A Lc-Ms/Ms Approach for Gangliosides Profiling in Brain And

A LC-MS/MS APPROACH FOR GANGLIOSIDES PROFILING IN BRAIN AND

RETINAL TISSUE OF MICE: APPLICATION TO GLAUCOMA MICE AGE

STUDIES

ASHTA LAKSHMI PRASAD GOBBURI

Bachelor in Pharmacy

Jawaharlal Nehru Technological University

Hyderabad, India, May 2011

Submitted in partial fulfillment of the requirements for the degree

DOCTOR OF PHILOSOPHY IN CLINICAL-BIOANALYTICAL CHEMISTRY

CLEVELAND STATE UNIVERSITY

December 2017

I dedicate this dissertation to:

Gobburi Family K. Kusuma Kumari David J Anderson

Thank you very much for your continuous support and understanding!

ACKNOWLEDGEMENTS

I am glad to take this opportunity to show my appreciation and love to the people in my life who have made this dissertation possible. First and foremost, I am extremely grateful to my Guru, research advisor, Dr. David J Anderson, who introduced me to research and walked me through high and low with his continuous guidance. It is hard to imagine pursuing my dream without his constant support. I would like to heartfully thank him for encouraging my research and for allowing me to grow as a researcher, and above all as a better human being.

I would like to specially thank our research collaborator, Dr. Denise Inman,

Assistant Professor at NorthEast Ohio Medical University (NEOMED). Her research ideas and inputs had a profound impact in shaping the project along with her immense support scientifically and financially for the project.

I would like to thank my dissertation committee members – Dr. Xue-Sun Long,

Dr. Aimin Zhou, Dr. Xiang Zhou and Dr. Nolan B. Holland for their consistent support throughout my research. I am honored to have a such a great committee who accommodated their valuable time whenever it is required. I owe to the committee for their valuable suggestions which helped in the progress of my research.

I take this opportunity to thank my high school class teacher, K. Kusuma Kumari, who remained my greatest support and inspiration throughout my life. I would like to specially thank Tutoring and Success Center (TASC) ex-director, Christine Vodicka, for her unconditional support for my interest in teaching which helped me gain a lot of confidence to pursue my interests in doctoral program. I thank Michael Kalafatis, Interim Chair of Chemistry Department, for his effort in continuous financial support. I also gratefully acknowledge Office of Graduate Studies and Research at CSU for financially supporting my research. I thank all the staff of Chemistry Department at Cleveland State

University who were part of my PhD journey from the past six years for their support.

I am thankful to my high school (All Saints High School, Hyderabad India), undergraduate (SLC’S College of Pharmacy, Hyderabad India) and graduate (Cleveland

State University, Cleveland USA) friends who were always supportive of my interests and goals. I am thankful to them for giving so many memories to relish.

I would like to take this moment to thank my parents, Gobburi Satyanarayana and Gobburi Padmavathi, for their constant love and encouragement. My brothers,

Gobburi Naga Prakash and Gobburi Narendra Dev, who believed in my decisions and pursuits. I am thankful to my niece, Gnana Siri, who brought a lot of love and happiness to our family from the past one year. And most of all for my loving, supporting and encouraging fiancée, Raziya Shaik, who is the main reason for my interest towards research. Her invaluable love and support has always been my strength. Her patience and sacrifice will be an inspiration throughout my life.

I gratefully acknowledge each one of you to allow me carryout my research successfully. I dedicate this dissertation to each one of you.

A LC-MS/MS APPROACH FOR GANGLIOSIDES PROFILING IN BRAIN AND

RETINAL TISSUE OF MICE: APPLICATION TO GLAUCOMA MICE AGE

STUDIES

ASHTA LAKSHMI PRASAD GOBBURI

ABSTRACT

Gangliosides are membrane lipids with a complex sugar head containing sialic acid (polar) and a ceramide moiety made up of a sphingosine and a fatty acid (non-polar).

Typically, gangliosides are part of the ganglion, a group of nerve cell bodies where they act as neuroprotective agents, support, and maintenance of the mature neuronal cells.

Glaucoma is an age-related neurodegenerative disorder of the eye that leads to blindness.

Retinal ganglion cells (RGCs) in retinal tissue of the eye, degeneration is seen followed by the optic nerve head damage. From immunohistochemistry studies with cholera toxin-

B, gangliosides which are part of the RGCs were lost during the RGCs degeneration.

Non-neuronal cells, astrocytes, which do not express gangliosides in normal retinal tissue surprisingly express/uptake gangliosides in correlation to the RGCs degeneration.

Astrocyte expression/uptake of gangliosides can play protective, adverse or combined effects at different times. An identification and quantitation of individual gangliosides

vi that are expressed/uptake provide an insight into understanding the pathophysiology of glaucoma.

A specific, sensitive and quantitative liquid-chromatography-mass spectrometry

(LC-MS/MS) analytical techniques were developed to identify heterogeneous molecular species of gangliosides. Of all of them, new LC-MS/MS method using a phenyl-hexyl reverse-phase chromatography was successful in resolving eight major gangliosides and ten minor gangliosides. An optimized ganglioside isolation technique was developed which is a combination of liquid-liquid extraction for lipid phase separation followed by a solid-phase extraction for desalting and removal of debris. This new LC-MS/MS method was validated for linearity, repeatability, matrix effect and recovery studies.

Age studies (young vs old mice) were performed for the first time to correlate the changes of gangliosides in retinal tissue level in comparison to the superior colliculus of the mid-brain using the new LC-MS/MS method in normal control mice (C57BL/6J) versus glaucoma model mice (DBA/2J). LC-MS/MS results show an increase of major gangliosides in the glaucoma retinal tissue from old to young age mice groups whereas little change in the superior colliculus. Nevertheless, these gangliosides changes were statistically insignificant as p-value is > 0.05. The optic nerve studies using electron microscopy was undertaken to establish the glaucoma disease progression. Electron microscopy study results show no change in the optic nerve count which represents no glaucoma in old age animal samples. Thus, a non-correlation was seen with the ganglioside changes compared to optic nerve counts. A greater animal sample with much older age provides a better understanding of gangliosides role in glaucoma.

vii

TABLE OF CONTENTS

ABSTRACT ...... vi

LIST OF TABLES ...... xiv

LIST OF FIGURES ...... xvi

ABBREVIATIONS ...... xxiii

CHAPTER Ⅰ ...... 1

1.1 General Introduction ...... 1

1.2 Lipids classification, structures and functions ...... 2

1.2.1 Fatty acyls ...... 5

1.2.2 Glycerolipids ...... 5

1.2.3 Glycerophospholipids...... 7

1.2.4 Sterol lipids ...... 7

1.2.5 Prenol lipids...... 9

1.2.6 Saccharolipids ...... 9

1.2.7 Polyketides ...... 11

1.2.8 Sphingolipids ...... 11

1.3 Gangliosides ...... 14

1.3.1 History ...... 14

1.3.2 Ganglioside – Structure ...... 14

1.3.2.1 Gangliosides – Carbohydrate complex[7] ...... 16

viii

1.3.2.2 Gangliosides – Ceramide group[7], [9] ...... 18

1.3.3 Gangliosides – Biosynthesis [10] ...... 21

1.3.4 Gangliosides – Position[11] ...... 23

1.3.5 Gangliosides – Biological functions ...... 25

1.3.6 Gangliosides – Diseases ...... 28

1.3.6.1 Genetic diseases ...... 28

1.3.6.2 Neurological diseases ...... 30

1.4 Glaucoma – A neurodegenerative eye disorder ...... 34

1.4.1 Glaucoma – Types and conditions ...... 34

1.4.2 Primary open-angle glaucoma (POAG) ...... 37

1.4.2.1 Glaucoma – Pathophysiology ...... 38

1.4.2.2 Retinal gliosis: Role of astrocytes ...... 40

1.4.2.3 Role of GM1 ganglioside in glaucoma – Preliminary experiments ...... 40

1.4.3 Research strategy...... 45

CHAPTER Ⅱ ...... 47

2.1 Introduction ...... 47

2.1.1 Ganglioside analysis – Immunological assays ...... 48

2.1.2 Ganglioside analysis by chromatography ...... 49

2.1.2.1 Thin layer chromatography (TLC) ...... 49

2.1.2.2 High performance thin layer chromatography (HPTLC) ...... 50

2.1.2.3 Liquid Chromatography ...... 51

2.1.3 Ganglioside analysis by mass spectrometry ...... 53

2.2 Experimental Section ...... 58

2.2.1 Materials ...... 58

2.2.2. Preparation of standard solutions ...... 59

2.2.3 HPLC Instruments, Columns and Experimental Parameters ...... 59

2.2.3.1 Optimized mixed-mode HPLC runs ...... 59

2.2.3.2 Optimized HILIC HPLC runs...... 60

2.2.3.3 Optimized phenyl-hexyl HPLC runs ...... 60

2.2.4 Mass spectrometer instruments and settings ...... 62

2.2.4.1 Absciex QTrap mass spectrometer ...... 62

2.2.4.2 Waters triple quadrupole mass spectrometer ...... 63

2.2.4.3 Shimadzu triple quadrupole mass spectrometer ...... 63

2.3 Results and discussion ...... 66

2.3.1 LC-MS/MS of gangliosides using mixed-mode column...... 66

2.3.2. LC-MS/MS of gangliosides using HILIC HPLC column ...... 68

2.3.3. LC-MS/MS determination of gangliosides using phenyl-hexyl column ...... 74

2.3.3.1 Optimization of chromatography using phenyl-hexyl column ...... 74

2.3.3.2 Phenyl-hexyl separation of ganglioside classes with fine separation of

hydrophobic subclasses ...... 79

2.3.3.3 Phenyl-hexyl column calibration curves, limit of quantification and limit of

detection...... 83

CHAPTER III ...... 86

3.1 Background ...... 86

3.2 Experimental ...... 87

3.2.1 Materials ...... 87

3.2.2 Schnaar ganglioside isolation procedure ...... 88

3.2.3 Modified Svennerholm and Fredman ganglioside isolation procedure ...... 89

3.2.4 LC-MS/MS Instrumentation and Parameters ...... 90

3.2.5 Recovery Studies ...... 91

3.2.6 Matrix Interference effect...... 91

3.2.7 Determination of gangliosides concentration in mouse brain and retinal samples

...... 91

3.3 Results and Discussion ...... 92

3.3.1 Schnaar isolation recovery studies ...... 92

3.3.2 Modified Svennerholm and Fredman ganglioside isolation procedure recovery

and matrix effect studies ...... 93

3.3.3 Application to brain and retinal tissues ...... 96

CHAPTER IV ...... 106

4.1 Background: Glaucoma and control mice ...... 106

4.1.2. Preliminary data ...... 107

4.1.3 Glaucoma age studies ...... 108

4.1.4 Glaucoma and control mice...... 109

4.2 Experimental section ...... 110

4.2.1 Optic nerve embedding for optical microscopy ...... 111

4.2.2 LC-MS/MS determination of gangliosides of mouse retinal and superior

colliculus samples ...... 113

4.3 Results and Discussion ...... 113

4.3.1 LC-MS/MS analysis on DBA/2J and C57BL/6J mice strains ...... 113

4.3.2 Optic nerve data ...... 131

4.4 Conclusion ...... 135

CHAPTER V ...... 137

5.1 Importance of analyte purity for quantification ...... 137

5.2 Materials and methods ...... 141

5.2.1 Materials ...... 141

5.2.2 Preparation of standard solutions ...... 141

5.2.3 HPLC conditions ...... 141

5.2.4 Mass spectrometry conditions ...... 142

5.2.5 Fatty acid analysis ...... 143

5.3 Results and Discussion ...... 145

xii

5.3.1 Method 1: Fatty acid and 290 m/z daughter ion MRM data ...... 145

5.3.2. Method 2: Fatty acid daughter ion MRM data ...... 150

5.3.3 Method 1 results ...... 153

5.3.4 Comparison of Method 1 with Method 2 results...... 153

5.4 Effect of sphinganine presence in GM1 standard ...... 156

5.5 Conclusions ...... 158

CHAPTER Ⅵ ...... 159

6.1 Summary ...... 159

6.2 Future directions ...... 162

6.2.1 Ganglioside distribution in glaucoma retinal tissue ...... 162

6.2.1.1 Expected results ...... 162

6.2.2 Ganglioside distribution at cellular level ...... 163

6.2.2.1 Methodology ...... 163

6.2.2.2 Expected results ...... 164

6.2.3 Studies of the gangliosides and Biosynthesis and metabolic pathways ...... 164

References ...... 168

Appendix………………………………………………………………………………..184

xiii

LIST OF TABLES

CHAPTER I

1.1 Lipid sub-categories and examples……………………………………….………...4

1.2 Recommended abbreviations for some monosaccharides, derivatives and related

compounds………………………………………………………………………...17

1.3 Abbreviations of gangliosides using the Svennerholm system…………………….20

CHAPTER II

2.1A Time gradient on Phenomenex HILIC HPLC column using the following mobile

phases - A% - 83% ACN; B% - 83% ACN + 5mM Ammonium Acetate; C% -

50% ACN + 50mM Ammonium Acetate………………………………..……....61

2.1B Time gradient on Imtakt HILIC HPLC column using the following mobile phases

- A% - 83% ACN; B% - 83% ACN + 5mM Ammonium Acetate; C% - 50% ACN

+ 50mM Ammonium Acetate; D% - 83%ACN………………………………….61

2.2 Individual ganglioside mass spectrometer parameters………….……………….65

CHAPTER III

3.1 Repeatability of all major gangliosides are shown based on their retention

times………………………………………………………………………….....100

3.2 Ganglioside distribution in adult mice brain (white and grey matter) and retinal

tissues (ng/mg)…………...……………………………………………..………101

CHAPTER Ⅳ

xiv

4.1 Concentrations of individual major gangliosides of retinal tissues grouped into

young (2 and 3 months) and old age (10 and 12 months) mice………………...118

4.2 Concentrations of individual major gangliosides of superior colliculus grouped

into young (2 and 3 months) and old age (10 and 12 months) mice ...…………122

CHAPTER V

5.1 Molecular Ganglioside Content of the Matreya Bovine Brain GM1

Standard………………………………………………………………………...156

LIST OF FIGURES

CHAPTER I

1.1 Example for fatty acids - Hexadecanoic acid…………………………………..…6

1.2 Example for glycerolipids - 1-hexadecanoyl-2-(9Z-octadecanoyl)-sn-glycerol…..6

1.3 Example of glycerophospholipids - 1-hexadecanoyl-2-(9Z-octadecenoyl)-sn-

glycero-3-phosphocholine…………………………………………………………8

1.4 Example of sterol lipids - cholest-5-en-3β-ol……………………………………..8

1.5 Example of prenol lipids - 2E,6E-farsenol...... 10

1.6 Example of saccharolipids - UDP-3-O-(3R-hydroxy-tetradecanoyl)-αD-N-

acetylglucosamine………………………………………………………………..10

1.7 Example of polyketides - aflatoxin B1…………………………………………..13

1.8 Example of sphingolipids - N-(tetradecanoyl)-sphing-4-enine…………………..13

1.9 A typical example of a ganglioside, GM1 monosialo ganglioside……...……….15

1.10 Ganglioside biosynthetic pathway……………………………………………….22

1.11 Position of gangliosides within the lipid rafts of cell membrane………………...24

1.12 Growth factor modulation by gangliosides………………………………………26

1.13 Genetic diseases causes accumulation of various gangliosides………………….29

1.14 Proposed scheme for the ganglioside-bound Ab (GAb) hypothesis. Soluble

amyloid b-protein (Ab) binds to the GM1 ganglioside cluster on the neuronal

membrane………………………………………………………………………...31

xvi

1.15 Circulation of aqueous humor: The anterior segment of the eye shows the

direction of aqueous humor secreted by the ciliary body and exits the trabecular

mesh network…………………...………………………………………………..36

1.16 Glaucoma pathology: A: Aqueous humor movement in normal eye; B: Normal

optic nerve head with no optic disc, healthy lamina cirbrosa and retinal ganglion

cells; C: Optic disc formed, retinal ganglion cells damage and posterior

displacement of lamina cirbrosa………………………………………………..39

1.17 Retina from 12-month-old DBA/2J showing the distribution of heavy chain

neurofilaments (red), (arrows) and degenerating, CTB-negative RGCs filled with

retrogradely transported FluoroGold (blue) and intraocularly injected CTB

(green). CTB labeled RGCs (left of star) exist among CTB-positive astrocyte cell

bodies phosphorylated neurofilaments (arrowheads). Scale bar=50µm…………42

1.18 Fixed retinas from 12-mo C57Bl/6 (left) or DBA/2J (right) mice incubated with

CTB conjugated to AlexaFluor-488; RGC somas (arrow) and axons 5arrowheads)

are CTB-positive. In the aged DBA/2J, most CTB was taken up by astrocytes.

Scale bar=50µm………………………………………………………………….43

CHAPTER II

2.1 Diagram of ESI ion source………………………………………………………56

2.2 Diagram of a triple quadrupole mass analyzer…………………………………..56

2.3 Ganglioside mixture separation on SCHERZO SM-C18 column mixed mode

column (IMTAKT). A gradient of methanol from 10% methanol in water to 90%

xvii

methanol in water for 7 minutes followed by 90% methanol in water for 2

minutes. Followed by equilibration for 10 minutes is performed…..……………67

2.4 Total ion chromatogram of gangliosides standard mixture (GM, GD, GT)

separation on Phenomenex HILIC HPLC column. The gradient of multiple

mobile phases from 6.00 min to 24 min is reported ……………..……………...69

2.5 GM1 ganglioside retention time shift on HILIC column (Phenomenex HPLC

column) from one run to another prior to the optimization of re-equilibration step

after the run………………………………………………………………………71

2.6 Total ion chromatogram of gangliosides standard mixture (GM, GD, GT)

separation on Imtakt HILIC HPLC column. The gradient of multiple mobile

phases from 6.00 min to 24 min is reported……………………………………..73

2.7 Phenyl-hexyl separation of gangliosides with methanol gradient, 45% methanol in

water to 95% methanol in water for 7 minutes. GM1 (d18:1-18:0 in blue and

d18:1-20:0 in red) were eluted between 7.5 – 9.0 min. GD1 (d18:1-18:0 in green

and d18:1-20:0 in grey) were eluted between 4.5 min to 10 min, GT1 (d18:1-18:0

in sky blue and d18:1-20:0 in pink) were eluted between 2.50 min to 9.0

min……………………………………………………………………………...75

2.8 Phenyl-Hexyl separation of gangliosides with methanol gradient with 5mM

ammonium hydroxide in 45%methanol to 5mM ammonium hydroxide in 95%

methanol for 7 minutes. GM1 (d18:1-18:0 and d18:1-20:0 in black), GD1 (d18:1-

18:0 and d18:1-20:0 in red) and GT1 (d18:1-18:0 and d18:1-20:0 in green)……77

xviii

2.9 Phenyl-Hexyl separation of gangliosides with methanol gradient, 10mM additive

ammonium hydroxide in 45%methanol to 10mM ammonium hydroxide in 95%

methanol for 7 minutes. GM1 (d18:1-18:0 and d18:1-20:0 in black), GD1 (d18:1-

18:0 and d18:1-20:0 in red) and GT1 (d18:1-18:0 and d18:1-20:0 in green)…..78

2.10 Phenyl-hexyl separation of 1000 ng/mL mixed gangliosides (GM, GD, GT and

GQ) standards with a methanol gradient of 0.1% ammonium hydroxide in 25%

methanol to 85% 0.1% ammonium hydroxide in methanol. The mobile phase pH

is 10.71. GM1 (d18:1-18:0 and d18:0-20:0 in red), GD1 (d18:1-18:0 and d18:0-

20:0 in yellow), GT1 (d18:1-18:0 and d18:0-20:0 in purple) and GQ (d18:1-18:0

and d18:0-20:0 in blue)………………………………………………....………..80

2.11 Calibration standard curve of GM1 ganglioside, 1545.80……………………….84

CHAPTER III

3.1 Percentage recovery of deuterated monosialo gangliosides GM1-D3 (d18:1-18:0

m/z 1550 in blue, d18:1-20:0 in orange) evaluating the Svennerholm and Fredman

procedure…………………………………………………………………….…...93

3.2 Matrix effect assessment by comparing the responses of GM1 d18:1-18:0 (A) and

GM1 d18:1-20:0/d20:1-18:0 (B) standards (solid circle) vs diluted pooled retinal

tissue samples (triangle)………………………………………………………….96

3.3 Matrix effect assessment by comparing the responses of GD1 d18:1-18:0 (A) and

GD1 d18:1-20:0/d20:1-18:0 (B) standards (solid circle) vs diluted pooled retinal

tissue samples (triangle)……………………………………………………….....97

xix

3.4 Matrix effect assessment by comparing the responses of GT1 d18:1-18:0 (A) and

GT1 d18:1-20:0/d20:1-18:0 (B) standards (solid circle) vs diluted pooled retinal

tissue samples (triangle)……………………………………………………….....98

3.5 Matrix effect assessment by comparing the responses of GQ1 d18:1-18:0 (A) and

GQ1 d18:1-20:0/d20:1-18:0 (B) standards (solid circle) vs diluted pooled retinal

tissue samples (triangle)……………………………………………………….....99

3.6 MRM chromatograms of total gangliosides in A: Brain white matter (10X), B:

Pooled retinal tissue (10X), C: Brain grey matter (10X)……………………….102

3.7 Ganglioside distribution in adult mice brain (white and grey matter) and retinal

tissues (ng/mg)………………………………………………………….………103

CHAPTER IV

4.1 Chromatograms of ganglioside profiling in glaucoma age studies, D2 glaucoma

mice 2 months: From top to bottom: Left eye; Right Eye; Left Superior colliculus;

Right Superior colliculus……………………………………………………….114

4.2 Chromatograms of ganglioside profiling in glaucoma age studies, D2 glaucoma

mice 12 months: From top to bottom: Left eye; Right Eye; Left Superior

colliculus; Right Superior colliculus……………………………………….……115

4.3 Chromatograms of ganglioside profiling in glaucoma age studies, D2G control

mice 2 months: From top to bottom: Left eye; Right Eye; Left Superior colliculus;

Right Superior colliculus……………...…………………………………..……116

4.4 Chromatograms of ganglioside profiling in glaucoma age studies, D2 glaucoma

mice 2 months: From top to bottom: Left eye; Right Eye; Left Superior colliculus;

Right Superior colliculus………………………………………………………117

4.5 Glaucoma (DBA/2J) vs Control (C57BL/6J) superior colliculus ganglioside

concentrations (ng/mg) in young adult mice………..…………………….……126

4.6 Glaucoma (DBA/2J) vs Control (C57BL/6J) superior colliculus ganglioside

concentrations (ng/mg) in old adult mice ...…………………………………....127

4.7 Glaucoma (DBA/2J) vs Control (C57BL/6J) retinal tissue ganglioside

concentrations (ng/mg) in young adult mice…………………………………...128

4.8 Glaucoma (DBA/2J) vs Control (C57BL/6J) retinal tissue ganglioside concentrations (ng/mg) in old adult mice………………………………………………129

4.9 Optical microscopy image of optic nerve of D2 glaucoma mice strain, 2 months

A: right eye, B: left eye, D2G glaucoma control mice C: left eye, D: right

eye…………………………………………………………………………....…131

4.10 Optical microscopy image of optic nerve of D2 glaucoma mice strain, 10 months

A: right eye, B: left eye, D2G glaucoma control mice C: left eye, D: right

eye………………………………………………………………………………132

4.11 Optical microscopy image of optic nerve of D2 glaucoma mice strain, 10 months

A: right eye, B: left eye, D2G glaucoma control mice C: left eye, D: right

eye……...……………………………………………………………………….133

CHAPTER V

xxi

5.1 GM1 ganglioside (m/z 1545) fragmentation pattern, the common daughter ion

(m/z 290) … …………………………………………………………………....143

5.2 Negative ion MRM chromatograms of 20 μL of 10 μg/mL GM1 standard (A) m/z

1545 → 290 and (B) m/z 1573 → 290………………………………………....146

5.3 Negative ion MRM chromatograms of 20 μL of 10 μg/mL GM1 standard (A) m/z

1545 → 283, [B] m/z 1573 → 283 and [C] m/z 1573 → 311……………...….151

5.4 Ratio of the percent of the two prominent GM1 gangliosides C18:0FA-

C18S/C18:0FA-C20S determined by Method 1 (triangles, dotted linear regression

line) and by Method 2 (circles, solid linear regression line) versus concentration

of GM1 standard injected………………………………………………………154

CHAPTER VI

6.1 Ganglioside biosynthetic pathway……………………………………………...165

xxii

ABBREVIATIONS

APP Amyloid precursor protein

AICD APP Intracellular domain

BBB Blood brain barrier

BDNF Brain- derived neurotrophic factor

CTB Cholera toxin B

CE Capillary Electrophoresis

ESI Electrospray ionization

FAB Fast atom bombardment

FID Flame ionization detection

FTICR Fourier transform ion cyclotron resonance

GM monosialo gangliosides

GBS Guillain-Barré syndrome

GC Gas chromatography

GD disialo gangliosides

GT trisialo gangliosides

GQ tetrasialo gangliosides

HCT High capacity ion trap

xxiii

HILIC Hydrophilic interaction chromatography

HPTLC High performance thin layer chromatography

IOP Intraocular pressure

MALDI Matrix assisted laser desorption ionization

MAG Myelin-associated Glycoprotein

NGF Nerve growth factor

NMJ Neuromuscular junction

NPC Neural progenitor cells

NSC Neural stem cells

ONH Optic Nerve Head

POAG Primary open angle glaucoma

PD Parkinson’s disease

PTFE Polytetrafluoro ethylene

Q-TOF Quadrupole time-of-flight

LC Liquid chromatography

MS Mass spectrometry

MAG Myelin associated glycoproteins

RGCs Retinal ganglion cells

xxiv

RNFL Retinal nerve fibre loss

SOAG Secondary open-angle glaucoma

SACG Secondary angle-closure glaucoma

SPE Solid phase extraction

TrkA Tyrosine kinase A

xxv

CHAPTER Ⅰ

INTRODUCTION

1.1 General Introduction

All living organisms are made up of one or more cells. The building blocks of these cells are nucleic acids, carbohydrates, proteins, and lipids. Lipids are primarily composed of carbon, hydrogen, and oxygen but also nitrogen, phosphorus and sulfur are seen. All lipids are classically defined as the biomolecules readily soluble in organic solvents such as hexane owing to their hydrophobicity.

Biological lipids are structurally and functionally diverse group of molecules.

Lipids are known as the energy repositories in plants and animals which serve as

1 metabolic fuels. The energy stored in lipids is supplied for cellular function through oxidation and facilitate metabolic flexibility. Lipids regulate mitochondrial electron transport chain flux and coupling efficiency (e.g., cardiolipin, fatty acids)[1]. On the fundamental level, lipids are the main components of biological membranes

(phospholipids) where they function to organize and distribute the molecular entities necessary for all cell processes. Lipid membranes promote cellular signaling to facilitate the transmission of biological information across cell membranes. Lipids serve as secondary messengers of signal transduction (e.g., eicosanoids, endocannabinoids) in biologic membranes which are activated by hydrolysis. Lipids interact with surface and internal membrane proteins and play a key role in their activation, modification and distribution depending on the cellular functions. As such, complete identification of the chemical diversity of lipids provides information about the specific functions of lipids and their importance.

In this chapter, three sub-chapters are included where the first sub-chapter describes the general classification of lipids, their chemical and functional importance.

The second sub-chapter discusses the sub-class of sphingolipids, gangliosides, their structure, position and functional role. The third sub-chapter introduces the role of gangliosides in the disease, glaucoma, along with the preliminary experiments.

1.2 Lipids classification, structures and functions

In addition to the overall classification of lipid as a hydrophobic or amphipathic molecule, lipids are sub-classified as simple and complex lipids. Simple lipids yield at most two byproducts upon hydrolysis (e.g., fatty acids, sterols, and acylglycerols) while the complex lipids yield three or more byproducts on hydrolysis (e.g., glycosphingolipids

2 and gycerophospholipids). A lipid classification based on the functions such as storage, cell signaling and growth regulation etc. is also seen. Examples of different classification of lipids based on their chemically functional backbone are given in Table 1.1.

Table 1.1: Lipid sub-categories and examples [2]

Category Abbreviation Example

Fatty acyls FA dodecanoic acid

1-hexdecanoyl-2-(9-Z-octadecenoyl)-sn- Glycerolipids GL glycerol

1-hexdecanoyl-2-(9-Z-octadecenoyl)-sn- Glycerophospholipids GP glycero-3-phosphocholine

Sphingolipids SP N-(tetradecanoyl)-sphing-4-enine

Sterol lipids ST cholest-5-en-3β-ol

Prenol lipids PR 2E,6E-farsenol

UDP-3-O-(3R-hydroxy-tetradecanoyl)- Saccharolipids SL αD-N-acetylglucosamine

Polyketides PK aflatoxin B1

Based on the lipid backbone, a total of eight primary sub-categories of lipids are presented[2].

