Abbreviations - Deakin University30023241/glenister-phospholipas… · Fellow PhD. Students, in...

parisr

Redacted stamp

Acknowledgements

My supervisors:

Dr. Charlie Corke. Thankyou for your never ending enthusiasm, support and

fantastic opportunities over the past five years.

Associate Professor Tom Watson, thanks for all of your support and encouragement.

Thanks for sparking my interest in biochemistry all those years ago. Thanks also for

playing the devil’s advocate in research meetings!

Laboratory assistance:

Dr. Kieran Scott, Megan Taberner, Dr. Katherine Bryant, Garvan Institute of Medical

Research. Thankyou, particularly to Kieran for your guidance, and help over many

years.

Australian Proteome Analysis Facility, in particular David Basseal and Dr. Stuart

Cordwell, thankyou for the opportunity to learn Proteomic techniques in such a

professional environment.

Dr. Wojtek Michalski, Brian Shiell, Dr. Mark Lanigan, Gary Beddome, Megan

Retallick, Australian Animal Health Laboratories, CSIRO. Thankyou for your

support over an extended period of time, and for making me feel like another member

of the lab.

Dr. Mark Raftery, Dr. Valerie Wassinger, BMSF, UNSW. Thankyou for the mass

spectrometry work.

Dr. Peter Hoffman, Professor Bruce Kemp, Sid Murthy, St. Vincent’s Institute of

Medical Research. Thankyou for the opportunity to use your mass spectrometer.

Rosemay Condron, La Trobe University. Thankyou for early sequencing attempts.

Carolina Lopez, Matt Constable, Cathy Aitken, Jason Hodge, Department of Clinical

and Biomedical Sciences, thankyou for the loan of various bits of equipment!

Personal Communications:

Professor Gregory Bulkley, for personal communications relating to

ischaemia/reperfusion injury.

IV

Acknowledgements

Dr. Ben Herbert, for personal communications relating to low abundance proteins and

Proteomics.

Dr. Margaret Henry, for personal communications relating to multivariate analysis.

Dr. Bruce Kemp, for personal communication relating to mass spectrometry of silver

stained proteins.

Phospholipase A2 practical assistance:

Dr. Tanweer Ahmad, Leeds University, thanks for the thesis, and ideas!

Dr. Anthony Lawrence, University of Glasgow, thankyou for ideas and hints.

Dr. David Wilton, University of Southampton, thanks for the problem solving session.

Laboratory members:

Fiona Collier, Caryll Waugh, Courtney Talbot, Dr. Liem Vo, Dr. Claudia Gregorio

King, Associate Professor Mark Kirkland, Karyn Bolton, Gavin van Der Meer, Dr.

Janet McLeod, Jane Hosking, Paul Fell, Sarah Roberts, Tamara Gough, Douglas

Hocking Research Institute. Thanks for everything!

Mark Rigby, Joanne Spence, Emilio Baldonado, Garth Stephenson, Andrew Krich,

Michael Lovelace, thanks for the laughs.

Other assistance:

Helen, Rosemary and Ruth, Intensive Care Unit, Geelong Hospital, Barwon Health.

Surgical Staff, Geelong Hospital, Barwon Health.

Dr. Harry Armstrong, Pathcare.

Peter Gumley, Pathcare.

Financial support:

Deakin University Postgraduate Association.

Geelong and Region Medical Research Foundation.

Moral support and Motivation:

Dr. Neil Barnett and Jane Pappin, for moral support over a number of years.

V

Acknowledgements

Debbie and Geoff Gill, the teachers and the kids at Geelong Aquatic Centre, thankyou

for a different perspective on life, and your support over the past five years.

Fellow PhD. Students, in particular Rachel, Pete, Benge and Emma, thanks for your

discussions and help. Thanks Caz for the emails, coffee and late night microscope

assistance.

My housemates (and mates) Weasel, Meaghan, Clairey, Rachelle, Tania, Michael and

Ellie, thanks for putting up with me!

My family: Mum, Dad, Jacqui, Dave, Grandpa G., Grandma and Grandpa W, where

do I start? Thanks for your support, I couldn’t have done it without you!

My friends: Lisa, Peter, Andrew, Jane P. & Adam, Jane W., Benge, Emma, Kit, Matt,

Rachel, Caz, Rachelle, Claire, Tania, Raelene, Jenny, Megan R., Heleen, Megan

McG., Meaghan & Ben, Kim, thanks for everything.

VI

Table of contents and figures.

Table of Contents: SECTION: Pages: Acknowledgements IV-VI Table of Contents VII List of Figures VIII-IX List of Tables X Abbreviations XI-XIII List of Publications arising from this Research XIV-XV Abstract XVI-XVIII Thesis summary XIX Aims of study XX CHAPTER ONE Introduction and Literature Review 1-18 CHAPTER TWO Methods section 19-41 CHAPTER THREE Plasma Proteomics

Introduction 42-60 Results 61-74 Discussion 75-82

CHAPTER FOUR Phospholipase activity in Mesenteric Ischaemia/Infarction


CHAPTER FIVE Protein Purification


CHAPTER SIX Cyclophilin B


CHAPTER SEVEN Alkaline Urea PAGE


Conclusions and Future Directions 202-212 Bibliography 213-236

VII


Table of Figures: Figure number: Description: Page: 1.1 The axis of ischaemic damage 4 1.2 The development of ischaemia/reperfusion injury. 5 3.1 The Plasma Proteome. 50 3.2 2D gels of control patients’ plasma proteins. 63 3.3 2D gels of bowel infarction patients. 64 3.4 A typical 2D gel with annotated protein identities. 65 3.5 Relative quantities of proteins of interest. 66 3.6 Serum amyloid A identification. 68 3.7 Sequence alignment of Serum Amyloid A subtypes. 69 3.8 Serum Amyloid A variant quantities. 70 3.9 Serum Amyloid A peptide functions. 71 3.10 PLA2 activity in plasma. 72 3.11 Five plasma variables. 73 3.12 ROC plot of five plasma variables. 74 3.13 The Acute Phase Proteins. 81 4.1 Sites of phospholipase activity on

a generalised phospholipid. 83 4.2 Products of phospholipase A2 hydrolysis. 86 4.3 Immunoprecipitation of PLA2-IIA from recombinant PLA2

standard. 97 4.4 Immunoprecipitation of PLA2 from human bowel. 98 4.5 Western blotting of PLA2 from human bowel. 99 4.6 PLA2 activity in infarcted bowel tissue and lumen content. 100 4.7 PLA2 activity in infarcted and ischaemic bowel tissue. 101 4.8 PLA2 activity versus tissue damage score. 102 5.1 Protein purification process 1. 118 5.2 Protein purification process 2. 119 5.3 Protein purification process 3. 120 5.4 Absence of PLA2 activity in haemoglobin. 121 5.5 Absence of PLA2 activity in haemoglobin & buffers. 122 5.6 Absence of PLA2 enhancing effects of haemoglobin. 123 5.7 Heparin binding of PLA2 isoforms. 124 5.8 Sequence alignment of sPLA2 isoforms. 125 5.9 Typical elution profile of total protein from a heparin affinity

column. 127 5.10 Elution of total protein and PLA2 activity from heparin affinity

column. 128 5.11 Effect of column volume on protein eluted. 129 5.12 Effect of dialysis. 130 5.13 Non-interference of Rotofor buffer in PLA2 assay. 131 5.14 Typical running conditions during a Rotofor run. 132 5.15 The pH and PLA2 profiles of a typical Rotofor experiment. 133 5.16 SDS PAGE gel of active fractions from a typical Rotofor

experiment. 134 5.17 Rotofor system reproducibility. 135 5.18 Rotofor refractionation. 136 5.19 Determination of isoelectric point by refractionation. 137

VIII


5.20 Electroelution of the protein of interest. 138 5.21 Optimal pH of the protein of interest. 139 5.22 Purification overview. 140 5.23 Silver stained SDS gel of protein of interest. 141 5.24 Protein purification system 4, the final purification process. 142 5.25 Mass spectrometry of the unknown protein, BisTris gel. 143 5.26 Mass spectrometry of the unknown protein, TrisGly gel. 144 5.27 Concentration of the protein of interest. 145 6.1 SDS gel of Jurkat cell lysate and cell culture medium. 159 6.2 Western blot of Cyclophilin B from infarcted human bowel

tissue. 160 6.3 Western blot of Cyclophilin B from infarcted human bowel

tissue. 161 6.4 Western blot of Cyclophilin B from normal human bowel

tissue. 162 6.5 Western blot of Cyclophilin B from infarcted human bowel

lumen samples. 163 6.6 Phospholipase activity before and after immunoprecipitation of cyclophilin B. 164 6.7 Western blot of Cyclophilin B from plasma. 165 6.8 Amino acid sequence of human Cyclophilin B. 166 6.9 Immunohistochemistry of cyclophilin B in human bowel. 167 6.10 Cyclosporin A does not alter phospholipase activity. 169 6.11 Cyclophilins and pathological pore opening. 175 7.1 Electrotransfer of proteins, appearance of membranes. 181 7.2 Electrotransfer of proteins, appearance of gels. 182 7.3 Transfer efficiency of CAPS and Towbin transfer buffers. 183 7.4 Amino acid sequencing of electrotransferred protein. 184 7.5 Running temperature of modified alkaline urea gels. 185 7.6 Protein mobility in modified alkaline urea gel. 186 7.7 Extraction of crude venom for SDS PAGE. 187 7.8 Protein size estimation. 188 7.9 Carbamylation of bee venom PLA2. 189 7.10 Alkaline urea gel of haemoglobin. 190 7.11 Extraction of PLA2 activity from alkaline urea gels. 191 7.12 PLA2 activity of tiger snake venom proteins. 192 7.13 PLA2 isoform mobility in alkaline urea gels. 193 7.14 Infarcted and normal bowel PLA2 protein mobility in alkaline

urea gels. 194 7.15 Alkaline urea gels of partially purified and crude bowel

phospholipase. 195 7.16 Alkaline urea gels of normal bowel and synovial fluid

phospholipase. 196 7.17 Effect of sodium chloride concentration on protein mobility.

197

IX


List of Tables: Table number: Description: Page(s): 1.1 Clinically assessed markers of intestinal ischaemia

and infarction. 9-11 1.2 Animal models of mesenteric ischaemia

and infarction. 12-13 2.1 Gross tissue rating system of Sun et al, 1997. 28 3.1 Advantages and disadvantages of nucleic acid

based global techniques. 42 3.2 Advantages and disadvantages of protein

based global techniques. 43 3.3 Acute phase Serum Amyloid Variants. 53 3.4 Protein spots present in 2D gels of plasma. 67 4.1 Phospholipase A2 in Intestinal Disease. 83 5.1 Purification of PLA2 using heparin

affinity chromatography. 111 5.2 Heparin affinity of PLA2 isoforms. 112 6.1 The Major Human Cyclophilins. 154

X

Abbreviations

2DE; two dimensional electrophoresis

ALP/AP: alkaline phosphatase

APACHE: acute physiology and chronic health evaluation

APS: Ammonium persulfate

ASB14: tetra decanol amidopropyl dimethylammonio propane sulfonate

ALT: alanine amino transferase

ARDS: adult respiratory distress syndrome

AST: aspartate amino transferase, formerly known as SGOT

BCA: Bicinochoninic Acid

BIS: N,N’-Methylene-bis-acrylamide

BMSF: Biomedical Mass Spectrometry Facility, University of New South Wales

BSA: Bovine serum albumin

CAPS: 3-[cyclohexylamino]-1-propanesulphonic acid

CBB: Coomassie brilliant blue

cDNA: complementary deoxyribonucleic acid

CHAPS: (3-[(3-cholamidopropyl)dimethylammonio]-1-propane sulfonate)

CHO: Chinese hamster ovary

CK: creatine kinase

CNBr: Cyanogen bromide

CRP: C-reactive protein

CsA: cyclosporin A

CT: computed tomography

CyPB: cyclophilin B

CyPB-CsA: cyclophilin B-cyclosporin A complex

DC: detergent compatible

DD-PCR: differential display polymerase chain reaction

DTE: dithioerythritol

DTT: dithiothreitol

ECL: enhanced chemiluminescence

EDTA: Ethylene diamine tetra-acetic acid

ELISA: Enzyme linked immunosorbent assay

ESI: electrospray ionization

EST: expressed sequence tag

GGT: γ glutamyl transferase

XI

Abbreviations

HCl: hydrochloric acid

H&E: Hematoxylin and eosin

HEK: human embryonic kidney

HPLC: High performance liquid chromatography

HRP: horseradish peroxidase

HUPO: human proteome organisation

IAA: iodoacetamide

ICU: intensive care unit

IEF: iso-electric focusing

I-FABP: intestinal fatty acid binding protein

IL 1, 6, 8: interlukin 1, 6, 8

IP: immunoprecipitation

IPG: immobilised pH gradient

I/R: ischaemia reperfusion

kDa: kilo Dalton

KLH: keyhole limpet hemocyanin

LC/MS/MS: liquid chromatography- tandem mass spectrometry

LDH: lactate dehydrogenase

LDS: lithium dodecylsulphate

MALDI-TOF: matrix assisted laser desorption ionization-time of flight

MDA: malondialdehyde

MES: 2-(N-morpholino)ethane sulphonic acid

MOF: multiple organ failure

MODS: multiple organ dysfunction syndrome

MPO: myeloperoxidase

MRI: magnetic resonance imaging

mRNA: messenger ribonucleic acid

MS: mass spectrometry

MS/MS: tandem mass spectrometry

MWCO: molecular weight cut off

NOGS: N-Octyl β-D-glucopyranoside or n-Octyl glucoside

NOMI: non-occlusive mesenteric ischaemia

PAF: platelet activating factor

PAGE: polyacrylamide gel electrophoresis

XII

Abbreviations

PBS: Phosphate buffered saline

PC: Phosphatidylcholine

pCO2: partial pressure carbon dioxide

PE: Phosphatidylethanolamine

PES: polyethersulphone

pI: isoelectric point

PLA2: Phospholipase A2

PMN: polymorphonuclear neutrophil

PMSF: Phenyl methyl sulfonyl fluoride

PNPP: p-nitrophenyl phosphate

PPO: Diphenyl oxazole

P&T: phloxine and tartrazine

PVDF: polyvinyl difluoride

ROC: receiver operating characteristic curve

RP-HPLC: reversed phase high performance liquid chromatography

RT-PCR: reverse trancriptase polymerase chain reaction

SAA: serum amyloid A

SDS: Sodium dodecyl sulfate or lauryl sulfate

SDS PAGE: Sodium dodecyl sulfate polyacrylamide gel electrophoresis

SGOT: serum glutamic oxaloacetic transaminase

sPLA2: secretory phospholipase A2

TBS: Tris buffered saline

TBST: Tris buffered saline, Tween-20

TCA: trichloroacetic acid

TEMED: N,N,N’,N’ tetramethyl ethylene diamine

TFA: trifluroacetic acid

TLC: Thin layer chromatography

TNFα: tumour necrosis factor α

TRICINE: N-tris[hydroxymethyl] methyl glycine

TRIS: Tris (hydroxy methyl) methylamine

XIII

Publications

Papers:

1. Corke C. and Glenister K., Monitoring intestinal ischaemia. Crit. Care Resuscitation,

2001. 3(3): p. 176-180.

2. Corke C., Glenister K., and Watson T., Circulating secretory phospholipase A2 in

critical illness- the importance of the intestine. Crit. Care Resuscitation, 2001. 3(4): p.

244-249.

3. Kristen M. Glenister, Charlie F. Corke. The Infarcted Intestine: a Diagnostic Void.

ANZ J. Surg. , 2004. 74 (4): p. 260-265.

4. Kristen M. Glenister, Charlie F. Corke (2003). Extensions of alkaline urea gels for

venom proteins. (in preparation).

Abstracts:

Australian Society of Medical Research, 2000:

Phospholipase A2 activity as a marker of bowel ischaemia.

Kristen Glenister, Charlie Corke, and Tom Watson, (2000).

Lorne Protein Structure and Function, 2001:

Phospholipase A2 isoenzymes as diagnostic markers of Bowel Ischaemia/Infarction.

K. Glenister, C.Corke and T. Watson (2001).

Australian and New Zealand Intensive Care Society, 2002:

Kristen Glenister, Charlie Corke & Tom Watson, (2002).

PLA2 in intestinal Infarction- a key to intestinal involvement in critical illness?

Melbourne protein group, Melbourne, 2002:

Glenister, K.M., Corke, C.F. and Watson, T.G. (2002).

Phospholipase A2 in Infarcted Human Bowel.

XIV

Publications

ComBio, Sydney, 2002:

Glenister, K.M., Corke, C.F. and Watson, T.G. (2002).

Phospholipase A2 in Infarcted Human Bowel.

Melbourne Protein group, Melbourne, 2003:

Kristen Glenister and Charlie Corke (2003).

Improved method for separation of basic venom proteins: alkaline urea PAGE.

ComBio, Melbourne, 2003:

Kristen Glenister and Charlie Corke (2003).

Improved method for separation of basic venom proteins: alkaline urea PAGE.

XV

Abstract

Currently, diagnostic tests for mesenteric ischaemia and infarction are inadequate due

to poor sensitivity and specificity. In addition, many potential markers appear too

late to be clinically useful. At present, definitive diagnosis can only be made at the

time of surgery, which is not ideal as surgery is often to be avoided in critically ill and

elderly patients. A clinically useful, minimally invasive test is likely to decrease the

currently very high mortality rate and allow monitoring of ‘at risk’ patients during

their hospital stay.

A two-dimensional electrophoresis based proteomic approach was undertaken to

assess plasma protein differences between patients with surgically confirmed bowel

infarction and control Intensive Care patients. The major protein differences were

found to be members or variants of acute phase proteins. Serum amyloid A showed

the largest difference between the two patient groups, and this protein was

investigated in greater depth. An analysis was performed to compare the diagnostic

ability of several commonly used indicators of critical illness and bowel infarction

with serum amyloid A and phospholipase A2. Although none of the variables were

ideal for clinical use, plasma phospholipase A2 activity showed the best

discriminatory power, as determined by Receiver Operating Characteristic curves.

From a review of the literature, phospholipase A2 (PLA2) appeared to be increased in

the bowel as a result of ischaemia and infarction. In one patient, matched tissues

were obtained, and PLA2 activity was found to be significantly higher in infarcted

bowel tissue compared to ischaemic bowel tissue. PLA2 activity was significantly

greater in bowel lumen than tissue, suggesting that the protein was being released, and

may enter the circulation. PLA2 activity was increased in the plasma of bowel

infarction patients compared with control patients, though the difference was not

significant. The phospholipase activity exhibited a number of similarities to typical

phospholipase A2 proteins, but also showed a number of inconsistent characteristics.

For this reason, we wished to identify the protein responsible for the increased

phospholipase activity in infarcted human bowel.

The PLA2 activity in human bowel could not be abolished by immunoprecipitation of

the PLA2 isoforms IIA (well described in bowel) and V (a closely related isoform).

To investigate these proteins, a native urea protein gel devised for snake venom

XVI

Abstract

phospholipase A2 was modified for use with mammalian phospholipase A2. The

modified gel was used to show that the protein with phospholipase activity from

infarcted gut was different from normal gut PLA2 and type IIA PLA2. A number of

extensions were devised for these native gels and were found to be useful both in this

investigation and for venom investigations.

Protein purification was undertaken to identify the protein responsible for the

increased phospholipase activity in infarcted bowel. Protein was purified from

infarcted human bowel using a number of techniques that exploited unusual

characteristics of the protein. The purification techniques each retained the native

activity of the protein and the purification could therefore be monitored with a

phospholipid hydrolysis assay at each stage.

The protein identified by mass spectrometry was an excellent match for cyclophilin B,

an inflammatory protein that had previously been identified in rat bowel at the mRNA

level (Hasel et al, 1991, Kainer & Doris, 2000). As the purification progress had

been monitored throughout with a phospholipid hydrolysis assay, cyclophilin B was

an unexpected identification, as it is not known to have phospholipase activity.

Cyclophilin B was removed from the highly purified samples via

immunoprecipitation and this process abolished all phospholipase activity. The

addition of cyclosporin A, (the pharmaceutical ligand of cyclophilin B), did not effect

the phospholipase activity. Cyclophilin B protein was found in normal and infarcted

human bowel using Western blotting. Cyclophilin B protein also appeared to be

present in the bowel lumen and plasma of several patients with bowel infarction, but

not in control patients. Immunohistochemistry confirmed the ubiquitous nature of

cyclophilin B that had been reported by other groups.

This project has investigated the use of two dimensional gel electrophoresis based

proteomics to identify proteins present in the plasma of patients with confirmed bowel

infarction and control intensive care patients. The major protein classes observed

were members of the acute phase proteins, which highlights the need for pre-

fractionation of plasma to identify lower abundance, disease associated proteins. A

series of potential plasma markers were compared using Receiver Operating

Characteristic Curves. Although no ideal marker was clear from this analysis,

XVII

Abstract

phospholipase activity appeared to warrant further investigation. Phospholipase

activity was investigated in human infarcted bowel. Protein purification identified

cyclophilin B as a bowel protein that showed unusual phospholipid hydrolysing

activity. Cyclophilin B is a ubiquitous protein in intestinal cell types in both normal

and infarcted tissue. There appears to be release of cyclophilin B into bowel lumen

and plasma under conditions of mesenteric ischaemia and infarction.

XVIII

Thesis Summary

DEAKIN UNIVERSITY.

Summary of Thesis Submitted for the Degree of:

DOCTOR OF PHILOSOPHY

Title of Thesis:

Phospholipase Activity in Human Mesenteric Ischaemia and Infarction. Diagnostic tests for mesenteric ischaemia and infarction are currently inadequate due

to low sensitivity and specificity. Many potential markers of mesenteric ischaemia

and infarction have been proposed but are not clinically useful as they appear too late

in the disease development. Proteomic techniques have great potential in medicine to

identify diagnostic markers and potential drug targets. Proteomic investigations have

previously identified proteins differentially expressed between benign and cancerous

tissues.

This thesis examines the proteins in the plasma of patients with surgically confirmed

bowel infarction and control Intensive Care patients by two dimensional

electrophoresis proteomics. A number of disease associated proteins are investigated

in greater depth using appropriate assays. This thesis examines the protein

purification of one of the proteins of interest, and the examination of this protein in

human mesenteric ischaemia and infarction. The investigations suggest that the

levels of a number of these disease associated proteins are increased during

mesenteric ischaemia and infarction in the bowel and in plasma.

Name: Kristen Glenister Signed: Date: Supervisors: Dr. Charlie Corke and Associate Professor Tom Watson.

XIX

Aims of this Study

The aims of this study are as follows:

1. To use Proteomics to assess whether an abundant and specific disease

associated protein was evident in the plasma of patients with confirmed bowel

infarction.

2. To assess the potential of a variety of disease associated proteins as clinically

useful diagnostic markers of mesenteric ischaemia and infarction.

3. To study potential disease associated proteins in greater depth in relation to

mesenteric ischaemia and infarction using appropriate biochemical and

antibody assays and protein purification techniques where necessary.

4. To develop or refine techniques that are useful to the study of disease

associated proteins where necessary.

XX

Chapter 1 Introduction

Chapter 1: Clinical introduction:

In 1926 A.J. Cokkinis stated starkly:

“occlusion of the mesenteric vessels is apt to be regarded as one of those conditions of which

diagnosis is impossible, the prognosis hopeless and the treatment almost useless”.

Despite the significant medical advances that have been made since this statement, the

problems associated with the diagnosis and prognosis of mesenteric ischaemia remain

substantial. Ischaemia results from the insufficient blood supply to an organ or tissue, in the

case of mesenteric ischaemia the organ affected is the intestine. If the ischaemia persists,

tissue necrosis, or infarction will invariably follow. Infarction of the intestine often leads to

gangrene and perforation of the gut wall.

As the period of ischaemia increases, the prognosis becomes more grave, and by the time

infarction has occurred, the condition has a 50-90% mortality rate during hospital treatment

(Aydin et al, 1998, Flinn & Bergan, 1997, Howard et al, 1996, Klempnauer et al, 1997).

Interestingly, not long after Cokkinis made his grave synopsis the mortality rate due to

mesenteric infarction was 70% (Hibbard, Swenson & Levin, 1931), which has not changed

significantly in the past 70 years. Early diagnosis is critical for effective treatment, and to

keep the mortality rate as low as possible but this remains difficult due to vague symptoms,

lack of clinically useful diagnostic tests (Bradbury et al, 1995) and broad groups of ‘at risk’

patients.

The occurrence of mesenteric ischaemia/infarction is quite low in comparison to other

circulatory disorders (Flinn & Bergan, 1997), accounting for approximately 0.1-0.2% of

hospital admissions (Aydin et al, 1998, Howard et al, 1996, Wadman, Syk & Elmstahl, 2000).

Mesenteric infarction can occur as a complication of hospitalization. In a study spanning six

years, Newman and colleagues (1998) found that a quarter of mesenteric infarction cases

occured as secondary events after hospitalisation, and, in the majority of cases, was the cause

of death. Approximately one half of these cases were due to non-occlusive mesenteric

ischaemia or NOMI (Newman et al, 1998). NOMI is a particularly severe condition due to a

high mortality rate and lack of any treatment other than resection of necrotic tissue. NOMI is

associated with a high mortality rate due to the critical nature of the illness and typically

advanced age of the patients involved (Boley et al, 1977). The use of drugs to reverse

1


hypotension in shock patients in ICU is common, but is likely to compromise blood supply to

the small bowel (Boley, Brandt & Sammartano, 1997, Newman et al, 1998).

Elderly people, particularly those with cardiac conditions, are most at risk of mesenteric

ischaemia. When considering the incidence of this condition, other factors must be

considered:

- the number of recognised cases has increased over the past few decades due to increased

awareness of the condition and an increased number of intensive care units (Bradbury et al,

1995, Rogers & David, 1995),

- the reported incidence is likely to be a substantial underestimation, as up to half of cases are

only detected at autopsy (Wilson et al, 1987), and the rates of autopsy in Victorian hospitals

have decreased by approximately 50% since the introduction of the Human Tissue Act of

1982 (Vic.), (McKelvie & Rode, 1992),

- the incidence is likely to continue to increase with the ageing population,

- patients with ‘low grade sepsis’ may have sub-clinical mesenteric ischaemia, particularly

those that later develop Multiple Organ Dysfunction Syndrome (MODS) (Montgomery,

Venbrux & Bulkley, 1997). The prevalence of sub-clinical mesenteric ischaemia cannot be

assessed without accurate diagnostic tests.

Mesenteric ischaemia can be divided into chronic or acute conditions. Chronic mesenteric

ischaemia is rare (Flinn & Bergan, 1997, Grendell & Ockner, 1989). This condition is

associated with marked weight loss due to avoidance of food as eating aggravates the

condition (Grendell & Ockner, 1989). Acute mesenteric ischaemia can be categorised as

being due to arterial occlusion (due to an embolus or thrombosis) or venous occlusion

(thrombosis or strangulation), or alternatively may be non occlusive in nature as a result of

inadequate blood flow (NOMI).

The high mortality rate is the result of not only the severity of the disease and its tendency to

cause remote organ injury (Montgomery, Venbrux & Bulkley, 1997) but also due to:

- the failure to make a diagnosis until infarction has already occurred

- the tendency of the gut to develop signs of ischaemic damage or infarction even after the

vascular problems have been corrected (Rogers & David, 1995)

2


- the proportion of NOMI cases, where the condition reflects poor general circulatory

perfusion and cannot be treated until infarction has occurred (Boley, Brandt & Sammartano,

1997).

The question of whether the gut is particularly susceptible to ischaemic episodes is

controversial. Villi tips have been observed to be hypoxic even under normal perfusion

states, due to short circuiting of oxygenated blood at the base of the villi (Kong et al, 1998).

The mucosa is much more dependent on constant blood supply than the underlying

connective tissue layers (Marshak & Lindner, 1967). In a normal resting state the mesenteric

vascular bed holds a third of the body’s total blood volume and receives a quarter of the

cardiac output (Schoenberg & Berger, 1993). Under conditions of low blood volume or

pressure this reserve is called upon, and for this reason many clinicians consider the bowel

susceptible to ischaemia during shock states (Deitch, 1992, Koike et al, 1992, Mansour, 1999,

Montgomery, Venbrux & Bulkley, 1997). The gut is thought to show poor tolerance to

hypoxic insult (Mansour, 1999).

Other groups consider the gut to be resistant to hypoxic damage. The bowel wall is capable

of extracting extra oxygen under conditions of decreased blood flow (Patel, Kaleya &

Sammartano, 1992, Rogers & David, 1995) and the splanchnic region in general has a high

capacity to increase oxygen extraction as required (Rowell et al, 1984). In an animal model

of septic shock, the microcirculation to the jejunal mucosa has been shown to remain constant

despite systemic blood flow reduction of approximately 50% (Hiltebrand et al, 2000).

Signs and symptoms of mesenteric ischaemia:

The signs and symptoms associated with mesenteric ischaemia are characteristically vague,

non specific and can vary greatly between patients. The first sign may be severe abdominal

pain which is perceived by the clinician as being ‘out of proportion to physical findings’ on

abdominal examination (Gredell & Ockner, 1989). Other signs that suggest the diagnoses

are an elevated white blood cell count, vomiting, diminished or hyperactive bowel sounds,

tachycardia and hypotension, however, all of these features are non-specific.

Intestinal pathophysiology in response to ischaemia:

The ischaemic bowel initially turns a paler colour (Rogers & David, 1995), but darkens as the

ischaemic period progresses (Carter & Camilleri, 1997). Affected regions of the bowel wall

3


then haemorrhage and appear purple to blackish red. The mucosa ulcerates and appears white

to green (Carter & Camilleri, 1997). The mucosa sloughs off into the lumen, which can

cause gastrointestinal bleeding (Flinn & Bergan, 1997). After 24 to 48 hours the bowel

becomes thin and friable (Carter & Camilleri, 1997). The lumen fills with blood and mucous

and ultimately perforates.

At a microscopic level, damage due to ischaemia begins at the villi tips and this superficial

damage is evident within minutes of occlusion (Carter & Camilleri, 1997). Histological signs

such as mucosal cell necrosis and submucosal venous congestion are evident within an hour

(Rogers & David, 1995). As the ischaemic insult continues damage progresses deeper into

the mucosa (Montgomery, Venbrux & Bulkley, 1997), see Figure 1.1 below.

Figure 1.1 The axis of ischaemic injury.

Mucosal damage during ischaemia occurs from the villi tips in toward the mucosa.

Acknowledgement to Jacqui Glenister for the design and drawing of this figure.

4


Biochemical level changes during ischaemia:

The insufficient blood supply to the gut during mesenteric ischaemia causes significant

cellular damage, but the resumption of blood supply may cause further damage (reperfusion

injury), see Figure 1.2. This became more evident when increasing numbers of patients who

had successful revascularisation of their gut developed intestinal and cardiopulmonary

problems post surgery (Schoenberg & Beger, 1993).

Figure 1.2. The development of Ischaemia Reperfusion injury.

Modified from Bulkley G.B. Free radical-mediated reperfusion injury: A selective

review. Br. J. Cancer 1987; 55: 66-73, with permission from author and Nature

Publishing Group.

If the period of ischaemia is relatively short, the damage caused by reperfusion is greater than

that caused by the ischaemia (Bradbury et al, 1995). As the period of ischaemia increases the

ischaemic component becomes more detrimental than reperfusion, and when infarction

occurs, there is no effect due to reperfusion (Bradbury et al, 1995).

5


During ischaemia xanthine dehydrogenase is converted to xanthine oxidase (Montgomery,

Venbrux & Bulkley 1997). When oxygen supply is restored the purines which have

accumulated during ischaemia are metabolised by xanthine oxidase (which requires oxygen,

unlike its precursor) (Montgomery, Venbrux & Bulkley 1997) forming excess amount of

tissue damaging superoxide O2•-. Reperfusion injury is thought to be primarily due to

reactive oxygen species and increased activity of phospholipase A2 (Schoenberg & Beger,

1993). Free radicals can cause cell membrane damage, mucosal edema and ulceration

(Schoenberg & Beger, 1993). Phospholipase A2 hydrolyses phospholipids to produce fatty

acids and the corresponding lysophospholipids. Lysophospholipids can cause cellular

damage, for example, lysophosphatidylcholine (LysoPC) is known to be cytotoxic

(Schoenberg & Beger, 1993) and increases intestinal permeability (Otamiri et al, 1987).

Inhibitors of phospholipase A2 have been shown to be protective against reperfusion damage

to gut tissue (Otamiri et al, 1987, Otamiri & Tagesson, 1989).

Efforts to limit tissue damage from reperfusion have been unsuccessful, which is likely to be

due to the lengthy and severe periods of hypoxia seen in clinical situations (Montgomery,

Venbrux & Bulkley, 1997).

Current diagnosis of mesenteric ischaemia:

Radiographic investigations are often of little use in the diagnosis of mesenteric ischaemia

(Grendell & Ockner, 1989) particularly in the early stages (Carter & Camilleri, 1997). Plain

films often only display definite signs when necrosis has already developed (Sutton, 1987).

Reviews of the value of computerised tomography (CT) in the diagnosis of mesenteric

ischaemia and infarction have been mixed. Rogers and David (1995) concluded that CT was

able to suggest intestinal ischaemia but was unable to provide a definite diagnosis.

Conversely, Klein and colleagues (1995) found CT to be highly sensitive (82%), able to

correctly diagnose 18 of 22 cases of intestinal ischaemia. This study also concluded that CT

was able to rule out other causes of surgical acute abdomen (Klein et al, 1995). For the

diagnosis of arterial occlusion, angiography is preferred due to the accuracy of diagnosis and

the ability to infuse vasodilatory drugs or even fibrinolytic agents in rare cases (Klein et al,

1995). Angiography is associated with a number of disadvantages including potential

nephrotoxicity, exposure to radiation and its invasive nature (Sreenarasimhaiah, 2003).

Angiography is of less value after the condition has proceeded to infarction (Corder & Taylor,

6


1993). Improvement in ultrasound technology and expertise mean that the absence of flow

can be reliably demonstrated in major abdominal arteries and veins (unless there is significant

gaseous distension). Where expertise exists this can be useful. In one small German study,

few cases (2 of 11 patients) were correctly diagnosed pre-operatively by imaging (plain X-

ray, sonography, CT or angiography) (Klempnauer et al, 1997).

In 1995 a paper stated that ‘little clinical work has been done to evaluate the use of MRI’

(Magnetic Resonance Imaging), (Rogers & David, 1995), but more recently, MRI has been

reported to be a ‘highly accurate and non invasive method of diagnosing mesenteric

ischaemia’ (Sreenarasimhaiah, 2003).

Intraluminal pCO2, measured by gastric tonometry, has been proposed as a way of monitoring

ischaemia and infarction. This technique has not been widely adopted because of the

relatively high cost of the catheters required, the inability of the probes to monitor large areas

of bowel and poor correlation between the disease state and measurement results (Corke &

Glenister, 2001).

As mesenteric infarction is a disease of older patients, often with significant co-morbidities

and a frequently unclear clinical picture, there may be a useful role for laparoscopy. Where

laparoscopy demonstrates profound, extensive infarction the patient is spared full laparotomy

(since resection in massive infarction is often impractical). When laparoscopy reveals

normal bowel laparotomy may also be avoided, though unfortunately bowel ischaemia cannot

be completely excluded by laparoscopy as the serosal surface can appear normal despite

mucosal necrosis (Kurland, Brandt & Delaney, 1992). Observation of limited bowel

infarction on laparoscopy can be followed by laparotomy and resection. Where the bowel

viability is unclear the laparoscope may be left in situ (with a purse ring suture to secure it)

and a second look bedside laparoscopy performed 12-24 hours later (Dr. C. Corke, personal

communication).

Diagnosis of mesenteric ischaemia via blood tests:

Many plasma based pathology tests have been proposed to diagnose mesenteric ischaemia but

none have been useful in a clinical setting because of low specificity, low sensitivity, but most

importantly markers appear only after infarction has already occurred. Lange and Jackel

(1994) concluded that elevated serum lactate was ‘the best marker of mesenteric ischemia to

7


date’. Raised serum lactate is not a specific finding but suggests that a life threatening

condition exists. The usefulness of lactate increases as a diagnostic marker when the possible

diagnoses of shock, diabetic ketoacidosis, renal and hepatic failure can be excluded (Lange &

Jackel, 1994).

Amylase was shown to be significantly elevated in a canine model of mesenteric ischaemia

(Aydin et al, 1998) but in a clinical setting has been shown to be only moderately elevated

(Corder & Taylor, 1993). In addition, approximately one half of patients with confirmed

infarction show normal levels of amylase (Wilson & Imrie, 1985). Total creatine kinase

(CK) is of little value in differentiating patients with gut infarction from other patients with

abdominal symptoms, or from healthy controls (Fried et al, 1991). The isoform of creatine

kinase CK-BB has been suggested to have some value in predicting gut infarction, with 63%

sensitivity at 100% specificity (Fried et al, 1991), but this protein is labile and prompt

analysis is critical, which limits its practicality for clinical use (Smirniotis, Labrou & Tsiftses,

1989). Serum phosphate has been shown to be a strong indicator of mesenteric ischaemia

(Jamieson et al, 1982) but normal concentrations cannot rule out ischaemia or infarction

(Carter & Camilleri, 1997). Hexosaminidase levels only become elevated after necrosis

(Polson, Mowat & Himal, 1981). Diamine oxidase (Bounous et al, 1984) and intestinal fatty

acid binding protein (I-FABP) (Kanda et al, 1995, Kanda et al, 1996) have shown some

promise but have only been investigated in very small groups of patients. Lactate

dehydrogenase has been found to be elevated during intestinal infarction compared with

‘surgical abdomens’, with a sensitivity of approximately 73% (Calman et al, 1958). In a

recent review, discussion of any blood based diagnostic tests of intestinal ischaemia was

notably lacking, highlighting the void in this area (Sreenarasimhaiah, 2003). The comparison

of the various markers that have been proposed is reviewed and summarised in the following

tables.

8


Potential marker Reference Control group(s) How accurate are these markers? Lactate dehydrogenase.

Calman et al, 1958.

Peritonitis, obstructive jaundice, distended abdomen, fractures & normal controls.

73% sensitive in patients with infarction. Elevated LDH could indicate necrotic intestine or another source of damaged tissue. A single normal LDH reading could not exclude intestinal infarction.

Hexosaminindase Polson,Mowat & Himal, 1981.

Abdominal pain or tenderness.

Indicator of necrosis.

Phosphate Jamieson et al, 1982.

Only patients with acute abdomen & gut ischaemia.

80% had elevated phosphate. The 20% that did not show elevated phosphate had blood taken late after onset of symptoms, so phosphate may have been cleared via urine.

Diamine oxidase Bounous et al, 1984.

Normal, healthy controls & cardiac patients with no abdominal symptoms & one NOMI patient.

One bowel infarction patient showed 7 times normal value, but two showed below normal levels. The elevated level seen in the bowel infarction patient fell to an approximately normal level even though necrotic tissue was not resected. High diamine oxidase levels could indicate pregnancy or lung carcinoma. Low levels are seen in celiac disease & Crohn’s disease.

Amylase, lactate dehydrogenase, phosphate.

Wilson et al, 1987.

Only acute mesenteric ischaemia patients.

Amylase elevated in 27 of 52 patients tested, lactate dehydrogenase elevated in 5 of 7 patients tested, phosphate elevated in 2 of 3 patients tested.

Creatine kinase (total CK) and isoforms MB and BB.

Fried et al, 1991.

Group I: normal healthy controls. Group II: acute abdominal signs.

Total CK and CK-MB levels were not significantly raised in patients with infarction. CK-BB showed 100% specificity, 63% sensitivity.

Table 1.1. Clinically assessed markers of intestinal ischaemia/infarction:

9


Lactate Lange &

Jackel, 1994.

Acute abdominal signs.

100% sensitive, 42% specific, but elevated levels were not indicative of bowel ischaemia/infarction.

Intestinal fatty acid binding protein

Kanda et al, 1995.

Healthy volunteers.

Only two patients in study. I-FABP was undetectable in healthy controls and after bowel resection.

Intestinal fatty acid binding protein, AST, LDH, ALP & CK.

Kanda et al, 1996.

Normal healthy controls and acute abdominal pain.

Small study group, (only 5 cases of mesenteric infarction). Patients with strangulated bowel and patients with mesenteric ischaemia showed high levels of I-FABP, significantly different from healthy controls and patients with other sources of abdominal pain. AST: 60% sensitivity, LDH: 60% sensitivity, ALP: 0% sensitivity, CK: 40% sensitivity.

Lactate Klein et al, 1995.

None. 17/22 bowel infarction patients showed elevated levels.

Lactate, amylase, alkaline phosphatase

Newman et al, 1998.

Mesenteric infarction patients only, divided into primary or secondary development.

Lactate was the only predictor of mortality.

α glutathione S-transferase, AST, ALT, amylase

Delaney et al, 1999.

Acute abdominal patients.

Sensitivity, specificity results: α glutathione S-transferase (100%, 86%, though there may be significant levels seen in liver insult, see Gearhart et al, 2003, below), AST (70%, 50%), ALT (73%, 60%), amylase (25%, 63%).

Lactate Klotz et al, 2001.

Open heart surgery patients +/- NOMI.

No significant difference found.

Table 1.1 continued. Clinically assessed markers of intestinal ischaemia/infarction:

10


Table 1.1 continued. Clinically assessed markers of intestinal ischaemia/infarction: α glutathione S-transferase, amylase, lactate, AST, ALT.

Gearhart et al, 2003.

Patients with suspected acute mesenteric ischaemia. 60% of patients showed some form of mesenteric ischaemia (small bowel ischaemia, colonic ischaemia or both).

α glutathione S-transferase was the most accurate predictor of acute mesenteric ischaemia (74% accuracy) compared with conventional tests (47-69% accuracy). α glutathione S-transferase could not distinguish between ischaemia and necrosis. Elevated levels of α glutathione S-transferase may reflect hepatic ischaemia. The negative predictive value of combined α glutathione S-transferase, white blood cell count and lactate was high.

11


Table 1.2. Animal models of mesenteric ischaemia/infarction:

Potential marker Reference Animal model & Controls used.

How accurate are these markers?

Alkaline phosphatase.

Barnett, Davidson & Bradley, 1976.

Dog. Control laparotomy.

50% of animals with necrotic bowel tissue had no detectable alkaline phosphatase. 2/6 dogs with induced pancreatitis also showed detectable de novo levels of intestinal alkaline phosphatase but also showed duodenal necrosis.

Diamine oxidase Wollin, Navert & Bounous, 1981.

Rat. Sham laparotomy.

Occlusion of superior mesenteric artery for either 30, 60 or 90 minutes resulted in increased levels of diamine oxidase in serum. Diamine oxidase was thought to be released from intestinal epithelium into interstitial fluid. A portion of this enzyme thought to then be released into blood via lymphatics.

Phosphate Jamieson et al, 1982.

Dog. Controls not described, SMA or SMV occluded.

Phosphate levels were elevated at 2 hours and continued to rise until 6 hours. Statistics were not reported.

ALP, CK, LDH & AST, GOT, amylase.

De Toma et al, 1983.

Dog. 6 laparotomy controls, 6 ligated superior mesenteric artery.

Non specific. There was no significant difference between control dogs and those dogs with bowel infarction.

ALP, creatine kinase, lactate dehydrogenase. SGOT, DAO.

Thompson, Bragg & West, 1990.

Dog. SMA ligation, or occluded then released & laparotomy controls.

LDH, SGOT and ALP had low sensitivity. Creatine kinase had greater sensitivity and overall accuracy than other enzymes tested.

ALP, LDH, AST & CK.

Kurland, Brandt & Delaney, 1992.

Dog. Controls not described.

No markers were found to be sensitive or specific to intestinal ischaemia.

12


Table 1.2 continued. Animal models of mesenteric ischaemia/infarction: MDA, ALP, LDH, amylase, AST, GGT & CK.

Aydin et al, 1998.

Dog. Sham laparotomy.

MDA levels significantly higher after ligation. Other proteins were significantly elevated compared with sham operated controls.

Von Willebrand’s factor, myeloperoxidase, protein carbonyl.

Abu-Zidan et al, 1999.

Rat. Sham laparotomy & anaesthesia controls.

Protein carbonyl significantly higher in I/R rats than sham controls. Protein carbonyl correlated with Von Willebrand’s factor. Myeloperoxidase levels were not significantly different between groups.

Cytosolic β-glucosidase.

Morris et al, 1999.

Guinea Pig. Sham laparotomy & anaesthesia controls.

Low specificity. Closed loop obstruction resulted in increased levels.

13


In conclusion successful early diagnosis of mesenteric ischaemia and infarction is

currently dependent on a high degree of suspicion by the physician. Recently, the

mortality rate associated with ischaemic bowel was 35%, but when the condition had

progressed to infarction the mortality rate increased to 68% (Rogers & David, 1995).

At present the only method of definitive diagnosis is surgery (Fried et al, 1991) but

this is often avoided in critically ill or elderly patients. However, an accurate

diagnostic marker would allow surgeons to make risk assessments prior to surgery.

A diagnostic marker would also allow post operative monitoring of patients that have

undergone bowel resection, to ensure that the condition has been rectified. At present

post operative monitoring is not practical and definitive post operative monitoring for

ongoing ischaemia depends upon clinical examination or ‘second look surgery’ 12 to

24 hours after resection (Rogers & David, 1995).

Were early diagnosis to become more attainable and reliable, treatment could be

administered earlier and mortality would decrease. Therefore the ability to make

accurate diagnosis at an early stage would decrease the mortality rate.

Treatment of mesenteric ischaemia:

When mesenteric ischaemia is suspected general treatments such as intravenous

fluids, thrombolytic agents (Flinn & Bergan, 1997), anticoagulants and broad

spectrum antibiotics (Mansour, 1999, Rogers & David, 1995) can be administered to

try to improve outcomes. Surgical treatment primarily aims to revascularize the

bowel when possible (in early ischemia), but irreversible ischaemia and/or gangrene

requires intestinal resection (Klempnauer et al, 1997).

