Blastocystis: Investigation of Parasite Adhesion and Its Role in Mediating Intestinal Barrier Compromise

BLASTOCYSTIS: INVESTIGATION OF PARASITE ADHESION AND ITS ROLE IN MEDIATING INTESTINAL BARRIER COMPROMISE

WU ZHAONA

(Bachelor of Sciences)

A THESIS SUBMITTED FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

DEPARTMENT OF MICROBIOLOGY

NATIONAL UNIVERSITY OF SINGAPORE

2014

“In quietness and trust is your strength.” Isaiah 30:15

DEDICATED WITH LOVE TO MY PARENTS &

FAMILY

Wu Zhaona 01.08.2014 ACKNOWLEDGEMENTS

Firstly, I would like to extend my deepest gratitude to my supervisor A/Prof

Kevin Tan, who has been highly supportive all along my PhD journey. He was a strict professor, holding onto guidelines when training students or staff in the way they should go. Sometimes I did not understand why he was strict, but soon it turned out to be extremely beneficial to me as a research student being trained. It would have been impossible for the completion of this project without his patient, timely, and kind help and encouragements, as well as wisdom which could not be understood in the beginning, but always speaks for itself later on. He encourages us to pursue projects we are passionate about, on condition that it comes with a good theoretical basis. It is such a blessing to pursue my PhD with his guidance.

We are very blessed to have a very good lab environment to work in and enjoy.

All the kind help has been given selflessly to me from my loving labmates, including past and present members of the Tan Lab. Specially, I would give thanks to Junhong, a brother-in-Christ, who has helped with every step of those experiments I used to have trouble with. His patient help has also been a great support. Also, for Chuu Ling, for her love which is not often verbally expressed but always abundantly given; Joshua, for his selfless support in experimental materials and devices; Yan Quan, for sharing lots of insights on projects in our discussions. My mentor, Haris, who is now at Yale University, has always been a good model for me in doing research. His passion towards research has been inspiring to me as a young research student. He also has

i shared a lot about his professional knowledge and insights in medicine which lends the research work clinical relevance.

I would also like to thank A/Prof Lee Yuan Kun for his kind help with the adhesion project, not only in knowledge input but also in advising on future directions; Dr Zhang Yongliang, for his valuable suggestions on various aspects of my project. His resourceful mind has provided new insights and approaches to tackle difficult questions. I highly appreciate their time and help given to me during the TAC meetings.

Many thanks be to Geok Choo, for her constant support and kindness which has made me feel at home in a foreign land; Mr. Rama, for his “harsh” words which urged me to have a good habit in experiment; Auntie Zina, for the so much love from her in treating us like her own children. I would like to thank

Mdm Siti Masnor for her patient and professional administrative assistance. I also want to thank all the lecturers of the Yong Loo Ling School of Medicine, for making the four-year PhD study a truly memorable journey. Speical thanks to Shuying and Wei an from Confocal Microscopy Unit, NUS, for their kind and timely assistance in my numerous bioimaging experiments.

I would like to thank National University of Singapore for granting me the scholarship to pursue this project.

Lastly, I extend my deepest gratitude to my dear Mommy and Daddy, and my dear brother for their unconditional love and support.

Wu Zhaona

2014

CONTENTS

ACKNOWLEDGEMENTS i TABLE OF CONTENTS iii SUMMARY ix LIST OF TABLES xii LIST OF FIGURES xiii LIST OF SYMBOLS xvii LIST OF PUBLICATIONS xxii

CHAPTER 1: INTRODUCTION 1-34 1.1 Introduction 1.2 Classification 1.3 Cell biology 1.4 Life cycle 1.5 Clinical presentation 1.6 Laboratory diagnosis 1.7 Epidemiology and prevalence 1.8 High risk population 1.9 Treatment 1.10 Pathogenesis 1.11 Objectives of the present study

CHAPTER 2: ADHESION OF BLASTOCYSTIS TO 35-53 ENTEROCYTES AND DEVELOPMENT OF AN ADHESION ASSAY 2.1 Introduction 2.2 Materials and methods 2.2.1 Culture of Caco-2 colonic epithelial cell line

iii

2.2.2 Parasite culture 2.2.3 Confocal microscopy 2.2.4 Scanning electron microscopy 2.2.5 Fluorescent labeling of live Blastocystis parasites with CFSE 2.3 Results 2.3.1 Apical attachment of Blastocystis to epithelial cells 2.3.2 Determination of Blastocystis attachment to Caco- 2 cell monolayer by Scanning Electron Microscopy 2.3.3 Development of an adhesion assay with CFSE staining for Blastocystis 2.4 Discussion

CHAPTER 3: FACTORS INVOLVED IN BLASTOCYSTIS 54-81 ADHESION TO ENTEROCYTES 3.1 Introduction 3.2 Materials and methods 3.2.1 Culture of Caco-2 colonic epithelial cell line 3.2.2 Parasite culture 3.2.3 Coincubation with different modulators and attachment assay 3.2.4 Statistical analysis 3.3 Results 3.3.1 Time dependent variations in attachment 3.3.2 Dose response studies 3.3.3 Effects of temperature, rotational shaking, fixation and trypsinization on Blastocystis attachment 3.3.4 Effects of inhibition of cytoskeletal components, parasite cysteine proteases and lipid raft on attachment 3.3.5 Effects of saccharides on Blastocystis adhesion to Caco-2 cells

3.4 Discussion

CHAPTER 4: HOST RESPONSES PART I: EFFECTS OF 82-111 BLASTOCYSTIS ON APICAL JUNCTIONAL COMPLEX AND EPITHELIAL BARRIER FUNCTION 4.1 Introduction 4.2 Materials and methods 4.2.1 Culture of Caco-2 colonic epithelial cell line 4.2.2 Parasite culture 4.2.3 Epithelial permeability 4.2.4 Western blot and cellular fractionation 4.2.5 Confocal microscopy 4.3 Results 4.3.1 Blastocystis induced time- and dose- dependent permeability increase in Caco-2 4.3.2 Effect of Blastocystis on the overall expression of the apical junctional complexes 4.3.3 Effect of Blastocystis on the distribution of the tight junction proteins by cellular fractionation 4.3.4 Effect of Blastocystis on the distribution of the tight junction proteins 4.3.5 Loss of co-localization between tight junction proteins and F-actin after exposure to Blastocystis 4.3.6 Possible association of host tight junction protein Claudin-1 with adhering Blastocystis parasites 4.4 Discussion

CHAPTER 5: HOST RESPONSES PART II: ROLE OF 112-133 BLASTOCYSTIS-INDUCED HUMAN ENTEROCYTE APOPTOSIS IN MEDIATING BARRIER DYSFUNCTION 5.1 Introduction

5.2 Materials and methods 5.2.1 Culture of Caco-2 colonic epithelial cell line 5.2.2 Parasite culture and lysate 5.2.3 Epithelial permeability 5.2.4 Flow cytometry 5.2.5 Confocal microscopy 5.2.6 Western blot 5.2.7 Statistical analysis 5.3 Results 5.3.1 ST-4 (WR-1) strain could not establish enteroadhesion and cause disruption of tight junction, in contrast to the enteroadhesive strain of Blastocystis ST-7 (B) 5.3.2 Blatocystis ST-7 (B) induces characteristics of apoptosis in Caco-2 cells; ST-4 (WR-1) does not 5.3.3 Blastocystis ST-7 (B) induces increase in permeability to FITC-conjugated Dextran; ST- 4 (WR-1) does not 5.3.4 Inhibition of caspase prevented Blastocystis ST-7-induced ZO-1 rearrangement and epithelial barrier dysfunction in Caco-2 5.4 Discussion

CHAPTER 6: STRAIN-DEPENDENT VARIATIONS IN 134-176 ADHESION AND ITS POSSIBLE ROLE IN MEDIATING BARRIER DYSFUNCTION 6.1 Introduction 6.2 Materials and methods 6.2.1 Parasite culture 6.2.2 Culture of Caco-2 colonic epithelial cell line 6.2.3 Parasite viability assay 6.2.4 Growth curve

6.2.5 Azocasein assay for cysteine protease activity 6.2.6 Adhesion assay 6.2.7 Epithelial permeability 6.2.8 Western blot 6.2.9 Confocal microscopy 6.2.10 Drug preparation 6.2.11 Statistical analysis 6.3 Results 6.3.1 Blastocystis exhibits intra-and inter- subtype variation in ability to induce epithelial permeability increase in human epithelium 6.3.2 Blastocystis exhibits intra-and inter-subtype variability in ability to disrupt epithelial tight junction proteins 6.3.3 Blastocystis isolates exhibit strain-to-strain variation in attachment to human enterocytes 6.3.4 Blastocystis exhibits extensive intra- and inter-subtype variation in cysteine protease activity 6.3.5 Blastocystis ST-7 isolates exhibit extensive variations in resistance to Mz and the level of Mz resistance is inversely correlated with an ability to attach and induce a permeability increase 6.3.6 Mz resistance in Blastocystis does not lead to decrease in cell proliferation in Blastocystis ST-7 6.3.7 Blastocystis ST-7 isolates exhibit extensive variations in resistance to nitrosative stress 6.4 Discussion

CHAPTER 7: APPLICATION OF THE ADHESION 177-194 MODULATOR IN BLASTOCYSTIS INFECTION 7.1 Introduction 7.2 Materials and methods

vii

7.2.1 Parasite culture 7.2.2 Culture of Caco-2 colonic epithelial cell line 7.2.3 Attachment assay 7.2.4 Western blot 7.2.5 Statistical analysis 7.3 Results 7.3.1 Displacing effect of galactose on Blastocystis adhesion to Caco-2 cells 7.3.2 Effect of the cytadherence inhibitor galactose on prevention of Blastocystis–induced tight junction degradation in Caco-2 cell monolayers 7.3.3 Evaluation of galactose on the adhesion of the highly adhesive Blastocystis ST-7 (H) isolate to Caco-2 cells: interference with adhesion, displacing effects as well as rescue of tight junction degradation 7.4 Discussion

CHAPTER 8: GENERAL DISCUSSIONS 195-202 8.1 Discussion

REFERENCES 203-253

viii

SUMMARY

Blastocystis is an emerging enteric protistan pathogen that induces intestinal disorders in a range of hosts including humans. Despite being discovered more than 100 years ago, it is difficult to argue the clinical significance and pathogenic potential of Blastocystis. Recent studies have shed light on its potential virulence factors and host-pathogen interactions; however, there have been major gaps in our understanding of its basic biology. Adhesion of

Blastocystis to host cells, as the first step to establish and maintain colonic infection, has never been studied. As an extracellular parasite, Blastocystis need to attach to avoid clearance by peristalsis. It would be important to investigate the ability of Blastocystis to attach and the factors that influence the attachment.

The aim of this study was to investigate the pathobiology of Blastocystis adhesion and study the role of adhesion in initiating pathogenesis. Here the study reports that a metronidazole resistant (Mzr) strain of Blastocystis was able to attach to human intestinal epithelial cells and attachment was relat ed to time, dose and temperature. By establishing an adhesion assay, this study made an extensive analysis of extrinsic and intrinsic factors influencing host- parasite adhesion and revealed that galactose-binding protein on the surface of the parasite was important in mediating parasite attachment.

Following attachment to the enterocytes, Blastocystis produces pathogenic effects on the host cells. The host responses upon adhesion were also characterized. A comprehensive analysis of the apical junctional complexes showed that the parasite altered the integrity of the tight junctions and

ix transferred Blastocystis legumain to the apical side of enterocytes. Cleavage of tight junction proteins caused a disruption in the colonic epithelial barrier and increased intestinal permeability, which might be related to the characteristic diarrhea observed in patients suffering from Blastocystis infections. The parasites could also induce host cell apoptosis in a caspase 3- and 9- dependent manner, and interestingly, pretreatment of Caco-2 monolayers with a pan-caspase inhibitor z-VAD-fmk, significantly inhibited the changes in tight junction protein and intestinal permeability increase. This suggested a role for enterocyte apoptosis in Blastocystis-mediated epithelial barrier compromise in the human intestine.

Diverse intestinal symptoms with Blastocystis infections have confused clinicians and microbiologists alike. While a large number of infected individuals present with clinical symptoms, asymptomatic carriage of the parasite is also common. Moreover, in symptomatic patients, the duration and severity of symptoms vary from acute enteritis to chronic, mild diarrhea.

Strain-to-strain variations in parasite pathogenicity have been proposed to contribute to this inconsistency. The study tested whether strain-to-strain variations in enteroadhesion may account for the differences in pathogenicity, and subsequently different outcomes of infection. The study showed that intra- and inter-subtype variations in cytopathogenicity were largely dependent on their adhesive properties, highlighting a major role of adhesion in pathogenesis. The study also showed impairment of parasite adhesion and consequently virulence as a possible fitness cost of drug resistance in

Blastocystis strains. Metronidazole resistant (Mzr) strains were found to be less pathogenic owing to compromised adhesion. In addition, enhanced

x tolerance of nitrosative stress was found in adhesive strains, suggesting parasites evolved for better survival fitness in adhesion as well as the defence against host immunity.

Another issue complicating the pathogenic potential of Blastocystis is reports of treatment failure. Although metronidazole is considered first-line therapy, physicians are often skeptical about prescribing antibiotics for Blastocystis infections due to frequent reports of non-responsiveness to chemotherapy.

There is a need to design alternative or adjunctive treatment options against antibiotic resistant Blastocystis infections. As attachment is crucial for

Blastocystis survival and important for its pathogenicity, and thus it would be an appealing drug target. Modulators that prevent robust attachment may allow Blastocystis to be flushed from the intestinal lumen by host peristalsis, and ultimately result in reduction in infection. The inhibition studies with the adhesion assay revealed that galactose significantly interfered with attachment and when applied in barrier dysfunction study, galactose was shown to rescue the degradation of tight junction protein ZO-1 induced by Blastocystis.

Interestingly, galactose was also able to dislodge adhered parasites from the host cells, suggesting potential application of galactose-based therapy in

Blastocystis infections.

Altogether, the findings from the study identified the importance of adhesion in Blastocystis pathogenesis, and provided a possible explanation for the cellular basis of strain-to-strain variations in pathogenicity, advanced our understanding of the mechanism of attachment and also pointed towards new directions for therapeutic interventions in Blastocystis infections.

LIST OF TABLES

Table 1.1 Clinically evaluated antibiotics for Blastocystis 24 infections

Table 3.1 Host or parasite structures or molecules mediating 57 attachment in common protistan parasites

Table 3.2 Effects of different treatments on attachment of 69 Blastocystis to Caco-2 cells (targeting host cells cytoskeleton, parasite proteases and lipid raft structure on both host and parasites)

Table 4.1 Mechanisms of modulating tight junction barrier by 87 some protistan parasites

Table 5.1 Studies on induction of apoptosis in protistan parasites 115 and their role in mediating barrier function

Table 6.1 Comparisons of effects of luminal pathogens/ toxins 146 on intestinal permeability increase

Table 6.2 IC50s of Mz and nitric oxide donors against 160 Blastocystis ST-7 isolates (µg/ml) Table 6.3 Correlation of Mz resistance with attachment, 161 permeability increase and nitric oxide IC-50 in Blastocystis ST-7

Table 7.1 Microbial carbohydrate receptors on the intestinal 180 microvillus membrane

xii

LIST OF FIGURES

Fig. 1.1 Phase contrast micrographs representing four 9 morphological forms of Blastocystis

Fig. 1.2 Proposed life cycle for Blastocystis suggesting the 11 existence of zoonotic genotypes (subtypes 1 to 7) with various host specificities

Fig. 2.1 Confocal differential interference contrast 43 micrographs of Blastocystis in contact with Caco-2 cells

Fig. 2.2 Fluorescence confocal micrographs of Blastocystis 44 in contact with Caco-2 cells

Fig. 2.3 Scanning electron micrographs of Blastocystis 46 attached to Caco-2 monolayers

Fig. 2.4 TEM semi-thin sections illustrating Blastocystis on 46 top of the epithelium

Fig. 2.5 Representative confocal micrograghs showing live 48 Blastocystis parasites after CFSE staining

Fig. 2.6 Representative confocal micrograghs showing 49 CFSE-labeled Blastocystis parasites adhering to Caco-2 cells

Fig. 2.7 Representative confocal micrograghs showing 50 CFSE-labeled Blastocystis parasites adhering to Caco-2 cells at a lower magnification

Fig. 3.1 Time course studies of Blastocystis adhesion to 62 Caco-2 monolayers

Fig. 3.2 Dose-response of Blastocystis adhesion to Caco-2 63 monolayers

Fig. 3.3 Effect of temperature on Blastocystis adhesion to 65 Caco-2 cells

Fig. 3.4 Effect of formaldehyde fixation on Blastocystis 66 adhesion to Caco-2 cells

Fig. 3.5 Effect of trypsinization on Blastocystis adhesion to 66 Caco-2 cells

xiii

Fig. 3.6 Effect of saccharides on adhesion of Blastocystis to 73 Caco-2 cells

Fig. 3.7 Dose-dependent inhibiton of galactose on 74 Blastocystis adhesion to Caco-2 monolayers

Fig. 3.8 Effects of pretreating either parasites or host cells 75 with galactose on Blastocystis adhesion to Caco-2 cells

Fig. 4.1 Time-dependent effect of Blastocystis ST-7 on 91 epithelial permeability of Caco-2 cell monolayers

Fig. 4.2 Dose-dependent effect of Blastocystis ST-7 on 92 epithelial permeability of Caco-2 cell monolayers

Fig. 4.3 Western blots showing distinct effects on tight 93 junction and adherens junction proteins by apical application of Blastocystis ST-7 (B)

Fig. 4.4 Representative Western blots showing the changes 95 in tight junction proteins and densitometric analysis of protein expressions in the insoluble fraction and soluble fraction of Caco-2 cells after treatment by Blastocystis ST-7 (B)

Fig. 4.5 Representative confocal z-projections showing 97 effects of Blastocystis exposure on the distribution of tight junction protein Occludin

Fig. 4.6 Representative confocal micrographs illustrating 98 ZO-1 integrity in Caco-2 monolayers

Fig. 4.7 Representative confocal micrographs with z- 99 projections showing the redistribution of Claudin-1 tight junction protein after 6h exposure to Blastocystis ST-7 (B)

Fig. 4.8 Representative confocal micrographs illustrating 101 Occludin and F-actin double staining in Caco-2 monolayers

Fig. 4.9 Co-imunostaining of tight junction protein ZO-1 102 and Claudin-1 in Caco-2 cells after 6h exposure to Blastocystis ST-7 (B)

Fig. 4.10 Representative confocal micrographs illustrating 104 Claudin-1 staining and deposit of Blastocystis legumain in Caco-2 monolayers after 6h exposure to Blastocystis ST-7 (B)

xiv

Fig. 4.11 Representative confocal micrographs illustrating the 105 spatial relationship of Claudin-1 tight junction protein and Blastocystis in Caco-2 monolayers after 6h exposure to Blastocystis ST-7 (B)

Fig. 5.1 Representative confocal micrographs illustrating 119 attachment of Blastocystis to Caco-2 monolayers resulted in changes in ZO-1 tight junction protein

Fig. 5.2 Flow cytometry analysis of annexin V-FITC and 121 propidium iodide staining

Fig. 5.3 Representative fluorescence micrographs after 122 DAPI staining showing apoptosis of Caco-2 cells

Fig. 5.4 Western blot analysis of caspase activation 123 (cleavage) in Caco-2 cells

Fig. 5.5 Transmission electron micrograph showing 124 ultrastructural changes of Caco-2 cell monolayer induced by Blastocysitis treatment

Fig. 5.6 Transmission electron micrograph showing 125 ultrastructural changes of Caco-2 cell monolayer induced by Blastocysitis treatment

Fig. 5.7 Representative confocal micrographs illustrating 127 ZO-1 integrity in Caco-2 monolayers

Fig. 5.8 Western blot analysis of ZO-1 integrity in Caco-2 128 epithelium

Fig. 5.9 Inhibition of Caco-2 caspases prevented 129 Blastocystis ST-7-induced epithelial barrier dysfunction

Fig. 6.1 Blastocystis ST-4 and ST-7 strains used in the study 139

Fig. 6.2 Intra-and inter- subtype variation in Blastocystis in 145 inducing permeability increase

Fig. 6.3 Differential effects on tight junction protein 149 degradation in Caco-2 cell monolayers by different strains of Blastocystis

Fig. 6.4 Blastocystis exhibits inter- and intra-subtype 153 variation in attachment to Caco-2 cells

Fig. 6.5 Relationship between attachment and permeability 154 increase by Blastocystis ST-4 and ST-7 parasites

Fig. 6.6 Blastocystis ST-7 (H) attaches to an intestinal 155 monolayer

Fig. 6.7 Intra- and inter-subtype variations in protease 158 activity in Blastocystis

Fig. 6.8 Blastocystis exhibits intra-subtype variation in 160 susceptibility and resistance to Mz

Fig. 6.9 Correlation analyses between Mz resistance and (A) 161 attachment, as well as (B) permeability increase in Blastocystis ST-7 parasites

Fig. 6.10 Mz resistance in Blastocystis does not lead to 162 decreased growth in Blastocystis ST-7

Fig. 6.11 Mz resistant isolates in Blastocystis ST-7 exhibit 165 susceptibility to nitric oxide

Fig. 6.12 Differential effects on myosin light chain 169 phosphorylation in Caco-2 cell monolayers by different strains of Blastocystis

Fig. 6.13 Differential consumption rate of glucose by 173 different strains of Blastocystis

Fig. 6.14 Summary model illustrating pathogenesis and 176 fitness cost in Blastocystis

Fig. 7.1 Displacement of Blastocystis from Caco-2 cells by 184 galactose

Fig. 7.2 Galactose prevented Blastocystis–induced ZO-1 186 tight junction degradation in a dose-dependent manner

Fig. 7.3 Dose-dependent inhibiton of galactose on 188 Blastocystis ST-7 (H) adhesion to Caco-2 monolayers.

Fig. 7.4 Displacement of Blastocystis ST-7 (H) from Caco-2 189 cells by galactose

Fig. 7.5 Galactose rescued Blastocystis ST-7 (H)–induced 190 ZO-1 tight junction degradation

xvi

LIST OF SYMBOLS

°C Degree Celsius

99mTc-DTPA Technetium-99m labeled diethyl

triamine penta acetic acid

AJ Adherens junctions

AJC Apical junctional complex

AIDS Acquired immune deficiency

syndrome

ANOVA Analysis of variance

APE-RR Carbobenzyloxy-Ala-Ala-AAsn-

epoxycarboxylate, ethyl ester

ATP Adenosine triphosphate

BD Bis die/ Twice daily

CFSE Carboxyfluorescein succinimidyl ester

CoA Coenzyme A

Cv Coefficient of variance

CP Cysteine protease

Cyt-D Cytochalasin D

DAPI 4',6-diamidino-2-phenylindole

DIC Differential interference contrast

DISC Death-inducing signalling complex

DMEM Dulbecco’s modified Eagle medium

DMSO Dimethyl sulfoxide

DTT Dithiothreitol

xvii

EF Elongation factor

ELISA Enzyme-linked immunosorbent assay

EPEC Enteropathogenic E. coli

FECT Formalin Ether Concentration

Technique

FITC Fluorescein isothiocyanate

FlavoHb Flavohemoglobin

Fuc Fucose

FUR Furazolidone

Gal Galactose

Gal/GalNAc Galactose-N-acetylgalactosamine

Glu Glucose

GM-CSF Granulocyte macrophage colony

stimulating factor

GOS Galacto-oligosaccharides

GSNO S-Nitrosoglutathione

GTP Guanosine triphosphate h Hour

HBMEC Human brain microvascular

endothelial cells

HIV Human immunodeficiency virus

HMG CoA 3-hydroxy-3-methyl-glutaryl-CoA

HMO Human milk oligosaccharides

IA Iodoacetamide

IBS Irritable bowel syndrome

xviii

IC Inhibitory concentration

IFN Interferon

Ig Immunoglobulin

IL Interleukin

IMDM Iscove’s modified Dulbecco’s medium iNOS Inducible nitric oxide synthase

IS Insoluble fraction

LPG Lipophosphoglycan

LPPG Lipophosphopeptidoglycans

LSEC Liver sinusoidal endothelial cells

Man Mannose

MAb Monoclonal antibody

MBCD Methyl-b-cyclodextrin

Min Minute ml Milliliter

MLC Myosin light chain

MLCK Myosin light chain kinase

MLCp Phophorylated myosin light chain

MLO Mitochondria like organelle mM Millimolar

MV Multivacuolar

Mz Metronidazole

Mzr Metronidazole-resistant

Mzs Metronidazole-susceptible

NADH Nicotinamide adenine dinucleotide

xix

NaOH Sodium hydroxide

NF-κB Nuclear factor kappa-light-chain-

enhancer of activated B cells

NHP Non human primates

NI Nitroimidazole nM Nanomolar nm Nanometer

NO Nitric oxide

NS Not susceptible

NTZ Nitazoxanide

OD Once daily

Oz Ornidazole

PAR-2 Proteinase activated receptor 2

PAR Paromomycin

PBS Phosphate buffered saline

PCR Polymerase chain reaction

Pen Strep Penicillin Streptomycin

PFOR Pyruvate:Ferredoxin oxidoreductase

PI Propidium iodide

PNO Pyruvate:ferredoxin

oxidoreductase/NADPH-cytochrome

P450 reductase

PS Phosphatidylserine

PYR Pyrimethamine

R Pearson Correlation

RAU Relative absorbance unit

RFU Relative fluorescence unit

ROCK Rho associated kinase

RT-PCR Reverse transcription PCR

S Soluble fraction

SEM Scanning electron microscopy

SIgA Secretory immunoglobulin A

SMZ Sulfamethoxazole

SNAP S-nitroso-N-acetyl-l,l-penicillamine ssrRNA Small subunit RNA

ST Subtype

TEM Transmission electron microscopy

TER Transepithelial resistance

TDS Ter die sumendum/ three times a day

TJ Tight Junction

TLCK N-Tosyl-L-Lysine chloromethyl

ketone

TMP Trimethoprim

TPCK N - Tosyl - Phenylalanine chloromethyl

ketone

Tz Tinidazole

V Vacuole

XIVC Xenic in vitro culture

ZO-1 Zonula occludens-1

xxi

LIST OF PUBLICATIONS

A. International Refereed Journals

1. Wu Zhaona, Haris Mirza, Kevin SW Tan. Intra-subtype variation in enteroadhesion accounts for differences in epithelial barrier disruption and is associated with metronidazole resistance in Blastocystis subtype-7. (2014) PLoS Negl Trop Dis 8: e2885.

2. Wu Zhaona, Haris Mirza, Joshua DW Teo, Kevin SW Tan. Strain- dependent induction of human enterocyte apoptosis by Blastocystis disrupts epithelial barrier and ZO-1 organization in a caspase 3- and 9- dependent manner. (2014) Biomed Res Int 2014: 209163.

3. Haris Mirza, Wu Zhaona, Joshua DW Teo, Kevin SW Tan. Statin pleiotropy prevents rho kinase-mediated intestinal epithelial barrier compromise induced by Blastocystis cysteine proteases (2012). Cellular Microbiology, 14: 1474-1484

4. Haris Mirza, Wu Zhaona, Fahad Kidwai, Kevin SW Tan. A metronidazole-resistant isolate of Blastocystis spp. is susceptible to nitric oxide and downregulates intestinal epithelial inducible nitric oxide synthase by a novel parasite survival mechanism (2011). Infection and Immunity, 79: 5019-5026.

B. Manuscripts Under Preparation

1. Wu Zhaona, Kevin SW Tan. In vitro model of attachment of Blastocystis to Caco-2 cells, a human intestinal cell line.

xxii

C. Conference Papers

1. Wu Zhaona, Haris Mirza, Ng Geok Choo, Kevin SW Tan. Trade-off

between drug resistance and survival fitness-Impaired epithelial

attachment and nitric oxide tolerance in metronidazole resistant

isolates of Blastocystis, 5th International Conference on Anaerobic

Protists, 6th – 9th September, 2012,University of California, Los

Angeles, USA

2. Wu Zhaona, Haris Mirza, Tao Fei, Joshua DW Teo, Kevin SW Tan.

Strain-dependent induction of human enterocyte apoptosis by

Blastocystis disrupts epithelial barrier and ZO-1 organization in a

caspase 3- and 9-dependent manner. Mitochondria, Apoptosis and

Cancer Conference (MAC), 27th – 29th October, 2011, National

University of Singapore, Singapore

3. Wu Zhaona, Haris Mirza, Joshua D. W. Teo, Kevin S.W. Tan.

Statins as a potential drug candidate in Blastocystis-induced intestinal

epithelial barrier dysfunction. ID-IRG Annual Workshop, Singapore-

MIT Alliance for Research and Technology (SMART), 2011,

Singapore

4. Kevin S.W. Tan, Haris Mirza, Wu Zhaona, Joshua D. W. Teo.

Updates on Blastocystis: Drug Resistance Detection and

Pathobiology. Paper presented at the 17th Japanese-German

Cooperative Symposium on Protozoan Diseases and Workshop of

Gastrointestinal Protozoan Diseases, 12-16 September, 2011, Nara

Women’s University, Nara, Japan. (Invited Paper)

xxiii

CHAPTER 1:

INTRODUCTION

1.1 INTRODUCTION

Blastocystis is a unicellular, genetically diverse protist, phylogenetically placed within the Stramenopiles (Scanlan, 2012) and it is the only member of this group associated with human pathology (Tan, 2008, Tan et al., 2010). It is a species complex comprising of 17 subtypes (STs) of which at least 9 are found in humans (Alfellani et al., 2013b). The prevalence of this parasite ranges from 1.5% to 15% in developed countries, and from 30% to 50% in developing countries (Tan, 2008). It is frequently reported in human fecal samples of symptomatic patients as well as healthy individuals (Stenzel et al.,

1996, Tan, 2004). The parasite induces general symptoms of intestinal inflammation such as mucous and watery diarrhea, bloating, abdominal pain and vomiting (Tan et al., 2010). Clinical studies also associate Blastocystis with intestinal and dermatological inflammatory disorders such as irritable bowel syndrome and urticaria (Garavelli et al., 1988, Kurniawan et al., 2009,

Llibre et al., 1989, Tasova et al., 2000), suggesting Blastocystis may be an opportunistic pathogen.

Despite being discovered more than 100 years ago (Alexeieff, 1911, Brumpt,

1912), it is difficult to argue the clinical significance and pathogenic potential of Blastocystis (Clark et al., 2013). Limitations of diagnostic techniques, variable phenotypes and genotypes of Blastocystis, large number of asymptomatic carriers, and frequent reports of treatment failure are blamed for most of the controversies surrounding the clinical management of the parasite infection (Tan et al., 2010). In recent years, significant advances have been made in the knowledge of its molecular and cellular biology (Clark et al.,

2013, Denoeud et al., 2011, Tan et al., 2010, Wu et al., 2010), as well as pathogenic mechanisms (Mirza et al., 2011b, Mirza et al., 2012b). However, many controversies remain unresolved regarding the life cycle, molecular biology and pathogenicity of the organism.

1.2 CLASSIFICATION

Blastocystis has been a taxonomic enigma since its initial description more than 100 years ago. The organism was initially classified as the cyst of a flagellate, vegetable, yeast, and fungus (Zierdt, 1991). The genus name

“Blastocystis” was coined by Alexeieff while Brumpt provided the species name, “hominis” (Zierdt, 1991). It was subsequently reclassified as a protist by

Zierdt and colleagues based on its morphology and phenotypic properties such as the presence of one or more nuclei, cellular organelles like mitochondria,

Golgi apparatus and endoplasmic reticulum, as well as its inabilty to grow on fungal medium, resistance to antifungal agents and susceptibility to antiprotozoal drugs (Zierdt et al., 1967). Blastocystis was later classified as a sporozoan and then as a sarcodine (Zierdt, 1991).

An earlier analysis of Blastocystis partial small-subunit rRNA (ssrRNA) showed that Blastocystis sp. is not monophyletic with the yeasts, fungi, sarcodines, or sporozoans or even yeast and fungi (Johnson et al., 1989).

Later, the complete Blastocystis ssrRNA gene was sequenced and phylogenetic analysis suggested that the parasite should be classified within the Stramenopiles (also known as Heterokonta) (Silberman et al., 1996).

Further analysis of Blastocystis ssrRNA, cytosolic-type 70-kDA heat shock protein, translation elongation factor 2, and the non-catalytic ‘B’ subunit of vacuolar ATPase confirmed that it is indeed a Stramenopile (Arisue et al.,

2002). In general, the stramenophiles have flagella surrounded by lateral hair like mastigonemes, but Blastocystis lacks a flagella and is non-motile

(Silberman et al., 1996, Boorom et al., 2008).

In the past, the species name of Blastocystis was given based on the host from which it was isolated such as B. hominis from humans, B. ratti from rats etc..

This outdated classification may have resulted in confusion regarding specificity, cell biology and pathogenicity of the parasite. Based on published ssrRNA sequence analysis (Noel et al., 2005), it was proposed Blastocystis isolates should be characterized by molecular subtyping of ssrRNA and designated as Blastocystis sp. (Stensvold et al., 2007b). More recently, 17 subtypes have been reported till date (Stensvold, 2013).

1.3 CELL BIOLOGY

Blastocystis is a polymorphic protist, and four major forms have been reported from in vitro culture and fecal samples (vacuolar, granular, amoeboid and cyst) (Fig. 1.1). The extensive heterogeneity of various forms of Blastocystis has led to the misinterpretation of findings from different studies (Tan, 2008).

There is limited information available on the physiological relevance of these forms. Mechanisms involved in transition of the parasite from one form to other are not known either.

1.3.1 Vacuolar form

The central vacuole form, also referred to as vacuolated or central body form, is the most frequently observed form in laboratory culture and in stool samples. It is spherical and may display large size variations, ranging from 2 to 200 μm (average of 4 to 15 μm) (Stenzel et al., 1996). The vacuolar form is characterized by a large central vacuole that occupies approximately 90% of the cell’s volume (Fig. 1.1) (MacPherson et al., 1994, Stenzel et al., 1996). In electron micrographs the peripheral cytoplasm appears to contain cellular organelles like nucleus, mitochondria-like organelle, endoplasmic-reticulum and Golgi bodies (Zierdt, 1973). Multiple nuclei can be seen in Blastocystis and an average of four nuclei is common (Zierdt, 1973). A surface coat or fibrillar layer of varying thickness often surrounds the organism. The exact role of the surface coat is not understood but it has been suggested to play a role in trapping and degrading bacteria for nutrition (Zaman et al., 1997,

Zaman et al., 1999) and protecting against osmotic shock (Cassidy et al.,

1994). The surface coat appears to be thicker in parasite freshly isolated from

6 stools, but it reduces with prolonged in vitro culturing (Cassidy et al., 1994).

For the central vacuole in Blastocystis, its exact function is currently unclear but it appears to act as a storage organelle (Puthia et al., 2008, Tan, 2008). It was suggested that the central vacuole may act as a repository for carbohydrates and lipids required for cell growth (Yoshikawa et al., 1996). A few studies also suggest the role of central vacuole as a depository for apoptotic bodies during Blastocystis programmed cell death (Tan et al., 2005,

Yin et al., 2010). Recently, the Blastocystis central vacuole was found to contain cysteine proteases (Puthia et al., 2008), which play an important role in parasites virulence and cell survival (Puthia et al., 2005, Puthia et al., 2008,

Mirza et al., 2011a, Mirza et al., 2012b).

1.3.2 Granular form

The granular form of Blastocystis is morphologically identical to the vacuolar form. The only differentiating characteristic is the presence of multiple granules, especially within the central vacuole (Fig 1.1). The granules might differ considerably in appearance and described as myelin-like inclusions, droplets of lipid, tiny vesicles or crystalline granules (Dunn et al., 1989). The size of this form ranges from 3-80 μm in diameter. Granular form is more frequently observed in nonaxenized, older, and antibiotic-treated cultures. The physiological role of this form is not clear, but it is suggested to play a role in parasite programmed cell death as well as storage (Tan, 2004).

1.3.3 Ameboid form

The amoeboid form (Fig 1.1) is the most infrequently observed Blastocystis form. These forms have been observed in antibiotic treated cultures, old

7 cultures or in fecal samples (Zierdt, 1973). The size is considerably smaller, ranging from 2.6-7.8 μm in diameter. There are conflicting reports about its description (Dunn et al., 1989, McClure et al., 1980). Amoeboid forms were also observed from Blastocystis colonies grown in soft agar (Chen et al.,

1997b, Tan et al., 1996b, Tan et al., 1996c), with a central vacuole, numerous

Golgi bodies and mitochondria. In contrast, Dunn et al. described amoeboid cells with extended pseudopodia and lysosome-like compartments containing ingested bacteria, yet lacked a central vacuole, a Golgi complex, a surface coat, and mitochondria (Dunn et al., 1989). The presence of bacteria and bacterial remnants within the amoeboid form suggested that this form may be phagocytic in nature.

1.3.4 Cyst form

The cyst form (Fig 1.1) is the most recently described form of the parasite.

The size ranges between 2-5 μm in diameter (Moe et al., 1996). It is protected by a multi-layered cyst wall which is sometimes covered with a loose surface coat (Moe et al., 1996). Unlike vacuolar and granular forms, It can survive up to 19 days in water at normal temperature, up to 1 month at 25°C (Moe et al.,

1996) and as long as 2 months at 4°C (Yoshikawa et al., 2004). A report of

Blastocystis cysts also suggests its resistance to common antiprotozoal agent metronidazole (Zaman et al., 1996). Therefore, this form is environmentally resistant and considered significant for the fecal-oral transmission of infection

(Tan, 2004).

Fig. 1.1 Phase contrast micrographs representing four morphological forms of Blastocystis. Vacuolar form (V), and Multivacuolar forms (MV), Scale bar = 10 μm; Granular form (G), Scale bar = 10 μm; Amoeboid form (arrow). Scale bar = 10 μm; Cyst (arrow), Scale bar = 5 μm

Adapted from Chen (1999).

1.4 LIFE CYCLE

There have been discrepancies over the life cycle of Blastocystis, which are due largely to the belief that Blastocystis exhibits multiple reproductive processes (Govind et al., 2002, Singh et al., 1995, Zhang et al., 2007). Modes of reproduction proposed such as schizogony, plasmotomy (budding) (Tan et al., 2007), endodyogeny (Zhang et al., 2007), and sac-like pouches (Govind et al., 2002, Suresh et al., 1997) are more likely due to the pleomorphic nature of the organism and not true modes of reproduction. Until proven otherwise, the only accepted mode of reproduction is binary fission (Tan, 2008).

Apart from humans, several animal hosts have been described for Blastocystis including rats, pigs, birds, reptiles and even insects. Blastocystis commonly exhibits zoonoses and to complete its life cycle through the fecal-oral route, the parasite appears to involve one of more host species (Tan, 2008). A modified life cycle (Fig. 1.2) must take into consideration the large reservoir of Blastocystis among various animal populations and that humans are potential hosts to numerous zoonotic genotypes (subtypes). Ingested cysts develop into vacuolar forms in the large intestine and later reproduce by binary fission. Encystation occurs during passage along the large intestines and is deposited in the feces (Chen et al., 1997a). Apart from a few studies, the transitions from one of the classically described forms to another are not well understood and remain to be elucidated.

Fig. 1.2 Proposed life cycle for Blastocystis suggesting the existence of zoonotic genotypes (subtypes 1 to 7) with various host specificities. Subtype 1 is cross infective among mammalian and avian isolates; subtypes 2, 3, 4, and 5 comprise primate/pig, human, cattle/pig, and rodent isolates, respectively; and subtypes 6 and 7 include avian isolates. The proposed scheme suggests that humans are potentially infected by seven or more species of Blastocystis and that certain animals represent reservoirs for transmission to humans. Recent studies have identified more subtypes.

Reproduced from Tan (2008).

1.5 CLINICAL PRESENTATION

Symptomatic infections with Blastocystis are called blastocystoses, which present with a variety of signs and symptoms (Tan et al., 2010), such as nonspecific intestinal symptoms (Mirza and Tan, 2012). Extraintestinal manifestations of Blastocystis infection including arthritis or skin disorders have also been described (Tejera et al., 2012, Vogelberg et al., 2010).

Intestinal Symptoms:

The most common symptom of blastocystosis is diarrhea with abdominal pain, with varying severity (Tan et al., 2010). Other nonspecific gastrointestinal symptoms include nausea, vomiting, constipation, dysentery, flatulence, bloating, anorexia and weight loss (Nagel et al., 2012, Stensvold et al., 2009c,

Tan et al., 2010). The duration and severity of symptoms are reported to be variable, ranging from mild and chronic diarrhea to acute gastroenteritis

(Mirza and Tan, 2012).

Although the parasite is frequently associated with intestinal symptoms, an unequivocal causal relationship between Blastocystis and gastrointestinal disorders has not been established thus far primarily because of a large number of asymptomatic carriers. Consequently, Koch’s postulates have not been fulfilled for Blastocystis (Coyle et al., 2012).

