Quick viewing(Text Mode)

Najju Ranjit Thesis (PDF 5MB)

Najju Ranjit Thesis (PDF 5MB)

CHARACTERISATION OF PROTEASES INVOLVED IN PROTEOLYTIC DEGRADATION OF HAEMOGLOBIN IN THE HUMAN AMERICANUS

Najju Ranjit BBiotech (Hons)

Queensland Institute of Medical Research Queensland University of Technology

A thesis submitted for the degree of Doctor of Philosophy 2008

LIST OF KEYWORDS

Necator americanus Hookworm Laser microscopy microdissection Haemoglobin degradation Haemoglobinases Cysteine protease Aspartic protease Metalloprotease Intestinal proteases Vaccine candidates

ii ABSTRACT

With over a billion people infected world wide, are considered as important human pathogens, particularly in developing countries which have the highest rates of infections. Hookworms reside in the of the where they continuously feed on blood, leading to conditions such as chronic iron- deficiency anaemia. The majority of blood-feeding parasites rely on proteins found in blood to provide many of their nutritional requirements for growth, reproduction and survival. Of the numerous proteins found in blood, haemoglobin (Hb) is one of the most abundant. In order to acquire amino acids for protein synthesis, it is thought that haematophagous parasites degrade Hb using various classes of endo- and exo- proteases, in a manner similar to that which occurs in catabolism of proteins in mammalian cellular lysosomes. This study identified and characterised proteases involved in the Hb degradation process in the human hookworm, , in order to identify potential candidate for a vaccine that interrupts blood-feeding. Red blood cells ingested by hookworms are lysed to release Hb, which is cleaved by various proteases into dipeptides or free amino acids and these are taken up through the gut membrane by amino acid transporters. Proteases expressed in the intestinal tract of hookworms are thought to play a major role in this process and would therefore make good targets for vaccine candidates aimed at interrupting blood-feeding. To identify these proteases, adult hookworms (both N. americanus and ) were sectioned and intestinal tissue was dissected via laser microdissection microscopy. RNA extracted from the dissected tissue was used to generate gut-specific cDNA, which then was used to create plasmid libraries. Each library was subjected to shotgun sequencing, and of the 480 expressed sequence tags (ESTs) sequenced from each species, 268 and 276 contigs were assembled from the N. americanus and A. caninum libraries, respectively. Nine percent of N. americanus and 6.5% of A. caninum contigs were considered novel as no homologues were identified in any published/accessible database. The gene ontology (GO) classification system was used to categorise the contigs to predicted biological functions. Only 17% and 38% of N. americanus and A. caninum contigs, respectively, were assigned GO categories, while the rest were classified as being of

iii unknown function. The most highly represented GO categories were molecular functions such as protein binding and catalytic activity. The most abundant transcripts encoded fatty acid binding proteins, C-type lectins and activation associated secreted proteins, indicative of the diversity of functions that occur in this complex organ. Of particular interest to this study were the contigs that encoded for cysteine and metalloproteases, expanding the list of potential N. americanus haemoglobinases. In the N. americanus cDNA library, four contigs encoding for B cysteine proteases were identified. Three contigs from the A. caninum and one contig from the N. americanus cDNA libraries encoded for metalloproteases, including astacin-like and O-sialoglycoprotein endopeptidases, neither of which had previously been reported from adult hookworms. Apart from haemoglobinases, other mRNAs encoding potential vaccine candidate molecules were identified, including anti-clotting factors, defensins and membrane proteins. This study confirmed that the gut of hookworms encodes a diverse range of proteases, some of which are likely to be involved in Hb digestion and have the potential to be hidden (cryptic) vaccine antigens. Four cysteine proteases (Na-CP-2, -3, -4 and -5) were identified from the gut cDNA library of N. americanus. All four proteases belong to the clan CA, family C1, share homology with human cathepsin B and possess a modified occluding loop. Real-time PCR indicated that all transcripts were up-regulated in the adult stage of the hookworm parasite with high levels of mRNA expression detected in gut cDNA. All four proteases were expressed in recombinant form, but only Na-CP-3 was successfully expressed in soluble form in the yeast Pichia pastoris. Proteolytic activity for Na-CP-3 was detected on a gelatin zymogen gel, however no catalytic activity was detected against the class-specific fluorogenic peptides Z-Phe-Arg-AMC and Z-Arg-Arg-AMC. Mass spectrometry analysis of the purified protein suggested that the pro-region had not been processed in trans when the protein was secreted by yeast. Incubation of Na-CP-3 in salt buffers containing dextran sulfate resulted in autoprocessing of the pro-region as detected by Western blot and catalytic activity was detected against Z-Phe-Arg-AMC. Activated Na-CP-3 did not digest intact tetrameric human Hb. The other three cysteine proteases (Na-CP-2, -4, and -5) were expressed in insoluble form in Escherichia coli. Antibodies to all four proteins (Na- CP-2 to 5) immunolocalised to the gut region of the adult worm, supporting mRNA

iv amplification results and strongly indicated that they might play a role in nutrient acquisition. Hb digestion in blood feeding parasites such as schistosomes and Plasmodium spp. occurs via a semi-ordered cascade of proteolysis involving numerous enzymes. In Plasmodium falciparum, at least three distinct mechanistic classes of endopeptidases have been implicated in this process, and at least two classes have been implicated in schistosomes. A similar process is thought to occur in hookworms. An aspartic protease, Na-APR-1, was expressed in P. pastoris and purified protein was shown to cleave the class-specific fluorogenic peptide 7- Methoxycoumarin-4-Acetyl-GKPILFFRLK(DNP)-D-Arg-Amide. Recombinant Na- APR-1 was able to cleave intact human Hb and completely degrade the 16 kDa monomer and 32 kDa dimer within one hour. Recombinant Na-CP-3 was not able to cleave intact Hb, but was able to further digest globin fragments that had previously been digested with Na-APR-1. A clan MA metalloprotease, Na-MEP-1, was identified in gut tissue of N. americanus and was expressed in recombinant form in Hi5 insect cells using the baculovirus expression system. Recombinant Na-MEP-1 displayed proteolytic activity when assessed by gelatin zymography, but was incapable of cleaving intact Hb. However, Na-MEP-1 did cleave globin fragments which had previously been incubated with Na-APR-1 and Na-CP-3. Hb digested with all three proteases was subjected to reverse phase HPLC and peptides were analysed using Liquid Chromatography-Mass Spectrometry (LC-MS). A total of 74 cleavage sites were identified within Hb α and β chains. Na-APR-1 was responsible for cleavage of Hb at the hinge region, probably unravelling the molecule so that Na- CP-3 and Na-MEP-1 could gain access to globin peptides. All three proteases were promiscuous in their subsite specificities, but the most common P1-P1′ residues were hydrophobic and/or bulky in nature, such as Phe, Leu and Ala. Antibodies to all three proteins (Na-APR-1, -CP-3, -MEP-1) immunolocalised to the gut region of the worm, further supporting their roles in Hb degradation. These results suggest that Hb degradation in N. americanus follows a similar pattern to that which has been described in Plasomdium falciparum. Studies conducted in this project have identified a number of potential haemoglobinases and have demonstrated that the gut region of the hookworm contains a multitude of proteases which could be targeted for production of new chemotherapies or as vaccine candidates. Results presented here also suggest that the

v Hb degradation process occurs in an ordered cascade, similar to those which have been reported in other haematophagous parasites. More importantly, it has been confirmed that Na-APR-1 plays a crucial role in the initiation of the Hb degradation process and therefore targeting this molecule as a vaccine candidate could provide high levels of protection against .

vi LIST OF PUBLICATIONS

Publication by candidate relevant to thesis: N. Ranjit, M.K. Jones, D.J Stenzel, R.B Gasser, A. Loukas (2006). A survey of the intestinal transcriptome of the hookworms, Necator americanus and Ancylostoma caninum using tissue isolated by laser microdissection microscopy. International Journal for Parasitology 36: 701-710 (Impact factor: 3.337)

N. Ranjit, B. Zhan, D. Stenzel, J. Mulvenna, R. Fujiwara, P. Hotez, A. Loukas (2008). A family of cathepsin B cysteine proteases expressed in the gut of the human hookworm, Necator americanus. Molecular and Biochemical Parasitology 160: 90-9 (Impact factor: 2.641)

N. Ranjit, B. Zhan, B. Hamilton, D. Stenzel, J Lowther, M. Pearson, J. Gorman, P. Hotez, A. Loukas (2008). Digestion of hemoglobin via an ordered cascade of proteolysis in the intestine of the human hookworm Necator americanus Submitted to Journal of Infectious Diseases (under review) (Impact factor: 6.035)

Additional publication by candidate relevant to the thesis but not forming part of it: A. Loukas, J. M.Bethony, S. Mendez, R. T. Fujiwara, G. N. Goud, N. Ranjit, B. Zhan, K. Jones, M. E. Bottazzi, P. J. Hotez (2005). Vaccination with recombinant aspartic hemoglobinase reduces parasite load and blood loss after hookworm infection in dogs. PLoS Medicine 2 (10): e295 (Impact factor: 13.750)

vii TABLE OF CONTENTS LIST OF KEYWORDS ...... ii ABSTRACT...... iii LIST OF PUBLICATIONS ...... vii TABLE OF CONTENTS...... viii LIST OF FIGURES AND TABLES...... xi LIST OF ABBREVATIONS ...... xiii LIST OF ABBREVATIONS ...... xiii STATEMENT OF ORIGINALITY...... xvi STATEMENT BY SUPERVISOR...... xvi ACKNOWLEDGEMENTS ...... xvii CHAPTER 1: INTRODUCTION ...... 1 1.1 DESCRIPTION OF THE SCIENTIFIC PROBLEM INVESTIGATED ...... 2 1.2 OVERALL OBJECTIVE OF THE STUDY ...... 3 1.3 SPECIFIC AIMS OF THE STUDY ...... 3 1.4 ACCOUNT OF SCIENTIFIC PROGRESS LINKING THE SCIENTIFIC PAPERS ...... 5 CHAPTER 2: LITERATURE REVIEW ...... 7 2.1 HOOKWORMS ...... 8 2.1.1 Introduction...... 8 2.1.2 Biology...... 8 2.1.3 Global Distribution...... 13 2.1.4 Clinical Aspects...... 14 2.1.4.1 Pathogenicity...... 14 2.1.4.2 Immunology ...... 17 2.1.4.3 Treatment ...... 19 2.1.4.4 Vaccines ...... 20 2.2 PROTEASES ...... 21 2.2.1 Hookworm Larval Proteases...... 22 2.2.2 Adult Hookworm Proteases ...... 26 2.2.2.1 Aspartic proteases...... 26 2.2.2.2 Cysteine proteases...... 28 2.2.2.3 Metalloproteases ...... 31 2.2.2.4 Exopeptidase and aminopeptidases...... 33 2.3 HAEMOGLOBIN DIGESTION CASCADE...... 34 2.4 SUMMARY ...... 37 2.5 THESIS HYPOTHESIS...... 38 CHAPTER 3: A SURVERY OF THE INTESTINAL TRANSCRIPTOMES OF THE HOOKWORMS, NECATOR AMERICANUS AND ANCYLOSTOMA CANINUM,USING TISSUES ISOLATED BY LASER MICRODISSECTION MICROSCOPY...... 40 3.1 CONTRIBUTIONS ...... 41 3.2 ABSTRACT...... 42 3.3 INTRODUCTION ...... 43 3.4 MATERIALS AND METHODS...... 44 3.4.1 Parasite material...... 44 3.4.2 Laser microdissection microscopy (LMM)...... 45 3.4.3 RNA extraction, cDNA synthesis and detection of known gut transcripts ...... 45

viii 3.4.4 Construction of cDNA libraries ...... 46 3.4.5 Bioinformatic analyses...... 46 3.4.6 Phylogenetic tree...... 47 3.5 RESULTS AND DISCUSSION ...... 47 3.5.1 Extraction of gut tissues from hookworms ...... 47 3.5.2 Tissue specificity of cDNA populations ...... 48 3.5.3 Characteristics of the EST dataset...... 49 3.5.4 Sequence analysis and gene ontology...... 49 3.5.5 Transcript abundance and highly represented genes...... 53 3.5.6 Molecules involved in feeding...... 54 3.5.7 Immunomodulation...... 58 3.5.8 Known and Potential Vaccine Antigens ...... 59 3.6 CONCLUSION...... 61 CHAPTER 4: A FAMILY OF CATHEPSIN B CYSTEINE PROTEASES EXPRESSED IN THE GUT OF THE HUMAN HOOKWORM, NECATOR AMERICANUS ...... 62 4.1 CONTRIBUTIONS ...... 63 4.2 ABSTRACT...... 64 4.3 INTRODUCTION ...... 65 4.4 MATERIALS AND METHODS...... 67 4.4.1 Phylogenetic tree...... 67 4.4.2 Amplification of cysteine proteases genes from gut cDNA...... 67 4.4.3 Quantitation of cysteine protease gene expression in different developmental stages ...... 68 4.4.4 Expression and purification of recombinant cysteine proteases ...... 68 4.4.5 Autoactivation, catalytic activity assays...... 69 4.4.6 Identification of the pro-mature Na-CP-3 junction...... 71 4.4.7 Antibody production...... 71 4.4.8 Immunolocalization ...... 71 4.5 RESULTS ...... 72 4.5.1 Sequence analysis of the cysteine proteases identified in N. americanus ...... 72 4.5.2 Phylogenetic analysis of cathepsin B-like proteases...... 73 4.5.3 Amplification of cysteine protease mRNAs from N. americanus gut cDNA ...... 73 4.5.4 Developmental expression of cysteine protease genes ...... 73 4.5.5 Expression of recombinant cysteine proteases...... 77 4.5.6 Catalytic activity of Na-CP-3...... 78 4.5.7 Antibody production and immunolocalization of proteins ...... 80 4.6 DISCUSSION ...... 81 CHAPTER 5: DIGESTION OF HEMOGLOBIN VIA AN ORDERED CASCADE OF PROTEOLYSIS IN THE INTESTINE OF THE HUMAN HOOKWORM, NECATOR AMERICANUS ...... 87 5.1 CONTRIBUTIONS ...... 88 5.2 ABSTRACT...... 89 5.3 INTRODUCTION ...... 90 5.4 MATERIALS AND METHODS...... 92 5.4.1 cDNA cloning...... 92 5.4.2 Protein expression and purification...... 92 5.4.3 Catalytic activity of recombinant hemoglobinases ...... 93

ix 5.4.4 Antibody production and immunolocalization...... 94 5.4.5 Proteolysis of Hb by Na-APR-1, Na-CP-3 and Na-MEP-1 ...... 95 5.4.6 LC-MS and MS-MS analysis of Hb hydrolysates ...... 95 5.5 RESULTS ...... 96 5.5.1 Cloning of cDNAs encoding N. americanus hemoglobinases...... 96 5.5.2 Expression and purification of Na-APR-1 and Na-MEP-1...... 98 5.5.3 Catalytic activity assays...... 99 5.5.4 Immunolocalization ...... 99 5.5.5 Hemoglobin degradation and LC-MS analysis of hemoglobin hydrolysates...... 100 5.6 DISCUSSION ...... 105 CHAPTER 6: GENERAL DISCUSSION, CONCLUSION AND FUTURE DIRECTIONS...... 109 6.1 GENERAL DISCUSSION...... 110 6.2 CONCLUSION AND FUTURE DIRECTIONS...... 122 REFERENCES...... 124 APPENDICES ...... 138

x LIST OF FIGURES AND TABLES Figure 2.1. Scanning electron microgrpahs of the buccal cavities of human hookworm species...... 9 Figure 2.2. Lifecycle of N. americanus ...... 11 Figure 2.3. Micrograph showing sectioned adult hookworm attached to intestinal microvilli...... 11 Figure 2.4. Micrographs of N. americanus adult and egg...... 12 Figure 2.5. Global distribution of human hookworm infection (both N. americanus and A. duodenale)...... 14 Figure 2.6. Relationship between hookworm burden and anaemia and amount of blood loss with different hookworm loads...... 17 Figure 2.7. Comparison of hookworm burden to other soil transmitted helminths. ....19 Figure 2.8. Percent reduction of A. caninum L3 that penetrated skin in an in vitro model of tissue migration by hookworm larvae...... 24 Figure 2.9. Schematic of the effects of anti-ASP-2 antibodies on host hookworm burdens...... 26 Figure 2.10. Schematic diagram of the cysteine protease superfamily from parasitic helminths...... 29 Figure 2.11. Families of zinc metalloproteases...... 32 Figure 2.12. Tertiary structure of the haemoglobin molecule and the two subunits....34 Figure 2.13. Proposed haemoglobin degradation pathway in P. falciparum and S. mansoni...... 36 Figure 2.14. Schematic of proposed haemoglobinase cascade in the intestine of blood-feeding ...... 37 Figure 2.15 Schematic for the development of a bivalent human .38 Figure 3.1. Light micrographs of adult hookworm section before and after laser microdissection...... 48 Figure 3.2. Detection in gut but not ovary cDNA of transcripts corresponding to proteins that were previously shown to be expressed in the intestine using immunolocalisation...... 49 Table 3.1. Summary of the gut EST datasets for A. caninum and N. americanus contigs with ORFs containing ≥ 30 amino acids...... 51 Table 3.2. Novel clones with no homologues in any datasets that contain ORFs with predicted signal peptides. Double-ended arrows denote the predicted cleavage site of the signal peptide...... 52 Figure 3.3. Pie charts depicting gene ontology classifications of Ancylostoma caninum and Necator americanus gut ESTs identified in this study...... 53 Table 3.3. The 10 most abundant contigs from the A. caninum and N. americanus gut expressed sequence tag (EST) datasets...... 55 Table 3.4. Hookoworm (Anyclostoma spp. plus Haemonchus and Caenorhabditis, of comparison) gut ESTs encoding proteolytic enzymes...... 57 Figure 3.4. Multiple sequence alignment of contig Ac129 with homologous members of the C-type lectin family...... 59 Figure 3.5. Neighbour joining phylogenetic tree showing the relationships of Ac173 and Na91 with other members of the Activation Associated Secretory Protein family...... 60 Table 3.5. Contigs identified in this study with potential as vaccine antigens...... 61 Table 4.1. General properties of N. americanus cathepsin B-like proteases ...... 73

xi Figure 4.1 Multiple sequence alignment of N. americanus cysteine proteases and human cathepsin B...... 75 Figure 4.2. Neighbour joining phylogenetic tree depicting the relationships of N. americanus cysteine proteases with homologues from other nematodes and other phyla...... 76 Figure 4.3. Amplification of cysteine protease mRNAs from N. americanus gut cDNA...... 77 Figure 4.4. Developmental expression profiles of N. americanus cysteine protease mRNAs...... 77 Figure 4.5. Expression and purification of recombinant N. americanus CatBs in yeast P. pastoris and E. coli...... 78 Figure 4.6. Gelatin zymogram showing catalytic activity of purified recombinant Na- CP-3...... 79 Figure 4.7. SDS-PAGE gel of purified recombinant pro-Na-CP-3 incubated with Pro- Q Emerald 300 glycoprotein stain (A). Activation of pro-Na-CP-3 (B)...... 80 Figure 4.8. pH profile of the catalytic activity of recombinant Na-CP-3 after auto- processing...... 80 Figure 4.9. Western blot showing recognition of recombinant Na-CP-2, CP-3, CP-4 and CP-5...... 81 Figure 4.10. Immunolocalization of Na-CP-2, -3, -4 and -5 in transverse sections of adult N. americanus...... 82 Figure 5.1. Multiple sequence alignment of Na-MEP-1 with Ac-MEP-1 from Ancylostoma caninum and human neprilysin 1...... 97 Figure 5.2. Expression and purification of N. americanus hemoglobinases...... 98 Figure 5.3. Catalytic activity of recombinant Na-APR-1 expressed in yeast and Na- MEP-1 expressed in insect cells...... 99 Figure 5.4. Immunolocalization of Na-APR-1, Na-CP-3 and Na-MEP-1...... 100 Figure 5.5. Hemoglobin digestion with recombinant Na-APR-1, Na-CP-3 and Na- MEP-1...... 101 Figure 5.6. LC trace of hemoglobin incubated with various recombinant proteins for 18 hours at 37oC at pH 4.5 ...... 102 Figure 5.7. Map of hemoglobin α and β chains highlighting the cleavages made by N. americanus recombinant haemoglobinases...... 103 Figure 5.8. P4-P4′ subsite specificities of Na-APR-1, Na-CP-3 and Na-MEP-1...... 104

xii LIST OF ABBREVATIONS α alpha aa amino acid Asn Asparagine APR aspartic protease ASP activation associated secreted protein β beta BLAST Basic Local Alignment Search Tool bp base pair BSA bovine serum albumin CAP cap analysis program cDNA complementary deoxyribonucleic acid CP cysteine protease CRD carbohydrate recognition domain C-TL C-type lectin Cys Cysteine Da Dalton DALY disability-adjusted life year DEPC diethyl pyrocarbonate DNA deoxyribonucleic acid dNTP dideoxyribonucleoside triphosphates E-64 trans-Epoxysuccinyl-L-leucylamido-(4 guanidino) butane EDTA ethylenediaminetetraacetic acid ELISA enzyme-linked immunosorbent assay epg eggs per gram ES excretory/secretory EST expressed sequence tag FPLC fast protein liquid chromatography g gram GAG glycosaminoglycans Gln Glutamine Glu Glutamatic acid GST Glutathione S-tranferase

xiii h hour(s) HAP histoaspartic protease Hb haemoglobin HCl hydrochloric acid Hi5 Trichoplusia ni His Histidine HPLC high performance liquid chromatography Igs immunoglobulins IL interleukin IPTG isopropyl β-D-thiogalactopyranoside kb kilobase kcat catalytic constant kDa kilo dalton Km Michaelis constant L litre L3 third-stage larvae LB Luria-Bertani media LC liquid chromatography M molar MEP metalloprotease mg milligram min minute(s) ml millilitre mM millimolar MQ Milli-Q-purified water mRNA messenger ribonucleic acid MS mass spectrophotometry MW molecular weight ng nanograms NIF neutrophil inhibitory factor nm nanometres NMS normal mouse serum nr non redundant nt nucleotide

xiv ORF open reading frame PBS phosphate buffered saline PBS/T phosphate buffered saline/Tween 20 PCR polymerase chain reaction PM plasmepsin PRP pathogenesis-related protein PSP polysaccharides RBC red blood cell RNA ribonucleic acid RNAi RNA interference RP-HPLC reverse-phase high-performance liquid chromatography RT room temperature RT-PCR reverse transcription PCR SDS sodium dodecyl sulfate SDS-PAGE SDS-polyacrylamide gel electrophoresis TFA trifluoroacetic acid TSBP thiol sepharose binding protein μg microgram μl microlitre μM micromolar μm micrometer WHO World Health Organization YPD yeast peptone dextrose

xv STATEMENT OF ORIGINALITY

The work contained in this thesis has not been previously submitted to meet requirements for an award at this or any other higher education institution. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person except where due reference is made.

Najju Ranjit 2008

STATEMENT BY SUPERVISOR

All co-authors have provided their consent for the inclusion of the papers presented in this thesis. The co-authors accept the student’s contribution to each manuscript and description of co-authors’ contribution.

Alex Loukas 2008

xvi ACKNOWLEDGEMENTS

First and foremost I would like to sincerely thank my principal supervisor Dr Alex Loukas for all his help, support and guidance throughout my PhD. This project would not have been possible without all his input. Thank you for allowing me to join your group and letting me work on this project, I am extremely grateful for having been given this opportunity to work in such a great lab.

Thanks to my associate supervisor Dr Deb Stenzel, who made sure that all the correct forms were filled and took time out of her busy schedule to make sure that my thesis was up to scratch. Thank you for your words of encouragement.

Many thanks to the past and present members of the Loukas lab who helped me enormously throughout the years. Special thanks to Mai Tran, Ben Datu, Soraya Gaze and Leanne Cooper who were always there to lend a helping hand and gave me plenty of great advice. Thank you for making my lab life so much more enjoyable.

To all my friends, thank you for being there for me and listening to me whenever I needed to vent or escape from the lab. To my PhD buddies, Meru, Melina and Louise thanks for keeping me company and always bringing a smile to my face. To my BoP buddies, you guys are the best bunch of people I have ever worked with and I hope we can all meet up again one day.

To everyone else who I met along this journey and who provided me with valuable guidance and assistance, thank you!

And last but not least a great big thanks to my family, especially my parents to whom I will forever be indebted to. Thank you for all your love and support throughout the years. Both of you are my source of inspiration and I dedicate this thesis to you.

xvii CHAPTER 1: INTRODUCTION

1.1 DESCRIPTION OF THE SCIENTIFIC PROBLEM INVESTIGATED Hookworms are parasitic nematodes that reside in the upper region of the of their mammalian hosts, where they attach to the mucosa and ingest extravasated blood from ruptured intestinal capillaries and arterioles. In many developing countries, hookworm infections are widespread, due to poor conditions and a lack of comprehensive chemotherapy control programs (Hotez, 2007). The two major species infecting people are Necator americanus and , the former being more prevalent and widespread. The main pathology associated with hookworm infection is iron deficiency anaemia, which is a direct result of intestinal blood loss that occurs in heavy infections. Although the overall prevalence and intensity of hookworm infections are higher in males than in females, in part because males have greater occupational exposure to infection, iron deficiency anaemia causes more adverse effects in young children and women of child bearing age, due to their low iron stores (Brooker et al., 2004). One of the main problems associated with the available chemotherapy options for hookworms is the high rate of re-infection after treatment. Although drugs used for treatment are highly effective in eliminating existing infections, they do not protect against rapid re-infection, which is a common occurrence in many endemic communities. Another concern of mass drug administration programs is the potential risk of drug resistance occurring, as is the case with parasitic nematodes of sheep and cattle where benzimidazole drug resistance is now well documented (Jabbar et al., 2006). As yet there is no solid evidence of drug resistance in human helminths but there have been reports of drug failure and diminished efficacy with treatments in a number of African countries (Albonico et al., 2003). Also in stark contrast to infection with other soil transmitted helminths, immunity against hookworms does not develop in most people - in fact the oldest people living in an endemic community sometimes have the heaviest worm burdens (Loukas et al., 2006). A highly desirable goal to combat hookworm infection would be the production of a prophylactic vaccine, one which reduces worm burden and leads to the decrease of intestinal blood loss, the main cause of the pathology associated with this infection. It has been suggested that a hookworm vaccine could be integrated into current chemotherapy control programs, with chemotherapy given first to treat

2 existing infections, followed by a vaccine administered to prevent or significantly delay reinfection (Bethony et al., 2005). Using this approach, vaccine-linked chemotherapy would not only diminish the requirement for frequent and periodic anthelmintic chemotherapy, but would also reduce the likelihood of the emergence of drug resistance.

1.2 OVERALL OBJECTIVE OF THE STUDY As hookworms are complex multicellular parasites, it has been suggested that an efficacious vaccine against this infection should consist of two antigens, one which targets the infective larval stage and inhibits migration through the host, and the second which targets defined physiological functions in the adult stage, such as blood feeding. An from the third-stage larvae, Na-ASP-2 (Bethony et al., 2005, Goud et al., 2005) has already been identified and has completed phase one clinical trials (Bethony et al., 2008). Therefore, the main objective of this study was to identify and investigate mRNAs encoding proteases from the gut of adult N. americanus and determine which of these enzymes are involved in haemoglobin (Hb) degradation, and could therefore be considered haemoglobinases. By including haemoglobinases in a cocktail hookworm vaccine, it is envisaged that the parasite’s ability to digest and therefore absorb nutrients would be compromised, therefore dramatically decreasing the viability of the parasite.

1.3 SPECIFIC AIMS OF THE STUDY There were three specific aims in this study. The first aim was to identify mRNAs encoding potential vaccine antigens, particularly haemoglobinases, from the intestinal cells of hookworms. The second aim was to express recombinant versions of potential haemoglobinases in catalytically active form. The third aim was to assess the abilities of these recombinant proteases to participate in a multi-enzyme cascade of haemoglobinolysis and determine whether this occurs via an ordered pathway. Finally, using tandem mass spectrometry, a Hb cleavage map was developed, allowing inferences to be made on the subsite specificities of each of the recombinant haemoglobinases. In order to address the aims of this study the following experiments were performed.

3 • Aim 1: Identification of the cDNAs encoding the major haemoglobinolytic proteases from the intestine of adult N. americanus and the common dog hookworm, Ancylostoma caninum. N. americanus and A. caninum gut tissues were dissected using laser microdissection microscopy in order to prepare gut specific mRNA for cDNA library construction and shot gun sequencing. The PALM microbeam microdissector plus laser catapult were used to dissect intestinal tissue from adult hookworms. RNA was isolated from the extracted tissue for subsequent cDNA synthesis followed by construction of gut cDNA plasmid libraries. Four hundred and eighty expressed sequence tags (ESTs) were generated from each library and these were subjected to extensive bioinformatics analyses including filtering, clustering, gene ontology analyses and characterisation of mRNAs encoding for proteases in particular. • Aim 2: Expression and characterisation of potential haemoglobinases. Proteases expressed by N. americanus that are orthologous to known/potential haemoglobinases from other haematophagous parasites were selected, expressed in recombinant form and purified using various expression systems. Recombinant cysteine proteases were expressed in the yeast Pichia pastoris and bacterium Escherichia coli. Recombinant aspartic protease was expressed in P. pastoris. Metalloproteases were expressed in Trichoplusia ni Hi5 insect cells using the baculovirus expression system. All recombinant proteins were purified via affinity chromatography using the C-terminal hexa-histidine tags encoded by the expression vectors. Polyclonal antibodies were generated in mice against N. americanus proteases and these were used to localize the anatomic sites of expression in tissue sections of adult N. americanus. Catalytic assays were conducted with purified recombinant proteases using either zymogram gels or class-specific fluorogenic peptides. • Aim 3: Assessment of the ability of recombinant proteases to digest Hb or globin peptides, development a cleavage map of human Hb and determine whether digestion occurs in an ordered pathway. Necator americanus haemoglobinases (both alone and in various combinations) were incubated with human Hb in vitro and assessed using SDS-PAGE and liquid chromatography-mass spectrometry (LC-MS).

4 1.4 ACCOUNT OF SCIENTIFIC PROGRESS LINKING THE SCIENTIFIC PAPERS It has been demonstrated in the blood-feeding of ruminants, Haemonchus contortus, that the intestine contains a myriad of hidden antigens, many of which are presumed to be involved in the feeding process and are therefore good targets for vaccine candidates. Following this concept the first paper presented in this thesis (A survey of the intestinal transcriptomes of the hookworms, Necator americanus and Ancylostoma caninum, using tissue isolated by laser microdissection microscopy) conducts a survey of the intestinal transcriptomes of the human and canine hookworms, N. americanus and A. caninum. This provided a snapshot of the genes that are highly expressed in this tissue and enabled me to identify intestinal proteases and assess their potential roles as haemoglobinases. At least eleven potential haemoglobinases were detected from both the N. americanus and A. caninum libraries, with cysteine and metalloproteases being the most abundantly represented. The second paper presented in this thesis (A family of cathepsin B cysteine proteases expressed in the gut of the human hookworm, Necator americanus) describes cysteine proteases (Na-CP-2, -3, -4 and -5) which were initially amplified from a third-stage larval N. americanus cDNA phage library by colleagues from George Washington University, USA. Gut cDNA (generated for paper #1) was used to verify that these cysteine proteases were expressed in the intestinal tissue of the adult worm, implying potential roles in blood-feeding. Immunolocalisation results supported mRNA amplification data, and specific antibodies to each recombinant cysteine protease bound to the gut microvillar surface, further verifying their involvement in digestion of the blood meal. Recombinant Na-CP-3 was successfully expressed in soluble form as a pro-enzyme and underwent auto-activation to a mature protease in the presence of dextran sulfate. Catalytic activity was detected using the fluorogenic peptide Z-Phe-Arg-aminomethylcoumarin, however activated Na-CP-3 was incapable of digesting intact Hb. Nonetheless, it was suggested that Na-CP-3 might have a role in nutrient acquisition by cleaving globin fragments after other haemoglobinases had made initial cleavages of the Hb tetramer. In the third paper (Digestion of hemoglobin via an ordered cascade of proteolysis in the intestine of the human hookworm Necator americanus), an

5 ordered pathway of Hb digestion in N. americanus was determined. The aspartic protease, Na-APR-1, was shown to cleave intact Hb. LC-MS was used to identify the sites where APR-1 cleaved Hb - one of these sites was the hinge region of Hb, implying that APR-1 unravels the tetrameric Hb protein, making it more susceptible to further cleavage by other proteases. The cysteine protease, Na-CP-3, was shown to cleave globin fragments after Hb had been digested with Na-APR-1, suggesting that cleavage of Hb is initiated by the aspartic protease, and then cysteine proteases (CP-3 at least) act to further digest the globin fragments. In similar fashion to Na-CP-3, the metalloprotease, Na-MEP-1, was also incapable of cleaving intact Hb but was able to further cleave globin fragments following incubation of Hb with Na-APR-1 and Na- CP-3. This study demonstrated that Hb digestion in N. americanus occurs in an ordered cascade, thus validating the targeting of the proteases involved in this process for vaccine development.

6 CHAPTER 2: LITERATURE REVIEW

2.1 HOOKWORMS 2.1.1 Introduction More than a dozen different species of soil-transmitted helminths infect humans in developing countries. Of these parasites, the two hookworm species Necator americanus and Ancylostoma duodenale, stand out because of their widespread prevalence and distribution, together contributing to hundreds of millions of infections (Hotez et al., 2003c). These parasites have evolved complex life histories with each stage expressing unique genes and gene families to promote its survival in distinct niches. Helminth parasites utilise a large and diverse range of proteases in order to infect, feed and reproduce in the host (Tort et al., 1999). Adult hookworms are voracious blood feeders and rely on the acquisition of proteins found in blood in order to meet their nutritional requirements. As with a number of other haematophagous parasites, it has been suggested that hookworms employ a cascade of proteolytic enzymes to break down proteins such as haemoglobin and serum albumin to release amino acids needed for protein synthesis (Williamson et al., 2003b). Unlike many mammalian acidic proteases which act in lysosomes, homologous enzymes in helminths are released into the gut lumen (or outside of the parasite altogether) where they digest their substrates extracellularly (Tort et al., 1999). As such, these enzymes are accessible to antibodies when the parasite takes a blood meal, making them viable targets for new therapies against hookworms and other blood-feeding worms.

2.1.2 Biology Hookworms are dioecious parasitic nematodes, with adult forms that generally reside in the intestinal tract of their host (Roche and Layrisse, 1966). They belong to the family and are part of the superfamily Strongyloidea. The two main genera that affect humans, Necator and Ancylostoma, are characterised by the presence of either blunt cutting plates or sharp “teeth” that line the adult parasite buccal capsule (Fig. 2.1). Necator americanus and Ancylostoma duodenale are responsible for most human infections with hookworms, although infection with the feline and canine hookworm Ancylostoma ceylancium also occurs in parts of Asia as a consequence of zoonotic (Hotez and Pritchard, 1995). Ancylostoma ceylancium does not reach maturity in human hosts

8 and hence it is not associated with the substantial blood loss, seen in humans infected with the other two hookworm species (Carroll and Grove, 1986). In north-eastern Australia, the canine hookworm, A. caninum, has been shown to cause occasional human intestinal infections, sometimes resulting in eosinophilic gastroenteritis (Prociv and Croese, 1990). Another canine and feline hookworm () can also infect humans and is the main cause of cutaneous migrans, a self-limiting dermatologic condition characterized by serpiginous burrows of 1 to 5 cm in length (Brooker et al., 2004, Hotez et al., 2004).

Figure 2.1. Scanning electron microgrpahs of the buccal cavities of human hookworm species Left to right: Necator americanus and Ancylostoma duodenale. N. americanus has two cutting plates along the anterior margin while A. duodenale has two pairs of teeth. (http://www. mercksource.com and http://www.nematode.net)

There are significant pathobiological differences between the two major human hookworms: N. americanus is smaller than A. duodenale, produces fewer eggs, and causes less blood loss from the host (Albonico et al., 1998). N. americanus is considered by some researchers to be better adapted to human because of its diminished virulence relative to A. duodenale (Brooker et al., 2004). It is also believed that N. americanus may be more adept at immune evasion (Pritchard and Brown, 2001). In contrast, A. duodenale is considered to be the more “opportunistic species” because of its ability to survive in more extreme environmental conditions, its ability to cause infections via the oral route and its greater fecundity (Hotez et al., 2003a). Ancylostoma duodenale L3 also have the unique ability to undergo arrested development in humans and may enter human mammary glands prior to lactogenic transmission (Yu et al., 1995). The life cycle of hookworms is direct. Infection of humans is by the infective third larval stage known as the L3. Generally, hookworms infect a host by larvae penetrating the skin, although A. duodenale is also orally infective (Brooker et al.,

9 2004). Necator americanus is an obligate skin-penetrating parasite, and larvae of this species cannot establish in the intestine when inoculated orally unless they first penetrate the oral mucosa to enter the blood-stream and then migrate in the same fashion as skin-penetrating larvae. The L3 attach to host skin on contact and penetrate through hair follicles, into the dermis, where they enter blood or lymphatic capillaries (Fig. 2.2). Skin penetration by L3 is facilitated by secretion of proteolytic enzymes that degrade macromolecules of the host tissue (Vetter and Leegwater-vd Linden, 1977, Williamson et al., 2006). It has been speculated that migration through the skin is required to reactivate developmental pathways and trigger expression of genes of the parasitic stage (Hawdon and Hotez, 1996). Following host entry, the L3 receive a signal present in mammalian serum and tissue that causes them to continue development (Hawdon and Datu, 2003, Hawdon and Hotez, 1996). The host- activated L3 then migrate through the vasculature and are swept via the afferent circulation to the heart and then the pulmonary vasculature (Fig. 2.2). The larvae then undergo tracheal migration by penetrating into the alveoli, to be coughed up into the airways and then swallowed into the gut. After the larvae enter the gastrointestinal tract they moult twice and differentiate into male and female adults. Approximately 5-8 weeks pass from the time L3 larvae penetrate human skin until the adult worms reach sexual maturity and mate (Hotez et al., 2004). The dioecious adult hookworms live in the upper small intestine where they attach to the mucosa via teeth (Ancylostoma species) or cutting plates (N. americanus) (Fig. 2.1), by sucking clumps of villi into their buccal capsules, effectively burying their anterior ends into the host intestinal mucosa or deeper (Loukas et al., 2005b) (Fig. 2.3).

10

Figure 2.2. Lifecycle of N. americanus Eggs are passed in the stool , and under favorable conditions (moisture, warmth, shade), larvae hatch in 1 to 2 days. The released rhabditiform larvae grow in the feces and/or the soil , and after 5 to 10 days (and two molts) they become become filariform (third-stage) larvae that are infective . On contact with the human host, the larvae penetrate the skin and are carried through the veins to the heart and then to the lungs. They penetrate into the pulmonary alveoli, ascend the bronchial tree to the pharynx, and are swallowed . The larvae reach the small intestine, where they reside and mature into adults. Adult worms live in the lumen of the small intestine, where they attach to the intestinal wall with resultant blood loss by the host. (http://www.dpd.cdc.gov/dpdx/HTML/Hookworm.htm)

Figure 2.3. Micrograph showing sectioned adult hookworm attached to intestinal microvilli The muscular pharynx is used to suck host tissue into the alimentary canal which is under hydrostatic pressure. (Loukas et al., 2005b)

11 The adult worms feed on host blood, and using radioactive tracers it has been estimated that approximately 30 μl of blood/day is lost to a single N. americanus and 260 μl to A. duodenale (Pritchard et al., 1991). Ingestion of blood by the worm, and associated intestinal blood loss from the host, begins just prior to egg production and continues for the life of the hookworm. The large, anterior cephalic glands of the adult hookworms secrete various products, including proteases and anti-coagulant peptides, into the worm’s buccal cavity and oesophagus, as well as into the attachment site in the host intestine to “predigest” host mucosal tissues (Loukas et al., 2005b). Adult hookworms are thought to rely mostly on blood for nutrition, and it has recently been shown that intact erythrocytes are ingested and lysed in the parasite’s intestine by haemolytic proteins (Don et al., 2004). Forty-five to sixty days after penetration of the host, female hookworms begin laying eggs (9,000-11,000/day). The eggs, measuring approximately 60 μm x 40 μm, are transparent, thin-shelled and ovoid, with blunt, rounded ends (Fig. 2.4). Eggs are passed in the host faeces, and embryonate in moist soil (25-28oC) before developing to L1 stage larvae (Bethony et al., 2006a, Brooker et al., 2004).

Figure 2.4. Micrographs of N. americanus adult and egg. Left to right: female adult worm, male adult worm and egg. Female worms are approximately 9 to 11 mm, male worms are approx. 7 to 9 mm and eggs measure approx. 60 μm x 40 μm. The posterior end of the male worm is equipped with a characteristic copulatory bursa (arrow). The eggs are transparent, thin shelled and ovoid with blunt rounded ends. (Bethony et al., 2006a)

The hatched, first-stage larvae feed on organic debris and bacteria in the soil. The larvae undergo two moults to become the infective third-stage larvae, which are enveloped in the loose outer cuticular sheath left over from the second moult and are approximately 600 μM in length (Bethony et al., 2006a, Brooker et al., 2004). At this stage, the larvae may undergo developmental arrest and can live in the soil for weeks if there is appropriate warmth, shade and moisture. The non-feeding, infective L3 migrate to the soil surface or low vegetation to maximize their chances of contacting

12 a new host (Hotez et al., 2004). The L3, in response to either an increase in temperature or other undetermined cues on contact with the host, emerge from developmental arrest, quickly shed their protective sheath and penetrate the host skin (Fig 2.2).

2.1.3 Global Distribution Hookworm infections predominantly occur in tropical and sub-tropical areas and are most prevalent in regions where humidity is high and there is a lack of proper sanitation (Bethony et al., 2006a). A major problem in these areas is the continued use of human faeces as a crop fertilizer, which provides ideal conditions for the perpetuation of the hookworm lifecycle. Adequate soil moisture and temperature are the primary environmental requirements for maintaining endemic infections (Brooker et al., 2004). Eggs that are shed in faeces can remain viable in moist soil provided with the necessary environmental conditions for development. The current global hookworm prevalence is shown in Figure 2.5. Hookworm infections are particularly prevalent throughout much of sub-Saharan Africa, China, Southeast Asia and the Pacific. Worldwide, N. americanus is the predominant agent of human hookworm infection, while A. duodenale occurs in more scattered focal environments where N. americanus cannot survive e.g. in colder and drier areas (Hotez et al., 2004). Necator americanus is widely spread throughout the Caribbean and Latin America and is a major pathogen in sub-Saharan Africa, Southeast Asia, the Indian subcontinent and the Pacific islands. Coastal areas of these regions are especially associated with high Necator transmission (Brooker et al., 2004). The predominant regions for A. duodenale include northerly latitudes of south and west China and India. Ancylostoma duodenale may survive in these harsher climates because of its ability to undergo arrested development in host tissues (Schad et al., 1973). Throughout the world, mixed infections with both the major hookworm species are common. It is estimated that nearly 1 billion people world wide harbour these parasites, with recent estimates indicating that there are over 500 million known clinically significant cases (Bethony et al., 2006a). Of these 500 million people, approximately 44 million are pregnant women (Bundy et al., 1995). In sub- Saharan Africa alone there are close to 200 million people infected with hookworms (Crompton, 2000). Not surprisingly, and like many other neglected tropical diseases,

13 there is a striking global relationship between hookworm prevalence and low socioeconomic status (de Silva et al., 2003).

