersieck 1985; Grossniklaus, Hutchinson et al. 1994), and changes in the physical characteristics of Bruchs membrane, such as an age-dependent increase in transmembrane hydraulic conductivity (Moore, Hussain et al. 1995) and age-related linear decline in collagen solubility, an index of deleterious cross-linking (Karwatowski, Jeffries et al. 1995), 3-nitrotyrosine is a specific marker for inflammation-induced oxidative damage to proteins. In addition to proteins, Bruch's membrane also contains lipids, lipofuscin and carbohydrates (Hageman, Luthert et al. 2001; Yasukawa, Wiedemann et al. 2007). Lipofuscin is a mixture of autofluorescent material that accumulates in the RPE cells and is reported to photochemically generate a series of reactive oxygen species, including singlet oxygen, hydrogen peroxide, and superoxide anions (Gaillard, Atherton et al. 1995; Rozanowska, Wessels et al. 1998) that may enhance oxidative stress in the RPE. One of the major organic soluble chromophores in lipofuscin is A2E (2-[2,6-dimethyl-8-(2,6,6-trimethyl-1cyclohexen-1-yl)-1E, 3E,5E,7E-octatetraenyl]-1-(2-hydroxyethyl)-4-[4-methyl-6-(2,6,6-trimethyl-1-cyclohexen-1-yl)-1E,3E,5E-hexatrienyl]pyridinium). In this chapter, liquid chromatography-mass spectrometry (LC-MS) was used to investigate the modifications to intrinsic and extrinsic proteins and A2E in human Bruch's membrane by reactive nitrogen species released during inflammation. We have identified an increasing accumulation of 3-nitrotyrosine and nitro-A2E in human Bruch's membrane with advancing patient age. Detection of nitro-A2E within

151 human Bruchs membrane may serve as a specific biomarker for inflammation and non-enzymatic nitration.

Results

Identification of tyrosine nitration in Bruch's membrane

To determine if tyrosine nitration occurs in Bruch's membrane, Bruch's membrane was acid hydrolyzed and analyzed by LC-MS. 3-nitrotyrosine (3-NT), which is an important biomarker of nonenzymatic nitration, is stable under acid hydrolysis (Crowley, Yarasheski et al. 1998). The m/z of the quasimolecular ion ([MH]+) of 3-nitrotyrosine is 227.0. This molecule easily loses a nitro group under collision-induced dissociation (CID), forming a fragment with m/z 181.0. Therefore, we used selective reaction monitoring (SRM) (parent ion m/z = 227.0 with daughter ion m/z = 181.0) to specifically monitor the presence of 3-nitrotyrosine. Figure 4.1 gives the results of selective reaction monitoring scans of the acid hydrolysate of Bruch's membrane and standard 3-nitrotyrosine. The SRM scan of the acid hydrolysate of Bruchs membrane has a peak with similar retention time to the peak of 3-nitrotyrosine. The tandem mass spectrum of the compound in this peak is also similar to the tandem mass spectrum of 3-nitrotyrosine (Figure 4.2) (Wang 2005). Identical experiments were performed on three samples of human Bruchs membranes from different donors to determine the relative concentration of 3nitrotyrosine within the human Bruchs membrane samples as a function of patient

152

1.2x10 1.0x10 8.0x10 6.0x10 4.0x10

5 4

Standard 3-NT

4 4

Intensity

2.0x10

0.0 0 1.0x10 8.0x10 6.0x10 4.0x10 2.0x10

5

20

40

60

80

100

120

Bruch's membrane

0.0 0 20 40 60 80 100 120

Retention time

Figure 4.1: Selected Reaction Monitoring (SRM) chromatogram of 3-NT and acid hydrolysate of BM (SRM 227.1181.1).3-NT and acid hydrolysate of BM was analyzed by LC/MS as described in method. The SRM scan of BM acid hydrolysate has a peak with m/z 227 and fragment 181 and similar retention time (51 minutes) to 3-NT which indicates the presence of 3-NT in BM acid hydrolysate (Wang 2005).

153

8.0x10 6.0x10 4.0x10 2.0x10

Standard 3-NT
HO

210
HO C O

O 2N
6

209.8

181

NH2

181.0

Intensity

m/z 227

0.0 100 5x10 4x10 3x10 2x10 1x10

4

120

140

160

180

200

220

Bruch's membrane

209.8

181.0

0 100 120 140 160 180 200 220

M/Z

Figure 4.2: The tandem mass spectra of standard 3-nitrotyrosine and component with m/z 227.0 at RT 51min in BM. The tandem mass spectrum of the component at RT 51mins from human BM extracted from 72-75 year old donors is very similar to the tandem mass spectrum of 3-NT. The inset gives the predicted fragmentation of 3-NT (Wang 2005).

226.9

226.9

154 age. Approximately six pieces of Bruchs membrane from four different donors from the decades <25 yrs, 40-60 yrs, and >65 yrs were obtained. These samples were then extracted as previously described in Chapter 2. To quantify the actual concentration of 3-nitrotyrosine, the standard addition of 50 M solution of 3nitrotyrosine was added to each of the samples before analysis with LC-MS. The Zoom and SRM scans for each sample are displayed in Figures 4.3 and 4.4, respectively. The peaks were then integrated (Figure 4.5) and and the concentration was calculated using standard addition with a calibration curve (Figure 4.6). Figure 4.7 displays the concentrations of 3-nitrotyrosine in the different decades. The presence of 3-nitrotyrosine is negligible in the < 25 yrs sample of BM. There was a small increase in the BM sample between the ages of 40 to 60 yrs followed by a substantial increase in the BM sample > 65 yrs. The exponential increase of 3nitrotyrosine in BM, observed in Figure 4.7, suggests that tyrosine nitration occurs in human Bruchs membrane as a function of age, which may be related to the inflammatory response.

Identification of nitro-A2E in Bruchs membrane

To investigate our hypothesis that one of the major components in lipofuscin, A2E, may be modified by reactive nitrogen species resulting in the formation of nitro-A2E, nitro-A2E was synthesized as described in Materials and Methods and then analyzed by mass spectrometry. To confirm the presence of A2E and nitro-

155

6.0x10 6 4.0x10 6 2.0x10 0.0 4.0x10

6 6

BM > 65 yrs

BM 40-60 yrs

Intensity (AU)

2.0x10

0.0 6.0x10 4.0x10

6 6 6

BM < 25 yrs

2.0x10 0.0

6.0x10 6 4.0x10 6 2.0x10 0.0 20 40 60

Standard

80

Time (min)

Figure 4.3: The zoom scan of BM with the standard addition of 3-nitrotyrosine (m/z 227).

156

4x10 6 3x10 6 2x10 6 1x10 0 2x10

6 6

BM > 65 yrs

BM 40-60 yrs

Intensity (AU)

1x10

0 3x10 2x10
5 5 5

BM < 25 yrs

1x10 0

5x10 6 4x10 6 3x10 6 2x10 6 1x10 0 20 40 60

Standard

80

Time (min)

Figure 4.4: The SRM scan of m/z 227 181 from the standard addition of 3nitrotyrosine in BM samples from different age groups.

157

4.5x10 4.0x10 3.5x10 3.0x10

Standard 1 BM < 25 yrs BM 40-60 yrs Standard 2 BM > 65 yrs Standard 3

Intensity (AU)

2.5x10 2.0x10 1.5x10 1.0x10

5.0x10

0.0 -5.0x10
5

-20

20

40

60

80

100

120

140

Time (min)

Figure 4.5: Integration of area under SRM scan from Figure 4.4.

158

Calibration Curve of 3-nitrotyrosine Standards

7.00E+06 6.00E+06 5.00E+06 4.00E+06 Area 3.00E+06 2.00E+06 1.00E+06 0.00E+00 -1.00E+06 -2.00E+06 Volume (ml) 0 5 10 15 20 25 30

Figure 4.6: Calibration curve for 3-nitrotyrosine

159

Concentration 3-nitrotyrosine in BM
2.00E-04 1.80E-04 1.60E-04 Concentration (M) 1.40E-04 1.20E-04 1.00E-04 8.00E-05 6.00E-05 4.00E-05 2.00E-05 0.00E+00 BM <25 yrs BM 40-60 yrs BM > 65yrs

Figure 4.7: The concentration of 3-nitrotyrosine in BM samples from ages of < 25, 40-60, and > 65 years.

160

4.0x10 3.5x10 3.0x10 2.5x10 2.0x10 1.5x10 1.0x10

8 8 8 8 8 8 8 7

52.86

m/z 592.5

Intensity

5.0x10 0.0 3.5x10 3.0x10 2.5x10 2.0x10 1.5x10 1.0x10 5.0x10

7 7 7 7 7 7 6

20

40

51.66

60

80

100

120

m/z 637.5

0.0 0 20 40 60 80 100 120

Time (min)

Figure 4.8: The selected ion chromatograms for synthetic A2E (top) and nitro-A2E

161 A2E, the total ion chromatogram for synthetic nitro-A2E filtered for m/z 592.5 and 637.5 are displayed in Figure 4.8 with corresponding retention times. The ultraviolet-visible absorption spectra of m/z 592.5 (A2E) and 637.5 (nitro-A2E) were also compared (Figure 4.9). A2E had absorption peaks at 335 and 430 nm (Figure 4.9a), which corresponds to previously reported results (Parish, Hashimoto et al. 1998). The absorption spectrum of nitro-A2E (Figure 4.9b) is very similar to the absorption spectrum observed in Figure 5a for A2E. However, the two most intense absorption peaks were located at 330 and 415 nm (Figure 4.9b), indicating that nonenzymatic nitration induces an expected slight blue shift in the absorption spectrum. The structure of A2E is compared to the predicted structure of nitro-A2E in Figure 4.10 with characteristic cleavages identified (Dillon, Wang et al. 2004). The m/z of synthetic nitro-A2E measured by mass spectrometry is 637.5, which is in agreement with this predicted structure. Figure 4.11 displays the tandem mass spectrum of synthetic nitro-A2E. The major fragments from the CID spectrum match the predicted structure and characteristic fragmentations shown in Figure 4.10. To investigate the possible presence of nitro-A2E in vivo, the organic soluble components in Bruchs membrane from 70-yr-old donor globes were extracted and analyzed by LC/MS. The total ion chromatogram revealed a peak with m/z 592, which was identified as A2E based on its absorption spectrum and characteristic fragmentation pattern. A peak with m/z 637.5 within the total ion chromatogram was also seen at approximately the same retention time as the peak with m/z 637.5

162

a
100 Relative Absorbance 80

# 335.0 430.0 395.0 235.0 60 40 20 0 200 250 300 350 400 m/z 450 500 550 600

m/ z 592.5

b
100 Relative Absorbance 80 60 40 20 0 200

# 235.0

m/ z 637.5
330.0 415.0 425.0

285.0

250

300

350

400 450 wavelength (nm)

500

550

600

Figure 4.9: The UV-Vis spectra for A2E (m/z 592.5) and for nitro A2E (m/z 637.5).

163

A2E (m/z 592)

Nitro-A2E (m/z 637)

Figure 4.10: Structures of A2E (m/z 592) and nitro-A2E (m/z 637) showing characteristic cleavage points and the resulting fragment molecular weights .

164

100 90 80

MS/MS 637.5 605.3 591.4 576.4 637.4 418.2 401.5 376.2 442.3 469.2 487.3 619.4

Relative Abundance (AU)

70 60 50 40 30 20 10 0 300

400

500

600

m/z

Figure 4.11: The tandem mass spectrum of synthetic nitro-A2E induced dissociation to confirm the identification of nitro-A2E.

165 located in the synthetic nitro-A2E sample, suggesting the presence of nitro-A2E within the Bruchs membrane sample. This peak was then fragmented by collisioninduced dissociation to confirm the identification of nitro-A2E. Figure 4.12 displays the tandem mass spectrum of the component with m/z 637.5 located within the Bruchs membrane extract. The major fragments correspond to characteristic cleavages illustrated in Figure 4.10 and also observed in the authentic nitro-A2E sample (Figure 4.11). The total ion chromatogram also contained samples with molecular weights of m/z 653 and 682 (Figure 4.13), which suggests the presence of oxidized-nitrated A2E and doubly nitrated A2E, but the amounts were insufficient to acquire a full CID spectrum. We then sought to determine whether A2E was nitrated within RPE lipofuscin and then transported to Bruchs membrane, or whether nitration of A2E occurred after A2E accumulation within Bruchs membrane. To address this issue, approximately ten samples of the organic soluble extract of lipofuscin and the organic soluble extract of Bruchs membrane from three donors were analyzed by LC-MS and compared. Figure 4.14 displays filtered total ion chromatograms for A2E (m/z 592) and nitro-A2E (m/z 637) in RPE lipofuscin and Bruchs membrane. The presence of several peaks in the chromatograms result from several isomers coexisting (Parish, Hashimoto et al. 1998). The highest concentration of A2E was observed in RPE lipofuscin followed by a significantly lower concentration (30-40 fold) within Bruchs membrane extract. Nitro-A2E was absent from the lipofuscin samples tested but nitro-A2E was detected within human Bruchs membrane, thus

166

MS/MS 637.8 in BM
100

619.4

80

591.5 469.3 487.2 637.8

Relative Abundance (AU)

60

418.4 252.2 358.1 317.6

442.4

40

517.2 535.4

20

273.2

0 300 400 500 600

m/z

Figure 4.12: The tandem mass spectrum of nitro-A2E isolated from 65 yrs and older BM. Box = mass same in synthetic nitro-A2E and nitro-A2E isolated from 65 yrs and older BM.

167

4x10 6 3x10 6 2x10 6 1x10 0 3x10

5 5 5

m/z 592.5

20

40

60

80

100

120

Intensity (AU)

2x10

m/z 637.2

1x10 0

2.0x10 4 1.5x10 4 1.0x10 3 5.0x10 0.0 1.0x10 3 8.0x10 3 6.0x10 3 4.0x10 3 2.0x10 0.0
4

20

40

60

80

100

120

m/z 653.4

20

40

60

80

100

120

m/z 682.5

20

40

60

80

100

120

Time (min)

Figure 4.13: The selected ion chromatogram of m/z 592.5 (A2E), m/z 637.5 (nitro A2E), m/z 653.4 (nitro A2E plus oxygen), and m/z 682.5 (A2E with 2 nitro substitutions).

168

5x10 7 4x10 7 3x10 7 2x10 7 1x10 0 0 4.0x10

6 6

m/z 592 in lipofuscin

20

40

60

80

100

120

m/z 637 in lipofuscin

Intensity

2.0x10

0.0 2.0x10 6 1.5x10 6 1.0x10 5 5.0x10 0.0

6 6

20

40

60

80

100

120

m/z 592 in BM

0 2.0x10

20

40

60

80

100

120

m/z 637 in BM

0.0 0 20 40 60 80 100 120

Time (min)

Figure 4.14: The selected ion chromatograms for A2E (m/z 592) and nitro-A2E (m/z 637) from RPE lipofuscin and BM extracts from human donor globes that were 65 yrs and older. Note that nitro A2E and A2E from the BM have similar concentrations, whereas no nitro-A2E could be detected from the RPE despite increasing the sensitivity of the detector.

169 providing strong evidence that the nitration of A2E is specific to Bruchs membrane and does not occur within RPE lipofuscin.

Concentration of Nitro-A2E in Bruchs membrane samples from different decades of life

We then determined the relative concentration of A2E within the human Bruchs membrane samples as a function of patient age. Approximately 8 pieces of Bruchs membrane from 4 different donors from each decade (including <20s, 40s, 50s, 60s, 70s, 80s), and clinically diagnosed nonexudative AMD were obtained. These samples were then extracted as previously described in Chapter 2. To quantify the actual concentration of A2E and nitro-A2E, an internal standard of 50 M tryptophan was added to each of the samples before analysis with LC-MS. The peaks were then integrated (Figure 4.15) and concentration calculated for A2E (Figure 4.16). The SRM scans of Nitro-A2E in each decade are displayed in Figure 4.17. Once analyzed by LC-MS, the corresponding peaks were integrated (Figure 4.18) and the concentration in each sample was calculated (Figure 4.19). The concentrations of A2E and nitro-A2E throughout the different decades were then compared to concentrations found in AMD samples (Figure 4.20). The accumulation of both A2E and nitro-A2E is negligible up to the 4th decade of life. However, between the 4th and 5th decades there is a substantial increase in the concentrations of both A2E and nitro-A2E, which continues to rise throughout the 6th, 7th, and 8th decades. To determine if these results were relevant to AMD, the

170

6.0x10 5.5x10 5.0x10 4.5x10 4.0x10

6 6 6 6 6 6 6 6 6 6 6 5

< 20 yrs 40 yrs 50 yrs 60yrs 70 yrs 80 yrs AMD

Intensity (AU)

3.5x10 3.0x10 2.5x10 2.0x10 1.5x10 1.0x10 5.0x10

0.0 0 20 40 60 80 100 120 140 160 180 200

TIme (min)

Figure 4.15: Integration of A2E in BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs)

171

0.0008 0.0007 0.0006 0.0005 0.0004 0.0003 0.0002 0.0001 0.0000 -0.0001 10 20 30 40 50 60 70 80

Concentration (M)

Time (years)

Figure 4.16: The concentration of A2E in BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs)

172

4x107 3x107 2x107 1x10 0 6.0x105 4.0x105 2.0x10 0.0 3x105 2x105 1x10 0
5 5

AMD 80 yrs 70 yrs 60 yrs 50 yrs 40 yrs <20 yrs

0 20 40 60 80 100

Intensity (AU)

2x104 1x10 0 1.2x103 8.0x103 4.0x10 0.0 1.2x103 8.0x103 4.0x10 0.0 1x10
4 4 4

Figure 4.17: The SRM scans of Nitro-A2E from BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs)

173

5.5x10 5.0x10 4.5x10 4.0x10

< 18 yrs 40 yrs 50 yrs 60 yrs 70 yrs 80 yrs AMD

Intensity (AU)

3.5x10 3.0x10 2.5x10 2.0x10 1.5x10 1.0x10 5.0x10

0.0 0 20 40 60 80 100 120 140 160 180 200

Time (min)

Figure 4.18: Integration of Nitro-A2E A2E in BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs)

174

0.00018 0.00016

Concentration of nitro-A2E (M)

0.00014 0.00012 0.00010 0.00008 0.00006 0.00004 0.00002 0.00000 -0.00002 10 20 30 40 50 60 70 80

Time (years)

Figure 4.19: The concentration of Nitro-A2E in BM samples from different decades of life (<20, 40, 50, 60, 70, and 80 yrs)

175

1.20E-03 1.00E-03 Concentration (M) 8.00E-04 6.00E-04 4.00E-04 2.00E-04 0.00E+00 18 40 50 60 Time (years) 70 80 AMD A2E Nitro-A2E

Figure 4.20: The concentration of A2E and Nitro-A2E in BM samples from <20, 40, 50, 60 70, and 80 decades of life and dry AMD.