No consensus has been reached regarding the possible reasons for variation in intestinal symptoms. Some studies associate the severity of disease with parasite density (Tan et al., 2010). Significantly higher concentrations of

Blastocystis were found in symptomatic patients than in asymptomatic ones

(Kaya et al., 2007). More than 5 Blastocystis on every 40x magnification field in fecal samples was accepted as pathogenity criterion (Stenzel et al., 1996).

In a recent case of life-threatening toxic megacolon, >10 Blastocystis/field were observed (Salvi et al., 2012).

Extraintestinal Symptoms:

Interestingly, various reports associate Blastocystis infections with cutaneous lesions particularly urticaria (Katsarou-Katsari et al., 2008, Vogelberg et al.,

2010), suggesting host allergic response to some unknown parasitic factors.

Treatment of the parasite leads to resolution of both the infection and cutaneous lesions. These cutaneous manifestations are likely immune- mediated although the exact mechanism is presently unclear. Recently, a specific association has been found between acute urticaria and the amoeboid forms of Blastocystis subtype 3 (Katsarou-Katsari et al., 2008). The authors suggested several possible modes through which Blastocystis may give rise to such cutaneous symptoms, namely through a disruption to immune homeostasis as the host mounts an inflammatory response against the amoeboid forms of Blastocystis. Possible antigens might include carbohydrates found on its surface coat, leading to inflammatory cell recruitment, which release histamine-activating factors that prime mast cells and basophils. IgE might also be a possible mediator in urticaria through mast cell activation and degranulation, and subsequent release of inflammatory mediators. However, the mechanism by which a Blastocystis infection gives rise to such symptoms remains to be elucidated. The identification of

Blastocystis-specific IgEs in patients with urticaria would be useful in defining the mechanism by which the parasite causes these lesions.

1.6 LABORATORY DIAGNOSIS

One of the reasons for the controversial pathogenic potential of Blastocystis is lack of standardized diagnostic techniques. Therefore, establishing authentic methods for diagnosis would be crucial for the field to properly evaluate the clinical significance of the parasite.

As reviewed in the previous sections, Blastocystis is an extremely diverse complex and it should be taken into consideration during laboratory investigations (Tan et al., 2010). Information on the density, morphology, subtype, as well as antibiotic susceptibility is important for appropriate management of Blastocystis infections. Technical advances in recent years have made it possible to assess these parameters. However, a number of

Blastocystis infections go unreported or ignored owing to its controversial pathogenesis. Another challenge for diagnosis of Blastocystis is the presence of diverse morphological forms, making it one of the most difficult organisms to identify in stool samples. Diagnostic staff should be trained to identify various stages of the parasite, especially for cyst form because of their small size.

Conventional diagnostic methods to identify the parasite in stools include

FECT and XIVC, while tests based on serology and molecular biology are being used increasingly. Blastocystis antibiotic susceptibility testing is yet to be instituted in clinical diagnostics, but should be used in cases refractory to antibiotic treatment.

1.6.1 FECT

The formol ether concentration technique (FECT), though commonly employed to detect parasite ova and cysts, is not recommended for laboratory identification of Blastocystis, due to extremely poor sensitivity (Rene et al.,

2009, Stensvold et al., 2007a). Using only FECT to isolate Blastocystis might probably contributes to the underestimation of parasite burden in epidemiological studies (Tan et al., 2010).

1.6.2 XIVC

Trichrome-stained fecal smears following short-term xenic in vitro culture

(XIVC) exhibits superior sensitivity in parasite detection. However, it should be noted that due to difference in generation time between Blastocystis strains

(Zierdt et al., 1981), slower growing subtypes and mixed infections of slower- and faster growing subtypes may be missed by this technique. Another consideration for better diagnosis of Blastocystis infections is to take multiple stool samples over a period of time since the parasite shedding by humans is irregular (Suresh et al., 2004).

1.6.3 Molecular Diagnostics

A quantitative TaqMan PCR was developed for the detection of Blastocystis and shown to be applicable to genomic DNAs extracted directly from feces

(Stensvold et al., 2012a). Emerging data suggests association of certain

Blastocystis subtypes or strains with pathogenicity and antibiotic resistance

(Stensvold et al., 2009c, Stensvold et al., 2010). Therefore, it’s important to determine which subtype is present in the specimen for better clinical management of blastocystosis. For subtyping, most epidemiological studies used a 600 bp barcoding region within the 18S small subunit ribosomal DNA

(18S rDNA) (Scicluna et al., 2006). The amplification of this 18S rDNA region is followed by sequencing of the PCR product to assign isolates to specific STs. A real-time quantitative PCR approach based on SYBR green detection of double-stranded DNA was reported (Poirier et al., 2011) and was designed as a genus-specific PCR targeting the SSU rRNA gene, enabling amplification of DNAs from Blastocystis strains. A recent study identified a single-copy subtyping rDNA marker in the genome of the mitochondria-like organelles (MLOs) and showed that it could be used not only for subtyping, but also to discriminate different isolates within a same ST (Poirier et al.,

2014).

1.6.4 Serology

There have only been a handful of reports employing serological techniques in identification of Blastocystis infections. Some studies involving enzyme- linked immunosorbent assay for Blastocystis correlate antibody titers with symptoms (Hussain et al., 1997, Mahmoud et al., 2003). Since serological approaches provide rapid, sensitive and quantitative detection of microbe- specific antibodies, they could be powerful tools for diagnosis of the parasite

(Tan et al., 2010). Development of such assays however should take into consideration the genetic diversity of Blastocystis. A panel of monoclonal antibodies against specific subtypes would be useful for diagnosis of

Blastocystis from stool specimens (Tan et al., 2010).

1.6.5 Antibiotic Susceptibility Testing

Chronic Blastocystis infections are often nonresponsive to conventional antibiotic regimens (Stensvold et al., 2010). Antibiotic susceptibility data, therefore, would be useful to manage such infections. Classical Blastocystis

16 antibiotic susceptibility evaluation utilized serial dilution or radioisotope inclusion methods (Dunn et al., 1991, Yakoob et al., 2004, Zierdt et al., 1983).

These techniques were expensive, time consuming, and often injurious to health. Recent development of redox-based assays (Mirza et al., 2011a), as well as improvements in anaerobic microculture methods, have made it possible to efficiently evaluate Blastocystis antibiotic susceptibilities within

24h (Mirza et al., 2011a). Redox-based assays are cheap, require a small parasite density, and could be adapted on high-throughput platforms.

Furthermore, these assays do not require sophisticated equipment and are ideal for use in resource-poor healthcare facilities (Mirza et al., 2011a). However, most of these assays were adapted for the vacuolar morphological form of the parasite, which is the most commonly reported form (Dunn et al., 2012, Mirza et al., 2011a). It is believed that cysts, which are resistant to metronidazole

(Zaman et al., 1996), and not the vacuolar form represent the infectious phenotype of the parasite (Tan, 2008). For more accurate assessment and management of the parasite antibiotic resistance, future studies should focus on adapting antibiotic susceptibility assays for non-vacuolar morphological forms of Blastocystis.

1.7 EPIDEMIOLOGY AND PREVALENCE

Despite the discrepancy in diagnostic methods used, there has been a renewed interest in Blastocystis epidemiology (Tan, 2008), and considerable numbers of the reports are now available for geographical and demographic distribution of Blastocystis infection in human (Yoshikawa, 2012).

1.7.1 Geographic distribution

Prevalence of human Blastocystis infection is different among countries or among different regions in the same countries. In general, prevalence of

Blastocystis is closely related to socio-economic development and higher in developing countries than in developed countries (Yoshikawa, 2012). The higher prevalence in developing countries is due to the poor sanitary infrastructure and/or low level of care to environmental hygiene in developing countries, because this parasite is known to be infected by the ingestion of the cyst form via fecal–oral route (Tan, 2008). High prevalence in human infection of the parasite was observed in developing countries, even using routine diagnostic technique in epidemiological survey. These results indicate that real prevalence of these countries is higher than that of reported data.

Although the prevalence in developed countries is generally lower than developing countries, several countries such as Turkey (32.5%) and USA

(22.9%) show relatively high prevalence. In contrast, the prevalence in

Singapore (3.3%) and Japan (2.5%) is very low among developed countries even using the most sensitive method in vitro culture.

1.7.2 Genotype distribution

Genotyping studies have highlighted the diversity in Blastocystis in humans and other animal hosts (Parkar et al., 2007, Stensvold et al., 2009a). Of the nine definite STs detected in humans, only four are common – ST1, ST2, ST3, and ST4. Together, these make up around 90% of all human Blastocystis in surveys that involve subtyping (Alfellani et al., 2013b). ST3 is the most common ST in humans worldwide, and its occurrence is a frequent finding in analyses of ST distribution, irrespective of the geographic origin of the population (Forsell et al., 2012, Malheiros et al., 2011, Meloni et al., 2011,

Nagel et al., 2012), supporting the idea that subtype distribution does not vary much between human populations from different geographical locations (Tan,

2008). The other STs are only sporadically reported in humans and may well prove to be the result of zoonotic transmission, as they are mostly much more common in nonhuman hosts. Though rare in humans, ST5 seems to be frequently found in livestock, particularly cattle, pigs, sheep, and camels

(Stensvold et al., 2009a). STs 6 and 7 have been detected primarily in ground- dwelling birds, with only single ST7 samples from a goat and an NHP having been found in nonhuman mammalian hosts (Alfellani et al., 2013a). Other than in humans, ST8 appears to be restricted to arboreal NHPs from Asia and

South America—it has not been reported from African NHPs (Alfellani et al.,

2013a). The STs with numbers above 9 are, as far as is known at present, confined to nonhuman hosts (Clark et al., 2013).

In the recent literature, it is still matter of debate whether disease is ST related.

There have been reports suggesting that pathogenic strains belong to subtypes

1, 4 and 7 (Dominguez-Marquez et al., 2009, Eroglu et al., 2009, Stensvold et

19 al., 2009c, Stensvold et al., 2009b) while subtypes 2 and 3 are likely to be non-pathogenic (Eroglu et al., 2009, Stensvold et al., 2009b). However, mixed subtype infections (Nagel et al., 2012) and high degree of genetic (Stensvold et al., 2012b) and pathobiological diversity within subtypes (Mirza and Tan,

2009) complicate the subtype-dependent pathogenecity hypothesis. Numerous studies suggested the existence of pathogenic and non-pathogenic variants in different STs (Dogruman-Al et al., 2009, Hussein et al., 2008, Souppart et al.,

2009). All these observations suggested that Blastocystis infection is likely not due to a particular ST even if some STs are frequently dominant in epidemiologic studies but rather to a conjunction of factors in the course of infection including environmental risk (transmission routes and contamination sources) and parasite (pathogenic and zoonotic potential) and host factors

(genotype, immunity, and age) (Souppart et al., 2010). Further studies should take these parameters into account.

1.8 HIGH RISK POPULATION

Blastocystis was considered to be an opportunistic pathogen and a number of reports suggested that certain populations might be more susceptible to

Blastocystis infections.

1.8.1 Patientswith HIV/AIDS

Patients immunocompromised due to HIV have been reported to have frequent

Blastocystis infections (Garavelli et al., 1988, Kurniawan et al., 2009, Llibre et al., 1989). Blastocystis was found in the stools in significantly pathogenic amounts in AIDS patients and induced intestinal disorders such as diarrhoea, abdominal pain. Blastocystis had disappeared from the stools with resolution of symptoms after metronidazole treatment (Garavelli et al., 1988). A serious complication of HIV-related Blastocystis infections is disseminated disease, not frequently observed in otherwise healthy individuals (Kurniawan et al.,

2009). These studies illustrate the possible role of Blastocystis in causing diarrhoea observed in AIDS and related conditions.

1.8.2 Cancer Patients

Blastocystis is one of the most frequently isolated from stools of cancer patients (Tasova et al., 2000). Infections present with abdominal pain, diarrhea, and flatulence (Tasova et al., 2000). Multiple reports suggest that the parasite thrives in cancer patients (Horiki et al., 1999, Patino et al., 2008).

Moreover, Blastocystis infections have also been reported in breast and colorectal cancer patients undergoing anticancer chemotherapy suggesting that the parasite is capable of surviving highly cytotoxic drugs, while debilitating immunosuppressive state induced by chemotherapy promotes its colonization

(Chandramathi et al., 2012). Therefore, Blastocystis colonization, particularly in patients undergoing anticancer therapy might contribute to post- chemotherapy intestinal inflammation and increased permeability (Tsuji et al.,

2003).

1.8.3 Pediatric Infections

Children are reported to be highly susceptible to Blastocystis infections

(Londono et al., 2009). High incidence of the parasite colonization is observed in immunocompetent (Londono et al., 2009) as well as immunocompromised

(Noureldin et al., 1999). Children receiving corticosteroid therapy for nephrotic syndrome were also reported to be susceptible to Blastocystis infections (Noureldin et al., 1999). Associations of low socioeconomic status

(Mehraj et al., 2008) and HIV positivity (Idris et al., 2010) with Blastocystis infections are also observed in pediatric patients.

1.8.4 Animal Exposure

Zoonosis is well reported in Blastocystis (Stensvold et al., 2009c, Yoshikawa et al., 2009). Blastocystis also has a higher prevalence in occupations frequently exposed to animals or animal products (Stensvold et al., 2009c).

Contact with pigs, poultry, primates, and canines increases the risk of

Blastocystis infections (Nagel et al., 2012, Navarro et al., 2008, Stensvold et al., 2009c, Yoshikawa et al., 2009). A recent study of symptomatic patients and their pets identified a concordance of parasite subtypes recovered from their stool samples suggesting a possible transmission of the parasite between them (Nagel et al., 2012).

1.9 TREATMENT

Due to the lack of knowledge of Blastocystis pathogenicity, treatments are not often offered (Mirza and Tan, 2012). The mild nature of the disease and self- limitation of symptoms also made clinicians skeptical of antibiotic prescription for Blastocystis infections (Stensvold et al., 2010), which may result in ongoing symptoms. Metronidazole, a position 1,5 nitroimidazole is the most commonly prescribed drug for the treatment of Blastocystis in which all other etiologies have been excluded (Tan et al., 2010). However, Multiple studies have suggested a large variation in the efficacy of metronidazole against Blastocystis has been reported ranging from 0%- 100% (Moghaddam et al., 2005, Nigro et al., 2003, Stensvold et al., 2008). Factors that might influence treatment outcomes include infection density, acquisition of mutations, resistance of certain developmental stages to metronidazole, and, most importantly, strain-to-strain variation in antibiotic susceptibility (Tan et al., 2010). Recent studies have identified metronidazole resistance in various

Blastocystis isolates (Mirza et al., 2011a). Other antimicrobial agents which have been used to treat Blastocystis infection include co-trimoxazole and

Paromomycin with varying results (Stensvold et al., 2010). Unfortunately, resistance to these drugs has also been reported (Stensvold et al., 2010).

Though prudent use and appropriate management may be able to slow the selection of resistant strains, there is a pressing need to identify alternative therapeutic strategies targeting Blastocystis infections. Table 1.1 summarizes the clinically proven treatment options for Blastocystis.

Table 1.1 Clinically evaluated antibiotics for Blastocystis infections

Clinical Clinically Drug Dose indication resistant subtypes

Metronidazole · 750 mg TDS for 10 days Intestinal ST -1, ST-4 and ST-6 and cutaneous

· 500 mg TDS for 10 days symptoms

· 1.5 g OD for 7 days

TMP-SMX · 320 mg TMP and Intestinal ST -1, ST-4 and ST-6 1600 mg SMX and TDS for 10 days cutaneous symptoms

Nitazoxanide · 500 mg BD for 3 days Intestinal __ symptoms Paromomycin · 25 mg/kg TDS for 10 days Cutaneous __ symptoms

Tinidazole · 2 g OD for 5 days __ __

Ketoconazole · 200 mg OD for 7 days __ __

TDS three times per day, BD twice daily, OD once daily

Adapted from Mirza and Tan, (2012).

1.10 PATHOGENESIS

1.10.1 Clinical

Most studies advocating a lack of association between Blastocystis infections and intestinal disease focus on the distribution of the parasite between asymptomatic and symptomatic groups (Tan, 2008). They either indicate no significant difference between the prevalence of the parasite between the two groups or a higher incidence in the asymptomatic group. Such studies wrongfully assume that Blastocystis infections are biologically and pathogenically homogeneous. Clinical outcomes of parasitic infections are multi-factorial and are influenced by diverse host and parasite factors; therefore such studies may not truly reflect the pathogenic potential of the parasite (Tan et al., 2010). A recent study highlighted the fact that Blastocystis should be considered a potential pathogen as all patients noted severe symptoms in the absence of any other infectious agents (Roberts et al., 2013). However, evidence is still limited. Results of placebo-controlled treatment trials involving symptomatic patients infected solely with Blastocystis would better reflect

Blastocystis pathogenic potential. There have been only two such reports and both concluding that chemotherapy successfully eradicated parasite with concomitant resolution of symptoms (Nigro et al., 2003, Rossignol et al.,

2005). One of these studies used metronidazole and reported clinical cure in

88% of the patients in the treatment group with parasitological clearance in

80% of the cases (Nigro et al., 2003). Similar outcomes were observed in studies using nitazoxanide (Rossignol et al., 2005) as a chemotherapeutic agent. Symptoms however persisted in patients whom Blastocystis infection could not be eradicated, suggesting that these patients were harboring 25

Blastocystis subtypes that are resistant or less responsive to the anti-protozoal agents used. Resolution of symptoms with concomitant parasite clearance in symptomatic patients with Blastocystis infection provides compelling evidence for the parasite’s pathogenic potential. However, the outcomes of such studies are also influenced by several factors (Tan, 2008). The drugs used in these studies are all broad-spectrum antibiotics and clinical cure could thus be attributed to the clearance of some unidentified enteric pathogen or due to lack of comprehensive exclusion of other possible enteric pathogens, especially viruses or bacterial toxins.

1.10.2 Animal infection studies

An absence of a good animal model to study has hindered our understanding of the pathobiology of the parasite and its pathogenic potential. Several experimental infection studies involving rats, mice, guinea pigs, and chickens have been reported (Tan, 2008). Animal surveys have reported that laboratory mice generally do not harbor Blastocystis suggesting that these are not suitable animal models for Blastocystis infections. Additionally, experimental infections in mice are generally mild and self-limiting. Rats and domestic fowl, particularly chickens, are often infected with the parasite. Due to this reason, animal infection studies in recent years have focused particularly on rats and chickens as potential animal models for Blastocystis infections. More recently, the infectivity of various zoonotic Blastocystis genotypes from humans was tested in rats and chickens (Iguchi et al., 2007). Variability in the infectivity of rodent subtype 4 and avian subtype 6 isolates was observed.

Curiously, isolates of subtype 3, the most common subtype found in humans could not infect chickens and rats, whereas avian subtype 7 isolates could only

26 infect chickens, suggesting that these subtypes exhibit some level of host specificity. In another study, Blastocystis isolates from symptomatic and asymptomatic individuals were used to experimentally infect rats (Hussein et al., 2008). Isolates from symptomatic patients induced moderate to severe pathological changes in infected rats, whereas parasites from asymptomatic individuals induced only mild pathology. Authors concluded that the subtype

1 isolate, that induced mortality in 25% of rats, was pathogenic, and pathogenic potential for subtypes 3 and 4 was variable. A recent Blastocystis animal infection study, using rat as host reported up-regulation of proinflammatory cytokine interferon-gamma, interleukin (IL)-12 and tumor necrosis factor alpha 2-3 weeks post infection (Iguchi et al., 2009). Despite significant up-regulation of pro-inflammatory cytokines, local tissue pathology was moderate and the authors suggested that the subtype 4 isolate used in their study was a weak pathogen.

The results of recent animal infection studies are encouraging. However, the observation that rodents do not naturally harbor subtypes other than subtype 4 while birds harbor mostly subtypes 6 and 7 may limit the usefulness of these animals in infection studies (Stensvold et al., 2009a). The lack of pathology might therefore reflect resistance to colonization in rats or chickens rather than avirulence of the parasite.

1.10.3 In vitro, host-pathogen interaction models

In the absence of a relevant animal model to study Blastocystis infections, in vitro culture systems provide useful alternatives to study Blastocystis–host interactions. These in vitro studies have shed lights on many aspects of

Blastocystis pathogenesis, such as induction of intestinal barrier compromise, immune evasion, immune response and virulence factors.

Evidence for Blastocystis-mediated barrier compromise has been reported

(Puthia et al., 2006, Mirza et al., 2012b). Using transwell systems, it was reported that a rat isolate of Blastocystis (ST4) mediated barrier dysfunction evidenced by decrease in transepithelial resistance and increase in transepithelial flux in IEC-6 cells. Pretreatment with caspase inhibitors failed to rescue Blastocystis-induced changes, suggesting that apoptosis was not a major factor contributing to the loss of barrier function (Puthia et al., 2006). It has been recently observed that cysteine proteases of Blastocystis disrupt

Caco-2 monolayer tight junctions in a Rho-kinase (ROCK)-mediated fashion resulting in colonic epithelial barrier compromise and that the pathological outcomes are mitigated when components of the Rho/ROCK pathways are blocked by the inhibitors Fasudil and Simvastatin (Mirza et al., 2012b).

Recent studies have provided evidence that the parasite is capable of modulating host immune response to its own advantage. At mucosal surfaces of the gastrointestinal tract, secretory immunoglobulin A (sIgA) is the major immune defense against microbial pathogens and their toxins. In the

Blastocystis study, both spent media and cell lysates from subtypes 4 and 7 parasites were able to degrade human sIgA (Puthia et al., 2005). A recent study observed that Blastocystis ST-7 (B) modulates host anti-microbial nitric oxide (NO) response (Mirza et al., 2011b). ST-7 (B) was found to be highly susceptible to nitrosative stress compared to an isolate of ST-4. In order to contain NO to a tolerable concentration in its microenvironment, it was found that ST-7 (B) down-regulated epithelial inducible nitric oxide synthase and

28 reduced epithelial, anti-parasitic NO secretion, suggesting suppression of the innate immune response of human enterocytes by Blastocystis. Nitrosative stress generated by enterocytes in the gut lumen keeps luminal colonization of pathogens in check (Eckmann et al., 2000). Suppression of enterocyte NO production not only helps Blastocystis to survive and reproduce, but it might also facilitate the survival of pathogenic microorganisms in the immediate vicinity. Frequent reports of Blastocysits co-infection with other pathogens support this idea (Hu et al., 2008, Kurniawan et al., 2009, Rayan et al., 2007).

Although reports of obvious mucosal inflammatory changes in human gut are inconclusive (Zuckerman et al., 1994), Blastocystis infections do induce host immune responses (Chandramathi et al., 2012, Iguchi et al., 2009, Puthia et al.,

2008). Recent studies have shown that the parasite causes up-regulation of pro-inflammatory cytokines IL-8 in human T-84 colonic epithelium mediated by NF-kB (Puthia et al., 2008). Similarly, up-regulation of IL-8 and GM-CSF has also been reported in the parasite-infected HT-29 colonic cells (Long et al.,

2001).

Collectively, in vitro studies in recent years have advanced our understanding in Blastocystis pathogenic mechanisms as well as virulence factors. However, many gaps remained unfilled. For example, one important aspect of the parasite pathogenesis, adhesion, has never been investigated. Another challenge for characterizing specific virulence factor is the lack of a workable transfection system for Blastocystis to facilitate gene knock-out and expression studies (Tan and Mirza, 2012) . The advances in such in experimental technologies will help elucidate pathogenic factors of

Blastocystis in the future.

1.11 OBJECTIVES OF THE PRESENT STUDY

Significant progress has been achieved in the field of Blastocystis research since its discovery as far back as the early 1910s. However, knowledge of its cell biology and virulence factors is still limited, which has impeded our understanding of its pathogenic potential and clinical relevance (Tan and

Mirza, 2012). Major gaps remain to be filled in the basic biology of

Blastocystis to gain deeper insights into its pathogenesis. So far, Blastocystis adhesion to epithelial cells, an important aspect of its pathogenesis, has never been studied. The current study will focus on the pathobiology of Blastocystis adhesion to host cells and investigate its role in mediating pathogenesis. Using in vitro approaches, the study will first identify a metronidazole resistant strain of Blastocystis capable of attaching to human intestinal epithelial cells and inducing direct pathological changes. The study will then establish a convenient adhesion assay for an in-depth analysis of the physiochemical factors involved in Blastocystis adhesion to Caco-2 cells. Following adhesion, the parasites produced several aspects of pathology in the host cells. The study will also characterize the host responses in the aspects of intestinal barrier defect and apoptotic effects induced by the parasite. Strain-to-strain variations in pathogenicity have been proposed to contribute to the clinical controversies concerning the parasite pathogenic potential (Mirza and Tan, 2012). As attachment is generally an important aspect of virulence factor affecting the infection outcomes (Twu et al., 2013), using a range of various strains, the study will address the question whether the attachment properties affect their pathogenicity. In light of treatment failures in Blastocystis infections with

30 commonly used antiprotozoal agents (Roberts et al., 2013), the study will also evaluate alternative treatment options by targeting adhesion, and apply the adhesion inhibitor to see its protective effects against Blastocystis-induced pathology.

Blastocystis adhesion to epithelial cells, which plays an integral role not only for parasite survival, but also for establishing and maintaining infections, has received little attention. An understanding of the basic pathobiology of

Blastocystis adhesion and factors involved in it is necessary to evaluate the virulence as well as clinical relevance of Blastocystis. Using a clinically important strain isolated from a symptomatic patient, the study will first test whether Blastocystis is able to attach to human intestinal epithelial cells and induce subsequent pathological changes in the host cells. The study will develop a convenient and easily-manipulated adhesion assay to investigate the molecular mechanisms involved in Blastocystis adhesion to Caco-2 cells and elucidate possible molecules mediating the attachment. Such an understanding will provide targets for therapeutic treatment and prevention of Blastocystis infections. The information acquired here will be also used for further investigation of the role of adhesion in mediating pathogenesis.

After establishing contact with the intestinal epithelium, the parasites survive and thrive in the gut and may lead to intestinal symptoms such as diarrhea, which indicates a functional disturbance of the intestine. How Blastocystis infection leads to the diseases is largely unknown. Host responses upon parasite adhesion in the changes of intestinal barrier function and enterocyte apoptosis will be characterized. The molecular basis for Blastocystis–induced epithelium injury is largely unknown. The study will provide for the first time

31 a comprehensive analysis of the changes in the apical junctional complexes and especially the tight junction proteins. As a role of apoptosis in barrier perturbation has been proposed in several studies, the effects of Blastocystis on enterocyte apoptosis will be evaluated. Besides, the study will also address the question whether apoptosis contributes to Blastocystis–induced barrier compromise.

Asymptomatic carriage and inconsistent outcomes of infection continue to contibute to the pathogenic potential of the parasite (Stensvold et al., 2010).

Several studies in recent years suggest that strain-dependent variations in

Blastocystis pathogenicity might have contributed to the controversy (Mirza and Tan, 2012). However, direct evidence of strain-to-strain variations in pathogenicity with a convincing infection model is still limited. Identification of virulent and avirulent strains of the parasite might provide an explanation for the inconsistencies in the clinical outcome of Blastocystis infections. As attachment of the parasite to host cells is a critical step for the pathophysiological events in infection and might be an important characteristic of virulence, the study will evaluate the attachment properties of various strains from different subtypes, and compare whether they differ in their capability in attachment and how these variations would affect their pathogenicity.

Another issue concerning the pathogenic potential of Blastocystis is treatment failure (Moghaddam et al., 2005, Roberts et al., 2013, Stensvold et al., 2008).

Metronidazole is the treatment of choice, but physicians are often skeptical about prescribing antibiotics for Blastocystis infections (Mirza et al., 2011a).

Recent studies have provided evidences of drug resistance of clinical isolates

(Mirza et al., 2011a). The lack of efficacy of commonly used antimicrobial regiments in the treatment of Blastocystis demonstrates the need for further research into alternative treatment options for Blastocystis infection. Since parasite attachment to the epithelial lining of the small intestine is a prerequisite in initiating and maintaining an infection, adhesion would serve as an appealing target. The study will utilize the knowledge acquired regarding the molecular mechanisms in the adhesion process of a metronidazole- resistant isolate of Blastocystis and select the effective adhesion modulator and apply it in host-pathogen interaction to see whether the pathogenic effects exerted by Blastocystis could be ameliorated. The study will provide basis for evaluating the therapeutic potential of antiadhesives in Blastocystis infections and might facilitate future design of preventative and interventional strategies against this disease.

Additionally, studies have implicated impaired attachment and compromised virulence as a possible fitness cost of drug resistance (Ben-Ami et al., 2012).

The interplay between drug resistance, fitness and virulence in Blastocystis has never been investigated. Using various strains which exhibit different levels of adhesion, the possible association of attachment with drug susceptibility will be tested. This knowledge will yield new insights into parasite biology from an evolutionary point of view.

In summary, the objectives of the current proposal focus on unraveling the role of adhesion in pathobiology of Blastocystis and determining the molecular mechanisms involved in adhesion. Thus, the specific aims for this study are:

1. To provide evidence of adhesion of Blastocystis to human intestinal epithelial cells and develop an easily-manipulated adhesion assay.

2. To determine extrinsic and intrinsic factors involved in Blastocystis adhesion to Caco-2 cells, from both Blastocystis and host cells perspectives.

3. To screen for effective adhesion modulator.

4. To characterize host responses upon adhesion leading to functional disturbance of the intestinal barrier: a. To identify the molecular alterations of the apical junctional complex. b. To evaluate Blastocystis–induced enterocyte apoptosis and its role in mediating barrier compromise.

5. To determine whether adhesion is important in mediating Blastocystis– induced barrier dysfunction by characterizing the intra- and inter-subtype variations in Blastocystis adhesion and investigating how it would affect their virulence.

6. To evaluate the therapeutic potential of adhesion inhibitors by applying the adhesion modulator in a pathogenesis study.

7. To investigate the interplay among drug resistance, fitness and virulence in

Blastocystis.

CHAPTER 2:

ADHESION OF BLASTOCYSTIS TO

ENTEROCYTES AND DEVELOPMENT OF

AN ADHESION ASSAY

2.1 INTRODUCTION

After introduction of Blastocystis to the gastrointestinal tract by faecal-oral route (Tan, 2008), a specific adherence event is required for the parasite to establish and maintain itself to resist shear stresses resulting from peristaltic flow, as it remains extracellular throughout infection (Boorom et al., 2008).

Besides, initial attachment of parasites to host cells is a prerequisite for the pathophysiological events in infection (Joe et al., 1998). Yet nothing is known about Blastocystis adhesion to host cells. The study firstly investigated whether Blastocystis could adhere to intestinal epithelial cells and characterized the basic features of Blastocystis adhesion.

A thorough examination of the mechanism of host-parasite adhesion will be important for our understanding of its pathogenesis, and also will help us design therapeutic options targeting adhesion in the future (Chen et al., 1998).

Identification of molecular mechanisms involved in adhesion between host and parasite requires a robust, easy-to-manipulate assay for quantifying adhesion. This in vitro assay will provide the stepping stone for more detailed investigation of parasite pathogenic mechanisms.

Adhesion of Giardia, Trichomonas and Acathomoeba has been measured qualitatively (discriminate cell types) and quantitatively by metabolic radiolabeling of parasites (Fichorova et al., 2006, McCabe et al., 1991,

Morton et al., 1991) or counting with a haemocytometer or under microscopy

(Magne et al., 1991). But these methods are laborious, time-consuming and therefore not suitable for extensive analysis with various conditions. More importantly, for Blastocystis–host interaction, it would not provide a good

36 resolution between the two cell types, as the morphological distinction was not obvious especially after co-incubation and washing steps. Thus, to accurately visualize and count the number of parasites in a mixed cell population, a cell tracker dye is required to label the parasites.

CFSE, carboxyfluorescein diacetate, succinimidyl ester, is a well-retained cell- tracing reagent (Lyons et al., 1994). CFSE passively diffuses into cells. It is colorless and nonfluorescent until the acetate groups are cleaved by intracellular esterases to yield highly fluorescent carboxyfluorescein succinimidyl ester. The succinimidyl ester group reacts with intracellular amines, forming fluorescent conjugates that are well retained. Excess unconjugated reagent and by-products passively diffuse to the extracellular medium, where they can be washed away.Therefore, CFSE would be suitable for staining Blastocystis to discriminate host cells and parasites.

Caco-2 cells, derived from a human colon adenocarcinoma, has been used extensively over the past couple of decades as an in vitro model system to study various biological interactions of parasitic pathogens and host cells, such as the adhesion, invasion, and cytotoxicity of pathogens (Humen et al., 2011,

Joe et al., 1998, Katelaris et al., 1995, Li et al., 1994, Magne et al., 1991,

Sambuy et al., 2005). Caco-2 cells are physiologically closer to the natural targets of Blastocystis, as in culture, they exhibit the differentiated features of enterocytes, such as cell polarity, apical brush border (Rigothier et al., 1991).

Caco-2 model has been shown to be well suited to investigations of the mechanisms involved in Blastocystis–induced intestinal barrier function

(Mirza et al., 2012b). Thus, this study would also use Caco-2 cell line for investigations of Blastocystis adhesion.

Using Caco-2 epithelial monolayers and ST-7 (B), a metronidazole-resistant

(Mzr) Blastocystis isolate recovered from a symptomatic patient (Mirza et al.,

2011b), the current study was designed to determine whether Blastocystis could adhere to intestinal epithelial cells and investigated whether adhesion of

Blastocystis causes lesions in the cells upon adhesion. In this chapter, various microscopic techniques, including differential interference microscopy, confocal microscopy, scanning electron microscopy, would be used to visualize adhesion of the parasite to host cells, to examine the features of adhesion and characterize the pathogenic effects it exerts on Caco-2 cells.

Also, by CFSE labelling of the parasites, an in vitro adhesion assay would be established to facilitate further study of the factors involved in adhesion between Blastocystis and enterocytes.

2.2 MATERIALS AND METHODS

2.2.1 Culture of Caco-2 colonic epithelial cell line

All Blastocystis-host interaction experiments were performed using Caco-2 human colonic cell line (ATCC). Caco-2 stock cultures were maintained in T-

75 flasks in a humidified incubator with 5% CO2 at 37°C. Cell cultures were grown in Dulbecco’s modified Eagle’s medium (HyClone) supplemented with

10% heat-inactivated fetal bovine serum (HyClone) and 1% each of sodium pyruvate, MEM and antibiotic Penicillin-Streptomycin (Gibco; final concentrations are 100 units/mL of penicillin and 100 µg/mL of streptomycin).

Culture health was evaluated using trypan blue assay and only cultures with

>95% viability, were used for the experiments. Cells were trypsinzed with

0.25% trypsin-EDTA. For immunofluorescence, confocal microscopy and scanning microscopy experiments, Caco-2 monolayers were cultured on Poly-

L-lysine-treated 15mm glass coverslips placed in standard 24-well culture plates.

2.2.2 Parasite culture

Axenized Blastocystis isolate B was previously recovered from a symptomatic patient at Singapore General Hospital (Ho et al., 1993). It belonged to ST-7 according to recent classification system (Stensvold et al., 2007b). Stock cultures were maintained as described previously (Mirza et al., 2011a). In brief, the parasites were maintained in 10 ml of prereduced Iscove's modified

Dulbecco's medium (IMDM) containing 10% heat-inactivated horse serum in an anaerobic jar (Oxoid) with an AnaeroGen gas pack (Oxoid) at 37°C.

2.2.3 Confocal microscopy

To visualize the spatial relationship between Blastocystis and Caco-2 monolayers after attachment, Caco-2 monolayers were grown to confluency on glass converslips and were then conincubated for 1 h with Blastocystis live cells. After 1h incubation, cells were washed 5 times with sterile PBS and fixed with 2% formaldehyde for 30 min, blocked with 10% FBS for 2 hours at room temperature and then incubated with legumain antibody-mAb1D5 (Wu et al., 2010), a murine IgM monoclonal antibody produced in this laboratory

(Tan et al., 1996a) and secondary Alexa Fluor® 594 goat anti-mouse IgM

(red). Phalloidin-FITC (green) was used to label F-actin of Caco-2 cells and

DAPI (blue) for nuclei. All monolayers were mounted onto glass slides using fluorescence mounting media (VECTASHIELD) before being observed using confocal microscope (Olympus FLUOVIEW-FV100; Olympus, Japan).

2.2.4 Scanning electron microscopy

For SEM, Caco-2 monolayers were grown to confluency on glass converslips and were then conincubated with Blastocystis live cells. After 1h incubation, cells were washed 5 times with sterile PBS and fixed with 2% Glutaradehyde

(double-strength fixative) for 1 h at 4°C; after three times of washing with sterile PBS, cells were post-fixed with 1 % Osmium Textroxide for 2 h at room temperature (RT) followed by another 3 times of washing with sterile

PBS. Dehydration was performed with 25%, 50%, 75% and 95% ethanol sequentially, for 5, 10, 10 and 10min, respectively; and with 100% ethanol for three rounds, 20min each. Samples were transferred to critical point dryer for infiltration with CO2 then the dried specimen were mounted on SEM stub and

40 transfer stubs to sputter-coater to coat sample with a thin layer of gold. The samples were then viewed under SEM.

2.2.5 Fluorescent labeling of live Blastocystis parasites with CFSE

1-day-old Blastocystis ST-7 (B) isolate culture was used for CFSE labeling.

Blastocystis pellets were collected and CFSE stain was added at a final concentration of 20μM. After 15min incubation at 37°C, pre-warmed sterile

PBS was added and spun down at 3,000 rpm for 10min to get rid of the excessive reagent. With wet mount, the stained parasites were then for examination under confocal microscope.

2.3 RESULTS

2.3.1 Apical attachment of Blastocystis to epithelial cells

When applied apically onto Caco-2 monolayers, attachment of Blastocystis to the epithelial monolayer could be readily seen under confocal microscopy.

Differential interference contrast image clearly illustrated individual ring- shaped Blastocystis cells with diameters about 10μm (Fig. 2.1), distinct from the Caco-2 cells. From the corresponding z-projections of the XY view, numerous parasites were shown to adhere closely to the apical regions of epithelial cells.

Double fluorescence labelling of Blastocystis with 1D5 antibody targeting parasite surface-localized legumain, and host cells with Phalloidin-FITC more clearly demonstrated their spatial relationship. The close contact at the interface between the host cell and the parasites could be seen (Fig. 2.2A).

Notably from the 3-D reconstruction of image slices from apical to basal of the speciment, parasites were found to preferentially adhere to the junctional area (Fig. 2.2B).

Fig. 2.1 Confocal differential interference contrast micrographs of Blastocystis in contact with Caco-2 cells. Typical Blastocystis cells (yellow arrows) are seen among larger Caco-2 epithelial cells. The micrograph corresponds to an optical plane at the top of the epithelial monolayer. In XZ and YZ transverse sections, the parasites are shown in close proximity (black arrowheads) to the apical regions of the epithelial cells.

Reproduced from Tan and Mirza, (2012)

Fig. 2.2 Fluorescence confocal micrographs of Blastocystis in contact with Caco-2 cells. (A) The micrograph corresponds to an optical plane at the top of the epithelial monolayer Blastocystis ST-7 (B) is labeled with Blastocystis legumain antibody-mAb1D5, a murine IgM monoclonal antibody and visualized with secondary Alexa Fluor® 594 goat anti-mouse IgM (red). Caco-2 cells are stained with actin-specific Phalloidin-FITC (green) and DAPI (blue). Scale bar = 10 μm (B) Representative confocal micrographs illustrating Blastocystis ST-7 (B) preferentially adhered to the junctional area of Caco-2 monolayers. Attachment of Blastocystis ST-7 to Caco-2 apical side could be seen clearly. Parasites could be seen intimately adhering to the epithelium and preferentially to the cell-cell junction site. The percentage of Blastocystis that located at the junctional area was 78% ± 4.6% of all attached cells (values are means ± SE). Scale bar = 10 μm

Fig. 2.2A was adapted from Tan and Mirza, (2012)

2.3.2 Determination of Blastocystis attachment to Caco-2 cell monolayer by Scanning Electron Microscopy

The untreated Caco-2 was shown to be a confluent monolayer under scanning electron microscopy, healthy and intact with clear borders between the cells.

In contrast, significant changes were noticed in the monolayers attached with

Blastocystis parasites. The cells appeared flat and exhibited a number of holes on the surface, indicating cell leisions (Fig. 2.3A, red arrows). Quite a few parasites could be seen adhering on top and again, they appeared to mainly bind along the regions reassembling the junctional areas (Fig. 2.3A).

At a higher magnification, dense brush border were found on the surface of

Caco-2 cell, which is a characteristic of mature intestinal epithelial cells.

However, this could not be seen in the infected monolayer with Blastocystis

(Fig. 2.3B). The parasites established close contact with the epithelium and caused lesions at the interface (Fig. 2.3B). Interestingly, membrane extensions were found between the parasite and the host. With methylene blue staining, examination of semi-thin sections of monolayers also revealed loss of brush border (Fig. 2.4).