Outcomes after mesenteric ischaemia:

Relatively few studies have focused on the long term prognosis of the patients

discharged from hospital after mesenteric ischaemia. One study conducted follow up

investigations of 31 patients for up to five years after discharge (Klempnauer et al,

1997). It was found that only one patient suffered recurring mesenteric ischaemia,

and died as a result. This patient, unlike the others was not receiving ongoing anti-

coagulant therapy. The major causes of mortality during this study were

cardiovascular events or malignancies. One fifth of the group suffered from short

14


bowel syndrome. The severity of this syndrome is inversely correlated to the length

of small bowel remaining (in this study there was an average of 170cm of small bowel

remaining). An earlier study of patients with infarction due to venous occlusion

found recurrence of thrombosis in up to 29% of cases, most commonly due to residual

thrombi (Clavien, 1990). Of the 31 patients studied, there was a 70% two year

survival rate and 50% 5 year survival rate. The average age of the patients was not

reported.

When large sections of the small intestine require resection, patients must receive

long term parenteral nutrition for survival. Parenteral nutrition allows the delivery of

nutrients independent of the alimentary tract, typically intravenously. Usually

parenteral nutrition is only administered on a short term basis (during a hospital stay

for instance) but longer term nutrition is possible. Prolonged parenteral nutrition is

associated with significant complication and expense (Pennington, 1997). It can be

difficult to estimate the nutritional requirements of patients, and prolonged nutrition

independent of the alimentary tract can cause mucosal atrophy (Braga & Gianotti,

2002), infectious complications, imbalance of gut flora (Deitch, 1993), and decreased

host immune function (Deitch, 1992).

Importance of gut ischaemia/reperfusion injury to other conditions:

Gut ischaemia/reperfusion injury (I/R) is an important condition in its own right, but

is also involved in the development of several other serious conditions. Under

conditions of mucosal damage and hypoxia the gut releases factors into circulation

that may trigger remote organ injury (for example, adult respiratory distress syndrome

(ARDS), or multiple organ dysfunction syndrome (MODS)) and sepsis. The identity

of these gut derived factors and the mechanism of their release are not yet understood.

Research has focussed on gut derived:

- bacteria,

- bacterial endotoxin,

- proteases,

- enzymes (for example, phospholipase A2),

- nitric oxide,

- free radicals,

- cytokines.

15


Some of the gut derived mediators (eg cytokines, free radicals) produced under

ischaemic conditions cause mucosal damage leading to a loss of gut barrier integrity

(Kong et al, 1998), and leakage into the peritoneal cavity. Other factors are

postulated to be transported via the portal circulation or mesenteric lymph. A

complex interplay of a variety of factors is likely. Conditions such as sepsis may

play a role in the development of gut dysfunction (Bradbury et al, 1995) or may

develop because of gut dysfunction (Kong et al, 1998).

Phospholipase A2 is thought to be involved in ischaemia induced remote organ injury,

as PLA2 inhibitors have been shown to reduce bacterial translocation and mucosal

injury in a canine model of shock (Xu, Lu & Deitch, 1995). Interestingly, PLA2

inhibition and calcium channel blocking failed to be fully protective, suggesting that

other factors, such as oxidising agents, are involved (Xu, Lu & Deitch, 1995).

Inhibitors of cyclooxygenase, phospholipase A2 or xanthine oxidase have been shown

to protect against the lung injury caused by laparotomy/intestinal handling in a rat

model (Thomas, Karnik & Balasubramanian, 2002). Animal studies suggest that the

gut serves as a priming bed for circulating polymorphonuclear neutrophils (PMN)

which then promote remote organ injury, a process that is likely to involve

phospholipase A2 (Koike et al, 1992, Koike et al, 1995).

Multiple Organ Dysfunction Syndrome:

Multiple organ failure syndrome (MOFS) was first recognised in 1973 (Tilney, Bailey

& Morgan, 1973) when it was observed that patients with ruptured aortic aneurysms

suffered injury to initially uninvolved organs after surgery. Later it was proposed

that the term MOFS be updated to Multiple Organ Dysfunction Syndrome (MODS) to

reflect the continuum of the syndrome (Fry, 2002). MODS is not an uncommon

event [incidence of 5-20% in ICU patients (Poole, 2002)] and accounts for up to 90%

of the deaths of surgical ICU patients (Montgomery, Venbrux & Bulkley, 1997), a

rate that has not changed significantly since the condition was first described.

MODS is a condition defined as:

“a generalized autoaggressive inflammatory process” (Goris, 1985, Nyman et al,

1996). MODS involves the ‘sequential failure of vital organs caused by generalized

cellular damage’ (Nyman et al, 1996). The most common predisposing factors are

16


shock and infection (Deitch, 1992). The exact cause of MODS is still unclear but

may be due to exogenous factors (bacteria and toxins), but more importantly,

mediators produced by the patient (Deitch, 1992), such as cytokines (Magnotti et al,

1998) or inflammatory mediators. MODS is diagnosed by the identification of the

level of function of several organs by general tests (elevated bilirubin, creatinine,

hematological changes) and clinical observations (requirement for ventilation,

intolerance to enteral feeding, progressive coma) (Deitch, 1992). In reference to the

diagnosis of MODS, Fry (2002) stated that: “Current diagnosis sophistication is

inadequate”.

The condition is characterized by three stages. The first stage involves sepsis, failure

of the pulmonary system and decreasingly utilisation of oxygen in the periphery

(McMenamy et al, 1981). The second stage involves the failure of the hepatic,

intestinal and renal systems (Deitch, 1992), and the third stage involves cardiac failure

(McMenamy et al, 1981). As the number of organs involved increases, so too does

the mortality rate (Deitch, 1992). When a single organ is involved, the mortality rate

is between 15 and 20%, when two organs are involved the mortality rate rises to 40%,

three organs shows 60-80% mortality, and four or more organs involved is associated

with close to 100% mortality (Poole, 2002). The realisation that the organs fail either

simultaneously or in similar sequences has prompted investigations into a common

cause of MODS (Goris et al, 1985). The contribution of gut dysfunction to remote

organ injury and MODS is yet unknown.

Gut derived factors; bacteria and endotoxin:

The suspected involvement of bacteria and bacterial endotoxin in the development of

MODS, have been studied in numerous animal models and in clinical settings. The

gut is thought to be the source of bacteria and toxins. Although healthy duodenum

and jejunum are relatively sterile (Welsh & Reynolds, 2002), bacteria are present in

the distal small bowel and colon at levels of 1010 anaerobic and 105 to 108 aerobic

organisms per gram of bowel tissue (Deitch & Sambal, 2002). It has been clearly

demonstrated that the integrity of the gut epithelial barrier is compromised after sepsis

(Ziegler et al, 1988). This theory may explain why no identifiable source of sepsis is

found in 30% of bacteremic patients with MODS (Goris, 1985). Research has also

focussed on the route that bacteria and endotoxin pass from the gut into the

17


circulation. Initially research focused on the portal circulation, but no bacteremia or

endotoxemia was detectable (Magnotti et al, 1998, Moore et al, 1991). It was also

found that endotoxin concentrations were not significantly different in gut I/R or

laparotomy controls in rats (Koike et al, 1994). This latter study also demonstrated

that endotoxin elimination failed to prevent lung injury due to gut I/R (Koike et al,

1994). A similar clinical study failed to find bacteria or endotoxin in the portal blood

of trauma patients, even those that went on to develop MODS (Moore et al, 1991).

Later studies in animals focussed on the mesenteric lymph as the route of release.

Entry into the circulation via the mesenteric lymph is an interesting possibility as it

reaches the lung before any other vascular bed (Deitch & Sambal, 2002, Magnotti et

al, 1998) (possibly explaining why the lung is the first organ to fail in MODS). It

was found that gut derived factors that increase cell permeability and contribute to

shock induced lung injury were more prevalent in mesenteric lymph than portal

plasma (Magnotti et al, 1998). Deitch and colleagues (2001) also concluded that gut

derived factors carried in mesenteric lymph appeared to be responsible for lung injury

after shock but bacteria were not responsible. It has been proposed that although

bacteria and/or endotoxin are rarely found in either the portal circulation or lymph

they may still play a role in the development of MODS. A change in the bacterial

balance or bacterial overgrowth may trigger a more potent cytokine response in the

gut after ischaemia/reperfusion injury or shock (Deitch et al, 2001). It has been

demonstrated that IL-6 (Biffl, Moore & Moore, 1995, Grotz et al, 1999) and TNF-α

are released from ischaemic gut and that there is a relationship between the magnitude

of the ischaemic injury and the cytokine response (Grotz et al, 1999). Cytokines

(IL-1, IL-6, TNF) and neutrophils can decrease oxygen supply to the gut by altering

microcirculation, possibly leading to a self sustaining cycle (Deitch, 1992).

18

Methods

POLYACRYLAMIDE GEL STAINING METHODS:

SILVER STAINING:

Silver Staining (BioRad method):

Immediately after electrophoresis, gels were placed in fixative (40% methanol v/v, 10% acetic

acid, 30 minutes). The gels were then placed in oxidising solution (1:10 dilution of BioRad

Silver Staining Kit stock solution oxidiser, 5 minutes, BioRad, Hercules, CA). The gels were

then washed with water (2L with several changes over 15 minutes). The gels were then

stained with silver reagent (1:10 dilution of BioRad Silver Staining Kit stock solution, 20

minutes) and then quickly rinsed with water. The gels were then developed (8g BioRad

Silver Staining developer/250mL water, with multiple changes of solution). The

development was stopped with an acetic acid solution (5% v/v).

Silver staining: (SilverQuest™, Invitrogen method):

Silver staining of NuPAGE gels was performed using SilverQuest™ mass spectrometry

compatible silver staining kit (Invitrogen, Carlsbad, CA). Immediately after the completion

of the electrophoretic run the gels were rinsed quickly (ultrapure water). The gels were then

fixed (40% ethanol, 10% acetic acid, 20 minutes). The gels were then washed (30% ethanol,

10 minutes) before the sensitizing step (10% sensitizing solution, 30% ethanol, 10 minutes).

The gels were then washed in ethanol (30%, 10 minutes) and then ultrapure water (10

minutes). The gels were then stained (100mL, 1% staining solution, 15 minutes) and quickly

rinsed (ultrapure water, 1 minute). The developing solution was then added (10% developer,

1 drop enhancer, 6 minutes) until the desired level of staining was achieved after which the

stop solution was added directly.

Silver staining (PlusOne Pharmacia Biotech method):

All solutions were prepared in ultrapure water. Immediately after electrophoretic run was

complete the gels were placed in fixative (40% ethanol, 10% acetic acid v/v, 60 minutes).

The gels were then placed in sensitizing solution (30% ethanol, 0.125% glutardialdehyde

solution, 0.2% sodium thiosulphate solution, 6.8% sodium acetate w/v, 30 minutes). The

gels were then washed with ethanol solution (20% v/v) followed by two washed with water (5

minutes per wash). The gels were then stained with silver reagent (0.25% solution of silver

nitrate, 0.04% formaldehyde, 20 minutes) and then washed with water (2x1 minute). The

19

Methods

gels were then developed (2.5% sodium carbonate, 0.02% formaldehyde, 10 minutes). The

development was stopped with an EDTA solution (1.5% w/v). Gels were then placed in

water before scanning.

MS Compatible Modification of the PlusOne method:

This method was modified by Yan et al, 2000 for use with ESI and MALDI.

The PlusOne method was followed with the following modifications:

Two fixing steps were employed (15 minutes each). The sensitiser did not contain

glutardialdehyde. The silver solution did not contain formaldehyde. Formaldehyde (100µL)

was added to the developer solution. Acrylamide (1%) was added to the samples after

heating, before loading onto the gel (Dr. Bruce Kemp, personal communication), to prevent

re-formation of disulfide bonds.

Protein bands and spots of interest stained by this method were excised from the gels and

rinsed briefly (ultrapure water). The gel pieces were destained (50µL, fresh 1:1 solution of

30mM potassium ferricyanide, 100mM sodium thiosulphate, 8 minutes, then repeated). The

destaining solution was discarded. The pieces were then washed (ultrapure water, two

washes). The gel pieces were then incubated in ammonium hydrogencarbonate (100µL,

50mM, 20 minutes room temperature). The gel pieces were then cut into small pieces,

washed in ultrapure water and dried (four changes of acetonitrile). The gel pieces were then

dried under vacuum, trypsin digested and submitted for nano-MS.

COOMASSIE STAINING:

Protein stains for gels of samples containing ampholytes:

Coomassie G250 (CBB-G250) containing stains (Neuhoff et al, 1988) were used in

preference to Coomassie R-250 (CBB-R250) stains when running gels of samples that

contained ampholytes. CBB-G250 did not show the high background and ampholyte staining

that CBB-R250 did. For quantitative studies gels were stained with Colloidal Coomassie for

three days with minimal destaining (Herbert, personal communication).

20

Methods

Copper protein stain:

Copper containing protein stain could also be used for staining gels that contained

ampholytes. Copper sulphate (0.5%) was dissolved in water. Ethanol (final concentration

27%) and acetic acid (final concentration 10%) and Coomassie R-250 (0.04%) were added

and solution made up to volume with water. Gels were destained with ethanol/acetic

acid/copper sulphate (12%v/v, 7%v/v, 0.5%w/v respectively). The ethanol could be replaced

with isopropanol.

ANTIBODY TECHNIQUES:

Preparation of immobilised antibodies, with CNBr:

Dialysis tubing (6-8MWCO) was boiled in buffer (100mM bicarbonate, 10mM EDTA, 3

minutes). Antibody (1mL) of 10B2 (~0.93mg/mL, Cayman Chemicals, Ann Arbor, MI) or

4A1 (1mg/mL, Boehringer Mannheim) in coupling buffer were dialysed against coupling

buffer (4L, 4oC, 18 hours). CNBr Sepharose 4B beads (Pharmacia Biotech, 0.06g) were

swelled in dilute HCl (1mM HCl, 5 changes of acidic solution). Beads and dialysed antibody

were mixed and rotated (4oC, 47 hours). The beads were then washed with coupling buffer

(5 volumes, 0.1M NaHCO3 pH 8.3, 0.5M NaCl) and active sites remaining were blocked

(0.1M Tris-HCl, pH 8.0, 2 hours).

Immunodepletion of PLA2:

CNBr sepharose-antibody beads (50µL) and samples suspended in PBS were mixed with

protease inhibitors (1mM PMSF, 50µg/mL aprotinin, 200µM leupeptin) and incubated with

rotation (4oC, 3.25hours). The mixture was then centrifuged (750g, 1 minute) and

supernatent removed. The beads were washed with PBS (4 times) and SDS loading buffer

was added (20µL). The beads were boiled (100oC, 7 minutes) and loaded onto a 4-20%

gradient Tris-Glycine gel (Invitrogen pre-cast mini-gel), and run according to manufacturer’s

instructions. Gels were then silver stained (BioRad silver staining kit).

Immunodepletion of Cyclophilin B:

Protein G agarose (50µL of 50% solution per sample) was mixed with ice cold PBS (500µL)

and centrifuged (12 000g, 4oC, 20 seconds). Supernatent was discarded and the washing

process was repeated twice.

21

Methods

Pre-clearing samples:

Samples (500µg total protein, cold PBS/protease inhibitor cocktail (Sigma, St. Louis, MO) to

a volume of 1mL) were added to the protein G agarose beads. The samples were incubated

with rotation (4oC, overnight).

Immunodepletion:

Pre-cleared samples were centrifuged (12 000g, 4oC, 30 seconds) and supernatent removed to

a fresh tube. Aliquots were removed for other assays. Antibody (1µg of polyclonal

cyclophilin B antibody, Abcam Ltd, Cambridge, UK.) was added to each sample and rotated

for one hour (4oC). Washed protein G-agarose (50µg) was then added and the samples were

again incubated with rotation (4oC, 3 hours). The supernatent was then collected (12 000g,

4oC, 30 seconds) for other assays. The beads were then washed (three washes in ice cold

PBS) before addition of SDS-PAGE sample buffer (50µL).

Cyclophilin B Immunohistochemistry:

Cyclophilin B was investigated in sections of human gut (control and infarcted) using the

Dako immunoperoxidase method, according to manufacturer’s instructions (Dako, Glostrup,

Denmark). The primary antibody dilution was optimised to 1:1200, by performing a

preliminary serial antibody dilution experiment. The sections were counterstained with

Gill’s haematoxylin. Acknowledgement must go to the Pathcare histology laboratory staff

for the preparation of tissue sections and slides for this work.

Phospholipase A2 sandwich ELISA:

ELISA plates (Nunc, Weisbaden, Germany) were coated with the monoclonal antibody 9C1

(100µL per well, 2mg/mL diluted 1:1000 in PBS, Cayman Chemicals, 4oC, 8 hours). The

plates were then blocked (1% skimmed milk powder, 0.1% BSA in PBS, 37oC, 16 hours).

The wells were then washed twice with wash buffer (200µL per well). A series of sPLA2

standards were prepared in the range 0-200ng/mL in PBS/0.1% BSA. The samples were also

prepared in a series of dilutions from 1/5 to 1/50 in PBS. The standards and samples were

added (100µL per well) and were incubated at 37oC for 2 hours. The wells were then washed

twice with wash buffer (200µL per well). Conjugate binding was performed by adding 4A1-

AP (100µL of 0.1% conjugate antibody in 0.1% BSA/PBS). The plates were incubated

22

Methods

(37oC, 60 minutes). The plates were washed three times with wash buffer and three times

with carbonate buffer (200µL per well). p-Nitrophenyl Phosphate (pNPP) (100µL of 15mg

pNPP in 15mL carbonate buffer) was added and incubated at room temperature for 10

minutes. The absorbance of the plates were then read at 405 nm.

PLA2 Western Blotting:

16% polyacrylamide gels (Invitrogen) were used in this section, and were run according to

manufacturer’s instructions under non-reducing conditions. Lanes contained 30µg of total

protein. Multimark Multi Coloured Standard (Invitrogen), (5µL) were used for size

estimation.

The transfer apparatus used was the X-cell-II Blot Module, (Invitrogen). The transfer was

completed at 30volts for 1½ hours (12mM Tris, 96mM glycine, 15% methanol). The

membrane was then washed with Tris buffered saline (TBS) and carefully dried. The

membrane was then blocked (50mL 5% skim milk powder in TBS) overnight at 4oC. The gel

was stained with Coomassie R-250.

The blocked membrane was washed in Tris buffered saline-Tween-20 (TBST) (50mL), and

then incubated in the primary antibody (10mL of 9C1 antibody (5µL) in 1%BSA, 1% skim

milk powder in TBS) for 1 hour. The antibody solution was then removed and the membrane

washed three times with TBST (50mL). The membrane was incubated in the secondary

antibody (3.5µL anti-mouse IgG-HRP in 1%BSA, 1% skim milk powder in TBS, 10 mL final

volume). The antibody solution was removed and the membrane washed in TBST (3x50mL,

3x15 minutes) and TBS (50mL, 15 minutes). Enhanced Chemiluminescence Reagents

(Renaissance, NEN, Irvine, CA), luminol (700µL) and oxidiser (700µL) were mixed and

placed on the membrane for 1 minute. The membrane was blotted dry and covered with

plastic wrap. The membrane was exposed to film (Hyperfilm-ECL, Amersham, Little

Chalfont, Buckinghamshire, UK) for varying time periods (1-60 minutes).

Serum Amyloid A ELISA Assay (TriDelta Phase™ range SAA kit):

The SAA ELISA (TriDelta, Greystokes, Ireland) assay was performed according to

manufacturer’s instructions. Samples of plasma were diluted 1:500, 1:1000 or 1:5000 in

diluent. Biotinylated anti-SAA (1:100 in diluent) was applied to wells containing SAA

23

Methods

antibody (50µL per well). Samples and standards then added (50µL per well in duplicate)

and incubated (60 minutes, 37oC). The wells were then washed four times with wash buffer.

Plates were read at 450nm.

Determination of plasma lactate, C-reactive protein and the APACHE-II score.

Acknowledgement must go to the laboratory staff at Pathcare Pathology, Geelong for the

analysis of plasma C-reactive protein and lactate. Acknowledgement must also go to the

staff of the Intensive Care Unit, Geelong Hospital for the APACHE-II scoring of patients.

Cyclophilin B Western blotting:

SDS PAGE gels (8-16% iGEL, TrisGlycine, Gradipore) were run according to manufacturers

instructions, under reducing conditions. Sample lanes contained 30µg of total protein.

Prestained protein molecular weight markers (MBI Fermentas, St. Leon Rot, Germany) were

included for size comparison. The positive and internal control was a whole cell lysate

(15µL) from Jurkat cells. Acknowledgement must go to Caryll Waugh, Douglas Hocking

Research Institute, for the culture of Jurkat cells. The same cell lysate was used in all

Western analyses. The Jurkat cell pellet (1× 106 cells) was collected (10 minutes, 4oC,

9000g). The supernatent (cell culture medium) was removed, and stored (-80oC) for the

positive and internal control in the Western analyses of plasma samples. SDS-PAGE sample

buffer (100µL, containing 0.05M DTT) was added to cell pellet. The solution was

thoroughly mixed and incubated (room temperature, one hour). The solution was then boiled

(100oC, 5 minutes). If the Jurkat cell lysate sample was stored, fresh DTT (10% v/v, 0.5M)

was added and the samples re-boiled before applying to a gel.

Proteins were transferred from SDS-PAGE to PVDF (Amersham) using the CAPS buffer

system (15% methanol, 0.01% SDS, 10mM CAPS) for 45 minutes at 180 mA. Membranes

were blocked (5% skim milk powder in TBS-T) overnight at 4oC. The primary antibody

(rabbit polyclonal anti-cyclophilin B, Abcam Ltd.) was applied (0.025µg/mL in TBS-T, 5%

skim milk powder) for one hour, room temperature with gentle agitation. The membrane was

washed extensively (TBS-T). The secondary antibody (goat anti-rabbit IgG conjugated to

HRP, Abcam Ltd.) was applied (diluted 1:50 000 in TBS-T, 5% skim milk powder) for one

hour at room temperature. The membrane was washed extensively (TBS-T). The signal was

detected with chemiluminescence (ECL plus kit, Amersham), according to manufacturer’s

24

Methods

instructions. Images were recorded and image analysis was performed using Kodak Digital

Science 1D, version 3.0.0.

Membranes were stripped (50oC, 30 minutes) with stripping buffer (2% SDS, 62.5mM Tris-

HCl, pH 6.8, 100mM β-mercaptoethanol), according to the method of Kaufman, Ewing &

Shaper (1987). Following stripping membranes were washed extensively (TBS-T), dried and

stored at 4oC until re-probing.

ALKALINE UREA GELS:

Gels for mammalian phospholipases A2:

The method used with these gels is a modification of Ahmad, Lawrence and Moores (1994).

The resolving gel contained acrylamide (12.5%), bis-acrylamide (0.5%), urea (8M) and

ethanolamine (2%) in ultrapure water. The gels were polymerised by the addition of

ammonium persulphate (0.005%) and TEMED (0.1%) under an iso-butanol overlay. The

gels were cast in mini gel format. The surface of the resolving gel was rinsed thoroughly

with water and then stacking gel solution. The stacking gel contained acrylamide (7%), bis-

acrylamide (0.3%), Tris (0.01M, pH 6.8) and urea (8M) in ultrapure water. The gels were

polymerised by the addition of ammonium persulphate (0.05%) and TEMED (0.2%). The

running buffer was dilute ethanolamine (2% in ultrapure water). The gels were pre-run at

200V for two hours to remove free radicals and prevent artifactual protein bands, after which

the buffer was changed. Wells were rinsed with running buffer before the application of

samples to remove the products of secondary polymerisation. Samples were applied to the

wells in sucrose (50%) or glycerol (50%) containing bromophenol blue as a tracking dye.

Typical lanes contained 4µg (purified proteins) to 20µg (crude venom) of total protein. Gels

were run at 200V in an ice bath until the tracking dye had migrated at least 7cm. The gels

were stained with the Neuhoff stain/Colloidal Coomassie (Neuhoff et al, 1988). Crude

venom samples were obtained from Sigma.

Blotting/ electrotransfer of proteins from alkaline urea gels to PVDF membrane:

The PVDF membrane (PALL BioTrace ™ PVDF (Pall, Ann Arbor, MI) or BioRad

SequiBLOT) was thoroughly wet with methanol before use but was not equilibrated in

transfer buffer. Proteins were transferred to PVDF membrane from alkaline urea gels with

25

Methods

the CAPS transfer buffer (10mM CAPS, pH 11.0 with 20M NaOH, 0.01% SDS, 10-15%

methanol) or the Towbin buffer (25mM Tris, 192mM glycine, 20% methanol, Towbin,

Staehelin & Gordon, 1970). CAPS transfer buffer is recommended for basic proteins

(Szewczyk & Kozloff, 1985). The transfer was performed at 100-180mA for 45 minutes

(CAPS) or 150mA for 60 minutes (Towbin) in a stirred blotting apparatus (Mini Trans Blot,

BioRad). After the transfer was complete the gel was stained in the Neuhoff stain (to check

transfer efficiency) and the membrane was washed in ultrapure water (5 minutes). The

membrane was then stained quickly (10 seconds) in Coomassie R-250 stain (0.1% w/v

Coomassie R-250, 40 % v/v methanol in deionised water) before rinsing the membrane with

methanol. If required the staining/methanol rinse was repeated until an appropriate level of

differential staining was achieved. Both the gel and the membrane were destained in

methanol solution (30% v/v methanol in ultrapure water).

Transfer of proteins of interest to SDS-PAGE:

Protein bands of interest from alkaline urea gels were excised with a clean scalpel. Bands

were more easily handled if frozen before transfer to SDS-PAGE. Gel pieces were placed in

wells of SDS-PAGE or Tricine gels and embedded in molten agarose (0.5% agarose, trace

bromophenol blue in appropriate running buffer), or reduced and then applied to the gels.

Gels were run according to the method of Laemmli (1970) for SDS-PAGE or Schagger & von

Jagow (1987) for Tricine gels.

Reducing method “A”: Gel pieces were placed in fresh microcentrifuge tubes

containing molten agarose (100µL, 0.5% agarose, trace bromophenol blue in SDS-PAGE

running buffer) containing DTT (1 part 0.5M DTT to 9 parts agarose). Tubes were then

heated (100oC, 10 minutes). Gel pieces and the agarose were then loaded into wells of the

SDS gels.

Reducing method “B”: According to the method of Davie, 1982. Briefly, gel pieces

were incubated (room temperature, 30 minutes) in equilibration buffer (10% glycerol, 5% β

mercaptoethanol, 2.3% SDS, 62.5mM Tris-HCL pH 6.8). The gel pieces were then placed in

the wells of the SDS gel and embedded in agarose (0.5%, agarose (Davie, 1982 reported the

use of 1% agarose) trace bromophenol blue in SDS running buffer).

26

Methods

Phospholipase A2 enzyme activity assay of gel pieces:

The gel was sliced into 1mm slices with a clean scalpel. Each piece was quickly transferred

to a microcentrifuge tube containing ethanolamine solution (100µL of 10mM ethanolamine in

ultrapure water), or Tris buffer (100µL, 50mM pH 6.8) or PBS (100µL) or ultrapure water

(100µL). Each piece was macerated with clean forceps. Aliquots (25µL) were removed

from each tube to analyse phospholipase A2 activity.

PHOSPHOLIPASE A2 ASSAYS:

Phospholipase A2 enzyme activity assay:

This method follows that published by Green and colleagues (1991).

Phosphatidylethanolamine L-1-palmitoyl, 2-arachidonyl, [arachidonyl-1-14C] (NEN), L-3-

Phosphatidyethanolamine 1-acyl-2-[1-14C arachidonyl] (specific activity 54 mCi/mmol)

(Amersham) or Phosphatidylcholine 1-stearoyl-2-[1-14C arachidonyl] (specific activity

55mCi/mmol) (Amersham) (5.5nmol) were dried under a stream of cold air or nitrogen. The

phospholipid was taken up in deoxycholate (2% w/v in ultrapure water). Assay buffer (25µL

per sample, 100mM Tris, pH 8.0, 300mM NaCl, 10mM CaCl2) was then added and this

mixture pre-warmed (37oC, 10 minutes). The sample solutions were also pre-warmed (37oC,

5 minutes). The substrate was then added to the samples (25µL per sample) and incubated

(37oC, 30 minutes). The reaction was terminated with EDTA (10µL per sample, 100mM).

The reaction solutions (each 30µL) were spotted onto TLC plates and dried with warm air.

The TLC plates were run in a chloroform/methanol/glacial acetic acid system (90:10:1 v/v/v).

Autoradiography was used to establish the position of the product (fatty acid) and then un-

reacted phospholipid. These regions were scraped from the plates and analysed by

scintillation counting (United technologies, Packard 2000CA Tricarb Liquid Scintillation

Analyzer).

When the pH profile was being examined the buffers used were: Tris (100mM) for pH 7-9,

Glycine (100mM) for pH 9.5-10 and sodium acetate (100mM) for pH 6-6.5 to ensure

adequate buffering capacity.

27

Methods

Samples were diluted as required to maintain the hydrolysis to less than 30%, which

represents the upper limit of linearity for this assay (Seilhamer et al, 1989). Where the

phospholipase activity of samples with very different quantities of total protein were

compared, results were expressed as the specific activity.

The addition of cyclosporin A to the assay:

Cyclosporin A (Sigma) was dissolved in ethanol (1mg/mL) to form the stock solution. The

stock solution was diluted in phospholipase A2 assay buffer (containing the phospholipid) to

the working concentration (2µg/mL). The cyclosporin A/assay buffer solution (25µL) was

then added to the enzyme source (25µL).

Phospholipase A2 and purified protein sources:

Pancreatic PLA2: Porcine, Sigma.

Bee venom PLA2: Sapphire Biosciences (Crow’s Nest, Sydney, Australia).

Haemoglobin: Human, Sigma.

Cytochrome C: Bovine heart, >99% pure, Sigma.

Bovine serum albumin: Fraction V, >96% pure, Sigma.

BOWEL TISSUE ANALYSIS:

Ethics approval for use of human bowel and plasma samples:

All experimental procedures were approved by the Deakin University (Ethics approval

number EC 116-2001) and Geelong Hospital (Ethics approval numbers 01/31 and 01/44)

ethics committees. Tissue was obtained from patients undergoing surgical removal of

infarcted bowel tissue or removal of normal bowel tissue from patients undergoing right

hemicolectomy. Acknowledgement must go to Geelong Hospital theatre staff for their help

in this regard.

Gross tissue damage rating system:

Damage to bowel tissue was assessed according to the scale proposed by Sun and colleagues

(1997).

28

Methods

Table 2.1: Gross tissue rating system of Sun and colleagues (1997):

Rating Features

1 Mild damage. Slight reddish discoloration.

2 Moderate damage. Red discoloration often with haemorrhage.

3 Severe damage. Grossly necrotic, blackish red, friable and lustreless.

ONE AND TWO DIMENSIONAL GEL ELECTROPHORESIS:

NuPAGE:

Bis-Tris gels (10%) were run with the MES buffer system according to manufacturer’s

instructions. SeeBlue®Plus2 (Invitrogen) prestained molecular weight markers and

Mark12™ wide range protein standards (Invitrogen) were used in these gels for protein size

estimation. Blotting from these gels was performed using the Mini Trans-Blot® apparatus

(BioRad) with a CAPS transfer buffer (10mM CAPS, pH 11, 15% methanol) at 180mA for 45

minutes. The PVDF membrane (ProBlot™, Applied Biosystems) was wet in methanol and

equilibrated in transfer buffer. The gel was rinsed briefly (Milli-Q) and also equilibrated in

transfer buffer. The gel was stained (0.25% Coomassie brilliant blue R-250 in 45%

methanol, 10% acetic acid). The membrane was stained quickly (60 seconds, 0.1%

Coomassie brilliant blue R-250 in 40% methanol, 1% aldehyde free acetic acid) before

quickly rinsing with methanol and soaking in ultrapure water.

Two-Dimensional Electrophoresis of heparin fraction samples:

Samples were desalted and concentrated by methanol or TCA precipitation. Samples

(100µL) were mixed with cold methanol (-20oC, 900µL) and stored overnight (-20oC).

Alternatively samples (100µL) were mixed with TCA (10mg) and left on ice (30 minutes).

The samples were then centrifuged (10 minutes, 4oC, 18000g). The pellets were rinsed with

cold methanol or acetone (TCA precipitation) before re-centrifugation and were then air dried

(room temperature, 30 minutes). 1st dimension rehydration buffer containing DTT (2.9mg

per mL) and biolytes (0.5%) was added (130µL per sample) and samples were vortexed

repeatedly and quickly centrifuged before placing in contact with IPG strips (pH 3-10, 7cm

strips). The proteins were focussed using a ramping current:

29

Methods

Stage Voltage (V) Time (hours)

1 50 12

2 500 1

3 1000 1

4 8000 3.5-8

TOTAL 18 500-26 000 volts

The IPG strips were then equilibrated firstly in DTT buffer (15 minutes) and then in IAA

buffer (15 minutes). The strips were then loaded onto NuPAGE® gels (4-12% BisTris,

1.0mm, MES running buffer) or Tris Glycine gels (4-20% Zoom™, SDS running buffer,

Invitrogen) along with Mark 12 ™Standards (strip loaded) and embedded in agarose (0.5%

agarose, trace bromophenol blue, in MES buffer). The gels were run according to

manufacturer’s instructions. The gels were then stained with either Coomassie R-250 (0.25%

CBB-R250 in 45% methanol, 10% acetic acid) or silver stained (PlusOne, Amersham

protocol).

Proteomics:

Samples of human plasma were thawed and added to chilled methanol (-80oC, 1.0mL).

Chilled methanol was added to a final volume of 2mL for each sample. Samples were

thoroughly vortexed and placed in a freezer overnight (-80oC). Samples were centrifuged in

a pre-cooled centrifuge (12 000g, -4oC, 60 minutes). The methanol supernatent was removed

and discarded. The pellet was resuspended in solubilisation solution containing 1% carrier

ampholytes (pH 3-10), (230µL solubilisation solution per 100µL original volume of plasma).

The mixture was repeatedly vortexed and sonicated until the pellet had been dissolved. If

required the samples were also subjected to bead beating (4 X 15 seconds, Bio 101 Fast

Protein™ blue tubes). Immobilised pH gradient strips (IPG), (pH 7-10, 7cm, BioRad) were

re-hydrated overnight in solubilisation solution containing 1% carrier ampholytes (pH 3-10,

200µL per strip). Samples were centrifuged quickly before cup loading onto the IPG strips.

The IPG strips were subjected to isoelectric focussing with a ramping current:

30

Methods

Stage Votage (kV) Time (hours)

1 0.1 5

2 0.3 3-7

3 0.6 2-3

4 1 1-3

5 2 1-3

6 3-3.5 9-12

At the completion of isoelectric focussing the IPG strips were equilibrated in equilibration

solution (20 minutes with agitation). The IPG strips were then embedded in 0.5% agarose in

cathode buffer containing bromophenol blue (0.001%) on the top of either pre-cast Criterion

gels (4-20%, Tris-HCl) or freshly prepared large format gradient gels (18x20cmx1.5mm, 8-

18% acrylamide). The second dimension was performed using the following programme for

Criterion gels:

Stage Current (mA) Time (hours)

1 3 per gel 0.5

2 30 per gel 2

The second dimension was performed using the following programme for large format gels:

Stage Current (mA) Time (hours)

1 3 per gel 2

2 50 overnight

The programme was continued until the dye front had begun to run off the gel. Gels were

then fixed for 30 minutes (10% methanol, 7% glacial acetic acid in ultrapure water). Gels

were then stained with Sypro Ruby Protein gel Stain (BioRad) for the required length of time.

Gels were destained (10% methanol, 7% glacial acetic acid in ultrapure water) for 30 minutes.

Gel images were captured with BioRad Molecular Imager® FX and BioRad Quantity One

version 4.1.1 software. Image analysis was performed using Melanie version 3. Gels

previously stained with Sypro Ruby were re-stained with colloidal Coomassie G-250

overnight. Coomassie stained gels were scanned with the Molecular Dynamics personal

31

Methods

densitometer SI and Corel Photo Paint version 8. Protein differences were assessed using

Melanie version 3.

PROTEIN IDENTIFICATION:

Montage™ In-Gel Digest96 Kit (Millipore):

Gel spots were cut and placed in the MultiScreen plate (Millipore, Bedford, MA). Destaining

solution (100µL) was added to each well and incubated (room temperature, 20 minutes). The

solution was removed under vacuum. This process was repeated twice. Acetonitrile

(100µL) was then added to each well and incubated (room temperature, 10 minutes). The

solvent was then removed under vacuum. The samples were reduced with DTT (100µL,

10mM in 0.2M ammonium bicarbonate) and incubated (room temperature for 60 minutes or

37oC for 30 minutes). The solution was removed under vacuum. The gel pieces were

washed using three changes of ammonium bicarbonate (100µL, 0.2M) followed by

acetonitrile (100µL) with removal of solutions under vacuum. The proteins were then

alkylated (100µL of 50mM IAA in 0.2M ammonium bicarbonate) for 20 minutes at room

temperature. The solution was then removed under vacuum. The gel pieces were washed

using three changes of ammonium bicarbonate (100µL of 0.2M) followed by acetonitrile

(100µL) with removal of solutions under vacuum. Trypsin solution (50µL diluted to

approximately 1/10 the mass of protein in the spot in 0.1M ammonium bicarbonate) was then

added to each well and the plate was incubated overnight (37oC). The peptides were then

extracted by adding extraction solution (50µL per well) and incubating (room temperature, 30

minutes) before collecting the solution in a microtitre plate. Extra extraction solution (50µL)

was added to each well, incubated (room temperature, 30 minutes) and then collected. This

process was repeated. The peptide solutions were then concentrated (SpeedVac, Savant).

Nano MS/MS:

Samples for nano MS/MS were run in 4-12% TrisGly gels (Invitrogen), according to

manufacturer’s instructions. Proteins were thoroughly reduced with DTT containing sample

buffer and heating (100oC, 7 minutes). After heating the samples were allowed to cool

before the addition of acrylamide (final concentration of 1%) to form cysteine adducts to

prevent reformation of disulphide bonds. Potassium ferricyanide (30mM) and sodium

thiosulphate (100mM) were mixed in equal proportions. Excised bands from Tris-Glycine

32

Methods

gels were incubated in the ferricyanide/thiosulphate solution (50µL per tube) until the brown

colour of the silver stain had been removed. The solution was removed and the gel pieces

were washed with two changes of water. The gel pieces were then incubated in ammonium

hydrogen carbonate (100µL, 50mM, room temperature, 20 minutes). The gel bands were cut

into small pieces, washed with water and transferred to new tubes. Acetonitrile was used to

rinse and shrink the gel pieces (3x100µL, 1x500µL). The gel pieces were dried (SpeedVac,

Savant). Trypsin (20µL, 0.1µg/mL in 1mM HCl, Promega, Madison, WI) was added,

followed by digestion buffer (100µL, 50mM NH4CO3, 1mM CaCl2, pH 7.8), and incubated

overnight at 37oC.

Extraction of peptides :

Ammonium bicarbonate (100-150µL, 25mM) was added to each tube, centrifuged and then

incubated, (37oC, with sonication). Acetonitrile (150µL) was added to each tube, vortexed

and incubated. Formic acid (50µL, 5%) was added, vortexed and incubated. Acetonitrile

(150µL) was added to each tube, vortexed and incubated. The pooled extracts were then

dried to approximately 10µL (SpeedVac, Savant).

The peptide solutions were desalted using miniature C18 columns constructed in gel loading

pipette tips (Eppendorf). The columns were prepared using C18 resin (Vydac) in

acetonitrile/water. The columns were rinsed extensively with methanol, then formic acid

(5%). The peptides were bound to the column, and then washed with formic acid (5%). The

peptides were eluted into the sample applicator with methanol.

The mass spectrometer used was the API Q-STAR Pulsar i, LC/MS/MS (Applied

Biosystems). MS/MS spectra were submitted to Mascot and SwissProt. Acknowledgment

must go to Dr. Peter Hoffman (St. Vincent’s Medical Research Institute) for assistance with

sample preparation and mass spectrometry analysis.

MALDI-TOF MS:

Protein spots of interest were excised from the gels and placed in microtitre plates with care

taken to avoid contamination between other spots and outside sources. Wash buffer (120µL

per gel piece, 50% acetonitrile, 25mM NH4CO3, pH 7.8) was added to each well. The plate

was incubated with shaking (37oC, 10 minutes). The wash solution was removed and

33

Methods

discarded. This washing process was repeated three times, after which there was no colour

remaining in the gel pieces. The gel pieces were then dried under vacuum for 15 minutes.

Sequencing grade trypsin (8µL, 15ng/mL porcine trypsin, (Promega) in NH4CO3, pH 7.8) was

added to each gel piece. The plate was then sealed and incubated overnight with agitation

(37oC). The plate was spun (SpeedVac Plus, Savant, 5 minutes). Extract solution (8µL,

50% acetonitrile, 1% TFA) was added to each well and sonicated for 20 minutes.

An aliquot of each sample (1.5µL) was spotted onto a MALDI plate. Matrix (1.0µL, α-

cyano-4-hydroxycinnamic acid, 8mg/mL in 50 % acetonitrile, 1% TFA) was added to each

sample spot on the MALDI plate. Samples were air dried and analysed. Acknowledgment

must go to Dr. Stewart Cordwell (Australian Proteome Analysis Facility) for analysis of

samples by MALDI-TOF.

Bioinformatics:

The protein sequences of the ‘theoretical proteins’ identified by MALDI were analysed using

software from the National Centre for Biotechnology Information (BLASTN, BLASTP) and

the EXPASy Molecular Biology Server of the Swiss Institute of Bioinformatics.

Nano LC-MS/MS:

Protein bands of interest were excised from BisTris (Invitrogen) and TrisGly (Gradipore) gels

with appropriate positive (protein markers) and negative (gel pieces from no protein regions)

controls. The protein bands were destained and digested with trypsin. The proteins were

analysed by nano LC-MS/MS. Acknowledgement must go to Dr. Mark Raftery, Biomedical

Mass Spectrometry Facility, University of New South Wales for the protein digestion and

mass spectrometry analysis.

Amino terminal amino acid sequencing;

Rotofor fractions were applied to a ProSorb sample preparation cartridge (Applied

Biosystems, Foster City). The cartridge membrane was washed several times with TFA

solution (0.1%), changing the filtration absorbance filter after each 750µL load.

Alternatively the samples were applied via PVDF membrane:

34

Methods

Protein bands of interest were excised from PVDF membranes and subjected to automated

amino acid sequencing (Edman degradation) using the Procise automated sequentor (Applied

Biosystems, Foster City, California). The phenylthiodantoin (PTH) derivatised amino acids

were analysed to determine the amino acid sequence. The Procise PTH system is comprised

of the Micro-gradient delivery system (model 140C), a UV detector (model 785) and data

analysis software (model 610A). Acknowledgement must go to Gary Beddome and Brian

Shiell (Australian Animal Health Laboratories, CSIRO) for amino acid sequencing.

PROTEIN PURIFICATION:

Ammonium sulphate precipitation:

The appropriate amount of ammonium sulphate to reach saturations of 15 to 80% (ACS

reagent grade) was added to samples and agitated by rotating (5 minutes). The samples were

then centrifuged (7500g, 4oC, 10 minutes) and supernatent removed. The precipitate was

resuspended in buffer or water to the original sample volume. The supernatent was diluted as

required to attain original sample volume. The supernatent and precipitate at each saturation

level were analysed for total protein (Bradford assay) and phospholipase A2 activity

(phosphatidylcholine hydrolysis).

Methanol precipitation:

Methanol (-80oC) was added to samples to equivalent or greater volume and mixed

thoroughly. Samples were then stored in freezer overnight (-80oC). Samples were then

centrifuged (12000g, 0oC, 60 minutes). The extended storage followed by extended

centrifugation at low temperature was essential for pellet formation. The supernatent was

removed and discarded. The pellet was then air dried for a maximum of 5 minutes. The

pellet was resuspended in appropriate buffer or electrophoresis sample buffer.

Homogenising bowel tissue:

The mesentery was removed from bowel tissue. Bowel tissue was roughly chopped with a

scalpel. Tissue was weighed and placed in 5vol/g buffer (10mM Tris, pH 7.4). Homogenate

was clarified by centrifugation (2000g, 4oC, 10 minutes ).

35

Methods

Heparin Sepharose Affinity Chromatography;

Heparin Sepharose columns (Amersham Pharmacia Biotech, HiTrap columns) were

equilibrated with binding buffer (10 volumes, 10mM Tris-HCl, pH 7.4, flow rate of

approximately 1mL/minute). Samples were centrifuged (7500g, 5 minutes) before being

loaded onto pre-equilibrated column. Non-bound proteins were eluted with buffer (10 to 25

column volumes, 10mM Tris, pH 7.4) until eluent was clear and contained no protein (by

detection of absorbance at 280nm). The bound fractions were then eluted with increasing

concentrations of NaCl (0-5M NaCl, 10mM Tris, pH 7.4) in stepwise fashion either using a

single concentrated salt solution or a series of increasingly concentrated salt solutions.

Fractions were collected at regular time intervals (typically 30 seconds). Fractions were

analysed for total protein, phospholipase A2 activity and protein sizes (Tricine PAGE or SDS-

PAGE).

Preparative Isoelectric focusing using the BioRad Rotofor®:

Heparin column fractions with PLA2 activity eluted under conditions of 0.5M NaCl were

pooled. Glycerol and n-octyl glucoside (NOGS) were added to the sample to final

concentrations of 1% and 0.5% respectively. Biolytes (pH 3-10, BioRad, 1.5mL) were added

to the sample (total volume 40 mL). The Rotofor was pre-run with ultrapure water (15W

constant power, 5 minutes) to clear any contaminants from the apparatus. The Rotofor was

set to run at constant power (15watts, 4oC). The current was monitored throughout the run

and the experiment was stopped when the current had reached a plateau (approximately 3.5

hours). The fractions were then harvested and analysed for PLA2 activity. The pH of each

fraction was determined. The fractions were also run on a gel (NuPAGE Bis-Tris 4-12% gel

with Mark 12™ protein standards). After completion of the gel run the gel was fixed (30%

methanol, 10% TCA) with sulfosalicylic acid (3.5%) for 1 hour to remove ampholytes. The

gel was fixed for a further 2 hours (30% methanol, 12% TCA) to remove excess sulfosalicylic

acid. These gels were then silver stained using the PlusOne protocol.