176 concentrations of A2E and nitro-A2E throughout the different decades were also compared to the concentrations found in nonexudative AMD, as shown in Figure 4.19. The nonexudative AMD samples had the highest concentration of A2E and nitro-A2E. Patients in the 8th decade of life displayed similar concentration of both the A2E and nitro-A2E, as seen in the nonexudative AMD samples.

Discussion

Bruch's membrane is located between the endothelium layer of the choriocapillaris and a monolayer of retinal pigment epithelium. In the normal eye Bruchs membrane serves as an attachment surface for the RPE. The outer bloodretinal barrier is formed by tight junctions between adjacent RPE; Bruchs membrane is partially responsible for limiting the movement of large molecules and cells from the choriocapillaris to the outer retina. This barrier is broken down during inflammation and inflammatory cells such as monocytes, macrophages, lymphocytes (Dua, McKinnon et al. 1991) and inflammatory mediators including complement components (Hollyfield, Bonilha et al. 2008) can traverse Bruchs membrane and accumulate within this structure. Nitric oxide released by these inflammatory cells together with the high oxygen concentration in the retina is expected to cause oxidative stress to many components in Bruchs membrane and could lead to nonenzymatic nitration of intrinsic proteins and extrinsic products that accumulate within Bruchs membrane as a function of age. To our knowledge, the finding of 3-nitrotyrosine and A2E nitration in Bruchs membrane in this study

177 provides the first clear demonstration of non-enzymatic nitration of proteins and age-related deposits (A2E) within human Bruchs membrane. Numerous changes develop within human Bruchs membrane as a function of increasing patient age, including collagen cross-linking (Yamamoto and Yamashita 1989) and the accumulation of abnormal deposits such as drusen (Ruberti, Curcio et al. 2003). Physiological collagen cross linking provides structural stability to this important structural protein, whereas nonphysiological collagen cross linking is an imprecisely controlled process that impairs collagen structure and function (Bailey, Paul et al. 1998). Nonenzymatic collagen cross linking can be induced by nitrite, and nitration of protein tyrosine residues to form 3-nitrotyrosine is a hallmark of tissue injury caused by inflammation. 3nitrotyrosine has been identified in many diverse pathological conditions such as atherosclerosis, pulmonary and heart disease, viral infections, and neurological disorders (Ischiropoulos 1998). Recent studies have established that 3-nitrotyrosine serves as a marker of reactive nitrogen species formation and can alter protein function. For example, modification of tyrosine residues can affect the phosphorylation and dephosphorylation of tyrosine, an important mechanism of cell regulation (Di Stasi, Mallozzi et al. 1999). Tyrosine nitration in Bruch's membrane may affect the degree of phosphorylation of some important proteins and further affect the migration of inflammatory cells through the blood retinal barrier (Erickson, Sundstrom et al. 2007). Nitrite-induced modification of extracellular proteins can be induced in vitro (Paik, Ramey et al. 1997; Paik, Dillon et al. 2001), and RPE cell viability and phagocytic ability decrease on nitrite-treated extracellular

178 matrix (Wang, Paik et al. 2005; Sun, Cai et al. 2007). Nitrite-induced changes in normal basement membrane mimic the deleterious effects of aging Bruchs membrane on RPE function (Wang, Paik et al. 2005; Sun, Cai et al. 2007). Lipofuscin and other RPE cellular components have been found in drusen, the extracellular deposits located between the basal lamina of the RPE and the inner collagenous layer of the Bruchs membrane (Hageman, Luthert et al. 2001; Crabb, Miyagi et al. 2002). One of the major components of lipofuscin is A2E, and this study demonstrates the presence of A2E in human Bruchs membrane. A2E is not normally a component of Bruchs membrane in young eyes, and we did not identify significant levels of A2E or nitro A2E in samples obtained from patients in the second decade of life (Figure 4.19). In addition, the concentration of A2E clearly increases with patient age (Figure 4.19), thus demonstrating that A2E deposition is a nonphysiological process that does not occur, or occurs to a very limited extent, in young individuals. The mechanism for A2E accumulation is not known. It is believed that RPE ordinarily does not extrude or exocytose active lysosomes or lysosomal enzymes although aged RPE extrude cytoplasm with active lysosomes into Bruch's membrane (Feeney-Burns, Gao et al. 1987). We could not determine if the A2E identified in Bruchs membrane is part of this normal extrusion process. Lipofuscin and other cellular debris accumulated in Bruchs membrane may contribute to the decreasing hydraulic conductivity observed with age (Moore, Hussain et al. 1995) and also may stimulate chronic inflammation. Our results clearly demonstrate that 3-nitrotyrosine is present within proteins that are present within human Bruchs membrane that is isolated using previously

179 described techniques. Previous studies using scanning and transmission electron microscopy of Bruchs membrane preparations demonstrate that the Bruchs membrane isolated in these preparations contains all 5 layers of Bruchs membrane (i.e., basal lamina of the RPE, inner collagen layer, elastin, outer collagen layer, and basal lamina of the choriocapillaris). Scanning electron microscopy demonstrates the preparation contains extracellular deposits on both the inner and the outer aspects of the RPE basal lamina (Del Priore and Tezel 1998; Tezel, Del Priore et al. 2004). Since our preparation contains intrinsic Bruchs membrane proteins as well as extracellular deposits, additional studies are required to determine if the 3nitrotyrosine that we have detected represents modifications of intrinsic Bruchs membrane proteins, proteins located in extracellular deposits such as drusen, or both. However, it should be noted that nitro-A2E is present within the Bruchs membrane preparation but we did not detect nitro-A2E in lipofuscin extracted from human RPE (Figure 4.13). This suggests that nitration of A2E occurs after A2E has accumulated within Bruchs membrane. Thus, non-enzymatic nitration of A2E must occur within Bruchs membrane, possibly due to nitric oxide and/or related nitrating agents such as peroxynitrite. It is likely that non-enzymatic nitration of both intrinsic Bruchs membrane proteins and extracellular deposits would occur by a similar mechanism. To our knowledge, the current study represents the first clear demonstration of inflammation-related chemical modifications detected in human Bruchs membrane. The presence of 3-nitrotyrosine and nitro-A2E may be important biomarkers for immune-mediated processes during aging, and the role of this

180 process in the development of age-related macular degeneration. Further experiments are needed to evaluate other aspects of this process, such as: (1) the relationship between the degree of nitration and the age/medical history of the donor; (2) the effect of nitration on the turnover of intrinsic extracellular matrix proteins; and (3) determination of the three-dimensional structural changes resulting from nitration and the effects of these changes on cellular function.

CHAPTER 5 MODIFICATIONS TO THE BASEMENT MEMBRANE PROTEIN LAMININ AND A2E: A MODEL FOR AGING IN BRUCHS MEMBRANE

Introduction

The dry form of age-related macular degeneration (AMD) is a multifactoral disease characterized by central vision loss attributed to photoreceptor cell death as a result of the degeneration of the retinal pigment epithelium (RPE). Oxidative stress and blue light-mediated damage related to lipofuscin, and particularly its major chromophore A2E, are considered some of the possible mechanisms underlying the degeneration of the RPE (Dorey, Wu et al. 1989; Winkler, Boulton et al. 1999; Sparrow, Nakanishi et al. 2000; Suter, Reme et al. 2000; Sparrow and Cai 2001; Liang and Godley Bernard 2003). The accumulation of drusen and basal deposits between the RPE and Bruchs membrane have also been shown to have deleterious effects on RPE cell viability (Nakaizumi 1964; Sarks 1976; Newsome, Huh et al. 1987; Pauleikhoff, Barondes et al. 1990; Ho and Del Priore 1997; Mullins, Russell et al. 2000). Furthermore, previous research has suggested that modification of basement membrane proteins may also affect the physiology and attachment of RPE cells (Ho and Del Priore 1997; Ivert, Keldbye et al. 2005; Wang, Paik et al. 2005). The accumulation of fluorescent lysosomal storage bodies, lipofuscin, in the

182 RPE is considered one of the major changes associated with age and may be related to the onset of AMD (Dorey, Wu et al. 1989). A2E (2-[2,6-dimethyl-8-(2,6,6trimethyl-1-cyclohexen-1-yl)-1E, 3E,5E,7E-octatetraenyl]-1-(2-hydroxyethyl)-4-[4methyl-6-- (2,6,6-trimethyl-1-cyclohexen-1-yl)-1E,3E,5E-hexatrienyl] pyridinium), a major fluorescent component of lipofuscin, is a bis-retinoid pyridinium salt that induces blue light damage in RPE cells (Sparrow, Nakanishi et al. 2000; Sparrow and Cai 2001) and nonphotochemically induces apoptosis in RPE cells at physiological concentrations (Suter, Reme et al. 2000). A2E, shown in Figure 1.9, has also been reported to photochemically initiate free radical reactions in organized media (Ragauskaite, Heckathorn et al. 2001) and in solution produce superoxide and peroxyl radicals in the presence of oxygen, which are toxic to cells (Reszka, Eldred et al. 1995). We have recently reported that A2E has also been detected in human BM extracts (Dillon, Wang et al. 2006). As a result of these age-related changes and the relationship between the RPE and Bruchs membrane, studies involving modifications to basement membrane proteins such as laminin and type IV collagen have been of interest. Wang et al. have shown that glycation and nitration of Type IV collagen leads to damaging effects on RPE cell function and viability (Wang, Paik et al. 2005). Also, nonenzymatic glycation of laminin has been shown to decrease self assembly, binding to type IV collagen, and binding of heparin sulfate proteoglycan due to structural deformations following glycation (Federoff, Lawrence et al. 1993). Handa et al. reported the increase of pentosidine and carboxymethyllysine, advanced glycation endproducts, in Bruchs membrane from older human samples

183 (Handa, Verzijl et al. 1999). The presence of AGEs has also been reported in drusen and basal laminar deposits. The accumulation of drusen is considered the predominant morphological change that occurs in aged Bruchs membrane, which may be related to inflammation (Anderson, Mullins et al. 2002). Hagemen et al. reported that several proteins located within drusen were related to the inflammatory or immune response (Hageman, Luthert et al. 2001). During inflammation, large fluxes of nitric oxide (NO) are released through the activation of inducible nitric oxide synthase (Marletta, Yoon et al. 1988; Carreras, Pargament et al. 1994). In addition, nitrite concentration is nearly doubled in the diabetic retina (El-Remessy, Behzadian et al. 2003); serum nitrite levels are elevated in people who smoke and cigarette smoking has been strongly associated with the development of AMD (Solberg, Rosner et al. 1998). Patients with AMD were reported to have significantly higher plasma NO levels over control subjects (Evereklioglu, Er et al. 2003). NO itself is a relatively unreactive radical; however, it is able to form other reactive intermediates including nitrite (NO2-), peroxynitrite (ONOO-), NO2, and N2O3 that can modify proteins, lipids and other compounds. Nitrite is one of the major NO metabolic products and has been used as a marker of NO production (Farrell, Blake et al. 1992; Gaston, Reilly et al. 1993). In addition, nonenzymatic nitration of long-lived proteins such as extracellular matrix (ECM) proteins is a well known pathway that has been associated with inflammation (Bailey, Paul et al. 1998; Paik, Dillon et al. 2001). The ECM proteins, such as collagen and elastin, have been reported to be nonenzymatically modified by nitrite at physiological pH (Paik, Ramey et al. 1997; Paik, Dillon et al. 2001).

184 Studies of basement membrane proteins are experimentally problematic, often hindered by small samples sizes and complex sample compositions. Previous studies have concentrated on creating antibodies that are generally expensive and frequently unreliable. To mitigate these complications, it is advantageous to create model systems that are biologically relevant and then compare these results to data obtained from living tissues. Therefore in this chapter, nonenzymatic glycation and nitration of the basement membrane protein laminin was performed using the Cys-laminin -chain synthetic peptide as a model compound for the modification of basement membrane proteins, which is comparable to the alpha chain of Laminin type 1 within Bruchs membrane (Kanemoto, Reich et al. 1990; Aisenbrey, Zhang et al. 2006). In addition, modifications by A2E are also studied. Following the enzymatic digests, all samples, including control groups, were analyzed using liquid chromatographyelectrospray ionization mass spectrometry (LC/ESI-MS). The results explicitly indicated that fragments containing lysine and arginine residues were preferentially modified in the glycated and irradiated samples. However, nitration of laminin fragments was not observed. Instead several of the fragments ending in a lysine residue appeared to bind to other fragments also ending in a lysine residue, indicating a polymerization-type reaction.

185 Results Laminin modification with glycolaldehyde

The non-enzymatic glycation of basement membrane proteins via the Maillard reaction is of particular interest since this type of modification has been implicated in retinal dysfunction. To simulate the generation of advanced glycation endproducts, one sample of the cys-laminin chain, CSRARKQAASIKVAVSADR (Figure 5.1), was incubated with glycolaldehyde. The reaction scheme for glycolaldeyhe modification of a primary amine is displayed in Figure 5.2. Following tryptic digests, both samples were analyzed via LC-MS and the resulting total ion chromatograms (TIC), were compared to elucidate the glycated fragments. Figure 5.3 displays a typical TIC of the unmodified tryptically digested laminin segment with corresponding mass-to-charge (m/z) ratios of identified fragments. Laminin fragments identified via SEQUEST software in the MS/MS experiments for the control and glycolaldehyde incubated samples were identified using the B and Y ions generated for each fragment (Figure 5.1). The identified fragments from each sample were then compared. The peptides generated were a result of fully and partially enzymatically digested protein. The most abundant peptides identified in the laminin control are displayed in Table 5.1. The most abundant fragments in the control were identified as CSR (Figure 5.4), ARK (Figure 5.5), QAASIK (Figure 5.6), VAVSADR (Figure 5.7), and CSRARKQAASIKVAVSADR (Figure 5.8). The unmodified laminin fragments from the glycolaldehyde incubated laminin

186

Figure 5.1: The amino acid sequence of laminin fragment with B and Y ions identified. (B2 and Y17 are examples of fragments generated after cleavage)

187

Figure 5.2: The reaction scheme for glycation of lysine and arginine within the laminin fragment or with A2E and A2E derived aldehydes

188

2.5x10

180 Relative Absorbance (AU)

2.0x10
8

404

1.5x10

246 592

1.0x10

5.0x10

0.0 0 20 40 60 80 100 120

Time (min)

Figure 5.3: The TIC for a typical enzymatically digested laminin control sample without modification is shown. The m/z ratios corresponding to fragments with amino acid sequences of CSRARK, AR, CSRARKQAASIKVAVSADR, and CSRAR are identified.

189

Table 5.1 Laminin Control: Laminin fragments identified in the control sample including the observed m/z, associated charge, parent ions (MH+), and corresponding amino acid sequences. m/z
365.431 592.696 180.217 329.89 403.667 246.288 324.385 417.982 248.63 722.332 617.718 1316.490 717.795

Charge (z)
1.000 1.000 4.000 4.000 5.000 1.000 3.000 4.000 3.000 2.000 1.000 1.000 1.000

(MH+)
365.431 592.696 720.868 1319.564 2018.335 246.288 973.156 1671.927 745.891 1444.663 617.718 1316.490 717.795

Sequence
[-]CSR[A] [-]CSRAR[K] [-]CSRARK[Q] [-]CSRARKQAASIK[V] [-]CSRARKQAASIKVAVSADR[-] [R]AR[K] [R]ARKQAASIK[V] [R]ARKQAASIKVAVSADR[-] [R]KQAASIK[V] [R]KQAASIKVAVSADR[-] [K]QAASIK[V] [K]QAASIKVAVSADR[-] [K]VAVSADR[-]

190

CSR B2 190.2
4x10
4

Intensity (AU)

3x10

Y1 175.2
2x10
4

B3 347.4 Y2 245.5

1x10

Y3 365.3

0 200 250 300 350

m/z

Figure 5.4: The MS/MS spectrum for digested laminin fragment, CSR (m/z 365), with B and Y ions identified.

