i

Functional study of hemolymph coagulation in Drosophila larvae

Zhi Wang

Doctoral thesis Stockholm 2012

Department of Molecular Biology and Functional Genomics Stockholm University

ii

©Zhi Wang, Stockholm 2012 ISBN 978-91-7447-501-2 Printed in Sweden by Universitetsservice US-AB, Stockholm 2012 Distributor: Stockholm University Library

III

Abstract

Many pathogen infections in nature are accompanied by injury and subsequent coagulation. Despite the contribution of hemolymph coagulation to wound sealing, little is known about its immune function. Based on the molecular knowledge of Drosophila innate immunity, this thesis investigated the immune function of clot both in vitro and in vivo, the immune relevant genes involved in a natural infection model, involving entomopathogenic nematodes (EPN) and the factors leading to crystal cell activation.

Transglutaminase (TG) and its substrate Fondue (Fon) have been identified as bona fide clot components in Drosophila larvae. By knocking down TG or Fon via RNAi, we observed an increased susceptibility to EPN in larvae. In addition, this increased susceptibility was associated with an impaired ability of hemolymph clots to entrap bacteria. Immunostaining revealed that both clot components (Fon and TG) were able to target microbial surfaces. All these data suggest an immune function for the Drosophila hemolymph clot. Strikingly, similar results were obtained when we ran parallel experiments with human FXIIIa, an ortholog of Drosophila TG, indicating a functional conservation. We also found evidence for the regulation on both clot and immunity by eicosanoids in Drosophila larvae.

The combination of EPN infection with the Drosophila model system allowed us to discover an immune function for TEP3 and Glutactin. However the molecular mechanism underlying the involvement of these two proteins in this particular host-pathogen interaction remains to be elucidated.

Prophenoloxidase, the pro-form of enzyme involved in hardening the clot matrix, has been shown to be released by rupture of crystal cells. This cell rupture is dependent on activation of the JNK pathway, Rho GTPases and Eiger. Our work further identified the cytoskeletal component, Moesin, and the cytoskeletal regulator Rac2 as mediators of cell rupture. Despite the possible role of caspases in crystal cell activation, such cell rupture was turned out to be different from apoptosis. The implication of Rab5 in this process indicated that proper endocytosis is required for cell activation and subsequent melanization. Our findings furthered not only our understanding of the release of proPO via cell rupture but also our knowledge on different paths of immune cell activation.

IV

List of papers

This thesis is based on the following publications and manuscript, which are referred to in the text by Roman numerals. Some additional results are presented. I. Pavel Dobes, Zhi Wang, Robert Markus, Ulrich Theopold, Pavel Hyrsl. An improved

method for nematode infection assays in Drosophila larvae. (in press in Journal Fly)

II. Wang Z*, Wilhelmsson C*, Hyrsl P*, Loof TG*, Dobes P, Klupp M, Loseva O, Mörgelin M, Iklé J, Cripps RM, Herwald H, Theopold U. Pathogen entrapment by transglutaminase--a conserved early innate immune mechanism. PLoS Pathog. 2010 Feb 12;6(2):e1000763.

III. Hyrsl P, Dobes P, Wang Z, Hauling T, Wilhelmsson C, Theopold U. Clotting factors and eicosanoids protect against nematode infections. J Innate Immun. 2011;3(1):65-70. Epub 2010 Oct 16.

IV. Zhi Wang, Gawa Bidla, Robert Markus and Ulrich Theopold . Crystal cell activation

through rupture. (Manuscript)

*These author contributed equally to the work Paper not included in the thesis

Waldholm J, Wang Z, Brodin D, Tyagi A, Yu S, Theopold U, Farrants AK, Visa N. SWI/SNF regulates the alternative processing of a specific subset of pre-mRNAs in Drosophila melanogaster. BMC Mol Biol. 2011 Nov 2;12:46.

Reprints were made with permission from the publishers.

V

Abbreviations

AA Arachidonic acid

AMP Antimicrobial peptide

Bsk Basket

Crq Croquemort

COX Cyclooxygenase

Cyt Cytochrome

DAMP Damage-associated molecular pattern

DAP-type meso-diaminopimelic acid-type

DCV Drosophila C virus

DSCAM Down syndrome cell adhesion molecule

dSR-CI Drosophila scavenger receptor- CI

EGF Epidermal growth factor

EPN Entomopathogenic nematode

Fon Fondue

GFP Green fluorescent protein

Glu Glutamine

GNBP Gram-negative binding protein

GRP Glucanases recognition protein

GTPase Guanosine triphosphate hydrolase enzyme

Hml Hemolectin

Ig Immunoglobulin

IJs Infective juveniles

Imd Immunodeficiency

JAK Janus kinases

JNK Jun N-terminal kinase

LOX Lipoxygenase

LPS Lipopolysaccharide

LTA Lipoteichoic acid

Lys Lysine

MP1/MP2 Melanization protein 1 and 2, respectively

NF-kB Nuclear factor-kappa B

PAF Platelet activating factor

PAMP Pathogen-associated molecular pattern

PGN Peptidoglycan

VI

PGRP Peptidoglycan recognition protein

PLA2 Phospholipase A2

PM Peritrophic matrix

PO Phenoloxidase

PPAS Prophenoloxidase activating system

proPO Prophenoloxidase

PRR Pattern recognition receptor

Rel Relish

Rho Ras homology

RNAi Ribonucleic acid (RNA) interference

RNS Reactive nitrogen species

ROS Reactive oxygen species

Serpin Serine protease inhibitor

TEP Thioester containing protein

TF Tissue factor

TG Transglutaminase

TLR Toll-like receptor

TNF Tumor necrosis factor

Tcs Toxin complex

Wt Wild type

7

Contents

Introduction ............................................................................................................................... 9

Insect models to study innate immunity .............................................................................. 10

Insect pathogens and entomopathogenic nematode (EPN) infection .............................. 10

Insect pathogens .......................................................................................................... 10

Entomopathogenic nematode infection ....................................................................... 11

The immune system of insects ........................................................................................ 13

Sensing and recognition – the first step of immune activation .................................... 14

Recognition receptors/molecules ............................................................................. 14

Danger signals ......................................................................................................... 17

Defense mechanisms-elimination of invaders ............................................................. 17

Phagocytosis and encapsulation .............................................................................. 17

Local and systemic induction of AMPs ................................................................... 18

Melanization and Coagulation ................................................................................. 19

Eicosanoids and immunity .......................................................................................... 23

The present study ..................................................................................................................... 25

Aim of this study ................................................................................................................. 25

Results and discussion ......................................................................................................... 26

Improvement of a natural entomopathogenic nematode infection in Drosophila larvae (Paper I) ........................................................................................................................... 26

Immune function of Drosophila clot in vivo (Paper II and paper III) ............................. 27

Recognition Molecules and immune related genes involved in EPN infection in Drosophila larvae ............................................................................................................ 29

Crystal cell activation via rupture (Paper IV) .................................................................. 33

Concluding remarks ................................................................................................................ 36

Acknowledgements ................................................................................................................. 39

References ............................................................................................................................... 41

8

9

Introduction

Multicellular organisms, including vertebrates, invertebrates and plants, are surrounded by a great variety of infectious agents including viruses, bacteria, fungi, protozoa and multicellular parasites. These infectious agents are usually prevented from entering the host body by a combination of physical, chemical and biochemical barriers. Once the barriers are breached by penetrating or through injuries, the infectious agents have the potential to multiply unchecked in their host bodies and cause disease, eventually killing their hosts. However, most of the infections are eliminated and leave little permanent damage. This is due to the immune system, which surveys the whole host body and combats infections.

Infectious microbes differ in size, lifestyle and the way they induce infection, so a wide variety of immune responses are required for the host to cope with all kinds of infections. Any immune response consists of two phases: recognition and elimination. Broadly, the immune responses fall into two categories: innate (or non-adaptive) immune responses and adaptive immune responses. The innate immune system is an ancient form of host defense found in all multicellular organisms (reviewed in (Medzhitov & Janeway 1998; Janeway & Medzhitov 2002)). It relies on germ-line encoded receptors that recognize conserved molecular patterns associated with pathogens (PAMPs for pathogen associated molecular patterns) as non-self (Janeway 1992). This interaction between the host receptors and PAMPs directly induces host defense responses, such as phagocytosis and induction of antimicrobial peptides and reactive oxygen species (ROS) (Janeway 1992), which constitute primary defense mechanisms against infections.

In addition to innate immune responses, vertebrates also rely on their adaptive immune responses, which are mainly mediated by lymphocytes. There are considerable interactions between these two systems. The activated innate immune system releases proinflammatory molecules like cytokines to guide and turn on a series of adaptive immune responses, including lymphocyte division, maturation and antibody secretion (reviewed in (Hoebe et al. 2004)). Unlike the innate immune system, whose recognition receptors are constitutively expressed, the adaptive immune system depends on specific antigen receptors created by a somatic mechanism of gene rearrangement and clonal selection to recognize pathogens. The clonal selection offers vertebrates exquisite specificity and memory, the ability to rapid response when re-exposed to the same pathogen. However, this takes time, which makes adaptive immune responses less effective during the initial defense against infections than innate immune responses (reviewed in (Dushay & Eldon 1998)).

The Immune system is complex. Although many the microbes are recognized as non-self by hosts, not all of them are eliminated by host immune system. For instance, the bacteria inhabiting the animal gut are not entirely eliminated and rather maintained

10

and regulated by the host despite the low induction of antimicrobial peptides in the gut epithelia (Ryu et al. 2008). In some cases, host cells are removed by activating the host immune system when they undergo apoptosis or cancer and are recognized as altered self (reviewed in (Miyake & Yamasaki 2012)). Thus, the immune system is not only a means to combat infectious agents but also the mechanism to maintain homeostasis (Lazzaro & Rolff 2011).

Insect models to study innate immunity

Insects, like other invertebrates, rely entirely on innate immune responses due to the lack of adaptive immune system. Therefore the study of the interactions of insects with microbes provides a powerful tool for discovering early immune reactions without interference with adaptive responses that are present in vertebrate systems.

The study of insect immunity dates back to the late 19 century when Louis Pasteur initiated the study of silkworm disease. Since then the study of infection and immunity in insects has grown and contributed greatly to an increasing knowledge of innate immunity for both invertebrates and vertebrates. The first concept of an inducible antibacterial defense was introduced to the whole immunity field in the 1970s by the pioneering studies in Drosophila melanogaster (Boman et al. 1972). Subsequently, the primary structure of the first antimicrobial peptide (AMP), designated cecropin, was identified in the giant silk moth Hyalophora cecropia (Steiner et al. 1981). This has led to the blooming of the biochemical characterization of antimicrobial peptides and other immune proteins.

Due to the conserved nature of many fundamental aspects of the immune system across all animals, insects have served as valuable models for the study of immune system function and the interaction between pathogen and host. The best example is the discovery of the Toll signaling pathway in Drosophila in 1990s (Lemaitre et al. 1996) , which paved the way for the subsequent identification of the Toll-like receptor pathway in mammals (Poltorak et al. 1998). With this work earning a share of the 2011 Nobel Prize in Physiology or Medicine, the importance of Drosophila as a model system was further emphasized.

Insect pathogens and entomopathogenic nematode (EPN) infection

Insect pathogens Insects are infected by an incredibly large number and diversity of pathogen species. The different types of pathogen include viruses, bacteria, fungi, protozoans, nematodes and parasitoids. They differ in size, lifestyle, and the way of causing infection; even the induced host defense responses are different depending on pathogen species.

11

Insect viruses may be double- or single-stranded DNA (dsDNA and ssDNA, respectively) or double- or single-stranded RNA. In Drosophila, viruses are important natural pathogens and can be transmitted horizontally such as Drosophila C Virus (DCV) or vertically like Sigma virus (reviewed in (Lemaitre & Hoffmann 2007)). They are eradicated by the host via the RNA interference machinery (van Mierlo et al. 2011). Most bacteria gain the entry into their insect hosts by injuries or ingestion. Some bacterial species get into the insect host via special vector-nematodes. Such bacteria are usually associated with their specific nematode species forming a symbiosis. Bacterial endosymbionts like Wolbachia infect their hosts by vertical transmission from a female host to her offspring. In the host body, bacteria usually induce immune responses like the production of antimicrobial peptides or phagocytosis, resulting in clearance of the bacteria. Note that some bacteria such as Mycobacterium marinum benefit from being phagocytosed because the immune cell provides a temporal shelter for them preventing contact with other effector molecules (Dionne et al. 2003).

Unlike viruses and bacteria, entomopathogenic fungi almost always invade their insect hosts by penetrating directly through the cuticle, and eventually the infecting hyphae reach the insect hemocoel. Beauveria bassiana is a widely used fungus in the study of insect immune responses to fungi. Recently an opportunistic human pathogen, Canida albicans has been applied to infect Drosophila (Davis et al. 2011; Glittenberg et al. 2011).