Active Rotofor fractions of pH 8.98 to 9.8 were pooled and NaCl was added to a

concentration of 1M to electrostatically remove ampholytes. This solution was mixed and

incubated (45 minutes, 4oC). This solution was dialysed against 0.5X PBS (24 hours, 4oC)

using Slide-A-Lyzer® cassettes (Pierce) with a 3,500MWCO.

36

Methods

Electroelution:

The Mini Whole Gel Eluter apparatus (BioRad) was used to electroelute protein from SDS

gels. The gels were run under non-reducing conditions. The gel was equilibrated in elution

buffer (20mM CAPS pH 11, 15 minutes with three changes of buffer). The elution was

performed (20 minutes, 100mA), and at completion the electrodes were briefly reversed. The

fractions were collected and analysed for PLA2 activity. Pre-stained molecular weight

standards (See Blue®, Invitrogen) were used to construct a standard curve of mobility and

protein size.

Vivaspin concentration and desalting:

Vivaspin centrifugal filter devices (PES membrane, 5 000 MWCO, 500µL capacity,

Sartorius) were used to concentrate protein samples and exchange buffers. Samples were

applied to the upper chamber and centrifuged (10 000g, 10 minutes, 4oC). PBS (500µL) was

added and centrifuged (10 000g, 10 minutes, 4oC). The washing process was then repeated.

If ampholytes needed to be removed NaCl was added to the sample to a final concentration of

1M to electrostatically strip protein bound ampholytes (BioRad Rotofor application guide).

TOTAL PROTEIN ASSAYS:

BioRad DC Protein High Range Assay:

Protein assays were performed according to manufacturer’s instructions. Briefly, a standard

curve was constructed using BSA diluted in PBS in the range 0-2mg/mL. Samples were

diluted 1:50, 1:100, 1:200 or 1:300 in PBS. Samples and standards (5µL per well) were

assayed in triplicate. The absorbance was read at 750nm.

Bradford Total Protein Assay:

The protein assay of Bradford (1976) was used routinely. Standard curve composed of BSA

in PBS in the range 1-10 µg per 100µL. Samples diluted in PBS until the absorbance of the

solution is less than the absorbance of the highest standard. Protein reagent was added to

samples and standard in the ratio of 10 parts reagent to 1 part sample or standard. The

solutions were mixed thoroughly and the absorbance read at 595nm within 60 minutes of

addition of protein reagent.

37

Methods

Pierce BCA Protein Assay:

The BCA Protein Assay was used for samples containing carrier ampholytes. The Pierce

BCA Protein Assay Kit standard protocol was followed using a sample to working reagent

ratio of 1:20. A standard curve was constructed using BSA in the range 0-2000µg/mL.

Samples were prepared using a method to remove carrier ampholytes (Brown, Jarvis and

Hyland, 1989). Briefly, samples containing carrier ampholytes (100µL) were diluted in

ultrapure water (900µL). Deoxycholate solution (100µL, 0.15%) was added and solutions

were incubated (room temperature, 10 minutes). TCA solution was then added (100µL of

72% solution) and samples vortexed, then centrifuged (2000g, room temperature, 15

minutes). Supernatent was discarded and pellet immediately resuspended in detergent

solution (5% SDS in 0.1M NaOH). The BCA working reagent was immediately added

(2mL). Absorbance was read at 562nm.

38

Methods

RECIPES:

Bradford Protein reagent:

Method according to Bradford, 1976.

Coomassie Brilliant Blue G-250 final concentration 0.01%

Ethanol final concentration 4.7%

Phosphoric acid final concentration 8.5%

In ultrapure water.

Colloidal Coomassie/Neuhoff reagent:

Ammonium sulphate (17% w/v)

Phosphoric acid (4.2% v/v)

Methanol (34% v/v)

Coomassie G-250 (0.1% w/v)

in ultrapure water.

Solubilisation solution:

5M urea

2M thiourea

2mM tributylphosphine

2% CHAPS

2% sulfobetaine 3-10

0.2% carrier ampholyes

40mM Tris

0.002% bromophenol blue

1% ASB14 (BioRad)

20mM DTT

0.2% carrier ampholytes pH 3-10

Proteomics: Equilibration solution:

2.5% acrylamide

5mM tributylphosphine

20% glycerol

6M urea

39

Methods

2% SDS

0.375M Tris-HCl pH 8.8

Tris Glycine SDS running buffer:

29g Tris base

144g Glycine

10g SDS

to 1L ultrapure water.

Tris Glycine sample buffer:

2.5mL 0.5M Tris HCl pH 6.8

2mL glycerol

4mL 10% SDS

0.5mL 0.1% bromophenol blue

ultrapure water to 10mL

Two-dimensional electrophoresis equilibration buffer:

50mM Tris pH 8.8

6M urea

30% glycerol

2% SDS

bromophenol blue

2mL per strip, containing either 10mg/mL DTT or 25mg/mL IAA.

ELISA wash buffer:

Amounts for 1Litre of buffer:

NaCl (8g)

KCl (0.2g)

NaH2PO4.H2O (0.2g)

Na2HPO4 (1.96g)

Tween 20 (0.5mL)

BSA (10g) added directly before use

40

Methods

Carbonate buffer:

Amount for 500mL, pH 9.8.

Na2CO3 (1.1g)

NaHCO3 (1.5g)

MgCl2 (0.2g)

Phosphate buffered saline:

Amounts for 1L, pH 7.4.

Na2HPO4 (1.44g)

NaH2PO4 (0.24g)

NaCl (8g)

KCl (0.2g)

NuPAGE 4X LDS sample buffer:

Glycerol 40%

Tris base 6.82% w/v

Tris HCl 6.66% w/v

LDS 8% w/v

EDTA 0.06% w/v

Serva Blue G-250 0.075% solution

Phenol red 0.025% solution

In ultrapure water

NuPAGE 20X MES running buffer:

MES 1M

Tris Base 1M

SDS 69.3mM

EDTA 20.5mM

In ultrapure water.

41


Chapter 3: Plasma and Proteomics.

Global techniques:

Global techniques aim to detect all of the expression products (RNA or protein) in a

particular sample in a non-targeted, single experiment. Global analysis may be

undertaken to identify drug targets, diagnostic markers, markers of infection, markers

of developmental stages, or organism specific markers.

Nucleic acid based examples of global analysis include differential display RT-PCR,

serial analysis of gene expression (SAGE) and cDNA micro-arrays. Protein based

examples of global analysis include surface enhanced laser desorption ionisation

(SELDI) protein chips, multi-dimensional HPLC, antibody arrays and two-

dimensional electrophoresis. There are several advantages and disadvantages

associated with either nucleic acid and protein based global techniques, as

summarised in the table below.

NUCLEIC ACID BASED GLOBAL TECHNIQUES:

Table 3.1: The advantages and disadvantages of nucleic acid based global

techniques:

Advantages: Disadvantages:

Less material required (~5X103cells)

(Seliger & Kellner, 2002).

No detection of isoforms is possible

(Seliger & Kellner, 2002).

Amplification steps are possible (Seliger

& Kellner, 2002).

Alternate gene splicing adds complexity

(Banks et al, 2000, Wilkins et al, 1995).

Only potential functional state is

indicated (Griffin & Aebersold, 2001).

42


PROTEIN BASED GLOBAL TECHNIQUES:

Table 3.2: The advantages and disadvantages of protein based global techniques:

Advantages: Disadvantages:

Drug targets generally work on proteins

(Banks et al, 2000).

Diversity in protein size, pI and

solubility.

Post translational modifications can be

investigated, these are not apparent from

nucleic acid information (Krishna &

Wold, 1993).

Dynamic range varies over 6 orders of

magnitude (Corthals & Nelson, 2001)

May provide information about function

(Griffin & Aebersold, 2001, Seliger &

Kellner, 2002).

More material is required (~5×106 cells)

Abundance and sub-cellular location are

identified (Griffin & Aebersold, 2001).

No amplification step is available.

Proteomics:

Proteomics describes the study of the ‘entire protein complement of the genome’.

Unlike the genome, the proteome is not a static entity but changes according to:

- tissue or cellular source or cell culture conditions,

- stage of development of the organism or cell,

- pathophysiological state.

Proteomics is a relatively new field, but is becoming increasingly important to

biologists with the completion of the Human Genome project and similar milestones

involving other organisms. It is believed that approximately one third of the gene

sequences in genome databases encode for proteins of yet unknown function (Banks

et al, 2000).

It is proposed that the number of genes in the human genome is in the order of 20 000

to 25 000 (Stein, 2004) and that at any one time, up to five thousand of these genes

are being transcribed. The number of genes provides an underestimate of the number

of proteins due to:

43


- post translational modifications (Herbert & Righetti, 2000), of which 200

have been described (Krishna & Wold, 1993),

- alternate gene splicing (Wilkins et al, 1995).

Preliminary studies suggest that the number of protein forms per gene ranges from

one to two in bacteria to three to six per gene in humans (Banks et al, 2000), resulting

in up to one million proteins (Herbert & Righetti, 2000). Another group has

proposed that up to 10 protein variants exist per locus [Traini, Herbert, Wilkins &

Williams, unpublished as cited by (Gabor Miklos & Maleszka, 2001)].

Proteomics involves:

- preparation of the sample to extract the maximum number of proteins in the

maximum yield,

- separation of the proteins, or their peptides,

- visualisation and differential analysis,

- identification of the proteins.

Proteomics has traditionally been associated with two dimensional electrophoresis as

the technique for separation.

Two dimensional electrophoresis involves the separation of the proteins firstly on the

basis of the isoelectric point (pI) and then on the basis of molecular weight. The first

dimension of two dimensional electrophoresis involves isoelectric focusing (IEF)

which separates proteins according to their isoelectric point. In the past this step has

been achieved using carrier ampholytes embedded in polyacrylamide filled tubes to

create the pH gradient. Immobilised pH gradient (IPG) strips have largely replaced

carrier ampholyte based IEF, as they have several advantageous features. IPG strips

are composed of a thin layer of polyacrylamide on a supporting plastic film. The pH

gradient is covalently coupled to the acrylamide. IPG strips can tolerate higher

protein loads, have stable and precisely determined pH gradients and are not

susceptible to distortion (which could occur with ampholyte based IEF gels). IPG

strips are also very accurate and commercially available which limits interlaboratory

reproducibility problems.

The second dimension involves polyacrylamide gel electrophoresis in the presence of

sodium dodecyl sulphate (SDS), which separates the isoelectrically focused proteins

44


by molecular weight. The gels used for the second dimension can be gradient gels or

gels of a single percentage of acrylamide, depending on the application. It has been

reported that two dimensional gels are able to resolve 3000 proteins on a single gel

(Wilkins et al, 1996). Although there are several disadvantages of 2DE based

proteomics it has remained the leading technique because of its ability to:

a) ‘visualise a very large number of proteins simultaneously’ (Haynes &

Yates, 2000) and

b) be used in a ‘differential display format’ (Haynes & Yates, 2000).

The deficiencies of two dimensional electrophoresis based Proteomics:

It is accepted that two dimensional gels are deficient in the resolution of several

classes of proteins including those with:

- extreme pI (<4 and >10) (Gygi & Aebersold, 2000, Gygi, Rist & Aebersold,

2000, Washburn, Wolters & Yates, 2000).

- extreme molecular weight (Gygi & Aebersold, 2000, Gygi, Rist &

Aebersold, 2000, Washburn, Wolters & Yates, 2000), for example proteins

>100kDa are under-represented and those >150kDa are usually not separated

in the 2nd dimension (Corthals et al, 1999), those <10kDa are unable to be

detected (Seliger & Kellner, 2002).

- low codon bias*/low abundance and therefore unable to be detected (Gygi &

Aebersold, 2000, Simpson et al, 2000, Washburn, Wolters & Yates, 2000),

- membrane associations (Gygi, Rist & Aebersold, 2000, Han et al, 2001,

Simpson et al, 2000)

- a high degree of hydrophobicity (Gygi, Rist & Aebersold, 2000, Han et al,

2001, Simpson et al, 2000)

* Codon bias is the tendency for a given gene to preferentially use one of several potential codons to incorporate a specific amino acid into a protein (Liebler, 2002). Highly expressed proteins have high codon bias value, conversely low abundance proteins have low codon bias values. Low abundance is defined by a codon bias of <0.1.

45


Various modifications have been proposed to address the limitations of two

dimensional gel based proteomics including:

- pre-fractionation of sample by size, solubility or isoelectric point,

- using a series of gels of restricted pH gradients (subproteomics),

- improved solubilization and extraction procedures.

Other approaches have sought alternatives for two dimensional electrophoretic gels in

the separation step. Technologies that have emerged involve the use of one or more

of the following techniques:

- multidimensional chromatography, or MudPit ,*

- ion exchange chromatography,

- size exclusion chromatography,

- capillary electrophoresis,

- fragmentation of proteins into peptides before separation,

- the isotope coded affinity tag peptide labelling (ICAT) approach.

Work has been completed to assess how much of a proteome is likely to be seen in

2DE based proteomics, using yeast as a model. Yeast is a relatively simple organism

in terms of its genome and proteome given that it contains only 6 000 open reading

frames and simplified post translational modifications (Liebler, 2002). Yeast

proteins vary over 5 orders of magnitude (Gygi et al, 2000), and an estimated 80% of

the proteome is composed of low abundance proteins (Dr. Ben Herbert, personal

communication). Almost 90% of the yeast proteome can be solubilised with existing

technology (Pedersen et al, 2003) but half of the proteins are too low in abundance to

be detected (Griffin, Goodlet & Aebersold, 2001). The analysis of low abundance

yeast proteins can be improved with pre-fractionation (Gygi et al, 2000).

* The multidimensional protein identification technology or MudPIT approach involves the digestion of all of the proteins in a sample, and then separation of the resulting peptides by strong cation exchange chromatography and reverse phase chromatography. The peptides are then submitted to tandem mass spectrometry and database searching (Link et al, 1999, Washburn, Wolters & Yates, 2001). The advantage of this approach is that the peptides generated have more uniform characteristics than the proteins from which they were derived, in terms of size, pI and hydrophobicity (Washburn, Wolters & Yates, 2001).

46


When the maximum amount of whole cell lysate was run in a 2D gel the average spot

detected represented 51 000 copies per cell (Gygi et al, 2000). When the sample was

pre-fractionated, proteins present at 1 000 copies per cell could be detected (Gygi et

al, 2000).

To allow the detection of low abundance proteins the relative concentrations must

obviously be increased (Simpson et al, 2000). Theoretically two approaches can be

used:

- increase the sample load (which can be impractical due to sample load

limitations of 2DE gels),

- increase the concentration of the low abundance proteins (which alters the

composition of the sample),

- fractionate the sample and analyse each fraction separately (pre-fractionation

or sub-proteomics).

If the sample composition is altered to allow the detection of low abundance proteins

the final sample will not

1. reflect the proteome

2. remain quantitative

Sample alterations may involve:

- the removal of high abundance proteins such as albumin,

- pre fractionation.

Proteomics in medicine:

There is significant potential for proteomic research in the biomedical field.

Proteomics can identify disease specific proteins which could then be used as

diagnostic markers, markers for the monitoring of disease progression, therapy

effectiveness, toxicity or drug targeting. Proteomics has several advantages over

genomic based research in the biomedical field. For example:

- examination of the genome cannot provide information on multi-gene

processes such as aging, stress or many diseases,

- most drugs target proteins rather than genes,

47


- post translational modifications may be critically important in disease and

these modifications cannot be predicted from the genetic sequence (Banks et

al, 2000).

The vast potential of proteomics in biomedical research is yet to be realised (Banks et

al, 2000). This is thought to be due to a lack of awareness of the advances that have

been made in the field (Banks et al, 2000) and the concept that the technique is

unrefined. Proteomic investigations of pathophysiological conditions have generally

focussed on a variety of cancers. Cancer proteome databases have been constructed

for a hepatocellular carcinoma cell line (Liang et al, 2002) and a colon cancer cell line

(Simpson et al, 2000). Human proteome maps also exist for plasma, urine,

cerebrospinal fluid, breast tissue and heart tissue (Banks et al, 2000). Proteomics has

allowed the identification of:

- several protein variants associated with the development of liver cancer

(Zeindl-Eberhart et al, 1994),

- proteins differentially expressed between non tumorigenic and metastatic

prostate epithelial cells (Griffin, Goodlet & Aebersold, 2001),

-several proteins that are differentially expressed between benign prostate

tissue and prostate carcinoma (Alaiya et al, 2001),

-a number of protein alterations in a neuropsychiatric disease (Merril &

Goldman, 1982),

- a number of novel serum markers in patients with lung cancer (Hanash,

Brichory & Beer, 2001),

- microheterogeneity of proteins (perhaps glycoproteins) in pancreatic juice of

patients with pancreatic cancer and pancreatitis (Scheele, 1982),

- a putative urinary marker of bladder squamous cell carcinoma (Celis &

Gromov, 1999).

The importance of low abundance proteins in medicine:

A large section of the proteome is composed of low abundance proteins (Gygi et al,

2000). The use of proteomics to study disease raises the issue of low abundance

proteins for a number of reasons:

- many important classes of proteins, for example transcriptional control

proteins are present in low abundance (Hoffman et al, 2001),

48


- many disease associated proteins are expected to be present in low copy

number (Griffin, Goodlet & Aebersold, 2001),

- the relative changes in the amount of a particular protein between a normal

and disease state are expected to be quite low (Gabor Miklos & Maleszka,

2001),

- micro-dissection of tissue and tumours is often used to obtain pure cell

populations, resulting in prohibitively small samples, in which low abundance

proteins cannot be identified (Adam et al, 2001).

Plasma Proteomics:

In this review, the focus will be on the plasma proteome rather than the serum

proteome, for simplicity and to include the largest range of proteins of the soluble

component of blood (Anderson & Anderson, 2002). An assessment of the

advantages and disadvantages of investigating proteins either in serum or plasma is

currently a priority of HUPO (the worldwide Human Proteome Organisation).

Plasma is the primary clinical specimen because it is one of the most easily obtained

and reflects many physiological and pathophysiological processes. The plasma

proteome can be regarded as the ‘largest and deepest’ version of the human proteome,

because plasma contains not only classical plasma proteins, but leakage proteins from

all tissues in addition to immunoglobulins (Anderson & Anderson, 2002). Plasma is

also one of the most difficult proteomes to work with, because of the number of

protein present, the number of protein variants (degradation products and post

translationally modified proteins) and the large dynamic range. It is common for

small number of proteins (albumin, α1-antitrypsin, α2 macroglobulin, transferrin, γ

globulins) to represent over 80% of the total protein in plasma (Georgiou, Rice &

Baker, 2001). The abundance of plasma proteins varies over ten orders of magnitude

(Anderson & Anderson, 2002). It is difficult to analyse a large proportion of the

plasma proteome using any single method. For example: two dimensional gel

electrophoresis based proteomics is likely to detect only the high abundance and/or

long lived proteins of a sample (Haynes & Yates, 2000), at the expense of the short

lived or low abundance proteins.

49


Detectable in 2D gels Likely to be not detectable in 2D gels

↓C-reactive protein

↑Haptoglobin

↑Serum amyloid A

↓PLA2

↑Haemoglobin

Figure 3.1:

The Plasma Proteome. Modified from Anderson N.L. & Anderson N.G. The

Human Plasma Proteome: History, Character and Diagnostic Prospects.

Molecular and Cellular Proteomics 2002; 1 (11): 845-867, with permission from

the author.

Plasma proteins encompass proteins with function within the plasma (haemoglobin,

albumin), proteins that have been released from tissues, immunoglobulins and

cytokines. Tissue leakage proteins are those proteins that are generally retained

within a particular tissue but are released into plasma during tissue damage because of

cell death or injury (Anderson & Anderson, 2002). In this way, cellular factors

including proteins can be used as markers of tissue damage in plasma (Noe, 2001).

An ideal marker of tissue damage would be present only in the tissue or organ of

interest (Noe, 2001). A number of highly tissue specific proteins have been

identified, such as pancreatic α amylase and cardiac muscle creatine kinase MB (Noe,

2001). The concentration of tissue leakage proteins in plasma varies greatly

according to the amount present in the tissue, the distribution of the protein within the

tissue, the mode of release into circulation and the clearance or catabolism of the

protein from circulation. A highly abundant protein in a particular tissue can become

a very attractive marker, as even moderately elevated levels in plasma can reflect

small amounts of tissue damage (Anderson & Anderson, 2002). For example,

cardiac myoglobin is released during myocardial infarction. A myocardial infarction

50


affecting less than 3g of cardiac tissue (a typical moderate infarct affects 45g) would

release cardiac myoglobin into plasma to a substantial concentration of 3µg/mL

(Anderson & Anderson, 2002).

Tissue damage proteins may be released from tissues during injury due to:

- leakage of cytoplasmic proteins,

- heat shock response,

- inflammatory response,

- signalling,

- protective mechanisms.

Plasma proteomics offers a method of evaluating protein differences between patients

with a disease of interest and control patients. If a suitable tissue damage marker can

be identified by a proteomic investigation, this becomes a significant result because

the test sample (plasma) is the same in a clinical situation. Obtaining plasma is a

minimally invasive procedure and samples would be routinely taken from patients for

a multitude of other tests, during diagnosis and treatment.

Serum Amyloid A, an important and abundant disease associated protein.

Serum amyloid A (SAA) is a protein family comprised of several isoforms, that have

high homology (50-95%) and similar size (see Figure 3.11, for sequence alignment of

human serum amyloid A variants). The SAA family can be divided into two groups-

the acute phase SAA isoforms (A-SAA) and the constituent isoforms (C-SAA). The

acute phase SAA are highly conserved and have been found in all vertebrates studied

(Uhlar & Whitehead, 1999). As the name suggests, the acute phase SAA are

involved in the acute phase response, where the concentration in blood can increase

up to 1000 fold (Uhlar & Whitehead, 1999). Acute phase SAA isoforms are among

the most responsive acute phase proteins (Ensenauer et al, 1994), with mRNA

detectable within 2 hours of inflammation (Pruzanski et al, 1995). Plasma levels of

acute phase SAA can increase from baseline levels (2-5µg/mL (Malle et al, 1997)) to

1mg/mL within 20 hours (de Beer et al, 1994). SAA is also one of the most rapidly

cleared acute phase proteins, with a half life of only 90 minutes under both normal

and inflammatory states (Husby et al, 1994). The acute phase SAA are the

51


precursors of the amyloid fibril protein A found in secondary or reactive amyloidosis

(Husby et al, 1994). Amyloidosis refers to a group of diseases characterised by

deposition of amyloid in a variety of tissues and organs (Husby et al, 1994), most

commonly the spleen, liver and kidney (Buxbaum & Tagoe, 2000). Constitutive

SAA (C-SAA) is maintained at a relatively constant concentration in blood at

approximately 50µg/mL (Yamada et al, 2001). C-SAA is only minimally inducible

during the acute phase response, for example, during renal allograft transplantation,

C-SAA only rises approximately 3 fold (Yamada et al, 2001). Constitutive SAA has

only been found in humans and mice (Uhlar & Whitehead, 1999). The relevance of

C-SAA is unknown (Husby et al, 1994). Unlike the acute phase SAA, the C-SAA

gene lacks IL-1 and IL-6 responsive elements (Watson, Coade & Woo, 1992).

In humans one isoform of SAA is predominant (SAA1), and this isoform comprises

95% of the total SAA (Malle et al, 1997). In humans four SAA genes have been

detected, SAA1 (five alleles), SAA2 (the acute phase isoforms, two alleles), SAA3 (a

pseudogene) and SAA4 (the constitutive isoform, one allele). SAA3 is a pseudogene

due to a single base insertion which causes an early stop signal. No mRNA or

protein have been found for this gene (Uhlar & Whitehead, 1999). These genes are

all located on chromosome 11, and are thought to be the result of gene duplication and

conversion (Buxbaum & Tagoe, 2000). The serum amyloid A proteins are produced

primarily in the liver, and circulate in blood complexed to lipids (Buxbaum & Tagoe,

2000). SAA synthesis has also been detected in epithelial and endothelial cells,

smooth muscle, macrophages and lymphocytes (Sipe, 2000). The precise function of

SAA is unknown, but the massive increase of SAA during the acute phase and high

degree of conservation across species suggests an important role in defence (Uhlar &

Whitehead, 1999). Possible SAA functions include lipid metabolism or transport,

regulation of extracellular matrix degrading enzymes, inflammatory cell recruitment

(Uhlar & Whitehead, 1999), bacteria clearance (Alsemgeest et al, 1995), tissue repair

and cholesterol metabolism during inflammation (Ensenauer et al, 1994).

There are six electrophoretic variants of human acute phase SAA which occur in three

possible phenotypes (Betts et al, 1991). Two of these isoforms occur in all

52


phenotypes, and in addition two or four other isotypes distinguish the phenotypes.

The acute phase SAA isoforms are summarised in Table 3.1:

Table 3.3. Acute Phase Serum Amyloid A Variants. Isoelectric point

N-terminal amino acids

Identification Reference

6.0 SFFSFL SAA-1α des arg Dwulet, Wallace & Benson, 1988.

6.0 FFSFL SAA-1α des arg minor variant

Strachan et al, 1989.

6.4/6.6 RSFFSFL SAA-1α Foyn Bruun et al, 1995, Strachan et al, 1989.

7.0 SFFSFLG SAA-2α des arg Beach et al, 1992.

7.0 FFSFL SAA-2α des arg minor variant

Strachan et al, 1989.

7.4 SFFSFLG SAA-2β des arg Beach et al, 1992.

7.5 RSFFSFL SAA-2α Beach et al, 1992.

8.0 RSFFSFL SAA-2β Beach et al, 1992.

All humans express the pI 6.0/6.4 pair, and individuals can be divided into three

groups based on the patterns of the additional acute phase SAA expressed:

Phenotype 1: isotypes with pI values of 7.0, 7.4, 7.5 & 8.0, exists in 33% of

individuals.

Phenotype 2: isotypes with pI values 7.0 & 7.5, exists in 61% of individuals.

Phenotype 3: isotypes with pI values 7.4 & 8.0, exists in 6% of individuals, (Strachan

et al, 1989).

Most Europeans and Americans show the SAA1 +/- N terminal arginine (pI 6) and

SAA 2α +/- N terminal arginine (pI 7 & 7.5) while the SAA 2β pair is rare (Husby et

al, 1994). Constitutive serum amyloid A (SAA4) also occurs as a number of

electrophoretic variants. The smaller variant (14kDa) is the non-glycosylated form

and has three alternate isoelectric points (7.3, 7.9 and 8.1) (Whitehead et al, 1992).

53


The larger variant (19kDa) is glycosylated and has two alterative isoelectric points

(7.3 and 7.9) (Whitehead et al, 1992).

Serum Amyloid A as a disease marker:

The elevation of SAA concentration in the acute phase response is a generalised

reaction to inflammation or infection, but several groups have sought to investigate

SAA as a disease marker.

In cattle with experimental respiratory infection SAA responded more quickly than

other acute phase proteins (Heegard et al, 2000). In a murine model of colitis SAA

levels in plasma preceded inflammatory signs in intestinal tissue and levels reflected

the severity of the disease (de Villiers et al, 2000). In the synovium of patients

suffering from rheumatoid arthritis, SAA appears to induce degradation of the

extracellular matrix (Migita et al, 1998). Patients with cerebral infarction showed

elevated SAA levels and the levels appeared to correlate with the clinical severity

(Ilzecka & Stelmasiak, 2000).

The previous examples examined the net level of SAA in disease. Several groups

have investigated the SAA subtype patterns in disease. To quote Alsemgeest and

colleagues (1995):

“if isoforms of SAA are differentially expressed during different disease

processes this might be of potential value in diagnostics”.

In a murine model different expression patterns of SAA1, SAA2 and SAA3 were

found and tissues could be categorised into those that expressed all three isoforms,

those expressing SAA1 and SAA2 and those with predominant SAA3 expression

(Meek & Benditt, 1986). In cows with a variety of diseases a heterogenous pattern of

isoforms was reported (Alsemgeest et al, 1995). In contrast, human patients with

sepsis, arthritis, renal or liver allograft rejection or administration of inteferon,

showed that SAA subtype response was similar irrespective of the ‘initiating

stimulus’ (Maury, Enholm & Lukka, 1985).

Phospholipase A2, a low abundance disease associated protein.

Phospholipase A2 is a protein implicated in several facets of disease. Its involvement

in inflammatory conditions has been widely studied because of its role in the

54


generation of eicosanoids by the liberation of arachidonic acid from phospholipid

molecules. Hydrolysis of phospholipids by phospholipase A2 also releases

lysophospholipid which can be metabolised to platelet activating factor, both of which

can elicit cell responses (Bomalaski & Clarke, 1993), see Figure 4.2. The

inactivation of phospholipase A2 by alkylation or heat prevents the inflammatory

response, which suggests that the inflammatory response is due to the enzyme’s

action rather than a non specific response to an inflammatory initiator such as

endotoxin (Bomalaski & Clarke, 1993).

Phospholipase A2 in bowel injury:

Phospholipase A2 levels have been shown to be increased in the plasma (Minami et

al, 1992) and the intestinal tissue (Lilja et al, 1995) of patients with Crohn’s disease.

Phospholipase A2 may be involved in the pathogenesis of bowel injury. The isoform

of phospholipase A2, type II is thought to be involved in the development of intestinal

inflammation in both Crohn’s disease and ulcerative colitis (Minami & Tojo, 1997).

Serum phospholipase A2 correlates well with the activity of ulcerative colitis and

Crohn’s disease (Nevalainen, Gronroos & Kortesuo, 1993). Inhibition of

phospholipase A2 has been shown to be partially protective against intestinal mucosal

injury in a model of shock (Xu, Lu & Deitch, 1995) and against lung injury caused by

intestinal handling (Thomas, Karnik & Balasubramanian, 2002).

Phospholipase A2 in gut ischaemia/reperfusion:

In rat models of intestinal ischaemia/reperfusion injury, increased levels of

phospholipase hydrolysis products have been observed in the mucosa (Otamiri et al,

1987, Otamiri & Tagesson, 1989). Phospholipase activity in the portal circulation of

rats with experimental ischaemia/reperfusion was ten fold higher than in the systemic

circulation suggesting that the enzyme was being released from the injured gut (Koike

et al, 2000). Inhibition of phospholipase A2 reduced mucosal injury and permeability

in a rat model of ischaemia/reperfusion (Otamiri, Lindahl & Tagesson, 1988). The

involvement of phospholipase A2 in disease including bowel injury is discussed in

greater detail in the introduction to Chapter 5.

55


Pre-fractionation to identify low abundance proteins:

The problem of analysing low abundance proteins in complex mixtures can be

overcome by pre-fractionation of the sample into a number of discrete sub-samples.

The sub-samples can then be further separated using two dimensional gel

electrophoresis based proteomics or another technique such as a series of

chromatographic columns. The pre-fractionation approach can be applied to a global

analysis scheme, which allows greater resolution, particularly in regions which are

usually distorted by high abundance proteins. Pre-fractionation can also be applied in

a targeted approach to purify a particular protein of interest. A number of pre-

fractionation techniques have been designed to overcome the limitations of current

proteomic, and include subproteomics or preparative isoelectric focussing prior to the

global protein separation.

Subproteomics:

Subproteomics involves the use of sequential protein extraction procedures followed

by several 2DE gels of each fraction. By using a series of protein extraction

solutions a biological sample can be fractionated according to relative solubility,

which also reflects the protein’s cellular compartment (secreted, cytosolic or

membrane). Numerous 2DE gels (down to a single pH unit IPG strips) can then be

run from each fraction and a composite map constructed (Cordwell et al, 2000). By

using this approach low abundance proteins can be identified because of increased

sample loading and resolving power (Cordwell et al, 2000). An additional advantage

if that the sequential protein extraction provides some information about the cellular

location of proteins identified (Cordwell et al, 2000).

Preparative IEF techniques:

The use of preparative iso-electric focussing as the preliminary pre-fractionation step

is a powerful approach that is particularly suited to low abundance and/or membrane

proteins. Isoelectric focussing (IEF) separates proteins on the basis of their

isoelectric point or pI, the pH at which there is no net charge on the protein.

Isoelectric focussing can be conducted using small amounts of sample on an IEF gel,

or an IPG strip. Preparative IEF uses large amount of sample and can be performed

using large gel tubes or in the liquid phase. The multi-compartment electrolyzer, free

flow electrophoresis and the Rotofor® apparatus are examples of preparative

56


isoelectric focussing systems. For further discussion of the Rotofor® apparatus, see

Chapter 4.

Multi-compartment electrolyzer:

The multi-compartment electrolyzer is a liquid based preparative IEF system

composed of a variable number of compartments (typically six) separated by

isoelectric membranes (Herbert & Righetti, 2000). Each compartment is capable of

producing a sub-sample with a discrete range of isoelectric points. The fractions can

then be analysed (without alteration of the chemical composition) using 2DE gels.

Disadvantages of this approach include possible loss of protein on the membrane

surface or precipitation of the proteins onto membranes at their pI (Herbert &

Righetti, 2000). This approach led to the identification of several acidic proteins

from human plasma that were unable to be seen in un-fractionated plasma (Herbert &

Righetti, 2000).

Free flow electrophoresis:

Free flow electrophoresis is another liquid based IEF system that can be used to

fractionate samples prior to 2DE gels. Sample recovery is optimal due to the absence

of solid membrane supports and sample load is less of a concern than with other

techniques due to continuous sample application (Hoffman et al, 2001). Free flow

electrophoresis has allowed the identification of a number of intact protein complexes

and membrane associated proteins (Hoffman et al, 2001).

Determining the Accuracy of Diagnostic Markers:

The effectiveness of a new diagnostic marker can be assessed in a number of different

ways. The accuracy of a test is assessed by the sensitivity and specificity.

Sensitivity is the ability of a test to correctly identify patients that have the condition

in question. Specificity is the ability of the test to correctly discount the patients who

do not have the condition in question. The positive predictive value of a test is the

probability that a subject is positive if a positive result is obtained. Conversely, the

negative predictive value is the probability that a negative subject results in a negative

result. Positive/negative predictive values are dependent on the prevalence of the

disease. Diagnostic likelihood ratios are independent on the prevalence of the

disease, and represent the odds ratio of obtaining a positive result among a negative

57


population, compared with the likelihood of obtaining a positive result among a

positive population. Other techniques to measure the effectiveness of a diagnostic

marker include univariate/multivariate analysis and receiver operating characteristic

(ROC) curves. The Receiver Operating Characteristic Plot:

The accuracy of a particular test in the diagnosis of a particular disease is often

evaluated using the concepts of sensitivity and specificity. The specificity and

sensitivity of a test depend on the decision threshold or cut-off point for the test.

ROC plots were first introduced to evaluate radar in the 1950s and have since been

employed in radiography, experimental psychology and pathology testing fields

(reviewed by Zweig & Campbell, 1993). ROC plots evaluate the test’s

‘ability to discriminate between alternating states of health over the complete

spectrum of operating conditions’ (Zweig & Campbell, 1993).

ROC graphs plot sensitivity against 1-specificity. Each point represents the

specificity/sensitivity pair for a particular cut-off point for a particular test. ROC

plots are useful in situations where

- there is no ‘gold standard’ with which to compare a new test (Zweig &

Campbell, 1993),

- if the accepted test has an associated bias (Zweig & Campbell, 1993),

- if the decision threshold has not been decided.

ROC plots are advantageous because

- the entire range of threshold values is assessed,

- a common scale is used, unlike a frequency diagram

which usually has different scales (Zweig & Campbell,

1993).

The ROC plot has a number of disadvantages:

- the decision threshold is not visually obvious from the graph (Zweig &

Campbell, 1993),

- the number of subjects is not shown by the graph (Zweig & Campbell, 1993).

58


When a plot is created the area under the curve is a measure of the discriminatory

power of that test. The closer the area under the curve is to 1, the better

discriminating power the test has. Conversely, an area of 0.5 has no more

discriminating ability than random chance.

Plasma Variables assessed in this Study:

APACHE II:

The APACHE system is a scoring system that was devised by a group at Washington

University and was first published in 1981 (Knaus et al, 1981). APACHE II

represented an improved version of the original APACHE system designed to predict

patients’ outcome in ICU based on twelve physiological variables, the Glasgow Coma

Score, chronic health status and age (Knaus et al, 1985, Siegel & Rixen, 2002).

APACHE II has been used as an indicator of patient status and to predict mortality in

patients with multiple trauma (Knaus et al, 1985, Rhee et al, 1990).

SAA:

SAA is one of the most reactive and most quickly cleared acute phase proteins in

humans. SAA can increase in concentration in the circulation from baseline levels to

200 to 500 times that level within 20 hours (de Beer et al, 1994). The half life of

SAA in normal and inflammatory states is only 90 minutes (Husby et al, 1994), and

therefore the clearance of this protein from circulation can be very rapid.

CRP:

CRP is the most widely used marker of inflammation in clinical practice at present

(Bellamy, Lansbury & Murdoch, 2002). CRP can reach levels of 500mg/L within 6

hours of the induction of the acute phase response (Bellamy, Lansbury & Murdoch,

2002). Ongoing infection or inflammation usually results in continually elevated

levels of CRP whereas declining levels usually indicate resolution of the condition

(Bellamy, Lansbury & Murdoch, 2002).

PLA2:

PLA2 is a rate limiting enzyme in the generation of a variety of inflammatory

mediators including the eicosanoids, lysophospholipid and platelet activating factor

(Dennis, 1997). Phospholipase A2 production and release into plasma/extracellular

59


fluid is increased during inflammation (Kallajoki & Nevalainen, 1997, Kudo et al,

1993). Extracellular phospholipase A2 levels can reach concentrations of up to 1000

times baseline during inflammation (Weinrauch et al, 1998).

LACTATE:

Elevated lactate concentration has been described as

“the best marker of mesenteric ischaemia to date” (Lange & Jackel, 1994).

A high lactate concentration is not a finding that is unique to mesenteric ischaemia,

but indicates the need for immediate surgery to avert life threatening deterioration.

Shock, diabetic ketoacidosis, convulsion and hepatic or renal failure can also lead to

elevated lactate levels, but if these conditions can be discounted then the likelihood of

mesenteric ischaemia increases (Lange & Jackel, 1994).

60

Results

Chapter 3: Proteomics and plasma investigations.

A proteomic investigation of potential differences between the plasma proteins of

patients with surgically confirmed bowel infarction and control patients was

undertaken. The control patients in this study were intensive care patients that were

being successfully fed, and therefore had functioning bowels. An initial investigation

of the neutral to acidic (pI of 4 to 7) proteins was undertaken as a starting point. The

major protein differences were found to be variants of haptoglobin (results not

shown). The basic plasma proteins (pI of 7 to 10) were investigated in greater depth

as it was thought that small basic proteins may be the first to transverse the gut barrier

into circulation. It is known that gut permeability increases as a result of ischaemia

(Kong et al, 1998). Well resolved plasma protein gels were achieved using cup

loading rather than in-gel rehydration, and a methanol precipitation cleanup step

(Figures 3.2 and 3.3). A number of protein differences were evident between the

samples from patients with bowel infarction and control patients (Figure 3.4).

Proteins of interest were identified using MALDI-TOF mass spectrometry and

database searching. If MALDI-TOF did not provide an adequate database match, the

protein was identified using Q-TOF tandem mass spectrometry and database

searching. The relative quantities of each of the proteins were assessed using image

analysis software (Figure 3.5). The presence or absence of each spot in each gel was

also noted (Figure 3.6). From the analysis of protein quantity, serum amyloid A

appeared to warrant further investigation. The mass spectrometry analysis of serum

amyloid A gave good sequence coverage (Figure 3.7), and Q-TOF was also able to

distinguish between SAA2 and SAA1 (Figure 3.8), two proteins that share very high

homology (96%). Three serum amyloid protein variants were identified in this study,

and the relative proportions of each is shown in Figure 3.9, though there is no clear

pattern that distinguishes the bowel infarction patients from controls.

An analysis was conducted to compare the ability of the following variables in

distinguishing patients with bowel infarction from control patients:

- a commonly used clinical marker (C-reactive protein)

- a commonly used clinical scoring system (APACHE II)

- an indicator of bowel infarction (lactate)

- a protein suspected to be released from infarcted bowel (phospholipase A2)

- serum amyloid A

61

Results

The analysis of serum amyloid A was conducted using an ELISA system, due to a

superior limit of detection than image analysis of gel spots. SAA levels were not

significantly different between the plasma of control and bowel infarction patients

(Figure 3.13).

Previous research had shown that phospholipase A2 activity was increased in the

intestinal mucosa in rat models of ischaemia/reperfusion injury (Koike et al, 1992,

Koike et al, 1995, Otamiri & Tagesson, 1989). In this investigation we wished to

investigate the possible release of phospholipase activity into the plasma of patients

with bowel infarction. Phospholipase A2 activity was measured in the plasma of

patients involved in this study (Figure 3.10). Each variable was assessed in groups of

patients with bowel infarction and control patients (Figure 3.11). C-reactive protein

showed a trend (p<0.025) toward increased levels in the patients with bowel

infarction, but the other variables showed poor discrimination between the two groups

of patients, as assessed by t-tests of the mean data. Receiver operating characteristic

curves of the five variables were constructed to allow comparison on a common scale

(Figure 3.12). The variable that showed the best discrimination power was

phospholipase A2, though none of the variables assessed were ideal for use in a

clinical setting. The next investigations focused on the phospholipase A2 activity in

normal and infarcted human bowel (Chapter 4) and the purification of the protein

responsible for the increased phospholipase A2 activity (Chapter 5).

62

63

A B C

D E F

Figure 3.2: Two Dimensional gels of plasma proteins from control patients A to F. Large

format gels (8-18% gradient) and basic IPG strips (pH 7-10) were employed. These gels

were stained with the colloidal Coomassie stain. These gels have been cropped to show

proteins smaller than 50kDa.

pH7 10

kDa

~50

~25

~15

64

1 2 3

6 8 9

kDa~50

~25

~15

pH7 10

Figure 3.3: Two Dimensional gels of plasma from patients with confirmed bowel infarction,

labelled here as patients G to L. Large format gels (8-18% gradient) and basic region IPG strips

(pH 7-10) were employed. These gels were stained with the colloidal Coomassie stain. These

gels have been cropped to show proteins smaller than 50kDa.

65

Figu

re 3

.4: T

ypic

al tw

o-di

men

sion

al g

el sh

owin

g pl

asm

a pr

otei

ns fr

om a

pat

ient

with

bow

el in

farc

tion

(left

hand

side

) and

pla

sma

from

a c

ontro

l pat

ient

(rig

ht h

and

side

).

The

gel h

as b

een

crop

ped

to sh

ow o

nly

the

smal

l pro

tein

s. I

mm

obili

sed

pH g

radi

ents

(pH

7-1

0) a

nd la

rge

form

at g

radi

ent g

els (

8-18

%) w

ere

empl

oyed

. M

ass s

pect

rom

etry

was

per

form

ed o

n sp

ots l

abel

led

in re

d (A

1-7)

. M

ass S

pect

rom

etry

ana

lysi

s em

ploy

ed

MA

LDI a

nd Q

-TO

F.

66

00.020.040.060.08

0.10.120.140.160.18

0.2

Haemoglobinbeta

fragment a

Haemoglobinbeta

fragment b

Haemoglobinalpha

Serumamyloid

Serumamyloid A2

A

Serumamyloid A2

Q9NV34

Qua

ntity

.

Infarcted bowel Normal bowel

* ** ***

Figure 3.5: Relative quantities of spots selected for Mass Spectrometry analysis. Analysis

was performed using BioRad Quantity One Software, version 4.1.1. Proteins that were

significantly different by the student’s t-test are highlighted with asterisks (* for p<0.05,

** for p<0.02, *** for p<0.01 levels of significance).

67

Protein: Control Plasma: Bowel Infarction plasma: Haemoglobin α 6/6 6/6 Haemoglobin β fragment a 6/6 6/6 Haemoglobin β fragment b 5/6 6/6 Q9NV34 theoretical protein 6/6 6/6 Q9P1I9 theoretical protein 1/6 0/6 SAA (insufficient peptide information to distinguish the specific variant)

5/6 6/6

SAA 2 3/6 6/6 SAA A2 5/6 6/6

Table 3.4: Protein spots present in 2DE gels of plasma.

68

Spot Type of Mass Spec.

Identity Number of peptides obtained

Sequence coverage

A2 MALDI SAA, (unknown variant)

7 58.70%

A5 MALDI SAA 2-α 7 61.51% A7 MALDI - 0 - A7 Q-TOF SAA2 3+sequence data 25.96% by amino

acids

Figure 3.6: Serum amyloid identification by Mass Spectrometry.

69

SAA1 MKLLTGLVFCSLVLGVSSRSFFSFLGEAFDGARDMWRAYSDMREANYI SAA2 MKLLTGLVFCSLVLGVSSRSFFSFLGEAFDGARDMWRAYSDMREANYI SAA3 MKLSTGIIFCSLVLGVSSQGWLTFLKAAGQGAKDMWRAYSDMKEANYK SAA4 MRLFTGIVFCSLVMGVTSESWRSFFKEALQGVGDMGRAYWDIMISNHQ SAA1 GSDKYFHARGNYDAAKRGPGGAWAAEVISDARENIQRFFGHGAEDSLA SAA2 GSDKYFHARGNYDAAKRGPGGAWAAEVISNARENIQRLTGHGAEDSLA SAA3 KSDKYFHARGNYDAVQRGPGGVWATEVISDARENVQRLTGDHAEDSLA SAA4 NSNRYLYARGNYDAAQRGPGGVWAAKLISRSRVYLQGLIDYYLFGNSS SAA1 DQAANEWGRSGKDPNHFRPAGLPEKY SAA2 DQAANKWGRSGRDPNHFRPAGLPEKY SAA3 GQATNKWGQSGKDPNHFRPAGLPEKY SAA4 TVLEDSKSNEKAEEWGRSGKDPDRFRPDGLPKKY PEPTIDE A SFFSFLGEAFDGAR PEPTIDE B SFLGEAFD PEPTIDE C GPGGAWAAEVISNAR

Figure 3.7: Sequence alignment of serum amyloid isoforms SAA1-4. Q-TOF Mass

Spectrometry allowed the spot ‘A7’ to be identified as SAA2 from peptide denoted ‘C’.

SAA1 and SAA2 differ by only 5 from 104 amino acids, and share the same Swiss-Prot entry

because of this homology. The high degree of homology between the isoforms

is likely to be the reason that it was not possible to determine the specific isoform

corresponding to spot A2.