Figure 2.5. Global distribution of human hookworm infection (both N. americanus and A. duodenale). The highest prevalence of hookworm occurs in sub-Saharan Africa and eastern Asia. High transmission also occurs in other areas of rural poverty in the tropics and southern China. (Hotez et al., 2005)

2.1.4 Clinical Aspects 2.1.4.1 Pathogenicity The morbidity associated with hookworm infection is varied and ranges from mild, transient clinical signs and symptoms to severe clinical disease. Moreover, chronic infections are associated with effects on the physical growth, cognition and productivity of individuals (Brooker et al., 2004). The overall impact of a hookworm infection ultimately depends on the underlying health status of the patient (Brooker et al., 2004). Although the adult hookworm elicits most of the pathological effects of hookworm disease, the infective larvae may also contribute to disease during host entry: the larvae release immunogenic and bioactive macromolecules, including allergens and tissue invasive enzymes (Brown et al., 1999, Zhan et al., 2002). In areas of high transmission, repeated L3 entry through the skin can result in a cutaneous syndrome known as ground itch (Hotez et al., 2004). This comprises a pruritic erythematous rash and appears mostly on the hands and feet where the parasite penetrates the host. Zoonotic infection with A. braziliense L3 results in which is characterised by serpiginous burrows appearing

14 most frequently on the feet, buttocks and abdomen (Blackwell and Vega-Lopez, 2001). Following entry into the human host, L3 undergo pulmonary migration which can be accompanied by cough, sore throat and fever (Hotez et al., 2004). This usually resolves when the L3 leave the lungs for the gastrointestinal tract. When A. duodenale infection occurs via the oral route, the L3 migration can sometimes produce a syndrome known as Wakana disease which is characterised by nausea, vomiting, pharyngeal irritation and cough (Hotez et al., 2004). The main pathogenic effect of hookworm infection is linked to the intestinal blood loss that occurs during adult worm attachment and feeding in the host small intestine (Hotez et al., 2004). Hookworms induce blood loss directly through mechanical rupture of host capillaries and arterioles, followed by the release of a battery of pharmacologically active polypeptides such as anticoagulants, anti-platelet agents and antioxidants (Furmidge et al., 1996). In many developing countries, where nutrition is inadequate, infection with these parasites is a leading cause of iron deficiency anaemia, which results from the chronic blood loss associated with the feeding of multiple adult worms (Loukas et al., 2006). In heavy infections, the blood loss brought about by feeding activity can cause shortness of breath, lassitude and angina, which can lead to congestive heart failure and, in the most severe cases, death (Brooker et al., 2004). Children are especially vulnerable to hookworm infection, and the effects of blood loss are exacerbated in children with nutrient deprivation. Hookworm-induced iron deficiency anaemia has been shown to cause developmental and mental retardation in children, and adverse maternal-foetal outcomes in pregnant women (Crompton, 2000). In many developing countries, anaemia in pregnancy is an important contributor to maternal mortality, especially around the time of delivery. Severe maternal anaemia is also associated with reduced birth weight which, in turn, is an important risk factor for infant mortality (Brooker et al., 2004). Morbidity is highest among patients that harbour large numbers of adult parasites (Hotez et al., 2004). As hookworms do not replicate in the human host, the number of adult worms present in the intestine of an infected person is directly related to the level of exposure to infective larvae in the environment. Estimates of the intensity of hookworm infections are typically obtained by using quantitative faecal egg counts as a surrogate marker for worm burden (Hotez et al., 2005). The World Health Organization (WHO) defines moderate-intensity infections as those

15 with 2,000-3,999 eggs per gram of faeces (epg) and heavy intensity as those with ≥4,000 epg (Hotez et al., 2005). Hookworm-induced blood loss is estimated to be as high as 9 mL of blood per day in heavy infections and hookworm burdens of 40–160 worms are usually sufficient to cause anaemia (Fig. 2.6). Even light hookworm infections (~300 epg) can contribute significantly to low haemoglobin and serum ferritin levels in nutritionally-compromised hosts (Hotez et al., 2004). Because hookworm infections cause more disability than death, the burden of disease is typically assessed by using a metric known as the DALY (disability- adjusted life year, i.e., the numbers of life years lost from premature death or disability) (Bethony et al., 2006a). For hookworm infection, DALYs are determined mainly on the basis of disability weights assigned to anaemia and cognition, with estimates varying widely depending on the level of severity assigned to each component (Hotez et al., 2003c). According to the 2002 global burden of disease study conducted by WHO, hookworms caused the loss of 1.8 million DALYs worldwide (Bethony et al., 2006a). However, estimates of the number of people infected do not translate into the disease burden caused by hookworm, because not everyone infected will develop clinically significant symptoms - morbidity is typically related to the intensity of infection. Therefore, there is a need to revise estimates of DALYs by combining data from attributable burdens of anaemia, retarded childhood development, and pregnancy related morbidities resulting from hookworm infection (Loukas et al., 2006).

16 A Hookworm infection Eggs per gram of Mean blood loss faeces (mL/day) (SD) Negative 0 1.24 (1.85) Light 1-999 1.46 (1.07) Moderate 1000-4999 2.96 (3.03) Heavy >5000 8.79 (1.10) B

Figure 2.6. Relationship between hookworm burden and anaemia and amount of blood loss with different hookworm loads. Quantitative egg counts serve as an indirect measure of the adult-hookworm burden. The heavier the infection, the higher the blood loss, thus haemoglobin levels drop in proportion to infection. (A) (Hotez et al., 2004) (B) Adapted from (Loukas et al., 2006)

2.1.4.2 Immunology The most studied aspect of the human immune response to hookworm infection is antibody levels to crude larval and adult soluble extracts or adult excretory/secretory (ES) products (Loukas and Prociv, 2001). As with most helminths, the antibody response to hookworm consists predominantly of the Th2 antibody isotypes IgG1, IgG4 and IgE as well as the production of Th2 cytokines, interleukin (IL)-4, -5, -9, -10, and -13 (Brooker et al., 2004). Adult hookworms also induce the production of secretory IgE, IgG and IgM, but not IgA, and the levels of these immunoglobulins (Igs) return to normal after successful anthelmintic treatment (Brooker et al., 2004). Despite the extensive antibody response to infection, there is limited evidence that these antibodies offer any protection by significantly reducing either larval or adult hookworm numbers. Another hallmark feature of the immune response to helminth infection is peripheral blood eosinophilia (Loukas and Prociv, 2001). also predominate in the inflammatory response to hookworm L3 in tissues (Maxwell et al., 1987). Hookworms appear to induce less intestinal inflammation than most other intestinal nematodes, perhaps reflecting their attachment and feeding strategies. Eosinophilia, mastocytosis and IgE stimulation are the three main immune alterations observed during a hookworm infection in humans (Brooker et al., 2004).

17 Hookworms excrete/secrete a myriad of products with the potential for immunomodulation (Loukas et al., 2005b). In the case of several other nematodes, injection of excretory/secretory products alone was shown to induce immune responses similar to those observed during infection with live parasites in laboratory (Allen and MacDonald, 1998). Characterisation of the composition of hookworm and other nematode ES products have demonstrated the presence of many different types of proteins, including proteases, protease inhibitors, C-type lectins, anti-oxidants and anti-inflammatory proteins that might contribute to the cellular hypo-responsiveness that is characteristic of chronic hookworm infections (Loukas and Prociv, 2001). Other secreted proteins from hookworms have also been shown to modulate the immune response, at least in vitro. An anti-inflammatory polypeptide, termed neutrophil inhibitory factor (NIF), that binds to the Mac-1 ligand has been identified in A. caninum (Moyle et al., 1994). This molecule was demonstrated to inhibit neutrophil adhesion to endothelial cells through blocking of the CD11/CD18 integrin on the neutrophil surface, as well as to inhibit peroxide release from active neutrophils (Moyle et al., 1994). NIF belongs to the pathogenesis-related protein (PRP) superfamily, a class of cysteine rich proteins that is abundantly expressed by all parasitic nematodes investigated to date (Datu et al., 2008). Published data suggest that the PRPs play diverse roles in nematode parasitism, perhaps via ligand-receptor interactions courtesy of a large “binding” groove that appears to accommodate peptide ligands (Asojo et al., 2005) Unlike most other human helminthiases, such as and , host resistance to infection and reinfection (after treatment) with N. americanus and A. duodenale is not clear cut (reviewed in (Loukas et al., 2005b)). In fact, a positive correlation between intensity of infection with N. americanus and age has been demonstrated in human populations (Fig. 2.7) (Bethony et al., 2002), which is in stark contrast to canine hookworm infections where age and exposure related immunity occurs (Loukas and Prociv, 2001). While human hookworm infections exhibit the hallmark features of Th2 responses, these immune responses clearly fail to protect most infected people. The reason of the observed failure of Th2 cells to mount effective anti-hookworm response remains unknown (Loukas et al., 2005b).

18

Figure 2.7. Comparison of hookworm burden to other soil transmitted helminths. The hookworm burden increases with age, in contrast to the burden of other soil-transmitted helminths (e.g. Ascaris lumbricoides and Trichuris trichiura), which is highest in childhood. (Hotez et al., 2004)

2.1.4.3 Treatment Because of the high transmission potential, hookworm infections have proven to be extremely difficult to eliminate or eradicate in areas of poverty and poor sanitation (Hotez, 2007). However, hookworm disease is easily treatable: oral doses of the anthelmintic drug, can significantly reduce or completely cure infection (Bethony et al., 2006a). Unfortunately, most hookworm infections occur in developing and economically poor countries, where access to these medications is limited or non-existent (Hotez, 2007). Additionally, anthelmintics do not provide lasting protection, and there is no naturally acquired immunity (in most people) to hookworms. Reinfection occurs rapidly, particularly in rural areas where activities such as farming often provide ample opportunity for re-exposure to infective hookworm larvae (Bethony et al., 2006a, Hotez, 2007). Because infection with hookworm generally relies on the penetration of host skin by infective larvae, simply wearing shoes can often provide an effective barrier to exposure. However, as larvae can penetrate other areas of exposed skin this practice is not 100% effective and, in the case of A. duodenale, infection can occur by ingestion of infective larvae (Hotez, 2007). Individual predispositions to hookworm infection appear to vary: some individuals have consistently higher (or lower) infection levels than others in the same population and, moreover, these individuals will re-acquire infection to previous levels after cure (Hotez et al., 2004). A study conducted in an Iranian village revealed that 1-3% of the population harboured the majority of the parasites (Croll and Ghadirain, 1981). It is well known

19 that host genetic and socio behavioural facts can contribute tremendously to hookworm epidemiology. The current widespread use of anthelmintics for treating hookworm and other soil transmitted helminths has raised concerns of the potential of drug resistance (Albonico, 2003). In some nematode species that parasitise livestock, benzimidazole resistance occurs due to a single point mutation in nematode beta-tubulin alleles (Schwenkenbecher et al., 2007, Albonico et al., 2004), and it is speculated that this might partially account for an observed failure of chemotherapy for human hookworm in southern Mali (De Clercq et al., 1997). Studies have also shown diminished efficacy with repeated targeted treatments of mebendazole in Zanzibar (Albonico et al., 2003) and pyrantel pamoate failure against A. duodenale in Western Australia (Reynoldson et al., 1997). However, as yet, there is not any conclusive evidence of drug resistance in the hookworm species that infect humans.

2.1.4.4 Vaccines Potential anthelmintic resistance in nematode populations and concerns about the effects of drug residues on consumer health and the environment have focused attention on developing effective anti-nematode vaccines. However, one of the major issues that has been raised is whether it is feasible to develop a vaccine for an infection in which natural exposure to the pathogen does not confer immunity (Hotez et al., 2003c). This is especially highlighted by the fact that the cornerstone for the development of first generation attenuated vaccines was based on the concept of natural acquisition of immunity (Hotez et al., 2003c). A central challenge for hookworm vaccine development will be to stimulate an artificial immune response that is unique and results in disease burden reduction. Despite the lack of naturally acquired immunity to hookworm infections in humans, there are two major lines of evidence that support the feasibility of developing a hookworm vaccine. Firstly, dogs immunised with radiation-attenuated larvae of A. caninum were protected from challenge infection, with up to 90% reduction in worm burdens compared to control dogs that did not receive the vaccine (Miller, 1967). Interestingly, immunisation with dead larvae or non-irradiated larvae did not confer similar levels of protection (Miller, 1978), inferring that protection was induced by excretory/secretory antigens released by live larvae. Moreover, attenuation of L3 stops their development to adulthood and likely interferes with their implementation

20 of immunomodulatory strategies, accounting for the protection generated by attenuated but not live larvae. What this vaccine provides is “proof of principle” that antigens secreted by L3 upon migration through the host are capable of inducing protective immune responses against healthy larvae during challenge infection (Loukas et al., 2005b). The second line of evidence comes from research conducted on the sheep nematode, Haemonchus contortus (the “barber’s pole worm”), which is phylogenetically related to hookworms. Sheep immunised with parasite extracts rich in proteases derived from the intestine of adult worms showed high levels of protection against infection (Knox and Smith, 2001). Among the antigens identified in these extracts were parasite gut-derived glycoproteins that comprise a complex of proteases and other components (designated H-gal-GP) (Smith et al., 1994). Although the H-gal-GP complex is not ordinarily recognised during natural infection and is considered a hidden antigen, vaccination with the complex provided high levels of protection with respect to adult H. contortus worm burdens and fecundity (Knox and Smith, 2001). In similar fashion, dogs vaccinated with extracts of the oesophagus of adult A. caninum acquired immunity to this hookworm (Loukas and Prociv, 2001). Antibodies from the vaccinated dogs had the capacity to neutralise parasite protease activity, indicating that anti-enzyme antibodies have importance in mediating protective immunity against hookworm infections (Loukas and Prociv, 2001). Unlike vaccines developed against viruses and bacteria, where pathogens reproduce within the host, and sterile immunity is imperative, the major goal of a hookworm vaccine program is to decrease worm burdens to a level that minimizes pathology (Loukas et al., 2005b). It is envisaged that the ultimate hookworm vaccine will consist of a cocktail that elicits an immune response to at least two hookworm proteins - probably one from the L3 larval stage to target penetrating and migrating larvae, and another from the adult stage to minimize blood loss and anaemia (Loukas et al., 2006).

2.2 PROTEASES Proteases encompass a broad class of hydrolytic enzymes that play essential roles in cellular development and digestive processes, blood coagulation, inflammation,

21 wound healing and hormone processing. Recent studies have indicated that numerous proteolytic enzymes are essential molecules in a wide range of processes that determine parasitism, such as tissue penetration, feeding and immune evasion (Tort et al., 1999). Proteases are now the current focus of drug and vaccine based control programs for protistan and multicellular parasites (Dalton et al., 2003). Proteolytic enzymes are classified into five major catalytic categories based on the structure of their active site (serine, cysteine, threonine, aspartic) and dependency on co-factors (metallopeptidases) (Rawlings et al., 2008). Proteases produced by larval and adult hookworms play an important role in assuring parasite invasiveness and completion of the hookworm lifecycle in the appropriate host (Williamson et al., 2003b). Generally, proteases produced by hookworm larval forms have been implicated in tissue penetration, immune evasion and moulting, while proteases from adult forms are primarily implicated in tissue and blood meal digestion (Williamson et al., 2003b). Until recently, not much was known about the proteases of nematodes and their roles in haemoglobin proteolysis and nutrient acquisition. It has been suggested that as proteins are abundant components of blood, the major proteases found in hematophagous parasites are most likely to be proteolytic digestive enzymes (Williamson et al., 2004). It also has been noted that stage specific developmental regulation of these molecules often occurs (Hotez et al., 2004).

2.2.1 Hookworm Larval Proteases Molecules associated with hookworm invasion of skin and tissue have been identified in all hookworm species, but several detected in dog hookworm, A. caninum, have been the most extensively studied (Hotez et al., 2003b, Zhan et al., 2002, Zhan et al., 2003). Secreted enzymes that have been identified in A. caninum hookworm larvae include a hyaluronidase, zinc metalloproteases and cysteine-rich secretory proteins (Hawdon et al., 1999, Hotez et al., 1992, Zhan et al., 1999). Hyaluronic acid is a major component of the extracellular matrix and is also associated with cell adhesion by ligand binding with the CD44 cell surface receptor (Miyake et al., 1990). The release of hyaluronidase by invading hookworm larvae would presumably facilitate migration of the larvae through the host dermal layers by degrading cellular adhesion mediated by hyaluronic acid (Hotez et al., 1992). The inability of A. braziliense to penetrate humans seems to be associated with the failure

22 of larvae to produce adequate digestive enzymes which could allow penetration (Hotez et al., 1992). On the other hand, larvae of A. duodenale produce digestive hydrolases that allow complete penetration of human skin and, therefore, completion of the parasite lifecycle (Hotez et al., 1992). Datu recently identified a mRNA encoding for a hyaluronidase from A. caninum that were activated with serum (Datu et al., 2008). The larval metalloprotease identified by (Hotez et al., 1990), termed Ac- MTP-1, has similar effects on hyaluronic acid, and aids skin penetration by migrating larvae. This molecule was shown to degrade fibronectin and tissue elastin in vitro (Hotez et al., 1990). Williamson et al. then demonstrated that recombinant Ac-MTP-1 was capable of cleaving connective tissue proteins (Williamson et al., 2006). Because this secreted enzyme displays activity against components of the extracellular matrix, it was hypothesised that MTP-1 is important in parasite larval invasion and aids in attachment of adult worms to the host intestinal wall (Zhan et al., 2002). Secretion of zinc metalloproteases during the hookworm lifecycle may indicate a key role for these enzymes in maintaining a parasitic relationship. Dogs vaccinated with Ac-MTP-1, followed by challenge infection with A. caninum L3 larvae, revealed a statistically significant inverse association between anti-Ac-MTP-1 IgG2 antibody titers and the canine intestinal adult hookworm burden and quantitative egg counts at necropsy (Hotez et al., 2003b). A recent vaccine trial conducted with hamsters indicated that recombinant Ac-MTP-1 reduced worm burden by up to 29% compared to a the control group which received adjuvant alone (Xiao et al., 2008), suggesting that this molecule offers promise as a recombinant vaccine. A similar metalloprotease was identified in the blood feeding nematode H. contortus, and is involved in digestion of the larval cuticle, thus allowing the anterior cap to open permitting larval escape (exsheathment) (Gamble et al., 1989). The most abundant molecules released by penetrating hookworm larvae are cysteine rich secretory proteins which belong to the pathogenesis related protein (PRP) superfamily, and are termed Ancylostoma Secreted Proteins (ASPs) (Hawdon and Hotez, 1996, Hotez et al., 2003c). ASP-1 is a 45 kDa protein with a double PRP domain and is the major secreted protein in larval ES products (Hawdon et al., 1996). Immunisation of mice with recombinant ASP-1 inhibited migration of A. caninum L3 larvae to the lungs, the endpoint of migration in this non-permissive host (Ghosh et al., 1996). Although the biological functions of the ASPs are unknown, it has been

23 speculated that ASP-1 may modulate the host immune system in a manner which may benefit the parasite (Zhan et al., 1999). Such an action would not be entirely novel, as neutrophil inhibitory factor (a PRP family member) from A. caninum has been hypothesized to modulate the host immune system (Moyle et al., 1994), and a secreted hookworm zinc metalloprotease has been shown to specifically cleave the chemoattractant eotaxin and is hypothesised to aid in immune evasion (Culley et al., 2000). ASPs share sequence identity with wasp venom allergens, and adopt a fold that is similar to that displayed by some chemokines (Asojo et al., 2005). A second ASP molecule, termed ASP-2, has also been identified in the ES products of L3 that have been activated by the addition of serum in vitro (Hawdon et al., 1999) ASP-2 is a 21 kDa protein with a single PRP domain. The molecule has been immunolocalised to the glandular oesophagus of A. caninum L3, the basal lamina of the body cavity, the channels that connect the glandular oesophagus to the L3 surface, and on the L3 cuticle and epicuticle (Bethony et al., 2005). Vaccination of hamsters with recombinant Ay-ASP-2 protein (from A. ceylanicum) resulted in reduced egg output by female worms after challenge infection with A. ceylanicum L3 (Goud et al., 2004). Sera recovered from Ac-ASP-2 vaccinated dogs inhibited tissue penetration by A. caninum L3 by 60% percent compared with control sera (Fig. 2.8) (Bethony et al., 2005). A vaccine trial conducted in hamsters indicated that vaccination with recombinant Na-ASP-2 protein (from N. americanus) reduced the worm burden by 39% compared to controls that received adjuvant alone (Xiao et al., 2008).

Figure 2.8. Percent reduction of A. caninum L3 that penetrated skin in an in vitro model of tissue migration by hookworm larvae. Na-ASP-2 recombinant protein was shown to inhibit larval invasion in vitro at similar levels to that of irradiated A. caninum L3 larvae. (Loukas et al., 2006)

24 Evidence to date indicates that anti-ASP-2 antibodies interact primarily with the tissue-invading L3 stage of hookworms (Fujiwara et al., 2006). There are three independent lines of evidence indicating that ASP-2 is a leading vaccine candidate for human hookworm infection: 1) increased levels of IgE antibodies against ASP-2 appear to protect against heavy hookworm infection in humans; 2) anti-ASP-2 antibodies immunoprecipitate native ASP-2 from larval extracts and inhibit hookworm larval invasion through tissue in vitro; and 3) vaccination with ASP-2 results in lower worm burdens and fecal egg counts in dogs (Bethony et al., 2005, Fujiwara et al., 2006). The mechanism by which antibodies against ASP-2 reduce host hookworm burdens and faecal egg counts is not known, but based on the reported crystal structure of ASP-2, it has been proposed that the molecule could function as an immunomodulator by mimicking chemokines (Asojo et al., 2005). As it is inferred that ASP-2 vaccination interferes with the early stages of parasite invasion of the host, vaccination with this molecule would decrease the number of L3 larvae that reach the gastrointestinal tract, leading to reduced adult hookworm burdens and, hence, reduced host blood loss (Bethony et al., 2006b) (Fig. 2.9). A phase 1 study evaluating the safety and immunogenicity of Na-ASP-2/Alhydrogel in healthy adults without evidence of hookworm infection and living in the United States was conducted from 2005 through 2006. The vaccine was safe and well- tolerated and induced significant anti–Na-ASP-2 IgG and cellular immune responses

(Bethony et al., 2008; Diemert et al., 2008). In addition to the L3 secreted proteins, another surface protein from A. caninum, Ac-16, also shows promise as a potential vaccine antigen (Diemert et al., 2008; Fujiwara et al., 2007). Although Ac-16 is expressed during all stages of hookworm development, it is an immunodominant larval surface protein that has been shown to reduce faecal egg counts and blood loss in vaccinated dogs (Fujiwara et al., 2007).

25

Figure 2.9. Schematic of the effects of anti-ASP-2 antibodies on host hookworm burdens. Anti-ASP-2 antibodies inhibit early host entry of hookworm L3 through tissues, resulting in reduced numbers of L3 that gain entry into the host gastrointestinal tract. (Bethony et al., 2005)

2.2.2 Adult Hookworm Proteases 2.2.2.1 Aspartic proteases Aspartic proteases belong to clan AA and have two aspartic acid residues in the active site cleft which are utilised for catalysis of peptide substrates (Rawlings et al., 2008). They are optimally active at acidic pH, have endopeptidase activity and nearly all known aspartic proteases are inhibited by pepstatin (Umezawa et al., 1970). Aspartic proteases have several functions in mammals, including the processing of hormones, growth factors and proteolytic enzymes (Tang and Wong, 1987). Aspartic proteases include pepsins, D and E, and renins, and are considered to be the most conserved group of the five classes of proteases (Tang and Wong, 1987). One of the most well known aspartic proteases belongs to the A1 family of enzymes, which is typified by the mammalian gastric enzymes pepsin and gastricsin. The A1 family also includes the lysosomal processing enzyme cathepsin D (Rawlings et al., 2008). Cathepsin D-like aspartic protease genes have been reported from human blood flukes ( japonicum and ), which are known to digest haemoglobin (Brindley et al., 2001, Tort et al., 1999). Gene knockout of SmCatD demonstrated significant growth reduction of S. mansoni in vitro,

26 suppression of aspartic protease enzyme activity and an absence of black-pigment heme in the gut, suggesting that this protein plays an important role in nutrient absorption (Morales et al., 2008). Cathepsin D-like aspartic proteases have also been identified from adult A. caninum and N. americanus. The A. caninum cathepsin-D- like aspartic protease, termed Ac-APR-1, is 47% identical to SmCatD from S. mansoni, and is expressed in the intestinal microvilli of adult hookworms (Williamson et al., 2002). Dogs vaccinated with recombinant Ac-APR-1 had significant reductions in hookworm burdens (33%) and fecal egg counts (70%), in comparison to control dogs (Loukas et al., 2005a). More importantly, vaccinated dogs had reduced blood loss and most did not develop anaemia, which is the main pathology of hookworm disease (Loukas et al., 2005a). In addition, IgG from vaccinated animals decreased the catalytic activity of the recombinant enzyme in vitro and the antibody bound in situ to the intestines of worms recovered from vaccinated dogs, implying that the antibody interferes with the ability of the parasite to digest blood (Loukas et al., 2005a). A vaccine trial conducted with Ac-ARP-1 in hamsters resulted in a 44% reduction in worm burden compared to control hamsters that received adjuvant only (Xiao et al., 2008), further supporting the development of this molecule as a recombinant vaccine for hookworm infection. A second family of aspartic proteases, termed nemepsins, has been identified from the intestines of blood-feeding strongyle nematodes (Williamson et al., 2003a). This group of proteases resemble mammalian pepsin more closely than cathepsin D (Williamson et al., 2003b). In N. americanus, a nemepsin termed Na-APR-2 was characterised and localised primarily to the intestinal microvillar surface in adult worms (Bethony et al., 2005). Mouse antisera to Na-APR-2 inhibited 50% of infective hookworm larvae from penetrating mouse skin in vitro and inhibited the ability of the protease to cleave peptide substrates in vitro (Williamson et al., 2003a). A protease from H. contortus (termed Pep1) which is similar to mammalian pepsinogen has been localized to the gut of H. contortus adult worms, where it is believed to be involved in digestion of the blood meal (Longbottom et al., 1997). Pep1 is a component of the highly host-protective integral membrane protein complex, H-gal-GP, which reduced faecal egg count by 93% and worm burden by 75% when used as a vaccine in sheep (Smith et al., 2000). The contribution of Pep1 to the vaccine efficacy of H-gal-GP has yet to be determined.

27 Four aspartic proteases termed plasmepsin I, II, IV and histoaspartic protease (HAP) have been identified in the erythrocytic stage of the malaria parasite, Plasmodium falciparum, and have been localised to the food vacuole indicating that they are involved in the blood digestion process (Banerjee et al., 2002). Studies aimed at chemically inhibiting the plasmepsins, as well as gene knockout experiments, showed that this group of proteases is important for parasite viability and replication (Liu et al., 2005).

2.2.2.2 Cysteine proteases Cysteine proteases play numerous roles in the biology of parasitic organisms, which can range from general catabolic functions and protein processing to parasite immune evasion, excystment/encystment, exsheathing, and cell and tissue invasion (Tort et al., 1999). Parasite cysteine proteases are usually very immunogenic and have been exploited as serodiagnostic markers and vaccine targets (Dalton et al., 2003). Although host homologues exist, parasite cysteine proteases have distinct structural and biochemical properties, including pH optima and stability, alterations in peptide loops or domain extensions, diverse substrate specificities and cellular locations (Sajid and McKerrow, 2002). Cysteine proteases of parasitic organisms are divided into two main groups referred to as clans CA and CD (Barrett, 1994). The most widely reported class of cysteine proteases from parasitic nematodes is clan CA (Fig. 2.10). The clan CA proteases are further divided into families: C1, which comprises cathepsins B and L-like proteases, and C2, which comprises calpain-like proteases (Fig. 2.10) (Sajid and McKerrow, 2002). Cysteine proteases possess an essential cysteine residue that forms a covalent intermediate complex with substrates. The papain superfamily has a catalytic triad comprising of Cys, His and Asn residues and a highly conserved Gln that forms the oxyanion hole (Sajid and McKerrow, 2002).

28

Figure 2.10. Schematic diagram of the cysteine protease superfamily from parasitic helminths. Cysteine proteases of parasitic organisms are divided into two main groups referred to as clans, CA and CD. Papain-like, or Clan CA proteases, are further divided into family C1 (cathepsin B and cathepsin L-like) and family C2 (calpain-like) (Sajid and McKerrow, 2002).

Cysteine proteases have been identified in whole worm extracts from A. caninum and N. americanus, using the synthetic fluorogenic substrates such as Z- Phe-Arg-AMC (Harrop et al., 1995, Loukas et al., 2000). Cysteine proteases have also been identified in ES products obtained from adult hookworms, with a five fold increase compared to whole worm extracts (Loukas et al., 2000). It is hypothesized that secreted cysteine proteases may be involved in causing eosinophilic enteritis in humans - an allergic response after zoonotic infection with A. caninum (Loukas et al., 2000). Two A. caninum cysteine proteases, termed Ac-CP-1 and Ac-CP-2, identified from the adult stage, are closely related to human and bovine cathepsin B (Harrop et al., 1995). Molecular models of the active site of Ac-CP-1 indicate that, although it is structurally similar to the cathepsin B gene, it may have cathepsin L-

29 like specificity (Harrop et al., 1995). Ac-CP-1 immunolocalises to the oesophageal, amphidal and excretory glands of adult worms, which suggests that Ac-CP-1 is available for secretion at the site of attachment to the host (Harrop et al., 1995). In contrast, Ac-CP-2 immunolocalised to the brush border membrane of the intestine of adult hookworms, suggesting that it is involved in blood meal digestion (Loukas et al., 2004, Williamson et al., 2004). Vaccination of dogs with Ac-CP-2 resulted in a marked decrease in faecal egg count, and the number of female hookworms present in the intestine was significantly reduced relative to control dogs (Loukas et al., 2004). Adult worms recovered from the intestine of dogs vaccinated with Ac-CP-2 were significantly smaller than hookworms from control dogs (Loukas et al., 2004). It was also shown that anti–Ac-CP-2 antibodies bound to the gut of hookworms retrived from vaccinated dogs, which suggests that these antibodies were ingested by the parasites with their blood meal. IgG from vaccinated dogs decreased proteolytic activity of the recombinant protein against a peptide substrate by 73%, which implies that neutralizing antibodies were induced by vaccination (Loukas et al., 2004). Cysteine proteases have also been identified in H. contortus. Five distinct cysteine protease cDNAs have been cloned and termed AC1 to AC5 (Pratt, 1992). Members of this gene family appear to be expressed only in the adult stage of this parasite, and, hence, it has been hypothesized that these proteases are involved in digestion of the blood meal (Pratt, 1990). These cysteine proteases have been demonstrated to inhibit blood clot formation and to degrade haemoglobin, fibrinogen, and IgG, suggesting a role for the enzymes in attachment, blood feeding and immune evasion by the adult worm (Roads and Fetterer, 1995). A cysteine protease enriched fraction (TSBP), prepared from membrane extracts of adult H. contortus, was shown to localise to the microvillar surface of intestinal cells of the worm. Lambs immunised with TSBP were substantially protected against a single challenge infection with H. contortus, with reductions in daily faecal egg outputs of 77% and final worm burdens of 47%, compared to the control group (Knox et al., 1999). The microvillar surface of worms that survived in, and were recovered from, vaccinated lambs was found to be coated with sheep immunoglobulin, and antibody harvested from vaccinated lambs functionally inhibited the cysteine protease components of TSBP (Knox et al., 2005).

30 Several classes of cysteine protease have also been identified from adult stage Schistosoma species: these include proteins with cathepsin B, C and L like functions (Caffrey et al., 2004). Antisera against the cathepsin L-like proteases SmCL1 and SmCL2 (from S. mansoni) recognize schistosome gut tissue, suggesting a role for these proteases in blood meal digestion (Bogitsh et al., 2001). Similarly, cathepsin B- like protease SmCB1 also immunolocalised to the gut region of adult schistosomes, and was shown to be able to cleave haemoglobin, thus is thought to play role in nutrient absorption (Delcroix et al., 2006, Lipps et al., 1996). Cysteine protease activity has also been detected in P. falciparum, with three proteases, termed falcipains 1-3, that are localised to the food vacuole and thought to have a role in haemoglobin digestion (Shenai et al., 2000, Sijwali et al., 2001). Experiments conducted with gene knockout of falcipain-2, showed that trophozoites developed swollen, haemoglobin filled food vacuoles indicative of a block in haemoglobin digestion. Gene disruption of falcipain-3 was unsuccessful, suggesting that this protein is essential for survival of the erythrocytic parasite (Sijwali et al., 2006).

2.2.2.3 Metalloproteases Zinc metalloproteases are enzymes that rely on the presence of a zinc atom coordinated by nucleophilic amino acids at the active site for catalytic activity (Hooper, 1994). This enables the polarisation of the target scissile peptide bond before nucleophilic attack and subsequent cleavage (Hooper, 1994). The majority of the metalloproteases identified to date belong to the clan MA (Fig. 2.11) (Rawlings et al., 2008) and are recognised by the presence of a HEXXH motif in which the two His residues are zinc ligands and the Glu has a catalytic function (Hooper, 1994). A major component of the H. contortus highly protective antigen complex, H-gal-GP, is a family of four zinc metalloendopeptidases, designated MEPs 1–4. These enzymes belong to the M13 zinc metalloendopeptidase family (EC 3.4.24.11), which are also known as neutral endopeptidases or neprilysins (Newlands et al., 2006). Vaccination of sheep with a combination of all four MEPs, separated chromatographically from the rest of the complex, reduced H. contortus egg counts by 45 to 50%, compared to the control group (Smith et al., 2003). Similarly, MEP3 alone or MEPs 1, 2 and 4 in combination each reduced egg counts by 33% (Smith et al., 2003). It was therefore suggested that the MEPs are the host-protective

31 components of the H-gal-GP complex, and that MEP3 is the most effective member of this metalloendopeptidase family (Smith et al., 2003). Jones and Hotez (2002) cloned a metalloprotease from A. caninum: termed Ac-MEP-1, this protease is a neprilysin-like zinc dependent enzyme and was shown to be expressed in the intestinal lumen of adult stage hookworms (Jones and Hotez, 2002). Ac-MEP-1 exhibits significant similarity to the H. contortus developmentally regulated metalloprotease, MEP1, which is expressed in L4 larvae and adult stages of the parasite, and immunolocalises to the gut microvilli of the adult worm, where it is hypothesized to play a role in blood meal digestion (Redmond et al., 1997). Metallopeptidases have also been detected in P. falciparum, but unlike the nematode metalloproteases, these proteases belong to clan ME, family M16, which includes an “inverted” HXXEH active site motif (Fig. 2.11) (Eggleson et al., 1999). P. falciparum falcilysin shares primary structural features with M16 family members such as insulysin, mitochondrial processing peptidase, nardilysin, and pitrilysin, and has been localised to the food vacuole where it is thought to play a role in haemoglobin degradation (Eggleson et al., 1999).

Figure 2.11. Families of zinc metalloproteases. This schematic shows the families of the zinc metalloproteases and their inter-relationships, based on the sequence around the zinc binding residues. (Hooper, 1994)

32 2.2.2.4 Exopeptidase and aminopeptidases In addition to the proteases mentioned above, which mainly have endopeptidase and oligopeptidase activities, it has been suggested that aminopeptidases and proteases with exopeptidase functions play a vital role in the blood digestion process in hematophagous parasites (Williamson et al., 2003b). Aminopeptidase activity was recently described in P. falciparum trophozoites, suggesting that this activity was responsible for generating free amino acids from haemoglobin peptides after prior digestion with plasmepsins, falcipains and falycilysin (Stack et al., 2007, Gavigan et al., 2001). Aminopeptidase activity has been detected in the intestine of N. americanus (McLaren et al., 1974). AcDNA encoding an aminopeptidase has been identified from A. caninum and is thought to be involved in the hookworm feeding process, given that it is highly similar to the aminopeptidase from P. falciparum 1(Williamson et al., 2004). A glycosylated gut membrane protein with aminopeptidase activity known as H11 has been identified from H. contortus (Smith and Smith, 1993). Sequence analysis of the H11 clone revealed that it is similar to mammalian microsomal aminopeptidases which mediate the terminal events of digestion by digesting small peptides (Smith et al., 1997). H11 is the most effective immunogen isolated from a parasitic nematode to date, inducing high levels of protection, with >90% reductions in H. contortus worm burdens compared to the control group (Smith et al., 1993). H11 has been immunolocalised exclusively to the intestinal brush-border of adult H. contortus and enzyme activity is inhibited by H11 antisera in vitro (Smith et al., 1997). Similarly, sheep vaccinated with a leucine aminopeptidase, a metalloprotease from (the sheep ), induced the production of neutralising antibodies and elicited 89% protection against fascioliasis compared to the control group (Piacenza et al., 1999). It has also been shown that both S. mansoni and S. japonicum express a gene encoding a member of the M17 family of leucine aminopeptidases, and immunolocalisation studies showed that this protein is synthesised in the gastrodermal cells surrounding the gut lumen (McCarthy et al., 2004). Accordingly, it was proposed that peptides generated by protein degradation in the lumen of the schistosome gut are absorbed into the gastrodermal cells and are cleaved intracellularly by the aminopeptidase to free

1 T. Don and A. Loukas unpublished observation.

33 amino acids before being distributed to the internal tissues of the parasite (McCarthy et al., 2004).

2.3 HAEMOGLOBIN DIGESTION CASCADE The main source of nutrient for haemotophagous parasites are proteins found in the blood that they ingest. Blood is a specialized fluid that is composed of cells suspended in plasma. The most abundant cells in blood are red blood cells, which contain haemoglobin, an iron-containing protein. Haemoglobin is a tetramer consisting of two α and two β subunits, non-covalently bound to each other and made of 141 and 146 amino acid residues, respectively (Fig. 2.12). Each subunit has a molecular weight of about 17 kDa, for a total molecular weight of the tetramer of about 68 kDa. Each subunit is composed of a protein chain tightly associated with a non-protein heme group. Each protein chain arranges into a set of alpha-helical structural segments connected together in a globin fold arrangement, this folding pattern contains a pocket which strongly binds the heme group.

A B Figure 2.12. Tertiary structure of the haemoglobin molecule and the two subunits Haemoglobin is a tetramer (A) consisting of two α and two β subunits (B) non-covalently bound to each other. (A) (sourced from http://chemistry.ewu.edu/jcorkill/biochem/HemoglobinMOM.jpg. (B) www.science.org.au/sats2004/images/mackay9.jpg)

In mammals, gastric digestion of proteins derived from food involves a cascade of mechanistically distinct proteolytic enzymes including pepsin, trypsin and gastricsin. Similarly, haemoglobin the major source of protein for blood-feeding parasites, is thought to be degraded by a cascade of proteases found in the intestine (or food vacuole in Plasmodium) (Williamson et al., 2003b). A proteolytic cascade of haemoglobin digestion has been identified in the malaria parasite P. falciparum,

34 where a semi-ordered catalytic pathway exists and includes members of at least three different mechanistic classes of endoproteases (Fig. 2.13A) (Goldberg, 2005). It is believed that plasmepsin I and II initiate the degradation process by cleaving haemoglobin at the hinge region. The process is then followed by cleavage of globin fragments by other plasmepsins, falcipains and falcilysins (Goldberg, 2005). Lastly aminopeptidases are thought to release free amino acids which can then be used for protein synthesis. Aspartic and cysteine proteases present in the lumen and surrounding gastrodermal cells of blood feeding adult schistosomes are also thought to play a role in the degradation of haemoglobin through an ordered pathway (Fig. 2.13B) (Brindley et al., 1997, Delcroix et al., 2006). However, unlike the digestive pathway that has been described in P. falciparum, there appears to be less redundancy in schistosomes, as only one aspartic protease (catD) has been implicated in initiating the process (Brindley et al., 2001). In schistosomes, both cathepsin B and L-like cysteine proteases are capable of digesting haemoglobin fragments and in addition, an asparaginyl endopeptidase has also been implicated in the process but cannot cleave haemoglobin and is thought to be necessary for processing some of the SmCatBs (Fig. 2.13B) (Bogitsh et al., 2001, Caffrey and Ruppel, 1997, Dalton et al., 1997, Sajid et al., 2003). As yet, no metalloendopeptidase has been identified from the gut of adult schistosomes, however a metallo leucine aminopeptidase has been identified and is hypothesised to be involved in release of free amino acids (McCarthy et al., 2004). In similar fashion to P. falciparum and S. mansoni, it has been suggested that hookworms also employ a cascade of haemoglobinolysis using multiple proteases of distinct mechanistic classes (Fig. 2.14) (Williamson et al., 2003b). Don et al. demonstrated that erythrocytes ingested by A. caninum are lysed in the gut using a pore-forming membrane-bound hemolysin, which releases the red cell contents into the intestinal lumen for proteolytic degradation (Don et al., 2004). Similarly to P. falciparum plasmepsins (I-II) and falcipains (2-3), the A. caninum aspartic protease Ac-APR-1 and the cathepsin B-like cysteine protease Ac-CP-2 were both shown to be capable of digesting haemoglobin in vitro (Williamson et al., 2004). While experiments conducted with Ac-MEP-1, indicated that like falcilysin, it was incapable of digesting native haemoglobin or heat-denatured globin, it could further digest globin fragments generated by initial cleavage with aspartic and cysteine proteases (Williamson et al., 2004).

35

A

B

Figure 2.13. Proposed haemoglobin degradation pathway in P. falciparum and S. mansoni. In P. falciparum (A) it is proposed that the degradative enzymes function in a semi-ordered pathway, with plasmepsins making the initial cleavage in intact hemoglobin, followed by secondary cleavages by falcipains, falcilysin, while the dipeptidylpeptidases and aminopeptidases are presumed to function most efficiently in terminal degradation/amino acid release. (Goldberg, 2005) In S. mansoni (B) the primary cleavage of hemoglobin is facilitated by cathepsin D, followed by the endopeptidases cathepsins B1, L1. Peptides are then further broken down by exopeptidase cathepsin B1 and aminopeptidases. (Delcroix et al., 2006)

Hydrostatic pressure and peristalsis of the nematode pseudocoelom ensures that blood ingested by adult hookworms has a rapid passage time through the intestine and out of the anus (Roche and Layrisse, 1966). It is therefore important for the parasite to quickly lyse ingested erythrocytes and employ fast acting haemoglobinases in the intestinal lumen. Complete digestion of the haemoglobin tetramer by Ac-APR-1 and Ac-CP-2 was evident after just 15 minutes in vitro (Williamson et al., 2004), however it is difficult to extrapolate these findings to those occurring in the gut of a worm in vivo; suffice to say that haemoglobinolysis occurs rapidly with hookworm haemoglobinases.

36

Figure 2.14. Schematic of proposed haemoglobinase cascade in the intestine of blood-feeding nematodes. It is postulated that host Hb is initially cleaved by aspartic proteases and degraded to smaller peptides by cysteine proteases and then metalloproteases. Finally exopeptidases complete the digestion to constituent amino acids. (Williamson et al., 2003b)

2.4 SUMMARY Hookworms are one of the most debilitating intestinal pathogens in the developing world. New control measures are required, the most important perhaps being the development of a prophylactic vaccine. Proteases expressed by the various life stages of blood-feeding parasites appear to be viable candidates for development of novel therapeutics, so a more comprehensive understanding of their roles in host- parasite interactions is warranted (Dalton et al., 2003). Unlike vaccines produced against viruses and bacteria which reproduce asexually in the host, sterile immunity against helminth parasites, which do not reproduce asexually, is not essential for an effective vaccine (Loukas et al., 2005b). Therefore, an efficacious vaccine would be one which decreases worm burden and egg output, thereby minimising pathology as well as lowering transmission rates (Fig. 2.15). Elucidation of the molecular mechanisms by which haematophagous worms digest blood should lead to the production of new generation control strategies. By exploiting the absolute requirement of hookworms for blood, a recombinant vaccine targeting the digestion of the bloodmeal of these complex multicellular pathogens is a realistic near term goal.