Protozoans are also among animal enemies. Some of them are responsible for serious human diseases, such as malaria, vectored by mosquitoes; and some of them are specific to and causing diseases in insects. The microsporidia, one group of insect protozoans, are mostly investigated as insecticides. They are ingested by insect host and infect host gut epithelial cells, resulting in the debilitation and eventual death of the host (Texier et al. 2010).

In nature, parasitoid wasps parasitize mainly insects. They oviposit into other insects’ host bodies. The suppression of the host defense aids the wasp offspring to complete their development, and the mature adult wasps emerge from the host pupal case, mate and disperse for another infection (Carton & Nappi 2001).

Because some stages of insect life cycle are in a humid environment, there is a considerable risk for contact with nematodes. The parasitic nematodes are also natural pathogens of insects, known as entomopathogenic nematodes (EPNs).

Entomopathogenic nematode infection Entomopathogenic nematodes (EPNs) are soil-dwelling parasitic worms that infect and kill insects (Bednarek & Gaugler 1997). In the life cycle (See figure 1), the non-feeding and developmentally arrested infective juvenile (IJ) is the only stage of EPN found outside of insect host in nature. Infective juveniles (IJs) are tightly associated

12

with their specific symbiotic bacteria inhabiting their intestinal lumen (Poinar 1977). During an infection, IJs enter the insect host through their natural openings such as spiracles, mouth or by penetrating the cuticle, and release the symbiotic bacteria that kill and digest host tissues, thereby providing nutrients for the development of both nematodes and bacteria. There is evidence to show that the symbiotic bacteria are essential to nematodes, since axenic nematodes do not cause insect mortality and do not proliferate inside the insect cadavers (Han & Ehlers 2000). The cues released by insect hosts, such as CO2, are sensed by IJs and guide them to find the host, indicating that IJs have evolved chemosensory mechanisms not only to detect insect hosts, but also to locate the hosts (Ciche 2007; Hallem et al. 2011).

Figure 1. Live cycle of entomopathogenic nematode. Infective juvenile nematodes enter the insect body through natural opening or by injury (right part of figure), then they release their symbiotic bacteria (green dots in lower part) to kill the host. Nematodes reproduce and produce offspring for 2 or 3 generations (left part). Infective juveniles leave the dead insect and seek a new insect host (upper part).

Although EPNs have long been widely applied as biologic control agents for insect pests, particularly the species in the genera Steinernema and Heterorhabditis (Steinernematidae and Heterorhabditae, respectively) (Ehlers 2001), the interactions between EPNs and their hosts, particularly at the level of the host immune response are not fully understood. Early studies on the interaction between Hyalophora cecropia and Neoaplectana carpocapsae/Xenorhabdus nematophilus have found that the nematodes helped the bacteria by excreting an immune inhibitor that selectively destroys immune proteins P5 and P9 (P. Gotz et al. 1981). In Manduca sexta, the nematode Heterorhabditis bacteriophora has been observed to be encapsulated without being (Li 2007). Recently, the EPN infection of Drosophila melanogaster has been performed by Hallem et al., who found that the infection of Drosophila larvae with Heterorhabditis/Photorhabdus resulted in the temporary dynamic expression of AMP genes, a response that was induced specifically by the symbiotic bacterium Photorhabdus. However, Drosophila larvae with defects in the induction of AMPs

13

show the same sensitivity towards EPNs as controls, indicating a dispensable role of AMPs and in fact of both the Imd and Toll pathway in this infection model (Hallem et al. 2007). Accumulating empirical evidences show how the two partners of EPN symbiosis evade or suppress host immune system to achieve a successful infection of insects.

The immune system of insects Insects are constantly exposed to various pathogens, but only a few invaders lead to infections. This is due to a robust and efficient immune system (Figure 2).

Figure 2. Schematic overview of insect immune system. The recognition of pathogens elicits a large array of immune responses, including cellular immune responses and humoral effector mechanisms (see text for more details).

Insect immune tissues comprise immunocytes, lymph gland (the “factory” of immunocytes), epithelia, and fat body which is equivalent to the mammalian liver. Drosophila produces three types of blood cells (hemocytes): plasmatocytes, crystal cells and lamellocytes (reviewed in (Crozatier & Meister 2007)). Plasmatocytes are the majority in the blood or hemolymph and account for 95% of total hemocytes. They are macrophage-like phagocytes and found in all life stages. Crystal cells are only found at embryonic and larval stages comprising 5% of the blood cell population. With crystallized Prophenoloxidase (PPO), the pro-enzyme required in melanin production, crystal cells are the main player in melanization. Lamellocytes differentiate in the lymph gland or from sessile hemocytes at the larval stage upon immune challenge by parasites or melanotic tumors (Sorrentino et al. 2002; Markus et al. 2009). As in vertebrates, Drosophila blood cells also have two developmental

14

origins: one early in the embryonic head mesoderm and another during larval development in the lymph gland (Holz et al. 2003; Crozatier & Meister 2007).

The first line of defense comprises the integument, the peritrophic matrix (PM), a layer composed of chitin and glycoproteins that lines the insect intestinal lumen, and the epithelia covering the outer and inner surfaces. For entry into the host body cavity (hemocoel), pathogens have to breach these barriers. Once in the host body, they are immediately recognized, leading to the activation of host immune cells and tissues and effective immune responses. The binding of plasmatocytes to the invading pathogens through the cell surface recognition receptors leads to the phagocytosis of microbes. Pathogens too big to be engulfed by plasmatocytes like wasp eggs, activate both plasmatocytes and the differentiation of lamellocytes and are eliminated by formation of multicellular layered capsules. In addition to the recognition of pathogens, even signals from damaged host cell can turn on humoral immunity, including melanization, coagulation and production of AMPs (reviewed in (Lemaitre & Hoffmann 2007)). Rather than being independent, immune responses are often interconnected. Some evidences have shown that the depletion of phagocytic plasmatocytes in Drosophila larvae largely increased the susceptibility of host to bacteria (Basset et al. 2000; Brennan et al. 2007). Thus, cellular effectors and humoral effectors work in synergy to mount a successful defense against pathogens.

Sensing and recognition – the first step of immune activation Each immune reaction is initiated by either the recognition of a pathogen or sensing the ‘danger’ due to the presence of pathogens or both. Thus the identification of these cues is essential to understand immunity and has been actively investigated.

Recognition receptors/molecules

The recognition of pathogen is achieved by the interaction of host recognition molecules with microbial conserved patterns, so called PAMPs which are only found on evolutionarily distant organisms like bacteria and absent on eukaryotes (Janeway 1989). PAMPs are essential components of microbial cell wall, including peptidoglycan (PGN) found in both Gram-positive and Gram-negative bacteria, lipopolysaccharide (LPS) from the outer membrane of Gram-negative bacteria, lipoteichoic acid (LTA) of Gram-positive bacteria and fungal β 1, 3-glucans (reviewed in (Royet 2004; Charroux et al. 2009)). The host receptors capable of recognizing or interacting with them are collectively called pattern recognition receptors (PRRs).

Peptidoglycan consists of long glycan chains crosslinked to each other by short peptide bridges which can be separated from glycan chain by cleavage of amidases (reviewed in (Chaput & Boneca 2007)). Peptidoglycan from Gram-negative bacteria differs from most Gram-positive PGN by the replacement of lysine (Lys) with meso-diaminopimelic acid (DAP) at the third position in the peptide bridge (Schleifer &

15

Kandler 1972), which allows host to differentiate different type of bacteria (Paredes et al. 2011). Peptidoglycan recognition proteins (PGRPs) have been shown to interact with PGNs in a wide range of species from insects to mammals, revealing the evolutionary conservation in this family proteins (Yoshida et al. 1996; Sang et al. 2005).

Drosophila contains 13 PGRPs. Some of them lack amidase activity, such as PGRP-SA,-SD, -LC, -LE, -LF. However their conserved PGRP domain is responsible for sensing of bacterial elicitors upstream of the Toll and Imd pathways, which regulate the production of AMPs (reviewed in (Ferrandon et al. 2007; Royet et al. 2011). Study of loss-of-function mutants in PGRP-SA known as Semmelweis reveals its role in activation of the Toll pathway in response to Gram-positive bacteria, but not to fungal infection (Michel et al. 2001). PGRP-LC and –LE function synergistically in the activation of the Imd pathway, since PGRP-LC and –LE (loss of function) double mutants show a more dramatic susceptibility to Gram-negative bacterial infection than either mutation alone(Takehana et al. 2004). In addition to its critical role in the induction of AMP, PGRP-LC is also involved in the microbial recognition in phagocytosis, as its depletion by RNA interference (RNAi) decreases the phagocytosis of E.coli but not S.aureus in S2 cells (Ramet et al. 2002). Other essential function have been attributed to PGRP-LE namely recognition of intracellular Listeria monocytogenes and activation of proPO cascade (Takehana et al. 2004; Yano et al. 2008). Recently, the discovery of the negative regulation of the Imd and JNK pathway by PGRP-LF in flies suggests that it has a function similar to decoy receptors for mammalian cytokine receptors (Maillet et al. 2008; Charroux et al. 2009).

The Drosophila genome encodes six catalytic PGRPs (PGRP-SC1A, -SC1B, -SC2, -LB, -SB1, and –SB2). PGRP–SB1 has been reported to show the amidase activity towards DAP-type PGN (Mellroth & Steiner 2006). Using deletion of six catalytic PGRPs separately or in combination, Lemaitre’s group has uncovered the role of PGRP-LB as a negative regulator of the Imd pathway and a synergy of PGRP-SCs with –LB in the systemic response (Paredes et al. 2011).

Gram-negative binding proteins (GNBPs) and β 1, 3-glucanases recognition protein (βGRP) form another family of PRR. They share sequence similarities with bacterial β 1, 3-glucanases but loss of the enzymatic activity due to the amino-acid substitutions in residues important for catalytic cleavage. Like non-catalytic PGRPs, they retain only the ability to bind to β 1, 3-glucans. (Yahata et al. 1990; Royet 2004). A mutation in the Gram-negative binding protein 1 (GNBP1) gene leads to compromised survival of mutant flies after Gram-positive infections, but not after fungal or Gram-negative bacteria challenge. This demonstrates that GNBP1 and PGRP-SA can jointly activate the Toll pathway (Gobert et al. 2003). The detection of fungal infections relies on both the binding of GNBP3 to fungal cell wall and on cleavage of Persephone by the fungal virulence factor PR1 (Gottar et al. 2006).

16

Like PGRPs, scavenger receptors are conserved across the animal kingdom. It is now becoming clearer that they recognize not only PAMPs but also endogenous ligands from damaged self cells (reviewed in (Pal & Wu 2009)) . In Drosophila, DScr-C1, a homologue of mammalian class C macrophage specific scavenger receptor (SR-C), shows a restricted expression on macrophages or hemocytes during embryonic development and is important for phagocytosis of bacteria (Pearson et al. 1995). Similar to DScr-C1, Croquemort (Crq) is required for phagocytosis by embryonic hemocytes, not of bacteria instead of dying cells (Franc et al. 1999). In mammals, macrophage receptor CD36 has been implicated in recognizing and internalizing Gram-positive bacteria, but its Drosophila homologue, Croquemort (Crq) has so far only been shown to identify and remove apoptotic cells in vivo (Franc et al. 1999; Stuart et al. 2005). In some cases, receptors show exclusive specificity for one type of bacteria, for example Peste, which promotes exclusively the phagocytosis of Mycobaterium fortuitum (Philips et al. 2005). Eater and NimrodC1, both possessing an epidermal growth factor-like motif (EGF-like) are new pattern recognition receptors. They are found on the plasmatocytes surface and play an active role in the removal of apoptotic cells and microorganisms (Kocks et al. 2005; Kurucz et al. 2007).

Thioester containing proteins (TEPs) are highly conserved. Due to the innate immune function of vertebrate complement proteins, belonging to TEPs, insect TEPs are thought to be involved in innate immunity. This has been corroborated by studies in the mosquito Anopheles gambiae where TEP1 (aTEP1) was shown to act as a bona fide opsonin to promote phagocytosis of bacteria (Levashina et al. 2001). The Drosophila genome encodes 6 TEPs. Except TEP5, which does not appear to be expressed, other TEPs have been actively studied in the context of immunity. So far, evidence only showed an up-regulation of TEPs expression upon immune challenge in Drosophila larvae and adult; however mutants in TEPs survived different bacterial infection as well as wild type flies. One possibility is a redundancy in the activity of TEPs (Bou Aoun et al. 2011) .

Lectins are carbohydrate binding proteins and proposed to participate in innate immunity. Some Drosophila C-type lectins (DL2 and DL3) have been identified to agglutinate Gram-negative E. coli to hemocytes in a calcium-dependent manner; and agarose beads showed enhanced encapsulation and melanization by Drosophila hemocytes in vitro after coating with the lectins (Ao et al. 2007). An interesting finding made early highlights the ability to induce cecropin A1 with a hemagglutinin from edible snail in the Drosophila blood cell-line mbn-2 (Theopold et al. 1996).