191

Y2* 286.3
5x10
5

ARK

4x10

Intensity (AU)

3x10

2x10

1x10

B2 228

B3 B3* Y3 339.1 356.4 374.4

240 260 280 300 320 340 360 380

0 220

m/z

Figure 5.5: The MS/MS spectrum for digested laminin fragment, ARK (m/z 374), with B and Y ions identified.

192

QAASIK
7x10 6x10 5x10 4x10 3x10 2x10 1x10
5

323.4 341.3

Y3*

Y4*

436.1 408.2

Intensity (AU)

B3*

582.3

B6*

Y4 Y2
260.5 254.3
250

426.3 313.5 347.4

300 350

418.4454.1

B5*

601.2
500 550 600 650

B6

618.2

0 400 450

m/z

Figure 5.6: The MS/MS spectrum for digested laminin fragment, QAASIK (m/z 618), with B and Y ions identified.

193

VAVSADR
1.4x10 1.2x10 1.0x10 8.0x10 6.0x10 4.0x10 2.0x10
6

B7*
699.5 718.4

Intensity (AU)

271.2
5

B3

431.3 428.4 448.4 373.4

300 400 500

Y3*

Y4*
530.5
544.5

Y2

B5

Y3

B6

Y5
618.2
600 700

0.0

m/z

Figure 5.7: The MS/MS spectrum for digested laminin fragment, VAVSADR (m/z 718), with B and Y ions identified.

194

CSRARKQAASIKVAVSADR
4.0x10 3.5x10 3.0x10 2.5x10 2.0x10 1.5x10 1.0x10 5.0x10
5

B17 865

Intensity (AU)

267

B6
5

308

Y7 457 B10 566 402 509

B9*

B11

B13 1009 B19 B Y 18 701 B * 14 14 Y15 922 1000 801 728 Y12 785 835 Y16 Y17 958 914.56 B15 637
700 800 900 1000

0.0 400 500 600

m/z

Figure 5.8: The MS/MS spectrum for digested laminin fragment, CSRARKQAASIKVAVSADR (m/z 1009), with B and Y ions identified.

195 sample that were identified vary slightly from the control (Table 5.2). This slight variation is a result of different charges associated with each peptide and the abundance of fragments with incomplete proteolytic digestion of a protein sample, resulting in fragments containing internal cleavage sites. The most abundant modified peptides from the glycolaldehyde incubated sample identified using MS/MS data are reported in Table 5.3, including the unmodified m/z ratios, charge state, and relative intensity of each fragment. In the third column of Table 5.3, the modified masses are reported as the m/z of the laminin fragment with the addition of glycolaldehyde and the loss of water. Fragments identified primarily ended with a lysine or arginine residue, which was expcted. The MS/MS spectra of the most abundant fragments, CSR (Figure 5.9), CSRAR (Figure 5.10), CSRARK (Figure 5.11), QAASIK (Figure 5.12) and CSRARKQAASIKVAVSADR (Figure 5.13) are presumably the result of an incomplete tryptic digest. Because trypsin cleaves peptides at arginine and lysine residues, the incomplete digest can be attributed to trypsin not being able to recognize the arginine and lysine residues of each fragment after modification. The additional fragments reported were consistent with a single site of glycation and a loss of water. The identified sites of modification are highlighted in column four with the identified sequence (Table 5.3). These sites of modification were identified using the B and Y ions generated for the MS/MS scan of each sequence (Hunt, Yates et al. 1986; Mann and Wilm 1994). Modified sites displayed B and Y ions

196

Table 5.2 Glycated Laminin Sample: Laminin fragments (without modifications) identified in the glycated laminin sample including the observed m/z, associated charge, the MH+, and corresponding amino acid sequence. m/z
365.431 197.56 180.217 263.913 403.667 246.288 187.231 324.385 417.982 248.63 722.332 717.795

Charge (z)
1.000 3.000 4.000 5.000 5.000 1.000 2.000 3.000 4.000 3.000 2.000 1.000

MH+
365.431 592.696 720.868 1319.564 2018.335 246.288 374.461 973.156 1671.927 745.891 1444.663 717.795

Sequence
[-]CSR[A] [-]CSRAR[K] [-]CSRARK[Q] [-]CSRARKQAASIK[V] [-]CSRARKQAASIKVAVSADR[-] [R]AR[K] [R]ARK[Q] [R]ARKQAASIK[V] [R]ARKQAASIKVAVSADR[-] [R]KQAASIK[V] [R]KQAASIKVAVSADR[-] [K]VAVSADR[-]

197

Table 5.3 Glycated Laminin: Most abundant laminin fragments modified with glycolaldehyde identified by LC-MS/MS including the observed m/z of the unmodified sequence, the associated charge, the observed m/z of the sequence after modification with glycolaldehyde, and the corresponding amino acid sequence with site of modification highlighted. m/z After Modification [((M+unmod) + 60 18)/ z) = (m/zmod)] 203.7 211.57 381.43 329.86 1051.17 Sequence of Laminin Fragment (Site of Modification Highlighted)

m/z 365.43 592.70 720.87 617.72 2018.34

Charge (z) 2 3 2 2 2

CSR CSRAR CSRARK QAASIK CSRARKQAASIKVAVSADR

198

CSR
6x10
3

5x10

Y2* 142.9

B3* 186.9

4x10

Intensity (AU)

3x10

Y2 152.1 B2* 87.7 204.2 Y3* 195.9

2x10

1x10

0 100 120 140 160 180 200

m/z

Figure 5.9: The MS/MS spectrum for digested laminin fragment, CSR (m/z 204), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue.

199

CSRAR
1.6x10 1.4x10 1.2x10 1.0x10 8.0x10 6.0x10 4.0x10
5

616.2

B5

B4
460.6

Intensity (AU)

Y3
4

402.7

B3
229.7
200 250 300

2.0x10

Y2*

294.7 335
350

389.7

634.6

Y4
531
400 450 500 550 600 650

0.0

m/z

Figure 5.10: The MS/MS spectrum for digested laminin fragment, CSRAR (m/z 634), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue.

200

CSRARK Y5 Y5* 659.8 642.7 762.8

1.6x10 1.4x10 1.2x10

Intensity (AU)

1.0x10 8.0x10 6.0x10 4.0x10

Y4 B4 418.2 B3* 330.1 399.4

300 400 500

Y6* 745.9 B6* 727.6

572.6 Y4* 555.2 B5 599.7

600

2.0x10

Y3*

0.0 700

m/z

Figure 5.11: The MS/MS spectrum for digested laminin fragment, CSRARK (m/z 762), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue.

201

Y5*
1.8x10 1.6x10 1.4x10
5

QAASIK

514.4

Intensity (AU)

1.2x10 1.0x10 8.0x10 6.0x10 4.0x10 2.0x10

Y3* 372.0 B3* 254.2

200 300

Y4 B5 471.2

659.2 B6 642.9

0.0

Y4* 341.1 442.1 B4 Y3

400 500

B4*

600

m/z

Figure 5.12: The MS/MS spectrum for digested laminin fragment, QAASIK (m/z 659), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue.

202

CSRARKQAASIKVAVSADR
4.0x10 3.5x10 3.0x10 2.5x10 2.0x10 1.5x10 1.0x10 5.0x10
3

B17 907

Y6* 350

Intensity (AU)

309 373

Y5

B6 B9* B11 Y7 499 B 10 608 444 551

1051 B Y15 19 B18 743 B * 14 843 Y 964 1043 770 Y8 828 87816 Y17 Y18 956 1000 679 B15 B13
700 800 900 1000

0.0 400 500 600

m/z

Figure 5.13: The MS/MS spectrum for digested laminin fragment, CSRARKQAASIKVAVSADR (m/z 1051), modified by glycolaldehyde. The site of glycation is highlighted in red and the B and Y ions are identified in blue.

203 that contained an additional mass of 42, which corresponded to the addition of glycolaldehyde with the loss of water. By reconstructing the peptide from the individual B and Y ions, the site of modification was determined. The corresponding B and Y ions identified for the most abundant fragments are displayed in blue in their corresponding spectra.

Laminin modification with Carboxymethyllysine (CML)

In addition to modifications of laminin by glycolaldehyde, the data generated for the glycated laminin sample was also analyzed to determine if common advanced glycation end products were present. Specifically, CML was identified in the sample, which forms from nonenzymatic glycation followed by oxidation of proteins (Figure 5.14). The most abundant CML modified peptides from the glycolaldehyde incubated sample identified using MS/MS data are reported in Table 5.4 including the unmodified m/z ratios, charge state, and relative intensity of each fragment. In the third column of Table 5.4, the modified masses are reported as the m/z of the laminin fragment with the addition of CML and the loss of water. Fragments identified primarily ended with a lysine residue. The MS/MS spectrum of CML (m/z 205) with characteristic cleavages is displayed in Figure 5.15. The most abundant laminin fragments modified by CML were ARK, CSRARK, and QAASIK. The MS/MS spectrum and proposed structure of ARK are displayed in Figures 5.16 and 5.17, respectively. The B and Y ions associated with

204

Figure 5.14: The reaction scheme of glycolaldehyde with lysine producing carboxymethyl lysine (CML) and then the modification of primary amines in laminin by CML

205

Table 5.4 - Glycated Laminin: Most abundant laminin fragments modified with CML identified by LC-MS/MS including the observed m/z of the unmodified sequence, the associated charge, the observed m/z of the sequence after modification with glycolaldehyde, and the corresponding amino acid sequence with site of modification highlighted. m/z After Modification [((M+unmod) + 204-18)/ z) = (m/zmod)] 560.43 906.8 803.87 580.36 1195.17

MH+unmod

Charge (z) 1 1 1 2 2

Sequence of Laminin Fragment (Site of Modification Highlighted)

374.43 720.80 617.87 974.72 2018.34

ARK CSRARK QAASIK ARKQAASIK CSRARKQAASIKVAVSADR

206

CML
7x10 6x10 5x10
4

OH

O H N

188.0
OH

187
4

159
4

145

130

NH 2

Intensity (AU)

4x10 3x10 2x10

130.2
4

1x10

145.8
0 140

159.3
160 180

189.8 205.8
200

m/z

Figure 5.15: The MS/MS spectrum of CML located in the glycated laminin sample. The inset is the structure of CML (m/z 205) with characteristic cleavages identified.

207

ARK

B2* 525.6

8.0x10

Intensity (AU)

6.0x10

4.0x10

543.6
2.0x10
4

B1 228.3

Y1 Y1* 333.4 316.3 358.3

300 350 400 450 500 550

0.0 250

m/z

Figure 5.16: The MS/MS spectrum for digested laminin fragment, ARK (m/z 543), modified by CML. The site of modification is highlighted in red and the B and Y ions are identified in blue.

208

Figure 5.17: The proposed structure for CML modification of ARK fragment

209 the cleavages of each amino acid are identified in blue and the site of modification is highlighted in red. The carbonyl carbon forms a double bond with the primary amine on lysines side chain with the subsequent loss of water. In addition to ARK, the MS/MS spectra for laminin fragments, CSRARK (Figure 5.18) and QAASIK (5.19), displayed similar additions in their identified B and Y ions, resulting in structural similarities (Figure 5.20 and 5.21). Both the CSRARK and QAASIK fragments showed that the carbonyl compound of CML formed a double bond with the primary amine on a lysine residue with the loss of water.

Laminin modification with A2E

The bis-retinoid pyridium compound A2E is also of interest as a potential modifier of basement membrane proteins. A2E has previously been shown to mediate blue light-induced apoptosis in RPE cells and to non-photochemically induce apoptosis in RPE cells at physiological concentrations (Sparrow, Nakanishi et al. 2000; Sparrow and Cai 2001). In addition, the photo-oxidation and autooxidation of A2E has been reported to yield a complex mixture of lower molecular weight products, which are generated from a series of cleavages along the acyclic chain forming aldehydes displayed in Figure 5.22 (Wang, Keller et al. 2006). The aldehydes formed could potentially modify basement membrane proteins and, therefore, cause damage to RPE cells. As a result, laminin was incubated with A2E

210

CSRARK
1.6x10 1.4x10 1.2x10
4

Y5* 786.8 704.7

906.6

Intensity (AU)

1.0x10 8.0x10 6.0x10 4.0x10 2.0x10

B*
3

347.3 Y1

B5 B4 574.6 Y3 Y2 560.2 489.4 Y4* 699.7

700

B6 889.9 Y5 803.6

401.2 B2 191.1
200

333.4
300 400

0.0

500

600

800

900

m/z

Figure 5.18: The MS/MS spectrum for digested laminin fragment, CSRARK (m/z 906), modified by CML. The site of modification is highlighted in red and the B and Y ions are identified in blue.

211

QAASIK
1.8x10 1.6x10 1.4x10
4

Y5* 658.76

Intensity (AU)

1.2x10 1.0x10 8.0x10 6.0x10 4.0x10

2.0x10

B3* 254.2
200

B4* 341.3 B4
300 400

B5 471.53

Y3* 516.6 615.4 Y4* 587.68

803.92 B6 785.9

0.0

500

600

700

800

m/z

Figure 5.19: The MS/MS spectrum for digested laminin fragment, QAASIK (m/z 803), modified by CML. The site of modification is highlighted in red and the B and Y ions are identified in blue.

212

Figure 5.20: The proposed structure for CML modification of CSRARK fragment

213

Figure 5.21: The proposed structure for CML modification of QAASIK fragment

214

Figure 5.22: The cleavage positions and the molecular masses of corresponding aldehydes in A2E and oxidized A2E are shown.

215 and subjected to blue light irradiation to compare these effects with modifications from glycolaldehyde. To differentiate between modifications from irriadiated A2E, a non-irradiated control sample was analyzed in conjunction with the irradiated sample. The unmodified fragments identified in the sample irradiated in the presence of A2E differ from the control sample only by the charge associated with each fragment and abundance of fragments with incomplete proteolytic digestion (Table 5.5). Once identified, these peaks were eliminated from both chromatograms, leaving only peaks that were possible sites of modification. The chromatogram of the A2E laminin sample without irradiation, the dark control, contained an abundant A2E peak and multiple laminin peaks; however, the presence of modifications from A2E on the protein fragments was absent, indicating that there are no dark reactions detectable with our experimental conditions. The data for the most abundant modifications resulting from the A2E-laminin sample after irradiation are highlighted in column 2 of Table 5.6. The A2E-laminin sample primarily displayed modifications to arginine and lysine residues, which agreed with the glycated laminin results. However, the number of modifications identified with A2E was significantly more extensive and is evidence of the high reactivity and structural diversity of A2E-derived oxidation products. The modified laminin fragments are consistent with additions of A2E derived aldehydes that predominantly result from cleavages closest to the pyridinium ring in A2E and oxidized A2E. These A2E derived aldehydes correspond to masses of 366, 382, 406, and 422 Da and are

216

Table 5.5 Laminin fragments identified in A2E incubated laminin samples including the observed m/z of the laminin fragment, the associated charge, the MH+, and the corresponding amino acid sequence. m/z
182.716 592.696 180.217 659.782 672.778 374.461 324.385 745.891 722.332 617.718 438.83 717.795

Charge (z)
2.000 1.000 4.000 2.000 3.000 1.000 3.000 1.000 2.000 1.000 3.000 1.000

MH+
365.431 592.696 720.868 1319.564 2018.335 374.461 973.156 745.891 1444.663 617.718 1316.490 717.795

Sequence
[-]CSR[A] [-]CSRAR[K] [-]CSRARK[Q] [-]CSRARKQAASIK[V] [-]CSRARKQAASIKVAVSADR[-] [R]ARK[Q] [R]ARKQAASIK[V] [R]KQAASIK[V] [R]KQAASIKVAVSADR[-] [K]QAASIK[V] [K]QAASIKVAVSADR[-] [K]VAVSADR[-]

217

Table 5.6 Laminin fragments modified with irradiated A2E including the observed m/z of the laminin fragment, the corresponding amino acid sequence with the site of modification highlighted, the associated charge, and the observed masses of laminin with modification A2E aldehydes. Laminin MH+
374.46 592.73 720.87 973.16 1319.56 592.73

Sequence Fragments (Site of

Modification Highlighted)

Charge (Z)
1 1 2 2 2 2

Masses of Known A2E Aldehydes (Da) 366

722.46 --------660.58 ----470.4

382
738.46 956.73 542.4 668.58 841.78 478.4

406
--------554.4 --------490.4

422
778.46 ----562.4 688.58 861.78 ---

ARK CSRAR CSRARK ARKQAASIK CSRARKQAASIK CSRAR

218

9x10 8x10 7x10 6x10

CSRAR (mod. 406)

589 654.7 B4 830.7 789.5 325.5 390 544.1 B3 735.4 B4*

Intensity (AU)

5x10 4x10 3x10 2x10 1x10

Y1 174.2 B2 191.3 Y2* 229.3

455 529.2 Y3*

400 500 600 700 800

945.6 Y5* Y4 877.6 963.7 Y5 980.4

0 200 300 900 1000

m/z

Figure 5.23: The MS/MS spectrum for digested laminin fragment, CSRAR (m/z 980), modified by A2E-derived aldehyde with m/z 406. The site of modification is highlighted in red and the B and Y ions are identified in blue.