1.2.1 Fatty acyls

Fatty acyls (FAs) are chemically made up of repetitive methylene groups which imparts the hydrophobicity. The first sub-class of fatty acyls are fatty acids, which are made up of long carbon chains with a carboxylic acid in it (Figure 1.1). Fatty acids include a broad range of carbon chains: straight-chain, branched, saturated and unsaturated (double and triple bonded) fatty acids. Other functional group substituents such as methyl, -oxo-, aldehyde and amino are also seen. Mycolic acids are a classic example of a highly complex branched-chain fatty acids. Hetero atoms such as oxygen, nitrogen, halogen and sulfur occur in conjunction with the carbon chains exists as specific fatty acid sub-classes. Eicosanoids are the enzymatic and non-enzymatic derivatives of arachidonic acid that includes prostaglandins, leukotrienes, thromboxanes and others play an important role in cell signaling, immune response and inflammation responses. Fatty acid esters are esters of fatty acid and alcohol includes wax monoesters, lactones and others which are fundamental energy repositories and biochemical intermediates.

Hydrocarbons are included in the fatty acyl class as they are the reduction products of fatty acids.

1.2.2 Glycerolipids

Glycerolipids are a class of lipids with glycerol as the backbone with various substitutions. The first sub-class are the fatty acids substitutions on the alcohols forming

Figure 1.1: Example of Fatty acids: Hexadecanoic acid [2]

Figure 1.2: Example for glycerolipids - 1-hexadecanoyl-2-(9Z-octadecanoyl)-sn- glycerol [2]

6 fatty acid esters, acylglycerols (Figure 1.2). The other sub-class includes glycan substitutions such as diacylglycerol glycans. The sugar residues are attached by a glycosidic linkage to the glycerol. Macrocyclic ether lipids are a distinctive class of glycerolipids are seen in archaebacteria. These ether lipids play a significant role in the signal transduction in these microorganisms.

1.2.3 Glycerophospholipids

Glycerophospholipids are the predominant form of lipids in the lipid bilayer cells.

They exist in many distinctive forms based on the polar head group types.

Glycerophosphocholines, glycerophosphoethanolamines, glycerophosphoserines and glycerophosphoinositols are the common glycerophospholipids which are not only part of biological membranes but play a crucial role in signal transduction, cellular protein binding and metabolic pathways. Each head group is further diversified based on the fatty acid substitutions on sn-1 and sn-2 carbons of glycerol (Figure 1.3). Ether linked fatty acids such as plasmalogens are enriched in vertebrate heart tissues. A separate class, called oxidized glycerophospholipids, where the side chains are oxidized from oxidative stress.

1.2.4 Sterol lipids

Sterols are fused ring structures along with having a long carbon chains attached to it. Sterols are classified based on the carbon number in the core skeleton of the fused four ring structures (Figure 1.4). Estrogen and testosterone families are made up of C18 and C19 steroids. Secosteroids are characterized by the open B ring structure, the main

Figure 1.3: Example of glycerophospholipids - 1-hexadecanoyl-2-(9Z-octadecenoyl)- sn-glycero-3-phosphocholine [2]

Figure 1.4: Example of sterol lipids - cholest-5-en-3β-ol [2]

8 components are vitamin D and its derivatives. Bile acids are the primary derivatives of cholesterol derivative, cholan-24-oic acid. Sterols functional role is as hormones, signaling molecules and electron carriers. Sterols have a significant role in the metabolism or physiology of animals.

1.2.5 Prenol lipids

Isopentenyl diphosphate and dimethylallyl diphosphate are the five carbon precursor molecules which are used in the synthesis of prenols (Figure 1.5). Prenols are the precursors of the mevalonic acid pathway present in eukaryotes, archaea, and some bacteria. The methylerythritol phosphate pathway is made up of isoprenoid precursors in a few bacteria and plants. Vitamin A and its derivatives are made up of C20 isoprenoids.

Carotenoids are simple isoprenoids, which function as antioxidants and as precursors of vitamin A. Quinones and hydroquinone’s which are attached to the chains of isoprenoids, are another important class of prenol lipids. The important quinones are ubiquinone, vitamins E and K. Polyprenols are another important class of prenol lipids where multiple isoprenols are linked together. Polyprenols play an important role in the transport of oligosaccharides across membranes.

1.2.6 Saccharolipids

Saccharolipids are the lipid class where glycerols are substituted with glycans in glycerolipids and glycerophospholipids. Saccharolipids can occur as glycan or as phosphorylated derivatives. Acylated forms of glucose and sucrose are also seen.

Saccharolipids are classified as mono-, di-, tri- and so on based on the number of acyl.

Figure 1.5: Example of prenol lipids - 2E,6E-farsenol [2]

Figure 1.6: Example of saccharolipids - UDP-3-O-(3R-hydroxy-tetradecanoyl)-αD-N- acetylglucosamine [2]

10 linkages. Fatty acyls are linked either as amides or esters. A classic example of saccharolipids are acylated glucosamine sugars, the precursors for lipid A in the cell membrane of gram-negative bacteria (Figure 1.6).

1.2.7 Polyketides

Polyketides are the lipid class which are usually biosyntesized by classic and multimodular enzymes. The two main classes of polyketides are macrolide and aromatic polyketides. Macrolides are macrocyclic lactones in the size ranging from 14 to 40 carbon chain atoms formed by class Ⅰ polyketide synthase enzymes. Aromatic classes are formed by one to many complex aromatic rings formed by class Ⅱ and Ⅲ polyketide synthase enzymes (Figure 1.7). Further modifications by glycosylation, methylation, hydroxylation and oxidation make these compounds more complex. Mycotoxins are the biologically made polyketides in fungi. Antimicrobial, antiparasitic, and anti-cancer agents are polyketides or polyketide derivatives such as erythromycins, tetracyclins, nystatins and antitumor epothilones.

1.2.8 Sphingolipids

Sphingolipids are a complex lipid family where a sphingoid base forms the backbone for all the molecules. Sphingoid base is synthesized de novo from fatty acyl-

CoA and serine. The first three carbons of the sphingoid base long chain acts as the glycerol molecule in glycerolipids. C1 carbon atom binds to the polar head groups. C2 carbon atom binds to the fatty acyl chain. The main class of sphingolipids are sphingoid bases which are long-chain alkyl with an amine and hydroxyl functional groups. The major sphingoid base in mammals is sphingosine. Sphingosine is a (2S, 3R,4E)-2-

11 aminooctadec-4-ene-1,3-diol. Sphingosine exists in various forms depending on the length of the carbon chain, branching, additional hydroxyl groups, and number and position of the double bonds. Ceramides are another class of sphingoid bases, where sphingosine and its derivatives are linked to the fatty acids by an amide linkage at the C2 carbon (Figure 1.8). A skin ceramide is seen with a 30-carbon fatty acid chain with a hydroxyl group on the terminal carbon atom. Ceramides are the precursors for the more complex sphingolipids such as phosphosphingolipids, phosphonosphingolipids and neutral and acidic glycosphingolipids. These complex lipid structures vary based on the polar head group attached to the C1 carbon atom of the sphingosine molecule.

Phosphosphingolipids has a phosphoric acid head group having derivatives such as phosphocholine, phosphoethanolamine, phosphoinositols and mannose.

Glycosphingolipids are important sphingolipids where the head group contains carbohydrate group. Neutral glycosphingolipids contain one or more uncharged sugars such as glucose (Glu), galactose (Gal), N-acetylglucosamine (GlcNAc), N- acetylgalactosamine (GalNAc) and fucose (Fuc). Acidic glycosphingolipids are the glycosphingolipids containing ionized functional groups such as sialic acid (N-acetyl or

N-glycolyl neuraminic acid). These acidic glycosphingolipids are also known as gangliosides. The number and the position of the sugars make gangliosides more highly complex polar lipids.

Figure 1.7: Example of polyketides - aflatoxin B1 [2]

Figure 1.8: Example of sphingolipids - N-(tetradecanoyl)-sphing-4-enine [2]

1.3 Gangliosides

1.3.1 History

The discovery of the first naturally occurring glycolipid, galactosyl ceramide or cerebroside and sphingomyelin in the brain by Thudichum, laid a broad pathway for the discovery of sugar lipids[3]. Ernest Klenk characterized the structure of hematoside in the red blood cells. The discovery of the new acidic lipids by Klenk et al. from the brains of amaurotic familial idiocy and Nieman-Pick disease patients opened a new field of study of lipid molecules that are rich in acidic carbohydrates. Klenk surmised this Substance

‘X’ was localized in the ganglion cells, hence the name as ganglioside[3]. Later, Blix identified that these lipids are observed in the normal brain. Carter et al. elucidated the sphingosine structure provided a greater understanding of gangliosides. Gottschalk et al. later discovered it contained sialic acid which established gangliosides as a distinctive polar lipid[3]. The first ganglioside structure was described by Kuhn and Wiengandt in bovine liver[3], [4]. Higher sialylation of gangliosides is seen in fish, reptile and mammalian brain tissues. This structural diversity is pertinent in the neurological studies, metabolism, cellular topology, biological functions, and pathobiological conditions[5].

Though a better understanding of gangliosides has been achieved from past work, study of gangliosides still remains a fertile field for future research.

1.3.2 Ganglioside – Structure

Glycosphingolipids are glycolipids in which a lipophilic sphingoid or ceramide moiety are bound to at-least one sugar molecule. Gangliosides are

Figure 1.9: A typical example of a ganglioside, GM1 monosialo ganglioside [2]

15 sialoglycosphingolipids having one or more sialic acids in the oligosaccharide head group along with the sphingosine containing ceramide moiety (Figure 1.9). Gangliosides have amphiphilic lipid composition with a distinctive carbohydrate complex and a varied ceramide moiety[6],[3].

1.3.2.1 Gangliosides – Carbohydrate complex[7]

The sugar head group in ganglioside is made up of five, six and nine membered ring sugar molecules (Table 1.2), of which, major neutral sugars are glucose (Glc), galactose (Gal), fucose (Fuc), N-acetylgalactosamine (GalNAc) and N-acetylglucosamine

(GlcNAc)[8]. The primary sugar molecule that differentiates gangliosides from other glycosphingolipids are N-acetyl neuraminic acid (Neu5Ac)/ N-glycolyl neuraminic acid

(Neu5Gc) or its derivative acidic groups. N-acetyl neuraminic acid, also known as sialic acid, gives a negative charge to the ganglioside at pH 7.0. In general, D and L configurations are not referred in the nomenclature of gangliosides. But all the monosaccharides in the polar head group have D configuration except the fucose and rhamnose which have L configuration.

A typical carbohydrate sequence of the ganglioside (Figure 1.10) consists of glucose, where the β-1 anomeric carbon in the glucose is attached to the O-1 of sphingoid base in the ceramide moiety. To this glucose, galactose is attached by a β1,4 linkages, followed by a N-acetyl galactosamine, attached to the galactose molecule by a β1,4 linkages. After the N-acetyl galactosamine there is another galactose molecule connected through a β1,3 linkage. Sialic acid is attached to one or more galactose molecules by a α- ketosidic linkage (α2,3). Additional sialic acids may be attached to galactose-linked sialic

Table 1.2: Recommended abbreviations for some monosaccharides, derivatives and related compounds [7]

Name Symbol N-acetylgalactosamine GalNAc N-acetylglucosamine GlcNAc N-acetylneuraminic acid* Neu5Ac or NeuAc 5,9-N-O-diacetylneuraminic acid* Neu5,9Ac2 fucose (6-deoxygalactose) Fuc Galactitol Gal-ol Galactosamine GalN galactopyranose-3-sulfate Galp3S Galactose Gal galacturonic acid GalA Glucitol Glc-ol Glucosamine GlcN Glucose Glc glucose-6-phosphate Glc6P glucuronic acid GlcA N-glycoloylneuraminic acid* Neu5Gc or NeuGc myo-inositol** Ins Mannose Man 4-O-methylgalactose Gal4Me Rhamnose Rha Xylose Xyl

17 acids by α2,8 linkages. The term ‘iso’ is used when the galactose and N-acetyl galactosamine is attached by a β1,3 linkages instead of by the usual β1,4 linkage. The term ‘neo’ denotes when N-acetyl galactosamine and galactose are linked β1,4 instead of

β1,3.

Most sialic acids are linked to a galactose molecule and rarely are linked to

N-acetyl galactosamine. Svennerholm designated a simplified ganglioside nomenclature based on the number of sialic acids and their position (Table 1.3).

Depending on the number of sialic acids, gangliosides are thus sub-classed as mono-, di-, tri- and quaternary (GM, GD, GT and GQ) gangliosides and so on. Furthermore, gangliosides are classified as o-, a-, b- and c- series based on the position of the sialic acid in the oligosaccharides (Figure 1.11). The variation in structure of the oligosaccharide group in gangliosides as summarized above is a factor that contributes to the extensive diversity of ganglioside structures. To this point one study identified 188 different oligosaccharide of gangliosides based on the number of sugars and their positions in the polar head[9].

1.3.2.2 Gangliosides – Ceramide group[7], [9]

The oligosaccharide moiety of gangliosides is attached to the O-1 of the sphingoid, which are long-chain aliphatic amino alcohols. The C-2 carbon of the sphingoid base attaches to the fatty acid by an amide linkage, forming a ceramide moiety.

Sphingoids are classified based on the presence of alcohol groups such as monohydro-, dihydro- and trihydrosphingosine and so on. The common sphingoids are dihydrosphingosines which are also referred as sphinganines are made up of 18 carbon chain length. A hydroxyl group is present on C3 carbon atom of the sphingoid long chain

18 bases. Similarly, carbon chain-length homologs of 20-carbon atoms are referred as icosasphinganine. Unsaturated spinganines or sphingoids are defined by the number and position of each olefinic center. The most abundant unsaturated sphingoid is sphingosine

[(E)-sphing-4-enine)] has an 18-carbon chain length and a trans double bond at C4-C5 carbon atoms. The IUPAC name for this sphingosine is (2S,3R,4E)-2-aminooctadec-4- ene-1,3-diol. Structural variants such as more double bonds or substituents hydroxy, oxo and methyl groups are less seen. Carboamidiacally attached fatty acids have usual carbon chain length of C16-C26. Most common fatty acids are stearic acid (C18:0) and arachidic acid (C20:0). Fatty acid moieties having double bonds and /or hydroxyl groups are only rarely present.

Table 1.3: Abbreviations of gangliosides using the Svennerholm system [7]

Abbreviatio Structure n

Neu5Acα3Galβ4GlcCer GM3

GalNAcβ4(Neu5Acα3)Galβ4GlcCer GM2

Galβ3GalNAcβ4(Neu5Acα3)Galβ4GlcCer GM1a

Neu5Acα3Galβ3GalNAcβ4Galβ4GlcCer GM1b

Neu5Acα8Neu5Acα3Galβ4GlcCer GD3

GalNAcβ4(Neu5Acα8Neu5Acα3)Galβ4GlcCer GD2

Neu5Acα3Galβ3GalNAcβ4(Neu5Acα3)Galβ4GlcCer GD1a

Galβ3GalNAcβ4(Neu5Acα8Neu5Acα3)Galβ4GlcCer GD1b

Neu5Acα8Neu5Acα3Galβ3GalNAcβ4(Neu5Acα3)Galβ4GlcCer GT1a

Neu5Acα3Galβ3GalNAcβ4(Neu5Acα8Neu5Acα3)Galβ4GlcCer GT1b

Galβ3GalNAcβ4(Neu5Acα8Neu5Acα8Neu5Acα3)Galβ4GlcCer GT1c

Neu5Acα8Neu5Acα3Galβ3GalNAcβ4(Neu5Acα8Neu5Acα3)Galβ4Gl GQ1b cCer

1.3.3 Gangliosides – Biosynthesis [10]

Synthesis of gangliosides is initiated in the endoplasmic reticulum of a cell. After initial ceramide synthesis, sugars are added in golgi apparatus. Ceramide is the precursor for the ganglioside biosynthetic pathway. All the gangliosides are derivatives of lactosylceramide (LacCer) except the GM4, which is derived from galactosylceramide

(GalCer). Attachment of sugars is catalyzed by a class of enzymes called glycosyltransferases (Figure 1.10). Sialyltransferase family (Ⅰ - Ⅶ) aids in the addition of sialic acid groups to the lactosyl/galactosyl ceramide moiety. GM3, is the first and the simplest ganglioside formed from a CMP-sialic acid being added to LacCer by LacCer

α2–3 sialyltransferase (ST-I or GM3 synthase). Subsequently CMP-sialic acid is added to

GM3 and GD3 to synthesize GD3 and GT3 by GM3 α2–8 sialyltransferase (ST-II or

GD3 synthase) and GD3 α2–8 sialyltransferase (ST-III or GT3 synthase), respectively.

Further gangliosides in the biosynthetic pathway are synthesized from the LacCer, GM3,

GD3 and GT3 precursors which form more complex gangliosides belonging to the o-, a-, b- and c-series, respectively. Other neutral sugars are catalyzed to the above gangliosides by UDP-GalNAc: LacCer/GM3/GD3/GT3 β1–4 N-acetylgalactosaminyltransferase

(GalNAcT or GA2/GM2/GD2/GT2 synthase), UDP-Gal: GA2/GM2/GD2/GT2 β1–3 galactosyltransferase (GalT-II or GA1/GM1/GD1b/GT1c synthase), CMP-sialic acid:

GA1/GM1/GD1b/GT1c α2–3 sialyltransferase (ST-IV or GM1b/GD1a/GT1b/GQ1c

Figure 1.10: Ganglioside biosynthetic pathway: Gangliosides are classified into asialo-, a, b- and c- series based on the number and position of sialic acid in the polar head [10]

22 synthase), and CMP-sialic acid: GM1b/GD1a/GT1b/GQ1c α2–8 sialyltransferase (ST-V or GD1α/GT1aα/GQ1bα/GP1cα).

Gangliosides are majorly present in the neurological cells of the brain. During the brain development, dramatic changes occurs in the ganglioside biosynthetic pathway.

From newborn brain tissue to an adult brain tissue, the simple gangliosides GM3 and

GD3, which predominate in newborn are replaced by the complex gangliosides such as

GM1, GD1a, GD1b and GT1b. These changes in the ganglioside reflects a pattern of expression levels of ganglioside synthases.

1.3.4 Gangliosides – Position[11]

In physiological conditions, biological membranes are made up of bilipid layer mostly made up by glycerophospholipids. Though gangliosides are under-represented lipids in the biological membranes they are an important lipid component of all cell membranes, being particularly abundant in cell membranes of mammalian nerve cells (6-

7% of the total lipids in brain)[12]. In cells, gangliosides are primarily, but not exclusively, localized in the outer leaflets of plasma membranes. The distinct molecular properties of gangliosides promote lateral associations within the bilayer forming groupings with other sphingolipids, cholesterol and selected proteins called lipid rafts or membrane microdomains. In this microdomains (Figure 1.11) the carbohydrate arm of the gangliosides is freely rotating in the extra cellular space which makes gangliosides functionally important. Within the cell membrane, the ceramide moiety, which is made up of sphingosine and a fatty acid, is anchored into the lipid membranes. The major

Figure 1.11: Position of gangliosides within the lipid rafts of cell membrane. Sugar head of gangliosides freely rotates whereas the ceramide moiety is embedded in the bilayered cellular membrane [11] (Malchiodi-Albedi et al, International Journal of Alzheimer’s Disease, 2010)

24 ceramide, made up of dihydrosphingosine and stearic acid is dominantly present enhancing the lateral self-associations. This capability to form self-associations in the microdomains make it more rigid, stable and less mobile aiding in the complementary binding with extracellular proteins in comparison to the fluidic bilipid membrane.

1.3.5 Gangliosides – Biological functions

Gangliosides regulate various biochemical and cellular functions. The regulation of the gangliosides expression during developmental stages strongly correlates with its functional roles. Gangliosides regulate these interactions through two molecular mechanisms: lateral interactions within the lipid bilayer proteins (cis) and complementary binding on the cell surface or in extracellular space[13] with glycan binding proteins.

Examples are given below- the lateral interactions of insulin receptors with GM3 ganglioside regulate the insulin-induced autophosphorylation and downstream signaling events is an example for cis interactions of gangliosides. An example of trans interactions is the glycan binding protein which binds specifically myelin-associated glycoprotein,

MAG, on the innermost layer of myelin GD1a and GT1b gangliosides, maintaining the axon-myelin stability long-term. Gangliosides play a critical role in the activation/inhibition of various growth factor regulators (Figure 1.12). GM3 inhibits the insulin receptors, while GD1a enhances activation of the EGF receptor. GM1 ganglioside has both inhibition and activation function with PDGF and TrkA receptors, respectively[14].

Neural stem cells (NSCs) and neural progenitor cells (NPCs) are the precursors to the major neural cell types. NSCs serve as a cellular reservoir for CNS development and

Figure 1.12: Growth factor modulation by gangliosides: Various gangliosides have either excitatory or inhibitory effect on the growth factor expression [14]

26 for replacement of lost normal cells. Gangliosides mediate and modulate the NSC regulation known as glycosignaling. Gangliosides regulate axon outgrowth as follows.

Sialidase Neu3 enzyme cleaves the major gangliosides such as GD1a, GD1b, GT1b except GM1[15]. This results in a sharp increase in the GM1 which enhances the neurotrophin receptor tyrosine kinase TrkA resulting in the increased axonal outgrowth.

Similarly, in the adult brain, when MAG binds to axonal gangliosides such as GD1a,

GD1b, and GT1b gangliosides, the Ras homolog gene family, member A, RhoA activates resulting axon outgrowth restriction.

Cell to cell interactions are one crucial functions of gangliosides. The glycans on one cell interacts with the glycan binding proteins (lectins) on an apposing cell to mediate cell-cell interactions. Glycan mediated interactions between the juxtaposition cells leads to the intracellular signaling pathways.

Gangliosides modulate natural killer (NK) cell cytotoxicity. Siglecs (sialic-acid- binding immunoglobulin-like lectins) are animal lectins that bind to sialic acid-containing glycans. In specific, gangliosides “b-series” GD3, GD1b and GT1b gangliosides have a preferential high binding affinity for Siglec-7. Siglecs have inhibitory receptors for natural killer cells. Siglec-ganglioside binding results in a suppressed NK cell-mediated cytotoxicity[16].

Gangliosides function as cell adhesion receptors in inflammation. During injury or inflammation circulating neutrophils bind to the blood vessels through selectins that are ganglioside site-specific.

Different tissues and cell-types will have variable distribution of ganglioside structures and thus serve as specific markers for cell types. GD1a and GM1 gangliosides were investigated to evaluate the direct effect on stem cell differentiation[17].

In summary, the functions of gangliosides include modulating glycoproteins, growth factors, brain-derived neurotrophic factors, immune response, cellular calcium concentrations, apoptosis (programmed cell death), inflammatory responses and cell adhesion. It also protects against inflammation and degeneration in the central nervous system, gangliosides have both broad application, as well as specificity according to cell type.

1.3.6 Gangliosides – Diseases

1.3.6.1 Genetic diseases

Gangliosides undergo constant metabolic turnover where the rate of synthesis is on par with the rate of breakdown[18]. Specific hydrolytic enzymes in lysosomes hydrolyze the specific sugars of the ganglioside. A genetic alteration resulting in the defect in any of these enzymes impairs the ganglioside degradation (Figure 1.13). This results in the accumulation of products in the cells/tissues causing a serious disease.

Niemann-Pick disease, a rare genetic defect in infants leads to accumulation of sphingomyelin in the brain, spleen and liver because of a defective enzyme, sphingomyelinase that cleaves phosphocholine from sphingomyelin. Similar defective enzymes in Sandhoff’s and Fabry’s disease results in the accumulation of globoside and galactose ceramide, respectively.

Figure 1.13: Genetic diseases causes by accumulation of various gangliosides [10]

Defective galactosidase and hexaminadase A enzymes lead in brain and spleen to the accumulation of GM1 and GM2 gangliosides, resulting in a general gangliosidosis and

Tay-Sachs disease, respectively. These genetic diseases develop progressive retardation, paralysis, blindness and death by the age of 3-4 years[10].

1.3.6.2 Neurological diseases

Gangliosides play a pivotal role in many neurological diseases[19]. Alzheimer’s disease (AD) is a chronic neurodegenerative disorder. It results in the slow progress of dementia and worsens with the age. Amyloid precursor protein (APP) regulates the nerve growth and repair. A genetic mutation results in the cleavage of APP into amyloid beta,

Aβ and APP intracellular domain, AICD peptides. Gangliosides are the potent stimulators of Aβ protein production. In the (Figure 1.14) Aβ excess production leads to formation of a Ganglioside-Aβ complex with the GM3 ganglioside, a precursor substrate of GD3S, a synthase enzyme which converts GM3 into a- and b- series gangliosides such as GD3 and so on. Along with that, AICD down regulates the GD3S transcription. These two molecular mechanisms alter the ganglioside metabolic turnover. The concentration and composition of gangliosides are altered in the brain studies of transgenic mouse models and AD patients. Alzheimer’s pathologically characterized by extracellular senile plaques and intracellular neurofibrillary tangles by the ganglioside-Aβ complex. These accumulations lead to the formation of severe dysphoric-form amyloid angiopathy. It would be of great interest in targeting ganglioside synthase enzymes to study the cognitive deficits, amyloid plaque formation, and neurodegeneration associated with

AD[20], [21].

Figure 1.14: Proposed scheme for the ganglioside-bound Ab (GAb) hypothesis in Alzheimer’s disease. Soluble amyloid β -protein (Aβ) binds to the GM3 ganglioside cluster on the neuronal membrane [20]

Parkinson’s disease (PD) is another neurodegenerative disorder which is inherited but sometimes it is sporadic. Though the etiology is not established, mitochondrial dysfunction, oxidative stress, and inflammation leads to degeneration of dopaminergic

(DA) neurons of the substantia nigra pars compacta (SNpc) and certain non-DA cells. N- acetyl galactosaminyl transferase (GalNAcT) disrupted gene in mutant mice were reported to express a highly elevated α-synuclein in the SNpc of the brain leading to motor disability due to loss of dopaminergic neurons. Symptoms were alleviated by administration of L-dopa or LIGA-20, a membrane permeable analog of lyso derivative

GM1 ganglioside. LIGA-20 can penetrate the blood-brain barrier (BBB) and access intracellular neurons serving as a replacement for the GM1 ganglioside. This suggests a role of GM1 in the PD disease. A clinical trial on PD patients were under study by the administration of GM1 ganglioside, results showing an improvement in the motor neurons function[22], [23].

Guillain-Barré syndrome (GBS) is an auto-immune disorder characterized by the immune system attack on the motor and peripheral nerves. Infectious agents such as

Campylobacter jejuni (C. jejuni) bacteria or some viruses causes a change in the nature of the nerve cells making it foreign to the immune system. The immune mediated attack on the myelin sheath of motor nerves affects the signal transduction. The IgG antibodies produced in lieu of the bacteria invasion reacts with the peripheral nerves. Gangliosides in the peripheral nerves cross react with the antibodies. The neuromuscular junction

(NMJ) is a rich source for gangliosides where the circulating autoantibodies interacts and damages the neuron and leads to a progressive neuropathy[24].

Huntington’s disease (HD) is another rare neurodegenerative disorder that is characterized by chorea and progressive motor, psychiatric and cognitive decline. HD is a member of the Type II trinucleotide repeat neurodegenerative disorders. Ganglioside changes in the fore brain of R6/1 transgenic mice were studied. In this study, alteration in the ganglioside biosynthetic pathway were identified resulting in GM1 gangliosides down regulation causes high elevation of the b-series gangliosides, especially GD3 ganglioside.

Correlations were also found between a decrease in gene expression of GalNAcT and a decrease in the downstream products GD1a, GD1b and GT1b. A similar correlation with the ganglioside changes were seen in the mice cerebellum. These results implicate a role of gangliosides in the pathogenesis of HD[25].

Ganglioside effects on eye physiology is comparable to the brain. Gangliosides show composition changes in neuronal development[26], [27]; involvement in neurotrophic and growth factor actions (maintaining various eye cell viability)[28],[29] with demonstration of GM1 ganglioside neuroprotection of eye tissue in a rat model[30]; and changes in ganglioside synthesis with exposure to light in various types of ocular cells[31], [32]. Demonstration of the neuroprotective action of gangliosides has been reported in ischemia (condition of reduced oxygen availability) of the retina[33]; for lens and retinas exposed to harsh oxidation conditions[34], and for retinal neurons exposed to neurotoxic agents[35]. In one animal study, aged retinas were restored to a younger status with GM1 ganglioside treatment, as assessed by biomarkers of age[36]. There have been, however, only a few studies investigating the involvement of gangliosides in clinical conditions of the eye. One report studied retinal dystrophy in a rat model[37]. These rats, which manifest a loss of photo receptors, and thus decreased vision, showed a marked

33 increase in retinal gangliosides. Other studies implicated increase levels of gangliosides in the pathological effects of diabetes on vision[38], [39].

The above examples show the importance of ganglioside metabolism, catalytic enzymes in the ganglioside biosynthetic pathway and glycogenes encoding these catalytic enzymes in various neurological disorders. Apart from the structural role of gangliosides, their biological roles in cell-mediation, cell recognition and signal transduction studies[40], [41] prove gangliosides to be vital in many genetic and neurological diseases.