70

SAA gel spots- control patients.

00.050.1

0.150.2

0.25pl

asm

a11

plas

ma1

2

plas

ma1

3

plas

ma1

4

plas

ma1

7

plas

ma2

0

Qua

ntity

.

spot A2spot A5spot A7

SAA gel spots- Bowel infarction patients.

00.050.1

0.150.2

0.25

plas

ma1

plas

ma2

plas

ma3

plas

ma6

plas

ma8

plas

ma9

Qua

ntity

.

spot A2spot A5spot A7

Figure 3.8: Serum amyloid proteins present in plasma of control patients and

patients with confirmed bowel infarction. Analysis was performed using

BioRad Quantity One software version 4.1.1. Spot A2= SAA, spot A5=

SAA 2α, spot A7= SAA 2.

71

SAA1 MKLLTGLVFCSLVLGVSSRSFFSFLGEAFDGARDMWRAYSDMREANYISAA2 MKLLTGLVFCSLVLGVSSRSFFSFLGEAFDGARDMWRAYSDMREANYISAA3 MKLSTGIIFCSLVLGVSSQGWLTFLKAAGQGAKDMWRAYSDMKEANYKSAA4 MRLFTGIVFCSLVMGVTSESWRSFFKEALQGVGDMGRAYWDIMISNHQ

SAA1 GSDKYFHARGNYDAAKRGPGGAWAAEVISDARENIQRFFGHGAEDSLASAA2 GSDKYFHARGNYDAAKRGPGGAWAAEVISNARENIQRLTGHGAEDSLASAA3 KSDKYFHARGNYDAVQRGPGGVWATEVISDARENVQRLTGDHAEDSLASAA4 NSNRYLYARGNYDAAQRGPGGVWAAKLISRSRVYLQGLIDYYLFGNSS

SAA1 DQAANEWGRSGKDPNHFRPAGLPEKYSAA2 DQAANKWGRSGRDPNHFRPAGLPEKYSAA3 GQATNKWGQSGKDPNHFRPAGLPEKYSAA4 TVLEDSKSNEKAEEWGRSGKDPDRFRPDGLPKKY

Leader peptide: residues 1-18Heparin binding sites: residues 78-104SAA4 octapeptide extension: residues 70-77

Figure 3.9: Serum amyloid A isoform sequence alignment. The leader peptide (18

amino acids in length) is cleaved to form the mature protein.

The serum amyloid 4 octapeptide extension is not present in the acute phase isoforms.

This octapeptide is the only potential N-linked glycosylation site in the molecule.

Studies have shown that 50% of the proteins are glycosylated giving rise to two size

classes differing by 5kDa (Whitehead et al, 1992).

72

01234567

BOWEL INFARCTION CONTROL

PLA

2 ac

tivity

(%

hyd

roly

sis)

Figure 3.10: PLA2 activity in the plasma of bowel infarction patients and

control patients. Results are expressed as the mean +/- standard error of

the PLA2 activity of bowel infarction (n=10) and control plasma (n=9).

The PLA2 activity in the plasma of bowel infarction patients and control

patients is not significantly different.

73

LACTATE

01234567

Bowel Infarction Control

LACT

ATE

(mm

ol/L

)

PLA2 ACTIVITY

0

2

4

6

8

Bowel Infarction ControlPLA

2 H

YDRO

LYSI

S (%

)

CRP

0

100

200

300

400


CRP

(ug/

mL)

APACHE

05

1015202530


APAC

HE

SCO

RE *

SAA

0

200

400

600

800


SAA

(ug/

mL)

Figure 3.11: A visual comparison of the five variables measured in the plasma of

control patients and patients with confirmed bowel infarction.

The differences for each of these variables was not significant except for CRP

levels, (* p<0.025, student’s t test). Data are represented as the mean +/- standard

error. Number of samples in bowel infarction group and control group

respectively: CRP n=7 in each group, Lactate n=7 & n=10, APACHE n=10 in each

group, PLA2 n=10 & n=9, and SAA n=10 & n=9. The numbers of samples in each

group varies because of differences in the availability of clinical testing.

74

Figure 3.12: The ROC plot and the area under the curve obtained for each variable

measured in the plasma of control patients and patients with surgically confirmed

bowel infarction. The separator variable was the presence or absence of bowel

infarction. As the area under the ROC curve approaches 1, the test approaches

perfect discrimination power.

Discussion

Chapter 3 Discussion:

Our proteomic investigation examined possible plasma protein differences between

patients with surgically confirmed bowel infarction and control patients. This

investigation was undertaken as a starting point for this project and to assess whether

a highly abundant marker of tissue damage or a diagnostic protein was evident in the

plasma of bowel infarction patients but not in controls. The control patients were

Intensive Care patients that were being successfully fed, and thus had functioning

bowels. The control patients had been admitted to intensive care for a wide variety

of conditions. Our investigation did not recruit normal healthy volunteers as

controls, to try to limit protein differences that were simply indicative of health status.

The majority of proteomic investigations of disease associated proteins have focussed

on a variety of cancers, often at the tissue level. The control patients in proteomic

investigations of plasma proteins tend to be healthy volunteers. We chose to use

plasma (rather than bowel tissue, for example) as our sample of choice, as a clinically

useful diagnostic test would be minimally invasive, and be able to be used to

continually monitor ‘at risk’ patients. Our pilot investigation focussed on the neutral

to acidic (pH 4-7) proteins. The specific proteins identified from this region in

patients with confirmed bowel infarction were variants of haptoglobin.

Haptoglobin variants have been observed in two dimensional electrophoretic gels of

human serum during and after the acute phase, and were predictably more abundant

during the acute phase (Bini et al, 1996). Haptoglobin is thought to aid in tissue

repair by stimulating angiogenesis, which may compensate for the decreased oxygen

supply during ischaemia (Cid et al, 1993). Haptoglobin also has antioxidant

properties and protects against damage by reactive oxygen species (Gabay &

Kushner, 1999). The increased levels of haptoglobin observed during infection and

inflammation are a non-specific response triggered by cytokines (Cid et al, 1993) and

is therefore not of specific diagnostic value for bowel infarction. Although the role

of haptoglobin in ischaemic injury to bowel may be an interesting area for further

research, it was not investigated in greater detail in this project due to its non specific

increase during the acute phase.

75

Discussion

Cup loading and methanol precipitation:

Our investigation then focused on the small basic proteins present in the plasma of

patients with and without surgically confirmed bowel infarction. The small, charged

proteins were thought to be the first to be released from injured bowel (into the lumen

and subsequently into general circulation). Bowel permeability is known to increase

as a result of ischaemia (Kong et al, 1998). Basic proteins are known to be difficult

to resolve by in gel rehydration, which is the usual method of uptake of proteins into

IPG strips. Cup loading is commonly recommended for uptake of proteins into

alkaline IPG strips (Gorg et al, 1999, Gorg et al, 2000). Our work confirmed that

cup loading of basic proteins was more successful than in gel rehydration (results not

shown). It is also thought that plasma proteins in general are more successfully

resolved with cup loading than in gel rehydration (Australian Proteome Analysis

Facility, personal communication). Methanol precipitation was also found to give

superior results, in terms of superior resolution of protein spots and less streaking.

Methanol precipitation has been reported to improve resolution by removing non-

protein contaminants such as lipids, polysaccharides, salts and nucleic acids

(Amersham Pharmacia applications manual). The improved resolution provided by a

methanol precipitation step was even more evident with gels of bowel content

samples, which are likely to contain more non-protein contaminants than plasma.

The gels of bowel content samples are not shown in this report, as the protein patterns

were too heterogenous to draw any conclusions or locate particular proteins for

identification by mass spectrometry.

PROTEIN DIFFERENCES FOUND BY 2D-PAGE:

A number of protein differences were evident from a composite synthetic gel

constructed from the gels of plasma proteins from the two groups of patients. A

number of haemoglobin variants were found, and, not surprisingly, the levels of these

proteins were similar between the two groups of patients. Two proteins that were

identified by MALDI-TOF mass spectrometry had been entered into the protein

databases as ‘theoretical proteins’. These two proteins had little potential as

diagnostic proteins as one was present in all samples at similar levels (Q9NV34), and

the other was present in only one sample (Q9P1I9), but were interesting nevertheless.

76

Discussion

Q9NV34 was present in all controls and all patients with confirmed bowel infarction.

Q9NV34 was an excellent match for v-maf musculoaponeurotic fibrosarcoma

oncogene family proteins (BLASTN, BLASTP). Maf genes are highly conserved in

vertebrates (Iwata et al, 1998) and have been implicated in a variety of physiological

processes including the stress response (Moran, Dahl & Mulcahy, 2002). The maf

proteins in disease would be an interesting area for further investigation. It is

interesting to note that the level of Q9NV34 was similar in the plasma of each control

patient, despite the heterogenous disease pattern of this group. Q9P1I9 was present

in only 1/6 control patients and no patients with bowel infarction. The amino acid

sequence of Q9P1I9 showed no matches to any entered sequence in either the

nucleotide or protein databases. Q9P1I9 showed limited similarity to a

metalloproteinase inhibitor (Propsearch), but matched no recognised motifs (Prosite).

Propsearch uses amino acid composition instead of sequence to find similar proteins.

Propsearch can provide information about molecular weight, hydrophobicity, charged

residues and distribution of residue sizes. Motif scan (Prosite) identifies similarities

to recognised protein motifs (Falquet et al, 2002).

SERUM AMYLOID A:

Three variants of serum amyloid A were identified in the plasma of patients with

surgically confirmed bowel infarction by mass spectrometry in this study. Serum

amyloid A exists as a complex family of proteins, and includes six electrophoretic

variants of the acute phase types (Betts et al, 1991) in addition to five electrophoretic

variants of the constitutive type (Whitehead et al, 1992). Serum amyloid A variants

exist in three recognised phenotypes, with either four and six acute phase variants, in

addition to up to five constitutive variants. All of the constitutive variants and either

two or four acute phase variants (depending on phenotype) will theoretically resolve

in the pH region 7-10 of a two dimensional gel, though some are likely to be below

the detection limit of the gel stains. In this study, the quantity of the SAA2 and

SAA2α variants appeared to be elevated in the plasma of patients with bowel

infarction, compared with control patients. This is an interesting finding as the level

of SAA and other acute phase proteins would be expected to be similar among

intensive care patients with similar levels of critical illness.

77

Discussion

If the expression profile of the SAA variants was different in various tissues and/or

different diseases then this protein family may be of diagnostic significance. This

idea has been proposed by groups studying a variety of diseases in both cattle

(Alsemgeest et al, 1995) and humans (Maury, Enholm & Lukka, 1985). Human

patients with a variety of diseases showed a similar SAA subtype pattern regardless of

the disease (Maury, Enholm & Lukka, 1985). Serum amyloid A variants have been

observed in two dimensional gels of human serum during the acute phase, but never

after recovery (Bini et al, 1996). In cattle, a heterogenous pattern of SAA isoforms

was observed (Alsemgeest et al, 1995). In our study the SAA variants were

generally higher in the patients with confirmed bowel infarction than in controls, but

no clear pattern of variants was evident between the two patient groups.

The function of serum amyloid A during the acute phase is not yet known, but

functions such as lipid metabolism, recruitment of inflammatory cells (Uhlar &

Whitehead, 1999), cell adhesion and chemotaxis of phagocytes and lymphocytes

(Berliner et al, 1995), tissue repair (Ensenauer et al, 1994) or bacterial clearance

(Alsemgeest et al, 1995) have been proposed.

Mass spectrometry analysis of serum amyloid A:

One variant of SAA (SAA2) was unable to be identified by MALDI-TOF mass

spectrometry, but was successfully identified by Q-TOF tandem mass spectrometry.

The variant submitted to Q-TOF analysis was identified as SAA2, which was a

fortuitous result, as SAA1 and SAA2 differ by only 4 out of 105 amino acids (Betts et

al, 1991), and share the same entry in protein databases because of their very high

homology. SAA1 is thought to be the major SAA variant in humans, comprising

over 95% of the total SAA in plasma (Malle et al, 1997), so it was interesting to

identify variants other than SAA1 in this study. The SAA1 variants have the

isoelectric points 6.0, 6.4 and 6.6, and therefore would not have resolved within the

IPG strip range employed in this study.

Serum Amyloid A ELISA:

The plasma levels of serum amyloid A in bowel infarction and control patients were

assessed in greater depth using an ELISA. It was expected that the levels of SAA

would be greater in the plasma of patients with bowel infarction compared with

78

Discussion

controls, from the proteomic results. The ELISA was employed because of its

superior limit of detection compared to the quantification of gel spots (ELISA limit of

detection 5ng/mL, compared to a gel spot which represents 0.8-1µg/mL). There was

a trend toward increased serum amyloid A levels in the plasma of bowel infarction

patients compared with controls, but the difference was not significant The antibody

used in the ELISA cross reacts with both SAA1 and SAA2 (TriDelta laboratories,

personal communication). Any subtle differences in the subtype patterns of SAA

between the two patient groups is therefore likely to have been overlooked with this

approach. To our knowledge, no antibodies have been raised that can distinguish the

SAA subtypes, therefore subtype specific ELISA are not yet possible. As SAA1 is

the predominant SAA isoform, this may have masked subtle differences in other SAA

variants. To test this hypothesis, western blotting could be undertaken using an

acute phase SAA antibody and proteins arrayed by pH 3-10 two dimensional gels.

Although the use of an acute phase protein as a possible diagnostic marker may seem

inappropriate, a number of examples of this approach have been reported:

- CRP has been recently recommended as a marker of coronary disease risk

(Pearson et al, 2003).

- All of the patients with bacterial infections in one study showed elevated

SAA, while only 6/16 patients with viral infection had elevated SAA levels

(Bini et al, 1996).

Until subtype specific SAA antibodies can be generated and corresponding ELISAs

developed, the increased levels of SAA and possible involvement of SAA variants in

bowel ischaemia/infarction are difficult areas to study in greater depth.

PHOSPHOLIPASE A2:

Another protein liberated in response to tissue injury, phospholipase A2, was

suggested in the literature to be of possible importance in bowel ischaemia and

infarction. For example, a rat model of ischaemia/reperfusion injury showed that

phospholipase A2 was increased in the portal circulation to a tenfold higher level than

in the systemic circulation, suggesting that circulating phospholipase A2 in ischaemia

was released by the injured gut (Koike et al, 2000). Phospholipase A2 could not be

studied using a two dimensional proteomic approach because of its low abundance,

and was studied using catalytic activity assay as an alternative. The patients with

79

Discussion

confirmed bowel infarction in this study showed higher plasma PLA2 activity than

control patients, but the difference was not significant. The phospholipase

complement of bowel is complex, and a subtle difference between bowel infarction

and control patients could not be examined using a catalytic activity assay. The

phospholipase A2 assay employed in this project, and many other publications

quantifies the net in vitro catalytic activity and is not able to determine the presence or

contribution of the various isoenzymes (Farrugia et al, 1993). The aim of these

experiments was to assay the catalytic activity of the enzyme (rather than the amount

of protein which may not be active or bound to inhibitors). It is possible that the

hydrolysis products of this enzyme may be of greater importance than the protein

isoform itself. Protein purification was undertaken to identify the protein responsible

for the increased phospholipase activity (Chapter 5).

APPLICATION OF PROTEOMICS TO BIOMARKER DISCOVERY:

A number of issues were presented by the proteomic investigations. Although the

number of patients in each group was large in terms of a proteomic investigation (10

samples from each group), this number is small in terms of a clinical investigation.

Biological variability results in large differences in protein expression between

patients. There are also significant fluctuations in protein levels within a patient.

For example; one study investigated C-reactive protein in a group of healthy

individuals and found that protein levels fluctuated over a six month period (Macy,

Hayes & Tracy, 1997). These fluctuations equated to coefficients of variation of

42% within an individual and 93% between individuals (Macy, Hayes & Tracy,

1997). There are also fluctuations in disease associated proteins, such as the acute

phase proteins (see Figure 3.14), over the time course of the development of the

disease. These fluctuations add even greater complexity to the development of

reference ranges and subsequent diagnostic tests.

80

Discussion

Figure 3.14 The Acute Phase proteins.

Modified from Gitlin J.D. & Colten H.R., Molecular Biology of the acute phase

plasma proteins. Lymphokines 1987; 14: 123-153 with permission from Elsevier.

Low abundance proteins:

Low abundance plasma proteins are more likely to be more clinically important than

highly abundant or long lived proteins (Griffin, Goodlet & Aebersold, 2001).

However, the majority of the proteins identified by proteomics in this study were high

abundance and/or acute phase proteins. There was a clear need for pre-fractionation

of a clinical sample prior to two-dimensional gels in order to identify low abundance

proteins, and this would become the basis for future proteomic investigations for this

project. Calculation based on our sample volumes and protein stain capabilities show

that a protein would need to present in a concentration of greater than 0.8µg/mL to be

identified in our two dimensional gel electrophoresis system, which is a relatively

81

Discussion

high level. This level of abundance is likely to represent very few tissue

damage/leakage proteins (as shown by Figure 3.1).

Multiple biomarker approach:

The process of identifying a diagnostic test for bowel infarction may benefit from the

simultaneous investigation of several potential markers. A diagnostic test for bowel

ischaemia/infarction need not be based on the detection of a single protein, but may

involve a combination of protein(s), radiographic features and/or clinical signs. The

use of a panel of diagnostic markers has been proposed by a number of groups.

Anderson and Anderson recently stated that there is a:

“significant theoretical problem with the notion that there should be a single

protein in plasma whose levels change in response to one specific disease”.

Adam and colleagues (2002) state that ‘no single marker is likely to prove sufficiently

predictive’, while Seliger and Kellner (2002) suggest that ‘preferably small clusters of

proteins represent the ideal diagnostic markers”. Conversely, the number of

diagnostic variable must be kept as low as possible to keep costs to a minimum.

ROC PLOTS:

Receiver Operating Characteristic curves are very useful in assessing a number of

diagnostic tests on a common scale. ROC plots are also very useful when the cut-off

point has not been decided, or even if a cut-off point is known it can be omitted from

the analysis and thus does not bias the evaluation. In this project, the numbers of

patients were too low to obtain an accurate assessment of the value of the variables by

univariate or multivariate analysis. It has been recommended that multivariate

analysis only be undertaken when the study involves more than thirty subjects in each

group (Dr. M. Henry, personal communication). If the numbers of patients recruited

in the future were increased, then multivariate analysis is likely to be valuable in

identifying the most useful diagnostic marker, or combination of diagnostic markers.

Higher numbers of patients in each group (bowel infarction and control) would also

begin to overcome problems associated with biological variation and protein

fluctuation.

82


Chapter 4: Phospholipase A2 in Mesenteric Ischaemia and Infarction: Phospholipase A2: The phospholipases are a diverse range of enzymes that have been grouped by their ability to

hydrolyse phospholipids. The phospholipase family is divided into the subgroups A1, A2, B,

C and D according to the site of action within the generalised phospholipid molecule (see

Figure 4.1). The subgroup A2 hydrolyses phospholipids at the middle or sn-2 fatty acid to

produce a free fatty acid and the corresponding lysophospholipid molecule. Phospholipase

A2 has been studied extensively because of the physiological importance of the hydrolysis

products and the role of these products in disease (see Figure 4.2). For example: the most

common fatty acid occurring in the sn-2 position is arachidonic acid, the fatty acid central to

the eicosanoid mediator pathway in inflammation. The lysophospholipids liberated by the

action of phospholipase A2 have also been implicated in various pathophysiological

processes.

CH

CH2

CH2

O C

O

R'

O C R'

O

OP

O

O-

XO

A1

B

A2

CD

Figure 4.1: Site of action of the phospholipase enzymes on a generalised phospholipid

molecule. X can represent choline, ethanolamine, inositol, serine, glycerol,

diacylglycerol etc (Mansbach, 1990).

The physiological roles that phospholipase A2 plays are diverse, ranging from digestion of

dietary phospholipids, cell membrane remodelling, host defence (Hanasaki & Arita, 1999),

cell signalling, biosynthesis of eicosanoids (Lambeau & Lazdunski, 1999) to envenomation

83


(Gelb et al, 2000). Phospholipase A2 exists as a family of isoforms, the recognised number

of which has increased rapidly over the last five to ten years. The isoforms can be grouped

into secretory isoforms, cytosolic isoforms, calcium independent isoforms and platelet

activating factor (PAF) acetylhydrolases (Hanasaki & Arita, 2002).

Phospholipase A2 proteins utilising a catalytic Histidine:

Typical phospholipase A2 isoforms (Group I, II, III, V, IX, X and XI) have phospholipid

hydrolysing capability due to a catalytic histidine which is conserved along with an aspartate

residue in the active site. All secretory phospholipase A2 (groups I, II, II, V, X, XII) isoforms

show a catalytic histidine and share a number of common features including:

- a conserved calcium binding loop,

- low molecular weight (13-18kDa1),

- low milli-molar calcium dependency (Hanasaki & Arita, 2002) and

- high sensitivity to reduction by chemicals such as DTT (Clark, Milona & Knopf,

1990).

The number of secretory phospholipase A2 isoforms recognised in mammals has increased

from two isoforms in 1990 (IB and IIA) to ten (Hanasaki & Arita, 1999, Scott, Graham &

Bryant, 2003). The mammalian secretory phospholipase A2 isoforms that have identified

(IB, IIA, IIC, IID, IIE, IIF, III, V, X and XII) all occur in humans except for IIC which

appears to be a pseudogene. The secretory phospholipases A2 have low homology and

Singer et al (2002) state ‘different sPLA2 paralogs are not closely related isoforms since

amino acid identity between any two is in the range <15% to 50%’.

Phospholipase A2 proteins utilising a catalytic Serine:

As discussed in the previous section, the catalytic activity of secretory phospholipase A2

isoforms is dependent on a conserved histidine that occurs in a Histidine-Aspartate dyad.

Several phospholipase A2 variants exist that do not exhibit a catalytic histidine, and the

phospholipase A2 activity of these ‘orphans’ is dependent on a catalytic serine (Six & Dennis,

2000). Platelet activating factor (PAF) acetylhydrolase is one of the serine phospholipase A2

variants. PAF acetylhydrolase (group VIIA PLA2) shows the lipase consensus motif Gly-X-

Ser-X-Gly and the classic hydrolase triad of Ser273Asp296His351 (Six & Dennis, 2000). Group

1 Human type III is an exception, the active protein forms a section of a larger protein of 55kDa (Scott, Graham & Bryant, 2003).

84


VIIIA and B are subunits of the PAF acetylhydrolase Ib protein and both exhibit the ‘pseudo

lipase’ consensus motif of Gly-X-Ser-X-Val (Six & Dennis, 2000). Group VIA

phospholipase A2 (iPLA2) also contains the hydrolase (Gly-X-Ser-X-Gly) motif (Six &

Dennis, 2000). Groups IVA (cPLA2), IVB and IVC phospholipases A2 also exhibit a serine

aspartate active site (Six & Dennis, 2000). 1-cysteine peroxiredoxin, (or bifunctional lung

enzyme) was another unusual protein reported to exhibit phospholipase A2 activity (Chen et

al, 2000, Kim et al, 1997). Six and Dennis (2000) argue that this protein should not be

classified as a phospholipase A2 because mutation of the serine (and cysteine) had no effect

on phospholipase A2 activity (Kang, Baines & Rhee, 1998). Six and Dennis (2000) proposed

that the classification of a phospholipase A2 may only occur when a protein can be shown to

catalyse the hydrolysis of the sn-2 bond of a phospholipid substrate, and when the complete

amino acid sequence is known.

The complexity of Phospholipase A2 Research:

The enzyme characteristics and tissue distribution of phospholipase A2 isoforms show

considerable overlap, making the study of specific isoforms at the protein level difficult. The

functions of many of the recently identified isoforms are yet to be elucidated and detailed

investigations of tissue and cellular distribution and enzyme characteristics are yet to be

performed. The source of circulating secretory phospholipase A2 activity is unknown

(Nevalainen, Gronroos & Kallajoki, 1995, Nevalainen, Haapamaki & Gronroos, 2000).

The precise roles of phospholipase A2 has not been clearly defined in physiology, membrane

maintenance or cellular signalling. Even less is understood about the roles and interactions

of the multitude of secretory PLA2 isoforms. The complexity of the functions of the

secretory PLA2 isoforms is likely to increase further during pathophysiological situations.

Organs (Scott, Graham & Bryant, 2003), cells and tissues (Yang et al, 1999) are all known to

contain multiple forms of phospholipase A2 proteins.

Phospholipase A2 in disease:

Since the physiological roles of recently described PLA2 isoforms is unclear, the role in

disease states is essentially unknown.

The study of phospholipase A2 in relation to disease has focussed on several key areas:

- the role of phospholipase A2 at the site of inflammation or tissue damage,

- the use of phospholipase A2 as a diagnostic or prognostic marker,

85


- the role of phospholipase A2 in the generation of eicosanoids,

- the possibility of inhibition of phospholipase A2 to treat inflammatory conditions.

Figure 4.2: The multitude of products of phospholipase A2 Hydrolysis and their

Cellular Response (Bomalaski & Clark, 1993).

PLA2 involved in the pathogenesis of disease:

Synovial fluid from patients with rheumatoid arthritis shows elevated levels of PLA2 (Hara et

al, 1989, Lai & Wada, 1988). Type II PLA2 is proposed to be involved in the pathogenesis

of intestinal inflammation in Crohn’s disease and ulcerative colitis (Minami & Tojo, 1997).

PLA2 levels correlate to mortality rates in patients with established MODS (Uhl et al, 1990,

Uhl et al, 1995). Recently it has been proposed that PLA2-II may “contribute to the

induction of organ dysfunction but less of a role in sustaining organ system failure” (Abraham

et al, 2003). Elevated levels of sPLA2 have been shown to be a significant risk factor for

presence of coronary artery disease (Kugiyama et al, 1999). PLA2 is thought to be a

circulating mediator of typhoid fever (Keuter et al, 1995).

Elevated levels of PLA2 in serum of patients:

Elevated levels of phospholipase A2 type II have been reported in the plasma of patients with

septic shock (Green et al, 1991, Nevalainen et al, 1993), Crohn’s disease (Minami et al,

1992), systemic lupus erythrematosus (Pruzanski et al, 1994) and ARDS (Arbibe, Vial &

86


Touqui, 1997, Romashin et al, 1992). Serum concentrations of PLA2-II are considerably

elevated in patients with surgical intensive care patients (Nyman et al, 1996).

PLA2 as a marker:

PLA2 is regarded as a sensitive marker for inflammatory activity in rheumatoid arthritis

(Nevalainen & Gronroos, 1997). Serum PLA2 correlates with the activity of Crohn’s disease

and ulcerative colitis (Minami et al, 1992). PLA2 type I in serum is useful in the diagnosis of

acute pancreatitis but does not aid in predicting the severity of the condition (Nevalainen,

Gronroos & Kortesuo, 1993). Enhanced PLA2-IIA expression has been correlated to poor 5

year survival in patients with prostate cancer (Graff et al, 2001, Jiang et al, 2002). Elevated

levels of secretory phospholipase A2 have been found to predict coronary events in patients

with coronary artery disease, independent of other risk factors (Kugiyama et al, 1999).

PLA2 as a eicosanoid producing enzyme:

Synovial fluid from patients with arthritis shows elevated levels of arachidonic acid and

downstream factors (PAF, leukotrienes and prostaglandins) derived from the hydrolysis by

PLA2 (Bomalaski & Clark, 1993).

PLA2 as a therapeutic target:

The inhibition of PLA2 as a therapeutic strategy has been proposed for many years. PLA2

could be directly inhibited, inactivated by dephosphorylation or indirectly regulated by

inhibiting mRNA translation of agents such as TNF or interlukins that activate PLA2

expression (Waage & Bakke, 1988). There are many considerations for the use of

phospholipase A2 inhibition as a therapeutic strategy including:

- difficulty in targeting single isoforms,

- potential disruption to essential phospholipase A2 functions (membrane remodelling,

digestion),

- the presence of multiple PLA2 receptors may mean that catalytically inactive

enzymes may still be able to act via signalling pathways (Scott, Graham & Bryant,

2003).

Recent results of the clinical trial of a type IIA selective inhibitor have shown no survival

benefit but a statistically significant dose dependent improvement in patients with sepsis

induced organ failure in the first 18 hours (Abraham et al, 2003).

Assay of Phospholipase A2.

Phospholipase A2 can be studied using biochemical enzyme activity assays based on the

hydrolysis of phospholipids. Biochemical assays are generally of limited value in the study

87


of phospholipase A2 isoenzymes in crude samples because of the difficulty in differentiating

isoforms and the low abundance of PLA2 in mammalian tissue. Conversely, biochemical

assays for PLA2 can be very sensitive and robust. PLA2 assays can be used for monitoring

protein purification progress and examining enzyme characteristics. Results obtained from

different types of PLA2 assays (vesicle assays, mixed micelle assays, synthetic or natural

membrane assays) can be difficult to compare on absolute terms, but one study showed that,

in general, different assays showed similar relative results (Ghomashchi et al, 1999).

Species differences:

Although the use of an animal model would seem advantageous to study such a complex

enzyme, there are significant limitations associated with phospholipase A2. The species

differences that exist encompass distinct enzyme characteristics, function and tissue

distribution (Dr. K. Scott, personal communication). Some examples of the species

differences in phospholipase A2 are noted below:

- In rats and humans the predominant secretory isoform is type IIA (Murakami et al,

1998), whereas in mice the major secretory isoform is type V (Chen et al, 1994 (a),

Sawada et al, 1999),

- In some strains of mice, type IIA is found only in the intestine (Sawada et al, 1999)

- Humans have a soluble form of PLA2 receptor which is not found in other animals

(Tischfield, 1997),

- Patterns of PLA2 receptors are distinctly different in rabbits and humans (Tischfield,

1997),

- Expression profiles show that genes are regulated differently in humans and rodents

(Scott, Graham & Bryant, 2003).

General PLA2 Assay difficulties:

The complexity of the kinetics of phospholipase A2 hydrolysis has ensured that this enzyme is

a popular kinetics model. Part of the complexity of the PLA2 kinetics arises from the ability

of PLA2 to act at the interface of the hydrophobic membrane and the hydrophilic extracellular

fluid. Phospholipase A2 hydrolysis is dependent on traditional enzyme activity variables (for

example, temperature, pH, cofactors, ionic strength, aqueous environment). In addition PLA2

activity is dependent on :

- substrate presentation (vesicles, micelles, bilayers),

- size of substrate aggregates (small or large vesicles),

88


- presence of detergents or other enhancing agents,

- whether substrate contains one or more phospholipid species.

These variables can make interpretation of results and the comparison of results from different

assays very difficult. For example:

- type V shows a preference for choline phospholipids over ethanolamine

phospholipids when deoxycholate is present and the opposite preference when the

detergent is omitted (Chen et al, 1994),

- the production of lysophospholipid can inhibit the hydrolysis reaction (Lawrence &

Moores, 1975).

Many groups have tried to differentiate PLA2 isoforms using hydrolysis assays with different

substrates, but it is not possible to draw conclusions from the results obtained due to

significant overlap of results. Studies of PLA2 inhibitors have shown that in general, PLA2

assays show similar patterns, even though absolute results cannot be compared (Ghomaschi et

al, 1999).

Phospholipase A2 in the gastrointestinal tract:

The study of phospholipases in the human gastrointestinal tract is complicated by difficulty

obtaining tissue in suitably good condition (as the gut undergoes spontaneous autolysis after

death). In animal models it is difficult to obtain gut tissue without sacrificing the animal.

Access to human tissue is very difficult but is essential in investigations of human gut

pathologies due to significant species differences.

Biochemical PLA2 assays;

Early investigations found clear patterns in the distribution of phospholipases throughout the

gastrointestinal tract, through cross sections of small intestinal mucosa and were able to

discount a number of possible sources of the phospholipase activity. Mansbach and

colleagues (1982) showed that the phospholipase A2 in rat gut did not originate from

pancreatic, salivary or gastric secretions by diverting relevant ducts away from the small

intestine. This group also showed that the phospholipase A2 activity was not bacterial by

repeating their experiments in germ free rat lines. They also found that the phospholipase A2

activity in the small intestinal mucosa was ten fold greater in the crypts than at the villi tips.

Secretory PLA2 was also found to increase with increasing distance from the ligament of

Treitz to the ileo-cecal valve in rats (Sonnino & Pigatt, 1996).

89


Type IIA Phospholipase A2 in Human Gut:

Human Paneth cells (particularly the secretory apparatus) have been shown to have strong

immunoreactivity to type IIA PLA2 (Kiyohara et al, 1992). Morita and colleagues (1999)

found that human ileal mucosa reacted with an antisynovial PLA2 antibody (and was therefore

likely to be type IIA). The Paneth cells were found to be the only cell type in the gut found

to secrete type II PLA2 mRNA (Nevalainen, Gronroos & Kallajoki, 1995). Differential

display polymerase chain reaction (DD-PCR) was used to examine genes that were

differentially expressed between the small and large intestine in mice (Keshav et al, 1997).

A secretory phospholipase A2 isoform was found using this technique (Keshav et al, 1997)

and was found to have significant homology to mouse enhancing factor and to rat and human

secretory PLA2 type II proteins.

Other phospholipase A2 isoforms in Human Gut:

In the past five years attempts have been made to study the phospholipase A2 isoenzymes in a

clear and systematic manner. With the identification of each new secretory isoform came the

probing of panels of human tissue for mRNA transcripts. In normal human gut the

transcripts corresponding to four of the isoforms have been detected (reviewed by Scott,

Graham & Bryant, 2003): namely IIA (Cupillard et al, 1997), IID (Ishizaki et al, 1999), V

(Cupillard et al, 1997) and XII-1 (Gelb et al, 2000). These results were obtained using either

RT-PCR or Northern blotting. The expression of type X in normal human gut has been

shown by one group (Suzuki et al, 2000) but not by another group (Cupillard et al, 1997).

These results are an indication of the complexity of phospholipase A2 in human gut but even

then may be an oversimplification because:

- northern hybridisation panels can be composed of a single sample (Scott, Graham & Bryant,

2003) which may not be representative of the typical pattern,

- they are an indication of the normal situation, likely to be complicated further in disease

states,

- as yet unidentified isoforms may exist (Scott, Graham and Bryant, 2003),

- isoforms expressed at very low levels may not have been detected.

In conclusion, Morita and colleagues (1999) state that there is a:

“complex involvement of several phospholipases A2 in normal gastrointestinal tissues”.

90


A summary of the research conducted on the topic of phospholipase A2 involvement in gut

pathology is presented in the following tables:

91


Table 4.1. Phospholipase A2 in Intestinal Disease Research.

Group Year Disease Studygroup

Conclusions

Otamiri et al 1987 Ischaemia/reperfusion injury

Rat Increased levels of lysophosphatidylcholine in ischaemic gut mucosa compared with non-ischaemic control. Increased lysophosphatidylcholine or PLA2 may mediate mucosal injury.

Otamiri, Lindahl & Tagesson

1988 Ischaemia/reperfusion injury

Rat Mucosal injury and increased mucosal permeability seen in I/R is prevented by inhibition of PLA2.

Otamiri & Tagesson

1989 Ischemia/reperfusion injury

Rat PLA2 increased in mucosa during I/R, and was significantly different to control mucosa.

Koike et al 1992 Ischemia/reperfusion injury

Rat Gut PLA2 was increased. Inhibition of PLA2 reduced distant organ injury by preventing PMN influx, PMN superoxide production and lung leak.

Minami et al 1993 Crohn’s disease Human Levels of PLA2 type II (mRNA and protein) were higher in ileal than cecal mucosa. Increased serum levels of PLA2 seen in Crohn’s patients may be due to leakage of protein from the gut mucosa.

Minami et al 1994 Crohn’s diseaseand ulcerative colitis

Human Actively inflamed mucosa of Crohn’s disease showed higher levels of PLA2 activity and type II immunoreactivity. Severely inflamed mucosa of ulcerative colitis patients showed higher levels of PLA2 and type II immunoreactivity than control mucosa.


Rat PLA2 activated by I/R. Remote organ damage caused by I/R could be reduced with a PLA2 inhibitor

92


Lilja et al 1995 Crohn’s disease Human Increased levels of PLA2 may contribute to

inflammation in Crohn’s disease and in recurrent Crohn’s after resection. Increased levels of mRNA of type I, II and cytosolic PLA2 observed, but type II was predominant.

Lilja et al 1995 Crohn’s disease Human Increased levels of PLA2 may contribute to inflammation in Crohn’s disease and in recurrent Crohn’s after resection. Increased levels of mRNA of type I, II and cytosolic PLA2 observed, but type II was predominant.

Nevalainen, Gronroos & Kallajoki

1995 Crohn’s diseaseand ulcerative colitis

Human Paneth cells were the only cells that showed mRNA for type II PLA2. Type II immunoreactivity in blood vessel walls likely to reflect transport of the protein rather than synthesis.

Qu et al 1996 LPS infusion Rat PLA2 release from Paneth cells was able to be triggered by LPS.

Sonnino et al 1997 Graft preservation Rat Calcium dependent, secretory PLA2 accumulated rapidly during preservation. PLA2 likely to be proinflammatory and a priming agent for intestinal tissue damage.

Arcuni et al 1999 Graft preservation Rat PLA2 released in response to other proteins, perhaps cytokines, but not due to cell death. PLA2 inhibition improves the outcome of preservation.

Haapamaki et al 1999 Crohn’s disease

Human Columnar epithelial cells in Crohn’s mucosa responsible for the synthesis of PLA2-II in colon and small intestine, but not in control mucosa.

Morita et al 1999 Inflammatorybowel disease

Human Isoform hydrolysing PE more closely related to inflammation than the PC hydrolysing PLA2 which was found irrespective of inflammation.

Table 4.1 continued. Phospholipase A2 in Intestinal Disease Research.

93

Chapter 4 Introduction Table 4.1 continued. Phospholipase A2 in Intestinal Disease Research.


Rat PLA2 activity in portal circulation was 10 fold higher than systemic circulation, suggesting that the serum PLA2 activity arises from ischaemic gut.

Weifeng et al 2001 Ischaemia/reperfusion injury

Rat Inhibition of PLA2 early in the ischaemia/reperfusion process decreased bacterial translocation, endotoxin release, pro-inflammatory mediator and cytokine levels. This reduced the injury to remote organs.

Thomas, Karnik & Balasubramanian

2002 Surgicalmanipulation of intestine.

Rat Inhibitors of PLA2 were protective against lung damage which may be due to reduced neutrophil infiltration, reduced oxidative stress or reduced permeability.

94


95

Results

Chapter 4: Phospholipase A2 activity in ischaemic/infarcted human bowel:

Previous research had detected phospholipase A2 in bowel and had identified the

major isoform as type IIA (Kiyohara et al, 1992, Morita et al, 1999, Nevalainen,

Gronroos & Kallajoki, 1995). There is some evidence that suggests that a complex

pattern of phospholipase proteins in the bowel (Morita et al, 1999, Scott, Graham &

Bryant, 2003). As a preliminary control experiment, immunodepletion was

conducted to remove the phospholipase A2 isoform type IIA and its closely related

isoform type V from a recombinant type IIA standard (Figure 4.3). Immunodepletion

of phospholipase A2 types IIA and V from human bowel showed that types IIA/V

were present in infarcted human bowel content (Figure 4.4), but did not account for

all of the phospholipase A2 activity (Figure 4.5).

The phospholipase A2 activity was measured in several patients with matched tissue

homogenate and bowel lumen content (Figure 4.6). These results are expressed as

specific activity of phospholipase A2 due to the large variation of total protein

between these samples. Phospholipase A2 activity was significantly increased in the

lumen content (Fisher’s exact test, p=0.048), compared with the matched tissue

homogenate. On one occasion ischaemic (tissue damage rating 2-3) and infarcted

tissue (tissue damage rating 3) were obtained from the one patient, and the infarcted

tissue showed significantly higher phospholipase A2 activity (p<0.0005), (Figure 4.7).

The tissue damage rating and the PLA2 activity were compared in several samples of

bowel tissue homogenate (Figure 4.8). The next section details the purification of the

protein responsible for this phospholipase A2 activity (Chapter 5).

96

97

Type IIA standard

0102030405060

Non depleted 4A1 10B2

PLA 2

act

ivity

, %

hyd

roly

sis

Figure 4.3 Immunodepletion of PLA2 from recombinant PLA2-IIA

protein with antibodies 4A1 (type IIA specific) and 10B2 (type IIA+V

specific). Results are expressed as the mean +/- standard error of

triplicate assays. The recombinant protein was a generous gift from

Dr. Kieran Scott.

98

Protein size, (kDa)

1 2 3 4

22

98148

Albumin

36

16 sPLA2

Figure 4.4 Silver stained gel of the antibody-bead-PLA2 complex after

immunodepletion from bowel samples. SDS-PAGE gel: 4-20% Tris Glycine

gel, Novex.

Lanes:

1- See Blue Plus2 markers.

2- immunodepletion with 4A1 antibody.

3-immunodepletion with 10B2 antibody.

4-recombinant sPLA2-IIA standard.

99

Mar

kers

sPLA

2

type

IIA

imm

un

odep

lete

d

Rep

eate

d de

plet

ion

53

34

23

17

0

5

10

15

20

25

Before depletion After depletion

PLA 2

act

ivit

y,

% h

ydro

lysi

s

Figure 4.5 Western blotting of immunodepleted bowel autolysis

sample, and phospholipase A2 activity before and after

immunodepletion of phospholipase A2 isoforms IIA and V.

The arrow indicates the immunoreactive protein of interest.

100

0

500

1000

1500

A B C D EPatient

Spec

ific

activ

ity (%

hy

drol

ysis/

min

/mg

prot

ein)

T issue homogenate Bowel lumen content

Figure 4.6 PLA2 activity in matched homogenised bowel tissue and

lumen content. These results are significantly different, Fisher’s

exact test, two tailed p=0.048. Fisher’s exact test was chosen due to

the small number of samples. These results are presented as the

mean +/- standard error of triplicate assays.

101

0

12

34

56

7

Infarcted tissue Ischaemic tissue

PLA 2

act

ivit

y (%

hyd

roly

sis)

Figure 4.7 Two homogenised bowel tissue samples from same

patient. The results are presented here as the mean +/- standard error

of triplicate assays. The tissue damage rating for these tissues was 3

for the infarcted tissue and 2-3 for the ischaemic tissue, using the

scale published by Sun et al, 1997.

102

020406080

100

a b c d e f g

PLA

2 ac

tivity

, %

hy

drol

ysis

1

2

3

Tiss

ue d

amag

e ra

ting

phospholipase activityTissue rating

Figure 4.8 Tissue damage rating of Sun et al, 1997 and corresponding

PLA2 activity of various homogenised bowel tissue samples. These

results are presented as the mean +/- standard error of at least triplicate

assays.

Discussion


Type IIA PLA2 had been described in the bowel by many other groups, using a

variety of techniques (immunohistochemistry, in situ hybridisation, differential

display PCR). Type V PLA2 is another secretory PLA2 isoform and shares very

similar enzyme characteristics to type IIA. Type V PLA2 has been detected at low

levels in normal human small bowel by Northern blotting (Cupillard et al, 1997). In

this study, we chose to investigate phospholipase A2 at the protein level, as we wanted

to focus on the activity of the enzyme and were concerned about mRNA degradation

that may occur as a result of bowel injury. Immunoprecipitation of secretory

phospholipase A2 types IIA and V from bowel samples was conducted. The

immunoprecipitation experiment was verified by duplicating the same set of

conditions using recombinant human type IIA PLA2 as the enzyme source. The

antibodies 4A1 (IIA specific) and 10B2 (cross reactive with types IIA and V) were

used. As expected, phospholipase A2 types IIA/V were found to be present in human

bowel samples as shown by silver stained gels of the proteins attracted to the resin-

immobilised antibodies. Also as expected, type V PLA2 is likely to be a minor PLA2

isoform in human bowel, as the difference between the enzyme activity remaining

after depletion with either antibody 4A1 or 10B2 was very similar. The more

surprising but interesting result was that removal of types IIA/V from the bowel

samples failed to abolish all of the phospholipase enzyme activity. An average of

approximately one third of the phospholipase activity remained after removal of types

IIA/V phospholipase A2. This result was not simply due to an inadequate amount of

antibody, as when an additional ten fold excess of immobilised antibody was used, the

enzyme activity remained. The immunoprecipitated samples were analysed by

Western blotting and the remaining immunoreactive band appeared to be of low

molecular weight. The gels for Western blotting were run under non-reducing

conditions, as the epitope recognised by the antibody 4A1 is reported to be lost upon

reduction (Rice et al, 1998). A protein purification process was then undertaken with

the aim of identifying the protein responsible for this unexpected phospholipase

activity (Chapter 5).

We were fortunate to have access to human bowel excised during surgery to examine

the phospholipase activity. Access to human bowel tissue in suitably good condition

can be difficult due to post mortem autolysis. We had access to infarcted small

103

Discussion

bowel tissue and lumen contents, and also to normal small bowel from patients

undergoing right hemicolectomy. The normal bowel samples were always from the

terminal ileum. The samples of infarcted bowel were obtained from a number of

regions along the length of the resected small bowel. As phospholipase activity is

reported to increase along the length of the small intestine (a study in rats, Sonnino et

al, 1996) we did not try to compare all of the absolute results between samples.

Samples of tissue and lumen content were frozen as soon after surgery as practical.

Phospholipase activity has been reported to remain constant with freezing and

extended storage. For example, the phospholipase A2 from peritoneal exudate was

reported to be stable for four months at 4oC (Chang et al, 1987). Our investigations

also verified that the phospholipase activity was stable with storage at –80oC for

several months and also with repeated freeze-thaw cycles (results not shown).

The study of phospholipase A2 in the bowel has generally focussed on normal rodent

bowel tissue. Human bowel phospholipase A2 research has typically focussed on

normal tissue, at the RNA level, due to the difficulties obtaining sufficient diseased

tissue for protein studies (for example, biopsy samples), and the complexity of this

protein family. Few publications, particularly in these past decade, have been

reported that combine:

- human bowel tissue

- phospholipase at the protein level

- diseased tissue

- bowel ischaemia/reperfusion injury.

We have chosen to investigate phospholipase enzyme activity, at the protein level in

normal and ischaemic/infarcted human bowel with the aim of better understanding the

role of this disease associated protein.