219 displayed in Figure 5.22. Each laminin fragment appeared in the TIC multiple times; however, the A2E aldehyde modification associated with each fragment differed. For example, the MS/MS spectrum for CSRAR (Figure 5.23) displayed a fragmentation pattern with B and Y ions indicating that the arginine residue was modified by A2E aldehyde with m/z 406 (Figure 5.24). An additional MS/MS spectrum for this same fragment, CSRAR (Figure 5.25), displayed a fragmentation pattern with B and Y ions indicating that the same arginine residue was modified by a different A2E derived aldehyde with m/z 382 (Figure 5.26). The MS/MS for laminin fragment KQAASIK (Figure 5.27) displayed a fragmentation pattern with ions indicating a modification to the lysine residue by A2E-derived aldehyde with m/z 360 (Figure 5.28). An additional MS/MS spectrum for the same fragment, KQAASIK (Figure 5.29), displayed a fragmentation pattern with B and Y ions indicating the modification to the same lysine residue by A2E-derived aldehyde with m/z 366 (Figure 5.30). Therefore, each laminin fragment displayed in Table 5.6 yielded multiple peaks in the TIC corresponding to modifications of the same amino acid in the same laminin fragment but with a different aldehyde attached. The PDA chromatograms for the laminin control, glycated laminin, and the irradiated A2E laminin samples are displayed in Figure 5.31. The UV-vis maxima for the laminin control and glycated sample were approximately 265, 270, 280, and 295 nm. However, the UV-maxima for the irradiated A2E laminin samples shifted to 330, 340, 360, and 380 nm. The CSR, CSRAR, CSRARK, and

CSRARKQAASIKVAVSADR fragments for each sample are identified in the individual chromatograms as representative peaks for sites of modification. These

220

Figure 5.24: The proposed structure for A2E derived aldehyde (m/z 406) modification of CSRAR fragment

221

9x10 8x10 7x10 6x10

CSRAR (mod. 382)

589 Y1 175.2 B3* 694.7 364

Intensity (AU)

5x10 4x10 3x10 2x10 1x10

B4* 765.7 789.5 B5* 921.6Y *

B2 191.3

325.5 Y3 402.5 544.3 529.1

500 600

296 Y2* 229.3

B3 711.4

5 Y4 939.7 853.6 Y4* 956.6 836.5

0 200 300 400 700 800 900

m/z

Figure 5.25: The MS/MS spectrum for digested laminin fragment, CSRAR (m/z 956), modified by A2E derived aldehyde with m/z 382. The site of modification is highlighted in red and the B and Y ions are identified in blue.

222

Figure 5.26: The proposed structure for A2E derived aldehyde (m/z 382) modification of CSRAR fragment

223

1.4x10 1.2x10 1.0x10

B3 676.9

KQAASIK (mod. 366)

Intensity (AU)

8.0x10 6.0x10 4.0x10 2.0x10

B5* 817.2 Y5* 472.1 Y6* Y5 Y * 352.1 3 489.6 600.3 Y6 401.4 617.3
400 500 600 700

B4* 728.2 Y7 745.2 B6 834.7 947.3 1058.1

900 1000

B5

0.0

800

m/z

Figure 5.27: The MS/MS spectrum for digested laminin fragment, KQAASIK (m/z 1058), modified by A2E derived aldehyde with m/z 366. The site of modification is highlighted in red and the B and Y ions are identified in blue.

224

Figure 5.28: The proposed structure for A2E derived aldehyde (m/z 366) modification of KQAASIK fragment

225

1.4x10 1.2x10 1.0x10

B3 693.9

KQAASIK (mod. 382)

Intensity (AU)

8.0x10 6.0x10 4.0x10 2.0x10

B5* 834.2 Y5* 472.0 Y7* 728.2 Y7 745.2

800

0.0

Y6* Y5 Y * 4 366.9 489.6 600.2 Y6 401.3 617.2

400 500 600 700

B5 B6 851.7 964.6
900 1000

B7* 1075 1092.3

1100

m/z

Figure 5.29: The MS/MS spectrum for digested laminin fragment, KQAASIK (m/z 1092), modified by A2E derived aldehyde with m/z 382. The site of modification is highlighted in red and the B and Y ions are identified in blue.

226

Figure 5.30: The proposed structure for A2E derived aldehyde (m/z 382) modification of KQAASIK fragment

227

3.0x105 2.5x105 2.0x105 1.5x105 1.0x104 5.0x10 0.0

365

180

296 Laminin Control 1009

Relative Abundnce (AU)

1.0x10 5 8.0x10 5 6.0x10 5 4.0x10 5 2.0x10 0.0 2.0x10 5 1.5x10 5 1.0x10 4 5.0x10 0.0 5x10 4 4x10 4 3x10 4 2x10 4 1x10 0 0
4 5

381

317

1051

Glycated Laminin

180 183 296

A2E

A2E Laminin Control

326

A2E 554 494

A2E Laminin Irradiated

20

40

60

Time (min)

Figure 5.31: The HPLC total PDA trace of the laminin control, glycated laminin, A2E laminin control, and irradiated A2E laminin samples are shown respectively. Selected fragments are identified in each chromatogram.

228 CSRARKQAASIKVAVSADR fragments for each sample are identified in the individual chromatograms as representative peaks for sites of modification. These three fragments only varied by specific mass modifications, elution times, and charge states, which are displayed in Table 5.7.

Laminin modification with nitrite

Because nonenzymatic nitration of long lived-proteins has also been associated with retinal dysfunction, laminin was also treated with nitrite and the control was treated with NaCl. After dialysis and tryptic digests, both samples were analyzed via LC-MS and the resulting total ion chromatograms (TIC) were compared to elucidate the nitrated fragments. Tables 5.8 and 5.9 display the unmodified laminin fragments that were found in the control sample and nitrated samples. The unmodified laminin fragments from the nitrated laminin sample that were identified vary slightly from the control. This slight variation is a result of different charges associated with each peptide and the abundance of fragments with incomplete proteolytic digestion of a protein sample, resulting in fragments containing internal cleavage sites. Once identified, peaks common to both samples were removed and the remaining peaks in the nitrated laminin sample were further analyzed. Initially the remaining spectra were analyzed to determine if an addition of 45 or multiple of 45, corresponding to the addition of NO2, to the parent mass was observed. However, this addition could not be found in any of the spectra, suggesting that

229

Table 5.7 Peptide fragments CSR, CSRAR, and CSRARK in the laminin control, glycated laminin, A2E laminin control, and irradiated A2E laminin samples including their corresponding observed m/z, associated charge, and retention time. The irradiated A2E sample also includes the mass of the corresponding A2E aldehyde modification. Peptide Fragment CSR
Observed m/z associated charge Retention time Mass of A2E aldehyde CSRAR Observed m/z associated charge Retention time Mass of A2E aldehyde CSRARK Observed m/z associated charge Retention time Mass of A2E aldehyde CSRARKQAASIKVAVSADR Observed m/z associated charge Retention time Mass of A2E aldehyde 365 1 8 mins 203.7 2 10 mins 183 2 10 mins Not Found -------

Laminin Control

Glycated Laminin

A2E Laminin Control

Irradiated A2E Laminin

278 2 10 mins

317.3 2 12 mins

Not Found -------

326 3 14 mins 406 Da 554.0 2 17 mins 406 Da 494 5 36 472

180 4 10 mins

381.4 2 14 mins

180 4 10 mins

1009 2 35

1051 2 38

Not Found -------

230

Table 5.8 Control Laminin Sample: Laminin fragments identified in the NaCl laminin sample including the observed m/z, associated charge, the MH+, and corresponding amino acid sequence.

m/z
182.416 592.696 240.217 659.782 403.778 374.461 973.385 745.891 722.332 617.718 438.83 717.795

Charge (z)
2.000 1.000 3.000 2.000 5.000 1.000 1.000 1.000 2.000 1.000 3.000 1.000

MH+
364.831 592.696 720.868 1319.564 2018.335 374.461 973.156 745.891 1444.663 617.718 1316.490 717.795

Sequence
[-]CSR[A] [-]CSRAR[K] [-]CSRARK[Q] [-]CSRARKQAASIK[V] [-]CSRARKQAASIKVAVSADR[-] [R]ARK[Q] [R]ARKQAASIK[V] [R]KQAASIK[V] [R]KQAASIKVAVSADR[-] [K]QAASIK[V] [K]QAASIKVAVSADR[-] [K]VAVSADR[-]

231

Table 5.9 Nitrated Laminin Sample: Laminin fragments identified in the nitrated sample including the observed m/z, associated charge, the MH+, and corresponding amino acid sequence. m/z
365.416 592.696 720.217 659.782 672.778 374.461 324.385 745.891 722.332 617.718 438.83 717.795

Charge (z)
1.000 1.000 1.000 2.000 3.000 1.000 3.000 1.000 2.000 1.000 3.000 1.000

MH+
365.431 592.696 720.868 1319.564 2018.335 374.461 973.156 745.891 1444.663 617.718 1316.490 717.795

Sequence
[-]CSR[A] [-]CSRAR[K] [-]CSRARK[Q] [-]CSRARKQAASIK[V] [-]CSRARKQAASIKVAVSADR[-] [R]ARK[Q] [R]ARKQAASIK[V] [R]KQAASIK[V] [R]KQAASIKVAVSADR[-] [K]QAASIK[V] [K]QAASIKVAVSADR[-] [K]VAVSADR[-]

232 nitration of the amino acids did not occur. This may be attributed to the lack of aromatic amino acids within the laminin sequence or possibly the fragments were undergoing another type of reaction. Next, the fragmentation patterns for several of the larger parent ions were analyzed. This data suggested that the laminin fragments ending in a lysine residue were further reacting with other laminin fragments also ending in a lysine residue. The primary amine on a lysine side chain binds to the carbonyl carbon on the second lysine residue with the loss of water. Table 5.10 displays the most abundant fragments that formed in this type of reaction, which include the sequences QAASIKKRA, CSRARKKRARSC, QAASIKKISAAQ, and ARKKRA. The MS/MS spectra for these fragments are displayed Figures 5.32, 5.33, 5.34, and 5.35 followed by the proposed structures in Figures 5.36, 5.37, 5.38, and 5.39. The corresponding B and Y ions for each fragment are identified in blue for each spectra and were used to determine each structure. The PDA chromatograms for the laminin control and the nitrated laminin samples are displayed in Figure 5.40. The UV-vis maxima for the laminin control and nitrated laminin sample were approximately 265, 270, and 280 nm. The ARK, CSRARK, and QAASIK fragments for each sample are identified in the individual chromatograms as representative peaks for sites of modification. These three fragments only varied by specific mass modifications, elution times, and charge states, which are displayed in Table 5.10.

233

Table 5.10 Peptide fragments ARK, CSRARK, and QAASIK in the laminin control and nitrated laminin sample including their corresponding observed m/z, associated charge, and retention time. Peptide Fragment ARK Observed m/z associated charge Retention time CSRARK Observed m/z associated charge Retention time QAASIK Observed m/z associated charge Retention time 374 1 23 mins 720 1 43 mins 618 1 35 mins 730 1 32 mins 711 2 52 mins 608 2 41 mins Laminin Control Laminin Dimer (nitration sample)

234

QAASIKKRA
1.6x10 1.4x10 1.2x10
4

973.1 625.5 B9 955.8 651.3 Y6 692.4 782.5 673.5 523.4 Y B3 341.1 742.4 4 Y5* 478.4 557.6 271.3 B8 599.5 384.5 725.4
200 300 400 500 600 700 800 900

Intensity (AU)

1.0x10 8.0x10 6.0x10 4.0x10 2.0x10

937.7

898.9
3

B4*

0.0

m/z

Figure 5.32: The MS/MS spectrum for digested laminin fragment QAASIKKRA (m/z 973). The B and Y ions are identified in blue.

235

CSRARKKRARSC
1.8x10 1.6x10 1.4x10
5 5

711.8

Intensity (AU)

1.2x10 1.0x10 8.0x10 6.0x10 4.0x10

Y5* 352.4 B8* Y6* 485.6 B4* B6* 416.9 B5* B9* 201.2 278.9 343.4 B * 7 494.1 521 407.5
200 300 400 500

B12 B11* 703.4 642.8 B10Y10* B 11 599.2 651.7 608

600 700

2.0x10

0.0

m/z

Figure 5.33: The MS/MS spectrum for digested laminin fragment CSRARKKRARSC (m/z 711). The B and Y ions are identified in blue.

236

QAASIKKISAAQ
2.0x10
4

B11 535.7

1.5x10

B6* 291.8

B10 491.6 Y11* 500.1 B11* 527.1

Intensity (AU)

B12* 591.2

1.0x10

5.0x10

Y4* 236.8

B9 B8* 412.5 464.6 B7* 421.3 Y8 B6 355.9 456.0 300.8

300 350 400 450 500 550

Y7* 364.9

600.2 608.7

0.0 250 600 650

m/z

Figure 5.34: The MS/MS spectrum for digested laminin fragment QAASIKKISAAQ (m/z 608). The B and Y ions are identified in blue.

237

1.0x10

ARKKRA

B6 712.8

8.0x10

Intensity (AU)

6.0x10

4.0x10

Y4* Y1*
4

2.0x10

0.0

B2* Y* B3* 2 467.6 485.6 583.9 B5* 211.2 622.7 B5 357.4 502.2531.3 228.3 339.4 641.8
200 300 400 500 600 700

B4*

Y4 729.9

m/z

Figure 5.35: The MS/MS spectrum for digested laminin fragment ARKKRA (m/z 729). The B and Y ions are identified in blue.

238

Figure 5.36: The proposed structure for QAASIKKRA

239

Figure 5.37: The proposed structure for CSRARKKRARSC

240

Figure 5.38: The proposed structure for QAASIKKISAAQ

241

Figure 5.39: The proposed structure for ARKKRA

242

Laminin control
5x10 4x10 3x10 2x10
5 5

720 374 618

0 20 40 60 80 100 120 140

Intensity (AU)

1x10

3x10 2x10

Nitrated Laminin 711 608 730

0 20 40 60 80 100 120 140

1x10

Time (min)

Figure 5.40: The HPLC total PDA trace of the laminin control and nitrated laminin samples are shown respectively. Selected fragments, ARK, CSRARK, AND QAASIK, are identified in each chromatogram

243 Discussion

Laminins are ubiquitous multifunctional basement membrane proteins that are the most abundant noncollagenous structural glycoproteins within extracellular matrices. Laminins are involved in copious intricate interactions with themselves, components of the basal lamina, and a variety of other cells facilitating the formation, design, and integrity of the basement membranes (Aumailley and Smyth 1998). The non-enzymatic modifications of proteins via the Maillard reaction have been shown to form advanced glycation endproducts and disrupt function (Ahmed, Dunn et al. 1988; Vlassara, Bucala et al. 1994; Stitt and Vlassara 1999). In this reaction scheme, the carbonyl on a reducing sugar reacts with the nucleophilic amino group on an amino acid, preferentially lysine or arginine, producing a variety of AGEs (Ulrich and Cerami 2001; Yan, Ramasamy et al. 2003). Initially, a condensation reaction produces a Schiff base followed by Amadori rearrangement to produce Amadori products. The Schiff-base and Amadori products can further react through cyclization, enolization, and oxidation to produce numerous AGEs. The type, quantity, and physiological relevance of AGEs produced will depend on the source of the primary amine and the type of sugar used as the reducing agent (Honda, Farboud et al. 2001; Nagaraj, Biswas et al. 2008). Modifications to the sequence or changes to the structure of laminin caused by AGEs can disrupt control of cellular functions (Charonis, Reger et al. 1990; Charonis and Tsilbary 1992; Federoff, Lawrence et al. 1993). Crosslinking mediated via AGEs decreases interactions between the components within the

244 basement membranes, leading to irreversible changes to the integrity and structure of the membrane, tissue rigidity, and reduction in enzymatic susceptibility (Tarsio, Reger et al. 1988; Tsilibary, Charonis et al. 1988; Charonis, Reger et al. 1990). Recent studies have shown that abnormal adhesion, spreading, and proliferation of vascular endothelial cells occurs when cultured on AGE-modified substrates (Haitoglou, Tsilibary et al. 1992; Kalfa, Gerritsen et al. 1995; Paul and Bailey 1999). Since adequate interactions between the vascular cells and the basement membrane are required for normal cell function, AGE accumulation contributes to cell dysfunction by altering receptor recognition (Grant, Kleinman et al. 1990). In addition, Glenn et al. have recently reported an increase in accumulation of autofluorescent material on AGE-modified BM attributed to a decrease in RPE lysosomal enzyme activity (Glenn, Mahaffy et al. 2008), providing evidence that components of the RPE and BM interact. Interestingly, Maeda et al. reported that undigested photoreceptor cell outer segments and basal laminar deposits within the RPE invaded BM (Maeda, Maeda et al. 2008). We have also observed the presence of A2E and its derivatives in organic soluble extracts of human BM, suggesting that A2E does come into contact with material within BM (Dillon, Wang et al. 2006). In addition, previous work has shown that the photo-oxidation of A2E, a bis-retinoid pyridinium compound, forms highly reactive aldehydes (Wang, Keller et al. 2006) that are small enough to diffuse to and in Bruchs membrane. These highly reactive species were hypothesized to cause comparable modifications to laminin as the glycated sample. As a result, studies were performed which showed that A2E preferentially modifies laminin fragments ending in a lysine or arginine residue with