Down Syndrome Cell Adhesion Molecule (Dscam) is composed of several variable exons flanked by constant exons, which can potentially generate more than 18,000 splice variants of immunoglobulin (Ig)-superfamily receptors (Schmucker et al. 2000). A possible immune function for Dscam was discovered by Watson et al. in 2005. The authors proposed a role of Dscam in phagocytosis based on the observation that

17

hemocyte-specific loss of Dscam impairs the efficiency of phagocytic uptake of bacteria, possibly due to reduced bacterial binding (Watson et al. 2005).

Danger signals In the absence of pathogens, the host immune system is also turned on such as in the case of an aseptic injury. This indicates that not only exogenous pathogens but also endogenous molecules released from damaged self, so called damage-associated molecular patterns (DAMPs) are capable of activating the immune system (reviewed in (Bianchi 2007; Miyake & Yamasaki 2012)). Examples of danger signals are heat shock proteins, reactive oxygen intermediates and extracellular-matrix breakdown products. For example, encapsulation of salivary glands is induced when tumorigenesis occurs by overexpressing of the Ras oncogen in glands in Drosophila larvae (unpublished data from Thomas Hauling). In flies, the turandot (Tot) peptide A are upregulated in fat body and secreted into the hemolymph under severe stress such as high temperature and exposure to oxidative agents (Ekengren et al. 2001). TotA has also been shown to be induced by bacterial challenge (Agaisse et al. 2003). Studies in Galleria mellonella where released collagen fragments and nucleic acid from damaged cell synergize with PAMPs to stimulate an immune response adds additional support to the idea that danger signals play a part in host recognition. It is becoming more accepted that both exogenous signals (PAMPs) and endogenous signals (DAMPs) work together to alarm the host for appropriate immune responses and to maintain host homeostasis.

Defense mechanisms-elimination of invaders

Phagocytosis and encapsulation Phagocytosis, thought to have evolved originally from engulfment of nutrients, is a primary defense mechanism highly conserved across the animal kingdom (reviewed in (Stuart & Ezekowitz 2005)). It is mediated by phagocytic cells, plasmatocytes in Drosophila. Phagocytic receptors on the cell surface (see above) recognize the invading microorganisms or apoptotic cells, followed by cell cytoskeleton remodeling. The ingested particles (phagosomes) fuse with lysosomes to form phagolysosome and finally are degraded. It is noteworthy that despite the increasing list of identified phagocytic receptors, Eater and NimC1 are the only ones for which convincing evidence for their activity has been provided (Kocks et al. 2005; Kurucz et al. 2007).

Bacterial pathogens have devised numerous and diverse strategies to overcome phagocytosis. Most are aimed at blocking one or more of the steps in phagocytosis, thereby halting the process. In EPN infection, the symbiotic bacteria Photorhabdus luminescens suppress phagocytosis by secreting toxin complex (Tcs) to alternate the actin cytoskeleton (reviewed in (Eleftherianos et al. 2010)). For intracellular bacteria, they exploit the engulfment of phagocytes to reach their destination within the cell.

18

Encapsulation is another type of cellular response directed against foreign objects too large to be phagocytosed by individual phagocytes, such as parasitic nematodes, wasp eggs, and host tissues undergoing apoptosis. The recognition of foreign intruders initiates the attachment of plasmatocytes to their surface leading to the formation of several layers; subsequently, the differentiated lamellocytes encompass the multicellular sheath and the parasite is eventually destroyed by the local production of ROSs (Nappi et al. 1995), reactive nitrogen species (RNSs) (Nappi et al. 2000) and melanization(Nappi & Vass 1993). However, some invaders like parasitic nematodes invading Manduca sexta escape from this reaction although parts of the body may be encapsulated (Park et al. 2007). The differentiation and release of lamellocytes is one of the hallmarks in this immune response, therefore an increased number of circulating lamellocytes is observed in tumor inducing Drosophila strains.

Local and systemic induction of AMPs Upon bacterial or fungal infections, antimicrobial peptides (AMPs) are transiently synthesized and serve as self-produced antibiotics. The production of AMPs is mainly regulated by two signaling pathways: the Imd (immune deficiency) and the Toll pathways. This immune response can be activated locally, such as in the epidermis at the site of infection, or systemically, meaning that the synthesis occurs in fat body. The Toll pathway is activated in response to most of Gram-positive bacteria and fungi (reviewed in (Lemaitre & Hoffmann 2007)). Unlike the Toll like receptors (TLRs) in mammals, Toll receptor in Drosophila does not directly bind to PAMPs. Instead it binds to the cleaved extracellular Spätzle, leading to the activation of two Relish factors, the nuclear transcriptional factors (NF-ƘB) Dif and Dorsor (Weber et al. 2003). Normally, Dif and Dorsal form an inactive complex with Cactus, an inhibitor of NF-ƘB. Upon the activation of Toll, the intracellular signal transduction leads to the degradation of Cactus, in turn resulting in the translocation of Dif and Dorsal to the nucleus where genes encoding AMPs like drosomycin and cecropin are expressed (reviewed in (Engstrom 1999)). The immune deficiency (Imd) pathway is turned on preferably by recognition of gram-negative bacteria or in some cases of certain positive bacteria (Leulier et al. 2003) . This recognition is mediated through the receptors PGRP-LC and PGRP-LE (Kaneko et al. 2006). As consequence, the activated receptors signal to Imd protein and leads to the cleavage of another member of NF-ƘB family, Relish (Stoven et al. 2003). The cleaved Relish translocates to the nucleus and activates antibacterial peptide gene expression (reviewed in (Ganesan et al. 2011)). The AMPs like diptericin and attacin induced by this pathway are different from the ones induced by the Toll pathway. The observation that some types of bacteria like S.aureus, L.monocytogens activate both signaling pathways (Hashimoto et al. 2009; Igboin et al. 2012) suggests that there is some functional overlap and synergy in gene regulation between the two them (reviewed in (Igboin et al. 2012)).

19

The Drosophila Toll and Imd pathways share some similarities both at the level of the recognition receptors (Toll) and in their signal transduction (Rel/NF-ƘB) with vertebrates despite differences at some level (reviewed in (Ganesan et al. 2011)). Once again this suggests that immune mechanisms have been conserved through evolution. The AMPs secreted by the fat body are released into the hemolymph 30min to 1h after the onset of infection and persist for several days during which time they target and kill the bacteria and fungi. Most of the AMPs have a broad specificity for both bacteria and even fungi like metchnikowin and cecropins, whereas some exhibit narrow specificity such as drosomycin which is a potent antifungal molecule (reviewed in(Meister et al. 2000)). Using GFP reporter transgenes, Tzou et.al have shown that the antimicrobial peptides can be induced in surface epithelia in a tissue-specific manner, meaning local activation of AMPs (Tzou et al. 2000). More importantly the Imd pathway, rather than the Toll pathway plays a critical role in the activation of this local response to infection. The evidence that drosomycin expression, which is regulated by the Toll pathway during the systemic response, is regulated by the Imd in the respiratory tract speaks for the existence of distinct regulatory mechanisms for local and systemic induction of antimicrobial peptide genes in Drosophila.

Melanization and Coagulation

Melanization Melanization is a specific defense mechanism for insects and many other invertebrates but not found in vertebrates. It is believed to contribute to wound healing, sclerotization and encapsulation (reviewed in (Cerenius et al. 2008)). It is also suggested that PO in Drosophila contributes to the strengthen of teh clot (Bidla et al. 2005).The reaction involves the rapid synthesis and deposition of melanin (brown-black pigments) at the site of infection or injury (reviewed in (Nappi 1973; Eleftherianos & Revenis 2011)). During the synthesis of melanin, the key enzyme phenoloxidase (PO) is activated from its pro-form (proPO) via a serine protease cascade(Kanost et al. 2004). The triggers responsible for proPO activation include the recognition of PAMPs by PGRP-SA or GNBP (Park et al. 2007), tissue damage inflicted by mechanical wounding (Galko & Krasnow 2004) and endogenous elicitors from aberrant tissue like breakdown of basement membrane (Brennan & Anderson 2004; Bidla et al. 2009). Following the stimulation, proPO stored mainly in crystal cells in case of Drosophila, is released into the extracellular milieu via the rupture of crystal cells (see below)(Bidla et al. 2007) and is activated via a serine protease cascade, proPO activation system (PPAS) (Kanost et al. 2004; Wang & Jiang 2004; Tang et al. 2006). Given the toxicity of the products and by-products of the

20

melanization reaction, a tight control of PO-activation is necessary. In Drosophila, one serine protease inhibitor (Spn27) has been identified to inhibit the proPO activating proteases M1 and M2 (De Gregorio et al. 2002; Tang et al. 2006). Mutants of Spn27A prevent uncontrolled melanization (systemic melanization) upon infection or injury, and resulting lethality (De Gregorio et al. 2002).

Although the effects of melanization on parasitoid killing have been well documented in both Drosophila and A. gambiae (Christensen et al. 2005), its effects on the clearance of bacteria and fungi have recently been debated (reviewed in (Cerenius et al. 2008)). Leclerc et.al have shown that Drosophila mutants that fail to activate PO in the hemolymph were as resistant to microbial infections as wild-type flies (Leclerc et al. 2006), whereas the importance of melanization in resistance to infections is supported by the increased susceptibility to fungal and some bacterial infection of the flies with defects in PO activation due to the silencing of serine protease MP1 (CG3066) (Tang et al. 2006; Ayres & Schneider 2008). The increased mortality of the Bc Drosophila mutants, which lack of active PO after aseptic injury, implies PO in wound healing.

PO released via the activation of crystal cells

The Drosophila genome encodes three POs, namely, DoxA1 (Bc/CG5779), DoxA2 (CG2952) and CG8193. Except DoxA2, which is expressed in lamellocytes, the other two POs are expressed in crystal cells (reviewed in (Meister 2004)). Due to the lack of signal peptides it has been unclear for decades how proPO was released into the hemolymph to function extracellularly. Recently crystal cell rupture has been put forward to explain this conundrum. Accompanying cell rupture the crystals, which are believed to contain proPO dissolve. This process is regulated by JNK pathway because the knock down of Bsk (Jun kinase) by RNAi blocks the rupture of crystal cell, resulting in loss of melanization in larval bleeds (Bidla et al. 2007). Moreover, it was found that the constitute activation of a RhoGTPase in blood cells also led to the loss of clot melanization and mutant crystal cells appeared incapable of rupture and dissolving crystals. This implies both the cytoskeleton and small GTPases in the activation of crystal cell.

Coagulation

An ancient defense mechanism Coagulation has historically been regarded as a process separate from the immune responses due to the role it plays in hemostasis upon injury. Only recently has it been proposed that coagulation and innate immunity evolved from a common ancestral immune system (Opal & Esmon 2003). Most animal species, share the involvement of both humoral and cellular mechanism during coagulation, a proteolytic cascade based on zymogen activation, dependence of calcium ions (Ca2+) and a requirement for transglutaminase (TG) dependent crosslinking of clot proteins (reviewed in (Cerenius & Soderhall 2011)). However, except TG and to a limited degree hemolectin from the insect Drosophila, factors participating in invertebrate coagulation lack a vertebrate

21

counterpart, suggesting that coagulation has evolved independently in different species (reviewed in (Jiang & Doolittle 2003; Theopold et al. 2004; Cerenius & Soderhall 2011)).

Since insects have an open circulatory system, coagulation must be activated quickly to prevent excessive loss of body fluid. Upon wounding or triggered by microbes, insect blood cells, mainly granulocytes (Galleria) or plasmatocytes (Drosophila) are attracted (Galko & Krasnow 2004) to the wound sites and undergo degranulation of procoagulants such as hemolectin and GP150 from Drosophila plasmatocytes (Lesch et al. 2007; Lindgren et al. 2008). At the presence of phenoloxidase and TG, cellular clot components are crosslinked to humoral/hemolymph clot proteins which are primarily secreted by fat body, forming a protein meshwork (Fig. 3). In Drosophila, using recombinant Fondue-GFP, the clot, could be labeled with indicating the incorporation of Fondue into clots (Lindgren et al. 2008). Moreover, knockdown of Fondue or TG causes larval bleeding defects (Scherfer et al. 2006; Lindgren et al. 2008), which functionally proves that both Fondue and TG are the true Drosophila lot components. However, the role of PO in coagulation differs between different insect species. The genetic mutation of PO in Drosophila does not affect the soft/primary clot formation, whereas these mutants display wounding defects, suggesting that PO may be not essential for primary clot formation but play a role at later stage in hardening the wound plug (Bidla et al. 2005). In contrast to Drosophila, PO in the mosquito Anopheles appears to be more critical in clot formation since blocking of PO by addition of PTU (PO inhibitor) impedes the procoagulant like lipophorin particles to coalesce into sheet structures (Agianian et al. 2007).