Fig. 2.3 Scanning electron micrographs of Blastocysitis attached to Caco-2 monolayers. Caco-2 cells were incubated with parasites for 10 h, normal culture media was used as a negative control. After 10 hours, unbound parasites were washed away and samples were fixed with 2% Glutaradehyde and processed for SEM. Compared to negative control, attachment of Blastocysitis to Caco-2 cells caused significant changes in the monolayer at 10 h post-infection. In (A), parasites attached to the monolayer intimately and preferentially to the junctional area (red arrows). The affected epithelial cells appeared to be damaged and exhibited holes on the cell surface and leisions especially at the junctional area. In (B), cells in the negative control have dense brush borders on the surface, Blastocysitis treatment seemed to distort microvilli on the cell surface. It also could be noted that filipod-like structures, with lengths ranging from 2-6 μm, extended at the host-parasite interface. Scale bar for (A) =20 μm, Scale bar for (B) = 5 μm

Fig. 2.4 TEM semi-thin sections illustrating Blastocystis on top of the epithelium. Light micrographs of a 1 μm section of Caco-2 monolayers without parasites (Control) and after incubation with ST-7 (B). A typical Blastocystis cell was found to adhere on top of the epithelium. The treated monolayer showed loss of brush borders at the apical regions compared to untreated control. Methylene blue stain. Scale bar = 20 μm 46

2.3.3 Development of an adhesion assay with CFSE staining for

Blastocystis isolate ST-7 (B)

The results above have shown that Blastocystis readily adhered to human epithelium. To develop a simple and fast adhesion assay to quantify the number of parasites adhered, the first step is to label the parasites and the host.

CFSE is a well-retained cell-tracing reagent widely used for live cell imaging and in vivo tracing and was therefore tested for Blastocystis labelling. The results showed that Blastocystis well retained the stain and light up brightly when probed with CFSE (Fig. 2.5). Most of the healthy cells were stained in round shape by the green fluorescence; some debris or dead cells also got stained but their irregular shapes could be distinguished from the healthy and intact cells based on the DIC images and thus be excluded (Fig. 2.5, red arrows). Therefore, CFSE labeling could be used to trace Blastocystis live cells. Also, CFSE at the concentration used to label the parasites was not toxic to the cells and did not alter the metabolic activity of Blastocystis (Results not shown).

The pre-stained parasites were then co-incubated with Caco-2 cells for attachment assay. After washing and fixing, the host cells were stained with

DAPI. When viewed under the green fluorescence channel, no green staining was detected for the negative control; in monolayers treated with ST-7 (B) isolate, bright green was detected on top of the host cells, indicating adherent parasites, as could be seen in the DIC image (Fig. 2.6). Interestingly again, it could be noted that parasites preferred to adhere at the sites of pericellular junctions (Fig. 2.6).

As shown in Fig. 2.6, in a particular field, the number of host cells could be counted from DAPI staining, the number of attached parasites visualized by green was also countable. Therefore a ratio of attached parasites with host cells could be calculated and thus quantitiated. In order to include a larger sample size, images were taken under 200× magnification (Fig. 2.7), so that more host cells could be captured for better statistical strength of the image analysis. 6 Images for each treatment were taken under confocal microscope and analyzed.

Fig. 2.5 Representative confocal micrograghs showing live Blastocystis parasites after CFSE staining. 1-day-old Blastocystis ST-7 (B) isolate cultures were stained with CFSE and viewed under confocal microscope. The parasite could retain the stain and light up brightly when probed with CFSE. Most of the healthy cells were stained by the green fluorescence in round shape; some debris or dead cells also got stained but their irregular shapes could be distinguished from the DIC images (red arrows). Therefore, CFSE labeling could be used to trace Blastocystis live cells. Images were taken at a magnification of 600×. Scale bar=20 μm

Fig. 2.6 Representative confocal micrograghs showing CFSE-labeled Blastocystis parasites adhering to Caco-2 cells. 1-day-old Blastocystis ST-7 (B) isolate cultures were stained with CFSE and co-incubated with Caco-2 cells, washed, and fixed for viewing under confocal microscope (maginification 1500×). From both the DIC images and the merged images, ST-7 (B) were found to adhere at the junctional area of Caco-2 cells, as shown the bright green staining (red arrows), whereas negative control did not show any green staining. Scale bar=5 μm

Fig. 2.7 Representative confocal micrograghs showing CFSE-labeled Blastocystis parasites adhering to Caco-2 cells at a lower magnification. One-day-old Blastocystis ST-7 (B) isolate cultures were stained with CFSE and co-incubated with Caco-2 cells, washed, and fixed for viewing under confocal microscope (maginification 200×). Green dots represented adherent parasites on top of Caco-2 cells. Absence of parasites was noticed in zones of clearance. Scale bar=100 μm

2.4 DISCUSSION

Attachment of Blastocystis, though being a crucial step to establish colonization and infections for an extracellular parasite, has never been studied (Tan and Mirza, 2012). By using a Blastocystis isolate from a symptomatic human patient, this isolate was found to readily adhere to the

Caco-2 human intestinal epithelial cell line, and induce various aspects of pathological changes. The current study provided the first visual evidences of

Blastocystis adhesion and that adherent parasites induced direct cell damage.

By TEM and SEM, Caco-2 cells exposed to Blastocystis underwent a series of changes. The features of cell lesions were readily seen, including accumulation of actin at the cell junctions, loss of microvilli, and disruption of the cell membrane. These morphological and structural alterations induced by the adhesion of parasites may be involved in pathological processes of diarrhea by affecting intestinal cell function (Puthia et al., 2006). For example, loss of apical brush borders might lead to malabsorption and thus might be related to diarrhea in some patients infected with Blastocystis. The morphological data with SEM and confocal microscopy clearly demonstrated the intimate contact between the parasite and the host cells, suggesting that there might be molecules of both host and parasite origins mediating their contact. This will be investigated in detail in Chapter 3. Besides, surface features of membrane extensions were observed. The function of this structure is unknown yet. In Entamoeba, pseudopods, filipods have also been observed and involved in trophozoite adhesion (Rigothier et al., 1991).

Another notable observation from the SEM and confocal images was the preferential attachment of the parasites at or near the intercellular junction.

Similar observations have also been made in other pathogens. Campylobacter concisus UNSWCD strain was shown to preferentially attach to intercellular junctional spaces (Man et al., 2010). Such spatial distribution was concomitantly associated with a striking barrier disruption, and a loss of tight junction proteins ZO-1 and occludin in Caco-2 monolayers after C. concisus

UNSWCD infection. It was suggested that the observed preferential attachment may allow rapid disruption of tight junctions, either by direct C. concisus–host contact or by release of secretory proteins directly to the target sites, resulting in damage to the structural integrity and function of intestinal epithelial cells. Naegleria fowleri also exhibited a tropism for cell junction regions and led to disruption of tight junctions (Shibayama et al., 2013). The same pattern of attachment at the cell junction has been seen in Brachyspira pilosicoli infection both in vitro and in vivo and it was suggested that this distribution may facilitate penetration of the epithelial layer (Naresh et al.,

2009). Blastocystis has not been shown to be invasive, but it is reasonable to speculate that preferential adherence of cytopathic Blastocystis strains to intercellular junctions might facilitate tight junction disruption leading to barrier dysfunction. The effects of Blastocystis on TJ changes have received little attention so far and will be described in detail in Chapter 4.

Altogether, these observations strongly suggest that Blastocystis is directly cytopathic. These data not only provide insights into the features and consequences of microbial interactions with human intestinal epithelia, but

52 may also be relevant to the mechanisms by which Blastocystis causes diarrhoeal disease in human infections.

Moreover, the study also developed an easy and fast adhesion assay for future investigations of possible factors involved in adhesion. Labeling the parasites is a necessary step in adhesion experiments to differentiate the parasite from the cell monolayer employed. Generally, radioactive labeling of the parasites is done with thymidine (Alderete et al., 1985, McCabe et al., 1991) or methionine (Krieger et al., 1990). This type of labeling requires radioisotopes and specialized and expensive equipment, like a cell harvester and scintillation counter; and there is a hazard in the handling of ‘‘hot’’ scintillation fluids

(Rojas et al., 2004). These problems led to the search for non-radioactive alternatives for adhesion assays. In the current study, the CFSE staining was found to be suitable for Blastocystis labeling and can be used to discriminate the parasite from Caco-2 cells, and therefore allowed the detection and quantification of parasites adhering to the intestinal cell monolayer. This process for labeling and detection is safer than a radioactive label, and much easier to be assayed. This convenient method would be suitable for detailed investigations of molecular mechanisms of host-parasite interactions.

It should be noted that ST-7 (B) is not only adhesive, but also Mz resistant. It will be particularly difficult to manage infections with such strains, demonstrating the need for new treatments. It is hoped that the use or adaptation of this adherence assay will prove useful in screening potential therapeutic or prophylactic compounds aimed at inhibiting parasite adherence to enterocytes.

CHAPTER 3:

FACTORS INVOLVED IN

BLASTOCYSTIS ADHESION TO

ENTEROCYTES

3.1 INTRODUCTION

Adherence of microorganisms to the intestinal cells is a prerequisite for colonization and infection of the intestinal epithelium (Twu et al., 2013). A thorough understanding of mechanisms of parasite-epithelial cell association requires study of this process under defined conditions. The in vitro adhesion assay developed in Chapter 2 would enable the study of various parameters influencing in the adherence of Blastocystis to this highly differentiated, Caco-

2 cell line.

Studies carried out in recent years on the process of parasite-host cell interaction in common protozoas such as Giardia, Entamoeba and

Trichomonas have indicated that adherence is time, temperature, and dose dependent (Katelaris et al., 1995, Li et al., 1994, Mendoza-Lopez et al., 2000) and that it is a multi-factorial process in which microtubules, microfilaments, different adhesins (Arroyo et al., 1992), cysteine proteinases (Arroyo et al.,

1989). A comprehensive summary of common structures or molecules mediating the attachment of protozoan parasites to host cells was listed in

Table 3.1. Among these factors investigated, recognition of host cell glycolipids and/or glycoproteins by a variety of microbial pathogens appears to be significant in adherence of organisms to their target tissues and in determination of tissue tropism, species specificity, and genetic specificity of pathogens (Hashim et al., 2006).

However, none of these factors mentioned above has been investigated in

Blastocystis adhesion to host cells. To mimic some in-vivo conditions in the colon, where Blastocystis usually colonize, it’s important to investigate

55 possible effects of physicochemical factors on attachment, i.e. temperature, rocking conditions. The above adhesion assay could be done with physiological manipulation for different experiment conditions.

An understanding of factors controlling these primary interactions between the pathogen and the enteric cell surface not only will shed new light on the pathogenesis of Blastocystis, but also will be crucial for developing new therapeutic strategies aimed at preventing the initial colonic establishment of this pathogen. Compounds that appear to specifically inhibit attachment will serve as promising tools for future identification of the cellular machinery involved in Blastocystis attachment.

To get a more comprehensive understanding about Blastocystis adhesion to host cells as well as to screen for effective adhesion modulators, in this chapter, the in vitro adherence assay developed in Chapter 2 would be used to characterize conditions for Blastocystis adherence to intestinal epithelium, to determine whether the binding is time-, dose- and temperature-dependent, and whether recognition of carbohydrate moieties on the surface of the corneal epithelial cells may play a role in Blastocystis adhesion to Caco-2 cells. The respective roles of the cytoskeleton, cysteine proteases and lipid raft structures in the recognition process would also be investigated with this model.

Table 3.1 Host or parasite structures or molecules mediating attachment in common protistan parasites

Molecules/ Structure Parasites Study Cell Line Host/Parasite Modulator Used involved in Adhesion

Morton et Corneal Mannose and Acanthamoeba al., 1991, mannose-binding epithelial Acanthamoeba methyl-α-D- keratitis Yang et protein cells mannopyranoside al., 1997 Joe et al., Gal/GalNAc- Cryptosporidium 1998, Joe Gal/GalNAc- Caco-2A Cryptospordium containing parvum et al., specific sporozoite glycoproteins 1994 surface lectin

HeLa and human Arroyo et cysteine vaginal Trichomonas al., 1989 TPCK epithelial proteinases Trichomonas cells vaginalis Human Gilbert et vaginal lipophosphoglycan Trichomonas periodate al.,2000 epithelial (LPG) cells

Liver Entamoeba Faust et sinusoidal Entamoeba Gal/GalNAc lectin galactose histolytica al., 2011 endothelial cells

methyl-b- Humen et lipid raft Caco-2 Giardia cyclodextrin al., 2011 membrane (MBCD) Sousa et Int-407 Host cytoskeletelon cytochalasin D al., 2001

Giardia lamblia colchicine cytoskeleton and mebendazole Katelaris et al., Caco-2 Giardia cytochalasin B 1995 D-mannose, lectin mannose-6- phosphate

3.2 MATERIALS AND METHODS

3.2.1 Culture of Caco-2 colonic epithelial cell line

Caco-2 stock cultures were maintained as previously described in Chapter 2.

For adhesion assay, Caco-2 monolayers were cultured on Poly-L-lysine- treated 15mm glass coverslips placed in standard 24-well culture plates. In order to synchronize cells before experiments, all cultures were serum-starved overnight in antibiotic-free and serum-free DMEM.

3.2.2 Parasite culture

Axenized isolates of Blastocystis B was maintained as previously described in

Chapter 2. 1-day-old healthy Blastocystis parasites were used in all experiment.

3.2.3 Coincubation with different modulators and attachment assay

1-day-old Blastocystis were collected and stained with CFSE at a final concentration of 20 M. Pellets were then added with PBS and centrifuged

@3,000 rpm for 10min to get rid of the stain and were diluted with pre- warmed PBS for inoculation onto Caco-2 cells. Attachment was assayed with coverslips in 24-well tissue culture plates. Unless stated otherwise, for each well, 1.25 × 107 parasites were added onto Caco-2 cell monolayer and incubated together for 1 hour at 37°C. Because at this time point and dosage for parasite-host incubation, Caco-2 cell monolayers remained confluent and appeared intact with normal morphology when examined by light microscopy.

Parasites produced lesions in the cell monolayer and at a later time point of coincubation, the cell monolayers usually would not stay intact and some areas

58 may be sloughed off and thus not suitable for evaluation of attachment. Also,

Blastocystis remained viable after treatments and co-incubation as they could be successfully subcultured after recovery from culture wells.

After the incubation, Caco-2 cell monolayers were washed 5 times with sterile

PBS and fixed with 2% formaldehyde for 30 min. The monolayers were then washed and stained with DAPI for 10min and washed again. All monolayers were mounted on a glass slide using fluorescence mounting media

(VECTASHIELD) before being observed using confocal microroscope

(Olympus FLUOVIEW-FV100; Olympus, Japan). Attachment was quantified by image analysis. The factors that might be involved in host-parasite attachment are being investigated as follows:

For time course studies, the parasites were incubated with Caco-2 for different periods of 10min, 30min, 1, 1.5, 2, 4, 6, 9 and 12 hours at 37°C. For dose response, different numbers of parasites at 0.04, 0.125, 0.4, 0.6, 0.8, 1.25, 1.6,

2.5, 3.3 and 4.2 × 107 were added to each well and incubated with host cells for 1 hour at 37°C.

An investigation of the impact of temperature on adherence was carried out concurrently at 4, 25 and 37°C. To assess the avidity of binding of Blastocystis to Caco-2 cells (McCabe et al., 1991), monolayers were incubated for 1 h on a rotational shaker (Orbital shaker) at 75 rpm at 37°C. To determine whether cells need to be alive for parasite and host binding (McCabe et al., 1991), either Blastocystis ST-7 (B) or Caco-2 cells were fixed with 2% formaldehyde for 30min, washed once and then incubated for 1h at 37°C in 5% CO2 with viable Caco-2 cells or Blastocystis parasites. To see whether Blastocystis

59 surface molecules mediating attachment are sensentive to trypsin, parasites were trypsinized for different lengths of time (15min, 30min and 1h) at 37°C before interacting with epithelial cells.

To determine the role of components of the host or parasite cytoskeleton in attachment, assays were done after 30 min of preincubation of hosts or parasites with the microfilament inhibitor cytochalasin D (1 μg/ml or 2 μg/ml;

Sigma) at 37°C (Sousa et al., 2001). To investigate whether Rho/ROCK signaling or MLCK pathway-mediated actin polymerization (Mirza et al.,

2012b) is involved in attachment, host cells were pre-incubated with MLCK inhibitor ML-9 or Rho/ROCK signaling pathway inhibitors Simavastatin,

Fasudil and Y-27632 (All were purchased from Sigma) before interaction with parasites (Table 3.2).

To see whether Blastocystis cysteine proteases play a role in cytoadherence to host cells, protease inhibitors TLCK, TPCK and E64 (All were purchased from Sigma) were used to pre-treat the parasites for 1 hour before the attachment assay (Table 1) (Arroyo et al., 1989, Lauwaet et al., 2004a, Mirza et al., 2012b). Particularly, a special Blastocystis cysteine protease legumain was inhibited by either monoclonal antibody 1D5, or legumain specific inhibitor APE-RR (Table 3.2) (Wu et al., 2010).

To investigate whether lipid raft structures are involved in Blastocystis adhesion (Humen et al., 2011, Laughlin et al., 2004), either parasites or Caco-

2 cells were pre-treated with MBCD (10 mM) for 1 hour, washed and then used for adhesion assay.

To define the possible role of surface glycoproteins in attachment, assays were done in the presence of monosaccharides (D-galactose, D-mannose, D-glucose and L-fucose) (5 to 100 mM). All monosaccharides were purchased from

Sigma. To further study whether the parasite or the host has sugar-binding protein or sugar containing protein, either the parasites or the host cells were pre-incubated with sugar for 1 h at 37°C, washed and then used for the attachment assay.

3.2.4 Statistical analysis

Results are expressed as means ± standard errors. Differences between means were compared using Student's t tests or analysis of variance where appropriate. Experiments were carried out on at least three separate occasions.

3.3 RESULTS

3.3.1 Time dependent variations in attachment

A time course of attachment was firstly conducted to analyze the kinetics of

Blastocystis attachment to Caco-2 cells (Fig. 3.1). Parasites were able to attach as early as 10min after addition onto the Caco-2 monolayer (59.8 ± 18.3 parasites for every 500 Caco-2 cells). The number of attached parasites increased steadily with maximal level reached at 4h post-infection (333.2 ±

32.4 parasites for every 500 Caco-2 cells) and then began to decrease.

Parasites remained bound to epithelial cells even at 12h post-infection (187.7

± 27.4 parasites for every 500 Caco-2 cells).

Fig. 3.1 Time course studies of Blastocystis adhesion to Caco-2 monolayers. The same number of Blastocystis ST-7 (B) parasites (1.25 × 107 parasites) interacted with Caco-2 cells for different lengths of incubation: 10min, 30min, 1, 1.5, 2, 4, 6,9,12 hours at 37°C. The number of parasites attached at different time points were counted and calculated for every 500 Caco-2 cells. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors.

3.3.2 Dose-response studies

The number of Blastocystis parasites attached to Caco-2 cells was directly related to the inoculum onto the monolayer and increased as more parasites were added (Fig. 3.2). 142.1 ± 31.4 parasites were observed for every 500

Caco-2 cells at the standard dose of 1.25 × 107parasites/well. At the highest infection dose (4.2 x 107parasites/well) tested in our study, no saturation point of attachment was observed (294.2 ± 38.9 parasites for every 500 Caco-2 cells).

Fig. 3.2 Dose response of Blastocystis adhesion to Caco-2 monolayers. Different doses of Blastocystis ST-7 (B) parasites (0.04, 0.125, 0.4, 0.6, 0.8, 1.25, 1.6, 2.5, 3.3, 4.2 × 107 parasites/ well respectively) were incubated with Caco-2 cells for the same length of incubation time of 1h at 37°C. The number of parasites attached at all doses were counted and calculated for every 500 Caco-2 cells. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors.

3.3.3 Effects of temperature, rotational shaking, fixation and trypsinization on Blastocystis attachment

Blastocystis adhesion to Caco-2 cells was temperature dependent (Fig. 3.3).

The number of parasite attachment at 37°C was assigned as a control value of

100%. Adherence was significantly reduced to 55% ± 1.9% at 25°C (p<0.01), and was almost abrogated at 4°C (13.7% ± 2.9% of control) Interestingly, incubation at a higher temperature of 40°C significantly increased attachment to 130.4% ± 5.4% of control (p<0.01).

With rotational shaking at 75rpm, no reduction in Blastocystis attachment to

Caco-2 cells was observed as compared to assays conducted in non-shaking condition (102% ± 12%).

Formaldehyde fixation of Caco-2 cells diminished the attachment to 57.6% ±

15.3% of control (p<0.01) (Fig. 3.4), and when parasites were fixed, attachment was reduced to only 19.4% ± 7.97% of control (p<0.01) (Fig. 3.4).

To see whether Blastocystis surface molecules mediating attachment are sensentive to trypsin, parasites were trypsinized for different lengths of time

(15min, 30min and 1h) at 37°C before interacting with epithelial cells.

Pretreatment of live parasites with trypsin reduced Blastocystis attachment to

59.4% ± 9% of control (p<0.01) (Fig. 3.5), and decreased even more (39.8% ±

10.6%) (p<0.01) if trypsinized for 30min. Similar level of diminished attachment (40.5% ± 5.3% of control) (p<0.01) was observed with 1h trypsinization of the parasites compared to 30min with trypsin.

Fig. 3.3 Effect of temperature on Blastocystis adhesion to Caco-2 cells. 1.25 × 107 Blastocystis ST-7 (B) parasites were incubated with Caco-2 cells for 1 hour at different temperatures (37°C, 25°C, 4°C). A value of 100% was assigned to the number of binding parasites incubated at 37°C. Blastocystis attachment for monolayers incubated at 25°C and 4°C are expressed as change in binding value with respect to control. *, p<0.01, versus control incubated at 37°C. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors. Scale bar=50 μm

Fig. 3.4 Effect of formaldehyde fixation on Blastocystis adhesion to Caco-2 cells. Either Blastocystis ST-7 (B) or Caco-2 cells were fixed with 2% formaldehyde for 30min, washed once and then incubated at 37°C in 5% CO2 with viable Caco-2 cells or Blastocystis parasites. Blastocystis ST-7 (B) or Caco-2 cells pretreated with PBS at the same time was used as control (100%), respectively. The numbers of attached parasites for all other treatments were normalized to control. *, p<0.01 versus control. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors.

Fig. 3.5 Effect of trypsinization on Blastocystis adhesion to Caco-2 cells. Blastocystis ST-7 (B) were pretreated with 0.15% trypsin at 37°C for 15min, 30min and 1h respectively, washed and added onto epithelial cells. Blastocystis ST-7 (B) pretreated with PBS at the same time was used as control (100%). The numbers of attached parasites for all other treatments were normalized to control. *, p<0.01 versus control. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors.

3.3.4 Effects of inhibition of cytoskeletal components, parasite cysteine proteases and lipid raft on attachment

To investigate whether host or parasite cytoskeletal components participate in attachment, Caco-2 cells or parasites were treated with actin inhibitors prior to co-incubation (Table 1). No significant modification in attachment was observed when either host cells or parasites were pretreated with microfilament inhibitor cytochalasin D (1 and 2 μg/ml). ML-9, which inhibits host cell actin polymerization via myosin light chain kinase, did not affect cytoadherence levels (Table 3.2). Similarly, inhibition of ROCK/MLC- mediated actomyosin contraction by Simvastatin, Fasudil or Y-27632, at concentrations which were able to prevent Blastocystis–induced barrier dysfunction (Mirza et al., 2012b). also did not produce any notable inhibitory effects on parasite attachment (Table 3.2).

To evaluate whether cysteine proteases mediate Blastocystis attachment to epithelial cells, parasites were pretreated with various cysteine proteases inhibitors (TPCK, TLCK and E-64). The results showed that the extent of cytoadherence of parasites after cysteine proteases inhibition equaled those of control parasites (vehicle control) (Table 3.2). Particularly inhibiting a surface-localized cysteine protease-Blastocystis legumain, with either monoclonal antibody 1D5, or legumain specific inhibitor APE-RR, did not alter the attachment significantly from that of control, suggesting legumain might not play a direct role in Blastocystis adhesion to host cells.

Lipid rafts are submembrane domains enriched with cholesterol and sphingolipid (Chen et al., 2006) and have been shown to play a crucial role in

67 adhesion of Entamoeba and Giardia (Humen et al., 2011, Laughlin et al.,

2004). The study therefore also tested whether disruption of lipid raft structure on the host cell or the parasite would result in any change in Blastocystis adhesion. Interestingly. an about 20 % inhibition (p<0.05) (Table 3.2) in adhesion was observed by preincubating Caco-2 cells with 10mM MBCD (a lipid raft disorganizing agent), but not when MBCD was used to treat parasites

(91.9 ± 10%).

Table 3.2 Effects of different treatments on attachment of Blastocystis to Caco-2 cells (targeting host cells cytoskeleton, parasite proteases and lipid raft structure on both host and parasites)

% Attachment after:

Agents and concn Pretreatment of

Blastocystis Caco-2

Cytochalasin Da 1 μg/ml 96.3 ± 3 100.8 ± 2 2 μg/ml 98.1 ± 5 ML-9b (50 μM) 105 ± 4 Simvastatinc 5 μg/ml 95.3 ± 6 10 μg/ml 91.2 ± 5 Fasudild 5 μg/ml 93.6 ± 9 10 μg/ml 96.4 ± 6 Y-27632e 5 μM 99.9 ± 4 10 μM 104.2 TPCKf (50 μM) 99.1 ± 10 TLCKg (50 μM) 98 ± 15 E64h (50 μM) 91.7 ± 12 APE-RRi (1mM) 86.8 ± 13 mAb 1D5j (1.2 μg/ml) 97.1 ± 8 MBCDk (10mM) 91.9 ± 10 81.2 ± 6

a Cytochalasin D: actin microfilament destabilizer

69 b ML-9: a myosin light chain kinase (MLCK) inhibitor c Simvastatin: the HMG-CoA reductase inhibitor, which prevents rho kinase-mediated intestinal epithelial barrier compromise induced by Blastocystis d Fasudil: Rho kinase (ROCK) inhibitor e Y-27632: Rho kinase (ROCK) inhibitor f TPCK: N-Tosyl-Phenylalanine chloromethyl ketone, inhibitor of papain-like cysteine proteinases g TLCK: N-Tosyl-L-Lysine chloromethyl ketone, cysteine protease inhibitor h E-64: trans-epoxysuccinyl-L-leucylamido-(4-guanidino) butane, a broad- spectrum cysteine protease inhibitor i mAb 1D5: a murine IgM monoclonal antibody targeting Blastocystis legumain j APE-RR: legumain-specific inhibitor, carbobenzyloxy-Ala-Ala-AAsn- epoxycarboxylate, ethyl ester (where AAsn is aza-asparagine) k MBCD: Methyl-beta-cyclodextrin, a lipid raft disorganizing agent

3.3.5 Effects of saccharides on Blastocystis adhesion to Caco-2 cells

Surface glycoconjugates were described as important mediators in adhesion of various parasitic pathogens (de Araujo Jorge et al., 1984, Lu et al., 2001,

Pegado et al., 1994). Therefore monosaccharides were used, i.e. D-Glucose

(Glu), L-Fucose (Fuc), D-Mannose (Man) and D-Galacose (Gal) (Fig. 3.6A), to test the participation of glycoconjugates on the surface of either the parasite or the epithelial cell in the adhesion process. Differential inhibitory effects by these four saccharides were observed (Fig. 3.6B). Galactose (100 mM) was the most effective inhibitor and reduced Blastocystis adhesion to Caco-2 cells to

13% ± 2% of control (p<0.01). Fucose and Mannose at the same concentration

(100mM) reduced attachment to 81.4% ± 6.6% and 76.9% ± 5.3% of control, respectively (p<0.01 for both values). Glucose, however, did not affect the attachment significantly compared with control (96.7% ± 4.7%).

The inhibitory effect of galactose on the adhesion of Blastocystis to epithelial cells was dose dependent (Fig. 3.7). Even at the lowest concentration tested (5 mM), Blastocystis binding to epithelium was inhibited by 41.2 % ± 9.5%

(p<0.01). The inhibition values were even higher at 10, 25 and 50mM (51.5%

± 12.1%, 63.1% ± 9.8%, 80% ± 2%, respectively) (p<0.01 for all values).

Glucose, as a control sugar, had little effect on attachment at concentrations ranging from 5 to 100mM.

The results above with galactose suggested the possible involvement of a galactosyl glycoconjugate in mediating adhesion. It was also of interest to determine whether the glycoconjugate is parasite or host cell origin.

Preincubation of parasites with galactose (100mM) reduced Blastocystis

71 adhesion to 6.7% ± 2% of control values (p<0.01) (Fig. 3.8), compared to 13%

± 2%, when galactose was added in the host-parasite coculture medium.

Interestingly, attachment was significantly increased when Caco-2 cells were pretreated with galactose at 100mM for 1h (144% ± 15% of control values)

(p<0.01), suggesting a galactose-binding protein may reside on the parasite surface.

Fig. 3.6 Effect of saccharides on adhesion of Blastocystis to Caco-2 cells. Blastocystis ST-7 (B) were incubated with epithelial cells in the presence of 100 mM different glucose, fucose, mannose and galactose. A value of 100% was assigned to number of binding parasites without addition of sugars. The values of Blastocystis attachment to Caco-2 cells performed in other conditions are expressed as change in binding value with respect to control. *, p<0.01 versus control. Values are the means ± standard errors from data of four experiments. Error bars represent standard errors. Glu, glucose; Fuc, Fucose; Man, Mannose; Gal, Galactose.

Fig. 3.7 Dose-dependent inhibiton of galactose on Blastocystis adhesion to Caco-2 monolayers. Blastocystis ST-7 (B) parasites were incubated with epithelial cells in the presence of different concentrations of galactose and glucose (5, 10, 25, 50 and 100mM, respectively). A value of 100% was assigned to number of binding parasites without addition of sugars. The values of Blastocystis attachment to Caco-2 cells performed in other conditions are expressed as change in binding value with respect to control. *, p<0.01 versus control. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors. Glu, glucose; Gal, Galactose.

Fig. 3.8 Effects of pretreating either parasites or host cells with galactose on Blastocystis adhesion to Caco-2 cells. For pretreatments, either Blastocystis ST-7 (B) parasites or host cells were pre-incubated with 100mM galactose for 1h at 37°C, washed once with warm PBS, then used for adhesion assay. Also, another treatment was to add galactose upon the coincubation of Blastocystis and Caco-2 (First bar). A value of 100% was assigned to number of binding parasites pretreated with warm PBS at the same time. The values of Blastocystis attachment to Caco-2 cells performed in other conditions are expressed as change in binding value with respect to control. *, p<0.01 versus control. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors. Glu, glucoe; Gal, Galactose.

3.4 DISCUSSION

As an immotile protist in the gut, attachment of Blastocystis would be critical for host parasitism and establishment of infection. Yet it has received little attention. Using the adhesion assay established in Chapter 2, this is the first study to investigate comprehensively the factors influencing Blastocystis attachment and the role of attachment in Blastocystis pathobiology. The study showed that, similar to Giardia, Entamoeba and Trichomonas (Katelaris et al.,

1995, Li et al., 1994, Mendoza-Lopez et al., 2000), Blastocystis adhesion is a multifactorial process and it is time-, temperature-, and dose-dependent.

It was observed that parasite attachment decreased after 4h incubation with the host. One possible reason is that the experiments were conducted under aerobic conditions needed for viability of the Caco-2 cells (McCabe et al.,

1991), whereas Blastocystis are anaerobes (Silberman et al., 1996). Thus the non-optimal condition for the parasite might contribute to the decreased attachment. This observation might also be due to the dynamic interactions between host and parasite. Findings from other parasitic pathogens suggested host cells develop various anti-parasitic strategies which are likely to decrease parasite colonization (Eckmann et al., 2000, Gobert et al., 2001). It has been shown that Caco-2 cells actively produce nitric oxide upon Blastocystis infection (Mirza et al., 2011b), as an aspect of host innate immune response which could induce parasite cell death. Unpublished data from our laboratory suggested that host cells also synthesized antimicrobial peptides against parasite invasion. Thus complicated host-parasite interactions occurring during

76 prolonged coincubation might possibly lead to the reduction in Blastocystis attachment.

The results of dose-response showed more parasites adhered to host cells as the infection dose increased. Similar trend has been shown in Giardia

(Katelaris et al., 1995, McCabe et al., 1991). Studies have reported overgrowth of Blastocystis in response to illnesses and other immunosuppressive conditions (Dogruman-Al et al., 2009, Rao et al., 2003,

Waghorn et al., 1991). Thus in the scenario of intestinal infestation with

Blastocystis, the number of parasites which are able to adhere and colonize the gut may continue to increase, which might contribute to subsequent intestinal pathology.

Lower temperature decreased adhesion. One reason might be, at low temperature, the alteration in the microenvironment of the cell surface might render the adhesins inaccessible to the binding partners from the host cells

(Panjwani et al., 1997). The dependence of Blastocystis attachment on temperature has its clinical relevance. Blastocystis is regarded as an opportunistic pathogen associated with a number of inflammatory disorders such as IBS and was more prevalent especially when there are other infections occurring (Dogruman-Al et al., 2009, Llibre et al., 1989, Poirier et al., 2012).

In a status of inflammation, the local temperature of infected sites might increase and thus the local microenvironment might facilitate the attachment of Blastocystis and leading to more severe pathology. Interestingly, the adherence was also significantly inhibited by formaldehyde, indicating that adherence is indeed a dynamic process, and both mammalian cells and

Blastocystis must be metabolically active for the attachment process to occur.

Cytoskeletal components of either host cells or pathogens have been shown to participate in the adhesion process of various bacterial and parasitic pathogens

(Katelaris et al., 1995, Knutton et al., 1989, McCabe et al., 1991), which include actin polymerization at sites of adhesion to tissue culture cells and reorganization of cytoskeleton (Knutton et al., 1989, Seigneur et al., 2005).

Host cells microfilament inhibition by cytochalasin D has been shown to inhibit adherence of G. lamblia to Int-407 cells (Sousa et al., 2001). Other studies have shown preincubation of Giardia trophozoites with cytochalasins also decreased attachment (Katelaris et al., 1995, McCabe et al., 1991). In the current study, neither inhibiting host nor parasite cytoskeleton resulted in significant modifications in attachment, indicating attachment of Blastocystis to Caco-2 cells might be independent of microfilaments.

Simvastatin, Fasudil and Y-27632 have been shown to effectively prevent

Blastocystis–induced barrier compromise by blocking actin polymerization via

Rho/ROCK signaling pathway (Mirza et al., 2012b). With the concentration that effectively prevented Blastocystis–induced barrier compromise, no change in attachment was observed, confirming that actin polymerization, though important in mediating barrier dysfunction, might not be crucial for parasite to establish initial contact with host cells.

Cysteine proteases are suspected to be Blastocystis virulence factors (Mirza et al., 2012b, Puthia et al., 2005, Puthia et al., 2008). In Trichomonas, cysteine proteases are involved in cytoadherence and have been suggested to be prerequisite for recognition and binding to host cell surfaces (Arroyo et al.,

1989, Mendoza-Lopez et al., 2000). Also, one particular legumain-like cysteine proteases has been recently shown to be crucial for cytoadherence

(Rendon-Gandarilla et al., 2013). Interestingly, Blastocystis legumain, an unusual cysteine protease located on the cell surface of subtype 7 parasites, was suggested to be a potential virulence factor (Wu et al., 2010). It is of interest to evaluate their role in cytoadherence to host cells. However, neither cysteine proteases inhibition nor specifically legumain inhibition, resulted in significant changes in adhesion, suggesting unlike that in Trichomonas, cysteine proteases may be insignificant in Blastocystis cytoadherence.

Lipid raft structures of Entamoeba and Giardia have been shown to play a crucial role in cytadherence (Humen et al., 2011, Laughlin et al., 2004, Mittal et al., 2008) and disruption of lipid rafts by MBCD treatment leads to the inhibition of adhesion and rescued parasite-induced pathology (Faust et al.,

2011, Humen et al., 2011). However, a reduction in attachment was observed only when host cells were treated with MBCD, but not when lipid raft of parasites were targeted, suggesting lipid raft structure in the host instead of the parasites might participate in adhesion. Indeed, lipid rafts in mammalian cells have been involved in several bacterial pathogens and was proposed to be the possible preferred entry points for Campylobacter (Chen et al., 2006,

Wooldridge et al., 1996). How the lipid raft of the host mediates Blastocystis adhesion needs to be investigated further. Lipid raft is known to control numerous protein–protein and lipid–protein interactions at the cell surface

(Manes et al., 2003). It is possible that some host proteins involved in adhesion are located in this structure and disruption of it might interfere with the binding to Blastocystis.

Recognition of host cell glycolipids and/or glycoproteins by a variety of microbial pathogens appears to be significant in adherence of organisms to

79 their target tissues (Morton et al., 1991). In the current study, drastic inhibition of Blastocystis adhesion by galactose suggested that Blastocystis bind to the surface of the epithelial cells by a recognition system mainly mediated by galactosyl glycoconjugates, with galactose-binding protein probably residing on the surface of Blastocystis. Similar inhibitory effects of galactose have been well documented in Entamoeba adhesion (Faust et al., 2011, Li et al., 1988) and it has been shown that recognition of epithelial glycoconjugates to an amebic cell surface Gal/GalNAc lectin is involved in adherence of Entameba histolytica to the colon (Ravdin et al., 1981). Gal or Gal/GalNAc-specific lectins have also been identified in several other protozoan parasites.

Trypanosoma cruzi expresses a 67-kDa surface-associated Gal-binding protein, which is implicated in mediating attachment to host cells (Silber et al.,

2002, Hashim et al., 2006). It would be important to characterize the galactose-binding protein on the surface membranes of Blastocystis as well as its reactive glycoprotein in epithelial cells, and investigate the role of their interaction in subsequent pathogenesis of the parasite. In Entamoeba, the virulence program was activated upon interaction of the lectin with host cells

(Blazquez et al., 2007). Thus it will be interesting to investigate the function of the galactose-binding protein and see whether the specific binding could initiate the sequence of events leading to barrier dysfunction in the epithelial cells after adhesion.

On the other hand, it should be noted that galactose did not abolish adhesion completely. This incomplete inhibition of adhesion may also suggest that additional galactose-glycoprotein-independent mechanisms are involved in adhesion to host cells. The observation that Mannose and Fucose could also

80 reduce adhesion, albeit to a much less significant extent, shows that there might be more than one type of glycoproteins mediating the process of adhesion. Previous studies on the identification of surface coat carbohydrates in Blastocystis indicate that the parasite has surface components containing D- mannose (Lanuza et al., 1996). Carbohydrate binding studies on Caco-2 cells also showed that Caco-2 binds to oligomannose structures (Arndt et al., 2011).

Thus it might explain why addition of mannose partially inhibited adhesion.

The results with the adhesion modulators suggested that the pathogen possesses multiple mechanisms to mediate the initial contact with the host. It appears likely that one major mechanism of Blastocystis adhesion to the epithelial cell surface may involve interactions between the galactose-binding protein of Blastocystis and galactose-containing proteins on the surface of

Caco-2 cells. Fucose-glycoproteins and mannose-glycoproteins and lipid raft of the host cell also seem to play a minor role in adhesion.

Altogether, the in vitro attachment model has proven to be reliable and useful for investigating the mechanisms by which Blastocystis attach to intestinal cells and facilitating the evaluation of the efficacy of antiadhesives. This study also screened out a potent inhibitor of adhesion-galactose and confirmed its role in ameliorating Blastocystis–induced pathology. Although studies performed in vitro may not mirror events in vivo, and further in vivo validation is needed, these investigations provide important clues regarding the initial event of Blastocystis pathogenesis. Future work on the identification of the parasite and host molecules that mediate the initial host-parasite interactions is crucial for designing preventive and interventional strategies to prevent

Blastocystis–induced diseases.

CHAPTER 4:

HOST RESPONSES-PART I:

EFFECTS OF BLASTOCYSTIS ON

APICAL JUNCTIONAL COMPLEX AND

EPITHELIAL BARRIER FUNCTION

Chapter 2 revealed that Blastocystis readily attaches to the epithelium. Studies have shown that pathogen adhesion led to several alterations within the host cells (da Costa et al., 2005, Lauwaet et al., 2004b, Maia-Brigagao et al., 2012,

Spitz et al., 1995). For luminal pathogens especially, this initial attachment between pathogen and intestinal epithelial cells triggers a series of events that culminate in the production of diarrhoea (Cotton et al., 2011, Peiffer et al.,

2000, Spitz et al., 1995). While the process leading to these abnormalities could be complicated, these pathophysiological stages are believed to involve heightened rates of enterocyte apoptosis and intestinal barrier dysfunction

(Cotton et al., 2011). Blastocystis infections have been frequently associated with intestinal symptoms such as diarrhea (Tan et al., 2010). The post- adhesion events leading to perturbations on host cell function, however, have not been determined. It would be important to characterize changes in epithelial-cell physiology after parasite infection. In Chapter 2, it was observed that adhesion of Blastocystis induced direct pathological changes in host cell. But these were mostly morphological observations. In the next two chapters, the functional and structural changes upon parasite adhesion would be characterized. Mainly, the study would focus on the intestinal barrier- disrupting events as well as induction of host cell death by Blastocystis. In the first part of host responses, alterations in barrier function would be investigated; while the second part would address the question whether enterocyte apoptosis occurred upon Blastocystis infections.