1.4 Glaucoma – A neurodegenerative eye disorder

Glaucoma is a chronic, degenerative optic neuropathy with a distinguished optic nerve appearance. It is characterized by the progressive degeneration and functional deterioration of the optic nerve which includes the optic nerve head (ONH) and the retinal nerve fibre layer. The damage to the optic nerve progressively leads to the loss of vision and in some patients, blindness. The damage to the optic nerve is irreversible. It is a major eye disease affecting more than 70 million people worldwide, being the major cause of worldwide blindness. Among the affected, 10% patients with glaucoma have bilateral blindness. The major drawback of the disease is it is asymptomatic and only found when it is severe.

1.4.1 Glaucoma – Types and conditions

Glaucoma is broadly classified based on the optic disc cupping, a characteristic deformation of the optic nerve head and damage to retinal ganglion axons. In the normal eye, the anterior segment circulatory system[42] is made up of the aqueous humor, a natural clear fluid that is produced by the ciliary body. It is circulated throughout the

34 anterior chamber which nourishes the crystalline lens and cornea. The aqueous humor drains out through the trabecular meshwork which passes into Schlemm’s canal, located at an angle formed by the iris and cornea. This angle is known as the iridocorneal angle.

Glaucoma is broadly distinguished as open-angle glaucoma and angle -closure glaucoma based on the aqueous humor anatomical configuration of outflow pathway. In open-angle glaucoma, the iridocorneal angle is open and there is no obstruction to the flow of aqueous fluid, but rather it is the trabecular meshwork which is often blocked internally[42]. In angle closure glaucoma, the iris blocks the angle at the iris and cornea leading to a blockage of the aqueous humor produced by the ciliary body (Figure 1.15).

Both open-angle and angle closure glaucoma lead to increase in the intraocular pressure

(IOP) leading to the damage of the optic nerve head. Glaucoma is further classified as primary and secondary. When the cause of glaucoma is unknown it is referred as primary glaucoma, where as if the cause is diagnosed clinically (such as cataract, diabetes or tumor) it is secondary.

Open-angle glaucoma is the common type of glaucoma. Open-angle glaucoma is highly prevalent in individuals of African and European descent and commonly seen in the adults of the age 40 or more. Though it is often associated with an increase in the

IOP, a normal IOP has been observed in primary open-angle glaucoma. Secondary open- angle glaucoma (SOAG) are various kinds based on the aqueous flow blockage.

Pigmentary glaucoma results from a blockage of the aqueous humor tracts by pigment granules released from the iris[43]. Exfoliative glaucoma is another kind of SOAG where the tracts are blocked by exfoliative materials from the lens and internal structures.

Figure 1.15: Circulation of aqueous humor: The anterior segment of the eye shows the direction of aqueous humor secreted by the ciliary body and exits the trabecular mesh network [42]

Angle-closure glaucoma is identified by an increase in IOP that is caused by the narrow end of the iridocorneal angle. Secondary angle-closure glaucoma (SACG) are several types such as neovascular, exfoliative or pigmentary glaucoma and others.

Neovascular glaucoma is an important angle closure glaucoma which results in angle closure when aqueous outflow tracts are blocked by the fibrovascular membranes. Angle- closure is prevalent among the Asian, particularly of Chinese descent[44].

There are other kinds of glaucoma. Glaucoma genetically inherited by children is known as childhood glaucoma. The glaucoma associated with the inflammation of the uvea is called uveitic glaucoma. Developmental forms of glaucoma, which are rare, include primary congenital glaucoma and glaucoma associated with syndromes (e.g., aniridia or the Axenfeld–Rieger syndrome).

1.4.2 Primary open-angle glaucoma (POAG)

POAG is the second most common eye disease which causes blindness in the

United States. The ethnic breakdown of POAG patients is as follows: 18% Latin and

Hispanics, 15% Black, 7% White and 5% Asian. African Americans develop POAG earlier in comparison to other races. There are various risk factors which affect the eye vision in the POAG patients. The major risk factor among them is the age. Adults above the age of 40 have significant risk, with the highest risk at the age 60 and above. Men have 30% more risk compared to women. The other important risk factor is the IOP. The prevalence of POAG is high with an increase in IOP. Nonetheless, many cases treated for

IOP did not show a reduction in the optic nerve head damage. A genetic inheritance is another major contributor. Other risk factors include high blood pressure and type 2 diabetes mellitus[43], [44].

1.4.2.1 Glaucoma – Pathophysiology

In general, intraocular pressure is generated between the aqueous humor secreted by the ciliary body and the aqueous humor outflow at trabecular meshwork (Figure 1.16).

The pressure generated in the anterior chamber spreads to all compartments of the eye. In normal eye, unmyelinated axons of the retinal ganglion cells (RGCs) merge at the ONH and exit the eye in the optic nerve, which is a myelinated axon. The ONH consists of neuro retinal rim which is made up of RGC axons, the optic cup (a central depression) and the lamina cirbrosa (collagenous structure). In glaucoma, a change of pressure leads to the damage of the ONH, along with the RGC axonal death and thinning of the lamina cirbrosa (Figure 1.16). The pathophysiology of POAG starts with retinal nerve fibre loss

(RNFL) which affects the visual field. RNFL damage shows the ONH as the basal point for glaucoma disease. Along with the deformation of ONH, loss of retinal ganglion cell

(RGC) axons is the characteristic of glaucoma compared to the other eye disease such as ischaemic or traumatic optic neuropathies. An increase in the IOP increases the damage to the ONH and RGCs death. The increase in IOP results in the calcium-dependent cell death, vascular ischaemia, oxidative stress, excitotoxicity, inflammation and reactive gliosis[45].

Figure 1.16: Glaucoma pathology: A: Aqueous humor movement in normal eye; B: Normal optic nerve head with no optic disc, healthy lamina cirbrosa and retinal ganglion cells; C: Optic disc formed, retinal ganglion cells damage and posterior displacement of lamina cirbrosa [42]

1.4.2.2 Retinal gliosis: Role of astrocytes

Reactive gliosis is a major contributing factor in the progression of glaucoma disease. A response by glial cells to a wide variety of insults on the central nervous system is referred to as reactive gliosis. Glial cells (in general, non-neuronal cells) are a heterogeneous pool of cell types in the central nervous system. They are particularly present in the synapse, giving primary support to neurons response in to damaging insults by providing energy substrates and eliminating excess neurotransmitters. Gliosis has been linked to the progression of pathology in various neurological disorders. Reactive gliosis is observed in the eye during glaucoma, which is termed retinal gliosis. Astrocytes are major glial cells which undergo reactive gliosis in glaucoma[46].

1.4.2.3 Role of GM1 ganglioside in glaucoma – Preliminary experiments

Gangliosides are a diverse group of sialic acid-containing glycosphingolipids that are abundant in the CNS. Often found in plasma membrane microdomains that include growth factor receptors and adhesion molecules, gangliosides have been implicated in cell recognition, neurite outgrowth and signal transduction. Type 1 and type 2 astrocytes can be differentiated by their affinity for A2B5, an antibody directed against gangliosides[47],[16]. Fibrous type 2 astrocytes also have GD1b and GT1b gangliosides on their membranes that can bind tetanus toxin[48]. Ganglioside expression is thought to be developmentally regulated since retinal Müller glia express tetanus toxin-binding gangliosides and A2B5 gangliosides between P5 and P7, the period over which retinal neurons and glia finish populating the retina[49]. These data suggest that the onset or termination of ganglioside expression signals important cellular transition. In keeping with this concept, a group of gangliosides, but not GM1, is upregulated in retinal glial

40 cells after optic nerve axotomy[50]. Knowing that glial cells express gangliosides when

RGC axons are cut suggests that ganglioside expression could be fundamental to glial response to degeneration.

Glaucoma is caused by the degeneration of retinal ganglion cells (RGCs), a group of nerve cell bodies, secondary to axon damage at the optic nerve head. RGCs express

GM1 ganglioside in normal retinal tissue. In glaucoma, RGCs are gradually degenerated with the GM1 gangliosides being up-regulated/increased in the astrocyte cells (a glial cell) resulting from retinal gliosis. We hypothesize that GM1 ganglioside up- regulation/uptake by the astrocyte cells is a crucial stage in the retinal gliosis, being responsive as a compensatory repair mechanism to glaucoma progression. The possibility that astrocytes express GM1 gangliosides to compensate for the RGCs loss during glaucoma degeneration has implications for the therapeutic development for glaucoma.

RGCs in a normal retina express GM1 gangliosides as confirmed by their efficient and specific uptake of cholera toxin-B (CTB). CTB binding to GM1 gangliosides occurs through specific protein-sugar interactions. In the DBA/2J mouse, a glaucoma model mouse, astrocytes, which do not express GM1 gangliosides, have been shown to bind

CTB, indicating the presence of GM1 ganglioside (Figure 1.17). Dysfunctional RGCs are identified by the accumulation of phosphorylated neuro-filament[51], [52] (Figure 1.17).

Other indicators of dysfunctional RGCs are lack in action potential[53] and dendritic arborization[54]. Astrocytes with bound CTB were not observed in aged normal mice however were expressed/accumulated by/in astrocytes in aged glaucoma model mice

(Figure 1.18).

Figure 1.17: Retina from 12-month-old DBA/2J showing the distribution of heavy chain neurofilaments (red), (arrows) and degenerating, CTB-negative RGCs filled with retrogradely transported FluoroGold (blue) and intraocularly injected CTB (green). CTB labeled RGCs

(D. Inman, Department of Pharmaceutical Sciences, Northeast Ohio Medical University, unpublished data)

Figure 1.18: Fixed retinas from 12-mo C57Bl/6 (left) or DBA/2J (right) mice incubated with CTB conjugated to AlexaFluor-488; RGC somas (arrow) and axons (arrowheads) are CTB-positive. In the aged DBA/2J, most CTB was taken up by astrocytes. Scale bar=50µm

(D. Inman, Department of Pharmaceutical Sciences, Northeast Ohio Medical University, unpublished data)

Gangliosides are associated with a wide variety of cellular functions including immune function modulation[55], growth factor receptor regulation and maintaining and repairing the CNS[56]. RGC degeneration during glaucoma can be associated with the ganglioside cellular functions particularly growth factor receptor regulation and innate immune regulation. Gangliosides regulate most of the growth factor receptors by altering their binding, dimerization and activation[56],[57] . Nerve growth factor (NGF) and brain- derived neurotrophic factor (BDNF), which is secondary to NGF, play an important role in protection, development and survival, respectively[58],[59]. Loss of

GM1 from RGCs as observed during the progression of the glaucoma contributes to the decreased activity of BDNF resulting in the death of the RGCs[60]. Thus, GM1 loss correlates with RGCs death in glaucoma progression.

GM1 gangliosides are neuroprotective agents. It has been shown that injection of

GM1 into the eye after optic nerve axotomy mitigates RGCs degeneration[61]. Studies performed at Dr. Inman’s lab (NEOMED) also show that decreased GM1 in RGCs is matched by an increase in GM1 in astrocytes, which could be either an acquisition from deteriorating RGCs or could be self-expression of GM1 in astrocytes in response to degeneration. Contrary to a neuroprotective role of GM1 in astrocytes a pathologic role should also be considered.

Astrocytes are enriched in specific phagocytic pathway proteins[62] suggesting their roles in phagocytosis. GM1 in astrocytes might be a form of antigen presentation, resulting in further RGCs death due to phagocytosis.

Summarizing the above discussion on earlier research, GM1 gangliosides are expressed by RGCs in normal retinal tissue while in glaucoma progression expression

44 declines as RGCs degenerate caused either by the growth factor deregulation or by astrocyte phagocytosis. At the same time, it is confirmed that GM1 are up-regulated

/increased by astrocytes in correlation with the decline of RGCs.

1.4.3 Research strategy

Gangliosides are sialic acid containing glycosphingolipids. The heterogeneity in both the carbohydrate arm and ceramide moiety presents analytical challenges.

Ganglioside-CTB binding studies in the preliminary results show a high specific binding of CTB to GM1 ganglioside. However, the binding affinity studies on CTB with the ganglioside family shows variability. Though CTB has high binding affinity to GM1 ganglioside, flow cytometry studies on neural precursor cells reveal the CTB binding to the other gangliosides such as GD1a, GD1b, GT1b, GM3, GM2, GM1a, GM1b and fucosyl-GM1[63]. Thus, relying on results of CTB binding to gangliosides cannot necessarily be interpreted as an over-expression of GM1, given the broad cross-reactivity of other gangliosides with CTB. Hence a highly sensitive, specific and quantitative analytical method is needed to identify and quantitate the different ganglioside molecular species.

The present work develops and evaluates a novel LC-MS/MS technique, including development of improved sample preparation technique, to quantitate different gangliosides in tissue samples including mouse retina. A preliminary profiling of gangliosides in a normal and glaucomatous retinal tissue needs to be assessed. These results lay a foundation for the age studies to identify the changes within the ganglioside family as the glaucoma disease progresses studies of the present work lay a foundation for cellular ganglioside studies in RGCs and astrocytes in normal and glaucoma mice to

45 determine whether the increase in astrocyte GM1 is due to an increased expression or an uptake of GM1 lost by the RGCs during the glaucoma disease progression. In addition, changes in other gangliosides may play a role in the disease which the newly developed

LC-MS/MS method will be able to assess.

CHAPTER Ⅱ

DEVELOPMENT OF NOVEL LC-MS/MS TECHNIQUE IN GANGLIOSIDE

ANALYSIS: SUPERIOR SEPARATION OF GANGLIOSIDES USING A

PHENYL-HEXYL HPLC COLUMN

2.1 Introduction

Gangliosides are diverse structural molecules with a variable carbohydrate skeleton and varied ceramide moiety. The molecule has two basic sections: 1) the carbohydrate “arm” consists of a chain with a branch of several sugar groups and 2) the ceramide group consists of two “arms”, one being a fatty acid group and the second being

47 a sphingosine group. Heterogeneity in gangliosides arise in all three “arms” of the ganglioside structure. In the carbohydrate arms from different sugar units (glucose, galactose, sialic acid, etc.), a different order of sugar units, a different number of sugars, and various locations of the branch. In the two arms of the ceramide section differences arise in the length (number of carbon atoms) in the sphingosine and fatty acid arm, in the presence or absence of double bonds between two of the carbons and the positions of these double bonds. Taken together this gives a huge variability in ganglioside structures challenging the methodology to measure each molecular structure. One report identified

188 subgroups of gangliosides based on differences in the carbohydrate arm[9]. This diversity is further increased by heterogeneity in the ceramide arms. As an example of the degree of heterogeneity to expect in neurological sample, human fetal brain tissue has identified 137 gangliosides[64] The various analytical methodologies used to determine gangliosides in biological samples are discussed below.

2.1.1 Ganglioside analysis – Immunological assays

Immunological studies have been applied to selectively identify gangliosides.

Antibodies to gangliosides have been applied to study neuromembranes, for ultrastructural localization in neurophysiology[3] The complex specificity of conventional antibodies poses limitations in ganglioside studies. Antibodies to carbohydrate determinants recognize a specific sugar sequence and bind regardless if the sugars are attached to a protein, lipid or nucleic acid. These antibodies can bind to either to the peripheral neutral sugars or internal sugar sequence and thus antibodies display cross- reactivity. The radio-immunoassay studies of anti-GM1 antibodies show a binding to multiple gangliosides such as GM1, GD1b and GM2[6] Monoclonal antibodies show

48 more selectivity towards gangliosides because of their uniformity, making them ideal agents for immunoassays. Some of the specific antibodies (biotinylated CTB or a specific binding protein) for gangliosides GM1, GD1b, GD3 and GQ1b are mAb GGR12, mAb

GMR19, and mAb GMR13. Enzyme-labeled antibody and immunofluorescence methods have been developed recently to fix selective gangliosides in paraformaldehyde-fixed brain tissues[65] However immunoassays can be less selective compared to other techniques and hence other analytical techniques are employed for ganglioside analysis[65].

2.1.2 Ganglioside analysis by chromatography

Chromatography is a physical technique to partition mixture of compounds between stationary and mobile phases. Depending on the type of stationary and mobile phases, chromatography is classified into multiple types. Ganglioside analysis studies in the chromatographic techniques along with their limitations are summarized below.

2.1.2.1 Thin layer chromatography (TLC)

TLC is the easiest and quickest chromatographic technique to separate ganglioside mixtures. Stationary phases are generally silica, alumina or cellulose on a flat inert plate. Several solvent systems, such as combinations of chloroform, methanol and water, are used to separate gangliosides based on their size and affinity for the stationary phase. TLC provides screening information regarding the possible structure in comparison to the standards. Detection is achieved through staining techniques, including colorimetric, cholera toxin and radio imaging techniques. TLC is performed both in one

49 and two dimensions for better separation and quantitation. Rat brain and human fibroblasts have been analyzed using TLC[66],[67].

2.1.2.2 High performance thin layer chromatography (HPTLC)

HPTLC is a high-resolution form of thin layer chromatography used to identify and quantify chemical compounds, separating them from impurities. Enhancements include high-quality TLC plates with much smaller particles containing the stationary phase than conventional TLC. HPTLC achieves better resolution and attains lower detection limits, as well as having a quantitative capability compared to conventional

TLC. HPTLC can also employ repeated development of the plate using a multiple development device. HPTLC is hyphenated with many detectors such as UV, diode-array and fluorescence spectroscopy for quantitative detection and structural information. Now- a-days, more advance detectors such as Fourier-transform infrared (FTIR), Raman spectroscopy and mass spectrometry have been applied to in situ detection of analyte zones on a HPTLC plate. Gangliosides from the brain tissue extractants which were separated on ion-exchange affinity columns into individual fractions were spotted on silica gel HPTLC plates. Similar to TLC, one and two dimensional HPTLC are performed on gangliosides for better resolution, identification and quantification. Anti- glycosphingolipid antibodies, cholera toxins and other antibodies are used on HPTLC plates for specific identification and confirmation of gangliosides.

TLC and HPTLC techniques are severely limited in their sensitivity and specificity. HPTLC can separate only 15 bands of major ganglioside groups which does not approach the resolution needed to profile the molecular heterogeneity of gangliosides[68].

2.1.2.3 Liquid Chromatography

Liquid chromatography separates chemical compounds through a partitioning mechanism between the solid stationary phase and the liquid mobile phase. There are various stationary phases based on their polarity and other physiochemical properties that are used to separate gangliosides.

Ion-exchange chromatography: The process of ion exchange can be considered as a competition between the solute ions and counter ions present in the mobile phase for fixed sites of opposite charge on the chromatographic support. The quality of a given separation can be manipulated by varying the nature and concentration of the counter ion or by changing the pH of the mobile phase. Silica-based phases with a chemically bonded primary amine group (anion exchanger) has been used to separate acidic gangliosides.

The main drawback of ion-exchange chromatography is its incompatibility with mass spectrometric detection.

Reverse-phase partition chromatography: In reverse-phase chromatography, separation is based on the selective interaction of solute molecules between a relatively non-polar stationary phase and a relatively polar mobile phase. The various non-polar stationary phases that are employed for ganglioside separation are octadecylsilyl-, octylsilyl-, butylsilyl carbon chain lengths. As the carbon chain lengths bring the non- polar nature, various combinations of polar mobile phases such as water, acetonitrile, methanol and isopropanol are applied for hydrophobic partitioning. The separation of gangliosides is based on the hydrophobic ceramide moiety which interacts with the non- polar stationary phase. A gradient of strong solvents such as acetonitrile, methanol or isopropanol are employed to elute gangliosides depending on the ceramide chain.

Additives such as ammonium acetate or ammonium formate are used. Separation of gangliosides employing C18 and C8 columns is based on the hydrophobicity characteristics of the ceramide moiety irrespective of the carbohydrate chain content

[69]–[75].

Specialty reverse-phase columns were investigated in the present work for separation of gangliosides. The specialty columns investigated were mixed-mode having both reverse-phase and ion-exchange ligands phenyl-hexyl and pentafluoro phenyl reverse phase columns.

Hydrophilic interaction chromatography: Hydrophilic interaction chromatography (HILIC) is a variation of normal phase liquid chromatography technique in which the stationary phase is polar and the mobile phase is an organic polar solvent

(>50%) for retention and separation of hydrophilic analytes. The water in the mobile phase adsorbs on to the polar stationary phase forms an aqueous rich layer which assists in the partition of the polar analytes. Analytes preferentially partition between the aqueous rich stationary phase and highly organic mobile phase based on its polarity and solvability. It is a complex mechanism of interactions of the analyte and stationary phase which includes hydrophilic partitioning, hydrogen-hydrogen bonding, electrostatic interaction and Van der Waals interactions contributing towards the separation.

Chromatography of gangliosides by HILIC separates gangliosides based on carbohydrate chain differences, differentiating gangliosides differing in ceramide composition[70],

[72], [73], [76], [77].

The advantages of HILIC are: 1) retention of highly polar analytes; 2) elution order depending on the polarity of the analytes; 3) improved signal/noise ratio (S/N) with

MS detection; 4) low back-pressures as mobile phase is highly organic solvent and 5) compatibility with solid phase extraction, helping in reducing sample preparation time.

Other separation techniques used for ganglioside analysis are gas chromatography and capillary electrophoresis. Gas chromatography analysis have been performed on trimethyl silyl derivatized gangliosides[78]. Limited volatility, degradation and a prolonged derivatization process makes GC analysis a less desirable technique in gangliosides analysis. Capillary electrophoresis (CE) is coupled to Electrospray

Ionization (ESI)-MS to analyze gangliosides separated per their electrophoretic mobility.

Efficient CE buffer and their relative incompatibility to mass spectrometric detection, as well as laborious sample preparation, has limited the CE technique in the determination of gangliosides[79]–[81].

2.1.3 Ganglioside analysis by mass spectrometry

Ganglioside analysis has been done with various detectors such as UV spectroscopy, colorimetry and flame ionization detector. All of them are limited in their selectivity and sensitivity. Mass spectrometry (MS) has revolutionized the analysis of gangliosides. A variety of MS ion sources such as Fast Atom Bombardment (FAB),

Matrix Assisted Laser Desorption Ionization (MALDI), Electrospray Ionization (ESI) and chip based ESI have been used in studies of gangliosides. High-resolution mass analyzers such as Fourier transform ion cyclotron resonance (FTICR)[82], high capacity ion trap

(HCT) Trap[83], and quadrupole time-of-flight (Q-TOF)[84], [85], [86] have been employed for structural characterization of gangliosides. Currently, Orbitrap MS has gained popularity in high resolution ganglioside studies[76], [77]. Ion mobility spectrometry has also been employed for the ionization and detection of higher sialylated

53 gangliosides showing great promise[87], [88]. Triple quadrupole (QQQ), quadrupole and ion-trap (Q-Trap) mass analyzers are highly preferred for quantitative ganglioside analysis[70]–[73], [89].

Recently LC-MS/MS methods have been developed to profile and quantify gangliosides in brain tissue[90]–[93]. LC-MS/MS is a superior technique in ganglioside analysis with respect to: 1) profiling of hundreds of gangliosides in a single run which helps in elucidating the heterogeneous complexity of gangliosides; 2) Having high specificity through its molecular weight basis of detection; 3) having molecular structure determination via fragmentation analysis; 4) having high sensitivity compared to TLC which quantifies at the micro-molar compared to the pico-molar level of LC-MS/MS in cerebrospinal fluid samples; 5) having superior quantification performance determination of individual gangliosides in terms of accuracy and precision.

Electrospray ionization (ESI)-triple quadrupole mass spectrometry: In all the

LC-MS/MS experiments in this chapter, electrospray ion source (ESI) (Figure 2.1) is used along with triple quadrupole (QqQ) mass analyzer (Figure 2.2). HPLC is interfaced to ESI-triple quadrupole mass spectrometer. A triple quadrupole mass spectrometer identifies the analyte based on their mass-to-charge ratio. The analyte is introduced into the mass spectrometer after it is transformed into gas phase ions by the ion source.

ESI is the ion source which produces ions by applying a strong electrical field, under atmospheric pressure, to the liquid flowing through it (see Figure 2.1). The field induces a charge accumulation at the end of the capillary which will break the effluent mobile phase to form highly charged droplets. These are transferred into vacuum chambers through a series of lenses (skimmers or cones) with very small orifices. Ion

54 desolvation is performed by a counter flow of heated dry gas called curtain gas (N2 gas) under the strong electrical field, deformation of the droplets occurs until the charged molecules undergo desorption. The gas phase ions are separated according to the mass- to-charge ratio (m/z).

Figure 2.1: Diagram of ESI ion source

Figure 2.2: Diagram of a triple quadrupole mass analyzer

A triple quadrupole mass analyzer is utilized in these experiments. The gas phase ions produced by the ion source enter into the first quadrupole, Q1 of mass spectrometer which separates the ions according to mass-to-charge ratio using the stability of the trajectories in oscillating electrical fields. In the collision cell, q2, ions are fragmented

(daughter ions) and these are again analyzed in the third quadrupole, Q3, as it does in the

Q1. The daughter ions are detected and converted into electrical signals proportional to the number ions of each daughter ions. There are several scan modes of the triple quadrupole mass spectrometer as given below.

Precursor ion scan (parent ion scan): An ion of selected mass-to-charge ratio

(m/z) is selected by Q1 and collided with in the collision cell (q2) resulting in fragments which are identified by scanning the third quadrupole, Q3.

Product ion scan (daughter ion scan): An ion of particular mass-to-charge ratio is selected in the third quadrupole (Q3) and the parent ions are scanned in the first quadrupole (Q1) for those that produce that particular product ion.

Multiple reaction monitoring (MRM): An ion is selected in the first quadrupole

(Q1) is fixed, selecting a particular m/z parent ion) and a particular fragment product ion of that parent ion is selected by the third quadrupole (Q3) is fixed, selecting a particular m/z fragment of the parent ion). This product ion is monitored by the detector for quantification of the original parent ion. Thus, the selectivity and sensitivity of the ion is increased many fold.

Current chromatographic techniques separating gangliosides use either reversed- phase chromatography (C18, C8), separating gangliosides based on the hydrophobicity of the ceramide moiety[70]–[75], [89] or hydrophilic interaction chromatography (HILIC -

57 amino, silica), separating gangliosides based on the polar sugar head.[70], [72], [76],

[77], [91]. Both of these techniques suffer from co-eluting gangliosides. For reversed- phase techniques, GM, GD, GT and GQ gangliosides with the same ceramide moiety co- elute, while for HILIC techniques gangliosides with different ceramide moieties but the same carbohydrate group co-elute. A phenyl-hexyl reversed-phase chromatography technique developed in the present work combines the best of the HILIC separation of polar head classes and the best of the reversed-phase separation of ceramide moiety, in addition to accomplishing isomeric separation. A sensitive and specific LC-MS/MS method has been developed that quantifies individual gangliosides. Results are also presented for experiments separating gangliosides on a mixed-mode and other reversed- phase columns with unique chemistries. The resolution of gangliosides is low with this special chemistry HPLC columns.

2.2 Experimental Section

2.2.1 Materials

Monosialo ganglioside GM1 (NH4+salt), disialo ganglioside GD1a (NH4+salt), trisialo ganglioside GT1b (NH4+salt), tetrasialo ganglioside GQ1b (NH4+salt) and N-

+ omega-CD3-Octadecanoyl monosialo ganglioside GM1 (NH4 salt) standards of bovine brain were obtained from Matreya LLC (State College, PA, USA). Methanol optima

LC/MS grade, acetonitrile LC/MS grade and isopropanol LC/MS grade were from Fisher

Scientific (Fair Lawn, NJ, USA). Chloroform, 99.8+ %, ACS reagent was from Across

Organics (New Jersey, USA). Ammonium acetate, ≥ 99.5 %, ammonium formate, ≥

99.95 %, and ammonium hydroxide, ≥ 99 %, were from Sigma-Aldrich (St. Louis, MO,

USA). HPLC grade water was from a Barnstead Nanopure water purification system

58 from Thermo Scientific (West Palm Beach, FL, USA) was used for mobile phase, standards and sample preparation.

2.2.2. Preparation of standard solutions

A stock solution of gangliosides (GM, GD, GT, GQ, 1 mg/mL) in 50% methanol in water was prepared. All stock solutions were filtered through 0.2 μm PTFE membrane syringe filters from EMD Millipore (Billerica, MA, USA). All the working ganglioside standard solutions in the range of 78 – 5000 ng/mL (this range is specific to phenyl-hexyl column experiments) were prepared from the serial dilutions of the stock solution (1 mg/mL) with 50% methanol in water. All the stock and working solutions were stored at

-20 °C.

2.2.3 HPLC Instruments, Columns and Experimental Parameters

2.2.3.1 Optimized mixed-mode HPLC runs

The analysis was performed using an LC-MS/MS-MS system in which a

Shimadzu UPLC system (Columbia, MD, USA) was interfaced to an AB Sciex QTrap

5500 mass spectrometer (Framingham, MA, USA). The UPLC system consisted of a

Prominence DGU-20A3R inline degasser, two LC-30AD pumps, a SIL-30AC auto sampler, and a CBM-20A controller. 10 μL of each prepared sample was injected on to a mixed-mode column, a Scherzo SM C18 HPLC column (100mm × 2.1 mm, 5µ particle size, 3000 A pore size) from Imtakt USA. An optimized linear gradient of mobile phase of A, deionized water, and mobile phase B, 100% methanol at 0.2 ml/min, was developed.