245 aldehydes generated from the two closest cleavages to the pyridinium ring of A2E (m/z = 592) and oxidized A2E (m/z = 608) displayed in Figure 5.22. The presence of modifications to A2E and oxidized A2E indicated that the irradiated sample underwent photo-oxidation and/or auto-oxidation. The extent of modification to laminin by A2E was greater than that of the glycated laminin sample because of the high reactivity of A2E after photo-oxidation. The major modifications were determined by comparing the laminin control, glycolaldehyde-incubated laminin, nitrated laminin and A2E irradiated laminin samples. Unlike typical post translational modifications, these changes result from oxidative stress and are therefore not present in current databases. As a result, sample specific modifications were determined by eliminating control spectra and analysis of remaining peaks instead of typical databases searches. The results for the glycated and A2E irradiated laminin indicated that fragments containing lysine and arginine residues were preferentially modified within this protein. However, in the nitrated laminin sample there was no evidence that nitration occurred. Instead, the lysine residues in the fragments appeared to react by forming longer chains of amino acids. This may suggest that in the presence of NO2, the lysine is reacting to form an adduct similar to the AGEs formed in the glycation reactions. Although several studies have reported that proteins can undergo deaminitation and nitriteinduced crossing linking (Paik and Dillon 2000; Deng 2006; Paik, Saito et al. 2006), studies involving polymierization reactions in the presence of nitrite could not be found, making this a novel reaction. In addition to the comparison of these modifications to laminin, the presence of CML was also identified in the glycated

246 sample. The CML, a common AGE, also caused modification to the laminin fragments, suggesting that oxidative stress accelerates the production of AGEs and modification to proteins, which can cause dysfunction. In a variety of retinal diseases, including AMD, basement membranes are susceptible to alterations in structure and function. These modifications can lead to a decrease in the ability of basement membrane proteins to function normally, leading to a decrease in the associations of components that make up the membrane network, a decrease in the protein present in the membrane, and possibly a more permeable membrane (Charonis, Reger et al. 1990; Charonis and Tsilbary 1992). It is also possible that the protein will exhibit abnormal crosslinking, a decrease in enzymatic susceptibility, and a decrease in solubility. These factors may have a harmful effect on Bruchs membrane, resulting in damage to RPE cells. When the cellular attachments between the RPE cells and basement membrane are disrupted, the RPE and photoreceptor cells die, leading to the onset of AMD. Previous studies have shown that numerous structural changes are induced in Bruchs membrane with age including thickening of the membrane (Pauleikhoff, Harper et al. 1990; Zarbin 2004) and the accumulation of hydrophobic patches of debris related to drusen (Moore, Hussain et al. 1995). The study of modifications from A2E, glycolaldehyde, or other components located within the extra cellular matrix is still incomplete; however, this study provides suggestive evidence that A2E may be involved in modifications to essential basement membrane proteins leading to deleterious changes within the RPE-ECM environment. These preliminary experiments are essential in the identification of such changes in vivo because they

247 give predictable chromatographic and spectroscopic characteristics of those modifications.

CHAPTER 6 CONCLUSIONS AND FUTURE WORK

As a result of the increased number of documented cases and severity of the disease, research focused specifically on the origin and progression of AMD is essential in order to treat patients effectively. Currently, medical treatment to stop the progression of the disease is limited. The characteristic central vision loss associated with AMD is caused by photoreceptor cell death. In wet AMD, the death of these cells is associatd with neovascularization, whereas in dry AMD the death of photoreceptor cells is attributed to damage of the RPE by extracellular deposits such as drusen. However, the exact mechanism leading to the death of these cells and the onset of AMD is still unknown. Recent research has suggested that age-related changes within the RPE and underlying BM may play a crucial role in the development of AMD. Therefore, in this work, the biochemical and cellular changes occurring in the RPE and Bruchs membrane were studied. Our results suggest that multiple factors, including age-related changes, contribute to the pathogenesis of AMD.

249 Compositional Studies of Human Retinal Lipofuscin

The formation and composition of lipofuscin and the major fluorophore A2E have received notable attention. However, the origin of the granules and the identity of most of the compounds and the consequence of A2E accumulation within the granules are still unknown. One hypothesis suggested that A2E could exist in a free or esterified form. In the RPE, all-trans retinol, produced from the visual cycle, is converted to all-trans retinyl ester, which then self-aggregates into a retinosome (Imanishi, Gerke et al. 2004). This prevents hydrophobic interactions with cellular components disrupting normal cell function. Since A2E is extremely hydrophobic and accumulates within RPE lysosomes, A2E was suggested to undergo a similar esterification reaction (Mandal 2008). In addition to the esterification reactions, a second hypothesis involving the modification of A2E by A2E-derived aldehydes was also suggested. Within the acidic lysosomal environment, A2E undergoes rearrangements and oxidation, generating aldehydes and ketones that are structurally similar to -carotene oxidation products. These aldehydes are extremely reactive and in the presence of A2E may interact, forming higher molecular weight products. Therefore, in this study, lipofuscin was analyzed to investigate the hydrophobic compounds that chromatographically elute later than A2E and that absorb radiation with wavelengths greater than 400 nm. The results indicate that a large quantity of the components of lipofuscin have mass spectra analogous to that of A2E, as determined by their fragmentation pattern with losses of 190, 174 and/or 150 amu and the formation of fragments of ca 592 amu but with higher molecular

250 weights. The vast majority of the relatively hydrophobic components correspond to derivatized A2E with discrete molecular weights of m/z 800-900, m/z 970-1080 and above m/z 1200 regions. The majority of modifications are much more hydrophobic than A2E, increasing its log P, and probably explain the formation of lipofuscin granules in the RPE. The spectroscopic characteristics and fragmentation patterns associated with these compounds supports the hypothesis that A2E is reacting with aldehydes such as all-trans-retinal, A2E-derived aldehydes, by studying model reactions with cinnamaldehyde and benzaldehyde to form the higher molecular weight compounds found in lipofuscin. This study is part of a continuing effort to identify the molecular modifications to the structure of A2E. Further experiments are being performed to confirm the structure of some of these higher molecular weight products such as the compounds with m/z 920 and 1188. However, this process is complicated by small sample sizes. One of the main challenges is to collect enough pure A2E to react with RAL, producing mM quantities of these higher molecular weight products so that 1H and 13C NMR can be performed. In addition, the fragmentation patterns of additional higher molecular weight compounds in the lipofuscin sample are still being analyzed to determine the structures.

Accumulation of 3-nitrotyrosine and nitro-A2E in Bruchs membrane

The inflammatory response has also been associated with the development of AMD. Recently four independent research groups have identified a commonly

251 inherited variant (Y402H) of the complement factor H gene in the genome from different groups of AMD patients (Edwards, Ritter et al. 2005; Hageman, Anderson et al. 2005; Haines, Hauser et al. 2005; Klein, Zeiss et al. 2005). The Y402H variant of CFH significantly increases the risk of AMD and links the genetics of the disease with inflammation. During inflammation there is activation of inducible nitric oxide synthase and release of nitric oxide (Marletta, Yoon et al. 1988; Carreras, Pargament et al. 1994), which in principle could lead to non-enzymatic nitration within extracellular deposits and/or intrinsic extracellular matrix protein components of human Bruchs membrane. Modifying ECM proteins in Bruchs membrane can result in changes to the ECM proteins similar to age-related changes, such as protein yellowing and cross-linking. Therefore, the second part of this work investigated the modifications to intrinsic proteins and extrinsic deposits and A2E in human Bruch's membrane by reactive nitrogen species released during inflammation. We have identified an increasing accumulation of 3-nitrotyrosine and nitro-A2E in human Bruch's membrane with advancing patient age. To our knowledge, the current study represents the first clear demonstration of inflammation-related chemical modifications detected in human Bruchs membrane. The presence of 3nitrotyrosine and nitro-A2E may be important biomarkers for immune-mediated processes during aging, and the role of this process in the development of agerelated macular degeneration. However, further experiments are needed to evaluate other aspects of this process, including the relationship between the degree of nitration and the age/medical history of the donor, the effect of nitration on the turnover of intrinsic extracellular matrix proteins, and determination of the three-

252 dimensional structural changes resulting from nitration and the effects these changes have on cellular function.

Modifications to Laminin

Chemical modifications to basement membrane proteins may deleteriously affect Bruchs membrane, leading to the development of AMD. These modifications can lead to a decrease in the basement membrane proteins ability to function normally, leading to a decrease in the association of components that make up the membrane network, a decrease in the protein present in the membrane, and possibly a more permeable membrane (Charonis, Reger et al. 1990; Charonis and Tsilbary 1992). It is also possible that the protein will exhibit abnormal crosslinking, a decrease in enzymatic susceptibility, and a decrease in solubility. These factors may have a harmful effect on Bruchs membrane, resulting in damage to RPE cells and the onset of AMD. Previous studies have shown that numerous structural changes are induced in Bruchs membrane with age including the accumulation of hydrophobic patches of debris, which may be related to drusen (Moore, Hussain et al. 1995). In addition, the formation of advanced glycation endproducts (AGEs) and associated damage has been associated with a variety of diseases including diabetic retinopathy. Therefore, the purpose of this study was to investigate modifications from AGEs (glycolaldehyde-derived AGEs and CML), inflammation, (nitrite), and AMD (A2E) on the membrane-like protein fragment laminin as a model for aging of

253 Bruchs membrane in age-related eye diseases. Modifications to laminin by glycolaldehyde, CML, and A2E occurred preferentially on lysine or arginine residues. The A2E-modified laminin fragments are consistent with additions of A2E-derived aldehydes resulting from cleavages closest to the pyridinium ring in A2E and oxidized A2E. Although direct nitration of the laminin fragment was not observed, the laminin fragments ending in a lysine residue appeared to undergo dimerization. This suggests that in the presence of NO2, the laminin fragments may react to create higher molecular weight products similar to AGEs, which may result in damage to Bruchs membrane and the overlying RPE. This is a novel reaction induced by nitration. These results provide evidence that A2E and AGEs may be involved in modifications to essential basement membrane proteins leading to deleterious changes in the retinal pigment epithelium extracellular matrix (RPEECM) environment. The study of modifications from A2E, glycolaldehyde, or other components located within the extracellular matrix is still incomplete. Further analysis of other AGEs, A2E, and A2E-derived aldehydes present in the laminin samples is being performed. However, this study provides suggestive evidence that A2E may be involved in modifications to essential basement membrane proteins leading to deleterious changes. The pathogenesis of AMD is multifactorial. RPE cells and Bruchs

membrane are functionally and structurally interdependent. This work focused on the study of biochemical and cellular changes to the RPE and Bruchs membrane in order to understand the modifications and mechanism of formation in vivo, which

254 may provide important information related to the development of AMD and the development of an effective treatment.

REFERENCES

(2002). "Biology of Amd." Retrieved May 1, 2009, from http://www.csmd.ucsb.edu/biology/drusen_tem.jpg. Ahmed, M. U., J. A. Dunn, et al. (1988). "Oxidative Degradation of Glucose Adducts to Protein. Formation of 3-(Ne-Lysino)-Lactic Acid from Model Compounds and Glycated Proteins." J. Biol. Chem. 263(18): 8816-21. Ahmed, S. S., M. N. Lott, et al. (2005). "The Macular Xanthophylls." Surv Ophthalmol 50(2): 183-93. Aisenbrey, S., M. Zhang, et al. (2006). "Retinal Pigment Epithelial Cells Synthesize Laminins, Including Laminin 5, and Adhere to Them through Alpha3- and Alpha6-Containing Integrins." Invest Ophthalmol Vis Sci 47(12): 5537-44. Alizadeh, M., M. Wada, et al. (2001). "Downregulation of Differentiation Specific Gene Expression by Oxidative Stress in ARPE-19 Cells." Invest Ophthalmol Vis Sci 42(11): 2706-13. Anderson, D. H., R. F. Mullins, et al. (2002). "A Role for Local Inflammation in the Formation of Drusen in the Aging Eye." Am J Ophthalmol 134(3): 411-31. Anderson, D. H., R. F. Mullins, et al. (2002). "A Role for Local Inflammation in the Formation of Drusen in the Aging Eye." Am. J. Ophthalmol. 134(3): 411431. Anderson, D. H., S. Ozaki, et al. (2001). "Local Cellular Sources of Apolipoprotein E in the Human Retina and Retinal Pigmented Epithelium: Implications for the Process of Drusen Formation." Am J Ophthalmol 131(6): 767-81. Armstrong, N. R., R. K. Quinn, et al. (1974). "Voltammetry in Sulfolane: The Electrochemical Behavior of Benzaldehyde and Substituted Benzaldehydes." Anal Chem(46): 1759. Aumailley, M. and N. Smyth (1998). "The Role of Laminins in Basement Membrane Function." J. Anat. 193(1): 1-21. Avalle, L. B., Z. Wang, et al. (2004). "Observation of A2E Oxidation Products in Human Retinal Lipofuscin." Exp Eye Res 78(4): 895-8.

Bailey, A. J., R. G. Paul, et al. (1998). "Mechanisms of Maturation and Ageing of Collagen." Mech Ageing Dev 106(1-2): 1-56. Bailey, T. A., N. Kanuga, et al. (2004). "Oxidative Stress Affects the Junctional Integrity of Retinal Pigment Epithelial Cells." Invest Ophthalmol Vis Sci 45(2): 675-84. Ballatori, N., C. L. Hammond, et al. (2005). "Molecular Mechanisms of Reduced Glutathione Transport: Role of the MRP/CFTR/ABCC and OATP/SLC21A Families of Membrane Proteins." Toxicol Appl Pharmacol 204(3): 238-55. Baynes, J. W., N. G. Watkins, et al. (1989). The Amadori Product on Protein: Structure and Reactions. The Maillard Reaction in Aging, Diabetes and Nutrition 43-67. Beatty, S., H. Koh, et al. (2000). "The Role of Oxidative Stress in the Pathogenesis of Age-Related Macular Degeneration." Surv Ophthalmol 45(2): 115-34. Behar-Cohen, F. F., O. Goureau, et al. (1996). "Decreased Intraocular Pressure Induced by Nitric Oxide Donors Is Correlated to Nitrite Production in the Rabbit Eye." Invest Ophthalmol Vis Sci 37(8): 1711-5. Ben-Shabat, S., Y. Itagaki, et al. (2002). "Formation of a Nonaoxirane from A2E, a Lipofuscin Fluorophore Related to Macular Degeneration, and Evidence of Singlet Oxygen Involvement." Angew Chem Int Ed Engl 41(5): 814-7. Bergmann, M., F. Schutt, et al. (2004). "Inhibition of the ATP-Driven Proton Pump in RPE Lysosomes by the Major Lipofuscin Fluorophore A2-E May Contribute to the Pathogenesis of Age-Related Macular Degeneration." FASEB J 18(3): 562-4. Borland, C. and T. Higenbottam (1987). "Nitric Oxide Yields of Contemporary UK, US and French Cigarettes." Int J Epidemiol 16(1): 31-4. Boulton, M., A. Dontsov, et al. (1993). "Lipofuscin Is a Photoinducible Free Radical Generator." J Photochem Photobiol B 19(3): 201-4. Boulton, M., N. M. McKechnie, et al. (1989). "The Formation of Autofluorescent Granules in Cultured Human RPE." Invest Ophthalmol Vis Sci 30(1): 82-9. Boyer, M. M., G. L. Poulsen, et al. (2000). "Relative Contributions of the Neurosensory Retina and Retinal Pigment Epithelium to Macular Hypofluorescence." Arch Ophthalmol 118(1): 27-31. Bridges, C. D., R. A. Alvarez, et al. (1982). "Vitamin A in Human Eyes: Amount, Distribution, and Composition." Invest Ophthalmol Vis Sci 22(6): 706-14.

257 Brown, G. C., M. M. Brown, et al. (2005). "The Burden of Age-Related Macular Degeneration: A Value-Based Medicine Analysis." Trans Am Ophthalmol Soc 103: 173-84; discussion 184-6. Brown, P. K. and G. Wald (1964). "Visual Pigments in Single Rods and Cones of the Human Retina. Direct Measurements Reveal Mechanisms of Human Night and Color Vision." Science 144: 45-52. Bui, T. V., Y. Han, et al. (2006). "Characterization of Native Retinal Fluorophores Involved in Biosynthesis of A2E and Lipofuscin-Associated Retinopathies." J Biol Chem 281(26): 18112-9. Bunn, H. F. and P. J. Higgins (1981). "Reaction of Monosaccharides with Proteins: Possible Evolutionary Significance." Science 213(4504): 222-4. Burke, J. M., F. Cao, et al. (1999). "Expression of E-Cadherin by Human Retinal Pigment Epithelium: Delayed Expression in Vitro." Invest Ophthalmol Vis Sci 40(12): 2963-70. Calvin, H. I., C. Medvedovsky, et al. (1986). "Near-Total Glutathione Depletion and Age-Specific Cataracts Induced by Buthionine Sulfoximine in Mice." Science 233(4763): 553-5. Carreras, M. C., G. A. Pargament, et al. (1994). "Kinetics of Nitric Oxide and Hydrogen Peroxide Production and Formation of Peroxynitrite During the Respiratory Burst of Human Neutrophils." FEBS Lett 341(1): 65-8. Cassin, B. and S. Solomon (1990). Dictionary of Eye Terminology. Gainesville, Florida, Triad Publishing Company. Charonis, A. S., L. A. Reger, et al. (1990). "Laminin Alterations after in Vitro Nonenzymatic Glycosylation." Diabetes 39(7): 807-14. Charonis, A. S. and E. C. Tsilbary (1992). "Structural and Functional Changes of Laminin and Type Iv Collagen after Nonenzymatic Glycation." Diabetes 41 Suppl 2: 49-51. Chen, M., J. V. Forrester, et al. (2007). "Synthesis of Complement Factor H by Retinal Pigment Epithelial Cells Is Down-Regulated by Oxidized Photoreceptor Outer Segments." Exp Eye Res 84(4): 635-45. Cheng, Y. and M. Gao (2005). "The Effect of Glycation of CD59 on ComplementMediated Cytolysis." Cell Mol Immunol 2(4): 313-7.