As the first line of defence, the main function of intestinal epithelial cells is to provide a physical and chemical barrier that keeps luminal contents separate from the interstitium (Turner, 2009). The functional integrity of the intestinal

83 epithelial barrier forms a major defense against unwelcome intrusion of pathogens (Cliffe et al., 2005). Therefore, an intact intestinal barrier is critical to normal physiological function and prevention of disease (Turner, 2006). In many intestinal diseases, such as inflammatory bowel disease, this barrier function can become compromised and a leaky barrier may result in gastrointestinal symptoms including diarrhea (Clayburgh et al., 2004). As reviewed in Chapter 1, there is ample evidence suggesting that Blastocystis causes intestinal epithelial barrier compromise in vitro and in vivo. However, the molecular alterations behind the functional disturbances have not yet been fully elucidated.

In polarized epithelial cells, it is well established that the apical junctional complex (AJC) is responsible for maintenance of the integrity of intercellular contacts (Baranwal et al., 2012). AJC is mainly composed of tight junctions and adherens junctions (Vogelmann et al., 2005), the former being most critical (Turner, 2006). Tight junctions (TJ) are multi-protein complexes composed of transmembrane proteins, peripheral membrane (scaffolding) proteins and regulatory molecules that include kinases. Together they form the continuous intercellular barrier between epithelial cells, which are required to separate tissue spaces and regulate selective movement of solutes across the epithelium (Anderson et al., 2009). Claudins are transmembrane proteins and are the key components for the structure and function of TJs (Furuse, 2010,

Itoh et al., 1999). Occludin is a transmembrane tight junction protein that interacts directly with claudins and actin, and plays a direct role in the paracellular permeability (Furuse et al., 1993). Peripheral membrane proteins, such as zonula occludens 1 (ZO-1) and ZO-2, are crucial to tight junction

84 assembly and maintenance (Turner, 2009) and these cytosolic adaptor proteins link tight junctions to a perijunctional actomyosin ring, which supports and regulates tight junction permeability (Steed et al., 2010). Adherens junctions

(AJs) are cell–cell adhesion complexes that are important for solid tissues maintenance (Harris et al., 2010). Cadherin adhesion molecules, including E- cadherin, are transmembrane proteins that are the core of AJ (Halbleib et al.,

2006). AJC provides a tight physical barrier that protects deeper tissues from external aggressions, including microbial infections. However, the first line of defense against infection has become one of the most exploited gates to access and invade epithelial cells, disrupt epithelial integrity, and colonize the organism (Bonazzi et al., 2011). Many enteropathogenic bacteria or parasitic pathogens disrupt the tight and adherence junctions to alter intestinal permeability (Groschwitz et al., 2009). The strategies to overcome the tight junction barrier employed by harmful microorganisms are diverse. Whereas some pathogens bind and modify junction components directly, others exert their effects through the actin cytoskeleton, which controls the integrity of tight junctions (O'Hara et al., 2008). The mechanisms of the pathogens reviewed are summarized in Table 4.1.

An association of this functional disturbance of the intestinal barrier by

Blastocystis with reorganization of ZO-1 tight junction protein was recently demonstrated (Mirza et al., 2012b), yet a comprehensive analysis of molecular alterations in the apical junctional complex for the barrier defects is still lacking. The aim of the current chapter was to conduct an in-depth analysis of molecular changes in apical junctional complex, which would facilitate our understanding of the pathogenesis of the understudied emerging organism.

The results showed that on direct contact with the apical side of the enteric cells, Blastocystis caused an increase in paracellular permeability as evidenced by an increase in Dextran-FITC probe flux. Protein expression profile by

Western blot revealed degradation of Occludin, loss of ZO-1 and Claudin-1, but none of E-cadherin. Therefore, Blastocystis-induced increase in paracellular permeability of enteric cell layers might be ascribed to disturbance of the molecular organization of multiple tight junction proteins.

Table 4.1 Mechanisms of modulating tight junction barrier by some protistan parasites

Proposed Mechanism of Barrier Cell Line Function Compromise through Parasite Study Structural Modification of AJC (mainly TJ)

Disturbance of the molecular organization of tight Leroy, A., et T84 junction proteins: Dephosphorylation of ZO-2, al., (2000). loss of ZO-1 from ZO-2; ZO-1 degradation

Entamoeba Amebic proteinases-mediated proteolysis of tight histolytica Lauwaet, T. et T84 junction molecules ZO-1 and ZO-2 and of villin, al., (2004a). ZO-1 degradation

Tight junctional ZO-1 disruption: Buret, A. G., et SCBN Focal disruption, punctate concentrations of ZO-1 al. (2002). along the pericellular junctions and cytoplasmic accumulation of ZO-1

Myosin light chain kinase-dependent Scott, K. G., et SCBN reorganization of cytoskeletal al. (2002). F-actin and tight junctional ZO-1 Giardia lamblia ZO-1 disruption mediated by apoptosis in a caspase-3-dependent manner: Focal disruptions Chin, A. C., et SCBN and punctate concentrations of ZO-1 along the al. (2002). pericellular junctions, and migration of ZO-1 towards the nuclei

Buret, A. G., A. Caco-2 cells, Apoptosis-induced tight junctional ZO-1 Cryptosporidium C. Chin, et al. SCBN, MDCK disruption andersoni (2003).

da Costa, R. F. Alterations in the junctional complex: distribution Trichomonas Caco-2 cells et al. (2005). for E-cadherin, occludin and ZO-1 vaginalis Human brain Khan, N. A. et microvascular Degradation of occludin and ZO-1 proteins in a Acanthamoeba al. (2009). endothelial cells Rho kinase-dependent manner (HBMEC)

IEC-6, a rat Puthia, M. K., Rearrangement of F-actin, decreases intestinal S. W. Sio, et al. transepithelial resistance, and increases epithelial epithelial cell (2006) permeability in IEC-6 cell monolayers line Blastocystis Rho kinase-mediated myosin light chain Mirza H, et al. Caco-2 cells phosphorylation and epithelial actin cytoskeleton (2012b) and ZO-1 reorganization

4.2 MATERIALS AND METHODS

4.2.1 Culture of Caco-2 colonic epithelial cell line

Caco-2 stock cultures were maintained as previously described in Chapter 2.

For cell fractionation and Western blot analysis, Caco-2 monolayers were grown on 6-well cell culture plates (corning) until 100% confluency. For immunofluorescence and confocal microscopy experiments, Caco-2 monolayers were cultured on Poly-L-lysine-treated 24mm glass coverslips placed in standard 6-well culture plates. For epithelial permeability experiments cells were grown on Millipore transwell filters with PET membranes of 3μm pore-size, placed in 24-well tissue culture plates. In order to synchronize cells before experiments, all cultures were serum-starved overnight in antibiotic-free and serum-free DMEM.

4.2.2 Parasite culture

Axenized isolate of Blastocystis ST-7 (B) was maintained as previously described in Chapter 2. 1-day-old healthy Blastocystis parasites were used in all experiment.

4.2.3 Epithelial permeability

Caco-2 monolayers were grown on Millipore transwell system till they reach confluency and tight junction maturation on day 21. After confirmation of maturation by TER measurement, monolayers were co-incubated with parasite live cells for 24 h. Following co-incubation, epithelial and basolateral compartments were washed twice, followed by addition of 400 μl of warm

HBSS at the basolateral compartments and 200 μl of 100 mg/ml FITC-

88 conjugated Dextran 4000 (Sigma) solution in HBSS to apical compartments.

After 1 h at 37°C, 300μl of HBSS was taken from basolateral compartments and was transferred to a black 96-well plate (NUNC) to estimate Dextran–

FITC flux across monolayers. Fluorescence was measured using an ELISA reader (Tecan Infinite M200) at excitation and emission wavelengths of 492 and 518nm respectively.

4.2.4 Western blot and cellular fractionation

Caco-2 cells were plated at a density of 2 × 104 /cm2 on a 6-well culture plate and allowed to reach confluency. After treatments, monolayers of cells were then rinsed three times with chilled sterile PBS (pH 7·4) and then were lysed on ice for 40 min with 150l of ice-cold radioimmunoprecipitation assay

(RIPA) buffer (150 mM NaCl, 1.0% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl, pH 8.0), including protease and phosphotase inhibitors for protein extraction. After centrifugation at 16,000 × g for 30 min at 4°C, the supernatant was collected for further analysis.

For cellular fractionation, Caco-2 monolayers were lysed with lysis buffer A

(0.5% Triton X-100, 150 mM NaCl, 50 mM Tris-HCl, and protease inhibitors, pH 7.5) for 30 min on ice. After centrifugation of cell lysates (14,000 × g for

10 min), the supernatant (“soluble fraction”) was separated from the pellet.

The pellet was further incubated with in lysis buffer B (lysis buffer A with

0.1% SDS) for 20 min, on ice, and centrifuged (14,000 × g for 30 min). The resulting supernatant was collected (“insoluble fraction”) for further analysis.

Equal amounts of total protein were separated on 10% SDS-polyacrylamide gels and then transferred to a nitrocellulose membrane. After blocking for 2 h

89 in PBS containing 0.1% Tween and 5% dry powdered milk, membranes were incubated overnight at 4°C, in primary antibody (mouse anti-ZO-1, mouse anti-Occludin, rabbit anti-Claudin-1; Zymed). After five washes for 5 min each with PBS-T, the membranes were incubated for 1 h with corresponding horseradish peroxidase-conjugated secondary antibody. Following five washes with PBS-T, the membranes were developed for visualization of protein by addition of enhanced chemiluminescence reagent (Amersham, Princeton, NJ,

USA).

4.2.5 Confocal microscopy

For visualization of tight junction proteins in Caco-2 cells, Caco-2 cells were grown on Poly-L-Lysine treated glass coverslips to 100% confluency. After treatments, monolayers of cells were then rinsed three times with chilled sterile PBS (pH 7·4), fixed with 2% formaldehyde for 30 min, permeabilized with 0.1% Triton X-100 for 10 min, blocked with 10% FBS for 2 hours at room temperature and then incubated with respective primary antibodies to

ZO-1, occludin and Claudin-1 overnight 4°C. After five washes for 5 min each with PBS, the monolayer were then incubated with Cy3-conjugated secondary antibodies. Coverslips were then mounted onto clean glass slides using fluorescence mounting media (VECTASHIELD) and viewed under confocal laser microscopy using an Olympus Fluoview FV1000 (Japan). Images were captured using Olympus Fluoview version 1.6b.

Legumain staining after coincubation of Caco-2 cells with Blastocystis was performed as described in Chapter 2.

4.3 RESULTS

4.3.1 Blastocystis induced time- and dose-dependent permeability increase in Caco-2 monolayer

Barrier function could be measured by intestinal permeability, thus the flux of

Dextran–fluorescein isothiocyanate (FITC) probe across the Caco-2 monolayer was measured. Blastocystis ST-7 (B) induced a time- and dose- dependent drop in Caco-2 epithelial permeability (Fig. 4.1). A significant increase in Caco-2 permeability was observed as early as after 6h of coincubation (p<0.05; Fig. 4.1) and continued at 12 and 24h (p<0.01; Fig. 4.1).

A minimum dose of 1.5 × 107 parasites induced a significant alteration in epithelial permeability after 24h of co-incubation (p<0.01, Fig. 4.2).

Fig. 4.1 Time-dependent effect of Blastocystis ST-7 on epithelial permeability of Caco-2 cell monolayers. Confluent Caco-2 cells grown on 24-well Millipore transwell inserts were coincubated with 2 × 107 Blastocystis ST-7 parasites or normal culture media as a negative control for the indicated lengths of time (3,6,12 and 24h). Permeability was determined by measurement of Dextran-FITC fluxes across the monolayer. ST-7 treatment, compared to negative control induced a significant increase in permeability from 6h onwards (p<0.05 for 6h; p<0.01 for 12 and 24h). Each value represents mean of twelve samples, taken from three independent experiments, four samples from each. Error bars represents standard error.

Fig. 4.2 Dose-dependent effect of Blastocystis ST-7 on epithelial permeability of Caco-2 cell monolayers. Confluent monolayers grown on 24-well Millipore transwell inserts were coincubated with varying doses of ST-7 for 24 h at 1, 1.5, and 2 × 107parasites. Compared to negative control Blastocystis ST-7 induced a significant permeability increase in Caco-2 at 1.5, and 2 × 107parasites (p<0.01). Each value represents mean of twelve samples, taken from three independent experiments, four samples from each. Error bars represents standard error.

4.3.2 Effect of Blastocystis on the overall expression of the apical junctional complexes

The functional disturbance by Blastocystis led us to further investigate the molecular changes. Western blot analyses were performed to determine the total protein expression levels of the apical junctional complex, including tight junction proteins Occludin, ZO-1 and claudin-1, as well as adherence junction protein E-Cadherin. An obvious reduction in the expression of tight junction proteins Occludin, ZO-1 and Claudin-1 was observed from the densitometry analysis (p<0.01; Fig. 4.3). However, no significant change was detected in the total amount of E-Cadherin, suggesting Blastocystis might induce

92 impairment of barrier function mainly via the tight junction proteins of the epithelial cells, instead of proteins at adherence junction.

Fig. 4.3 Western blots showing distinct effects on tight junction and adherens junction proteins by apical application of Blastocystis ST-7 (B). Confluent Caco-2 monolayers were incubated with Blastocystis ST-7 (B) and normal culture media as a negative control. Total protein was extracted and run for Western Blot. Tight junction proteins Occludin, ZO-1 and Claudin-1, as well as adherence junction protein E-Cadherin were analyzed. The densitometry values revealed obvious reduction in the expression of all the tight junction proteins, but not for E-Cadherin. *, p<0.01 versus control for all tight junction proteins, but not for E-Cadherin. Values are the means ± standard errors from data of three experiments.

4.3.3 Effect of Blastocystis on the distribution of the tight junction proteins by cellular fractionation

As tight junction is the most important regulator for barrier function, and there is no significant change observed in E-Cadherin, the study went on to make an in-depth analysis of the tight junction proteins by cellular fractionation. The proteins were analyzed in cytosolic (soluble) and membrane-associated

(insoluble) fractions (Fig. 4.4). Occludin, ZO-1, Claudin-1 and β-actin in uninfected controls were detected as clear bands in both fractions. Our results showed Blastocystis treatment induced an obvious decrease in Occludin expression in the membrane-associated fraction compared with control

(p<0.01; Fig. 4.4). ZO-1 protein level in both fractions were greatly affected and significantly reduced (p<0.01; Fig. 4.4). Interestingly, for Claudin-1, while the expression in the cytosol fraction was decreased, a prominent increase in membrane-associated fraction was observed concomitantly after exposure to Blastocystis (p<0.01; Fig. 4.4).

Fig. 4.4 Representative Western blots showing the changes in tight junction proteins and densitometric analysis of protein expressions in the insoluble fraction and soluble fraction of Caco-2 cells after treatment by Blastocystis ST-7 (B). Confluent Caco-2 cells cultured on 6-well plates were exposed to Blastocystis ST-7 (B) for 6h or normal culture media as a negative control. Cells were prepared and analyzed in cytosolic (soluble; S) and membrane-associated (insoluble; IS) fractions, and also at the same time for total protein extraction after the same treatments. Western blots were probed with anti-Occludin, anti-ZO-1 and ani-Claudin-1 antibody. β-actin was used as a loading control. Data shown is representative of 3 separate experiments. Each experiment had duplicate cultures for each experimental condition. The measurements for each fraction were determined from the ratio of the integrated band intensity of the target proteins to that of β-actin in the same fraction of the same sample. *, p< 0.01 versus control. Values are the means ± standard errors from data of three experiments.

4.3.4 Effect of Blastocystis on the distribution of the tight junction proteins

Meanwhile, to visualize the distribution and integrity of the tight junction proteins after exposure to Blastocystis ST-7 (B), indirect immunolabelling of occludin, ZO-1 and Claudin-1 were performed and analyzed by confocal microscopy. All three tight junction proteins were continuously distributed in a belt-like manner along membranes in the control Caco-2 intestinal monolayers. From the cross sectional view, tight junction proteins were present at the intercellular borders, encircling the cells and delineating the cellular borders with bright staining.

Treatment of Blastocystis ST-7 (B) led to significant changes in the distribution of all three proteins. In contrast to the uniform membrane staining in controls, a drastic decrease in Occludin fluorescence intensity in Caco-2 cell line was observed (Fig. 4.5). ZO-1 proteins, instead of a sharp definitive localization at intercellular junction, became punctate and migrated from the apical region to the cytoplasm (Fig. 4.6) after apical infection with

Blastocystis ST-7 (B). For Claudin-1, the amount of tight junction proteins was significantly decreased compared with that of the control (Fig. 4.7).

Interestingly, the distribution of Claudin-1 was noticed to migrate from the basolateral region to the apical side.

Fig. 4.5 Representative confocal z-projections showing effects of Blastocystis exposure on the distribution of tight junction protein Occludin. Caco-2 cells were grown to confluency on 24mm glass coverslips in 6-well culture plates. After 6h coincubation with Blastocystis ST-7 (B) and normal culture media as a negative control, monolayers were fixed, permabilized, and labelled with DAPI (Blue) and an antibody targeting tight junction protein Occludin and Cy3® goat anti-mouse IgG (Red). Control monolayer (C) exhibited a typical occludin pericellular organization at the periphery of enterocytes. Monolayers incubated with ST-7 (B) exhibited a great loss of Occludin along pericellular junctions (arrows). Scale bar = 10 μm

Fig. 4.6 Representative confocal micrographs illustrating ZO-1 integrity in Caco-2 monolayers. Monolayers were grown to confluency on glass converslips. Caco-2 cells were then conincubated for 6h with Blastocystis ST- 7. Normal culture media were used as a negative control. Control monolayers exhibited a clear and sharp ZO-1 staining at intercellular junctions, as well as uniform localization at the apical region (yellow arrows), whereas ST-7 treatment resulted in focal disruption of ZO-1 tight junction protein (white arrows) and the apical localization of ZO-1 was also re-distributed in the monolayer. Scale bar = 10 μm

Fig. 4.7 Representative confocal micrographs with z-projections showing the redistribution of Claudin-1 tight junction protein after 6h exposure to Blastocystis ST-7 (B). Representative confocal micrographs illustrating Claudin-1 distribution in Caco-2 monolayers. Caco-2 cells were grown to confluency on glass converslips. Monolayers were fixed, permealized and immunostained with rabbit polyclonal anti-Claudin-1 antibody and analyzed by confocal microscopy. Control monolayers showed strong staining at pericellular regions, the intensity in ST-7 (B)-treated cells, however, has decreased significantly. Besides, Instead of being localized basolaterally in untreated cells, Claudin-1 was re-distributed to the apical of the monolayer (yellow arrow). Scale bar= 10 μm

4.3.5 Loss of co-localization between tight junction proteins and F-actin after exposure to Blastocystis

Double immunostaining was performed to see the interaction between tight junction proteins and cytoskeleton. The F-actin staining pattern of control

Caco-2 cells showed a continuously lined distribution at the cell borders and cytoskeletal regions, with negligible stress fiber formation in the cytoplasmic area. Occludin and F-actin showed yellow merged patterns in the negative control; however, after treatments, a pronounced formation of stress fibers and actin condensation was noticed in the monolayers incubated with parasites

(Fig. 4.8). With alteration of F-actin filaments as well as loss of Occludin protein at the pericellular regions, the colocalization also disappeared.

ZO-1 and Claudin-1 showed a spatial overlap at regions of cell-to-cell contact, as seen in the merged view of control cells. After Blastocystis ST-7 (B) treatment, both ZO-1 and Claudin-1 staining have decreased drastically leading to loss of co-localization of these two tight junction proteins (Fig. 4.9).

100

Fig. 4.8 Representative confocal micrographs illustrating Occludin and F- actin double staining in Caco-2 monolayers. Caco-2 cells were grown to confluency on glass converslips. After 6h coincubation with Blastocystis ST-7 (B) and normal culture media as a negative control, monolayers were fixed, permabilized, and labelled with DAPI (Blue) for nuclei, fluoresceinphalloidin (Green) for F-actin and an antibody targeting tight junction protein Occludin and Cy3® goat anti-mouse IgG (Red). The pericelluar colocalization of Occludin and F-actin was disrupted in ST-7 (B)-treated cells. Scale bar = 10 μm

101

Fig. 4.9 Co-imunostaining of tight junction protein ZO-1 and Claudin-1 in Caco-2 cells after 6h exposure to Blastocystis ST-7 (B). Caco-2 cells were grown to confluency on glass converslips. Monolayers were fixed, permealized and immunostained with rabbit polyclonal anti-Claudin-1 antibody and mouse monoclonal anti-ZO-1 antibody. Staining intensity has greatly decreased after treatments with ST-7 (B) for both proteins. The colocalization pattern seen in control monolayers also became ambiguous in infected cells (white arrow). Scale bar = 10 μm

102

4.3.6 Possible association of host tight junction protein Claudin-1 with adhering Blastocystis parasites

As an increase in Claudin-1 protein in the membrane-associated fraction was observed, more experiment were performed to see the relationship of host tight junction protein Claudin-1 and the parasite using two methods of staining. One was to label the parasite with legumain antibody, targeting a surface-localized protease of Blastocystis; the other labeling was to stain the parasites with

CFSE as described in Chapter 2. Both staining methods showed similar patterns regarding the relationship of the adhered parasites with host protein

Claudin-1. Instead of being localized basolaterally, Claudin-1 was re- distributed to the apical of the monolayer, some of which were found to colocalize with the parasites, indicating that parasites might interact with the host tight junction protein (Fig. 4.10 and Fig. 4.11). Additionally, 1D5- binding protein, Blastocystis legumain, was transferred to apical surface of enterocytes, and preferentially deposited onto the junctional area (Fig. 4.10).

103

Fig. 4.10 Representative confocal micrographs illustrating Claudin-1 staining and deposit of Blastocystis legumain in Caco-2 monolayers after 6h exposure to Blastocystis ST-7 (B). Caco-2 monolayers were grown to confluency on glass converslips and were then incubated with Blastocystis ST- 7 (B). Normal culture media was used as a negative control. Blastocysis ST-7 was labeled with legumain antibody-mAb1D5, and secondary Alexa Fluor® 594 goat anti-mouse IgM (red). Claudin-1 of Caco-2 cells was at the same time stained in green, and DAPI (blue) was used to counterstain nuclei. No 1D5-binding legumain can be seen in the untreated monolayers. After infection, parasites were seen adhering to the monolayer. Besides, legumain deposit was found to bind to junctional area of Caco-2 cells (green arrow). White arrows indicate spatial overlap (colocalization) of the red-stained parasites with the green Claudin-1 signal. Scale bar = 10 μm

104

Fig. 4.11 Representative confocal micrographs illustrating the spatial relationship of Claudin-1 tight junction protein and Blastocystis in Caco-2 monolayers after 6h exposure to Blastocystis ST-7 (B). Confluent Caco-2 monolayers were incubated with Blastocystis ST-7 (B) pre-stained with CFSE (green). Normal culture media was used as a negative control. After coincubation, the non-attached parasites were washed away. The Caco-2 monolayers were then stained with Claudin-1 (red) and DAPI (blue) and then were viewed using confocal microscope. Green in the field represents parasites attached to the monolayer, and some were found to colocalize the Claudin-1 protein which migrated from basolateral to the apical region of enterocytes. Scale bar = 10 μm

105

4.4 DISCUSSION

It was demonstrated here that after a close interaction with polarized intestinal epithelial cells, Blastocystis could directly alter the epithelial barrier evidenced by an increase of paracellular permeability, suggesting that Blastocystis induced a perturbation of the epithelial barrier that may lead to Blastocystis- related gastrointestinal symptoms.

As one of the major physiological functions of intestinal epithelium, barrier function is crucial to regulate the back and forth flow of contents between the gut lumen and sub-epithelial tissue (Turner, 2009). A breakdown of epithelial barrier function has been the cause of a wide range of gastrointestinal disorders particularly bacterial enteritis, celiac disease and inflammatory bowel disease (Bjarnason et al., 1995, Chin et al., 2002). Increased permeability of tight junctions may cause or contribute to diarrhea by exposing the sub-epithelial cells to parasitic virulence factors and to help the invasion of the organism (Spitz et al., 1995). Modulation of intestinal epithelial barrier function is one of the major mechanisms employed by pathogens to induce host pathology and several extracellular luminal parasites have evolved mechanisms to sabotage this gate function and cause bowel pathology by increasing intestinal epithelial permeability. There are a few reports of Blastocystis compromising intestinal epithelial barrier in rats, in- vitro and in-vivo (Mirza et al., 2012b, Puthia et al., 2006). There is also one clinical study correlating increased intestinal absorption of Technetium-99m labeled diethyl triamine penta acetic acid (99mTc-DTPA), with Blastocystis infections in humans (Dagci et al., 2002). The cause of Blastocystis-associated diarrhea is so far unidentified, but an association between increase in epithelial

106 permeability and diarrhea has been described for other intestinal diseases

(Madara, 1988). It is possible that Blastocystis-induced epithelial barrier function alteration might lead to electrolyte imbalance and thus diarrhea.

It was found that the functional disturbance in intestinal barriers by

Blastocystis was associated with molecular changes in the tight junction complex. These changes are coupled to disruption of ZO-1, degradation of

Occludin in the membrane-associated fraction, Claudin-1 degradation as well as redistribution to the apical region. No obvious change was observed in adherens junctions. Tight junctions are key regulators of epithelial barrier function (Turner, 2006). Numerous enteric pathogens disrupt epithelial tight junctions and alter intestinal permeability (Berkes et al., 2003), or utilize tight junctions during the pathogenesis (O'Hara et al., 2008). The strategies to overcome the tight junction barrier employed by enteropathogens are diverse.

Whereas some pathogens bind and modify junction components directly, by downregulating at translational or protein levels by degradation (Wan et al.,

1999, Wu et al., 2000) or at transcriptional level (Nazli et al., 2010), others exert their effects through the actin cytoskeleton (Scott et al., 2002), which controls the integrity of tight junctions. For example, Giardia intestinalis and

Trichomonas vaginalis caused a defect in barrier function by altering the distribution of the ZO-1 and ZO-2 proteins (da Costa et al., 2005, Maia-

Brigagao et al., 2012). Entamoeba histolytica triggered disruption of the TJ proteins via the dephosphorylation and degradation of ZO-1 (Leroy et al.,

2000). Acathamoeba castellanii, a pathogenic free-living amoeba, releases secretory products (mainly proteases) into the culture medium, activates

Rho/Rho-kinase pathway resulting in the degradation of ZO-1 and occludin in

107 human brain microvascular endothelial cells (HBMEC) (Khan et al., 2009).

However, the reports are not always consistent even for one microorganism, which not only depends on the experimental conditions or models used (Maia-

Brigagao et al., 2012), but also because a single pathogen could engage diverse strategies to sabotage the TJ (Sakaguchi et al., 2002). For example, level of ZO-1 expression was affected in G. duodenalis infection (Buret et al.,

2002), but in another study, ZO-1 migration from the cellular margins into the cytoplasm was the mainly observed mechanism (Scott et al., 2002). reiterating the complexity of the mechanisms used by pathogens in disassembly of epithelial TJ.

How adhesion leads to tight junction disruption should be investigated further.

The adhesion of the parasites to the host cells may be activating or deactivating a signalling cascade (Maia-Brigagao et al., 2012), leading to the destabilisation of the junctional proteins with a loss of their functionality.

These effects may be due to release of secreted products, like toxins or proteases by the parasite during adhesion on epithelial cells. However, further studies will be necessary to support these hypotheses in Blastocystis.

The current study for the first time provided a comprehensive analysis with cellular fractionation, which enables us to see the distribution of specific proteins upon Blastocystis infection, as some microorganisms induced a partitioning of those proteins to induce permeability changes (Man et al., 2010,

Scott et al., 2002). The results not only provided molecular basis for the epithelial functional disturbance, mediated mainly by degradation of tight junction protein, but also unraveled interesting events in parasite pathogenesis and revealed interesting patterns of changes in tight junction protein-Claudin-

108

1. An overall reduction in protein expression was found in Claudin-1 after infection with Blastocystis. Interestingly enough, it also showed redistribution from the cytosol to the membrane-associated fractions. Immunofluorescence revealed that Claudin-1 migrated to the apical regions and was found to colocalize with the parasites, indicating interaction between the parasite and the host tight junction proteins. The redistribution of Claudin-1, however, was not without precedent. Similar observations have been found in

Campylobacter jejuni infection (Chen et al., 2006), where Claudin-1 was shown to increase in raft-associated fraction. So far, the reason is not clear why parasites want to up-regulate Claudin-1 in the membrane-associated fraction. It is intriguing to think Claudin-1 could be a potential interacting partner of Blastocystis for it to stabilize and develop further pathogenesis. The imaging data also supported the association of parasite with host protein Claudin-1. Future work on the protein-protein interaction between parasite factors and host tight junction protein Claudin-1 will shed new light on the strategy the parasite uses to target the host cells.

The study also implicates an unusual parasitic cysteine protease, 1D5-binding legumain in pathogenesis. The result showed that legumain was transferred onto the enterocytes upon interaction. Blastocystis legumains have been shown to localize on the cell surface as well as in the cytoplasm and central vacuole

(Wu et al., 2010), and have also been found to be secreted. They were described as possible Blastocysistis virulence factors (Wawrzyniak et al.,

2012). Similar phenomena of antigen transfer have been reported in

Entamoeba to host cells where Gal/GalNAc specific lectin as well as lipophosphopeptidoglycans (LPPG) were deposited by the trophozoites to the

109 enterocytes following adhesion and prior to dysfunction of tight junctions, suggesting antigen transfer could play a role also in the breakdown of target cell membrane integrity (Lauwaet et al., 2004b, Leroy et al., 1995). Besides, the deposited legumain in infected cells appeared to be most prominent in regions of epithelial cell-cell contact, indicating transfer site specificity. How it is specifically bound to the junctional area of enterocytes remains to be investigated. Legumain in cancer cells has been shown on the cell surface and colocalizes with β1 integrins (Liu et al., 2003). Although integrins are found generally associated with the basolateral cell surfaces of enterocytes (Beaulieu,

1999, Hynes, 1992), data has indicated that β1 integrins are exposed also apically in the tight junction area in Caco-2 cells (Levy et al., 1998, Tafazoli et al., 2000). For example, Yersinia was shown to bind to apical β1 integrins, and consequently modified host F-actin organization and affected tight junction structure and function (Tafazoli et al., 2000). Another possibility is that when tight junction is disrupted and epithelial cells loose polarization, integrin molecules could migrate to the apical (Muza-Moons et al., 2003).

Disruption of tight junctions after Blastocystis infection was clearly shown in the results above, whether integrin is redistributed from basolateral to apical, serving as a binding site for legumain would be interesting to investigate.

Interestingly, integrin regulates the activity of several members of the Rho family of small GTPases, which control the contraction of filamentous actin fibers (DeMali et al., 2003, Huveneers et al., 2009). A recently study showed

Blastocystis was able to induce reorganization of actin cytoskeleton in a

Rho/ROCK-dependent fashion (Mirza et al., 2012b). Future study addressing whether Blastocystis legumain binds to integrin molecules on the enterocytes

110 and initiates integrin signaling leading to actin polymerization via Rho/ROCK pathway would help elucidate the role of Blastocystis legumain binding to human enterocytes.

Altogether, these results demonstrate the functional disturbance as one significant resultant host cell effects by parasite attachment and also revealed the complexity of molecular events and quite a few molecules participating in early host-parasite relationship.

111

CHAPTER 5:

HOST RESPONSES-PART II:

ROLE OF BLASTOCYSTIS-INDUCED

HUMAN ENTEROCYTE APOPTOSIS IN

MEDIATING BARRIER DYSFUNCTION

112

5.1 INTRODUCTION

Chapter 4 characterized the molecular basis for barrier disruption following

Blastocystis adhesion. Interestingly, recent evidence has suggested various luminal parasites sabotage this gate function by utilizing the apoptotic machinery of the host cell (Betanzos et al., 2013, Chin et al., 2002). It would be important to investigate if an increased incidence of cell death is a cellular basis contributing to the intestinal barrier defect upon parasite infection.

Apoptosis, a regulated programmed cell death, is a physiological process which plays an important role in the normal turnover of cells, maintaining a correct cell number in organisms by balancing cell growth and death, and is also involved in various pathological processes (Earnshaw, 1995).

Biochemically, apoptosis is the orderly, immunologically silent dismantling of a cell following the activation of a family of cysteine proteases known as caspases (Danial et al., 2004). There are two central pathways which activate caspases (Elmore, 2007). The extrinsic pathway is activated following the binding of a ligand to a death receptor on the cell surface (e.g. FasL/Fas, TNF-

α/TNFR), and activation of the receptor results in the formation of the death- inducing signalling complex (DISC) (Muzio et al., 1996), which recruits and activates the initiator caspase, caspase 8 (Peter et al., 2003). Active caspase 8 then cleaves and activates an executioner caspase, caspase 3, which is responsible for cleaving multiple proteins and the subsequent death of the cell

(Peter et al., 2003). The intrinsic pathway, initiated by intracellular signals (e.g.

DNA damage), relies on the release of cytochrome c from mitochondria to activate the initiator caspase, caspase 9, in a macromolecular complex called

113 the apoptosome (Green, 2005, Opferman et al., 2003). As with the extrinsic pathway, the initiator caspase then activates caspase 3 and the cell undergoes apoptosis (Green, 2005). The activation of caspase 8 can also lead to the release of cytochrome c due to the cleavage of the Bcl-2-family protein Bid – serving as a nexus between the intrinsic and extrinsic cascades (Kuwana et al.,

1998, Li et al., 1998). Apoptosis or programmed cell death is a mechanism of clearing up unwanted cells by the body while simultaneously limiting overt immune response (Gao et al., 1998). Recently, a variety of intestinal pathogens were found to induce apoptosis in intestinal epithelial cells, and their association with pathogenicity was suggested (Crane et al., 1999, Huston et al., 2003, McCole et al., 2000, Valenti et al., 1999). Entamoeba induces enterocyte apoptosis to facilitate the parasite infection of gut (Becker et al.,

2010), while Giardia induces epithelial barrier compromise by activating caspase 3-mediated enterocyte programmed cell death (Chin et al., 2002). A list of parasite-induced apoptosis and the role of apoptosis in mediating barrier function was listed in Table 5.1. Blastocystis was reported to induce caspase- mediated apoptosis in vitro (Puthia et al., 2006), with no effect on the barrier function of rat intestinal epithelium, but this finding was limited to rodent models. An association of parasite-induced apoptosis with disruption of epithelial barrier function, as observed in Giardia and Entamoeba infections is not known in human intestinal epithelium.

Using ST-7 (B) isolate, recovered from a symptomatic patient, which was demonstrated to adhere to intestinal cells and cause epithelial damage in

Chapter 2, the study aimed to find whether Blastocystis treatment induces apoptosis in Caco-2 cells and whether apoptosis could contribute to the

114 observed impairment in barrier function. A control strain ST-4 (WR-1), which was recovered from a rat and not adhesive to Caco-2 cells, would also be used to test whether there exists a difference in adhesive strain and non-adhesive strain in their capability of inducing enterocyte aopotosis and barrier dysfunction. The study reported for the first time that Blastocystis ST-7 (B) induced caspase-3- and 9-mediated apoptosis, breakdown of epithelial barrier function, and rearrangement of the tight junction protein ZO-1. Caspase inhibition in human gut epithelium prevented the effect of parasite on barrier dysfunction. In contrast, WR-1, previously reported to induce rat epithelial injury (Puthia et al., 2006), did not cause any pathology in human epithelium.

Table 5.1 Studies on induction of apoptosis in protistan parasites and their role in mediating barrier function

Association with Intracellular Parasite Study Barrier Apoptotic Pathway Dysfunction Caspase-3; independent Huston et al, of the upstream -- (2000) caspases 8 and 9 Ragland et al., Entamoeba Independent of Bcl-2 -- (1994) histolytica Faust et al., -- -- (2011) Betanzos A, et -- Yes al., (2013) Panaro et al., caspases -8, -9 and -3 -- (2007) Giardia lamblia Chin et al., Caspase-3 Yes (2002) Sasahara et al., -- -- (2003) Cryptosporidium Chen et al., -- Yes parvum (2002) McCole et al., -- -- (2000) Trichomonas Sommer, et al., Caspase-3 -- vaginalis (2005)

Puthia et al., Blastocystis Caspase-3 No (2006)

115

5.2 MATERIALS AND METHODS

5.2.1 Culture of Caco-2 colonic epithelial cell line

Caco-2 stock cultures were maintained as previously described in Chapter 2.

Cell cultures for Western blotting and flow cytometry experiments were grown on standard cell culture 24-well plates (Corning). For confocal imaging, cells were cultured on poly-L-lysine-treated 24-mm glass coverslips, placed in standard 6-well culture plates (Corning). For permeability experiments, cells were grown on Millipore transwell filters as previously described in Chapter 4.

For caspase inhibition experiments, cell cultures were pretreated with 40 μM broad-spectrum caspase inhibitor, z-VAD-fmk (Sigma) for 4 h. Cytochalasin

D (Sigma) was used as a positive control for permeability experiments at a concetration of 1 μg/ml. Staurosporine was used as a positive control for induction of apoptosis and ZO-1 rearrangement experiments at a concentration of 0.5 μM.

5.2.2 Parasite culture and lysate

Two axenized Blastocystis isolates belonging to different subtypes (ST) were used in this study. Axenized isolates of Blastocystis ST-7 (B) was maintained as previously described in Chapter 2. Isolate WR-1, belonging to ST-4, was isolated from a Wistar rat during an animal survey (Chen et al., 1997a). Both parasite cultures were maintained in the same conditions. One-day-old parasite cultures were harvested for lysate preparation. The cultures were washed twice in sterile PBS. Parasitic lysates were prepared by three freeze-thaw cycles in liquid nitrogen and 37°C water bath.

116

5.2.3 Epithelial permeability

Epithelial permeability measurements after co-incubated with lysate of ST-7

(B) and ST-4 (WR-1) were performed as described in Chapter 4.

5.2.4 Flow cytometry

Annexin-V binding assay was used to observe early apoptotic changes in

Caco-2 cells. Caco-2 monolayers were grown in 24-well culture plates and coincubated with Blastocystis ST-7 or ST-4 for 3 h. Annexin-V-FITC apoptosis detection kit (BioVision) was used according to manufacturer's instructions. Propidium iodide (PI) was used to exclude necrotic cells. After cell staining, samples were analyzed using a flow cytometer (Dako

Cytomation; Cyan LX) at 488-nm excitation wavelength, with a 515-nm band- pass filter for fluorescein detection, and a 600-nm filter for PI detection. The lower right quadrant was defined to represent the apoptotic cells showing annexin-V-FITC-positive and PI-negative staining.

5.2.5 Confocal microscopy

Immunostaining for ZO-1 tight junction protein was performed as described in

Chapter 4. The ImageJ software was used for intensity measurement.

5.2.6 Western blot

ZO-1 Western blot analysis was done as previously described in Chapter 4.

5.2.7 Statistical analysis

The ANOVA test was used to confirm the statistical significance of the results.

117

5.3 RESULTS

5.3.1 ST-4 (WR-1) strain could not establish enteroadhesion and cause disruption of tight junction, in contrast to the enteroadhesive strain of

Blastocystis ST-7 (B)

The ability of different Blastocystis strains in adhesion and causing leisions in tight junction protein was firstly investigated. Chapter 2 characterized ST-7

(B) as an adhesive strain and Chapter 4 showed that it could cause degradation of ZO-1 tight junction proteins upon adhesion. Here with a strain isolated from rat, we demonstrated compared to ST-7 (B), ST-4 (WR-1) is not adhesive to

Caco-2 cells and did not produce obvious changes in ZO-1.

Similar to the changes shown in Chapter 4, ST-7 (B) disrupted integrity in ZO-

1 tight junction protein (Fig. 5.1), and caused focal disruption (white arrows) and obvious reduction in ZO-1 apical localization. The ZO-1 proteins in WR-

1-treated monolayers remained well preserved and apically localized in the monolayer, similar to that in the control monolayer.

Therefore, by using a control strain, the results provide the first visual evidence that the ability in causing leisions in tight junction proteins of

Blastocystis is directly associated with their enteroadhesive ability.