37

Figure 2.15 Schematic for the development of a bivalent human hookworm vaccine. A bivalent recombinant human hookworm vaccine should consist of a protein that targets (1) invasion and migration of the L3 (eg, Na-ASP-2) and (2) blood-feeding by the adult hookworm (eg haemoglobinase) in order to be efficacious. (Loukas et al., 2005b)

2.5 THESIS HYPOTHESIS The main underlying hypothesis of this study is that human hookworm, N. americanus, digests haemoglobin using a semi-ordered cascade of mechanistically distinct proteases, termed haemoglobinases, which are expressed in the gut region of the adult worm. Furthermore, it is hypothesised that haemoglobinases are excellent targets for an anti-hookworm vaccine, as interfering with the parasite’s ability to digest blood, will decrease their growth, fecundity and survival. It has previously been demonstrated that proteases expressed in the gut of other parasitic helminths, such as H. contortus and S. mansoni, are viable vaccine candidates, providing significant levels of protection by reducing both worm burden and faecal egg output. It is proposed that the ideal hookworm vaccine will consist of a cocktail of two recombinant proteins - one targeting penetration and migration of infective larvae, and a second protein (such as a haemoglobinase) that will interrupt blood-feeding by adult worms, thereby reducing blood loss and anaemia.

38 This thesis has been submitted by publication. As such, each research chapter (Chapters 3 to 5) presents a manuscript which has been re-formatted to suit the style of the thesis. Published versions of the papers are attached as appendices.

39 CHAPTER 3: A SURVERY OF THE INTESTINAL TRANSCRIPTOMES OF THE HOOKWORMS, NECATOR AMERICANUS AND ANCYLOSTOMA CANINUM,USING TISSUES ISOLATED BY LASER MICRODISSECTION MICROSCOPY (International Journal for Parasitology 36: 701-710)

Najju Ranjita,b, Malcolm Jonesa,c, Deborah Stenzelb, Robin Gasserd, Alex Loukasa,e aDivision of Infectious Diseases and Immunology, Queensland Institute of Medical Research, Brisbane, QLD, Australia, bSchool of Life Sciences, Queensland University of Technology, Brisbane, QLD, Australia, cSchool of Molecular and Microbial Sciences, The University of Queensland, Brisbane, QLD, Australia, dDepartment of Veterinary Science, The University of Melbourne, Werribee, VIC, Australia, eAustralian Centre for International and Tropical Health and Nutrition, The University of Queensland, Brisbane, QLD, Australia.

3.1 CONTRIBUTIONS Contributor Statement of contribution Najju Ranjit -Designed all the experiments -Conducted all the experiments -Interpreted and analysed all the data -Drafted manuscript

Malcolm Jones -Aided in experimental design – LMM experiments -Provided feedback on manuscript

Deborah Stenzel -Discussed experimental design -Provided feedback on manuscript

Robin Gasser -Investigator on the grant that funded this project -Arranged for sequencing of ESTs -Provided feedback on manuscript

Alex Loukas -Aided in experimental design -Aided in data analysis -Aided in drafting manuscript

Principal Supervisor Confirmation

I have sighted email or other correspondence from all Co-authors confirming their certifying authorship.

Alex Loukas 27-5-08 Name Signature Date

41 3.2 ABSTRACT The gastrointestinal tracts of multi-cellular blood-feeding parasites are targets for vaccines and drugs. Recently, recombinant vaccines that interrupt the digestion of blood in the hookworm gut have shown efficacy, so we explored the intestinal transcriptomes of the human and canine hookworms, Necator americanus and Ancylostoma caninum, respectively. We used Laser Microdissection Microscopy (LMM) to dissect gut tissue from the parasites, extracted the RNA and generated cDNA libraries. A total of 480 expressed sequence tags (ESTs) were sequenced from each library and assembled into contigs, accounting for 268 N. americanus genes and 276 A. caninum genes. Only 17% of N. americanus and 36% of A. caninum contigs were assigned Gene Ontology classifications. Twenty-six (9.8%) N. americanus and 18 (6.5%) A. caninum contigs did not have homologues in any databases including dbEST – of these novel clones, seven N. americanus and three A. caninum contigs had Open Reading Frames (ORFs) with predicted secretory signal peptides. The most abundant transcripts corresponded to mRNAs encoding cholesterol- and fatty acid-binding proteins, C-type lectins, Activation-Associated Secretory Proteins, and proteases of different mechanistic classes, particularly astacin-like metallopeptidases. ESTs corresponding to known and potential recombinant vaccines were identified and these included homologues of proteases, anti-clotting factors, defensins and integral membrane proteins involved in cell adhesion.

Keywords: Laser microdissection microscopy; Hookworm; Expressed sequence tag; Intestine; Protease; Vaccine.

42 3.3 INTRODUCTION Hookworms are blood-feeding nematodes which inhabit the small intestines of their definitive mammalian hosts. Infective, third-stage larvae (L3) penetrate the host’s skin and migrate via the circulatory system to reside in the duodenum as adult stage worms (1-1.5 cm in length). Adult parasites bury their anterior ends beneath the mucosa of the bowel, rupture capillaries and feed on the extravasated blood. The pathogenesis of hookworm infection is a direct consequence of the blood loss which occurs during attachment and feeding. In developing countries, hookworms are the leading cause of iron deficiency anaemia, which, in heavy infections, can cause developmental and mental retardation in children as well as adverse maternal-foetal outcomes in pregnant women (Hotez et al., 2004). Current control strategies for hookworm are limited mainly to the treatment of infected patients with anthelmintic drugs. However, due to increasing drug resistance in parasitic nematodes of livestock and the perception that this may occur in helminths of humans, as well as the absence of naturally acquired immunity in most exposed people (Loukas et al., 2005b), the major focus of research has shifted towards developing an effective hookworm vaccine. Through the auspices of the Human Hookworm Vaccine Initiative, an antigen (Na-ASP-2) derived from the L3 of Necator americanus, a major hookworm species of humans, was selected for progression to clinical trials (Bethony et al., 2005, Goud et al., 2005). Na-ASP-2 is expressed exclusively by the L3 and is only partially efficacious at reducing worm burdens in vaccinated animals. Therefore, a useful human hookworm vaccine will likely require a second antigen, preferably one derived from the adult blood-feeding stage of the parasite (Loukas et al., 2005b, Hotez et al., 2003c). Hookworms ingest red blood cells, lyse the cells in their intestines via pore- forming proteins (Don et al., 2004) and digest the liberated haemoglobin using a semi-ordered cascade of proteases, some of which have been characterised in vitro (Hsieh et al., 2004). Vaccine trials in dogs with some of the recombinant haemoglobin-degrading proteases (haemoglobinases) have shown encouraging levels of efficacy (Don et al., 2004, Asojo et al., 2005). Moreover, a related nematode that parasitizes livestock, Haemonchus contortus, can be successfully vaccinated against using extracts enriched for haemoglobinases (Knox et al., 2005, Knox et al., 2001), and a recombinant vaccine against cattle tick targets a gut membrane glycoprotein

43 (de la Fuente et al., 1999), lending support to the targeting of gut proteins for the development of vaccines against blood-feeding parasites. Expressed sequence tags (ESTs) have been characterised from hookworms (Mitreva et al., 2005, Daub et al., 2000), but the majority of these sequences are derived from the L3 stage for Ancylostoma sp. (approximately 20,000 - www.nematode.net). Less than 5,000 ESTs from N. americanus have been described (see www.nematodes.org) and are deposited in dbEST (www.ncbi.nlm.gov/dbEST) (Parkinson et al., 2004). With a view to identifying mRNAs encoding potential new vaccine antigens from the hookworm gut, we characterised gut-specific transcripts of the two major hookworms of humans and canines, N. americanus and Ancylostoma caninum, respectively. Adult hookworms are small, which makes it very difficult to accurately dissect individual tissues, unlike the situation for larger nematodes such as H. contortus (Jasmer et al., 2001). However, with the development of Laser Microdissection Microscopy (LMM), a technique which allows dissection of tissues and even individual cells (Jones et al., 2004), it is now possible to dissect defined organs and cells from small parasites for subsequent isolation of tissue-specific proteins and nucleic acids. The potential of this application for extracting individual cells/tissues from histological sections of helminths has been proposed (Jones et al., 2004). Here, we describe the first application of LMM to the dissection of gut tissue from adult N. americanus and A. caninum, extraction of RNA for production of cDNA libraries, and comparative analyses of ESTs from each library.

3.4 MATERIALS AND METHODS 3.4.1 Parasite material A Shanghai strain of N. americanus was maintained in hamsters at The George Washington University, and worms were a kind gift from Drs Bin Zhan and Peter Hotez. Adult A. caninum were collected from dogs in Brisbane, Queensland, as described previously (Don et al., 2004). The recovered worms were washed 3× in PBS, individually positioned in Optimal Cutting Temperature (OCT, Tissue-tek) compound and snap-frozen on dry ice. Frozen blocks were kept at -80oC until sectioned onto glass slides which were coated with poly-ethylene naphthalene membrane. Frozen blocks were sectioned at a thickness of 7 microns.

44 3.4.2 Laser microdissection microscopy (LMM) After sectioning, each slide was individually wrapped in plastic wrapping pretreated with RNAase Zap (Ambion) and stored at -20oC until needed. Slides were washed with diethylpyrocarbonate (DEPC) water to remove OCT, fixed in 100% methanol, stained with hematoxylin to allow visualisation of intestinal tissue and dehydrated with 70% ethanol, 100% ethanol and xylene respectively. Slides were dried in a fume hood for ~ 2 hours. Intestinal or gonadal tissues were selected and catapulted separately into 0.6 ml microfuge tube lids containing 30 μl of Trizol (Invitrogen) using a PALM MicroBeam Laser Catapult Microscope (P.A.L.M. Microlaser Technologies, Bernried, ).

3.4.3 RNA extraction, cDNA synthesis and detection of known gut transcripts Total RNA was isolated from catapulted sections using Trizol (according to the manufacturer’s instructions), with a yield of 200 ng for A. caninum and 130 ng for N. americanus. The cDNA was generated using a SuperSMART PCR cDNA kit (BD Clontech) where full-length cDNAs are generated by oligocapping. Second- strand synthesis and PCR amplification were conducted according to the manufacturer’s instructions. cDNA fragments >1 kb were size selected on a 0.8% agarose gel, excised and gel extracted with a kit (Qiagen). Gut cDNA was used as a template for PCR amplification of transcripts which were already known to be expressed in the gut (via immunlocalisation) using gene specific oligonucleotide primers that corresponded to the following cDNAs - Ac-mep-1 5’ CCGAAAAGGGACCACTTCCTG 3’ (forward primer), 5’ AGTCGCTAAGGCTTCCGTCG 3’ (reverse primer); Na-mep-1 5’ CGCGCCGGATCCGACAACGATAACCCACCA 3’ (forward primer), 5’ CGCGCCCTCGAGCTCTTGAACCCAAACTGA 3’ (reverse primer) (Jones and Hotez, 2002) (N. Ranjit and A. Loukas, unpublished); Na-apr-1 5’ CGCGCCGAGCTCAGCGTTCATCGACGACTC 3’ (forward primer), 5’ CGCGCCAAGCTTAAAAAAGTAA 3’ (reverse primer) (Hotez et al., 2002); Na- apr-2 5’ CGCGCCGAGCTCGGTGTATATAAAATCCCA 3’ (forward primer), 5’ CGCGCCAAGCTTAAATGTTTTACAGCTGCAAAACC 3’ (reverse primer) (Bethony et al., 2005). Primers corresponding to a gene with a presumed ubiquitous expression pattern, N. americanus β−tubulin (AF453524) - 5’ CGAATCTCGTGCCATATCGT 3’ (forward primer), 5’

45 TTCCTCCATACCCTCACCGA 3’ (reverse primer) - were used to amplify transcripts from both cDNA populations, and cDNA recovered from whole adult worms. Primers corresponding to mRNAs for proteins that localise to sites other than the gut were also used in the PCR, employing as a template cDNA from gut, gonad or whole adults. Non-gut derived cDNAs included Ac-asp-4 5 ‘AAGCCAGTGTCTCACAGGAGGGT 3’ (forward primer), 5’ TCGGGTCTTGGTCATAGATGGGG 3’ (reverse primer) and Ac-asp-6 5’ TTTGTGGACCATAACAGTGCGGC 3’ (forward primer), 5’ TTTTGGGGAGTAGGGCAGACGA 3’ (reverse primer) (Zhan et al., 2003).

3.4.4 Construction of cDNA libraries Non-directional cDNA libraries were constructed from gut cDNA in the plasmid vector pGEM-T (Promega) using T-ended cloning. One hundred and fifty ng of size-selected cDNA was added to 50 ng of vector. Colonies were screened using blue-white selection. Five hundred clones from each library were randomly picked and patched onto grided Luria Bertani (LB) agarose plates containing 100 μg/ml ampicillin. Sequencing reactions were performed at AgGenomics (Melbourne, Australia), utilising ABI BigDye terminator v3.1 and an ABI 3730xl DNA analyser. Single-pass sequencing was performed on each template using the T7 primer.

3.4.5 Bioinformatic analyses Sequence corresponding to vector was trimmed from the raw sequence data using BioEdit software (http://www.mbio.ncsu.edu/BioEdit/bioedit.html). The sequence of each EST was also manually edited in order to remove poly A tails and poor quality sequence before further analyses. All edited sequences were condensed into contigs or singletons using the contig analysis program (CAP) in BioEdit with parameters of 100 bp overlap and a minimum of 95% identity at the nucleotide level. Sequences were compared to other sequences in GenBank (nr), Wormbase (http://blast.wormbase.org/db/searches/blat) and nematode.net (http://www.nematode.net/index.php) using BLASTx and BLASTn (http://www.ncbi.nlm.nih.gov/BLAST/), and WU-BLAST (http://www.ebi.ac.uk/blast2) respectively. BLAST alignments with an E-value of ≤ 1 × 10-5 were reported. Clusters were functionally categorised using InterProScan (http://www.ebi.ac.uk/InterProScan) and clusters were mapped to the three

46 organising principles of gene ontology (http://www.geneontology.org). Files were submitted as batch sequences at the Goblet web browser (www.goblet.molgen.mpg.de/cgi-bin/goblet-batch.cgi), sequences were queried against Wormbase (http://www.wormbase.org/) and alignments with an E-value of ≤ 1 × 10-20 were reported. Multiple sequence alignments were conducted using ClustalW at the BCM Search Launcher (http://searchlauncher.bcm.tmc.edu/multi- align/multi-align.html). Predictions of signal peptides were conducted using SignalP 3.0 (http://www.cbs.dtu.dk/services/SignalP/) incorporating both neural networks and hidden Markov models. Predictions of transmembrane domains were conducted using TMPred (http://www.ch.embnet.org/software/TMPRED_form.html). The clan and family assignments of proteolytic enzymes were analysed via the MEROPS protease database (Rawlings et al., 2008).

3.4.6 Phylogenetic tree Phylogenetic relationships were assessed and trees were generated by the neighbour joining method using PAUP 4.0 beta version (Swofford, 1993). Robustness was assessed by bootstrap analysis using 100 replicates; clades with more than 50% support are denoted with bootstrap values on the branches.

3.5 RESULTS AND DISCUSSION 3.5.1 Extraction of gut tissues from hookworms Intestinal and gonad tissues were readily identified by light microscopy in both longitudinal and transverse sections of both A. caninum and N. americanus (Fig. 3.1). Each slide contained sections of 5-7 worms and tissue was extracted from 12 slides per species, corresponding to a total of ~ 720,000 μm2 tissue. Two hundred ng and 130 ng of total RNA were extracted from A. caninum and N. americanus gut tissue, generating 2.5 μg and 1.7 μg of double stranded cDNA respectively.

47 A B

C D Figure 3.1. Light micrographs of adult hookworm section before and after laser microdissection. Longitudinal sections of A. caninum before (A) and after (B) removal of highlighted tissues by laser microdissection. Transverse sections of N. americanus before (C) and after (D) removal of highlighted tissues by laser microdissection. lu – gut lumen; in – intestine; ov – ovary.

3.5.2 Tissue specificity of cDNA populations To verify that extracted tissue catapulted from intestinal tissues was gut- derived and did not contain control tissue dissected from gonad, we used cDNA populations extracted from gut and gonad as templates in the PCR employing oligonucleotide primers corresponding to mRNAs for which the anatomic expression sites of the corresponding proteins have previously been demonstrated using immunofluorescence with specific antibodies against the recombinant proteins (Hotez et al., 2002, Bethony et al., 2005, Jones and Hotez, 2002). Partial transcripts corresponding to these mRNAs were successfully amplified from gut-derived but not gonad-derived tissues (Fig. 3.2). An mRNA for a protein presumed to be ubiquitously expressed throughout all tissues, β−tubulin (AF45352), was amplified from both gut and gonad cDNAs. Two cDNAs encoding the activation associated secreted proteins – Ac-ASP-4 which was immunolocalised to the cuticle surface and Ac-ASP-6 which was immunolocalised to cephalic and excretory glands (Zhan et al.,

48 2003) - were amplified from gonad (and whole worm cDNA – not shown) but not from gut cDNA (Fig. 3.2).

A B

1200

300 300

1 2 3 4 5 6 1 2 3 4 5 Figure 3.2. Detection in gut but not ovary cDNA of transcripts corresponding to proteins that were previously shown to be expressed in the intestine using immunolocalisation. Agarose gels stained with ethidium bromide showing PCR products amplified from N. americanus gut cDNA (A) or gonad cDNA (B). Panel A - lanes 1-6: Na-apr-2, Ac-mep-1, Na-apr-1, Na-β- tubulin, Ac-asp-6, Ac-asp-4. Panel B – lanes 1-5: Na-mep-1, Na-apr-1, Na- β -tubulin, Ac-asp-6, Ac-asp-4.

3.5.3 Characteristics of the EST dataset ESTs were derived from the cDNA libraries representing the intestinal tissues from each of N. americanus and A. caninum. Four hundred and eighty ESTs were generated from each library and assembled into contigs to remove redundant sequences and establish the quality and length of the sequences. Contigs were further grouped into clusters to provide a non-redundant catalogue of partial genes represented. Clusters ranged in size from a single EST to 20 and 30 ESTs for N. americanus and A. caninum respectively, but generally most clusters consisted of five or less ESTs. The average sizes of inserts were 400 bp for N. americanus and 490 bp for A. caninum. After the contig assembly, there were 268 N. americanus cDNAs and 276 A. caninum cDNAs, achieving cDNA discovery rates of 55% and 57%, respectively.

3.5.4 Sequence analysis and gene ontology In addition to the data provided in Table 3.1, 230 N. americanus contigs (86% of the total) and 249 A. caninum contigs (90%) had ORFs ≥ 30 amino acids. Of these contigs with ORFs of ≥ 30 amino acids, 136 N. americanus and 120 A. caninum contigs had significant homologues in dbEST only, and did not have homologues in GenBank nr; the vast majority of these homologues were derived from parasitic nematodes. Of those N. americanus contigs with homologues in dbEST only, 41 corresponded to previously identified N. americanus ESTs in dbEST (≥ 95% identity

49 at the nt level over at least 100 nt). Of those A. caninum contigs with homologues in dbEST only, 34 corresponded to previously identified A. caninum ESTs in dbEST. Gene ontology (GO) and BLAST analyses revealed some interesting features for the two datasets, particularly the relatively large number of contigs with no homology to molecules with known functions (in any of the public databases interrogated – Table 3.1). Of these, 7 N. americanus and 3 A. caninum contigs had ORFs with putative N-terminal signal peptides (Table 3.2), indicating that the corresponding mRNAs encoded secreted and/or transmembrane proteins with unknown functions and no known homologues thus far identified in any organism. Fig. 3.3 provides a summary of representatives categorized by major Gene Ontology (GO). Level II GO hierarchy was chosen for the classification of ESTs. GO categories revealed numerous transcripts encoding kinase activity, heat shock proteins, and metabolic enzymes. Interestingly, only 36% of A. caninum and 17% of N. americanus contigs mapped to GO categories, with the most common GO category for both species being molecular functions, in particular binding and catalytic activity. The two main predicted binding functions were protein- and ion- binding, whereas the main catalytic activity was that of hydrolases. These GO assignments were similar to those reported by Mitreva et al. (Mitreva et al., 2005) for different life history stages of A. caninum and A. ceylanicum, where major GO categories represented binding, catalytic activity, transporter activity and structural molecular activity.

50 Table 3.1. Summary of the gut EST datasets for A. caninum and N. americanus contigs with ORFs containing ≥ 30 amino acids.

No. identical to No. similar to No. similar to No. No. novel with signal Species No. of contigsa existing ESTsb existing ESTsc GenBank nr novel peptide 18 A. caninum 249 (276) 52 (18.8%) 206 (74.6%) 122 (44.2%) (6.5%) 3 (1.6%) 26 N. americanus 230 (268) 80 (29.8%) 184 (68.8%) 84 (31.3%) (9.8%) 7 (2.6%)

a Numbers in brackets are total numbers of contigs irrespective of length of ORF. b Identity determined as ≥95% identity at the nucleotide level over at least 100 nt. c Similarity determined as E-value of ≤1.0x105.

51

Table 3.2. Novel clones with no homologues in any datasets that contain ORFs with predicted signal peptides. Double-ended arrows denote the predicted cleavage site of the signal peptide.

Clone Predicted Signal Sequence Ca S Y s D HMM internal TM domainsb Ac92 MRRRSVLLKPTTTNLSIVMLGFILGSPAVS↔VI No Yes Yes Yes Yes SP (0.905) 1 Ac94 MIQNIYLMLILNPHILFL↔YK No No Yes Yes Yes NS 0 Ac198 MAPGSRTSLLLAFALLCLPWLQEAGA↔VQ Yes Yes Yes Yes Yes SP (1.000) 0 Na60 MVRSAVCCSLLFLAPSTTT↔TI No Yes Yes Yes Yes SP (0.998) 1 Na130 MLTFIELLIGVVVIVGVA↔RY Yes Yes Yes Yes Yes SA (0.688) 1 Na138 MLFIYSVNSKTCLLLRFFHPEVVA↔SC Yes Yes No Yes No NS 0 Na144 MEIQKGVEGIGKVKDFLRIFQRFKFFNNCFGW Yes Yes Yes No No SA (0.841) 0 VFFWLFMFFCWCES↔FF Na152 MRQRMFLVLMRASYGGL↔ED No No No Yes No NS 0 Na157 MFFCSLFLFSLVFA↔WY Yes No Yes Yes Yes SP (0.850) 0 Na239 MFAYPPVYPLCTLCLGGIRGKSA↔GT Yes No Yes No No SP (0.857) 0 a The SignalP algorithm incorporates both neural networks and hidden Markov models. Output scores are predicted as being above (Yes) or below (No) a defined cut-off. C, ‘cleavage site’; S, ‘signal peptide’ score, Y, the combined C and S scores, s is the mean S score between the N-terminus and the cleavage site and D is the average of the s and Y scores.HMMis the hidden Markov model prediction with the prediction probability in parentheses. SP, signal peptide; SA, signal anchor; NS, non-secretory. b Number of predicted transmembrane domains C-terminal to the predicted signal sequence.

52 A. caninum N. americanus

Binding 7.49% Catalytic activity 5.62% 20% Structural molecular activity 1.50% Transporter activity 1.12% 1.12% Antioxidant activity Unknown

13%

64% 3% 83.15%

Figure 3.3. Pie charts depicting gene ontology classifications of Ancylostoma caninum and Necator americanus gut ESTs identified in this study.

3.5.5 Transcript abundance and highly represented genes The 10 most abundant clusters from both libraries are presented in Table 3.3 and account for 17% and 24% of ESTs for N. americanus and A. caninum, respectively. Transcripts abundantly represented in both of the libraries included genes predicted to encode proteins which carry out (and facilitate) key energetic and metabolic processes, such as feeding. In the A. caninum dataset, transcripts encoding proteases and proteins which bind to both fatty acids and cholesterol were predominant. Vitellogenin was the most abundant transcript accounting for 6% of ESTs sequenced and was also one of the most abundant gut transcripts from N. americanus (Table 3.3). Vitellogenin transcripts are also abundant in the intestine of H. contortus (Jasmer et al., 2001) and Pristionchus pacificus (http://www.nematode.net/index.php). A fatty acid-binding protein with invertebrate (including nematodes) and vertebrate homologues was also represented by an abundant A. caninum transcript. A homologous hookworm protein has been suggested to be involved in the scavenging, transport and metabolism of fatty acids and sterols (Basavaraju et al., 2003), necessary for metabolic and developmental processes, such as embryogenesis, glycoprotein synthesis, growth and cellular differentiation (Kennedy, 2000).

53 3.5.6 Molecules involved in feeding cDNAs encoding known and new potential anticoagulants were identified in the gut ESTs. Contig Ac72 was 99% identical to anticoagulant peptide 3 (AcAP3), while contig Ac166 was 97% identical to AcAP4. Immunolocalisation of AcAP3 showed exclusive expression in the oesophagus, and it has been hypothesised that these anticoagulants are critical during the blood meal as well as assisting in digestion by keeping the blood in a liquid state once it has entered the digestive tract (Mieszczanek et al., 2004). To avoid the formation of clots at the site of attachment, hookworms secrete a platelet inhibitor (HPI) which inhibits platelet aggregation and adhesion (Chadderdon and Cappello, 1999, Hotez et al., 2003c, Del Valle et al., 2003). HPI has been identified from the cephalic glands of adult worms (Del Valle et al., 2003), and the identification herein of contig Ac93 (the ninth most abundant transcript in the A. caninum dataset) suggests that HPI is expressed in the gut as well as the cephalic glands. The presence of these transcripts in the gut is unlikely to be accounted for by contamination of dissected gut tissue with oesophagus, because none of the tissue sections used for laser microdissection contained oesophagus or pharynx. Moreover, Ac93 was abundantly represented in the gut ESTs yet another mRNA from the oesophagus, Ac-cp-1 (Sawangjaroen et al., 1995), was not detected in the gut ESTs.

54 Table 3.3. The 10 most abundant contigs from the A. caninum and N. americanus gut expressed sequence tag (EST) datasets

Contig Closest homologue E-value ESTs no. Accession no. per (GenBank) contig Ac153 A. ceylanicum Vitellogenin CB176085 6.10E-130 30 Ac81 A. ceylanicum Excretory/Secretory protein AY046590 4.10E-14 26 Ac201 A. ceylanicum Heat Shock Protein 20 CB338919 7.30E-63 16 Ac242 C. elegans Hypothetical protein CB037581 2.4E-31 16 Ac196 N. americanus 18s small subunit AY295811 3.70E-123 14 Ac129 C. elegans C-type lectin CB176035 2.70E-107 10 Ac197 C. elegans Astacin metalloprotease BM130887 7.60E-17 6 Ac8 A. ceylanicum Fatty acid binding protein CA341320 1.80E-78 5 Ac93 A. caninum Platelet inhibitor AF399709 1.40E-124 5 Ac155 A. caninum Cytochrome c oxidase subunit 1 AW627047 9.40E-45 5

Na205 N. americanus 18s ribosomal RNA AY295811 8.80E-132 20 Na85 N. americanus Heat shock protein 20 BG734476 1.60E-57 12 Na155 A. ceylanicum Lysozyme protein 8 CB190629 8.10E-41 12 Na160 A. ceylanicum Vitellogenin CB176264 7.70E-39 12 Na96 S. ratti MDR protein (ATP-binding cassette) CB098543 4.60E-23 10 Na265 X. index Non-functional folate binding protein CV579873 9.80E-31 9 Na220 N. americanus Amyloid precursor protein BG467555 1.30E-11 6 Na182 A. caninum NADH-ubiquinone oxidoreductase BM077443 1.20E-46 4 Na230 N. americanus Cathepsin B cysteine proteinase BI744492 4.00E-14 4 Na10 N. americanus Cytochrome c oxidase subunit 1 BU088443 1.20E-66 4

Hookworms are thought to digest the proteinaceous contents of lysed red cells with a semi-ordered cascade of proteolysis, consisting of aspartic, cysteine and metalloproteases, some of which have been characterised in vitro (Hsieh et al., 2004). Three haemoglobinases expressed in the gut of A. caninum were all detected by PCR in cDNA derived from the gut (Fig. 3.2). Two of these (Ac-APR-1 and Ac- CP-2) were also detected as ESTs in the present study. At least twelve contigs encoded for proteases from all four major mechanistic classes between the datasets of the two species, and nine contigs corresponded to newly identified enzymes

55 (Table 3.4). Three new metalloproteases of the astacin family were identified. An astacin-like enzyme, Ac-MTP-1 is secreted by L3 during the invasion process (Zhan et al., 2002) and digests connective tissue substrates (Williamson et al., 2006), but this class of enzyme has not been described in adult hookworms. Compared with other metalloprotease families, astacin-like enzymes (MEROPS – family M12A) are the most abundantly expressed metalloproteases in C. elegans, and subgroup I of the M12A family are secreted by pharyngeal cells into the lumen of the alimentary tract of the nematode, where the protease is thought to be involved in the digestion of food (Mohrlen et al., 2003). The M12A proteases identified amongst the A. caninum ESTs might also be involved in digestion of the blood meal. Contig Ac70 encoded an O- sialoglycoprotein endopeptidase, a family of metallopeptidases which only cleave proteins that are O-sialoglycosylated. This family of proteases has not been reported previously for nematodes which parasitize animals. Cathepsin B-like cysteine proteases were also abundantly represented in the gut and were the ninth most abundant gene family in the N. americanus dataset. Four cathepsins B were identified from N. americanus (Table 3.4), two of which were not found in the existing N. americanus ESTs derived from entire adult worm cDNA. The H. contortus intestinal transcriptome is dominated by cysteine proteases (~16%) (Jasmer et al., 2004, Jasmer et al., 2001), and while abundantly expressed in N. americanus, the diversity of this gene family appears not to be as extensive as it is for H. contortus.

56 Table 3.4. Hookoworm (Anyclostoma spp. plus Haemonchus and Caenorhabditis, of comparison) gut ESTs encoding proteolytic enzymes

Protease class Contig no. Closest nr homologue Closest EST a ORF (MEROPS i.d.) length (aa) b Aspartic Clan AA, Family A1 Ac132 Ac-APR-1c A. caninum 84 Cathepsin D U34888 (6.00x10-66) Clan AA, Family A1 Na48 Na-APR-2c N. americanus 43 AJ245458 (1.90x10-36) Metallo- Clan MA, Family M12 Ac110 C. elegans Anyclostoma ceylanicum 62 Astacin CAB05814.2 (2x10-15) BM130887 (3.10x10-13) Clan MA, Family M12 Ac197 C. briggsae A. caninum 45 CAE60270.1 (7x10-5) AF397162 (7.60x10-17) Clan MA, Family M12 Na250 C. briggsae A. ceylanicum 45 CAE60270.1 (7x10-5) BM130887 (7.60x10-17) Clan MK, Family M22 Ac70 C. elegans Pristionchus pacificus 142 O-sialoglycoprotein AAK29978.4 (3ex10-57) AI989172 (0.019) endopeptidase Cysteine Clan CA, Family C1A Na56 H. contortus N. americanus 53 Cathepsin B CAA93278.1 (3x10-18) BG468101 (1.70x10-12) Clan CA, Family C1A Na105 Glossina morsitans H. contortus 66 AAK07477.2 (3ex10-20) BM873292 (3.70x10-11) Clan CA, Family C1A Na191 Fasciola hepatica N. americanus 52 CAD32937.1 (7x10-8) BG468129 (1.10x10-40) Clan CA, Family C1A Na230 Lonomia obliqua A. caninum 53 AAV91452.1 (1x10-19) BI744492 (7.60x10-17) Clan CA, Family C1A Ac193 Ac-CP-2 A.caninum 88 U18912 (4.30x10-80) Serine Clan PA, Family S1A Ac78 Manduca sexta A. ceylanicum 81 AAV91012.1 (8.00x10-4) CA341432 (2.5 x10-4)

Clan and family assignments and generic family names from the MEROPS database are provided. BLAST E-values and GenBank accession numbers are provided for homologues. a EST, expressed sequence tags. b ORF, open reading frames. c Previously identified protease known to be expressed in the gut using immunlocalisation

Contig Ac78 encoded an S1A family of serine proteases with similarity to Manduca sexta prophenoloxidase-activating endopeptidase and vertebrate coagulation factors VIIa and complement factors C1. A text search of nematode.net (http://www.nematode.net/index.php) using “serine protease” or “serine proteinase”

57 did not reveal any serine protease-encoding clusters from hookworms or other nematodes that parasitise humans; three clusters corresponding to trypsin-like serine proteases were identified from H. contortus. (A. Loukas, unpublished observation).

3.5.7 Immunomodulation Another highly-represented gene (sixth most abundant in the A. caninum dataset) encoded a C-type lectin (C-TL). C-TLs are proteins with a carbohydrate recognition domain (CRD) which bind to glycoprotein ligands in a calcium-dependent manner. They are important in a multitude of physiological processes in animals, including innate and acquired immunity, haemostasis and wound repair. C-TLs are an abundant gene family in C. elegans (Drickamer and Dodd, 1999), and some are expressed in the intestine where they are upregulated in response to bacterial infection (Mallo et al., 2002). Together with cysteine proteases, C-TLs are the most abundant gene family in the gut ESTs from H. contortus (Jasmer et al., 2001), although the functions of these molecules have yet to be determined. Identified in the present study was a full length cDNA (contig Ac129) encoding an N-terminal secretory signal peptide followed by a long form C-TL domain that shared homology with Na-CTL-2, C. elegans lectins and CD69, an early leukocyte activation molecule expressed at sites of chronic inflammation where it is thought to down regulate the immune response through the production of TGF-β (Sancho et al., 2005) (Fig. 3.4). A total of six contigs encoded additional C-TLs in the two hookworm EST datasets. Activation-associated Secreted Proteins (or ASPs), are a family of nematode- specific cysteine-rich, secreted proteins belonging to the pathogenesis-related protein superfamily (Hawdon and Hotez, 1996). ASPs expressed by hookworm L3 are efficacious recombinant vaccines (Bethony et al., 2005, Ghosh and Hotez, 1999). Adult hookworms also secrete at least four other ASPs, and each of them localises to a unique organ in the parasite (Zhan et al., 2003). In the present study, two novel ASPs were identified - contigs Na91 and Ac173 (Fig. 3.5). The functions of most of the ASPs in adult hookworms are still unknown, however the recent determination of the crystal structure of Na-ASP-2 from N. americanus revealed a protein conformation similar to that adopted by some chemokines, prompting speculation about an immunomodulatory function (Asojo et al., 2005).

58 Ac129 1 ----MFFKSSLLFCVLTTALSVTIN------Na-CTL-2 1 ----MLFISSLFFCVLSTVSSTIIN------CeCBG03358 1 ----MRFFRFLVFPVIAGLSSVLAAPITSNDTVDGSGEAPETLLQNSEEQPHQRLKF--- CD69 1 ------NKR-P1B 1 MDSTTLVYADLNLARIQEPKHDSPPSLSPDTCRCPRWHRLALKFGCAGLILLVLVVIGLC

* * Ac129 22 ------TTELKCPTGWFEYRDSCYFIDNPLAEYDRAQA Na-CTL-2 22 ------TTELTCPPGWFGYRDSCYFFDNPLLEHDKAEI CeCBG03358 54 -YNWDYKDLGTTAFEDISFPARQPPVAVNQSEQCPDGWLRFADSCYWIETELMGFAKAER CD69 1 ------VSSCSEDWVGYQRKCYFISTVKRSWTSAQN NKR-P1B 61 VLVLSVQKSSVQKICADVQENRTHTTGCSAKLECPQDWLSHRDKCFHVSQVSNTWKECRI

* Ac129 54 RCWEQGATFLVAETPEEYTYVTEHSKPS-TWSWAGITQED----ENHLPKWSNNGGVDPA Na-CTL-2 54 KCWEMGSTLLVAETLDEYELITDRAKES-AWSWVGLTQSDDL--EHHIPQWSTSGGVDP- CeCBG03358 113 KCFEKQSTLFVANSLEEWDSIRSHSKEA-YFSWIGLVRFTHYEKSEQLPRWQTEGAINP- CD69 31 ACSEHGATLAVIDSEKDMNFLKRYAGR--EEHWVGLKKEP-----GHPWKWSNGKEFNN- NKR-P1B 121 DCDKKGATLLLIQDQEELRFLLDSIKEKYNSFWIGLSYTLT----DMNWKWINGTAFNS-

* * * Ac129 109 TMINWLVKPFTPVANGWSTTAKCAAHLNVPVS--FAAYTFFLPCNIQINSICEKNFTLFP Na-CTL-2 110 ILINWLVKPYLAVSNGWTTQAKCAAHLNVPAGP-SASYTFFLPCTVQTYSICEKNATLFP CeCBG03358 171 TKMNWLIKPYKPIVNGWTSFANCAASYKSPATLESASYTFFYPCTYLLYSICERNSTIVN CD69 83 ----WFNV---TGSD------KCVFLKNTEVS------SMECEKNLYWICNKPYK--- NKR-P1B 176 ----DVLKITGDTENG-----SCASISGDKVT------SESCSTDNRWICQKELNHET

Ac129 167 RIWDHGLVNLK Na-CTL-2 169 RIWEHGLIGL- CeCBG03358 231 VMQ------CD69 ------NKR-P1B 219 PSNDS------

Figure 3.4. Multiple sequence alignment of contig Ac129 with homologous members of the C- type lectin family. Arrow denotes signal peptide and asterisks denote conserved cysteine residues involved in disulfide bond formation. The conserved WIG motif is boxed. Black boxes denote identical and grey boxes denote similar residues. GenBank accession numbers for homologues are as follows: Na-CTL-2 - AF388311; CeCBG03358 - CAAC01000013; CD69 - NP_001772; NKR-P1B - AAK39100.

3.5.8 Known and Potential Vaccine Antigens Secreted and membrane-bound proteins are targets for the development of vaccines because they are exposed to the host immune response. Clones identified in this study encoded proteins with proven and predicted potential as vaccine antigens (Table 3.5). For example, vaccination with recombinant ASPs (Bethony et al., 2005, Goud et al., 2005) and astacin-like metalloproteases (Hotez et al., 2003b) partially protect dogs against challenge infection with hookworm L3 by targeting invading larvae, and contigs encoding members of these protein families were identified in this study. Vaccination of dogs with recombinant haemoglobinases results in reduced worm burdens, worm fecundity and blood loss (Asojo et al., 2005, Loukas et al., 2004). ESTs corresponding to two known haemoglobinase vaccine antigens from adult hookworms (Ac-CP-2 and Ac-APR-1) were identified from the A. caninum dataset. In addition, ESTs corresponding to four new cathepsin B cysteine proteases were identified, and some of these might be involved in digestion of the blood meal.

59 Other interesting molecules from a vaccine perspective include contigs with sequence similarities to anti-microbial alpha-defensins, tissue factor inhibitors, and integral membrane proteins involved in cell adhesion.

Figure 3.5. Neighbour joining phylogenetic tree showing the relationships of Ac173 and Na91 with other members of the Activation Associated Secretory Protein family. Numbers on branches refer to bootstrap values from 100 replicates. The names and the GenBank accession numbers of the sequences used in the phylogenetic tree are as follows: A. caninum Ac-ASP- 4 - AY217005; A. caninum Ac-ASP-6 - AY217007; N. americanus Na-ASP-2 - AY288089; Caenorhabditis elegans F49E11.5 and F49E11.9 - Z70308 (cosmid entry); Ov-VAL, venom allergen homologue (VAL) - AAB97282; Bm-VAL - AAK12274; H. contortus Hc24 - AAC47714; A. caninum neutrophil inhibitory factor (NIF) - L27427; A. caninum platelet inhibitor (HPI) - AAK81732; Homo sapiens cysteine-rich secretory protein 2 (CRISP2) - S68682.

60 Table 3.5. Contigs identified in this study with potential as vaccine antigens. Contig Function Ac8,-11,-39 Fatty acid-binding proteins Ac110,-197 Astacin metalloproteases Na250 Ac27,-41,-62 -129, C-type lectins -158,-261 Ac161,-173,-214 Activation associated secreted proteins Na91 Na56,-105,-191,-230 Cathepsin B cysteine proteases Ac247,-202 Alpha defensins Ac78 Serine protease Na58,-143,-220,-251 Tissue factor inhibitors

3.6 CONCLUSION Here, we show that LMM is a very useful technique for the dissection of defined tissues from helminth parasites. Moreover, the quality of nucleic acids recovered from these tissues is sufficient to yield at least some full-length cDNAs. The ESTs presented in this study add to the expanding catalogue of hookworm genes, but importantly, provide the first set of tissue/organ-specific cDNAs from these important blood-feeding parasites.

Acknowledgements We thank Mary Lee for technical assistance, Bennett Datu, David McMillan and Geoff Gobert for helpful discussions and advice, and Bin Zhan and Peter Hotez for providing N. americanus. We also thank Clare Hopkins from AgGenomics for conducting sequencing and Ian Smith for his support. This research was funded by QIMR, a grant from the Bill and Melinda Gates Foundation awarded to the Sabin Vaccine Initiative, and ARC Linkage Grant LP0667795.

61

CHAPTER 4: A FAMILY OF CATHEPSIN B CYSTEINE PROTEASES EXPRESSED IN THE GUT OF THE HUMAN HOOKWORM, NECATOR AMERICANUS (Molecular and Biochemical Parasitology In press)

Najju Ranjita,b, Bin Zhanc, Deborah J. Stenzela, Jason Mulvennab, Ricardo Fujiwarad, Peter J. Hotezc, Alex Loukasb aSchool of Life Sciences, Queensland University of Technology, Brisbane, QLD, Australia; bDivision of Infectious Diseases, Queensland Institute of Medical Research, Brisbane, QLD 4006, Australia; cDepartment of Microbiology, Immunology and Tropical Medicine, George Washington University Medical Center, Washington DC 20037, USA; dCellular and Molecular Immunology Laboratory, Centro de Pesquisas René Rachou, FIOCRUZ, Minas Gerais, Brazil

4.1 CONTRIBUTIONS Contributor Statement of contribution* Najju Ranjit -Designed all the experiments -Conducted all experiments except of the ones mentioned below -Interpreted and analysed all the data -Drafted manuscript

Bin Zhan -Provided cDNA clones -Provided feedback on manuscript

Deborah Stenzel -Discussed experimental design -Provided feedback on manuscript

Jason Mulvenna -Conducted proteomic analyses -Provided feedback on manuscript

Ricardo Fujiwara -Provided L3 larvae -Provided feedback on manuscript

Peter Hotez -Investigator on the grant that funded the project -Provided feedback on manuscript

Alex Loukas -Aided in experimental design -Aided in data analysis -Aided in drafting manuscript

Principal Supervisor Confirmation

I have sighted email or other correspondence from all Co-authors confirming their certifying authorship.

Alex Loukas 27-5-08 Name Signature Date

63 4.2 ABSTRACT mRNAs encoding cathepsin B-like cysteine proteases (CatBs) are abundantly expressed in the genomes of blood-feeding nematodes. Recombinant CatBs have been partially efficacious in vaccine trials in models of hookworm infection, supporting further investigation of these enzymes as new control tools. We recently described a family of four distinct CatBs (Na-CP-2,-3,-4,-5) from the human hookworm, Necator americanus. Here we show that these N. americanus CatBs form a robust clade with other hookworm CatBs and are most similar to intestinal CatBs from Haemonchus contortus. All four mRNAs (Na-cp-2, -3, -4 and -5) are upregulated during the transition from a free-living larva to a blood-feeding adult worm and are also expressed in gut tissue of adult N. americanus that was dissected using laser microdissection microscopy. Recombinant Na-CP-3 was expressed in soluble, secreted form in the yeast, Pichia pastoris, while Na-CP-2, -4 and -5 were expressed in insoluble inclusion bodies in Escherichia coli. Recombinant Na-CP-3 was not catalytically active when secreted by yeast but underwent auto-activation to an active enzyme at low pH in the presence of dextran sulphate. Activated Na-CP-3 digested gelatin and cleaved the fluorogenic cysteine protease substrate Z-Phe-Arg- aminomethylcoumarin (AMC) but not Z-Arg-Arg-AMC. Recombinant Na-CP-3 did not digest intact hemoglobin, but digested globin fragments generated by prior hydrolysis with N. americanus aspartic hemoglobinases. Antibodies raised in mice to all four recombinant proteins showed minimal cross-reactivity with each other, and each antiserum bound to the intestine of adult N. americanus, supporting the intestinal expression of their mRNAs. These data show that N. americanus expresses a family of intestinal CatBs, many of which are likely to be involved in nutrient acquisition and therefore are potential targets for chemotherapies and vaccines.