The gradient program is given in Figure 2.3.

2.2.3.2 Optimized HILIC HPLC runs

HPLC analysis was performed on a Waters Alliance 2695 quaternary pump system (Milford, MA, USA). The chromatographic separation was performed on a hydrophilic interaction liquid chromatography (HILIC) columns, amino-propyl ligand

(50 × 1 mm, 3 um particle size) with amino-propyl ligand guard column (5 × 1 mm, 3 um particle size) from Phenomenex (Torrance, CA, USA) and IMTAKT USA (Portland, OR,

USA). A volume of 20 µl of the working standard was injected with an auto sampler at 4 oC. A multiple linear gradient of mobile phases (A: 83% acetonitrile, B: 83% acetonitrile and 5 mM ammonium acetate, C: 50% acetonitrile and 50 mM ammonium acetate) were used with Phenomenex HILIC HPLC column. A multiple linear gradient of mobile phases (A: 83% acetonitrile, B: 83% acetonitrile and 5 mM ammonium acetate, C: 50% acetonitrile and 50 mM ammonium acetate, D: 50% acetonitrile) was used with Imtakt

HILIC HPLC column. See Table 2.1 A&B for multi-step gradient of mobile phases on both the HILIC HPLC columns. The flow rate was 0.1 mL/min. Mobile phases were filtered through 0.45 µm membrane filters from Millipore (Billerica, MA, USA).

2.2.3.3 Optimized phenyl-hexyl HPLC runs

A Shimadzu Nexera X2 LC-30AD binary pump system (Columbia, MD, USA) with high pH capability was used for the LC analysis. A volume of 20 μl of samples were injected with a Shimadzu auto sampler (SIL-30AC), also having high pH capability, was maintained at 4 0C. The column oven (CTO-20A) was maintained at 40 0C. The column

Table 2.1.A: Time gradient on Phenomenex HILIC HPLC column using the following mobile phases - A% - 83% ACN; B% - 83% ACN + 5mM Ammonium Acetate; C% - 50% ACN + 50mM Ammonium Acetate

Time (mins) A% B% C%

0-2 100 0 0

2-10 0 100 0

10-13 0 0 100

13-24 0 0 100

24-26 100 0 0

26-35 100 0 0

Table 2.1.B: Time gradient on Imtakt HILIC HPLC column using the following mobile phases - A% - 83% ACN; B% - 83% ACN + 5mM Ammonium Acetate; C% - 50% ACN + 50mM Ammonium Acetate; D% - 83% ACN

Time (mins) A% B% C% D%

0-2 100 0 0 0

2-10 0 100 0 0

10-13 0 0 100 0

13-24 0 0 100 0

24-24.1 0 0 0 100

24.1-32 0 0 0 100

32-35 100 0 0 0

35-40 100 0 0 0

61 was a phenyl-hexyl reversed-phase HPLC column (Waters XBridge BEH 2.1 x 50 mm,

3.5 μm, 110 A0 pore size) along with a Waters vanguard cartridge (XBridge BEH, phenyl-hexyl, 2.1 x 50 mm, 3.5 μm, 110 A0 pore size). Mobile phase A was 0.1% ammonium hydroxide (NH4OH) in water while mobile phase B was 0.1% NH4OH in methanol. The gradient was 25% - 100% B in 20 min and maintained at 100% B for an additional 8 min. The total run time was 30 min, which included equilibration for 10 min.

The flow rate was 0.2 mL/min. The above are the final optimized solvent system conditions. The mobile phase solvents were filtered through 0.2 μm membrane filters from EMD Millipore (Billerica, MA, USA).

2.2.4 Mass spectrometer instruments and settings

2.2.4.1 Absciex QTrap mass spectrometer

Absciex QTrap 5500 mass spectrometer was employed for ganglioside studies with mixed-mode HPLC column and initial ganglioside studies on phenyl-hexyl HPLC column. The ESI ion source was used in negative mode. The optimized ionization MS parameters were as follows: ion spray voltage (-4500 V), declustering potential (60 V), entrance potential (12 V), collision energy (individual setting), collision exit potential (10

V), and nebulizer temperature (450 ºC). Data acquisition and processing were done using the Analyst software (version 1.6.1). The multiple reaction monitoring (MRM) transitions were doubly charged gangliosides as follows m/z 772.0 → 290.0 (GM1 d18:1-18:0),

786.0 → 290.0 (GM1 d18:1-20:0), m/z 917.6 → 290.1 (GD1a d18:1-18:0), m/z 931.5 →

290.0 (GD1a d18:1-20:0), m/z 708.7 → 290.0 (GT1b d18:1-18:0) and m/z 717.80 →

290.0 (GT1b d18:1-20:0) monitoring the sialic acid daughter ion.

2.2.4.2 Waters triple quadrupole mass spectrometer

A triple quadrupole Waters Micromass Quattro Ultima instrument (Milford, MA,

USA) with an electrospray ionization (ESI) source was employed for ganglioside studies with HILIC HPLC columns. The ESI source in a negative ionization mode was optimized as follows: capillary potential (-3 KV), cone potential (-40 V), source temperature (150 oC), desolvation temperature (300 oC), cone gas flow (144 L/hr) and desolvation gas flow

(756 L/hr). The triple quadrupole analyzer was optimized as follows: ion energy 1 and 2 were 1.0 and 3.0 eV respectively, collision energy was 70 eV, entrance and exit potential were 120 V and multiplier potential was -650 V. The multiple reaction monitoring

(MRM) transitions were m/z 1545.3 → 290.3 (GM1 d18:1-18:0), m/z 1573.3 → 290.3

(GM1 d18:1-20:0), m/z 917.5 → 290.1 (GD1a d18:1-18:0), m/z 931.6 → 290.0 (GD1a d18:1-18:0), m/z 708.7 → 290.0 (GT1b d18:1-18:0) and m/z 717.80 → 290.1 (GT1b d18:1-18:0) monitoring the dehydrated sialic acid fragment.

2.2.4.3 Shimadzu triple quadrupole mass spectrometer

The following was used for all the phenyl-hexyl HPLC column studies, except for the optimization studies discussed in section 2.3.3.1. The Shimadzu chromatographic system was interfaced to a Shimadzu QQQ LCMS-8040 instrument (Columbia, MD,

USA) via a ESI ion source. The ESI ion source was operated in the negative ionization mode. Parameters were optimized as follows: nebulizer gas flow and drying gas flow were 2.5 L/min and 12 L/min, respectively. The desolvation line (DL) temperature was

250 0C and the heat block temperature was 400 0C. The interface voltage was 3.5 KV.

Argon gas was used for the collision ion dissociation with 230 kPa pressure. Collision energy and pre-bias voltages were optimized for individual multiple reaction monitoring

(MRM) transitions (Table 2.2). GM and GD were doubly-charged [M-H]2-, GT were triply-charged [M-H]3-, and GQ were quadruple charged [M-H]4- parent ions, while singly-charged negative daughter ions were selected.

Table 2.2: Individual ganglioside mass spectrometer parameters Gangliosid Q1 Pre-bias Q3 Pre-bias Ceramide group M (m/z) Charge state MRM CE e class (V) (V)

1 GM11 d18:1-18:0 1545.80 [M-H]2₋ 771.90>289.90 24.0 45.0 28.0

2 GM1 d18:1-20:0/d20:1-18:0 1573.90 [M-H]2₋ 785.95>289.85 30.0 36.0 14.0

3 GM2 d18:1-18:0 1382.80 [M-H]2₋ 690.40>289.90 32.0 45.0 28.0

4 GM2 d18:1-20:0/d20:1-18:0 1410.80 [M-H]2₋ 704.40>289.90 32.0 45.0 28.0

5 GM3 d18:1-18:0 1179.40 [M-H]1₋ 1179.40>290.0 48.0 53.0 18.0

6 GD1 d18:1-16:0/d16:1-18:0 1809.00 [M-H]2₋ 903.50>289.90 22.0 47.0 10.0

7 GD1 d18:1-18:0 1836.80 [M-H]2₋ 917.40>289.90 32.0 49.0 23.0

2 8 GD1 d18:1-20:0/d20:1-18:0 1865.00 [M-H] ₋ 931.50>289.90 38.0 41.0 10.0

2₋ 9 GD1 d18:1-22:0/d22:1-18:0 1893.00 [M-H] 945.50>289.90 32.0 49.0 29.0 10 GT1 d18:1-16:0/d16:1-18:0 2100.80 [M-H]3₋ 699.30>289.90 28.0 30.0 10.0

11 GT1 d18:1-18:0 2128.80 [M-H]3₋ 708.60>289.90 28.0 30.0 10.0

12 GT1 d18:1-20:0/d20:1-18:0 2156.40 [M-H]3₋ 717.80>289.90 40.0 33.0 10.0

13 GQ1 d18:1-16:0/d16:1-18:0 2379.20 [M-H]4₋ 593.80>289.90 32.0 27.0 17.0

14 GQ1 d18:1-18:0 2417.60 [M-H]4₋ 603.40>289.90 32.0 27.0 17.0

15 GQ1 d18:1-20:0/d20:1-18:0 2446.40 [M-H]4₋ 610.60>289.90 32.0 26.0 10.0

16 GQ1 d18-22:0/d22:1-18:0 2475.20 [M-H]4₋ 617.8>289.90 32.0 27.0 17.0

Note: Designation of all gangliosides in this dissertation does not imply a single molecular species but could include any isomer of the particular ganglioside class (GM1 in this case). Thus, GM1 d18:1-18:0 also includes GM1 d20:1-16:0, GM1 d18:0-18:1, GM1 d20:0-16:1, etc.

2.3 Results and discussion

2.3.1 LC-MS/MS of gangliosides using mixed-mode column

Since gangliosides have both a hydrophobic and an ionic nature, a stationary phase with both ionic and hydrophobic ligands may result in unique separation capabilities in ganglioside analysis. A mixed-mode column was employed for ganglioside separation for the first time. The stationary phase of the SCHERZO SM-C18 column mixed mode column (IMTAKT) evaluated in the present work consists of C18, weak-cation and weak-anion exchange ligands. Each stationary phase ligand would selectively interact with either the hydrophobic portion (C18 ligand) or the ionic/polar portion (anion- exchange ligand binding, cation-exchange repelling) of the ganglioside. The mobile phase components/characteristics of organic modifier strength, buffer concentration and pH would modulate interaction with the stationary phase ligands. Given in Figure 2.3 are the chromatograms of gangliosides differing in the ceramide carbon number for each class of ganglioside (GT, GD, GM). As can be seen in Figure 2.3 there is co-elution of the gangliosides of each particular class (for example the two GT gangliosides have similar retention times), as well as peak tailing for most of the ganglioside peaks. Also, there is significant overlap of the GTs and GDs peaks. Addition of additives such as ammonium formate resulted in the ion suppression. A gradient in pH was done with ammonium formate buffered with formic acid but the results were not desirable.

Figure 2.3: Ganglioside standard mixture (1000ng/mL) separation on SCHERZO SM-C18 column mixed mode column (IMTAKT). A gradient of methanol from 10% methanol in water to 90% methanol in water for 8 minutes followed by 90% methanol in water for 2 minutes. Followed by equilibration for 10 minutes is performed. From top to bottom GT1 d18:1/C18:0, GT1 d18:1/C20:0, GD1 d18:1/C18:0, GD1 d18:1/C20:0 and GM1 d18:1/C18:0, GM1 d18:1/C20:0,

2.3.2. LC-MS/MS of gangliosides using HILIC HPLC column

Separation of GM, GD, and GT ganglioside classes on an amino-propyl HILIC column was achieved. As shown in Figure 2.4 gangliosides elute in the order from low polarity/ionic charge to high polarity/ionic charge: monosialo (GM) < disialo (GD) < trisialo (GT). Although the GM, GD and GT classes of gangliosides are baseline separated, the subclasses for any particular class consisting of ganglioside differing in ceramide moieties, which would have different hydrophobicity based on the number of carbons, all co-elute in their particular class peak. Although these different subclass gangliosides can be identified by the specific MRM determinations, co-eluting peaks are not ideal due to suppression effects.

Initially the run started with 83% ACN, where the water forms a polar layer on the stationary phase as the ligands are polar, water retains on the stationary phase forming a partitioning layer. A gradient to 83% ACN + 5mM ammonium acetate allows the partitioning of the gangliosides into the stationary phase. The gradient decreases the organic content of the mobile phase and increases its buffer strength (50% ACN + 50mM ammonium acetate) causing the elution of the analytes, as the mobile phase becomes higher in polarity and thus can compete with the stationary phase for the polar gangliosides.

Figure 2.4: Total ion chromatogram of gangliosides standard mixture (GM, GD, GT) separation on Phenomenex HILIC HPLC column. The gradient of multiple mobile phases from 6.00 min to 24 min is reported.

Ganglioside precipitation: In HILIC chromatography, acetonitrile is a commonly used solvent as it is aprotic and does not interact with the water unlike the other polar organic compounds (alcohols) which compete with water to solvate the stationary phase composition. Hence in standards solution, 83% ACN was used to match the mobile phase composition (83% ACN). We observed precipitation of gangliosides in stock and higher working ACN standard solutions, as well as its absorption onto the glass vials. To overcome this problem, 83 % methanol was used in the preparation of all stock and working standard solutions. The change in the solvent solved the problem as no precipitation was observed either in the stock solution or in working standard solutions.

Retention time shift: Reproducible chromatography is a requirement for any analytical method. In HILIC chromatography, the kinetics at the stationary phase are slow which results in higher equilibration times compared to reverse-phased chromatography. A minimum of 20 column volumes is usually required for optimum conditions. In our method, we equilibrated column for nearly 30 volumes with mobile phase-A (83% ACN). However, we observed retention shifts of nearly 2.5 minutes (see

Figure 2.5) with GM1 gangliosides after 30-40 runs on the same column done. We inferred this retention shift due to the insufficient equilibration of the column even after optimal washing. This may be due to the equilibration with 83% ACN which takes long time to wash off end run ammonium acetate buffer (50% ACN + 50mM ammonium acetate), with the ammonium acetate being retained on the stationary phase. To eliminate this problem, the column was first washed with 20 column volumes of 50% ACN after a run to better wash off the ammonium acetate (less organic content in the mobile phase), followed by another 20 column volumes of 83% ACN. Reproducibility of the retention

Figure 2.5: GM1 ganglioside retention time shift on HILIC column (Phenomenex HPLC column) from one run to another prior to the optimization of re-equilibration step after the run. The gradient of multiple mobile phases from 6.00 min to 24 min is reported.

71 time was then obtained. However, long-term reproducibility of the GM1 retention time was found to be problematic, even with the optimized re-equilibration procedure, as the retention time was seen to progressively decrease with age of the HPLC column.

Secondary Interactions: Although hydrophilic interaction is the primary mechanism in HILIC, electrostatic interactions with the packing material also plays a significant role affecting retention of ionized analytes. When silica is used as the support for the amino ligand, as is the case for the HILIC columns used in the present work, there is ionic interactions of ionic analyte with the base silica. To minimize these secondary interactions a HILIC column (Imtakt) having the same dimensions and ligand (propyl amine) as the Phenomenex column, but instead of silica base has polymer end-capping support, thus reducing secondary interactions. See figure 2.6, a significant shift in the retention time proves the effect of secondary interactions. The HILIC Imtakt column has improved resolution and increased sensitivity compared to Phenomenex column.

16.85 min GD1 (di-sialo)

9.26 min GM1 (mono-sialo)

18.19

GT1 (tri-sialo)

10.58

GM1

Figure 2.6: Total ion chromatogram of gangliosides standard mixture (GM, GD, GT) separation on Imtakt HILIC HPLC column. The gradient of multiple mobile phases from 6.00 min to 24 min is reported.

2.3.3. LC-MS/MS determination of gangliosides using phenyl-hexyl column

2.3.3.1 Optimization of chromatography using phenyl-hexyl column

Solvent Selectivity: The phenyl-hexyl ligand has reversed-phase characteristics, having a structure of a six-carbon atom linear chain with a phenyl ring on the terminal carbon atom. To this non-polar stationary phase, gradients employing different organic modifiers (methanol, acetonitrile and isopropanol) in water were used to elute the gangliosides based on the hydrophobic partitioning. Isocratic and gradient time programs were tested for better separation. Water and methanol solvents gave optimized results

(Figure 2.7).

Additives: Though gangliosides were separated based on the number of sialic acids (retention time increases with decreasing number of sialic acid groups in the ganglioside) there is significant peak overlap due to peak tailing. Ammonium acetate was added to the mobile phase to see if it could improve peak shape and/or sensitivity. As the concentrations of ammonium acetate increased from 1 mM, 5 mM and 10 mM the signal for the gangliosides diminished more. The results were similar with ammonium formate.

The phenomena could be a result of electrospray ionization where the buffer components ionization dominates the analytes ionization. This is supported by the greater analyte ion suppression as the concentration of ammonium acetate/ammonium formate were increased in the mobile phase solvent system.

Given that the tailing was most prominent for GD and GT in Figure 2.7, it is thought that the tailing may be caused by a heterogeneous charge state of the GD and GT

Figure 2.7: Phenyl-hexyl separation of gangliosides with methanol gradient, 45% methanol in water to 95% methanol in water for 7 minutes. GM1 (d18:1-18:0 in blue and d18:1-20:0 in red) were eluted between 7.5 – 9.0 min. GD1 (d18:1-18:0 in green and d18:1-20:0 in grey) were eluted between 4.5 min to 10 min, GT1 (d18:1-18:0 in sky blue and d18:1-20:0 in pink) were eluted between 2.50 min to 9.0 min

75 gangliosides resulting from heterogeneous de-protonation of the multiple sialic acids.

Thus, depending on the pH, there could be 0, 1, 2, 3 or 4 deprotonated sialic acids in the

GT gangliosides and at any particular pH there could be several charge forms of the GT gangliosides that are chromatographed leading to tailing. Given this hypothesis, a volatile base, ammonium hydroxide, was added to the mobile phase to deprotonate all sialic acids so that there would be only one ionic form for a particular class of gangliosides.

Chromatographing one ionic form, and not multiple forms, of any individual ganglioside molecule would address the tailing issue, if heterogeneity of the sialic acid deprotonation state was the cause of the tailing. Thus, with addition of ammonium hydroxide to the mobile phase, GQ molecules would only have a -4-charge state (all four sialic acids deprotonated), GT molecules would only have a -3 charge state (all three sialic acids deprotonated), GD molecules would only have a -2 charge state (both sialic acids deprotonated) and GM molecules would only have a -1 charge state (the one sialic acid is deprotonated).

LC-MS/MS analyses of gangliosides with a phenyl-hexyl column employing mobile phases with different concentrations of ammonium hydroxide were evaluated.

Given in Figures 2.8 and 2.9 are chromatograms of gangliosides employing gradients of increasing methanol with mobile phases that were 5 mM and 10 mM ammonium hydroxide in the mobile phase throughout the gradient, respectively.

GD1a-1836 Max. 4.9e4 1.4e5

1.0e5

5.0e4 Intensity, cps Intensity,

1.0e4 2.0 4.0 6.0 8.0 10.0 12.0 14.0

Time, min Figure 2.8: Phenyl-Hexyl separation of gangliosides with methanol gradient with 5mM ammonium hydroxide in 45%methanol to 5mM ammonium hydroxide in 95% methanol for 7 minutes. GM1 (d18:1-18:0 and d18:1-20:0 in black), GD1 (d18:1-18:0 and d18:1-20:0 in red) and GT1 (d18:1-18:0 and d18:1-20:0 in green)

Max. 9.9e4 cps. GD1a-1836 1.8e5

GM1-1545

GD1a-1864

1.0e5 GT1b-2155

Intensity, cps Intensity,

GT1b-2127 GM1-1573

5.0e4

1.0e4 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Time, min Figure 2.9: Phenyl-Hexyl separation of gangliosides with methanol gradient, 10mM additive ammonium hydroxide in 45%methanol to 10mM ammonium hydroxide in 95% methanol for 7 minutes. GM1 (d18:1-18:0 and d18:1-20:0 in black), GD1 (d18:1-18:0 and d18:1-20:0 in red) and GT1 (d18:1-18:0 and d18:1- 20:0 in green)

It is seen that tailing of the GT and GD gangliosides occurs with 5 mM ammonium hydroxide mobile phases (Figure 2.8). However, chromatography employing 10 mM ammonium hydroxide gave narrow peaks, resolving individual molecular species (having different carbon number in the ceramide moiety) within each class of ganglioside, as seen in Figure 2.9.

2.3.3.2 Phenyl-hexyl separation of ganglioside classes with fine separation of hydrophobic subclasses

Standards of GQ, GT, GD and GM gangliosides were pooled and chromatographed on a Waters Xbridge BEH phenyl-hexyl HPLC column. The chromatogram in Fig 2.10 shows baseline separation of most of the 18 gangliosides that were monitored. Of note is that all the hydrophobic subclasses of each class of gangliosides (GQ, GT, GD, GM) were well-resolved while being clustered together in a retention window specific for each ganglioside class. Thus, GD subclasses were separated into four major subclasses: GD d18:1-16:0/d16:1-18:0, GD d18:1-18:0, GD d18:1-20:0/d20:1-18:0 and GD d18:1-22:0/d22:1-18:0 with retention times of 13.84,

14.47, 15.05 and 15.67 min, respectively, in order of increasing hydrophobicity. The retention times of isomers 1, 2 and 3 of GD d18:1-18:0 were noted to be 14.18, 14.47 and

14.73 min, respectively, with isomer 3 eluting as a shoulder peak on major isomer 2. A similar isomer pattern was seen for GD d18:1-20:0/d20:1-18:0, having retention times of

14.77, 15.05 and 15.31 min.

Figure 2.10: Phenyl-hexyl separation of 1000 ng/mL mixed gangliosides (GM, GD, GT and GQ) standards with a methanol gradient of 0.1% ammonium hydroxide in 25% methanol to0.1% ammonium hydroxide in 85% methanol. The mobile phase pH is 10.71. GM1 (d18:1-18:0 and d18:0-20:0 in red), GD1 (d18:1-18:0 and d18:0-20:0 in yellow), GT1 (d18:1-18:0 and d18:0-20:0 in purple) and GQ (d18:1-18:0 and d18:0-20:0 in blue).

A total of eight disialo gangliosides were identified. There are three major subclasses for the trisialo gangliosides: GT d18:1-16:0/d16:1-18:0, GT d18:1-18:0 and GT d18:1-

20:0/d20:1-18:0, with retention times of 11.36, 11.92 and 12.67 min, respectively, again in order of increasing hydrophobicity. The isomers 1 and 2 of both GT d18:1-18:0 and

GT d18:1-20:0/d20:1-18:0 eluted at 11.92, 12.67 and 12.67, 13.43 min, respectively. In total, five trisialo gangliosides were identified. For monosialo gangliosides, GM d18:1-

18:0 and GM d18:1-20:0/d20:1-18:0 subclasses eluted at 17.60 and 18.10 min, respectively. GM d18:1-18:0 had isomers 1 and 2, which eluted at 17.37 and 17.60 min, respectively. A total of three GMs was identified. Finally, GQ d18:1-18:0 and GQ d18:1-

20:0/d20:1-18:0 subclasses eluted at 10.49 and 11.26 min, respectively.

The separation of ganglioside species on phenyl-hexyl HPLC column combines the advantage of the sialic acid class separation of HILIC techniques, which are based on the ionic characteristic of the ganglioside, with that of C18/C8 reversed-phase HPLC techniques, separating the hydrophobic subclasses, which are based on the ceramide moiety of the gangliosides. The elution order of the ganglioside classes for the phenyl- hexyl technique is in the order of decreasing number of sialic acid groups, mimicking

HILIC by being able to separate the ganglioside classes, although in reverse retention order compared to the HILIC technique.

Thus, the GQ gangliosides elute the earliest in the gradient, given that they have the highest negative charge and thus are the least retained of the ganglioside classes on the non-polar phenyl-hexyl column. Conversely, the GM gangliosides elute the latest in the gradient, being the strongest retained, given that it has the highest hydrophobic-to- ionic character of the ganglioside classes studied. Furthermore, within each class of gangliosides, there is further separation of the hydrophobic subclasses based on the number of carbon atoms in the ceramide moiety. This is the reversed–phase separation aspect of the phenyl-hexyl column. Thus, the combined separating capabilities of the phenyl-hexyl column technique separates the gangliosides into clusters of gangliosides, with each cluster consisting of resolved gangliosides having different ceramide groups but are of the same sialic acid class. This is helpful over the HILIC and the C18/C8 reversed-phase techniques. The HILIC technique does not have the ability to separate the ceramide subclasses of a particular ganglioside class (such as GD d18:1-20:0/d20:1-18:0 and GD d18:1-22:0/d22:1-18:0), while the C8/C18 columns cannot separate the sialic acid classes of gangliosides with the same ceramide moiety (such as GD d18:1-18:0 and

GM d18:1-18:0).

The phenyl-hexyl technique was also able to separate ganglioside isomers. Exact identification of the ganglioside isomers will require future high-end mass spectrometry studies employing an orbitrap mass spectrometer, which has better capabilities in doing structural studies compared to triple quadrupole mass spectrometry, which was used in the present work.

Temperature: The column oven temperature has been tested for a better peak shape. At

40 C, better peak shapes were seen.

2.3.3.3 Phenyl-hexyl column calibration curves, limit of quantification and limit of detection

The LC-MS/MS method employing a phenyl-hexyl column will use ganglioside standards to quantify gangliosides in mouse brain and retinal tissue samples (Chapters 3 and 4). Calibration plot of GM1 ganglioside (other ganglioside calibration curves are reported in appendix A1-A7) is given in Figure 2.11. The calibration range of all the gangliosides were in the range of 30.0 – 10,000 ng/mL. The theoretical concentrations were corrected to their individual gangliosides (for details, see chapter 5) as all the standards were a mixture of gangliosides varying in their ceramide chain. The correction factors were used to correct the concentrations of individual gangliosides. The slopes of eight major gangliosides vary due to the differences in the individual ionization efficiencies. The slope values of individual gangliosides are as follows: GM1 (1545.8) –

1216.9, GM1 (1573.9) – 1382.9, GD1 (1836.8) – 9250.4, GD1 (1865.0) - 7206.7, GT1

(2128.8) – 4277.6, GT1 (2156.4) – 4653.6, GQ1 (2417.6) – 1524.2 and GQ1 (2446) –

1458.2 shows GD1, disialo gangliosides have greater ionization among the ganglioside family. The ceramide chain length has little influence on ionization efficiency as observed by the slopes of individual ganglioside standard curves. The ionization efficiency trend is GD > GT > GQ > GM, as the sialic acid number increases the ionization efficiency decreases except the GM1 ganglioside which has the least ionization. The square of regression coefficient (r2) of all the gangliosides were >0.995.

The lower limit of quantitation (LLOQ, S/N = 10) and the limit of detection (LOD, S/N =

3) for the GM1, monosialo gangliosides (1545.80) were 38.92 ng/mL & 19.46 ng/mL respectively.

A highly sensitive and selective LC-MS/MS technique using a phenyl-hexyl column, which achieved base-line separation of all major individual gangliosides is reported for the first time. This highly selective LC-MS/MS technique separating gangliosides combines the sialic acid class separation capability of HILIC with the hydrophobic separation capability of C8/C18 reverse-phase HPLC. A high-resolution mass spectrometer in combination with the phenyl-hexyl separation will be a gold standard technique for structural elucidation and quantification of gangliosides. This new technique will have wide application to study the changes of ganglioside distribution in the neurological diseases such as Parkinson’s disease, Alzheimer’s disease, Huntington’s disease, glaucoma and others.

GM1-1545.80 1800000

1600000 y = 1216.9x + 12119 R² = 0.9993 1400000

1200000

1000000

800000 Peak Area Peak 600000

400000

200000

0 0 200 400 600 800 1000 1200 1400 Concentration (ng/mL)

Figure 2.11: Calibration standard curve of GM1 ganglioside, 1545.80

CHAPTER III

ISOLATION AND QUANTIFICATION OF GANGLIOSIDES FROM MOUSE

BRAIN AND RETINAL TISSUES

3.1 Background

Gangliosides are a large family of acidic glycosphingolipids. Isolation of gangliosides from complex tissues is necessary and challenging. Quantitative isolation of lipids free of non-lipid contaminants must ideally be achieved before the lipid analysis.

This is important in the study of biological functions of gangliosides and their role in various diseases. Various liquid-liquid extraction techniques have been developed to isolate gangliosides. Tettamanti and co-workers have extensively worked on the

86 glycosphingolipids isolation with a tetrahydrofuran/phosphate buffer solvent system[94].

Fredman and Svennerholm optimized it to separate gangliosides by including a large proportion of water in the extraction solution, along with methanol and chloroform[95].

A similar modified ganglioside isolation technique was developed by Schnaar to isolate gangliosides.

The extraction protocol by Schnaar. and a modified Svennerholm and Fredman ganglioside isolation protocol are evaluated in the present work. The latter procedure, while using many of the steps of the established procedure, incorporates modifications that were worked out in the present work. The modified procedure utilizes both liquid- liquid extraction and solid-phase extraction. The sample preparation procedures were evaluated by a LC-MS/MS analytical method using phenyl-hexyl column. Validation studies include evaluation of repeatability of chromatographic peak retention times, extraction recovery, and matrix effect. The quantitative distribution of gangliosides in adult mouse brain and retinal tissues was also found.