118

Fig. 5.1 Representative confocal micrographs illustrating attachment of Blastocystis to Caco-2 monolayers resulted in changes in ZO-1 tight junction protein. Caco-2 monolayers were grown to confluency on glass converslips and were then co-incubated with the same number of Blastocystis ST-4 (WR-1) and ST-7 (B) parasites pre-stained with CFSE (green). Normal culture media was used as a negative control. After co-incubation, the non- attached parasites were washed away. Caco-2 cells were then stained with DAPI and labeled with an antibody targeting tight junction protein ZO-1 and Cy3® goat anti-mouse IgG (Red). Blastocysis ST-7 (B) attached to Caco-2 to a greater extent than ST-4 (WR-1) as seen by more parasites stained with CFSE. Compared to negative control, ST-7 (B) treatment resulted in disruption of ZO-1 tight junction protein, as shown by the decreased intensity, focal disruption (white arrows) and obvious reduction in ZO-1 apical localization (yellow arrows) in Caco-2 cell monolayer. However, Blastocystis ST-4 (WR1) isolate, which could not adhere, did not produce obvious changes in ZO-1 tight junction protein. It could also be noted that parasites preferably adhere to the sites of tight junctions where ZO-1 protein localizes. Scale bar = 10 μm

119

5.3.2 Blatocystis ST-7 (B) induces characteristics of apoptosis in Caco-2 cells; ST-4 (WR-1) does not

(i) PS-flip indicated by annexin-FITC binding

One of the early indicators of apoptosis is the flipping of phosphatidylserine

(PS) molecules from the inner to outer leaflet of the plasma membrane. A 35- to 36-kDa molecule, Annexin-V, binds to PS with high specificity in the presence of calcium. Viable, apoptotic, and necrotic cells can be distinguished when FITC conjugated-annexin-V is used in conjunction with PI. The lower right quadrants of the dot plots represent the apoptotic cell population, positive for annexin binding but PI-negative (Fig. 5.2). Upper quadrants represent necrotic cells due to permeability to PI (Fig. 5.2). After treatment with

Blastocystis ST-7, Caco-2 cells exhibited a significant rise in percentage of apoptotic cells compared to the negative control and ST-4 treatment (p<0.01)

(Fig. 5.2). These findings are in agreement with a strain-dependent pathogenecity of the parasite.

120

Fig. 5.2 Flow cytometry analysis of annexin V-FITC and propidium iodide staining. Representative dot plots of Caco-2 cells pre-incubated with culture media as negative control, Blastocystis ST-7, ST-4 and 0.5 μM staurosporin as positive control. 2 × 104 cells were analysed in each sample. Values represent mean ± standard error (error bars; n=3). Caco-2 cells treated with ST-7 exhibited significantly higher percentage of annexin V+ and PI- cells (lower right quadrant) compared to cells co-incubated with ST-4 or culture media only (*, p< 0.01 versus control and ST-4). Values are means ± standard error (error bars) (n=6).

121

(ii) Nuclear fragmentation indicated by DAPI staining

One of the most distinctive features of apoptosis is morphological change in the nucleus, easily observed under florescence microscopy (Fig. 5.3). After 24 hr treatment with Blastocystis ST-7, nuclei of the Caco-2 cells exhibited nuclear condensation and fragmentation (Fig. 5.3) typical of apoptotic cells.

Significantly higher number of apoptotic cells was observed in ST-7-treated membranes compared to ST-4 (p<0.01) and control (p< 0.01) (Fig. 5.3).

Fig. 5.3 Representative fluorescence micrographs after DAPI staining showing apoptosis of Caco-2 cells. Caco-2 cells were incubated for 24 h with culture media, ST-4, ST-7 or 0.5 μM staurosporine as positive control. Cells co-incubated with ST-7 and staurosporine exhibit nuclear fragmentation and condensation (arrow) typical of apoptotic cells. Histogram represents percentage of apoptotic cells after DAPI fluorescence assay. Caco-2 monolayers coincubated with ST-7 exhibited significantly higher percentage of apoptosis (*, p<0.01) compared with ST-4 and the negative control. Values are means ± standard error (n=6 per group). For each sample, ~100 cells were counted at 1000× magnification.

122

(iii) Caspase 3 and 9 activation

Caspases are pro-enzymes, which are activated by being cleaved into active fragments in apoptotic cells. In this study, Western blot analysis revealed that

Blastocystis ST-7 coincubation with Caco-2, resulted in cleavage of caspase 3 and 9 (Fig. 5.4). No cleavage of caspase 8 was observed even after 12 h treatment with ST-7 (Fig. 5.4) suggesting that Blastocystis ST-7 induced

Caco-2 programmed cell death by activation of the intrinsic apoptotic pathway. ST-4, as expected, did not activate any of the three caspases tested in this study (Fig. 5.4).

Fig. 5.4 Western blot analysis of caspase activation (cleavage) in Caco-2 cells. Caco-2 cells were grown on cell-culture plates and harvested after coincubation with Blastocystis ST-4 and ST-7 for 6 and 12 h, as described in Materials and methods. Treatment with ST-7 resulted in loss of Caspase 3 and 9 bands suggesting activation, while caspase 8 remained unchanged. ST-4 did not cause any cleavage of caspases in Caco-2.

123

(iv) Ultrastructural changes

Morphological changes were observed by transmission electron microscope in

Caco-2 cells treated with Blastocysitis for 6h. The cells displayed obvious characteristic changes of apoptosis, including chromatin condensation and mitochondrial denaturation such as swelling and disappearance of mitochondrial cristae in the Blastocysitis-treated cells (Fig. 5.5, Fig. 5.6).

Fig. 5.5 Transmission electron micrograph showing ultrastructural changes of Caco-2 cell monolayer induced by Blastocysitis treatment. Caco-2 cells were co-incubated with Blastocysitis for 6 hours and then prepared for viewing under a transmission electron microscope. One prominent feature is the changes in mitochondria. Compared to the condensed mitochondria in the normal control, abnormal mitochondria with swollen appearances were observed after Blastocysitis treatment (red aroow). The cristae structure in the mitochondria appeared to be dirsrupted. The surface microvilli also were erased at the apical side in the Blastocysitis-infected cells. Scale bar=1μm

124

Fig. 5.6 Transmission electron micrograph showing ultrastructural changes of Caco-2 cell monolayer induced by Blastocysitis treatment. Caco-2 cells were co-incubated with Blastocysitis for 6 hours and then prepared for viewing under a transmission electron microscope. The changes in mitochondria again were noticed, with swollen appearances and disrupted cristae (red arrow). Chromatin condensation as well as formation of vacuoels were found in the nuclei of infected cells (yellow arrows). Scale bar=5 μm

125

5.3.3 Blastocystis ST-7 (B) induces increase in permeability to FITC- conjugated Dextran; ST-4 (WR-1) does not

Chapter 4 demonstrated ST-7 (B) was able to induce a significant increase in epithelial permeability. For monolayers treated by ST-4 (WR-1), on the other hand, no significant flux of Dextran-FITC probe across Caco-2 monolayers was noticed (Fig. 5.9).

5.3.4 Inhibition of caspase prevented Blastocystis ST-7-induced ZO-1 rearrangement and epithelial barrier dysfunction in Caco-2

It has been reported previously that tight junction disruption and host epithelial dysfunction induced by luminal parasites is caused by increased apoptosis in enterocytes (Chin et al., 2002). Confocal micrographs and

Western blot analysis showed that inhibition of host caspases by z-VAD-fmk prevented ST-7-induced ZO-1 changes (Fig. 5.7, and Fig. 5.8). Also, the increase in permeability of Caco-2 monolayers induced by ST-7 (B) was significantly prevented (p<0.01) (Fig. 5.9). These findings suggest a role of parasite-induced apoptosis in epithelial barrier dysfunction.

126

Fig. 5.7 Representative confocal micrographs illustrating ZO-1 integrity in Caco-2 monolayers. Caco-2 monolayers were grown to confluency on poly-L-lysine treated converslips. Caco-2 cells were then conincubated for 12h with either Blastocystis ST-4 or ST-7. Some monolayers were treated with broad-spectrum caspase inhibitor z-VAD-fmk before coincubation with ST-7. Normal culture media and 0.5 μM staurosporine were used as negative and positive controls respectively. Compared to negative control, ST-7 treatment resulted in obvious reduction in ZO-1 apical localization in Caco-2 cell line. ST-4 did not alter ZO-1 integrity. Pretreatment with z-VAD-fmk rescued ST-7 induced ZO-1 changes in the epithelium. (B) Quantification of ZO-1 staining shown as graphs. Each cell layer (1–25) corresponds to series of images from Z-stack sections taken at 1µm thickness through the cell monolayer shown in (A). X-axis illustrates cell layers from apical to basolateral. Y-axis illustrates the number of pixels present over the entire area of image. Monolayers treated

127 with ST-7 and staurosporine resulted in marked reduction in number of pixels in cell layers representing apical region, compared to ST-4 treated epithelium and normal control. Pre-treatment of epithelium with broad-spectrum caspase inhibitor, z-VAD-fmk resulted in inhibition of ST-7 induced ZO-1 changes in the monolayer. Results shown are mean of 4 separate Z-stacks for each treatment.

Fig. 5.8 Western blot analysis of ZO-1 integrity in Caco-2 epithelium. Caco-2 monolayers were cultured in cell-culture plates and harvested after coincubation with Blastocystis ST-4, ST-7, normal growth media and 0.5 μM staurosporine. Monolayers were also treated with broad-spectrum caspase inhibitor z-VAD-fmk before incubation with ST-7. After coincubation with ST-7 and staurosporine for 6 h, loss of ZO-1 band was observed. The ZO-1 band remained unchanged in ST-4 treated and control Caco-2 cells. Caspase inhibition resulted in rescue of ST-7-induced loss of ZO-1.

128

Fig. 5.9 Inhibition of Caco-2 caspases prevented Blastocystis ST-7-induced epithelial barrier dysfunction. Flux measurement with FITC-conjugated Dextran. Confluent monolayers of Caco-2 cells were co-incubated for 24 h with Blastocystis ST-4 or ST-7. Some monolayers were co-incubated with ST- 7 after pretreatment of cells with the broad-spectrum caspase inhibitor Z- VAD-fmk. Permeability was determined by measurement of Dextran-FITC fluxes across the monolayer as described in Materials and Methods. A significant increase in the epithelial permeability can be noticed after incubation with ST-7, compared to negative control and ST-4 coincubation (p<0.01). Pretreatment of Caco-2 cells with caspase inhibitor significantly rescued these cells from Blastocystis-induced effect on permeability. Values are means ± standard error (error bars) (n=6). (*, p<0.01 vs. control and ST-4; #, p<0.01 vs. ST-7)

129

5.4 DISCUSSION

The study reported that Blastocystis ST-7 (B) isolate recovered from a symptomatic patient adhered to the intestinal epithelium and induced caspase- mediated enterocyte apotosis and epithelial barrier dysfunction in Caco-2 human epithelial cell-line. Interestingly an ST-4 isolate, WR-1, recovered from an asymptomatic host did not adhere to the epithelium or induce enterocyte pathology. The study provides the evidence of a strain-to-strain variation in Blastocystis induced intestinal epithelial barrier compromise and suggests that this variation might be determined by enterocyte adhesion.

This was also the first report that Blastocystis caused apoptosis in human epithelial cells. Apoptosis or programmed cell death is a pre-programmed method of clearing up unwanted cells by the body (Savill, 1997) without overtly stimulating a host immune response (Gao et al., 1998). Cells undergo apoptosis even under physiological conditions, but several pathogens including parasites have evolved mechanisms to disrupt this machinery, by either up-regulating (Chin et al., 2002, Huston et al., 2003) or down-regulating

(Yamada et al., 2011) it for their survival in the host. Blastocystis ST-7 induced both early and late hallmarks of apoptosis, i.e. PS-externalization and nuclear fragmentation respectively, in host enterocytes. Increased enterocyte apoptosis in some cases is also proposed to be a host response to infection by increasing the cell turnover in order to rid the body of the infected cells (Cliffe et al., 2005). Mucosal sloughing reported during Blastocystis infections (Chen et al., 1997a) might be a result of this increased turnover. Anti-inflammatory effects of apoptosis are also well recognized (Haslett, 1997). The relatively

130 moderate level of apoptosis induced by Blastocystis in intestinal epithelium, is similar to programmed cell death of enterocytes when exposed to

Cryptosporidium (McCole et al., 2000) and Giardia (Chin et al., 2002). It is suggested that up-regulation of host cell apoptosis might be a reason for lack of overt host inflammatory response during parasitic infections (Chin et al.,

2002, McCole et al., 2000) and might assist them in colonizing the hostile host environment. Although further investigation is needed in this area, the absence of obvious gut inflammatory changes in Blastocystis infected hosts might be due to the ability of the parasite to down-regulate host inflammatory response by induction of enterocyte apoptosis.

The first study also identified that Blastocystis induced epithelial apoptosis by activation of the intrinsic pathway. Caspase 3 activation lies at the center of the caspase-mediated apoptosis (Savill, 1997). It is preceded by activation of either intrinsic pathway, involving mitochondrial injury and capase 9 activation or extrinsic pathway due to Fas/FasL receptor mediated caspase 8 activation. Parasites have evolved complex mechanisms to activate epithelial caspase 3. Giardia activates extrinsic as well as intrinsic apoptotic pathways

(Panaro et al., 2007). Entamoeba on the other hand does not require either caspase 8 or 9 for enterocyte caspase 3 activation (Huston et al., 2000). An earlier study reported that Blastocystis ST-4 induced rat epithelial apoptosis by caspase 3 activation (Puthia et al., 2006). There is no data available suggesting a similar outcome in human epithelium. Upstream pathways involved in caspase 3 activation by the parasite are not known either. In this study the rodent strain had no effect on Caco-2 cell line, but ST-7-induced caspase 3 activation. Blastocystis unlike Giardia or Entamoeba only activated caspase

131

9. A recent study suggested the involvement of Rho kinases in selective activation of caspase 9, leading to apoptosis (Del Re et al., 2007). A role of

Rho kinase has been suggested in Blastocystis-induced breakdown of host epithelial barrier function, cytoskeletal rearrangement (Mirza et al., 2012b).

Although more data is required, selective activation of caspase 9 by

Blastocystis in this study might be a result of epithelial Rho kinase modulation by the parasite. Further understanding of the unique cellular mechanisms employed by Blastocystis to induce host-cell apoptosis might help us develop targeted therapeutics to prevent parasite induced host pathology.

Apoptosis also plays a diverse role in the modulation of epithelial barrier function and tight junction reorganization. On one hand enterocyte apoptosis ensures that epithelial barrier remains sealed (Marchiando et al., 2011), while in other cases induction of apoptosis is employed by pathogens to increase epithelial permeability (Chin et al., 2002) and cause host pathology (Becker et al., 2010). Inhibition of host caspases resulted in prevention of Giardia- induced modulation of epithelial permeability and ZO-1 organization (Chin et al., 2002) while, inhibition of caspase-mediated apoptosis in enteropathogenic

E. coli (EPEC) infections did not prevent the epithelial ZO-1 alterations (Buret et al., 1998). In a recent study on rodent epithelium, pretreatment of host epithelium with pan-caspase inhibitor z-VAD-fmk did not rescue Blastocystis

ST-4-induced epithelial barrier dysfunction (Puthia et al., 2006). In the current study, z-VAD-fmk treatment of epithelium significantly inhibited Blastocystis

ST-7-induced epithelial barrier compromise. Parasite-induced ZO-1 alteration was also prevented significantly by host caspase inhibition, reiterating that role of enterocyte apoptosis in parasite-induced epithelial barrier dysfunction.

132

Interestingly, changes in ZO-1 were also shown to be prevented by Rho kinase

(Mirza et al., 2012b), again raising the question of the role of Rho kinase in capase-mediated apoptosis.

To conclude, this is the first study showing that enteroadhesive Blastocystis

ST-7 (B), an isolate recovered from a symptomatic patient, induced enterocyte apoptosis by activating caspase 3 and 9, suggesting the involvement of the intrinsic apoptotic pathway in pathogenesis. Inhibition of host caspases prevented parasite induced epithelial barrier dysfunction as well as ZO-1 rearrangement suggesting the role of caspase-dependent enterocyte apoptosis, induced by Blastocystis, in host epithelial barrier dysfunction. Furthermore, the inability of rodent subtype ST-4 to adhere and induce any changes in

Caco-2 provides evidence of host-specificity and strain-dependency in

Blastocystis-induced human epithelial pathology. The strain-to-strain variation in parasite virulence is a plausible explanation for the large number of asymptomatic human carriers of Blastocystis.

133

CHAPTER 6:

STRAIN-DEPENDENT VARIATIONS IN

ADHESION AND ITS POSSIBLE ROLE

IN MEDIATING BARRIER

DYSFUNCTION

134

6.1 INTRODUCTION

As reviewed in Chapter 1, it is difficult to argue the clinical significance and pathogenic potential of Blastocystis, since infections do not consistently lead to intestinal symptoms (Eroglu et al., 2009, Kaneda et al., 2000, Stensvold et al., 2009c). While a large number of infected individuals present with clinical symptoms (Eroglu et al., 2009), asymptomatic carriage of the parasite is also common (Boorom et al., 2008). Moreover, in symptomatic patients, the duration and severity of symptoms vary from acute enteritis to chronic mild diarrhea (Nagel et al., 2012, Stensvold et al., 2009c). There is no consensus on the possible reasons for the observed diverse intestinal symptoms. A number of reports have suggested a strain- or subtype-dependent variation in parasite pathogenicity (Dominguez-Marquez et al., 2009, Mirza et al., 2011b, Nagel et al., 2012, Scanlan, 2012, Stensvold et al., 2009b, Stensvold et al., 2009c).

Studies have associated ST-1, -4 and -7 with pathological alterations in humans, as reviewed in Chapter 1. The presence of both pathogenic and nonpathogenic strains within one subtype has also been well reported (Hussein et al., 2008, Iguchi et al., 2007). Yet the factors determining this variation in pathogenicity across different Blastocystis strains have not been resolved.

Attachment is a crucial step in the expression of the cytopathogenicity of extracellular parasites. Studies have suggested the level of adhesion is directly linked to the virulence properties of pathogens (Brooks et al., 2013, Park et al.,

2009). Highly adhesive strains of Giardia, Entamoeba, Trichomonas and other eukaryotes have been shown to result in more severe damage of the epithelium compared with less adherent strains (Flores-Romo et al., 1997, Muller et al.,

135

2005, Rojas et al., 2004). Chapter 2 identified a strain which adhered and caused pathological changes. Chapter 4 and 5 investigated in-depth the structural and functional changes in epithelial cells after parasite adhesion. It is important to determine adhesiveness of the parasite with enterocytes across different Blastocystis strains and investigate its association with parasite pathogenicity. In Chapter 5, with two strains exhibiting varied levels of enteroadhesion, there was also a variation in the induction of apoptosis and barrier compromise. However, it was limited to two strains. To see how the ability in attachment would affect their pathogenicity, more strains need to be included.

Aside from adhesion, which is an important measurement of virulence in inducing intestinal pathology, it’s also important to compare specific virulence factors contributing to pathogenicity. Cysteine proteases are suspected to be

Blastocystis virulence factors and have been shown to play a role in epithelial barrier compromise (Mirza et al., 2012b). Proteases from Entamoeba,

Acanthamoeba and other pathogens were reported to induce epithelial barrier dysfunction (Groeger et al., 2010, Lauwaet et al., 2004a). Both qualitative and quantitative differences in cysteine protease profiles have also been reported between pathogenic and non-pathogenic strains of E. histolytica (Davis et al.,

2007, Hirata et al., 2007, Reed et al., 1989) and G. intestinalis (DuBois et al.,

2006, Guimaraes et al., 2003). Previously, it was shown that Blastocystis ST-4 and ST-7 exhibited extensive variations in protease activity (Mirza and Tan,

2009). But it was limited to two strains from each subtype, moreover, a comparison of the cytopathic effects of these strains is still lacking. It’s necessary to investigate the differences in protease activity among a wider

136 range of Blastocystis strains and see whether it correlates with the parasites’ ability to induce epithelial pathology.

Another issue complicating the pathogenic potential of Blastocystis is reports of treatment failure (Stensvold et al., 2008, Stensvold et al., 2010). Non- responsiveness to Mz chemotherapy has been frequently reported (Stensvold et al., 2009c). Strain-to-strain variation within Blastocystis in susceptibility to

Mz and other antiparasitic agents is commonly reported and has been proposed to be the reason for frequent treatment failures in parasite infections

(Mirza et al., 2011a). However, from an evolutionary standpoint, mutations associated with drug resistance may impair essential biological functions or impose energy demands on the organism, leading to decreased microbial fitness (Andersson et al., 2010, Ben-Ami et al., 2012). Studies of a variety of pathogens including different species of viruses, bacteria, and parasites, indicate that antimicrobial resistance places a toll on the organisms’ fitness as well as virulence (Muregi et al., 2011). A recent study in an intestinal protozoan parasite, Giardia, revealed impaired attachment and decreased infectivity in Mz resistant (Mzr) strains compared with parental Mz sensitive

(Mzs) strains (Tejman-Yarden et al., 2011) and was given as a possible reason for the scarcity of treatment failure in people with symptomatic parasite infections. The interplay among drug resistance, fitness and virulence in

Blastocystis has never been studied. Considering the frequent reports of treatment failure in symptomatic Blastocystis infections, it will be interesting to establish whether drug resistance exerts any effects on the pathogenicity as well as other aspects of parasite fitness.

137

In the present study, using seven isolates from Blastocystis ST-4 and ST-7, extensive intra- and inter-subtype variability in inducing intestinal barrier dysfunction was firstly demonstrated, and the question whether adhesion contributed to this variation was investigated. It was found that Mz resistance correlated with impairment of parasite adhesion and adhesion-associated cytopathic effects. Additionally, Mz resistance also seemed to be associated with nitrosative stress tolerance, an important factor assisting parasite survival in the gut lumen.

138

6.2 MATERIALS AND METHODS

6.2.1 Parasite culture

Seven axenized isolates of Blastocystis (designated B, C, E, G, H, S-1 and

WR-1) were used (Figure 6.1). Isolates B, C, E, G and H were originally recovered from symptomatic patients at the Singapore General Hospital (Ho et al., 1993). All five isolates belonged to ST-7 according to recent classification system as B (Stensvold et al, 2007b). ST-4 (WR-1 and S-1) strains were isolated from rats during animal survey (Chen et al., 1997a). Isolates were subtyped previously by small-subunit rRNA gene analyses (Noel et al, 2005).

Both ST-4 and ST-7 are well characterized zoonotic Blastocystis isolates commonly detected in humans with gastrointestinal symptoms (Stensvold et al,

2009c). Stock cultures of all seven isolates were maintained as isolate B as described in Chapter 2.

Fig. 6.1 Blastocystis ST-4 and ST-7 strains used in the study. Scale bar= 20 μm

139

6.2.2 Culture of Caco-2 colonic epithelial cell line

Caco-2 stock cultures were maintained as previously described in Chapter 2.

For Western blot analysis, Caco-2 monolayers were grown on 6-well cell culture plates (corning) until 100% confluency. For immunofluorescence and confocal microscopy experiments, Caco-2 monolayers were cultured on Poly-

L-lysine-treated 15mm glass coverslips placed in standard 24-well culture plates. For epithelial permeability experiments cells were grown on Millipore transwell filters with PET membranes of 3μm pore-size, placed in 24-well tissue culture plates. In order to synchronize cells before experiments, all cultures were serum-starved overnight in antibiotic-free and serum-free

DMEM.

6.2.3 Parasite viability assay

The optimized Blastocystis viability assay was used to measure 50% inhibitory concentrations (IC50s) of different drugs for the Blastocystis strains

(Mirza et al., 2011a). Briefly, stock solutions of drugs were prepared in dimethyl sulfoxide (DMSO), diluted in pre-reduced Blastocystis culture medium, transferred to 96-well plates and pre-reduced in anaerobic jar for 4 h at 37°C. Parasites were counted and 0.5×106 cells were incubated in each well of a standard 96-well plate with dilutions of different drugs ranging between 0 and 100 μg/ml. The drug concentration range was reduced or increased depending on the results of the first test. A range of parasite counts between zero and 0.5×106 per well were used for viability control. The final DMSO concentration was kept constant at 0.5% in each well. Since the parasite redox

140 activity varies with volume, the final total volume of each well was kept constant at 200 µl. After 24 h of drug exposure, resazurin solution (Sigma) was added to each well at a final concentration of 10% (v/v); 3 h after incubation under anaerobic condition at 37°C, fluorescence readings of resazurin were taken at 550-nm excitation and 580-nm emission wave lengths using a Tecan Infinite M200 reader. Values were imported into GraphPad

Prism5 software and IC50s were calculated.

6.2.4 Growth curve

The growth curve of each of the seven isolates was made by culturing over a period of 96h. Briefly, 5×106 parasites of each isolate were inoculated into

IMDM and cultured under anaerobic conditions at 37°C. At each of the five time points (0, 24, 48, 72 and 96 h), parasite pellets were resuspended and cell numbers were counted using appropriate dilutions in a hemocytometer.

6.2.5 Azocasein assay for cysteine protease activity

Parasite protease activity was determined using an azocasein assay as described previously (Sio et al., 2006). Briefly, lysates of 4×106 parasites were co-incubated with 2 mM dithiothreitol (DTT) (Sigma) at 37°C for 10 min to activate protease activity. Azocasein (Sigma), 100 μl of 5 mg/ml, solution was prepared in PBS (pH 7.4) and incubated with 100 μl of parasite lysate for 1 h at 37°C. The reaction was stopped by adding 300 μl of 10% trichloroacetic acid (TCA) and samples were incubated on ice for 30 min. Undigested azocasein was removed by centrifugation (5,000 × g for 5 min) and the resultant supernatant was transferred to a clean tube containing 500 μl of 525 mM NaOH. Absorbance was measured using a spectrophotometer at 442 nm

141

(Tecan Magellan). PBS was used as a negative control. For inhibition experiment, 2 mM iodoacetamide (IA) were added to the parasite lysate and incubated for 1 h at room temperature (22°C-24°C) to inhibit the cysteine proteases (Puthia et al., 2005).

6.2.6 Adhesion assay

Adhesion assay with all Blastocystis isolates were performed as described in

Chapter 3.

6.2.7 Epithelial permeability

Caco-2 monolayers were co-incubated with parasite live cells of all

Blastocystis isolates for 24 h. Epithelial permeability was then measured as described in Chapter 4.

6.2.8 Western blot

Western blot analysis for ZO-1, Occludin were performed as described in

Chapter 4. MLC and MLCp were done with primary antibody (mouse anti-

MLC, rabbit anti-MLCp; Sigma).

6.2.9 Confocal microscopy

Nitric oxide cytotoxicity against Blastocystis was investigated by confocal microscopy (Annexin-FITC/PI staining). Live Blastocystis were treated for 3 h with a 50 μg/ml concentration of GSNO (S-Nitrosoglutathione), a nitric oxide donor. After drug exposure, the parasites were washed and re-suspended in

Annexin V binding buffer (BioVision). Fluorescein isothiocyanate (FITC)- labeled annexin V and propidium iodide (PI) (BioVision) were then added to

142 the cell suspension at the ratio of 1:100. Imaging of cell suspensions was conducted using an Olympus Fluoview FV1000 (Japan). Images were captured using Olympus Fluoview v. 1.6b.

Visualization of tight junction proteins in Caco-2 cells were performed as described in Chapter 4.

To explore the spatial relationship between Blastocystis and Caco-2 monolayers after attachment, immunofluorescence was performed as described in Chapter 2.

6.2.10 Drug preparation

Stock solutions of each compound to be tested were prepared fresh in dimethyl sulfoxide (DMSO). For drug sensitivity determination, stock solutions were diluted in pre-reduced Blastocystis medium and transferred to 96-well plates and pre-reduced again for 4 h. The final DMSO concentration was kept constant at 0.5%.

6.2.11 Statistical analysis

The ANOVA test was used to confirm the statistical significance of the results. Correlation analyses between Mz resistance and fitness parameters (i. e. attachment, permeability increase and nitric oxide tolerance) were performed using the Pearson correlation method in GraphPad Prism 6

(GraphPad Software Inc., San Diego, CA, USA).

143

6.3 RESULTS

6.3.1 Blastocystis exhibits intra-and inter- subtype variation in ability to induce epithelial permeability increase in human epithelium

It has been previously reported that Blastocystis induces human intestinal barrier dysfunction in an in vitro model (Mirza et al., 2012b). Chapter 4 also identified the barier-disruptiong effects of Blastocystis, the study investigated the variability in the ability of different parasite isolates to breach the epithelial barrier. Live parasites of all Blastocystis isolates used in the study were applied apically on to Caco-2 cell monolayers differentiated on transwell inserts. Permeability changes caused by the parasites were analyzed by measuring the flux of Dextran–fluorescein isothiocyanate (FITC) probe across the intestinal epithelial barrier from the apical to the basolateral compartment.

Exposure to ST-4 isolates WR-1 and S-1 did not change permeability significantly compared with that of the control (Fig. 6.2). In contrast, all five

ST-7 strains of Blastocystis could induce significant increase in epithelial permeability compared with the control and ST-4-treated monolayers

(p<0.01). Within ST-7, extensive variations in the ability to induce permeability increase were observed (Fig. 6.2). Compared with the negative control monolayers, the fold increase in permeability to FITC-dextran in ST-7- infected monolayers were 16, 11.8, 14.5, 5.5, 4.2, respectively, for isolates H,

G, C, B and E (Table 6.1). Among them, C, G and H induced a significantly higher permeability increase compared with isolates B and E (p<0.01) (Fig.

6.2). The permeability change caused by isolate H was the most prominent and was more than three times higher than that by isolate E (Table 6.1).

144

Altogether, it suggested intra- and inter-subtype variation in barrier disruptive activity of different Blastocystis isolates.

Fig. 6.2 Intra-and inter- subtype variation in Blastocystis in inducing permeability increase. Graph representing epithelial permeability of Caco-2 monolayers after co-incubation with live cells of Blastocystis ST-7 and ST-4 for 24h. All strains in ST-7 induced significant increase in flux of Dextran– FITC across epithelial monolayer of Caco-2 cells compared to control monolayers (p<0.01). No significant change in permeability was observed in Blastocystis ST-4 -infected Caco-2.Within ST-7, an intra-subtype variation in the capability of inducing permeability change was observed. Isolates C, G and H induced much higher permeability increase compared with isolates B and E within the same subtype (p<0.01). **, p<0.01 vs. ST-7 (B, E)-infected cells; ##, p<0.01 vs. ST-4-infected cells; ‡, p<0.01 vs. non-infected cells. Each value represents mean of twelve samples, taken from three independent experiments, four samples from each. Error bars represent the standard errors.

145

Table 6.1 Comparisons of effects of luminal pathogens/ toxins on intestinal permeability increase

Permeability Increase (fold Pathogen/ Toxin Cell Line Strain Infection Infection change Reference Dose time normalized to control) ST-7 (H) 16 ST-7 (G) 11.8 ST-7 (C) 14.5 ST-7 (B) 5.5 Current Blastocystis Caco-2 108 /ml 24h ST-7 (E) 4.2 study ST-4 (WR- a 1) NS ST-4 (S-1) NSa Clostridium Clostridium Chen et T84 difficile 10 nM 5h 3 difficile Toxin A al., (2002) 2×105 2.14 Cryptosporidium GCH1, /ml Griffith et Caco-2 48h parvum UCP 2×106 al., (1994) 18.6 /ml

Human Entamoeba HM1:IMSS Stenson et intestinal 107 /ml 24h >100 histolytica strain al., (2001) xenografts

2×107 24h >4 Chin et al., SCBN NF strain /ml 48h 9 (2002) Giardia lamblia Buret et SCBN S2 isolate 107 /ml 24h >40 al., (2002) Scott et al., SCBN S2 isolate 106 /ml 2h 5.6 (2002)

146

6.3.2 Blastocystis exhibits intra-and inter-subtype variability in ability to disrupt epithelial tight junction proteins

As in Chapter 4, we proceeded to conduct an analysis of the changes in Caco-

2 tight junction proteins Occludin and ZO-1 after exposure to all the seven strains of Blastocystis. Consistent with the permeability results, the profile of tight junction degradation shown by Western blot analysis differed from strain to strain and correlated with the degree of permeability increase caused by each strain (Fig. 6.3). ST-4 treatments did not alter the two tight junction proteins significantly which might explain the insignificant permeability increase seen above (Fig. 6.2).

Expression of occludin, was significantly decreased compared with the control and ST-4 treatments by exposure to ST-7 isolates C, G, and H (p<0.01) (Fig.

6.3). But no changes in occludin were observed in ST-7 isolates B- and E- treated cells, which might contribute to the less pronounced permeability increase by B and E compared with C, G and H. The other tight junction protein ZO-1, a multi-domain scaffold protein localized at the tight junction, however, showed a different pattern of disruption from that of occludin. All

ST-7 strains induced a significant decrease in ZO-1 band intensity compared with the negative control, which might explain the prominent permeability increase in B-, E-treated cells compared with control monolayers; nevertheless, the degree of ZO-1 degradation by B and E was again significantly less than C,

G and H (p<0.01) (Fig. 6.3), suggesting a more potent ability in disrupting

ZO-1 by C, G and H among ST-7 isolates.

147

In addition, immunofluorescence was performed to observe the distribution and integrity of the tight junction proteins. Indirect immunolabelling of occludin in control monolayers showed a bright and continuous band lining cell-to-cell contact (Fig. 6.3C). The staining pattern in cells treated by ST-4

(WR-1 and S-1) and ST-7 (B and E) isolates did not differ from that of the controls. In contrast, infection of Caco-2 cell monolayers with ST-7 (C and G) led to focal disruptions and punctate concentration along pericellular junctions, while there was almost a complete loss of Occludin in H-treated cells. As expected, ZO-1 proteins in cells after ST-4 infection appeared to be similar to those of control monolayers, where they showed a typical, continuous pericellular organization. Treatments with ST-7 (B and E), however, resulted in both a reduction in ZO-1 intensity and reorganization of

ZO-1 (Fig. 6.3D), as reported previously for ST-7 (B)-treated Caco-2 cells

(Mirza et al., 2012b)，whereas exposure to ST-7 (C, G and H) isolates led to a complete degradation of ZO-1 at the apical junctions (Fig. 6.3D).

148

Fig. 6.3 Differential effects on tight junction protein degradation in Caco- 2 cell monolayers by different strains of Blastocystis. (A) Representative Western blot analysis of Occludin and ZO-1 level in Caco-2 epithelium. Monolayers were harvested after infection with Blastocystis ST-4 and ST-7 isolates; normal culture media was used as a negative control. No obvious change in Occludin and ZO-1 could be noticed in ST-4-treated samples. Within ST-7, a more prominent decrease or loss of Occludin band could be seen in cells treated by C, G and H strains, while the changes in B, E-treated

149 cells were not obvious; for ZO-1 tight junction protein, loss of band was observed in Caco-2 cells after coincubation with ST-7 (C, G and H), whereras a decrease in band intensity was also noticed in B, E-treated cells. (B) Quantification of tight junction levels through densitometry analysis of Western blot radiographs. Densitometric values of Occludin and ZO-1 signals were quantified and expressed as the ratio to α-Tubulin. ST-4-treated samples did not differ significantly from control in Occludin and ZO-1 level; ST-7 (C, G and H) induced significant Occludin and ZO-1 degradation compared with ST-4 and ST-7(B, E) isolates (p<0.01). ST-7 (B, E), however, were able to induce significant ZO-1 degradation compared with control and ST-4-treated samples, but not for Occludin. **, p<0.01 vs. ST-7 (B, E)-infected cells; #, p<0.05 vs. ST-4-infected cells; ##, p<0.01 vs. ST-4-infected cells; ‡, p<0.01 vs. non-infected cells. Results were from three independent experiments. Error bars represent the standard errors. (C) Representative confocal micrographs illustrating Occludin integrity in Caco-2 monolayers. Caco-2 cells are labelled with DAPI (Blue) and an antibody targeting tight junction protein Occludin and Cy3® goat anti-mouse IgG (Red). Corresponding with Western blot results, in ST-4 and ST-7 (B, E)-treated samples, no obvious change in Occludin was observed. Infection with ST-7 (C, G, H), however, induced focal disruptions and degradation in Occludin, as shown by the decreased Occludin staining intensity in Caco-2 cell line. Scale bar= 10 μm. (D) Representative confocal micrographs illustrating ZO-1 integrity in Caco-2 monolayers. Caco- 2 cells are labelled with DAPI (Blue) and an antibody targeting tight junction protein ZO-1 and Cy3® goat anti-mouse IgG (Red). Compared with the negative control and ST-4-treated samples, where ZO-1 appeared as continuous with sharp pericellular staining patterns, all ST-7 isolates resulted in focal disruption as well as degradation of ZO-1 in Caco-2 cells. In ST-7 (C, G, H)-treated samples, ZO-1 staining almost became invisible, indicating more degradation of the protein, which corresponded with the Western blot results. Notably, treatments with ST-7 (B, E) resulted in both focal disruptions (green arrows) as well as reorganizations of ZO-1 (yellow arrows). Scale bar= 10 μm.

150

6.3.3 Blastocystis isolates exhibit strain-to-strain variation in attachment to human enterocytes

In many parasites and other pathogens, adhesion is directly associated with the virulence properties of the strains (Brooks et al., 2013, Rojas et al., 2004). The study therefore tested whether there was a difference in the adhesion properties of different isolates and whether it correlated with the ability of the parasites to induce a permeability increase in intestinal monolayers. To detect and quantify adhesion of different Blastocystis isolates to host cells, an adhesion assay was developed using Caco-2 cell line. Both of the ST-4 strains

WR-1 and S-1, showed negligible adhesion (Fig. 6.4). Within ST-7, isolates B and E adhered to host cells in a significantly lower number than C, G and H

(p<0.05) (Fig. 6.5). From the merged image, it was clear that C, G and H parasites attached in large numbers (Fig. 6.4). Isolate H, which induced highest epithelial permeability increase, was also the most adhesive to host cells (Fig. 6.5). The results suggested an association between the level of attachment and the ability to induce permeability increase. An analysis of these two phenotypes showed a significant positive correlation, suggesting that adherence might play a role in virulence (R2=0.8506, p<0.01) (Fig. 6.5). Even though isolates S-1 and E exhibited similar level of attachment (Fig. 6.4), the former failed to induce a permeability increase in Caco-2 monolayers (Figure

6.2), suggesting that additional factors of isolate E might contribute to the permeability increase. Furthermore, the study also provided visual evidence of association between Blastocystis-induced host-cell pathology with parasite- enterocyte contact (Fig. 6.6). Blastocystis ST-7 (H) preferentially adhered at the apical junction region (Fig. 6.6A), as seen previously in Chapter 2 for

151 isolate ST-7 (H). The percentage of parasites preferentially attached to the intercellular junctions was 80% ± 3.8% of all attached cells (values are means

± SE). Parasites adhered intimately with host cells at the cell-cell junction site and induced increase in actin polymerization (Fig. 6.6B and C). Besides, loss of cellular symmetry could be noticed in the host cells compared to negative control which displayed a properly organized epithelium (Fig. 6.6B), indicating adhesion-mediated intestinal epithelial injury.

152

Fig. 6.4 Blastocystis exhibits intra- and inter-subtype variation in attachment to Caco-2 cells. (A) Representative confocal micrographs illustrating intra- and inter-subtype variation in Blastocystis attachment to Caco-2 cells. Caco-2 monolayers were grown to confluency on glass converslips and were then co-incubated with the same number of parasites of different strains of Blastocystis pre-stained with CFSE (green). Normal culture media was used as a negative control. After co-incubation, the non-attached parasites were washed away. The Caco-2 monolayers were then stained with DAPI and then were viewed using confocal microscope (Olympus FLUOVIEW-FV100; Olympus, Japan). More green in the field represents more parasites attached to the monolayer. Both ST-4 strains adhere with a negligible number. An intra-subtype variation in the number of attachment within ST-7 is obvious. Isolates C, G, H appeared to attach at a much higher level than B and E to Caco-2 cells. Scale bar= 100 μm. (B) Graph representing number of Blastocystis parasites attached to host cells. ST-7strains C, G and H exhibited a significantly higher number of attached parasites than ST-4 strains

153 and ST-7 isolates B and E. *, p<0.05 vs. ST-7 (B, E); **, p<0.01 vs. ST-7 (B, E); ##, p<0.01 vs. ST-4. Each value represents a mean of six readings derived from 3 independent experiments. Error bar represents standard error.

Fig. 6.5 Relationship between attachment and permeability increase by Blastocystis ST-4 and ST-7 parasites. The data points indicate individual strains. x and y error bars indicate the standard error for the respective measurements (n=3). The R2 for the trend line shown is 0.8506, and the p value is 0.0031. There is a positive correlation between the level of attachment and permeability increase (R=0.9223).

154

Fig. 6.6 Blastocystis ST-7 (H) attaches to an intestinal monolayer. Caco-2 monolayers were grown to confluency on glass coverslips and were then coincubated with Blastocystis ST-7 isolate H. Normal culture media was used as a negative control. (A) Representative confocal micrographs illustrating attachment of Blastocystis ST-7 (H) to Caco-2 monolayers. Blastocysis ST-7 was labelled with legumain antibody-mAb1D5and secondary Alexa Fluro® 594 goat anti-mouse IgM (red). Phalloidin-FITC (green) was used to label F- actin of Caco-2 cells and DAPI (blue) for nuclei. Compared with the negative

155 control, attachment of Blastocystis ST-7 to Caco-2 apical side could be seen clearly. (B) Expanded section views of epithelium of control and ST-7 (H)- treated-monolayers. Parasites could be seen intimately adhering to the epithelium and it could also be noted that the parasites adhered preferentially to the cell-cell junction site (yellow arrows) and induced an increase in actin polymerization. Compared with the negative control which displayed a properly organized epithelium, there was loss of cellular symmetry in the cells treated by ST-7 (H) (black arrowheads). Scale bar =20 μm. (C) Quantification of F-actin staining in Blastocystis-infected Caco-2 monolayers. Each cell slice (1–25) corresponds to series of images from Z-stack sections taken at 1μm thickness through the cell monolayer. X-axis illustrates cell layers from basolateral to apical. Y-axis illustrates the number of pixels present over the entire area of image. Monolayers treated with ST-7 (H) resulted in marked increase in actin intensity compared with the negative control (p<0.01).