Keywords: Necator americanus; Cysteine protease; Cathepsin B; Hookworm; Hemoglobin

64 4.3 INTRODUCTION Proteolytic enzymes are essential components of biological systems, ranging from viruses to vertebrates. They have been divided into groups on the basis of the catalytic mechanisms used during their hydrolytic processes (Rawlings et al., 2008). There are five major catalytic types: aspartic, cysteine, metallo, serine and threonine proteases. In addition to these five classes, it is believed that there are others which are as yet undefined. The most abundantly represented family of proteases in parasitic nematodes, particularly the blood-feeding strongyle nematodes, is the papain-like cysteine protease family (family C1, MEROPS classification) (Sajid and McKerrow, 2002). Cysteine proteases from parasitic helminths perform numerous functions, including house-keeping roles such as protein processing and turnover, degradation of host proteins such as the extracellular dermal matrix during skin penetration, hydrolysis of hemoglobin (Hb) for nutritional uptake, and inhibition of host protective immune responses, including cleavage of immunoglobulin and complement components to evade host immunity (reviewed by (Beckham et al., 2006, Knox et al., 1999, Bethony et al., 2005)). Cysteine proteases of parasitic organisms are divided into two main groups; clan CA and clan CD. Clan CA or ‘papain like’ proteases are further divided into two families, C1 and C2. Most parasitic helminth proteases belong to the C1 family, which contains the cathepsin B- and cathepsin L-like enzymes (Sajid and McKerrow, 2002). In contrast, the C2, or calpain-like family of proteases appear to be less well represented in parasitic helminths. Clan CD contains the C13 family, which are referred to as legumain-like or asparaginyl endopeptidases. Cathepsin B-like cysteine proteases (CatBs) generally constitute large multi- gene families in both parasitic and non-parasitic helminths, including Haemonchus contortus, Schistosoma sp. and Caenorhabditis sp. In parasitic helminths, CatBs are often expressed in the alimentary canal of the worm, where they play proven or suspected roles in digestion of protein for nutrient acquisition. In blood-feeding nematodes, the CatB gene family has undergone enormous expansion. CatB- encoding genes account for 16-17% of intestinal transcripts from H. contortus (Jasmer et al., 2001, Jasmer et al., 2004), representing the most abundant and diverse protein family in the gut of this parasite. Schistosoma mansoni ingests and lyses red

65 blood cells then digests Hb with a multi-enzyme network of proteases including the CatB, SmCB1 (Lipps et al., 1996, Delcroix et al., 2006, Caffrey et al., 2004, Brindley et al., 1997). CatBs are also highly expressed in the hookworm gut. In a previous study, we used laser microdissection microscopy to dissect gut tissue from adult hookworms and showed that mRNAs encoding for C1 cysteine proteases (and other classes of proteases) were abundantly represented in mRNAs from the gastrointestinal tract of the blood-feeding stage (Ranjit et al., 2006). Intestinal cysteine proteases are showing particular promise as efficacious vaccine antigens against numerous helminth parasites (reviewed in (Bethony et al., 2005, Dalton et al., 2003)). Cattle and sheep can be protected against infection with the liver fluke, Fasciola hepatica, by vaccination with the major secreted cysteine proteases (Dalton et al., 2003). Vaccination with protease-rich gut membrane extracts of H. contortus results in upwards of 90% reduction in adult worms and fecal egg burdens in lambs (reviewed in (Knox et al., 2001, Dalton et al., 2003)). The adult stage of the dog hookworm, Ancylostoma caninum, digests Hb using a semi-ordered pathway of proteolysis that involves the CatB, Ac-CP-2 (Williamson et al., 2004). Vaccination of dogs with Ac-CP-2 provided proof-of- concept that hemoglobinolytic proteases (hemoglobinases) are efficacious vaccine antigens against hookworm infection (Loukas et al., 2004). Catalytically active recombinant Ac-CP-2 conferred partial protection to laboratory beagles, resulting in a significant decrease in fecal egg counts and worm sizes. Moreover, the worms that were recovered from CP-2-vaccinated dogs at necropsy had anti-CP-2 IgG adhered to the parasite gut, and IgG from vaccinated dogs neutralized the hemoglobinolytic activity of recombinant Ac-CP-2 in vitro (Loukas et al., 2004). Here, we characterize the transcriptional profile, anatomical expression sites and recombinant expression of four CatBs from adults of the human hookworm, Necator americanus, and discuss their potential as vaccine antigens for human .

66 4.4 MATERIALS AND METHODS 4.4.1 Phylogenetic tree The phylogenetic relationships between N. americanus CatBs and other C1 family peptidases were assessed based on amino acid sequence. The full length ORFs of Na- CP-2, CP-3, CP-4 and CP-5 were aligned with homologues from other species using ClustalW. A phylogenetic tree was constructed with distance matrix using the neighbor-joining method with 1,000 bootstrap samplings in PAUP beta version 8.0 for Macintosh.

4.4.2 Amplification of cysteine proteases genes from gut cDNA Cloning of Na-cp-2, -3, -4 and -5 cDNAs has been reported previously (Xiao et al., 2008). We began our terminology from “Na-cp-2” and avoided the term “Na-cp-1”, because the first N. americanus cysteine protease cDNA from N. americanus deposited in GenBank is called “necpain” (GenBank - AJ132421) - there is no corresponding publication in the literature describing necpain, but to avoid confusion, we did not use the term “Na-cp-1” when naming the cDNAs described herein (Xiao et al., 2008). To confirm that Na-cp-2, -3, -4, -5 mRNAs were expressed in the gut of adult N. americanus, gut tissue extracted by laser microdissection microscopy (LMM) (Ranjit et al., 2006) was used as template for PCR. The following primers were used: Na-cp-2F: 5’- TCTGTTTCGCTGGTTGAGCC (nt 61); Na-cp-2R: 5’- TCCTGTCGGTCATCACCTCTGC (nt 435); Na-cp-3F: 5’- GAGGAGGCAGAGAATCTTTC (nt 82); Na-cp-3R: 5’- GCAACAGGCAAGGATGTCCG (nt 456); Na-cp-4F: 5’- CGTCGTCCTTCTGGCAATAAAC (nt 123); Na-cp-4R: 5’- GCGAATGAGACCGATGGAGG (nt 423); Na-cp-5F: 5’- TTCCCCGCTGGTTGAACAG (nt 300); Na-cp-5R: 5’ – CCATCGCATCCCATTCCG (nt 617). Necator americanus alpha tubulin mRNA (GenBank accession no. BG467527) was amplified as a constitutively expressed control using the primers Na-α tubulinF: 5’ – CGAATCTCGTGCCATATCCT (nt 157), Na-α tubulinR: 5’ – TTCCTCCATACCCTCACCGA (nt 628). Gonadal tissue cDNA which was

67 extracted by LMM in similar fashion as that of intestinal tissue (Ranjit et al., 2006) was used as a template in PCR as a negative control.

4.4.3 Quantitation of cysteine protease gene expression in different developmental stages Real-time PCR reactions were conducted simultaneously for all candidate genes using a Rotor-Gene 6000 thermal cycler (Corbett). Amplified products were detected with SYBR Green I DNA binding dye. Single-stranded cDNA was prepared from infective third stage larvae (L3) and adult worms of N. americanus, as reported elsewhere (Datu et al., 2008), quantified spectrophotometrically, diluted to 50 ng/μl with water and used as template for PCR. In each 20 μl reaction, 10 μl of SYBR green super-mix (Applied Biosystems) was added to 100 nM of each primer with 250 ng of cDNA. All experiments were repeated twice with three replicates in each run using the following cycle conditions: 3 min at 95°C, 45 cycles of 1 min at 95°C, 30 secs at 55°C and 30 secs at 72°C. A melt curve analysis step, included at the end of each run, verified the absence of primer–dimers and non-specific products. Changes in the expression of transcripts between L3 and adult worms were normalized to the 60S acidic ribosomal protein gene (GenBank accession no. BG734493). The following primers were used: Na-cp-2F: 5’- GCTCAAGAACGCATGAAATC (nt 205); Na-cp-2R: 5’- GAAGGACATTCTGGCCATT (nt 359); Na-cp-3F: 5’- CCGACGACAAATACTACGC (nt 680); Na-cp-3R: 5’- GCCTGAAGTCACATAAACTCC (nt 834); Na-cp-4F: 5’- CGTCCTTCTGGCAATAAACC (nt 126); Na-cp-4R: 5’- TGTTCATTGGTTGGCGAATA (nt 272); Na-cp-5F: 5’- CGATAGACAATGGTGTATGC (nt 647); Na-cp-5R: 5’ – CTATTTCTGGTTTTGGTGGC (nt 747), Na-60SF: 5’ – GTCGGAATCGTCGGAAAGTA (nt 39); Na-60SR: 5’ – GTCTTGTTGCACTTCGAGCA (nt 205).

4.4.4 Expression and purification of recombinant cysteine proteases We attempted to express all four proteases in the yeast, Pichia pastoris, however only Na-CP-3 was produced in sufficient yields for use in this study. The entire ORF encoding the pro-enzyme of Na-CP-3 (GenBank accession no. ABL825237), excluding the predicted signal peptide, was cloned into the expression

68 vector pPICZαA (Invitrogen) according to the manufacturer’s instructions. The correct reading frame was confirmed by sequencing the plasmid using the α-factor and 3’AOX1 vector-derived primers. The recombinant plasmid was linearized with the enzyme PmeI and transformed into P. pastoris X-33 strain by electroporation. The transformants were selected on Zeocin (Invitrogen) Yeast Peptone Dextrose (YPD) plates. Colonies were screened for cDNA insertion by PCR using gene specific primers. Positive colonies were tested for protein expression as recommended by the manufacturer using Western blot with an anti-hexa His monoclonal antibody (Invitrogen). The clone exhibiting the highest expression was scaled-up to 1.5 L suspension culture. Cells were harvested 96 hours post-induction and supernatant was collected, concentrated to 200 ml by ultrafiltration using a 10 kDa cut off membrane (Pall Scientific) and buffer exchanged into binding buffer

(50mM NaH2PO4, 300mM NaCl and 10 mM imidazole). The recombinant protein was purified by affinity chromatography on a Ni-NTA agarose column (Qiagen), and purified protein was buffer exchanged into phosphate buffered saline (PBS) and protein concentration determined using the bicinchoninic acid assay kit (Pierce). For expression of Na-CP-2 (GenBank accession no. ABL85236), Na-CP-4 (GenBank accession no. ABL85238) and Na-CP-5 (GenBank accession no. ABL85239), the ORFs without the signal peptides were cloned in frame into the pET41a vector (Novagen) and the recombinant products were transformed into E. coli strain BL21-DE2 (Invitrogen). Cultures were induced with isopropyl β-D- thiogalactopyranoside (IPTG) at a final concentration of 1mM. Four hours post- induction, the culture media were centrifuged at 20,000 g for 20 mins and cell pellets were collected. The cell pellets were resuspended in lysis buffer (20 mM NaH2PO4, 500 mM NaCl) and subjected to two cycles of disruption in a French press (SLM Instruments). Lysates were centrifuged at 20,000 g for 15 mins and pellets were incubated in denaturing solubilization buffer (6 M GuHCl, 0.5 M NaCl, 50 mM Tris, 10 mM imidazole) for 2-4 hrs at RT. Recombinant proteins were purified by affinity chromatography using Ni-NTA agarose (Qiagen) according to the manufacturer’s instructions.

4.4.5 Autoactivation, catalytic activity assays. To trans activate recombinant Na-CP-3 from its pro-form (as secreted by P. pastoris) to its mature form, 100 μg of recombinant protein was incubated for 24

69 hours in AMT buffer (100mM sodium acetate, 100mM MES, 200 mM Tris, 4 mM EDTA, 200 mM NaCl), 50 μg/ml dextran sulfate (DS 500K) and 10 mM DTT at one pH unit increments from pH 4-7. The processed Na-CP-3 protein was then assayed for a shift in molecular weight by western blot analysis using an anti-hexa His antibody and catalytic activity against the fluorogenic peptide substrates benzyloxycarbonyl-L-phenylalanine-L-arginine-7-amido-4-methyl-coumarin (Z-Phe- Arg-AMC; Bachem) or Z-Arginine-ArginineAMC (Z-Arg-Arg; Bachem). Briefly, 10 μg of protein was added to the assay buffer containing 100 mM sodium acetate, 100 mM NaCl, 500 μM DTT to which either fluorogenic peptide was added to a final concentration of 10 μM. The cysteine protease inhibitor E64 (Sigma) was included in some reactions at a final concentration of 5 μM. Papain (Sigma) at a final concentration of 1 μM and adult hookworm Excretory/Secretory (ES) products were used as positive controls. Reactions were incubated at 37oC for up to 6 hrs and cleavage of AMC was measured on a Fluostar Galaxy microtitre plate reader (BMG Labtech) using excitation and emission wavelengths of 370 nm and 440 nm respectively. To further assess the catalytic activity of Na-CP-3, recombinant protein was electrophoresed on precast gelatin zymogram gels (Invitrogen) as recommended by the manufacturer with a slight modification - protease assay buffer (100 mM sodium acetate, 100 mM NaCl and 10 mM DTT) was added to the zymogram developing reagent. Adult hookworm ES products were used as a positive control for the zymogram, and E64 was included in some assays to assess inhibition of enzymatic activity. To verify if recombinant Na-CP-3 was being glycosylated by P. pastoris, the purified protein was electrophoresed on a 10% SDS-PAGE gel and was incubated with Pro-Q Emerald 300 glycoprotein gel stain (Molecular Probes), as per the manufacturer’s instructions. To determine whether activated recombinant Na-CP-3 cleaved intact Hb or globin peptides, 100 μg of Hb tetramer or Hb that had been pre-digested with recombinant N. americanus aspartic protease Na-APR-1 (GenBank accession number. CAC00543) (Acosta et al.2008) was incubated with activated Na-CP-3 protein (20 μg) in 0.1 M sodium acetate at pH 4.5 at 37oC. Samples were assessed visually for cleavage of Hb by SDS-PAGE under native conditions, or using LC-MS to assess cleavage of globin peptides (as described in section 4.4.5).

70

4.4.6 Identification of the pro-mature Na-CP-3 junction To identify the cleavage site between the pro-region and mature protease of Na-CP-3, protein was reduced and alkylated with DTT and ioadoacetamide and digested with trypsin. Briefly, the peptides were injected onto a 150 μm C18 reverse phase capillary column (Alltech) attached to an Ultimate 3000 nanoLC system (Dionex) and eluted using a 50 min gradient from 5 to 55% acetonitrile. The mass spectrometer (MicrOTOF-Q, Bruker) used an autoMSn methodology that collected MS2 spectra for the two most intense ions in each full scan spectrum. Post- acquisition spectral analysis was conducted using Biotools (Bruker).

4.4.7 Antibody production Antibodies against all four purified recombinant proteins were raised in female BALB/c mice (three mice per group). Pre-immune sera were collected by tail bleed two days prior to the first injection. For the first immunization, mice were injected in the tail subcutaneously with 25 μg of protein emulsified with an equal volume of Freund’s complete adjuvant. Mice were boosted intraperitoneally, twice at two weekly intervals, with 25 μg of protein emulsified with Freund’s incomplete adjuvant. Two weeks after the final boost, mice were euthanized and blood was collected via cardiac puncture. Blood from all 3 mice in each group was pooled. To separate serum from cells, blood was incubated at 37oC for one hour and then centrifuged at 10,000 g for 10 mins. Antibody endpoint titers were determined by enzyme linked immunosorbent assay (ELISA), following the method of (Beckham et al., 2006). To test for immunologic cross-reactivity between the different anti-CatB sera, Western blot analyses were conducted by probing 1 μg of all four recombinant cysteine proteases with mouse antisera to each enzyme, diluted 1:10,000 in antibody dilution buffer (PBS/0.05% Tween-20/5% skimmed milk powder) for 1 hr at RT. After washing 3 times for 5 mins each in PBS/0.05% Tween-20, blots were incubated with goat anti-mouse IgG conjugated to horse radish peroxidase (Chemicon) diluted 1:2,000 in antibody dilution buffer.

4.4.8 Immunolocalization Adult N. americanus recovered from euthanized hamsters were fixed in 4% formaldehyde and placed in OCT compound. OCT blocks were frozen on dry ice for

71 5 mins. Frozen blocks of fixed worms were cut by cryostat into 7 μm thick transverse sections and mounted onto superfrosted slides and stored at -20oC until needed. To rehydrate the sections, slides were incubated in PBS for 5 mins. Non-specific binding was inhibited by blocking the sections with 5% fetal calf serum in PBS for one hour at room temperature (RT). Slides were incubated with mouse antisera at various dilutions (1:100, 1:250, 1:500, 1:1,000) with 1% Bovine Serum Albumin (BSA) in PBST at RT for 1 hr. Anti-mouse IgG conjugated to Alexa Fluor 555 (Invitrogen) was used as fluorescent secondary antibody at a dilution of 1:500 in 1% BSA/PBST at RT for 1 hour. Slides were washed three times with PBST for 5 mins each, air dried and coversliped with mounting media (Sigma). Slides were viewed and photographed using a Leica IM100 fluorescent microscope.

4.5 RESULTS 4.5.1 Sequence analysis of the cysteine proteases identified in N. americanus Four cDNAs encoding distinct CatBs were cloned from a N. americanus L3 cDNA library by Xiao et al. (2008); these were designated Na-cp-2, Na-cp-3, Na-cp- 4 and Na-cp-5, but their sequence features were not assessed in depth. All of the ORFs consisted of a hydrophobic signal peptide at the N-terminus, followed by a pro- region of between 72-77 amino acids (aa) and a mature protease sequence (Table 4.1). All four proteases possessed the highly conserved residues of the catalytic triad (Cys, His, Asn) as well as the oxyanion Gln. None of the four proteases had the ERFNIN motif in their pro-regions (characteristic of non-cathepsin B-like C1 proteases (Karrer et al., 1993)). All four proteases had the occluding loop that is diagnostic of CatBs (Illy et al., 1997), however the loop was modified in Na- CP-2 and CP-5, both of which contained just one of the conserved His doublet. Of the four CatBs, only Na-CP-5 did not possess the haemoglobinase motif, as described by (Baig et al., 2002). Interestingly, none of the CatBs had the Glu residue at the base of the S2 pocket which is influential in determining substrate specificity (Fig. 4.1) (Sajid and McKerrow, 2002). The four proteases shared 50-70% amino acid identities with each other.

72 4.5.2 Phylogenetic analysis of cathepsin B-like proteases A neighbour joining tree was constructed to compare the phylogenetic relationships between the CatBs of N. americanus and other blood-feeding parasitic and non-parasitic nematodes (Fig. 4.2). The outgroup for the tree was human cathepsin F. All of the nematode CatBs formed one clade but it did not receive greater than 50% bootstrap support. Within the nematode clade, the proteins were further divided into more robust clades based on the species from which they were derived. The five N. americanus proteases grouped with CatBs from Ancylostoma spp. in a clade that obtained 69% bootstrap support. The hookworm CatBs grouped with a clade of H. contortus CatBs to form a blood-feeding nematode clade with 61% bootstrap support.

Table 4.1. General properties of N. americanus cathepsin B-like proteases cDNA Size (bp) ORF size Mature protein properties Predicted (aa) N-glycoslyation (aa, MW, pI) sites Na-cp-2 1134 347 262aa, 27.9 kDa, pI 6.36 140 (NGT) Na-cp-3 1174 360 270aa, 30.1 kDa, pI 8.8 32 (NLS) 136 (NGT) 242 (NET) 298 (NGT) Na-cp-4 1181 339 252aa, 28.3 kDa, pI 5.56 134 (NGT) 296 (NGT) Na-cp-5 1236 342 254aa, 28.4 kDa, pI 8.15 101 (NCT) 135 (NGT) 245 (NET)

4.5.3 Amplification of cysteine protease mRNAs from N. americanus gut cDNA All four mRNAs were amplified by PCR from cDNA which had been synthesised from tissue dissected from the gut of adult N. americanus (Fig. 4.3). None of the mRNAs were amplified from dissected gonad tissue (not shown) which served as a control.

4.5.4 Developmental expression of cysteine protease genes Expression patterns of Na-cp-2-, -3, -4 and -5 mRNAs were determined from N. americanus L3, whole adults and dissected adult gut using real-time PCR. In L3 cDNA, comparatively low expression levels were detected for Na-cp-3, -4 and -5,

73 and Na-cp-2 could not be detected above the negative control signal (Fig. 4.4). This is somewhat surprising given that all four cDNAs were cloned from an L3 cDNA library. All four mRNAs were highly upregulated in the adult stage, with increases of 229-, 60- and 70-fold for Na-cp-3, cp-4 and cp-5 respectively (Fig. 4.4). All four mRNAs were more highly expressed in gut tissue than whole worm - both Na-cp-2 and cp-5 had 4-fold increases, while Na-cp-3 and -cp-4 underwent 20- and 70-fold increases respectively (Fig. 4.4).

74 Na-CP-2 MLTLAALLISVSLVEPTGIGEFLAQPAPAYARRLTGQALVDYVNSHHSLYKAKYSPDAQE Na-CP-4 MKANFALVVVLLAINQLYADELLHKQESEHG--LSGQALVDYVNSHQSLFKTEYSPTNEQ Na-CP-5 MITIITLLLIASTVKSLTVEEYLARPVPEYATKLTGQAYVDYVNQHQSFYKAEYSPLVEQ Na-CP-3 -LILIALVVTALAQQPLSLKEYLEQPIPEEAENLSGEAFAEFLNKRQSFFTAKYTPNALN Necpain MLLFLTLFVAILAAD----EKILQDAVKKESKALTGHALAEFLRTLQSLFEVKKSEEVPV Human_CatB MWQLWASLCC-LLVLAN------ARSR-PSFHPLSDEL-VNYVNKRNTTWQAGHNFYNVD

Na-CP-2 RMKSRIMDLSFMVDAEVMMEEMDQQEDIDLAVSLPESFDAREKWPECPSIG-LIRDQSAG Na-CP-4 FVKARIMDIKYMTEASHKYPRK----GINLNVELPERFDAREKWPHCASIG-LIRDQSAC Na-CP-5 YAKAVMRSEFMTKPNQNYVVKD-----VDLNINLPETFDAREKWPNCTSIR-TIRDQSNC Na-CP-3 ILKMRVMESRFLDNEEGEMLKE---EDMDFSEEIPVSFDARDKWPKCTSIG-FIRDQSHC Necpain RMKYLLPKHFMVKPKEEDRTKIQ------LDKEPPEKFDARDAWPYCREIIGHVRDQSRC Human_CatB MSYLKRLCGTFLGGPKPPQRVMFT-----EDLKLPASFDAREQWPQCPTIK-EIRDQGSC

Na-CP-2 GGCWAVSSAEVMTDRICIQSNGTKQVYVSETDILSCCGQRCGSGCTSGVPRQAFNYAIRK Na-CP-4 GSCWAVSAASVMSDRLCIQTNGTNQKILSSADILACCGEDCGSGCEGGYPIQAYFYLENT Na-CP-5 GSCWAVSAASVMSDRLCIQSNGTIQSWASDTDILSCCWN-CGMGCDGGRPFAAFFFAIDN Na-CP-3 GSCWAVSSAETMSDRLCVQSNGTIKVLLSDTDILACCPN-CGAGCGGGHTIRAWEYFKNT Necpain GSCWAVSAASVMSDRLCVQSNGKIKLHVSDTDILACCGEFCGDGCSGGWPFQAWEWVRKY Human_CatB GSCWAFGAVEAISDRICIHTNAHVSVEVSAEDLLTCCGSMCGDGCNGGYPAEAWNFWTRK

Na-CP-2 GVCSGGPYGTKGVCKPYPFYPCGYHAHLPYYGPCP-DGMWPTPTCEKACQSDYTVPYNDD Na-CP-4 GVCSGGEYREKNVCKPYPFYPCDG-----NYGPCPKEGAFDTPKCRKICQFRYPVPYEED Na-CP-5 GVCTGGPFREPNVCKPYAFYPCGRHQNQKYFGPCP-KELWPTPKCRKMCQLKYNVAYKDD Na-CP-3 GVCTGGLYGTKDSCKPYAFYPCKD----ESYGKCP-KDSFPTPKCRKICQYKYSKKYADD Necpain GVCTGGDYRAKGVCKPYAFHPCGNHENQVYYGVCP-KGSWPTPRCEKFCQRGYIKPYKKD Human_CatB GLVSGGLYESHVGCRPYSIPPCEHHVN---GSRPPCTGEGDTPKCSKICEPGYSPTYKQD

Na-CP-2 RIFG--SKTIVLTGEEKIKREIFNNGPLVATYTVYEDFAYYKNGIYMTGLGRATGAHAVK Na-CP-4 KVFGKNSHILLQDNEARIRQEIFINGPVGANFYVFEDFIHYKEGIYKQTYGKWIGVHAIK Na-CP-5 KIYG-NDAYSLPNNETRIMQEIFTNGPVVGSFSVFADFAIYKKGVYVSNGIQQNGAHAVK Na-CP-3 KYYA-NSAYRIPQNETWIKLEIMRNGPVTASFRIYPDFGFYEKGVYVTSGGRELGGHAIK Necpain KFYA-KKSYWLPNDEKEIRLDIMKNGPVQAAFDVYEDFKLYKRGIYKHKEGIQTGGHAVK Human_CatB KHYG-YNSYSVSNSEKDIMAEIYKNGPVEGAFSVYSDFLLYKSGVYQHVTGEMMGGHAIR

Na-CP-2 IIGWGEENG----VKYWLIANSWNTDWGEN-GFFRMLRGTNLCDIELSATGGTFKV---- Na-CP-4 LIGWGTENG----TDYWLVANSWNYDWGEN-GTFRILRGTNHCLIESQVIATEMIV---- Na-CP-5 IIGWGVQDG----LKYWLIANSWNNDWGDE-GYVRFLRGDNHCGIESRVVTGTMKV---- Na-CP-3 IIGWGTEKVNGTDLPYWLIANSWGTDWGENNGYFRILRGQNHCQIEQKVIAGMIKVPQPK Necpain IIGWGKDNG----TDYWLIANSWSKDWGES-GFFRMVRGENDCEIEDMITAGIMMV---- Human_CatB ILGWGVENG----TPYWLVANSWNTDWGDN-GFFKILRGQDHCGIESEVVAGIPRTDQYW

Na-CP-2 ------Na-CP-4 ------Na-CP-5 ------Na-CP-3 SAGPPLQPNPSS Necpain ------Human_CatB EKI------

Figure 4.1 Multiple sequence alignment of N. americanus cysteine proteases and human cathepsin B. Blue font: Predicted signal peptide cleavage points, Aqua shading: Predicted cleavage points of mature and pro-domains, Red shading: Glutamine oxyanion hole, Pink shading: Catalytic triad - Cysteine, Histidine, Asparagine, Green shading: Occluding loop (Histidine doublet is underlined), Yellow shading: Haemoglobinase motif, Grey shading: Protease S2 pocket (arrow denoting Gln residue) Necpain GenBank accession number: CAB53364, Human CatB GenBank accession number: AAH10240

75

Figure 4.2. Neighbour joining phylogenetic tree depicting the relationships of N. americanus cysteine proteases with homologues from other nematodes and other phyla. GenBank accession numbers are listed next to the protein names. Sm: S. mansoni, Ce: C. elegans, Hc: H. contortus

76 . Figure 4.3. Amplification of cysteine protease mRNAs from N. americanus gut cDNA. Lane 1: Na-cp-2, Lane 2: Na-cp-3, Lane 3: Na-cp-4, Lane 4: Na-cp-5.

108 107

106 105 L3 104 Adult

3 10 Gut 102 1 mRNA transcript level level transcript mRNA

(copies per per reaction) (copies 10

0 Na-cp-2 Na-cp-3 Na-cp-4 Na-cp-5 Na-60s N. americanus cysteine proteases

Figure 4.4. Developmental expression profiles of N. americanus cysteine protease mRNAs. Quantitation of mRNA levels for Na-cp-2, cp-3, cp-4 and cp-5 by real time PCR from infective larvae (L3), adult worms and adult worm gut tissue. Expression profiles are depicted as mRNA copy numbers using N. americanus 60S ribosomal RNA as a constitutively expressed control. Star denotes statistically significant difference of expression between L3 and adult cDNA (p≤0.05). Diamond denotes statistically significant difference of expression between adult and gut cDNA (p≤0.05).

4.5.5 Expression of recombinant cysteine proteases Na-CP-3 was expressed in soluble form in P. pastoris X-33 strain. Protein was detected by Western blot analysis using anti-hexa-His and anti-c-myc monoclonal antibodies (Invitrogen). The protein was secreted into the culture supernatant at ~3.5 mg/L and was purified via the nickel-NTA affinity chromatography under native conditions. Na-CP-2, -4 and -5 were expressed as recombinant proteins in E. coli strain BL21-DE2 in the vector pET41a. They were all expressed in insoluble inclusion bodies and purified by nickel-NTA affinity chromatography under denaturing conditions. Na-CP-2, CP-4 and CP-5 expressed at 4 mg/L, 5.5 mg/L and 7.5 mg/L respectively (Fig. 4.5). Attempts to remove urea and refold the proteins into a soluble form were unsuccessful (not shown).

77

Figure 4.5. Expression and purification of recombinant N. americanus CatBs in yeast P. pastoris and E. coli. Na-CP-2 (A), CP-3 (B), CP-4 (C) and CP-5 (D). Na-CP-2, CP-4 and CP-5 were expressed in E. coli; Na-CP-3 was expressed in P. pastoris. Lanes show protein purification steps on Ni-NTA resin. Lanes 1: Before binding, Lanes 2: Flow through, Lanes 3: Wash 1, Lanes 4: Wash 2, Lanes 5: Elution.

4.5.6 Catalytic activity of Na-CP-3 Purified recombinant Na-CP-3 did not cleave either Z-Phe-Arg-AMC or Z- Arg-Arg-AMC, and the molecular weight of the purified protein (55kDa) indicated that the enzyme had not undergone post-translational processing from its pro- form. LC-MS analysis of the purified protein revealed that the 77 amino acid pro-domain was still present. When the same recombinant protein was electrophoresed on a gelatin gel, however, a zone of hydrolysis was visible at the predicted size of the pro- enzyme, indicating that the renaturation process employed in gelatin zymography was at least partially activating the pro-enzyme (Fig. 4.6). The gelatinolytic activity was inhibited by E64 (cysteine protease inhibitor), confirming that this activity was due to recombinant Na-CP-3. The recombinant CP-3 bound Pro-Q Emerald 300

78 glycoprotein stain, indicating that the protein had been glycosylated (Fig. 4.7a). In order to facilitate trans processing of the pro- domain of Na-CP-3, the purified protein was incubated in AMT buffer (containing dextran sulphate) at a range of pH values. Western blotting of the processed protein samples indicated a shift in molecular weight of the expected size from 55 kDa to ~30 kDa at pH 4 (Fig 4.7b). Catalysis assays with the various samples indicated that auto-activated Na-CP-3 cleaved Z-Phe-Arg-AMC but not Z-Arg-Arg-AMC. Pro-enzyme that was activated at pH 4 yielded the highest catalytic activity, followed by pro-enzymes activated at pH 5 and pH 6 respectively (Fig. 4.8). No activity was detected from samples which had been trans activated at pH 3, 7 or 8 (data not shown). Activated proteases displayed catalytic activity against Z-Phe-Arg-AMC from pH 4 to 7, with highest activity between pH 6 and 7 (Fig 4.8). Na-CP-3 did not digest intact Hb tetramer but did digest globin peptides that were generated by hydrolysis of Hb with the aspartic protease, Na-APR-1 (data not shown), indicating a downstream role in the proposed hemoglobinolysis cascade (Williamson et al., 2003b). The nature/identities of the peptides that were generated by digestion of globin with Na-CP-3 are not shown here and have been submitted elsewhere for publication as just one component of a multi-enzyme hemoglobinolytic cascade (Ranjit et al., manuscript submitted).

Figure 4.6. Gelatin zymogram showing catalytic activity of purified recombinant Na-CP-3. Lane 1: 10 ng Na-CP-3, Lane 2: 1 ng Na-CP-3, Lane 3: 10 ng Na-CP-3 + E64

79

Figure 4.7. SDS-PAGE gel of purified recombinant pro-Na-CP-3 incubated with Pro-Q Emerald 300 glycoprotein stain (A). Activation of pro-Na-CP-3 (B). Western blot of purified recombinant Na-CP-3 activated in AMT buffer. Lane 1: no incubation, Lane 2: incubated O/N in AMT buffer, pH 4, Lane 3: incubated O/N in AMT buffer, pH 5, Lane 4: incubated O/N in AMT buffer pH 6.

Figure 4.8. pH profile of the catalytic activity of recombinant Na-CP-3 after auto-processing. Autoprocessing was facilitated in AMT buffer and conducted at pH 4, 5 or 6. The processed proteases were then assessed for catalytic activity at different pH values (x axis) against the fluorogenic substrate Z-Phe-Arg-aminomethylcoumarin. RFU – relative fluorescence units.

4.5.7 Antibody production and immunolocalization of proteins Antisera were raised in BALB/c mice to all four proteases. Mice generated antibody endpoint titers of 1:30,000 to 1:45,000 as determined by ELISA. Western blot analyses conducted with all four antisera indicated specificity of each antibody for its homologous enzyme, but there was slight cross reactivity between anti-Na- CP-3 serum which strongly recognized Na-CP-3 and weakly bound to Na-CP-4, -CP-

80 5 and –CP-2. Anti-Na-CP-4 serum strongly bound to Na-CP-4 and very weakly bound to Na-CP-5 (Fig. 4.9). Immunolocalization of each protease within adult N. americanus tissue sections demonstrated binding of all four antibodies to the gut of the parasite (Fig. 4.10), supporting the intestinal expression of the corresponding mRNAs. Antisera to Na-CP-2 and Na-CP-4 bound strongly to the gut but also bound weakly to the cuticle and/or hypodermis. We routinely see weak binding with some antibodies to the cuticle/hypodermis region, and believe that this represents non- specific antibody binding. Anti-Na-CP-3 showed the strongest immunofluorescent labelling and this was clearly localised to the microvillar surface of the gut (Fig 4.10). The pre-immune serum did not bind to any structures (Fig. 4.10).

Figure 4.9. Western blot showing recognition of recombinant Na-CP-2, CP-3, CP-4 and CP-5. (listed under each panel) by antisera raised to each recombinant enzyme (listed above each panel).

4.6 DISCUSSION The intestine of blood-feeding helminths is a protease-rich site (Ranjit et al., 2006, Jasmer et al., 2001, Caffrey et al., 2004) and at least some of these proteins play defined roles in nutrient acquisition. Cysteine proteases have been shown to be one of the most abundantly expressed protease families in the gastrointestinal tracts of parasitic helminths. These proteases are often developmentally regulated throughout the complex parasitic life cycles, are frequently up-regulated in actively feeding stages and have been shown to be commonly expressed in organs or organelles involved in feeding (Jasmer et al., 2004). Here, we examined the transcriptional expression, anatomical localisation and characterisation of four different N. americanus cathepsin B- like proteases in order to gain an understanding of what roles they might play and whether they would be effective as chemotherapy or vaccine targets. .

81 A B

cut

cut cut cut

int int

int mv int mv

C D

cut int int int int ceph ceph

cut

E

cut cut

int int 50 µm Figure 4.10. Immunolocalization of Na-CP-2, -3, -4 and -5 in transverse sections of adult N. americanus. Fluorescence images are on the left and corresponding bright field images are on the right for each antibody. Arrows indicate the intestine (int), intestinal microvillar surface (mv), cephalic gland (ceph) and cuticle (cut). A: α-Na-CP-2, B: α-Na-CP-3, C: α-Na-CP-4, D: α-Na-CP-5, E: Pre-vaccination serum.

82 All four N. americanus proteinases have been classified as CatBs due to their structure and sequence homology to cathepsin B proteins. They all have the essential catalytic triad residues of cysteine, histidine and asparagine as well the highly conserved glutamine that forms the oxyanion hole. In addition the conserved glutamine is a crucial element in forming an electrophilic centre to stabilise the tetrahedral intermediate during hydrolysis (Sajid and McKerrow, 2002). In our study, three out of the four N. americanus CatBs, Na-CP-2, CP-3 and CP-4, contained the so called ‘hemoglobinase motif’ described by (Baig et al., 2002). This motif is located around the catalytic Asn and is thought to be diagnostic of hemoglobinase activity. These authors suggested that CatBs which lack this motif would not be able to degrade Hb readily and may play more generalised housekeeping functions. Although there is sufficient evidence to indicate that CatBs which lack this motif have limited abilities to degrading Hb, the reverse scenario does not apply, i.e. not all CatBs which possess the motif will necessarily be involved in Hb degradation. We found numerous CatBs from non-parasitic organisms in GenBank which possessed the “hemoglobinase motif” (not shown), including the free living protozoan Tetrahymena thermophila, the free living nematode Caenorhabditis briggsae, the neuropathogenic schistosome regenti (which feeds on nerve tissue) and the liver fluke (which generally feeds on bile and epithelial cells rather than blood). Although possession of this motif does not necessarily confirm that the protein will have Hb degrading function, it is still a valuable tool for identifying potential hemoglobinases. Na-CP-3 possesses the motif but seems incapable of cleaving intact Hb, although it does digest globin fragments i.e. Hb that has been pre-digested with other N. americanus proteases, indicating that Na-CP-3 plays a downstream role in the Hb digestion cascade, functioning as a “globinase” rather than a hemoglobinase. In addition to containing the “hemoglobinase motif”, all four N. americanus CatBs were localized to the gut of adult worms and, moreover, mRNAs for all four proteases were up-regulated in the transition from infective L3 to blood-feeding adult worms. This strongly suggests that these enzymes are involved in blood-feeding, and that vaccines that target this family of molecules might be efficacious against human necatoriasis. Two CatBs, Ac-CP-1 and Ac-CP-2, have previously been characterised from the adult stage of the dog hookworm, A. caninum (Harrop et al., 1995). Although these two proteases share 86% amino acid sequence identity with each

83 other, they are expressed at distinct sites and are thought to have different functions - Ac-CP-1 is expressed in the cephalic and excretory glands (Loukas et al., 2004) and is detected in ES products (J. Mulvenna, A. Loukas, J. Gorman, unpublished), suggesting an extracorporeal digestive function at the site of attachment. Ac-CP-2 is localized to the brush border membrane of the intestine (Loukas et al., 2004) and is involved in Hb digestion (Williamson et al., 2004). Vaccine trials with recombinant Ac-CP-2 in the canine model resulted in a decrease in the number and fecundity of female worms, stunted growth of adult worms and production of antibodies that bound to the parasite gut in vivo and neutralised the activity of the enzyme in vitro (Loukas et al., 2004). A recent study conducted by (Xiao et al., 2008) showed that immunising with Na-CP-2, expressed as a denatured protein in E. coli, provided 29% reduction in adult worm burdens (P < 0.05) in hamsters that were experimentally challenged with N. americanus. Significant levels of protection against H. contortus have been achieved in sheep by vaccination with a cysteine proteinase-enriched fraction, TSBP (thiol sepharose binding protein) isolated from the gut of adult parasites. This protection is associated with three CatBs (hmcp 1, 4 & 6). Sheep immunized with a cocktail of these proteins expressed in bacteria had reduced faecal egg counts and worm burdens compared to controls. Sera from immunized animals also bound to the microvillar surface of the gut of adult H. contortus (Redmond and Knox, 2006). The results of this haemonchosis vaccine trial, when coupled with the protective efficacies in the canine hookworm model of Ac-CP-2 (Loukas et al., 2004) and the cathepsin D aspartic protease Ac-APR-1 (Asojo et al., 2005), present a strong case for exploitation of intestinal proteases as vaccine targets for hematophagous nematodes. The occluding loop is a diagnostic feature of CatBs, and modifications of this loop were present in all four N. americanus proteases. In addition to its role in conferring exopeptidase activity to CatBs, the occluding loop also governs the pH dependence of auto-activation. In a study conducted on Fasciola hepatica, recombinant FhCatB1 did not auto-activate upon secretion by yeast, but could be auto-activated in a low pH buffer containing glycosaminoglycans (GAGs) or polysulfated polysaccharides (PSPs) (Beckham et al., 2006). In similar fashion to FhCatB1, we found that recombinant Na-CP-3 was secreted from Pichia as a proenzyme rather than a processed mature protease, implying that the enzyme was not auto-processed during its secretion from yeast, however in the presence of

84 dextran sulphate and acidic pH, it underwent auto-processing and became catalytically active. A study conducted with human procathepsin B suggested that GAGs and PSPs bind to positively charged residues in the pro- peptide as well as the His residues in the occluding loop, inducing a conformational change which unmasks the active site and enables the access of a substrate molecule, which can then lead to the intermolecular cleavage of the pro-domain (Caglic et al., 2007). Activation of Na-CP-3 via this method supports the concept that processing of the proenzyme to the mature form could be linked to the occluding loop. In another study, parasite CatBs which were unable to undergo auto-processing required the presence of another cysteine protease, asparaginyl endopeptidase (which cleaves on the C- terminal side of Asn residues), to initiate the activation process (Sajid et al., 2003). Na-CP-4 and Na-CP-5 are likely candidates for asparaginyl endopeptidase processing, as both contain an Asn residue in the vicinity of the predicted pro-mature junction (Fig. 4.1), however Na-CP-2 and Na-CP-3 do not contain an equivalent Asn. It is well established that mammalian cathepsins B hydrolyse small peptides at the P2 position, but cathepsins L are not efficient at cleaving peptides with a P2 Arg and, instead, prefer a bulky residue at P2 (Musil et al., 1991). Many CatBs of parasitic nematodes display cathepsin B-like primary sequences, but their substrate specificities are more reminiscent of cathepsins L in that they do not cleave substrates with a P2 Arg. Indeed, ES products of A. caninum show a strong preference for peptides with a P2 Phe, and CatBs predominate in ES products (Mulvenna, Loukas, Gorman, personal communication) and in gene survey studies (Mitreva et al., 2005, Ranjit et al., 2006, Datu et al., 2008). The substrate specificity of mammalian cysteine proteases is thought to be determined by interactions in the

S2 pocket, particularly residue Glu-205 in human cathepsin B and the equivalent Ala- 205 in cathepsin L (reviewed in (Sajid and McKerrow, 2002)). The Glu residue can accommodate and stabilise the polar group of Arg in the peptide, but Ala at this position cannot bind to Arg. While many parasite proteases share sequence identities with mammalian cathepsins B and L, many do not possess either an acidic or hydrophobic residue at this site of the S2 pocket, making it difficult to classify these proteases based solely on substrate specificity or sequence identity. For example, all four N. americanus CatBs described in this study are structurally similar to cathepsins B but all of them have substituted the S2 Glu for Gln, Ser, Arg or Lys,

85 none of which are acidic or hydrophobic. This possibly accounts for the inability of Na-CP-3, at least, to cleave Z-Arg-Arg. In this study, we have investigated four cathepsin B-like cysteine proteases expressed by the human hookworm N. americanus. The roles of CP-2, -4 and -5 in blood-feeding have yet to be determined, but their developmental expression patterns, anatomic sites of expression and sequence features suggest that, like CP-3, they are involved in digestion of the blood meal. These enzymes therefore present as attractive targets for development of an anti-blood-feeding vaccine that will reduce worm viability and fecundity, which in turn could lessen the transmission of and morbidity associated with hookworm infection.

Acknowledgements This research was supported by grants from the National Health and Medical Research Council (NHMRC, Australia) and Bill and Melinda Gates Foundation. NR was supported by a QUTBLU award and funding from the ARC/NHMRC Research Network for Parasitology. AL was supported by a Senior Research Fellowship from NHMRC.