3.2 Experimental

3.2.1 Materials

Adult mouse brain (n=1 normal mouse, grey and white matter) and pooled mouse retinal tissue samples (n= 3 normal mice, 6 retinas pooled) were used for all studies.

Chloroform anhydrous, ≥99%, contains 0.5-1.0% ethanol as stabilizer, from Sigma

Aldrich (St. Louis, MO USA), methanol Optima, LC-MS/MS grade 99.99%, Fisher

Scientific, Potter- Elvehjem glass-teflon homogenizer tube, Waters C18 SPE cartridges.

3.2.2 Schnaar ganglioside isolation procedure

The Schnaar protocol for isolating polar lipids, was evaluated in the present for ganglioside isolation. The procedure is described below.

Normal adult mice (n=3) retinal tissues were pooled. The phase partitioning of gangliosides was performed with polar water and methanol and non-polar chloroform solvents. The pooled tissues were weighed and placed in a Potter- Elvehjem glass-teflon homogenizer tube, adding four volumes of ice-cold water to it. The sample was homogenized with ten strokes on ice and methanol was added to achieve a methanol

(MeOH)-to-water (H2O) ratio of 8:3. The suspension was mixed thoroughly, brought to ambient room temperature (RT) and chloroform (CHCl3) was added at half the methanol volume to achieve a CHCl3: MeOH: H2O ratio of 4:8:3. The suspension was mixed and centrifuged at 1200xg for 15 min at RT. The supernatant was collected and the cell debris discarded. The supernatant volume was measured and 0.173 volumes of water was added to the supernatant. This was mixed vigorously (vortexed) and centrifuged at 1200xg for

15 min. The upper (polar) phase was collected, discarding the lower phase. The upper phase was evaporated at 45°C under a stream of nitrogen to near dryness (not complete dryness). Water was added drop-wise to the residue until it dissolved, keeping the volume of water used to a minimum. Micro dialysis tubes (250µL) from GE Healthcare Life

Sciences were used to perform the desalting. The volume was dialyzed against water for

≥24 h using 1000-MW cut-off dialysis tubing, recovering the aqueous solution and evaporating to near dryness (not complete). This was reconstituted 200 µL with

83%ACN and a 20 µL volume aliquot was injected onto the LC-MS/MS.

3.2.3 Modified Svennerholm and Fredman ganglioside isolation procedure

The Svennerholm and Fredman ganglioside isolation protocol is a standard extraction procedure with phase separation of non-polar and polar solvents [95]. This established technique was changed by replacing a dialysis step with a solid phase extraction as well as the solid phase extraction step for desalting resulted in reduced time and improved recovery. The details of the modified Svennerholm and Fredman procedure are given below.

Liquid-Liquid Extraction: Fresh brain or retinal tissue samples from mice were collected, combined and weighed. To a 1 mg of the tissue sample, 50 μL of 40% methanol in water (ice cold) was added prior to homogenization. The tissue sample was homogenized in a tissue sample grinder (2 mL) for 10 min in an ice bath. A volume of

100 μL of homogenized tissue was subject to a biphasic liquid extraction employing a modified Svennerholm - Fredman lipid extraction process;[95] as summarized below. To every 1 volume of tissue sample MeOH (ice cold), CHCl3 (ice cold) and water (ice cold) were added in the ratio 5.4:2.7:1. The mixed solution was vortexed for 2 min and incubated at room temperature on a shaker for 30 min. After this it was centrifuged at

2000 g for 30 min and the supernatant collected. The pellet was resuspended in the

MeOH, CHCl3, and water in the above ratios, followed by vortexing, incubation, centrifugation, and collection of supernatant, as described above. The supernatants were pooled and the pellet discarded. A volume of 260 µL of water was added to the pooled supernatant for solvent partitioning. Gentle inversion (3 to 4 times) was done followed by centrifugation for 30 min at 2000 g. The upper-phase containing the gangliosides was collected and saved. A volume of 100 μL of 0.01 M potassium chloride (KCl) in water

89 was added to the bottom chloroform phase, which was then incubated, and centrifuged, as given above. The supernatant from this was then collected, combined with the previously collected upper phase and further processed with solid phase extraction (SPE), as described below.

Solid phase extraction (SPE): SPE was done using Waters C18 SPE cartridges

(10 μm particle size, 3 mL in volume) to desalt the extracted samples. SPE cartridges were conditioned with 3 mL of 100% methanol, twice, followed by 2 mL of 30% methanol in water, twice. The following steps of SPE were done under gravitational force, as vacuum elution resulted in poor recoveries. The extracted sample was loaded and the collected effluent was again passed through the cartridge to bind any unretained gangliosides on the first pass. Two 3 mL volumes of 30% methanol in water were then passed through SPE cartridges. Gangliosides were then eluted with two portions of 2 mL100% methanol, with the eluent being collected. The eluent finally was near dried under vacuum and reconstituted with 200 µl of 50% methanol in water for LC-MS/MS analysis.

3.2.4 LC-MS/MS Instrumentation and Parameters

Characterization of the modified Svennerholm and Fredman procedure and quantification of gangliosides in the mouse retinal and brain samples were done on the same LC-MS/MS instruments and hexyl-phenyl column, using the optimized procedures and settings as described in Chapter 2. The injection volume was 12 µl for all studies.

The MRM transitions used are also given in Chapter 2. For the Schnaar procedure, the

HILIC LC-MS/MS technique described in Chapter 2 was used.

3.2.5 Recovery Studies

For the modified Svennerholm and Fredman recovery studies, 5 μL of 0.5 mg/mL deuterated mono-sialo ganglioside (GM1-D3) internal standard was added to the 100 μL homogenized sample prior to the extraction process. LC-MS/MS determinations were done comparing the peak area of GM1-D3 of the prepared sample with that of a GM1-D3 standard prepared in 50% methanol in water. A replicate of three per run was performed.

3.2.6 Matrix Interference effect

Gangliosides from mouse brain grey and white matter and pooled mouse retinal tissue samples extracted by the modified Svennerholm and Fredman ganglioside isolation procedure were serially diluted with reconstitution solvent 50% methanol in water and its

MRM peak area versus concentration (based on the value of the undiluted sample) plot

(n=3) the compared with the standard calibration plot (n=3) of the respective gangliosides.[77] Calibrators were made up in solution of 50% Superimposable plots of the calibration and the serial diluted sample plots confirm a minimum matrix effect.

3.2.7 Determination of gangliosides concentration in mouse brain and retinal samples

Samples were prepared according to the modified Svennerholm and Fredman ganglioside isolation procedure. MRM transitions for 16 gangliosides were monitored as given in the Results and Discussion section. Gangliosides in the tissue samples (n=3 for each tissue) were quantified using external calibration of standard calibrators (n=3 calibrations) in Calibrators were from 78 to 5000 ng/mL.

3.3 Results and Discussion

3.3.1 Schnaar isolation recovery studies

Mouse retinal tissues samples were extracted using the Schnaar procedure. The recovery of standard gangliosides from the retinal tissue of normal mice was poor 0.8% for GM1 (1573) and 9% for GM1 (1545).

The Schnaar procedure was further characterized with respect to the steps at which gangliosides loss occurred. Four stages in the isolation procedure were identified as potential steps where loss of gangliosides could take place. A GM1 ganglioside standard of 1 µg/mL was processed using the Schnaar procedure the collected sample was infused on Bruker Daltonics HCT ion trap mass spectrometer and MS signal compared to that of a 1 µg/mL GM1 ganglioside standard calibrator not taken through the extraction procedure. The samples were collected at the following stages:1) after homogenization, 2) after first supernatant collection, 3) before dialysis and 4) after dialysis. The percent loss of GM1 ganglioside was calculated by comparing the ion intensities with standard GM1. With sample 1, 2, 3 and 4 the recovery was approximately

95%, 70%, 40% and 11%. At all these stages, GM1 loss is observed which contributes to the poor recovery. A loss of approximately 90% was observed by the sample procedure stage after dialysis. The steps at which significant loss occurred are after the supernatant collection and micro dialysis steps.

Since the Schnaar procedure was assessed to have poor recovery a modified ganglioside isolation technique was developed, with the results of the characterization studies described below.

3.3.2 Modified Svennerholm and Fredman ganglioside isolation procedure recovery and matrix effect studies

Recovery: Recovery studies assessing the performance of the modified Svennerholm and Fredman ganglioside isolation procedure were performed on the adult mouse brain (grey and white matter) and pooled mouse retinal tissue samples.

Figure 3.1: Percentage recovery of deuterated monosialo gangliosides GM1-D3 (d18:1-18:0 m/z 1550 in blue, d18:1-20:0 in orange) evaluating the Svennerholm and Fredman procedure

This was assessed by evaluating the recovery of the deuterated monosialo ganglioside

(GM1-D3), a surrogate analyte, which was included in the sample preparation. The results are summarized in Figure 3.1. The combined liquid-liquid and solid-phase extraction achieved substantially greater recovery. compared to the Schnaar technique.

The recovery of GM1-D3 d18:1-18:0 and GM1-D3 d18:1-20:0/d20:1-18:0 in brain grey matter, brain white matter and pooled retinal tissues were 105.6%, 91.8%, 101.6% and

105.7%, 127.7%, 106.8% respectively (Figure 3.1). The recovery of other higher sialic gangliosides in tissue studies was assessed based on deuterated monosialo gangliosides,

GM1-D3 because of unavailability of each individual deuterated gangliosides.

Matrix effect: Gangliosides extracted from mouse brain grey and white matter and pooled mouse retinal tissue samples were diluted and compared with the standards of respective gangliosides.[77] The response of diluted samples of pooled retinal tissue

(triangles) is compared to the standard calibration plot (dots) for gangliosides GM1, GD1,

GT1 and GQ1 in Figures 3.2, 3.3, 3.4 and 3.5 respectively. The tissue sample responses were identical with the standard calibrators indicating a negligible matrix effect for the

GN1, GD1, and GT1 gangliosides (the d18:1-18:0 and d18:1-20:0/d20:1-18:0 gangliosides for each of these classes) However, tetrasialo ganglioside (GQ d18:1-

20:0/d20:1-18:0) showed matrix suppression for the retinal tissue sample as the diluted retinal tissue responses are lower than the standard calibration plot (Figure 3.5). In contrast, the brain GQ gangliosides shown a little matrix effect (data not shown).

3.3.3 Application to brain and retinal tissues

The chromatograms of gangliosides from adult mouse brain (grey and white matter) and pooled mouse retinal tissues are shown in Figure 3.6. Given in Table 3.1 are the retention times of 16 gangliosides monitored. Of those, the ones found, along with the determined concentrations using external calibrators were listed in Table 3.2. There are also minor isomers of the gangliosides monitored that were found in the mice tissue samples (data not shown). For mouse brain, white matter 14 of the 16 gangliosides monitored were found, with four additional isomers of the monitored gangliosides also found (Figure 3.6A). For mouse retinal tissue (Figure 3.6B), 12 of the 16 gangliosides monitored were found, with an additional five isomers noted. For mouse brain grey matter, 14 of the 16 gangliosides monitored were found, with four additional isomers of the monitored gangliosides also found, (Figure 3.6C). A varied distribution of the ganglioside family is seen in the different tissue. The distribution of all the major individual gangliosides in adult mouse brain (grey and white matter) and pooled mouse retinal tissues are shown in Figure 3.7.

A: GM1 d18:1-18:0 1600000 1400000 1200000 1000000 800000

Peak Peak Area 600000 400000 200000 0 0 500 1000 1500 2000 2500 3000 concentration ng/mL

B: GM1 d18:1-C20:0 1600000

1400000

1200000

1000000

800000

Peak Peak Area 600000

400000

200000

0 0 500 1000 1500 2000 2500 3000 concentration ng/mL

Figure 3.2: Matrix effect assessment by comparing the responses of GM1 d18:1-18:0 (A) and GM1 d18:1-20:0/d20:1-18:0 (B) standards (solid circle) vs diluted pooled retinal tissue samples (triangle)

A: GD1 d18:1-18:0 10000000 9000000 8000000 7000000 6000000 5000000 4000000 Peak Peak Area 3000000 2000000 1000000 0 0 500 1000 1500 2000 2500 3000 Concentration ng / mL

B: GD d18:1-20:0/d20:1-18:0 6000000

5000000

4000000

3000000 Peak Peak Area 2000000

1000000

0 0 200 400 600 800 1000 1200 1400 Concentration ng / mL

Figure 3.3: Matrix effect assessment by comparing the responses of GD1 d18:1-18:0 (A) and GD1 d18:1-20:0/d20:1-18:0 (B) standards (solid circle) vs diluted pooled retinal tissue samples (triangle)

A: GT d18:1-18:0 3500000

3000000

2500000

2000000

1500000 Peak Peak Area 1000000

500000

0 0 500 1000 1500 2000 2500 3000 Concentration ng / mL

B: GT d18:1-20:0/d20:1-18:0 14000000

12000000

10000000

8000000

6000000 Peak Peak Area 4000000

2000000

0 0 1000 2000 3000 4000 5000 6000 Concentration ng / mL

Figure 3.4: Matrix effect assessment by comparing the responses of GT1 d18:1-18:0 (A) and GT1 d18:1-20:0/d20:1-18:0 (B) standards (solid circle) vs diluted pooled retinal tissue samples (triangle)

A: GQ d18:1-18:0 1400000

1200000

1000000

800000

600000 Peak Peak Area 400000

200000

0 0 500 1000 1500 2000 2500 3000 Concentration ng / mL

B: GQ d18:1-20:0/d20:1-18:0 1800000 1600000 1400000 1200000 1000000 800000

Peak Peak Area 600000 400000 200000 0 0 500 1000 1500 2000 2500 3000 Concentration ng / mL

Figure 3.5: Matrix effect assessment by comparing the responses of GQ1 d18:1-18:0 (A) and GQ1 d18:1-20:0/d20:1-18:0 (B) standards (solid circle) vs diluted pooled retinal tissue samples (triangle)

100

Table 3.1: Repeatability of all major gangliosides are shown based on their retention times

Ganglioside Brain grey Brain white Pooled retinal Ceramide group M (m/z) class matter matter tissue 1. GM1 d18:1-18:0 1545.80 17.6 ± 0.00 17.6 ± 0.02 17.6 ± 0.02 2. GM1 d18:1-20:0/d20:1-18:0 1573.90 18.1 ± 0.00 18.1 ± 0.02 18.1 ± 0.02 3. GM2 d18:1-18:0 1382.80 17.6 17.6 ND 4. GM2 d18:1-20:0/d20:1-18:0 1410.80 18.1 18.1 ND 5. GM3 d18:1-18:0 1179.40 17.6 17.6 17.6 6. GD1 d18:1-16:0/d16:1-18:0 1809.00 13.8 13.8 13.8 7. GD1 d18:1-18:0 1836.80 14.5 ± 0.04 14.5 ± 0.06 14.5 ± 0.06 8. GD1 d18:1-20:0/d20:1-18:0 1865.00 15.1 ± 0.05 15.1 ± 0.07 15.1 ± 0.05 9. GD1 d18:1-22:0/d22:1-18:0 1893.00 15.7 15.7 15.7 10. GT1 d18:1-16:0/d16:1-18:0 2100.80 11.4 11.4 11.4 11. GT1 d18:1-18:0 2128.80 11.9 ± 0.02 12.1 ± 0.12 12.0 ± 0.13 12. GT1 d18:1-20:0/d20:1-18:0 2156.40 12.7 ± 0.12 12.7 ± 0.12 12.7 ± 0.13 13. GQ1 d18:1-16:0/d16:1-18:0 2379.20 ND ND ND 14. GQ1 d18:1-18:0 2417.60 10.5 ± 0.28 10.5 ±0.29 10.5 ± 0.26 15 GQ1 d18:1-20:0/d20:1-18:0 2446.40 11.3 ± 0.36 11.3 ± 0.38 11.3 ± 0.34 16 GQ1 d18-22:0/d22:1-18:0 2475.20 ND ND ND D: Detected ND: Not detected n=3 over two to three-day period

101

Table 3.2: Ganglioside distribution in adult mice brain (white and grey matter) and retinal tissues (ng/mg)

Ganglio- Brain grey Brain white Pooled Ceramide group M (m/z) side class matter matter retinal tissue 1. GM1 d18:1-18:0 1545.80 528.7 1271.8 358.2 2. GM1 d18:1-20:0/d20:1-18:0 1573.90 258.8 798.1 332.5 3. GM2 d18:1-18:0 1382.80 D D ND 4. GM2 d18:1-20:0/d20:1-18:0 1410.80 D D ND 5. GM3 d18:1-18:0 1179.40 D D D 6. GD1 d18:1-16:0/d16:1-18:0 1809.00 D D D 7. GD1 d18:1-18:0 1836.80 951.3 229.7 68.2 8. GD1 d18:1-20:0/d20:1-18:0 1865.00 179.7 177.5 31.5 9. GD1 d18:1-22:0/d22:1-18:0 1893.00 D D D 10. GT1 d18:1-16:0/d16:1-18:0 2100.80 D D D 11. GT1 d18:1-18:0 2128.80 1133.2 1120.2 222.7 12. GT1 d18:1-20:0/d20:1-18:0 2156.40 273.9 795.3 89.3 13. GQ1 d18:1-16:0/d16:1-18:0 2379.20 ND ND ND 14. GQ1 d18:1-18:0 2417.60 320.0 805.0 112.5 15. GQ1 d18:1-20:0/d20:1-18:0 2446.40 135.6 544.3 84.0 16. GQ1 d18-22:0/d22:1-18:0 2475.20 ND ND ND

D: Detected ND: Not detected

102

Figure 3.6: MRM chromatograms of total gangliosides in A: Brain white matter (10x dilution), B: Pooled retinal tissue (10x dilution), C: Brain grey matter (10x dilution)

103

Ganglioside Concentrations (ng/mg)

Brain grey matter Brain white matter Pooled retinal tissue

1600

1400

1200

1000

800

600

400

200

-200

Figure 3.7: Ganglioside distribution in adult mice brain (white and grey matter) and retinal tissues (ng/mg)

104

The precision of the peak retention of the gangliosides in adult mouse brain grey and white matter, and pooled mouse retinal tissue, n=3 over a two to three period is given in

Table 3.1. The percentage coefficient of variation (%CV) for GM, GD, GT gangliosides was < 1.05% and < 3.4% GQ gangliosides. The retention times of the ganglioside peaks are stable compared to the retention time shift observed in the HILIC chromatography of gangliosides (Chapter 2).

105

CHAPTER IV

GLAUCOMA MICE AGE STUDIES: APPLICATION TO LC-MS/MS ANALYSIS

ON PHENYL-HEXYL HPLC COLUMN AND OPTIC NERVE STUDIES

4.1 Background: Glaucoma and control mice

Primary open angle glaucoma is an age related common glaucoma among old adults. Though the pathophysiology of the disease is not clear, the blockage to the trabecular meshwork leads to the obstruction of aqueous humor secreted by ciliary body into the anterior chamber of the eye. This obstruction results in the increase of intraocular pressure (in most cases) which leads to the damage of the optic nerve head and retinal

106 tissue. The retinal axons of the optic nerve head and retinal tissue are made up of retinal ganglion cells which are lost during this damage. The retinal ganglion cells have gangliosides in their membranes, which can act as neuroprotective agents of the retinal tissue. Hence loss of gangliosides may play a crucial role in the glaucoma disease.

Studies of ganglioside changes during the glaucoma progression would thus assess whether gangliosides have a functional role in the disease. In the current studies, DBA/2J glaucoma model mice, compared with C57BL/6J control model mice, were employed to assess ganglioside changes at various stages.

4.1.2. Preliminary data

Glaucoma is caused by the degeneration of retinal ganglion cells (RGCs), a group of nerve cell bodies, secondary to axon damage at the optic nerve head. RGCs express

Pharmaceutical Sciences, Northeast Ohio Medical University) hypothesizes that GM1 ganglioside up-regulation/uptake by the astrocyte cells is a crucial stage in the retinal gliosis, being responsive as a compensatory repair mechanism to glaucoma progression.

The possibility that astrocytes express GM1 gangliosides to compensate for the RGCs loss during glaucoma degeneration has implications for use as a therapeutic agent to treat.

RGCs in normal retina express GM1 gangliosides as confirmed by their efficient and specific uptake of cholera toxin-B (CTB). CTB binding to GM1 gangliosides occurs through specific protein-sugar interactions. In the DBA/2J mouse, a glaucoma model mouse, astrocytes, which do not express GM1 gangliosides, have been shown to bind

107

CTB, indicating the presence of GM1 ganglioside (see Chapter 1, Figure 1.17).

Dysfunctional RGCs are identified by the accumulation of phosphorylated neuro- filament[51,52] (see Chapter 1, Figure 1.17). Other indicators of dysfunctional RGCs are lack in action potential[53] and dendritic arborization [54]. Astrocytes with bound CTB were not observed in aged normal mice however were expressed/accumulated by/in astrocytes in aged glaucoma model mice (see Chapter 1, Figure 1.18). Studies performed at Dr. Inman’s lab also show that decreased GM1 in RGCs is matched by an increase in

GM1 in astrocytes, which could be either an uptake of GM1 from deteriorating RGCs or could be self-expression of GM1 in astrocytes in response to degeneration.

4.1.3 Glaucoma age studies

Preliminary ganglioside-CTB binding studies show changes in the GM1 ganglioside landscape in the retinal tissue. Though CTB binding affinity is high with

GM1, CTB has variable affinity with other gangliosides classes [63]. Accounting for changes of all the major gangliosides will provide a better insight on the role of gangliosides in glaucoma progression. Glaucomatous mice, DBA/2J and the glaucoma control mice, C57BL/6J are preferred as the animal models for these glaucoma age studies. Both DBA/2J and C57BL/6J mice were selected at various ages such as young adults (2 and 3months), middle age adults (6 and 10 months) and old adults (12 months).

Retinal ganglion call axons from the retinal tissue bundle form the optic nerve. These optic nerve tracts are projected into the mid-brain and connect to the superior colliculus.

Hence in these studies ganglioside changes were studied in retinal tissue and the superior colliculus. Retinal tissues and superior colliculus of all the mice were collected for ganglioside extraction. In these glaucoma age studies, using the LC-MS/MS analytical

108 method developed on phenyl-hexyl column as discussed in Chapter 2.2 is applied for ganglioside distribution studies. Optic nerve studies were conducted to establish the damage to the optic nerve head and estimate the glaucoma progression in the mice.

4.1.4 Glaucoma and control mice

Glaucoma mice: DBA/2J inbred mice strain is widely used in a large number of research areas, including cardiovascular biology, neurobiology and sensorineural research. DBA/2J mice show a low susceptibility to developing atherosclerotic aortic lesions (20 to 350 um2 atherosclerotic aortic lesions /aortic cross-section) following 14 weeks on an atherogenic diet (1.25% cholesterol, 0.5% cholic acid and 15% fat). DBA/2J mice have extreme intolerance to alcohol and morphine. DBA/2J strain also exhibit a high frequency for hearing loss by the early adulthood around the age of three to four weeks. By the age two to three months the loss is severe. DBA/2J strain possesses three recessive alleles that cause progressive cochlear pathology initially affecting the Organ of

Corti. There is high incidence of calcareous pericarditis, and calcified lesions of the testes, tongue and skeletal muscle. This strain is among the least responsive to phytohemagglutinin.

DBA/2J is a popular choice for age related studies, specifically glaucoma. This strain develops progressive eye abnormalities with age progression. The glaucoma disease in aging DBA/2J mimics human hereditary glaucoma. Two alleles contribute to the eye phenotype, GpnmbR150X and Tyrp1isa; both are present in DBA/2J mice. The common symptoms are iris pigment dispersion, iris atrophy, anterior synechia (adhesion of the iris to the cornea), and elevated intra-ocular pressure (IOP), which is similar to primary open angle glaucoma. The disease onset is seen in adult mice, between the age of

109 three and four months with 56% of females and 15% of males. At this age, there are signs of iris pigment epithelium loss and trans illumination of the peripheral iris. By age of six to seven months of age, all mice demonstrate significant widespread transillumination and thickening of the iris border. By nine months of age, IOP is highly elevated with pressures higher in females (mean: 20.3 +/-79; 1.8 mmHg) compared to males (mean:

16.2 +/-79; 1.4 mmHg). Immunohistopathology of the retinal tissues reveal the loss of retinal ganglion cells, cholinergic amacrine, and GABAergic cells.

Glaucoma control mice: C57BL/6J, glaucoma control mice have characteristics that are often contrasted with those of the inbred strain DBA/2J. This strain provides a genetically matched control for DBA/2J. This coisogenic strain has a functional allele of Gpnmb. Homozygous mice do not develop elevated intraocular pressure or glaucoma, although they exhibit a mild iris stromal atrophy (ISA). The inbred strain DBA/2J is homozygous for the glaucoma-related GpnmbR150X and Tyrp1isa mutations.

Single retinal tissue studies were performed for the first time as the glaucoma progression is specific to individual retinal tissue. The correlation studies are performed with right and left retinal tissues along with the cross-linked right and left superior colliculus at each age group. In the present work, correlation studies were performed assessing the ganglioside changes between the glaucoma and glaucoma control mice at each specific age. The optic nerve count at various ages of both glaucoma and control mice were also performed to assess the disease progression.

4.2 Experimental section

For the current age studies, 2 months (2), 3 months (1), 10 months (2) and 12 months (3) aged DBA/2J glaucomatous mice and 3 months (2), 10 months (2) and 12

110 months (2) aged C57BL/6J glaucoma control mice were used (number of mice is indicated by the number in the parentheses). Dr. Inman group (NEOMED) performed the animal handling. The mice were sacked and fresh retinal and superior colliculus tissues were collected and processed for ganglioside extraction (as described in chapter 3, the modified Svennerholm ganglioside isolation technique). The optic nerves were processed using the optic nerve protocol (described in section 4.2.1) for further studies.

4.2.1 Optic nerve embedding for optical microscopy

Mice were overdosed with sodium pentobarbital (300mg/kg, Beuthanasia-D) and euthanized. Individual retinal tissue along with the optic nerve were removed. Later, retinal tissues and optic nerves were separated and the removed optic nerves were immersion fixed in 2% paraformaldehyde and 2.5% glutaraldehyde in 0.1M sodium cacodylate buffer, pH 7.4. Immersed optic nerves were stained with 2% osmium. This process is referred to as osmication. Nerves were dehydrated in graded ethanol from

50%, 70%, 95% and 100% combined with incubation and shaking. The nerve was removed from 100% EtOH and placed in scintillating glass vial with of 1:1 mixture of ethanol and propylene oxide. After incubation and shaking, propylene oxide was replaced by 1:1 mixture of propylene oxide/PolyBed and incubated at 4°C with shaking overnight.

This process is referred as embedding. This process was repeated and incubated for 4 hours. Followed by it, increasing concentrations of Araldite 502/ PolyBed® 812

(Polysciences, Inc) in propylene oxide was done. Nerves were incubated in 100%

Araldite 502/ PolyBed® 812 under vacuum, switching embedding medium three times until being placed in an embedding mold and then cured over three successive days in an oven at 35, 40, then 60°C. Araldite-Poly/Bed mix (without DMP-30) can be stored in 10

111 or 20cc syringes at -20C for up to two months, or at 4C in an airtight syringe for two weeks.

Embedded Nerve Cutting: Molds are trimmed under a dissecting scope using a razor blade into a four-sided pyramid with the nerve in the center. Semi-thin sections of approximately 1-2 µm are then cut using an ultramicrotome and a diamond knife. The sections are collected on a slide using distilled water and dried in a 60°C oven until all water has evaporated. The slides are then cooled at room temperature.

Staining: PPD: Slides containing sections are immersed in 1% paraphenylenediamine (PPD) in a 1:1 mixture of methanol and 2-propanol for 28 minutes. The slides are then rinsed in two consecutive mixtures of equal parts methanol and 2-propanol for 1 to 2 minutes each. The slides are then rinsed a final time in 100% ethanol for one minute. Allow slides to air dry.

Mounting: Slides with sections are then cover slipped using Permount mounting media. It is important to remove all excess Permount from under cover slip. You may find it helpful to first place a drop of Xylene on the slide before adding Permount to it.

The slides are then allowed to dry overnight. It may be necessary to scrape the cover slip with a razor blade to remove excess or stray dried Permount.

Using a 100X oil aim and the optical fractionator approach within Stereo

Investigator (MBF Bioscience, VT) running on a Leica DM4 B upright microscope optic nerve count was performed. Roughly 30 sampling sites (10%) of each optic nerve cross- section were counted. The coefficient of error (Gunderson) was 0.05 or below, ensuring sufficient sampling rate.

112

4.2.2 LC-MS/MS determination of gangliosides of mouse retinal and superior colliculus samples

Ganglioside analysis by LC –MS/MS on retinal and superior colliculus samples were done after sample preparation by the modified Svennerholm and Fredman ganglioside isolation procedure (see Chapter 3) on the same LC-MS/MS instruments and hexyl-phenyl column, using the optimized procedures and settings as described in

Chapter 2. The injection volume was 20 µL for all samples. The mass spectrometer parameters and MRM transitions used are also given in Chapter 2.