Figure 3. Schematic overview of coagulation in Drosophila and humans. Trauma elicits clotting in both species. In humans, clotting activates two proteolytic cascades (the contact and tissue factor systems), which culminate in the activation of thrombin and FXIII and the formation of a stable clot. In Drosophila larvae, hemolymph clotting also involves a proteolytic cascade (activation of PO) and secretion of clot components like TG and Fondue.

22

Mammalian coagulation is initiated by trauma. First, circulating platelets adhere to the injured site by binding of integrin to collagen exposed on damaged tissue (Clemetson et al. 1999). As a result, activated platelets degranulate and release factors that recruit more platelets aggregating at the wound site and form a cellular plug. This is also known as primary hemostasis (Clemetson 1999). However the complete arrest of bleeding requires the proteolytic blood coagulation cascade leading to the formation of a fibrin clot, a process that is called secondary hemostasis.

During secondary hemostasis, proteolytic cascades are activated by two pathways (fig. 3): contact activation or the intrinsic pathway initiated by damaged host cells or foreign objects and the tissue factor or extrinsic activation pathway initiated by tissue factor from wound site (Spronk et al. 2003). Both of them converge onto a common pathway leading to the activation of thrombin, which in turn converts plasma fibrinogen to fibrin. The loose fibrin mesh is finally thickened and strengthened by crosslinking via FactorXIIIa (FXIIIa), a homologue to Drosophila transglutaminase (TG) (Lorand & Graham 2003; Theopold et al. 2004).

As many infections in nature are concurrent with blood or hemolymph coagulation, it is generally accepted that clotting serves as an immune effector, in some cases, as the mediator of other immune reactions, such as during the reciprocal interaction between complement and coagulation in mammals (reviewed in (Markiewski et al. 2007)). The in vitro evidence from Drosophila strongly supports that coagulation facilitates a localized immune response by immobilizing the pathogens and to some extent killing them (Bidla et al. 2005). Therefore coagulation is integrated into host immunity.

Transglutaminase, the best conserved protein in coagulation As a crosslinking enzyme, TG catalyzes the isopeptide bond formation between lysine (Lys) and glutamine (Glu) residues from appropriated substrates in a Ca2+ dependent manner (Shibata et al. 2010). Among the different clotting factors in both invertebrate and vertebrate coagulation systems, transglutaminase is the only one which is highly conserved (reviewed in (Theopold et al. 2004; Loof et al. 2011)). In mammals, FXIII is converted from pro-transglutaminase by thrombin into activated FXIII in the presence of Ca2+ and fibrin. As a consequence, fibrin fibrils are assembled tightly in a spotwelding-like fashion. The FXIIIa required stabilization of clot structure is very important, because FXIII deficiency and other disorders of stable fibrin meshes can be life threatening (Lorand & Graham 2003; Hsieh & Nugent 2008). Similar to FXIII, TG in horseshoe crabs is not required to form coagulin gel, but the gel is stabilized by its crosslinking of proxin and stablin, two hemocyte derived proteins (Matsuda et al. 2007). Different from the former, in insects like Drosophila and in crustaceans, TG is apparently necessary for producing the clot matrix, which is based on the fact that knockdown of TG in Drosophila or blocking TG activity prevents clot formation (Wang et al. 2001; Lindgren et al. 2008).

23

Eicosanoids and immunity Eicosanoids are a group of biologically active metabolites of certain C20 polyunsaturated fatty acids. The biosynthesis of eicosanoids starts with the hydrolysis of arachidonic acids (AAs) from membrane phospholipid pool via phospholipase A2 (PLA2). Depending on the cell-specific complement of enzymes, the free AA is a substrate for cyclooxygenases (COXs), lipoxygenases (LOXs) and cytochrome P450(Cyt P450) (reviewed in (Stanley et al. 2009)). The biosynthesis of eicosanoids can be blocked by inhibition of PLA2 or of the three pathways.

The regulatory role of eicosanoids in immunity was first observed when inhibitor of eicosanoid biosynthesis were shown to impair the host ability to form microaggregates and nodules following the bacterial injection (Miller et al. 1994). Later, the same group found that eicosanoids mediate the immune reaction to LPS (Bedick et al. 2000) and to fungi (Lord et al. 2002). Besides chemical inhibitors, several bacteria such as Xenorhabdus nematophilus suppress host immunity by blocking eicosanoids biosynthesis (Park et al. 2003). Similarly, prevention of eicosanoids production reduces encapsulation of parasitoids in Drosophila (Carton et al. 2002; Stanley 2011). It is clear that eicosanoids interact with cellular immune responses.

Beyond their contribution to cellular immunity, eicosanoids have also been reported to be functionally coupled to the Imd pathway in Drosophila (Yajima et al. 2003). In mammals, eicosanoids have been demonstrated to regulate blood coagulation (reviewed in (Simmons et al. 2004)). It is tempting to speculate that eicosanoid play a similar role during invertebrate coagulation.

24

25

The present study

The presented work in my thesis is based on the collaboration with my colleagues at Stockholm University and external collaborators during my PhD journey.

Aim of this study

Despite our increased understanding of the molecular bases for Drosophila clot formation, there is no solid evidence from in vivo studies demonstrating the immune function of the Drosophila clot. This study aimed to fill this gap using the genetically tractable model organism, Drosophila melanogaster. The advantages offered by natural infection with entomopathogenic nematode prompted us to apply and improve this infection model in Drosophila larvae (Paper I). Since TG and other proteins like Fondue and GP150 have been identified and confirmed as true components of the Drosophila clot, we have studied their functional role in host immunity in vivo and provided evidence for an immune function of the Drosophila clot (Paper II and III). Additional work has focused on the study of the molecular mechanism by which crystal cells release proPO to contribute to clot formation (Paper IV).

26

Results and discussion

Improvement of a natural entomopathogenic nematode infection in Drosophila larvae (Paper I)

When studying immunity, a common procedure to induce an immune response is to inject microbes and immune elicitors into the host. This can be time consuming and technically tricky, especially in larvae. In the wild, infective juveniles (IJs) of entomopathogenic nematode (EPN) infect insect hosts naturally and release their symbiotic bacteria to kill insects and complete their life cycle (reviewed in (Eleftherianos et al. 2010)). Therefore, EPNs have been applied to study interactions between the insect immune system and insect-pathogenic nematodes and their symbiotic bacteria since the early 1980’s (P. Gotz et al. 1981). However, the lack of genetic knowledge in most of insect model organisms, like Hyalophora cecropia and Maduca sexta, hampers our understanding of the host-pathogen interactions, in particular the host immune responses. The completion of the Drosophila genome sequencing project, microarray analysis as well as the use of genetic screen, have broadened our knowledge of host immune responses (reviewed in (Tzou et al. 2002; Lemaitre & Ausubel 2008)). The aim of this paper was to improve the EPN infection in Drosophila larvae based on the technique established by Hallem et al (Hallem et al. 2007) and to develop a fast and efficient strategy for nematode infection, which is suitable for large-scale screens to identify genes that modulate host-pathogen interactions.

To study the specific host preferences displayed by different nematode species, we compared the infectivity of Steinernema and Heterorhabditis towards Drosophila larvae. Both nematode species displayed the ability to infect and kill Drosophila larvae with the latter being less pathogenic (Fig. 1). Given the different foraging strategies employed by these two species, we next assessed the effect of media (agarose and tissue paper) on nematode infection. Tissue paper appeared more efficient for both nematodes to kill the host larvae (Fig. 2), but only with Steinernema, these effects were significant. The more passive waiting for hosts by Steinernema may require host exposed more often than possible on agarose. Then we tested a key parameter for Heterorhabditis infection, i.e. the dosage of the nematode. Host mortality displayed dose dependence (Fig. 3). To identify negative effects on the immune response against EPNs, 100 IJs/larva is the optimal dose for a single host challenged at room temperature, while a ten times higher concentration may be suitable to identify dominant activators of the immune responses.

Since the improved nematode infection was performed with 96-well plates, which requires individual inspection of the larvae in each well, we tried to improve scoring the infection by mixing 50 host larvae with H.bacteriophora/GFP-expressing P.luminescens in a transparent plastic bag. Due to the GFP labeled symbiotic bacteria, both infected and killed host larvae are scored as GFP positive (Fig. 4 B). In addition

27

to that, GFP-expressing P.luminescens allowed us to monitor the infection in real time. Using the novel assay, we could fully confirm the increase in susceptibility in previously described lines (Fig. 4 A).

So, this newly established infection model in Drosophila larvae will be applicable in large scale screens aimed at identifying novel genes/pathways involved in innate immune responses.

Immune function of Drosophila clot in vivo (Paper II and paper III)

Since many infections in nature are accompanied by breaching body surfaces leading to the hemolymph or blood coagulation, coagulation has been proposed to prevent not only blood loss but also dissemination of infectious agents from entry to the host body (reviewed in (Sun 2006)). In insects, this is supported by the observation that both Galleria and Drosophila clot entrap bacteria in vitro (Bidla et al. 2005). However, evidence from in vivo studies demonstrating an immune function for the Drosophila clot is missing, despite the increased susceptibility to certain bacteria observed in the larvae where the expression of clot component was reduced or lost (Scherfer et al. 2006; Lesch et al. 2007).

To study the immune function of the Drosophila clot in vivo, we focused on previously identified clot components. Since transglutaminase (TG) has been shown to play pivotal role in Drosophila clot formation based on the clotting defects caused by reduced TG level in Drosophila larvae (Lindgren et al. 2008), we wondered whether TG knockdown larvae displayed any defects in wound healing or combating microbial infection. To our surprise, no wounding defects were observed in larvae with reduced TG level (Fig. 5 A; paper II). In contrast, they were more sensitive to pathogenic bacterial infection by S.aureus or P.luminescens but not to E.coli (Fig. 5 B; paper II), which implicates TG in fighting pathogenic bacterial infection. More interestingly, in an EPN natural infection model, TG knockdown larvae succumbed more quickly to H.bacteriophora/P.lum complex while mutants for the Toll or the Imd pathway survived as well as wild type (wt) (Fig. 5 C; paper II), despite the fact that both the Toll and the Imd pathway-dependent AMPs are induced after infection (Hallem et al. 2007). In addition, we tested the contribution of other host immune responses including melanization and phagocytosis to the defense against this natural infection. It appeared that none of these immune reactions were effective. Therefore, it is very likely that this natural infection model revealed a previously uncharacterized immune mechanism, namely the immune function of the Drosophila clot. To further confirm this, we infected knockdown (RNAi) lines for previously identified clotting factors, including Hemomucin, GP150, Fbp1, Tiggrin and Fondue (Schmidt & Theopold 1997; Karlsson et al. 2004; Scherfer et al. 2004; Lindgren et al. 2008). Similar to TG, reduced level of Fondue, or GP150 rendered the host more susceptible to nematode infection (Fig. 1 A and B; paper III). This lends further support for a

28

function of the Drosophila clot in the response against nematodes and their bacteria. Note that the influence of genes that did not show an effect on mortality cannot be excluded, for example due to a partial knockdown.

Given the ability of the clot to entrap microbes, we speculated that the increase in mortality after knocking down clotting factors may be caused by an impaired sequestration of bacteria. Clots from TG-RNAi larvae did in fact entrap fewer bacteria compared to those from wt larvae (Fig. 6 C; paper II), in line with quantification data in Fig. 6 B; paper II. Moreover, a more brittle appearance of clots from TG-RNAi larvae was observed, similar to what was found after inhibition of TG activity by addition of MDC (Lindgren et al. 2008).

TG crosslinks selected glutamines and lysines in proteins leading to ε-(γ-glutamyl) lysine bridges (reviewed in (Lorand & Graham 2003)). Therefore TG may target the proteins from both pathogens and Drosophila itself. Using an antibody recognizing ε-(γ-glutamyl) lysine bridges, we detected TG activity present as a punctuate pattern or aggregates on the surface of zymosan beads that had been mixed with hemolymph (Fig. 1 A; paper II). Such a punctuate pattern was also observed when we replaced the antibody by biotin-cadaverine (B-cad), a small primary amine capable of replacing lysine during TG-mediated crosslinking and which can serve to mark host proteins involved in crosslinking. Other than on the surface of zymosan beads, TG activity was also detected on the surface of both bacteria and nematodes in contact with hemolymph. In all cases, the pattern after B-cad incorporation appeared to localize to small deposits on the microbial surfaces (Fig. 1 B and C; paper II). Using GFP-tagged Fondue, we traced Fondue not only to the clot but also to the surface of P.luminescens as local aggregates (Fig. 2 C-F; paper III). The similarities in the patterns on pathogen surfaces for both clot factors, TG and Fondue, suggest that the same mechanism is active during sequestration of all pathogens.