On a number of occasions, matched infarcted bowel tissue and lumen content were

obtained from the same patient. The phospholipase activity in the lumen was

significantly higher than its corresponding tissue homogenate. This may reflect the

release of the enzyme into the bowel lumen from a storage site in the tissue, and may

provide a route for entry into general circulation. This idea is supported by evidence

that intestinal permeability increases in response to ischaemia and infarction, and

results in the release of gut derived factors into circulation (Kong et al, 1998).

104

Discussion

Alternatively, the enzyme may be actively synthesised and released into the lumen in

response to the ischaemic damage and may subsequently enter circulation. The

alternate possibility is that the presence of increased phospholipase activity in the

lumen simply reflects cell death and therefore leakage of the protein. A study

investigating the release of phospholipase into bowel preservation fluid concluded

that the release of the enzyme was not due to leakage of protein from dying cells

(Sonnino et al, 1997). This group provided data showing the absence of release of

lactate dehydrogenase to support their argument (Sonnino et al, 1997). An approach

like this would be valuable to assess whether the increased phospholipase activity is

due to cell death, or active release, and requires further elucidation.

The increased phospholipase activity observed as a consequence of tissue injury,

infection or inflammation may be a beneficial response to the patient in a number of

ways. Phospholipase A2 is known to have anti-bacterial properties (Nevalainen,

Haapamaki & Gronroos, 2000). Phospholipases are likely to aid in the removal or

rebuilding of peroxidised or damaged phospholipids (Vadas et al, 1993).

Conversely, phospholipase hydrolysis products have many tissue and cell damaging

effects, and inflammatory properties, as discussed in the introduction to this chapter

(Figure 4.2). For example, lysophospholipids are cytotoxic (Schoenberg & Beger,

1993). The roles of phospholipase A2 and its corresponding hydrolysis products in

inflammation, infection and tissue injury require detailed investigation. Conceivably

the important concept is the balance between the beneficial roles and damaging

effects of phospholipase during disease development.

On one occasion two bowel tissue samples were obtained from one patient, each

showing a different level of ischaemic damage. It is not unusual for the bowel to

show areas of severe damage and areas of mild damage in close proximity because of

the arrangement of the blood supply to the bowel. The infarcted tissue was given a

tissue damage rating of 3, according to the scale proposed by Sun and colleagues

(1997). The infarcted tissue showed significantly higher phospholipase activity than

the adjacent ischaemic tissue (tissue damage rating of 2-3). This is an interesting

finding (even though this is sample size of only one) as there may be a positive

relationship between the severity of tissue damage (from normal to severe infarction)

and the phospholipase activity of the tissue homogenate. Many biomarkers of bowel

105

Discussion

infarction have been found to appear too late to be clinically useful, or decrease

before the condition has been rectified.

106


Chapter 5: Protein purification.

Purification of non-pancreatic Phospholipase A2 from mammalian tissue:

The primary structure of pancreatic phospholipase A2 was reported by Puijk, Verheij

& de Haas (1977). Venom and pancreatic tissue contain high amounts of

phospholipase A2 (typically 1-10% of total protein) (Verheij, Slotboom & de Haas,

1981). In 1982, a phospholipase A2 of similar size to the pancreatic isoform was

isolated from the ileum of pigs, and was found to have distinctly different

characteristics (N-terminal amino acid sequence, isoelectric point, high proportion of

basic to acidic amino acid residues)(Verger et al, 1982). Extracellular phospholipase

A2 found at the sites of inflammation sparked great interest. A soluble phospholipase

A2 from the synovial fluid of patients with rheumatoid arthritis was purified and its

characteristics were found to be distinct from the pancreatic isoform (Stefanski et al,

1986). Phospholipase A2 is thought to be central in the eicosanoid biosynthetic

pathway due to the release of arachidonic acid. The purification of secretory

phospholipase A2 from rheumatoid arthritis synovial fluid (to obtain sequence data or

amino acid composition) was achieved almost simultaneously by three groups (Hara

et al, 1989, Kramer et al, 1989, Seilhamer et al, 1989). The latter two groups also

reported the identification of the N-terminal amino acid sequence:

Group: Source of PLA2: Manuscript

submitted:

Amino acid sequence

published:

Kramer et al,

1989

Platelets and

Rheumatoid

arthritis synovial

fluid

December 20,

1988

NLVNFHRMIKLTTGKEAAL

Hara et al, 1989 Rheumatoid

arthritis synovial

fluid

September 28,

1988

NO SEQUENCE DATA.

Seilhamer et al,

1989

Rheumatoid

arthritis synovial

fluid

December 27,

1988.

?LVNFHRMIKLTTGKEAAL

107


Type II phospholipase A2 has been purified to homogeneity and at least partial amino

acid sequence determined from:

Normal animal tissues/fluids:

Porcine ileum (Verger et al, 1982).

Stimulated rat platelets (Hayakawa et al, 1987).

Rat liver (Aarsman et al, 1989).

Rat spleen (Ono et al, 1988).

Normal human tissues/fluids:

Platelet concentrate (Kramer et al, 1989).

Spleen (Kanda et al, 1989).

Ileal mucosa (Minami et al, 1993).

Diseased animal tissue/fluids:

In models of peritonitis:

-rabbit ascitic fluid treated with glycogen IP (Forst et al, 1986).

-rat peritoneal fluid treated with casein IP (Chang et al, 1986, Chang et al,

1987).

Diseased human tissue/fluids:

Synovial fluid from rheumatoid arthritis patients (Lai & Wada, 1988).

Synovial fluid from arthritis patients (Kramer et al, 1989).

The purification schemes used in the research cited above generally involved the use

of a series of different chromatography columns (ion exchange, hydrophobic

interaction, gel filtration, affinity, RP-HPLC). The purification processes have

monitored phospholipase A2 using activity assays of column fractions at all stages of

the purifications. These purification schemes are laborious, time consuming and

expensive.

The purification of phospholipase A2 from animal tissues or fluids can be difficult due

to the low abundance of the protein in samples and unusual characteristics such as

tendency to adhere to glass and some column packings. The number of isoforms,

difficulty in differentiating isoforms and lack of many antibodies further compound

these problems. Phospholipase A2 enzymes have some characteristics that are

advantageous during purification, such as heat and acid stability. Phospholipase A2

108


enzymes often have characteristics that are quite different to the majority of proteins

in a sample, such as isoelectric point. The majority of proteins have acidic to neutral

isoelectric points (pI 4-7) whereas many phospholipase A2 isoforms tend to have basic

isoelectric points.

Heparin Sepharose affinity chromatography:

Heparin sepharose chromatography can be used to purify several proteins types based

on affinity binding and ion exchange interactions. Heparin affinity is exhibited by

many coagulation factors, enzymes, growth factors and receptor proteins (reviewed by

Farooqui, Yang & Horrocks, 1994). Affinity chromatography is advantageous due to

its ability to quickly bind proteins of interest and elute the majority of contaminating

proteins. The specifically bound proteins can be washed extensively and eluted in

suitable buffers. The affinity of a protein to a ligand such as heparin can be quite

specific and slight changes in the elution conditions can allow very narrow sample

fractions, and therefore high purity fractions.

Heparin is a group of naturally occurring glycosaminoglycans. The heparin members

vary greatly in size (5 to 40 kDa) but share characteristics such as being highly

charged, highly sulphated, and composed of disaccharide repeating units of

glucuronic acid and D-glucosamine (Dua & Cho, 1994). In vivo, heparin occurs only

in mast cells (Farooqui, Yang & Horrocks, 1994).

The attraction of proteins to immobilised heparin in a chromatography situation is

based on either specific interactions or electrostatic interactions (Farooqui, Yang &

Horrocks, 1994). It has been proposed that the specific binding of a protein to heparin

is due to a cluster of cationic residues on the outer surface of an amphiphilic α helix

(Dua & Cho, 1994). Heparin can be immobilised on resin such as sepharose or

agarose, and packed in columns for affinity purification of proteins. The proteins

with an affinity for the heparin are retained on the column and the rest of the material

is washed through. The proteins with heparin affinity can then be eluted with an

excess of heparin or changing the elution buffer composition with salts or denaturants

(Farooqui, Yang & Horrocks, 1994). There are some considerations for the use of

heparin in protein purification, firstly the affinity of a particular protein to the heparin

109


may be too high, meaning that protein is lost with each chromatographic run.

Secondly, heparin is known to inhibit some classes of protein, though this is generally

reversible. Heparin (at levels used for anticoagulation) reversibly inhibits

phospholipase A2 activity. For this reason, serum is preferable in assays when

undiluted samples are used (such as antibacterial plate assays). In highly sensitive

assays where diluted samples are used, both plasma and serum are suitable

(Weinrauch et al, 1998). Affinity columns such as heparin columns can become

fouled with protein build-up, but columns can be cleaned with a variety of agents to

prevent this. Eluted fractions can contain high concentrations of salts and may

require dialysis before further purification steps.

Heparin sepharose has been used in the purification of a variety of lipases, kinases

and phospholipases (reviewed by Farooqui, Yang & Horrocks, 1994). In the case

of phospholipase A2, the enzyme can be eluted from the column with increasing

concentrations of salts (KCl, NaCl) and thus retain native confirmation and enzyme

activity. Several of the secretory isoforms of phospholipase A2 have high affinity for

heparin (IIA, IID, IIE), while others have less affinity (IB, IIF, V), (see Table 5.2).

The strength of the affinity of phospholipase A2 isoforms generally follows the degree

of basicity, therefore in general isoforms with high pI values have higher heparin

affinity (Murakami et al, 2001), see Figure 5.7. Heparin binding of phospholipase A2

is due to a cluster of 5 to 7 C-terminal amino acid residues (Arni & Ward, 1996,

Murakami, Nakatani & Kudo, 1996), see Figure 5.8. The cluster of amino acids

responsible for heparin binding may only be evident on examination of the three

dimensional structure (Koduri et al, 1998). The heparin binding site is also a

substrate binding region, but is independent of active site residues (Dua & Cho,

1994).

Purification schemes

Several phospholipase A2 purification schemes have employed a heparin sepharose

column, usually preceeded by a crude sample clean-up step and followed by one or

more RP-HPLC columns.

110


Table 5.1 Purification of PLA2 by Heparin Affinity Chromatography:

Author(s) Year Source of PLA2 Purification scheme

Hara et al 1989 Human rheumatoid

arthritis synovial fluid

Heparin sepharose.

Hydrophobic interaction.

RP-HPLC.

Diccianni et al 1991 Porcine pancreatic PLA2

(Sigma)

Heparin Affigel.

Kim, Kudo &

Inoue

1991 Rabbit platelet enriched

plasma, cytosolic PLA2

Heparin sepharose.

Ion exchange.

Hydrophobic interaction.

Ion exchange.

Gel filtration.

Murakami et al 1998 Culture supernatent

CHO/HEK stable

transfectants.

Heparin sepharose.

Shimbara et al 1999 Baculovirus derived

recombinant rat and

mouse PLA2.

Heparin Sepharose.

Anti-sPLA2-IIA

immunoaffinity column.

111


Table 5.2. Heparin affinity of Phospholipase A2 Isoforms: Phospholipase A2 isoform Heparin affinity

(Concentration of NaCl

required for elution)

Isolectric point (Human

isoform), calculated from

the Swiss Prot database)

IB No binding 7.95

IIA 0.8M 9.38

IID 40% recovery at 1M 8.75

IIE 1M 8.53

IIF No binding 4.51

III* Unknown 9.1

V 0.4M 8.72

X No binding 5.1

XII Unknown 6.3

The Rotofor ® preparative isoelectric focussing system:

The BioRad Rotofor® cell is a liquid phase preparative isoelectric focussing

apparatus characterised by the short length of time required for separation

(approximately 4 hours for an average analysis). The Rotofor® cell has been

successfully used in the study of proteome of cerebrospinal fluid, specific enzymes in

whole plasma, purification of haemoglobin variants A, C and F, and separation of

peanut lectin isoenzymes.

Liquid based isoelectric focussing such as the Rotofor® cell establish a pH gradient

with carrier ampholytes in a liquid medium containing a low concentration of

detergent. Urea and denaturants such as DTE can be added as required. Urea can be

added to aid in the prevention of precipitation. Glycerol can be added to maintain

solubility and stability of proteins. A constant current is then applied which causes

proteins and ampholytes to migrate toward the anode or cathode according to their net

charge until they reach their pI. The liquid fractions are collected simultaneously and

quickly to avoid loss to the pH gradient using vacuum. The fractions can then be

*Mammalian sPLA2 type III occurs as a central PLA2 like domain flanked by two protein domains of unknown function. The central sPLA2 like domain has no heparan sulphate proteoglycan binding (Scott, Graham & Bryant, 2003).

112


analysed for pH, total protein, a specific target protein or run on one or two

dimensional gels. The use of a liquid phase for the protein separation is

advantageous as there is no need for elution from a solid medium. Another

advantage of liquid based IEF is that toxic polymerisation agents are avoided and

ampholyes could be omitted in favour of creating the pH gradient by ‘seeding’ the

IEF fluid with a pure protein of interest. This may have potential in the

pharmaceutical industry for production of purified proteins (Righetti, Wenisch &

Faupel, 1989).

Advantages of Preparative IEF:

The advantages of preparative isoelectric focussing include the retention of biological

activity, a high degrees of purification in a single step and flexibility in terms of

sample type and size. Isoelectric focussing is capable of high resolution separations,

is able to concentrate the protein of interest and defines a physical parameter of the

protein. The use of the Rotofor ® as a prefractionation step before 2DE has been

reported to improve the mass spectrometry sequence coverage in comparison to 2DE

alone and aids in the identification of low abundance and post translationally modified

proteins (Westman-Brinkmalm & Davidsson, 2002).

Disadvantages of Preparative IEF:

The introduction of ampholytes can be disadvantageous. Ampholytes can be difficult

to remove (Bloomster & Watson, 1983) and can interfere with post separation

analysis. Ampholytes can be removed with dialysis, gel filtration, ultrafiltration, ion

exchange chromatography or protein precipitation, though removal may be

incomplete. It has been suggested that ampholytes may bind irreversibly with

proteins, but Baumann and Chrambach (1975) found that this was not the case.

Considerations for the use of Preparative IEF:

The pH gradient in a liquid phase isoelectric focussing apparatus is achieved with a

mixture of carrier ampholytes. Ampholytes are small, amphoteric molecules that

have a high buffering capacity centered at their pI. Because ampholytes behave in a

similar fashion to proteins or peptides, care must be taken to avoid interference in

protein analysis. For example, carrier ampholytes will interfere with total protein

113


assays and give inflated estimates of protein concentrations (Bloomster & Watson,

1983). Ampholytes are incompatible with the commonly used total protein assays

bicinchoninic acid (BCA) (Brown, Jarvis & Hyland, 1989), Bradford and Lowry

assays (Guttenberger, Neuhoff & Hampp, 1991). The precipitation of the protein or

immobilisation of the protein on a membrane can overcome the interference by

ampholytes.

Detergent is added to the isoelectric focussing buffer to solubilise the proteins and to

prevent aggregation. The detergent used must carry no net charge (non-ionic or

zwitterionic) to enable the detergent to remain in the system under high currents.

Detergents may interfere with some biological assays, for example total protein assays

or those based on lipid or membrane substrates. The interference of detergent in total

protein assays can be overcome by precipitating or immobilising the protein and

thoroughly washing the sample. The interference in other assays can be more

difficult to overcome.

Some proteins are inherently unstable and may not tolerate complicated, multiple step

purification procedures. The isoelectric focussing procedure can lead to protein loss

for the following reasons:

-proteins may be unstable at their pI and may precipitate

-proteins may not tolerate very dilute solutions

-the detergent, reductants etc may alter the biological activity or structure of the

protein

114

Results

Chapter 5: Protein purification:

The protein of interest was purified from bowel tissue rather than from blood because

the tissue showed higher specific activity for phospholipase A2, was available in

larger quantities and was more likely to express a tissue specific protein. The

separation techniques were chosen because they retained the native enzyme activity,

exploited an unusual protein characteristic and removed large amounts of

contaminating protein at each step. The purification progress was monitored by

analysis of all fractions for PLA2 activity with a 14carbon labelled phospholipid mixed

micelle assay.

Unfortunately, a number of preliminary purification processes, each with different

separation steps, resulted in the identification of haemoglobin as the protein of

interest. These purification schemes are summarised in Figures 5.1 to 5.3. To verify

that haemoglobin was not responsible for the phospholipase A2 activity, a variety of

concentrations were assayed for activity (Figure 5.4), and, as expected, none showed

PLA2 activity. The absence of PLA2 activity in haemoglobin was verified by

comparison to ultrapure water, the diluting buffer used (acetate) and an equivalent

concentration of albumin (Figure 5.5). To show that haemoglobin did not

synergistically enhance phospholipase A2 activity, various amounts of haemoglobin

were incubated with pancreatic phospholipase A2 and then assayed for enzyme

activity (Figure 5.6).

Heparin sepharose affinity chromatography was performed using a step gradient of

sodium chloride to elute bound proteins. The major peak of total protein was eluted

from the chromatography column with 0.05M sodium chloride (Figure 5.9 and 5.10).

The majority of phospholipase A2 enzyme activity required 0.5M sodium chloride for

elution (Figure 5.10). The fractions with the highest phospholipase A2 activity were

then pooled for the next purification step. A step gradient of the elution buffers was

used, and 20 column volumes were routinely employed due to the low contaminating

protein levels that resulted (Figure 5.11). Due to the high salt concentration of the

heparin sepharose column fractions, the pooled fractions were dialysed against

phosphate buffer to reduce the molarity to 10mM. The dialysis step did not reduce

the enzyme activity of the samples (Figure 5.12).

115

Results

Preparative isoelectric focussing was performed using the Rotofor® system. A

preliminary experiment was conducted to confirm that the isoelectric focussing buffer

(detergent, ampholytes, glycerol) did not interfere with the phospholipase A2 activity

assay (Figure 5.13). The Rotofor® experiments were run until the power output had

stabilised for an hour (Figure 5.14). The fractions obtained from Rotofor®

experiments were analysed for pH, phospholipase A2 activity (Figure 5.15) and were

run on reducing SDS-PAGE (Figure 5.16). Five consecutive Rotofor® experiments

were conducted to investigate the reproducibility of the system, which was found to

be very good (Figure 5.17). Gels of typical fractions show that the majority of the

contaminating protein focussed in the neutral to acidic range of isoelectric points

(Figure 5.18), while the phospholipase A2 activity focussed in the basic region (Figure

5.15). The active fractions from a representative Rotofor® experiment were pooled

and re-fractionated (Figure 5.18), which allowed a better estimation of the isoelectric

point of the protein of interest (~9.7). The re-fractionation experiment extended the

number of fractions with a pH of between 7.5 and 10 (Figure 5.19). The basic

fractions of the re-fractionation experiment were run on SDS-PAGE, but no protein

band was evident of the expected size (Figure 5.18), and for this reason, a scaled up

purification was undertaken to obtain enough protein for identification .

Active fractions from a Rotofor® run were pooled and run under non-reducing

conditions, to retain phospholipase A2 activity. Proteins were electro-eluted from the

gel and analysed for phospholipase A2 activity. Pre-stained protein molecular weight

markers were run at the same time and their mobilities were used to construct a

standard curve (Figure 5.20). Enzyme activity was present in fractions 8&9,

corresponding to a protein of between 16 and 20 kilodaltons. The pH of optimal

enzyme activity was 9 (Figure 5.21). An overview of the purification scheme is

presented in Figure 5.22, and this represents an approximately 9700 fold purification.

The scaled up purification system resulted in highly purified material for protein

identification. A typical silver stained gel of active Rotofor® fractions is shown in

Figure 5.23, and this indicates that the protein was >95% pure. The fractions

exhibiting phospholipase A2 activity matched well with the fractions exhibiting a

protein band at approximately 21 kDa. The active fractions were pooled and

116

Results

concentrated by Vivaspin centrifugal cartridges. These cartridges appeared to

concentrate the protein of interest more effectively than ethanol precipitation,

although the latter was superior for removing ampholytes (Figure 5.24). The active

fractions concentrated using the Vivaspin cartridges were run on two different gel

systems, (BisTris, Invitrogen and TrisGly, Gradipore), digested with trypsin and

analysed by mass spectrometry. Positive (various protein markers) and negative (gel

pieces from no-protein lanes) controls were also digested and analysed in the same

manner. The unknown protein from the BisTris system was identified as peptidyl

prolyl isomerase/cyclophilin B (Figure 5.26). The positive control from the BisTris

system was correctly identified as lysozyme (Figure 5.26). The unknown protein

from the TrisGly system was also identified as peptidyl prolyl isomerase/cyclophilin

B (Figure 5.27). The positive control from the BisTris system was correctly

identified as restriction endonuclease (Figure 5.27). The presence of cyclophilin B in

a variety of human samples was then investigated (Chapter 6).

117

Results

118

118

Bowel content samples

Ammonium sulphate precipitation

Alkaline urea gel

Gel slices assayed for PLA2 activity.

Electrotransfer of proteins to PVDF membrane.

Region corresponding to PLA2 activity submitted to N-terminal sequencing.

Results: proteins identified as Haemoglobin α and β.

Figure 5.1

Purification scheme 1.

119


Heparin sepharose chromatography

Fractions with highest PLA2activity retained.

Alkaline urea gel to verify mobility of PLA2 activity.

Concentrated ten fold (SpeedVac).

SDS-PAGE and electrotransfer of proteins to PVDF membrane.

N-terminal sequencing

Results: proteins identified as Haemoglobin α and β.Figure 5.2


120


Heparin sepharose chromatography


Preparative isoelectric focussing (Rotofor®).

Fractions with highest PLA2 activity retained.

Concentrated up to three fold (SpeedVac).

SDS-PAGE, silver stained, trypsin digest of protein bands.

Nano ES/MS/MS

Figure 5.3


Results: proteins identified as Haemoglobin β.

121

0

1

2

3

4

5

12.5 25 50

Haemoglobin concentration (micromolar).

PLA 2

act

ivit

y, %

hyd

roly

sis.

Figure 5.4: The lack of phospholipase A2 activity of

haemoglobin. Haemoglobin (human, Sigma) was

diluted in acetate buffer (pH 5.5). These results are

presented as the mean +/- standard error of

triplicate assays.

122

0

1

2

3

4

5

Haemoglobin Albumin Water Acetate buffer

PLA 2

act

ivit

y, %

hyd

roly

sis.

Figure 5.5: The lack of phospholipase A2 activity in

haemoglobin (25µM), bovine serum albumin (25µM), ultrapure

water and acetate buffer (pH 5.5).

123

0

20

40

60

80

100

0 12.5 25 50

Haemoglobin concentration (micromolar).

PLA 2

act

ivity

, % h

ydro

lysi

s.

Figure 5.6: Pancreatic phospholipase A2 (0.1µg/mL

containing 1mg/mL BSA, 10mM Tris) activity with

various amounts of added haemoglobin.

124

R2 = 0.75

0

2

4

6

8

10

0 0.5 1 1.

NaCl concentration (M) required for elutionis

oele

ctri

c po

in5

t

R2 = 0.84

0

2

46

8

10

12

0 0.5 1 1.5

NaCl concentration (M) required for elution

pI o

f 25

C-t

erm

inal

res

idue

s

R2 = 0.77

0

2

4

6

8

10

0 0.5 1 1.5

NaCl concentration (M) required for elution

Num

ber o

f bas

ic re

sidu

es in

25

C-te

rmin

al a

min

o ac

ids

a)

b)

c)

Figure 5.7: Heparin binding of PLA2 isoforms.

The sodium chloride concentration required to elute the PLA2 isoform from

heparin sepharose shows high correlation to the isoelectric point (pI) of

the isoform (Murakami et al, 2001, graph a) and to the pI (graph b) and

number of basic residues in the C-terminal portion of 25 amino acid

residues (graph c). Characteristics of the isoforms IIA, IID, IIE, IIF, V &

X were used to construct these graphs.

125

Figure 5.8: Sequence Alignment of Secretory Phospholipase A2 isoenzymes &

Heparin binding regions. The C-terminal 25 amino acids shown in red. The

basic residues in the C-terminal amino acids are underlined. Additional

amino acids that are likely to be involved in heparin binding are shown in blue

(Koduri et al, 1998).

Isoform

IIA

NLVNFHRMIKLTTGKEA-ALSYGFYGC ALSYGFYGC--HCGVGGR-GS-

IID

GILNLNKMVKQVTGKMP-ILSYWPYGC ILSYWPYGC--HCGLGGR-GQ-

IIE

NLVQFGVMIEKMTGKS—ALQYNDYGC ALQYNDYGC--YCGIGGS-HW-

IIF

SLLNLKAMVEAVTGRSA-ILSFVGYGC ILSFVGYGC--YCGLGGR-GQ-

V

GLLDLKSMIEKVTGKNA-LTNYGFYGC LTNYGFYGC--YCGWGGR-GT-

X

GILELAGTVGCVGPRT—PIAYMKYGC PIAYMKYGC--FCGLGGH-GQ-

IIA

PKDATDRCCVTHDCCYKRLEKRG-CGTKFLSY----KF SNSGS-RITCAKQ-----

IID

PKDATDWCCQTHDCCYDHLKTQG-CSIYKDYY----RY NFSQG-NIHCSDK-G---

IIE

PVDQTDWCCHAHDCCYGRLEKLG-CEPKLEKY----LF SVSER-GIFCAGR-----

IIF

PKDEVDWCCHAHDCCYQELFDQG-CHPYVDHY----DH TIENNTEIVCSDLNK---

V

PKDGTDWCCWAHDHCYGRLEEKG-CNIRTQSY----KY RFAWG-VVTCEPG-----

X

PRDAIDWCCHGHDCCYTRAEEAG-CSPKTERY---SWQ CVNQS--VLCGPA----

IIA

DSCRSQLCECDKAAATCFA RNKTTYNKKYQYYSNKH-CRGST P---------RC

IID

SWCEQQLCACDKEVAFCLK RNLDTYQKRLRFYWRPH-CRGQT P---------GC

IIE

TTCQRLTCECDKRAALCFRRNLGTYNRKYAHYPNKL-CTGPT P---------PC

IIF

TECDKQTCMCDKNMVLCLMN—QTYREEYRGFLNVY -CQGPT P---------NCSIY EPPPEEVTCSHQSPAPPAPP

V

PFCHVNLCACDRKLVYCLKRNLRSYNPQYQYFPNIL-CS--

X

ENKCQELLCKCDQEIANCLAQ--TEYNLKYLFYPQFL-CEPDS P---------KC

126

127

00.10.20.30.40.50.6

1 4 7 10 13 16 19 22 25 28 31Fraction

NaC

l co

ncen

trat

ion

(M)

0

200

400

600

800

1000

Tota

l pro

tein

(u

g/m

L)

NaCl concentration (M) Total protein (ug/mL)

Figure 5.9: Typical elution of the total protein a homogenised

infarcted bowel sample from a heparin sepharose column. The

majority of the total protein eluted with 0.05M NaCl.

128

Phospholipase A2 activity

NaCl 0M

NaCl 0.05MNaCl 0.1M

NaCl 0.2M

NaCl 0.5M

Total protein

NaCl 0M

NaCl 0.05M

NaCl 0.1M

NaCl 0.2M

NaCl 0.5M

Figure 5.10: The elution of total protein and PLA2 activity

from a typical heparin sepharose chromatography column.

There was no detectable total protein or phospholipase A2

activity with either 2 or 5M NaCl elution.

129

Protein size (kDa)

Protein size (kDa)

200

116

97

66

55

36

31

21

14

6

3

200

116

97

66

55

36

31

21

14

6

3

Column volumes per elution buffer 15 20

Total protein in active fractions (µg/mL)

26.57 3.91

PLA2 activity, % hydrolysis. 24.5 15.6

Figure 5.11: Effect on total protein content of elution from heparin

sepharose chromatography column using either 15 or 20 column

volumes per elution buffer. The use of twenty column volumes

decreased the amount of contaminating total protein while retaining the

majority of the enzyme activity.

130

0

20

4060

80

100

Dialysed Non dialysedPhos

phol

ipas

e A 2

act

ivit

y,

% h

ydro

lysi

s.

Figure 5.12: PLA2 activity before and after dialysis against

phosphate buffer (10mM). Dialysis of the active heparin

column fractions was required before application to the

Rotofor IEF system.

131

02040

6080

100

Water Bee sPLA2 Bowel Type IIAsPLA2

PLA 2

act

ivit

y (%

hyd

roly

sis)

Water Rotofor buffer

Figure 5.13: Confirmation that the Rotofor® buffer did not

interfere with assay of PLA2 activity. Three sources of PLA2

(bee venom sPLA2, bowel homogenate and purified type IIA

PLA2) and a negative control (water) were diluted in either

water or Rotofor® buffer.

Results are expressed as the mean +/- standard error of

triplicate assays.

132

0

200

400

600

800

1000

0 0.5 1 1.5 2 2.5 3 3.5Time (hours)

Volts

0246810121416

mAM

Ps, w

atts

Volts

mAmps

Watts

Figure 5.14: The power, current & voltage changes over

the duration of a typical Rotofor® run. The system was

run until the plateau phase had been reached and

maintained for an hour. A typical run took 3.5 hours.

133

0

5

10

15

1 3 5 7 9 11 13 15 17 19

Rotofor fraction

pH

0

20

40

60

80

PLA

2 ac

tivity

,%

hyd

roly

sis.

pH PLA2 activity

Figure 5.15: The pH and PLA2 activity of the fractions of a

typical Rotofor® experiment.

134

Mark 12

stand

ards

Mark 12

stand

ards

1 2 3 4 5 6 7 8 9 10 Origina

l sample

Origina

l sample

11 12 13 14 15 16 17 18 19 20

Exhibited PLA2 activity

Ampholytes

21

14

Figure 5.16: Typical Rotofor® fractions from pooled active

heparin fractions of infarcted bowel content eluted by 0.5M NaCl

and dialysed against 10mM phosphate. Rotofor fractions 15-18

exhibited PLA2 activity.

135

02468

1012

1 3 5 7 9 11 13 15 17 19

Fraction number

pH

2.5.03

5.5.03

7.5.03

8.5.03

9.5.03

Figure 5.17: The reproducibility of Rotofor® system. The

pH of each of the 20 fractions of five consecutive

Rotofor® experiments is shown in this graph.

136

Figure 5.18: Refractionation of a typical Rotofor® experiment. No

extra ampholytes were added, and the active fractions were 10-17.

The maximum phospholipase A2 activity was observed in fraction 17

which corresponded to a pH of 9.7. The arrow points to the region of

gel where phospholipase A2 was predicted to run to, as determined by

electroelution. The red line shows the phospholipase activity in each

Rotofor fraction, while the blue line shows the phospholipase activity

in each re-fractionated Rotofor fraction.

137

02468

101214

1 3 5 7 9 11 13 15 17 19Fraction number

pH

Rorofor 1

Re-Rotofor

Figure 5.19: The pH profile of Rotofor® fractions before

and after re-fractionation.

138

0204060

80100120

0 5 10 15Electroelution fraction number.

Size

of

prot

ein

(kD

a)

Figure 5.20: A standard curve representing the electroelution of

PLA2 from active Rotofor® fractions. The sample was run on a

non-reducing gel. PLA2 activity was focussed in fractions 8 & 9,

corresponding to a protein of 16-20kDa in size.

139

0

1

2

3

4

5

6.5 7 7.5 8 8.5 9 9.5pH

PLA 2

act

ivit

y (%

hyd

roly

sis )

Figure 5.21: The pH of maximum activity of the semi purified bowel

phospholipase. Results are expressed as the mean +/- standard

error of triplicate assays.

140

0

20000

40000

60000

80000

100000

Crude samp leHeparin co lumn Ro to fo r Electroelut ion0

2000

4000

6000

8000

10000

Specific act ivity To tal p ro tein

Figure 5.22: Purification summary of a typical experiment.

This purification experiment is equivalent to an average 9700

fold purification from the crude sample.

141

31

21

14

Prote

in

markers

11 12 13 14 15 1 6 17 18 19 20

Rotofor® fraction

Exhibited PLA2 activity

Size, (kDa)

Figure 5.23: A typical result from a Rotofor® run during the

scaled up purification, as shown by a silver stained gel of

active Rotofor® fractions. The arrow indicates a protein of

approximately 21kDa which was present in the fractions with

phospholipase activity.

142

Protein of interest

Ampholytes

Ethano

l prec

ipitat

ion

Vivasp

inco

ncen

tratio

n

Figure 5.24: Gel showing the effect of concentrating

protein and removing ampholytes using either

VivaSpin centrifuge filters or ethanol precipitation.

143

Sample Top 1-2 identifications

Number of peptides

Mowsescore

Unknown protein ~20kDa.

Peptidylprolylisomerase.*Keratin.

15

8

465

414

Positive control, lysozyme.

Lysozyme (chicken). 5 204

Negative control.

Keratin. 15 588

*Proteins matching the same set of peptides:

•Cyclophilin B (ICYNA)

•Cyclophilin B (Q9BVK5)

•Human cyclophilin

Figure 5.25: Mass spectrometry of the unknown protein from trypsin

digests of gel pieces (BisTris gel, Invitrogen).

144

Sample Top 1-2 identifications.

Number of peptides.

Mowse score.

Unknown protein ~20kDa-A.

Peptidylprolylisomerase.*Keratin.

13

5

427

272

Unknown protein ~20kDa-B.

Keratin.Peptidylprolylisomerase.*

810

435386

Unknown protein ~20kDa-C.

Peptidylprolylisomerase.*Bovine cyclophilin B.

10

16

426

331

Postive control, Restriction endonuclease E.Coli.

Restriction endonuclease Bacillus sp.

48 723

Negative control Keratin. 6 333

*Proteins matching the same set of peptides:

•Cyclophilin B (ICYNA)

•Cyclophilin B (Q9BVK5)

•Human cyclophilin

Figure 5.26: Mass spectrometry of the unknown protein from trypsin

digests of gel pieces (TrisGly gel, Gradipore).

145

Figure 5.27: Final purification scheme. This protein purification scheme

resulted in sufficient purified protein for identification of the protein of interest

by mass spectrometry.

146


Heparin sepharose chromatography, 20 column volumes.


Preparative isoelectric focussing (Rotofor®).

Fractions with highest PLA2 activity retained.

Concentrated ten fold (VivaSpin).

SDS-PAGE, TrisGlycineiGEL, trypsin digest of protein bands.

SDS-PAGE, BisTris, trypsin digest of protein bands.

LC/MS/MS

Results: proteins identified as Cyclophilin B.

Discussion


The aim of the protein purification experiments was to identify the protein responsible

for the phospholipase activity that remained when phospholipase A2 isoforms IIA and

V were removed from human bowel. The protein of interest was purified from

infarcted gut tissue homogenate rather than blood for a number of reasons:

- a larger amount of tissue was available,

- the tissue showed higher specific activity,

- a bowel tissue specific protein was more likely to be found.

The separation techniques were chosen because they retained the native enzyme

activity of proteins and they removed large quantities of contaminating protein at each

step. The separation techniques also exploited unusual characteristics of the protein

of interest (namely heparin affinity and very alkaline isoelectric point). The progress

of the purification was monitored by analysing all fractions obtained for phospholipid

hydrolysing activity. The assay used for monitoring phospholipase activity (the

mixed micelle assay) has been well described and is very sensitive (Farrugia et al,

1993, Green et al, 1991). This assay was chosen for its sensitivity, ability to be used

with crude and purified samples and ability to detect net phospholipase activity. A

series of PLA2 isoenzyme specific assays have been published (Yang et al, 1999), but

this was not suitable as we required an assay to monitor net phospholipase activity for

the widest range of phospholipase proteins.

Because of the low abundance of non-pancreatic phospholipase A2 in biological

samples, proteins need to be purified several thousand fold (Stefanski et al, 1986). In

the past, purification of phospholipase A2 from animal tissues has involved the use of

several sequential chromatographic columns, often in conjunction with phospholipase

assays of the fractions obtained where compatible. A typical phospholipase A2

purification process may involve one or more anion or cation exchange columns, a

hydrophobic interaction column, one or more gel filtration columns, and a final

reversed phase high performance liquid chromatography (RP-HPLC) column. For

example, Minami and colleagues used a cation exchange column (run and then

repeated) followed by a gel filtration column (also run twice), another cation

exchange column and a final RP-HPLC column to purify phospholipase A2 type II

from human ileal mucosa (Minami et al, 1993). For tissues such as spleen, multiple

ion exchange columns (cation and anion), a hydrophobic interaction column and RP-

147

Discussion

HPLC column have been used (Kanda et al, 1989, Ono et al, 1988). For biological

fluids such as synovial fluid, generally fewer columns are required, for example an

ion exchange and a gel filtration column followed by RP-HPLC and PAGE (Kramer

et al, 1989) or hydrophobic interaction, HPLC and PAGE (Seilhamer et al, 1989).

The use of multiple chromatography columns is associated with the risk that

phospholipase activity may be lost because of the buffers or solvents used or the

numerous manipulations, and therefore the activity of the protein of interest may be

reduced or lost during purification. For this project we chose to employ a small

number of purification techniques to exploit the unusual characteristic of

phospholipase A2 instead of a large number of chromatography columns.

Heparin sepharose chromatography was used for a number of reasons:

- it removed a large amount of contaminating protein in one experiment, and

retained enzyme activity,

- discrete fractions could be achieved with high specific activity by using a

step gradient of salt concentrations,

- the high salt concentrations required to elute the tightly bound protein could

be reduced by dialysis without altering the enzyme activity,

- the system could be scaled up or down relatively easily.

Groups that have used heparin affinity chromatography in protein purification have

generally employed a linear gradient of salt rather than a step gradient, but we found a

step gradient useful in terms of reproducibility and obtaining highly purified fractions.

Preparative isoelectric focussing was conducted using the BioRad Rotofor®

apparatus. The buffer used in the Rotofor® system did not appear to interfere with

the PLA2 assay, which was an advantage because the fractions could be analysed

without any modification, such as dialysis. The Rotofor® system was useful for a

number of reasons:

- the system was highly reproducible,

- the fractions obtained could be analysed in a number of different ways (pH,

enzyme activity, total protein, SDS-PAGE, electroelution),

148

Discussion

- the interference of the ampholytes in total protein assays and in protein

staining in SDS-PAGE could be overcome via a number of different strategies

(Coomassie G-250 stains with or without copper and protein precipitation

prior to total protein analysis).

The majority of the publications involving the Rotofor® system have investigated the

proteins of the cerebrospinal fluid which may reflect the development of Alzheimer’s

disease (Davidsson, BioRad technote 2859). Westman-Brinkmalm and Davidsson

(2002) found that the Rotofor® apparatus could be used as a pre-fractionation step for

cerebrospinal fluid proteins, and this pre-fractionation resulted in increased sequence

coverage and greater identification of low abundance proteins. The Rotofor®

apparatus is an excellent tool in protein purification and is particularly suited to

enzymes with compatible assays. At present, such a valuable tool in protein

purification and proteomics appears to be under-utilised.

The initial purification system of heparin sepharose chromatography, Rotofor®

fractionation and Rotofor® refractionation did not provide sufficient protein for

identification so the system was scaled up approximately twenty fold. The scaled up

purification resulted in a protein that was >95% pure on a silver stained gel, and

represented a 9700 fold purification. Throughout the purification system, a number

of protein characteristics were recognized:

- the protein size appeared to be between 16 and 20kDa,

- the phospholipase activity of the protein appeared to be sensitive to reduction

by DTT,

- the phospholipase activity of the protein appeared to be calcium dependent,

- the pH of optimal phospholipase activity appeared to be approximately 9,

- the isoelectric point of the protein appeared to be between 8.6 and 9.8,

- the protein appeared to have heparin affinity up to conditions of

approximately 0.5M sodium chloride.

Many of these features were suggestive of a secretory phospholipase isoform, but no

clear match was evident. Therefore a conclusive identification was required,

preferably by mass spectrometry.

149

Discussion

Identification of the protein of interest:

Haemoglobin was identified on a number of occasions (by Edman degradation and

mass spectrometry), due to haemoglobin either:

-co-purifying with the protein of interest,

- contaminating the samples simply because it was present in high

concentration or because of its ‘sticky’ nature (Associate Professor Mark

Kirkland, personal communication).

The problems with proteins co-purifying with phospholipase A2 has been encountered

previously, for example, histone proteins co-purified with phospholipase A2-IIA from

synovial fluid (Dr. Kieran Scott, personal communication). The possibility that

haemoglobin may have intrinsic phospholipase A2 activity was dismissed by assaying

a range of haemoglobin concentrations for phospholipase activity. None of the

haemoglobin concentrations showed phospholipase activity above background levels

(similar to the background seen in ultrapure water, buffers or albumin solution). In

addition, haemoglobin did not appear to enhance phospholipase activity.

Haemoglobin was not irreversibly bound to the protein of interest, as the size of the

protein as determined by electroelution under non reducing conditions was too small

to be complexed to haemoglobin (molecular weight of ~16kDa). When a solution of

standard haemoglobin and phospholipase A2 from either pancreas or bee venom was

run on a gradient SDS-PAGE, distinct bands for each protein were visible, suggesting

that:

- if the haemoglobin binds to phospholipase A2 proteins, the association is

disrupted by SDS,

- a gradient SDS-PAGE gel was required as the final purification step to limit

the possibility of identifying haemoglobin in subsequent experiments.

The final purification system was comprised of heparin sepharose affinity

chromatography (with twenty column volume elution), preparative IEF (Rotofor®

apparatus) and a gradient SDS gel. The final identification of the protein of interest

was made using nano-LC MS/MS. The final identification was made from two

highly purified fractions, each run in a different gel system with separate positive

controls. The two different gel systems were used to add strength to the

150

Discussion

identification, and to overcome the possibility that one type of gel may be

incompatible with the mass spectrometry system. The protein of interest was an

excellent match for cyclophilin B, a protein that is also referred to as peptidylprolyl

isomerase. Cyclophilin B is an inflammatory protein (Allain et al, 2002, Carpentier

et al, 1999) that had been previously identified in rodent intestine at the mRNA level

(Hasel et al, 1991, Kainer & Doris, 2000).

Cyclophilin B is known to have heparin sulphate binding ability (De Ceuninck et al,

2003). Heparin affinity chromatography has been used to semi-purify cyclophilin

proteins previously (Carpentier et al, 1999, De Ceuninck et al, 2003, Sherry et al,

1992). Cyclophilin A (64% homology to cyclophilin B) cannot interact with

glycosaminoglycans such as heparin because it lacks the required N-terminal binding

residues, likely to involve a triplet lysine (Allain et al, 2002, Bukrinsky, 2002). The

binding of cyclophilin B with glycosaminoglycans is thought to contribute to

cyclophilin involvement in inflammation (Allain et al, 2002). The affinity of

cyclophilin B to heparin is quite strong, and this protein is not eluted from a heparin

column until sodium chloride concentration exceeds 0.6M sodium chloride

(Carpentier et al, 1999).

The size (20.289kDa, Carpentier et al, 1999), isoelectric point (9.25) and heparin

affinity (0.6M NaCl, Carpentier et al, 1999) of cyclophilin B were consistent with the

characteristics that had been determined throughout the protein purification

experiments. The next chapter details our investigations into cyclophilin B in human

bowel.

151


Chapter 6: Cyclophilin B.

Immunophilins:

The immunophilins are a family of heterogenous proteins grouped by their ability to

bind the immunosuppressive agents cyclosporin A (the cyclophilin proteins), FK506

(FK506 binding proteins) or rapamycin (target of rapamycin or Tor proteins, (Ivery,

2000)). The FK506 binding proteins have no sequence homology to the cyclophilins

(Snyder & Sabatini, 1995). Immunophilins have peptidyl prolyl isomerase activity,

which catalyses folding of proline containing proteins in vivo and in vitro. This

isomerase activity is inhibited on binding to the relevant immunosuppressive drug

(Bochelen, Rudin & Sauter, 1999). Immunophilins have the ability to interact with

many signal transduction molecules, in particular those involving calcium or

phosphorylation (Snyder & Sabatini, 1995).

Cyclophilins:

Cyclophilin was first purified from bovine thymus and human spleen (Harding,

Handschumacher & Speicher, 1986). To date, more than forty gene encoded

cyclophilins have been recognised (Fanghanel & Fischer, 2003). The major human

isoforms are described in Table 6.1. Cyclophilins have been found in all cell types

investigated (Price et al, 1991) and are highly conserved (Allain et al, 1995). The

widespread distribution and significant levels of expression suggest an important

biological role (Ivery, 2000), but the relevance of their biological activity is unknown

(Denys et al, 1998). Cyclophilins may function as catalysts of protein folding (Kay,

1996), or protein stabilisation (Galat & Metcalfe, 1995), as molecular chaperones

(Yoo et al, 1997) or cytokines (Tegeder et al, 1997).

The two major isoforms of cyclophilin, cyclophilins A and B, are only 64%

homologous (Price et al, 1991), and the major difference between the two is the signal

sequence in cyclophilin B directing it into a secretory pathway (Price et al, 1991,

Price et al, 1994). Cyclophilin A is a cytosolic protein (Price et al, 1994), and can

comprise up to 0.1-0.4% of the soluble protein in the cytoplasm (Harding,

Handschumacher & Speicher, 1986). Cyclophilin A is the most abundant cyclophilin

in human cells (Fanghanel & Fischer, 2003). It is assumed that cyclophilin A is non-

secretory, although this protein has been detected in the synovial fluid from patients

152


with rheumatoid arthritis (Billich et al, 1997). Cyclophilin A is present in leukocytes

at around 10 fold higher levels than cyclophilin B, but is not released into the plasma

(Allain et al, 1995). Cyclophilin B is a secreted protein which has been detected in

human blood (Allain et al, 1995) and breast milk (Spik et al, 1991).

Cyclophilin C (23kDa) localises to the endoplasmic reticulum (Snyder & Sabatini,

1995) and has been shown to be a secretory protein (Price et al, 1994). Cyclophilin

40 is a component of the inactive form of the glucocorticoid receptor (Synder &

Sabatini, 1995). Cyclophilin NK plays an important role in natural killer cell

cytotoxicity (Anderson et al, 1993).

153


Table 6.1. The Major Human Cyclophilins: Protein Alternate names Molecular

weight, kDa

Isoelectric

point

Signal

peptide

Abundant

sources

Cyclophilin A Cyclophilin 1

Cyclophilin 18

18 7.8 None. Cytosolic,

ubiquitous

(Hasel et al,

1991).