4.3 Results and Discussion

4.3.1 LC-MS/MS analysis on DBA/2J and C57BL/6J mice strains

A comparative study of major gangliosides (GM, GD, GT, GQ) was done between eye retinal tissue and superior colliculus of young and old age groups. Along with it comparative studies were performed between glaucomatous and glaucomatous control mice strain. A multi model age study was done by correlating the LC-MS/MS ganglioside composition results and optic nerve studies to establish the GM1 ganglioside role in the glaucoma disease progression.

Gangliosides were quantitated using the LC-MS/MS analytical method developed on

Phenyl-hexyl column (Figure 4.1 – 4.4). Right and left retinal and superior colliculus tissues were analyzed individually. Based on the optic nerve studies, there is minor difference in the number of axons from right and left. Hence, these individual ganglioside concentrations are grouped, averaged and reported. A total of eight major gangliosides which include 36 and 38 carbon chain ceramide moieties of tetrasialo-, trisialo-, disialo- and monosialo gangliosides. Table 4.1 summarizes the amounts of individual

113 gangliosides in both young/old and glaucoma/control mice retinal tissues. And Table 4.2 summarized the amounts of individual gangliosides in both young/old and glaucoma/control mice superior colliculus brain tissue. Figure 4.5 & 4.6 shows a comparison of ganglioside changes in superior colliculus of glaucoma and control mice.

The comparison studies show insignificant changes in ganglioside superior colliculus from glaucomatous mice to control mice both at young adult and old adult stages. Figure

4.7 & 4.8, represents the retinal gangliosides comparison between glaucoma and control mice both young and old age mice. The ganglioside changes in the young are nearly equal in glaucoma and control mice. But in old mice studies of retinal tissues, a significant amount of all the major individual gangliosides upregulation is observed. All the 36-carbon ceramide moiety GM1, GD1, GT1 and GQ1 shows double increment compared to the control of the respective ganglioside components. The statistical difference between the old and young mice (control vs disease) in retinal and superior colliculus tissues were calculated as two independent sample sets with a null hypothesis.

The p-value is found to be > 0.05.

114

Figure 4.1: Chromatograms of ganglioside profiling in glaucoma age studies, D2 glaucoma mice 2 months: From top to bottom: Left eye; Right Eye; Left Superior colliculus; Right Superior colliculus

115

Figure 4.2: Chromatograms of ganglioside profiling in glaucoma age studies, D2 glaucoma mice 12 months: From top to bottom: Left eye; Right Eye; Left Superior colliculus; Right Superior colliculus

116

Figure 4.3: Chromatograms of ganglioside profiling in glaucoma age studies, D2G control mice 2 months: From top to bottom: Left eye; Right Eye; Left Superior colliculus; Right Superior colliculus

117

Figure 4.4: Chromatograms of ganglioside profiling in glaucoma age studies, D2G control mice 12 months: From top to bottom: Left eye; Right Eye; Left Superior colliculus; Right Superior colliculus

118

Table 4.1: Concentrations of individual major gangliosides of retinal tissues grouped into young (2 and 3 months) and old age (10 and 12 months) mice.

Concentration (ng/mg) Animal GM 1 GM1 d18:1- GD1 GD1 d18:1- GT1 GT1 d18:1- GQ1 GQ1 d18:1- d18: C20:0/d20:1- d18:1/C C20:0/d20:1- d18:1/C18 C20:0/d20:1 d18:1/C18: C20:0/d20: 1/C1 C18:0 18:0 C18:0 :0 -C18:0 0 1-C18:0 8:0

Control mice OLD 12mo- 37.90 11.03 23.82 12.74 54.32 18.60 25.60 7.73 D2G-2R 12mo- 31.45 9.57 15.79 9.06 36.71 12.53 17.57 5.82 D2G-2L 12mo- 25.56 17.06 25.54 15.13 58.57 21.37 26.84 10.18 D2G-1R 12mo- 28.62 22.59 37.37 21.01 102.82 34.23 44.12 15.44 D2G-1L 10mo- 20.35 6.83 13.73 8.27 44.94 12.89 21.68 6.08 D2G-2R 10mo- 20.89 12.58 9.84 7.85 29.26 9.98 14.89 5.13 D2G-2L 10mo- 34.82 13.03 16.70 9.30 32.97 8.99 14.83 4.84 D2G-1R

119

10mo- 32.86 15.89 24.97 12.80 55.06 15.38 22.65 6.86 D2G-1L AVG 29.06 13.57 20.97 12.02 51.83 16.75 23.52 7.76 STD 5.98 4.58 8.18 4.17 21.80 7.67 8.87 3.31

YOUNG 3mo- 43.95 19.10 22.89 13.50 61.24 21.56 21.55 8.42 D2G-2R 3mo- 42.21 15.38 20.47 11.62 52.57 18.75 23.09 7.60 D2G-2L 3m D2G 29.71 14.46 17.82 9.71 49.17 13.40 20.24 7.17 1R 3mo- 88.33 60.88 82.01 44.70 192.23 66.77 80.59 28.37 D2G-1L AVG 51.05 27.45 35.79 19.88 88.80 30.12 36.37 12.89 STD 22.21 19.38 26.74 14.39 59.87 21.36 25.55 8.95

Glaucom a mice OLD 12mo-D2- 84.64 47.02 60.84 37.70 131.05 58.21 57.53 21.19 1R 12mo-D2- 28.20 9.33 17.21 9.71 54.14 16.80 21.84 6.88 1L 12mo-D2- 54.33 26.92 25.28 17.42 46.28 13.45 23.14 8.22 2R

120

12mo-D2- 33.26 17.34 16.34 10.69 27.36 8.28 12.97 4.82 2L 12mo-D2- 29.85 17.20 16.76 13.56 33.73 13.14 13.13 7.26 3R 12mo-D2- 20.29 12.44 11.28 8.72 21.04 6.95 10.02 9.60 3L 10mo-D2- 17.54 14.02 9.87 8.12 28.48 9.08 14.01 5.39 2R 10mo-D2- 25.26 26.54 31.94 19.84 83.29 35.50 39.64 14.29 2L 10mo-D2- 160.3 106.24 171.10 95.04 357.09 133.98 149.82 59.81 1R 2 10mo-D2- 51.68 33.09 50.81 29.28 116.37 43.87 56.51 22.58 1L AVG 50.54 31.01 41.14 25.01 89.88 33.93 39.86 16.00 STD 41.35 27.27 46.22 25.08 96.22 37.22 40.35 15.78

YOUNG 3mo-D2- 29.69 13.21 27.39 13.12 56.42 16.41 24.11 7.48 1R 3mo-D2- 48.73 33.83 49.34 24.61 116.62 35.73 47.63 15.01 1L 2mo-D2- 39.35 14.92 26.71 13.75 74.11 21.46 27.02 9.78 2R 2mo-D2- 56.58 12.96 42.86 16.40 95.75 21.06 28.52 8.35 2L 2mo-D2- 42.93 16.90 37.24 16.16 102.03 26.29 40.17 10.93 1R

121

2m D2-1L 38.37 23.11 49.46 22.43 120.22 32.79 44.72 13.29 AVG 42.61 19.16 38.83 17.74 94.19 25.62 35.36 10.81 STD 8.45 7.39 9.31 4.30 22.63 6.79 9.17 2.65

122

Table 4.2: Concentrations of individual major gangliosides of superior colliculus grouped into young (2 and 3 months) and old age (10 and 12 months) mice

Concentration (ng/mg) Animal GM1 GM1 d18:1- GD1 GD1 d18:1- GT1 GT1 d18:1- GQ1 GQ1 d18:1- d18:1/ C20:0/d20:1- d18:1/C C20:0/d20:1- d18:1/C C20:0/d20:1- d18:1/C C20:0/d20:1- C18:0 C18:0 18:0 C18:0 18:0 C18:0 18:0 C18:0

Control mice OLD 12mo- 33.91 17.02 134.95 78.77 505.93 286.71 188.10 100.35 D2G-2R 12mo- 43.00 18.29 82.86 56.04 322.70 189.19 153.51 87.59 D2G-2L 12mo- 45.53 15.78 62.32 42.44 277.48 176.14 111.70 70.04 D2G-1R 12mo- 71.87 21.33 94.97 53.15 423.44 267.83 114.10 64.85 D2G-1L 10mo- 57.99 6.83 71.17 51.95 326.06 177.25 189.57 114.54 D2G-2R 10mo- 48.12 16.57 71.47 47.83 349.14 196.60 244.14 135.20 D2G-2L 10mo- 36.07 16.61 112.72 71.08 488.51 268.91 370.71 192.50 D2G-1R

123

10mo- 47.98 26.01 117.03 81.33 513.55 294.92 273.54 151.92 D2G-1L AVG 48.06 17.30 93.44 60.32 400.85 232.19 205.67 114.62 STD 11.41 5.08 24.27 13.75 87.70 48.48 82.03 40.74

YOUNG 3mo- 39.44 20.78 109.40 70.99 459.64 235.30 294.71 147.40 D2G-2R 3mo- 38.57 19.87 136.65 83.95 623.20 302.40 591.17 264.01 D2G-2L 3m D2G 43.01 10.14 106.86 61.66 402.29 210.42 320.47 164.98 1R 3mo- 108.2 27.27 268.98 151.56 1060.77 566.28 205.29 92.60 D2G-1L 7 AVG 57.32 19.52 155.47 92.04 636.47 328.60 352.91 167.24 STD 29.46 6.12 66.57 35.26 258.03 141.29 144.05 61.92

Glaucom a mice OLD 12mo- 74.10 38.05 177.98 116.47 670.80 373.27 148.76 75.68 D2-1R 12mo- 36.29 21.38 135.59 89.12 539.99 324.59 289.08 146.85 D2-1L 12mo- 84.08 33.14 203.95 132.89 889.36 478.00 260.55 138.02 D2-2R

124

12mo- 60.84 20.92 70.85 50.17 331.53 178.50 272.63 138.40 D2-2L 12mo- 64.38 23.69 115.13 78.99 415.00 242.38 193.12 122.93 D2-3R 12mo- 56.89 15.24 91.09 53.66 335.41 174.40 434.48 236.22 D2-3L 10mo- 43.77 15.39 50.11 35.35 181.72 81.60 104.21 54.83 D2-2R 10mo- 59.91 20.35 51.93 39.97 172.06 83.30 90.58 37.73 D2-2L 10mo- 259.8 75.73 236.13 187.43 755.02 313.50 311.69 122.89 D2-1R 0 10mo- 84.11 28.11 60.30 44.36 172.18 67.88 54.12 23.61 D2-1L AVG 82.42 29.20 119.31 82.84 446.31 231.74 215.92 109.72 STD 60.92 16.98 63.63 47.03 244.46 131.77 112.65 60.10

YOUNG 3mo-D2- 18.60 9.58 162.61 84.25 676.33 288.30 422.20 173.95 1R 3mo-D2- 40.07 16.59 221.30 102.27 729.33 309.37 382.06 167.61 1L 2mo-D2- 48.01 27.95 199.93 120.86 983.98 468.87 296.28 134.08 2R 2mo-D2- 36.92 22.13 202.77 113.94 1035.20 480.59 365.67 165.05 2L 2mo-D2- 61.80 23.72 146.77 89.69 614.60 295.49 227.09 101.87 1R

125

2m D2- 54.47 17.26 154.72 78.68 620.73 284.52 182.32 77.34 1L AVG 43.31 19.54 181.35 98.28 776.69 354.52 312.61 136.65 STD 13.85 5.89 27.86 15.42 169.66 85.42 85.83 36.26

126

Major ganglioside changes (ng/mg) in young control vs disease mice superior colliculus

GQ1 d18:1-C20:0/d20:1-C18:0

GQ1 d18:1/C18:0

GT1 d18:1-C20:0/d20:1-C18:0

GT1 d18:1/C18:0

GD1 d18:1-C20:0/d20:1-C18:0

GD1 d18:1/C18:0

GM1 d18:1-C20:0/d20:1-C18:0

GM1 d18:1/C18:0

-200.00 0.00 200.00 400.00 600.00 800.00 1000.00 DISEASE CONTROL

Figure 4.5: Glaucoma (DBA/2J) vs Control (C57BL/6J) superior colliculus ganglioside concentrations (ng/mg) in young adult mice

127

Major ganglioside changes (ng/mg) in old control vs disease mice superior colliculus

GQ1 d18:1-C20:0/d20:1-C18:0

GQ1 d18:1/C18:0

GT1 d18:1-C20:0/d20:1-C18:0

GT1 d18:1/C18:0

GD1 d18:1-C20:0/d20:1-C18:0

GD1 d18:1/C18:0

GM1 d18:1-C20:0/d20:1-C18:0

GM1 d18:1/C18:0

0.00 100.00 200.00 300.00 400.00 500.00 600.00 DISEASE CONTROL

Figure 4.6: Glaucoma (DBA/2J) vs Control (C57BL/6J) superior colliculus ganglioside concentrations (ng/mg) in old adult mice

128

Major ganglioside changes (ng/mg) in young glaucoma disease vs control mice retinal tissues

GQ1 d18:1-C20:0/d20:1-C18:0

GQ1 d18:1/C18:0

GT1 d18:1-C20:0/d20:1-C18:0

GT1 d18:1/C18:0

GD1 d18:1-C20:0/d20:1-C18:0

GD1 d18:1/C18:0

GM1 d18:1-C20:0/d20:1-C18:0

GM1 d18:1/C18:0

0 20 40 60 80 100 120 DISEASE CONTROL

Figure 4.7: Glaucoma (DBA/2J) vs Control (C57BL/6J) retinal tissue ganglioside concentrations (ng/mg) in young adult mice

129

Major ganglioside changes (ng/mg) in old glaucoma disease vs control mice retinal tissues

GQ1 d18:1-C20:0/d20:1-C18:0

GQ1 d18:1/C18:0

GT1 d18:1-C20:0/d20:1-C18:0

GT1 d18:1/C18:0

GD1 d18:1-C20:0/d20:1-C18:0

GD1 d18:1/C18:0

GM1 d18:1-C20:0/d20:1-C18:0

GM1 d18:1/C18:0

0 20 40 60 80 100 120 DISEASE CONTROL

Figure 4.8: Glaucoma (DBA/2J) vs Control (C57BL/6J) retinal tissue ganglioside concentrations (ng/mg) in old adult mice

130

4.3.2 Optic nerve data

DBA/2J glaucoma model mice at ages two-month (2), three-month (1), ten-month

(2) and twelve-month (3) and C57BL/6J, glaucoma control mice at ages three-month (3), ten-month (2) and twelve-month (2) were selected for age studies. Figure 4.5, Figure 4.6 and Figure 4.7 shows the light microscopy results of the optic nerve studies for the above-mentioned mice. To reduce the age variance among the selected mice, all of them were grouped as young (two and three-month) and old (ten and twelve-month) age groups. From the light microscopy studies of the glaucoma and control mice, optic nerve count was done. In DBA/2J, a glaucomatous strain, the average number of optic nerves in young-age group was found to be 34225.8 with a standard deviation of 6520. In old age group, optic nerve number was found to be 36031.6 with a standard deviation of 2301. In

C57BL/6J, glaucoma control mice strain, the average number of optic nerves in young- age group was estimated to 36561.3 with a standard deviation of 6658 and in old age group, the average optic nerve number was 36061.1 with a standard deviation of 6459.

The optic nerve count in DBA/2J shows a marginal damage to the optic nerve contradicting the theoretical values. DBA/2J mice strain usually have elevated intraocular pressure by nine months irrespective of the gender. But the results show less damage to the optic nerve/ healthy optic nerve even at old age. In the control strain, optic nerve count stays nearly same thus proves the control mice do not develop glaucoma as the age progresses.

131

A B

C D Figure 4.9: Optical microscopy image of optic nerve of D2 glaucoma mice strain, 2 months A: right eye, B: left eye, D2G glaucoma control mice C: left eye, D: right eye (D. Inman, Department of Pharmaceutical Sciences, Northeast Ohio Medical University, unpublished data)

132

A B

C D Figure 4.10: Optical microscopy image of optic nerve of D2 glaucoma mice strain, 10 months A: right eye, B: left eye, D2G glaucoma control mice C: left eye, D: right eye. (D. Inman, Department of Pharmaceutical Sciences, Northeast Ohio Medical University, unpublished data)

133

A B

C D Figure 4.11: Optical microscopy image of optic nerve of D2 glaucoma mice strain, 10 months A: right eye, B: left eye, D2G glaucoma control mice C: left eye, D: right eye. (D. Inman, Department of Pharmaceutical Sciences, Northeast Ohio Medical University, unpublished data)

134

The correlation of optic nerve studies was performed across different age groups.

However, the axon counts showed no change from young to old age mice both in disease and control strains taking into account the standard deviation. With retinal tissue, though a comparable deviation in the amounts of gangliosides was seen, it is less significant and inconclusive. Retinal tissue ganglioside results are supportive as the optic nerve studies indicated a healthy optic nerve even at old age glaucoma mice shows a healthy retinal tissue indicating retinal ganglion cells are viable and gangliosides are intact. A correlation of optic damage to the ganglioside levels, specifically GM1 ganglioside was anticipated. The interesting revelation in these studies were the ganglioside changes in the superior colliculus. Superior colliculus of DBA/2J glaucomatous mice shows a significant loss of higher sialylated gangliosides and on contrary the GM1 gangliosides shows a significant growth.

4.4 Conclusion

The tissue level expression of ganglioside changes in DBA/2J glaucomatous mice were studied to ascertain the role of gangliosides, GM1 gangliosides distinctively in glaucoma disease. The optic nerve shows no damage in old age groups of glaucomatous mice. This shows the mice models used in these studies did not develop glaucoma. The

LC-MS/MS data on individual ganglioside changes represents a distinctive change in ganglioside concentrations especially in the old glaucoma retinal tissue compared to the old control mice. A non-correlation between the optic nerve studies data and LC-MS/MS data is observed. No change in the optic nerve count shows a healthy optic nerve though changes in the ganglioside of retinal tissue upregulation is distinctively observed. The changes in the ganglioside amounts shows the neuroprotective action in the retinal tissue.

135

Whereas, statistically the ganglioside changes are insignificant. The p-value is found to be >0.05 which shows the null hypothesis is true. The higher p-value is a result of ganglioside changes observed from individual mice tissues where a high variability in the tissue amounts is seen. Conversely, the ganglioside amounts are variable from animal to animal. The age of the old mice were near 12 months, which might have not developed glaucoma completely. Hence a greater sample study which includes more older mice models is needed to confirm the ganglioside role in glaucoma age disease. Another variability was using a wet tissue for LC-MS/MS tissue analysis resulted in varying amounts. A greater variability can be eliminated by using dry tissue for analysis where results will be more accurate.

136

CHAPTER V

QUANTITATION OF HETEROGENOUS GM1 GANGLIOSIDE WITH HILIC LC-MS/MS ANALYTICAL TECHNIQUE WITH GC-FID

5.1 Importance of analyte purity for quantification

An essential component of most analytical techniques is the use of standards

(calibrators) to generate a calibration curve for determining the concentration or amount of analyte in a sample[96], [97]. Standards are usually prepared from pure analytes or analytes with an established purity. Additionally, standards can be prepared from reference materials in which the concentration of analyte is directly measured, or traceable, to a reference technique and a primary calibrator. Although pure analyte or certified reference materials are available for many analytes, they are not available for all analytes, particularly biological compounds with inherent heterogeneity.

137

This heterogeneity issue is particularly prominent in the analysis of gangliosides.

One study identified 137 different ganglioside and asialo-ganglioside components in human fetal brain tissue[64]. Gangliosides have a backbone structure of sphingosine which is a long chain carbon molecule with a hydroxyl group (at C3) and a double bond

(at C4-C5) and, prior to being incorporated into a ganglioside, have another hydroxyl group (at C1) and an amino group (at C2). Neuronal gangliosides consist almost entirely of a C18 and C20 carbon length sphingosine. It should be noted, however, that a small percentage of neuronal gangliosides (less than 5%) have a C18- or C20-sphinganine backbone which is a saturated carbon chain not containing a double bond[98]–[100].

Moieties added to sphingosine (or sphinganine) to form gangliosides are an oligosaccharide bonded via a glycosidic linkage at C1 and a fatty acid bonded via an amide linkage at C2[9]. Heterogeneity in the gangliosides results from heterogeneity in the oligosaccharide, fatty acid and/or sphingosine, giving rise to the vast number of ganglioside variants.

Commercially-available standards are also heterogeneous in composition, which is problematic in LC-MS/MS as well as other analytical techniques. For example, a commercially-available GM1 ganglioside standard has three major gangliosides and is only quantified in terms of fatty acid content and not according to the individual ganglioside molecular components. This is an issue in ganglioside analysis. Current analytical techniques do not determine intact molecular components but rather are based solely on fatty acid content. Basing ganglioside analysis on fatty acid has substantial limitations, as a particular fatty acid is usually a part in multiple ganglioside molecules.

In particular, a technique which can differentiate among individual iso-molecular weight

138 components of ganglioside needs to be developed. The present work is the first report addressing this issue.

GM1 is a subclass of gangliosides homogeneous in a particular oligosaccharide

(containing one sialic acid group) but varying in the ceramide (sphingosine plus bonded fatty acid) composition. The three major ganglioside components in the particular commercially-available GM1 standard characterized in the present work has the following different ceramide structures: C18:0FA-C18S, and two iso-molecular components, C18:0FA-C20S and C20:0FA-C18S, where FA and S indicate fatty acid and sphingosine, respectively, the 18 and 20 indicate the number of carbons, and the :0 indicates that there are no carbon-carbon double bonds in the fatty acid.

The standard multiple reaction monitoring (MRM) LC-MS/MS techniques used in the determination of gangliosides monitors the m/z 290 dehydrated sialic acid daughter ion of the oligosaccharide. The disadvantage of this MRM technique is that the iso- molecular weight components C18:0FA-C20S and C20:0FA-C18S cannot be differentiated. Present characterization of the gangliosides in the standard is usually limited to the gas chromatography-flame ionization detection (GC-FID) analysis of the hydrolyzed/methylated fatty acids. This analysis only determines the percentage of all gangliosides containing the C18 saturated fatty acid (in this case, the combined amounts of C18:0FA-C18S, C18:0FA-C20S and the C18:0FA-spinganine components), the C20:0 saturated fatty acid [in this case only C20:0FA-C18S and the C20:0FA-spinganine components, as there was no noted signal for C20:0FA-C20S (m/z 1601)] and other fatty acids. However, it is unable to parse out the percentages of the individual GM1 components.

139

The importance of determining the individual molecular gangliosides in a standard to be used in analysis of biological samples is underscored by noted physiologic differences of the gangliosides differing in their sphingosine/sphinganine carbon length, commonly referred to as long chain bases. Physiologic differences have been noted for

C18- and C20-sphingosine/sphinganine gangliosides. Gangliosides in human brain consist entirely of C18-sphingosine at birth, changing to approximately equal proportions of C18- and C20-sphingosine gangliosides in adulthood[99], [101], [102]. Other examples of physiologic significance of the C18-/C20-sphingosine portion of the ganglioside are as follows. It has been suggested that membrane characteristics such as fluidity, thickness and microdomain properties are related to the C18-/C20-sphingosine content[102]. A study reported a considerable effect of increasing the C20-sphingosine ganglioside proportion of gangliosides on micelle size and aggregation characteristics, with implications for neuronal membrane formation and characteristics[103]. Differences have been noted in the rate of association of the particular ganglioside with HeLa cells and intracellular accumulation of cyclic AMP in HeLa cells depending on the length and saturation/unsaturation status of C18- and C20-sphingosine/sphinganine GM1 gangliosides[104]. In another study, different association and metabolism effects were noted for C18 compared to C20 long chain bases in rat cerebellar granule cells[105].

Thus, determining individual molecular gangliosides according to their long chain base identity is important.

In the present work a sensitive MRM LC-MS/MS method for determining the individual gangliosides in a heterogeneous GM1 standard was developed and assessed for accuracy. This MRM method (Method 1) monitors the m/z 290 dehydrated sialic acid

140 daughter ion and incorporated the fatty acid content data determined by (and readily available from) the manufacturer by a GC-FID technique. Equations are derived combining the MRM m/z 290 daughter ion data with the fatty acid content data to calculate the amounts of the three individual molecular components in the GM1 standard.

The accuracy was then assessed employing a less sensitive but specific MRM technique monitoring the fatty acid daughter ion of the individual GM1 molecular components.

5.2 Materials and methods

5.2.1 Materials

GM1 ganglioside standard from bovine brain (Cat. No. 1061) was obtained from

Matreya LLC (Pleasant Gap, PA, USA). Acetonitrile and methanol were Optima LC/MS grade from Fisher Scientific (Fair Lawn, NJ, USA). Ammonium acetate (≥99.99%) was from Sigma-Aldrich (St. Louis, MO, USA). HPLC grade water was from a Barnstead

Nanopure water purification system from Thermo Scientific (West Palm Beach, FL,

USA).

5.2.2 Preparation of standard solutions

A stock solution of the GM1 standard 1mg/ml with 83% methanol was prepared.

All the working GM1 standard solutions in the range 0.05-10 μg/mL were prepared from the serial dilution of the stock solution with 83% methanol. All the stock solutions and working solutions were stored at -20 oC.

5.2.3 HPLC conditions

HPLC analysis was performed on a Waters Alliance 2695 quaternary pump system (Milford, MA, USA). The chromatographic separation was performed on a

141 hydrophilic interaction liquid chromatography (HILIC) column (amino-propyl ligand, 50

× 1 mm, 3 um particle size) with a guard column (amino-propyl ligand, 5 × 1 mm, 3 um particle size) from IMTAKT USA (Portland, OR, USA). A volume of 20 µl of the working standard was injected with an auto sampler at 4 oC. A multiple linear gradient of mobile phases (A: 83% acetonitrile, B: 83% acetonitrile and 5 mM ammonium acetate,

C: 50% acetonitrile and 50 mM ammonium acetate, D: 50% acetonitrile) was used.

Sample was injected into 100% mobile phase A run for 1 min, followed in succession by a linear gradient for 1 min to 100% mobile phase B, 100% mobile phase B for 4 min, a linear gradient for 6 min to 100% mobile phase C and finally 100% mobile phase C for 8 min. Mobile phase was directed to the mass spectrometer after 4 min of the run. The column was re-equilibrated as follows: 50% acetonitrile was run for 12 min followed by

83% acetonitrile for 8 min. Flow rate was 0.1 mL/min. Mobile phases were filtered through 0.45 µm membrane filters from Millipore (Billerica, MA, USA).

5.2.4 Mass spectrometry conditions

The eluent from the chromatographic system was introduced into a triple quadrupole Waters Micromass Quattro Ultima instrument (Milford, MA, USA) with an electrospray ionization (ESI) source. The ESI source in a negative ionization mode was optimized as follows: capillary potential (-3 KV), cone potential (-40 V), source temperature (150 oC), desolvation temperature (300 oC), cone gas flow (144 L/hr) and desolvation gas flow (756 L/hr). The triple quadrupole analyzer was optimized as follows: ion energy 1 and 2 were 1.0 and 3.0 eV respectively, collision energy was 70 eV, entrance and exit potential were 120 V and multiplier potential was -650 V. The multiple reaction monitoring (MRM) transitions were m/z 1545.3 → 290.3 and m/z 1573.3. →

142

290.3 monitoring the dehydrated sialic acid fragment of the GM1 components and m/z

1545.3 → 283.1 (C18:0 fatty acid), m/z 1573.3 → 283.1 (C18:0 fatty acid), m/z 1573.3

→ 311.3 (C20:0 fatty acid) monitoring the respective fatty acids of the GM1 components.

5.2.5 Fatty acid analysis

Fatty acids of gangliosides were esterified with acidic methanol heating to 100 oC for 4 hrs followed by extraction with hexane and quantified with respect to fatty acid methyl ester standards using gas chromatography with flame ionization detection employing a polar phase GC column[101], [106]. The results provided by the manufacturer for the GM1 standard are 90%, 3% and 7% for C18:0, C20:0 and other fatty acid content in the GM1 standard, respectively. No C18:1 or C20:1 fatty acids were present as determined by this method.

143

Figure 5.1: GM1 ganglioside (m/z 1545) fragmentation pattern, the common daughter ion (m/z 290)

144

5.3 Results and Discussion

The GM1 standard consists almost entirely of gangliosides with two sphingosine backbones: C18S and C20S, as has been established for gangliosides from animal brain.

As established by the fatty acid analysis studies, the two predominant fatty acids attached to these sphingosine backbones via amide bonds were C18:0-FA (stearic acid, 90%) and

C20:0-FA (arachidic acid, 3%). The remainder 7% are other gangliosides containing a variety of other length fatty acids. In Method 1 these data were combined with m/z 290 daughter ion MRM LC-MS/MS data to determine the amounts of the three predominant

GM1 gangliosides: C18:0FA-C18S, C18:0FA-C20S and C20:0FA-C18S and compared with results of Method 2 fatty acid daughter MRM LC-MS/MS technique. These methods assume that the MRM response factors are the same for the three molecular ganglioside components, compared between the m/z 290 MRM peaks or among the fatty acid daughter ion MRM peaks. This is a reasonable assumption given that the gangliosides have very similar structures that vary only by 2 carbon units.The fragmentation generating the fatty acid daughter ion is shown in Figure 5.3[107]. The fragmentation mechanism generating the m/z 290 negative ion from the sialic acid has been reported to be cleavage of the glycosidic bond linking the sialic acid group to the oligosaccharide chain followed by the elimination of water from the sialic acid[108].