258 Chong, N. H., J. Keonin, et al. (2005). "Decreased Thickness and Integrity of the Macular Elastic Layer of Bruch's Membrane Correspond to the Distribution of Lesions Associated with Age-Related Macular Degeneration." Am J Pathol 166(1): 241-51. Chung, H. Y., H. J. Kim, et al. (2001). "The Inflammation Hypothesis of Aging: Molecular Modulation by Calorie Restriction." Ann N Y Acad Sci 928: 32735. Coffey, A. J. and S. Brownstein (1986). "The Prevalence of Macular Drusen in Postmortem Eyes." Am J Ophthalmol 102(2): 164-71. Cohen, S. M., K. L. Olin, et al. (1994). "Low Glutathione Reductase and Peroxidase Activity in Age-Related Macular Degeneration." Br J Ophthalmol 78(10): 791-4. Coral, K., R. Raman, et al. (2006). "Plasma Homocysteine and Total Thiol Content in Patients with Exudative Age-Related Macular Degeneration." Eye 20(2): 203-7. Cornwall, M. C. (2009). "Department of Physiology and Biophysics Boston University School of Medicine." Retrieved May 1, 2009, from http://biophysics.bumc.bu.edu/faculty/cornwall/images/carter.jpg. Council, N. R. (1986). Envirnmental Tobacco Smoke: Measuring Esxposure and Assessing Health Effects. Washington D.C., National Academy Press. Crabb, J. W., M. Miyagi, et al. (2002). "Drusen Proteome Analysis: An Approach to the Etiology of Age-Related Macular Degeneration." Proc Natl Acad Sci U S A 99(23): 14682-7. Crane, I. J. and J. Liversidge (2008). "Mechanisms of Leukocyte Migration across the Blood-Retina Barrier." Semin Immunopathol 30(2): 165-77. Crowley, J. R., K. Yarasheski, et al. (1998). "Isotope Dilution Mass Spectrometric Quantification of 3-Nitrotyrosine in Proteins and Tissues Is Facilitated by Reduction to 3-Aminotyrosine." Anal Biochem 259(1): 127-35. Curcio, C. A., C. L. Millican, et al. (2001). "Accumulation of Cholesterol with Age in Human Bruch's Membrane." Invest Ophthalmol Vis Sci 42(1): 265-74. Curcio, C. A., J. B. Presley, et al. (2005). "Esterified and Unesterified Cholesterol in Drusen and Basal Deposits of Eyes with Age-Related Maculopathy." Exp Eye Res 81(6): 731-41.

259 D'Souza, Y. B., C. J. Jones, et al. (2009). "Comparison of Lectin Binding of Drusen, RPE, Bruch's Membrane, and Photoreceptors." Mol Vis 15: 906-11. Das, A., R. N. Frank, et al. (1990). "Ultrastructural Localization of Extracellular Matrix Components in Human Retinal Vessels and Bruch's Membrane." Arch Ophthalmol 108(3): 421-9. David, L. L. and T. R. Shearer (1984). "State of Sulfhydryl in Selenite Cataract." Toxicol Appl Pharmacol 74(1): 109-15. Davies, S., M. H. Elliott, et al. (2001). "Photocytotoxicity of Lipofuscin in Human Retinal Pigment Epithelial Cells." Free Radic Biol Med 31(2): 256-65. Davis, A. A., P. S. Bernstein, et al. (1995). "A Human Retinal Pigment Epithelial Cell Line That Retains Epithelial Characteristics after Prolonged Culture." Invest Ophthalmol Vis Sci 36(5): 955-64. Davis, M. D., R. E. Gangnon, et al. (2005). "The Age-Related Eye Disease Study Severity Scale for Age-Related Macular Degeneration: Areds Report No. 17." Arch Ophthalmol 123(11): 1484-98. Dawson, V. L. (1995). "Nitric Oxide: Role in Neurotoxicity." Clin Exp Pharmacol Physiol 22(4): 305-8. Del Priore, L. V. and T. H. Tezel (1998). "Reattachment Rate of Human Retinal Pigment Epithelium to Layers of Human Bruch's Membrane." Arch Ophthalmol 116(3): 335-41. Deng, H. (2006). "Nitrite Anions Induce Nitrosative Deamination of Peptides and Proteins." Rapid Commun Mass Spectrom 20(24): 3634-8. Di Stasi, A. M., C. Mallozzi, et al. (1999). "Peroxynitrite Induces Tryosine Nitration and Modulates Tyrosine Phosphorylation of Synaptic Proteins." J Neurochem 73(2): 727-35. Dillon, J., Z. Wang, et al. (2004). "The Photochemical Oxidation of A2E Results in the Formation of a 5,8,5',8'-Bis-Furanoid Oxide." Exp Eye Res 79(4): 53742. Dillon, J., L. Zheng, et al. (1999). "The Optical Properties of the Anterior Segment of the Eye: Implications for Cortical Cataract." Exp Eye Res 68(6): 785-95. Dillon, J. P., Z. Wang, et al. (2006). "Detection of Nitro-Adducts by Lc/Ms in Aged Human Bruch's Membrane." Invest Ophthalmol Vis Sci 47: 1399.

260 Dorey, C. K., G. Wu, et al. (1989). "Cell Loss in the Aging Retina. Relationship to Lipofuscin Accumulation and Macular Degeneration." Invest Ophthalmol Vis Sci 30(8): 1691-9. Dorey, C. K., G. Wu, et al. (1989). "Cell Loss in the Aging Retina. Relationship to Lipofuscin Accumulation and Macular Degeneration." Invest Ophthalmol Vis Sci 30(8): 1691-9. Dua, H. S., A. McKinnon, et al. (1991). "Ultrastructural Pathology of the 'Barrier Sites' in Experimental Autoimmune Uveitis and Experimental Autoimmune Pinealitis." Br J Ophthalmol 75(7): 391-7. Edwards, A. O., R. Ritter, 3rd, et al. (2005). "Complement Factor H Polymorphism and Age-Related Macular Degeneration." Science 308(5720): 421-4. El-Remessy, A. B., M. A. Behzadian, et al. (2003). "Experimental Diabetes Causes Breakdown of the Blood-Retina Barrier by a Mechanism Involving Tyrosine Nitration and Increases in Expression of Vascular Endothelial Growth Factor and Urokinase Plasminogen Activator Receptor." Am J Pathol 162(6): 19952004. Eldred, G. E. and M. L. Katz (1988). "Fluorophores of the Human Retinal Pigment Epithelium: Separation and Spectral Characterization." Exp Eye Res 47(1): 71-86. Eldred, G. E. and M. R. Lasky (1993). "Retinal Age Pigments Generated by SelfAssembling Lysosomotropic Detergents." Nature 361(6414): 724-6. Erickson, K. K., J. M. Sundstrom, et al. (2007). "Vascular Permeability in Ocular Disease and the Role of Tight Junctions." Angiogenesis 10(2): 103-17. Evereklioglu, C., H. Er, et al. (2003). "Nitric Oxide and Lipid Peroxidation Are Increased and Associated with Decreased Antioxidant Enzyme Activities in Patients with Age-Related Macular Degeneration." Doc Ophthalmol 106(2): 129-36. Farrell, A. J., D. R. Blake, et al. (1992). "Increased Concentrations of Nitrite in Synovial Fluid and Serum Samples Suggest Increased Nitric Oxide Synthesis in Rheumatic Diseases." Ann Rheum Dis 51(11): 1219-22. Fawcett, R. D. and A. Lasia (1981). "The Electroreduction of Aromatic Aldehydes in Aprotic Solvents." Canadian Journal Chemistry(59): 3256. Federoff, H. J., D. Lawrence, et al. (1993). "Nonenzymatic Glycosylation of Laminin and the Laminin Peptide Cikvavs Inhibits Neurite Outgrowth." Diabetes 42(4): 509-13.

261 Feeney-Burns, L. and G. E. Eldred (1983). "The Fate of the Phagosome: Conversion to 'Age Pigment' and Impact in Human Retinal Pigment Epithelium." Trans Ophthalmol Soc U K 103 ( Pt 4): 416-21. Feeney-Burns, L. and M. R. Ellersieck (1985). "Age-Related Changes in the Ultrastructure of Bruch's Membrane." Am J Ophthalmol 100(5): 686-97. Feeney-Burns, L., C. L. Gao, et al. (1988). "The Fate of Immunoreactive Opsin Following Phagocytosis by Pigment Epithelium in Human and Monkey Retinas." Invest Ophthalmol Vis Sci 29(5): 708-19. Feeney-Burns, L., C. L. Gao, et al. (1987). "Lysosomal Enzyme Cytochemistry of Human RPE, Bruch's Membrane and Drusen." Invest Ophthalmol Vis Sci 28(7): 1138-47. Feeney-Burns, L., E. S. Hilderbrand, et al. (1984). "Aging Human RPE: Morphometric Analysis of Macular, Equatorial, and Peripheral Cells." Invest Ophthalmol Vis Sci 25(2): 195-200. Feeney, L. (1978). "Lipofuscin and Melanin of Human Retinal Pigment Epithelium. Fluorescence, Enzyme Cytochemical, and Ultrastructural Studies." Invest Ophthalmol Vis Sci 17(7): 583-600. Ferris, F. L., M. D. Davis, et al. (2005). "A Simplified Severity Scale for AgeRelated Macular Degeneration: Areds Report No. 18." Arch Ophthalmol 123(11): 1570-4. Finnemann, S. C., L. W. Leung, et al. (2002). "The Lipofuscin Component A2E Selectively Inhibits Phagolysosomal Degradation of Photoreceptor Phospholipid by the Retinal Pigment Epithelium." Proc Natl Acad Sci U S A 99(6): 3842-7. Fliesler, S. J. (2002). Sterols and Oxysterols : Chemistry, Biology, and Pathobiology. Trivandrum, Research Signpost. Friedman, D. S., B. J. O'Colmain, et al. (2004). "Prevalence of Age-Related Macular Degeneration in the United States." Arch Ophthalmol 122(4): 564-72. Gaillard, E. R., S. J. Atherton, et al. (1995). "Photophysical Studies on Human Retinal Lipofuscin." Photochem Photobiol 61(5): 448-53. Gaillard, E. R., L. Zheng, et al. (2000). "Age-Related Changes in the Absorption Characteristics of the Primate Lens." Invest Ophthalmol Vis Sci 41(6): 14549.

262 Gaston, B., J. Reilly, et al. (1993). "Endogenous Nitrogen Oxides and Bronchodilator S-Nitrosothiols in Human Airways." Proc Natl Acad Sci U S A 90(23): 10957-61. Gates, P. (2004). "Electrospray Ionization." 2009, from http://www.chm.bris.ac.uk/ms/images/esi-mechanism.gif. Gates, P. (2004). "Quadruple Ion Trap (Qit) Mass Analysis." 2009, from http://www.chm.bris.ac.uk/ms/images/iontrap-schematic.gif. Glenn, J. V., J. R. Beattie, et al. (2007). "Confocal Raman Microscopy Can Quantify Advanced Glycation End Product (AGE) Modifications in Bruch's Membrane Leading to Accurate, Nondestructive Prediction of Ocular Aging." FASEB J 21(13): 3542-52. Glenn, J. V., H. Mahaffy, et al. (2008). "AGE-Modified Substrate Induces Global Gene Expression Changes in ARPE-19 Monolayers: Relevance to Lysosomal Dysfunction and Lipofuscin Accumulation." Invest Ophthalmol Vis Sci. Glenn, J. V., H. Mahaffy, et al. (2009). "Advanced Glycation End Product (AGE) Accumulation on Bruch's Membrane: Links to Age-Related RPE Dysfunction." Invest Ophthalmol Vis Sci 50(1): 441-51. Goldstein, B. E. (2007). Sensation & Perception Canada, Thompson Wadsworth. Grant, D. S., H. K. Kleinman, et al. (1990). "The Role of Basement Membranes in Vascular Development." Ann N Y Acad Sci 588: 61-72. Grossniklaus, H. E., A. K. Hutchinson, et al. (1994). "Clinicopathologic Features of Surgically Excised Choroidal Neovascular Membranes." Ophthalmology 101(6): 1099-111. Grossniklaus, H. E., J. X. Ling, et al. (2002). "Macrophage and Retinal Pigment Epithelium Expression of Angiogenic Cytokines in Choroidal Neovascularization." Mol Vis 8: 119-26. Gu, X., X. Yuan, et al. (2009). "Comparison of Proteins in Dry and Wet Amd Bruch's Membrane " Invest Ophthalmol Vis Sci(50): 2343. Hageman, G. S., D. H. Anderson, et al. (2005). "A Common Haplotype in the Complement Regulatory Gene Factor H (Hf1/Cfh) Predisposes Individuals to Age-Related Macular Degeneration." Proc Natl Acad Sci U S A 102(20): 7227-32.

263 Hageman, G. S., P. J. Luthert, et al. (2001). "An Integrated Hypothesis That Considers Drusen as Biomarkers of Immune-Mediated Processes at the RPE-Bruch's Membrane Interface in Aging and Age-Related Macular Degeneration." Prog Retin Eye Res 20(6): 705-32. Hageman, G. S., P. J. Luthert, et al. (2001). "An Integrated Hypothesis That Considers Drusen as Biomarkers of Immune-Mediated Processes at the RPE-Bruch's Membrane Interface in Aging and Age-Related Macular Degeneration." Prog. Retinal Eye Res. 20(6): 705-732. Hageman, G. S. and R. F. Mullins (1999). "Molecular Composition of Drusen as Related to Substructural Phenotype." Mol Vis 5: 28. Hageman, G. S., R. F. Mullins, et al. (1999). "Vitronectin Is a Constituent of Ocular Drusen and the Vitronectin Gene Is Expressed in Human Retinal Pigmented Epithelial Cells." Faseb J 13(3): 477-84. Haines, J. L., M. A. Hauser, et al. (2005). "Complement Factor H Variant Increases the Risk of Age-Related Macular Degeneration." Science 308(5720): 419-21. Haitoglou, C. S., E. C. Tsilibary, et al. (1992). "Altered Cellular Interactions between Endothelial Cells and Nonenzymatically Glucosylated Laminin/Type Iv Collagen." J Biol Chem 267(18): 12404-7. Hammond, C. L., T. K. Lee, et al. (2001). "Novel Roles for Glutathione in Gene Expression, Cell Death, and Membrane Transport of Organic Solutes." J Hepatol 34(6): 946-54. Handa, J. T., N. Verzijl, et al. (1999). "Increase in the Advanced Glycation End Product Pentosidine in Bruch's Membrane with Age." Invest Ophthalmol Vis Sci 40(3): 775-9. Hansen, J. M., E. W. Carney, et al. (2001). "Altered Differentiation in Rat and Rabbit Limb Bud Micromass Cultures by Glutathione Modulating Agents." Free Radic Biol Med 31(12): 1582-92. Haralampus-Grynaviski, N. M., L. E. Lamb, et al. (2003). "Spectroscopic and Morphological Studies of Human Retinal Lipofuscin Granules." Proc Natl Acad Sci U S A 100(6): 3179-84. Hayase, T., K. Yamamoto, et al. (1996). "Grazing Bullet Wounds on the Tongue and Liver." Nihon Hoigaku Zasshi 50(4): 268-71.

264 Hayes, K. C. (1974). "Retinal Degeneration in Monkeys Induced by Deficiencies of Vitamin E or A." Invest Ophthalmol 13(7): 499-510. Heller, J. and D. Bok (1976). "Transport of Retinol from the Blood to the Retina: Involvement of High Molecular Weight Lipoproteins as Intracellular Carriers." Exp Eye Res 22(5): 403-10. Hewitt, A. T., K. Nakazawa, et al. (1989). "Analysis of Newly Synthesized Bruch's Membrane Proteoglycans." Invest Ophthalmol Vis Sci 30(3): 478-86. Ho, T. C. and L. V. Del Priore (1997). "Reattachment of Cultured Human Retinal Pigment Epithelium to Extracellular Matrix and Human Bruch's Membrane." Invest Ophthalmol Vis Sci 38(6): 1110-8. Hogan, M. J. (1972). "Role of the Retinal Pigment Epithelium in Macular Disease." Trans Am Acad Ophthalmol Otolaryngol 76(1): 64-80. Hollyfield, J. G., V. L. Bonilha, et al. (2008). "Oxidative Damage-Induced Inflammation Initiates Age-Related Macular Degeneration." Nat Med 14(2): 194-8. Holz, F. G., C. Bellman, et al. (2001). "Fundus Autofluorescence and Development of Geographic Atrophy in Age-Related Macular Degeneration." Invest Ophthalmol Vis Sci 42(5): 1051-6. Holz, F. G., F. Schutt, et al. (1999). "Inhibition of Lysosomal Degradative Functions in RPE Cells by a Retinoid Component of Lipofuscin." Invest Ophthalmol Vis Sci 40(3): 737-43. Honda, S., B. Farboud, et al. (2001). "Induction of an Aging Mrna Retinal Pigment Epithelial Cell Phenotype by Matrix-Containing Advanced Glycation End Products in Vitro." Invest Ophthalmol Vis Sci 42(10): 2419-25. Howes, K. A., Y. Liu, et al. (2004). "Receptor for Advanced Glycation End Products and Age-Related Macular Degeneration." Invest Ophthalmol Vis Sci 45(10): 3713-20. Hu, J., L. Dong, et al. (2008). "The Redox Environment in the Mitochondrial Intermembrane Space Is Maintained Separately from the Cytosol and Matrix." J Biol Chem 283(43): 29126-34. Hunt, D. F., J. R. Yates, 3rd, et al. (1986). "Protein Sequencing by Tandem Mass Spectrometry." Proc Natl Acad Sci U S A 83(17): 6233-7. Huotari, V., R. Sormunen, et al. (1995). "The Polarity of the Membrane Skeleton in Retinal Pigment Epithelial Cells of Developing Chicken Embryos and in Primary Culture." Differentiation 58(3): 205-15.