86

CHAPTER 5: DIGESTION OF HEMOGLOBIN VIA AN ORDERED CASCADE OF PROTEOLYSIS IN THE INTESTINE OF THE HUMAN HOOKWORM, NECATOR AMERICANUS

Najju Ranjita,d, Bin Zhanc, Brett Hamiltonb, Deborah Stenzeld, Jonathan

Lowthere, Mark Pearsona, Jeffrey Gormanb, Peter Hotezc, Alex Loukasa aHelminth Biology Laboratory and bProtein Discovery Centre, Division of Infectious Diseases,

Queensland Institute of Medical Research, Brisbane, Australia; cDepartment of Microbiology,

Immunology and Tropical Medicine, George Washington University, Washington DC, USA; dSchool of Life Sciences, Queensland University of Technology, Brisbane, Australia; eInstitute for the

Biotechnology of Infectious Diseases, University of Technology, Sydney, Australia.

5.1 CONTRIBUTIONS Contributor Statement of contribution* Najju Ranjit -Designed all the experiments expect for the ones mentioned below -Conducted all the experiments expect for the ones mentioned below -Analysed all the data -Drafted manuscript, Bin Zhan -Provided Na-APR-1 recombinant protein -Submitted Na-MEP-1 sequence to GenBank -Provided feedback on manuscript Brett Hamilton -Ran hemoglobin samples on mass spectrometer

Deborah Stenzel -Provided feedback on manuscript -Discussed experimental design

Jonathan Lowter -Designed Na-APR-1 kinetic activity assay -Calculated kinetics for Na-APR-1 activity -Provided feedback on manuscript Mark Pearson -Designed Na-APR-1 activity assay -Helped conduct Na-APR-1 activity assay -Provided feedback on manuscript Jeffery Gorman -Aided in proteomics aspects of experimental design for Figure 5.6 -Provided feedback on manuscript

Peter Hotez -Investigator on the grant that funded this project -Provided feedback on manuscript

Alex Loukas -Aided in experimental design -Aided in data analyses -Aided in drafting manuscript

Principal Supervisor Confirmation

I have sighted email or other correspondence from all Co-authors confirming their certifying authorship.

Alex Loukas 27-5-08 Name Signature Date

88 5.2 ABSTRACT Blood-feeding parasites utilize mechanistically distinct proteases in a vast array of biological processes. Whilst the majority of these proteases have specialised roles and generally function independently, many of them also work in a synergistic fashion as components of multi-protease networks or cascades. In this study, we investigate the roles of three distinct proteases, all of which are expressed in the gut of adult Necator americanus hookworms. The A1 family aspartic protease, Na-APR- 1, and the C1 family cysteine protease, Na-CP-3, were expressed in recombinant form in yeast and shown to be catalytically active towards synthetic peptide substrates. A cDNA encoding an M13 family metalloprotease called Na-mep-1 was cloned. Recombinant Na-MEP-1 was expressed in insect cells and displayed catalytic activity in gelatin gels that was inhibited by 1,10-phenanthroline. Antibodies raised to all three recombinant proteins were used to localize each native enzyme to the intestine of adult N. americanus using immunofluorescence microscopy. Recombinant Na-APR-1 cleaved intact Hb. In contrast, recombinant Na-CP-3 and Na-MEP-1 could not cleave Hb but, instead, cleaved globin fragments after Hb hydrolysis by Na-APR-1, implying an ordered process of hemoglobinolysis. Using tandem mass spectrometry, 74 cleavage sites within Hb α and β chains were characterised after digestion with all three proteases. All of the proteases demonstrated a promiscuous subsite specificity within Hb – noteworthy subsite preferences included bulky aromatic (Phe) and hydrophobic P1′ residues for Na- APR-1, and hydrophobic (Ala and Leu) P1′ residues for Na-MEP-1. We conclude that Hb digestion in N. americanus occurs in an ordered fashion, similar to that described in the digestive vacuole of Plasmodium falciparum, and this provides a potential mechanism by which these proteases exert their efficacy as recombinant vaccines against hookworm infection.

Keywords: Hemoglobin digestion cascade; Hemoglobinase; Necator americanus; aspartic protease; cysteine protease; metalloprotease

89 5.3 INTRODUCTION Over 700 million people in developing countries are infected with the human hookworms, Necator americanus and Ancylostoma duodenale (Hotez et al., 2004). The main pathology associated with hookworm infection stems from intestinal blood loss which can lead to iron deficiency anaemia in heavy infections. Hookworms are voracious blood feeders, causing an estimated loss of up to 9 ml of blood per day in heavily infected subjects (Lwambo et al., 1999). While anthelmintic chemotherapies such as benzimidazoles are generally effective at removing adult worms from the gut, reinfection occurs rapidly and often to equal or even higher infection intensities, leading to concerns about the long-term viability of such practices (Hotez et al., 2006). It has also been reported that, unlike other human helminthiases, clear-cut protective immunity does not occur in most individuals exposed to hookworms (Loukas et al., 2005b). This has culminated in efforts to develop a prophylactic vaccine against hookworm infection as a sustainable solution for long-term control (Bungiro and Cappello, 2004, Loukas et al., 2006). Hematophagous parasites depend on the catabolism of blood proteins such as hemoglobin (Hb) for survival. Schistosome blood flukes and the malaria parasite, Plasmodium falciparum, digest Hb in the gastrodermis and digestive vacuole respectively, using synergistic cascades consisting of enzymes belonging to distinct mechanistic classes. In schistosomes, the intact Hb tetramer is cleaved initially by aspartic proteases (Brindley et al., 2001, Delcroix et al., 2006) and to a lesser degree by papain-like cysteine proteases; the cysteine proteases are thought to exert most of their activity downstream of the aspartic protease by further catabolising globin fragments (Delcroix et al., 2006). On the other hand, treatment of schistosomes with double stranded RNA for the gastrodermal cathepsin B cysteine protease SmCB1 interrupted the ability of worms to digest serum albumin (Delcroix et al., 2006), suggesting that the order in which the different hemoglobinases act differs for each distinct protein substrate. There are varying degrees of redundancy in hemoglobinolysis in blood- feeding parasites. The roles and hierarchical positions of various Schistosoma and Plasmodium proteases in the hemoglobinolytic processes have been debated, but the application of gene knockout (Plasmodium) and gene silencing (Schistosoma) technologies have resolved some of these issues. For example, hemoglobinase

90 knockout and double knockout P. falciparum have been used to show that Hb degradation is partially redundant and relies on dual protease families with overlapping function (Liu et al., 2006). In S. mansoni, silencing of the mRNAs for different hemoglobinases has shown an ordered and somewhat less redundant pathway of Hb degradation (Delcroix et al., 2006, Morales et al., 2008). Gene silencing techniques have not yet been successfully utilized with hookworms and, indeed, some have questioned whether parasitic nematodes have the required enzymes to process double stranded RNA (Knox et al., 2007, Viney and Thompson, 2008). We have therefore taken a different approach to explore hemoglobinolysis in blood-feeding hookworms. We previously identified proteases involved in digestion of Hb (hemoglobinases) in the canine hookworm, Ancylostoma caninum, and used recombinant enzymes to reveal a semi-ordered cascade of proteolysis in vitro, whereby an aspartic and a cathepsin B-like cysteine protease both digested intact Hb followed by further digestion of globin fragments into small peptides by a metalloprotease (Williamson et al., 2004). Two of the enzymes involved in this semi-ordered cascade were then shown to confer protection as recombinant proteins against hookworm infection in dogs (Loukas et al., 2005a, Loukas et al., 2004). To characterize the hemoglobinolysis pathways used by human hookworms, we used laser capture microdissection to isolate intestinal tissue from the adult stage of N. americanus and identified the most highly expressed mRNAs encoding proteases (Ranjit et al., 2006), including homologues of some of the A. caninum hemoglobinases. Here, we describe the cDNA cloning and recombinant expression of intestinal aspartic, cysteine and metalloproteases from N. americanus. We then provide evidence for an ordered cascade of hemoglobinolysis whereby aspartic (Na- APR-1), cysteine (Na-CP-3) and metalloproteases (Na-MEP-1), in that order, digest Hb and globin fragments, and provide a map of the cleavage sites using liquid chromatography – mass spectrometry (LC-MS). Our findings shed light on the molecular mechanisms used by hookworms to obtain nutrition from human blood, and further support the pursuit of hemoglobinases as targets for the development of new drugs and vaccines against blood-feeding nematode parasites of humans and livestock.

91 5.4 MATERIALS AND METHODS 5.4.1 cDNA cloning Na-apr-1 was first reported by Williamson et al. (Williamson et al., 2002) and is assigned GenBank accession number CAC00543. Na-cp-3 was first reported by Xiao et al. (Xiao et al., 2008) and is assigned GenBank accession number ABL85237. The Na-mep-1 cDNA was initially identified as a truncated EST (BG467946) by using Ac-mep-1 as the query in a blast search of the Parasite Genomes Database through Wu-Blast2 (http://www.ebi.ac.uk/blast2/parasites.html). In order to clone the 5’ and 3’ ends of the cDNA, gene-specific primers (Na-MEPF2: GAGCTTCAATCCACCGTACT; NaMEPF1: TAGTCAACTTCTCACCGACC); NaMEPR4: CCAAGGAATTCCGCATCTTC; Na-MEPR7: TCAAAGTGGGCAGATCGTAG; Na-MEPR8: ATAGCTCCGTAACGACTGAC) were designed for 5′ and 3′ rapid amplification of cDNA ends (RACE) using adult N. americanus RNA and a modified RNA ligase-mediated rapid amplification technique (GeneRacer, Invitrogen) as described previously (Zhan et al., 2004). The full-length cDNA sequence of Na-mep-1 was obtained by creating a consensus consisting of the 5′ and 3′ RACE products and the BG467946 EST sequence. The final sequence of Na-mep-1 was confirmed by designing primers at either end of the ORF and amplifying the full-length ORF by PCR. Multiple amplicons were generated and sequenced. The cDNA sequence of Na-mep-1 has been deposited in GenBank with accession number EU523699.

5.4.2 Protein expression and purification We expressed Na-APR-1 in the yeast Pichia pastoris, as published earlier for Ac-APR-1 from A. caninum (Loukas et al., 2005a). Expression of Na-CP-3 in P. pastoris was reported by us recently (Ranjit et al., in press). Attempts to express Na- MEP-1 in yeast were unsuccessful (not shown), so we expressed it in a baculovirus/insect cell system. The entire open reading frame minus the predicted signal peptide was amplified from adult N. americanus cDNA using PCR. Amplicons were ligated into the baculovirus shuttle vector pHotWax, a modified version of pMelBac (Invitrogen), where C-terminal V5 and 6×His were inserted (Bethony et al., 2005). The plasmid contained an N-terminal melittin signal peptide to direct secretion of the recombinant protein. Recombinant plasmids were

92 sequenced to verify their identities and transfected into Spodoptera frugiperda Sf9 insect cells following the Bac-N-Blue transfection protocol (Invitrogen). After generation of a high titer viral stock, Trichoplusia ni Hi5 cells were infected with recombinant virus. Cells were maintained at 27oC with constant shaking at 120 rpm for 96 hours post infection. Supernatant from the baculovirus Hi5 cell culture was collected by centrifugation at 20,000 g for 20 mins. The supernatant was then concentrated to one-fifth original volume using a 10 kDa molecular weight cut off membrane (Pall scientific) and was buffer exchanged by dialysis into binding buffer

(50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole). The concentrated supernatant was applied onto a 2 ml Ni-NTA agarose column (Qiagen). The column was washed with increasing concentrations of imidazole (10-60 mM) and protein was eluted with 250 mM imidazole. Eluate fractions were analysed by SDS-PAGE and recombinant protein was detected by Western blot using an anti-V5 antibody (Invitrogen). Protein concentration was determined using a bicinchoninic acid (BCA) assay kit (Pierce). Na-MEP-1 was also expressed in E. coli to obtain sufficient protein for antibody production. The entire open reading frame without the signal peptide was ligated into pBAD/Thio-TOPO vector (Invitrogen) and transformed into E. coli strain BL21-DE2 (Invitrogen). To express recombinant protein, cultures were induced with arabinose (0.2% final concentration) once they had reached an OD wavelength of 0.6 nm. Culture medium was collected 4-6 hours post induction and centrifuged at 25,000 g for 20 mins. Cell pellets were collected and resuspended in lysis buffer (20 mM

NaH2PO4, 500 mM NaCl). The cells were subjected to two cycles of disruption in a French press (SLM Instruments). Lysates were centrifuged at 20,000 g for 15 mins, supernatant was retained and pellets were incubated in solubilisation buffer (6 M GuHCL, 50 mM Tris, 10 mM imidazole) for 2 hours at room temperature, and then centrifuged again. Solubilised pellet was applied onto Ni-NTA agarose column. Wash buffer (8 M Urea, 50 mM Tris, 10-60 mM imidazole gradient) was applied to the column, as recommended by the manufacturer and protein was eluted with 250 mM imidazole. Purified protein was electrophoresed on SDS-PAGE gels and detected by Western blot using an anti-His antibody (Invitrogen)

5.4.3 Catalytic activity of recombinant hemoglobinases Recombinant Na-MEP-1 expressed in insect cells was electrophoresed on precast gelatin zymogen gels (Invitrogen) as per the manufacturer’s instructions with slight

93 modifications – 10 mM CaCl2 or 10 mM ZnCl2 were added to the zymogram developing buffer, and the pH of the developing buffer was titrated in single pH units from pH 3-7. Excretory/secretory proteins of adult A. caninum were used as a positive control for the gelatin gel and 1,10-phenanthroline (10 μM final concentration) was included to inhibit metalloprotease activity. Gels were incubated overnight at 37oC and stained with Coomassie Brilliant Blue (CBB) (BioRad) followed by destaining. A zone of clearance in the gel after destaining was indicative of proteolytic activity. Catalytic activity of Na-APR-1 towards the flourogenic substrate 7- Methoxycoumarin-4-Acetyl-GKPILFFRLK(DNP)-D-Arg-Amide (MoCAc- GKPILFFRLK) (Sigma) was assayed in a Fluostar Optima microplate reader (BMG Labtech) at 37°C. Rates of hydrolysis were recorded by monitoring the increase in fluorescence measured in arbitrary units (relative fluorescence units – rfu) at excitation and emission wavelengths of 330 nm and 390 nm, respectively. The pH optimum for enzyme (1.0 μg) activity was determined by assaying in 50 mM sodium acetate buffers at half-unit pH increments from pH 2-6. The final substrate concentration was 1.0 μM and the final volume of each reaction was 100 µl. Enzyme (0.105 nM) efficiency was assessed at pH 3.5 by measuring initial rates over a range of substrate concentrations (0.2-25 μM). The catalytic constants kcat, Km and kcat/Km were derived from the resulting Michaelis-Menten plot. Catalytic activity of Na-CP-3 against the dipeptidyl substrate, Z-Phe-Arg- aminomethylcoumarin (AMC), was described by us previously (Ranjit et al., in press).

5.4.4 Antibody production and immunolocalization Antibodies were raised in female BALB/c mice against the Na-MEP-1 pBAD/Thio-TOPO construct as previously described (Tran et al., 2006). Briefly, pre- immune sera were collected two days prior to immunization and pooled. The mice were injected with 25 μg of protein subcutaneously and boosted twice at two week intervals. Whole blood was collected via cardiac puncture, sera were separated and stored at -20oC. Production of a mouse antiserum raised to recombinant Na-CP-3 was reported earlier (Ranjit et al., in press). A rabbit antiserum was produced to recombinant Na-APR-1 as described for Ac-APR-1 (Loukas et al., 2005a).

94 Immunolocalization was performed on 7 μm thick cryosections of formaldehyde (4%) fixed adult N. americanus worms. Localization was performed as described elsewhere (Don et al., 2007) with minor modifications. Briefly, slides were rehydrated in PBS and non-specific binding was blocked by incubating the slides with 5% fetal calf serum/PBS for one hour at room temperature (RT). Slides were incubated with antisera at various concentrations (1:100 to 1:1,000) in 1% Bovine Serum Albumin (BSA) for 1 hour at RT. Goat anti-mouse IgG Alexa Fluor 555 (Invitrogen) was used as fluorescent secondary antibody at a dilution of 1:500 in 1% BSA/PBST; slides were incubated for 1 hour at RT. Slides were washed in PBST, air dried and coverslips applied with mounting media (Sigma). Slides were viewed and photographed using a Leica IM100 fluorescent microscope.

5.4.5 Proteolysis of Hb by Na-APR-1, Na-CP-3 and Na-MEP-1 Human Hb was prepared by lysing whole red blood cells in a hypotonic buffer (1% PBS) (Brindley et al., 2001). Hb was incubated with individual recombinant proteases at molar ratios of 72:1 Hb to Na-APR-1, 5:1 and 50:1 Hb to Na-CP-3, and 14:1 and 140:1 Hb to Na-MEP-1, in 0.1 M sodium acetate at 0.5 pH unit increments from pH 3-7 for up to 18 hours at 37oC. Hb was also incubated with various combinations of recombinant proteases at 37oC for 6 -18 hours. Each sample was electrophoresed on a 15% SDS-PAGE gel under native conditions to observe the amount of intact Hb that remained. Hb subsite preferences for each enzyme were determined using the web logo program (http://weblogo.berkeley.edu/).

5.4.6 LC-MS and MS-MS analysis of Hb hydrolysates Hb samples that had been incubated with various combinations of hookworm proteases were subjected to reverse phase HPLC (RP-HPLC) and the Hb-derived peptides were identified by liquid chromatography mass spectrometry (LC-MS). LC- MS analysis was performed using an Ultimate 3000 nanoLC system (Dionex, Germany) with a CAP-LC flow splitter and a variable wavelength UV-VIS detector scanning at 214 nm coupled to a quadrupole time-of-flight mass spectrometer (MicrOTOFq, Bruker Daltonics) operated with a low flow electrospray needle. The column used was a Vydac monomeric C18 300 Å 3 μm 150 μm x 150 mm, at a flow rate of 1.2 μl.min-1. The mobile phase buffers used for the gradient program were (A) water with 0.1% formic acid and (B) acetonitrile:water

95 (4:1) with 0.1% formic acid. The gradient program consisted of 5% B for 5 minutes, linear ramping to 55% B over 29 min, linear ramping to 90% B over 1 min, holding at 90% B for 9 min, ramping back to 5% B over 1 min, and holding at 5% B for 20 min. The mass spectrometer scanned 50-3000 m/z and acquired data for 50 mins of each analysis. Data acquisition was facilitated using Hystar (Bruker) and data was processed using Data Analysis (Bruker). The mass spectrometer used an autoMSn methodology that collected MS2 spectra for the two most intense ions in each full scan spectrum. The scan time was set to 0.5 seconds for the Survey Scan and the MS2 spectra were recorded were the result of 2 microscans, giving an overall duty cycle of 2.5 seconds. In addition, dynamic exclusion was used such that after 2 MS2 spectra, the precursor would be added to an exclusion list for 1 min. This allowed the collection of the maximum number of MS2 spectra during the analysis. Calibration was performed immediately prior to the analysis. The data was prepared into a format suitable for Mascot database searching using Data Analysis, and Biotools (Bruker, Germany) was used to store and further scrutinize the data and search results. Mascot searches were performed using Swiss-Prot database and a 20 ppm tolerance on the precursor, 0.2 Da tolerance on the product ions, methionine oxidation as a variable modification, and charge states 1, 2 and 3. Searches were performed using no enzyme so that we could identify novel protease cleavage sites generated by the hookworm hemoglobinases.

5.5 RESULTS 5.5.1 Cloning of cDNAs encoding N. americanus hemoglobinases Cloning of Na-apr-1 (Williamson et al., 2002) and Na-cp-3 (Xiao et al., 2008) have been reported elsewhere. Na-mep-1 was submitted to GenBank under accession number EU523699. The ORF is comprised of 846 amino acids (excluding the signal peptide of 26 amino acid residues) with a predicted molecular mass of 96.77 kDa and pI of 6.69. It contains three predicted N-glycosylation sites (118-NRT, 251-NHT and 545-NMT), two zinc binding motifs and is a member of the M13 peptidase family (Fig. 5.1). Closest homologues at the protein sequence level included predicted M13 family members from other hookworms (Ancylostoma sp.), the free-living nematode Caenorhabditis elegans and the jewel wasp Nasonia vitripennis. Na-MEP-1 is 56% identical at the amino acid level to Ac-MEP-1 from A. caninum.

96

Na-MEP-1 MTKLLVSTAGLTGVVAALFITSLVFSILTWTRVKNDNDNPPRPKEPLSRPVVQLSSSIQT Ac-MEP-1 MAKLLEVTTGLVVLLGVLGVISVVFNVLTWLKLNENKDDSS-PAPKIWNVGEQDNTPVLT neprilysin ------MGKSESQMDITDINTPKPKKK----QRWTPLEI : . : : : * * :.: Na-MEP-1 TVTENVVTEPIVTVPTVSRTRVSAKTISPRSSATTSTRTLRTLTTPKFVATEAAP--RRN Ac-MEP-1 NLLVLEKEELAAKLKKTPYEEVDEQTVR-QSSVMKLRNIKNALFTPIEPVASALPPLRVN neprilysin SLSVLVLLLTIIAVTMIALYATYDDGICKSSDCIKS------.: : . . . : *. . Na-MEP-1 RTIMCPNYGVSDNSYAYQEAASFILSGLDERVNPCEDFYAFTCNKFLKDHKAEEHGVSRY Ac-MEP-1 DPKYCPSYGEPDKKYAYQEAASYLLSGLDQTVDPCEDLYAFTCNTYLRNHNATDIGVNRI neprilysin ------AARLIQNMDATTEPCTDFFKYACGGWLKRNVIPET-SSRY *: ::..:* .:** *:: ::*. :*: : : .* Na-MEP-1 GAIKELQDAVNTEIVDALFDVDVNDKKRSETERITKALLHDCVYHISPN-VPTETIINFL Ac-MEP-1 GTYKDAQDDVNAEIVEALEEVNVSDTKWSETERLVKATLFTCVHHTRAR-KPIDNSKNVL neprilysin GNFDILRDELEVVLKDVLQEPKTED---IVAVQKAKALYRSCINESAIDSRGGEPLLKLL * . :* ::. : :.* : ...* : : .** *: . : :.* Na-MEP-1 EEIARMFGGIPFLNHTLKEDFDVFAAMGEVEQNHAMGTLFSAMVSVDYKKIKQNSLFLSQ Ac-MEP-1 IEMRDLFGGIPFLNHTLKKDIDFFDIMGKFEQNHAMGTLLGAMVSVDFKNVNKHSLFLSQ neprilysin PDIYGWPVATENWEQKYGASWTAEKAIAQLNSKYGKKVLINLFVGTDDKNSVNHVIHIDQ :: . ::. . :.:.:.::. .*:. :*..* *: :: :.:.* Na-MEP-1 PRLPMP-REFYVLPQFTMKLKKRGLQIADVLKKFAEKILEEPDKYRDMIEKAAQDVVELE Ac-MEP-1 PYLPMA-RDFYVFPQHTKMVENRVSLINSVLRSFAEAVLDDPSPYLDLMSRSARDVVKLE neprilysin PRLGLPSRDYYECTGIYKEACTAYVDFMISVARLIRQEERLP-IDENQLALEMNKVMELE * * :. *::* . . : : : . * : : ..*::** Na-MEP-1 RRIALASWADAEMRNYAQQYNPYDLPTLKKAY------PSVKWESYLRSLLSTVGPVDFS Ac-MEP-1 MQIAMASWPESELRNYAQQHNPRTLNQLKAAY------PAIKWDSYFNALLSSVQGVDMN neprilysin KEIANATAKPEDRNDPMLLYNKMTLAQIQNNFSLEINGKPFSWLNFTNEIMSTVNISITN .** *: : .: :* * :: : ...* .: . ::*:* . Na-MEP-1 GPHKRLIISQPSYFGWLNALFNGNVVDENTIVNYIITHLIFEDAEFLGGIFKESAEDLNY Ac-MEP-1 --RQNIILTQPSYFGWLNALFNG-GADDKTIANYLLVHLILEEADFLGGALKTMVQKSDY neprilysin --EEDVVVYAPEYLTKLKPILTK--YSARDLQNLMSWRFIMDLVSSLSRTYKESRNAFRK .: ::: *.*: *:.::. . . : * : ::*:: .. *. * : Na-MEP-1 VRYAQRSGRGVARVGRQLMHQR-DTRGDPNIPCMNFIMTYMPYGPGYVYVRSKQQRNDVQ Ac-MEP-1 VPYALGRGKGVTRVGQQLTRSHDDTVEDANIQCLNSMMTYMPFGPGYVYVKSRKNRDDVV neprilysin ALYGTTSETATWRR------CANYVNGNMENAVGRLYVEAAFAG-ESK . *. .. * * * : * . * :**.: : Na-MEP-1 ADIRKQTELVIESFLNMTSGLKWMSSDSKEKARQKAKGMVRNYGWPQKLFGDFKSSEEID Ac-MEP-1 KDIEHQTELVFKNFVNMIGNLNWMTDASLELAMEKADTMVKNYGWPKDLFGNFRDSSKID neprilysin HVVEDLIAQIREVFIQTLDDLTWMDAETKKRAEEKALAIKERIGYPDDIVSNDNKLN--N :.. : : *:: ..*.** : : * :** : .. *:*..:..: .. . : Na-MEP-1 EYHKKDYAEILELTKTERSSLRYYRMRRVLIKGYSNRESLRLLLQDADRSNFLLSPALVS Ac-MEP-1 AYHKKDYGNIINLYK-ENITHNYYHIRRTMIKGYSNHESLRLLTEAPKRDHFLLSPALVN neprilysin EYLELNYK------EDEYFENIIQNLKFSQSKQLKKLREKVDKDEWISGAAVVN * : :* * . :*: :.*: * : .:..:: ..*:*. Na-MEP-1 AWYQPERNSITFPYASFNPPYYSYEYPQAYNYGGQGGTAGHELVHGFDDQGVQFGPDGSL Ac-MEP-1 AWYIPERNSIAFPYAFWNPPYYNYEYPQACNYAGQGGTAGHELVHGFDDQGVQFAADGSL neprilysin AFYSSGRNQIVFPAGILQPPFFSAQQSNSLNYGGIGMVIGHEITHGFDDNGRNFNKDGDL *:* . **.*.** . :**::. : .:: **.* * . ***:.*****:* :* **.* Na-MEP-1 SRCTWYDCGWMDKRSKDGFNDMAQCVVTHYSTFCCPEQEGNIHCANGATTQGENIADIGG Ac-MEP-1 SDCTWIECGWLEEKSKKGFSDMAQCVVTQYSTQCCPQTGGVTHCANGATTQGENIADLGG neprilysin ------VDWWTQQSASNFKEQSQCMVYQYGNFSWDLAGG--QHLNGINTLGENIADNGG .* ::* ..*.: :**:* :*.. . * : ** .* ****** ** Na-MEP-1 EHAAYIAYREYIKS-LGHEEKRLPGLERYTPNQIFWITYGYSWCRSVTEEYLISQLLTDP Ac-MEP-1 QLAAYRAYREYITKERGEEEKRLPGLEQYTPNQIFWITYGYSWCMSQTDSSLIRQLLTDV neprilysin LGQAYRAYQNYIKK--NGEEKLLPGLDLN-HKQLFFLNFAQVWCGTYRPEYAVNSIKTDV ** **::**.. . *** ****: :*:*::.:. ** : . : .: ** Na-MEP-1 HAPSACRTNQVVQSIPAFGRDFGCSLGDRMYPAPEQRCSVWVQE Ac-MEP-1 HSPGSCRVNQVMQDIPEFALDFGCTMGQKMYPEPEQRCPVWVAE neprilysin HSPGNFRIIGTLQNSAEFSEAFHCRK--NSYMNPEKKCRVW--- *:*. * .:*. . *. * * . * **::* **

Figure 5.1. Multiple sequence alignment of Na-MEP-1 with Ac-MEP-1 from Ancylostoma caninum and human neprilysin 1. Signal peptides of the hookworm proteins are italicised. Putative N-linked glycosylation sites of Na- MEP-1 are shown in bold font. Black boxes surround the catalytic residues that bind to divalent metal ions (zinc binding motif). Asterisks denote conserved residues in all three sequences. Two dots denote residues conserved in two of the three sequences. Single dots denote similar amino acids in at least two of the sequences

97 5.5.2 Expression and purification of Na-APR-1 and Na-MEP-1 Na-APR-1 was secreted by yeast as a zymogen into culture medium. The recombinant protein was purified on a nickel-NTA column and was produced at an approximate concentration of 2.0 mg.L-1 of purified protein (Fig. 5.2A). Na-CP-3 was expressed in yeast and purified on nickel-NTA columns as described (Ranjit et al., in press). Na-MEP-1 was expressed in Hi5 insect cells, secreted into culture medium then purified by nickel-NTA chromatography at a yield of 0.2 mg.L-1 (Fig. 5.2B). The identities of the recombinant proteins were confirmed by western blot analysis using anti-Hexa His and anti-cmyc for Na-APR-1 and anti-V5 for Na-MEP- 1 antibodies (not shown). Na-MEP-1 was also transformed into E. coli and protein was expressed in insoluble inclusion bodies (Fig. 5.2C). The protein was purified under denaturing conditions and verified by Western blot analysis using anti-hexa His antibody. The yield of purified, denatured protein was 1.5mg.L-1. Attempts to refold denatured Na-MEP-1 expressed in E. coli were unsuccessful (not shown). Na- MEP-1 expressed in E.coli pBAD-Thio vector included a glutathione S-transferase (GST) fusion thus the purified protein was approximately 129 kDa (33 kDa larger than the protein purified from Hi5 insect cells).

Figure 5.2. Expression and purification of N. americanus hemoglobinases. Na-APR-1 expressed in the yeast Pichia pastoris X33 (A). Lane 1 – culture supernatant; 2 - flow through from nickel-NTA column; 3 - 20 mM imidazole wash; 4 – 60 mM imidazole wash; 5 -1 M imidazole eluate. Na-MEP-1 expressed in Hi5 insect cells (B). Lane 1 – culture supernatant; 2 – flow through from Ni-NTA column; 3 – 40 mM imidazole wash; 4 – 60 mM imidazole wash; 5 – 250mM imidazole eluate. Na-MEP-1 expressed in insoluble form in E. coli (C). Lane 1 – urea solubilised pellet; 2 – flow through from Ni-NTA column under denaturing conditions (6 M urea); 3 – 40 mM imidazole/6 M urea wash; 4 – 60 mM imidazole/6 M urea wash; 5 – 250 mM imidazole/6 M urea eluate. Arrows indicate purified protein.

98 5.5.3 Catalytic activity assays Na-APR-1 was secreted from P. pastoris in its pro-form and underwent auto- activation at acid pH. Optimal cleavage of MoCAc-GKPILFFRLK was at pH 3.5 (not shown). MoCAc-GKPILFFRLK was hydrolysed by 0.105 nM recombinant Na-

APR-1 at pH 3.5 with a Km = 15.4 ± 0.7 μM, kcat = 32.2 ± 0.7 and kcat/Km = 2,090,909 M-1s-1 (Fig. 5.3A). As little as 5 ng of recombinant Na-MEP-1 displayed gelatinolytic activity. The catalytic activity was dependent on the presence of divalent cations (10 mM CaCl2) in the developing buffer; when CaCl2 was replaced with ZnCl2, catalytic activity was no longer observed (Fig. 5.3B). Activity was observed at pH 4.5-6.5 (not shown), and addition of 1,10- phenanthroline completely inhibited catalytic activity.

Figure 5.3. Catalytic activity of recombinant Na-APR-1 expressed in yeast and Na-MEP-1 expressed in insect cells. Catalytic activity of Na-APR-1 was determined using the fluorogenic peptide MoCAc-GKPILFFRLK in 50 mM sodium acetate (A). A range of concentrations of peptide were hydrolysed by 0.105 nM of recombinant Na-APR-1 at pH 3.5. Gelatin zymogram showing catalytic activity of purified recombinant Na-MEP-1 (B). Lane 1: 10 ng Na-MEP-1+ CaCl2; 2: 5 ng Na-MEP-1 + CaCl2; Lane 3: 10 ng Na-MEP-1 + 1,10-phenanthroline

5.5.4 Immunolocalization Immunofluorescence experiments were conducted with tissue sections of adult N. americanus that had been removed from euthanised hamsters. Antibodies to Na-APR-1, Na-CP-3 and Na-MEP-1 all localized to the intestine of the adult worm (Fig. 5.4) and confirmed earlier reports of intestinal localization for Na-APR-1 (Williamson et al., 2002) and Na-CP-3 (Ranjit et al., in press). Pre-vaccination control serum did not bind specifically to any structures (Fig 5.4D).

99

Figure 5.4. Immunolocalization of Na-APR-1, Na-CP-3 and Na-MEP-1. Na-APR-1, Na-CP-3 and Na-MEP-1 in transverse (A, C, D) and longitudinal (B) sections of adult N. americanus. Arrows indicate the intestine. A: mouse α-Na-APR-1; B: mouse α-Na-CP-3; C: mouse α- Na-MEP-1; D: normal mouse serum. White bar denotes 50 μm.

5.5.5 Hemoglobin degradation and LC-MS analysis of hemoglobin hydrolysates. Human Hb was readily degraded by Na-APR-1, with digestion starting to occur within 5 mins at pH 3.5 and pH 4.5; digestion was slower at pH 5.5 but still occurred (Fig. 5.5A). In contrast, Na-CP-3 and Na-MEP-1 were not able to digest intact Hb, even after 24 hour incubation (Fig. 5.5B). To observe whether Na-CP-3 and Na-MEP-1 acted downstream in the hemoglobinolysis pathway to digest globin fragments after initial digestion of Hb with Na-APR-1, hydrolysates were observed by LC-MS (Fig. 5.6). Nineteen cleavage sites were detected in the Hb α chain and a further 18 in the Hb β chain after digestion with Na-APR-1. Supporting the SDS- PAGE observations, neither Na-CP-3 nor Na-MEP-1 digested Hb when assessed by

100 LC-MS (data not shown). However, when Na-APR-1 and Na-CP-3 were added to Hb together, an additional 4 cleavage sites in the Hb α chain and 9 in the β chain were detected. Na-MEP-1 did not digest Hb, but when added to Hb in the presence of Na- APR-1 and Na-CP-3, a further 16 cleavage sites in the Hb α and 8 in the β chain were detected (Fig. 5.7). None of the proteases displayed a distinct preference for defined residues at the P4-P4′ subsites. Na-APR-1 preferred bulky aromatic and hydrophobic residues at the P1 position, cleaving at 8 sites with a P1 Phe, 10 sites with a P1 Leu and 5 sites with a P1 Ala. Na-CP-3 was also promiscuous in its preferred P1 residues, mostly cleaving after hydrophobic and neutral amino acids. Na-MEP-1 preferred P1′ Ala (5 sites), but also readily cleaved sites with P1′ Lys, Ser, Asp and Leu (Fig. 5.8). Na-MEP-1 also showed a preference for hydrophobic residues at P2 (Leu, Val and Ala).

AB

35 35

25 25

15 15

10

10

1 2 3 4 1 2 3 4

Figure 5.5. Hemoglobin digestion with recombinant Na-APR-1, Na-CP-3 and Na-MEP-1. (A) One hundred μg of hemoglobin was incubated with 1 μg of recombinant Na-APR-1 various pHs at 37oC for 5 mins. Lane 1: 10 μg Hb; L2: Hb + Na-APR-1 pH 3.5; L3: Hb + Na-APR-1 pH 4.5; L4: Hb + Na-APR-1 pH 5.5 (B) One hundred μg of hemoglobin was incubated with various recombinant proteases at pH 4.5 at 37oC. Lane 1: 10 μg Hb; L2: Hb + Na-APR-1 (5 mins); L 3: Hb + Na-CP-3 (24 hrs); L4: Hb + Na- MEP-1 (24 hrs).

101 Intens. [m AU]

50

40 30

20

10

0 A

0 10 20 30 40 50 60 Time [min]

In te n s . [m A U ]

10

5

0 B

0 10 20 30 40 50 60 T im e [m in ] In te n s . [m A U ] 30

20

10

0 C

0 10 20 30 40 50 60 T im e [m in ]

In te n s . [m A U ] 50 40

30

20

10

0 D

0 10 20 30 40 50 60 Time [min]

In te n s . [m A U ]

50

40 30

20

10

0 E 0 10 20 30 40 50 60 T im e [m in ] Figure 5.6. LC trace of hemoglobin incubated with various recombinant proteins for 18 hours at 37oC at pH 4.5 Colored lines represent the UV trace (214 nm) of hemoglobin. (A) Overlapping UV trace of all four samples; (B) Hb (C) Hb + Na-APR-1; (D) Hb + Na-APR-1 + Na-CP-3; (E) Hb + Na-APR-1 + Na-CP- 3 + Na-MEP-1

102

Figure 5.7. Map of hemoglobin α and β chains highlighting the cleavages made by N. americanus recombinant haemoglobinases. All three enzymes were co-incubated with hemoglobin at pH 4.5. Black arrows – Na-APR-1 cleavage sites; green arrows – Na-CP-3 cleavage sites; red arrows- Na-MEP-1 cleavage sites. Stars denote cleavage in Hb hinge region.

103

A B C

Figure 5.8. P4-P4′ subsite specificities of Na-APR-1, Na-CP-3 and Na-MEP-1. (A) Na-APR-1, (B) Na-CP-3; (C) Na-MEP-1. MS/MS data compiled from Mascot searches was used to create an alignment of peptide cleavage sites which was submitted to the web logo program to generate the image. The letters denote each amino acid, the larger the letter the more frequently that residue occurred at each subsite.

104 5.6 DISCUSSION To liberate Hb into the gut lumen, hookworms must first lyse the ingested erythrocytes. They do this via a protein that creates pores in the erythrocyte membrane (Don et al., 2004). Hookworms then, like other blood-feeding parasites, digest blood via a cooperative and ordered cascade of hemoglobinolysis. Here, using an in vitro approach we show that the human hookworm, N. americanus, can utilize a pathway of mechanistically distinct proteases (hemoglobinases) to digest Hb and the subsequent globin peptides, and that the pathway is similar to that used by the malaria parasite, P. falciparum. It is ordered, consists of at least three distinct classes of endopeptidases, aspartic, cysteine and metallo-enzymes, and can be a major target for the development of new therapies. Plasmodium spp. are no more closely related to nematodes than they are to humans, inferring that convergent evolution has resulted in phylogenetically distant pathogens adopting similar ordered cascades of Hb hydrolysis. P. falciparum makes initial cleavages of Hb with aspartic proteases, called plasmepsins. Four distinct plasmepsins (PM I, II, IV and HAP) cleave Hb in a semi-ordered fashion - the process is initiated by PM I and II, allowing PM IV and HAP to then act downstream in the cascade (Banerjee et al., 2002). However, P. falciparum clones with deletions of each of these plasmepsins or combinations thereof remained viable, despite prolonged doubling times, suggesting that at least the early phases of the hemoglobinolysis process is redundant (Goldberg, 2005, Liu et al., 2006). This is in contrast to the P. falciparum cysteine hemoglobinase, falcipain-2, which is essential for Hb digestion and parasite growth (Sijwali and Rosenthal, 2004). Moreover, falcipain-2 knockout parasites were over 1000-fold more sensitive to the aspartic protease inhibitor, pepstatin, than were wild type controls, highlighting the cooperative action of cysteine and aspartic hemoglobinases in this parasite (Sijwali et al., 2006). In contrast to Plasmodium, hemoglobinolysis is less redundant in the parasitic . Current data suggests that only one aspartic protease and three clan CA cysteine proteases are present in the gastrodermis of schistosomes, and only the aspartic and two of the cysteine proteases (SmCB1 and SmCL1) have been shown to digest the intact Hb tetramer (Brindley et al., 2001, Lipps et al., 1996). Further evidence of an ordered cascade in schistosomes comes from studies employing RNA

105 interference (RNAi) (Delcroix et al., 2006). Silencing of the S. mansoni mRNA for cathepsin D aspartic hemoglobinase resulted in an inability of larval worms to digest Hb in vitro, and, more importantly, none of the worms treated with cathepsin D double stranded RNA matured to adulthood when injected into mice (Morales et al., 2008). Suppression of mRNA encoding the cathepsin B cysteine protease, SmCB1, did not overtly affect Hb degradation, but slowed parasite growth considerably (Correnti et al., 2005). In the study described herein, as well as our earlier findings with A. caninum (Williamson et al., 2004), we show that hookworms express intestinal homologues of the major aspartic and cysteine hemoglobinases of schistosomes and Plasmodium spp. Moreover, hookworms express M13 metalloprotease that digests globin fragments but not intact Hb; only P. falciparum has an analogous metalloendopeptidase (albeit an M16 peptidase), and a homologue has yet to be reported from schistosomes. Of the three hemoglobinases that we explored in this study, only one could cleave intact Hb, the aspartic protease Na-APR-1. In our assays, Na-APR-1 was responsible for making the initial cuts in the tetramer, thereby unravelling it and making the globin fragments accessible to the other enzymes that act downstream in the pathway. Na-APR-1 is by no means the only hemoglobinase that cleaves the intact Hb molecule in hookworms. A pepsin-like aspartic protease, Na-APR-2, can also cleave Hb, and does so at distinct sites to Na-APR-1 (Williamson et al., 2003a). Moreover, the A. caninum cysteine protease, Ac-CP-2, cleaves intact Hb (Williamson et al., 2004), implying at least some redundancy whereby distinct mechanistic classes of enzyme can initiate the cascade, although they cleave Hb at distinct sites. We recently reported a family of 5 cathepsin B cysteine proteases from the gut of N. americanus (Ranjit et al., in press), most of which were identified from just 480 expressed sequence tags (ESTs). Only one of these enzymes, Na-CP-3, was expressed in active form. Moreover, Na-CP-3, but not the other intestinal cathepsins B, had a C-terminal insertion in the same region as the falcipain-2 Hb-interacting motif (Pandey et al., 2005), suggesting that not all of the N. americanus intestinal cathepsins B are involved in Hb digestion. Indeed, adult hookworms can be kept alive for months in medium containing just serum in the absence of Hb or blood (D. Smyth & A. Loukas, personal communication), implying that Hb is not the only food source that hookworms can utilise for survival, at least in vitro. Although Na-CP-3

106 did not digest intact Hb and only digested globin fragments, it might be capable of digesting other full-length serum proteins, such as albumin. In support of this, Hb digestion in schistosomes is inhibited most effectively by aspartic protease inhibitors, while digestion of serum albumin is most effectively inhibited by cysteine protease inhibitors, implying that different classes of enzymes play distinct roles in the digestion of different protein substrates in the parasitic helminths (Delcroix et al., 2006). Exopeptidases are involved in hemoglobinolysis in Plasmodium (Dalal and Klemba, 2007, Stack et al., 2007) and schistosomes (Caffrey et al., 2004, McCarthy et al., 2004). They are thought to exert their activities downstream of the endopeptidases, removing terminal amino acids and dipeptides for transport into the gastrodermal cells (schistosomes) or in the cytosol (Plasmodium). Hookworm ESTs encoding amino- and carboxy-peptidases are present in public databases, and we have actively expressed an A. caninum aminopeptidase and are currently exploring its role in the Hb digestion process (T. Don, J. Lowther and A. Loukas, personal communication). The degree of redundancy, both in terms of numbers of proteases and overlapping functions/substrate preferences, has not been addressed in any depth for hookworms or other parasitic nematodes. Progress in this area has been hampered by a number of roadblocks: the lack of a genome sequence and, until recently, a substantive number of expressed sequence tags (ESTs); and absence of RNAi and gene knockout technologies. As a result, studies on the functions of proteases of parasitic nematodes have been restricted to in vitro observations only. For example, the barber’s pole worm, Haemonchus contortus, is a blood-feeding nematode of sheep that causes enormous economic losses worldwide. Like hookworms, the gut of H. contortus contains a plethora of mechanistically distinct endo- and exopeptidases (Jasmer et al., 2004, Knox et al., 2003, Williamson et al., 2003b). However, most have not been expressed in catalytically active form and have yet to be unequivocally assigned hemoglobinolytic functions, despite being expressed in the gut. Nonetheless, parasite-derived H. contortus antigen complexes that are enriched for intestinal proteases are highly efficacious vaccines (Knox et al., 2003, Knox and Smith, 2001). This strongly supports the case for dissecting the molecular basis of the haemoglobinolytic pathway in blood-feeding nematodes.