156

6.3.4 Blastocystis exhibits extensive intra- and inter-subtype variation in cysteine protease activity

As cysteine proteases (CPs) have been suspected to be virulence factors in

Blastocystis and have been shown to play a role in inducing barrier compromise (Mirza et al., 2012b), hence the study compared the cysteine protease activity of all the seven Blastocystis isolates. CP activities in respective parasite lysates were determined using azocasein assay. All strains tested showed significantly higher protease activity than the PBS control.

Variation in protease activity was observed among different Blastocystis isolates (Fig. 6.7). ST-4 isolates WR-1 and S-1 showed protease activity of

13.86±2.5 and 13.25±4.2 azocasein units. Except isolate G, most of the ST-7 strains showed significantly higher protease activity than ST-4 isolates

(p<0.01), with 30.41±0.94, 34.99±3.04, 26.03±2.10, 26.83±4.17 azocasein units for B, E, C, and H respectively. Lysate of G isolate showed activity of

16.77±3.98 units, which was not significantly different from ST-4 strains.

There was no correlation between the total CP activities and the virulence of each strain, suggesting that total cellular cysteine protease is not responsible for the variation in pathogenicity within ST-7 isolates in inducing intestinal barrier dysfunction.

157

Fig. 6.7 Intra- and inter-subtype variations in protease activity in Blastocystis. Protease activity of Blastocystis ST-4 and ST-7 isolates was determined by azocasein assays. Except isolate G, most of ST-7 strains exhibited significantly higher protease activities than the two ST-4 isolates (p<0.01). PBS showed activity that was significantly lower when compared to protease activities of all the isolates (p<0.01). Note that cysteine protease inhibitor iodoacetamide (IA) abolished protease activity of all the isolates, which is comparable to PBS. ##, p<0.01 vs. ST-4; ‡, p<0.01 vs. PBS control. Each value represents mean of six samples, taken from three independent experiments. Error bars represent the standard errors.

158

6.3.5 Blastocystis ST-7 isolates exhibit extensive variation in resistance to

Mz, and the level of Mz resistance is inversely correlated with an ability to attach and induce a permeability increase

It was previously reported that isolates B and E belonging to ST-7 were Mz resistant (Mirza et al., 2011a). As both of them also exhibited impaired ability to adhere to host cells compared with other ST-7 isolates, the study tested whether there was an association between attachment and Mz resistance in

ST-7 as seen in Giardia (Tejman-Yarden et al., 2011). IC-50s of C, G, H isolates (2.98±0.97, 3.63±0.96, 1.05±0.32 µg/ml, respectively; Table 6.2) were significantly lower than the previously reported IC50s of isolates B, E

(p<0.01; Fig. 6.8), indicating an intra-subtype variation in Mz resistance in

Blastocystis ST-7 parasites. Correlation analysis showed a significant negative correlation between the level of drug resistance of an isolate and its ability to adhere to host cells (R2=0.8908, p=0.0158) (Fig. 6.9A and Table 6.3). The results suggested Mz resistance might entail an attachment defect in

Blastocystis ST-7 strains.

As the result above showed that the level of attachment was correlated with the ability to induce permeability increase (Fig. 6.5), the impaired attachment in Mzr strains might indicate a less potent ability in inducing permeability increase. Indeed, the correlation analysis between drug resistance and permeability increase showed a significant negative correlation (R2=0.9644, p<0.01) (Fig. 6.9B and Table 6.3). The results suggested Mz resistance might also compromise virulence of Blastocystis in causing inducing barrier disruption.

159

Fig. 6.8 Blastocystis exhibits intra-subtype variation in susceptibility and resistance to Mz. Graph representing IC50s of Mz against Blastocystis ST-4 and ST-7 isolates tested in the study using the resazurin assay. Y axis was presented on a logarithmic scale in base 5. The IC50s of Mz against ST-7 isolates C, G and H were found to be significantly lower than those of isolates B, E within the same subtype (p<0.01). **, p<0.01 vs. ST-7 (B, E). Each point represents a mean of nine readings derived from three independent experiments, triplicate each. Error bars represent the standard errors.

Table 6.2 IC50s of Mz and nitric oxide donors against Blastocystis isolates (µg/ml)

ST-4 ST-7

Mz-susceptible Mz-resistant

WR-1 S-1 H G C B E

Mz 5.5±2.89a 0.75±0.04a 1.05±0.32 3.63±0.96 2.98±0.97 32.5±3.4a 115.6±21.6

GSNO 83.3±2.5b 98.29±0.73 121.73±8.2 75.24±3.99 61.79±4.03 30.63±1.67b 30.92±1.74

NaNO2 233±9.2 240.6±10.4 217.45±16 162.15±34 199.75±5.8 86.47±12.8 106.35±4.8 a, See Reference (Mirza et al., 2011a) b, See Reference (Mirza et al., 2011b)

160

Fig. 6.9 Correlation analyses between Mz resistance and (A) attachment, as well as (B) permeability increase in Blastocystis ST-7 parasites. (A) Relationship between Mz resistance and attachment for Blastocystis ST-7 parasites. The data points indicate individual strains. Error bars indicate the standard error for the respective measurements (n=3). There was a negative correlation between the level of resistance and attachment (p<0.05, R2=0.8908). (B) Relationship between Mz resistance of Blastocystis ST-7 parasites and their ability in inducing permeability increase. The data points indicate individual strains. Error bars indicate standard errors for the respective measurements (n=3). There was a negative correlation between the level of resistance and permeability increase (p< 0.01, R2=0.9644).

Table 6.3 Correlation of Mz resistance with attachment, permeability increase and nitric oxide IC-50 in Blastocystis ST-7

Pearson R2 P value Correlation (R)

Attachment -0.9438 0.8908 0.0158a Permeability Increase -0.9821 0.9644 0.0029a Nitric Oxide IC-50 -0.8929 0.7972 0.0414a

a. Correlation is significant at the 0.05 level (2-tailed).

161

6.3.6 Mz resistance in Blastocystis does not lead to decrease in cell proliferation in Blastocystis ST-7

Most drug resistance mechanisms are associated with a fitness cost that is typically observed as a reduced growth rate (Andersson et al., 2010,

Andersson, 2006). To investigate the functional consequences of Mz resistance in Blastocystis, the study examined the proliferative potential of Mzr and Mzs strains from growth curves assayed over 96 h. The growth rates of

Mzr isolates were not always lower than Mzs isolates (Fig. 6.10). Isolate E, being highly resistant to Mz, consistently grew more rapidly than Mzs isolates

C and G, suggesting that drug resistance in Blastocystis ST-7 might not necessarily result in slower growth.

Fig. 6.10 Mz resistance in Blastocystis does not lead to decreased growth in Blastocystis ST-7. Graph representing growth curve of Blastocystis ST-7 isolates tested in the study over a period of 96h. The growth rates of Mzr isolates B, E were not significantly different from that of Mzs isolates C, G and H within the same subtype. Each point represents a mean of 6 readings derived from two independent experiments, triplicate each. The error bar represents standard errors.

162

6.3.7 Blastocystis ST-7 isolates exhibit extensive variations in resistance to nitrosative stress

Aside from adhesion, microorganisms also need to evolve means to counter the host immune system for effective colonization in the intestinal lumen (Tan and Mirza, 2012). Nitric oxide (NO) is a potent innate immune response against a range of pathogens and nitrosative stress prevents colonization of host tissues by parasites and bacteria susceptible to NO (Eckmann et al., 2000,

Lewis et al., 2010). Previous studies have already suggested NO induces cell death in Blastocystis (Eida et al., 2008, Mirza et al., 2011b). Thus, the ability of Blastocystis to tolerate nitrosative stress is likely to be important for its survival in the gut lumen. It was previously observed that a Mzr strain of

Blastocystis exhibited lower tolerance to nitric oxide toxicity compared with a

Mzs strain (Mirza et al., 2011b). In this study, the IC50s of NO donors against the Mzr and Mzs strains were tested to examine their ability to tolerate nitrosative stress. Interestingly, Mzs isolates C, G and H of Blastocystis ST-7, all exhibited significantly higher IC-50s of GSNO (61.79±4.03, 75.24±3.99,

121.73±8.17µg/ml, respectively; p<0.01; Fig. 6.11A and Table 6.2) than Mzr

B and E isolates (30.63±1.67, 30.92±1.73 µg/ml, respectively). Therefore, Mzr strains B and E were more susceptible to GSNO. An analysis again showed a significant negative correlation between the level of Mz resistance of an isolate and its ability to tolerate GSNO toxicity (R2=0.7972, p<0.05) (Fig.

6.11B and Table 6.3). IC50s with another NO donor NaNO2 for all these strains exhibited similar trends to those with GSNO (Table 6.2). Taken together, the results suggested that, although Mz resistance helps Blastocystis

163 survive stress from chemotherapy, it makes Mzr strains less capable of coping with nitrosative stress.

Concomitantly, to determine the morphological changes resulting from nitrosative stress, isolates H and B were stained with propidium iodide and annexin V-FITC after exposure to NO. Necrotic cells incorporated both PI and annexin V-FITC stain and cells undergoing programmed cell death bind annexin V-FITC alone. After treatment with 50 μg /ml concentration of

GSNO, features of both necrosis and apoptosis were observed in the Mzr isolate B, whereas the Mzs isolate H were less affected (Fig. 6.11C).

Compared with cultures of C, G and H, Mzr isolate B and E exhibited a significantly higher percentage of cells undergoing cell death (37.54% and

32.93% 7.21% for B and E respectively; 12.38%, 10.2%, 7.21% for isolates C,

G and H respectively; p<0.01) (Fig. 6.11D), which also indicates that the parasite cell death induced by NO was predominantly by necrosis with a minority of cells dying by apoptosis.

164

Fig. 6.11 Mz resistant isolates in Blastocystis ST-7 exhibit susceptibility to nitric oxide. (A) Graph representing IC50s of NO donor GSNO against Blastocystis ST-7 strains tested. ST-7 isolates C, G and H showed significantly higher tolerance to GSNO (p<0.01) than Mzr strains B and E. **, p<0.01 vs. ST-7 (B, E). Each point represents a mean of nine readings derived from three independent experiments, triplicate each. The error bar represents standard errors. (B) Relationship between Mz resistance and GSNO resistance in Blastocystis ST-7 parasites. The data points indicate individual strains. Error bars indicate the standard error for the respective measurements (n=3). There

165 was a negative correlation between level of Mz resistance and tolerance of nitric oxide toxicity (p<0.05, R2=0.7972). (C) Confocal micrographs illustrating cell death features of Blastocystis H and B isolates under nitrosative stress. Blastocystis were cultured under anearobic culture conditions, in the presence of 50μg/ml of GSNO. The cells were then stained with Annexin-V-FITC and PI to identify cells undergoing cell death. Features of necrosis (yellow arrows) and programmed cell death (red arrows) were observed. Blastocystis cells undergoing necrosis incorporated both PI and annexin V-FITC stain, cells undergoing programmed cell death, on the other hand bind annexin V-FITC alone. A significantly higher number of dying cells were observed in cultures of ST-7 (B) compared with ST-7 (H) cultures. Scale bar= 10 μm. The micrographs are representative of 15 pictures taken in three separate experiments (five from each). (D) Graph representing percentage of cell death in Blastocystis ST-7 strains after treatment with NO donor GSNO. Compared with Mzs ST-7 isolates (C, G, H), percentages of dead cells in ST-7 (B, E) are significantly higher (p<0.01).**, p<0.01 vs. ST-7 (B, E). Each value represents a mean of nine readings derived from three independent experiments, triplicate each. Error bars represent the standard errors.

166

6.4 DISCUSSION

Although it is one of the commonest eukaryotic organisms present in the alimentary tract of human and nonhuman hosts worldwide (Stensvold et al.,

2009d), the debate about the pathogenicity of Blastocystis continues (Garcia,

1984, Waghorn et al., 1991, Barry, 2004, Tan et al., 2010, Clark et al., 2013) and arguments for and against its pathogenicity both abound (Stensvold et al.,

2009d). The enigmatic role of Blastocystis has mainly been due to a lack of basic knowledge about its biology and convincing evidence of its pathogenicity (Stensvold et al., 2009d). A major obstacle or challenge is the potential for intra- and inter-subtype variation in Blastocystis pathogenicity

(Scanlan, 2012, Clark et al., 2013). Although previous research attempted to validate this argument by using animal models (Iguchi et al., 2007), the number of strains used were limited and no consensus on the appropriate infection model was reached for Blastocystis, thus the infection outcomes might be reflective of infectivity rather than pathogenicity. The current study, using a well-established in-vitro system mimicking the host-pathogen interplay, provides, for the first time, a comprehensive analysis of different strains from two clinically relevant subtypes, and gives evidence that

Blastocystis exhibited not only inter- but also intra-subtype variability in causing barrier dysfunction. Disturbances in epithelial barrier function are commonly associated with intestinal inflammatory disorders and have been reported in symptomatic cases of Blastocystis infection (Turner, 2009, Tan,

2008, Dagci et al., 2002, Poirier et al., 2012). Thus the findings from the current study might help explain the conflicting data concerning the

167 inconsistency in reports of Blastocystis-induced intestinal inflammatory disorders.

Paracellular permeability was regulated by tight junctions, which seal the space between neighboring cells, generating an impermeable barrier between the epithelium and the extracellular environment, protecting deeper tissues from external aggressions including microbial infections (Bonazzi et al.,

2011). However, this first line of defense against infection has become one of the most exploited gates for pathogens to access and colonize the host organism (Bonazzi et al., 2011). Enteric pathogens have developed a broad and complex range of mechanisms to subvert host tight junction (O'Hara et al.,

2008). As discussed in Chapter 4, key mechanisms identified to date include direct rearrangement or degradation of specific tight junction proteins, reorganization of the cell cytoskeleton, and activation of host cell signaling events (Sears, 2000, Berkes et al., 2003, Fasano et al., 2004). The study revealed disruption of tight junctions (TJs) which mainly resulted from degradation in at least two TJ proteins: ZO-1 protein, which is linked to the cytoskeleton and plays a pivotal role in the TJ architecture (Peiffer et al.,

2001); and occludin, which is important in maintaining the integrity and barrier function (Furuse et al., 1993, McCarthy et al., 1996). Previously, it was reported that Blastocystis phophorylates MLC and leads to cytoskeletal and

ZO-1 rearrangement (Mirza et al., 2012b). Indeed, the study also observed myosin light chain phosphorylation and TJ rearrangement in B and E-treated

Caco-2 cells besides degradation of ZO-1 protein (Fig. 6.12). However, for

Blastocystis isolates C, G, and H, although MLC was phosphorylated, degradation of TJs appeared to be the main mechanism, indicating the

168 strategies utilized by Blastocystis to induce permeability increase are likely to be diverse and different isolates may have developed different major mechanism to compromise barrier function. In a recent study, the observation that ROCK inhibition did not completely rescue the increase in the epithelial permeability induced by the parasite (Mirza et al., 2012b) also suggested that

Blastocystis uitilizes more than one mechanism to breach the epithelial barrier.

Fig. 6.12 Differential effects on myosin light chain phosphorylation in Caco-2 cell monolayers by different strains of Blastocystis. (A) Representative Western blots exhibiting level of MLCp after incubation with Blastocystis ST-4 and ST-7 parasites. Total MLC was used as a loading control. An obvious increase in MLCp could be noticed in ST-7-treated samples. (B) Graph represents MLCp/MLC ratio normalized to the negative control in Caco-2 cells calculated through densitometry analysis of Western blot radiographs. A significant increase in was observed in all ST-7-treated Caco-2 epithelium compared with the negative control and ST-4-treated samples (p<0.01). ‡, p<0.01 vs. control. Error bars represent the standard error of the mean of 3 values obtained from three independent Western blots.

169

Epithelial attachment is an important factor determining the persistence and virulence of luminal pathogens (Elmendorf et al., 2003, Tavares et al., 2005).

The study for the first time showed Blastocystis isolates exhibits strain-to- strain variation in attachment to human enterocytes. Differential production of the adhesins or other factors could modulate the parasite–cell surface interaction (Gilbert et al. 2000). Results in Chapter 3 suggested a galactose- binding protein as an important element involved in cell surface recognition and attachment to host cells. It would be interesting to investigate whether similar proteins could be found on other isolates and compare the expression.

More importantly, the results showed for the first time pathogenicity of

Blastocystis was correlated with their ability to attach to epithelial cells. How adhesion leads to intestinal barrier dysfunction needs to be investigated further. Giardia intestinalis was shown to produce harmful substances as a result of the host-pathogen contact (Maia-Brigagao et al., 2012). Several enzymes from the secreted products have been identified in G. intestinalis, which are suggested to facilitate effective parasite adhesion and colonization of the human small intestine (Ringqvist et al., 2011, Rodriguez-Fuentes et al.,

2006). Whether the adhesion of Blastocystis with host cells also enables particular enzymes to be released and to participate in pathogenesis would be interesting to investigate. The Blastocystis genome has been available and in silico analysis of the ST-7 secretome predicted 75 putative secreted proteins, some of which may have a direct connection with pathogenicity (Denoeud et al., 2011). Recently, two cysteine proteases (legumain and cathepsin B) have been characterized in the ST-7 culture supernatant. The enzymes showed proteolytic activities by gelatin zymograms (Wawrzyniak et al., 2012).

170

However, whether these secreted proteins act on intestinal cells and disturb gut function has not been studied. It would be interesting to investigate whether adhesion of Blastocystis to the epithelium might trigger the parasite to actively produce virulence factors that possibly activate signaling cascade leading to tight junction disruption. Preferential adherence of cytopathic Blastocystis strains to intercellular junctions leading to tight junction degradation observed in this chapter as well as in Chapter 2 also supports this idea.

Parasites such as Giardia intestinalis and Entamoeba histolytica have distinct virulent and non-virulent strains that may be attributable to qualitative and quantitative variation in their cysteine proteases activity (Reed et al., 1989,

Guimaraes et al., 2003, Puthia et al., 2005) and are also implicated in the activation of NF-kB in colonic epithelium, leading to proinflammatory cytokine IL-8 upregulation (Puthia et al., 2008). It was shown that cysteine proteases could cause human epithelial barrier compromise (Mirza et al.,

2012b). Indeed, isolate E, being the least adhesive in ST-7, has the highest cysteine protease activity, which might explain the permeability increase observed in E-infected epithelium; while isolate S-1, with equivalent level of attachment to E, and at the same time the lowest CP activity, failed to induce any increase in intestinal permeability. Interestingly, the highly adhesive G isolate, though with similar CP activity to ST-4 isolates, could induce prominent epithelial barrier disruption, highlighting the important role of adhesion in mediating intestinal pathology. Taken together, both adhesion and

CP activity might contribute to Blastocystis–induced barrier dysfunction. ST-4 isolates, low in both adhesion and CP activity, appeared to be avirulent for human intestinal monolayer; whereas all ST-7 isolates studied are capable of

171 infection because of being either adhesive or exhibiting high CP activity, or both. Moreover, adhesion is a major determinant in the pathogenicity of a strain. The results provided evidence that the pathogenesis of blastocystosis is complex and parasite adhesion and cysteine proteinases might be important virulence factors. Specific virulence factors and their mechanisms remain to be explored in depth.

Besides the observation that Blastocystis ST-7 exhibited intra-subtype variation in epithelial attachment, the association between epithelial attachment and Mz susceptibility was also observed. Mzr isolates in ST-7 exhibited a defect in epithelial attachment. How Mz resistance would affect

Blastocystis adhesion to host cells remains to be addressed. Drug resistance has been shown to modulate adhesion in some bacterial and parasitic agents such as Escherichia coli, Giardia, and Stenotrophomonas (Deneke et al.,

1985, Klein et al., 1985, Carlone et al., 1987, Li Pira et al., 1987, Gismondo et al., 1994, Nilsson et al., 2003, Pompilio et al., 2010, Tejman-Yarden et al.,

2011). Tejman-Yarden et al. showed that in Giardia lamblia, epithelial attachment was linked to glucose metabolism (Tejman-Yarden et al., 2011).

Glucose promoted parasite attachment and Mzr lines were shown to consume less glucose than their parental Mzs lines, providing a possible explanation for the attachment defect observed in Mzr lines. In Blastocystis, the glucose consumption of all seven strains used in the study was also measured.

However, their respective rate did not correlate with the level of Mz resistance

(Fig. 6. 13), suggesting that glucose metabolism may play a different role in

Blastocystis from that in Giardia. In Chapter 3, a galactose-binding protein was suggested to play a role in Blastocystis adhesion. Whether the synthesis or

172 expression of this adhesion protein was affected by Mz selection would be worth exploring.

Fig. 6.13 Differential consumption rate of glucose by Blastocystis ST-4 and ST-7 strains. Different strains of Blastocystis from ST-4 and ST-7 were cultured in medium containing 5 mM glucose and incubated under anaerobic conditions, and levels of glucose remaining in the medium were measured after a 24-h incubation period. Glucose consumption was expressed as the mean glucose consumption for 106 cells over the period (mmol/day). No correlation was found between the rate of glucose consumption and the level of attachment or Mz resistance.

Mzr strains not only exhibited a defect in attachment, they also showed impaired NO tolerance in Blastocystis ST-7. NO secretion is an important aspect in human innate immunity and limits microbial persistence in the gut by inducing cell death and preventing their growth and encystations (Eckmann et al., 2000). Therefore, the ability to tolerate nitrosative stress would be an important aspect for parasite survival. The results suggested Mz resistance might limit its survivability to host innate immune response such as coping with nitrosative stress. How Mz resistance affects the susceptibility to NO needs to be investigated further. Development of Mz resistance in

173 microorganisms is a multifactorial process and the mechanisms are diverse

(Leitsch et al., 2011). In Giardia and Trichomonas, Mz resistance usually develops due to impairment of Mz activation enzymes in the parasite. For example, resistance to Mz has been associated with down-regulation or decreased activity of pyruvate: ferredoxin oxidoreductase (PFOR), which activates the 5-nitroimidazole (5-NI) pro-drugs to the toxic radical states inside the parasite in conjunction with the electron acceptor ferredoxin

(Tejman-Yarden et al., 2011, Leitsch et al., 2010, Upcroft et al., 2001). These

Mz activation enzymes are oxygen sensitive and are also important components of parasite anaerobic respiratory system. Anaerobic respiratory enzymes also play an essential role in free radical (e.g. H2O2 or NO) scavenging and if these enzymes are impaired, as expected in metronidazole resistant parasite strains, it would lead to decreased free radical tolerance in these isolates (Ellis et al., 1993, Rasoloson et al., 2001). However, the mechanism of Mz activation has not been studied in Blastocystis, although

PFOR and other oxidoreductase enzymes are present in the organism

(Denoeud et al., 2011, Wawrzyniak et al., 2008). Further investigation of Mz activation and/or resistance mechanisms as well as NO-scavenging mechanisms in Blastocystis would be necessary to elucidate the association between NO susceptibility and Mz resistance. Recent studies of Giardia and

Trichomonas suggested that Mz was activated by the nitroreductase activity of flavin-dependent thioredoxin reductase and Mz resistance was associated with decreased NADPH oxidase activity (Leitsch et al., 2011, Leitsch et al., 2009).

A flavohemoglobin (FlavoHb) which plays a pivotal role in NO detoxification has been characterized recently in Giardia intestinalis (Mastronicola et al.,

174

2010). Interestingly, other studies have suggested flavoHb also demonstrates

NADPH oxidase activity (Rafferty et al., 2010). Furthermore, flavoHb seems to be important in the pathogenesis of disease (Sebbane et al., 2006, Roos et al., 2006). It would be intriguing to explore whether flavoHb or its homolog exists in Blastocystis and whether it plays any function in Mz resistance, NO- scavenging or pathogenesis.

In conclusion, this chapter showed the molecular and cellular basis for extensive intra- and inter- subtype variability in pathogenicity in Blastocystis

ST-4 and ST-7 in vitro. The existence of both pathogenic and apathogenic isolates might help explain the disparity in symptoms among patients with

Blastocystis infections. The modes of pathogenesis of blastocystosis are diverse and the range of clinical symptoms observed is unlikely to be attributed to a single virulence factor or pathogenic mechanism. Moreover, the study indicates that adhesion is an important biomarker for the pathogenicity of the parasite and subsequent outcome of infection/disease and that adhesion can be impaired by Mz resistance. The proposed model for Blastocystis pathogenesis based on findings from this chapter is summarized in Fig. 6.14.

Interestingly, the five ST-7 strains tested here were morphologically indistinguishable from each other, but varied significantly in many aspects of their pathobiology, including attachment and drug resistance. A deeper analysis and systematic comparison of the genome and protein expression profiles among these Blastocystis strains could aid in the identification and analysis of the key factors associated with the pathobiology of blastocystosis.

175

Fig. 6.14 Summary model illustrating pathogenesis and fitness cost in Blastocystis. Both adhesion and cysteine proteases contribute to Blastocystis– induced barrier dysfunction. Adhesion leads to degradation of TJ proteins possibly by triggering release of virulence factors. The factors may also contribute to cysteine protease activity. CP induces intestinal barrier function by rearranging TJ proteins, but might also cause degradation of TJ. Development of Mz resistance entails impairment in adhesion as well as tolerance to nitrosative stress. TJ, tight junction; MLCp, myosin light chainphosphorylated at p-ser 19; ROCK, rho-associated kinase.

176

CHAPTER 7:

APPLICATION OF THE ADHESION

MODULATOR IN BLASTOCYSTIS

INFECTION

177

7.1 INTRODUCTION

Chapters 2 and 4 demonstrated that Blastocystis isolate ST-7 (B) is capable of adhering to human intestinal epithelium and causing epithelial barrier compromise. This isolate also displays metronidazole resistance as seen in

Chapter 6. Patients infected with antibiotic resistant virulent Blastocystis strains might be difficult to manage (Mirza et al., 2012b). There is a need to develop alternative or adjunctive treatment options for such infections.

Currently too few drugs are available to treat parasitic infections, and most of these were developed over 50 years ago (Sibley et al., 2003). Some still retain efficacy, but these are frequently so toxic that the treatment itself is life- threatening (Jackson et al., 2010, Marino et al., 1990). Prudent use and appropriate management can slow the selection of resistant microbial populations, but in all cases, the development of new therapeutic strategies is necessary. Because adhesion to host cells is the key first step in causing microbial infections and Chapter 6 highlights its important role in Blastocystis pathogenesis, it should be possible to prevent such infections by blocking adhesion. The current study aims to address this issue by evaluating a treatment strategy targeting Blastocystis adhesion to host cells.

Anti-adhesion therapy is meant to reduce the contact between host tissues and pathogens, either by prevention or reversal of adhesion of the infectious agent

(Ofek et al., 2003). The adherent state is not only a key step in pathogenesis, but also advantageous for pathogen survival. After adhering to host cells, pathogens are able to initiate certain biological processes like proliferation, toxin secretion and cell invasion and assume a significantly greater resistance

178 to clearance by normal cleansing mechanisms and to killing by normal immune factors, and antibiotics (Dai et al., 2000). Pathogens may be better able to acquire nutrients, further enhancing their ability to survive and infect the host (Ofek et al., 2003). Therefore, prevention of adhesion at an early stage following the exposure of the host to pathogens would be a good strategy to reduce the infection and disease.

It was found that a large number of bacterial, viral, and protozoan pathogens bind in vitro to carbohydrate structures of glycoconjugates present on the intestinal cell surface (Table 6.1) and that this binding can be readily inhibited by suitable mono-or oligosaccharides (Dai et al., 2000), which compete with the host-cell glycoproteins or glycolipids on the carbohydrate binding domain of the pathogen adhesin lectin. Inhibition of adhesion by suitable carbohydrates could lead to the development of anti-adhesion therapy as a novel approach against pathogen infections. In Chapter 3, galactose was identified as an effective modulator for ST-7 (B) adhesion to Caco-2 cells. The current chapter would evaluate the therapeutic potential of galactose by testing its displacing effects on adhered parasites and also investigate whether application of galactose in the host-pathogen interaction system could ameliorate the pathogenic effects exerted by the parasite. Chapter 6 revealed that Blastocystis parasites of the same subtype exhibited extensive variations in the level of attachment. It would be interesting to test whether the same modulator galactose, which interfered with the adhesion of one strain ST-7 (B), could also be applied to another strain of the same subtype, to see whether similar molecules are involved in the host-pathogen recognition process across different strains. Therefore, another important goal of the current study was to

179 apply galactose in the highly adhesive strain H, identified in Chapter 6, and evaluate whether it could also work for H in displacement and rescue of tight junction degradation.

It’s described here for the first time that addition of galactose could dislodge parasites which already adhered to host cells, and that disruption of

Blastocystis adhesion by galactose treatment leads to the abolition of the

Blastocystis-induced tight junction degradation in Caco-2 cells. Importantly, the results showed galactose was also effective for the most adhesive and virulent H strain within Blastocystis ST-7.

Table 7.1 Microbial carbohydrate receptors on the intestinal microvillus membrane

a, Adapted from Reference (Dai et al., 2000)

180

7.2 MATERIALS AND METHODS

7.2.1 Parasite culture

Axenized isolates of Blastocystis B and H were maintained as previously described in Chapter 2. 1-day-old healthy Blastocystis parasites were used in all experiment.

7.2.2 Culture of Caco-2 colonic epithelial cell line

Caco-2 stock cultures were maintained as previously described in Chapter 2.

For Western blot analysis, Caco-2 monolayers were grown on standard 24- well cell culture plates (Corning) until 100% confluency.

7.2.3 Attachment assay

Attachment assay was performed as previously described in Chapter 3.

To define the possible role of surface galactose glycoproteins in attachment, assays were done in the presence of monosaccharides (D-galactose) (50 and

100 mM).

To see whether monosaccharides have a displacing effect for parasites already adhered to the host cells. Parasites were incubated with Caco-2 for 1 hour at

37°C to adhere first, then monosaccharides or PBS control were added to the coculture, then continue to incubate for another 1 hour. At the end of incubation, unbound parasites were washed as previously described.

7.2.4 Western blot

Monolayers were treated with 1.25 × 107 parasites/ well in the presence or absence of sugar at different concentrations and co-incubated for 1 h at 37°C.

181

ZO-1 Western blot analysis was then performed as previously described in

Chapter 4.

7.2.5 Statistical analysis

Results are expressed as mean ± SEM. Differences between means were compared using Student's t tests or analysis of variance where appropriate.

Experiments were carried out on at least three separate occasions.

182

7.3 RESULTS

7.3.1 Displacing effect of galactose on Blastocystis adhesion to Caco-2 cells

To examine the ability of galactose to displace already adhered Blastocystis parasites, galactose was applied to the medium of Blastocystis-epithelial cell interaction after the parasites have established adhesion with Caco-2 cells. The results showed that addition of galactose significantly interfered with the parasite-cell attachment and displaced adhered parasites to 13.9% ± 4.2% of the control (p<0.01) (Fig. 7.1), while no obvious change was observed with addition of glucose (compared to PBS control).

183

Fig. 7.1 Displacement of Blastocystis from Caco-2 cells by galactose. Blastocystis ST-7 (B) were incubated with Caco-2 for 1 hour at 37°C to allow parasites to adhere first, then galacose (100mM), glucose (100mM) or PBS control were added to the coculture medium, then continue to incubate for another hour at 37°C. The number of parasites attached with addition of PBS control was assigned as 100%. The numbers of attached parasites with saccharides addition were normalized to control. *, p<0.01 versus control. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors. Glu, glucoe; Gal, Galactose.

184

7.3.2 Effect of the cytadherence inhibitor galactose on prevention of

Blastocystis–induced tight junction degradation in Caco-2 cell monolayers

In Chapter 6, it was proposed adhesive capacity of a Blastocystis strain might be an important determining factor of its pathogenicity. Inhibition of parasite adhesion to host cells, therefore, might be expected to concomitantly diminish the pathogenic effects. For these reasons, it would be good to test the effect of galactose on Blastocystis-induced barrier dysfunction. Degradation of tight junction protein ZO-1 was prevented by addition of galactose in a dose- dependent manner (Fig. 7.2). In contrast, glucose with no inhibitory effects for

Blastocystis cytadherence was unable to ameliorate the tight junction changes, supporting the view that adhesion was an important prerequisite for

Blastocystis–induced cytopathogenicity. Galactose and glucose alone

(saccharide control) did not change the expression of ZO-1 tight junction protein in the current experiment set-up.

185

Fig. 7.2 Galactose prevented Blastocystis–induced ZO-1 tight junction degradation in a dose-dependent manner. (A) Representative Western blot analysis of ZO-1 level in Caco-2 epithelium. Caco-2 monolayers were infected with Blastocystis ST-7 (B) in the presence of saccharides at different concentrations (25, 50 and 100mM) and incubated for 1 hour. Monolayers were washed and prepared for Western blot. Normal culture media with no sugar addtion was used as a negative control. (B) Quantification of ZO-1 levels through densitometry analysis of Western blot radiographs. Densitometric values of ZO-1 signals were quantified and expressed as the ratio to β-actin*, p<0.01 versus control. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors. Glu, glucoe; Gal, Galactose.

186

7.3.3 Evaluation of galactose on the adhesion of the highly adhesive

Blastocystis ST-7 (H) isolate to Caco-2 cells: interference with adhesion, displacing effects as well as rescue of tight junction degradation

It was important to test the efficacy of galactose in other representative isolates of Blastocystis with known differences in cytadherence levels. ST-7

(H) has been been shown to be highly adhesive among ST-7 in Chapter 6. The results showed that galactose did effectively decrease H adhesion in a dose- dependent manner (67.7% ± 14.1%, 33% ± 0.3% of control at 50 and 100mM, respectively) (Fig. 7.3) (p<0.01 for both values), suggesting there might be common receptor-ligand mediating parasite-host adhesion among different isolates. Displacement effect of galactose on ST-7 (H) adhesion was also significant (reduced by 51% ± 14%) (p=0.012) (Fig. 7.4), albeit to a lesser extent than that on ST-7 (B) as indicated above (inhibited by 86.1% ± 4.2% compared to control) (Fig. 7.1). Also, the ZO-1 tight junction protein was prevented from degradation by ST-7 (H) in the presence of galactose at

100mM (Fig. 7.5).

187

Fig. 7.3 Dose-dependent inhibiton of galactose on Blastocystis ST-7 (H) adhesion to Caco-2 monolayers. Blastocystis ST-7 (H) were incubated with epithelial cells in the presence of different concentrations of galactose (50 and 100mM, respectively). A value of 100% was assigned to number of binding parasites without addition of sugars. The numbers of attached parasites with galactose addition were normalized to control. *, p<0.01 versus control.

188

Fig. 7.4 Displacement of Blastocystis ST-7 (H) from Caco-2 cells by galactose. Blastocystis ST-7 (H) were incubated with Caco-2 for 1 hour at 37°C to allow parasites to adhere first, then galacose, glucose or PBS control were added to the coculture medium, then continue to incubate for another hour. The number of parasites attached with addition of PBS control was assigned as 100%. The numbers of attached parasites with saccharides addition were normalized to control. #, p<0.05 versus control.

189

Fig. 7.5 Galactose rescued Blastocystis ST-7 (H) –induced ZO-1 tight junction degradation. (A) Representative Western blot analysis of ZO-1 level in Caco-2 epithelium. Caco-2 monolayers were infected with Blastocystis ST- 7 (H) in the presence of saccharides galactose and glucose at 100mM and incubated for 1 hour. Monolayers were washed and prepared for Western blot. Normal culture media with no sugar addtion was used as a negative control. (B) Quantification of ZO-1 levels through densitometry analysis of Western blot radiographs. Densitometric values of Occludin and ZO-1 signals were quantified and expressed as the ratio to β-actin. *, p<0.01 versus control. #, p<0.05 versus control. Values are the means ± standard errors from data of three experiments. Error bars represent standard errors. Glu, glucose; Gal, Galactose.

190

7.4 DISCUSSION

The results in Chapter 3 showed galactose was able to greatly inhibit the binding of Blastocystis ST-7 (B) to Caco-2 cell. This chapter highlights its ability to displace adhered parasites. Eradication has been challenging in

Blastocystis infections (Scanlan et al., 2013). In the presence of galactose, rather than binding to host cell surface and initiating the infection process, the parasites would be detached and then flushed from the gastrointestinal tract.

Because of its displacing effects, galactosewould probably help with this difficult problem. Cytadherence inhibition by galactose also significantly reduced disruption of tight junction protein caused by Blastocystis. This observation not only confirmed the important role of adhesion in initiating pathogenesis, but also pointed towards potential application of galactose-based antiadhesives in Blastocystis infections.

Altogether, the results provided a rationale for developing therapeutic agents that act by either prevention of attachment or induction of detachment of

Blastocystis from intestinal epithelial cells to ameliorate the harmful effects of

Blastocystis infections. There have been frequent reports of treatment failures using common anti-protozoals (Roberts et al., 2013). With the identification of antibiotic resistant strains, alternative therapies against infectious diseases are urgently needed (Mirza et al., 2011a, Mirza et al., 2012b). Chapter 6 highlights that adherence to host cells might be an important determining factor in its pathogenesis and thus might serve as a good therapeutic target.

Because the action of antiadhesives does not require the blocking of any fundamental metabolic processes of micro-organisms, emergence of resistance

191 is unlikely (Dai et al., 2000). Antiadhesives represent a viable alternative for antibiotic-resistant Blastocystis infections.

A number of anti-adhesion strategies have been suggested (Dai et al., 2000), including mucosal immunity (to induce secretory immunoglobulin A (SIgA) antiadhesin antibodies), metabolic inhibitors of the expression of adhesions

(e.g., sublethal concentration of antibiotics), dietary inhibitors (e.g., human milk), lectin drugs (consisting of lectin-like molecules that can block attachment by competitively occupying the receptor), and receptor analogues

(oligosaccharides and glycoconjugates) (Dai et al., 2000). Among these, the key role of lectin-carbohydrate interactions in microbial adhesion has been increasingly recognized (Ofek et al., 2003, Sharon et al., 2000). Inhibition of bacterial adhesion by suitable carbohydrates could lead to an anti-adhesion therapy as a novel approach against pathogen infections. Entamoeba histolytica employs a Gal/GalNAc lectin that is involved in parasite attachment and host cell toxicity (Saffer et al., 1991). Gal, GalNAc and lactose have been shown to inhibit trophozoite attachment and host cell cytotoxicity in vitro (Ravdin et al., 1981). The results suggested the possible role of galactose in reducing Blastocystis infections. However, it should be noted that mono- and disaccharides are mostly digested and absorbed in the small intestine and are usually not found in the colon (Jantscher-Krenn et al., 2012), the site of

Blastocystis colonisation. Recent studies have suggested the use of oligosaccharides as antiadhesives against parasitic and bacterial infections. For example, it has been shown that Human milk oligosaccharides (HMO) reduce

Entamoeba histolytica attachment and cytotoxicity in vitro (Jantscher-Krenn et al., 2012), with the major role played by galacto-oligosaccharides (GOS). Also,

192 studies indicate that GOS and perhaps other galactose-containing prebiotic oligosaccharides can act as antiadhesives against EPEC adherence to intestinal epithelial cells (Shoaf et al., 2006). It is suggested that certain exogenous oligosaccharides structurally resemble the receptor sites coating the intestinal epithelial cells to which intestinal pathogens recognize and adhere (Kunz et al.,

2000). Thus, these oligosaccharides may act as molecular receptor decoys or antiadhesives that can competitively inhibit adherence and prevent infection.

Moreover, HMO and GOS, resist digestion and reach the colon (Albrecht et al., 2011), and thus might be more clinically useful. Another alternative is glycomimetics (Hartmann et al., 2012), compounds that structurally mimic the inhibitory carbohydrates but that may be more readily obtainable and could be designed to resist digestion. It would be interesting to test some these compounds in Blastocystis adhesion and barrier function studies.

More interestingly, the reduction in Blastocystis parasitism of epithelial cells by galactose was observed in a highly adhesive ST-7 isolate H. The fact that the two isolates belonging to the same subtype are affected by the same inhibitor in a key step during infection suggests that the attachment of the different isolates of parasite belonging to the same subtype to the host may be mediated by the same or similar galactose-binding proteins. The reasons for the apparently great difference in attachment and virulence of the strains within the same subtype as seen in Chapter 6 are still unknown. It would be interesting to characterize the galactose-binding protein and compare among strains to see whether it is equally present or expressed in adhesive and non- adhesive Blastocystis strains. Further exploration of the functional consequences of the antigenic differences for the protein may lead to a better

193 understanding of its role in pathogenesis. More importantly, in view of the potential for pharmacologic intervention, the observation also points out the possibility of applying the same or structurally similar modulator in infections with different Blastocystis isolates.

Despite the efficacy of galactose in ST-7 (H), the inhibitory activity of galactose at the same concentration was much lower in ST-7 (H) than in ST-7

(B), suggesting there might be other unknown mechanisms important for mediating the contact between ST-7 (H) and Caco-2 cells. It indicates that pathogens may express more than one adhesin during the infectious process.