265 Hwang, C., A. J. Sinskey, et al. (1992). "Oxidized Redox State of Glutathione in the Endoplasmic Reticulum." Science 257(5076): 1496-502. Ignarro, L. J. (1996). "Physiology and Pathophysiology of Nitric Oxide." Kidney Int Suppl 55: S2-5. Imanishi, Y., V. Gerke, et al. (2004). "Retinosomes: New Insights into Intracellular Managing of Hydrophobic Substances in Lipid Bodies." J Cell Biol 166(4): 447-53. Iriyama, A., R. Fujiki, et al. (2008). "A2E, a Pigment of the Lipofuscin of Retinal Pigment Epithelial Cells, Is an Endogenous Ligand for Retinoic Acid Receptor." J Biol Chem 283(18): 11947-53. Ischiropoulos, H. (1998). "Biological Tyrosine Nitration: A Pathophysiological Function of Nitric Oxide and Reactive Oxygen Species." Arch Biochem Biophys 356(1): 1-11. Ishibashi, T., T. Murata, et al. (1998). "Advanced Glycation End Products in AgeRelated Macular Degeneration." Arch Ophthalmol 116(12): 1629-32. Ivert, L., H. Keldbye, et al. (2005). "Age-Related Changes in the Basement Membrane of the Retinal Pigment Epithelium of Rpe65 -/- and Wild-Type Mice." Graefes Arch Clin Exp Ophthalmol 243(3): 250-6. Jha, P., P. S. Bora, et al. (2007). "The Role of Complement System in Ocular Diseases Including Uveitis and Macular Degeneration." Mol Immunol 44(16): 3901-8. Johnson, L. V., W. P. Leitner, et al. (2001). "Complement Activation and Inflammatory Processes in Drusen Formation and Age-Related Macular Degeneration." Exp Eye Res 73(6): 887-96. Johnson, L. V., S. Ozaki, et al. (2000). "A Potential Role for Immune Complex Pathogenesis in Drusen Formation." Exp Eye Res 70(4): 441-9. Kalfa, T. A., M. E. Gerritsen, et al. (1995). "Altered Proliferation of Retinal Microvascular Cells on Glycated Matrix." Invest Ophthalmol Vis Sci 36(12): 2358-67. Kanemoto, T., R. Reich, et al. (1990). "Identification of an Amino Acid Sequence from the Laminin a Chain That Stimulates Metastasis and Collagenase Iv Production." Proc Natl Acad Sci U S A 87(6): 2279-83. Karwatowski, W. S., T. E. Jeffries, et al. (1995). "Preparation of Bruch's Membrane and Analysis of the Age-Related Changes in the Structural Collagens." Br J Ophthalmol 79(10): 944-52.

266 Katakura, K., K. Kishida, et al. (2004). "Changes in Rat Lens Proteins and Glutathione Reductase Activity with Advancing Age." Int J Vitam Nutr Res 74(5): 329-33. Katz, M. L., C. M. Drea, et al. (1986). "Influence of Early Photoreceptor Degeneration on Lipofuscin in the Retinal Pigment Epithelium." Exp Eye Res 43(4): 561-73. Kennedy, C. J., P. E. Rakoczy, et al. (1995). "Lipofuscin of the Retinal Pigment Epithelium: A Review." Eye 9 ( Pt 6): 763-71. King, M. W. (2009). "Amino Acid Derivatives." 2009, from http://themedicalbiochemistrypage.org/aminoacidderivatives.html. Klein, R. J., C. Zeiss, et al. (2005). "Complement Factor H Polymorphism in AgeRelated Macular Degeneration." Science 308(5720): 385-9. Kobayashi, S., M. Nomura, et al. (2007). "Overproduction of N(Epsilon)(Carboxymethyl)Lysine-Induced Neovascularization in Cultured Choroidal Explant of Aged Rat." Biol Pharm Bull 30(1): 133-8. Kolb, H., R. Nelson, et al. (2001). "Cellular Organization of the Vertebrate Retina." Prog Brain Res 131: 3-26. Koldunov, V. V., D. S. Kononov, et al. "Oscillations of Photon Emission Accompanying the Oxidative Process in Aqueous Solutions of Glycin with Ribose or Glucose ", 2009, from http://www.photobiology.com/photoiupac2000/koldunov/Image377.gif. Kvech, S. (2000). "ICP-MS." 2009, from http://www.cee.vt.edu/ewr/environmental/teach/smprimer/icpms/horn.gif. Labuza, T. P. and W. M. Baisier (1992). The Kinetics of Nonenzymatic Browning. Physical Chemistry of Foods: 595-689. Lagunowich, L. A. and G. B. Grunwald (1989). "Expression of Calcium-Dependent Cell Adhesion During Ocular Development: A Biochemical, Histochemical and Functional Analysis." Dev Biol 135(1): 158-71. Laine, M., H. Jarva, et al. (2007). "Y402h Polymorphism of Complement Factor H Affects Binding Affinity to C-Reactive Protein." J Immunol 178(6): 3831-6. Lamb, L. E., M. Zareba, et al. (2001). "Retinyl Palmitate and the Blue-LightInduced Phototoxicity of Human Ocular Lipofuscin." Arch Biochem Biophys 393(2): 316-20.

267 Lang, C. A., B. J. Mills, et al. (2000). "Blood Glutathione Decreases in Chronic Diseases." J Lab Clin Med 135(5): 402-5. Lewis, J. E., J. K. Wahl, 3rd, et al. (1997). "Cross-Talk between Adherens Junctions and Desmosomes Depends on Plakoglobin." J Cell Biol 136(4): 919-34. Liang, F.-Q. and F. Godley Bernard (2003). "Oxidative Stress-Induced Mitochondrial DNA Damage in Human Retinal Pigment Epithelial Cells: A Possible Mechanism for RPE Aging and Age-Related Macular Degeneration." Exp Eye Res 76(4): 397-403. Liang, F. Q. and B. F. Godley (2003). "Oxidative Stress-Induced Mitochondrial DNA Damage in Human Retinal Pigment Epithelial Cells: A Possible Mechanism for RPE Aging and Age-Related Macular Degeneration." Exp Eye Res 76(4): 397-403. Ligget, T. (2007). In Vitromodels for the Study of the Mechanisms of Damage in Age-Related Macular Degeneration and Stargardt's Disease. . Biological Sciences. DeKalb, Northern Illinois University. Doctorate of Philosophy. Liu, J., Y. Itagaki, et al. (2000). "The Biosynthesis of A2E, a Fluorophore of Aging Retina, Involves the Formation of the Precursor, A2-PE, in the Photoreceptor Outer Segment Membrane." J Biol Chem 275(38): 29354-60. Lyda, W., N. Eriksen, et al. (1957). "Studies of Bruch's Membrane; Flow and Permeability Studies in a Bruch's Membrane-Choroid Preparation." Am J Ophthalmol 44(5, Part 2): 362-9; discussion 369-70. Maeda, A., T. Maeda, et al. (2008). "Retinopathy in Mice Induced by Disrupted AllTrans-Retinal Clearance." J Biol Chem 283(39): 26684-93. Maher, P. (2005). "The Effects of Stress and Aging on Glutathione Metabolism." Ageing Res Rev 4(2): 288-314. Mandal, S. (2008). Possible Alternative Pathway for A2E Biosynthesis and Compositional Studies on Human Lipofuscin Extract. Chemistry and Biochemistry. DeKalb, Northern Illinois University. Master of Science: 189. Mann, M. and M. Wilm (1994). "Error-Tolerant Identification of Peptides in Sequence Databases by Peptide Sequence Tags." Anal Chem 66(24): 4390-9. Marletta, M. A., P. S. Yoon, et al. (1988). "Macrophage Oxidation of L-Arginine to Nitrite and Nitrate: Nitric Oxide Is an Intermediate." Biochemistry 27(24): 8706-11.

268 Marrs, J. A., C. Andersson-Fisone, et al. (1995). "Plasticity in Epithelial Cell Phenotype: Modulation by Expression of Different Cadherin Cell Adhesion Molecules." J Cell Biol 129(2): 507-19. Marshall, J. (1987). "The Ageing Retina: Physiology or Pathology." Eye 1 ( Pt 2): 282-95. McCarthy, R. D. (2009). "Anatomy of the Eye." Retrieved May 1, 2009, from http://mangastart.com/blog/wp-content/uploads/eye-anatomy.jpg. McKay, B. S., P. E. Irving, et al. (1997). "Cell-Cell Adhesion Molecules and the Development of an Epithelial Phenotype in Cultured Human Retinal Pigment Epithelial Cells." Exp Eye Res 65(5): 661-71. Miki, H., M. B. Bellhorn, et al. (1975). "Specializations of the Retinochoroidal Juncture." Invest Ophthalmol 14(9): 701-7. Miller, D. J. and N. G. MacFarlane (1995). "Intracellular Effects of Free Radicals and Reactive Oxygen Species in Cardiac Muscle." J Hum Hypertens 9(6): 465-73. Mishima, H., Hasebe, H., Kondo, K. (1978). "Age Changes in the Fine Structure of the Human Retinal Pigment Epithelium." Jpn J Ophthalmol 22: 476. Miyagi, M., H. Sakaguchi, et al. (2002). "Evidence That Light Modulates Protein Nitration in Rat Retina." Mol Cell Proteomics 1(4): 293-303. Molavi, D. W. (1997). "Neuroscience Tutorial: Eye and Retina." Retrieved May 1, 2009, from http://thalamus.wustl.edu/course/eyeret.html. Moore, D. J., A. A. Hussain, et al. (1995). "Age-Related Variation in the Hydraulic Conductivity of Bruch's Membrane." Invest Ophthalmol Vis Sci 36(7): 1290-7. Moore, D. J., A. A. Hussain, et al. (1995). "Age-Related Variation in the Hydraulic Conductivity of Bruch's Membrane." Invest Ophthalmol Vis Sci 36(7): 1290-7. Mozaffarieh, M., S. Sacu, et al. (2003). "The Role of the Carotenoids, Lutein and Zeaxanthin, in Protecting against Age-Related Macular Degeneration: A Review Based on Controversial Evidence." Nutr J 2: 20. Mullins, R. F. and G. S. Hageman (1999). "Human Ocular Drusen Possess Novel Core Domains with a Distinct Carbohydrate Composition." J Histochem Cytochem 47(12): 1533-40.

269 Mullins, R. F., S. R. Russell, et al. (2000). "Drusen Associated with Aging and AgeRelated Macular Degeneration Contain Proteins Common to Extracellular Deposits Associated with Atherosclerosis, Elastosis, Amyloidosis, and Dense Deposit Disease." FASEB J. 14(7): 835-846. Mullins, R. F., S. R. Russell, et al. (2000). "Drusen Associated with Aging and AgeRelated Macular Degeneration Contain Proteins Common to Extracellular Deposits Associated with Atherosclerosis, Elastosis, Amyloidosis, and Dense Deposit Disease." Faseb J 14(7): 835-46. Nagai, R., K. Matsumoto, et al. (2000). "Glycolaldehyde, a Reactive Intermediate for Advanced Glycation End Products, Plays an Important Role in the Generation of an Active Ligand for the Macrophage Scavenger Receptor." Diabetes 49(10): 1714-23. Nagaraj, R. H., A. Biswas, et al. (2008). "The Other Side of the Maillard Reaction." Ann N Y Acad Sci 1126: 107-12. Nakaizumi, Y. (1964). "The Ultrastructure of Bruch's Membrane. I. Human, Monkey, Rabbit, Guinea Pig, and Rat Eyes." Arch Ophthalmol 72: 380-7. Nakajou, K., S. Horiuchi, et al. (2005). "Renal Clearance of Glycolaldehyde- and Methylglyoxal-Modified Proteins in Mice Is Mediated by Mesangial Cells through a Class a Scavenger Receptor (Sr-a)." Diabetologia 48(2): 317-27. Naranjo, G. B., L. S. Malec, et al. (1998). Reducing Sugar Effect on Available Lysine Loss of Casein by Moderate Heat Treatment. Food Chem: 309-313. Nelson, W. J., E. M. Shore, et al. (1990). "Identification of a MembraneCytoskeletal Complex Containing the Cell Adhesion Molecule Uvomorulin (E-Cadherin), Ankyrin, and Fodrin in Madin-Darby Canine Kidney Epithelial Cells." J Cell Biol 110(2): 349-57. New Objective. (2004). "Electrospray Ionization." 2009, from http://www.newobjective.com/images/electro/spraytip_bw.jpg. Newsome, D. A., W. Huh, et al. (1987). "Bruch's Membrane Age-Related Changes Vary by Region." Curr Eye Res 6(10): 1211-21. Newsome, D. A., W. Huh, et al. (1987). "Bruch's Membrane Age-Related Changes Vary by Region." Curr Eye Res 6(10): 1211-21. Olver, J. M. (1990). "Functional Anatomy of the Choroidal Circulation: Methyl Methacrylate Casting of Human Choroid." Eye 4 ( Pt 2): 262-72. Ostergaard, H., C. Tachibana, et al. (2004). "Monitoring Disulfide Bond Formation in the Eukaryotic Cytosol." J Cell Biol 166(3): 337-45.

270 Outten, C. E. and V. C. Culotta (2004). "Alternative Start Sites in the Saccharomyces Cerevisiae GLR1 Gene Are Responsible for Mitochondrial and Cytosolic Isoforms of Glutathione Reductase." J Biol Chem 279(9): 7785-91. Oyster, C. W. (1999). The Human Eye : Structure and Function. Sunderland, Mass., Sinauer Associates. Paik, D. C. and J. Dillon (2000). "The Nitrite/Alpha Crystallin Reaction: A Possible Mechanism in Lens Matrix Damage." Exp Eye Res 70(1): 73-80. Paik, D. C., J. Dillon, et al. (2001). "The Nitrite/Collagen Reaction: Non-Enzymatic Nitration as a Model System for Age-Related Damage." Connect Tissue Res 42(2): 111-22. Paik, D. C., W. G. Ramey, et al. (1997). "The Nitrite/Elastin Reaction: Implications for in Vivo Degenerative Effects." Connect Tissue Res 36(3): 241-51. Paik, D. C., L. Y. Saito, et al. (2006). "Nitrite-Induced Cross-Linking Alters Remodeling and Mechanical Properties of Collagenous Engineered Tissues." Connect Tissue Res 47(3): 163-76. Parish, C. A., M. Hashimoto, et al. (1998). "Isolation and One-Step Preparation of A2E and iso-A2E, Fluorophores from Human Retinal Pigment Epithelium." Proc Natl Acad Sci U S A 95(25): 14609-13. Paul, R. G. and A. J. Bailey (1999). "The Effect of Advanced Glycation EndProduct Formation Upon Cell-Matrix Interactions." Int J Biochem Cell Biol 31(6): 653-60. Pauleikhoff, D., M. J. Barondes, et al. (1990). "Drusen as Risk Factors in AgeRelated Macular Disease." Am J Ophthalmol 109(1): 38-43. Pauleikhoff, D., C. A. Harper, et al. (1990). "Aging Changes in Bruch's Membrane. A Histochemical and Morphologic Study." Ophthalmology 97(2): 171-8. Pauleikhoff, D., S. Wojteki, et al. (2000). "[Adhesive Properties of Basal Membranes of Bruch's Membrane. Immunohistochemical Studies of AgeDependent Changes in Adhesive Molecules and Lipid Deposits]." Ophthalmologe 97(4): 243-50. Pepperberg, D. R., T. L. Okajima, et al. (1993). "Interphotoreceptor RetinoidBinding Protein (IRBP). Molecular Biology and Physiological Role in the Visual Cycle of Rhodopsin." Mol Neurobiol 7(1): 61-85.