107 N. americanus intestinal cells produce an array of proteolytic enzymes; our study has now attributed a hemoglobinolytic function to several of these. This is by no means the complete set of enzymes involved in the pathway, but serves to highlight the complexity of the process and shows that at least some level of order exists. The pathway is similar to that described for the canine hookworm, A. caninum, but several key differences exist; Na-CP-3 was incapable of cleaving intact Hb and instead cleaved globin peptides generated by hydrolysis with Na-APR-1, however, Ac-CP-2 is the only cysteine hemoglobinase identified thus far from A. caninum, that cleaves intact Hb. Numerous cysteine proteases are present in the gut of N. americanus (Ranjit et al., in press), so we cannot yet discount the presence of an intestinal cathepsin B that digests intact Hb. Unlike Plasmodium hemoglobinases, which are within the digestive vacuole and inaccessible to antibodies, hookworm hemoglobinases are extracellular and exposed to host antibodies when the worm ingests blood. Antibodies against Ac- APR-1 and Ac-CP-2 bind to the intestine of feeding hookworms, neutralize the catalytic activity of the target enzymes and damage the epithelial surface (Loukas et al., 2005a, Loukas et al., 2004). By understanding the functions of these enzymes in the intestine of the worm we can obtain a more in-depth understanding of blood- feeding in parasitic nematodes and make more informed decisions on hemoglobinase antigen selection and mechanisms for their delivery as recombinant vaccines.

Acknowledgements We thank Tristan Wallis for his technical assistance and helpful suggestions. This research was supported by grants from the National Health and Medical Research Council (NHMRC, Australia), The Sabin Vaccine Institute and the Bill and Melinda Gates Foundation. NR was supported by a QUTBLU award and funding from the ARC/NHMRC Research Network for Parasitology. AL was supported by a Senior Research Fellowship from NHMRC.

108

CHAPTER 6: GENERAL DISCUSSION, CONCLUSION AND FUTURE DIRECTIONS

6.1 GENERAL DISCUSSION Hookworm infections are highly endemic in many developing countries. The nature of the infection varies from person to person, ranging from no overt symptoms (usually light infections) to lethal, depending on the intensity of infection as well as the nutritional and physiological status of the patient. At present, the control of hookworm infections is dependent on anthelmintic drugs, and while this mode of treatment provides successful cure, there are growing concerns that continued use of these therapies could potentially lead to the emergence of drug resistant parasites (Hotez et al., 2006). Vaccines are an attractive alternative to anthelmintics and, while they may not achieve the rapid parasite clearance obtainable by drug treatment, the effects of vaccination are prolonged, offering long-term protection against infection and reinfection. In developing a hookworm vaccine, it is essential to identify one or a few out of the numerous potential antigens, as it is not feasible to assess all possible molecules for vaccine efficacy in animal models. Immunological studies conducted with the sheep nematode H. contortus have suggested that parasitic nematode antigens can be categorised into two groups - either natural antigens or hidden antigens (Munn, 1997). ES products and exposed somatic antigens which induce an immune response in the host during the course of infection are termed natural antigens, whilst antigens that do not induce an immune response during infection (because they are hidden from the afferent immune system) are referred to as hidden antigens (Newton and Munn, 1999). Each type of antigen can induce a robust immune response in isolation, however it has been suggested that by combining members of the two groups, their effects may be synergistic and they could target distinct physiological pathways and developmental stages (Munn, 1997). It is believed that an efficacious hookworm vaccine should consist of a combination of two antigens, one of which is expressed/secreted by the infective larvae and is involved in skin penetration and migration, and a second one which targets proteins expressed in the gut of the adult stage, such as those involved in blood-feeding (Loukas et al., 2006). A secreted protein, termed Na-ASP-2, from L3 N. americanus larvae, has already been selected and tested in phase I clinical trials (Diemert et al., 2008) on the basis of its ability to partially protect laboratory animals against hookworm challenge infections (Bethony et al., 2005, Goud et al., 2004). In

110 addition to Na-ASP-2, it has been proposed that a second antigen, one from the adult blood-feeding stage, would also be needed for an optimal vaccine. The process by which haemoglobin (Hb) is broken down into free amino acids is a focus of investigation in a number of haematophagous parasites, as it has long been thought that targeting this process could lead to identification of new chemotherapeutic strategies or vaccines (Delcroix et al., 2006). In lysosomes of mammalian cells, endopeptidases act cooperatively with exopeptidases in the catabolism of proteins. There is now mounting evidence that haematophagous parasites digest Hb using a similar cooperative cascade, whereby endopeptidases that are similar to mammalian cathepsins B and D act together with exopeptidases that are similar to mammalian dipeptidyl peptidases and aminopeptidases (Williamson et al., 2003b). Studies conducted with hematophagous parasites such as P. falciparum and S. mansoni have shown that there is a hierarchical system whereby defined proteases initiate the cleavage of the intact Hb tetramer, which is subsequently cleaved by other endo-proteases and then exo-proteases, with the endpoint being small globin peptides and free amino acids that are readily absorbed to provide nutrients for the parasite. A similar network of enzymes has been shown to digest Hb in the canine hookworm, A. caninum, and this model of Hb digestion has been loosely termed a ‘haemoglobin digestion cascade’ (Delcroix et al., 2006, Williamson et al., 2004). The main objective of this thesis was to identify and characterise potential vaccine candidate molecules from the adult stage of N. americanus, in particular seeking those proteins that are involved in feeding and nutrient acquisition. To date, there has been a paucity of literature on the feeding process in human hookworms, mostly due to the difficulty in maintaining them in laboratory animals. If the blood- feeding process is to be targeted as a potential vaccine strategy for human hookworms, it is imperative to gain a comprehensive understanding of the molecular events in this process and to determine which proteases are involved in the enzymatic digestion of the blood-meal. It is thought that the main source of nutrition for hookworms comes from proteins contained in the blood that they ingest from ruptured capillaries in the host intestinal wall - one of most highly abundant proteins in blood is haemoglobin (Hb). In this study, I have investigated the roles of various intestinal proteases which are thought to play a role in haemoglobin degradation, aiming to establish which

111 protease would be the most viable target as a vaccine candidate. The first manuscript from this study (A survey of the intestinal transcriptome of the hookworms, Necator americanus and Ancylostoma caninum using tissue isolated by laser microdissection microscopy) details the investigation of the intestinal transcriptome from both human and canine hookworm species. This study was conducted in order to profile the proteins expressed in the intestine of the worms and gain a general snapshot of the functions that occur in this specialised organ. As it is thought that Hb digestion processes takes place in intestine of the worm, I determined which groups of proteases were expressed in this region, and selected several of these for more detailed characterisation. The aim here was to obtain better knowledge of which proteins might be the most important to target in a vaccine development program. The N. americanus gut cDNA generated in this study was not only used for the identification of novel protease mRNAs, but was also valuable in verifying which of the previously identified proteases such as Na-APR-2, Na-CP-2-5, and other proteins with potential roles in blood-feeding were expressed in the gut of this hookworm, further strengthening the claim that they are involved in digestion of the blood-meal. The data I have presented in the following two manuscripts (A family of cathepsin B cysteine proteases expressed in the gut of the human hookworm, Necator americanus and Digestion of haemoglobin via an ordered cascade of proteolysis in the intestine of the human hookworm, Necator americanus), provides the first detailed evidence of the roles of several individual proteins, belonging to three distinct classes of proteases, expressed by N. americanus. The majority of the proteases I found are orthologs of proteases expressed by other hematophagous helminths and protozoa, some of which have been already been shown to play a role in Hb digestion process in these parasites (Delcroix et al., 2006, Goldberg, 2005). I postulated that, due to their similarity in protein sequence and structure, the N. americanus proteases identified in this study would also be involved in the Hb degradation process.

Aspartic proteases Studies conducted with several hematophagous parasites have implied that aspartic proteases play a significant role in the initial degradation of tetrameric Hb by cleaving it at the hinge region. In S. mansoni and P. falciparum, it has been shown that cathepsin D-like enzymes make initial cuts in Hb (Banerjee et al., 2002,

112 Delcroix et al., 2006). Moreover, Williamson et al. demonstrated that Ac-APR-1 and Na-APR-1 cleave intact dog and human Hb, and suggested that this enzyme was likely to be responsible for the initiation of the Hb digestion cascade in hookworms (Williamson et al., 2002). My results confirm this hypothesis, as MS/MS data showed that Na-APR-1 cleaved the Hb α chain at Phe-33 – Leu-34 and Hb β chain at Phe-41 – Phe-42, which are critical sites for the maintenance of Hb conformation. Recent published studies on cathepsin D-like proteases from S. mansoni and P. falciparum demonstrated that suppression of these enzymes, either at the RNA expression level or with chemical inhibitors of aspartic proteases, resulted in the parasites being incapable of digesting haemoglobin properly, leading to retarded growth and reduction in fecundity in S. mansoni (Delcroix et al., 2006, Morales et al., 2008) and slower replication time in P. falciparum (Liu et al., 2005). Vaccination of dogs with Ac-APR-1 resulted in significant reductions in worm burden and faecal egg output, but most importantly, vaccination protected the host against substantial blood loss (Loukas et al., 2005a). These results clearly indicate that this group of proteases play a critical role in Hb degradation and therefore warrant targeting as vaccine candidates.

Cysteine proteases Although it has been shown that aspartic proteases play a principal role in the Hb digestion process, studies investigating the effects of specific inhibitors on proteolytic activity of hookworm extracts have revealed that the inhibition of aspartic proteases alone does not completely ablate Hb degradation (Williamson personal communication). In fact, complete inhibition of Hb degradation was only observed when all four protease inhibitors were utilised, suggesting that other classes of proteases are necessary for the Hb degradation process to occur efficiently. A number of studies have now shown that cysteine proteases also play a substantial role in Hb digestion process (Sajid et al., 2003, Sijwali et al., 2006). Cathepsin B-like cysteine proteases have been found to be abundantly expressed in a number of blood-feeding parasites, particularly in the gut of the nematode H. contortus. Indeed, this gene family has undergone enormous expansion in Haemonchus, and mRNAs encoding cathepsins account for approximately 16% of all transcripts in the adult female worm intestine (Jasmer et al., 2004). For this reason, I have proposed that cathepsins B also play a major role in

113 haemoglobinolysis in the gut of hookworms, and they were therefore a major focus of this study. Cysteine proteases of parasites are often developmentally regulated, showing highest expression levels in actively feeding stages (Jasmer et al., 2004). Moreover, they are often primarily localised in the region where digestion of food occurs, such as the intestine of helminths. Kumar et al. characterised a cysteine protease from the exsheathing fluid of N. americanus larvae (Kumar and Pritchard, 1992), but this is the only report in the literature on cysteine proteases from human hookworms. A N. americanus cDNA sequence, encoding a cathepsin B protease called necpain, has been deposited in GenBank, but a publication has not accompanied this submission. In the second manuscript from this thesis (A family of cathepsin B cysteine proteases expressed in the gut of the human hookworm, Necator americanus), I characterised four additional mRNAs encoding cathepsins B and showed that all four are expressed in the intestine of N. americanus. A comparison of the ORFs of the four N. americanus cathepsins B (Na-CP-2, -3, -4, -5) showed that these proteins shared the greatest sequence homology with the A. caninum cysteine proteases Ac- CP-2 and Ac-CP-1. Despite these two A. caninum proteases sharing 86% amino acid identity, they are expressed in two different locations and are thought to play distinct roles: Ac-CP-1 has been localised to cephalic and excretory glands, which suggests that it is secreted into host tissue by the adult worm during attachment and feeding, while Ac-CP-2 has been localised to the brush border membrane of the worm intestine and is involved in Hb digestion (Loukas et al., 2004, Williamson et al., 2004). Therefore, to determine the roles of the N. americanus cysteine proteases described earlier in my study, immunolocalisation studies and functional assays were undertaken. Published studies of cysteine proteases from a range of parasites have demonstrated high levels of expression of secreted recombinant proteins from yeast, in particular, Pichia pastoris (Beckham et al., 2006, Sajid et al., 2003). I therefore adopted this strategy for the expression of all four N. americanus cysteine proteases, initially transforming plasmids into the X-33 strain of P. pastoris. However, only Na- CP-3 and CP-5 were expressed and secreted into the culture supernatant, and only Na-CP-3 was expressed at sufficiently high yield to pursue further characterisation. Numerous experimental parameters were modified in an attempt to obtain or improve expression of Na-CP-2, -4 and -5. These included transforming into the KM71H

114 strain of P. pastoris, addition of protease inhibitor cocktails (to prevent digestion of the recombinant protein by itself or by yeast proteases), varying the pH of the culture medium, and expression of the mature form (in the absence of the pro-region). None of these variables affected the levels of protein produced (data not shown). Further characterisation (in terms of catalytic activity) of the proteases Na-CP-2, -4 and -5 was therefore deemed unfeasible. As Na-CP-3 was the only recombinant protein which expressed in yeast at high yield, I restricted further analyses of enzymatic activity to this protein. When the pro-form of the protease was electrophoresed in a gelatin gel with dithiothreitol as the reducing agent, a zone of clearance was detected in the gel, indicating that cis- processing was occurring in the gel, to yield an active mature enzyme. In contrast to this result, catalytic activity could not be detected using clan CA-specific peptide substrates bearing a fluorochrome, indicating an absence of catalytic activity. MS analysis of the secreted recombinant protease revealed that the pro-region of the protein had not been cleaved off during secretion from P. pastoris. Incubation of recombinant Na-CP-3 for extended periods in low pH buffers with thiol agents did not result in auto-activation (data not shown). Similar scenarios have been described for F. hepatica FhCatB1 and S. mansoni SmCB1, where the pro-regions were not cleaved during secretion, thereby inhibiting catalytic activity (Beckham et al., 2006, Sajid et al., 2003). In the case of FhCatB1, the pro-region was removed by incubating the protein with a large negatively charged molecule such as dextran sulfate, while the SmCB1 pro-region was removed via a trans-processing event with an asparaginyl endopeptidase, an enzyme that belongs to another family of cysteine proteases and is also expressed in the schistosome gut (Beckham et al., 2006, Sajid et al., 2003). An asparaginyl endopeptidase has been reported from A. caninum (D. Smyth & A. Loukas, personal communication) and has also been shown to be expressed in the gut of H. contortus (Oliver et al., 2006), so it is likely that this enzyme is also expressed in the gut of N. americanus. Asparaginyl endopeptidases cleave on the C-terminal side of Asn residues. It is unlikely that Na-CP-3 is activated by asparaginyl endopeptidase due to the absence of an Asn residue in the vicinity of the predicted pro-mature junction. However, Na-CP-4 and CP-5 do possess Asn residues in this vicinity and may be candidates for trans-processing by asparaginyl endopeptidase.

115 To further address the activation status of pro-Na-CP-3, a similar strategy to that used for activation of FhCatB1 was employed (Beckham et al., 2006). This involved incubation of pro-Na-CP-3 with dextran sulfate, which resulted in cis- processing of the protein and auto-catalytic removal of the pro-region. The processed protease demonstrated catalytic activity against the diagnostic fluorogenic peptide Z- Phe-Arg-AMC. This result verified that Na-CP-3 was catalytically activity, thus the next step was to determine whether or not it was capable of cleaving intact Hb. Previous studies on cathepsin B proteases from blood-feeding parasites, such as S. mansoni SmCB1 and A. caninum Ac-CP-2, indicated that these proteins were capable of cleaving native Hb. However, Na-CP-3 could not cleave intact Hb, although it was capable of further digesting globin fragments which had initially been cleaved by the aspartic protease, Na-APR-1. Na-CP-3 is expressed in the hookworm gut and the cp- 3 mRNA is more highly expressed in gut tissue than any of the other cathepsin B mRNAs examined in this study, further suggesting that Na-CP-3 plays a pivitol role in nutrient acquisition. Although not tested in this study, it is possible that Na-CP-3 might play a more upstream role in cleavage of other blood proteins such as serum albumin, fibrinogen and immunoglobulins. Delcroix et al. reported that gene knockdown of S. mansoni cysteine proteases did not result in a significant reduction of haemoglobin digestion, but did substantially impact on digestion of albumin (Delcroix et al., 2006). Similarly Correnti et al, demonstrated that schistosomes treated with SmCB1-dsRNA were viable and developed intestinal heme pigmentation indicative of haemoglobin digestion, but showed significant growth retardation when compared to control parasites, indicating that SmCB1 function is not essential for haemoglobin digestion but is necessary for normal parasite growth (Correnti et al., 2005). Although catalytic activity was not established for the other three cysteine proteases (Na-CP-2, -4 and -5) because recombinant protein could not be expressed in native form in my study, it is speculated that these proteases are also involved in nutrient acquisition. Localisation studies conducted with antibodies produced against the denatured proteins demonstrated that all three proteases are expressed in the gut of the worm. Moreover, real time PCR results demonstrated that the mRNAs encoding these proteases were upregulated in adult worms, and were most highly expressed in the gut tissue. In addition, both Na-CP-2 and CP-4 possess the so called “haemoglobinase” motif as described by (Baig et al., 2002) and, although this does

116 not guarantee that these proteins are haemoglobinolytic, it is strongly suggestive of this function. Knox et al. demonstrated that vaccination of sheep with a thiol-sepharose binding protein (TSBP) extract from adult H. contortus reduced worm burdens in sheep intestines by 47% (Knox et al., 2005). Redmond et al. further demonstrated that vaccination with a cocktail of three different cysteine proteases resulted in 38% reduction in numbers of H. contortus in sheep compared to controls, indicating that the majority of the protective immunity generated by TSBP was due to the combined effects of a few cysteine proteases (Redmond and Knox, 2006). In similar fashion, all four N. americanus cysteine proteases described in this study might work synergistically, and present a better vaccine target when combined, rather than focusing on each isolated enzyme. Similarly, their roles could be redundant, as seen for the P. falciparum plasmepsins, and to a lesser extent, the falcipains (Liu et al., 2006), so that if one mRNA was suppressed, others could compensate with minimal or no effect on overall feeding of the worm. In like fashion, if redundancy exists amongst the cathepsins B, antibodies targeting just one protease might not directly interrupt blood feeding. A vaccine trial conducted with Ac-CP-2 did not result in reduced worm burden, but did cause a significant decrease in the size of adult worms and the fecundity of female worms (Loukas et al., 2004). Whilst the Ac-CP-2 vaccine was not as efficacious (in terms of reductions in adult worms) as the Ac-APR-1 vaccine (which had a reduction of worm burden by 33%) (Loukas et al., 2005a), it did have a significant effect on fecundity in particular, indicating that this group of proteases is a valid target for new chemotherapies. A recent study by Xiao et al showed that vaccination of hamsters with Na- CP-2 resulted in a 28% reduction in worm burdens (Xiao et al., 2008). While the level of protection was not particularly high, the Na-CP-2 immunogen that was used was expressed in E. coli in denatured form and could not be refolded in soluble form. As a result, the recombinant protein was not properly folded and would not have displayed many of the potential conformational epitopes of the native protein. If this vaccine trial was repeated with correctly folded protease, it is reasonable to assume that the level of protection would increase.

117 Metalloproteases In addition to aspartic and cysteine proteases, a number of hematophagous parasites, including the nematodes A. caninum and H. contortus, and the malaria parasite P. falciparum, express a third mechanistic class of proteases - the metalloproteases - that play a downstream role in haemoglobinolysis. Surprisingly, a metalloendopeptidase involved in haemoglobinolysis has not yet been reported from schistosomes, implying that the convergent evolution of this digestive process in blood-feeding parasites is more obvious in Plasmodium spp. and the blood-feeding nematodes than it is in the schistosomes. None of the metalloenzymes described so far from haematophagous parasites can digest intact Hb, and all appear to act downstream in the cascade after aspartic and/or cysteine proteases have generated globin peptides (Eggleson et al., 1999, Williamson et al., 2004). Presented here is the first report of identification and characterisation of a metalloprotease from N. americanus. Previously, an eotaxin-cleaving metalloprotease activity was described from N. americanus ES products, but neither the protein nor the cDNA were isolated (Culley et al., 2000). In my study, it was demonstrated that a metalloprotease (termed Na-MEP-1) is expressed in the gut of adult N. americanus and, in similar fashion to other metallo-haemoglobinases, it cannot digest intact Hb but instead digests globin fragments, providing further support for the concept of an ordered digestive process in hookworms. A vaccine trial has been conducted by Smith et al. with the four metalloproteases fractionated from the H. contortus H-gal-GP complex, which provided up to 50% reduction in egg count, however no significant protection was reported with bacterially expressed forms of the metalloproteases indicating that conformational epitopes are required for immunity (Smith et al., 2003). Unlike other proteases, little is know about what level of effective immunity can be provided by individual metalloprotease antigens. Nonetheless, vaccine trials with these molecules should be undertaken as they have been implicated in Hb degradation and in immune evasion.

Hidden antigens The strategy of identifying vaccine candidate molecules using hidden antigens such as gut molecules has been successfully implemented in both H. contortus and the cattle tick, Boophilus microplus (Munn, 1997). In the case of B. microplus, a gut membrane structural protein termed Bm86 has been produced in

118 recombinant form as a commercial vaccine and provides high levels of protection against tick (de la Fuente et al., 1999). While a commercial vaccine has yet to be produced from H. contortus gut proteins, a number of semi-purified extracts from the gut membrane have been shown to be highly protective in vaccine trials, with worm burden reductions as high as 90% (Knox and Smith, 2001). Interestingly, the protective gut antigen complexes from H. contortus comprise various classes of proteases including aspartic, cysteine, metalloendopeptiodases and microsomal aminopeptidases, indicating the complexity of the process involved in nutrient acquisition. What has hampered progress in the development of a commercial haemonchosis vaccine has been the inability to produce many of the haemoglobinases in recombinant soluble form, with appropriate post-translational modifications (Dalton et al., 2003), and at sufficient yields for viable scale-up. Nonetheless, what these studies clearly establish is that targeting proteases which are expressed in the gut and are involved in feeding is a valid approach to vaccine (and drug) development. The distinct benefit of targeting hidden antigens for a vaccine is that there has been no selective pressure placed upon these molecules by the host immune system, so many of the polymorphisms that are found in exposed extracellular antigens (Good et al., 2004), are not evident in gut proteins (Munn, 1997). Hookworms are smaller than Haemonchus sp., making dissection of gut tissue via standard methods (Jasmer et al., 2001) extremely difficult. Instead, in my study laser microscopy microdissection (LMM) was used to remove the gut – this was the first report of the use of LMM to dissect tissues from a parasitic helminth. The approach was highly successful and proved to be an excellent method for isolating specific tissue from defined hookworm organs – intestine and gonad. RNA isolated from the catapulted gut sections was used to synthesise cDNA which was then transformed into a plasmid library. Shotgun sequencing of clones from these gut tissue specific libraries from both N. americanus and A. caninum provided a snapshot of the intestinal transcriptome, particularly the cDNAs that encode proteases. Many of the H. contortus gut proteases, particularly those that form the H-gal-GP complex, are hidden from the host’s immune system (Newton and Munn, 1999), and are not ES products. By analogy, the haemoglobinases in the hookworm gut are also likely to be hidden antigens, but this has not been comprehensively addressed. In unpublished data from our laboratory (J. Mulvenna & A. Loukas, personal

119 communication), very few cysteine proteases were detected in the ES products of adult A. caninum using a proteomics approach. In my study, the transcriptomes of gut tissue from both N. americanus and A. caninum were investigated in order to gain a better understanding of the nature, abundance and diversity of the major proteases which are expressed in this organ and to identify potential vaccine candidates, particularly those that shared sequence identities with H. contortus proteases present in H-gal-GP.

Abundant hookworm gut transcripts The gene ontology (GO) classification system was used to identify the functional roles of some of the intestinal contigs. A large number of contigs could not be placed into specific GO categories, but of the ~30% of contigs which did receive a GO assignment, these had diverse predicted functions. The most frequently predicted GO functions for cDNAs from the gut of both species of hookworms was protein/ion binding and catalytic activity, indicative of nutrient digestion and uptake processes that occur in this tissue. Some mRNAs were particularly abundant, such as those encoding vitellogenin, fatty acid binding proteins and heat shock proteins. Vitellogenin was one of the most highly abundant transcripts in both A. caninum and N. americanus libraries, and while this protein is generally associated with embryogenesis, Caenorhabditis elegans vitellogenin is expressed in the intestine where it transports cholesterol from the gut to the oocytes (Matyash et al., 2001). As nematodes are auxotrophic for sterols, uptake and transportation of cholesterol to the gonads is essential for viable reproduction (Matyash et al., 2001). Heat shock protein 20 was another highly abundant gut transcript, which aligns with reports on the importance of this family of proteins during the transition of nematodes from the external environment into the mammalian host. In H. contortus, HSP-20 has been localised to both the intestine and reproductive organs and is thought to play a role in preparing parasites for the stress of moving from the pastures into a warm blooded host – specifically, the sudden rise in temperature and exposure to the host immune response (Hartman et al., 2003). Whilst a large number of gut mRNAs with GO assignments encoded general house-keeping proteins, numerous contigs that encoded proteins with suggested or proven roles in feeding were identified in my study. These included anticoagulants and platelet inhibitors, as well as all of the proteases described earlier. Hookworm

120 anticoagulants and platelet inhibitors are expressed in the oesophageal and cephalic glands from where they are secreted into host tissue at the attachment site (Del Valle et al., 2003), but their expression in the gut has not been reported previously. It is possible that they function in the parasite gut by inhibiting clot formation of ingested blood.

Novel hookworm gut transcripts Prior to this study, only one metalloprotease had been identified in the gut of adult hookworms – the M13 family member Ac-MEP-1 from A. caninum (Jones and Hotez, 2002). In my study, the N. americanus orthologue, Na-MEP-1, was identified. However, an additional four novel metalloproteases were identified in the intestinal EST datasets (Ranjit et al., 2006) - three of these belonged to the M12 family, also known as astacins, and one belonged to the M22 family of O-sialoglycoprotein endopeptidases. To date, M12 proteases have only been found in hookworm L3 larval stage ES products (Zhan et al., 2002) where they are thought to assist in tissue migration during entry into the mammalian host (Williamson et al., 2006). In C. elegans, M12 proteases are secreted into the alimentary tract where they are thought to be involved in digestion of food (Mohrlen et al., 2003), so it is feasible that these hookworm M12 astacin-like proteases are also involved in nutrient acquisition from blood. The identification of an O-sialoglycoprotein endopeptidase represents the first report of an M22 protease in parasitic nematodes. These enzymes are highly specific for O-sialoglycosylated proteins such as glycophorin A which is a component of red blood cells (Rawlings et al., 2008). Serine proteases do not appear to be widely expressed in parasitic nematodes. In my study an EST encoding a serine protease was detected from A. caninum gut cDNA library, this protease belongs to the clan CA, S1A family which is classified as chymotrypsin (Rawlings et al., 2008). In vertebrates, intestinal protein digestion is largely the work of serine proteases, primarily members of the trypsin family Apart from the serine protease activity which was detected in the H. contortus TBSP complex (Knox et al., 1999), there have not been any other reports of serine proteases involved in the feeding process in helminths, possibly indicating that this group of proteases does not play a major role in the blood digestion process in parasites. In contrast, serine proteases are integral to the digestion process in blood feeding insects such as Anopheles species (Muller et al., 1993). It is believed that the

121 evolutionary transition from cysteine/aspartic to serine proteases occurred in arthropods or molluscs (Delcroix et al., 2006). Other than APR-1 and APR-2, additional aspartic protease cDNAs were not identified in gut tissue from either hookworm species in my study, suggesting that only a small number of aspartic proteases are involved in Hb digestion, and this early phase of Hb digestion might not be as redundant as it is in Plasmodium, where at least four aspartic proteases are found in the digestive vacuole. My results further highlight the validity of targeting these enzymes as vaccines and drug targets. However it is important to note that only a small number of ESTs were sequenced from the gut cDNA libraries – it is noteworthy that I did not identify intestinal ESTs encoding aminopeptidases or haemolysins, two groups of proteins that are thought to be important in blood-feeding parasites. However, using PCR, Don et al. amplified two cDNAs encoding pore-forming saposins (Don et al., 2007) and an aminopeptidase (T. Don & A. Loukas, personal communication) from the A. caninum gut cDNA library, indicating the presence but not abundance of these enzymes in hookworm intestinal tissue.

6.2 CONCLUSION AND FUTURE DIRECTIONS Results presented in this thesis have verified the complex nature of the Hb degradation process in hookworms, as it was demonstrated that this process involves numerous proteases from various mechanistic classes. I have shown that aspartic proteases play a principal role in initiating the cleavage of the tetrameric Hb protein in the gut of adult N. americanus, while cysteine proteases and metalloproteases also play essential, but downstream, roles in this process after aspartic proteases have made initial cuts. The data presented implies that Hb digestion in N. americanus occurs in an ordered fashion, similar to that described in the digestive vacuole of Plasmodium falciparum, although there appears to be less redundancy in hookworms, at least in the early stages of the process. The conclusions drawn have established that all three mechanistic classes of proteases are needed for Hb degradation to occur efficiently. The aspartic protease Na-APR-1 plays the pivotal role in initiating the process, and therefore might be considered a good target for vaccine or drug development. Vaccine trials with Ac-APR-1 followed by a heterologous challenge with N. americanus L3 have already been conducted in a

122 hamster model of necatoriasis, and a 44% reduction in adult worms was obtained (Xiao et al., 2008). My studies suggest that even greater reductions might be achieved when hamsters are immunized with Na-APR-1 and receive a homologous N. americanus larval challenge. These trials, including assessment of the vaccine efficacy of Na-CP-3 and –MEP-1 are now underway at the Institute of Parastic Diseases in China. My investigation of the gut transcriptional profile of adult hookworms indicated that proteases are, as expected, abundantly represented in this tissue. There were numerous additional proteases identified from gut tissue ESTs, other than those investigated in detail in this study. Many of these might have roles in haemoglobinolysis process, thus further investigation of these enymes is necessary. Moreover, there were numerous other transcripts identified which are predicted to be involved in various other processes that could be targeted as vaccines, including haemolysins, lipid/cholesterol binding proteins and amino acid transporters, some of which are now under examination as vaccine candidates. Further research into the up and downstream process of Hb degradation in hookworms is required as it is still unknown how red blood cells are lysed to release Hb or the process by which amino acids are transported across gut cells. As yet, no exopeptidases (caboxy and amino) have been indentified in the Hb digestion process in hookworms, however these enzymes have been implicated in a down stream role in other haematophagous parasites such as S. manosoni and P. falciparum. Thus it is highly probable that homologous proteases are also utilised by hookworms in this process, and idenfication of such enzymes is warranted. Due to the sheer number of people infected with hookworms worldwide, it is imperative that an efficacious vaccine is produced in order to combat this insidious disease. The ultimate hookworm vaccine would be one that targets both the larval and adult stages, assaulting the parasite at both major developmental stages. As with other haematophagous parasites, Hb degradation is essential for the survival of these parasites and, therefore, further dissection of the molecular mechanisms of this process is essential if we are to exploit the requirement of hookworms for blood feeding. This will, and indeed already has, facilitated the identification and development of target molecules that will form the basis of molecular vaccines against this and other neglected diseases of humans and animals.

123

REFERENCES

ACOSTA, D., CANCELA, M., PIACENZA, L., ROCHE, L., CARMONA, C. & TORT, J. F. (2008) Fasciola hepatica leucine aminopeptidase, a promising candidate for vaccination against ruminant . Mol Biochem Parasitol, 158, 52-64. ALBONICO, M. (2003) Methods to sustain drug efficacy in helminth control programmes. Acta Trop, 86, 233-42. ALBONICO, M., BICKLE, Q., RAMSAN, M., MONTRESOR, A., SAVIOLI, L. & TAYLOR, M. (2003) Efficacy of mebendazole and levamisole alone or in combination against intestinal nematode infections after repeated targeted mebendazole treatment in Zanzibar. Bull World Health Organ, 81, 343-52. ALBONICO, M., STOLTZFUS, R. J., SAVIOLI, L., TIELSCH, J. M., CHWAYA, H. M., ERCOLE, E. & CANCRINI, G. (1998) Epidemiological evidence for a differential effect of hookworm species, Ancylostoma duodenale or Necator americanus, on iron status of children. Int J Epidemiol, 27, 530-7. ALBONICO, M., WRIGHT, V. & BICKLE, Q. (2004) Molecular analysis of the beta-tubulin gene of human hookworms as a basis for possible benzimidazole resistance on Pemba Island. Mol Biochem Parasitol, 134, 281-4. ALLEN, J. E. & MACDONALD, A. S. (1998) Profound suppression of cellular proliferation mediated by the secretions of nematodes. Parasite Immunol, 20, 241-7. ASOJO, O. A., GOUD, G., DHAR, K., LOUKAS, A., ZHAN, B., DEUMIC, V., LIU, S., BORGSTAHL, G. E. & HOTEZ, P. J. (2005) X-ray structure of Na- ASP-2, a pathogenesis-related-1 protein from the nematode parasite, Necator americanus, and a vaccine antigen for human hookworm infection. J Mol Biol, 346, 801-14. BAIG, S., DAMIAN, R. T. & PETERSON, D. S. (2002) A novel cathepsin B active site motif is shared by helminth bloodfeeders. Exp Parasitol, 101, 83-9. BANERJEE, R., LIU, J., BEATTY, W., PELOSOF, L., KLEMBA, M. & GOLDBERG, D. E. (2002) Four plasmepsins are active in the Plasmodium falciparum food vacuole, including a protease with an active-site histidine. Proc Natl Acad Sci U S A, 99, 990-5. BARRETT, A. J. (1994) Classification of peptidases. Methods Enzymol, 244, 1-15. BASAVARAJU, S., ZHAN, B., KENNEDY, M. W., LIU, Y., HAWDON, J. & HOTEZ, P. J. (2003) Ac-FAR-1, a 20 kDa fatty acid- and retinol-binding protein secreted by adult Ancylostoma caninum hookworms: gene transcription pattern, ligand binding properties and structural characterisation. Mol Biochem Parasitol, 126, 63-71. BECKHAM, S. A., LAW, R. H., SMOOKER, P. M., QUINSEY, N. S., CAFFREY, C. R., MCKERROW, J. H., PIKE, R. N. & SPITHILL, T. W. (2006) Production and processing of a recombinant Fasciola hepatica cathepsin B- like enzyme (FhcatB1) reveals potential processing mechanisms in the parasite. Biol Chem, 387, 1053-61. BETHONY, J. M., SIMON, G., DIEMERT, D. J., PARENTI, D., DESROSIERS, A., SCHUCK, S., FUJIWARA, R., SANTIAGO, H. & HOTEZ, P. J. (2008) Randomized, placebo-controlled, double-blind trial of the Na-ASP-2 hookworm vaccine in unexposed adults. Vaccine, 26, 2408-17. BETHONY, J., BROOKER, S., ALBONICO, M., GEIGER, S. M., LOUKAS, A., DIEMERT, D. & HOTEZ, P. J. (2006a) Soil-transmitted helminth infections: , , and hookworm. Lancet, 367, 1521-32.

125 BETHONY, J., CHEN, J., LIN, S., XIAO, S., ZHAN, B., LI, S., XUE, H., XING, F., HUMPHRIES, D., YAN, W., CHEN, G., FOSTER, V., HAWDON, J. M. & HOTEZ, P. J. (2002) Emerging patterns of hookworm infection: influence of aging on the intensity of Necator infection in Hainan Province, People's Republic of China. Clin Infect Dis, 35, 1336-44. BETHONY, J., LOUKAS, A., SMOUT, M., BROOKER, S., MENDEZ, S., PLIESKATT, J., GOUD, G., BOTTAZZI, M. E., ZHAN, B., WANG, Y., WILLIAMSON, A., LUSTIGMAN, S., CORREA-OLIVEIRA, R., XIAO, S. & HOTEZ, P. J. (2005) Antibodies against a secreted protein from hookworm larvae reduce the intensity of hookworm infection in humans and vaccinated laboratory animals. Faseb J, 19, 1743-5. BETHONY, J. M., LOUKAS, A., HOTEZ, P. J. & KNOX, D. P. (2006b) Vaccines against blood-feeding nematodes of humans and livestock. Parasitology, 133 Suppl, S63-79. BLACKWELL, V. & VEGA-LOPEZ, F. (2001) Cutaneous larva migrans: clinical features and management of 44 cases presenting in the returning traveller. Br J Dermatol, 145, 434-7. BOGITSH, B. J., DALTON, J. P., BRADY, C. P. & BRINDLEY, P. J. (2001) Gut- associated immunolocalization of the Schistosoma mansoni cysteine proteases, SmCL1 and SmCL2. J Parasitol, 87, 237-41. BRINDLEY, P. J., KALINNA, B. H., DALTON, J. P., DAY, S. R., WONG, J. Y., SMYTHE, M. L. & MCMANUS, D. P. (1997) Proteolytic degradation of host hemoglobin by schistosomes. Mol Biochem Parasitol, 89, 1-9. BRINDLEY, P. J., KALINNA, B. H., WONG, J. Y., BOGITSH, B. J., KING, L. T., SMYTH, D. J., VERITY, C. K., ABBENANTE, G., BRINKWORTH, R. I., FAIRLIE, D. P., SMYTHE, M. L., MILBURN, P. J., BIELEFELDT- OHMANN, H., ZHENG, Y. & MCMANUS, D. P. (2001) Proteolysis of human hemoglobin by schistosome cathepsin D. Mol Biochem Parasitol, 112, 103-12. BROOKER, S., BETHONY, J. & HOTEZ, P. J. (2004) Human hookworm infection in the 21st century. Advances in Parasitology, Vol 58. BROWN, A., GIROD, N., BILLETT, E. E. & PRITCHARD, D. I. (1999) Necator americanus (human hookworm) aspartyl proteinases and digestion of skin macromolecules during skin penetration. Am J Trop Med Hyg, 60, 840-7. BUNDY, D. A., CHAN, M. S. & SAVIOLI, L. (1995) Hookworm infection in pregnancy. Trans R Soc Trop Med Hyg, 89, 521-2. BUNGIRO, R. & CAPPELLO, M. (2004) Hookworm infection: new developments and prospects for control. Curr Opin Infect Dis, 17, 421-6. CAFFREY, C. R., MCKERROW, J. H., SALTER, J. P. & SAJID, M. (2004) Blood 'n' guts: an update on schistosome digestive peptidases. Trends Parasitol, 20, 241-8. CAFFREY, C. R. & RUPPEL, A. (1997) Affinity isolation and characterization of the cathepsin B-like proteinase Sj31 from . J Parasitol, 83, 1112-8. CAGLIC, D., PUNGERCAR, J. R., PEJLER, G., TURK, V. & TURK, B. (2007) Glycosaminoglycans facilitate procathepsin B activation through disruption of propeptide-mature enzyme interactions. J Biol Chem, 282, 33076-85. CARROLL, S. M. & GROVE, D. I. (1986) Experimental infection of humans with : clinical, parasitological, haematological and immunological findings. Trop Geogr Med, 38, 38-45.

126 CHADDERDON, R. C. & CAPPELLO, M. (1999) The hookworm platelet inhibitor: functional blockade of integrins GPIIb/IIIa (alphaIIbbeta3) and GPIa/IIa (alpha2beta1) inhibits platelet aggregation and adhesion in vitro. J Infect Dis, 179, 1235-41. CORRENTI, J. M., BRINDLEY, P. J. & PEARCE, E. J. (2005) Long-term suppression of cathepsin B levels by RNA interference retards schistosome growth. Mol Biochem Parasitol, 143, 209-15. CROLL, N. A. & GHADIRAIN, E. (1981) Wormy persons: contribuitons to the nature and patterns of overdipersion with Ascaris lumbricoides, Ancylostoma duodenale, Necator americanus and Trichuris trichiura. Trop Geogr Med, 33, 241-8. CROMPTON, D. W. (2000) The public health importance of hookworm disease. Parasitology, 121 Suppl, S39-50. CULLEY, F. J., BROWN, A., CONROY, D. M., SABROE, I., PRITCHARD, D. I. & WILLIAMS, T. J. (2000) Eotaxin is specifically cleaved by hookworm metalloproteases preventing its action in vitro and in vivo. J Immunol, 165, 6447-53. DALAL, S. & KLEMBA, M. (2007) Roles for two aminopeptidases in vacuolar hemoglobin catabolism in Plasmodium falciparum. J Biol Chem, 282, 35978- 87. DALTON, J. P., BRINDLEY, P. J., KNOX, D. P., BRADY, C. P., HOTEZ, P. J., DONNELLY, S., O'NEILL, S. M., MULCAHY, G. & LOUKAS, A. (2003) Helminth vaccines: from mining genomic information for vaccine targets to systems used for protein expression. Int J Parasitol, 33, 621-40. DALTON, J. P., CLOUGH, K. A., JONES, M. K. & BRINDLEY, P. J. (1997) The cysteine proteinases of Schistosoma mansoni cercariae. Parasitology, 114 ( Pt 2), 105-12. DATU, B. J., GASSER, R. B., NAGARAJ, S. H., ONG, E. K., O'DONOGHUE, P., MCINNES, R., RANGANATHAN, S. & LOUKAS, A. (2008) Transcriptional changes in the hookworm, Ancylostoma caninum, during the Transition from a Free-Living to a Parasitic Larva. PLoS Negl Trop Dis, 2, e130. DAUB, J., LOUKAS, A., PRITCHARD, D. I. & BLAXTER, M. (2000) A survey of genes expressed in adults of the human hookworm, Necator americanus. Parasitology, 120 ( Pt 2), 171-84. DE CLERCQ, D., SACKO, M., BEHNKE, J., GILBERT, F., DORNY, P. & VERCRUYSSE, J. (1997) Failure of mebendazole in treatment of human hookworm infections in the southern region of Mali. Am J Trop Med Hyg, 57, 25-30. DE LA FUENTE, J., RODRIGUEZ, M., MONTERO, C., REDONDO, M., GARCIA-GARCIA, J. C., MENDEZ, L., SERRANO, E., VALDES, M., ENRIQUEZ, A., CANALES, M., RAMOS, E., BOUE, O., MACHADO, H. & LLEONART, R. (1999) Vaccination against ticks (Boophilus spp.): the experience with the Bm86-based vaccine Gavac. Genet Anal, 15, 143-8. DE SILVA, N. R., BROOKER, S., HOTEZ, P. J., MONTRESOR, A., ENGELS, D. & SAVIOLI, L. (2003) Soil-transmitted helminth infections: updating the global picture. Trends in Parasitology, 19, 547-551. DEL VALLE, A., JONES, B. F., HARRISON, L. M., CHADDERDON, R. C. & CAPPELLO, M. (2003) Isolation and molecular cloning of a secreted

127 hookworm platelet inhibitor from adult Ancylostoma caninum. Mol Biochem Parasitol, 129, 167-77. DELCROIX, M., SAJID, M., CAFFREY, C. R., LIM, K. C., DVORAK, J., HSIEH, I., BAHGAT, M., DISSOUS, C. & MCKERROW, J. H. (2006) A multienzyme network functions in intestinal protein digestion by a platyhelminth parasite. J Biol Chem, 281, 39316-29. DIEMERT, D. J., BETHONY, J. M. & HOTEZ, P. J. (2008) Hookworm vaccines. Clin Infect Dis, 46, 282-8. DON, T. A., JONES, M. K., SMYTH, D., O'DONOGHUE, P., HOTEZ, P. & LOUKAS, A. (2004) A pore-forming haemolysin from the hookworm, Ancylostoma caninum. Int J Parasitol, 34, 1029-1035. DON, T. A., OKSOV, Y., LUSTIGMAN, S. & LOUKAS, A. (2007) Saposin-like proteins from the intestine of the blood-feeding hookworm, Ancylostoma caninum. Parasitology, 134, 427-36. DRICKAMER, K. & DODD, R. B. (1999) C-Type lectin-like domains in Caenorhabditis elegans: predictions from the complete genome sequence. Glycobiology, 9, 1357-69. EGGLESON, K. K., DUFFIN, K. L. & GOLDBERG, D. E. (1999) Identification and characterization of falcilysin, a metallopeptidase involved in hemoglobin catabolism within the malaria parasite Plasmodium falciparum. J Biol Chem, 274, 32411-7. FUJIWARA, R. T., LOUKAS, A., MENDEZ, S., WILLIAMSON, A. L., BUENO, L. L., WANG, Y., SAMUEL, A., ZHAN, B., BOTTAZZI, M. E., HOTEZ, P. J. & BETHONY, J. M. (2006) Vaccination with irradiated Ancylostoma caninum third stage larvae induces a Th2 protective response in dogs. Vaccine, 24, 501-9. FUJIWARA, R. T., ZHAN, B., MENDEZ, S., LOUKAS, A., BUENO, L. L., WANG, Y., PLIESKATT, J., OKSOV, Y., LUSTIGMAN, S., BOTTAZZI, M. E., HOTEZ, P. & BETHONY, J. M. (2007) Reduction of worm fecundity and canine host blood loss mediates protection against hookworm infection elicited by vaccination with recombinant Ac-16. Clin Vaccine Immunol, 14, 281-7. FURMIDGE, B. A., HORN, L. A. & PRITCHARD, D. I. (1996) The anti- haemostatic strategies of the human hookworm Necator americanus. Parasitology, 112 ( Pt 1), 81-7. GAMBLE, H. R., PURCELL, J. P. & FETTERER, R. H. (1989) Purification of a 44 kilodalton protease which mediates the ecdysis of infective Haemonchus contortus larvae. Mol Biochem Parasitol, 33, 49-58. GAVIGAN, C. S., DALTON, J. P. & BELL, A. (2001) The role of aminopeptidases in haemoglobin degradation in Plasmodium falciparum-infected erythrocytes. Mol Biochem Parasitol, 117, 37-48. GHOSH, K., HAWDON, J. & HOTEZ, P. (1996) Vaccination with alum- precipitated recombinant Ancylostoma-secreted protein 1 protects mice against challenge infections with infective hookworm (Ancylostoma caninum) larvae. J. Infect. Dis, 174, 1380-1383. GHOSH, K. & HOTEZ, P. J. (1999) Antibody-dependent reductions in mouse hookworm burden after vaccination with Ancylostoma caninum secreted protein 1. J Infect Dis, 180, 1674-81. GOLDBERG, D. E. (2005) Hemoglobin degradation. Curr Top Microbiol Immunol, 295, 275-91.