Cyclophilin B S-cyclophilin

SCYLP

Cyclophilin 2

Cyclophilin 20

22.74,

unprocessed

20.29

(mature)

9.25 (mature) 1-25 signal

26-208 mature

protein

Liver,

intestine and

placenta

(Hasel et al,

1991).

Cyclophilin C Not applicable. 23 8.48 None. Liver, lung,

heart,

pancreas,

skeletal

muscle

(Schneider et

al, 1994).

Cyclophilin D Cyclophilin 3

Cyclophilin M

Mitochondrial

cyclophilin.

18 8.97 1-29 potential

signal, 30-207

chain.

Muscle, heart,

liver, kidney,

brain

(ExPASY

database).

Cyclophilin

40

Not applicable. 40 6.77 None Heart, brain,

placenta,

skeletal

muscle,

pancreas, liver

(Kieffer et al,

1993).

Cyclophilin

NK

Not applicable. 150 (mature

protein),

19.5

(cyclophilin

like domain).

10.01 None. Membrane

anchored on

natural killer

cells. Spleen,

lung, liver

(Anderson et

al, 1993).

154


Cyclophilin B in human disease:

The levels of cyclophilin B have been investigated in relation to sepsis, HIV and

organ transplant in clinical settings. A 1997 study investigated the levels of

cyclophilins A and B in the plasma of patients with severe sepsis, other diagnoses and

healthy controls using western blotting and a peptidyl prolyl isomerase activity assay

(Tegeder et al, 1997). The healthy control plasma showed a weak but distinct

cyclophilin B signal, but no cyclophilin A signal. Patients with severe sepsis showed

high levels of cyclophilin B and some (6/22) patients also showed a cyclophilin A

signal. Patients with diagnoses other than sepsis showed variable levels of

cyclophilin B, but no cyclophilin A (Tegeder et al, 1997).

Organ transplant patients showed significantly elevated levels of cyclophilin B in

plasma compared with health controls (Denys et al, 1998). It was postulated that

cyclophilin B levels may effect a patient’s sensitivity to the immunosuppressive drug

cyclosporin A (Denys et al, 1998). Levels of cyclophilins A and B were found to be

increased in the serum of patients with HIV infection (Endrich & Gehring, 1998).

The cyclophilin B-Cyclosporin A complex:

The pharmacological ligand of cyclophilins, cyclosporin A is known for its

immunosuppressive effects. The binding of cyclosporin A to a cyclophilin is

required for the immunosuppressive effect associated with this drug (Gonzalez-

Cuadrado et al, 1996). The cyclophilin-cyclosporin A complex then interacts with

protein targets such as calcinuerin (Ivery, 2000), to exert the immunosuppressive

effect. All of the therapeutic and toxic effects of cyclosporin A (and FK506) have

been found to be due to the inhibition of calcinuerin (Ho et al, 1996). The

cyclophilin B-cyclosporin A complex shows the most potent binding to calcineurin of

all cyclophilins (Kay, 1996).

Cyclosporin A and cyclophilin B appear to exert effects on each other. Cyclosporin

A has been shown to increase levels of cyclophilin B mRNA in a dose dependent

manner (Galat & Metcalfe, 1995, Gonzalez-Cuadrado et al, 1996, Price et al, 1994).

Cyclosporin A has been shown to significantly increase levels of cyclophilin B

protein in plasma compared with healthy donors (Denys et al, 1998). Cyclosporin A

has been shown to increase the secretion of cyclophilin B in cultured chondrocytes

155


(De Ceuninck et al, 2003). Cyclophilin B modifies the distribution of cyclosporin A

in blood, channelling the drug into peripheral blood mononuclear cells (PBMC) or

retaining the drug-protein complex extracellularly (Denys et al, 1998). Cyclosporin

A redirects cyclophilin B from the endoplasmic reticulum into a secretory pathway

(Price et al, 1994). The CyPB-CsA complex inhibits T cell proliferation more

effectively than cyclosporin A alone (Denys et al, 1998).

Use of cyclophilin B antibodies:

Due to the high homology of cyclophilins, antibodies generated to intact cyclophilin

proteins tend to cross react with multiple cyclophilin forms. For example antibodies

raised to recombinant cyclophilin B crossreact with cyclophilin A (Allain et al, 1995).

Useful antibodies to cyclophilin B have been generated by using a 14 residue peptide

from the C-terminal, variable region, coupled to keyhole limpet hemocyanin (KLH) to

improve the immunogenicity of the peptide (Vaitukaitis et al, 1971).

Western blotting for cyclophilin B:

One study employed an antibody raised in the mode discussed above to investigate

cyclophilin B in Jurkat cells (Bergsma et al, 1991). Jurkat cells showed four

immunoreactive bands in Western blots, but only one band could be removed by

incubating the extract with an excess of the peptide used to generate the antibody

(Bergsma et al, 1991).

ELISA studies of cyclophilin B:

Another study employing a peptide derived cyclophilin B antibody devised an ELISA

to assay blood cells and serum from blood transfusion samples (Allain et al, 1995).

A modified version of this ELISA was used to investigate cyclophilin B in organ

transplant patients (Denys et al, 1998).

Cyclophilin B in gut:

Studies in rats have shown evidence of cyclophilin B mRNA transcripts in the

intestine using competitive RT-PCR (Kainer & Doris, 2000) and Northern blotting

(Hasel et al, 1991). Competitive RT-PCR of rat tissues showed that the highest

expression of cyclophilin B occurred in the kidney, followed by the ventricle, liver

and small intestine (Kainer & Doris, 2000). Northern blotting of rat tissues showed

156


that the highest expression was in the placenta and liver, followed by the intestine and

uterus (Hasel et al, 1991).

157

Results

Chapter 6: Cyclophilin B in human bowel:

Previous research had identified cyclophilin B in Jurkat cells, (a T-cell lymphoma cell

line) therefore whole cell lysate was used as a positive control in these investigations.

Jurkat cell lysate and cell culture medium were run on SDS-PAGE to estimate protein

loading (Figure 6.1). Bowel tissue from patients with surgically confirmed bowel

infarction and control patients with normal bowel all showed cyclophilin B protein by

Western blotting (Figures 6.2, 6.3 & 6.4). Plasma from patients with confirmed

bowel infarction showed variable levels of cyclophilin B, but control patients showed

no cyclophilin B (Figure 6.7). Some lumen content from infarcted bowel showed

cyclophilin B protein, though at much lower levels than tissue homogenates (Figure

6.5). Immunoprecipitation of cyclophilin B from highly purified fractions removed

all of the phospholipase activity from these samples (Figure 6.6). The phospholipase

activity was not altered by the addition of cyclosporin A (Figure 6.10).

Immunohistochemistry was performed using an immunoperoxidase method and the

commercially available polyclonal antibody raised to a cyclophilin B peptide. After

the antibody dilution had been optimised, the ubiquitous expression of cyclophilin B

was confirmed (Figure 6.9).

158

159

1 2 3 4 5 6 7 8 9

116

66.2

45

35

25

18.4

Figure 6.1: Colloidal Coomassie stained gel of Jurkat cell lysate and cell culture

medium.

Lanes 1-3 Jurkat cell lysate, increasing amounts of sample loaded

Lane 4 No protein loaded

Lanes 5-8 Jurkat cell culture medium, increasing amounts of sample loaded

Lane 9 Protein Markers

160

1 2 3 4 5 6 7 8 9 10

26kDa

0100002000030000400005000060000700008000090000

2 3 4 5 6 7 8 9 10

Net

inte

nsity

of s

igna

l

Figure 6.2: Western blotting of cyclophilin B in infarcted human bowel samples, 30µg total protein per lane.1 Markers2 Jurkat cell lysate3 Jurkat cell culture medium4 no protein loaded5 Bowel tissue A6 Bowel mucosa A7 Bowel tissue B8 Bowel mucosa B9 Bowel tissue C10 Bowel mucosa C

161

1 2 3 4 5 6 7 8 9 1026kDa

0

20000

40000

60000

80000

2 3 4 5 6 7 8 9 10

Net

inte

nsity

of s

igna

l

Figure 6.3: Western blotting of cyclophilin B in infarcted human bowel samples, 30µg total protein per lane.1 Markers2 Jurkat cell lysate3 Jurkat cell culture medium4 no protein loaded5 Bowel tissue D6 Bowel tissue E7 Bowel tissue F8 Bowel tissue G9 Bowel tissue (infarcted)10 Bowel tissue (ischaemic)

162

2 3 4 5 6 7 8 9 10

26kDa

0

20000

40000

60000

80000

100000

2 3 4 5 6 7 8 9 10

Net

inte

nsity

of s

igna

l

Figure 6.4: Western blotting of cyclophilin B in control (normal) human bowel samples, 30µg total protein per lane.2 Jurkat cell lysate3 Jurkat cell culture medium4 no protein loaded5 Control bowel tissue A6 Control bowel tissue B7 Control bowel tissue C8 Control bowel tissue D9 Control bowel tissue E10 Control bowel content

163

2 3 4 5 6 7 8 9 10

26kDa

0

40000

80000

120000

160000

2 3 4 5 6 7 8 9 10

Net

inte

nsit

y of

sig

nal

Figure 6.5: Western blotting of cyclophilin B in bowel content samples, 40µg total protein loaded per lane.2 Jurkat cell culture medium3 Infarcted bowel content A4 Infarcted bowel content B5 Infarcted bowel content C6 Infarcted bowel content D7 Infarcted bowel content E8 Infarcted bowel content F9 Infarcted bowel content G10 Infarcted bowel content H

164

05

1015202530

Rotofor fraction 7.5 Rotofor fraction 8.5

PLA 2

act

ivit

y, %

hyd

roly

sis.

Before IP After IP

Figure 6.6: Phospholipase activity before and after immunoprecipitation

(IP) of cyclophilin B from Rotofor® fractions. Results are expressed as

the mean +/- standard error of duplicate assays.

165

0

1000

2000

3000

4000

5000

2 3 5 6 7 8 9 10 13 14 15 16 17 18 19 20

Patient

Net

inte

nsity

of s

igna

l

Bowel infarction patients:Positive control 2 3 5 6 7 8 9 10

26kDa

Control patients:

Positive control 13 14 15 16 17 18 19 20

26kDa

Figure 6.7: Western blotting of cyclophilin B in plasma, 30µg total protein per lane.

Patients 1-10 had surgically confirmed bowel infarction while patients 11-20 were

control ICU patients. The positive control used in these blots was Jurkat cell culture

medium.

MKVLLAAALI

AGSVFFLLLP

GPSAADEKKK

GPKVTVK

VY

FD

LRIGDEDVG

RV

IF

GL

FG

KT

VPKTVDNFVA

LATGEKGFGY

KNSKFHR

VI

KD

FM

IQ

GG

DF

TRGDGTGGKSI

YGER

FP

DE

NF

KLKHYGPGWV

SMANAGK

DT

NG

SQ

FF

IT

TV

KT

AW

LD

GK

HV

VF

GK

VL

EG

ME

VV

RKVESTKTD

SRDKPLKDVI

IADCGK

IE

VE

KP

FA

IA

KE

Figu

re 6

.8: A

min

o ac

ids s

how

n in

blu

e ar

e th

e try

ptic

pep

tides

obt

aine

d in

this

stud

y (B

MSF

mas

s

spec

trom

etry

). A

min

o ac

ids i

n re

d: si

gnal

pep

tide

clea

ved

in t

he m

atur

e pr

otei

n.

166

167

Figure 6.9: Immunohistochemistry of cyclophilin B in normal and infarcted

human intestine. Picture a) shows the normal bowel test slide (cyclophilin B

immunohistochemistry (brown staining), counterstained with haemotoxylin

(blue staining)) . Picture b) shows the infarcted bowel test slide. Picture c)

shows the negative control slide (haematoxylin stained). These

photomicrographs were taken at a 200X magnification. The scale bars

represent 100µm.

168

a)

b)

c)

169

02468

10

Fraction 7.5 Fraction 8.5Phos

phol

ipas

e ac

tivity

, %

hydr

olys

is

Cyclosporin A No cyclosporin A

Figure 6.10: Cyclosporin A (1µg/mL, Sigma) does not affect

phospholipase activity of cyclophilin B containing samples. Results

are expressed as the mean +/- standard error of triplicate assays.

Discussion


The unknown protein purified in this project was identified as cyclophilin B by nano-

LC tandem mass spectrometry. The accuracy of the mass spectrometry was verified

by the addition of positive controls (lysozyme and restriction endonuclease), all of

which gave very good matches upon database searching. The identification was

repeated from two different highly purified fractions, in two different gel systems.

Each of the samples from the two gel systems showed that the best match was

cyclophilin B, also referred to as peptidyl prolyl isomerase. This was an unexpected

finding because the progress of the protein purification had been monitored at each

stage by phospholipase hydrolysis assays, and cyclophilin B has not been reported to

have phospholipid hydrolysing activity. To investigate the possibility that

cyclophilin B had the ability to hydrolyse phospholipids, the cyclophilin B was

removed from highly purified fractions via immunoprecipitation with a commercially

available polyclonal antibody raised to a cyclophilin B peptide. The peptide used to

raise the antibody (194GKIEVEKPFAIAKE208) was one of the peptides identified by

the tandem mass spectrometry, which added clarity to the experiment. The presence

of cyclophilin B in the highly purified samples was verified by Western blotting of the

protein attracted to the resin immobilised antibody. This result also suggested that

the conditions for immunoprecipitation (for example: ionic strength of the buffer,

incubation times) were adequate for the removal of cyclophilin B from solution. The

samples devoid of cyclophilin B were then analysed for phospholipid hydrolysing

activity, and found to have no remaining activity. This result suggests that

cyclophilin B is able to hydrolyse phospholipid substrates. The converse experiment

(examining the possible phospholipid hydrolysing activity of a recombinant source of

cyclophilin B) would be a valuable exercise, but an application to the only known

pharmaceutical supplier of this protein was unsuccessful. To our knowledge there is

no commercial source of the recombinant protein.

A similar experiment employing the commercially available cyclophilin A protein

was not performed, as cyclophilins A and B have only moderate homology (64%) and

quite distinct roles in inflammation. The addition of cyclosporin A did not appear to

affect the phospholipase activity of the highly purified cyclophilin samples. This

suggests that amino acids involved in the proposed phospholipid hydrolysis are likely

to be separate from residues involved in cyclosporin binding.

170

Discussion

Western blotting was undertaken to assess the presence of cyclophilin B in a variety

of bowel samples. For these experiments, Jurkat cell lysate and the culture medium

form Jurkat cell culture were used as positive controls. Jurkat cells (a T-cell

lymphoma cell line) have been shown to express significant quantities of cyclophilin

B (Bergsma et al, 1991). An appropriate negative control tissue could not be found,

as the expression of cyclophilin is reported to be ubiquitous (Allain et al, 1995,

Tegeder et al, 1997). Jurkat cell extracts were also used as an internal control in the

Western blots, to control for signal intensity in the semi-quantitative ECL Western

system employed in this project. The Jurkat cell lysate and cell culture fluid were

used as the controls in blots of samples with high (eg tissue) and low (eg plasma)

levels of cyclophilin B respectively.

Western blots of cyclophilin B protein exhibited a number of non-specific bands of

different sizes than expected for cyclophilin B protein. The non specific bands could

be removed by more stringent washing procedures or stripping the membranes and re-

probing. The non-specific bands seen with this polyclonal cyclophilin B antibody

have also been shown by Bergsma and colleagues (1991). This group showed that

only one band was due to cyclophilin B by incubating the samples with an excess of

the peptide used to generated the antibody, which cleared the cyclophilin B band from

the blot (Bergsma et al, 1991). This approach would be a valuable control

experiment in future blotting experiments.

Bowel tissue homogenates from both normal bowel and infarcted bowel all showed

very strong signal for cyclophilin B of the expected size. This is not a surprising

result given the reported ubiquitous expression of cyclophilin B, and the reported high

mRNA expression in rodent gut (Hasel et al, 1991, Kainer & Doris, 2000). However,

it is supporting evidence of similar expression patterns in human tissue, at the protein

level. The infarcted bowel lumen content samples contained variable signal intensity

of cyclophilin B, but were generally lower than the signal intensity of tissue

homogenates. Although only one sample of lumen content from normal human

bowel could be obtained, this sample contained no signal for cyclophilin B. This is

an interesting finding as it supports the concept that cyclophilin B is present in large

amount in bowel tissue and may be released in response to the ischaemic conditions.

171

Discussion

In the two tissue samples obtained from one patient, the infarcted tissue showed a

higher cyclophilin B signal than the ischaemic tissue. None of the control patients’

plasma displayed cyclophilin B, while several samples of plasma from bowel

infarction patients exhibited this protein. The possible role of cyclophilin B as a

mediator, or marker of ischaemic damage to the human bowel is an interesting

concept that requires further investigation. Interestingly, this observation is in

accordance with previous reports in other human diseases.

THE ROLE OF CYCLOPHILIN B IN INFLAMMATION:

Increased levels of cyclophilin B protein have been observed in the plasma of patients

with sepsis (Tegeder et al, 1997), HIV infection (Endrich & Gehring, 1998) and organ

transplant (Denys et al, 1998). As such, it has been postulated that cyclophilin B has

an important role in inflammation (Allain et al, 2002), and this role is due in part to

the induction of chemotaxis and proliferation of T lymphocytes (Denys et al, 1997).

Chemotaxis:

Cyclophilins induce chemotactic activity with neutrophils (Billich et al, 1997), T

lymphocytes (Allain et al, 2002), polymorphonuclear leukocytes (Sherry et al, 1992)

and monocytes (Xu et al, 1992). Cyclophilin B, in particular, induces the migration

of neutrophils and T lymphocytes (De Ceuninck et al, 2003), which induces the

inflammatory cascade (Denys et al, 1997). Chemotactic inducing activity of

cyclophilins B is thought to be due to the same binding site as cyclosporin A, as the

addition of cyclosporin A abolishes chemotactic ability (Allain et al, 2002).

Cyclophilin binding sites:

Cyclophilin B shows pro-inflammatory effects and enhances integrin mediated

adhesion of CD4+ T-lymphocytes to the extracellular matrix, through interaction with

glycosaminoglycans via the type II binding site (Allain et al, 2002). The binding

sites for glycosaminoglycans and cyclosporin A are independent and located on

opposite sides of the cyclophilin B molecule (De Ceuninck et al, 2003, Mikol, Kallen

& Walkinshaw, 1994). The cyclosporin A binding site, or type I binding site is due

to a cluster of residues in the central core of the cyclophilin B molecule (De Ceuninck

et al, 2003, Denys et al, 1998). The type II binding site is responsible for the binding

172

Discussion

of glycosaminoglycans, and is due to a cluster of N terminal cationic residues (De

Ceuninck et al, 2003, Denys et al, 1998).

The possible involvement of cyclophilin B in oxidative stress:

Cyclophilin B has been implicated in the development of oxidative stress. Oxidative

stress in vascular smooth muscle cells increases cyclophilin B secretion (Liao et al,

2000). More extensive research has focussed on the role of other cyclophilins (A and

D) during oxidative stress. Cyclophilin D appears to have an important role in

mitochondrial membrane pore opening and the consequential necrosis or apoptosis

(Halestrap, McStay & Clarke, 2002). Mitochondria are known to have a critical role

in apoptosis, which is a favourable alternative to the inflammatory necrotic process.

Under normal conditions membrane pores are non permeable. Under conditions of

moderate stress membrane pores remain closed and cells are channelled into the

favourable apoptotic pathway. Under conditions of high calcium (mM

concentrations), pores become non-specifically permeable, cells swell and necrosis

ensues. Under conditions of moderated calcium (µM concentrations), in the presence

of cyclophilin D the permeable, non-specific pore opening also occurs and cells are

channelled into the necrotic pathway (see Figure 6.11). This process is greatly

enhanced by oxidative stress, such as reperfusion after ischaemic episodes. The pore

opening can be inhibited by cyclosporin A and its analogues, at similar affinities to

peptidyl prolyl isomerase activity inhibition (Halestrap, McStay & Clarke, 2002). It

is unknown whether this phenomenon is specific to cyclophilin D, or a general

cyclophilin process.

When antisense technology was used to knockout cyclophilin A, cardiac myocytes

were more sensitive to oxidative stress, but the protective action of cyclosporin A was

more pronounced (Doyle, Virji & Crompton, 1999). This study concluded that two

cyclophilin isoforms were involved in the oxidative stress response. Cyclophilin A

had a protective role, and a second, unknown cyclophilin isoform had a detrimental

effect on cells (Doyle, Virji & Crompton, 1999). The protective role of cyclophilin A

was also observed in a model of oxidative stress in vascular smooth muscle cells

where this protein prevented apoptosis (Jin et al, 2000).

173

Discussion

It is unclear whether the effects shown by cyclophilins A and D are specific

responses, or general cyclophilin responses to oxidative stress. It would be valuable

to investigate the role of each of the major cyclophilin isoforms in oxidative stress and

ischaemic episodes, and the possible protective role of cyclophilin inhibitors.

174

Discussion

Figure 6.11. Cyclophilins and Pathological Pore opening. Modified from

Halestrap A. The Mitochondrial Permeability Transition- A Pore Way for the

Heart to Die. J. Clin. Basic Cardiol. 2002; 5:29-41, with permission from the

author.

Cyclophilin B inhibition.

There are several examples of the inhibition of cyclophilins by cyclosporin A having

a protective or anti-inflammatory role, for example:

- inhibition of cyclophilin B reduced T-cell proliferation (Denys et al, 1998),

-inhibition of cyclophilin D avoided pore formation, preventing cellular

necrosis (Halestrap, 2002),

-inhibition of murine macrophage cyclophilin inhibited chemotactic induction

in monocytes (Sherry et al, 1992),

175

Discussion

- disruption of the binding of cyclophilin to a HIV capsid protein which

resulted in non infectious virions (Francke, Yuan & Luban, 1994, Thali et al,

1994),

- inhibition of chemotactic enhancing effect on eosinophils and neutrophils

(Xu et al, 1992).

Cyclosporin A can have unwanted effects such as inducing increased secretion of

cyclophilin B and has uncontrolled cellular distribution. Analogues of cyclosporin A

that specifically target one cyclophilin isoform (such as cyclophilin B) would be

advantageous, and may help to:

- prevent the drug compartmentalisation/distribution changes in presence of

cyclophilin B,

- retain the drug extracellularly (Endrich & Gehring, 1998),

- prevent the increased secretion of cyclophilin B seen when cyclosporin A is

administered (De Ceuninck et al, 2003),

- prevent the inhibition of possibly protective cyclophilin isoforms, such as

cyclophilin A (Halestrap, McStay & Clarke, 2002, Jin et al, 2000).

A synthetic analogue of cyclosporin A (SDZ-NIM811) was devised to have no

immunosuppressive effects. SDZ-NIM811 has been shown to disrupt the binding of

cyclophilin A to a critical protein domain in the HIV virion, and as a result the virion

was less infectious (Yoo et al, 1997).

The source of circulating cyclophilin B is unknown (Endrich & Gehring, 1998), and

this would be an area of importance in understanding role in inflammation. The

patterns of secretion also need to be better understood. The possible protective

features of cyclophilin B inhibitors in inflammation and infection require further

investigation.

176


Chapter 7: Alkaline Urea gels:

Traditional protein electrophoresis (SDS-PAGE) involves the separation of proteins in their

denatured (via SDS interactions) and reduced (via DTT or β-mercaptoethanol) state. The

separation of proteins in SDS-PAGE is dependent exclusively on the molecular weight of the

protein. Protein bands in SDS gels are sharply focussed, because the proteins exist as

monomers, are unfolded and interactions between proteins are not possible.

An alternate protein electrophoresis approach retains the native state of the protein (by the

omission of reducing agents), and any denaturation of proteins is either mild, and/or

reversible. Native gels are useful for investigations of protein subunit formation, protein

degradation and binding events (Alliance protein lab). Native gels are also useful when the

protein of interest has enzymic activity that can be detected post electrophoresis. Native gels

have a number of associated problems including:

- the mobility of proteins is not simply due to molecular weight*, [so protein band patterns

can be difficult to interpret (Chettibi & Lawrence, 1989)],

- bands can be indistinct, broadened or blurred due to protein-protein interactions,

polymers or multiple forms of proteins resulting from disulfide bonding.

The high concentrations of urea that are often used in native gels help to achieve high

resolution of proteins by eliminating protein-protein interactions and ensuring that proteins

exist in their unfolded state (Whittaker & Simpson, 1983). The denaturation of proteins by

urea is a reversible process, which allows for analysis of the native protein after renaturation.

High concentrations of urea can result in the formation of cyanate ions which can cause

artifactual spots or bands in gels via carbamylation of proteins (Gorg et al, 1997, Harris &

Angal, 1989) and N-terminal blocking (Harwig et al, 1993). Chemical modification as a

result of urea is a particular concern at high temperature and at alkaline pH.

Ahmad and colleagues (1994) employed a urea gel system of alkaline pH and high acrylamide

to resolve alkaline snake venom proteins, in particular venom phospholipase A2 isoforms.

Some 150 phospholipase A2 isoforms have been characterised in venoms (Valentin &

Lambeau, 2000). A single species of snake can express up to 15 distinct phospholipase A2

isoforms (Valentin & Lambeau, 2000). Phospholipase A2 is one of the most active venom

* Mobility in native gels is due to a combination of a protein’s size, shape and number of charged residues.

177


components and can exhibit a number of toxic effects including myonecrotic effects,

cardiotoxicity, anticoagulent effects, and hemorrhagic effects (Fortes-Dias et al, 1994).

Phospholipase A2 isoforms of high pI are of particular interest as it is thought that the most

potent venom toxins are either basic phospholipase A2 proteins or complexes involving

phospholipase A2 subunits (Karlsson, 1979). Ahmad and colleagues (1994) used these gels to

identify several PLA2 isoforms that had been unable to be resolved by IEF or ion exchange

chromatography. The high acrylamide content of these gels favoured the resolution of small

proteins and large peptides. These gels were particularly suited to the investigation of

phospholipase A2 because:

- they allow post separation analysis of enzyme activity,

- phospholipase A2 enzymes from snake venom are not denatured by high

concentrations of urea,

- PLA2 enzymes are stable at extreme pH, particularly high pH,

- PLA2 can be extracted at high yield (>60%) from the gels (Ahmad & Lawrence,

1993),

but most importantly:

- PLA2 isoforms show variable numbers of charged residues, particularly numbers of

acidic residues (Ahmad & Lawrence, 1993), which result in different mobilities in

these gels.

The relative mobility of phospholipase A2 isoforms in these gels is given by:

Number of Aspartic acid residues + number of Glutamic acid residues + number of

Tyrosine residues – number of Arginine residues + 1 (Ahmad, Lawrence & Moores,

1994).

These gels were also advantageous because they are rapid, inexpensive, use small amounts of

sample and allowed post electrophoretic enzyme activity analysis. These gels are also more

suited to comparative studies than IEF.

178

Results.

Chapter 7: Alkaline urea gels:

The two part alkaline urea gels devised by Ahmad, Lawrence and Moores (1993), for

the analysis of venom phospholipase A2 was investigated for use in the analysis of

mammalian phospholipase A2. A number of extensions were possible and may

broaden the applications available for these gels. Transfer of proteins from alkaline

urea gels to membranes would allow identification of proteins of interest by N-

terminal amino acid sequencing. Proteins could be transferred to PVDF membranes

from alkaline urea gels with no apparent loss of protein through the membrane

(Figure 7.1). No protein was evident in the gel after transfer (Figure 7.2). The

protein bands transferred to the membrane were sharp and very similar to the same

sample run on a gel without transfer (Figure 7.2). The transfer of small basic proteins

from alkaline urea gels to PVDF membrane was more efficient using the CAPS

transfer buffer than the Towbin transfer buffer (Figure 7.3). Proteins transferred

from alkaline urea gels to PVDF could be successfully sequenced (Figure 7.4), which

suggests that the process did not N-terminally block the proteins. Operating the gel

system in an ice bath and chilling the buffer aided in lowering the temperature of the

system (Figure 7.5), which may help to prevent protein modification. The length of

the gel could be more efficiently used by decreasing the acrylamide concentration and

decreasing the pH of the stacking gel (Figure 7.6). The protein bands in an alkaline

urea gel could be removed and run in an SDS-PAGE system to estimate the protein

size (Figure 7.7). This approach appeared to give a reasonable estimation of protein

size, and accuracy did not appear to be improved by reducing the proteins before

electrophoresis (Figure 7.8), thought this process may lead to loss of proteins.

These gels could be used to examine carbamylated and non-carbamylated protein

(Figure 7.9). These gels could also be used to examine forms of haemoglobin

(Figure 7.10). The extraction efficiency could be improved by lowering the pH of

the extraction buffer (Figure 7.11). Phospholipase A2 could be extracted from crude

tiger snake venom and analysed for enzyme activity (Figure 7.12). These gels could

also be used to examine the mobility of a variety of available of phospholipase A2

isoforms, which are very similar in size and difficult to resolve using SDS-PAGE

(Figure 7.13). These gels were able to separate the proteins with phospholipase A2

activity from normal and infarcted gut samples (Figure 7.14). The protein from

infarcted gut appeared to have higher mobility than that from normal gut. When the

179

Results.

bowel protein of interest was partially purified by heparin sepharose chromatography

it retained high mobility in these gels (Figure 7.15). The phospholipase A2 activity

from synovial fluid and normal human gut appear to run to a similar position in these

gels (Figure 7.16). The different mobilities of proteins in these gels did not appear

to be due to different salt concentrations/ionic strength of samples (Figure 7.18).

180

181

Figure 7.1: A typical PVDF membrane after blotting. The membrane surface in

contact with the gel is shown in the upper picture. The opposite side of the

membrane is shown in the lower picture. The sample shown in these pictures is

a bowel content sample, precipitated with ammonium sulphate (60%

saturation).

182

Figure 7.2: Appearance of gel (upper picture) after blotting and an equivalent gel

without blotting (lower picture). The sample shown in these pictures is a bowel

content sample, precipitated with ammonium sulphate (60% saturation).

183

1 2 3 4 5 6

1 2 3 4 5 6

Figure 7.3: Transfer of crude snake venom proteins and small basic proteins from alkaline urea

PAGE to PVDF. Transfer efficiency of CAPS buffer (upper membrane) and Towbin (lower

membrane) are compared.

Lanes 1&2 Crotalus scutulatus (Mojave rattlesnake) venom, 20µg protein.

Lanes 3&4 cytochrome c- (bovine heart) 4 µg protein. (Cytochome C- Mr= 12.33kDa, pI=10.6).

Lanes 5&6 sPLA2-III (Apis mellifera) 4 µg protein (sPLA2-III Mr=15.25kDa, pI=8.07).

184

1 2 3 4 5 6

Theoretical: IIYPGTLWCGHGN

Experimental: IIYPGTL……

Figure 7.4: N-terminal amino acid sequencing of proteins transferred from alkaline urea gels. Lanes 1&2 Crotalus scutulatus (Mojave rattlesnake), crude venom, 20µg per lane.Lanes 3&4 Cytochrome c, 4µg per lane.Lanes 5&6 Bee venom sPLA2, 4µg per lane.

185

05

1015202530

0 1 2 3 4 5

Time elapsed (hours)

Tem

pera

ture

(o C

)

Original gel

Modified gel

Figure 7.5: Running temperature of alkaline urea gels according to

the original method or the modified method. The running buffer was

replaced with fresh buffer after two hours.

186

1 2 3 4 5 6 1 2 3 4 5 6

Modified technique Original technique

Figure 7.6: Urea gels of crude snake venoms, run according to the original method or the modified method.Lanes 1&2: Agkistrodon rhodostoma (Malayan pit viper), 20µg protein per lane.Lanes 3&4: Crotulus scutulatus (Mojave rattlesnake), 20µg protein per lane.Lanes 5&6: Echis carinatus (Saw scaled viper), 20µg protein per lane.

187

Alkaline

urea PAGEProtein markers

664535

251814

SDS-PAGE

Figure 7.7: Extraction of protein bands from alkaline urea gels for SDS

PAGE size estimation. Sample used was crude venom of Crotalus

Scutulatus (Mojave rattlesnake). SDS-PAGE: 4-20% iGEL, TrisGlycine

gel (Gradipore). Protein bands could also be extracted for Tricine gel for

the estimation of size of very small proteins (results not shown).

188

1 2 3 4 5 6 7 8 9

Cyc

toch

rom

eC

Bee

ven

om

phos

phol

ipas

e A

2

116

66

45

35

25

18

14

116

66

45

35

25

18

14

1 2 3 4 5 6 7 8 9

Figure 7.8: An assessment of the accuracy of protein size determination after alkaline urea PAGE. The gel shown is an SDS-PAGE gel, 8-16% iGEL, Gradipore.Lanes:1 Protein markers2&3 Gel pieces from urea gel, reduced with method A4&5 Gel pieces from urea gel, non reduced6&7 Gel pieces from urea gel, reduced with method B (Davie, 1982)8 No prior urea gel, reduced 9 No prior urea gel, non reduced

189

1 2 3 4

1 2 3 4116

66

45

35

25

18

14

Figure 7.9: Carbamylation of bee venom sPLA2. The upper gel is an alkaline

urea gel. The lower gel is an SDS-PAGE gel of gel pieces from the alkaline

urea gel.

Lanes 1&2 carbamylated PLA2.

Lanes 3&4 non-carbamylated PLA2.

190

1 2 3 4

1 2 3 4

118

85

50

33

26

20

Alkaline urea gel

SDS-PAGEPr

otein

markers

Figure 7.10: Haemoglobin in alkaline urea gels. The upper gel strip is from

an alkaline urea gel. The lower gel is an SDS-PAGE gel (4-20% iGEL,

Gradipore).

191

0

20

40

60

80

100

Control Extraction withEthanolamine

Extraction withTris

Extraction withMilli-Q

Phos

phol

ipas

e A 2

act

ivity

, %

of c

ontr

ol

Figure 7.11: Extraction of phospholipase A2 enzyme activity from alkaline

urea gels. Results are expressed as the mean +/- std error as a percentage of

the control (no prior gel separation). The extraction buffers were

10mM ethanolamine, Milli-Q water, or 10mM Tris pH 6.8. The enzyme

sample employed was bee venom sPLA2.

192

02040

6080

100

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

Slice number

Phos

phol

ipas

e A 2

activ

ity

Figure 7.12: Tiger snake, crude venom, run in alkaline urea gel. Gel was sliced

into 1mm sections and protein was eluted into solution. Each section assayed for

PLA2 activity. The upper picture shows the gel stained for protein and the graph

below shows the PLA2 activity in each 1mm slice.

193

0

20

40

60

80

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Slice number

PLA 2

act

ivity

of

stan

dard

s

0

5

10

15

20

PLA 2

act

ivity

of

bow

el s

ampl

e

IB IIA III BC

Figure 7.13: Native urea gel of available PLA2 isoenzymes and bowel PLA2.

Type IB: purified porcine pancreatic phospholipase A2 (Sigma).

Type IIA: recombinant human phospholipase A2, (a kind donation from Dr.

Kieran Scott)

Type III: purified bee venom phospholipase A2 (Sapphire Biosciences).

194

0

2

4

6

Distance from well (mm)

Phos

phol

ipas

e A 2

acti

vity

(no

rmal

)

0

10

20

30

Phos

phol

ipas

e A 2

acti

vity

(i

nfar

cted

)

Normal bowel Infarcted bowel

Figure 7.14: The mobility of phospholipase A2 proteins from infarcted and

normal bowel in alkaline urea gels. Infarcted bowel content (precipitated

with ammonium sulphate at 60% saturation) vs normal bowel extract.

195

Heparin column:

0

1

2

3

4

5

1 5 9 13 17 21 25 29 33 37 41

Fraction

NaC

l con

cent

rati

on,

(M)

051015202530

PLA 2

act

ivit

y, %

hy

drol

ysis

NaCl PLA2

0

2

4

6

8

10

1 4 7 10 13 16 19 22 25 28

Distance into gel (mm)

PLA

2 ac

tivity

(%

hyd

roly

sis)

Alkaline urea gel:

Figure 7.15: A heparin column of crude bowel content eluted with increasing

NaCl. The fraction that exhibited the highest phospholipase A2 activity was

retained and run on an alkaline urea gel.

196

01234567

1 3 5 7 9 11 13 15 17 19 21 23 25 27

bow

el P

LA2 a

ctiv

ity

0

5

10

15

20

syno

vial

PLA

2 ac

tivity

Normal bowel Synovial fluid

Figure 7.16: PLA2 activity extracted from alkaline urea gels of

synovial fluid and supernatent from normal bowel.

197

0

0.2

0.4

0.6

0.8

1

0 0.25 0.5 0.75 1

NaCl (M)

mob

ility

Figure 7.17: Mobility of bee venom sPLA2 (10µg per lane) with increasing

concentrations of NaCl. Data represented as mean +/- standard deviation (error

bars <1%) of duplicate values. Mobility is represented as the distance that the

protein band ran from origin divided by the distance of bromophenol blue marker

dye from origin.

Discussion


Alkaline urea gels were used in this project to investigate whether the protein

responsible for the increased phospholipase activity in infarcted human bowel had

different characteristics from normal human bowel, and other well described sources

of phospholipase A2. These gels were more useful than SDS-PAGE gels in this type

of comparative analysis because the native activity of the protein was retained. The

process of extracting the enzyme and assaying it for phospholipase activity was far

more sensitive that protein gel staining, and therefore able to specifically localise the

protein of interest within the gel. These gels were devised by Ahmad and colleagues

(1994) for the analysis of isoforms of phospolipase A2 from snake venoms. The

connection between snake venom and intestinal proteins may appear questionable, but

others have noted the:

‘suprising similarities between the intestinal phospholipase A2 and some basic

venom proteins’ (Verheij et al, 1981).

These gels would be of use in many applications where the native activity of an

enzyme could be detected after electrophoresis. The mobility of proteins in these

gels is a function of size, shape and number of charged residues. This type of native

gel is also more useful than SDS-PAGE in the study of isoforms of similar sizes. For

example, a similar native gel (contained 8M urea and no SDS) was able to resolve

variants of lysyl oxidase from bovine aorta that were unable to be resolved by SDS-

PAGE (Williams & Kagan, 1985).

Alkaline urea gels have been criticised for a number of reasons including;

- the inability to assess protein size from these gels (Chettibi & Lawrence,

1989),

- the potential risk of carbamylation which may cause N-terminal blocking,

and therefore inability to N-terminally sequence proteins of interest,

- the potential risk of artifactual protein bands.

We considered that a number of extensions may be possible to address these issues

and increase the number of applications available for these gels.

198

Discussion

Transfer of proteins to membranes:

If proteins could be immobilised on a membrane, then they may be able to be N-

terminally sequenced by Edman chemistry or proteins of interest may be detected

with Western blotting. For this to be possible two issues had be addressed:

a) The transfer of small basic proteins (which are typical of phospholipase A2

isoforms and venom proteins) can be problematic

b) It was unknown whether proteins could be electro-transferred from urea gels.

The traditional transfer solution for proteins, the Tris Glycine buffer, (or Towbin

buffer, pH 8.3) was devised for the transfer of proteins up to a pI of approximately 7

(Towbin, Staehlin & Gordon, 1970). The efficiency of transfer of proteins varies

greatly, and it thought that the transfer buffer pH should be 1 to 2 pH units above the

isoelectric point of the protein of interest. The CAPS transfer buffer (pH 11) had

been devised for the transfer of basic proteins from SDS-PAGE gels to nitrocellulose

(Szewczyk & Kozloff, 1985).

The CAPS transfer buffer was found to transfer small basic proteins and crude venom

proteins from alkaline urea gels to PVDF more effectively than the Towbin buffer,

probably because of the high pH of the CAPS buffer. The proteins transferred to the

membrane could be successfully sequenced by N-terminal amino acid determination

by Edman chemistry (Edman & Begg, 1967). This is a positive finding because it not

only extends the applications available for these gels but suggests that N-terminal

blocking of proteins by carbamylation is minimal. The use of the membranes in

Western blotting would be an interesting extension for these gels, but was not within

the scope of this project.

Carbamylation:

The risk of carbamylation increases with increasing temperature, time and urea

breakdown, therefore efforts were taken to reduce each of these factors. Firstly, as

recommended by Ahmed & Lawrence (1994), high quality urea was used and gels

were run for two hours before the addition of the protein samples. These steps were

likely to reduce the risk of urea breakdown and to remove any breakdown products if

they were present. Secondly we chose to run the gels in an ice bath to reduce the

buffer temperature. Additionally, buffers could be pre-chilled. The acrylamide

concentration and pH of the stacking gel could both be reduced which appeared to

199

Discussion

increase the mobility of the proteins in the gel. These modifications had two results:

firstly the increased protein mobility reduced the running time and secondly the length

of the gel was used more efficiently. Bee venom phospholipase A2 contains 28

potential carbamylation sites. Bee venom PLA2 was carbamylated for 48 hours at

37oC, and was then run on an alkaline urea gel and SDS-PAGE. Carbamylation adds

only 43 atomic mass units, and this small difference could not be detected in an SDS

gel, as expected. Carbamylation results in the loss of positive charges from a protein,

and therefore mobility in alkaline urea gels should theoretically be increased. A

slight increase in mobility was suggested in the alkaline urea gels.

Native gels are excellent tools for detecting protein aggregates, unfolded

conformations and binding events (Alliance Protein Laboratories, 2003). A number

of haemoglobin forms were evident in the alkaline urea gels, and the nature of each

could be estimated by determining the size of the aggregate by SDS-PAGE.

The size of the proteins of interest could be estimated by excising protein bands from

the alkaline urea gel and re-running in an SDS-PAGE gel. The protein bands were

excised easily and cleanly from alkaline urea gels stained with either Coomassie R-

250 or Coomassie G-250 based stains. The gel pieces could be stored in the freezer,

and freezing aided in the manipulation into SDS-PAGE gels. The proteins could be

reduced with either a β-mercaptoethanol/equilibration technique (Davie, 1982) or a

DTT-boiling technique (that we developed to avoid the more toxic and odorous β

mercatoethanol). The reduction step did not appear to improve the accuracy of the

size determination, but may have resulted in protein losses. The transfer of protein

from alkaline urea gels to SDS gels was advantageous in two ways, firstly the size of

proteins of interest could be estimated and the process added an extra dimension of

resolution.

The extraction efficiency of proteins from these gels was reported to be approximately

60% (Ahmad & Lawrence, 1993). We found that the extraction efficiency was

improved by lowering the pH of the extraction buffer. Active PLA2 could be

extracted from crude tiger snake venom. These gels were used to examine a number

of different purified isoforms of PLA2 that were available to our group. The proteins

200

Discussion

with phospholipase activity from normal and infarcted human bowel were compared,

and the mobility of each of these proteins was quite distinct. Not surprisingly, the

phospholipase from normal bowel appeared to have similar mobility to the

phospholipase in synovial fluid from patient with rheumatoid arthritis. Normal gut

(Cupillard et al, 1997, Nevalainen, Gronroos & Kallajoki, 1995) and rheumatoid

arthritis (Seilhamer et al, 1989) are both known to be rich sources of type IIA

phospholipase A2. The infarcted gut protein with phospholipase activity had high

mobility in alkaline urea gels, and partially purified protein showed similar mobility.

This confirmed that the unusual protein was being followed successfully by the

protein purification process. The isoelectric point of a protein is dependent on the

environment (pH, ionic strength etc), so it was possible that different mobility in

alkaline urea gels was not due to protein differences but the ionic strength or pH of

the protein solution. To discount this possibility a series of protein solutions were

prepared containing 0 to 1M sodium chloride. Even under conditions of 1M NaCl,

the protein maintained the same mobility. This was a reassuring finding because

different protein mobility strongly suggests different protein composition. The ions

present in protein samples are likely to have high mobility under the running

conditions and are likely to be lost quickly.

201

Conclusions and Future Directions.

Chapter 3, Plasma Proteomics:

The proteomic investigation undertaken as part of this project identified a number of

proteins that appeared to be increased in the plasma of patients with confirmed bowel

infarction compared with control patients. The aim of the investigation was to assess

whether a tissue damage marker specific to bowel was present in high abundance in

plasma. The proteins that appeared to be the most differentially expressed were

variants of haptoglobin (in the region of pH 4-7) and serum amyloid A (in the region

pH 7-10). These two proteins are well recognised as being disease associated

proteins and have been shown to be present in high concentrations in inflammatory

diseases, infection and trauma. The elevated levels of these two proteins in disease

suggest that they may have a role in host defence, tissue repair and inflammation.

The existence of several highly homologous variants of serum amyloid A and lack of

variant specific assays makes the study of this protein difficult. The commercially

available serum amyloid A ELISA (TriDelta) detects the total acute phase serum

amyloid A (SAA 1 & 2 subtypes), with a good limit of detection and reproducibility.

However, if subtle patterns of SAA subtype levels exist that are indicative of a

particular disease, an ELISA such as this would not detect these patterns. At present

there are no serum amyloid A subtype specific antibodies available. It may be

interesting to probe two dimensional gel electrophoresis membranes (pH 3-10) for

SAA variants using Western blotting to assess protein changes due to disease states.

At present, the assay of haptoglobin and serum amyloid A for diagnostic purposes is

not a clinically attractive prospect because of their likely non-specific expression in

response to inflammation, injury or infection. There may be haptoglobin or SAA

subtype specific increases that are indicative of bowel infarction, but these cannot be

studied without subtype specific techniques such as subtype specific ELISA.

202

A proteomic approach for diagnostic marker identification is useful, but a number of

modifications to the techniques such as those used in this project are required to

improve the chance of finding a useful diagnostic protein. These modifications

include pre-fractionation of samples prior to two dimensional electrophoresis, or an

alternate technique such as the isotope coded affinity tag (ICAT) system.

Firstly, pre-fractionation of plasma (for example, by preparative isoelectric focussing

or sub-proteomics) would allow the generation of a set of sub samples. The sub

samples could each be run in a two dimensional electrophoresis gel system. Pre-

fractionation is likely to allow the identification of lower abundance proteins, which

are more likely to be disease associated proteins or markers of specific tissue damage.