To learn more about their composition, we performed affinity purification of bacterial lysates after incubation with the biotinylated cadaverine. We identified hexamerin subunits as the major constituent of the aggregates, while less abundant protein components included phenoloxidase and lipophorin (Fig. 2 C; paper II). Taken together, we propose a hypothetic mode for TG mediated immune function of hemolymph clot: upon contact with hemolymph, microbes are almost instantaneously targeted by TG activity and crosslinked to humoral clot factors, leading to the formation of small aggregates and ultimately to sequestration by the clot matrix (Fig. 7; paper II).

Due to the evolutionary similarity between Drosophila TG and human Factor XIII, we speculated that human Factor XIII might play the same role as Drosophila TG. To test our hypothesis, we performed parallel experiments to the ones in Drosophila using human blood. First, we asked whether human Factor XIII was able to target microbes. This was shown to be the case using artificial substrate. Both E.coli and S.aureus were targeted by human Factor XIII in contact with normal human plasma

29

although the reaction did not show the same pattern as on microbial surface as in Drosophila (Fig. 3 A; paper II). Next we found that a fibrin meshwork formed from Factor XIII deficient plasma, but it appeared less crosslinked compared to normal plasma. And the ability to sequester bacteria differed substantially between Factor XIII deficient and normal plasma, with a significant reduction of bacterial entrapment in Factor XIII deficient clot (Fig. 3 B; paper II). Thus, human Factor XIII function similarly as Drosophila TG, by targeting microbial surfaces and crosslinking them to the clot fiber leading to their sequestration. This also indicates a functional conservation between Drosophila TG and human Factor XIII.

Finally, the fact that eicosanoids are involved in regulation of vertebrate blood coagulation (reviewed in (Simmons et al. 2004)) and insect immune responses (reviewed in (Stanley 2011)) made them potential mediators in insect clotting. In the nematode infection model, we detected increased mortality in Drosophila larvae where eicosanoid biosynthesis had been blocked by injecting chemical inhibitors (Fig. 3 A; paper III), strongly suggesting the role of eicosanoids in the defense against nematode infection. Furthermore the injection of a chemical inhibitor was also found to interfere with the targeting of microbial surface by Fondue-GFP (Fig. 3A inset; paper III), which provides evidence for eicosanoids as mediators in insect hemolymph coagulation. Additionally, RNAi knockdown of the phospholipase A2 (CG11124/sPLA2), a potential upstream enzyme in the eicosanoids biosynthesis pathway, also caused high mortality in the nematode infection model (Fig. 3 B; paper III). Thus both genetic and chemical inhibition of eicosanoids biosynthesis impeded the host immune responses against nematodes in Drosophila, which is in line with the observation that some nematode symbiotic bacteria like P.lum evade host immune reactions by suppressing PLA2 in other insect models (Kim et al. 2005).

Taken together we could show for the first time that the TG-dependent clotting system in insects has an important role in the immune response to bacterial infection or during a natural nematode infection. This immune function seems conserved during evolution. Eicosanoids, appear to be involved in host immune responses potentially by regulating the clotting system in Drosophila, although, further investigations into their role in clotting are required.

Recognition Molecules and immune related genes involved in EPN infection in Drosophila larvae

Entomopathogenic nematode infection as a natural infection model probably reflects the natural order of microbe detection better than artificial injection of microbes, which may bypass early stages of microbial encounter. Although the studies from Hallem et.al and ours have shown that both phagocytosis and induction of AMPs appeared not effective in the host defense against EPN, this does not exclude a potential role of recognition molecules in combating EPN infection. To look for the

30

recognition molecules involved in this particular interaction between host and pathogen, we performed EPN infection in mutants for several known pattern recognition receptors (Fig. 4).

Figure 4. Identification of PRRs that contribute to the defense against Heterorhabditis- /Photorhabdus. Mutants of PRRs were infected by Heterorhabditis/Photorhabdus with a dose of 25IJs per larvae at 29 °C and mortality was scored after 48. The double mutant Bc/Imd served as a positive control. Data presented are means ±SD. ** P<0.01.

Mutants of pattern recognition receptors like PGRP-SA, -LC, -LE and GNBP3, which sense directly the bacterial elicitors upstream of the Toll or the Imd pathway seemed not sensitive to H.bacteriophora/P.luminescens complex. This result is in line with the observation that mutants for the Toll and the Imd pathway survived EPN infection as well as Wt animals (Hallem et al. 2007). Surprisingly, a lack of PGRP-LF caused increased mortality of Drosophila larvae. PGRP-LF is a membrane associated PGRP and has been demonstrated to act as a negative regulator of the Imd pathway by interacting with and sequestering PGRP-LCx (Maillet et al. 2008; Basbous et al. 2011). In addition, PGRP-LF negatively regulates the activation of the JNK pathway. Therefore, mutants of PGRP-LF show not only the activation of both the Imd and JNK pathway, but also displayed developmental defects, which made the increased mortality of these mutants in EPN infection more difficult to interpret. More investigations have to been performed to confirm the role of PGRP-LF in the host defense against the H.bacteriophora/P.luminescens complex. Thus far, there is no solid prove that PRRs are required for the host to combat EPN natural infection.

TEPs are complement like proteins with structural similarity to vertebrate complement C3 protein. These genes have been shown to be expressed in Drosophila larvae and adult at a low level, but significantly upregulated after an immune challenge (Lagueux et al. 2000; Bou Aoun et al. 2011). Recent microarray study on the induction of immune related gene upon H.bacteriophora/P.luminescens infection revealed the up-regulation of TEP2 (unpublished work from Michal Zurovec). This prompted us to investigate the involvement of TEPs in activation of immunity mechanisms with nematode infection model.

0

20

40

60

80

100

Mor

talit

y (%

) Wt PGRP-SA PGRP-LC PGRP-LE PGRP-LF GNBP3 BcImd

** **

31

Figure 5. TEP3 contributes to host defense against Heterorhabditis/Photorhabdus. Tep mutants were subjected to the infection by Heterorhabditis/Photorhabdus with a dose of 25IJs per larvae at 29 °C and mortality was scored after 48. Data presented are means ±SD. ** P<0.01. As shown in Fig. 5, the mutation of TEP2 appeared not effective for host in surviving nematode infection despite its up-regulation shown in the microarray data. This is not surprising since the upregulation of TEPs upon immune stimulation may not directly account for their immune function. Contrarily, the TEP3 single mutants and the double mutants for TEP2 and TEP3 (TEP23) displayed a clearly increased susceptibility to nematode infections. To our knowledge this is the first time it could be shown that a Drosophila TEP has an immune function in vivo. Although the exact underlying mechanism remains to be elucidated, based on our previous study of immune function of the clot, we postulate that Drosophila TEP3 may act as part of the clotting system rather than as an opsonin. To test our hypothesis, we will next look into the contribution of TEPs to hemolymph coagulation and to clot components-mediated targeting of microbial surface.

Figure 6. Hemoless larvae are more susceptible to nematode infection. Apoptosis-induced depletion of hemocytes in Drosophila larvae (Hemoless) led to significantly higher mortality 48h after infection with Heterorhabditis/Photorhabdus at a dose of 25IJs per larvae at 29 °C. Data presented are means ±SD. ** P<0.01.

The Immune defects observed in hemocytes depleted Drosophila larvae (Hemoless) reveal the role of hemocytes in multiple aspects of host immune defenses (Charroux & Royet 2009). To evaluate the contribution of hemocytes to the host defense against

0

20

40

60

80

100

Mor

talit

y (%

) W1118 TEP2 TEP3 TEP4 TEP23 TEP234

** **

0

20

40

60

80

100

Mor

talit

y (%

) Wt Hemoless

**

32

nematode, we infected the hemoless larvae where the majority of hemocytes are depleted by induction of apoptosis with the H.bacteriophora/P.luminescens complex. Remarkably, such larvae showed strong susceptibility to nematode infection (Fig. 6). This result raised the question what was missing from the hemocytes to account for the increased mortality. As a reservoir of immune genes, hemocytes express over 2,500 gene transcripts with significantly high hemocyte enrichment (Irving et al. 2005). To look for an answer to this question, we focused on a list of genes with enriched expression in hemocytes including recognition molecules like Eater and Lectins as well as extracellular matrix components such as Glutactin. We performed targeted screen for genes involved using nematode infections. Strikingly, a reduction of several genes’ expression compromised host immunity, leading to higher mortality, although some of the results are at the border of significance and have to be confirmed. The interesting genes comprise those encoding Glutactin, a protein containing leucine rich repeats (CG4950) and an immunoglobulin-like protein (CG7607) (Fig. 7).

Figure 7. Targeted screen for hemocyte enriched genes involved in defense against nematode infection. RNAi lines for hemocyte-enriched genes were crossed with a hemocyte specific driver (HeGAL4) and subjected to the infection by Heterorhabditis/Photorhabdus at a dose of 25IJs per larvae at 29 °C, and mortality was scored after 48. Data presented are means ±SD. ** P<0.01.

These in vivo data may therefore indicate an immune function for the three genes. Glutactin, one of the basement membrane components, has been found associated with the insect body wall, predominantly along the digestive tract, including the proventriculus (a specialization of the anterior alimentary canal), midgut, and hindgut. Since P.luminescens has been shown to occupy the specific niche between

0

20

40

60

80

100

Mor

talit

y (%

)

Wt PGRP-SA PGRP-LC

Galectin Lectin24Db Lectin28C

SPARC Eater SR-CI

Glutactin CG4950(LRR) CG7607(Ig-like)

** ** **

33

extracellular matrix (ECM) and basal membrane of the folded midgut epithelium in Manduca sexta (Silva et al. 2002), we speculate that reduced production of Glutactin in hemocytes in Drosphila larvae might weaken the protection by the ECM or impede entry more generally , resulting in the immune compromised phenotype. It is worth looking into the role of other ECM components like Viking in host immune defense. The function of the other two hits from screen, namely CG4950 and CG7607 are unknown, however the protein domain they contain might indicate their involvement in immunity.

Crystal cell activation via rupture (Paper IV)

Upon wounding or bleeding, Drosophila hemolymph clot is melanized (Bidla et al. 2005). And this melanization is tightly associated with crystal cell rupture, which is regulated by the JNK pathway, RhoGTPases and Eiger (Bidla et al. 2007). The fact that many factors leading to crystal cell activation remain unknown prompted us to look for such factors. By using the melanization of hemolymph clot as readout we analyzed mutants or RNAi lines for defects in this process.

Loss of clot melanization in the larvae overexpressing dominant active Rho1, implicates small RhoGTPase in the process but due to the known side effects of dominant active GTPases, are hard to interpret (Williams et al. 2007). To identify possible small RhoGTPase, we analyzed available classical mutants for Rac1, Rac2 and Mtl. We found that clot melanization was lost in preparation from Rac2 mutants (Fig. 2A, C and C’). This result is in agreement with the loss of melanization on parasitoid eggs in Rac2Δ mutants (Williams et al. 2005). Thus, both findings suggest a role for Rac2 in regulating localized melanization. Interestingly, the loss of control in localized melanization was also reflected in vivo by a less concentrated melanization at wound sites in Rac2Δ mutants, and it was pronounced in a Spn27A mutant background where the inhibition of proPO activating system (PAS) is lost (Fig. S2).

On the other hand, diffuse melanization was observed in the remainder of the hemolymph in Rac2Δ mutants and could be measured photometrically (Fig. 2D). This diffuse melanization was strongly inhibited in a JNK (Bsk) knockdown background (Fig. 2D), adding more supports for the previous finding that crystal cell activation is under control of JNK pathway, moreover, putting Rac2 genetically upstream of JNK in the regulation of crystal cell rupture.

Previously, ectopic expression of the viral inhibitor of apoptosis protein p35 driven by a hemocyte-specific Gal4 led to the inhibition of crystal cell rupture (Fig. 3B), but partial melanization in the clot was observed (Fig. 3A). In this study overexpresssion of p35 specifically in crystal cells caused the complete loss of clot melanization (Fig. 3A). This excludes plasmatocytes as the primary source for the observed defects and

34

indicates that apoptosis-like process may contribute to the activation of crystal cell (Bertin et al. 1996; Kester & Nambu 2011). Next, we tested the role of all proapoptotic proteins and caspases in melanization. Knockdown of two caspases, namely Strica and Decay caused reduced melanization in the clot (Table. S1). The reduced rather than complete loss of clot melanization may be explained by the redundancy of caspases in crystal cell activation or incomplete silencing of caspase gene. In the knockdown of proapoptotic genes by RNAi, we found that both Grim-RNAi and Reaper-RNAi larvae displayed abolished clot melanization, with difference that a diffuse melanization in hemolymph was observed in Grim-RNAi, reminiscent of Rac2 phenotype. This lends support to the idea that deregulation of crystal cell may activate the cell but this activation is not directed towards the clot and leads to diffuse rather than localized melanization.