128 GOOD, M. F., STANISIC, D., XU, H., ELLIOTT, S. & WYKES, M. (2004) The immunological challenge to developing a vaccine to the blood stages of malaria parasites. Immunol Rev, 201, 254-67. GOUD, G. N., BOTTAZZI, M. E., ZHAN, B., MENDEZ, S., DEUMIC, V., PLIESKATT, J., LIU, S., WANG, Y., BUENO, L., FUJIWARA, R., SAMUEL, A., AHN, S. Y., SOLANKI, M., ASOJO, O. A., WANG, J., BETHONY, J. M., LOUKAS, A., ROY, M. & HOTEZ, P. J. (2005) Expression of the Necator americanus hookworm larval antigen Na-ASP-2 in Pichia pastoris and purification of the recombinant protein for use in human clinical trials. Vaccine, 23, 4754-64. GOUD, G. N., ZHAN, B., GHOSH, K., LOUKAS, A., HAWDON, J., DOBARDZIC, A., DEUMIC, V., LIU, S., DOBARDZIC, R., ZOOK, B. C., JIN, Q., LIU, Y. Y., HOFFMAN, L., CHUNG-DEBOSE, S., PATEL, R., MENDEZ, S. & HOTEZ, P. J. (2004) Cloning, yeast expression, isolation, and vaccine testing of recombinant Ancylostoma-secreted protein (ASP)-1 and ASP-2 from Ancylostoma ceylanicum. J Infect Dis, 189, 919-929. HARROP, S. A., SAWANGJAROEN, N., PROCIV, P. & BRINDLEY, P. J. (1995) Characterization and localization of cathepsin B proteinases expressed by adult Ancylostoma caninum hookworms. Mol Biochem Parasitol, 71, 163-71. HARTMAN, D., COTTEE, P. A., SAVIN, K. W., BHAVE, M., PRESIDENTE, P. J., FULTON, L., WALKIEWICZ, M. & NEWTON, S. E. (2003) Haemonchus contortus: molecular characterisation of a small heat shock protein. Exp Parasitol, 104, 96-103. HAWDON, J. M. & DATU, B. (2003) The second messenger cyclic GMP mediates activation in Ancylostoma caninum infective larvae. Int J Parasitol, 33, 787- 93. HAWDON, J. M. & HOTEZ, P. J. (1996) Hookworm: developmental biology of the infectious process. Curr Opin Genet Dev, 6, 618-23. HAWDON, J. M., JONES, B. F., HOFFMAN, D. R. & HOTEZ, P. J. (1996) Cloning and characterization of Ancylostoma-secreted protein. A novel protein associated with the transition to parasitism by infective hookworm larvae. J Biol Chem, 271, 6672-8. HAWDON, J. M., NARASIMHAN, S. & HOTEZ, P. J. (1999) Ancylostoma secreted protein 2: cloning and characterization of a second member of a family of nematode secreted proteins from Ancylostoma caninum. Mol Biochem Parasitol, 99, 149-65. HOOPER, N. M. (1994) Families of zinc metalloproteases. FEBS Lett, 354, 1-6. HOTEZ, P., DE SILVA, N., BROOKER, S. & BETHONY, J. (2003a) Soil transmitted helminth infections: The nature, causes and burdent of the condition. DCPP Working Paper, 3. HOTEZ, P. J. (2007) Hookworm and Poverty. Ann N Y Acad Sci. HOTEZ, P. J., ASHCOM, J., BIN, Z., BETHONY, J., WILLIAMSON, A., HAWDON, J. M., JIANJUN, F., DOBARDZIC, A., RIZO, I., BOLDEN, J., JIN, Q., YAN, W., DOBARDZIC, R., CHUNG-DEBOSE, S., CROWELL, M., DATU, B., DELANEY, A., DRAGONOVSKI, D., JIANG, Y., LIU, Y. Y., GHOSH, K., LOUKAS, A., BRANDT, W., RUSSELL, P. K. & ZOOK, B. C. (2002) Effect of vaccinations with recombinant fusion proteins on Ancylostoma caninum habitat selection in the canine intestine. Journal of Parasitology, 88, 684-690.

129 HOTEZ, P. J., ASHCOM, J., ZHAN, B., BETHONY, J., LOUKAS, A., HAWDON, J., WANG, Y., JIN, Q., JONES, K. C., DOBARDZIC, A., DOBARDZIC, R., BOLDEN, J., ESSIET, I., BRANDT, W., RUSSELL, P. K., ZOOK, B. C., HOWARD, B. & CHACON, M. (2003b) Effect of vaccination with a recombinant fusion protein encoding an astacinlike metalloprotease (MTP-1) secreted by host-stimulated Ancylostoma caninum third-stage infective larvae. Journal of Parasitology, 89, 853-855. HOTEZ, P. J., BETHONY, J., BOTTAZZI, M. E., BROOKER, S. & BUSS, P. (2005) Hookworm: "The great infection of mankind". Plos Medicine, 2, 187- 191. HOTEZ, P. J., BETHONY, J., BOTTAZZI, M. E., BROOKER, S., DIEMERT, D. & LOUKAS, A. (2006) New technologies for the control of human hookworm infection. Trends Parasitol, 22, 327-31. HOTEZ, P. J., BROOKER, S., BETHONY, J. M., BOTTAZZI, M. E., LOUKAS, A. & XIAO, S. H. (2004) Hookworm infection. New England Journal of Medicine, 351, 799-807. HOTEZ, P. J., HAGGERTY, J., HAWDON, J., MILSTONE, L., GAMBLE, H. R., SCHAD, G. & RICHARDS, F. F. (1990) Metalloproteases of infective Ancylostoma hookworm larvae and their possible functions in tissue invasion and ecdysis. Infect Immun, 58, 3883-92. HOTEZ, P. J., NARASIMHAN, S., HAGGERTY, J., MILSTONE, L., BHOPALE, V., SCHAD, G. & RICHARDS, F. F. (1992) Hyaluronidase from infective Ancylostoma hookworm larvae and its possible function as a virulence factor in tissue invasion and in cutaneous larvae migrans. Infect Immun, 60, 1018- 23. HOTEZ, P. J. & PRITCHARD, D. I. (1995) Hookworm infection. Sci Am, 272, 68- 74. HOTEZ, P. J., ZHAN, B., BETHONY, J. M., LOUKAS, A., WILLIAMSON, A., GOUD, G. N., HAWDON, J. M., DOBARDZIC, A., DOBARDZIC, R., GHOSH, K., BOTTAZZI, M. E., MENDEZ, S., ZOOK, B., WANG, Y., LIU, S., ESSIET-GIBSON, I., CHUNG-DEBOSE, S., XIAO, S., KNOX, D., MEAGHER, M., INAN, M., CORREA-OLIVEIRA, R., VILK, P., SHEPHERD, H. R., BRANDT, W. & RUSSELL, P. K. (2003c) Progress in the development of a recombinant vaccine for human hookworm disease: the Human Hookworm Vaccine Initiative. Int J Parasitol, 33, 1245-58. HSIEH, G. C., LOUKAS, A., WAHL, A. M., BHATIA, M., WANG, Y., WILLIAMSON, A. L., KEHN, K. W., MARUYAMA, H., HOTEZ, P. J., LEITENBERG, D., BETHONY, J. & CONSTANT, S. L. (2004) A secreted protein from the human hookworm Necator americanus binds selectively to NK cells and induces IFN-gamma production. J Immunol, 173, 2699-704. ILLY, C., QURAISHI, O., WANG, J., PURISIMA, E., VERNET, T. & MORT, J. S. (1997) Role of the occluding loop in cathepsin B activity. J Biol Chem, 272, 1197-202. JABBAR, A., IQBAL, Z., KERBOEUF, D., MUHAMMAD, G., KHAN, M. N. & AFAQ, M. (2006) Anthelmintic resistance: the state of play revisited. Life Sci, 79, 2413-31. JASMER, D. P., MITREVA, M. D. & MCCARTER, J. P. (2004) mRNA sequences for Haemonchus contortus intestinal cathepsin B-like cysteine proteases display an extreme in abundance and diversity compared with other adult mammalian parasitic nematodes. Mol Biochem Parasitol, 137, 297-305.

130 JASMER, D. P., ROTH, J. & MYLER, P. J. (2001) Cathepsin B-like cysteine proteases and Caenorhabditis elegans homologues dominate gene products expressed in adult Haemonchus contortus intestine. Mol Biochem Parasitol, 116, 159-69. JONES, B. F. & HOTEZ, P. J. (2002) Molecular cloning and characterization of Ac- mep-1, a developmentally regulated gut luminal metalloendopeptidase from adult Ancylostoma caninum hookworms. Mol Biochem Parastiol, 119, 107- 116. JONES, M. K., RANDALL, L. M., MCMANUS, D. P. & ENGWERDA, C. R. (2004) Laser microdissection microscopy in parasitology: meet thermocyclers. Trends Parasitol, 20, 502-6. KARRER, K. M., PEIFFER, S. L. & DITOMAS, M. E. (1993) Two distinct gene subfamilies within the family of cysteine protease genes. Proc Natl Acad Sci U S A, 90, 3063-7. KENNEDY, M. W. (2000) The polyprotein lipid binding proteins of nematodes. Biochim Biophys Acta, 1476, 149-64. KNOX, D. P., GELDHOF, P., VISSER, A. & BRITTON, C. (2007) RNA interference in parasitic nematodes of animals: a reality check? Trends Parasitol, 23, 105-7. KNOX, D. P., REDMOND, D. L., NEWLANDS, G. F., SKUCE, P. J., PETTIT, D. & SMITH, W. D. (2003) The nature and prospects for gut membrane proteins as vaccine candidates for Haemonchus contortus and other ruminant trichostrongyloids. Int J Parasitol, 33, 1129-37. KNOX, D. P., REDMOND, D. L., SKUCE, P. J. & NEWLANDS, G. F. (2001) The contribution of molecular biology to the development of vaccines against nematode and trematode parasites of domestic ruminants. Vet Parasitol, 101, 311-35. KNOX, D. P., SMITH, S. K., REDMOND, D. L. & SMITH, W. D. (2005) Protection induced by vaccinating sheep with a thiol-binding extract of Haemonchus contortus membranes is associated with its protease components. Parasite Immunol, 27, 121-6. KNOX, D. P., SMITH, S. K. & SMITH, W. D. (1999) Immunization with an affinity purified protein extract from the adult parasite protects lambs against infection with Haemonchus contortus. Parasite Immunol, 21, 201-10. KNOX, D. P. & SMITH, W. D. (2001) Vaccination against gastrointestinal nematode parasites of ruminants using gut-expressed antigens. Vet Parasitol, 100, 21-32. KUMAR, S. & PRITCHARD, D. I. (1992) The partial characterization of proteases present in the excretory/secretory products and exsheathing fluid of the infective (L3) larva of Necator americanus. Int J Parasitol, 22, 563-72. LIPPS, G., FULLKRUG, R. & BECK, E. (1996) Cathepsin B of Schistosoma mansoni. Purification and activation of the recombinant proenzyme secreted by Saccharomyces cerevisiae. J Biol Chem, 271, 1717-25. LIU, J., GLUZMAN, I. Y., DREW, M. E. & GOLDBERG, D. E. (2005) The role of Plasmodium falciparum food vacuole plasmepsins. J Biol Chem, 280, 1432-7. LIU, J., ISTVAN, E. S., GLUZMAN, I. Y., GROSS, J. & GOLDBERG, D. E. (2006) Plasmodium falciparum ensures its amino acid supply with multiple acquisition pathways and redundant proteolytic enzyme systems. Proc Natl Acad Sci U S A, 103, 8840-5.

131 LONGBOTTOM, D., REDMOND, D. L., RUSSELL, M., LIDDELL, S., SMITH, W. D. & KNOX, D. P. (1997) Molecular cloning and characterisation of a putative aspartate proteinase associated with a gut membrane protein complex from adult Haemonchus contortus. Mol Biochem Parasitol, 88, 63-72. LOUKAS, A., BETHONY, J., BROOKER, S. & HOTEZ, P. (2006) Hookworm vaccines: past, present, and future. Lancet Infect Dis, 6, 733-41. LOUKAS, A., BETHONY, J. M., MENDEZ, S., FUJIWARA, R. T., GOUD, G. N., RANJIT, N., ZHAN, B., JONES, K., BOTTAZZI, M. E. & HOTEZ, P. J. (2005a) Vaccination with recombinant aspartic hemoglobinase reduces parasite load and blood loss after hookworm infection in dogs. PLoS Med, 2, e295. LOUKAS, A., BETHONY, J. M., WILLIAMSON, A. L., GOUD, G. N., MENDEZ, S., ZHAN, B., HAWDON, J. M., ELENA BOTTAZZI, M., BRINDLEY, P. J. & HOTEZ, P. J. (2004) Vaccination of dogs with a recombinant cysteine protease from the intestine of canine hookworms diminishes the fecundity and growth of worms. J Infect Dis, 189, 1952-61. LOUKAS, A., CONSTANT, S. L. & BETHONY, J. M. (2005b) Immunobiology of hookworm infection. FEMS Immunology and Medical Microbiology, 43, 115- 124. LOUKAS, A., DOWD, A. J., PROCIV, P. & BRINDLEY, P. J. (2000) Purification of a diagnostic, secreted cysteine protease-like protein from the hookworm Ancylostoma caninum. Parasitol Int, 49, 327-33. LOUKAS, A. & PROCIV, P. (2001) Immune responses in hookworm infections. Clin Microbiol Rev, 14, 689-703. LWAMBO, N. J., SIZA, J. E., BROOKER, S., BUNDY, D. A. & GUYATT, H. (1999) Patterns of concurrent hookworm infection and in schoolchildren in Tanzania. Trans R Soc Trop Med Hyg, 93, 497-502. MALLO, G. V., KURZ, C. L., COUILLAULT, C., PUJOL, N., GRANJEAUD, S., KOHARA, Y. & EWBANK, J. J. (2002) Inducible antibacterial defense system in C. elegans. Curr Biol, 12, 1209-14. MATYASH, V., GEIER, C., HENSKE, A., MUKHERJEE, S., HIRSH, D., THIELE, C., GRANT, B., MAXFIELD, F. R. & KURZCHALIA, T. V. (2001) Distribution and transport of cholesterol in Caenorhabditis elegans. Mol Biol Cell, 12, 1725-36. MAXWELL, C., HUSSAIN, R., NUTMAN, T. B., POINDEXTER, R. W., LITTLE, M. D., SCHAD, G. A. & OTTESEN, E. A. (1987) The clinical and immunologic responses of normal human volunteers to low dose hookworm (Necator americanus) infection. Am J Trop Med Hyg, 37, 126-34. MCCARTHY, E., STACK, C., DONNELLY, S. M., DOYLE, S., MANN, V. H., BRINDLEY, P. J., STEWART, M., DAY, T. A., MAULE, A. G. & DALTON, J. P. (2004) Leucine aminopeptidase of the human blood flukes, Schistosoma mansoni and Schistosoma japonicum. Int J Parasitol, 34, 703- 14. MCLAREN, D. J., BURT, J. S. & OGILVIE, B. M. (1974) The anterior glands of adult Necator americanus (Nematoda: Strongyloidea). II. Cytochemical and functional studies. Int J Parasitol, 4, 39-46. MIESZCZANEK, J., HARRISON, L. M., VLASUK, G. P. & CAPPELLO, M. (2004) Anticoagulant peptides from Ancylostoma caninum are immunologically distinct and localize to separate structures within the adult hookworm. Mol Biochem Parasitol, 133, 319-23.

132 MILLER, T. A. (1967) Immunity of dogs to Ancylostoma braziliense infection following vaccination with X-irradiated Ancylostoma caninum larvae. J Am Vet Med Assoc, 150, 508-15. MILLER, T. A. (1978) Industrial development and field use of the canine hookworm vaccine. Adv Parasitol, 16, 333-42. MITREVA, M., MCCARTER, J. P., ARASU, P., HAWDON, J., MARTIN, J., DANTE, M., WYLIE, T., XU, J., STAJICH, J. E., KAPULKIN, W., CLIFTON, S. W., WATERSTON, R. H. & WILSON, R. K. (2005) Investigating hookworm genomes by comparative analysis of two Ancylostoma species. BMC Genomics, 6, 58. MIYAKE, K., UNDERHILL, C. B., LESLEY, J. & KINCADE, P. W. (1990) Hyaluronate can function as a cell adhesion molecule and CD44 participates in hyaluronate recognition. J Exp Med, 172, 69-75. MOHRLEN, F., HUTTER, H. & ZWILLING, R. (2003) The astacin protein family in Caenorhabditis elegans. Eur J Biochem, 270, 4909-20. MORALES, M. E., RINALDI, G., GOBERT, G. N., KINES, K. J., TORT, J. F. & BRINDLEY, P. J. (2008) RNA interference of Schistosoma mansoni cathepsin D, the apical enzyme of the hemoglobin proteolysis cascade. Mol Biochem Parasitol, 157, 160-8. MOYLE, M., FOSTER, D. L., MCGRATH, D. E., BROWN, S. M., LAROCHE, Y., DE MEUTTER, J., STANSSENS, P., BOGOWITZ, C. A., FRIED, V. A., ELY, J. A. & ET AL. (1994) A hookworm glycoprotein that inhibits neutrophil function is a ligand of the integrin CD11b/CD18. J Biol Chem, 269, 10008-15. MULLER, H. M., CRAMPTON, J. M., DELLA TORRE, A., SINDEN, R. & CRISANTI, A. (1993) Members of a trypsin gene family in Anopheles gambiae are induced in the gut by blood meal. EMBO J, 12, 2891-900. MUNN, E. A. (1997) Rational design of nematode vaccines: hidden antigens. Int J Parasitol, 27, 359-66. MUSIL, D., ZUCIC, D., TURK, D., ENGH, R. A., MAYR, I., HUBER, R., POPOVIC, T., TURK, V., TOWATARI, T., KATUNUMA, N. & ET AL. (1991) The refined 2.15 A X-ray crystal structure of human liver cathepsin B: the structural basis for its specificity. EMBO J, 10, 2321-30. NEWLANDS, G. F., SKUCE, P. J., NISBET, A. J., REDMOND, D. L., SMITH, S. K., PETTIT, D. & SMITH, W. D. (2006) Molecular characterization of a family of metalloendopeptidases from the intestinal brush border of Haemonchus contortus. Parasitology, 133, 357-68. NEWTON, S. E. & MUNN, E. A. (1999) The development of vaccines against gastrointestinal nematode parasites, particularly Haemonchus contortus. Parasitol Today, 15, 116-22. OLIVER, E. M., SKUCE, P. J., MCNAIR, C. M. & KNOX, D. P. (2006) Identification and characterization of an asparaginyl proteinase (legumain) from the parasitic nematode, Haemonchus contortus. Parasitology, 133, 237- 44. PANDEY, K. C., WANG, S. X., SIJWALI, P. S., LAU, A. L., MCKERROW, J. H. & ROSENTHAL, P. J. (2005) The Plasmodium falciparum cysteine protease falcipain-2 captures its substrate, hemoglobin, via a unique motif. Proc Natl Acad Sci U S A, 102, 9138-43. PARKINSON, J., MITREVA, M., WHITTON, C., THOMSON, M., DAUB, J., MARTIN, J., SCHMID, R., HALL, N., BARRELL, B., WATERSTON, R.

133 H., MCCARTER, J. P. & BLAXTER, M. L. (2004) A transcriptomic analysis of the phylum Nematoda. Nat Genet, 36, 1259-67. PIACENZA, L., ACOSTA, D., BASMADJIAN, I., DALTON, J. P. & CARMONA, C. (1999) Vaccination with cathepsin L proteinases and with leucine aminopeptidase induces high levels of protection against fascioliasis in sheep. Infect Immun, 67, 1954-61. PRATT, D., ET AL. (1990) A developmentally regulated cysteine protease gene family in Haemonchus contortus. Mol Biochem Parasitol, 43, 181-91. PRATT, D., ET AL. (1992) Cloning and sequence comparison of four distinct cysteine proteases expressed by Haemonchus contortus adult worms. Mol Biochem Parasitol, 51, 209-18. PRITCHARD, D. I. & BROWN, A. (2001) Is Necator americanus approaching a mutualistic symbiotic relationship with humans? Trends Parasitol, 17, 169- 72. PRITCHARD, D. I., QUINNELL, R. J., MOUSTAFA, M., MCKEAN, P. G., SLATER, A. F., RAIKO, A., DALE, D. D. & KEYMER, A. E. (1991) Hookworm (Necator americanus) infection and storage iron depletion. Trans R Soc Trop Med Hyg, 85, 235-8. PROCIV, P. & CROESE, J. (1990) Human eosinophilic enteritis caused by dog hookworm Ancylostoma caninum. Lancet, 335, 1299-302. RANJIT, N., JONES, M. K., STENZEL, D. J., GASSER, R. B. & LOUKAS, A. (2006) A survey of the intestinal transcriptomes of the hookworms, Necator americanus and Ancylostoma caninum, using tissues isolated by laser microdissection microscopy. Int J Parasitol, 36, 701-10. RAWLINGS, N. D., MORTON, F. R., KOK, C. Y., KONG, J. & BARRETT, A. J. (2008) MEROPS: the peptidase database. Nucleic Acids Res, 36, D320-5. REDMOND, D. L. & KNOX, D. P. (2006) Further protection studies using recombinant forms of Haemonchus contortus cysteine proteinases. Parasite Immunol, 28, 213-9. REDMOND, D. L., KNOX, D. P., NEWLANDS, G. & SMITH, W. D. (1997) Molecular cloning and characterisation of a developmentally regulated putative metallopeptidase present in a host protective extract of Haemonchus contortus. Mol Biochem Parasitol, 85, 77-87. REYNOLDSON, J. A., BEHNKE, J. M., PALLANT, L. J., MACNISH, M. G., GILBERT, F., GILES, S., SPARGO, R. J. & THOMPSON, R. C. (1997) Failure of pyrantel in treatment of human hookworm infections (Ancylostoma duodenale) in the Kimberley region of north west Australia. Acta Trop, 68, 301-12. ROADS, M. L. & FETTERER, R. H. (1995) Developmentally regulated secretion of cathepsin L-like cysteine proteases by Haemonchus contortus. J Parasitol, 81, 505-12. ROCHE, M. & LAYRISSE, M. (1966) The nature and causes of "hookworm anemia". Am J Trop Med Hyg, 15, 1029-102. SAJID, M. & MCKERROW, J. H. (2002) Cysteine proteases of parasitic organisms. Mol Biochem Parasitol, 120, 1-21. SAJID, M., MCKERROW, J. H., HANSELL, E., MATHIEU, M. A., LUCAS, K. D., HSIEH, I., GREENBAUM, D., BOGYO, M., SALTER, J. P., LIM, K. C., FRANKLIN, C., KIM, J. H. & CAFFREY, C. R. (2003) Functional expression and characterization of Schistosoma mansoni cathepsin B and its

134 trans-activation by an endogenous asparaginyl endopeptidase. Mol Biochem Parasitol, 131, 65-75. SANCHO, D., GOMEZ, M. & SANCHEZ-MADRID, F. (2005) CD69 is an immunoregulatory molecule induced following activation. Trends Immunol, 26, 136-40. SAWANGJAROEN, N., OPDEBEECK, J. P. & PROCIV, P. (1995) Immunohistochemical localization of excretory/secretory antigens in adult Ancylostoma caninum using monoclonal antibodies and infected human sera. Parasite Immunol, 17, 29-35. SCHAD, G. A., CHOWDHURY, A. B., DEAN, C. G., KOCHAR, V. K., NAWALINSKI, T. A., THOMAS, J. & TONASCIA, J. A. (1973) Arrested development in human hookworm infections: An adaptation to a seasonally unfavorable external environment. Science, 180, 502-504. SCHWENKENBECHER, J. M., ALBONICO, M., BICKLE, Q. & KAPLAN, R. M. (2007) Characterization of beta-tubulin genes in hookworms and investigation of resistance-associated mutations using real-time PCR. Mol Biochem Parasitol, 156, 167-74. SHENAI, B. R., SIJWALI, P. S., SINGH, A. & ROSENTHAL, P. J. (2000) Characterization of native and recombinant falcipain-2, a principal trophozoite cysteine protease and essential hemoglobinase of Plasmodium falciparum. J Biol Chem, 275, 29000-10. SIJWALI, P. S., KOO, J., SINGH, N. & ROSENTHAL, P. J. (2006) Gene disruptions demonstrate independent roles for the four falcipain cysteine proteases of Plasmodium falciparum. Mol Biochem Parasitol, 150, 96-106. SIJWALI, P. S. & ROSENTHAL, P. J. (2004) Gene disruption confirms a critical role for the cysteine protease falcipain-2 in hemoglobin hydrolysis by Plasmodium falciparum. Proc Natl Acad Sci USA, 101, 4384-9. SIJWALI, P. S., SHENAI, B. R., GUT, J., SINGH, A. & ROSENTHAL, P. J. (2001) Expression and characterization of the Plasmodium falciparum haemoglobinase falcipain-3. Biochem J, 360, 481-9. SMITH, T. S., GRAHAM, M., MUNN, E. A., NEWTON, S. E., KNOX, D. P., COADWELL, W. J., MCMICHAEL-PHILLIPS, D., SMITH, H., SMITH, W. D. & OLIVER, J. J. (1997) Cloning and characterization of a microsomal aminopeptidase from the intestine of the nematode Haemonchus contortus. Biochim Biophys Acta, 1338, 295-306. SMITH, T. S., MUNN, E. A., GRAHAM, M., TAVERNOR, A. S. & GREENWOOD, C. A. (1993) Purification and evaluation of the integral membrane protein H11 as a protective antigen against Haemonchus contortus. Int J Parasitol, 23, 271-80. SMITH, W. D., NEWLANDS, G. F., SMITH, S. K., PETTIT, D. & SKUCE, P. J. (2003) Metalloendopeptidases from the intestinal brush border of Haemonchus contortus as protective antigens for sheep. Parasite Immunol, 25, 313-23. SMITH, W. D. & SMITH, S. K. (1993) Evaluation of aspects of the protection afforded to sheep immunised with a gut membrane protein of Haemonchus contortus. Res Vet Sci, 55, 1-9. SMITH, W. D., SMITH, S. K. & MURRAY, J. M. (1994) Protection studies with integral membrane fractions of Haemonchus contortus. Parasite Immunol, 16, 231-41.

135 SMITH, W. D., SMITH, S. K., PETTIT, D., NEWLANDS, G. F. & SKUCE, P. J. (2000) Relative protective properties of three membrane glycoprotein fractions from Haemonchus contortus. Parasite Immunol, 22, 63-71. STACK, C. M., LOWTHER, J., CUNNINGHAM, E., DONNELLY, S., GARDINER, D. L., TRENHOLME, K. R., SKINNER-ADAMS, T. S., TEUSCHER, F., GREMBECKA, J., MUCHA, A., KAFARSKI, P., LUA, L., BELL, A. & DALTON, J. P. (2007) Characterization of the Plasmodium falciparum M17 leucyl aminopeptidase. A protease involved in amino acid regulation with potential for antimalarial drug development. J Biol Chem, 282, 2069-80. SWOFFORD, D. I. (1993) PAUP:Phylogenetic Analysis Using Parsimony, Champaign, IL, Illinois Natural History Society. TANG, J. & WONG, R. N. (1987) Evolution in the structure and function of aspartic proteases. J Cell Biochem, 33, 53-63. TORT, J., BRINDLEY, P. J., KNOX, D., WOLFE, K. H. & DALTON, J. P. (1999) Proteinases and associated genes of parasitic helminths. Adv Parasitol, 43, 161-266. TRAN, M. H., PEARSON, M. S., BETHONY, J. M., SMYTH, D. J., JONES, M. K., DUKE, M., DON, T. A., MCMANUS, D. P., CORREA-OLIVEIRA, R. & LOUKAS, A. (2006) Tetraspanins on the surface of Schistosoma mansoni are protective antigens against schistosomiasis. Nat Med, 12, 835-40. UMEZAWA, H., AOYAGI, T., MORISHIMA, H., MATSUZAKI, M. & HAMADA, M. (1970) Pepstatin, a new pepsin inhibitor produced by Actinomycetes. J Antibiot (Tokyo), 23, 259-62. VETTER, J. C. & LEEGWATER-VD LINDEN, M. E. (1977) Skin penetration of infective hookworm larvae. III. Comparative studies on the path of migration of the hookworms Ancylostoma braziliense, Ancylostoma ceylanicum, and Ancylostoma caninum. Z Parasitenkd, 53, 155-8. VINEY, M. E. & THOMPSON, F. J. (2008) Two hypotheses to explain why RNA interference does not work in animal parasitic nematodes. Int J Parasitol, 38, 43-7. WILLIAMSON, A. L., BRINDLEY, P. J., ABBENANTE, G., DATU, B. J., PROCIV, P., BERRY, C., GIRDWOOD, K., PRITCHARD, D. I., FAIRLIE, D. P., HOTEZ, P. J., ZHAN, B. & LOUKAS, A. (2003a) Hookworm aspartic protease, Na-APR-2, cleaves human hemoglobin and serum proteins in a host-specific fashion. J Infect Dis, 187, 484-94. WILLIAMSON, A. L., BRINDLEY, P. J., ABBENANTE, G., PROCIV, P., BERRY, C., GIRDWOOD, K., PRITCHARD, D. I., FAIRLIE, D. P., HOTEZ, P. J., DALTON, J. P. & LOUKAS, A. (2002) Cleavage of hemoglobin by hookworm cathepsin D aspartic proteases and its potential contribution to host specificity. Faseb J, 16, 1458-60. WILLIAMSON, A. L., BRINDLEY, P. J., KNOX, D. P., HOTEZ, P. J. & LOUKAS, A. (2003b) Digestive proteases of blood-feeding nematodes. Trends Parasitol, 19, 417-23. WILLIAMSON, A. L., LECCHI, P., TURK, B. E., CHOE, Y., HOTEZ, P. J., MCKERROW, J. H., CANTLEY, L. C., SAJID, M., CRAIK, C. S. & LOUKAS, A. (2004) A multi-enzyme cascade of hemoglobin proteolysis in the intestine of blood-feeding hookworms. J Biol Chem, 279, 35950-7. WILLIAMSON, A. L., LUSTIGMAN, S., OKSOV, Y., DEUMIC, V., PLIESKATT, J., MENDEZ, S., ZHAN, B., BOTTAZZI, M. E., HOTEZ, P. J. & LOUKAS,

136 A. (2006) Ancylostoma caninum MTP-1, an astacin-like metalloprotease secreted by infective hookworm larvae, is involved in tissue migration. Infect Immun, 74, 961-7. XIAO, S., ZHAN, B., XUE, J., GOUD, G. N., LOUKAS, A., LIU, Y., WILLIAMSON, A., LIU, S., DEUMIC, V. & HOTEZ, P. (2008) The evaluation of recombinant hookworm antigens as vaccines in hamsters (Mesocricetus auratus) challenged with human hookworm, Necator americanus. Exp Parasitol, 118, 32-40. YU, S. H., JIANG, Z. X. & XU, L. Q. (1995) Infantile hookworm disease in China. A review. Acta Trop, 59, 265-70. ZHAN, B., HAWDON, J., QIANG, S., HAINAN, R., HUIQING, Q., WEI, H., SHU- HUA, X., TIEHUA, L., XING, G., ZHENG, F. & HOTEZ, P. (1999) Ancylostoma secreted protein 1 (ASP-1) homologues in human hookworms. Mol Biochem Parasitol, 98, 143-9. ZHAN, B., HOTEZ, P. J., WANG, Y. & HAWDON, J. M. (2002) A developmentally regulated metalloprotease secreted by host-stimulated Ancylostoma caninum third-stage infective larvae is a member of the astacin family of proteases. Mol Biochem Parasitol, 120, 291-296. ZHAN, B., LIU, Y. Y., BADAMCHIAN, M., WILLIAMSON, A., FENG, J. J., LOUKAS, A., HAWDON, J. M. & HOTEZ, P. J. (2003) Molecular characterisation of the Ancylostoma-secreted protein family from the adult stage of Ancylostoma caninum. Int J Parasitol, 33, 897-907. ZHAN, B., WANG, Y., LIU, Y., WILLIAMSON, A., LOUKAS, A., HAWDON, J. M., XUE, H. C., XIAO, S. H. & HOTEZ, P. J. (2004) Ac-SAA-1, an immunodominant 16 kDa surface-associated antigen of infective larvae and adults of Ancylostoma caninum. Int J Parasitol, 34, 1037-45.

137 APPENDICES

138

APPENDIX 1 A survey of the intestinal transcriptome of the hookworms, Necator americanus and Ancylostoma caninum using tissue isolated by laser microdissection microscopy. International Journal for Parasitology 36: 701-710 N. Ranjit, M.K. Jones, D.J Stenzel, R.B Gasser, A. Loukas (2006).

APPENDIX 2 A family of cathepsin B cysteine proteases expressed in the gut of the human hookworm, Necator americanus. Molecular and Biochemical Parasitology 160: 90-9 N. Ranjit, B. Zhan, D. Stenzel, J. Mulvenna, R. Fujiwara, P. Hotez, A. Loukas (2008).

APPENDIX 3

Vaccination with recombinant aspartic hemoglobinase reduces parasite load and blood loss after hookworm infection in dogs. PLoS Medicine 2 (10): e295 A. Loukas, J. M.Bethony, S. Mendez, R. T. Fujiwara, G. N. Goud, N. Ranjit, B. Zhan, K. Jones, M. E. Bottazzi, P. J. Hotez (2005). Open access, freely available online PLoS MEDICINE Vaccination with Recombinant Aspartic Hemoglobinase Reduces Parasite Load and Blood Loss after Hookworm Infection in Dogs

Alex Loukas1*, Jeffrey M. Bethony2, Susana Mendez2, Ricardo T. Fujiwara2, Gaddam Narsa Goud2, Najju Ranjit1, Bin Zhan2, Karen Jones2, Maria Elena Bottazzi2, Peter J. Hotez2* 1 Division of Infectious Diseases and Immunology, Queensland Institute of Medical Research, Brisbane, Queensland, Australia, 2 Department of Microbiology and Tropical Medicine, The George Washington University Medical Center, Washington, District of Columbia, United States of America

Competing Interests: The authors have declared that no competing ABSTRACT interests exist. Background Author Contributions: AL, JMB, MEB, and PJH designed the study. Hookworms infect 730 million people in developing countries where they are a leading cause AL, RTF, GNG, NR, BZ, performed of intestinal blood loss and iron-deficiency anemia. At the site of attachment to the host, adult experiments. AL, PJH, JMB, SM, and hookworms ingest blood and lyse the erythrocytes to release hemoglobin. The parasites KJ analyzed the data. AL, PJH, JMB, and SM contributed to writing the subsequently digest hemoglobin in their intestines using a cascade of proteolysis that begins paper. with the Ancylostoma caninum aspartic protease 1, APR-1. Academic Editor: Maria Yazdanbakhsh, Leiden University Methods and Findings Medical Center, the Netherlands

Citation: Loukas A, Bethony JM, We show that vaccination of dogs with recombinant Ac-APR-1 induced antibody and cellular Mendez S, Fujiwara RT, Goud GN, et responses and resulted in significantly reduced hookworm burdens (p ¼ 0.056) and fecal egg al. (2005) Vaccination with counts (p ¼ 0.018) in vaccinated dogs compared to control dogs after challenge with infective recombinant aspartic hemoglobinase reduces parasite larvae of A. caninum. Most importantly, vaccinated dogs were protected against blood loss (p ¼ load and blood loss after hookworm 0.049) and most did not develop anemia, the major pathologic sequela of hookworm disease. infection. PLoS Med 2(10): e295. IgG from vaccinated animals decreased the catalytic activity of the recombinant enzyme in vitro Received: April 8, 2005 and the antibody bound in situ to the intestines of worms recovered from vaccinated dogs, Accepted: July 13, 2005 implying that the vaccine interferes with the parasite’s ability to digest blood. Published: October 4, 2005 DOI: Conclusion 10.1371/journal.pmed.0020295 To the best of our knowledge, this is the first report of a recombinant vaccine from a Copyright: Ó 2005 Loukas et al. This is an open-access article distributed hematophagous parasite that significantly reduces both parasite load and blood loss, and it under the terms of the Creative supports the development of APR-1 as a human hookworm vaccine. Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Abbreviations: APR-1, Ancylostoma caninum aspartic protease 1; AS03, GlaxoSmithKline Adjuvant System 01; ELISA, enzyme-linked immunosorbent assay; epg, eggs per gram of feces; Hb, hemoglobin; L3, third stage larvae

*To whom correspondence should be addressed. E-mail: alex.loukas@ qimr.edu.au (AL); mtmpjh@gwumc. edu (PJH)

PLoS Medicine | www.plosmedicine.org1008 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase

Introduction and particularly fetal, Hb demands are considerable, render- ing these populations most vulnerable to the parasite [1]. Hookworms infect more than 700 million people in Here we describe vaccination of dogs with the aspartic tropical and subtropical regions of the world. The major hemoglobinase of A. caninum, Ac-APR-1 [21,23] and show that species infecting humans are Necator americanus and Ancylos- vaccination resulted in the production of neutralizing anti- toma duodenale. The parasites feed on blood, causing iron- bodies, significantly reduced egg counts, and significantly deficiency anemia, and as such, are a major cause of disease reduced adult worm burdens. Most importantly, Hb levels of burden in developing countries [1]. Unlike other human vaccinated dogs were significantly higher than those of dogs helminthiases, worm burdens do not generally decrease with that were vaccinated with adjuvant alone after parasite age; in fact, recent findings revealed that the heaviest worm challenge. These data show that aspartic hemoglobinases, burdens are found among the elderly [2,3]. Whereas particularly APR-1, are efficacious vaccines against canine anthelminthic chemotherapy with benzimidazole drugs is hookworm disease, providing strong support for further effective in eliminating existing adult parasites, re-infection investigation and development of APR-1 as a recombinant occurs rapidly after treatment [4], making a vaccine against vaccine against human hookworm disease. hookworm disease a desirable goal. Canines can be successfully vaccinated against infection Methods with the dog hookworm, Ancylostoma caninum, by immunization with third-stage infective larvae (L3) that have been attenu- Expression of Recombinant Ac-APR-1 in Pichia pastoris ated with ionizing radiation [5–7]. Subsequently, varying levels The entire open reading frame of Ac-APR-1 encoding the of vaccine efficacy have been reported for the major antigens zymogen (spanning Ser-17 to the C-terminal Phe-446) but secreted by hookworm L3 using hamsters [8,9] and dogs [10]. excluding the predicted signal peptide was cloned into the Despite obtaining encouraging levels of protection with larval expression vector pPIC-Za (Invitrogen, Carlsbad, California, antigens, only partial reductions in parasite load (fecal egg United States) using the XbaI and EcoRI sites. Yeast, P. pastoris counts and adult worm burdens) were reported. Moreover, X 33, was transformed with the vector encoding the Ac-APR-1 protective antigens from the larval stage are only expressed by zymogen as recommended by the manufacturer (Invitrogen) L3, and not adult worms, rendering antibodies against these with modifications. Protein disulfide isomerase (PDI) gene in L3 secretions useless against parasites that have successfully the vector pPIC3.5 (a gift from Mehmet Inan, University of reached adulthood in the gut and begun to feed on blood. We Nebraska, Lincoln, Nebraska, United States) was cut with SacI and transformed into P. pastoris X 33 cells which were already therefore suggest that an ideal hookworm vaccine would transformed with Ac-apr-1 following the manufacturer’s require a cocktail of two recombinant proteins, one targeting instructions. Eight transformed colonies were picked from the infective larva and the second targeting the blood-feeding YPD plates containing Geneticin (0.5–1.0 mgml1)and adult stage of the parasite [11]. Zeocin (1.0 mgml1) and tested for Ac-APR-1 expression Of the different families of proteins expressed by blood- following the manufacturer’s instructions. The highest feeding parasitic helminths, proteolytic enzymes have shown expressing colony was selected and transferred to suspension promise as intervention targets for vaccine development culture in flasks containing BMG medium (buffered minimal [12,13]. Proteases are pivotal for a parasitic existence, glycerol: 1.34% yeast nitrogen base, 0.00004% d-biotin, 1% w/ mediating fundamental physiologic processes such as molting, v glycerol, and 100 mM potassium phosphate, [pH 6.0]). tissue invasion, feeding, embryogenesis, and evasion of host Suspension cultures were then transferred to a Bioflo 3000 immune responses [12,14]. Parasite extracts enriched for fermentor (New Brunswick Scientific, Edison, New Jersey, proteases protect sheep against the blood-feeding nematodes United States) utilizing a 5-l vessel as described [8]. The Haemonchus contortus [15–18] and Ostertagia ostertagi [19]; how- recombinant protein was secreted into culture medium and ever, significant protective efficacy has not been shown with a affinity purified on nickel-agarose as described elsewhere [8]. purified recombinant protease from nematodes of livestock. Progress of purification was monitored using SDS-PAGE gels Hookworms feed by burying their anterior ends in the stained with Coomassie Brilliant Blue and immunoblots using intestinal mucosa of the host, rupturing capillaries and monoclonal antibodies to the vector-derived myc . ingesting the liberated blood. Erythrocytes are lysed by pore Recombinant Ac-APR-1 was treated with PNGase F and O- formation [20], releasing hemoglobin (Hb) into the lumen of glycosidase, according to the manufacturer’s instructions the parasite’s intestine, where it is degraded by a semi- (Enzymatic CarboRelease kit; QA-Bio, San Mateo, California, ordered pathway of catalysis that involves aspartic, cysteine, United States), under denaturing conditions to remove any and metalloproteases [21]. Vaccination of dogs with a N-linked and O-linked oligosaccharides. Deglycosylation was catalytically active recombinant cysteine hemoglobinase, Ac- performed only to confirm the presence of N-linked sugars CP-2, induced antibodies that neutralized proteolytic activity on the recombinant molecule. All remaining studies were and provided partial protection to vaccinees by reducing egg conducted with the glycoprotein. output (a measure of intestinal worm burden) and worm size, but significant reductions of adult worm burdens and/or Activation and Hemoglobinolytic Activity of Recombinant blood loss were not observed [22]. Anemia is the primary APR-1 pathology associated with hookworm infection, and an The unactivated zymogen was used for vaccination. A small ultimate human hookworm vaccine would limit the amount amount of the purified protein, however, was buffer of blood loss caused by feeding worms and maintain normal exchanged into 100 mM sodium formate (pH 3.6)/0.15 M levels of Hb. This is particularly important in young children NaCl using a PD10 desalting column (Amersham Biosciences, as well as women of child-bearing age, in whom menstrual, Little Chalfont, United Kingdom) to facilitate proteolytic