Secondly, an approach such as the ICAT (isotope coded affinity tag) system may

provide very interesting insights into protein differences between plasma from bowel

infarction and control patients. The ICAT technique was developed to achieve

quantitative protein profiling to examine differences between the proteins from a

matched pair of samples. Briefly, this technique involves the labelling of the cysteine

residues in sample ‘A’ with a ‘light’ (hydrogen, d0) biotin linked reagent and those in

sample ‘B’ with a ‘heavy’ (deuterium, d8) biotin linked reagent. Samples A and B

are then mixed, trypsin digested and the ICAT peptides isolated by avidin affinity

chromatography (Gygi & Aebersold, 2000). The peptides are then identified by LC-

MS/MS and quantified by the ratio of the ICAT labelled peptide pairs. The

preliminary ICAT experiments focussed on the protein differences between yeast

grown with a variety of carbon sources (Gygi et al, 1999). Recent ICAT publications

have reflected the diversity of potential applications (cortical neurons, wheat and

membrane proteins).

The advantages of the ICAT technique include the highly specific and robust labelling

process and the ability to use a wide variety of protein sources and quantities. Other

separation techniques could be added to the ICAT procedure to identify lower

abundance proteins (Gygi, Rist & Aebersold, 2000). The disadvantages of the ICAT

technique include some difficulty in database searching because of the modification

of peptides by the tag, and some peptides are missed because they contain no cysteine

residues (~8% of proteins in yeast contain no cysteine residues, (Gygi, Rist &

203

Aebersold, 2000)). In this type of project, the ICAT approach would be

advantageous because large amounts of protein could be analysed which would

increase the likelihood of identifying a low abundance, disease associated protein.

The ICAT technique also allows the identification of minor protein differences

between two states, in this case bowel ischaemia/infarction and control plasma.

In a different issue, proteomics may be a valuable tool to identify protein changes

associated with the development of remote organ injury that may be seen as a

consequence of bowel injury, either in blood or in mesenteric lymph. It would be of

great interest to study protein factors in mesenteric lymph that may be:

- mediators of remote organ injury

- potential therapeutic targets for remote organ injury

- expressed at different time points during the development of remote organ

injury

-involved in the induction of related processes such as increased gut

permeability, neutrophil priming and intestinal mucosal damage.

The study of lymph would be potentially less difficult than plasma, as lymph has only

half the total protein of plasma (Anderson & Anderson, 2002). Previous research

into the potential mediators of tissue/cellular damage carried in the mesenteric lymph

has generally been conducted in rodent models of hemorrhagic shock. Bacteria and

endotoxin are unlikely to be important factors, as these could not be detected in the

post shock lymph (Deitch et al, 2001, Magnotti et al, 1998). The potential

importance of a neutral lipid (rather than protein or carbohydrate) factor was

highlighted by another study (Gonzalez et al, 2001). It would be of great interest to

identify the nature and source of this lipid factor, and the enzyme(s) responsible for its

production.

204


Chapter 4, Phospholipase Activity in Mesenteric

Ischaemia/Infarction:

Increased mucosal phospholipase A2 activity associated with gut ischaemia

reperfusion injury had been previously reported in rat models (Koike et al, 1992,

Koike et al, 1995, Otamiri et al, 1987, Otamiri, & Tagesson, 1989). The involvement

of phospholipase A2 in the pathogenesis of ischaemia/reperfusion injury had also been

proposed by several groups, based on the protective qualities of phospholipase

inhibitors (Koike et al, 1992, Koike et al, 1995, Otamiri et al, 1987, Otamiri, Lindahl

& Tagesson, 1988). Many groups had identified the presence of the isoform of

phospholipase A2 type II, in normal gut tissue using a variety of techniques

(immunohistochemistry, in situ hybridisation, PCR), but few studies have investigated

PLA2 isoforms in diseased tissue. We were in the unusual but fortunate position to

have access to human bowel tissue from both bowel infarction and control patients.

In this project, early experiments suggested that while type IIA phospholipase A2 was

present in injured human bowel, this isoform was not responsible for all of the

phospholipase activity in these samples. Phospholipase activity was significantly

higher in the lumen of infarcted human bowel than in the corresponding bowel tissue.

This finding suggests a possible route for the protein to enter general circulation. It

would be a valuable future direction to examine the synthesis and/or release of this

protein in response to the development of the ischaemic injury. The possibility that

the protein is simply being leaked from dying cells needs to be eliminated. This

could be achieved by measuring the rate/percentage of cell lysis by assaying

cytoplasmic proteins or cell membrane integrity. An in depth investigation of the

hydrolysis products of phospholipase activity and their effects during

ischaemia/reperfusion injury and remote organ injury would also be valuable.

It would also be interesting to examine the phospholipase enzyme activity in matched

human portal and systemic circulation, ischaemic and infarcted tissue, to examine the

205

relative enzyme activity at each stage of the possible release into circulation. It may

also be a valuable exercise to use a labelled protein to assess intestinal permeability

changes in response to ischaemic injury, perhaps at the same time as the examination

of phospholipase activity. In a rat model of ischaemia/reperfusion, phospholipase

activity was ten fold higher in portal than systemic circulation, and this suggested that

the circulating phospholipase activity arose from the injured gut (Koike et al, 2000).

Our results showed that phospholipase activity was significantly higher in infarcted

than matched ischaemic tissue. Although this was only a sample size of one, this

result may be of significance if a spectrum of phospholipase activity reflects a

spectrum of tissue damage. In turn this may be of relevance in the development of a

diagnostic test, as the level of the protein in circulation may reflect the severity of the

condition. In future investigations, many more matched tissue samples would assist

in answering these questions. Many potential diagnostic markers of bowel infarction

have not found clinical utility because they either:

- decrease before the condition has been rectified (Bounous et al, 1984,

Jamieson et al, 1982).

- continue to increase even after the condition is rectified or

- cannot distinguish between ischaemia and infarction (Gearhart et al, 2003).

A suitable animal model could assist in examining the correlation between the degree

of tissue damage and protein response. However, there are significant species

differences in the distribution of phospholipase A2 isoforms making an animal model

unreliable, however this may not be an issue with other potential markers of bowel

injury. A time course study of protein changes at the various stages of the disease

development would also be a valuable future direction, and an animal model would be

essential for this investigation.

206


Chapter 5, Protein Purification:

The protein purification experiments in this project were undertaken to identify the

protein with phospholipase activity that was increased as a result of infarction.

Immunodepletion results suggested that the well described phospholipase A2 isoforms

of type IIA and V did not account for all of the phospholipase activity in infarcted

human bowel. The protein purification techniques used in this project were chosen

because of their ability to exploit an unusual characteristic of the protein of interest,

and in doing so, remove large amounts of contaminating protein at each step. In

addition, each purification technique used in this project retained the native enzyme

activity of the protein of interest, and therefore the protein could be tracked at each

stage. A number of preliminary purification attempts identified the purified protein

as haemoglobin. It was quickly confirmed that haemoglobin does not possess

phospholipase activity, nor does it enhance the activity of low levels of

phospholipase. It was concluded that haemoglobin was likely to have been co-

purifying with the protein of interest. To overcome this problem, a greater number of

column volumes were employed with the heparin affinity chromatography column.

The purification system was also scaled up approximately twenty fold, and this is

likely to have assisted in targeting the protein of interest for identification. The

scaled up purification with more stringent washing steps allowed the isolation of

several highly purified fractions, each with sufficient protein content for identification

by mass spectrometry.

The protein purification techniques used in this project (heparin affinity

chromatography, preparative isoelectric focussing with the Rotofor® apparatus) were

very useful for this application. Heparin affinity chromatography can be used to

easily semi purify a protein with heparin binding ability, whilst generally retaining the

native activity of the protein. It would be of interest to automate this system, which

may result in higher reproducibility and larger quantity purifications while decreasing

the currently highly labour intensive nature of this technique. It would be of interest

207

to optimise the protein desalting and concentration steps that are required at the

interface of the heparin column and the next technique (in this case, preparative

isoelectric focussing). The optimal desalting/concentration step would remove all of

the interfering salts or other contaminants and the majority of the water without any

protein alteration or loss. The dialysis step used in this project was time consuming

and did not concentrate the protein which resulted in the need for several Rotofor®

runs of dilute solutions, instead of one run of a pooled and concentrated sample.

The final identification of the protein of interest as cyclophilin B was made by nano

LC-MS/MS. An unusual aspect of the identification of cyclophilin B was that the

purification progress had been monitored at every stage by phospholipid hydrolysis

assays, but cyclophilin B was not known to possess phospholipase activity. A set of

experiments were undertaken to assess the possibility that cyclophilin B possessed

phospholipid hydrolysing ability. Immunodepletion of cyclophilin B from the highly

purified final fractions removed all of the phospholipase activity. This indirect

evidence was interesting and supported the idea that cyclophilin B had phospholipase

activity, but the direct investigation would have been preferable. However, an

application to the only known supplier of human cyclophilin B protein was

unsuccessful. In the future it would be of interest to produce recombinant human

cyclophilin B to examine the enzyme characteristics in detail, and to investigate its

pro-inflammatory properties. Recombinant human cyclophilin B has been produced

previously (Bergsma et al, 1991, Price et al, 1991, Spik et al, 1991) and these

methods would form the basis of this future direction.

208


Chapter 6, Cyclophilin B:

Cyclophilin B had been reported to be a ubiquitously expressed protein as detected by

northern blotting of mouse tissue panels (Hasel et al, 1991). Cyclophilin B had been

observed previously in rodent intestinal tissue at the mRNA level (Hasel et al, 1991,

Kainer & Doris, 2000). Very little research has focussed on cyclophilin B at the

protein level in tissues, but some work has detected cyclophilin B protein in cell

culture (Bergsma et al, 1991).

In this project, western blotting of cyclophilin B was performed using a commercially

available polyclonal antibody. Immunoreactive cyclophilin B protein was detected in

all bowel tissues investigated, from both control and bowel infarction patients.

Immunohistochemistry results supported the western blotting results, and showed

widespread distribution of cyclophilin B in many cell types in both normal and

infarcted tissue sections. Future directions in this area include the investigation of

cyclophilin B protein in a variety of normal and diseased human tissues via

immunohistochemistry, and the corresponding plasma by ELISA.

Another future direction for this project may include the use of recombinant human

cyclophilin B in the development of a sensitive and specific ELISA that is suitable for

clinical samples. To our knowledge, two cyclophilin B ELISAs have been

developed. These two ELISA protocols used the same cyclophilin B peptide

generated polyclonal antibody and constructed a standard curve using recombinant

cyclophilin B protein.

Allain et al, 1995, published a cyclophilin B ELISA with the following format:

-anti-cyclophilin B coated plates

-anti-cyclophilin B peroxidase conjugate

-O-phenylenediamine substrate

209

The limit of detection of this ELISA was reported to be 10ng/mL and the average

normal plasma level of cyclophilin B was found to be approximately 150ng/mL.

Denys et al, 1998, published a modification of the above technique with the following

format:

-anti-cyclophilin B coated plates

-anti-cyclophilin B biotin conjugate

-avidin-peroxidase conjugate

-O-phenylenediamine substrate

This ELISA was reported to have better sensitivity than the above method, because of

the likely biotin signal amplification effect. This system would be a valuable starting

point for the development of an ELISA in this project. Average normal plasma levels

were reported to be approximately 87ng/mL.

A sensitive and specific ELISA for cyclophilin B would allow the large scale

screening of patients samples and this may lead to a better understanding of the

correlation between increased cyclophilin B levels and disease states. In this project,

plasma cyclophilin B was not detected in any control patients, but several samples of

plasma from patients with confirmed bowel infarction contained cyclophilin B (via

Western blotting). An ELISA with a very low level of detection may be able to

establish typical ranges for plasma levels of cyclophilin B in bowel infarction, healthy

volunteers and control ICU patients. At present, this is not an option because

recombinant protein could not be obtained, and therefore a standard curve could not

be constructed.

Another future direction would be the development of novel cyclophilin B inhibitors

that blocked the inflammatory activity of this protein without altering the distribution

of cyclosporin A in blood or extracellular fluid or initiating the expression of more

cyclophilin B. It would also be of great interest to examine the potential role of

cyclophilin B as a mediator of inflammation or tissue injury, perhaps with an

emphasis on bowel ischaemia or infarction. Furthermore, an investigation into the

protective roles of cyclophilin B inhibitors in inflammatory conditions, oxidative

stress and ischemia/reperfusion injury could be extended to include bowel conditions

such as bowel ischaemia/infarction.

210


Chapter 7, Alkaline Urea PAGE:

The alkaline urea gel system of Ahmed, Lawrence & Moores (1994) was found to be

useful in this project, to examine differences in protein characteristics between a

control and disease state. This application was more suited to an alkaline urea gel

system than SDS PAGE because:

-the protein of interest was present in very low abundance and could not be

detected by protein staining, but could be detected by extracting enzyme

activity.

-the proteins from normal and healthy tissue were of similar size, and

therefore unable to be resolved by SDS PAGE, but as mobility in alkaline urea

gels is due to a number of variables, this technique could resolve these

proteins.

These gels were modified to allow greater mobility of proteins, and therefore greater

use of the entire length of the gel. The modified alkaline urea gel showed that the

proteins with phospholipase activity from normal human bowel and rheumatoid

arthritis synovial fluid (both well described and abundant sources of type II PLA2)

were similar and focussed in a region close to the site of application. In contrast, the

protein with phospholipase activity from infarcted human bowel lumen content had

distinctly different, and higher mobility. It would be of interest to use these gels to

examine mammalian phospholipase from normal and disease state samples.

A number of limitations were associated with these gels and attempts were made to

overcome these problems in order to extend the applications available. We have

proposed a method to estimate protein size of the proteins of interest from alkaline

urea gels. This technique resulted in a reasonable estimation of protein size which

may give information about a protein of interest and may allow a suitable purification

process to be planned (for example the use of gel filtration chromatography or

211

preparative PAGE). An additional, but unexpected, advantage of this technique was

the extra dimension of resolution provided by this step. For example, one protein

band observed in an alkaline urea gel of a crude snake venom resolved into two

protein spots when run in an SDS PAGE gel. This may be advantageous in very

complex protein samples, in a similar fashion to two dimensional gel electrophoresis.

Future directions for this area include an investigation of the suitability of western

blotting as an extension for these gels. Proteins separated by alkaline urea gels can

be successfully transferred to PVDF membrane, and it may be possible to probe these

membranes for immunoreactive proteins including venom toxins. This approach may

be useful for the identification of isoforms of proteins of interest that may share

similar size and therefore may be unable to be separated by SDS PAGE based western

blotting.

212

Bibliography

1. Aarsman A.J., de Jong J.G.N., Arnoldussen E., Neys F.W., van Wassenaar P.D., and Van den

Bosch H. 1989. Immunoaffinity Purification, Partial Sequence, and Subcellular Localization of Rat

Liver Phospholipase A2. J. Biol. Chem. 264:10008-10014.

2. Abraham E., Naum C., Bandi V., Gervich D., Lowry S.F., Wunderink R., and et al. 2003.

Efficacy and safety of LY315920Na/S-5920, a selective inhibitor of 14-kDa group IIA secretory

phospholipase A2, in patients with suspected sepsis and organ failure. Crit. Care Med. 31:718-728.

3. Abu Zidan F. M., Winterbourn C. C., Bonham M. J., Simovic M. O., Buss H., and Windsor J.

A. 1999. Small bowel ischaemia-reperfusion increases plasma concentrations of oxidised proteins in

rats. Eur. J. Surg. 165:383-389.

4. Adam B.L., Vlahou A., Semmes O.J., and Wright G.L. 2001. Proteomic approaches to

biomarker discovery in prostate and bladder cancers. Proteomics 1:1264-1270.

5. Ahmad T., and Lawrence A.J. 1993. Purification and activation of Phospholipase A2 isoforms

from Naja mossambica mossambica (Spitting Cobra) Venom. Toxicon 31:1279-1291.

6. Ahmad T., Lawrence A. J., and Moores G. 1994. High-resolution two-part basic urea gels for

analysis of venom phospholipase A2 isoforms. Toxicon 32:1627-1639.

7. Alaiya A.A., Poppermann M., Langridge J., Roblick U., Brandstedt S., Hellstrom M., Linder

S., Bergman T., Jornvall H., and Auer G. 2001. Identification of proteins in human prostate tumor

material by two-dimensional gel electrophoresis and mass spectrometry. Cell Mol. Life Sci. 58:307-

311.

8. Allain F., Boutillon C., Mariller C., and Spik G. 1995. Selective Assay for CyPA and CyPB in

human blood using highly specific anti-peptide antibodies. J. Immunol. Meth. 178:113-120.

9. Allain F., Vanpouille C., Carpentier M., Slomianny M.C., Durieux S., and Spik G. 2002.

Interaction with glycosaminoglycans is required for cyclophilin B to trigger integrin-mediated

adhesion of peripheral blood T lymphocytes to extracellular matrix. Proc. Natl. Acad. Sci. USA

99:2714-2719.

10. Alliance Protein Laboratory. 2003. Native gels.http://www.ap-lab.com/native_gels.htm

11. Alsemgeest S.P.M., Horadagoda A., Hulskamp C.k., Tooten P.C.J., Kim D.H., Niewold

T.H.A., and Gruys E. 1995. First Evidence for the Existence of Multiple Isoforms of Bovine Serum

Amyloid A (apo-SAA). Scand. J. Immunol. 41:407-413.

12. Anderson N.L., and Anderson N.G. 2002. The Human Plasma Proteome. Mol. Cell

Proteomics 1:845-867.

13. Anderson S.K., Gallinger S., Roder J., Frey J., Young H.A., and Ortaldo J.R. 1993. A

cyclophilin related protein involved in the function of natural killer cells. Proc. Natl. Acad. Sci. USA

90:542-546.

14. Aragona., De Caro., Parenti., Artibani., Bassi., Munari., and Pagano. 1998. Structural and

ultrastructural changes in ileal neobladder mucosa: a 7-year follow up. Br. J. Urol. 81:55.

213

Bibliography

15. Arbibe L., Vial D., and Touqui L. 1997. Phospholipase A2 and Acute Respiratory Distress

Syndrome. In Phospholipase A2: Basic and Clinical Aspects in Inflammatory Disease. Uhl W.,

Nevalainen T. J., and Buchler M., editors. Basel: Karger.

16. Arcuni J., Wang, L., Yousef, K., Chiu, S., Mikkelson, K., Franson, R.D., and Sonnino R. E.

1999. Secretory event in intestinal grafts during preservation ischemia. J. Surg. Res 84:233-239.

17. Arni R.K., and Ward R.J. 1996. Phospholipase A2- A Structural Review. Toxicon 34:827-841.

18. Aydin M., Guler O., Ugras S., Bakir B., and Sekeroglu R. 1998. Blood and tissue findings in

the diagnosis of mesenteric ischaemia: an experimental study. Clin. Chem. Lab. Med. 36:93-98.

19. Banks R.E., Dunn M.J., Hochstrasser D.F., Sanchez J.C., Blackstock W., Pappin D.J., and

Selby P.J. 2000. Proteomics: new perspectives, new biomedical opportunities. Lancet 356:1749-1756.

20. Barnett S.M., Davidson E.D., and Bradley E.L. 1976. Intestinal Alkaline Phosphatase and

Base Deficit in Mesenteric Occlusion. J. Surg. Res 20:243-246.

21. Baumann G., and Chrambach A. 1975. Quantitative Removal of Carrier Ampholytes from

Protein Fractions Derived from Isoelectric Focussing. Anal. Biochem. 69:649-651.

22. Beach C.M., de Beer M.C., Sipe J.D., Loose L.D., and de Beer F.C. 1992. Human Serum

Amyloid A Protein. Complete amino acid sequence of a new variant. Biochem. J. 282:615-620.

23. Bellamy M.C., Lansbury A., and Murdoch S.D. 2002. Inflammatory Serum Markers in Organ

Dysfunction. In Sepsis and Multiple Organ Dysfunction. Deitch E. A., Vincent J.L., and Windsor A.,

editors. London: W.B. Saunders. 187-196.

24. Bergsma D.J., Elder C., Gross M., Kersten H., Sylvester D., Appelbaum E., Cusimano D.,

Livi G.P., McLaughlin M.M., Kasyan K., et al. 1991. The Cyclophilin Multigene Family of Peptidyl-

Prolyl Isomerases, Characterization of Three Separate Human Isoforms. J. Biol. Chem. 266:23204-

23214.

25. Berkowitz D.B., and Webert D.W. 1981. The Inactivation of Horseradish Peroxidase by a

Polystyrene Surface. J. Immunol. Meth. 47:121-124.

26. Berliner J.A., Navab M., Fogelman A.M., Frank J.S., Demer L.L., Edwards P.A., Watson

A.D., and Lusis A.J. 1995. Atherosclerosis: Basic Mechanisms. Oxidation, Inflammation and

Genetics. Circulation 91:2488-2496.

27. Betts L.C., Edbrooke M.R., Thakker R.V., and Woo P. 1991. The Human Acute Phase Serum

Amyloid A gene Family: Structure, Evolution and Expression in Hepatoma Cells. Scand. J. Immunol.

34:471-482.

28. Biffl W.L., Moore E.E., and Moore F.A. 1995. Gut derived mediators of multiple organ

failure: platelet activating factor and interlukin-6. British J. Hosp. Med. 54:134-138.

29. Billich A., Winkler G., Ashauer H., Rot A., and Peichl P. 1997. Presence of Cyclophilin A in

Synovial Fluids of patients with Rheumatoid Arthritis. J. Exp. Med. 185:975-980.

214

Bibliography

30. Bini L., Magi B., Marzocchi B., Cellesi C., Berti B., Raggiaschi R., Rossolini A., and Pallini

V. 1996. Two-dimensional electrophoretic patterns of acute-phase human serum proteins in the course

of bacterial and viral diseases. Electrophoresis 17:612-616.

31. Bloomster T.G., and Watson D.W. 1983. Effects of Carrier Ampholyte Contamination on the

Biological and Biochemical Properties of Streptococcal Pyrogenic Endotoxin Type C. Infect. Immun.

39:311-314.

32. Bochelen D., Rudin M., and Sauter A. 1999. Calcineurin Inhibitors FK506 and SDZ ASM 981

Alleviate the Outcome of Focal Cerebral Ischemic/Reperfusion Injury. J. Pharmacol. Exp. Ther.

288:653-659.

33. Boley S.J., Sprayregan S., Siegelman S.S., and Veith F.J. 1977. Initial results from an

aggressive roentgenological and surgical approach to acute mesenteric ischemia. Surgery 82:848-855.

34. Boley S.J., Brandt L.J., and Sammartano R. J. 1997. History of mesenteric ischemia. Surg.

Clin. North Am. 77:275-288.

35. Bomalaski J.S., and Clark M.A. 1993. Phospholipase A2 and Arthritis. Arthritis Rheum.

36:190-198.

36. Bounous G., Echave V., Vobecky S.J., Navert H., and Wollin A. 1984. Acute Necrosis of the

Intestinal Mucosa with High Serum Levels of Diamine Oxidase. Dig. Dis. Sci. 29:872-874.

37. Bradbury A. W., Brittenden J., McBride K., and Ruckley C. V. 1995. Mesenteric ischaemia: a

multidisciplinary approach. Br. J. Surg. 82:1446-1459.

38. Bradford M.M. 1976. A Rapid and Sensitive Method for the Quantitation of Microgram

Quantities of protein Utilizing the Principle of Protein-Dye binding. Anal. Biochem. 72:248-254.

39. Braga M., and Gianotti L. 2002. Nutritional Support- Current and Future. In Sepsis and

Multiple Organ Dysfunction. Deitch E. A., Vincent J.L., and Windsor A., editors. London: W.B.

Saunders. 262-270.

40. Brown R.E., Jarvis K.L., and Hyland K.J. 1989. Protein Measurement Using Bicinchoninic

Acid: Elimination of Interfering Substances. Anal. Biochem. 180:136-139.

41. Bukrinsky M.I. 2002. Cyclophilins: unexpected messengers in intercellular communications.

Trends Immunol. 23:323-325.

42. Bulkley G.B. 1987. Free radical-mediated reperfusion injury: A selective review. Br. J.

Cancer 55:66-73.

43. Buxbaum J.N., and Tagoe C.E. 2000. The genetics of the Amyloidoses. Annu. Rev. Med.

51:543-569.

44. Calman C., Hershey F.B., Skaggs J.O., and Spencer A. 1958. Serum Lactic Dehydrogenase in

the Diagnosis of the Acute Surgical Abdomen. Surgery 44:43-52.

45. Carpentier M., Allain F., Haendler B., Denys A., Mariller C., Benaissa M., and Spik G. 1999.

Two distinct Regions of Cyclophilin B are Involved in the Recognition of a Functional Receptor and

of Glycosaminoglycans on T-Lymphocytes. J. Biol. Chem. 274:10990-10998.

215

Bibliography

46. Carter D.C., and Camilleri M. 1997. Intestinal ischemia and Vasculitis. In Diseases of the

Gastrointestinal Tract and Liver. Shearman D.J.C., Finlayson N.D.C., and Camilleri M., editors.

Edinburgh: Churchill Livingstone. 543-547.

47. Celis J.E., and Gromov P. 1999. 2D protein electrophoresis: can it be perfected? Curr.

Opinion Biotech. 10:16-21.

48. Chang H.W., Kudo I., Hara S., Karasawa K., and Inoue K. 1986. Extracellular Phospholipase

A2 Activity in Peritoneal Cavity of Casein Treated rats. J. Biochem. 100:1099-1101.

49. Chang H.W., Kudo I., Tomita M., and Inoue K. 1987. Purification and Characterization of

Extracellular Phospholipase A2 from Peritoneal Cavity of Caseinate-Treated rats. J. Biochem.

102:147-154.

50. Chen J. (a), Engle S. J., Seilhamer J. J., and Tischfield J. A. 1994. Cloning and

characterization of novel rat and mouse low molecular weight Ca2+-dependent phospholipase A2s

containing 16 cysteines. J. Biol. Chem. 269:23018-23024.

51. Chen J. (b), Engle S. J., Seilhamer J. J., and Tischfield J. A. 1994. Cloning and recombinant

expression of a novel human low molecular weight Ca2+-dependent phospholipase A2. J. Biol. Chem.

269:2365-2368.

52. Chen J.W., Dodia C., Feinstein S.I., Jain M.K., and Fisher A.B. 2000. 1-Cys Peroxiredoxin, a

Bifunctional Enzyme with Glutathione Peroxidase and Phospholipase A2 Activities. J. Biol. Chem.

275:28421-28427.

53. Chettibi S., and Lawrence A. 1989. High resolution of honey bee (Apis meliffera) venom

peptides by proprionic acid/urea polyacrylamide gel electrophoresis after ethanol precipitation.

Toxicon 27:781-787.

54. Chiu C.J., McArdle A.H., Brown R., Scott H.J., and Gurd F.N. 1970. Intestinal Mucosal

Lesion in Low Flow States. Arch. Surg. 101:478-483.

55. Cid M.C., Grant D.S., Hoffman G.S., Auerbach R., Fauci A.S., and Kleinman H.K. 1993.

Identification of Haptoglobin as an Angiogenic Factor in Sera from Patients with Systemic Vasculitis.

J. Clin. Invest. 91:977-985.

56. Clark J.D., Milona N., and Knopf J.L. 1990. Purification of a 110-kilodalton cytosolic

phospholipase A2 from the human monocytic cell line U937. Proc. Natl. Acad. Sci. USA 87:7708-

7712.

57. Clavien P.A. 1990. Diagnosis and management of mesenteric infarction. Br. J. Surg. 77:601-

603.

58. Cokkinis A.J. 1926. Mesenteric Vascular Occlusion. London: Bailliere, Tindall and Cox. 1-

93. pp.

59. Corder A.P., and Taylor I. 1993. Acute mesenteric ischaemia. Postgrad. Med. J. 69:1-3.

216

Bibliography

60. Cordwell S.J., Nouwens A.S., Verrills N.M., Basseal D.J., and Walsh B.J. 2000.

Subproteomics based upon protein cellular location and relative solubilities in conjunction with

composite two-dimensional electrophoresis gels. Electrophoresis 21:1094-1103.

61. Corke C., and Glenister K. 2001. Monitoring intestinal ischaemia. Crit. Care Resuscitation.

3:176-180.

62. Corthals G.L., Gygi S.P., Aebersold R., and Patterson S.D. 1999. Identification of proteins by

mass spectrometry. In Proteome research: 2D gel electrophoresis and detection methods. Rabilloud

T., editor. New York.: Springer. 197-231.

63. Corthals G.L., and Nelson P.S. 2001. Large-scale proteomics and its future impact on

medicine. The Pharmacogenetics Journal. 1:15-22.

64. Crowl R. M., Stoller T. J., Conroy R. R., and Stoner C. R. 1991. Induction of phospholipase

A2 gene expression in human hepatoma cells by mediators of the acute phase response. J. Biol. Chem.

266:2647-2651.

65. Cupillard L., Koumanov K., Mattei M.G., Lazdunski M., and Lambeau G. 1997. Cloning,

Chromosomal Mapping, and Expression of a Novel Human Secretory Phospholipase A2. J. Biol.

Chem. 272:15745-15752.

66. Davie J.R. 1982. Two-Dimensional Gel Systems for Rapid Histone Analysis for Use in

Minislab Polyacrylamide Gel Electrophoresis. Anal. Biochem. 120:276-281.

67. de Beer M.C., Kindy M.S., Lane W.S., and de Beer F.C. 1994. Mouse Serum Amyloid A

Protein (SAA5) Structure and Expression. J. Biol. Chem. 269:4661-4667.

68. De Ceuninck F., Allain F., Caliez A., Spik G., and Vanhoutte P.M. 2003. High Binding

Capacity of Cyclophilin B to Chrondrocyte Heparan Sulphate Proteoglycans and Its Release From the

Cell Surface by Matrix Metalloproteinases. Arthritis Rheum. 48:2197-2206.

69. De Toma G., Marzano D., Salvatore P., Cerza F., De Cesare E., Giacovazzo M., Martelletti P.,

and Antonucci M. 1983. Enzymatic and metabolic changes in peripheral serum after superior

mesenteric artery ligation in dogs. Ital. J. Surg. Sci. 13:269-273.

70. de Villiers W.J.S., Varilek G.W., de Beer F.C., Guo J., and Kindy M.S. 2000. Increased

Serum Amyloid A levels reflect Colitis Severity and precede Amyloid Formation in IL-2 Knockout

mice. Cytokine 12:1337-1347.

71. Deitch E. A. 1992. Multiple Organ Failure. Ann. Surg. 216:117-134.

72. Deitch E.A. 1993. Nutrition and the gut mucosal barrier. In Current Opin. Gen. Surg. Daly J.,

editor. Philadelphia: Current Science. 85-91.

73. Deitch E.A., Adams C., Li Q., and Xu D.Z. 2001. A time course study of the protective effect

of mesenteric lymph duct ligation on hemorrhagic shock-induced pulmonary injury and the toxic

effects of lymph from shocked rats on endothelial cell monolayer permeability. Surgery 129:39-47.

217

Bibliography

74. Deitch E.A., and Sambal J.T. 2002. The gut-origin hypothesis of MODS. In Sepsis and

Multiple Organ Dysfunction. Deitch E. A., Vincent J.L., and Windsor A., editors. London: W.B.

Saunders. 105-114.

75. Delaney C. P., O'Neill S., Manning F., Fitzpatrick J. M., and Gorey T. F. 1999. Plasma

concentrations of glutathione S-transferase isoenzyme are raised in patients with intestinal ischaemia.

Br. J. Surg. 86:1349-1353.

76. Dennis E. A. 1997. The growing phospholipase A2 superfamily of signal transduction

enzymes. Trends. Biochem. Sci. 22:1-2.

77. Denys A., Allain F., Foxwell B., and Spik G. 1997. Distribution of cyclophilin B binding sites

in the subsets of human peripheral blood lymphocytes. Immunol. 91:609-617.

78. Denys A., Allain F., Masy E., Dessaint J.P., and Spik G. 1998. Enhancing the Effect of

Secreted Cyclophilin B on Immunosuppressive Activity of Cyclosporine. Transplantation. 65:1076-

1084.

79. Diccianni M.B., McLean L.R., Stuart W.D., Mistry M.J., Gil C.M., and Harmony J.A.K. 1991.

Porcine pancreatic phospholipase A2 isoforms: differential regulation by heparin. Biochim. Biophys.

Acta 1082:85-93.

80. Doyle V., Viji S., and Crompton M. 1999. Evidence that cyclophilin A protects cells against

oxidative stress. Biochem. J. 341:127-132.

81. Dua R., and Cho W. 1994. Inhibition of human secretory class II phospholipase A2 by heparin.

Eur. J. Biochem. 221:481-490.

82. Dwulet F.E., Wallace D.K., and Benson M.D. 1988. Amino Acid Structures of Multiple

Forms of Amyloid-Related Serum Protein SAA from a Single Individual. Biochemistry. 27:1677-

1682.

83. Edman P., and Begg G. 1967. A Protein Sequentor. Eur. J. Biochem. 1:80-91.

84. Endrich M.M., and Gehring H. 1998. The V3 loop of human immunodeficiency virus type-1

envelope protein is a high affinity ligand for immunophilins present in human blood. Eur. J. Biochem.

252:441-446.

85. Ensenauer R., Puttmann M., Quintel M., Katterman R., and Aufenanger J. 1994. Comparison

of serum phospholipase A2, polymorphonuclear granulocyte elastase, C-reactive protein and serum

amyloid A with the APACHE II score in the prognosis of multiple injured patients. Clin. Invest.

72:843-849.

86. Erlanger B.F., Borek F., Beiser S.M., and Lieberman S. 1957. Steroid Protein Conjugates. J.

Biol. Chem. 228.

87. Evans G.S., Chwalinski S., Owen G., Booth C., Singh A., and Potten C.S. 1994. Expression of

Pokeweed Lectin Binding in Murine Intestinal Paneth Cells. Epith. Cell Biol. 3:7-15.

88. Falquet L., Pagni M., Bucher P., Halo N., Sigrist C.J., Hofman K., and Bairoch A. 2002. The

PROSITE database, its status in 2002. Nucleic Acids Res. 30:235-238.

218

Bibliography

89. Fanghanel J., and Fischer G. 2003. Thermodynamic characterization of the interaction of

human cyclophilin 18 with cyclosporin A. Biophys. Chem. 100:351-366.

90. Farooqui A.A., Yang H.C., and Horrocks L.A. 1994. Purification of lipases, phospholipases

and kinases by Heparin Sepharose chromatography. J. Chrom. A. 673:149-158.

91. Farrugia W., Aitken M.A., van Dunne F., Wong M.H., Brennecke S.P., Scott K.F., and Rice

G.E. 1993. Type II phospholipase A2 in human gestational tissues: Subcellular distribution of placental

immuno- and catalytic activity. Biochim. Biophys. Acta 1166:77-83.

92. Flinn W.R., and Bergan J.J. 1997. Visceral Ischemic Syndromes: Obstruction of the Superior

Mesenteric Artery, Celiac Axis, and Inferior Mesenteric Artery. In Textbook of Surgery. The

Biological Basis of Modern Surgical Practice. Sabiston D.C., and Lyerly H.K., editors. Philadephia:

W.B. Saunders. 1750-1759.

93. Forst S., Weiss J., Blackburn P., Frangione B., Goni F., and Elsback P. 1986. Agistrodon halys

blomhoffi Phospholipase A2. Possible role of NH2 terminal lysines in action on Phospholipids of E.

Coli. Biochemistry 25:8381-8385.

94. Fortes-Dias C.L., Lin Y., Ewell J., Diniz C.R., and Liu T.Y. 1994. A Phospholipase A2

Inhibitor from the Plasma of the South American Rattlesnake (Crotalus durissus terrificus). J. Biol.

Chem. 269:15646-15651.

95. Foyn Bruun C., Nordstoga K., Sletten K., Husby G., and Marhaug G. 1995. Serum amyloid A

protein in humans and four animal species: a comparison by two dimensional electrophoresis. Comp.

Biochem. Physiol. 112B:227-234.

96. Francke E.K., Yuan H.E.H., and Luban J. 1994. Specific incorporation of cyclophilin A into

HIV-1 virions. Nature. 372:359-362.

97. Fried M. W., Murthy U. K., Hassig S. R., Woo J., and Oates R. P. 1991. Creatine kinase

isoenzymes in the diagnosis of intestinal infarction. Dig. Dis. Sci. 36:1589-1593.

98. Fry D.E. 2002. MODS: An Introduction. In Sepsis and Multiple Organ Dysfunction. Deitch

E.A., Vincent J.L., and Windsor A., editors. London: W.B. Saunders. 19-25.

99. Gabay C., and Kushner I. 1999. Acute Phase Proteins and Other Systemic Responses to

Inflammation. New Engl. J. Med. 340:448-454.

100. Gabor Miklos G.L., and Maleszka R. 2001. Protein functions and biological contexts.


101. Galat A., and Metcalfe S.M. 1995. Peptidylproline Cis/Trans Isomerases. Prog. Biophys.

Molec. Biol. 63:67-118.

102. Garabedian E. M., Roberts L. J., McNevin M. S., and Gordon J. I. 1997. Examining the role of

Paneth cells in the small intestine by lineage ablation in transgenic mice. J. Biol. Chem. 272:23729-

23740.

219

Bibliography

103. Gearhart S.L., Delaney C.P., Senagore A.J., Banbury M.K., Remzi F.H., Kiran R.P., and Fazio

V.W. 2003. Prospective Asssessment of the Predictive Value of alpha-Glutathione S-Transferase for

Intestinal Ischemia. Am. Surg. 69:324-329.

104. Gelb M.H., Valentin E., Ghomaschi F., Lazdunski M., and Lambeau G. 2000. Cloning and

Recombinant Expression of a Structurally Novel Human Secreted Phospholipase A2. J. Biol. Chem.

275:39823-39826.

105. Georgiou H.M., Rice G.E., and Baker M.S. 2001. Proteomic analysis of human plasma:

Failure of centrifugal ultrafiltration to remove albumin and other high molecular weight proteins.


106. Ghomaschi F., Loo R., Balsinde J., Bartoli F., Apitz-Castro R., Clark J.D., Dennis E.A., and

Gelb M.H. 1999. Trimethyl ketones and methyl fluorophosphonates as inhibitors of group IV and VI

phospholipases A2: structure-function studies with vesicle, micelle and membrane assays. Biochim.

Biophys. Acta 1420:45-56.

107. Gitlin J.D., and Colten H.R. 1987. Molecular Biology of the acute phase plasma proteins. In

Lymphokines. Pick E., and Landy M., editors. San Diego: Academic Press. 125-153.

108. Gonzalez R.J., Moore E.E., Ciesla D.J., Biffl W.L., Offner P.J., and Silliman C.C. 2001.

Phospholipase A2 derived neutral lipids from posthemorrhagic shock mesenteric lymph prime the

neutrophil oxidative burst. Surgery 130:198-203.

109. Gonzalez-Cuadrado S., Bustos C., Ruiz-Ortega M., Oritz A., Guijarro C., Plaza J.J., and Egido

J. 1996. Expression of leucocyte chemoattractants by interstitial renal fibroblasts: up-regulation by

drugs associated with interstitial fibrosis. Clin. Exp. Immunol. 106:518-522.

110. Gorg A., Obermaier C., Boguth G., Csordas A., Diaz J.J., and Madjar J.J. 1997. Very alkaline

immobilized pH gradients for two-dimensional electrophoresis of ribosomal and nuclear proteins.

Electrophoresis 18:328-337.

111. Gorg A., Obermaier C., Boguth G., and Weiss W. 1999. Recent developments in two-

dimensional gel electrophoresis with immobilized pH gradients: Wide pH gradients up to pH 12,

longer separation distances and simplified procedures. Electrophoresis 20:712-717.

112. Gorg A., Obermaier C., Boguth G., Harder A., Scheibe B., Wildgruber R., and Weiss W.

2000. The current state of two-dimensional electrophoresis with immobilized pH gradients.


113. Goris R. J., te Boekhorst T. P., Nuytinck J. K., and Gimbrere J. S. 1985. Multiple-organ

failure. Generalized autodestructive inflammation? Arch. Surg. 120:1109-1115.

114. Graff J.R., Konicek B.W., Deddens J.A., Chedid M., Hurst B.M., Colligan B., Neubauer B.L.,

Carter H.W., and Carter J.H. 2001. Expression of Group IIA Secretory Phospholipase A2 Increases

with Prostate Tumor grade. Clin. Cancer Res. 7:3857-3861.

220

Bibliography

115. Green J. A., Smith, G.M., Buchta, R., Lee, R., Ho, K.Y., Rajkovic, I.A., and Scott, K.F. 1991.

Circulating phospholipase A2 activity associated with sepsis and septic shock is indistinguishable from

that associated with rheumatoid arthritis. Inflammation. 15:355-367.

116. Grendell J.H., and Ockner R.K. 1989. Vascular Diseases of the Bowel. In Gastrointestinal

Disease. Pathophysiology, Diagnosis, Management. Sleisenger, and Fordtran, editors. Philadephia:

W.B. Saunders. 1904-1929.

117. Griffin T.J., Goodlett D.R., and Aebersold R. 2001. Advances in proteome analysis by mass

spectrometry. Curr. Opinion Biotech. 12:607-612.

118. Griffin T.J., and Aebersold R. 2001. Adavnces in Proteome Analysis by Mass Spectrometry.

J. Biol. Chem. 276:45497-44550.

119. Gronroos J.M., and Nevalainen T. J. 1992. Increased Concentrations of Synovial Type

Phospholipase A2 in Serum and Pulmonary and Renal Complications in Acute Pancreatitis. Digestion.

52.

120. Grotz M.R.W., Deitch E.A., Ding J., Xu D., Huang Q., and Regel G. 1999. Intestinal Cytokine

Response After Gut Ischemia: Role of Gut Barrier Failure. Ann. Surg. 229:478-486.

121. Guttenberger M., Neuhoff V., and Hampp R. 1991. A Dot-Blot Assay for Quantitation of

Nanogram Amounts of Protein in the Presence of Carrier Ampholytes and Other Possibly Interfering

Substances. Anal. Biochem. 196:99-103.

122. Gygi S.P., Rist B., Gerber S.A., Turecek F., Gelb M.H., and Aebersold R. 1999. Quantitative

analysis of complex protein mixtures using isotope-coded affinity tags. Nat. Biotechnol. 17:994-997.

123. Gygi S.P., Rist B., and Aebersold R. 2000. Measuring gene expression by quantitative

proteome analysis. Curr. Opin. Biotech. 11:396-401.

124. Gygi S.P., Corthals G.L., Zhang Y., Rochon Y., and Aebersold R. 2000. Evaluation of two

dimensional gel electrophoresis based proteome analysis technology. Proc. Natl. Acad. Sci. USA

97:9390-9395.

125. Gygi S.P., and Aebersold R. 2000. Mass Spectrometry and Proteomics. Curr. Opinion Chem.

Biol. 4:489-494.

126. Haapamaki M. M., Gronroos, J.M., Nurmi, H., Alanen, K., and Nevalainen, T.J. 1999. Gene

expression of group II phospholipase A2 in intestine in Crohn's disease. Am. J. Gastroenterol. 94:713-

720.

127. Habermann E., and Hardt K.L. 1972. A Sensitive and Specific Plate Test for the Quantitation

of Phospholipases. Anal. Biochem. 50:163-173.

128. Haglund U., Hulten L., Ahren C., and Lundgren O. 1975. Mucosal lesions in the small

intestine in shock. Gut 16:979-984.

129. Halestrap A.P. 2002. The Mitochondrial Permeability Transition- A Pore Way for the Heart to

Die. J. Clin. Basic Cardiol. 5:29-41.

221

Bibliography

130. Halestrap A.P., McStay G.P., and Clarke S.J. 2002. The permeability transition pore complex:

another view. Biochemie 84:153-166.

131. Han D.K., Eng J., Zhou H., and R., A. 2001. Quantitative profiling of differentiation induced

microsomal proteins using isotope coded affinity tags and mass spectrometry. Nature Biotech. 19:946-

951.

132. Hanasaki K., and Arita H. 1999. Biological and Pathological Functions of Phospholipase A2

Receptor. Arch. Biochem. Biophys. 272:215-223.

133. Hanasaki K., and Arita H. 2002. Biological and Pathological Functions of Novel Types of

Secretory Phospholipase A2s and Their Receptors. Annual Report of Shinogi Research Laboratories.

52:1-22.

134. Hanash S., Brichory F., and Beer D. 2001. A proteomic approach to the identification of lung

cancer markers. Disease Markers. 17:295-300.

135. Hara S., Kudo, I., Chang, H.W., Matsuta, K., Miyamoto, T., and Inoue, K. 1989. Purification

and characterization of extracellular phospholipase A2 from human synovial fluid in rheumatoid

arthritis. J. Biochem. Tokyo 105:395-399.

136. Harding M.W., Handschumacher R.E., and Speicher D.W. 1986. Isolation and Amino Acid

Sequence of Cyclophilin. J. Biol. Chem. 261:8547-8555.

137. Harris E.L.V., and Angal S. 1989. Protein purification methods: a practical approach.

Oxford: IRL Press.

138. Harwig S.S.L., Chen N.P., Park A.S.K., and Lehrer R.I. 1993. Purification of Cysteine Rich

Bioactive Peptides from Leukocytes by Continuous Acid-Urea Polyacrylamide Gel Electrophoresis.

Anal. Biochem. 208:382-386.

139. Hasel K.W., Glass J.R., Godbout M., and Sutcliffe J.G. 1991. An Endoplasmic Reticulum

Specific Cyclophilin. Mol. Cell Biol. 11:3484-3491.

140. Hayakawa M., Horigome K., Kudo I., Tomita M., Nojima S., and Inoue K. 1987. Amino acid

Composition and NH2 Terminal Amino Acid Sequence of rat Platelet Secretory Phospholipase A2. J.

Biochem. 101:1311-1314.

141. Haynes P.A., and Yates J.R. 2000. Proteome profiling- pitfalls and progress. Yeast 17:81-87.

142. Heegard P.M.H., Godson D.L., Toussaint M.J.M., Tjornehoj K., Larsen L.E., Viuff B., and

Ronsholt L. 2000. The acute phase response of haptoglobin and serum amyloid A (SAA) in cattle

undergoing experimental infection with bovine respiratpry syncytial virus. Vet Immunol.

Immunopathol. 77:151-159.

143. Herbert B., and Righetti P.G. 2000. A turning point in proteome analysis: Sample

prefractionation via multicompartment electrolyzers with isoelectric membranes. Electrophoresis

21:3639-3648.

144. Hibbard J.S., Swenson P.C., and Levin A.G. 1931. Roentegenology of Experimental

Mesenteric Vascular Occlusion. Arch. Surg. 26:20-26.