Since Rho1 is proven involved in both regulation of cytoskeleton (Hall 1998) and activation crystal cell (Bidla et al. 2007), this triggered us to look for potential interactors of Rho1 and cytoskeletal effectors, including Moesin, Rock, Zipper, which might be involved in the same context. It turned out that reduced expression of Moesin (Moe), an anchoring protein between cell membrane and cortex actin (Edwards et al. 1997), caused not only loss of clot melanization and inactivated crystal cells but also rendered the animal more susceptible to aseptic injury (Fig. 4 A, B, G). Thus a functional role of Moe in the regulation of crystal cell activation gets additional support. Using GFP-tagged Moe, we performed live imaging on crystal cell rupture and revealed that there is an apparent increase in cell diameter, right before the cell lysis (Fig. 4C-E). It has been shown that Moe functions antagonistically to the Rho pathway to maintain epithelial integrity (Speck et al. 2003), however this interaction between Rho1 and Moe in regulation of crystal cell rupture remains to be shown.

A recent study by Mukherjee et al revealed the importance of endocytosis to maintain crystal cells integrity in lymphgland (Mukherjee et al. 2011). This finding is consistent with what our observations when we constitutively deactivated Rab5 in crystal cell. Conversely the expression of Rab5 dominant active form specifically in crystal cells led to the accumulation of vesicles (Fig. S4), in line with Rab5’s function in vesicular traffic (Nielsen et al. 2008). As a result, the rupture of crystal cell was disturbed, indicating the proper endocytosis is required for their immune activation.

Melanization requires tight control to keep it local and avoid unwanted or systemic activation (reviwed in (Tang 2009)). This control is achieved by setting up checkpoints at different levels, such as by using serpins which controls the serine proteases cascade dependent PAS (De Gregorio et al. 2002; Ligoxygakis et al. 2002; Gubb et al. 2010). Our findings on the regulators on crystal cell rupture identify another check point to control melanization, namely the regulation of crystal cell rupture. Although there are indications to imply p35 and caspases in melanization, we have not been able to test if crystal cell rupture is has further similarities to apoptosis. It is clear though that crystal cell activation is not the same as mammalian

35

neutrophil activation forming neutrophil extracellular traps (NETs), which involve extracellular DNA fibers (Papayannopoulos & Zychlinsky 2009), since we observe intact nuclei after crystal cell rupture (Fig. 1G-I). It is noteworthy that rupture may not be the exclusive way to activate crystal cell and release proPO, in fact the proPO containing crystals are observed dissolved in intact crystal cells in normal clot preparation and bleeds. They may release proPO towards the targets in case of septicemia via other means like exocytosis.

36

Concluding remarks

In this thesis, I focused mainly on studies of the immune function of the hemolymph clot in Drosophila larvae and crystal cell activation.

Our understanding of the Drosophila clot has gone through different stages, from morphological studies to molecular level, and now this work brought functional insight furthering our knowledge of blood coagulation. For the first time, clot formation as an immune defense mechanism in Drosophila was evident from our in vivo infection studies on clot factors knockdown larvae. Reduced level of some clot factors rendered larvae more sensitive to both bacterial and natural, namely EPN infections. And this immune defect was shown to correlate with impaired bacterial entrapment mediated by clot factors. Studies on two phylogenetically conserved clot components, Drosophila TG and human FXIII, led to the conclusion that the clot acts as a conserved early innate immune mechanism. Using the natural infection model, we also proved the involvement of eicosanoids in Drosophila immunity. Moreover, the fact that blocking of eicosanoinds biosynthesis inhibited clot factor (Fondue) to target microbial surface implicated eicosanoids in the clot’s immune function.

Even though we have made progress in understanding of hemolymph coagulation, some questions are left unanswered. Previous work has shown that entrapped bacteria were eventually killed and this was independent of PO activity. An open question is what in the clot causes the bacterial killing. It would be of interest to find whether clot factors may gain antimicrobial activity after being processed perhaps by proteolytic cleavage. As both Drosophila TG and human FXIII are able to target microbial surfaces, to identify the specific membrane targets on pathogens would contribute to the knowledge of the interaction between host and pathogen. In the mammalian system, blood coagulation interacts constantly with lipids and eicosanoids are proven regulators of blood coagulation. Whether this hold true in insect such as Drosophila remains to be elucidated. Thus far, the clot factors identified in Drosophila larvae are most likely incomplete. This is because the isolation and characterization of clot constituents has been hampered by the rapid nature of clotting. One future task is to complete the list of clot components. This will contribute to a better understanding of the clot as an integral part of innate immunity in both vertebrates and invertebrates.

With the EPN infection model, we have identified TEP3, Glutactin and other two genes (CG4950 and CG7607) that possibly act in the specific defense. These genes are not novel but their role in Drosophila immunity is novel except for TEP3, which had been implied in immune reactions. Due to the sequence similarity with vertebrate complement proteins, insect TEPs have attracted attention for over a decade. However, immune studies primarily found evidence for their up-regulation after immune challenge and the ability to bind to specific bacteria, therefore promoting their phagocytosis by S2 cells (Stroschein-Stevenson et al. 2006; Stroschein-Stevenson et al. 2009). Here we observed immune defects in larvae with loss of TEP3 expression,

37

strongly suggesting a functional role for TEP3 in the response against EPNs. It remains unknown at this stage whether TEP3 is involved in the recognition of the pathogen, in this case P.luminescens, or whether it takes part in the barrier defense. The possible immune function for Glutactin in this study may further our understanding of the extracellular matrix (ECM), particularly whether it contributes to host immunity. The fact that we blocked Glutactin expression in hemocytes, indicates that hemocyte derived ECM components participate host defense. One possible explanation is that hemocytes sourced ECM components as readymade ‘building blocks’ are required for sealing and repairing of wound, even sequestrations of invading microbes. Looking at other ECM components will provide more information on their immune function. It is quite clear that the natural infection model we applied in our studies is able to discover new immune relevant genes. We believe that the combination of EPN infections and the genetically tractable Drosophila system will continue to help us gain a more comprehensive picture of insect immune defense.

Our studies on crystal cell activation provided more information on the regulation of crystal cell rupture, one way to release proPO. We identified cytoskeletal components, Moe, and cytoskeletal regulator Rac2 to as mediators of cell rupture. Following up findings on the inhibition of crystal cell rupture by overexpression of p35, we tested the role of all caspases and proapoptotic proteins in the activation of crystal cell. There was signs of both caspases and proapoptotic proteins taking part in the process, but it is an open question how many features crystal cell rupture shares with apoptosis. Lastly Rab5 was also characterized as a regulator of cell rupture. One piece of important information we learned from this study was that crystal cell activation is not the same as the activation of neutrophil. Although both involve terminal differentiation of cell, the nucleus of crystal cell remains intact while the nucleus of neutrophil burst to release their content during the activation.

Crystal cell rupture was observed to initiate always at one point on cell membrane. It would be interesting to look at the rearrangement of cytoskeleton or depletion of lipids at the initial point upon sensing of the directional signal. It has been shown that Moe regulates Rho1 activity upstream during JNK dependent apoptosis. We would like to know if this Moe and Rho regulated JNK pathway is shared during activation of crystal cell. The answer to this would add to a more comprehensive picture of crystal cell activation.

In summary, this thesis work furthers our understanding of clot’s immune function in both Drosophila and humans. Additionally, the new insight into host-pathogen interaction provided by this study may broaden our understanding of the medical implications of clot, and lead to the new strategies for enhancing the clearance of pathogens.

38

39

Acknowledgements

Many people have contributed to my work with support of many kinds.

First, my sincere gratitude goes to my supervisor, Ulrich Theopold. Thank you so much for your optimism and open mindedness, for being always supportive in trying new ideas and techniques, for your constant encouragement of scientific independence. I have been greatly enjoyed being a member of your research group, which, among other things, has provided me the opportunity to meet many fascinating people. I am also very grateful to you and your wife Gaby for the help with my life here and for the summer BBQ and winter fondue at your place.

I also want to thank Christos Samakovlis for offering me the opportunity to come to Stockholm to pursue my study and for introducing me into fly world.

Thanks Michael, for offering me the one-week training on wasp parasitization, dissection and immunostaining in Aberdeen University.

I express my deep gratitude to past and present lab colleagues: Thomas, for all the laughs we have shared during these years. Being your partner at work has really entertained us and even people around us. I cannot imagine my PhD journey without you. Robert M, for teaching me all the skills in photography and microscopy, and sharing all your PhD experience and wisdom with us. Robert K, for being such an active and young gentleman to remind me how ‘old lady’ I am from time to time. Thank you for being so open to share all your interest and knowledge in science and concerts. I wish you could have joined our group earlier, so we can have more our two-person journal club. Hope you can keep it with other people. Badrul, thank you for the help at my last stay of PhD study, wish you good luck with your PhD study. Maja, your coming gave me the opportunity to talk with girl in the lab and office again. Thank you for showing me how efficiently a woman can work when she becomes mom and for all the discussion in science and life in general. Gawa, for the guidance you gave to me at the beginning and showing me that science can be pursued without protocol to some extent. Christine, it was wonderful experience to share project with you. And thank you for inviting me for meal at your place and meeting your nice family. Tine, for presenting me the importance of being organized and for being so helpful whenever I needed your help. Malin, for showing me the complicated Gateway system when I first joined the group and your expertise in dancing. Also there were many project students who had contributed to the accomplishment of the projects. They are Martina Klupp, Carina Krützmann, Kathrin Handge, Raha Riazi, Ashti Amin Said, Yvonne de Graaf and Marietherese Wuelbern. Thanks for all their work.

Thanks to all the collaborators involved in this thesis. Pavel Hyrsl and Pavel Dobes from Masaryk University, Czech, thank both of you for introducing me to the world of nematodes and for the great time we shared inside and outside lab. Special thanks

40

gave to Pavel H and your wife Allena for guesting me in Brno, that’s really kind of you. Pavel D, for teaching me to pick up the delicious mushrooms. Heiko Herwald,Torsten G. Loof and Matthias Mörgelin from Lund University, for the work on FXIII. Jennifer Iklé and Richard M. Cripps for sending us antibody against Transglutaminase. Olga Loseva, for the technique support for proteomics.

My special appreciations go to the ‘fly people’: Shiva, for sharing your smiles and nice chats. Monica, for giving me so many good advices regarding science and life in Sweden. Widad, for the nice trip to Washington D.C. Anne, thank you for taking off some of the teaching tasks from me. Gunnel, for showing me how to keep our important working space, insect room, in order and sharing the musics from FM107.1. Britta, thank you so much for taking so much work from us and for making our study go smoothly.

Thanks to all the staff from department of molecular biology and functional genomics for the times we shared in our kitchen and for creating the conductive and very friendly working environment. Johan and Simei, it was nice experience to share the project with you.

A ‘big thanks’ to all my friends outside the lab. Your smiles, greetings and caring sent by phone calls or emails, information and advices made my stay in Sweden one of the best time period in my life.

My deepest gratitude to my family in China. 感谢妈妈， 爸爸，姐姐，姐夫还有甜

甜对我的坚持不懈的信任和支持。 尤其感谢姐姐姐夫这么多年来在家对妈妈的

照顾。谢谢我所有的亲戚对我一直以来的关心和爱护。 感谢我的公公婆婆对我

们在外生活的体谅和对我由衷的关爱。

Finally, I am so grateful to my husband, Liang, for always being such a considerate man and having faith in me. Thank you for your endless love of me and of our little family. Thanks for your coming, my little daughter to be, you will be the biggest treasure in the world for me. Love both of you so much!

41

References

Agaisse H, Petersen UM, Boutros M et al. (2003) Signaling role of hemocytes in Drosophila JAK/STAT-dependent response to septic injury. Developmental cell 5, 441-450.

Agianian B, Lesch C, Loseva O et al. (2007) Preliminary characterization of hemolymph coagulation in Anopheles gambiae larvae. Developmental and comparative immunology 31, 879-888.

Ao J, Ling E, Yu XQ (2007) Drosophila C-type lectins enhance cellular encapsulation. Molecular immunology 44, 2541-2548.

Ayres JS, Schneider DS (2008) A signaling protease required for melanization in Drosophila affects resistance and tolerance of infections. PLoS biology 6, 2764-2773.

Basbous N, Coste F, Leone P et al. (2011) The Drosophila peptidoglycan-recognition protein LF interacts with peptidoglycan-recognition protein LC to downregulate the Imd pathway. EMBO reports 12, 327-333.

Basset A, Khush RS, Braun A et al. (2000) The phytopathogenic bacteria Erwinia carotovora infects Drosophila and activates an immune response. Proceedings of the National Academy of Sciences of the United States of America 97, 3376-3381.

Bedick JC, Pardy RL, Howard RW et al. (2000) Insect cellular reactions to the lipopolysaccharide component of the bacterium Serratia marcescens are mediated by eicosanoids. Journal of insect physiology 46, 1481-1487.

Bednarek A, Gaugler R (1997) Compatibility of soil amendments with entomopathogenic nematodes. Journal of nematology 29, 220-227.

Bertin J, Mendrysa SM, LaCount DJ et al. (1996) Apoptotic suppression by baculovirus P35 involves cleavage by and inhibition of a virus-induced CED-3/ICE-like protease. Journal of virology 70, 6251-6259.