Other than just adhesin-receptor interactions, adhesion may also involve factors such as hydrophobic and other non-specific interactions under different shear-forces (Dai et al., 2000). Therefore, for anti-adhesion therapy to be effective, it will probably be necessary to use multiple agents specifically inhibiting each type of adhesin of the infecting pathogen or by interfering with other interactions involved in adhesion. Preparation of suitable cocktails of inhibitory sugars have been used for the treatment of bacterial infections

(Sharon et al., 2000). More detailed information is needed about the molecular mechanisms mediating in adhesion of Blastocystis H strain to host cells.

Results from the current study indicated antiadhesives as a novel approach for treatment of metronidazole-resistant Blastocystis infections. More detailed information about the specificity of Blastocystis surface lectins and the elucidation of the structure of their receptors will certainly be of benefit in the design of such drugs. It would be interesting to clinically investigate whether these findings translate into a potent treatment option of antibiotic-resistant intestinal infections.

194

CHAPTER 8:

GENERAL DISCUSSION

195

8.1 DISCUSSION

As one of the most frequently encountered protist in human and animal fecal samples, the controversy over the pathogenic potential of Blastocystis has never ceased since it was firstly described (Barry, 2004). With the use of a variety of classic and modern molecular methods as well as the establishment of in vitro infection models during the last decade, there have been great advancements in the area of its molecular and cellular biology (Clark et al.,

2013, Denoeud et al., 2011, Tan et al., 2010, Wu et al., 2010), as well as pathogenic mechanisms (Mirza et al., 2011b, Mirza et al., 2012b). However, many important aspects of Blastocystis pathogenesis remain unidentified.

Adhesion is a crucial step in initiating and maintaining infections, yet nothing is known about the molecules mediating this initial interaction of Blastocystis and host. The chain of events leading to parasite-induced pathology after

Blastocystis-enterocyte contact is largely unknown. The overall aim of this thesis was to study the pathobiology of Blastocystis adhesion. This study identified a strain capable of adhering to host cells and causing pathological changes. It also provided an in-depth analysis of the factors involved in adhesion. The study also characterized events in the host cells in response to parasite adhesion in barrier functional disturbances. It also provided analysis of adhesion properties across different strains from two subtypes and compared their pathogenic potential. Furthermore, the study not only selected galactose as an effective inhibitor for adhesion, but also evaluated its efficacy in attenuating the pathogenic effects produced by Blastocystis.

196

The uptake of Blastocystis may occur by faecal-oral route from contaminated drinking water or food supplies (Tan, 2008). Once ingested, the parasite then enter the lumen of the large intestine where parasite replication and colonization occur. The ability of Blastocystis to attach itself to the epithelial lining of the large intestine would be critical for its survival in an environment where peristaltic flow and shear stress are constant. To initiate an infection and induce epithelial pathology, a microorganism must be able to effectively adhere to and colonize the intestinal lumen. Failure to do such would result in parasite clearance by peristalsis, and consequently a reduction of both infection and transmission. Results in Chapter 2 for the first time showed that

Blastocystis avidly bound to human intestinal epithelium and induced direct pathological changes. Importantly, by establishing an easily manipulated adhesion assay, the study was the first to investigate the physiochemical factors involved in Blastocystis adhesion. This model served as a valuable tool in elucidating potential mechanisms involved in parasite-host cell attachment.

A comprehensive investigation of the attachment process would help us understand clinically relevant questions. Blastocystis has been remarkable in establishing chronic as well as persistent infections, either symptomatic or asymptomatic, for which there is no known eradication strategy (Scanlan et al., 2013). Understanding the basis of persistent infections will aid the development of new treatments. Many factors contribute to the ability of pathogens to establish persistent infections, including both host and pathogen factors, among which one important characteristic is adhesion of the microorganism (Rhen et al., 2003). Our results showing the high affinity in the host-parasite attachment indicates that adhesion might impede clearance of the

197 parasite, which explains why eradication has been a long-term problem in

Blastocystis treatment. It was previously shown that Blastocystis was able to suppress of the innate immune response of human enterocytes by downregulating enterocyte NO production (Mirza et al., 2011b). Altogether, these findings suggest that the parasite may have evolved several strategies to survive and reproduce in the colon, which might contribute to the recurrent infections commonly seen in the clinical setting.

Our results revealed that adhesion was temperature-dependent and increased as the temperature was slightly elevated. The dependence of Blastocystis attachment on temperature also has its clinical relevance. Blastocystis is regarded as an opportunistic pathogen associated with a number of inflammatory disorders such as IBS and was more prevalent especially when there are other infections occurring (Dogruman-Al et al., 2010, Llibre et al.,

1989, Poirier et al., 2012). It was also suggested that even if Blastocystis is primarily a commensal, it could become pathogenic under certain conditions such as immunosuprression, poor nutrition or concurrent infections (Boreham et al., 1993). It might be because under such conditions of low-grade inflammation, the local temperature of infected sites might increase and thus the local microenvironment might facilitate the attachment of Blastocystis and leading to more severe pathology. Further understanding whether attachment might have contributed to the higher prevalence of Blastocystis in inflammatory disorders would provide insights into the implication of attachment in clinical management.

Following adhesion and colonization, Blastocystis produced pathogenic effects on the host cells. Chapter 4 conducted for the first time a comprehensive

198 analysis of molecular and cellular basis for the epithelial functional disturbance, including molecular alterations in the tight junction complex and enterocyte apoptosis and Interestingly, Blastocystis–induced apoptosis was shown to contribute to the alterations in epithelial permeability. Understanding the molecular and cellular basis for Blastocystis–induced defect in barrier function will enable us to devise therapies targeting these changes. In addition,

Chapter 4 also revealed an interesting redistribution of Claudin-1 tight junction protein, and pointed towards a possible association of the parasite with host cell proteins. Future exploration needs to focus on identification of molecules involved in the interaction of the parasite and host, which will reveal new insights into the molecular pathogenesis of this emerging protozoan microorganism.

One reason for the controversial pathogenic potential of Blastocystis is the diverse clinical presentations. The duration and severity of symptoms vary from acute enteritis to chronic mild diarrhea (Nagel et al., 2012, Stensvold et al., 2009c). There is no consensus on the possible reasons for variation in intestinal symptoms, but strain-to-strain variations in pathogenicity have been implicated. Similar variations have been observed in Giardia, and it was proposed that different isolates of G. duodenalis were shown to differ in their abilities to trigger pathological changes in vitro (Solaymani-Mohammadi et al.,

2011). Chapter 6 revealed that there are wide intra- and inter-subtype variations in the ability to induce intestinal permeability, which is largely related to their attachment capability, suggesting that the diverse symptoms seen in the clinical setting might be ascribed to the attachment features of various Blastocystis strains. Therefore, attachment could be a biomarker for

199 the virulence of a particular strain and is an important virulence factor in establishing infection as well as determining the diverse infection outcomes of

Blastocystis. This is the first study highlighting the cellular basis for strain-to- strain variations in pathogenicity, providing a possible explanation for the diverse clinical outcomes of Blastocystis infections. The study may also explain why some strains were more likely to be isolated in infections. It would be interesting to characterize more clinically important strains and conduct comparative study on their attachment capability.

Epithelial attachment is generally regarded as an important factor determining the persistence and virulence of luminal pathogens (Elmendorf et al., 2003,

Tavares et al., 2005). We showed for the first time pathogenicity of

Blastocystis was correlated with their ability to attach to epithelial cells. How adhesion leads to intestinal barrier dysfunction needs to be investigated further. Whether the adhesion of Blastocystis with host cells also enables certain enzymes to be released and thus participate in pathogenesis would be interesting to investigate. Recently the Blastocystis genome has been available and in silico analysis of the Blastocystis ST-7 secretome predicted 75 putative secreted proteins, some of which may have a direct connection with pathogenicity (Denoeud et al., 2011). Two cysteine proteases (legumain and cathepsin B) have recently been characterized in the ST-7 culture supernatant, which showed proteolytic activities by gelatin zymograms (Wawrzyniak et al.,

2012). However, whether these secreted proteins act on intestinal cells and disturb gut function have not been studied. It is reasonable to speculate that adhering to the epithelium might trigger Blastocystis to actively produce virulence factors which possibly activate signaling cascade leading to tight

200 junction disruption. Preferential adherence of cytopathic Blastocystis strains to intercellular junctions observed in this study also supports this idea.

In addition, in this study, using two clinically relevant Blastocystis subtypes, it’s shown that ST-7 Blastocystis isolates recovered from symptomatic human patients attached and induced epithelial barrier dysfunction in Caco-2 human epithelial cell-line; while ST-4 isolates, recovered from rats, did not induce enterocyte pathology. The results with ST-4 strains suggested that the interaction between Blastocystis and host cells is species specific. The isolates may have adapted to their hosts for survival. However, clinical reports of ST-4 infections are also very common. Regarding the contrasting results, it might probably be due to the different isolates from ST-4. It would be important to characterize more clinical strains from ST-4 and test their host cell specificity in adhesion and inducing pathological changes.

Understanding the physiochemical factors and molecular mechanisms involved in adhesion not only enabled us to know the clinical relevance of adhesion, but also lent us insights into devising anti-adhesion therapy. Recent studies have highlighted the lack of efficacy of several commonly used antimicrobial regimens in the treatment of Blastocystis infections, as well as the chronic nature of the infections, pointing out the need for further research into treatment options for Blastocystis infection. It’s necessary to design alternative or adjunctive treatment options against antibiotic resistant

Blastocystis infections. As attachment is crucial for Blastocystis survival and important for its pathogenicity, and thus would be an appealing drug target.

Modulators that prevent robust attachment may allow Blastocystis to be flushed from the intestinal lumen by host peristalsis, and ultimately result in

201 reduction in infection. Chapter 3 revealed that galactose was an effective inhibitor for attachment of the metronidazole-resistant isolate ST-7 (B). With addition of galactose, the results showed that the ZO-1 tight junction degradation by ST-7 (B) was prevented, pointing towards the possible application of adhesion inhibitors. Interestingly, the adhesion of the highly adhesive strain ST-7 (H) was also significantly prevented and the pathogenic effects of H on Caco-2 cells were also ameliorated, suggesting galactose therapy might have broader applications for Blastocystis infections by different strains. The result also suggested that similar proteins involved in adhesion and other processes related to pathophysiology of Blastocystis. It would be interesting to perform comparative genome sequencing and compare the genomic or expression variations within the same ST or between STs, which would yield insights into the identification and analysis of the key factors associated with the pathobiology of blastocystosis.

Altogether, these investigations have revealed fresh insight into the molecular mechanisms of Blastocystis pathogenesis, reiterating the pathogenic potential of the parasite, as well as providing an impetus for therapeutic targets. The study also points out interesting directions to pursue in the future of

Blastocystis research.

202

REFERENCES

203

Albrecht, S., Schols, H.A., van Zoeren, D., van Lingen, R.A., Groot Jebbink,

L.J., van den Heuvel, E.G., et al. (2011). Oligosaccharides in feces of breast- and formula-fed babies. Carbohydrate research 346, 2173-2181.

Alderete, J.F. and Garza, G.E. (1985). Specific nature of Trichomonas vaginalis parasitism of host cell surfaces. Infect Immun 50, 701-708.

Alexeieff, A. (1911). Sur la nature des formations dites "Kystes de

Trichomonas intestinalis". . C. R. Soc. Biol. 71, 296-298.

Alfellani, M.A., Jacob, A.S., Perea, N.O., Krecek, R.C., Taner-Mulla, D.,

Verweij, J.J., et al. (2013a). Diversity and distribution of Blastocystis sp. subtypes in non-human primates. Parasitology 140, 966-971.

Alfellani, M.A., Taner-Mulla, D., Jacob, A.S., Imeede, C.A., Yoshikawa, H.,

Stensvold, C.R. and Clark, C.G. (2013b). Genetic diversity of Blastocystis in livestock and zoo animals. Protist 164, 497-509.

Anderson, J.M. and Van Itallie, C.M. (2009). Physiology and function of the tight junction. Cold Spring Harb Perspect Biol 1, a002584.

Andersson, D.I. (2006). The biological cost of mutational antibiotic resistance: any practical conclusions? Curr Opin Microbiol 9, 461-465.

Andersson, D.I. and Hughes, D. (2010). Antibiotic resistance and its cost: is it possible to reverse resistance? Nat Rev Microbiol 8, 260-271.

204

Arisue, N., Hashimoto, T., Yoshikawa, H., Nakamura, Y., Nakamura, G.,

Nakamura, F., et al. (2002). Phylogenetic position of Blastocystis hominis and of stramenopiles inferred from multiple molecular sequence data. The Journal of eukaryotic microbiology 49, 42-53.

Arndt, N.X., Tiralongo, J., Madge, P.D., von Itzstein, M. and Day, C.J. (2011).

Differential carbohydrate binding and cell surface glycosylation of human cancer cell lines. Journal of cellular biochemistry 112, 2230-2240.

Arroyo, R. and Alderete, J.F. (1989). Trichomonas vaginalis surface proteinase activity is necessary for parasite adherence to epithelial cells. Infect

Immun 57, 2991-2997.

Arroyo, R., Engbring, J. and Alderete, J.F. (1992). Molecular basis of host epithelial cell recognition by Trichomonas vaginalis. Mol Microbiol 6, 853-

862.

Baranwal, S., Naydenov, N.G., Harris, G., Dugina, V., Morgan, K.G.,

Chaponnier, C. and Ivanov, A.I. (2012). Nonredundant roles of cytoplasmic beta- and gamma-actin isoforms in regulation of epithelial apical junctions.

Molecular biology of the cell 23, 3542-3553.

Barry, M. (2004). Travelers’ Diarrhea. The New England journal of medicine

350, 1801-1803.

205

Beaulieu, J.F. (1999). Integrins and human intestinal cell functions. Front

Biosci 4, D310-321.

Becker, S.M., Cho, K.N., Guo, X., Fendig, K., Oosman, M.N., Whitehead, R., et al. (2010). Epithelial cell apoptosis facilitates Entamoeba histolytica infection in the gut. The American journal of pathology 176, 1316-1322.

Ben-Ami, R. and Kontoyiannis, D.P. (2012). Resistance to echinocandins comes at a cost: the impact of FKS1 hotspot mutations on Candida albicans fitness and virulence. Virulence 3, 95-97.

Berkes, J., Viswanathan, V.K., Savkovic, S.D. and Hecht, G. (2003). Intestinal epithelial responses to enteric pathogens: effects on the tight junction barrier, ion transport, and inflammation. Gut 52, 439-451.

Betanzos, A., Javier-Reyna, R., Garcia-Rivera, G., Banuelos, C., Gonzalez-

Mariscal, L., Schnoor, M. and Orozco, E. (2013). The EhCPADH112 complex of Entamoeba histolytica interacts with tight junction proteins occludin and claudin-1 to produce epithelial damage. PLoS One 8, e65100.

Bjarnason, I., MacPherson, A. and Hollander, D. (1995). Intestinal permeability: an overview. Gastroenterology 108, 1566-1581.

Blazquez, S., Rigothier, M.C., Huerre, M. and Guillen, N. (2007). Initiation of inflammation and cell death during liver abscess formation by Entamoeba

206 histolytica depends on activity of the galactose/N-acetyl-D-galactosamine lectin. International journal for parasitology 37, 425-433.

Bonazzi, M. and Cossart, P. (2011). Impenetrable barriers or entry portals?

The role of cell-cell adhesion during infection. J Cell Biol 195, 349-358.

Boorom, K.F., Smith, H., Nimri, L., Viscogliosi, E., Spanakos, G., Parkar, U., et al. (2008). Oh my aching gut: irritable bowel syndrome, Blastocystis, and asymptomatic infection. Parasit Vectors 1, 40.

Boreham, P.F. and Stenzel, D.J. (1993). The current status of Blastocystis hominis. Parasitology today 9, 251.

Brooks, A.E., Parsamand, T., Kelly, R.W. and Simoes-Barbosa, A. (2013). An improved quantitative method to assess adhesive properties of Trichomonas vaginalis to host vaginal ectocervical cells using flow cytometry. J Microbiol

Methods 92, 73-78.

Brumpt, E. (1912). Blastocystis hominis n sp. et formes voisines. Bull Soc

Pathol Exot 5, 725-730.

Buret, A., Olson, M.E., Gall, D.G. and Hardin, J.A. (1998). Effects of orally administered epidermal growth factor on enteropathogenic Escherichia coli infection in rabbits. Infect Immun 66, 4917-4923.

207

Buret, A.G., Mitchell, K., Muench, D.G. and Scott, K.G. (2002). Giardia lamblia disrupts tight junctional ZO-1 and increases permeability in non- transformed human small intestinal epithelial monolayers: effects of epidermal growth factor. Parasitology 125, 11-19.

Buret, A.G., Chin, A.C. and Scott, K.G. (2003). Infection of human and bovine epithelial cells with Cryptosporidium andersoni induces apoptosis and disrupts tight junctional ZO-1: effects of epidermal growth factor. Int J

Parasitol 33, 1363-1371.

Carlone, N.A., Borsotto, M., Cuffini, A.M. and Savoia, D. (1987). Effect of fosfomycin trometamol on bacterial adhesion in comparison with other chemotherapeutic agents. Eur Urol 13 Suppl 1, 86-91.

Cassidy, M.F., Stenzel, D.J. and Boreham, P.F. (1994). Electron microscopy of surface structures of Blastocystis sp. from different hosts. Parasitol Res 80,

505-511.

Chandramathi, S., Suresh, K., Anita, Z.B. and Kuppusamy, U.R. (2012).

Infections of Blastocystis hominis and microsporidia in cancer patients: are they opportunistic? Trans R Soc Trop Med Hyg 106, 267-269.

Chen, M.L., Ge, Z., Fox, J.G. and Schauer, D.B. (2006). Disruption of tight junctions and induction of proinflammatory cytokine responses in colonic epithelial cells by Campylobacter jejuni. Infection and immunity 74, 6581-

6589.

208

Chen, M.L., Pothoulakis, C. and LaMont, J.T. (2002). Protein kinase C signaling regulates ZO-1 translocation and increased paracellular flux of T84 colonocytes exposed to Clostridium difficile toxin A. J Biol Chem 277, 4247-

4254.

Chen, X.M., Levine, S.A., Tietz, P., Krueger, E., McNiven, M.A., Jefferson,

D.M., et al. (1998). Cryptosporidium parvum is cytopathic for cultured human biliary epithelia via an apoptotic mechanism. Hepatology 28, 906-913.

Chen, X.Q. (1999) Studies on rat Blastocystis. In Department of microbiology.

(PhD Thesis), National University of Singapore.

Chen, X.Q., Singh, M., Ho, L.C., Moe, K.T., Tan, S.W. and Yap, E.H.

(1997a). A survey of Blastocystis sp. in rodents. Lab Anim Sci 47, 91-94.

Chen, X.Q., Singh, M., Ho, L.C., Tan, S.W., Ng, G.C., Moe, K.T. and Yap,

E.H. (1997b). Description of a Blastocystis species from Rattus norvegicus.

Parasitol Res 83, 313-318.

Chin, A.C., Teoh, D.A., Scott, K.G., Meddings, J.B., Macnaughton, W.K. and

Buret, A.G. (2002). Strain-dependent induction of enterocyte apoptosis by

Giardia lamblia disrupts epithelial barrier function in a caspase-3-dependent manner. Infect Immun 70, 3673-3680.

209

Clark, C.G., van der Giezen, M., Alfellani, M.A. and Stensvold, C.R. (2013).

Recent developments in Blastocystis research. Advances in parasitology 82, 1-

32.

Clayburgh, D.R., Shen, L. and Turner, J.R. (2004). A porous defense: the leaky epithelial barrier in intestinal disease. Laboratory investigation; a journal of technical methods and pathology 84, 282-291.

Cliffe, L.J., Humphreys, N.E., Lane, T.E., Potten, C.S., Booth, C. and Grencis,

R.K. (2005). Accelerated intestinal epithelial cell turnover: a new mechanism of parasite expulsion. Science 308, 1463-1465.

Cotton, J.A., Beatty, J.K. and Buret, A.G. (2011). Host parasite interactions and pathophysiology in Giardia infections. Int J Parasitol 41, 925-933.

Coyle, C.M., Varughese, J., Weiss, L.M. and Tanowitz, H.B. (2012).

Blastocystis: to treat or not to treat. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 54, 105-110.

Crane, J.K., Majumdar, S. and Pickhardt, D.F., 3rd (1999). Host cell death due to enteropathogenic Escherichia coli has features of apoptosis. Infect Immun

67, 2575-2584.

da Costa, R.F., de Souza, W., Benchimol, M., Alderete, J.F. and Morgado-

Diaz, J.A. (2005). Trichomonas vaginalis perturbs the junctional complex in epithelial cells. Cell research 15, 704-716.

210

Dagci, H., Ustun, S., Taner, M.S., Ersoz, G., Karacasu, F. and Budak, S.

(2002). Protozoon infections and intestinal permeability. Acta Trop 81, 1-5.

Dai, D., Nanthkumar, N.N., Newburg, D.S. and Walker, W.A. (2000). Role of oligosaccharides and glycoconjugates in intestinal host defense. Journal of pediatric gastroenterology and nutrition 30 Suppl 2, S23-33.

Danial, N.N. and Korsmeyer, S.J. (2004). Cell death: critical control points.

Cell 116, 205-219.

Davis, P.H., Schulze, J. and Stanley, S.L., Jr. (2007). Transcriptomic comparison of two Entamoeba histolytica strains with defined virulence phenotypes identifies new virulence factor candidates and key differences in the expression patterns of cysteine proteases, lectin light chains, and calmodulin. Mol Biochem Parasitol 151, 118-128.

de Araujo Jorge, T.C. and de Souza, W. (1984). Effect of carbohydrates, periodate and enzymes in the process of endocytosis of Trypanosoma cruzi by macrophages. Acta Trop 41, 17-28.

Del Re, D.P., Miyamoto, S. and Brown, J.H. (2007). RhoA/Rho kinase up- regulate Bax to activate a mitochondrial death pathway and induce cardiomyocyte apoptosis. J Biol Chem 282, 8069-8078.

DeMali, K.A. and Burridge, K. (2003). Coupling membrane protrusion and cell adhesion. J Cell Sci 116, 2389-2397.

211

Deneke, C.F., Thorne, G.M., Larson, A.D. and Gorbach, S.L. (1985). Effect of tetracycline on the attachment of K88+ enterotoxigenic Escherichia coli to porcine small-intestinal cells. J Infect Dis 152, 1032-1036.

Denoeud, F., Roussel, M., Noel, B., Wawrzyniak, I., Da Silva, C., Diogon, M., et al. (2011). Genome sequence of the stramenopile Blastocystis, a human anaerobic parasite. Genome Biol 12, R29.

Dogruman-Al, F., Kustimur, S., Yoshikawa, H., Tuncer, C., Simsek, Z.,

Tanyuksel, M., et al. (2009). Blastocystis subtypes in irritable bowel syndrome and inflammatory bowel disease in Ankara, Turkey. Mem Inst

Oswaldo Cruz 104, 724-727.

Dogruman-Al, F., Simsek, Z., Boorom, K., Ekici, E., Sahin, M., Tuncer, C., et al. (2010). Comparison of methods for detection of Blastocystis infection in routinely submitted stool samples, and also in IBS/IBD Patients in Ankara,

Turkey. PLoS One 5, e15484.

Dominguez-Marquez, M.V., Guna, R., Munoz, C., Gomez-Munoz, M.T. and

Borras, R. (2009). High prevalence of subtype 4 among isolates of

Blastocystis hominis from symptomatic patients of a health district of

Valencia (Spain). Parasitol Res 105, 949-955.

DuBois, K.N., Abodeely, M., Sajid, M., Engel, J.C. and McKerrow, J.H.

(2006). Giardia lamblia cysteine proteases. Parasitol Res 99, 313-316.

212

Dunn, L.A. and Boreham, P.F. (1991). The in-vitro activity of drugs against

Blastocystis hominis. J Antimicrob Chemother 27, 507-516.

Dunn, L.A., Boreham, P.F. and Stenzel, D.J. (1989). Ultrastructural variation of Blastocystis hominis stocks in culture. Int J Parasitol 19, 43-56.

Dunn, L.A., Tan, K.S., Vanelle, P., Juspin, T., Crozet, M.D., Terme, T., et al.

(2012). Development of metronidazole-resistant lines of Blastocystis sp.

Parasitol Res 111, 441-450.

Earnshaw, W.C. (1995). Apoptosis: lessons from in vitro systems. Trends in cell biology 5, 217-220.

Eckmann, L., Laurent, F., Langford, T.D., Hetsko, M.L., Smith, J.R., Kagnoff,

M.F. and Gillin, F.D. (2000). Nitric oxide production by human intestinal epithelial cells and competition for arginine as potential determinants of host defense against the lumen-dwelling pathogen Giardia lamblia. J Immunol 164,

1478-1487.

Eida, O.M., Hussein, E.M., Eida, A.M., El-Moamly, A.A. and Salem, A.M.

(2008). Evaluation of the nitric oxide activity against Blastocystis hominis in vitro and in vivo. J Egypt Soc Parasitol 38, 521-536.

Ellis, J.E., Wingfield, J.M., Cole, D., Boreham, P.F. and Lloyd, D. (1993).

Oxygen affinities of metronidazole-resistant and -sensitive stocks of Giardia intestinalis. Int J Parasitol 23, 35-39.

213

Elmendorf, H.G., Dawson, S.C. and McCaffery, J.M. (2003). The cytoskeleton of Giardia lamblia. Int J Parasitol 33, 3-28.

Elmore, S. (2007). Apoptosis: a review of programmed cell death. Toxicologic pathology 35, 495-516.

Eroglu, F., Genc, A., Elgun, G. and Koltas, I.S. (2009). Identification of

Blastocystis hominis isolates from asymptomatic and symptomatic patients by

PCR. Parasitology research 105, 1589-1592.

Fasano, A. and Nataro, J.P. (2004). Intestinal epithelial tight junctions as targets for enteric bacteria-derived toxins. Adv Drug Deliv Rev 56, 795-807.

Faust, D.M., Marquay Markiewicz, J., Danckaert, A., Soubigou, G. and

Guillen, N. (2011). Human liver sinusoidal endothelial cells respond to interaction with Entamoeba histolytica by changes in morphology, integrin signalling and cell death. Cell Microbiol 13, 1091-1106.

Fichorova, R.N., Trifonova, R.T., Gilbert, R.O., Costello, C.E., Hayes, G.R.,

Lucas, J.J. and Singh, B.N. (2006). Trichomonas vaginalis lipophosphoglycan triggers a selective upregulation of cytokines by human female reproductive tract epithelial cells. Infect Immun 74, 5773-5779.

Flores-Romo, L., Estrada-Garcia, T., Shibayama-Salas, M., Campos-

Rodriguez, R., Bacon, K., Martinez-Palomo, A. and Tsutsumi, V. (1997). In

214 vitro Entamoeba histolytica adhesion to human endothelium: a comparison using two strains of different virulence. Parasitol Res 83, 397-400.

Forsell, J., Granlund, M., Stensvold, C.R., Clark, C.G. and Evengard, B.

(2012). Subtype analysis of Blastocystis isolates in Swedish patients.

European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 31, 1689-1696.

Furuse, M. (2010). Molecular basis of the core structure of tight junctions.

Cold Spring Harb Perspect Biol 2, a002907.

Furuse, M., Hirase, T., Itoh, M., Nagafuchi, A., Yonemura, S. and Tsukita, S.

(1993). Occludin: a novel integral membrane protein localizing at tight junctions. J Cell Biol 123, 1777-1788.

Gao, Y., Herndon, J.M., Zhang, H., Griffith, T.S. and Ferguson, T.A. (1998).

Antiinflammatory effects of CD95 ligand (FasL)-induced apoptosis. The

Journal of experimental medicine 188, 887-896.

Garavelli, P.L., Orsi, P. and Scaglione, L. (1988). Blastocystis hominis infection during AIDS. Lancet 2, 1364.

Garcia, L.S. (1984). Clinical relevance of Blastocystis hominis. Lancet 1,

1233-1234.

215

Gismondo, M.R., Drago, L., Fassina, C., Garlaschi, M.L., Rosina, M. and

Lombardi, A. (1994). Escherichia coli: effect of fosfomycin trometamol on some urovirulence factors. J Chemother 6, 167-172.

Griffiths, J.K., Moore, R., Dooley, S., Keusch, G.T. and Tzipori, S. (1994).

Cryptosporidium parvum infection of Caco-2 cell monolayers induces an apical monolayer defect, selectively increases transmonolayer permeability, and causes epithelial cell death. Infect Immun 62, 4506-4514.

Gobert, A.P., McGee, D.J., Akhtar, M., Mendz, G.L., Newton, J.C., Cheng,

Y., et al. (2001). Helicobacter pylori arginase inhibits nitric oxide production by eukaryotic cells: a strategy for bacterial survival. Proc Natl Acad Sci U S A

98, 13844-13849.

Govind, S.K., Khairul, A.A. and Smith, H.V. (2002). Multiple reproductive processes in Blastocystis. Trends Parasitol 18, 528.

Green, D.R. (2005). Apoptotic pathways: ten minutes to dead. Cell 121, 671-

674.

Groeger, S., Doman, E., Chakraborty, T. and Meyle, J. (2010). Effects of

Porphyromonas gingivalis infection on human gingival epithelial barrier function in vitro. Eur J Oral Sci 118, 582-589.

216

Groschwitz, K.R. and Hogan, S.P. (2009). Intestinal barrier function: molecular regulation and disease pathogenesis. The Journal of allergy and clinical immunology 124, 3-20; quiz 21-22.

Guimaraes, S., Sogayar, M.I. and Franco, M.F. (2003). Protease activity in

Giardia duodenalis trophozoites of axenic strains isolated from symptomatic and asymptomatic patients. Mem Inst Oswaldo Cruz 98, 77-81.

Halbleib, J.M. and Nelson, W.J. (2006). Cadherins in development: cell adhesion, sorting, and tissue morphogenesis. Genes & development 20, 3199-

3214.

Harris, T.J. and Tepass, U. (2010). Adherens junctions: from molecules to morphogenesis. Nature reviews. Molecular cell biology 11, 502-514.

Hartmann, M., Papavlassopoulos, H., Chandrasekaran, V., Grabosch, C.,

Beiroth, F., Lindhorst, T.K. and Rohl, C. (2012). Inhibition of bacterial adhesion to live human cells: activity and cytotoxicity of synthetic mannosides. FEBS letters 586, 1459-1465.

Hashim, A., Mulcahy, G., Bourke, B. and Clyne, M. (2006). Interaction of

Cryptosporidium hominis and Cryptosporidium parvum with primary human and bovine intestinal cells. Infect Immun 74, 99-107.

Haslett, C. (1997). Granulocyte apoptosis and inflammatory disease. British medical bulletin 53, 669-683.

217

Hirata, K.K., Que, X., Melendez-Lopez, S.G., Debnath, A., Myers, S.,

Herdman, D.S., et al. (2007). A phagocytosis mutant of Entamoeba histolytica is less virulent due to deficient proteinase expression and release. Exp

Parasitol 115, 192-199.

Ho, L.C., Singh, M., Suresh, G., Ng, G.C. and Yap, E.H. (1993). Axenic culture of Blastocystis hominis in Iscove's modified Dulbecco's medium.

Parasitology research 79, 614-616.

Horiki, N., Kaneda, Y., Maruyama, M., Fujita, Y. and Tachibana, H. (1999).

Intestinal blockage by carcinoma and Blastocystis hominis infection. Am J

Trop Med Hyg 60, 400-402.

Hu, K.C., Lin, C.C., Wang, T.E., Liu, C.Y., Chen, M.J. and Chang, W.H.

(2008). Amoebic liver abscess or is it? Gut 57, 627, 683.

Humen, M.A., Perez, P.F. and Lievin-Le Moal, V. (2011). Lipid raft- dependent adhesion of Giardia intestinalis trophozoites to a cultured human enterocyte-like Caco-2/TC7 cell monolayer leads to cytoskeleton-dependent functional injuries. Cell Microbiol 13, 1683-1702.

Hussain, R., Jaferi, W., Zuberi, S., Baqai, R., Abrar, N., Ahmed, A. and

Zaman, V. (1997). Significantly increased IgG2 subclass antibody levels to

Blastocystis hominis in patients with irritable bowel syndrome. Am J Trop

Med Hyg 56, 301-306.

218

Hussein, E.M., Hussein, A.M., Eida, M.M. and Atwa, M.M. (2008).

Pathophysiological variability of different genotypes of human Blastocystis hominis Egyptian isolates in experimentally infected rats. Parasitol Res 102,

853-860.

Huston, C.D., Boettner, D.R., Miller-Sims, V. and Petri, W.A., Jr. (2003).

Apoptotic killing and phagocytosis of host cells by the parasite Entamoeba histolytica. Infect Immun 71, 964-972.

Huston, C.D., Houpt, E.R., Mann, B.J., Hahn, C.S. and Petri, W.A., Jr. (2000).

Caspase 3-dependent killing of host cells by the parasite Entamoeba histolytica. Cell Microbiol 2, 617-625.

Huveneers, S. and Danen, E.H. (2009). Adhesion signaling - crosstalk between integrins, Src and Rho. J Cell Sci 122, 1059-1069.

Hynes, R.O. (1992). Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69, 11-25.

Idris, N.S., Dwipoerwantoro, P.G., Kurniawan, A. and Said, M. (2010).

Intestinal parasitic infection of immunocompromised children with diarrhoea: clinical profile and therapeutic response. Journal of infection in developing countries 4, 309-317.

219

Iguchi, A., Ebisu, A., Nagata, S., Saitou, Y., Yoshikawa, H., Iwatani, S. and

Kimata, I. (2007). Infectivity of different genotypes of human Blastocystis hominis isolates in chickens and rats. Parasitol Int 56, 107-112.

Iguchi, A., Yoshikawa, H., Yamada, M., Kimata, I. and Arizono, N. (2009).

Expression of interferon gamma and proinflammatory cytokines in the cecal mucosa of rats experimentally infected with Blastocystis sp. strain RN94-9.

Parasitol Res 105, 135-140.

Itoh, M., Furuse, M., Morita, K., Kubota, K., Saitou, M. and Tsukita, S.

(1999). Direct binding of three tight junction-associated MAGUKs, ZO-1, ZO-

2, and ZO-3, with the COOH termini of claudins. J Cell Biol 147, 1351-1363.

Jackson, Y., Alirol, E., Getaz, L., Wolff, H., Combescure, C. and Chappuis, F.

(2010). Tolerance and safety of nifurtimox in patients with chronic chagas disease. Clinical infectious diseases : an official publication of the Infectious

Diseases Society of America 51, e69-75.

Jantscher-Krenn, E., Lauwaet, T., Bliss, L.A., Reed, S.L., Gillin, F.D. and

Bode, L. (2012). Human milk oligosaccharides reduce Entamoeba histolytica attachment and cytotoxicity in vitro. The British journal of nutrition 108,

1839-1846.

Joe, A., Verdon, R., Tzipori, S., Keusch, G.T. and Ward, H.D. (1998).

Attachment of Cryptosporidium parvum sporozoites to human intestinal epithelial cells. Infect Immun 66, 3429-3432.

220

Joe, A., Hamer, D.H., Kelley, M.A., Pereira, M.E., Keusch, G.T., Tzipori, S. and Ward, H.D. (1994). Role of a Gal/GalNAc-specific sporozoite surface lectin in Cryptosporidium parvum-host cell interaction. The Journal of eukaryotic microbiology 41, 44S.

Johnson, A.M., Thanou, A., Boreham, P.F. and Baverstock, P.R. (1989).

Blastocystis hominis: phylogenetic affinities determined by rRNA sequence comparison. Exp Parasitol 68, 283-288.

Kaneda, Y., Horiki, N., Cheng, X., Tachibana, H. and Tsutsumi, Y. (2000).

Serologic response to Blastocystis hominis infection in asymptomatic individuals. Tokai J Exp Clin Med 25, 51-56.

Katelaris, P.H., Naeem, A. and Farthing, M.J. (1995). Attachment of Giardia lamblia trophozoites to a cultured human intestinal cell line. Gut 37, 512-518.

Katsarou-Katsari, A., Vassalos, C.M., Tzanetou, K., Spanakos, G.,

Papadopoulou, C. and Vakalis, N. (2008). Acute urticaria associated with amoeboid forms of Blastocystis sp. subtype 3. Acta dermato-venereologica 88,

80-81.

Kaya, S., Cetin, E.S., Aridogan, B.C., Arikan, S. and Demirci, M. (2007).

Pathogenicity of Blastocystis hominis, a clinical reevaluation. Turkiye parazitolojii dergisi / Turkiye Parazitoloji Dernegi = Acta parasitologica

Turcica / Turkish Society for Parasitology 31, 184-187.

221

Khan, N.A. and Siddiqui, R. (2009). Acanthamoeba affects the integrity of human brain microvascular endothelial cells and degrades the tight junction proteins. Int J Parasitol 39, 1611-1616.

Klein, U., Pawelzik, M. and Opferkuch, W. (1985). Influence of beta-lactam antibiotics, fosfomycin and vancomycin on the adherence (hemagglutination) of Escherichia coli-containing different adhesins. Chemotherapy 31, 138-145.

Knutton, S., Baldwin, T., Williams, P.H. and McNeish, A.S. (1989). Actin accumulation at sites of bacterial adhesion to tissue culture cells: basis of a new diagnostic test for enteropathogenic and enterohemorrhagic Escherichia coli. Infect Immun 57, 1290-1298.

Krieger, J.N., Wolner-Hanssen, P., Stevens, C. and Holmes, K.K. (1990).

Characteristics of Trichomonas vaginalis isolates from women with and without colpitis macularis. J Infect Dis 161, 307-311.

Kunz, C., Rudloff, S., Baier, W., Klein, N. and Strobel, S. (2000).

Oligosaccharides in human milk: structural, functional, and metabolic aspects.

Annual review of nutrition 20, 699-722.

Kurniawan, A., Karyadi, T., Dwintasari, S.W., Sari, I.P., Yunihastuti, E.,

Djauzi, S. and Smith, H.V. (2009). Intestinal parasitic infections in HIV/AIDS patients presenting with diarrhoea in Jakarta, Indonesia. Transactions of the

Royal Society of Tropical Medicine and Hygiene 103, 892-898.

222

Kuwana, T., Smith, J.J., Muzio, M., Dixit, V., Newmeyer, D.D. and

Kornbluth, S. (1998). Apoptosis induction by caspase-8 is amplified through the mitochondrial release of cytochrome c. J Biol Chem 273, 16589-16594.

Kwak, Y.K., Vikstrom, E., Magnusson, K.E., Vecsey-Semjen, B., Colque-

Navarro, P. and Mollby, R. (2012). The Staphylococcus aureus alpha-toxin perturbs the barrier function in Caco-2 epithelial cell monolayers by altering junctional integrity. Infect Immun 80, 1670-1680.

Lanuza, M.D., Carbajal, J.A. and Borras, R. (1996). Identification of surface coat carbohydrates in Blastocystis hominis by lectin probes. Int J Parasitol 26,

527-532.

Laughlin, R.C., McGugan, G.C., Powell, R.R., Welter, B.H. and Temesvari,

L.A. (2004). Involvement of raft-like plasma membrane domains of

Entamoeba histolytica in pinocytosis and adhesion. Infect Immun 72, 5349-

5357.

Lauwaet, T., Oliveira, M.J., Callewaert, B., De Bruyne, G., Mareel, M. and

Leroy, A. (2004a). Proteinase inhibitors TPCK and TLCK prevent Entamoeba histolytica induced disturbance of tight junctions and microvilli in enteric cell layers in vitro. International journal for parasitology 34, 785-794.

Lauwaet, T., Oliveira, M.J., De Bruyne, G., Bruchhaus, I., Duchene, M.,

Mareel, M. and Leroy, A. (2004b). Entamoeba histolytica trophozoites transfer lipophosphopeptidoglycans to enteric cell layers. Int J Parasitol 34, 549-556.

223

Leitsch, D., Burgess, A.G., Dunn, L.A., Krauer, K.G., Tan, K., Duchene, M., et al. (2011). Pyruvate:ferredoxin oxidoreductase and thioredoxin reductase are involved in 5-nitroimidazole activation while flavin metabolism is linked to 5-nitroimidazole resistance in Giardia lamblia. J Antimicrob Chemother 66,

1756-1765.

Leitsch, D., Kolarich, D., Binder, M., Stadlmann, J., Altmann, F. and

Duchene, M. (2009). Trichomonas vaginalis: metronidazole and other nitroimidazole drugs are reduced by the flavin enzyme thioredoxin reductase and disrupt the cellular redox system. Implications for nitroimidazole toxicity and resistance. Mol Microbiol 72, 518-536.

Leitsch, D., Kolarich, D. and Duchene, M. (2010). The flavin inhibitor diphenyleneiodonium renders Trichomonas vaginalis resistant to metronidazole, inhibits thioredoxin reductase and flavin reductase, and shuts off hydrogenosomal enzymatic pathways. Mol Biochem Parasitol 171, 17-24.

Leroy, A., De Bruyne, G., Mareel, M., Nokkaew, C., Bailey, G. and Nelis, H.

(1995). Contact-dependent transfer of the galactose-specific lectin of

Entamoeba histolytica to the lateral surface of enterocytes in culture. Infect

Immun 63, 4253-4260.