222

Bibliography

145. Hiltebrand L.B., Krejci V., Banic A., Erni D., Wheatley A.M., and Sigurdsson G.H. 2000.

Dynamic study of the distribution of microcirculatory blood flow in multiple splanchnic organs in

septic shock. Crit. Care Med. 28:3233-3241.

146. Ho S., Clipstone N., Timmermann L., Northrop J., Graef I., Fiorentino D., Nourse J., and

Crabtree G.R. 1996. The Mechanism of Action of Cyclosporin A and FK506. Clin. Immunol.

Immunopathol. 80:S40-45.

147. Hoffman P., Ji H., Moritz R.L., Connolly L.M., Frecklington D.F., Layton M.J., Eddes J.S.,

and Simpson R.J. 2001. Continuous free-flow electrophoresis separation of cytosolic proteins from the

human colon carcinoma cell line LIM 1215: A non two dimensional gel electrophoresis-based

proteome analysis strategy. Proteomics. 1:807-818.

148. Holzman T.F., Egan D.A., Edalji R., Simmer R.L., Helfrich R., Taylor A., and Burres N.S.

1991. Preliminary Characterization of a Cloned Neutral Isoelectric Form of the Human Peptidyl Prolyl

Isomerase Cyclophilin. J. Biol. Chem. 266:2474-2479.

149. Howard T. J., Plaskon L. A., Wiebke E. A., Wilcox M. G., and Madura J. A. 1996.

Nonocclusive mesenteric ischemia remains a diagnostic dilemma. Am. J. Surg. 171:405-408.

150. Husby G., Marhaug G., Dowton B., Sletten K., and Sipe J.D. 1994. Serum Amyloid A (SAA)

biochemistry, genetics and the pathogenesis of AA amyloidosis. Amyloid: Int. J. Exp. Clin. Invest.

1:119-137.

151. Ilzecka J., and Stelmasiak Z. 2000. Prognostic Importance of monitoring serum amyloid A

protein (SAA) in patients with cerebral infarction. Acta Clin. Croat. 39:139-145.

152. Ishizaki J., Suzuki, N., Higashino, K., Yokota, Y., Ono, T., Kawamoto, K., Fujii, N., Arita, H.,

and Hanasaki, K. 1999. Cloning and characterization of novel mouse and human secretory

phospholipase A2s. J. Biol. Chem. 274:24973-24979.

153. Ivery M.T.G. 2000. Immunophilins: Switched on Protein Binding Domains? Med. Res. Rev.

20:452-484.

154. Iwata T., Kogame K., Toki T., Yokoyama M., Yamamoto M., and Ito E. 1998. Structure and

chromosome mapping of the human small maf genes MAFG and MAFK. Cytogenet. Cell Genet.

82:88-90.

155. Jamieson W.G., Marchuk S., Rowsom J., and Durland D. 1982. The early diagnosis of

massive acute intestinal ischaemia. Br. J. Surg. 69(Supp.):S52-53.

156. Jiang J., Neubauer B.L., Graff J.R., Chedid M., Thomas J.E., Roehm N.W., Zhang S., Eckert

G.J., Koch M.O., Eble J.N., et al. 2002. Expression of Group IIA Secretory Phospholipase A2 Is

Elevated in Prostatic Intraepithelial Neoplasia and Adenocarcinoma. Am. J. Pathol. 160:667-671.

157. Jungblut P. 1997. Two Dimensional Electrophoresis. In Protein Structure Analysis,

Preparation, Characterization and Microsequencing. Kamp R.M., Choli-Papadopolou T., and

Wittman-Liebold B., editors. Berlin: Springer. 183-214.

223

Bibliography

158. Kainer D.B., and Doris P.A. 2000. Cyclophilin B Expression in Renal Proximal Tubules of

Hypertensive rats. Hypertension 35:958-964.

159. Kallajoki M., and Nevalainen T. J. 1997. Expression of Group II Phospholipase A2 in Human

Tissues. In Phospholipase A2: Basic and Clinical Aspects in Inflammatory Diseases. Uhl W.,

Nevalainen T. J., and Buchler M.W., editors. Basel: Karger. 8-16.

160. Kanda A., Ono, T., Yoshida N., Tojo H., and Okamoto M. 1989. The Primary structure of a

Membrane associated Phospholipase A2 from Human Spleen. Biochem. Biophys. Res. Comm. 163:42-

48.

161. Kanda T., Fujii H., Fujita M., Sakai Y., Ono T., and Hatakeyama K. 1995. Intestinal fatty acid

binding protein is available for diagnosis of intestinal ischaemia: immunochemical analysis of two

patients with ischaemic intestinal diseases. Gut 36:788-791.

162. Kanda T., Fujii H., Tani T., Murakami H., Suda T., Sakai Y., Ono T., and Hatakeyama K.

1996. Intestinal fatty acid-binding protein is a useful diagnostic marker for mesenteric infarction in

humans. Gastroenterology 110:339-343.

163. Kang S.W., Baines I.C., and Rhee S.G. 1998. Characterization of a Mammalian Peroxiredoxin

That Contains One Conserved Cysteine. J. Biol. Chem. 273:6303-6311.

164. Karlsson E. 1979. Chemistry of Protein Toxins in Snake Venoms. In Snake venoms. Lee C.Y.,

editor. Berlin: Springer Verlag. 159-212.

165. Kaufmann S.H., Ewing C.M., and Shaper J.H. 1987. The Erasable Western Blot. Anal.

Biochem. 161:89-95.

166. Kay J.E. 1996. Structure-function relationships in the FK506-binding protein (FKBP) family

of peptidylprolyl cis-trans isomerases. Biochem. J. 314:361-385.

167. Kern S. E., Keren D. F., Beals T. F., and Varani J. 1987. A model for Paneth cell study: tissue

culture of the hyperplastic Paneth cell population of rabbit Thiry-Vella ileal loops. Adv. Exp. Med.

Biol. 216A:419-426.

168. Keshav S., McKnight A. J., Arora R., and Gordon S. 1997. Cloning of intestinal

phospholipase A2 from intestinal epithelial RNA by differential display PCR. Cell Prolif. 30:369-383.

169. Keuter M., Dharmana E., Kullberg B.J., Schalkwijk C., Gasem M.H., Seuren L.,

Djokomoeljanto R., Dolmans W.M.V., van den Bosch H., and van der Meer J.W.M. 1995.

Phospholipase A2 Is a Circulating Mediator in Typhoid Fever. J. Infect. Dis. 172:305-308.

170. Kieffer L.J., Seng T.W., Li W., Osterman D.G., Handschumacher R.E., and Bayney R.M.

1993. Cyclophilin-40, a Protein with Homology to the P59 Component of the Steroid Receptor

Complex. J. Biol. Chem. 268:12303-12310.

171. Kim D.K., Kudo I., and Inoue K. 1991. Purification and characterization of rabbit platelet

cytosolic phospholipase A2. Biochim. Biophys. Acta 1083:80-88.

172. Kim T.S., Sundaresh C.S., Feinstein S.I., Dodia C., Skach W.R., Jain M.K., nagase T., Seki

N., Ishikawa K., Nomura N., et al. 1997. Identification of a Human cDNA Clone for Lysosomal Type

224

Bibliography

Ca2+ independent Phospholipase A2 and Properties of the Expressed Protein. J. Biol. Chem. 272:2542-

2550.

173. Kiyohara H., Egami H., Shibata Y., Murata K., Ohshima S., and Ogawa M. 1992. Light

microscopic immunohistochemical analysis of the distribution of group II phospholipase A2 in human

digestive organs. J. Histochem. Cytochem. 40:1659-1664.

174. Klein H. M., Lensing, R., Klosterhalfen, B., Tons, C., and Gunther, R.W. 1995. Diagnostic

imaging of mesenteric infarction. Radiology 197:79-82.

175. Klempnauer J., Grothues F., Bektas H., and Wahlers T. 1997. Acute mesenteric ischemia

following cardiac surgery. J. Cardiovasc. Surg. 38:639-643.

176. Klotz S., Vestring T., Rotker J., Schmidt C., Scheld H.H., and Schmid C. 2001. Diagnosis and

Treatment of Nonocclusive Mesenteric Ischemia After Open Heart Surgery. Ann. Thorac. Surg.

72:1583-1586.

177. Knaus W.A., Zimmerman J.E., Wagner D.P., Draper E.A., and Lawrence D.E. 1981.

APACHE- acute physiology and chronic health evaluation: a physiologically based classification

system. Crit. Care Med. 9:591-597.

178. Knaus W.A., Draper E.A., Wagner D.P., and Zimmerman J.E. 1985. APACHE II: A severity

of disease classification system. Crit. Care Med. 13:818-829.

179. Koduri R. S., Baker, S.F., Snitko, Y., Han, S.K., Cho, W., Wilton, D.C., and Gelb, M.H. 1998.

Action of human group IIA secreted phospholipase A2 on cell membranes. Vesicle but not heparinoid

binding determines rate of fatty acid release by exogenously added enzyme. J. Biol. Chem. 273:32142-

32153.

180. Koike K., Moore E. E., Moore F. A., Carl V. S., Pitman J. M., and Banerjee A. 1992.

Phospholipase A2 inhibition decouples lung injury from gut ischemia-reperfusion. Surgery 112:173-

180.

181. Koike K., Moore E.E., Moore F.A., Read R.A., Carl V.S., and Banerjee A. 1994. Gut

Ischemia/reperfusion produces lung injury independent of endotoxin. Crit. Care Med. 22:1438-1444.

182. Koike K., Moore E. E., Moore F. A., Kim F. J., Carl V. S., and Banerjee A. 1995. Gut

phospholipase A2 mediates neutrophil priming and lung injury after mesenteric ischemia-reperfusion.

Am. J. Physiol. 268:G397-403.

183. Koike K., Yamamoto Y., Hori Y., and Ono T. 2000. Group IIA Phospholipase A2 Mediates

Lung Injury in Intestinal Ischemia-Reperfusion. Ann. Surg. 232:90-97.

184. Kong S. E., Blennerhassett L. R., Heel K. A., McCauley R. D., and Hall J. C. 1998.

Ischaemia-reperfusion injury to the intestine. Aust. N.Z. J. Surg. 68:554-561.

185. Kramer R. M., Hession C., Johansen B., Hayes G., McGray P., Chow E. P., Tizard R., and

Pepinsky R. B. 1989. Structure and properties of a human non-pancreatic phospholipase A2. J. Biol.

Chem. 264:5768-5775.

225

Bibliography

186. Krishna R.G., and Wold F. 1993. Post Translational Modification of Proteins. Adv. Enzymol.

Relat. Areas Mol. Biol. 67:265-298.

187. Kudo I., Murakami M., Hara S., and Inoue K. 1993. Mammalian non-pancreatic

phospholipases A2. Biochim. Biophys. Acta 1170:217-231.

188. Kugiyama K., Ota Y., Takazoe K., Moriyama Y., Kawano H., Miyao Y., Sakamoto T.,

Soejima H., Ogawa H., Doi H., et al. 1999. Circulating levels of Secretory Type II Phospholipase A2

predict Coronary Events in Patients with Coronary Artery Disease. Circulation 100:1280-1284.

189. Kurland B., Brandt L.J., and Delaney H.M. 1992. Diagnostic tests for Intestinal Ischemia.

Surg. Clin. North Am. 72:85-105.

190. Laemmli U.K. 1970. Cleavage of structural proteins during the assembly of the head of

bacteriophage T4. Nature (London) 227:680-685.

191. Lai C.Y., and Wada K. 1988. Phospholipase A2 From Human Synovial Fluid; Purification and

Structural Homology to the Placental Enzyme. Biochem. Biophys. Res. Comm. 157:488-493.

192. Lambeau G., and Lazdunski M. 1999. Receptors for a growing family of secreted

phospholipases A2. Trends Pharm. Sci. 20:162-170.

193. Lange H., and Jackel R. 1994. Usefulness of Plasma Lactate Concentration in the Diagnosis of

Acute Abdominal Disease. Eur. J. Surg. 160:381-384.

194. Lawrence A.J., and Moores G.R. 1975. Activation of bee venom Phospholipase A2 by Fatty

Acids, Aliphatic Anhydrides and Glutaraldehyde. FEBS Lett. 49:287-291.

195. Lawson A. J., Smit R. A., Jeffers N. A., and Osborne J. W. 1982. Isolation of rat intestinal

crypt cells. Cell. Tissue Kinet. 15:69-80.

196. Liang R.C.M.Y., Neo J.C.H., Lo S.L., Tan G.S., Seow T.K., and Chung M.C.M. 2002.

Proteome database of hepatocellular carcinoma. J. Chrom. B. 771:303-328.

197. Liao D.F., Jin Z.G., Baas A.S., Daum G., Gygi S.P., Aebersold R., and Berk B.C. 2000.

Purification and Identification of Secreted Oxidative Stress Induced factors from vascular Smooth

Muscle Cells. J. Biol. Chem. 275:189-196.

198. Liebler D.C. 2002. Introduction to Proteomics. Tools for the New Biology. Totowa: Humana

Press.

199. Lilja I., Smedh K., Olaison G., Sjodahl R., Tagesson C., and Gustafson Svard C. 1995.

Phospholipase A2 gene expression and activity in histologically normal ileal mucosa and in Crohn's

ileitis. Gut 37:380-385.

200. Link A.J., Eng J., CSchieltz D.M., Carmack E., Mize G.J., Morris D.R., Garvik B.M., and

Yates J.R. 1999. Direct analysis of protein complexes using mass spectrometry. Nature Biotech.

17:676-682.

201. Macy E.M., Hayes T.E., and Tracy R.P. 1997. Variability in the measurement of C-reactive

protein in healthy subjects: implications for reference intervals and epidemiological applications. Clin.

Chem. 43:52-58.

226

Bibliography

202. Magnotti L. J., Upperman J. S., Xu D. Z., Lu Q., and Deitch E. A. 1998. Gut-derived

mesenteric lymph but not portal blood increases endothelial cell permeability and promotes lung

injury after hemorrhagic shock. Ann. Surg. 228:518-527.

203. Malle E., Bollmann A., Steinmetz A., Gemsa D., Leis H.J., and Sattler W. 1997. Serum

amyloid A (SAA) protein enhances formation of cyclooxygenase metabolites of activated human

monocytes. FEBS Lett. 419:215-219.

204. Mansbach C. M., Pieroni G., and Verger R. 1982. Intestinal phospholipase, a novel enzyme. J.

Clin. Invest. 69:368-376.

205. Mansbach C.M. 1990. Phospholipases: Old Enzymes With New Meaning. Gastroenterol.

98:1369-1382.

206. Mansour M.A. 1999. Management of Acute Mesenteric Ischemia. Arch. Surg. 134:328-331.

207. Marshak R.H., and Lindner A.E. 1967. Vascular Disease of the small bowel and colon. In

Alimentary Tract Roentgenology. Marulis, and Burhenne, editors. St. Louis: Mosby. 1144-1150.

208. Martin Porter E., Poles M.A., Lee J.S., Naitoh J., Bevins C.L., and Ganz T. 1998. Isolation of

human intestinal defensins from ileal neobladder urine. FEBS Lett. 434:272-276.

209. Maury C.P.J., Ehnholm C., and Lukka M. 1985. Serum Amyloid A protein (SAA) subtypes in

acute and chronic inflammatory conditions. Ann. Rheum. Dis. 44:711-715.

210. McKelvie P.A., and Rode J. 1992. Autopsy rate and a clinicopathological audit in an

Australian metropolitan hospital- cause for concern? Med. J. Aust. 156:456-462.

211. McMenamy R.H., Birkhahn R., Oswald G., Reed R., Rumph C., Vaidyanath N., Yu L., Cerra

F.B., Sorkness R., and Border J.R. 1981. Multiple Systems Organ Failure: I. The Basal State. J.

Trauma. 21:99-114.

212. Meek R.L., and Benditt E.P. 1986. Amyloid A Gene Family Expression in Different Mouse

Tissues. J. Exp. Med. 164:2006-2017.

213. Merril C.R., and Goldman D. 1982. Quantitative Two-Dimensional Protein Electrophoresis

for Studies of Inborn Errors of Metabolism. Clin. Chem. 28:1015-1020.

214. Migita K., Kawabe Y., Tominaga M., Origuchi T., Aoyagi T., and Eguchi K. 1998. Serum

Amyloid A protein induces production of matrix metalloproteinases by human synovial fibroblasts.

Lab. Invest. 78:535-539.

215. Mikol V., Kallen J., and Walkinshaw M.D. 1994. X-ray structure of a cyclophilin

B/cyclosporin complex: Comparison with cyclophilin A and delineation of its calcineurin binding

domain. Proc. Natl. Acad. Sci. USA 91:5183-5186.

216. Minami T., Tojo H., Shinomura Y., Tarui S., and Okamoto M. 1992. Raised serum activity of

phospholipase A2 immunochemically related to group II enzyme in inflammatory bowel disease: its

correlation with disease activity of Crohn's disease and ulcerative colitis. Gut 33:914-921.

227

Bibliography

217. Minami T., Tojo H., Shinomura Y., Matsuzawa Y., and Okamoto M. 1993. Purification and

characterization of a phospholipase A2 from human ileal mucosa. Biochim. Biophys. Acta 1170:125-

130.

218. Minami T., Tojo H., Matsuzawa A., and Okamoto M. 1994. Increased group II phospolipase

A2 in colonic mucosa of patients with Crohn's disease and ulcerative colitis. Gut 35:1593-1598.

219. Minami T., and Tojo H. 1997. Phospholipase A2 in Chronic Inflammatory Disease of the

Intestine. In Phospholipase A2: Basic and Clinical Aspects in Inflammatory Disease. Basel: Karger.

205-213.

220. Montgomery R.A., Venbrux A.B., and Bulkley G.B. 1997. Mesenteric Ischemia. New York:

Churchill Livingstone. 541-553. pp.

221. Moore F.A., Moore E.E., Poggetti R., McAnena O.J., Peterson V.M., Abernathy C.M., and

Parsons P.E. 1991. Gut Bacterial Translocation via the Portal Vein; A Clinical Perspective with major

Torso Trauma. J. Trauma 31:629-643.

222. Moran J.A., Dahl E.L., and Mulcahy R.T. 2002. Differential induction of mafF, mafG and

mafK expression by electrophile response element activators. Biochem. J. 361:371-377.

223. Morita H., Nakanishi K., Dohi T., Yasugi E., and Oshima M. 1999. Phospholipid turnover in

the inflamed intestinal mucosa: arachidonic acid-rich phosphatidyl/plasmenyl-ethanolamine in the

mucosa in inflammatory bowel disease. J. Gastroenterol. 34:46-53.

224. Morley J.J., and Kushner I. 1982. Serum C-reactive protein levels in Disease. Ann. N.Y. Acad.

Sci. 389:406-418.

225. Morris S., Hays W., Enomoto M., Glew R., Feddersen R., Fry D., and Morris D. 1999. Serum

cytosolic beta-glucosidase elevation and early ischemic injury to guinea pig small intestine. Surgery

125:202-210.

226. Mulherkar R., Rao, R.S., Wagle, A.S., Patki, V., and Deo, M.G. 1993. Enhancing factor, a

Paneth cell specific protein from mouse small intestines: predicted amino acid sequence from RT-PCR

amplified cDNA and its expression. Biochem. Biophys. Res. Commun. 195:1254-1263.

227. Mulherkar R., Rao R., Rao L., Patki V., Chauhan V.S., and Deo M.G. 1993. Enhancing factor

protein from mouse small intestines belongs to the phospholipase A2 family. FEBS Lett. 317:263-266.

228. Murakami M., Nakatani Y., and Kudo I. 1996. Type II Secretory Phospholipase A2 Associated

with Cell Surfaces via C-terminal Heparin binding Lysine Residues Augments Stimulus initiated

Delayed Prostaglandin Generation. J. Biol. Chem. 271:30041-30061.

229. Murakami M., Shimbara S., Kambe T., Kuwata H., Winstead M. V., Tischfield J. A., and

Kudo I. 1998. The functions of five distinct mammalian phospholipase A2s in regulating arachidonic

acid release. Type IIA and type V secretory phospholipase A2s are functionally redundant and act in

concert with cytosolic phospholipase A2. J. Biol. Chem. 273:14411-14423.

230. Murakami M., Koduri R. S., Enomoto A., Shimbara S., Seki M., Yoshihara K., Singer A.,

Valentin E., Ghomaschi F., Lambeau G., et al. 2001. Distinct Arachidonate-releasing Functions of

228

Bibliography

Mammalian Secreted Phospholipase A2s in Human Embryonic Kidney 293 and Rat Mastocytoma

RBL-2H3 Cells through Heparan Sulphate Shuttling and External Plasma Membrane Mechanisms. J.

Biol. Chem. 276:10083-10096.

231. Murakami M., Yoshihara K., Shimbara S., Lambeau G., Gelb M. H., Singer A., Sawada M.,

Inagaki N., Nagai H., Ishihara M., et al. 2002. Cellular Arachidonate-releasing Function and

Inflammation-associated Expression of Group IIF Secretory Phospholipase A2. J. Biol. Chem.

277:19145-19155.

232. Neuhoff V., Arnold N., Taube D., and Ehrhardt W. 1988. Improved staining of proteins in

polyacrylamide including isoelectric focussing gels with clear background at nanogram sensitivity

using the Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 9:255-262.

233. Nevalainen T. J., Gronroos J.M., and Kortesuo P.T. 1993. Pancreatic and synovial type

phospholipase A2 in serum samples from patients with sever acute pancreatitis. Gut. 34:1133-1136.

234. Nevalainen T. J., Gronroos J. M., and Kallajoki M. 1995. Expression of group II

phospholipase A2 in the human gastrointestinal tract. Lab. Invest. 72:201-208.

235. Nevalainen T. J., and Gronroos J. M. 1997. Serum phospholipase A2 in Inflammatory Disease.

In Phospholipase A2: Basic and Clinical Aspects in Inflammatory Disease. Uhl W., Nevalainen T. J.,

and Buchler M., editors. Basel: Karger. 104-109.

236. Nevalainen T. J., Haapamaki M. M., and Gronroos J. M. 2000. Role of secretory

phospholipases A2 in inflammatory diseases. Biochim. Biophys. Acta 1488:83-90.

237. Newman T.S., Magnuson T.H., Ahrendt S.A., Smith-Meek M.A., and Bender J.S. 1998. The

Changing face of mesenteric Infarction. Am. Surg. 64:611-616.

238. Noe D.A. 2001. Tissue Injury.: Noe D.A.,.

239. Nyman K. M., Uhl W., Forsstrom J., Buchler M., Beger H. G., and Nevalainen T. J. 1996.

Serum phospholipase A2 in patients with multiple organ failure. J. Surg. Res. 60:7-14.

240. Ogawa M., Arakawa H., Yamahita S., Sakamoto K., and Ikei S. 1992. Postoperative levels of

Serum Interlukin 6 and Group II Phospholipase A2: Group II Phospholipase A2 is an Acute Phase

Reactant. Res. Commun. Chem. Pathol. Pharmacol. 75:109-115.

241. Ono T., Tojo H., JKuramitsu S., Kagamiyama H., and Okamoto M. 1988. Purification and

Characterization of a Membrane associated Phospholipase A2 from rat Speen. J. Biol. Chem.

263:5732-5738.

242. Otamiri T., Franzen L., Lindmark D., and Tagesson C. 1987. Increased phospholipase A2 and

decreased lysophospholipase activity in the small intestinal mucosa after ischaemia and

revascularisation. Gut 28:1445-1453.

243. Otamiri T., Lindahl M., and Tagesson C. 1988. Phospholipase A2 inhibition prevents mucosal

damage associated with small intestinal ischaemia in rats. Gut 29:489-494.

244. Otamiri T., and Tagesson C. 1989. Role of phospholipase A2 and oxygenated free radicals in

mucosal damage after small intestinal ischemia and reperfusion. Am. J. Surg. 157:562-565;.

229

Bibliography

245. Otamiri T. 1989. Oxygen radicals, lipid peroxidation, and neutrophil infiltration after small-

intestinal ischemia and reperfusion. Surgery 105:593-597.

246. Patel A., Kaleya R.N., and Sammartano R. J. 1992. Pathophysiology of Mesenteric Ischaemia.

Surg. Clin. North Am. 72:31-41.

247. Pearson T.A., Mensah G.A., Alexander R.W., Anderson J.L., Cannon R.O., Criqui M., Fadl

Y.Y., Fortmann S.P., Hong Y., Myers G.L., et al. 2003. Markers of Inflammation and Cardiovascular

Disease. Circulation 107:499-511.

248. Pedersen S.K., Harry J.L., Sebastian L., Baker J., Traini M.D., McCarthy J.T., Monoharan A.,

Wilkins M.R., Gooley A.A., Righetti P.G., et al. 2003. Unseen Proteome: Mining below the Tip of the

Iceberg to Find Low Abundance and Membrane Proteins. J. Proteome Research 2:303-311.

249. Pennington C.R. 1997. Intestinal failure and Artificial Nutritional Support. Proc. R. Coll.

Physicians Edinb. 27:418-431.

250. Phillpotts C.J. 1986. Histopathological changes in the epithelial cells of rat duodenum

following chronic dietary exposure to cadmium, with particular reference to Paneth cells. Br. J. Exp.

Path. 67:505-516.

251. Polson H., Mowat C., and Himal H. S. 1981. Experimental and clinical studies of mesenteric

infarction. Surg. Gynecol. Obstet. 153:360-362.

252. Poole G.V. 2002. MODS in the Septic/Inflammatory Patient. In Sepsis and Multiple Organ

Dysfunction. Deitch E. A., Vincent J.L., and Windsor A., editors. London: W.B. Saunders. 34-38.

253. Price E.R., Zydowsky L.D., Jin M., Baker C.H., McKeon F.D., and Walsh C.T. 1991. Human

cyclophilin B: a second cyclophilin gene encodes a peptidyl-prolyl isomerase with a signal sequence.

Proc. Natl. Acad. Sci. USA 88:1903-1907.

254. Price E.R., Jin M., Lim D., Pati S., Walsh C.T., and McKeon F.D. 1994. Cyclophilin B

Trafficking through the secretory pathway is altered by binding of cyclosporin. Proc. Natl. Acad. Sci.

USA 91:3931-3935.

255. Pruzanski W., Goulding N.J., Flower R.J., Gladman D.D., Urowitz M.B., Goodman P.J., Scott

K.F., and Vadas P. 1994. Circulating Group II Phospholipase A2 Activity and Antilipocortin

Antibodies in Systemic Lupus Erythematosus. Correlative Study with Disease Activity. J. Rheumatol.

21:252-257.

256. Pruzanski W., de Beer F.C., de Beer M.C., Stefanski E., and Vadas P. 1995. Serum amyloid A

protein enhances the activity of secretory non-pancreatic phospholipase A2. Biochem. J. 309:461-464.

257. Puglisi R. N., Whalen T. V., and Doolin E. J. 1995. Computer analyzed histology of ischemic

injury to the gut. J. Pediatr. Surg. 30:839-844.

258. Puijk W.C., Verheij H.M., and De Haas G.H. 1977. The Primary Structure of Phospholipase

A2 From Porcine Pancreas. Biochim. Biophys. Acta 492:254-259.

259. Qu X. D., Lloyd K. C., Walsh J. H., and Lehrer R. I. 1996. Secretion of type II phospholipase

A2 and cryptdin by rat small intestinal Paneth cells. Infect. Immun. 64:5161-5165.

230

Bibliography

260. Rhee K.J., Baxt W.G., Mackenzie J.R., Willits N.H., Burney R.E., O'Malley R.J., Reid N.,

Schwabe D., Storer D.L., and Weber R. 1990. APACHE II scoring in the injured patient. Crit. Care

Med. 18:827-830.

261. Rice G. E., Wong, M.H., Farrugia, W., and Scott, K.F. 1998. Contribution of type II

phospholipase A2 to in vitro phospholipase A2 enzymatic activity in human term placenta. 157:25-31.

262. Righetti P.G., Wenisch E., and Faupel M. 1989. Preparative Protein Purification in a Multi-

Compartment Electrolyser with Immobiline Membranes. J. Chrom. 475:293-309.

263. Rogers A.I., and David S. 1995. Intestinal Blood Flow and Diseases of Vascular Impairment.

In Gastroenterology. Habrich W.S., Schaffner F., and Berk J.E., editors. Philadelphia: W.B. Saunders.

1212-1234.

264. Romaschin A.D., DeMajo W.C., Winton T., D'Costa M., Chang G., Rubin B., Gamliel Z., and

Walker P.M. 1992. Systemic Phospholipase A2 and Cachectin Levels in Adult Respiratory Distress

Syndrome and Multiple Organ Failure. Clin. Biochem. 25:55-60.

265. Rowell L.B., Blackmon J.R., Kenny M.A., and Escourrou P. 1984. Splanchnic vasomotor and

metabolic adjustments to hypoxia and exercise in humans. Am. J. Physiol. 247:H251-258.

266. Rozenfeld R.A., Liu X., De Plaen I., and Hsueh W. 2001. Role of gut flora on intestinal group

II phospholipase A2 activity and intestinal injury in shock. Am. J. Physiol. 281:G957-963.

267. Rycyzyn M.A., and Clevenger C.V. 2002. The intranuclear prolactin/cyclophilin B complex

as a transcriptional inducer. Proc. Natl. Acad. Sci. USA 99:6790-6795.

268. Sawada H., Murakami M., Enomoto A., Shimbara S., and Kudo I. 1999. Regulation of type V

phospholipase A2 expression and function by proinflammatory stimuli. Eur. J. Biochem. 263:826-835.

269. Schagger H., and Von Jagow G. 1987. Tricine Sodium Dodecy Suphate-Polyacrylamide Gel

Electrophoresis for the separation of Proteins in the range from 1 to 100kDa. Anal. Biochem. 166:368-

379.

270. Scheele G.A. 1982. Two-Dimensional Electrophoresis in Basic and Clinical Research, as

Exemplified by Studies on the Exocrine Pancreas. Clin. Chem. 28:1056-1061.

271. Schneider H., Charara N., Schmitz R., Wehrli S., Mikol V., Quesniaux F.L., and Movva N.R.

1994. Human Cyclophilin C: Primary Structure, Tissue Distribution, and Determination of Specificity

for Cyclosporins. Biochemistry 33:8218-8224.

272. Schoenberg M. H., and Beger H. G. 1993. Reperfusion injury after intestinal ischemia. Crit.

Care Med. 21:1376-1386.

273. Scott K.F., Graham G.G., and Bryant K.J. 2003. Secreted phospholipase A2 enzymes as

therapeutic targets. Expert Opin. Ther. Targets 7:427-440.

274. Seilhamer J. J., Plant, S., Pruzanski, W., Schilling, J., Stefanski, E., Vadas, P., and Johnson,

L.K. 1989. Multiple forms of phospholipase A2 in arthritic synovial fluid. J. Biochem. Tokyo 106:38-

42.

231

Bibliography

275. Seliger B., and Kellner R. 2002. Design of proteome-based studies in combination with

serology for the identification of biomarkers and novel targets. Proteomics 2:1641-1651.

276. Sherry B., Yartlett N., Strupp A., and Cerami A. 1992. Identification of cyclophilin as a

proinflammatory secretory product of lipopolysaccharide acivated macrophages. Proc. Natl. Acad. Sci.

USA 89:3511-3515.

277. Shimbara S., Murakami M., Kambe T., and Kudo I. 1999. Comparison of recombinant types

IIA, V and IIC phospholipase A2s, the three related mammalian secretory phospholipase A2 isozymes.

In Eicosanoids and Other Bioactive Lipids in Cancer, Inflammation, and Radiation Injury, 4. Honn et

al., editor. New York: Kluwer Academic/Plenum Publishers. 209-214.

278. Siebert R., and Stirling J. 2000. Paneth Cells: Siebert R.,

Stirling J.,.http://home.primus.com.au/royellis/pancell.htm

279. Siegel J.H., and Rixen D. 2002. Clinical and Physiologic Scoring Systems for Sepsis and

Organ Dysfunction. In Sepsis and Multiple Organ Dysfunction. Deitch E. A., Vincent J.L., and

Windsor A., editors. London: W.B. Saunders. 167-178.

280. Simpson R.J., Connolly L.M., Eddes J.S., Pereira J.J., Moritz R.L., and Reid G.E. 2000.

Proteomic analysis of the human colon carcinoma cell line (LIM 1215): Development of a membrane

protein database. Electrophoresis 21:1707-1732.

281. Singer A.G., Ghomaschi F., Le Calvez C., Bollinger J., Bezzine S., Rouault M., Sadilek M.,

Lazdunski M., Lambeau G., and Gelb M. H. 2002. Interfacial Kinetic and Binding Properties of the

Complete Set of Human and Mouse Groups I, II, V. X and XII Secreted Phospholipases A2. J. Biol.

Chem.

282. Sipe J.D. 2000. Serum amyloid A: from fibril to function. Current Status. Amyloid: Int. J. Exp.

Clin. Invest. 7:10-12.

283. Six D. A., and Dennis E. A. 2000. The expanding superfamily of phospholipase A2 enzymes:

classification and characterisation. Biochim. Biophys. Acta 1488:1-19.

284. Smirniotis V. E., Labrou A. T., and Tsiftses D. D. 1989. Plasma level of the creatine

phosphokinase BB isoenzyme during experimental intestinal ischemia. Ann. Vasc. Surg. 3:8-10.

285. Snyder S.H., and Sabatini D.M. 1995. Immunophilins and the nervous system. Nature

Medicine. 1:32-37.

286. Sonnino R. E., Pigatt, L., Schrama, A., Burchett, S., and Franson, R. 1997. Phospholipase A2

secretion during intestinal graft ischemia. Dig. Dis. Sci. 42:972-981.

287. Sonnino R.E., and Pigatt L.A. 1996. Secretory Phospholipase A2 levels in Rat Small Bowel. J.

Invest. Surg. 9:313-319.

288. Spik G., Haendler B., Delmas O., Mariller C., Chamoux M., Maes P., Tartar A., Montreuil J.,

Stedman K., Kocher H.P., et al. 1991. A Novel Secreted Cyclophilin-like Protein (SCYLP). J. Biol.

Chem. 266:10735-10738.

232

Bibliography

289. Sreenarasimhaiah J. 2003. Diagnosis and management of intestinal ischaemic disorders. Br.

Med. J. 326:1372-1376.

290. Stefanski E., Pruzanski W., Sternby B., and Vadas P. 1986. Purification of a Soluble

Phospholipase A2 from Synovial Fluid in Rheumatoid Arthritis. J. Biochem. 100:1297-1303.

291. Stein L.D. 2004. Human genome: End of the beginning. Nature. 431: 915-916.

292. Strachan A.F., Brandt W.F., Woo P., van der Westhuyzen D.R., Coetzee G.A., de Beer M.C.,

Shephard E.G., and de Beer F.C. 1989. Human Serum Amyloid A Protein. J. Biol. Chem. 264:18368-

18373.

293. Sun X., Rozenfeld R.A., Qu X., Huang W., Gonzalez-Crussi F., and Hsueh W. 1997. P-

selectin-deficient mice are protected from PAF-induced shock, intestinal injury, and lethality. Am. J.

Physiol. 273.

294. Sutton D. 1987. A Textbook of radiology and imaging. Edinburgh: Churchill Livingstone.

295. Suzuki N., Ishizaki J., Yokota Y., Higashino K., Ono T., Ikeda M., Fujii N., Kawamoto K.,

and Hanasaki K. 2000. Structures, enzymatic properties, and expression of novel human and mouse

secretory phospholipase A2s. J. Biol. Chem. 275:5785-5793.

296. Szewczyk B., and Kozloff L.M. 1985. A Method for the efficient Blotting of Strongly Basic

Proteins from Sodium Dodecyl Sulphate Polyacrylamide gels to Nitrocellulose. Anal. Biochem.

150:403-407.

297. Tanaka M., Riddell H., Soma Y., Hidaka H., and Kudo H. 1999. Morphologic Criteria

Applicable to Biopsy Specimens for Effective Distinction of Inflammatory Bowel Disease from Other

forms of Colitis and of Crohn's Disease from Ulcerative Colitis. Scand. J. Gastroenterol. 1:55-67.

298. Tegeder I., Schumacher A., John S., Geiger H., Geisslinger G., Bang H., and Brune K. 1997.

Elevated Serum Cyclophilin Levels in Patients with Severe Sepsis. J. Clin. Immunol. 17:380-386.

299. Thali M., Bukovsky A., Kondo E., Rosenwirth B., Walsh C.T., Sodroski J., and Gottlinger

H.G. 1994. Functional association of cyclophilin A with HIV-1 virions. Nature. 372:363-365.

300. Thomas S., Karnik S., and Balasubramanian K. A. 2002. Surgical Manipulation of the Small

Intestine and Its Effect on the Lung. J. Surg. Res 106:145-156.

301. Thompson J. S., Bragg L. E., and West W. W. 1990. Serum enzyme levels during intestinal

ischemia. Ann. Surg. 211:369-373.

302. Tilney N.L., Bailey G.L., and Morgan A.P. 1973. Sequential System failure after Rupture of

Abdominal Aortic Aneurysms: An Unsolved Problem in Postoperative Care. Ann. Surg. 178:117-122.

303. Tischfield J. A. 1997. A reassessment of the low molecular weight phospholipase A2 gene

family in mammals. J. Biol. Chem. 272:17247-17250.

304. Towbin H., Staehelin T., and Gordon J. 1970. Electrophoretic transfer of proteins from

polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci.

USA 76:4350-4354.

233

Bibliography

305. Uhl W., Buchler M., Nevalainen T. J., Deller A., and Beger H. G. 1990. Serum Phospholipase

A2 in Patients with Multiple Injuries. J. Trauma 30:1285-1290.

306. Uhl W., Beger H.G., Hoffmann G., Hanisch E., Schild A., Waydhas C., Entholzner E., Muller

K., Kellermann W., and Vogeser M. 1995. A multicenter study of phospholipase A2 in patients in

intensive care units. J. Am. Coll. Surg. 180:323-331.

307. Uhl W., and Buchler M.W. 1997. Phospholipase A2 in Surgical Intensive Care Patients. In

Phospholipase A2, Basic and Clinical Aspects in Inflammatory Diseases. Uhl W., Nevalainen T.J., and

Buchler M.W., editors. Basel: Karger. 187-199.

308. Uhlar C.M., and Whitehead A.S. 1999. Serum amyloid A, the major vertebrate acute-phase

reactant. Eur. J. Biochem. 265:501-523.

309. Vadas P., Browning, J., Edelson, J., and Pruzanski, W. 1993. Extracellular phospholipase A2

expression and inflammation: the relationship with associated disease states. J. Lipid Mediators 8:1-

30.

310. Vadas P., and Pruzanski W. 1997. Inter-Relationship of Phospholipase A2 and Lipid

Peroxidation in Multisystem Organ Failure in Septic Shock. In Phospholipase A2. Basic and Clinical

Aspects in Inflammatory Diseases. Uhl W., Nevalainen T. J., and Buchler M.W., editors. Basel:

Karger. 176-181.

311. Vaitukaitis J., Robbins J.B., Nieschlag E., and Ross G.T. 1971. A Method for producing

Specific Antisera with Small Doses of Immunogen. J. Clin. Endocr. 33:988-991.

312. Valentin E., Koduri R.S., Scimeca J.C., Carle G., Gelb M.H., Lazdunski M., and Lambeau G.

1999. Cloning and Recombinant Expression of a Novel Mouse-secreted Phospholipase A2. J. Biol.

Chem. 274:19152-19160.

313. Valentin E., and Lambeau G. 2000. Increasing molecular diversity of secreted phospholipases

A2 and their receptors and binding proteins. Biochim. Biophys. Acta 1488:59-70.

314. Valentin E., Ghomashchi F., Gelb M. H., Lazdunski M., and Lambeau G. 2000. Novel human

secreted phospholipase A2 with homology to the group III bee venom enzyme. J. Biol. Chem.

275:7492-7496.

315. Verger R., Ferrato F., Mansbach C. M., and Pieroni G. 1982. Novel intestinal phospholipase

A2: purification and some molecular characteristics. Biochemistry 21:6883-6889.

316. Verheij H.M., Slotboom A.J., and de Haas G.H. 1981. Structure and Function of

Phospholipase A2. Rev. Physiol. Biochem. Pharmacol. 91:91-203.

317. Waage A., and Bakke O. 1988. Glucocorticoids suppress the production of tumour necrosis

factor by lipopolysaccharide-stimulated human monocytes. Immunol. 63:299-302.

318. Wadman M., Syk I., and Elmstahl S. 2000. Survival after Operations for Ischaemic Bowel

Disease. Eur. J. Surg. 166:872-877.

319. Washburn M.P., Wolters D., and Yates J.R. 2001. Large scale analysis of the yeast proteome

by multidimensional protein identification technology. Nature Biotech. 19:242-247.

234

Bibliography

320. Watson G., Coade S., and Woo P. 1992. Analysis of the Genomic and Derived Protein

Structure of a Novel Human Serum Amyloid A Gene, SAA4. Scand. J. Immunol. 36:703-712.

321. Weifeng T.U., Jun C., and Guangxia X. 2001. Effects of quinacrine on gut origin

bacteria/endotoxin translocation in rats with gut ischemia/reperfusion injury. Chin. J. Burns 17:301-

303.

322. Weinrauch Y., Abad C., Liang N. S., Lowry S. F., and Weiss J. 1998. Mobilization of potent

plasma bactericidal activity during systemic bacterial challenge. Role of group IIA phospholipase A2.

102:633-638.

323. Weiss L., and Greep R.O. 1977. Histology. New York: McGraw Hill.

343. Welsh F.K.S., and Reynolds J.V. 2002. Gut Barrier Failure. In Sepsis and Multiple Organ

Dysfunction, A multi-disciplinary approach. Deitch E.A., Vincent J.L., and Windsor A., editors.

London: W.B. Saunders.

325. Westman-Brinkmalm A., and Davidsson P. 2002. Comparison of Preparative and Analytical

Two-Dimensional Electrophoresis for Isolation and Matrix-Assisted Laser Desorption/Ionization-

Time of Flight Mass Spectrometric Analysis of Transthyretin in Cerebrospinal Fluid. Anal. Biochem.

301:161-167.

326. Whitehead A.S., de Beer M.C., Steel D.M., Rits M., Lelias J.M., Lane W.S., and de Beer F.C.

1992. Identification of novel Members of the Serum Amyloid A Protein Superfamily as Constitutive

Apolipoproteins of High Density Lipoprotein. J. Biol. Chem. 267:3862-3867.

327. Whittaker M., and Simpson C.F. 1983. Molecular Characterization of Proteins. In

Electrophoretic techniques. Simpson C.F., and Whittaker M., editors. London: Academic Press. 38-69.

328. Wiklund O., Mattson-Hulten L., Hurt-Camejo E., and Oscarsson J. 2002. Effects of simvastin

and atorvastin on inflammation markers in plasma. J. Internal Med. 251:338-347.

329. Wilkins M.R., Sanchez J.C., Gooley A.A., Appel R.D., Humphrey-Smith I., Hochstrasser

D.F., and Williams K.L. 1995. Progress with Proteome Projects: Why all Proteins Expressed by a

Genome Should be Identified and How to do it. Biotechnology and Genetic Engineering Reviews.

13:19-50.

330. Wilkins M.R., Sanchez J.C., Williams K.L., and Hochstrasser D.F. 1996. Current challenges

and future applications for protein maps and post-translational vector maps in proteome projects.


331. Williams M.A., and Kagan H.K. 1985. Assessment of Lysyl Oxidase Variants by Urea Gel

Electrophoresis: Evidence against disulfide Isomers as Bases of the Enzyme heterogeneity. Anal.

Biochem. 149:430-437.

332. Wilson C., and Imrie C.W. 1986. Amylase and gut infarction. Br. J. Surg. 73:219-221.

333. Wilson C., Gupta R., Gilmour D.G., and Imrie C.W. 1987. Acute mesenteric ischaemia. Br. J.

Surg. 74:279-281.

235

Bibliography

334. Wollin A., Navert H., and Bounous G. 1981. Effect of intestinal ischemia on diamine oxidase

activity in rat intestinal tissue and blood. Gastroenterol. 80:349-355.

335. Xu D., Lu Q., and Deitch E.A. 1995. Calcium and phospholipase A2 appear to be involved in

the pathogenesis of hemorrhagic shock-induced mucosal injury and bacterial translocation. Crit. Care

Med. 23:125-131.

336. Yamada T., Miyake N., Itoh K., and Igari J. 2001. Further Characterization of Serum Amyloid

A4 as a minor Acute Phase Reactant and a possible Nutritional Marker. Clin. Chem. Lab. Med. 39:7-

10.

337. Yan J.X., Wait R., Berkelman T., Harry R.A., Westbrook J.A., Wheeler C.H., and Dunn M.J.

2000. A modified silver staining protocol for visualization of proteins that is compatible with matrix

assisted laser desorption ionization and electrospray ionization mass spectrometry. Electrophoresis

21:3666-3672.

338. Yang H. C., Mosior, M., Johnson, C.A., Chen, Y., and Dennis, E.A. 1999. Group-specific

assays that distinguish between the four major types of mammalian phospholipase A2. Anal. Biochem.

269:278-288.

339. Yoo S., Myszka D.G., Yeh C., McMurray M., Hill C.P., and Sundquist W.I. 1997. Molecular

Recognition in the HIV-1 Capsid/Cyclophilin A Complex. J. Mol. Biol. 269:780-795.

340. Zeindl-Eberhart E., Jungblut P., Otto A., and Rabes H.M. 1994. Identification of Tumor

Associated Protein variants during Rat Hepatocarcinogenesis. J. Biol. Chem. 269:14589-14594.

341. Ziegler T.R., Smith R.J., O'Dwyer S.T., Demling R.H., and Wilmore D.W. 1988. Increased

Intestinal Permability Associated With Infection in Burn Patients. Arch. Surg. 123:1313-1319.

342. Zweig M.H., and Campbell G. 1993. Receiver-Operating Characteristic (ROC) Plots: A

Fundamental Evaluation Tool in Clinical Medicine. Clin. Chem. 39:561-577.

236

Abbreviations - Deakin University30023241/glenister-phospholipas… · Fellow PhD. Students, in...

Documents

Transcript of Abbreviations - Deakin University30023241/glenister-phospholipas… · Fellow PhD. Students, in...