Bianchi ME (2007) DAMPs, PAMPs and alarmins: all we need to know about danger. Journal of leukocyte biology 81, 1-5.

Bidla G, Dushay MS, Theopold U (2007) Crystal cell rupture after injury in Drosophila requires the JNK pathway, small GTPases and the TNF homolog Eiger. Journal of cell science 120, 1209-1215.

Bidla G, Hauling T, Dushay MS et al. (2009) Activation of insect phenoloxidase after injury: endogenous versus foreign elicitors. Journal of innate immunity 1, 301-308.

Bidla G, Lindgren M, Theopold U et al. (2005) Hemolymph coagulation and phenoloxidase in Drosophila larvae. Developmental and comparative immunology 29, 669-679.

Boman HG, Nilsson I, Rasmuson B (1972) Inducible antibacterial defence system in Drosophila. Nature 237, 232-235.

Bou Aoun R, Hetru C, Troxler L et al. (2011) Analysis of thioester-containing proteins during the innate immune response of Drosophila melanogaster. Journal of innate immunity 3, 52-64.

Brennan CA, Anderson KV (2004) Drosophila: the genetics of innate immune recognition and response. Annual review of immunology 22, 457-483.

Brennan CA, Delaney JR, Schneider DS et al. (2007) Psidin is required in Drosophila blood cells for both phagocytic degradation and immune activation of the fat body. Current biology : CB 17, 67-72.

Carton Y, Frey F, Stanley DW et al. (2002) Dexamethasone inhibition of the cellular immune response of Drosophila melanogaster against a parasitoid. The Journal of parasitology 88, 405-407.

Carton Y, Nappi AJ (2001) Immunogenetic aspects of the cellular immune response of Drosophilia against parasitoids. Immunogenetics 52, 157-164.

42

Cerenius L, Lee BL, Soderhall K (2008) The proPO-system: pros and cons for its role in invertebrate immunity. Trends in immunology 29, 263-271.

Cerenius L, Soderhall K (2011) Coagulation in invertebrates. Journal of innate immunity 3, 3-8.

Chaput C, Boneca IG (2007) Peptidoglycan detection by mammals and flies. Microbes and infection / Institut Pasteur 9, 637-647.

Charroux B, Rival T, Narbonne-Reveau K et al. (2009) Bacterial detection by Drosophila peptidoglycan recognition proteins. Microbes and infection / Institut Pasteur 11, 631-636.

Charroux B, Royet J (2009) Elimination of plasmatocytes by targeted apoptosis reveals their role in multiple aspects of the Drosophila immune response. Proceedings of the National Academy of Sciences of the United States of America 106, 9797-9802.

Christensen BM, Li J, Chen CC et al. (2005) Melanization immune responses in mosquito vectors. Trends in parasitology 21, 192-199.

Ciche T (2007) The biology and genome of Heterorhabditis bacteriophora. WormBook : the online review of C elegans biology, 1-9.

Clemetson JM, Polgar J, Magnenat E et al. (1999) The platelet collagen receptor glycoprotein VI is a member of the immunoglobulin superfamily closely related to FcalphaR and the natural killer receptors. The Journal of biological chemistry 274, 29019-29024.

Clemetson KJ (1999) Primary haemostasis: sticky fingers cement the relationship. Current biology : CB 9, R110-112.

Crozatier M, Meister M (2007) Drosophila haematopoiesis. Cellular microbiology 9, 1117-1126.

Davis MM, Alvarez FJ, Ryman K et al. (2011) Wild-type Drosophila melanogaster as a model host to analyze nitrogen source dependent virulence of Candida albicans. PloS one 6, e27434.

De Gregorio E, Han SJ, Lee WJ et al. (2002) An immune-responsive Serpin regulates the melanization cascade in Drosophila. Developmental cell 3, 581-592.

Dionne MS, Ghori N, Schneider DS (2003) Drosophila melanogaster is a genetically tractable model host for Mycobacterium marinum. Infection and immunity 71, 3540-3550.

Dushay MS, Eldon ED (1998) Drosophila immune responses as models for human immunity. American journal of human genetics 62, 10-14.

Edwards KA, Demsky M, Montague RA et al. (1997) GFP-moesin illuminates actin cytoskeleton dynamics in living tissue and demonstrates cell shape changes during morphogenesis in Drosophila. Developmental biology 191, 103-117.

Ehlers RU (2001) Mass production of entomopathogenic nematodes for plant protection. Applied microbiology and biotechnology 56, 623-633.

Ekengren S, Tryselius Y, Dushay MS et al. (2001) A humoral stress response in Drosophila. Current biology : CB 11, 714-718.

Eleftherianos I, ffrench-Constant RH, Clarke DJ et al. (2010) Dissecting the immune response to the entomopathogen Photorhabdus. Trends in microbiology 18, 552-560.

Eleftherianos I, Revenis C (2011) Role and importance of phenoloxidase in insect hemostasis. Journal of innate immunity 3, 28-33.

Engstrom Y (1999) Induction and regulation of antimicrobial peptides in Drosophila. Developmental and comparative immunology 23, 345-358.

Ferrandon D, Imler JL, Hetru C et al. (2007) The Drosophila systemic immune response: sensing and signalling during bacterial and fungal infections. Nature reviews Immunology 7, 862-874.

Franc NC, Heitzler P, Ezekowitz RA et al. (1999) Requirement for croquemort in phagocytosis of apoptotic cells in Drosophila. Science 284, 1991-1994.

43

Galko MJ, Krasnow MA (2004) Cellular and genetic analysis of wound healing in Drosophila larvae. PLoS biology 2, E239.

Ganesan S, Aggarwal K, Paquette N et al. (2011) NF-kappaB/Rel proteins and the humoral immune responses of Drosophila melanogaster. Current topics in microbiology and immunology 349, 25-60.

Glittenberg MT, Silas S, MacCallum DM et al. (2011) Wild-type Drosophila melanogaster as an alternative model system for investigating the pathogenicity of Candida albicans. Disease models & mechanisms 4, 504-514.

Gobert V, Gottar M, Matskevich AA et al. (2003) Dual activation of the Drosophila toll pathway by two pattern recognition receptors. Science 302, 2126-2130.

Gottar M, Gobert V, Matskevich AA et al. (2006) Dual detection of fungal infections in Drosophila via recognition of glucans and sensing of virulence factors. Cell 127, 1425-1437.

Gubb D, Sanz-Parra A, Barcena L et al. (2010) Protease inhibitors and proteolytic signalling cascades in insects. Biochimie 92, 1749-1759.

Hall A (1998) Rho GTPases and the actin cytoskeleton. Science 279, 509-514. Hallem EA, Dillman AR, Hong AV et al. (2011) A sensory code for host seeking in parasitic

nematodes. Current biology : CB 21, 377-383. Hallem EA, Rengarajan M, Ciche TA et al. (2007) Nematodes, bacteria, and flies: a tripartite

model for nematode parasitism. Current biology : CB 17, 898-904. Han R, Ehlers RU (2000) Pathogenicity, development, and reproduction of Heterorhabditis

bacteriophora and Steinernema carpocapsae under axenic in vivo conditions. Journal of invertebrate pathology 75, 55-58.

Hashimoto Y, Tabuchi Y, Sakurai K et al. (2009) Identification of lipoteichoic acid as a ligand for draper in the phagocytosis of Staphylococcus aureus by Drosophila hemocytes. J Immunol 183, 7451-7460.

Hoebe K, Janssen E, Beutler B (2004) The interface between innate and adaptive immunity. Nature immunology 5, 971-974.

Holz A, Bossinger B, Strasser T et al. (2003) The two origins of hemocytes in Drosophila. Development 130, 4955-4962.

Hsieh L, Nugent D (2008) Factor XIII deficiency. Haemophilia : the official journal of the World Federation of Hemophilia 14, 1190-1200.

Igboin CO, Griffen AL, Leys EJ (2012) The Drosophila melanogaster host model. Journal of oral microbiology 4.

Irving P, Ubeda JM, Doucet D et al. (2005) New insights into Drosophila larval haemocyte functions through genome-wide analysis. Cellular microbiology 7, 335-350.

Janeway CA, Jr. (1989) Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harbor symposia on quantitative biology 54 Pt 1, 1-13.

Janeway CA, Jr. (1992) The immune system evolved to discriminate infectious nonself from noninfectious self. Immunology today 13, 11-16.

Janeway CA, Jr., Medzhitov R (2002) Innate immune recognition. Annual review of immunology 20, 197-216.

Jiang Y, Doolittle RF (2003) The evolution of vertebrate blood coagulation as viewed from a comparison of puffer fish and sea squirt genomes. Proceedings of the National Academy of Sciences of the United States of America 100, 7527-7532.

Kaneko T, Yano T, Aggarwal K et al. (2006) PGRP-LC and PGRP-LE have essential yet distinct functions in the drosophila immune response to monomeric DAP-type peptidoglycan. Nature immunology 7, 715-723.

Kanost MR, Jiang H, Yu XQ (2004) Innate immune responses of a lepidopteran insect, Manduca sexta. Immunological reviews 198, 97-105.

44

Karlsson C, Korayem AM, Scherfer C et al. (2004) Proteomic analysis of the Drosophila larval hemolymph clot. The Journal of biological chemistry 279, 52033-52041.

Kester RS, Nambu JR (2011) Targeted expression of p35 reveals a role for caspases in formation of the adult abdominal cuticle in Drosophila. The International journal of developmental biology 55, 109-119.

Kim Y, Ji D, Cho S et al. (2005) Two groups of entomopathogenic bacteria, Photorhabdus and Xenorhabdus, share an inhibitory action against phospholipase A2 to induce host immunodepression. Journal of invertebrate pathology 89, 258-264.

Kocks C, Cho JH, Nehme N et al. (2005) Eater, a transmembrane protein mediating phagocytosis of bacterial pathogens in Drosophila. Cell 123, 335-346.

Kurucz E, Markus R, Zsamboki J et al. (2007) Nimrod, a putative phagocytosis receptor with EGF repeats in Drosophila plasmatocytes. Current biology : CB 17, 649-654.

Lagueux M, Perrodou E, Levashina EA et al. (2000) Constitutive expression of a complement-like protein in toll and JAK gain-of-function mutants of Drosophila. Proceedings of the National Academy of Sciences of the United States of America 97, 11427-11432.

Lazzaro BP, Rolff J (2011) Immunology. Danger, microbes, and homeostasis. Science 332, 43-44.

Leclerc V, Pelte N, El Chamy L et al. (2006) Prophenoloxidase activation is not required for survival to microbial infections in Drosophila. EMBO reports 7, 231-235.

Lemaitre B, Ausubel FM (2008) Animal models for host-pathogen interactions. Current opinion in microbiology 11, 249-250.

Lemaitre B, Hoffmann J (2007) The host defense of Drosophila melanogaster. Annual review of immunology 25, 697-743.

Lemaitre B, Nicolas E, Michaut L et al. (1996) The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 86, 973-983.

Lesch C, Goto A, Lindgren M et al. (2007) A role for Hemolectin in coagulation and immunity in Drosophila melanogaster. Developmental and comparative immunology 31, 1255-1263.

Leulier F, Parquet C, Pili-Floury S et al. (2003) The Drosophila immune system detects bacteria through specific peptidoglycan recognition. Nature immunology 4, 478-484.

Levashina EA, Moita LF, Blandin S et al. (2001) Conserved role of a complement-like protein in phagocytosis revealed by dsRNA knockout in cultured cells of the mosquito, Anopheles gambiae. Cell 104, 709-718.

Li XC, RS.; Cowles, EA.; Gaugler, R.; Cox-Foster, DL.. (2007) Relationship between the successful infection by entomopathogenic nematodes and the host immune response. Int J Parasitol 37, 10.

Ligoxygakis P, Pelte N, Ji C et al. (2002) A serpin mutant links Toll activation to melanization in the host defence of Drosophila. The EMBO journal 21, 6330-6337.

Lindgren M, Riazi R, Lesch C et al. (2008) Fondue and transglutaminase in the Drosophila larval clot. Journal of insect physiology 54, 586-592.

Loof TG, Schmidt O, Herwald H et al. (2011) Coagulation systems of invertebrates and vertebrates and their roles in innate immunity: the same side of two coins? Journal of innate immunity 3, 34-40.

Lorand L, Graham RM (2003) Transglutaminases: crosslinking enzymes with pleiotropic functions. Nature reviews Molecular cell biology 4, 140-156.

Lord JC, Anderson S, Stanley DW (2002) Eicosanoids mediate Manduca sexta cellular response to the fungal pathogen Beauveria bassiana: a role for the lipoxygenase pathway. Archives of insect biochemistry and physiology 51, 46-54.

45

Maillet F, Bischoff V, Vignal C et al. (2008) The Drosophila peptidoglycan recognition protein PGRP-LF blocks PGRP-LC and IMD/JNK pathway activation. Cell host & microbe 3, 293-303.