Leroy, A., Lauwaet, T., De Bruyne, G., Cornelissen, M. and Mareel, M.

(2000). Entamoeba histolytica disturbs the tight junction complex in human enteric T84 cell layers. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 14, 1139-1146.

224

Levy, P., Robin, H., Kornprobst, M., Capeau, J. and Cherqui, G. (1998).

Enterocytic differentiation of the human Caco-2 cell line correlates with alterations in integrin signaling. J Cell Physiol 177, 618-627.

Lewis, N.D., Asim, M., Barry, D.P., Singh, K., de Sablet, T., Boucher, J.L., et al. (2010). Arginase II restricts host defense to Helicobacter pylori by attenuating inducible nitric oxide synthase translation in macrophages. J

Immunol 184, 2572-2582.

Li, E., Becker, A. and Stanley, S.L., Jr. (1988). Use of Chinese hamster ovary cells with altered glycosylation patterns to define the carbohydrate specificity of Entamoeba histolytica adhesion. The Journal of experimental medicine 167,

1725-1730.

Li, E., Stenson, W.F., Kunz-Jenkins, C., Swanson, P.E., Duncan, R. and

Stanley, S.L., Jr. (1994). Entamoeba histolytica interactions with polarized human intestinal Caco-2 epithelial cells. Infection and immunity 62, 5112-

5119.

Li, H., Zhu, H., Xu, C.J. and Yuan, J. (1998). Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 94,

491-501.

Li Pira, G., Pruzzo, C. and Schito, G.C. (1987). Monuril and modification of pathogenicity traits in resistant microorganisms. Eur Urol 13 Suppl 1, 92-97.

225

Liu, C., Sun, C., Huang, H., Janda, K. and Edgington, T. (2003).

Overexpression of legumain in tumors is significant for invasion/metastasis and a candidate enzymatic target for prodrug therapy. Cancer Res 63, 2957-

2964.

Llibre, J.M., Tor, J., Manterola, J.M., Carbonell, C. and Foz, M. (1989).

Blastocystis hominis chronic diarrhoea in AIDS patients. Lancet 1, 221.

Londono, A.L., Mejia, S. and Gomez-Marin, J.E. (2009). [Prevalence and risk factors associated with intestinal parasitism in preschool children from the urban area of Calarca, Colombia]. Revista de salud publica 11, 72-81.

Long, H.Y., Handschack, A., Konig, W. and Ambrosch, A. (2001).

Blastocystis hominis modulates immune responses and cytokine release in colonic epithelial cells. Parasitol Res 87, 1029-1030.

Lu, L. and Walker, W.A. (2001). Pathologic and physiologic interactions of bacteria with the gastrointestinal epithelium. The American journal of clinical nutrition 73, 1124S-1130S.

Lyons, A.B. and Parish, C.R. (1994). Determination of lymphocyte division by flow cytometry. Journal of immunological methods 171, 131-137.

MacPherson, D.W. and MacQueen, W.M. (1994). Morphological diversity of

Blastocystis hominis in sodium acetate-acetic acid-formalin-preserved stool samples stained with iron hematoxylin. J Clin Microbiol 32, 267-268.

226

Madara, J.L. (1988). Tight junction dynamics: is paracellular transport regulated? Cell 53, 497-498.

Magne, D., Favennec, L., Chochillon, C., Gorenflot, A., Meillet, D., Kapel,

N., et al. (1991). Role of cytoskeleton and surface lectins in Giardia duodenalis attachment to Caco2 cells. Parasitology research 77, 659-662.

Mahmoud, M.S. and Saleh, W.A. (2003). Secretory and humoral antibody responses to Blastocystis hominis in symptomatic and asymptomatic human infections. J Egypt Soc Parasitol 33, 13-30.

Maia-Brigagao, C., Morgado-Diaz, J.A. and De Souza, W. (2012). Giardia disrupts the arrangement of tight, adherens and desmosomal junction proteins of intestinal cells. Parasitol Int 61, 280-287.

Malheiros, A.F., Stensvold, C.R., Clark, C.G., Braga, G.B. and Shaw, J.J.

(2011). Short report: Molecular characterization of Blastocystis obtained from members of the indigenous Tapirape ethnic group from the Brazilian Amazon region, Brazil. The American journal of tropical medicine and hygiene 85,

1050-1053.

Man, S.M., Kaakoush, N.O., Leach, S.T., Nahidi, L., Lu, H.K., Norman, J., et al. (2010). Host attachment, invasion, and stimulation of proinflammatory cytokines by Campylobacter concisus and other non-Campylobacter jejuni

Campylobacter species. J Infect Dis 202, 1855-1865.

227

Manes, S., del Real, G. and Martinez, A.C. (2003). Pathogens: raft hijackers.

Nat Rev Immunol 3, 557-568.

Marchiando, A.M., Shen, L., Graham, W.V., Edelblum, K.L., Duckworth,

C.A., Guan, Y., et al. (2011). The epithelial barrier is maintained by in vivo tight junction expansion during pathologic intestinal epithelial shedding.

Gastroenterology 140, 1208-1218 e1201-1202.

Marino, A., Costa, R. and De Natale, G. (1990). [Cardiotoxicity of emetine].

La Clinica terapeutica 133, 131-143.

Mastronicola, D., Testa, F., Forte, E., Bordi, E., Pucillo, L.P., Sarti, P. and

Giuffre, A. (2010). Flavohemoglobin and nitric oxide detoxification in the human protozoan parasite Giardia intestinalis. Biochem Biophys Res Commun

399, 654-658.

McCabe, R.E., Yu, G.S., Conteas, C., Morrill, P.R. and McMorrow, B. (1991).

In vitro model of attachment of Giardia intestinalis trophozoites to IEC-6 cells, an intestinal cell line. Antimicrob Agents Chemother 35, 29-35.

McCarthy, K.M., Skare, I.B., Stankewich, M.C., Furuse, M., Tsukita, S.,

Rogers, R.A., et al. (1996). Occludin is a functional component of the tight junction. J Cell Sci 109 ( Pt 9), 2287-2298.

228

McClure, H.M., Strobert, E.A. and Healy, G.R. (1980). Blastocystis hominis in a pig-tailed macaque: a potential enteric pathogen for nonhuman primates.

Lab Anim Sci 30, 890-894.

McCole, D.F., Eckmann, L., Laurent, F. and Kagnoff, M.F. (2000). Intestinal epithelial cell apoptosis following Cryptosporidium parvum infection. Infect

Immun 68, 1710-1713.

Mehraj, V., Hatcher, J., Akhtar, S., Rafique, G. and Beg, M.A. (2008).

Prevalence and factors associated with intestinal parasitic infection among children in an urban slum of Karachi. PloS one 3, e3680.

Meloni, D., Sanciu, G., Poirier, P., El Alaoui, H., Chabe, M., Delhaes, L., et al. (2011). Molecular subtyping of Blastocystis sp. isolates from symptomatic patients in Italy. Parasitology research 109, 613-619.

Mendoza-Lopez, M.R., Becerril-Garcia, C., Fattel-Facenda, L.V., Avila-

Gonzalez, L., Ruiz-Tachiquin, M.E., Ortega-Lopez, J. and Arroyo, R. (2000).

CP30, a cysteine proteinase involved in Trichomonas vaginalis cytoadherence.

Infect Immun 68, 4907-4912.

Mirza, H. and Tan, K.S. (2009). Blastocystis exhibits inter- and intra-subtype variation in cysteine protease activity. Parasitology research 104, 355-361.

Mirza, H. and Tan, K.S. (2012a) Clinical Aspects of Blastocystis Infections:

Advancements Amidst Controversies. In Blastocystis: Pathogen or

229

Passenger?, H. Mehlhorn, K.S. Tan, H. Yoshikawa (eds.). Verlag Berlin

Heidelberg, Springer, pp. 65-84.

Mirza, H., Teo, J.D., Upcroft, J. and Tan, K.S. (2011a). A rapid, high- throughput viability assay for Blastocystis spp. reveals metronidazole resistance and extensive subtype-dependent variations in drug susceptibilities.

Antimicrobial agents and chemotherapy 55, 637-648.

Mirza, H., Wu, Z., Kidwai, F. and Tan, K.S. (2011b). A metronidazole- resistant isolate of Blastocystis spp. is susceptible to nitric oxide and downregulates intestinal epithelial inducible nitric oxide synthase by a novel parasite survival mechanism. Infection and immunity 79, 5019-5026.

Mirza, H., Wu, Z., Teo, J.D. and Tan, K.S. (2012b). Statin pleiotropy prevents rho kinase-mediated intestinal epithelial barrier compromise induced by

Blastocystis cysteine proteases. Cellular microbiology 14, 1474-1484.

Mittal, K., Welter, B.H. and Temesvari, L.A. (2008). Entamoeba histolytica: lipid rafts are involved in adhesion of trophozoites to host extracellular matrix components. Experimental parasitology 120, 127-134.

Moe, K.T., Singh, M., Howe, J., Ho, L.C., Tan, S.W., Ng, G.C., et al. (1996).

Observations on the ultrastructure and viability of the cystic stage of

Blastocystis hominis from human feces. Parasitol Res 82, 439-444.

230

Moghaddam, D.D., Ghadirian, E. and Azami, M. (2005). Blastocystis hominis and the evaluation of efficacy of metronidazole and trimethoprim/sulfamethoxazole. Parasitol Res 96, 273-275.

Morton, L.D., McLaughlin, G.L. and Whiteley, H.E. (1991). Effects of temperature, amebic strain, and carbohydrates on Acanthamoeba adherence to corneal epithelium in vitro. Infect Immun 59, 3819-3822.

Muller, N. and von Allmen, N. (2005). Recent insights into the mucosal reactions associated with Giardia lamblia infections. Int J Parasitol 35, 1339-

1347.

Muregi, F.W., Ohta, I., Masato, U., Kino, H. and Ishih, A. (2011). Resistance of a rodent malaria parasite to a thymidylate synthase inhibitor induces an apoptotic parasite death and imposes a huge cost of fitness. PLoS One 6, e21251.

Muza-Moons, M.M., Koutsouris, A. and Hecht, G. (2003). Disruption of cell polarity by enteropathogenic Escherichia coli enables basolateral membrane proteins to migrate apically and to potentiate physiological consequences.

Infect Immun 71, 7069-7078.

Muzio, M., Chinnaiyan, A.M., Kischkel, F.C., O'Rourke, K., Shevchenko, A.,

Ni, J., et al. (1996). FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death--inducing signaling complex. Cell 85, 817-827.

231

Nagel, R., Cuttell, L., Stensvold, C.R., Mills, P.C., Bielefeldt-Ohmann, H. and

Traub, R.J. (2012). Blastocystis subtypes in symptomatic and asymptomatic family members and pets and response to therapy. Intern Med J 42, 1187-

1195.

Naresh, R., Song, Y. and Hampson, D.J. (2009). The intestinal spirochete

Brachyspira pilosicoli attaches to cultured Caco-2 cells and induces pathological changes. PLoS One 4, e8352.

Navarro, C., Dominguez-Marquez, M.V., Garijo-Toledo, M.M., Vega-Garcia,

S., Fernandez-Barredo, S., Perez-Gracia, M.T., et al. (2008). High prevalence of Blastocystis sp. in pigs reared under intensive growing systems: frequency of ribotypes and associated risk factors. Vet Parasitol 153, 347-358.

Nazli, A., Chan, O., Dobson-Belaire, W.N., Ouellet, M., Tremblay, M.J.,

Gray-Owen, S.D., et al. (2010). Exposure to HIV-1 directly impairs mucosal epithelial barrier integrity allowing microbial translocation. PLoS Pathog 6, e1000852.

Nigro, L., Larocca, L., Massarelli, L., Patamia, I., Minniti, S., Palermo, F. and

Cacopardo, B. (2003). A placebo-controlled treatment trial of Blastocystis hominis infection with metronidazole. Journal of travel medicine 10, 128-130.

Nilsson, A.I., Berg, O.G., Aspevall, O., Kahlmeter, G. and Andersson, D.I.

(2003). Biological costs and mechanisms of fosfomycin resistance in

Escherichia coli. Antimicrob Agents Chemother 47, 2850-2858.

232

Noel, C., Dufernez, F., Gerbod, D., Edgcomb, V.P., Delgado-Viscogliosi, P.,

Ho, L.C., et al. (2005). Molecular phylogenies of Blastocystis isolates from different hosts: implications for genetic diversity, identification of species, and zoonosis. J Clin Microbiol 43, 348-355.

Noureldin, M.S., Shaltout, A.A., El Hamshary, E.M. and Ali, M.E. (1999).

Opportunistic intestinal protozoal infections in immunocompromised children.

J Egypt Soc Parasitol 29, 951-961.

O'Hara, J.R. and Buret, A.G. (2008). Mechanisms of intestinal tight junctional disruption during infection. Front Biosci 13, 7008-7021.

Ofek, I., Hasty, D.L. and Sharon, N. (2003). Anti-adhesion therapy of bacterial diseases: prospects and problems. FEMS immunology and medical microbiology 38, 181-191.

Opferman, J.T. and Korsmeyer, S.J. (2003). Apoptosis in the development and maintenance of the immune system. Nature immunology 4, 410-415.

Panaro, M.A., Cianciulli, A., Mitolo, V., Mitolo, C.I., Acquafredda, A.,

Brandonisio, O. and Cavallo, P. (2007). Caspase-dependent apoptosis of the

HCT-8 epithelial cell line induced by the parasite Giardia intestinalis. FEMS immunology and medical microbiology 51, 302-309.

233

Panjwani, N., Zhao, Z., Baum, J., Hazlett, L.D. and Yang, Z. (1997).

Acanthamoebae bind to rabbit corneal epithelium in vitro. Investigative ophthalmology & visual science 38, 1858-1864.

Park, B.J., Haines, T. and Abu-Lail, N.I. (2009). A correlation between the virulence and the adhesion of Listeria monocytogenes to silicon nitride: an atomic force microscopy study. Colloids Surf B Biointerfaces 73, 237-243.

Parkar, U., Traub, R.J., Kumar, S., Mungthin, M., Vitali, S., Leelayoova, S., et al. (2007). Direct characterization of Blastocystis from faeces by PCR and evidence of zoonotic potential. Parasitology 134, 359-367.

Patino, W.D., Cavuoti, D., Banerjee, S.K., Swartz, K., Ashfaq, R. and

Gokaslan, T. (2008). Cytologic diagnosis of Blastocystis hominis in peritoneal fluid: a case report. Acta cytologica 52, 718-720.

Pegado, M.G. and de Souza, W. (1994). Role of surface components in the process of interaction of Giardia duodenalis with epithelial cells in vitro.

Parasitol Res 80, 320-326.

Peiffer, I., Bernet-Camard, M.F., Rousset, M. and Servin, A.L. (2001).

Impairments in enzyme activity and biosynthesis of brush border-associated hydrolases in human intestinal Caco-2/TC7 cells infected by members of the

Afa/Dr family of diffusely adhering Escherichia coli. Cell Microbiol 3, 341-

357.

234

Peiffer, I., Blanc-Potard, A.B., Bernet-Camard, M.F., Guignot, J., Barbat, A. and Servin, A.L. (2000). Afa/Dr diffusely adhering Escherichia coli C1845 infection promotes selective injuries in the junctional domain of polarized human intestinal Caco-2/TC7 cells. Infect Immun 68, 3431-3442.

Peter, M.E. and Krammer, P.H. (2003). The CD95(APO-1/Fas) DISC and beyond. Cell death and differentiation 10, 26-35.

Poirier, P., Meloni, D., Nourrisson, C., Wawrzyniak, I., Viscogliosi, E.,

Livrelli, V. and Delbac, F. (2014). Molecular subtyping of Blastocystis spp. using a new rDNA marker from the mitochondria-like organelle genome.

Parasitology 141, 670-681.

Poirier, P., Wawrzyniak, I., Albert, A., El Alaoui, H., Delbac, F. and Livrelli,

V. (2011). Development and evaluation of a real-time PCR assay for detection and quantification of Blastocystis parasites in human stool samples: prospective study of patients with hematological malignancies. J Clin

Microbiol 49, 975-983.

Poirier, P., Wawrzyniak, I., Vivares, C.P., Delbac, F. and El Alaoui, H.

(2012). New insights into Blastocystis spp.: a potential link with irritable bowel syndrome. PLoS pathogens 8, e1002545.

Pompilio, A., Catavitello, C., Picciani, C., Confalone, P., Piccolomini, R.,

Savini, V., et al. (2010). Subinhibitory concentrations of moxifloxacin

235 decrease adhesion and biofilm formation of Stenotrophomonas maltophilia from cystic fibrosis. J Med Microbiol 59, 76-81.

Puthia, M.K., Lu, J. and Tan, K.S. (2008). Blastocystis ratti contains cysteine proteases that mediate interleukin-8 response from human intestinal epithelial cells in an NF-kappaB-dependent manner. Eukaryot Cell 7, 435-443.

Puthia, M.K., Sio, S.W., Lu, J. and Tan, K.S. (2006). Blastocystis ratti induces contact-independent apoptosis, F-actin rearrangement, and barrier function disruption in IEC-6 cells. Infection and immunity 74, 4114-4123.

Puthia, M.K., Vaithilingam, A., Lu, J. and Tan, K.S. (2005). Degradation of human secretory immunoglobulin A by Blastocystis. Parasitology research

97, 386-389.

Rafferty, S., Luu, B., March, R.E. and Yee, J. (2010). Giardia lamblia encodes a functional flavohemoglobin. Biochem Biophys Res Commun 399, 347-351.

Ragland, B.D., Ashley, L.S., Vaux, D.L. and Petri, W.A., Jr. (1994).

Entamoeba histolytica: target cells killed by trophozoites undergo DNA fragmentation which is not blocked by Bcl-2. Exp Parasitol 79, 460-467.

Rao, K., Sekar, U., Iraivan, K.T., Abraham, G. and Soundararajan, P. (2003).

Blastocystis hominis--an emerging cause of diarrhoea in renal transplant recipients. The Journal of the Association of Physicians of India 51, 719-721.

236

Rasoloson, D., Tomkova, E., Cammack, R., Kulda, J. and Tachezy, J. (2001).

Metronidazole-resistant strains of Trichomonas vaginalis display increased susceptibility to oxygen. Parasitology 123, 45-56.

Ravdin, J.I. and Guerrant, R.L. (1981). Role of adherence in cytopathogenic mechanisms of Entamoeba histolytica. Study with mammalian tissue culture cells and human erythrocytes. The Journal of clinical investigation 68, 1305-

1313.

Rayan, H.Z., Ismail, O.A. and El Gayar, E.K. (2007). Prevalence and clinical features of Dientamoeba fragilis infections in patients suspected to have intestinal parasitic infection. J Egypt Soc Parasitol 37, 599-608.

Reed, S.L., Keene, W.E. and McKerrow, J.H. (1989). Thiol proteinase expression and pathogenicity of Entamoeba histolytica. J Clin Microbiol 27,

2772-2777.

Rendon-Gandarilla, F.J., Ramon-Luing Lde, L., Ortega-Lopez, J., Rosa de

Andrade, I., Benchimol, M. and Arroyo, R. (2013). The TvLEGU-1, a legumain-like cysteine proteinase, plays a key role in Trichomonas vaginalis cytoadherence. BioMed research international 2013, 561979.

Rene, B.A., Stensvold, C.R., Badsberg, J.H. and Nielsen, H.V. (2009).

Subtype analysis of Blastocystis isolates from Blastocystis cyst excreting patients. The American journal of tropical medicine and hygiene 80, 588-592.

237

Rhen, M., Eriksson, S., Clements, M., Bergstrom, S. and Normark, S.J.

(2003). The basis of persistent bacterial infections. Trends in microbiology 11,

80-86.

Rigothier, M.C., Coconnier, M.H., Servin, A.L. and Gayral, P. (1991). A new in vitro model of Entamoeba histolytica adhesion, using the human colon carcinoma cell line Caco-2: scanning electron microscopic study. Infect

Immun 59, 4142-4146.

Ringqvist, E., Avesson, L., Soderbom, F. and Svard, S.G. (2011).

Transcriptional changes in Giardia during host-parasite interactions. Int J

Parasitol 41, 277-285.

Roberts, T., Ellis, J., Harkness, J., Marriott, D. and Stark, D. (2013).

Treatment failure in patients with chronic Blastocystis infection. J Med

Microbiol.

Rodriguez-Fuentes, G.B., Cedillo-Rivera, R., Fonseca-Linan, R., Arguello-

Garcia, R., Munoz, O., Ortega-Pierres, G. and Yepez-Mulia, L. (2006).

Giardia duodenalis: analysis of secreted proteases upon trophozoite-epithelial cell interaction in vitro. Mem Inst Oswaldo Cruz 101, 693-696.

Rojas, L., Sariego, I., Fraga, J., Sarria, C.A., Sarracent, J. and Sanchez, L.

(2004). Use of in vitro cytoadherence assays in the comparative study of the virulence of isolates of Trichomonas vaginalis. Parasitol Res 93, 332-337.

238

Roos, V. and Klemm, P. (2006). Global gene expression profiling of the asymptomatic bacteriuria Escherichia coli strain 83972 in the human urinary tract. Infect Immun 74, 3565-3575.

Rossignol, J.F., Kabil, S.M., Said, M., Samir, H. and Younis, A.M. (2005).

Effect of nitazoxanide in persistent diarrhea and enteritis associated with

Blastocystis hominis. Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association 3,

987-991.

Saffer, L.D. and Petri, W.A., Jr. (1991). Role of the galactose lectin of

Entamoeba histolytica in adherence-dependent killing of mammalian cells.

Infect Immun 59, 4681-4683.

Sakaguchi, T., Kohler, H., Gu, X., McCormick, B.A. and Reinecker, H.C.

(2002). Shigella flexneri regulates tight junction-associated proteins in human intestinal epithelial cells. Cell Microbiol 4, 367-381.

Salvi, P.F., Virgilio, E., Bocchetti, T., Costa, G., Pascarella, G. and Balducci,

G. (2012). Blastocystis hominis and recurrent megacolon: a causative or fortuitous association? The American surgeon 78, E198-199.

Sambuy, Y., De Angelis, I., Ranaldi, G., Scarino, M.L., Stammati, A. and

Zucco, F. (2005). The Caco-2 cell line as a model of the intestinal barrier: influence of cell and culture-related factors on Caco-2 cell functional characteristics. Cell biology and toxicology 21, 1-26.

239

Sasahara, T., Maruyama, H., Aoki, M., Kikuno, R., Sekiguchi, T., Takahashi,

A., et al. (2003). Apoptosis of intestinal crypt epithelium after

Cryptosporidium parvum infection. Journal of infection and chemotherapy : official journal of the Japan Society of Chemotherapy 9, 278-281.

Savill, J. (1997). Apoptosis in resolution of inflammation. Journal of leukocyte biology 61, 375-380.

Scanlan, P.D. (2012). Blastocystis: past pitfalls and future perspectives.

Trends in parasitology 28, 327-334.

Scanlan, P.D. and Stensvold, C.R. (2013). Blastocystis: getting to grips with our guileful guest. Trends Parasitol, 523-529.

Scicluna, S.M., Tawari, B. and Clark, C.G. (2006). DNA barcoding of

Blastocystis. Protist 157, 77-85.

Scott, K.G., Meddings, J.B., Kirk, D.R., Lees-Miller, S.P. and Buret, A.G.

(2002). Intestinal infection with Giardia spp. reduces epithelial barrier function in a myosin light chain kinase-dependent fashion. Gastroenterology

123, 1179-1190.

Sears, C.L. (2000). Molecular physiology and pathophysiology of tight junctions V. assault of the tight junction by enteric pathogens. Am J Physiol

Gastrointest Liver Physiol 279, G1129-1134.

240

Sebbane, F., Lemaitre, N., Sturdevant, D.E., Rebeil, R., Virtaneva, K.,

Porcella, S.F. and Hinnebusch, B.J. (2006). Adaptive response of Yersinia pestis to extracellular effectors of innate immunity during bubonic plague.

Proc Natl Acad Sci U S A 103, 11766-11771.

Seigneur, M., Mounier, J., Prevost, M.C. and Guillen, N. (2005). A lysine- and glutamic acid-rich protein, KERP1, from Entamoeba histolytica binds to human enterocytes. Cell Microbiol 7, 569-579.

Sharon, N. and Ofek, I. (2000). Safe as mother's milk: Carbohydrates as future anti-adhesion drugs for bacterial diseases. Glycoconj J 17, 659-664.

Shibayama, M., Martinez-Castillo, M., Silva-Olivares, A., Galindo-Gomez, S.,

Navarro-Garcia, F., Escobar-Herrera, J., et al. (2013). Disruption of MDCK cell tight junctions by the free-living amoeba Naegleria fowleri. Microbiology

159, 392-401.

Shoaf, K., Mulvey, G.L., Armstrong, G.D. and Hutkins, R.W. (2006).

Prebiotic galactooligosaccharides reduce adherence of enteropathogenic

Escherichia coli to tissue culture cells. Infect Immun 74, 6920-6928.

Sibley, C.H. and Hunt, S.Y. (2003). Drug resistance in parasites: can we stay ahead of the evolutionary curve? Trends Parasitol 19, 532-537.

241

Silber, A.M., Tonelli, R.R., Martinelli, M., Colli, W. and Alves, M.J. (2002).

Active transport of L-proline in Trypanosoma cruzi. The Journal of eukaryotic microbiology 49, 441-446.

Silberman, J.D., Sogin, M.L., Leipe, D.D. and Clark, C.G. (1996). Human parasite finds taxonomic home. Nature 380, 398.

Singh, M., Suresh, K., Ho, L.C., Ng, G.C. and Yap, E.H. (1995). Elucidation of the life cycle of the intestinal protozoan Blastocystis hominis. Parasitol Res

81, 446-450.

Sio, S.W., Puthia, M.K., Lee, A.S., Lu, J. and Tan, K.S. (2006). Protease activity of Blastocystis hominis. Parasitology research 99, 126-130.

Solaymani-Mohammadi, S. and Singer, S.M. (2011). Host immunity and pathogen strain contribute to intestinal disaccharidase impairment following gut infection. J Immunol 187, 3769-3775.

Sommer, U., Costello, C.E., Hayes, G.R., Beach, D.H., Gilbert, R.O., Lucas,

J.J. and Singh, B.N. (2005). Identification of Trichomonas vaginalis cysteine proteases that induce apoptosis in human vaginal epithelial cells. J Biol Chem

280, 23853-23860.

Souppart, L., Moussa, H., Cian, A., Sanciu, G., Poirier, P., El Alaoui, H., et al.

(2010). Subtype analysis of Blastocystis isolates from symptomatic patients in

Egypt. Parasitology research 106, 505-511.

242

Souppart, L., Sanciu, G., Cian, A., Wawrzyniak, I., Delbac, F., Capron, M., et al. (2009). Molecular epidemiology of human Blastocystis isolates in France.

Parasitol Res 105, 413-421.

Sousa, M.C., Goncalves, C.A., Bairos, V.A. and Poiares-Da-Silva, J. (2001).

Adherence of Giardia lamblia trophozoites to Int-407 human intestinal cells.

Clinical and diagnostic laboratory immunology 8, 258-265.

Spitz, J., Yuhan, R., Koutsouris, A., Blatt, C., Alverdy, J. and Hecht, G.

(1995). Enteropathogenic Escherichia coli adherence to intestinal epithelial monolayers diminishes barrier function. Am J Physiol 268, G374-379.

Steed, E., Balda, M.S. and Matter, K. (2010). Dynamics and functions of tight junctions. Trends in cell biology 20, 142-149.

Stenson, W.F., Zhang, Z., Riehl, T. and Stanley, S.L., Jr. (2001). Amebic infection in the human colon induces cyclooxygenase-2. Infect Immun 69,

3382-3388.

Stensvold, C.R. (2013). Blastocystis: Genetic diversity and molecular methods for diagnosis and epidemiology. Tropical parasitology 3, 26-34.

Stensvold, C.R., Ahmed, U.N., Andersen, L.O. and Nielsen, H.V. (2012a).

Development and evaluation of a genus-specific, probe-based, internal- process-controlled real-time PCR assay for sensitive and specific detection of

Blastocystis spp. J Clin Microbiol 50, 1847-1851.

243

Stensvold, C.R., Alfellani, M. and Clark, C.G. (2012b). Levels of genetic diversity vary dramatically between Blastocystis subtypes. Infection, genetics and evolution : journal of molecular epidemiology and evolutionary genetics in infectious diseases 12, 263-273.

Stensvold, C.R., Alfellani, M.A., Norskov-Lauritsen, S., Prip, K., Victory,

E.L., Maddox, C., et al. (2009a). Subtype distribution of Blastocystis isolates from synanthropic and zoo animals and identification of a new subtype. Int J

Parasitol 39, 473-479.

Stensvold, C.R., Arendrup, M.C., Jespersgaard, C., Molbak, K. and Nielsen,

H.V. (2007a). Detecting Blastocystis using parasitologic and DNA-based methods: a comparative study. Diagnostic microbiology and infectious disease

59, 303-307.

Stensvold, C.R., Arendrup, M.C., Nielsen, H.V., Bada, A. and Thorsen, S.

(2008). Symptomatic infection with Blastocystis sp. subtype 8 successfully treated with trimethoprim-sulfamethoxazole. Annals of tropical medicine and parasitology 102, 271-274.

Stensvold, C.R., Arendrup, M.C., Nielsen, H.V. and Molbak, K. (2009b).

[Blastocystis--an enigmatic parasite]. Ugeskr Laeger 171, 2388-2390.

Stensvold, C.R., Lewis, H.C., Hammerum, A.M., Porsbo, L.J., Nielsen, S.S.,

Olsen, K.E., et al. (2009c). Blastocystis: unravelling potential risk factors and

244 clinical significance of a common but neglected parasite. Epidemiology and infection 137, 1655-1663.

Stensvold, C.R., Nielsen, H.V., Molbak, K. and Smith, H.V. (2009d).

Pursuing the clinical significance of Blastocystis--diagnostic limitations.

Trends Parasitol 25, 23-29.

Stensvold, C.R., Smith, H.V., Nagel, R., Olsen, K.E. and Traub, R.J. (2010).

Eradication of Blastocystis carriage with antimicrobials: reality or delusion?

Journal of clinical gastroenterology 44, 85-90.

Stensvold, C.R., Suresh, G.K., Tan, K.S., Thompson, R.C., Traub, R.J.,

Viscogliosi, E., et al. (2007b). Terminology for Blastocystis subtypes--a consensus. Trends Parasitol 23, 93-96.

Stenzel, D.J. and Boreham, P.F. (1996). Blastocystis hominis revisited.

Clinical microbiology reviews 9, 563-584.

Suresh, K., Mak, J.W., Chuong, L.S., Ragunathan, T. and Init, I. (1997). Sac- like pouches in Blastocystis from the house lizard Cosymbotus platyurus.

Parasitol Res 83, 523-525.

Suresh, K. and Smith, H. (2004). Comparison of methods for detecting

Blastocystis hominis. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical

Microbiology 23, 509-511.

245

Tafazoli, F., Holmstrom, A., Forsberg, A. and Magnusson, K.E. (2000).

Apically exposed, tight junction-associated beta1-integrins allow binding and

YopE-mediated perturbation of epithelial barriers by wild-type Yersinia bacteria. Infect Immun 68, 5335-5343.

Tan, K.S. (2004). Blastocystis in humans and animals: new insights using modern methodologies. Veterinary parasitology 126, 121-144.

Tan, K.S. (2008). New insights on classification, identification, and clinical relevance of Blastocystis spp. Clinical microbiology reviews 21, 639-665.

Tan, K.S. and Mirza, H. (2012) Blastocystis–Host Interactions: Insights from

In Vitro Model Systems. In Blastocystis: Pathogen or Passenger?, H.

Mehlhorn, K.S. Tan, H. Yoshikawa (eds.). Verlag Berlin Heidelberg, Springer, pp. 51-63.

Tan, K.S., Mirza, H., Teo, J.D., Wu, B. and Macary, P.A. (2010). Current

Views on the Clinical Relevance of Blastocystis spp. Current infectious disease reports 12, 28-35.

Tan, K.S. and Nasirudeen, A.M. (2005). Protozoan programmed cell death-- insights from Blastocystis deathstyles. Trends Parasitol 21, 547-550.

Tan, S.W., Ho, L.C., Moe, K.T., Chen, X.Q., Ng, G.C., Yap, E.H. and Singh,

M. (1996a). Production and characterization of murine monoclonal antibodies to Blastocystis hominis. Int J Parasitol 26, 375-381.

246

Tan, S.W., Singh, M., Thong, K.T., Ho, L.C., Moe, K.T., Chen, X.Q., et al.

(1996b). Clonal growth of Blastocystis hominis in soft agar with sodium thioglycollate. Parasitol Res 82, 737-739.

Tan, S.W., Singh, M., Yap, E.H., Ho, L.C., Moe, K.T., Howe, J. and Ng, G.C.

(1996c). Colony formation of Blastocystis hominis in soft agar. Parasitol Res

82, 375-377.

Tan, T.C. and Suresh, K.G. (2007). Evidence of plasmotomy in Blastocystis hominis. Parasitol Res 101, 1521-1525.

Tasova, Y., Sahin, B., Koltas, S. and Paydas, S. (2000). Clinical significance and frequency of Blastocystis hominis in Turkish patients with hematological malignancy. Acta Med Okayama 54, 133-136.

Tavares, P., Rigothier, M.C., Khun, H., Roux, P., Huerre, M. and Guillen, N.

(2005). Roles of cell adhesion and cytoskeleton activity in Entamoeba histolytica pathogenesis: a delicate balance. Infect Immun 73, 1771-1778.

Tejera, B., Grados, D., Martinez-Morillo, M. and Roure, S. (2012). [Reactive arthritis caused by Blastocystis hominis]. Reumatologia clinica 8, 50-51.

Tejman-Yarden, N., Millman, M., Lauwaet, T., Davids, B.J., Gillin, F.D.,

Dunn, L., et al. (2011). Impaired parasite attachment as fitness cost of metronidazole resistance in Giardia lamblia. Antimicrob Agents Chemother 55,

4643-4651.

247

Tsuji, E., Hiki, N., Nomura, S., Fukushima, R., Kojima, J., Ogawa, T., et al.

(2003). Simultaneous onset of acute inflammatory response, sepsis-like symptoms and intestinal mucosal injury after cancer chemotherapy.

International journal of cancer. Journal international du cancer 107, 303-308.

Turner, J.R. (2006). Molecular basis of epithelial barrier regulation: from basic mechanisms to clinical application. The American journal of pathology 169,

1901-1909.

Turner, J.R. (2009). Intestinal mucosal barrier function in health and disease.

Nat Rev Immunol 9, 799-809.

Twu, O., de Miguel, N., Lustig, G., Stevens, G.C., Vashisht, A.A.,

Wohlschlegel, J.A. and Johnson, P.J. (2013). Trichomonas vaginalis

Exosomes Deliver Cargo to Host Cells and Mediate HostratioParasite

Interactions. PLoS Pathog 9, e1003482.

Upcroft, P. and Upcroft, J.A. (2001). Drug targets and mechanisms of resistance in the anaerobic protozoa. Clin Microbiol Rev 14, 150-164.

Valenti, P., Greco, R., Pitari, G., Rossi, P., Ajello, M., Melino, G. and

Antonini, G. (1999). Apoptosis of Caco-2 intestinal cells invaded by Listeria monocytogenes: protective effect of lactoferrin. Experimental cell research

250, 197-202.

248

Vogelberg, C., Stensvold, C.R., Monecke, S., Ditzen, A., Stopsack, K.,

Heinrich-Grafe, U. and Pohlmann, C. (2010). Blastocystis sp. subtype 2 detection during recurrence of gastrointestinal and urticarial symptoms.

Parasitology international 59, 469-471.

Vogelmann, R. and Nelson, W.J. (2005). Fractionation of the epithelial apical junctional complex: reassessment of protein distributions in different substructures. Molecular biology of the cell 16, 701-716.

Waghorn, D.J. and Hancock, P. (1991). Clinical significance of Blastocystis hominis. Lancet 337, 609.

Wan, H., Winton, H.L., Soeller, C., Tovey, E.R., Gruenert, D.C., Thompson,

P.J., et al. (1999). Der p 1 facilitates transepithelial allergen delivery by disruption of tight junctions. J Clin Invest 104, 123-133.

Wawrzyniak, I., Roussel, M., Diogon, M., Couloux, A., Texier, C., Tan, K.S., et al. (2008). Complete circular DNA in the mitochondria-like organelles of

Blastocystis hominis. Int J Parasitol 38, 1377-1382.

Wawrzyniak, I., Texier, C., Poirier, P., Viscogliosi, E., Tan, K.S., Delbac, F. and El Alaoui, H. (2012). Characterization of two cysteine proteases secreted by Blastocystis ST7, a human intestinal parasite. Parasitol Int 61, 437-442.

249

Wooldridge, K.G., Williams, P.H. and Ketley, J.M. (1996). Host signal transduction and endocytosis of Campylobacter jejuni. Microbial pathogenesis

21, 299-305.

Wu, B., Yin, J., Texier, C., Roussel, M. and Tan, K.S. (2010). Blastocystis legumain is localized on the cell surface, and specific inhibition of its activity implicates a pro-survival role for the enzyme. J Biol Chem 285, 1790-1798.

Wu, Z., Nybom, P. and Magnusson, K.E. (2000). Distinct effects of Vibrio cholerae haemagglutinin/protease on the structure and localization of the tight junction-associated proteins occludin and ZO-1. Cell Microbiol 2, 11-17.

Yakoob, J., Jafri, W., Jafri, N., Islam, M. and Asim Beg, M. (2004). In vitro susceptibility of Blastocystis hominis isolated from patients with irritable bowel syndrome. British journal of biomedical science 61, 75-77.

Yamada, T., Tomita, T., Weiss, L.M. and Orlofsky, A. (2011). Toxoplasma gondii inhibits granzyme B-mediated apoptosis by the inhibition of granzyme

B function in host cells. Int J Parasitol 41, 595-607.

Yang, Z., Cao, Z. and Panjwani, N. (1997). Pathogenesis of Acanthamoeba keratitis: carbohydrate-mediated host-parasite interactions. Infect Immun 65,

439-445.

Yin, J., Ye, A.J. and Tan, K.S. (2010). Autophagy is involved in starvation response and cell death in Blastocystis. Microbiology 156, 665-677.

250

Yoshikawa, H. (2012) Epidemiology, Transmission, and Zoonotic Potential of

Blastocystis in Human and Animals. In Blastocystis: Pathogen or Passenger?

, H. Mehlhorn, K.S. Tan, H. Yoshikawa (eds.). Verlag Berlin Heidelberg,

Springer, pp. 37-49.

Yoshikawa, H. and Hayakawa, A. (1996). Freeze-fracture cytochemistry of membrane cholesterol in Blastocystis hominis. Int J Parasitol 26, 1111-1114.

Yoshikawa, H., Wu, Z., Kimata, I., Iseki, M., Ali, I.K., Hossain, M.B., et al.

(2004). Polymerase chain reaction-based genotype classification among human Blastocystis hominis populations isolated from different countries.

Parasitol Res 92, 22-29.

Yoshikawa, H., Wu, Z., Pandey, K., Pandey, B.D., Sherchand, J.B., Yanagi, T. and Kanbara, H. (2009). Molecular characterization of Blastocystis isolates from children and rhesus monkeys in Kathmandu, Nepal. Vet Parasitol 160,

295-300.

Zaman, V., Howe, J. and Ng, M. (1997). Observations on the surface coat of

Blastocystis hominis. Parasitol Res 83, 731-733.

Zaman, V., Howe, J., Ng, M. and Goh, T.K. (1999). Scanning electron microscopy of the surface coat of Blastocystis hominis. Parasitol Res 85, 974-

976.

251

Zaman, V. and Zaki, M. (1996). Resistance of Blastocystis hominis cysts to metronidazole. Tropical medicine & international health : TM & IH 1, 677-

678.

Zhang, X., Qiao, J.Y., Zhou, X.J., Yao, F.R. and Wei, Z.C. (2007).

Morphology and reproductive mode of Blastocystis hominis in diarrhea and in vitro. Parasitol Res 101, 43-51.

Zierdt, C.H. (1973). Studies of Blastocystis hominis. The Journal of protozoology 20, 114-121.

Zierdt, C.H. (1991). Blastocystis hominis--past and future. Clin Microbiol Rev

4, 61-79.

Zierdt, C.H., Rude, W.S. and Bull, B.S. (1967). Protozoan characteristics of

Blastocystis hominis. American journal of clinical pathology 48, 495-501.

Zierdt, C.H. and Swan, J.C. (1981). Generation time and growth rate of the human intestinal parasite Blastocystis hominis. The Journal of protozoology

28, 483-485.

Zierdt, C.H., Swan, J.C. and Hosseini, J. (1983). In vitro response of

Blastocystis hominis to antiprotozoal drugs. The Journal of protozoology 30,

332-334.

252

Zuckerman, M.J., Watts, M.T., Ho, H. and Meriano, F.V. (1994). Blastocystis hominis infection and intestinal injury. The American journal of the medical sciences 308, 96-101.

253