271 Ragauskaite, L., R. C. Heckathorn, et al. (2001). "Environmental Effects on the Photochemistry of A2-E, a Component of Human Retinal Lipofuscin." Photochem Photobiol 74(3): 483-8. Ragauskaite, L., R. C. Heckathorn, et al. (2001). "Environmental Effects on the Photochemistry of A2-E, a Component of Human Retinal Lipofuscin." Photochem. Photobiol. 74(3): 483-488. Ramrattan, R. S., T. L. van der Schaft, et al. (1994). "Morphometric Analysis of Bruch's Membrane, the Choriocapillaris, and the Choroid in Aging." Invest Ophthalmol Vis Sci 35(6): 2857-64. Rando, R. R. (2001). "The Biochemistry of the Visual Cycle." Chem Rev 101(7): 1881-96. Rebrin, I., S. Zicker, et al. (2005). "Effect of Antioxidant-Enriched Diets on Glutathione Redox Status in Tissue Homogenates and Mitochondria of the Senescence-Accelerated Mouse." Free Radic Biol Med 39(4): 549-57. Reszka, K., C. E. Eldred, et al. (1995). "The Photochemistry of Human Retinal Lipofuscin as Studied by EPR." Photochem. Photobiol. 62(6): 1005-8. Reszka, K., G. E. Eldred, et al. (1995). "The Photochemistry of Human Retinal Lipofuscin as Studied by EPR." Photochem Photobiol 62(6): 1005-8. Robey, P. G. and D. A. Newsome (1983). "Biosynthesis of Proteoglycans Present in Primate Bruch's Membrane." Invest Ophthalmol Vis Sci 24(7): 898-905. Rodrigues, E. B. (2007). "Inflammation in Dry Age-Related Macular Degeneration." Ophthalmologica 221(3): 143-52. Romitelli, F., S. A. Santini, et al. (2007). "Comparison of Nitrite/Nitrate Concentration in Human Plasma and Serum Samples Measured by the Enzymatic Batch Griess Assay, Ion-Pairing HPLC and Ion-Trap GC-MS: The Importance of a Correct Removal of Proteins in the Griess Assay." J Chromatogr B Analyt Technol Biomed Life Sci 851(1-2): 257-67. Rozanowska, M., J. Jarvis-Evans, et al. (1995). "Blue Light-Induced Reactivity of Retinal Age Pigment. In Vitro Generation of Oxygen-Reactive Species." J Biol Chem 270(32): 18825-30. Rozanowska, M., A. Pawlak, et al. (2004). "Age-Related Changes in the Photoreactivity of Retinal Lipofuscin Granules: Role of ChloroformInsoluble Components." Invest Ophthalmol Vis Sci 45(4): 1052-60.

272 Rozanowska, M., J. Wessels, et al. (1998). "Blue Light-Induced Singlet Oxygen Generation by Retinal Lipofuscin in Non-Polar Media." Free Radic Biol Med 24(7-8): 1107-12. Ruberti, J. W., C. A. Curcio, et al. (2003). "Quick-Freeze/Deep-Etch Visualization of Age-Related Lipid Accumulation in Bruch's Membrane." Invest Ophthalmol Vis Sci 44(4): 1753-9. Rudolf, M., M. E. Clark, et al. (2008). "Prevalence and Morphology of Druse Types in the Macula and Periphery of Eyes with Age-Related Maculopathy." Invest Ophthalmol Vis Sci 49(3): 1200-9. Sakai, N., J. Decatur, et al. (1996). "Ocular Age Pigment "A2-E": An Unprecedented Pyridinium Bisretinoid." J Am Chem Soc 118: 1559-1560. Sarici, S. U., F. Alpay, et al. (1999). "Comparison of the Efficacy of Conventional Special Blue Light Phototherapy and Fiberoptic Phototherapy in the Management of Neonatal Hyperbilirubinaemia." Acta Paediatr 88(11): 124953. Sarks, S. H. (1976). "Ageing and Degeneration in the Macular Region: A ClinicoPathological Study." Br J Ophthalmol 60(5): 324-41. Sarks, S. H. (1982). "Drusen Patterns Predisposing to Geographic Atrophy of the Retinal Pigment Epithelium." Aust J Ophthalmol 10(2): 91-7. Sarks, S. H., J. J. Arnold, et al. (1999). "Early Drusen Formation in the Normal and Aging Eye and Their Relation to Age Related Maculopathy: A Clinicopathological Study." Br J Ophthalmol 83(3): 358-68. Sastre, J., J. A. Martin, et al. (2005). "Age-Associated Oxidative Damage Leads to Absence of Gamma-Cystathionase in over 50% of Rat Lenses: Relevance in Cataractogenesis." Free Radic Biol Med 38(5): 575-82. Schafer, F. Q. and G. R. Buettner (2001). "Redox Environment of the Cell as Viewed through the Redox State of the Glutathione Disulfide/Glutathione Couple." Free Radic Biol Med 30(11): 1191-212. Schaumberg, D. A., W. G. Christen, et al. (2007). "High-Sensitivity C-Reactive Protein, Other Markers of Inflammation, and the Incidence of Macular Degeneration in Women." Arch Ophthalmol 125(3): 300-5. Schraermeyer, U. and K. Heimann (1999). "Current Understanding on the Role of Retinal Pigment Epithelium and Its Pigmentation." Pigment Cell Res 12(4): 219-36.

273 Schutt, F., M. Bergmann, et al. (2003). "Proteins Modified by Malondialdehyde, 4Hydroxynonenal, or Advanced Glycation End Products in Lipofuscin of Human Retinal Pigment Epithelium." Invest Ophthalmol Vis Sci 44(8): 3663-8. Sellner, P. A. (1986). "The Movement of Organic Solutes between the Retina and Pigment Epithelium." Exp Eye Res 43(4): 631-9. Sethna, S. S., A. M. Holleschau, et al. (1982). "Activity of Glutathione Synthesis Enzymes in Human Lens Related to Age." Curr Eye Res 2(11): 735-42. Shaban, H., P. Gazzotti, et al. (2001). "Cytochrome C Oxidase Inhibition by NRetinyl-N-Retinylidene Ethanolamine, a Compound Suspected to Cause Age-Related Macula Degeneration." Arch Biochem Biophys 394(1): 111-6. Shen, D., T. P. Dalton, et al. (2005). "Glutathione Redox State Regulates Mitochondrial Reactive Oxygen Production." J Biol Chem 280(27): 2530512. Skerka, C., N. Lauer, et al. (2007). "Defective Complement Control of Factor H (Y402h) and Fhl-1 in Age-Related Macular Degeneration." Mol Immunol 44(13): 3398-406. Solberg, Y., M. Rosner, et al. (1998). "The Association between Cigarette Smoking and Ocular Diseases." Surv Ophthalmol 42(6): 535-47. Sommerburg, O., C. D. Langhans, et al. (2003). "Beta-Carotene Cleavage Products after Oxidation Mediated by Hypochlorous Acid--a Model for NeutrophilDerived Degradation." Free Radic Biol Med 35(11): 1480-90. Sparrow, J. R. and B. Cai (2001). "Blue Light-Induced Apoptosis of A2EContaining RPE: Involvement of Caspase-3 and Protection by Bcl-2." Invest Ophthalmol Vis Sci 42(6): 1356-62. Sparrow, J. R. and B. Cai (2001). "Blue Light-Induced Apoptosis of A2EContaining RPE: Involvement of Caspase-3 and Protection by Bcl-2." Invest Ophthalmol Vis Sci 42(6): 1356-62. Sparrow, J. R., K. Nakanishi, et al. (2000). "The Lipofuscin Fluorophore A2E Mediates Blue Light-Induced Damage to Retinal Pigmented Epithelial Cells." Invest Ophthalmol Vis Sci 41(7): 1981-9. Sparrow, J. R., K. Nakanishi, et al. (2000). "The Lipofuscin Fluorophore A2E Mediates Blue Light-Induced Damage to Retinal Pigmented Epithelial Cells." Invest Ophthalmol Vis Sci 41(7): 1981-9.

274 Sparrow, J. R., C. A. Parish, et al. (1999). "A2E, a Lipofuscin Fluorophore, in Human Retinal Pigmented Epithelial Cells in Culture." Invest Ophthalmol Vis Sci 40(12): 2988-95. Sparrow, J. R., H. R. Vollmer-Snarr, et al. (2003). "A2E-Epoxides Damage DNA in Retinal Pigment Epithelial Cells. Vitamin E and Other Antioxidants Inhibit A2E-Epoxide Formation." J Biol Chem 278(20): 18207-13. Sternberg, P., Jr., P. C. Davidson, et al. (1993). "Protection of Retinal Pigment Epithelium from Oxidative Injury by Glutathione and Precursors." Invest Ophthalmol Vis Sci 34(13): 3661-8. Stitt, A. W. and H. Vlassara (1999). Advanced Glycation End-Products: Impact on Diabetic Complications. Current Perspectives in Diabetes. D. J. Betteridge. New York, Martin Dunitz Inc.: 67-92. Streilein, J. W. (2003). "Ocular Immune Privilege: The Eye Takes a Dim but Practical View of Immunity and Inflammation." J Leukoc Biol 74(2): 179-85. Sun, K., H. Cai, et al. (2007). "Bruch's Membrane Aging Decreases Phagocytosis of Outer Segments by Retinal Pigment Epithelium." Mol Vis 13: 2310-9. Sundelin, S., U. Wihlmark, et al. (1998). "Lipofuscin Accumulation in Cultured Retinal Pigment Epithelial Cells Reduces Their Phagocytic Capacity." Curr Eye Res 17(8): 851-7. Suter, M., C. Reme, et al. (2000). "Age-Related Macular Degeneration. The Lipofusion Component N-Retinyl-N-Retinylidene Ethanolamine Detaches Proapoptotic Proteins from Mitochondria and Induces Apoptosis in Mammalian Retinal Pigment Epithelial Cells." J Biol Chem 275(50): 3962530. Suter, M., C. Reme, et al. (2000). "Age-Related Macular Degeneration: The Lipofuscin Component N-Retinyl-N-Retinylidene Ethanolamine Detaches Proapoptotic Proteins from Mitochondria and Induces Apoptosis in Mammalian Retinal Pigment Epithelial Cells." J. Biol. Chem. 275(50): 39625-39630. Suthanthiran, M., M. E. Anderson, et al. (1990). "Glutathione Regulates ActivationDependent DNA Synthesis in Highly Purified Normal Human T Lymphocytes Stimulated Via the CD2 and CD3 Antigens." Proc Natl Acad Sci U S A 87(9): 3343-7. Swamy-Mruthinti, S., K. C. Miriam, et al. (2002). "Immunolocalization and Quantification of Advanced Glycation End Products in Retinal Neovascular Membranes and Serum: A Possible Role in Ocular Neovascularization." Curr Eye Res 25(3): 139-45.

275 Tarsio, J. F., L. A. Reger, et al. (1988). "Molecular Mechanisms in Basement Membrane Complications of Diabetes. Alterations in Heparin, Laminin, and Type Iv Collagen Association." Diabetes 37(5): 532-9. Tate, D. J., Jr., M. V. Miceli, et al. (1995). "Phagocytosis and H2O2 Induce Catalase and Metallothionein Gene Expression in Human Retinal Pigment Epithelial Cells." Invest Ophthalmol Vis Sci 36(7): 1271-9. Taylor, H. R., B. Munoz, et al. (1990). "Visible Light and Risk of Age-Related Macular Degeneration." Trans Am Ophthalmol Soc 88: 163-73; discussion 173-8. Tezel, T. H., L. V. Del Priore, et al. (2004). "Reengineering of Aged Bruch's Membrane to Enhance Retinal Pigment Epithelium Repopulation." Invest Ophthalmol Vis Sci 45(9): 3337-48. Thorpe, S. R. and J. W. Baynes (2003). "Maillard Reaction Products in Tissue Proteins: New Products and New Perspectives." Amino Acids 25(3-4): 27581. Tsikas, D., F. M. Gutzki, et al. (1997). "Measurement of Nitrite and Nitrate in Biological Fluids by Gas Chromatography-Mass Spectrometry and by the Griess Assay: Problems with the Griess Assay--Solutions by Gas Chromatography-Mass Spectrometry." Anal Biochem 244(2): 208-20. Tsilibary, E. C., A. S. Charonis, et al. (1988). "The Effect of Nonenzymatic Glucosylation on the Binding of the Main Noncollagenous Nc1 Domain to Type Iv Collagen." J Biol Chem 263(9): 4302-8. Ulrich, P. and A. Cerami (2001). "Protein Glycation, Diabetes, and Aging." Recent Prog Horm Res 56: 1-21. Vlassara, H., R. Bucala, et al. (1994). "Pathogenic Effects of Advanced Glycosylation: Biochemical, Biologic, and Clinical Implications for Diabetes and Aging." Lab. Invest. 70(2): 138-51. Wald, G. (1961). "Participation of Rods and Cones in Visual Responses. (Reply to the Comments of C. S. Bridgman)." J Opt Soc Am 51: 241-3. Wang, J., K. Ohno-Matsui, et al. (2008). "Altered Function of Factor I Caused by Amyloid Beta: Implication for Pathogenesis of Age-Related Macular Degeneration from Drusen." J Immunol 181(1): 712-20. Wang, Z. (2005). Biochemical Study of Mechanism of Damage to the Retinal Pigment Epithelium. Chemistry and Biochemistry. DeKalb, Northern Illinois University. Doctorate of Philosophy: 225.

276 Wang, Z., L. M. Keller, et al. (2006). "Oxidation of A2E Results in the Formation of Highly Reactive Aldehydes and Ketones." Photochem Photobiol 82(5): 1251-7. Wang, Z., L. M. M. Keller, et al. (2006). "Oxidation of A2E Results in the Formation of Highly Reactive Aldehydes and Ketones." Photochem. Photobiol. 82(5): 1251-1257. Wang, Z., D. C. Paik, et al. (2005). "Nitrite-Modified Extracellular Matrix Proteins Deleteriously Affect Retinal Pigment Epithelial Cell Function and Viability: A Comparison Study with Nonenzymatic Glycation Mechanisms." Curr Eye Res 30(8): 691-702. Wang, Z., D. C. Paik, et al. (2005). "Nitrite-Modified Extracellular Matrix Proteins Deleteriously Affect Retinal Pigment Epithelial Cell Function and Viability: A Comparison Study with Nonenzymatic Glycation Mechanisms." Curr. Eye Res. 30(8): 691-702. Wassell, J., S. Davies, et al. (1999). "The Photoreactivity of the Retinal Age Pigment Lipofuscin." J Biol Chem 274(34): 23828-32. Weiter, J. J., F. C. Delori, et al. (1986). "Retinal Pigment Epithelial Lipofuscin and Melanin and Choroidal Melanin in Human Eyes." Invest Ophthalmol Vis Sci 27(2): 145-52. Winkler, B. S., M. E. Boulton, et al. (1999). "Oxidative Damage and Age-Related Macular Degeneration." Mol. Vision 5: No pp Given. Winkler, B. S., M. E. Boulton, et al. (1999). "Oxidative Damage and Age-Related Macular Degeneration." Mol Vis 5: 32. Wu, Z., T. W. Lauer, et al. (2007). "Oxidative Stress Modulates Complement Factor H Expression in Retinal Pigmented Epithelial Cells by Acetylation of Foxo3." J Biol Chem 282(31): 22414-25. Yamamoto, T. and H. Yamashita (1989). "Scanning Electron Microscopic Observation of Bruch's Membrane with the Osmium Tetroxide Treatment." Br J Ophthalmol 73(3): 162-7. Yan, S. F., R. Ramasamy, et al. (2003). "Glycation, Inflammation, and RAGE: A Scaffold for the Macrovascular Complications of Diabetes and Beyond." Circ Res 93(12): 1159-69. Yasukawa, T., P. Wiedemann, et al. (2007). "Glycoxidized Particles Mimic Lipofuscin Accumulation in Aging Eyes: A New Age-Related Macular Degeneration Model in Rabbits." Graefes Arch Clin Exp Ophthalmol 245(10): 1475-85.

277 Yeboah, F. K., I. Alli, et al. (1999). "Reactivities of D-Glucose and D-Fructose During Glycation of Bovine Serum Albumin." J Agric Food Chem 47(8): 3164-72. Yeboah, F. K., I. Alli, et al. (2000). "Monitoring Glycation of Lysozyme by Electrospray Ionization Mass Spectrometry." J Agric Food Chem 48(7): 2766-74. Yeh, J. (1977). "Reductive Polymerization of Para-Cyanobenzaldehyde in Dimethyl Sulfoxide Solutions Using the Rotating Ring-Disk Electrode Technique " Electroanal Chem(84). Yeh, K. L., B. Liu, et al. (2004). "Ruthenium-Catalyzed Transformation of 3-Benzyl but-1-Ynyl Ethers into 1,3-Dienes and Benzaldehyde Via Transfer Hydrogen." J Org Chem 69(14): 4692-4. Yin, D. (1996). "Biochemical Basis of Lipofuscin, Ceroid, and Age Pigment-Like Fluorophores." Free Radic Biol Med 21(6): 871-88. Young, R. W. (1971a). "Shedding of Discs from Rod Outer Segments in the Rhesus Monkey." J Ultrastruct Res 34(1): 190-203. Young, R. W. (1971b). "The Renewal of Rod and Cone Outer Segments in the Rhesus Monkey." J Cell Biol 49(2): 303. Zarbin, M. A. (2004). "Current Concepts in the Pathogenesis of Age-Related Macular Degeneration." Arch Ophthalmol 122(4): 598-614. Zhou, J., B. Cai, et al. (2005). "Mechanisms for the Induction of HNE- MDA- and AGE-Adducts, RAGE and VEGF in Retinal Pigment Epithelial Cells." Exp Eye Res 80(4): 567-80.