Markiewski MM, Nilsson B, Ekdahl KN et al. (2007) Complement and coagulation: strangers or partners in crime? Trends in immunology 28, 184-192.

Markus R, Laurinyecz B, Kurucz E et al. (2009) Sessile hemocytes as a hematopoietic compartment in Drosophila melanogaster. Proceedings of the National Academy of Sciences of the United States of America 106, 4805-4809.

Matsuda Y, Osaki T, Hashii T et al. (2007) A cysteine-rich protein from an arthropod stabilizes clotting mesh and immobilizes bacteria at injury sites. The Journal of biological chemistry 282, 33545-33552.

Medzhitov R, Janeway CA, Jr. (1998) An ancient system of host defense. Current opinion in immunology 10, 12-15.

Meister M (2004) Blood cells of Drosophila: cell lineages and role in host defence. Current opinion in immunology 16, 10-15.

Meister M, Hetru C, Hoffmann JA (2000) The antimicrobial host defense of Drosophila. Current topics in microbiology and immunology 248, 17-36.

Mellroth P, Steiner H (2006) PGRP-SB1: an N-acetylmuramoyl L-alanine amidase with antibacterial activity. Biochemical and biophysical research communications 350, 994-999.

Michel T, Reichhart JM, Hoffmann JA et al. (2001) Drosophila Toll is activated by Gram-positive bacteria through a circulating peptidoglycan recognition protein. Nature 414, 756-759.

Miller JS, Nguyen T, Stanley-Samuelson DW (1994) Eicosanoids mediate insect nodulation responses to bacterial infections. Proceedings of the National Academy of Sciences of the United States of America 91, 12418-12422.

Miyake Y, Yamasaki S (2012) Sensing necrotic cells. Advances in experimental medicine and biology 738, 144-152.

Mukherjee T, Kim WS, Mandal L et al. (2011) Interaction between Notch and Hif-alpha in development and survival of Drosophila blood cells. Science 332, 1210-1213.

Nappi AJ (1973) Hemocytic changes associated with the encapsulation and melanization of some insect parasites. Experimental parasitology 33, 285-302.

Nappi AJ, Vass E (1993) Melanogenesis and the generation of cytotoxic molecules during insect cellular immune reactions. Pigment cell research / sponsored by the European Society for Pigment Cell Research and the International Pigment Cell Society 6, 117-126.

Nappi AJ, Vass E, Frey F et al. (1995) Superoxide anion generation in Drosophila during melanotic encapsulation of parasites. European journal of cell biology 68, 450-456.

Nappi AJ, Vass E, Frey F et al. (2000) Nitric oxide involvement in Drosophila immunity. Nitric oxide : biology and chemistry / official journal of the Nitric Oxide Society 4, 423-430.

Nielsen E, Cheung AY, Ueda T (2008) The regulatory RAB and ARF GTPases for vesicular trafficking. Plant physiology 147, 1516-1526.

Opal SM, Esmon CT (2003) Bench-to-bedside review: functional relationships between coagulation and the innate immune response and their respective roles in the pathogenesis of sepsis. Crit Care 7, 23-38.

P. Gotz, and AB, Boman HG (1981) Interactions between Insect Immunity and an Insect-Pathogenic Nematode with Symbiotic Bacteria. Proc R Soc Lond B 212, 18.

Pal S, Wu LP (2009) Pattern recognition receptors in the fly: lessons we can learn from the Drosophila melanogaster immune system. Fly 3, 121-129.

Papayannopoulos V, Zychlinsky A (2009) NETs: a new strategy for using old weapons. Trends in immunology 30, 513-521.

46

Paredes JC, Welchman DP, Poidevin M et al. (2011) Negative regulation by amidase PGRPs shapes the Drosophila antibacterial response and protects the fly from innocuous infection. Immunity 35, 770-779.

Park JW, Kim CH, Kim JH et al. (2007) Clustering of peptidoglycan recognition protein-SA is required for sensing lysine-type peptidoglycan in insects. Proceedings of the National Academy of Sciences of the United States of America 104, 6602-6607.

Park Y, Kim Y, Putnam SM et al. (2003) The bacterium Xenorhabdus nematophilus depresses nodulation reactions to infection by inhibiting eicosanoid biosynthesis in tobacco hornworms, Manduca sexta. Archives of insect biochemistry and physiology 52, 71-80.

Pearson A, Lux A, Krieger M (1995) Expression cloning of dSR-CI, a class C macrophage-specific scavenger receptor from Drosophila melanogaster. Proceedings of the National Academy of Sciences of the United States of America 92, 4056-4060.

Philips JA, Rubin EJ, Perrimon N (2005) Drosophila RNAi screen reveals CD36 family member required for mycobacterial infection. Science 309, 1251-1253.

Poinar GOJ, Thomas, G.M., and Hess, R. (1977) Characteristics of the specific bacterium associated with Heterorhabditis baceriophora heterorhabditidae rhabditida. Nematologica 21, 8.

Poltorak A, He X, Smirnova I et al. (1998) Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282, 2085-2088.

Ramet M, Manfruelli P, Pearson A et al. (2002) Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature 416, 644-648.

Royet J (2004) Infectious non-self recognition in invertebrates: lessons from Drosophila and other insect models. Molecular immunology 41, 1063-1075.

Royet J, Gupta D, Dziarski R (2011) Peptidoglycan recognition proteins: modulators of the microbiome and inflammation. Nature reviews Immunology 11, 837-851.

Ryu JH, Kim SH, Lee HY et al. (2008) Innate immune homeostasis by the homeobox gene caudal and commensal-gut mutualism in Drosophila. Science 319, 777-782.

Sang Y, Ramanathan B, Ross CR et al. (2005) Gene silencing and overexpression of porcine peptidoglycan recognition protein long isoforms: involvement in beta-defensin-1 expression. Infection and immunity 73, 7133-7141.

Scherfer C, Karlsson C, Loseva O et al. (2004) Isolation and characterization of hemolymph clotting factors in Drosophila melanogaster by a pullout method. Current biology : CB 14, 625-629.

Scherfer C, Qazi MR, Takahashi K et al. (2006) The Toll immune-regulated Drosophila protein Fondue is involved in hemolymph clotting and puparium formation. Developmental biology 295, 156-163.

Schleifer KH, Kandler O (1972) Peptidoglycan types of bacterial cell walls and their taxonomic implications. Bacteriological reviews 36, 407-477.

Schmidt O, Theopold U (1997) Helix pomatia lectin and annexin V, two molecular probes for insect microparticles: possible involvement in hemolymph coagulation. Journal of insect physiology 43, 667-674.

Schmucker D, Clemens JC, Shu H et al. (2000) Drosophila Dscam is an axon guidance receptor exhibiting extraordinary molecular diversity. Cell 101, 671-684.

Shibata T, Ariki S, Shinzawa N et al. (2010) Protein crosslinking by transglutaminase controls cuticle morphogenesis in Drosophila. PloS one 5, e13477.

Silva CP, Waterfield NR, Daborn PJ et al. (2002) Bacterial infection of a model insect: Photorhabdus luminescens and Manduca sexta. Cellular microbiology 4, 329-339.

Simmons DL, Botting RM, Hla T (2004) Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition. Pharmacological reviews 56, 387-437.

47

Sorrentino RP, Carton Y, Govind S (2002) Cellular immune response to parasite infection in the Drosophila lymph gland is developmentally regulated. Developmental biology 243, 65-80.

Speck O, Hughes SC, Noren NK et al. (2003) Moesin functions antagonistically to the Rho pathway to maintain epithelial integrity. Nature 421, 83-87.

Spronk HM, Govers-Riemslag JW, ten Cate H (2003) The blood coagulation system as a molecular machine. BioEssays : news and reviews in molecular, cellular and developmental biology 25, 1220-1228.

Stanley D (2011) Eicosanoids: progress towards manipulating insect immunity. J Appl Entomol 135, 12.

Stanley D, Miller J, Tunaz H (2009) Eicosanoid actions in insect immunity. Journal of innate immunity 1, 282-290.

Steiner H, Hultmark D, Engstrom A et al. (1981) Sequence and specificity of two antibacterial proteins involved in insect immunity. Nature 292, 246-248.

Stoven S, Silverman N, Junell A et al. (2003) Caspase-mediated processing of the Drosophila NF-kappaB factor Relish. Proceedings of the National Academy of Sciences of the United States of America 100, 5991-5996.

Stroschein-Stevenson SL, Foley E, O'Farrell PH et al. (2006) Identification of Drosophila gene products required for phagocytosis of Candida albicans. PLoS biology 4, e4.

Stroschein-Stevenson SL, Foley E, O'Farrell PH et al. (2009) Phagocytosis of Candida albicans by RNAi-treated Drosophila S2 cells. Methods Mol Biol 470, 347-358.

Stuart LM, Deng J, Silver JM et al. (2005) Response to Staphylococcus aureus requires CD36-mediated phagocytosis triggered by the COOH-terminal cytoplasmic domain. The Journal of cell biology 170, 477-485.

Stuart LM, Ezekowitz RA (2005) Phagocytosis: elegant complexity. Immunity 22, 539-550. Sun H (2006) The interaction between pathogens and the host coagulation system.

Physiology (Bethesda) 21, 281-288. Takehana A, Yano T, Mita S et al. (2004) Peptidoglycan recognition protein (PGRP)-LE and

PGRP-LC act synergistically in Drosophila immunity. The EMBO journal 23, 4690-4700.

Tang H (2009) Regulation and function of the melanization reaction in Drosophila. Fly 3, 105-111.

Tang H, Kambris Z, Lemaitre B et al. (2006) Two proteases defining a melanization cascade in the immune system of Drosophila. The Journal of biological chemistry 281, 28097-28104.

Texier C, Vidau C, Vigues B et al. (2010) Microsporidia: a model for minimal parasite-host interactions. Current opinion in microbiology 13, 443-449.

Theopold U, Samakovlis C, Erdjument-Bromage H et al. (1996) Helix pomatia lectin, an inducer of Drosophila immune response, binds to hemomucin, a novel surface mucin. The Journal of biological chemistry 271, 12708-12715.

Theopold U, Schmidt O, Soderhall K et al. (2004) Coagulation in arthropods: defence, wound closure and healing. Trends in immunology 25, 289-294.

Tzou P, De Gregorio E, Lemaitre B (2002) How Drosophila combats microbial infection: a model to study innate immunity and host-pathogen interactions. Current opinion in microbiology 5, 102-110.

Tzou P, Ohresser S, Ferrandon D et al. (2000) Tissue-specific inducible expression of antimicrobial peptide genes in Drosophila surface epithelia. Immunity 13, 737-748.

van Mierlo JT, van Cleef KW, van Rij RP (2011) Defense and counterdefense in the RNAi-based antiviral immune system in insects. Methods Mol Biol 721, 3-22.

48

Wang R, Liang Z, Hal M et al. (2001) A transglutaminase involved in the coagulation system of the freshwater crayfish, Pacifastacus leniusculus. Tissue localisation and cDNA cloning. Fish & shellfish immunology 11, 623-637.

Wang Y, Jiang H (2004) Prophenoloxidase (proPO) activation in Manduca sexta: an analysis of molecular interactions among proPO, proPO-activating proteinase-3, and a cofactor. Insect biochemistry and molecular biology 34, 731-742.

Watson FL, Puttmann-Holgado R, Thomas F et al. (2005) Extensive diversity of Ig-superfamily proteins in the immune system of insects. Science 309, 1874-1878.

Weber AN, Tauszig-Delamasure S, Hoffmann JA et al. (2003) Binding of the Drosophila cytokine Spatzle to Toll is direct and establishes signaling. Nature immunology 4, 794-800.

Williams MJ, Habayeb MS, Hultmark D (2007) Reciprocal regulation of Rac1 and Rho1 in Drosophila circulating immune surveillance cells. Journal of cell science 120, 502-511.

Williams MJ, Sutherland WH, McCormick MP et al. (2005) Aged garlic extract improves endothelial function in men with coronary artery disease. Phytotherapy research : PTR 19, 314-319.

Yahata N, Watanabe T, Nakamura Y et al. (1990) Structure of the gene encoding beta-1,3-glucanase A1 of Bacillus circulans WL-12. Gene 86, 113-117.

Yajima M, Takada M, Takahashi N et al. (2003) A newly established in vitro culture using transgenic Drosophila reveals functional coupling between the phospholipase A2-generated fatty acid cascade and lipopolysaccharide-dependent activation of the immune deficiency (imd) pathway in insect immunity. The Biochemical journal 371, 205-210.

Yano T, Mita S, Ohmori H et al. (2008) Autophagic control of listeria through intracellular innate immune recognition in drosophila. Nature immunology 9, 908-916.

Yoshida H, Kinoshita K, Ashida M (1996) Purification of a peptidoglycan recognition protein from hemolymph of the silkworm, Bombyx mori. The Journal of biological chemistry 271, 13854-13860.