5.3.1 Method 1: Fatty acid and 290 m/z daughter ion MRM data

Given below are equations 1 – 6 in which peak area data from MRM method monitoring the m/z 290 dehydrated sialic acid daughter ion and fatty acid data from GC-

FID analysis of methylated fatty acids hydrolyzed from the gangliosides are inserted into the appropriate equations to solve for the percentages of the three-predominant molecular

145

GM1 ganglioside components. These equations do not account for any C18 or C-20 sphinganine components that may be present in the standard, which have been measured to be less than 5% of the total gangliosides present in animal brain samples, which introduces a small accuracy error for the Method 1 results, as discussed below.

Given in Figure 5.2 is the MRM chromatogram monitoring the 290 m/z dehydrated sialic acid fragment of the GM1 gangliosides for the C18:0FA-C18S (m/z

1545  290) and for the total C20:0FA-C18S and C18:0FA-C20S content (m/z 1573 

290). The percent content of each ganglioside components was determined as follows by combining the methylated fatty acid hydrolysis data determined by the manufacturer (Eq.

1) and the peak area data for the two MRM m/z 290 daughter ion channels incorporated into 3 of the 4 terms in Eq. 2.

146

Figure 5.2: Negative ion MRM chromatograms of 20 μL of 10 μg/mL GM1 standard (A) m/z 1545 → 290 and (B) m/z 1573 → 290

147

The fatty acid content of the GM1 standard was determined by the manufacturer with the following relationship of the various gangliosides having a particular fatty acid, as given by Eq. 1.

%C18:0FA + %C20:0FA + % (other) FA = 100% Eq. 1 where %C18:0FA, % C20:0FA and % (other) FA were the percentages of gangliosides having: C18:0 fatty acid (stearic acid), C20:0 fatty acid (arachidic acid) and all the other fatty acids, respectively, determined by the manufacturer for the particular lot of GM1 standard used. In this work the percentages were 90%, 3% and 7%, respectively.

Eq. 2 is written in terms of the GM1 ganglioside molecular components

%(C18:0FA-C18S) + %(C18:0FA-C20S) + %(C20:0FA-C18S) + %(other gangliosides) = 100% Eq.2 where %(C18:0FA-C18S), %(C18:0FA-C20S), %(C20:0FA-C18S) and %(other gangliosides) are the percentages of the various molecular GM1 components with the particular fatty acid (FA) and carbon length of the sphingosine (S) given.

The terms in Eq. 2 are determined from the peak areas of the m/z 290 daughter ion MRM chromatograms and the fatty acid content data as given by Eqs. 3 – 6.

The first term in Eq. 2 is determined by Eq. 3:

148

%(퐂ퟏퟖ: ퟎ퐅퐀 − 퐂ퟏퟖ퐒)

퐏퐀 ퟏퟓퟒퟓ → ퟐퟗퟎ 퐌퐑퐌 = (ퟏ + 퐱) ∗ [(퐏퐀 ퟏퟓퟒퟓ → ퟐퟗퟎ 퐌퐑퐌) + (퐏퐀 ퟏퟓퟕퟑ → ퟐퟗퟎ 퐌퐑퐌)]

∗ ퟏퟎퟎ% 퐄퐪. ퟑ where x = %(other) FA)/ [%C18:0FA + %C20:0FA] (from fatty acid analysis, see Eq. 1 for term definitions), where the PA terms are the peak areas for the indicated MRM transitions. In the GM1 standard used in this work x= 1.075. This is calculated from the values obtained from the methanolic hydrolysis experiments determining the fatty acid percentage, which yielded %C18:0FA being 90%, %C20:0FA being 3% and %(other) FA being 7%. The factor (1 +x) in the denominator takes into account the other gangliosides in the standard besides the three predominant forms, which must be added to the denominator as a calculated peak area so that the denominator is the total of all the ganglioside peak areas (whether measured or calculated). This factor utilizes the data from the fatty acid analysis.

In Eq. 3 the area of the 1545 290 MRM chromatogram peak is due only to the

C18:0FA-C18S ganglioside, while the area of the 1573 290 MRM chromatogram peak is due to both the C18:0FA-C20S and the C20:0FA-C18S.

The second term in Eq. 2 is determined by Eq. 4:

%(C18:0FA-C20S) = %C18:0FA - %(C18:0FA-C18S) = 90% - %(C18:0FA-C18S)

Eq. 4 (90% is present work)

149

The value for C18:0FA was determined by the methanolic hydrolysis experiments

(assuming no sphinganines present) and the value for % (C18:0FA-C18S) was obtained from Eq. 3. These values were then inserted in Eq. 4 to solve for % (C18:0FA-C20S).

The third term in Eq. 2 is obtained directly from the fatty content acid data as given in

Eq. 5. It was experimentally confirmed by MS that there was no C20:0FA-C20S in the

GM1 standard and thus only the C20:0FA-C18 ganglioside had a C20:0 fatty acid.

%(C20:0FA-C18S) = %C20:0FA = 3% (3% in present work) Eq. 5

The final term in Eq. 2 is obtained directly from the fatty acid content data as given in Eq.

%(other gangliosides)=%(other)F=7% (present work) Eq.6

5.3.2. Method 2: Fatty acid daughter ion MRM data

The following MRM transitions monitoring fatty acid daughter ion fragments are not normally used in ganglioside MS analysis because of low sensitivity:

C18:0FA-C18S, m/z 1545  283; C18:0FA-C20S, m/z 1573  283; and C20:0FA-

C18S, m/z 1573  311.

However, these MRM parameters were monitored in the present work to determine the percentages of the three predominant GM1 gangliosides to compare with

150 that obtained by the more sensitive Method 1 results. The MRM chromatograms monitoring these negative fatty acid daughter ions are given in Figure 5.3.

The percentages of these three predominant GM1 gangliosides can be directly calculated from the ratio of each ganglioside to the total peak area as given in Eqs. 7-9.

퐏퐀 ퟏퟓퟒퟓ→ퟐퟖퟑ 퐌퐑퐌 %(퐂ퟏퟖ: ퟎ퐅퐀 − 퐂ퟏퟖ퐒) = ∗ (ퟏ+퐱)∗[(퐏퐀 ퟏퟓퟒퟓ→ퟐퟖퟑ 퐌퐑퐌)+(퐏퐀 ퟏퟓퟕퟑ→ퟐퟖퟑ 퐌퐑퐌)+(퐏퐀 ퟏퟓퟕퟑ→ퟑퟏퟏ)]

ퟏퟎퟎ% Eq. 7

퐏퐀 ퟏퟓퟕퟑ→ퟐퟖퟑ 퐌퐑퐌 %(퐂ퟏퟖ: ퟎ퐅퐀 − 퐂ퟐퟎ퐒) = ∗ (ퟏ+퐱)∗[(퐏퐀 ퟏퟓퟒퟓ→ퟐퟖퟑ 퐌퐑퐌)+(퐏퐀 ퟏퟓퟕퟑ→ퟐퟖퟑ 퐌퐑퐌)+(퐏퐀 ퟏퟓퟕퟑ→ퟑퟏퟏ)]

ퟏퟎퟎ% Eq. 8

%(퐂ퟐퟎ: ퟎ퐅퐀 − 퐂ퟏퟖ퐒)

퐏퐀 ퟏퟓퟕퟑ → ퟑퟏퟏ 퐌퐑퐌 = (ퟏ + 퐱) ∗ [(퐏퐀 ퟏퟓퟒퟓ → ퟐퟖퟑ 퐌퐑퐌) + (퐏퐀 ퟏퟓퟕퟑ → ퟐퟖퟑ 퐌퐑퐌) + (퐏퐀 ퟏퟓퟕퟑ → ퟑퟏퟏ)]

∗

ퟏퟎퟎ% Eq. 9

where the parameters are previously defined, with PA being the peak area of the indicated MRM peaks. For the GM1 standard used in the present work the factor in the denominator was 1.075, the same as for Method 1. Equations 7 -9 do not account any presence of C-18 and C-20 sphinganines, which will lead to small accuracy errors as discussed in section 3.5.

151

Figure 5.3: Negative ion MRM chromatograms of 20 μL of 10 μg/mL GM1 standard (A) m/z 1545 → 283, [B] m/z 1573 → 283 and [C] m/z 1573 → 311

152

5.3.3 Method 1 results

Duplicate injections of GM1 standards at concentrations 50, 100, 250, 500, 1000,

2000 and 10000 ng/ml on the LC-MS/MS system were done monitoring the MRM m/z

290 dehydrated sialic acid daughter ion. The peak area data were processed according to the Method 1 utilizing Eqs. 3-6 to determine the percentage of each of the three predominant GM1 gangliosides in the standard, with the average of the results and associated standard deviation given in Table 5.1. Example calculations for determining the percentages of the three major ganglioside components in the GM1 standard according to Method 1 are given in the Supplementary Information section.

A plot of the ratio of the Method 1 determined values for C18:0FA-C18S to

C18:0FA-C20S versus standard GM1 concentration is shown in Figure 4 to assess if there were any concentration effects or trends. The slope of the regression line (given

Figure 5.4) is close to zero, indicating no concentration trend in the data. However, the data at lower concentration does show some variability.

5.3.4 Comparison of Method 1 with Method 2 results

Similarly, duplicate injections of GM1 standard at concentrations 1000, 2000,

5000 and 10000ng/ml on the LC-MS/MS system were done monitoring the MRM fatty acid daughter ion to assess the accuracy of Method 1. Note, only higher concentrations could be determined for this MRM daughter ion monitoring, a less sensitive transition.

The peak area data were processed according to Method 2 utilizing Eqs. 7-9. with the results given in Table 5.1. Comparing the results for Method 1 and Method 2 in Table

5.1 for the predominant GM1 gangliosides shows the absolute % values do not differ by more than 2% for any of the three-predominant molecular GM1 gangliosides. Statistical

153 assessment shows no significant difference for the percent values for the C18:0FA-C18S

GM1 ganglioside in comparing the two methods. However, the percent values for the

C18:0FA-C20S GM1ganglioside is statistically lower for Method 1 (m/z 290 daughter ion MRM technique) than Method 2 (fatty acid daughter ion MRM technique) although differing only by 2% absolute percent. Example calculations for determining the percentages of the three major ganglioside components in the GM1 standard according to

Method 2 are given in the Supplementary Information section.

A plot of the ratio of the Method 2 determined values for C18:0FA-C18S to

C18:0FA-C20S versus standard GM1 concentration is also shown in Figure 5.4 The regression line slope is close to zero (see Figure 5.4) and thus no significant concentration trend is noted, as was see for Method 1. The average of ratios of C18:0FA-

C18S and C18:0FA-C20S at different concentrations by Method 1 and Method 2 were

1.163 (SD ± 0.035) and 1.125 (SD ± 0.018) respectively. These ratio means are not statistically different (p value = 0.07).

154

2 -

1.5

C20S -

1 C18S/C18FA

0.5 Ratiopercentagesof of C18FA

0 0 2000 4000 6000 8000 10000 12000 GM1 ng/ml

Figure 5.4: Ratio of the percent of the two prominent GM1 gangliosides C18:0FA- C18S/C18:0FA-C20S determined by Method 1 (triangles, dotted linear regression line) and by Method 2 (circles, solid linear regression line) versus concentration of GM1 standard injected

155

5.4 Effect of sphinganine presence in GM1 standard

Since the GM1 standard most likely contains a small amount of sphinganine (up to 5% of the gangliosides found in brain samples calculations were done assuming 5% sphinganine content and compared with the previous calculated results to determine the magnitude of the error resulting from the unmeasured presence of the sphinganine. The sphinganine gangliosides can be considered to having exclusively C18 fatty acid (being

90% of the fatty acids present) since this is the only components that can affect the sphingosine ganglioside results. With 5% sphinganine content the factor (1+x) in Eqs. 3,

7-9 increases from 1.075 to 1.136, the % (other) FA in Eqs. 1 and 6 increases from 7% to

12%, and the %C18FA bonded to the sphingosine ganglioside in Eq. 4 is decreased from

90% to 85%. The results assuming 5% sphingosine content are given in Table 5.1. There is a relative error of approximately +5-6% in the sphingosine ganglioside if an assumed

5% sphinganine content in the standard is not taken into account.

It should be noted that the sphingosine gangliosides (m/z 1545 and 1573) can be differentiated from sphinganine gangliosides (m/z 1547 and m/z 1575) because the selection of the parent ion is specific (m/z ± 0.5). However, the sphinganine gangliosides cannot be determined in this mass spectrometric experiment because the +2 isotopes of the sphingosine gangliosides of 1547 and 1575 masks the monoisotopic sphinganine gangliosides. Thus, the results need to be corrected with assumed sphinganine content or viewed as having up to 6% relative error.

156

Table 5.1: Molecular Ganglioside Content of the Matreya Bovine Brain GM1 Standard

% C18:0FA- % C18:0FA- % C20:0FA- % Other C18S C20S C18S Gangliosides (± SD) (± SD)

100% C18- and C20-Sphingosine Long Chain Bases Method 1a 48.3 (± 0.7) 41.7 (± 0.7) 3 7 Method 2b 47.8 (± 0.4) 43.6 (± 0.4) 1.6 7 95% C18- and C20-Sphingosine and 5% C18- and C20-Sphinganine Long Chain Bases Method 1a 45.7 (±0.7) 39.3 (±0.7) 3 12 Method 2b 45.3 (±0.4) 41.3 (±0.5) 1.5 12 % Error comparing 100% and 95% Long Chain Base Sphingosine Content Results Method 1 +5.4 % error +5.8 % error 0.0 % error Method 2 +5.2 % error +5.3 % error +6.3 % error

a Results for the C18:0FA-C18S and the C18:0FA-C20S for Method 1 are based on averaged duplicate runs of 8 standards varying in concentration from 50 ng/mL to 10,000 ng/mL of the GM1 standard. The numbers for the % C20:0FA-C18S and the % other gangliosides is from the fatty acid analysis performed by the manufacturer, for which no SD data is available. b Results for Method 2 are based on the duplicate runs of the highest standard 10,000 ng/mL, as only this standard gave an MRM peak for %C20:0FA-C18S. SDs are calculated for C18:0FA-C18S and the C18:0FA-C20S based on MRM peaks for averaged duplicate runs for 4 standards varying in concentration from 1000 ng/mL to 10,000 ng/mL. No SDs are calculated for the others as there were only 2 runs for C20:0FA-C18S and the other gangliosides was determined by fatty acid analysis, for which no standard deviation data is available.

157

5.5 Conclusions

A simple procedure for quantitatively determining the ganglioside composition of a commercially- available heterogeneous GM1 standard (extracted from a natural source, bovine) has been reported for the first time and the results have been confirmed by a comparative method. The significance of the present study is that the analysis utilizes intact ganglioside molecules data using a sensitive MRM measuring the dehydrated sialic acid fragment to determine individual molecular weight components of gangliosides, combining this with fatty acid content data. Peak area data as low as 50 ng/mL was established, which is relevant for measuring GM1 gangliosides at physiological concentrations[109]. The present work was done to confirm the validity of the method employing standard GM1 solutions. Future studies will involve assessing the method for determining these individual molecular GM1 ganglioside components in physiological samples.

158

CHAPTER Ⅵ

FUTURE DIRECTIONS

6.1 Summary

Glia are a heterogeneous collection of cell types in the CNS that continue to surprise in their diversity and functional capability. As intermediaries between vasculature and energy-intensive neurons, and because of their presence at the synapse, glia has established a reputation as primary support cells for neurons, providing energy substrate and eliminating excess neurotransmitters. Recent studies have shown that in reactive gliosis during glaucoma, astrocytes express gangliosides during the degeneration of retinal ganglion cells. Preliminary results with DBA/2J glaucoma mice show efficient

159

Cholera Toxin-B (CTB) uptake by the astrocytes which could be resulting from an increased expression of GM1 gangliosides. If we are able to confirm these initial observations, demonstrating that gangliosides are upregulated specifically in astrocytes as a result of local neural degeneration, we will have identified a potentially significant alteration in these cells that could have profound implications for the survival and function of RGCs. The CTB affinity is variable to various classes of gangliosides differing in the sialic acid sugar groups. Also, the heterogeneity with the ganglioside lipids requires a more specific and sensitive analytical method to identify and quantify individual gangliosides which plays a crucial role in the glaucoma disease.

The central research goal is to identify and quantify the gangliosides that alter during glaucoma age progression. For this, a highly sensitive, specific and high throughput analytical method is necessary. In my research, I developed two LC-MS/MS analytical methods where the gangliosides were by HPLC separated and quantified by

MRM mass spectrometry. A novel robust LC-MS/MS method was developed where the gangliosides were separated based on both their hydrophobicity and their sialic acid content on a phenyl-hexyl HPLC column. This is the first analytical method reported which includes polar and non-polar retention. Major gangliosides were baseline separated within a specific retention window based on the sialic acid classes and ceramide moiety sub-classes and isomer sub-sub classes. A total of 16 gangliosides in monosialo-, disialo-, trisialo- and tetrasialo gangliosides were monitored and eight major gangliosides were quantified. This LC-MS/MS technique was validated for recovery and matrix effect limit of quantification and limit of detection, linearity and precision. Another separation technique, HILIC, in which ganglioside retention is based on the polar head of the

160 gangliosides, was also investigated. The HILIC column technique separated the ganglioside classes but could not separate the hydrophobic subclasses. To quantitate co- eluted isomers in GM1 ganglioside, HILIC LC-MS/MS method was combined with the fatty acid analysis from GC-FID method. These results were confirmed with a less- sensitive HILIC fatty acid MRM LC-MS/MS method. The complexity of combining two methods and less robustness of separation on HILIC column. Thus phenyl-hexyl technique was chosen for further analysis, because it could separate both the sialic acid classes but also the subclasses of the gangliosides.

Ganglioside isolation was another challenge to address. The established isolation technique, the Schnaar protocol, was found to yield low recoveries. Another isolation technique, Svennerholm and Fredman method was adopted and optimized. The optimized technique is a combination of liquid-liquid extraction followed by a solid-phase extraction. The sample preparation was validated by deuterated mono-sialo ganglioside,

GM1-D3, in brain and pooled retinal tissues. Single retinal tissue extraction was also done for the first time. A total distribution of major gangliosides in brain (grey and white matter) and pooled retinal tissues was reported.

Using the above LC-MS/MS analytical method using a phenyl-hexyl column, after ganglioside isolation using the modified extraction technique, studies were done determining ganglioside distribution in retinal and superior colliculus tissues of glaucomatous DBA/2J and C57BL/6J control mice from young and old mice. These results were discussed in chapter 4. This is the first report on the ganglioside distribution in glaucoma disease progression. While the optic nerve results indicated no glaucoma- like development in the mice, there were significant differences in percentage of change

161 of each class of gangliosides particularly in the retinal tissue comparing the DBA/2J glaucoma mice with the control mice. The total gangliosides increased, being a protective response preventing optic nerve changes.

6.2 Future directions

6.2.1 Ganglioside distribution in glaucoma retinal tissue

Normally, age six to seven months all glaucomatous mice show significantly damaged optic nerves. Our age study experiments, measured gangliosides and assessed optic nerve status in10 and 12-month-old mice. However, the optic nerve study showed no damage. In future studies, older mice will be assessed to assure glaucoma development as evidence by optic nerve damage.

DBA/2J glaucomatous mice strain will be acquired from The Jackson Laboratory.

This strain is very popular and commonly used for many neurodegeneration diseases.

Although the procedure to measure the intraocular pressure is available, the drawback to this procedure to establish the disease progress is that in primary open-angle glaucoma the optic nerve can be healthy even though the intraocular pressure reaches a maximum.

Thus, other parameters contribute to the damage of the optic nerve. The only way to confirm the glaucoma disease stage is to do optic nerve studies. Thus, plan will be to study ganglioside changes in mice that are assured to develop optic nerve damage by using even older mice.

6.2.1.1 Expected results

These results will establish the changes in gangliosides, specifically GM1 ganglioside, during the retinal tissue damage. An elevated level confirms CTB binding

162 results which showed astrocytes expressing gangliosides on their surface and functioning as neuronal cells during retinal ganglion call damage.

6.2.2 Ganglioside distribution at cellular level

There is a need for understanding the cells which promote the GM1 expression as

RGCs degenerate in glaucoma progression. To study this, RGCs and astrocyte cells will be separated from the retinal tissue of glaucoma mice using an immuno-panning technique. Gangliosides will be extracted from the isolated astrocyte cells and RGCs in retinal tissue of glaucoma mice at three different ages at 3, 10 and 14 months. This will assess the role of astrocytes in GM1 expression during glaucoma neurodegeneration.

Similarly, the levels of GM1 gangliosides will be expressed by the RGCs and astrocytes measured at the different ages correlating a decrease in the expression of GM1 by RGCs with an increase in their expression in astrocytes as glaucoma progresses.

6.2.2.1 Methodology

Mice will be anesthetized with isoflurane (2.5%) and receive a fluorogold (FG) injection which retrogradely label RGCs. Three days prior to sacrifice, CTB will be injected which adsorbs on to the fluorophore. Mice retinas will be dissociated and CTB bound cells isolated by immuno-panning using antibody against CTB. Cell suspensions will be held at 370C for about an hour to remerge on the cell surface and incubated with

CTB-antibody coated culture dishes. Dishes will then be washed to remove the CTB- negative cells. After the CTB positive cells are left over on the culture dishes, it will be incubated with GLAST antibodies which bind specifically to astrocytes. Thus, the astrocytes will be separated from the RGCs and unbound RGCs washed off. These

163 isolated populations will then undergo sample preparation to extract GM1 gangliosides as specified previously. The age progression of DBA/2J mice will mimic glaucoma progression as follows: 3-month mice having no disease manifestation; 10 month mice being at the beginning/intermediate stages of the disease; and 14-month being at the stage of the disease when most ganglion cells are degenerated.

6.2.2.2 Expected results

GM1 gangliosides will be quantified in RGCs and astrocyte cell populations in normal and glaucoma model mice at various stages of age. These experiments are expected to show ganglioside changes correlating with glaucoma progression and give insight into the role of astrocytes in glaucoma. As the glaucoma disease progresses, preliminary results with CTB binding shows astrocyte cells express gangliosides while a decrease of gangliosides in the RGCs. Thus, cellular studies with LC-MS/MS analysis confirm the immunohistochemistry results and affirms the neuroprotective action of GM1 gangliosides.

6.2.3 Studies of the gangliosides and Biosynthesis and metabolic pathways

Ceramides are synthesized in endoplasmic reticulum through condensation, reduction and acetylation by a series of enzymes. These are transferred to golgi apparatus for further modifications in the carbohydrate complex to form gangliosides by UDP-Glc: ceramide glucosyltransferase, UDP-Gal: ceramide galactosyltransferase and UDP-

NeuAc: lactosylceramide sialyltransferase, N-acetylgalactosaminyltransferase and Gal-

T2, UDP-Gal: GA2/GM2/GD2/GT2 galactosyltransferase. GM2 galactosyltransferase transfers galactose to GM2 ganglioside to synthesize GM1 ganglioside. During glaucoma

164 expression, GM1 gangliosides on astrocytes are synthesized by these enzymes. An up- regulation of GM1 gangliosides is a result of the reflects an upregulation of these specific enzymes which are part of biosynthetic pathway (Figure 6.1).

Determination of enzyme activities or protein levels of Glc-T, UDP-Glc: ceramide glucosyltransferase; Gal-T1, UDP-Gal: glucosylceramide galactosyltransferase;

Sial-T1, CMP-NeuAc: lactosylceramide sialyltransferase and GalNAc-T, UDP-GalNAc: lactosylceramide/GM3/GD3/GT3 N-acetylgalactosaminyltransferase enzymes which take part in the biosynthetic pathway of GM1 ganglioside will provide an insight into up- regulation ganglioside synthesis. The up-regulation of these enzymes is expected to correlate with the up-regulation of GM1 gangliosides. During glaucoma disease progression, these enzyme analyses will confirm the expression of GM1 ganglioside during glaucoma disease.

Gangliosides are transported to the plasma membrane by vesicles. At the cell surface, gangliosides either undergo catabolism by glycohydrolases or can undergo further biosynthesis through the activity of sialyl transferases. The catabolic pathway of gangliosides involves sequential removal of sugars from the gangliosides. β- galactosidase, β-N-acetyl- hexosaminidase and β-glucosidase removes galactose, N- acetyl-galactosamine and glucose, respectively. The metabolized surface gangliosides can be internalized, endocytosed by endosomes where they can be reglycosylated and be substrates in biosynthetic and metabolic pathways. The turn-over of gangliosides on the

165

Figure 6.1: Ganglioside biosynthetic pathway

166 plasma membrane is a continuous process where complex gangliosides are converted into simpler gangliosides and endocytosed.

Again, these molecules participate in the biosynthesis and reach the plasma membranes. Hence, enzymes that breakdown gangliosides also play a critical role during regulation of gangliosides levels. Protein and enzyme activity studies of these breakdown enzymes are also important to understand changes in ganglioside expression.

As gangliosides undergo anabolic and catabolic processes where membrane enzymes play a critical role in the regulation of ganglioside classes on the membrane. A detailed study of enzymes involved in biosynthetic and metabolic pathway helps to better understand ganglioside changes. These studies will affirm the role of GM1 gangliosides and its up-regulation in glaucoma disease progression.

167

References

[1] Richard W. Gross, Xianlin Han, “Lipidomics at the Interface of Structure and

Function in Systems Biology,” Chemistry & Biology., vol. 18, pp. 284-291,

March 2011.

[2] E. Fahy et al., “A Comprehensive Classification System for Lipids,” J. Lipid Res.,

vol. 46, no. 5, pp. 839–861, May 2005.

[3] Lars Svennerholm, “Structure and Function of Gangliosides,” Advances in

Experimental Medicine and Biology., 1980.

[4] H. Wiegandt, “The Structure and the Function of Gangliosides,” Anew. Chem.

Int. Ed. Engl., vol. 7, no. 2, pp. 87–96, Feb 1968.

[5] Hinrich Rahmann, “Gangliosides and Modulation of Neuronal Functions,” NATO

Advanced Research Workshop, vol. H7, 1987.

[6] R. W. Ledeen, R. K. Yu, M. M. Rapport, and K. Suzuki, Eds., Ganglioside

Structure, Function, and Biomedical Potential, vol. 174. Boston, MA: Springer

US, 1984.

[7] M. Alan Chester, “Nomenclature of Glycolipids,” IUPAC-IUB Joint Commission

on Biochemical Nomenclature, 1997.

[8] Y. D. Vankar and R. R. Schmidt, “Chemistry of Glycosphingolipids—

Carbohydrate Molecules of Biological Significance,” Chem. Soc. Rev., vol. 29,

no. 3, pp. 201–216, Jan. 2000.

168

[9] R. K. Yu, Y.-T. Tsai, T. Ariga, and M. Yanagisawa, “Structures, Biosynthesis,

and Functions of Gangliosides—An overview,” J. Oleo Sci., vol. 60, no. 10, pp.

537–544, 2011.

[10] David L Nelson; Michael M Cox, “Lehninger Principles of Biochemistry” edition

6, pp. 353-357, Nov. 2012.

[11] F. Malchiodi-Albedi, S. Paradisi, A. Matteucci, C. Frank, and M. Diociaiuti,

“Amyloid Oligomer Neurotoxicity, Calcium Dysregulation, and Lipid Rafts,” Int.

J. Alzheimer’s Dis., vol. 2011, Feb. 2011.

[12] R. W. Ledeen and G. Wu, “The Multi-Tasked Life of GM1 Ganglioside, A True

Factotum of Nature,” Trends Biochem. Sci., vol. 40, no. 7, pp. 407–418, Jul.

2015.

[13] S. Sonnino, L. Mauri, V. Chigorno, and A. Prinetti, “Gangliosides as Components

of Lipid Membrane Domains,” Glycobiology, vol. 17, no. 1, p. 1R–13R, Jan.

2007.

[14] S. Julien, M. Bobowski, A. Steenackers, X. Le Bourhis, and P. Delannoy, “How

Do Gangliosides Regulate RTKs Signaling?” Cells, vol. 2, no. 4, pp. 751–767,

Dec. 2013.

[15] R. K. Yu, Y.-T. Tsai, and T. Ariga, “Functional Roles of Gangliosides in

Neurodevelopment--An Overview of Recent Advances,” Neurochem. Res., vol.

37, no. 6, pp. 1230–1244, Jun. 2012.

169

[16] G. S. Eisenbarth, F. S. Walsh, and M. Nirenberg, “Monoclonal Antibody to A

Plasma Membrane Antigen of Neurons,” Proc. Natl. Acad. Sci. U. S. A., vol. 76,

no. 10, pp. 4913–4917, Oct. 1979.

[17] S. Bergante et al., “Gangliosides as A Potential New Class of Stem Cell Markers:

The Case of GD1a in Human Bone Marrow Mesenchymal Stem Cells,” J. Lipid

Res., vol. 55, no. 3, pp. 549–560, Mar. 2014.