PLoS Medicine | www.plosmedicine.org1009 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase

activation and removal of the pro-region. One microgram of humidified 5% CO2 atmosphere at 37 8C for 2 d (ConA- purified, activated protease was then added to 10 lg of dog stimulated cultures) and 5 d (APR-1). Cells were pulsed for 6 h Hb in the same buffer and incubated at 37 8C for 2 h. with 1.0 lCi of [3H] thymidine (PerkinElmer Life And Cleavage of Hb was assessed visually by staining SDS-PAGE Analytical Sciences, Boston, Massachusetts, United States) gels with Coomassie Brilliant Blue. and harvested onto glass fiber filters. Radioactive incorpo- ration was determined by liquid scintillation spectrometry. Animal Husbandry Proliferation responses were expressed as stimulation indices, 6 Purpose-bred, parasite naive, male beagles aged 8 1wk SI (where SI ¼ mean proliferation of stimulated cultures/mean were purchased from Marshall Farms (North Rose, New York, proliferation of unstimulated cultures). For cytokine analyses, United States), identified by ear tattoo, and maintained in the whole blood (collected as described above) was diluted 1:8 in George Washington University Animal Research Facility as RPMI supplemented with 3% antibiotic/antimycotic solution previously described [24]. The experiments were conducted in a 48-well flat-bottomed culture plate with a final volume of according to a protocol approved by the University Animal 1.0 ml per well. Cells were stimulated by the addition of 25 Care and Use Committee (IACUC 48–12,0 [12,1]E). Before the lgml1 of recombinant APR-1. After 48 h of incubation at 37 first vaccination and after each subsequent one, a blood 8C, 700 ll of supernatant was removed from each well and sample was obtained from each dog. stored at 20 8C until required for the cytokine assay. IL-4, IL- Vaccine Study Design and Antigen-Adjuvant Formulation 10, and IFN-c were measured using a capture ELISA assay for The vaccine trial was designed to test Ac-APR-1 zymogen dogs (R & D Systems, Minneapolis, Minnesota, United States) formulated with the adjuvant AS03 [25], obtained from following the manufacturer’s instructions. Biotin-labeled 1 GlaxoSmithKline (a kind gift from Drs. Joe Cohen and Sylvie detection antibodies were used (100 ngml ), revealed with Cayphas; GSK Biologicals, Rixensart, Belgium). To make six streptavidin-HRP (Amersham Biosciences), and plates were doses of Ac-APR-1 formulated with AS03, 600 lgof developed with OPD (O-Phenylenediamine) substrate system recombinant protein (1.5 ml of Ac-APR-1 at a concentration (Sigma-Aldrich). of 0.4 mgml1) was mixed with 1.2 ml of 20 mM Tris-HCl, 0.5 Hb Measurements M NaCl (pH 7.9), and 1.5 ml of AS03; the contents of the tube were vortex mixed for 30 sec then shaken at low speed for 10 To determine Hb concentrations of experimental dogs, 1–2 min. Dogs were immunized with 100 lg of formulated antigen ml of blood were collected in EDTA and analyzed using a in a final volume of 0.5 ml. AS03-only control was prepared as QBC VetAutoread Hematology System and VETTEST Soft- described above, with PBS included instead of Ac-APR-1. ware (IDEXX Laboratories, Westbrook, Maine, United States). Canine Immunizations and Antibody Measurements Hookworm Infections and Parasite Recovery Five beagles were immunized three times with AS03- Two weeks after the final immunization, dogs were formulated Ac-APR-1 by intramuscular injection. The vaccine anaesthetized using a combination of ketamine and xylazine 1 1 was administered on days 0, 21, and 42, beginning when the (20 mgkg and 10 mgkg respectively) and infected via the dogs were 62 6 4 d of age. As negative controls, five beagles footpad with 500 A. caninum L3 as described elsewhere [22]. were also injected intramuscularly with an equivalent amount Quantitative hookworm egg counts (McMaster technique) of AS03 using the identical schedule. Blood was drawn at least were obtained for each dog 3 d per wk from days 12–26 once every 21 d and serum was separated from cells by postinfection. Four weeks postinfection, the dogs were killed centrifugation. Enzyme-linked immunosorbent assays (ELI- by intravenous injection of barbiturate, and adult hookworms SA) were performed as previously described [24]. Recombi- were recovered and counted from the small and large nant Ac-APR-1 was coated onto microtiter plates at a intestines at necropsy [24]. The sex of each adult worm was concentration of 5.0 lgml1. Dog sera were titrated between determined as described elsewhere [8]. Approximately 1–2 cm 1:100 and 1:2 3 106 to determine endpoint titers (the highest lengths of small intestine were removed and stored in dilution of test group [APR-1] sera that gave a mean O.D. of formalin for future histopathologic analysis. 33 the mean optical density (OD) of sera from the control Statistical Methods group). Anti-canine IgG1, IgG2, and IgE antibodies conju- gated to horseradish peroxidase (Bethyl Laboratories, Mont- In most cases, the small size of the samples did not enable gomery, Texas, United States) were used at a dilution of us to determine if values were normally distributed, so the 1:1,000. Blood was collected from dogs before immunizations following non-parametric tests were used: Mann-Whitney U and 7 d after the third vaccination but before L3 challenge. was used to test whether two independent samples (groups) came from the same population, and the Kruskal Wallis H test Stimulation of and Cytokine Measurements from Cultured was used to determine if several independent samples came Whole Blood from the same population. Normally distributed variables Lymphoproliferation assays were performed using a whole were tested in the following manner: The independent- blood microassay as previously described [26]. Briefly, 25 llof samples t-test procedure was used to compare the means for heparinized blood was diluted in 200 ll of RPMI 1640 two groups, and an analysis of variance was used to test the medium (Gibco, Invitrogen) supplemented with 3% anti- hypothesis that several means are equal, followed by a Dunnet biotic/antimycotic solution (Gibco). All tests were performed post hoc multiple comparison t-test to compare the vaccine in triplicate in 96-well flat-bottomed culture plates using treatment groups against the control group. Differences were recombinant APR-1 at a concentration of 25 lgml1 and considered statistically significant if the calculated p-value concanavalin A (ConA; Sigma-Aldrich, St. Louis, Missouri, was equal to or less than 0.1 (two-sided). The percentage United States) at 80 lgml1. Incubation was carried out in a reduction or increase in adult hookworm burden in the

PLoS Medicine | www.plosmedicine.org1010 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase

Figure 2. The Geometric Mean Titers of the IgG1 and IgG2 Antibody Responses of Dogs Vaccinated with Recombinant Ac-APR-1 Formulated with AS03 or AS03 Alone LC, day on which dogs were challenged with hookworm L3; N, day of necropsy; V1, V2, and V3, days on which animals were vaccinated. DOI: 10.1371/journal.pmed.0020295.g002 Effect of Anti–Ac-APR-1 IgG on Proteolytic Activity Canine IgG was purified from sera of vaccinated dogs using protein A-agarose (Amersham Biosciences) as previously described [23]. Purified IgG (0.2 lg) was incubated with 1.0 lg of recombinant Ac-APR-1 for 45 mins prior to assessing catalytic activity of APR-1 against the fluorogenic substrate o- aminobenzoyl-IEF-nFRL-NH2 as described previously [23]. The aspartic protease inhibitor, pepstatin A, was included at a final concentration of 1.0 lM as a positive control for enzymatic inhibition. Data was recorded from triplicate experiments and presented as relative fluorescence units using a TD700 fluorometer (Turner Designs, Sunnyvale, Figure 1. P. pastoris Secrete Ac-APR-1 Zymogen that Autoactivates at California, United States). Low pH and Degrades Canine Hb SDS-PAGE gel stained with Coomassie Brilliant Blue showing purification Results of recombinant APR-1 zymogen from P. pastoris culture supernatant. (A) Lane 1, molecular weight markers; lane 2, concentrated culture Secretion of Catalytically Active Ac-APR-1 by P. pastoris supernatant; lane 3, flow-through from a nickel-IDA column; lane 4, 5 Yeast secreted the APR-1 zymogen into culture medium at mM imidazole wash; lane 5, 20 mM imidazole column eluate; lane 6, 60 1 mM imidazole eluate; and lane 7, 1 M imidazole eluate. Purified an approximate concentration of 1.0 mgl (Figure 1A). In the recombinant APR-1 zymogen was activated by buffer exchange into 0.1 absence of co-expression with the PDI chaperone, the amount M sodium formate/0.1 M NaCl (pH 3.6). of APR-1 secreted by P. pastoris was approximately half that (B) Lane 1, molecular weight markers; lane 2, 5.0 lg of canine Hb (pH obtained here (not shown). Ac-APR-1 has one potential 3.6); and lane 3, 5.0 lg of canine Hb (pH 3.6) incubated with 0.2 lgof recombinant APR-1. glycosylation site at Asn-29 of the zymogen (after removal of DOI: 10.1371/journal.pmed.0020295.g001 the signal peptide), and treatment with PNGase F decreased the size of the recombinant protein by the expected size (2–3 kDa; not shown). The activated recombinant protease readily vaccinated group was expressed relative to the control group digested canine Hb at acidic pH (Figure 1B), confirming that as described elsewhere [24]. Ac-APR-1 expressed in yeast is catalytically active and digested Hb with similar efficiency to recombinant Ac-APR-1 produced Immunohistochemistry in baculovirus (data not shown). Adult hookworms were recovered at necropsy from vacci- nated dogs and control dogs, washed briefly, then fixed and Recombinant Ac-APR-1 Is Immunogenic in Dogs sectioned as previously described [22]. To observe whether IgG AS03 was used as an adjuvant based on its ability to induce a from vaccinated but not control dogs, bound to APR-1 lining higher IgG1 response and greater reduction in hookworm egg the intestinal microvillar surface of worms in situ, sections counts when used to vaccinate dogs in a head-to-head were probed with Cy3-conjugated rabbit anti-dog IgG (Jackson comparison of a cysteine hemoglobinase formulated with Immunoresearch, West Grove, Pennsylvania, United States) at four different adjuvants [22]. Dogs immunized with recombi- a dilution of 1:500 as described elsewhere [27]. Sections were nant Ac-APR-1 formulated with AS03 produced IgG1 and IgG2 visualized using a Leica IM 100 inverted fluorescence micro- antibody responses as measured by ELISA using the recombi- scope (Leica Microsystems, Wetzlar, Germany). nant protein (Figure 2). IgE titers were low (,1:1,500) and

PLoS Medicine | www.plosmedicine.org1011 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase

Figure 3. Canine Cellular Immune Response to Vaccination with Recombinant Ac-APR-1 Cell proliferation of whole blood cells from vaccinated (APR-1) and control dogs (AS03) when stimulated with concanavalin A (A) or recombinant Ac- APR-1 (B) before (day 0) and after the final immunization (day 51). The p-value comparing the mean differences between the vaccinated group and controls is denoted. Detection of secreted IFN-c in whole blood cultures taken from vaccinated and control dogs before and after immunization (C). Mean cytokine concentrations are indicated in pgml1 with standard error bars. Statistically significant differences are indicated above the bars by p- values. APR, stimulated with recombinant APR-1; NS, non-stimulated cultures. DOI: 10.1371/journal.pmed.0020295.g003 were not sustained past challenge. We did not adsorb IgG from IL-4 or IL-10 after stimulation with APR-1 in either serum before measuring IgE in this study; however, in previous vaccinated or control groups (not shown). trials IgG was removed and we did not see a difference in antigen-specific IgE titers. For vaccinated dogs, maximum Vaccination with Ac-APR-1 Decreases Fecundity of Female IgG2 titers of 1:121,500 were attained by all five dogs after the Hookworms second vaccination. High titers persisted through challenge Dogs develop age- and exposure-related immunity to A. and decreased to 1:26,098 by necropsy. IgG1 titers peaked at caninum [5], so we therefore observed egg counts from 1:13,500 after the third vaccination in all four dogs and vaccinated animals up to 26 d postchallenge, after which we dropped to 1:3,600 by necropsy. Dogs immunized with often observe a significant decrease in egg counts in some adjuvant alone did not generate detectable immune responses dogs. Because of daily variation in egg counts from infected greater than 1:500, even after larval challenge. dogs (A. Loukas. S. Mendez, and P. Hotez, unpublished data), Dogs rapidly acquire resistance to hookworm with matur- we analyzed the data in two ways. Firstly, the median egg ity. A single dog was therefore removed from the control counts for days 21, 23, and 26 postinfection were used to group (for all analyses) because its weight was greater than the compare worm fecundity between vaccinated and control acceptable range at all time points after the first vaccination groups. A 70% decrease in median egg counts was observed (mean plus or minus three standard errors). in dogs vaccinated with Ac-APR-1 (2,650 eggs per gram of feces [epg]) compared with dogs that were vaccinated with Vaccination Induces Antigen-Specific Cell Proliferation adjuvant alone (8,725 epg) when median egg counts were and Cytokine Production calculated for the three time points measured after larval Vaccination with APR-1 induced a high level of lympho- challenge (Figure 4A). We then compared geometric mean cyte/leukocyte proliferation compared with control dogs values of egg counts between the two groups (Figure 4B), and when cells were stimulated with APR-1 (p , 0.01, t-test). Cells showed that mean egg counts of the vaccinated animals from both vaccinated and control dogs proliferated equally remained lower than the control animals as worms became when stimulated with mitogen (Figure 3A and 3B). No fecund by day 21, implying that fecundity of female worms significant proliferation to APR-1 was observed before the diminished significantly as they began to feed on blood immunization process. Immunization with APR-1 elicited containing anti–APR-1 antibodies. By day 26 postchallenge, antigen-specific production of IFN-c (p ¼ 0.03, t-test) (Figure there was an 85% reduction in mean egg counts between the 3C). In contrast, we did not detect significant production of two groups. For statistical analyses, we transformed egg

PLoS Medicine | www.plosmedicine.org1012 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase

using a Mann-Whitney test (APR-1 versus control), revealed a statistically significant difference (p ¼ 0.018). Vaccination with Ac-APR-1 Significantly Reduces Adult Hookworm Burdens A statistically significant difference at the p 0.1 level (p ¼ 0.095; Mann-Whitney U test) was detected for a one-sided test between median adult worm burdens recovered from vaccinated dogs (182) compared with control dogs (270) but not for a two-sided test (p ¼ 0.190) (Figure 5). Percentage reduction of the median worm counts was 33% when data from both sexes of worms were combined, 30% for male worms (p ¼ 0.111 [2-sided] or p ¼ 0.056 [1-sided]) and 40% for female worms (p ¼ 0.1905 [2-sided] or p ¼ 0.0952 [1-sided]), again supporting the enhanced effect of the vaccine on female worms given their increased nutritional requirements for egg production. Vaccination with APR-1 Protects against Anemia Hb levels in four of the five dogs that were vaccinated with APR-1 were significantly elevated when compared with control dogs (adjuvant alone) after challenge infection (Figure 6). The median Hb concentration of vaccinated dogs for the last two time points (0 and 7 d prior to necropsy) was 12.45 gdl1 compared with 9.5 gdl1 for the control dogs that were immunized with adjuvant alone (p ¼ 0.049; Mann- Whitney U test). A decline in Hb levels was seen in all of the control dogs after challenge infection; the decline was marked in three of the four dogs. Four of the five dogs that were vaccinated with APR-1 did not show a similar decline, and had Hb levels within (or very close to) the normal clinical Figure 4. Vaccination with APR-1 Reduces Fecal Egg Counts of Dogs 1 after Challenge Infection with Hookworms range of 12–14 gdl . One dog (C5) from the vaccinated 1 Statistically significant reduction (p ¼ 0.018) in median fecal egg counts group did become anemic (Hb concentration was 9.6 gdl ), sampled on days 21, 23, and 26 of dogs vaccinated with APR-1 compared and this animal had more female worms (120 compared with to dogs that received adjuvant alone. a mean of 88 female worms for the group) and more male (A). Geometric mean values of fecal egg counts from vaccinated and control dogs between challenge infection and necropsy. worms (87 compared with a mean of 80 male worms for the (B). Error bars refer to the standard error of the mean. group). However, using both Spearman and Pearson tests, we DOI: 10.1371/journal.pmed.0020295.g004 did not detect a significant correlation between worm burdens (for either or both sexes) and Hb status of the counts into log values and ran the test in two ways: (1) vaccinated dogs. comparing the log transformed epgs in the last three egg counts by analysis of variance (Kruskall-Wallis) revealed no Anti–APR-1 Antibodies Are Ingested by and Bind to the significant differences among the groups for the last three egg Intestine of Feeding Hookworms counts when each time point was considered individually; and The site of anatomical expression of Ac-APR-1 within adult (2) comparing pooled data from the last three egg counts hookworms has been previously reported by us to be the

Figure 5. Vaccination with APR-1 Reduces Adult Worm Burdens of Dogs after Challenge Infection with Hookworms Statistically significant reduction at the p , 0.1 level (p ¼ 0.065) in median adult worm (both sexes) burdens of dogs vaccinated with APR-1 compared to dogs that received adjuvant alone (A). Reductions are also shown when only male (B) (p ¼ 0.111) and only female (C) (p ¼ 0.1905) worms were considered; however, statistically significant reductions were not achieved for single sex analyses. Bars represent the median value for each group. DOI: 10.1371/journal.pmed.0020295.g005

PLoS Medicine | www.plosmedicine.org1013 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase

APR-1 antibodies were ingested with the blood-meal of the worm and subsequently bound specifically to the intestine of the parasite in situ. IgG from Dogs Vaccinated with Ac-APR-1 Neutralizes Proteolytic Activity In Vitro Purified IgG from dogs that were immunized with Ac-APR-1 reduced the catalytic activity of the enzyme by 71%, compared with just 6% reduction when an equivalent amount of IgG from dogs immunized with adjuvant alone Figure 6. Vaccination of Dogs with APR-1 Reduces Blood Loss and was assessed (Table 1). The aspartic protease inhibitor, Protects against Anemia pepstatin A, inhibits catalytic activity of APR-1 [23] and was Hb concentrations of vaccinated dogs were significantly (p ¼ 0.049) therefore used as a positive control to obtain 100% inhibition greater than those of control dogs when blood was drawn after larval for comparative purposes. challenge (0 and 7 d before necropsy [post]) but not when blood was drawn 5 d before larval challenge (pre). DOI: 10.1371/journal.pmed.0020295.g006 Discussion Here we describe protective vaccination of dogs with a microvillar surface of the gut [21,23]. To determine whether recombinant aspartic hemoglobinase, a pivotal enzyme in the vaccination of dogs induced circulating antibodies that initiation of Hb digestion in the gut of canine hookworms bound to the intestinal lumen during infection, parasites [12,21]. We show that APR-1 provides the best efficacy thus were removed from vaccinated dogs, fixed, sectioned, and far reported for a recombinant vaccine aimed at reducing probed with anti-dog IgG conjugated to Cy3. Worms hookworm egg counts, intestinal worm burdens, and hook- recovered from dogs immunized with Ac-APR-1 but not from worm-induced blood loss. dogs immunized with adjuvant alone reacted with Cy3- The vaccine efficacy of recombinant Ac-APR-1 expressed in conjugated anti-dog IgG (Figure 7), indicating that anti– baculovirus-infected insect cells was described earlier by us

Figure 7. Antibodies Bind In Situ to the Intestines of Hookworms that Feed on Vaccinated Dogs Detection of antibodies that bound to the gut of worms recovered from vaccinated dogs (A and B) but not control dogs (C and D) by immunofluorescence. Binding was detected using Cy3-conjugated rabbit anti-dog IgG, allowing only detection of antibodies that had bound in situ while parasites were feeding on blood from vaccinated or control dogs. ic. intestinal contents; in, hookworm intestine; mv, intestinal microvillar surface; ro, reproductive organs. DOI: 10.1371/journal.pmed.0020295.g007

PLoS Medicine | www.plosmedicine.org1014 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase

The immunological parameters required for vaccine- Table 1. Reduction in Cleavage of the Fluorogenic Substrate o- induced protection against hookworm infection were, until Aminobenzoyl-IEF-nFRL-NH2 When 1.0 lg of Recombinant Ac- recently, poorly defined. Protection against A. caninum by APR-1 Was Pre-Incubated with 0.2 lg of IgG Purified from Sera of vaccination of dogs with radiation-attenuated L3 was Dogs Vaccinated with APR-1/AS03 or AS03 Alone (Control) reported many years ago [5]; however, it was not until recently that murine [8,33] and canine [34] studies revealed Protease and Corrected Relative Mean Percent the protective mechanisms of the irradiated larval vaccine at Treatment Fluorescence Units Reduction in Cleavage a cellular level. These studies suggested that a T-helper type-2 of IEF-nFRL-NH2 response is induced by vaccination with irradiated L3; however the authors did not prove that a T-helper type-1 APR-1 þ buffer 362 6 13 0 response abrogates protection. In our study reported here, APR-1 þ a-APR-1 IgG 104 6 24 71 APR-1 þ control IgG 340 6 41 6 dogs vaccinated with APR-1 generated strong memory APR-1 þ pepstatin 0 100 responses to the recombinant antigen and did not secrete Th-2 cytokines but instead secreted IFN-c in response to

Percent reductions caused by incubation of APR-1 with IgGs were determined using 1.0 lM pepstatin as positive stimulation with recombinant APR-1. Moreover, the domi- (100% reduction) control. Baseline was set at zero using the relative fluorescence of the positive control. nant antibody isotype induced by vaccination was IgG2, DOI: 10.1371/journal.pmed.0020295.t001 suggesting that a Th-1-like response was generated. Unlike the clear association between IgG2 and type I cytokines such [24]; however, this initial vaccine trial was hampered by as IFN-c in mice and humans, little is known about this limited availability of the recombinant protein: Suboptimal association in dogs. Experimental evidence using the canine doses were used and antibody responses (titers ,10,000) were model suggests that immune responses (Th1 versus Th2) are, first observed just 1 wk following the third (and final) however, linked to isotype production. For example, animals immunization, and only in some dogs. Despite the weak infected with and protected against visceral leishmaniasis antibody responses, a statistically significant reduction in (Th1 response) or Salmonella (also a Th1 response) mount a mean (18%, p , 0.05) and median (23%) hookworm burdens higher IgG2 than IgG1 response [35,36]. Our data [34] show were observed. In addition there was a shift of adult thatdogsimmunizedwithirradiatedhookwormlarvae hookworms from the small intestine to the colon [24]. demonstrated a stronger production of IgG1 (also supported However, no reduction in the mean fecal egg counts were by [37]) which accompanied IL-4 production, implying a Th2 observed, and hematologic parameters were not assessed. The cytokine response in dogs is accompanied by the same improved immunogenicity of APR-1 observed in this study immunoglobulin isotypes seen in humans and mice. Based on might also be attributed to use of the adjuvant AS03 the current data, we cannot conclude that a Th-1 response to compared with alhydrogel in the previous study. We have APR-1 is required to obtain protection; however, it does not shown in a head-to-head comparison of a hookworm cysteine inhibit the development of a protective memory response. It hemoglobinase formulated with different adjuvants (includ- should also be considered that successful immunity to the different developmental stages of hookworms might require ing alhydrogel) that AS03-formulated protein generated very different immune response phenotypes, not unlike those higher antibody titers and afforded greater protection to seen in schistosomiasis [38]. Further studies will explore the vaccinated dogs [22]. In this study, we show that yeast-derived effects of vaccination with APR-1 formulated with different APR-1 provides the best efficacy thus far reported for a adjuvants and co-factors (e.g., cytokines) that will promote a recombinant vaccine aimed at reducing hookworm load and Th2 response. potential transmission. Moreover, we show that vaccination Hematophagous helminths require blood as a source of protects against the pathology associated with worm-induced nutrients to mature and reproduce. Female schistosomes blood loss, or hookworm disease. ingest 13 times as many erythrocytes and ingest them about Hookworms bury their anterior ends into the intestinal nine times faster than male worms [39]. Moreover, mRNAs mucosa to feed, secreting anticoagulants to promote blood encoding Hb-degrading proteases of schistosomes are over- flow and stop clot formation at the site of attachment expressed in female worms [40]. Although similar studies have (reviewed in [28]). Numerous anticoagulant peptides have yet to be performed for hookworms, female hookworms are been reported from hookworms [29–31], and their combined bigger than males and lay up to 10,000 eggs per day, implying activities result in ‘‘leakage’’ of blood around the attachment that they have a greater metabolism and therefore greater site and into the host intestine [32]. It is not known whether demand for erythrocytes. Ac-APR-1 degrades Hb in the gut the majority of blood loss during a hookworm infection is due lumen of the worm, and it is therefore not surprising that to leakage around the feeding site or from ingested blood interruption of the function of APR-1 via the action of that enters the parasite’s alimentary canal for nutritional neutralizing antibodies has a deleterious effect on the purposes. To address this, attempts have been made to establishment of worms, particularly females and their measure blood lost from the anus of A. caninum (i.e., blood subsequent egg production. We observed a similar (although that has passed through the parasite’s alimentary canal); not as pronounced) phenomenon when dogs were vaccinated varying calculations have been proposed ranging from 0.14– with the cysteine hemoglobinase, Ac-CP-2, followed by 0.8 ml blood expelled over 24 h per adult worm (reviewed in challenge infection with A. caninum L3 [22]. Vaccination with [32]). Whatever the true figure is, significant blood loss occurs CP-2, however, did not result in reduced adult worm burdens via this route, supporting the hypothesis that vaccination with or reduced blood loss, essential attributes of an efficacious APR-1 damages that parasite’s intestine and results in hookworm vaccine. decreased blood intake (and blood loss) by feeding worms. Vaccination of livestock and laboratory animals with

PLoS Medicine | www.plosmedicine.org1015 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase aspartic proteases of other nematodes, as well as trematode a hemoglobinase [23]. For this reason, we believe that APR-1 helminths, has resulted in antifecundity/antiembryonation is now the major vaccine antigen from the adult stage of the effects. Immunization of sheep with the intestinal brush parasite, and as such, Na-APR-1 should undergo process border complex, H-gal-GP, confers high levels of protection development and enter into Phase I clinical trials as a vaccine (both antiparasite and antifecundity) against H. contortus and for human hookworm infection. This vaccine strategy is now at least three different protease activities, including aspartic being implemented for a larval hookworm antigen, with proteases, have been detected in this extract [16,41]. Phase 1 human trials using ASP-2 formulated with Alhydrogel Immunization of sheep with aspartic protease-enriched already underway [49]. Based on the data reported here, APR- fractions of H. contortus membranes resulted in 36% 1 may also be selected for downstream process development, reduction in adult worms and 48% reduction in fecal egg manufactured under good clinical manufacturing processes, output [17]. Vaccination of sheep with denatured H. contortus and tested in the clinic. proteases or recombinant proteases expressed in bacteria, however, did not confer protection, suggesting that con- Supporting Information formational epitopes are important in protection [17]. Vaccination of mice with recombinant aspartic protease of Accession Numbers The GenBank (http://www.ncbi.nlm.nih.gov/Genbank) accession num- the human blood fluke, Schistosoma mansoni, resulted in 21%– bers for the gene products mentioned in this paper are Ac-APR-1 38% reduction in adult parasites after challenge with (U34888) and Na-APR-1 (AJ245459). infective cercariae; however a reduction in eggs deposited in the liver (the cause of most pathology in schistosomiasis) Acknowledgments was not detected [42]. Protective efficacy of aspartic proteases has been observed against fungal pathogens as well. Vacci- This work was supported by a grant from the Bill and Melinda Gates Foundation awarded to the Sabin Vaccine Institute. AL is supported nation of mice with secreted aspartic proteases of Candida by a Career Development Award from the National Health and albicans, known virulence factors in candidiasis, protected Medical Research Council of Australia. JMB is supported by an animals against a lethal challenge infection and inhibited International Research Scientist Development Award (1K01 colonization of fungi in the kidneys [43]. Moreover, passive TW00009) from the Fogarty Center. For technical assistance and/or helpful advice, we thank Yan Wang, Lilian Bueno, Azra Dobardzic, transfer of serum from vaccinated animals conferred pro- Reshad Dobardzic, Andre Samuel, Sonia Ahn, Aaron Witherspoon, tection, pointing towards an antibody-mediated protective Clay Winters, Estelle Schoch, John Hawdon, and Philip Russell. We mechanism. would like to acknowledge Joe Cohen and Sylvie Cayphas of GlaxoSmithKline Biologicals (Rixensart, Belgium) for providing Almost all of the pathology and morbidity of human AS03 and technical assistance with formulation. hookworm infection results from intestinal blood loss caused by large numbers of adult hookworms. Depending on host References iron and protein stores, a range of hookworm intensities, 1. Hotez PJ, Brooker S, Bethony JM, Bottazzi ME, Loukas A, et al. (2004) Hookworm infection. N Engl J Med 351: 799–807. equivalent to burdens of 40 to 160 worms, is associated with 2. Bethony J, Chen J, Lin S, Xiao S, Zhan B, et al. (2002) Emerging patterns of 1 Hb levels below 11 gdl , the World Health Organization hookworm infection: influence of aging on the intensity of Necator infection threshold for anemia. In Tanzania, Nepal, and Vietnam where in Hainan Province, People’s Republic of China. Clin Infect Dis 35: 1336– 1344. host iron stores are generally depleted, there is a direct 3. Loukas A, Constant SL, Bethony JM (2005) Immunobiology of hookworm correlation between the number of adult hookworms in the infection. FEMS Immunol Med Microbiol 43: 115–124. intestine and host blood loss [1,44]. Therefore the optimal 4. Albonico M, Smith PG, Ercole E, Hall A, Chwaya HM, et al. (1995) Rate of reinfection with intestinal nematodes after treatment of children with hookworm vaccine will be one that either prevents L3 from mebendazole or albendazole in a highly endemic area. Trans R Soc Trop developing into adult blood-feeding hookworms, or one that Med Hyg 89: 538–541. blocks the establishment, survival, and fecundity of the adult 5. Miller TA (1965) Effect of age of the dog on immunogenic efficiency of double vaccination with x-irradiated Ancylostoma caninum larvae. Am J Vet parasites in the intestine [3,45]. Achieving both goals will Res 26: 1383–1390. likely require a vaccine cocktail comprised of an L3 antigen, 6. Miller TA (1965) Influence of age and sex on susceptibility of dogs to primary infection with Ancylostoma caninum. J Parasitol 51: 701–704. such as ASP-2 now under clinical development [46,47], and an 7. Miller TA (1965) Persistence of immunity following double vaccination of adult gut protease, such as APR-1. pups with x-irradiated Ancylostoma caninum larvae. J Parasitol 51: 705–711. An effective hookworm vaccine need not attain 100% 8. Goud GN, Zhan B, Ghosh K, Loukas A, Hawdon JM, et al. (2004) Cloning, yeast expression, isolation and vaccine testing of recombinant Ancylostoma- efficacy. Unlike many unicellular organisms that reproduce secreted protein 1 (ASP–1) and ASP-2 from Ancylostoma ceylanicum. J Infect asexually within the host, nematodes need to sexually Dis 189: 919–929. reproduce. Therefore, small numbers of adult worms will 9. Mendez S, Zhan B, Goud GN, Ghosh K, Dobardzic A, et al. (2005) Effect of combining the larval antigens Ancylostoma secreted protein 2 (ASP-2) and generate fewer eggs to contaminate the environment, and metalloprotease 1 (MTP-1) in protecting hamsters against hookworm subsequently reduce transmission. More importantly, because infection and disease caused by Ancylostoma ceylanicum. Vaccine 23: 3123– hookworms are blood feeders, a partial reduction in adult 3130. 10. Hotez PJ, Ashcom J, Zhan B, Bethony J, Loukas A, et al. (2003) Effect of worm burden equates to a decrease in pathology, notably vaccination with a recombinant fusion protein encoding an astacin-like iron-deficiency anemia [44]. Mathematical modeling of metalloprotease (MTP-1) secreted by host-stimulated Ancylostoma caninum schistosomiasis in China showed that elimination of the third-stage infective larvae. J Parasitol 89: 853–855. 11. Brooker S, Bethony JM, Rodrigues L, Alexander N, Geiger S, et al. (2005) parasite could be attained using an antifecundity vaccine that Epidemiological, immunological and practical considerations in develop- targets egg output with 75% efficacy [48], and it is likely that a ing and evaluating a human hookworm vaccine. Expert Rev Vacc 4: 35–50. 12. Williamson AL, Brindley PJ, Knox DP, Hotez PJ, Loukas A (2003) Digestive similar scenario applies to long-term elimination of soil- proteases of blood-feeding nematodes. Trends Parasitol 19: 417–423. transmitted helminths such as hookworms. An orthologue of 13. Dalton JP, Brindley PJ, Knox DP, Brady CP, Hotez PJ, et al. (2003) Helminth Ac-APR-1 has been reported from the major human hook- vaccines: From mining genomic information for vaccine targets to systems used for protein expression. Int J Parasitol 33: 621–640. worm, N. americanus [23]. Na-APR-1 is structurally and 14. Tort J, Brindley PJ, Knox D, Wolfe KH, Dalton JP (1999) Proteinases and antigenically very similar to Ac-APR-1 and also functions as associated genes of parasitic helminths. Adv Parasitol 43: 161–266.

PLoS Medicine | www.plosmedicine.org1016 October 2005 | Volume 2 | Issue 10 | e295 Vaccination with Hookworm Hemoglobinase

15. Knox DP, Redmond DL, Newlands GF, Skuce PJ, Pettit D, et al. (2003) The biomedical implications of a Schistosoma japonicum complementary DNA nature and prospects for gut membrane proteins as vaccine candidates for resource. Nature Genet 35: 139–147. Haemonchus contortus and other ruminant trichostrongyloids. Int J Parasitol 41. Knox DP, Smith WD (2001) Vaccination against gastrointestinal nematode 33: 1129–1137. parasites of ruminants using gut-expressed antigens. Vet Parasitol 100: 21–32. 16. Knox DP, Skuce PJ, Newlands GF, Redmond DL (2001) Nematode gut 42. Verity CK, McManus DP, Brindley PJ (2001) Vaccine efficacy of peptidases, proteins and vaccination. In: Kennedy MW, Harnett W, editors. recombinant cathepsin D aspartic protease from Schistosoma japonicum. Parasitic nematodes: Molecular biology, biochemistry and immunology. Parasite Immunol 23: 153–162. New York: CAB International. pp. 247–268 43. Vilanova M, Teixeira L, Caramalho I, Torrado E, Marques A, et al. (2004) 17. Smith WD, Skuce PJ, Newlands GF, Smith SK, Pettit D (2003) Aspartyl Protection against systemic candidiasis in mice immunized with secreted proteases from the intestinal brush border of Haemonchus contortus as aspartic proteinase 2. Immunology 111: 334–342. protective antigens for sheep. Parasite Immunol 25: 521–530. 44. Stoltzfus RJ, Dreyfuss ML, Chwaya HM, Albonico M (1997) Hookworm 18. Smith WD, Newlands GF, Smith SK, Pettit D, Skuce PJ (2003) Metal- control as a strategy to prevent iron deficiency. Nutr Rev 55: 223–232. loendopeptidases from the intestinal brush border of Haemonchus contortus 45. Hotez PJ, Zhan B, Bethony JM, Loukas A, Williamson A, et al. (2003) as protective antigens for sheep. Parasite Immunol 25: 313–323. Progress in the development of a recombinant vaccine for human 19. Geldhof P, Claerebout E, Knox D, Vercauteren I, Looszova A, et al. (2002) hookworm disease: The Human Hookworm Vaccine Initiative. Int J Vaccination of calves against Ostertagia ostertagi with cysteine proteinase Parasitol 33: 1245–1258. enriched protein fractions. Parasite Immunol 24: 263–270. 46. Bethony J, Loukas A, Smout M, Brooker S, Mendez S, et al. (2005) 20. Don TA, Jones MK, Smyth D, O’Donoghue P, Hotez P, et al. (2004) A pore- Antibodies against a secreted protein from hookworm larvae reduce the forming haemolysin from the hookworm, Ancylostoma caninum.IntJ intensity of infection in humans and vaccinated laboratory animals. FASEB Parasitol 34: 1029–1035. J. E-pub ahead of print. 21. Williamson AL, Lecchi P, Turk BE, Choe Y, Hotez PJ, et al. (2004) A multi- 47. Goud GN, Bottazzi ME, Zhan B, Mendez S, Deumic V, et al. (2005) enzyme cascade of hemoglobin proteolysis in the intestine of blood-feeding Expression of the Necator americanus hookworm larval antigen Na-ASP-2 in hookworms. J Biol Chem 279: 35950–35957. Pichia pastoris and purification of the recombinant protein for use in human 22. Loukas A, Bethony JM, Williamson AL, Goud GN, Mendez S, et al. (2004) clinical trials. Vaccine 23: 4754–4764. Vaccination of dogs with a recombinant cysteine protease from the 48. Williams GM, Sleigh AC, Li Y, Feng Z, Davis GM, et al. (2002) Mathematical intestine of canine hookworms diminishes the fecundity and growth of modelling of schistosomiasis japonica: Comparison of control strategies in worms. J Infect Dis 189: 1952–1961. the People’s Republic of China. Acta Trop 82: 253–262. 23. Williamson AL, Brindley PJ, Abbenante G, Prociv P, Berry C, et al. (2002) 49. Hotez P, Bethony J, Bottazzi ME, Brooker S, Buss P (2005) Hookworm: ‘‘The ’’ Cleavage of hemoglobin by hookworm cathepsin D aspartic proteases and great infection of mankind. PLoS Med 2: 0177–0181. its potential contribution to host specificity. FASEB J 16: 1458–1460. 24. Hotez PJ, Ashcom J, Bin Z, Bethony J, Williamson A, et al. (2002) Effect of vaccinations with recombinant fusion proteins on Ancylostoma caninum Patient Summary habitat selection in the canine intestine. J Parasitol 88: 684–690. Background Hookworms are parasites of the intestines. They can infect 25. Stoute JA, Slaoui M, Heppner DG, Momin P, Kester KE, et al. (1997) A many animals, including dogs, cats, and people. Worldwide, about one preliminary evaluation of a recombinant circumsporozoite protein vaccine person in five has a hookworm infection. Most of these one billion against Plasmodium falciparum malaria. RTS,S Malaria Vaccine Evaluation Group. N Engl J Med 336: 86–91. people live in tropical countries. Hookworm is not spread from person to 26. Shifrine M, Taylor NJ, Rosenblatt LS, Wilson FD (1978) Comparison of person, because at one stage of its lifecycle, the parasite needs to be in whole blood and purified canine lymphocytes in a lymphocyte-stimulation the soil. In areas where hookworm is common, people who have contact microassay. Am J Vet Res 39: 687–690. with soil that contains human feces are at high risk of infection; because 27. Williamson AL, Brindley PJ, Abbenante G, Datu BJ, Prociv P, et al. (2003) children play on soil and often go barefoot, they have the greatest risk. Hookworm aspartic protease, Na-APR-2, cleaves human hemoglobin and Infection leads to blood loss and a decrease in the amount of iron, and serum proteins in a host-specific fashion. J Infect Dis 187: 484–494. this causes anemia (i.e., because of a lack of iron, the blood cannot carry 28. Bungiro R, Cappello M (2004) Hookworm infection: New developments and oxygen efficiently). There are effective drugs to treat the infection, but prospects for control. Curr Opin Infect Dis 17: 421–426. they do not prevent the patient from becoming re-infected. Making a 29. Stassens P, Bergum PW, Gansemans Y, Jespers L, Laroche Y, et al. (1996) vaccine against hookworm is therefore a priority. Some vaccines for use Anticoagulant repertoire of the hookworm Ancylostoma caninum. Proc Natl in animals have already been developed, but their effectiveness is limited Acad Sci U S A 93: 2149–2154. to one stage of the hookworm’s lifecycle. The aim is to find a vaccine that 30. Cappello M, Vlasuk GP, Bergum PW, Huang S, Hotez PJ (1995) Ancylostoma works against more than one of the stages that the parasite passes caninum anticoagulant peptide: A hookworm-derived inhibitor of human through in its lifecycle. coagulation factor Xa. Proc Natl Acad Sci U S A 92: 6152–6156. 31. Chadderdon RC, Cappello M (1999) The hookworm platelet inhibitor: What Did the Researchers Do and Find? The researchers focused on Functional blockade of integrins GPIIb/IIIa (alphaIIbbeta3) and GPIa/IIa two enzymes the parasite needs in order to live. Building on earlier (alpha2beta1) inhibits platelet aggregation and adhesion in vitro. J Infect research and using a species of hookworm that affects dogs, the Dis 179: 1235–1241. researchers aimed to make these enzymes the ‘‘target’’ of a vaccine. 32. Roche M, Layrisse M (1966) The nature and causes of ‘‘hookworm anemia’’. They first vaccinated dogs, then infected them with hookworm. These Am J Trop Med Hyg 15: 1029–1102. dogs had fewer parasites than dogs that had not been vaccinated. Most 33. Girod N, Brown A, Pritchard DI, Billett EE (2003) Successful vaccination of importantly, vaccinated dogs were protected against blood loss, and BALB/c mice against human hookworm (Necator americanus): The immuno- most did not develop anemia. Laboratory tests confirmed that the target logical phenotype of the protective response. Int J Parasitol 33: 71–80. enzymes had been damaged. 34. Fujiwara RT, Loukas A, Mendez S, Williamson AL, Bueno LL, et al. (2005) Vaccination with irradiated Ancylostoma caninum third stage larvae induces a What Do These Findings Mean? This is the best result so far for a Th2-like response in dogs. Vaccine. In press. hookworm vaccine used in dogs. The authors believe that, as well as 35. de Oliveira Mendes C, Paraguai de Souza E, Borja-Cabrera GP, Maria Melo reducing parasite numbers, the vaccine reduces the ability of the Batista L, Aparecida dos Santos M, et al. (2003) IgG1/IgG2 antibody parasite to take in blood, which would explain the reduction in anemia. dichotomy in sera of vaccinated or naturally infected dogs with visceral The researchers have called for trials to begin with a vaccine targeted leishmaniosis. Vaccine 21: 2589–2597. against similar enzymes in the species of hookworm that most 36. Chabalgoity JA, Moreno M, Carol H, Dougan G, Hormaeche CE (2000) commonly affects humans. Salmonella typhimurium as a basis for a live oral vaccine. Vaccine 19: 460–469. Where Can I Get More Information Online? The US Centers for Disease 37. Boag PR, Parsons JC, Presidente PJ, Spithill TW, Sexton JL (2003) Control have a fact sheet on hookworm: Characterisation of humoral immune responses in dogs vaccinated with http://www.cdc.gov/ncidod/dpd/parasites/hookworm/factsht_hookworm. irradiated Ancylostoma caninum. Vet Immunol Immunopathol 92: 87–94. htm. 38. Pearce EJ, MacDonald AS (2002) The immunobiology of schistosomiasis. The Sabin Vaccine Institute has an overview of the Human Hookworm Nat Rev Immunol 2: 499–511. Vaccine Initiative: 39. Lawrence JD (1973) The ingestion of red blood cells by Schistosoma mansoni.J http://www.sabin.org/hookworm.htm. Parasitol 59: 60–63. 40. Hu W, Yan Q, Shen DK, Liu F, Zhu ZD, et al. (2003) Evolutionary and

PLoS Medicine | www.plosmedicine.org1017 October 2005 | Volume 2 | Issue 10 | e295

Top View