[18] J. L. Daniotti and R. Iglesias-Bartolomé, “Metabolic Pathways and Intracellular

Trafficking of Gangliosides,” IUBMB Life, vol. 63, no. 7, pp. 513–520, Jul. 2011.

[19] M. Tatsumoto et al., “Spectrum of Neurological Diseases Associated with

Antibodies to Minor Gangliosides GM1b and Galnac-GD1a,” J. Neuroimmunol.,

vol. 177, no. 1–2, pp. 201–208, Aug. 2006.

[20] Grimm et al., “Amyloid Precursor Protein (APP) Mediated Regulation of

Ganglioside Homeostasis Linking Alzheimer’s Disease Pathology with

Ganglioside Metabolism.” PLOS one., vol. 7, issue 3, March 2012.

[21] K. Yanagisawa, “Pathological significance of ganglioside clusters in Alzheimer’s

disease,” J. Neurochem., vol. 116, no. 5, pp. 806–812, Mar. 2011.

[22] J. S. Schneider, S. M. Gollomp, S. Sendek, A. Colcher, F. Cambi, and W. Du, “A

Randomized, Controlled, Delayed Start Trial of GM1 Ganglioside in Treated

Parkinson’s Disease Patients,” J. Neurol. Sci., vol. 324, no. 1–2, pp. 140–148, Jan.

2013.

170

[23] G. Wu, Z.-H. Lu, N. Kulkarni, and R. W. Ledeen, “Deficiency of Ganglioside

GM1 Correlates with Parkinson’s Disease in Mice and Humans,” J. Neurosci.

Res., vol. 90, no. 10, pp. 1997–2008, Oct. 2012.

[24] I. Nachamkin, B. M. Allos, and T. Ho, “Campylobacter Species and Guillain-

Barré Syndrome,” Clin. Microbiol. Rev., vol. 11, no. 3, pp. 555–567, Jul. 1998.

[25] V. Maglione et al., “Impaired Ganglioside Metabolism in Huntington’s Disease

and Neuroprotective Role of GM1,” J. Neurosci. Off. J. Soc. Neurosci., vol. 30,

no. 11, pp. 4072–4080, Mar. 2010.

[26] P. Panzetta and M. L. Allende, “Ganglioside Expression During Differentiation of

Chick Retinal Cells In Vitro,” Neurochem. Res., vol. 25, no. 1, pp. 163–169, Jan.

2000.

[27] J. L. Daniotti, C. A. Landa, and H. J. Maccioni, “Regulation of ganglioside

composition and synthesis is different in developing chick retinal pigment

epithelium and neural retina,” J. Neurochem., vol. 62, no. 3, pp. 1131–1136, Mar.

1994.

[28] E. Meuillet, G. Crémel, D. Hicks, and H. Dreyfus, “Ganglioside Effects on Basic

Fibroblast and Epidermal Growth Factor Receptors in Retinal Glial Cells,” J.

Lipid Mediat. Cell Signal., vol. 14, no. 1–3, pp. 277–288, Sep. 1996.

[29] E. Meuillet, G. Cremel, H. Dreyfus, and D. Hicks, “Differential Modulation of

Basic Fibroblast and Epidermal Growth Factor Receptor Activation by

Ganglioside GM3 in Cultured Retinal Müller glia,” Glia, vol. 17, no. 3, pp. 206–

216, Jul. 1996.

171

[30] H. Dreyfus et al., “Gangliosides and Neurotrophic Growth Factors in the Retina:

Molecular Interactions and Applications as Neuroprotective Agentsa,” Ann. N. Y.

Acad. Sci., vol. 845, no. 1, pp. 240–252, Jun. 1998.

[31] D. F. Bussolino, M. E. Guido, and B. L. Caputto, “Light Exposure Stimulates the

Ativity of Ganglioside Glycosyltransferases of Retina Ganglion Cells,”

Neurochem. Int., vol. 31, no. 1, pp. 105–111, Jul. 1997.

[32] R. Caputto, “The Labeling of the Retina and Optic Tectum Gangliosides and

Glycoproteins of Chickens in Darkness or Exposed to Light,” Adv. Exp. Med.

Biol., vol. 174, pp. 147–154, 1984.

[33] S. Mohand-Said, M. Weber, D. Hicks, H. Dreyfus, and J. Sahel, “Gangliosides: A

Potential Therapeutic Approach in Retinal Ischemia,” Ann. N. Y. Acad. Sci., vol.

845, no. 1, pp. 421–421, Jun. 1998.

[34] C. Bucolo, L. R. Lin, L. Dang, F. J. Giblin, and V. N. Reddy, “The Effect of

Ganglioside on Oxidation-Induced Permeability Changes in Lens and in

Epithelial Cells of Lens and Retina,” Exp. Eye Res., vol. 58, no. 6, pp. 697–704,

Jun. 1994.

[35] L. Facci, A. Leon, and S. D. Skaper, “Excitatory Amino Acid Neurotoxicity in

Cultured Retinal Neurons: Involvement of N-Methyl-D-Aspartate (NMDA) And

Non-NMDA Receptors and Effect of Ganglioside GM1,” J. Neurosci. Res., vol.

27, no. 2, pp. 202–210, Oct. 1990.

[36] V. M. Goettl, T. A. Wemlinger, T. G. Fong, N. H. Neff, and M.

Hadjiconstantinou, “Retinal Cholinergic and Dopaminergic Deficits of Aged Rats

172

Are Improved Following Treatment with GM1 Ganglioside,” Brain Res., vol.

877, no. 1, pp. 1–6, Sep. 2000.

[37] T. N. Seyfried, R. K. Yu, N. Miyazawa, and Y. L. Lai, “Retinal Gangliosides in

RCS Mutant Rats,” J. Neurochem., vol. 39, no. 1, pp. 277–279, Jul. 1982.

[38] Masson et al, “a-Series Gangliosides Mediate the Effects of Advanced Glycation

End Products on Pericyte and Mesangial Cell Proliferation a Common Mediator

for Retinal and Renal Microangiopathy?” Diabetes, Vol. 54, January 2005

[39] E. Masson, N. Wiernsperger, M. Lagarde, and S. El Bawab, “Involvement of

Gangliosides in Glucosamine-induced Proliferation Decrease of Retinal

Pericytes,” Glycobiology, vol. 15, no. 6, pp. 585–591, Jun. 2005.

[40] P. H. H. Lopez and R. L. Schnaar, “Gangliosides in Cell Recognition and

Membrane Protein Regulation,” Curr. Opin. Struct. Biol., vol. 19, no. 5, pp. 549–

557, Oct. 2009.

[41] M. L. Allende and R. L. Proia, “Lubricating cell signaling pathways with

gangliosides,” Curr. Opin. Struct. Biol., vol. 12, no. 5, pp. 587–592, Oct. 2002.

[42] Kwon et al, “Primary Open-Angle Glaucoma,” The New England Journal of

Medicine, vol. 360, pp. 1113-1124, March 12, 2009.

[43] R. N. Weinreb et al., “Primary open-angle glaucoma,” Nat. Rev. Dis. Primer, vol.

2, p. 16067, Sep. 2016.

173

[44] S. Yamamoto et al., “Primary open-angle glaucoma in a Population Associated

with High Prevalence of Primary angle-closure glaucoma: the Kumejima Study,”

Ophthalmology, vol. 121, no. 8, pp. 1558–1565, Aug. 2014.

[45] H. Takahashi et al., “Reactive Gliosis of Gstrocytes and Müller glial Cells in

Retina of POMGnT1-Deficient Mice,” Mol. Cell. Neurosci., vol. 47, no. 2, pp.

119–130, Jun. 2011.

[46] D. M. Inman and P. J. Horner, “Reactive Non-Proliferative Gliosis Predominates

in a Chronic Mouse Model of Glaucoma,” Glia, vol. 55, no. 9, pp. 942–953, Jul.

2007.

[47] M. C. Raff, E. R. Abney, J. Cohen, R. Lindsay, and M. Noble, “Two types of

astrocytes in cultures of developing rat white matter: differences in morphology,

surface gangliosides, and growth characteristics,” J. Neurosci. Off. J. Soc.

Neurosci., vol. 3, no. 6, pp. 1289–1300, Jun. 1983.

[48] W. E. van Heyningen, “Gangliosides as Membrane Receptors for Tetanus Toxin,

Cholera Toxin and Serotonin,” Nature, vol. 249, no. 5456, p. 415, May 1974.

[49] R. Huba and H. D. Hofmann, “Tetanus Toxin Binding to Isolated and Cultured

Rat Retinal Glial Cells,” Glia, vol. 1, no. 2, pp. 156–164, 1988.

[50] M. Bähr and B. Schlosshauer, “JONES Ganglioside Expression on Retinal Glia

Increases After Axotomy,” J. Neurocytol., vol. 18, no. 4, pp. 553–563, Aug. 1989.

174

[51] B. P. Buckingham et al., “Progressive Ganglion Cell Degeneration Precedes

Neuronal Loss in A Mouse Model of Glaucoma,” J. Neurosci. Off. J. Soc.

Neurosci., vol. 28, no. 11, pp. 2735–2744, Mar. 2008.

[52] I. Soto et al., “Retinal Ganglion Cells Downregulate Gene Expression and Lose

Their Axons Within the Optic Nerve Head in A Mouse Glaucoma Model,” J.

Neurosci. Off. J. Soc. Neurosci., vol. 28, no. 2, pp. 548–561, Jan. 2008.

[53] S. Baltan, D. M. Inman, C. A. Danilov, R. S. Morrison, D. J. Calkins, and P. J.

Horner, “Metabolic Vulnerability Disposes Retinal Ganglion Cell Axons to

Dysfunction in A Model of Glaucomatous Degeneration,” J. Neurosci. Off. J.

Soc. Neurosci., vol. 30, no. 16, pp. 5644–5652, Apr. 2010.

[54] T. C. Jakobs, R. T. Libby, Y. Ben, S. W. M. John, and R. H. Masland, “Retinal

Ganglion Cell Degeneration Is Topological but Not Cell Type Specific in DBA/2J

Mice,” J. Cell Biol., vol. 171, no. 2, pp. 313–325, Oct. 2005.

[55] Y. Ohmi et al., “Gangliosides Play Pivotal Roles in The Regulation of

Complement Systems and in The Maintenance of Integrity in Nerve Tissues,”

Proc. Natl. Acad. Sci. U. S. A., vol. 106, no. 52, pp. 22405–22410, Dec. 2009.

[56] Y. Ohmi et al., “Gangliosides Are Essential in The Protection of Inflammation

and Neurodegeneration Via Maintenance of Lipid Rafts: Elucidation by A Series

of Ganglioside-Deficient Mutant Mice,” J. Neurochem., vol. 116, no. 5, pp. 926–

935, Mar. 2011.

175

[57] E. A. Miljan and E. G. Bremer, “Regulation of Growth Factor Receptors by

Gangliosides,” Sci. STKE Signal Transduct. Knowl. Environ., vol. 2002, no. 160,

p. re15, Nov. 2002.

[58] R. Butowt and C. S. von Bartheld, “Anterograde Axonal Transport of BDNF and

NT-3 By Retinal Ganglion Cells: Roles of Neurotrophin Receptors,” Mol. Cell.

Neurosci., vol. 29, no. 1, pp. 11–25, May 2005.

[59] E. Vecino et al., “Rat Retinal Ganglion Cells Co-Express Brain Derived

Neurotrophic Factor (BDNF) and Its Receptor TrkB,” Vision Res., vol. 42, no. 2,

pp. 151–157, Jan. 2002.

[60] M. Pitto, T. Mutoh, M. Kuriyama, A. Ferraretto, P. Palestini, and M. Masserini,

“Influence of Endogenous GM1 Ganglioside On Trkb Activity, In Cultured

Neurons,” FEBS Lett., vol. 439, no. 1, pp. 93–96, Nov. 1998.

[61] J.-S. Choi, J. Kim, and C.-K. Joo, “Activation of MAPK and CREB by GM1

Induces Survival of RGCs in the Retina with Axotomized Nerve,” Invest.

Ophthalmol. Vis. Sci., vol. 44, no. 4, pp. 1747–1752, Apr. 2003.

[62] J. D. Cahoy et al., “A Transcriptome Database for Astrocytes, Neurons, And

Oligodendrocytes: A New Resource for Understanding Brain Development and

Function,” J. Neurosci. Off. J. Soc. Neurosci., vol. 28, no. 1, pp. 264–278, Jan.

2008.

[63] M. Yanagisawa, T. Ariga, and R. K. Yu, “Cholera toxin B Subunit Binding Does

Not Correlate with GM1 Expression: A Study Using Mouse Embryonic Neural

Precursor Cells,” Glycobiology, vol. 16, no. 9, p. 19G–22G, Sep. 2006.

176

[64] C. Flangea, A. Serb, E. Sisu, and A. D. Zamfir, “Chip-Based Nanoelectrospray

Mass Spectrometry of Brain Gangliosides,” Biochim. Biophys. Acta, vol. 1811,

no. 9, pp. 513–535, Sep. 2011.

[65] Y. Sakumoto, H. Ueta, N. Yuki, and K. Matsuno, “Simultaneous

Immunohistochemical Detection of Gangliosides and Neuronal Markers in

Paraformaldehyde-Fixed Nervous Tissues by Acetone Etching,” Arch. Histol.

Cytol., vol. 72, no. 2, pp. 77–90, Jul. 2009.

[66] F. Scandroglio, N. Loberto, M. Valsecchi, V. Chigorno, A. Prinetti, and S.

Sonnino, “Thin Layer Chromatography of Gangliosides,” Glycoconj. J., vol. 26,

no. 8, pp. 961–973, Nov. 2009.

[67] R. K. Yu and T. Ariga, “Ganglioside analysis by high-performance thin-layer

chromatography,” Methods Enzymol., vol. 312, pp. 115–134, 2000.

[68] J. Müthing, “High-Resolution Thin-Layer Chromatography of Gangliosides,” J.

Chromatogr. A, vol. 720, no. 1, pp. 3–25, Jan. 1996.

[69] I. M. Rivas-Serna, R. Polakowski, G. K. Shoemaker, V. C. Mazurak, and M. T.

Clandinin, “Profiling Gangliosides from Milk Products and Other Biological

Membranes Using LC/MS,” J. Food Compos. Anal., vol. 44, no. Supplement C,

pp. 45–55, Dec. 2015.

[70] E. A. Y. Masson, E. Sibille, L. Martine, F. Chaux-Picquet, L. Bretillon, and O.

Berdeaux, “Apprehending Ganglioside Diversity: A Comprehensive

Methodological Approach,” J. Lipid Res., vol. 56, no. 9, pp. 1821–1835, Sep.

2015.

177

[71] J. Zhang, Y. Ren, B. Huang, B. Tao, M. Ransborg Pedersen, and D. Li,

“Determination of Disialoganglioside GD3 and Monosialoganglioside GM3 in

Infant Formulas and Whey Protein Concentrates by Ultra-Performance Liquid

Chromatography/Electrospray Ionization Tandem Mass Spectrometry,” J. Sep.

Sci., vol. 35, no. 8, pp. 937–946, Apr. 2012.

[72] R. L. Shaner et al., “Quantitative Analysis of Sphingolipids for Lipidomics Using

Triple Quadrupole and Quadrupole Linear Ion Trap Mass Spectrometers,” J. Lipid

Res., vol. 50, no. 8, pp. 1692–1707, Aug. 2009.

[73] K. Ikeda, T. Shimizu, and R. Taguchi, “Targeted Analysis of Ganglioside and

Sulfatide Molecular Species by LC/ESI-MS/MS with Theoretically Expanded

Multiple Reaction Monitoring,” J. Lipid Res., vol. 49, no. 12, pp. 2678–2689,

Dec. 2008.

[74] G. Shui, A. K. Bendt, K. Pethe, T. Dick, and M. R. Wenk, “Sensitive Profiling of

Chemically Diverse Bioactive Lipids,” J. Lipid Res., vol. 48, no. 9, pp. 1976–

1984, Sep. 2007.

[75] J. Huang et al., “Glycomic Analysis of High Density Lipoprotein Shows a Highly

Sialylated Particle,” J. Proteome Res., vol. 13, no. 2, pp. 681–691, Feb. 2014.

[76] D. H. Park, L. Wang, P. Pittock, G. Lajoie, and S. N. Whitehead, “Increased

Expression of GM1 Detected by Electrospray Mass Spectrometry in Rat Primary

Embryonic Cortical Neurons Exposed to Glutamate Toxicity,” Anal. Chem., vol.

88, no. 15, pp. 7844–7852, Aug. 2016.

178

[77] B. Fong, C. Norris, E. Lowe, and P. McJarrow, “Liquid Chromatography-High-

Resolution Mass Spectrometry for Quantitative Analysis of Gangliosides,” Lipids,

vol. 44, no. 9, pp. 867–874, Sep. 2009.

[78] L. Hejazi, D. Ebrahimi, M. Guilhaus, and D. B. Hibbert, “Determination of the

Composition of Fatty Acid Mixtures Using GC x FI-MS: A Comprehensive Two-

Dimensional Separation Approach,” Anal. Chem., vol. 81, no. 4, pp. 1450–1458,

Feb. 2009.

[79] D. D. Ju, C. C. Lai, and G. R. Her, “Analysis of Gangliosides by Capillary Zone

Electrophoresis and Capillary Zone Electrophoresis-Electrospray Mass

Spectrometry,” J. Chromatogr. A, vol. 779, no. 1–2, pp. 195–203, Aug. 1997.

[80] A. Zamfir and J. Peter-Katalinić, “Capillary Electrophoresis-Mass Spectrometry

for Glycoscreening in Biomedical Research,” ELECTROPHORESIS, vol. 25, no.

13, pp. 1949–1963, Jul. 2004.

[81] M.-C. Tseng, Y.-R. Chen, and G.-R. Her, “A Low-Makeup Beveled Tip Capillary

Electrophoresis /Electrospray Ionization Mass Spectrometry Interface for Micellar

Electrokinetic Chromatography and Nonvolatile Buffer Capillary

Electrophoresis,” Anal. Chem., vol. 76, no. 21, pp. 6306–6312, Nov. 2004.

[82] P. B. O’Connor and C. E. Costello, “A High Pressure Matrix-Assisted Laser

Desorption/Ionization Fourier Transform Mass Spectrometry Ion Source for

Thermal Stabilization of Labile Biomolecules,” Rapid Commun. Mass Spectrom.,

vol. 15, no. 19, pp. 1862–1868, Oct. 2001.

179

[83] R. Almeida et al., “Coupling of Fully Automated Chip-Based Electrospray

Ionization to High-Capacity Ion Trap Mass Spectrometer for Ganglioside

Analysis,” Anal. Biochem., vol. 378, no. 1, pp. 43–52, Jul. 2008.

[84] H. Park, Y. Zhou, and C. E. Costello, “Direct Analysis of Sialylated or Sulfated

Glycosphingolipids and Other Polar and Neutral Lipids Using TLC-MS/MS

Interfaces,” J. Lipid Res., vol. 55, no. 4, pp. 773–781, Apr. 2014.

[85] S. N. Jackson, H.-Y. J. Wang, and A. S. Woods, “Direct Profiling of Lipid

Distribution in Brain Tissue Using MALDI-TOFMS,” Anal. Chem., vol. 77, no.

14, pp. 4523–4527, Jul. 2005.

[86] W. Metelmann, Z. Vukelić, and J. Peter-Katalinić, “Nano-Electrospray Ionization

Time-of-Flight Mass Spectrometry of Gangliosides from Human Brain Tissue,” J.

Mass Spectrom. JMS, vol. 36, no. 1, pp. 21–29, Jan. 2001.

[87] M. Sarbu, A. C. Robu, R. M. Ghiulai, Ž. Vukelić, D. E. Clemmer, and A. D.

Zamfir, “Electrospray Ionization Ion Mobility Mass Spectrometry of Human

Brain Gangliosides,” Anal. Chem., vol. 88, no. 10, pp. 5166–5178, May 2016.

[88] S. N. Jackson, B. Colsch, T. Egan, E. K. Lewis, J. A. Schultz, and A. S. Woods,

“Gangliosides’ Analysis by MALDI-ion Mobility MS,” The Analyst, vol. 136, no.

3, pp. 463–466, Feb. 2011.

[89] I. M. Rivas-Serna, R. Polakowski, G. K. Shoemaker, V. C. Mazurak, and M. T.

Clandinin, “Profiling Gangliosides from Milk Products and Other Biological

Membranes Using LC/MS,” J. Food Compos. Anal., vol. 44, pp. 45–55, Dec.

2015.

180

[90] J. Gu, C. J. Tifft, and S. J. Soldin, “Simultaneous Quantification of GM1 and

GM2 Gangliosides by Isotope Dilution Tandem Mass Spectrometry,” Clin.

Biochem., vol. 41, no. 6, pp. 413–417, Apr. 2008.

[91] K. Ikeda and R. Taguchi, “Highly Sensitive Localization Analysis of

Gangliosides and Sulfatides Including Structural Isomers in Mouse Cerebellum

Sections by Combination of Laser Microdissection and Hydrophilic Interaction

Liquid Chromatography/Electrospray Ionization Mass Spectrometry with

Theoretically Expanded Multiple Reaction Monitoring,” Rapid Commun. Mass

Spectrom., vol. 24, no. 20, pp. 2957–2965, Oct. 2010.

[92] Z. Vukelić et al., “Human Gliosarcoma-Associated Ganglioside Composition Is

Complex and Distinctive as Evidenced by High-Performance Mass Spectrometric

Determination and Structural Characterization,” Glycobiology, vol. 17, no. 5, pp.

504–515, May 2007.

[93] Ž. Vukelić, M. Zarei, J. Peter-Katalinić, and A. D. Zamfir, “Analysis of Human

Hippocampus Gangliosides by Fully-Automated Chip-Based Nanoelectrospray

Tandem Mass Spectrometry,” J. Chromatogr. A, vol. 1130, no. 2, pp. 238–245,

Oct. 2006.

[94] G. Tettamanti, F. Bonali, S. Marchesini, and V. Zambotti, “A New Procedure for

The Extraction, Purification and Fractionation of Brain Gangliosides,” Biochim.

Biophys. Acta BBA - Lipids Lipid Metab., vol. 296, no. 1, pp. 160–170, Jan.

1973.

181

[95] L. Svennerholm and P. Fredman, “A Procedure for the Quantitative Isolation of

Brain Gangliosides,” Biochim. Biophys. Acta BBA - Lipids Lipid Metab., vol.

617, no. 1, pp. 97–109, Jan. 1980.

[96] Paul De Bièvre1, René Dybkaer, Aleš Fajgelj, and D. Brynn Hibbert.

“Metrological Traceability of Measurement Results in Chemistry: Concepts and

Implementation (IUPAC Technical Report),” Pure Appl. Chem. Vol. 83, No.10,

pp. 1873-1935, 2011.

[97] L. Cuadros-Rodríguez, M. G. Bagur-González, M. Sánchez-Viñas, A. González-

Casado, and A. M. Gómez-Sáez, “Principles of Analytical

Calibration/Quantification for the Separation Sciences,” J. Chromatogr. A, vol.

1158, no. 1, pp. 33–46, Jul. 2007.

[98] P. Palestini, M. Masserini, S. Sonnino, A. Giuliani, and G. Tettamanti, “Changes

in the Ceramide Composition of Rat Forebrain Gangliosides with Age,” J.

Neurochem., vol. 54, no. 1, pp. 230–235, Jan. 1990.

[99] M. Valsecchi, P. Palestini, V. Chigorno, and S. Sonnino, “Age-Related Changes

of the Ganglioside Long-Chain Base Composition in Rat Cerebellum,”

Neurochem. Int., vol. 28, no. 2, pp. 183–187, Feb. 1996.

[100] L. Cantù, M. Corti, S. Sonnino, and G. Tettamanti, “Light Scattering

Measurements On Gangliosides: Dependence of Micellar Properties on Molecular

Structure and Temperature,” Chem. Phys. Lipids, vol. 41, no. 3–4, pp. 315–328,

Nov. 1986.

182

[101] A. Rosenberg and N. Stern, “Changes in Sphingosine and Fatty Acid Components

of the Gangliosides in Developing Rat and Human Brain,” J. Lipid Res., vol. 7,

no. 1, pp. 122–131, Jan. 1966.

[102] S. Sonnino and V. Chigorno, “Ganglioside Molecular Species Containing C18-

And C20-Sphingosine in Mammalian Nervous Tissues and Neuronal Cell

Cultures,” Biochim. Biophys. Acta BBA - Rev. Biomembr., vol. 1469, no. 2, pp.

63–77, Sep. 2000.

[103] H. C. Yohe, D. E. Roark, and A. Rosenberg, “C20-Sphingosine as A Determining

Factor in Aggregation of Gangliosides,” J. Biol. Chem., vol. 251, no. 22, pp.

7083–7087, Nov. 1976.

[104] M. Masserini, P. Palestini, M. Pitto, V. Chigorno, M. Tomasi, and G. Tettamanti,

“Cyclic AMP accumulation in HeLa cells induced by cholera toxin. Involvement

of the ceramide moiety of GM1 ganglioside,” Biochem. J., vol. 271, no. 1, pp.

107–111, Oct. 1990.

[105] M. Valsecchi, V. Chigorno, S. Sonnino, and G. Tettamanti, “Rat Cerebellar

Granule Cells in Culture Associate and Metabolize Differently Exogenous GM1

Ganglioside Molecular Species Containing a C18 or C20 Long Chain Base,”

Chem. Phys. Lipids, vol. 60, no. 3, pp. 247–252, Feb. 1992.

[106] A. Hara and T. Taketomi, “Long Chain Base and Fatty Acid Compositions of

Equine Kidney Sphingolipids,” J. Biochem. (Tokyo), vol. 78, no. 3, pp. 527–536,

Sep. 1975.

183

[107] A. Serb, C. Schiopu, C. Flangea, E. Sisu, and A. D. Zamfir, “Top–down

Glycolipidomics: Fragmentation Analysis of Ganglioside Oligosaccharide Core

and Ceramide Moiety by Chip-Nanoelectrospray Collision-Induced Dissociation

MS2–MS6,” J. Mass Spectrom., vol. 44, no. 10, pp. 1434–1442, Oct. 2009.

[108] L. K. Sørensen, “A Liquid Chromatography/Tandem Mass Spectrometric

Approach for The Determination of Gangliosides GD3 and GM3 in Bovine Milk

and Infant Formulae,” Rapid Commun. Mass Spectrom., vol. 20, no. 24, pp.

3625–3633, Dec. 2006.

[109] A. D. Garcia, J. L. Chavez, and Y. Mechref, “Rapid and Sensitive LC-ESI-MS of

Gangliosides,” J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci., vol. 947–

948, pp. 1–7, Feb. 2014.

184

APPENDIX

GM1-1573.90 1600000

1400000 y = 1382.9x + 2730.7 R² = 0.9999 1200000

1000000

800000

Peak Area Peak 600000

400000

200000

0 0 200 400 600 800 1000 1200 Concentration (ng/mL)

Figure A1: Calibration standard curve of GM1 ganglioside, 1573.90

185

GD1-1836.80 10000000

9000000 y = 9250.4x - 139447 R² = 0.9994 8000000

7000000 6000000 5000000 4000000 Area Peak 3000000

2000000

1000000

0 0 200 400 600 800 1000 1200 Concentration (ng/mL)

Figure A2: Calibration standard curve of GD1 ganglioside, 1836.80

186

GD1-1865.00

10000000 y = 7206.7x - 71930 9000000 R² = 0.9981 8000000

7000000

6000000 5000000

4000000 Area Peak 3000000 2000000

1000000

0 0 200 400 600 800 1000 1200 1400 Concentration (ng/mL)

Figure A3: Calibration standard curve of GD1 ganglioside, 1865.00

187

GT1-2128.80 16000000 y = 4277.6x - 158100 14000000 R² = 0.9991

12000000

10000000

8000000

6000000 area Peak 4000000

2000000

0 0 500 1000 1500 2000 2500 3000 3500 4000 -2000000 Concentration (ng/mL)

Figure A4: Calibration standard curve of GT1 ganglioside, 2128.80

188

GT1-2156.40

30000000

25000000 y = 4563.6x - 264072 R² = 0.9995

20000000

15000000

Peak Area Peak 10000000

5000000

0 0 1000 2000 3000 4000 5000 6000 Concentration (ng/mL)

Figure A5: Calibration standard curve of GT1 ganglioside, 2156.40

189

GQ1-2417.60

6000000 y = 1524.2x - 62075 5000000 R² = 0.9995

4000000

3000000

2000000

Area Peak 1000000

0 0 500 1000 1500 2000 2500 3000 3500 4000

-1000000 Concentration (ng/mL)

Figure A6: Calibration standard curve of GQ1 ganglioside, 2417.60

190

GQ1-2446.40 9000000

8000000 y = 1458.2x - 109348 R² = 0.9984 7000000 6000000 5000000 4000000 3000000 Area Peak 2000000

1000000

0 0 1000 2000 3000 4000 5000 6000 -1000000 Concentration (ng/mL)

Figure A7: Calibration standard curve of GQ1 ganglioside, 2446.40

191